Sedation Therapy - CareFusion · pain management and sedation in intensive care patients. This...

Conference report published by carefusion.com

Center for Safety and Clinical ExcellenceSan Diego, CA

2005

Sedation Therapy: Improving Safety and

Quality of CareProceedings from

The Sixth Conference

Center for Safety and Clinical Excellence

November 17-18, 2005, San Diego, CA

Philip J. Schneider, MS, FASHP, Editor

Issues and Opportunities

Assessment of Sedation/Delirium/Pain

Sedation Management: Medications/Techniques

Administration and Monitoring: New Technologies

Inter-professional Conference on Pain Management and Sedation

The sixth conference at the Center for Safety and Clinical Excellence

in San Diego, held November 17-18, 2005, brought together a

distinguished faculty from clinical practice, academia, organizations and

government. Philip J. Schneider, MS, FASHP, Director of the Latiolais

Leadership Program and Clinical Professor at The Ohio State University,

chaired the conference. Nationally recognized experts from different health

professions focused on current issues and opportunities in postoperative

pain management and sedation in intensive care patients. This document

summarizes the information presented on sedation therapy with regard to

safety concerns, criteria for determining best practices, guidelines,

assessment, nursing issues, and new administration and monitoring

technologies that can help improve safety and quality of care. A second

document summarizes the presentations and roundtable discussion of topics

related to pain management and patient-controlled analgesia.

1 Executive Summary Conference Report

Contents Editorial

P3 Asleep at the WheelPhilip J. Schneider, MS, FASHP, The Ohio State University

Issues and Opportunities

P4 Issues and Opportunities in Sedation and Pain Management: Perspectives from the Anesthesia Patient Safety Foundation

Robert Stoelting, MD, APSF

Assessment of Sedation/Delirium/Pain

P6 Use of Sedation Assessment Scales in the ICU:Paying Attention to What We Are Doing

Curtis Sessler, MD, Virginia Commonwealth University

P10 Assessment and Treatment of Delirium in the ICUBrenda Truman Pun, RN, MSN, ACNP, Vanderbilt University Medical Center

P14 Managing Pain in the ICU PatientKathleen Puntillo, RN, DNSC, FANN, University of California at San Francisco

P18 The Role of Brain Monitoring in the ICUMichael Ramsay, MD, Baylor Health Care System

Sedation Management: Medications/Techniques

P21 Clinical Pharmacology of Sedatives in the Critically IllBradley A. Boucher, PharmD, University Of Tennessee Health Science Center

P25 SCCM Guidelines for Sedation Management: Development and Use in Clinical Practice

Judith Jacobi, PharmD, Methodist Hospital / Clarian Health

P29 Nursing Issues Related to Sedation ManagementAnne Pohlman, RN, MSN, CCRN, University of Chicago

P32 Daily Interruption of Sedation in Mechanically Ventilated Patients: The ICU Sedation Vacation

John P. Kress, MD, University of Chicago

P35 ICU Sedation with Propofol: Measuring and Reducing VariationRay Maddox, PharmD, St. Joseph’s/Candler Health System

Administration and Monitoring: New Technologies

P39 Automation in Critical Care: Future Directions in Sedation DeliveryGeoff Shaw, MD, Christchurch, New Zealand

P43 Target-controlled Infusion: IntroductionMichel Struys, MD, University of Ghent, Belgium

P46 Target-controlled Infusions for ICU and Procedural SedationAnthony Absalom, MD, Addenbrookes Hospital, Cambridge, England

P51 Physiology of Oxygen and Carbon Dioxide MonitoringBhavani-Shankar Kodali, MD, Brigham and Women’s Hospital


6th Conference–Sedation Therapy: Improving Safety and Quality of Care

EDITORIAL

The expression “asleep at the

wheel” means one has fallen asleep

while driving a vehicle, or “behind the

wheel.” It can also describe someone

who is blissfully unaware of a dan-

gerous situation. As we heard and

discussed the presentations at the

6th Conference of the Center for

Safety and Clinical Excellence

Improvement, I was struck by my

own lack of awareness of the dan-

gerous situation that seems to arise

more commonly than I would have

thought with the use of drugs for

sedating patients. Technology that

makes it possible to more closely

monitor patients who are being

treated with sedation medications

for either procedures or agitation

have shown that respiratory depres-

sion is much more common that pre-

viously thought. This is especially

true in procedure areas and med-

ical/surgical units outside the inten-

sive care unit, where patients may be

unattended for longer periods of

time.

While Stoelting points out that we

do not always have randomized con-

trolled studies to provide evidence

for all medical practices, we do have,

as Bouchard presents, a wonderful

array of drugs that can be used to

sedate patients. Jacobi summarizes

guidelines for the safe and effective

use of these medications based on

the opinions of experts. Sessler

describes sedation assessment

scales that can be used by clinicians

to monitor the patient’s response to

treatment. Ramsay points out the

value of using these scales to

improve the use of sedatives.

Maddox, Shaw, Struys, and Absalom

describe existing and developing

technologies for the more effective

administration of these medications.

Yet, we still have problems.

There may be several reasons for

this. In discussing these presenta-

tions, several points were made.

First, there was a debate over the

importance of evidence-based med-

icine. Some participants felt that

there needed to be randomized con-

trolled trials before any guidelines or

technologies are adopted as a stan-

dard of practice. Without standardi-

zation, variability in practice exists,

which increases the chances of

harm. A second issue was the current

problems in the workforce. The

nursing shortage was cited as one

barrier to quality of care. Proper edu-

cation on the use of sedatives, scales

to assess the patient response, and

training to use new drug administra-

tion and monitoring technologies

are critical to achieving the desired

outcomes.

The components to effective

sedation management are available.

What is missing is execution. Just

like driving a car, the use of potent

medications requires our full atten-

tion. A better understanding of the

drugs used to sedate patients, the

tools used to assess patients, and

effective use of the technologies that

more effectively deliver the doses of

sedatives and monitor patient

response is needed. The information

presented at this conference and

contained in these proceedings is an

effective contribution to improving

sedation therapy.

Asleep at the Wheel Philip J. Schneider, MS, FASHP, Clinical Professor and Director, Latiolais Leadership Program, College of Pharmacy,

The Ohio State University, Columbus, OH


PROCEEDINGS

Issues and Opportunities in Sedation and Pain Management:

Perspectives from the Anesthesia Patient Safety Foundation

Robert K. Stoelting, MD, President, APSF, Indianapolis, IN

Although significant progress hasbeen made in making the periopera-tive-period experience safer, adverseevents still occur. There are stillopportunities to make changes andimprovements that translate intobetter perioperative care andimproved patient safety. This confer-ence is an opportunity to discusssafety concerns associated with post-operative pain management andsedation in the intensive care unit(ICU), and the criteria that should beused to determine best practices forimproving patient safety.

Key Points

• Opportunities exist to make significant changes and improvements in perioperative care that can translate into better patient care and improved patient safety.

• In addressing such concerns, the criteria to determine best practices for improving patient safety need to be considered.

• Safety changes reflect decreases in the incidence of rare events, so that randomized controlled trials to determine efficacy of interventions intended to prevent rare events are not always possible or needed to endorse new patient safety practices.

• Anesthesia safety was achieved by making changes that made sense, were based on an understanding of human factors principles, and had been documented to be effective in other settings.

• A common theme in these safety changes is they made sense and were the right thing to do.

Postoperative PainManagement

In the Summer 2005 issue of theAnesthesia Patient Safety Foundation(APSF) Newsletter, Frank J. Overdyk,MD, wrote that more deaths havebeen reported to the Food and DrugAdministration with patient-con-trolled analgesia (PCA) than with allother intravenous infusions com-bined.1 In this article, he alsoexpressed the opinion that thou-sands of patients are at risk for opi-oid-induced depression of ventila-tion. It was suggested that routine

patient monitoring with pulse oxime-try and capnography would preventmore than 90% of PCA-associatedcomplications – a significant oppor-tunity for improved patient safety. Anunanswered question is who (physi-cians, nurses, respiratory therapists)should monitor PCA infusions?

Sedation in the ICU

Sedation in the ICU may beaccomplished with multiple drugsoften administered in combinationwith or without infusion devices.There is currently great interest inusing propofol to provide sedationbased on its unique pharmacokinet-ics (rapid onset and prompt recov-ery). Disadvantages of propofolinclude its dose-related depressionof ventilation and circulation and itsability to produce general anesthe-sia. Concern that administration ofpropofol by individuals not trained inthe management of anesthetizedpatients may result in adverse eventsis a controversial topic. Developmentof infusion devices that administerpropofol based on its pharmacoki-netics combined with pulse oximetry and capnography presentsa potential opportunity to improvepatient safety.

Executive Summary Conference Report 4


Issues and Opportunities in Sedation and Pain Management:Perspectives from the Anesthesia Patient Safety Foundation

Achievement of Conference Goals

Achievement of the goals of thisconference will depend on the appli-cation of principles shown by otherorganizations such as APSF to be effec-tive in achieving improved patientsafety. These principles include(1) practitioner education and dissem-ination of information (expert confer-ences, educational publications), (2) research (medical and technology), (3) industry participation and support,(4) specialty society endorsement(standards), (5) support of accredit-ing organizations (Joint Commissionon Accreditation of HealthcareOrganizations), and (6) mobilizationof public opinion via the media.

Adoption of Safety Changes

An overriding question is “whatcriteria should be used to determinebest practices for improving patientsafety?”2 Safety changes reflectdecreases in the incidence of rareevents. Controlled and randomizedstudies to determine efficacy of inter-ventions intended to prevent rareevents are not always possible orneeded to endorse new patient safe-ty practices. In fact, there will neverbe complete evidence for everythingthat must be done in medicine toimprove patient safety. The prudentalternative to evidence-based data isto make reasonable judgmentsbased on the best available evidencecombined with successful experi-ences in health care.

Evidence-based Medicine

Rigorous proof of efficacy is nei-ther necessary nor in many cases suf-ficient for recommending wide-spread use of a safety practice.2

Although continued outcomesresearch is needed to establish andimprove the scientific evidence basefor medical practice, formal evidencemay neither be appropriate or essen-tial for all of the interventions need-ed to improve patient safety.

Aviation safety was not built onevidence that certain practicesreduced the frequency of crashes.Aviation relied on widespread imple-mentation of hundreds of smallchanges in procedures, equipment,training and organization that aggre-gated to establish a strong safety cul-ture and effective practices. Thesechanges made sense, were usuallybased on sound principles, technicaltheory or experience, and addressedreal-life problems, but few were sub-jected to controlled experiments.

In health care, the progress inanesthesia safety is similar to that inaviation. The achievement ofimproved anesthesia patient safety isnot attributable to a single practiceor development of new anestheticdrugs or even any type of improve-ment (such as technologicaladvances) but to application of abroad array of changes in process,equipment, organization, supervi-sion, training, and teamwork.2 No sin-gle one of these changes has been

proven to have a clear-cut impact onmortality. Anesthesia safety wasachieved by applying many changesthat made sense, were based on anunderstanding of human factorsprinciples, and had been shown to beeffective in other settings. A commontheme in all these safety changes isthey made sense and were the rightthing to do. To say that convincing evidence of progress and effect is lacking because randomized trials of all safe anesthesia practices have not been conducted would be counterproductive.

Summary

Development and ultimateachievement of the safety opportuni-ties in postoperative pain manage-ment and sedation in the ICU willrequire persistence and patience. It iscritical to recognize that evidence-based data are not necessary forinstituting patient safety changes.There is an opportunity to mergetechnology with clinical care inachieving the goals of this confer-ence. Ultimately, dissemination ofthe results of this conference will becritical in educating health professionals who will ultimately be responsible for advocating safety changes.

References1. Overdyk FJ. PCA presents serous risks. Letter. APSF

Newsletter. Summer 2005:33 (www.apsf.org)

2. Leape LL, Berwick DM, Bates DW. What practices will most improve safety? Evidence-based medicine meets patient safety. JAMA 2002;228:501-7.


PROCEEDINGS

Use of Sedation Assessment Scales in the ICU: Paying Attention to What We Are Doing

Curtis N. Sessler, MD, Orhan Muren Professor of Medicine, Division of Pulmonary and Critical Care Medicine, Virginia Commonwealth University Health System, Medical Director of Critical Care,

Medical College of Virginia Hospitals, Richmond, VA

Monitoring the level of conscious-ness of critically ill patients receivingsedative and analgesic medicationsis an important component of rou-tine patient care in the intensive careunit (ICU).1,2 This is important to satis-fy the goals of minimizing pain, dis-comfort, and intolerance of ICU-relat-ed therapy while avoiding excessiveor unnecessarily prolonged sedation.There are important differences inclinical characteristics amongpatients that may influence sedationmanagement and the desired depthof sedation, and the sedation targetmust be adjusted as the individualpatient’s condition evolves.

Key Points

• Monitoring the level of consciousness of critically ill patients receiving sedative and analgesic medications is important to minimize pain and dis-comfort and to avoid excessive or unnecessarily prolonged sedation.

• A sedation goal should be established that is regularly redefined, and avalidated assessment scale used to measure the level of sedation.

• Validated sedation assessment scales include the Ramsay Sedation Scale(RSS), Sedation Agitation Scale (SAS), Motor Activity Assessment Scale(MAAS), Vancouver Interactive and Calmness Scale (VICS), RichmondAgitation-Sedation Scale (RASS), Adaptation to Intensive Care Environment(ATICE) scale, and Minnesota Sedation Assessment Tool (MSAT).

• The use of sedation scales improves communication among clinicians,reduces the frequency of over-sedation, and aids in implementation of successful sedation management strategies.

Monitoring the Level of Consciousness:Assessment Scales

Among the various tools availablefor monitoring the level of conscious-ness, sedation assessment scales arethe most widely used. Expert con-sensus recommendations includethe routine use of a validated seda-tion assessment tool and establish-ing a sedation goal that should beregularly redefined.1 The assessmentof sedation and response to therapyshould also be systematically docu-mented. In spite of these guidelines,in a recent study of United States aca-

demic medical centers, a sedationscale was used in only 55% ofmechanically ventilated patients.

For a sedation assessment scale tobe used successfully, clinicians mustbe confident that it measures what itintends to, is reliable, and is easy touse. Desirable features includea) development by a multidiscipli-nary group that is representative ofthose who use and interpret it, e.g.,nurses, physicians, and pharmacists,b) logical structure to the scale suchthat it adequately covers the fullrange of responses in a graded fash-ion, c) rapid to perform and easy torecall and interpret, d) well-definedand mutually exclusive criteria foreach level such that there is littlelikelihood of overlap or confusion,e) sufficient sedation levels to permitmedication titration to various end-points as a patient’s conditionchanges, f ) an assessment ofagitation, and g) rigorous testing ofreliability and validity in the patientpopulation for which the scale isintended to be used.3

When one examines the varioussedation assessment scales, thedomains that are to be evaluated andthe structure of the instrumentshould be considered. The principle




domain of most scales is conscious-ness, with the range of responsesextending from alert to comatose.The most commonly measured sub-domain within the domain of con-sciousness is arousal or awakeness,often in response to stimuli ofincreasing intensity. In some instru-ments higher states of consciousnessare further defined by testing cogni-tion or comprehension. A sufficientnumber of sedation levels are need-ed in the consciousness domain, sothat sedative medications can betitrated more precisely to the desiredstate of consciousness.

A second important domain is agi-tation or, conversely, calmness, typi-cally anchored at opposite ends by acalm state and by frankly combativebehavior. Whereas agitation is rela-tively infrequent (5-10%) if a group ofpatients is examined at a single pointin time,4 as many as 70% of patientswill display agitated behavior atsome point during their ICU stay.5

Agitation is important to recognizesince it may reflect inadequatelytreated pain, unrecognized delirium,or other problems such as angina or amal-positioned endotracheal tube.Many conditions and medicationscan contribute to the developmentof delirium. Recognition and docu-mentation of agitation may lead toimproved management. Somerecent sedation assessment instru-ments broaden the range of assess-ments to include tolerance of the ICUenvironment, such as patient-ventila-tor synchrony.6

There are several sedation assess-ment scales that have been validatedin adult ICU populations (Table).

Although all scales include somemeasure of arousal or awakeness,there are differences in the domainsexamined, the construction of theinstrument, and the extent of testingof validity and inter-rater reliability.

Ramsay Sedation Scale (RSS)

In 1974, Ramsay and colleaguespublished the RSS, designed to sub-jectively grade level of sedation in aclinical trial of sedative agents.7

Although not validated nor tested forinter-rater reliability until recently,this scale has been and continues tobe widely used. This 6-level scaleincludes 4 levels of sedation or arous-al (levels 3-6), a “cooperative, orient-ed, and tranquil” level (2), and a sin-gle level for an anxious, agitated, orrestless state (1). A light-to-moderatelevel of sedation is depicted byresponse to spoken word in a singlelevel (3), and deeper levels of seda-tion are established by examiningthe response to physical stimulation(a light glabellar tap), ranging frombrisk (4), to sluggish (5), to none (6) ina sleeping patient. Inter-rater reliabil-ity has recently been shown to beexcellent, and RSS correlates wellwith other sedation scales.4,8

Sedation Agitation Scale (SAS)

In 1994, Riker and co-workerspublished the SAS that was subse-quently modified to its current statusin 1999, with an emphasis onexpanding the assessment of agita-tion.9,10 This 7 level scale (1-7) includes3 levels of agitation (5-7), a “calm andcooperative” level (4), and 3 levels ofsedation (1-3). In contrast to RSS, thesedation level is established by satisfying one or more measures ofarousal and cognition. The SAS hasbeen extensively validated in differ-ent patient populations10-12 and iswidely used.

Motor Activity AssessmentScale (MAAS)

The motor activity assessmentscale (MAAS), also published in 1999,has structure nearly identical to SASwith 3 levels for agitation (4-6), onelevel for “calm and cooperative”, and 3levels for sedation (0-2).13 Like SAS,similar criteria are used for the vari-ous sedation levels, although thespecific criteria differ. The use of sev-eral criteria for each level for thesescales raises the likelihood that some

Table. Sedation Assessment Scales Validated in Adult ICU Populations*

Ramsay Sedation Scale (RSS)

Sedation Agitation Scale (SAS)

Motor Activity Assessment Scale (MAAS)

Vancouver Interactive and Calmness Scale (VICS)

Richmond Agitation-Sedation Scale (RASS)

Adaptation to Intensive Care Environment (ATICE)

Minnesota Sedation Assessment Tool (MSAT)

* Listed in order of publication



patients may satisfy criteria frommultiple levels (and not all criteriafrom any one level), and might impairease of recall from memory.

Vancouver Interactive andCalmness Scale (VICS)

The structure of the VICS, pub-lished in 2000, differs in that separatescores for “interaction” and “calm-ness” domains are established bysumming the caregiver grading of aseries of questions.14 For eachdomain, there are 5 questions towhich the caregiver gives one of 6responses, varying from “stronglyagree” (1 point) to “strongly disagree”(6 points). Patient stimulation isrequired to address some questions.The scores for the questions aresummed with higher scores (maxi-mum 30) corresponding to highinteraction and calmness. Whiledemonstrating good reliability andface and construct validity, the com-plexity of calculation may present achallenge for bedside application.

Richmond Agitation-Sedation Scale (RASS)

In 2002, Sessler and colleaguesintroduced the RASS, a 10-point scalethat addresses both agitation (scoresof +1 to +4) and consciousness(scores of -1 to -5) with a “0”score cor-responding to a calm and alert state.4

Agitation scores range from “anxious’(+1) to frankly combative (+4), basedupon observation. Deeper levels ofsedation are established using 2additional steps. First the subject’sresponse to the tester’s verbalinstructions to “open your eyes andlook at me” is noted. If the subjectopens their eyes (arousal) and follows

the command to look in their eyes(cognition), they receive a score of -1if the actions are sustained > 10 sec(with prompting) or -2 if < 10 sec. Ascore of -3 is assigned if there is eyeopening, or any other movement tothe verbal command, but making eyecontact is not performed (arousalwithout cognition). If there is noresponse to verbal stimulation,observation of any response to phys-ical stimulation (shoulder shakingfollowed by sternal rub if no responseto shoulder shaking), a score of -4 isassigned, or absence of any move-ment a score of -5 is assigned. Therehas been extensive testing of reliabil-ity among physician, nurse, and phar-macy caregivers in all adult ICUpatient populations, and after imple-mentation in a medical ICU.4 Inter-rater reliability at other institutionsand over time has also been estab-lished.8,15 Extensive evidence of vali-dation, including face and constructvalidity testing, comparison withother sedation scales, and correlationwith processed EEG and limb move-ment technology has been demon-strated.4,8,16

Adaptation to Intensive CareEnvironment (ATICE)

The ATICE presents a comprehen-sive approach with a consciousnessdomain and a tolerance domain.6

There are 5 tests (awakeness andcomprehension in the consciousnessdomain, and calmness, ventilatorsynchrony, and face relaxation in thetolerance domain), and 20 steps arerequired to complete testing. Thereare 6 levels in the awakeness domain,including an alert (eyes openingspontaneously) level. The tests are

performed in a step-wise fashion toaddress tolerance first, followed byconsciousness, and then pain. Thisstrategy has been demonstrated toreduce duration of mechanical venti-lation in a longitudinal studydesign.17

Minnesota SedationAssessment Tool (MSAT)

The MSAT contains a 6-level arous-al scale which examines eye openingspontaneously or in response to ver-bal or physical stimuli, and a motoractivity scale, a 4-level scale basedupon patient movement rangingfrom no spontaneous movement tomovement of the back or abdomen.18

While the arousal scale was demon-strated to be valid using a number ofparameters, the motor scale resultswere less impressive.

All of these sedation assessmentinstruments have been demonstrat-ed to be reliable among research per-sonnel, and SAS, RASS, ATICE, andMSAT have been tested for reliabilityin clinical practice.4,15,17-19 These instru-ments have been tested for face, con-struct, and /or criterion validity usinga broad range of comparators,including visual analogue scales,other sedation scales and instru-ments, the quantity of sedative drugadministered, and expert opinion bycaregivers. Some scales (RSS, SAS,RASS) have also been correlated withprocessed electroencephalography,as well as limb acceleration andmovement using actigraphy or digi-tal imaging.8,16,20



Summary

The use of sedation scalesimproves communication amongcaregivers, reduces the frequency ofover-sedation, and aids in implemen-tation of successful sedation man-agement strategies. The measure-ment of the level of sedation shouldbe performed in all ICUs, particularlyin mechanically ventilated patients,and used as a target for sedative drug titration.

References1. Jacobi J, Fraser GL, Coursin DB, et al. Clinical

practice guidelines for the sustained use of sedatives and analgesics in the critically ill adult. Crit Care Med 2002;30:119-41.

2. Sessler CN, Grap MJ, Brophy GM. Multidisci-plinary management of sedation and analgesia incritical care. Seminars Respir Crit Care Med 2001; 22:211-25

3. Sessler CN. Sedation scales in the ICU. Chest2004;126:1727-30.

4. Sessler CN, Gosnell MS, Grap MJ, et al. The Richmond Agitation-Sedation Scale: validity and reliability in adult intensive care unit patients. Am J Respir Crit Care Med2002;166:1338-44.

5. Fraser GL, Prato BS, Riker RR, et al. Frequency, severity, and treatment of agitation in young versus elderly patients in the ICU. Pharmacotherapy 2000;20:75-82.

6. De Jonghe B, Cook D, Griffith L, et al. Adaptation to the Intensive Care Environment (ATICE): development and validation of a new sedation assessment instrument. Crit Care Med2003;31:2344-54.

7. Ramsay MA, Savege TM, Simpson BR, et al. Controlled sedation with alphaxalone-alphadolone. Br Med J 1974;2:656-9.

8. Ely EW, Truman B, Shintani A, et al. Monitoring sedation status over time in ICU patients: reliability and validity of the Richmond Agitation-Sedation Scale (RASS). JAMA2003;289:2983-91.

9. Riker RR, Fraser GL, Cox PM. Continuous infusion of haloperidol controls agitation in critically ill patients. Crit Care Med 1994;22:433-40.

10.Riker RR, Picard JT, Fraser GL. Prospective evaluation of the Sedation-Agitation Scale for adult critically ill patients. Crit Care Med1999;27:1325-9.

11.Simmons LE, Riker RR, Prato BS, et al. Assessing sedation during intensive care unit mechanical ventilation with the Bispectral Index and the Sedation-Agitation Scale. Crit Care Med1999;27:1499-1504.

12.Riker RR, Fraser GL, Simmons LE, et al. Validating the Sedation-Agitation Scale with the Bispectral Index and Visual Analog Scale in adult ICU patients after cardiac surgery. Intensive Care Med 2001;27:853-8.

13.Devlin JW, Boleski G, Mlynarek M, et al. Motor Activity Assessment Scale: a valid and reliable sedation scale for use with mechanically ventilated patients in an adult surgical intensive care unit. Crit Care Med 1999;27:1271-5.

14.de Lemos J, Tweeddale M, Chittock D. Measuring quality of sedation in adult mechanically ventilated critically ill patients. The Vancouver Interaction and Calmness Scale. Sedation Focus Group. J Clin Epidemiol 2000;53:908-19.

15.Pun BT, Gordon SM, Peterson JF, et al. Large-scale implementation of sedation and delirium monitoring in the intensive care unit: a report from two medical centers. Crit Care Med2005;33:1199-1205.

16.Grap MJ, Borchers CT, Munro CL, et al. Actigraphy in the critically ill: correlation with activity, agitation, and sedation. Am J Crit Care2005;14:52-60.

17.De Jonghe B, Bastuji-Garin S, Fangio P, et al. Sedation algorithm in critically ill patients without acute brain injury. Crit Care Med2005;33:120-7.

18.Weinert C, McFarland L. The state of intubated ICU patients: development of a two-dimension-al sedation rating scale for critically ill adults. Chest 2004;126:1883-90.

19 Brandl KM, Langley KA, Riker RR, et al. Confirming the reliability of the sedation-agitation scale administered by ICU nurses without experience in its use. Pharmacotherapy2001;21:431-6.

20 Chase JG, Agogue F, Starfinger C, et al. Quantifying agitation in sedated ICU patients using digital imaging. Comput Methods Programs Biomed 2004;76:131-41.


PROCEEDINGS

Assessment and Treatment of Delirium in the ICU

Brenda Truman Pun, RN, MSN, ACNP, Divisions Pulmonary/Critical Care and Center for Health Services Research, Vanderbilt University Medical Center, Nashville, TN

In 2002 the Society of Critical CareMedicine (SCCM) published a revi-sion of the clinical practice guidelinesfor the sustained use of sedation andanalgesia that included an entire sec-tion on the treatment of ICU deliri-um.1 The inclusion of this sectionunderscores the importance of mon-itoring and treatment of patientswith delirium to promote optimalcomfort for ICU patients. Manyhealthcare providers think that cog-nitive impairment is an expected out-come in the ICU patient that is tem-porary and of little consequence (i.e.,part of the “ICU psychosis”). ICU delir-ium can occur in up to 80% of ICUpatients and is an independent riskfactor for increased length of stay2

and six-month mortality.3 It is there-fore vital to consider patient safety in

Key Points

• Delirium is an independent risk factor for increased length of stay, six-month mortality, intensive care unit (ICU) and hospital costs.

• The Confusion Assessment Method for the ICU is a reliable and easy-to-use tool for delirium assessment.

• Society of Critical Care Medicine guidelines include recommendations for the treatment of ICU delirium.

• To reduce the risk associated with delirium, clinicians should first consider discontinuing or reducing doses of anticholinergics, sedatives and analgesics.

addition to patient comfort whenmonitoring and treating delirium.

Definition, Prevalence andOutcomes of Delirium

Many terms have been used todescribe delirium including ICU psy-chosis, ICU syndrome, acute confu-sional state, septic encephalopathy,and acute brain failure. TheDiagnostic and Statistical Manual ofMental Disorders (DSM IV), definesdelirium as a disturbance of con-sciousness with inattention accom-panied by a change in cognition orperceptual disturbance that devel-ops over a short period of time (hoursto days) and fluctuates over time.4

Delirium can be divided into threesubtypes according to level of psy-chomotor activity and alertness:hypoactive, hyperactive, and mixed.5-7

Hypoactive delirium is characterizedby lethargic level of consciousnessand is often referred to as “quiet”delirium.5,6,8 Hyperactive delirium isassociated with agitation and charac-terized by restlessness, fidgeting,pulling out tubes and lines, andsometimes even combativeness.5-7

The mixed subtype is characterizedby a fluctuating between hyper andhypoactive.

Delirium is present in up to 80% ofICU patients.9-12 Mixed-subtype deliri-um is the most common (54.9%), fol-lowed by hypoactive delirium(43.5%) and purely hyperactive delir-ium (1.6%).13 Despite the high preva-lence, delirium is unrecognized in66% to 84% of patients, whether theyare in an ICU, a hospital ward, or anemergency department.14-16

Delirium is associated with manynegative outcomes. Patients with ICUdelirium have a three-fold increasedrisk of death in six months whencompared to those without deliriumeven after controlling for pre-existingcomorbidities, severity of illness,coma and the use of sedative andanalgesic medicines. Each additionalday of delirium was associated with a20% increased risk of remaining inhospital and a 10% increased risk of




death.3 Delirium is also associatedwith higher ICU costs ($22,346 vs.$13,332) and hospital costs ($41,836vs. $27,106).17 Delirium may also pre-dispose ICU survivors to prolongedneuropsychological deficits.18

Assessment

Currently there are two ICU deliri-um assessment tools: the IntensiveCare Delirium Screening Checklist19

and the Confusion AssessmentMethod for the ICU (CAM-ICU).10 TheIntensive Care Delirium ScreeningChecklist is an eight-item checklistwith a sensitivity of 99%, specificityof 64%, and inter-rater reliability of0.94.19 Each of the eight items isscored as absent or present (1 or 0,respectively) and the item scores aresummed for a total score. Patientswho score >4 on the total score areconsidered delirious. The CAM-ICU(Figure), adapted from the ConfusionAssessment Method20 for use in non-verbal ICU patients, is a well-validat-ed, widely used delirium assessmentscale.9-11 The CAM-ICU was designedto be a serial assessment tool fornurses and physicians and is easy touse, taking a half minute on averageto complete. A recent report fromtwo institutions regarding the feasi-bility of implementing this newassessment found that even afterminimal training nurses were bothcompliant and accurate with per-forming the CAM-ICU.21 A completedescription of the CAM-ICU andtraining materials can be found onwww.ICUdelirium.org.

Prevention and Treatment

Although the exact pathophysio-logical mechanisms involved in delir-ium are unknown, imbalances ofneurotransmitters such as dopamine,γ-aminobutyric acid, and acetyl-choline are thought to be the rootcause.7,22 Imbalances in these neuro-transmitters can result from severalfactors: reduction in cerebral metab-olism, primary intracranial disease,systemic diseases, secondary infec-tion of the brain, exogenous toxicagents, withdrawal from substancesof abuse such as alcohol and seda-tive-hypnotic agents, hypoxemia andmetabolic disturbances, and theadministration of psychoactive medications such as benzodi-azepines and narcotics.

Possible causes of delirium must beconsidered when formulating treat-ment plans. For example, neurotrans-mitter levels are affected by drugswith anticholinergic properties, andpsychoactive medications are theleading iatrogenic risk factor for delir-ium.23,24 Benzodiazepines, narcotics,and other psychoactive drugs areassociated with a 3- to 11-foldincrease in relative risk for the devel-

opment of delirium.25 A recent studyfound the use of opiates was strong-ly related to the development ofdelirium but the use of benzodi-azepines and propofol was not.26 Inanother study, lorazepam was foundto be significantly associated withthe development of delirium but nosignificant relationship was foundwith fentanyl, morphine, or propo-fol.27 More research is needed todetermine the specific medicationsassociated with delirium. When con-sidering treatment options, it isimportant to consider the variousmechanisms that can lead to neuro-transmitter imbalances.

Early detection of delirium isimportant. Treatment efforts shouldthen focus on identifying the causeand removing or reducing thecausative factors to improve patients’mental status and reduce safety risks(Table). There are no data regardingprimary prevention in the ICU.Outside the critical care area, the fol-lowing strategies have been shownto decrease delirium: repeatedlyreorienting patients, providing cog-nitively stimulating activities, non-pharmacological sleep protocols,30

early mobilization activities, range-

Figure. CAM-ICU Features 9,10

and

and

or

= Delirium

1. Acute onset of mental statuschanges or a fluctuating course

2. Inattention

3. Disorganizedthinking

4. Altered level of consciousness



of-motion exercises, removal ofcatheters and physical restraints in atimely manner, use of patients' eyeglasses and magnifying lenses, use ofhearing aids and removing earwax,correcting dehydration, scheduledpain protocols, and minimizingunnecessary noise and stimuli.28,29

When considering pharmacologi-cal treatment of delirium, it is essen-tial to first perform a thoroughreview of the patient's current med-ications to detect any agents thatmay be may be causing or contribut-ing to delirium. This review will revealany sedatives, analgesics, or anti-cholinergic drugs that can be discon-tinued or decreased in dosage.

Although no drugs are approvedby the FDA for the treatment of delir-ium, the SCCM guidelines recom-mend haloperidol (Table). This GradeC recommendation is based onsparse outcomes data from nonran-domized case series and anecdotalreports.1 Neither haloperidol norsimilar agents (e.g., droperidol andchlorpromazine) have been exten-

sively studied in ICU patients.Patients receiving haloperidol shouldbe monitored for adverse effectssuch as QT prolongation, arrhyth-mias, and extrapyramidal effects.1

Other antipyschotic and neurolepticagents (e.g., risperidone and olanza-pine) may also be helpful for deliri-um.1 In fact, a recent study, showedthat although both olanzapine (anatypical antipsychotic) and haloperi-dol were associated with a decreasein delirium, haloperidol was associat-ed with more side effects. Althoughthis was a small, unblinded studylacking a placebo control and theresults should be interpreted withcaution, it represents the first studyto compare the two drugs.Prospective, randomized, controlledtrials are needed to evaluate theeffectiveness and safety of theseagents relative to one another.

Summary

Delirium is a significant problemfor critical care patients, affecting50% of non-ventilated and 80% ofventilated ICU patients. It is associat-

ed with many negative clinical out-comes, including increased mortality.Two valid and reliable tools to moni-tor for delirium are the Intensive CareDelirium Screening Checklist and theCAM-ICU. Current guidelines recom-mend the use of haloperidol to treatdelirium. It is important to first con-sider removing or reducing doses ofmedications that may be contribut-ing factors, such as anticholinergics,sedatives and analgesics. There aremany unanswered questions andmuch room for future study. Givenits prevalence and association withnegative outcomes, delirium in theICU cannot be ignored and patientsshould be actively monitored for thiscondition. There is a need to find bet-ter ways to decrease delirium andthus improve both patient comfortand safety.

References1. Jacobi J, Fraser GL, Coursin DB, et al. Clinical

practice guidelines for the sustained use of sedatives and analgesics in the critically ill adult. Crit Care Med 2002;30(1):119-41.

2. Ely EW, Gautam S, Margolin R, et al. The impact ofdelirium in the intensive care unit on hospital length of stay. Intensive Care Med 2001;27(12):1892-1900.

3. Ely EW, Shintani A, Truman B, et al. Delirium as a predictor of mortality in mechanically ventilated patients in the intensive care unit. JAMA2004;291(14):1753-62.

4. Diagnostic and Statistical Manual of Mental Disorders. Fourth Edition ed. Washington D.C.: American Psychiatric Association, 1994.

5. Milisen K, Foreman MD, Godderis J, et al. Deliriumin the hospitalized elderly: Nursing assessment and management. Nurs Clin N Amer1998;33:417-36.

6. Meagher DJ, Hanlon DO, Mahony EO, et al. Relationship between symptoms and motoric subtype of delirium. J Neuropsychiatry Clin Neurosci 2000;12:51-6.

7. Meagher DJ, Trzepacz PT. Motoric subtypes of delirium. Semin Clin Neuropsychiatry2000;5(2):75-85.

8. Holliday JE, Myers TM. The reduction of weaning time from mechanical ventilation using tidal volume and relaxation biofeedback. Am Rev Respir Dis 1990;141:1214.

9. Ely EW, Margolin R, Francis J, et al. Evaluation of delirium in critically ill patients: validation of the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU). Crit Care Med2001;29(7):1370-9.

Table. Treatment of Delirium

1. Identify the etiology.

2. Modify risk factors.

3. SCCM Grade C recommendation1:Haloperidol (Haldol ) 2–10mg IVP q20–30 min,then 25% of loading dose every 6h (possible side effects:extrapyramidal symptoms, QT prolongation, torsades de pointes)

4. The atypical antipsychotics (e.g., olanzapine, ziprasidone, andseroquel) may also be helpful in treating delirium, but more research is needed to evaluate these drugs and their effect on this condition.


10. Ely EW, Inouye SK, Bernard GR, et al. Delirium in mechanically ventilated patients: validity and reliability of the confusion assessment method for the intensive care unit (CAM-ICU). JAMA 2001; 286(21):2703-10.

11. Lin SM, Liu CY, Wang CH, et al. The impact of delirium on the survival of mechanically ventilated patients. Crit Care Med2004; 32(11):2254-9.

12. McNicoll L, Pisani MA, Zhang Y, et al. Delirium in the intensive care unit: occurrence and clinical course in older patients. J Am Geriatr Soc2003;51(5):591-8.

13. Peterson JF, Pun BT, Dittus RS, et al. Delirium andits motoric subtypes: a study of 614 critically ill patients. J Am Geriatr Soc 2006; 54:479-84.

14. Inouye SK. The dilemma of delirium: clinical and research controversies regarding diagnosis and evaluation of delirium in hospitalized elderly medical patients. Am J Med 1994;97(3):278-88.

15. Sanders AB. Missed delirium in older emergency department patients: a quality-of-care problem. Ann Emerg Med 2002;39:338-41.

16. Hustey FM, Meldon SW. The prevalence and documentation of impaired mental status in elderly emergency department patients. Ann Emerg Med 2002;39:248-53.

17. Milbrandt EB, Deppen S, Harrison PL, et al. Costs associated with delirium in mechanically ventilated patients. Crit Care Med 2004;32(4):955-62.

18. Jackson JC, Gordon SM, Hart RP, et al. The association between delirium and cognitive decline: a review of the empirical literature. Neuropsychol Rev 2004;14(2):87-98.

19. Bergeron N, Dubois MJ, Dumont M, et al. Intensive Care Delirium Screening Checklist: evaluation of a new screening tool.Intensive Care Med 2001;27(5):859-64.

20. Inouye SK, van Dyck CH, Alessi CA, et al. Clarifying confusion: the confusion assessment method. A new method for detection of delirium.Ann Intern Med 1990;113(12):941-8.

21. Pun BT, Gordon SM, Peterson JF, et al. Large-scale implementation of sedation and delirium monitoring in the intensive care unit: A report from two medical centers. Crit Care Med 2005;33(6):1199-1205.

22. Justic M. Does "ICU psychosis" really exist? Crit Care Nurse 2000;20:28-37.

23. Francis J. Drug-induced delirium: diagnosis and treatment. CNS Drugs 1996;5:103-14.

24.Brodaty H, Pond D, Kemp NM, et al. The GPCOG: anew screening test for dementia designed for general practice. JAGS 2002;50:530-4.

25. Lipowski ZJ. Delirium in the elderly patient. N Engl J Med 1989;320:578-82.

26. Dubois MJ, Bergeron N, Dumont M, et al. Delirium in an intensive care unit: a study of risk factors. 2001;27(8):1297-1304.

27. Pandharipande PP, Shintani A, Peterson J, et al. Lorazepam is an independent risk factor for transitioning to delirium in intensive care unit patients. Anesth 2006;104:21-6.

28.Inouye SK, Bogardus ST, Jr., Charpentier PA, et al. A multicomponent intervention to prevent delirium in hospitalized older patients.N Engl J Med 1999;340(9):669-76.

29. Milisen K, Foreman MD, Abraham IL, et al. A nurse-led interdisciplinary intervention program for delirium in elderly hip-fracture patients. JAGS2001;49:523-32.

30. Skrobik YK, Bergeron N, Dumont M, et al. Olanzapine vs haloperidol: treating delirium in a critical care setting. Intens Care Med 2004;30(3):444-9.



PROCEEDINGS

Intensive care units (ICUs) havebeen developed to provide an envi-ronment for the care of critically illpatients. There have been tremen-dous advances in technology, spe-cialization, and ICU professionals’skills, which have increased the likeli-hood of ICU patients' surviving theirserious illnesses. ICU patients are stillat risk for suffering considerable painfrom their diseases, injuries, and/orclinical procedures. Several studieshave documented the physiological1

and psychological1-3 costs of ICUpatients' unrelieved pain (Table 1).

Key Points

• Advances in intensive care unit (ICU) technology, specialization, and clinical skills have increased the likelihood of patients surviving critical illness; however ICU patients are still at risk for suffering unrelieved pain.

• No valid, reliable physiologic or biochemical measures of pain are currently appropriate for use in the ICU setting, but patient behaviors often can indicate the presence and causes of pain.

• Many studies have documented pain associated with common diagnostic and treatment procedures.

• Pre-emptive administration of analgesics can be used to help to reduce procedural pain and avoid the development of persistent pain.

• Balancing the positive and adverse effects of analgesia and sedation is essential but difficult; daily interruptions of sedative infusions may reduce the incidence of complications, but can result in acute withdrawal syndrome following rapid discontinuation.

Pain Assessment

One of the primary causes of inad-equate pain management in the ICUis the lack of appropriate pain assess-ment (Table 2). The use of well vali-dated pain assessment instru-ments4,5,7 can be difficult in ICUpatients. Professional practice stan-dards and regulatory mandates8

require that pain assessments beattempted for all patients. No valid orreliable physiologic or biochemicalmeasures of pain are currently appro-priate for use in the ICU setting, butpain-associated behaviors often indi-cate the presence and causes of pain.

Managing Pain in the ICU PatientKathleen Puntillo RN, DNSc, FAAN, Professor of Nursing and Faculty, Critical Care/Trauma Program,

Department of Physiological Nursing, University of California, San Francisco, CA

Table 1. Challenges of Unrelieved Pain

Physical suffering

• 40% of 80 post-ICU ARDS patients recalled having pain while in ICU

• 40% higher frequency of chronic pain than in controls1

• 87% of 97 post-ventilation patients remembered being moderatelyto extremely bothered by pain2

Psychic suffering

• Significantly higher PSTD scores in post-ICU ARDS patients than in controls (28% vs. 12%)1

• Traumatic memories associated with pain after cardiac surgery(81.6% of 184 patients)3



Managing Pain in the ICU Patient

Puntillo and colleagues9 noted asignificant relationship betweenbehavioral indicators of painobserved by nurses and the nurses'ratings of postoperative patients'pain intensity. Payen and colleagues10

reported validation of a pain behav-ior scale (PBS) in a sample of 30 intu-bated, sedated ICU patients. Using amodel in which patients underwentcommon noxious and non-noxiousprocedures, they found that the per-cent of patients exhibiting no painbehaviors was significantly lower inthe patient group that underwent anon-noxious procedure than in thegroup that underwent a noxious pro-cedure. More recently a behaviorobservation scale was developedand tested in a large sample ofpatients undergoing common proce-dures such as wound care, wounddrain removal, and turning.11 By com-paring behaviors exhibited beforeand during the procedure, andbehaviors in patients with and with-out procedural pain (as noted on a 0-10 numeric rating scale), theresearchers identified specific proce-dural pain behaviors that includedgrimacing, rigidity, wincing, shuttingof eyes, verbalization, moaning, andclenching of fists. Patients with pro-cedural pain were 2.8 times morelikely to have increased facialresponses, 4.1 times more likely tohave increased body movementresponses, and 10.3 times more likelyto have increased verbal responsesduring the procedure than patientswithout procedural pain. Despitethese studies, there is still a need fora valid, reliable, and feasible behav-ioral assessment methods for ICUpatients.12

Pain Management

Pain assessment is an essential stepbefore pain intervention, but painintervention can be provided evenbefore a pain stimulus. Preemptiveanalgesia is the administration of ananalgesic agent before the patientexperiences a noxious stimulus toprevent amplification and hyperex-citability of the central nervous sys-tem (CNS). Hyperexcitability maylead to CNS sensitization, which canresult in persistent pain. Patients maynot receive preemptive analgesiabefore procedures because healthprofessionals are unaware of thedegree of pain associated with vari-ous common procedures. Manystudies have documented the painassociated with procedures such asarterial blood gas draws, nasogastrictube insertion, intravenous (IV)catheter insertion, mechanical venti-lation,13 turning, wound drainremoval, wound care, tracheal suc-tioning,14,15 and chest tuberemoval.14,16,17 Planning for these pro-cedures should include considera-tion of the optimal preemptive anal-gesic intervention. Analgesic admin-istration should be timed so that theselected drug's peak effect is

obtained at the time of the proce-dure. Puntillo and Ley18 demonstrat-ed that pain associated with chesttube removal was minimal whenpatients were given either IV ketorolac or IV morphine before theprocedure at a time that will achieve a drug’s peak effect duringthe procedure.

Opioid and SedativeTherapies

Opioid analgesic therapy remainsthe primary pharmacologic treat-ment modality for ICU patients.Based on a review of the most recentevidence, Jacobi and colleagues4

have developed clinical practiceguidelines for the sustained use ofsedatives and analgesics in the criti-cally ill. Organ dysfunction andhemodynamic abnormalities in criti-cally ill patients can result in signifi-cant inter-individual variability in thepharmacokinetics and pharmacody-namics of drugs. For a given patient,a standard drug dose could be toxic,subtherapeutic, or effective.

It is not unusual for critically illpatients to receive a combination ofopioid and sedative therapies, soproper use of multiple, potentiallyinteracting medicines can be chal-

Table 2. Pain Assessment in Patients Who Cannot Self-report

• No valid or reliable physiological or biochemical measure of pain

• Pain behaviors have been validated in acutely/critically ill patients undergoing procedures

• Behavioral Pain Scale validated in critically ill, sedated patients10


lenging (Table 3). One or both typesof agents may adversely affect thepatient's hemodynamic or respirato-ry status. The use of both treatmentsmay have synergetic adverse effects.The clinical challenge is to use theright type and combination of med-ications while avoiding adverseeffects. Another challenge is to usetools that differentiate among pain,anxiety, and agitation, so that phar-macological interventions can be tar-geted to more clearly defined goals.Without the use of such tools,patients may be under-, over-, or“mis”-dosed.

Daily interruptions of sedativeinfusions for the purpose of assessingthe patient have become standardpractice in many ICUs. Kress and col-leagues19 showed that in ICU patientsdaily interruptions of sedative infu-sions reduced the incidence of manycomplications and did not result innegative psychological outcomes.20

Interruptions in sedative infusionsshould include pain and agitationassessments, and be minimized in

patients who respond adversely tothem. In patients who have beenreceiving opioids and benzodi-azepines for more than a few days,interruptions can result in a with-drawal syndrome. Cammerano andcolleagues21 showed that ICUpatients who received analgesic andsedative medications for longer thanseven days experienced acute with-drawal syndrome after rapid discon-tinuation of medication.

Summary

Although advances have been madein pain assessment and treatment forpatients in the ICU, gaps remain.There are no comparative trials ofopioids,4 and the evidence for mostrecommendations made by the mostcurrent guideline panel4 is based onobservational studies rather thanrandomized clinical trials. Until moreevidence-based practice tools andguidelines become available, clini-cians will need to use available prac-tice-based assessment tools to pro-mote patient comfort and safety.

References1. Schelling G, Stoll C, Haller M et al. Health-related

quality of life and posttraumatic stress disorder insurvivors of the acute respiratory distress syndrome. Crit Care Med 1998;2:651-9.

2. Rotondi A, Lakshmipathi C, Sirio C et al. Patients' recollections of stressful experiences while receiving prolonged mechanical ventilation in an intensive care unit. Crit Care Med 2002;30:746-52.

3. Schelling G, Richter M, Roozendall B et al. Exposure to high stress in the intensive care unit may have negative effects on health-related quality-of-life outcomes after cardiac surgery. Crit Care Med 2003;31:1971-80.

4. Jacobi J, Fraser GL, Coursin DB et al. Clinical practice guidelines for the sustained use of sedatives and analgesics in the critically ill. Crit Care Med 2002; 30:119-41.

5. Puntillo K. Critically Ill Patients in Pain: The CriticalIssues. Crit Connections 2004;3(2):14.

6. Puntillo KA, Weiss SJ. Pain: Its mediators and associated morbidity in critically ill cardiovascularsurgical patients. Nurs Research 1994;43:31 6.

7. Puntillo KA. Pain Management. In, H. Schell-Chapel and K. A.Puntillo (Eds.). Crit Care Nurs Secrets. 2000. Philadelphia: Hanley & Belfus Inc. .

8. Comprehensive Accreditation Manual for Hospitals. Joint Commission on the Accreditation of Healthcare Organizations; 1999. Accessed November 24, 2001 from http://www.jcaho.com/standards_frm.html.

9. Puntillo KA, Miaskowski CA, Kehrle K, et al.The relationship between behavioral andphysiological indicators of pain, critical carepatients' self reports of pain, and opioidadministration. Crit Care Med 1997;25: 1159-66.

10.Payen JF, Bru O, Bosson JL, et al. Assessing pain in critically ill sedated patients by using a behavioral pain scale. Crit. Care Med2001;29:2258-63.

11.Puntillo KA, Morris AB, Thompson CL et al. Pain Behaviors Observed During Six Common Procedures: Results from Thunder Project II. Crit Care Med 2004;32(2):412-27.

12.Gélinas C, Fortier M, Viens C et al. Pain assessment and management in critically-ill intubated patients: a retrospective study. Am J Crit Care 2004;13(2):126-35.

13.Morrison RS, Ahronheim JC, Morrison GR, et al.Pain and discomfort associated with commonhospital procedures and experiences. J PainSymptom Manage 1998;15:91-101.

14.Puntillo, K.A. Dimensions of procedural pain and its analgesic management in critically ill surgical patients. Amer J of Crit Care 1994;3:116 22.

15.Puntillo KA, White C, Morris A, et al. Patients'perceptions and responses to procedural pain:Results from Thunder Project II. Amer J of Crit Care2001;10:238-51.

16.Puntillo, KA. Effect of interpleural bupivacaine on pleural chest tube removal pain: A randomized controlled trial. Amer J of Crit Care1996;5:102-8.

17.Owen S, Gould D. Underwater seal chest drains: the patient’s experience. J Clin Nurs 1997;6:215-25.

18.Puntillo KA, Ley J. Appropriately timedanalgesics control pain due to chest tuberemoval. Amer J Crit Care 2004;13:292-301.

19.Kress JP, Pohlman AS, O'Connor MF, et al. Daily interruption of sedative infusions in critically ill

Table 3. Balancing Analgesia and Sedation isEssential but Elusive

• Combination of analgesics and sedatives common in ICU

• Challenges:

– Tools that differentiate pain from anxiety/agitation exist, with limitations14,15

– Choosing right type and combination of medications

– Need new ways to assess pain and discomfort in all ICU patients




patients undergoing mechanical ventilation. NEJM 2000;342:1471-7.

20.Kress JP, Gehlbach B, Lacy M, et al. The long-term psychological effects of daily sedative interrup-tion on critically ill patients. Amer J Resp Crit CareMed 2003;168:1457-61.

21.Cammarano WB, Pittet JF, Weitz S, et al. Acutewithdrawal syndrome related to the administra-tion of analgesic and sedative medications inadult intensive care unit patients. Crit Care Med1998;26:676-84.


PROCEEDINGS

The Role of Brain Monitoring in the ICUMichael Ramsay MD, FRCA, Chair, Department of Anesthesia,

Baylor Healthcare System, Dallas, TX

Monitoring mental well-being isas important as monitoring the per-formance of other major organ functions. Cerebral function, howev-er, is not always monitored in the critical care or perioperative setting.Monitoring brain function effectivelyand reliably is the next major challenge of intensivists and anesthesiologists.

Cerebral cortical function may beaffected by cerebral perfusion,hypoxia, pharmacological agents,seizure activity, cerebral metabolism,and neuronal dysfunction.Instruments for monitoring brainfunction include subjective meas-ures, such as the assessment of depthof sedation or severity of delirium byscoring systems, and objective meas-ures that assess cerebral corticalactivity, responses to evoked poten-tials, cerebral oximetry, and transcra-nial Doppler ultrasound. The success

Key Points

• Mental well-being is as important than physical well-being.

• Brain monitoring can detect, prevent and provide a prognosis for cerebral insults.

• Cerebral function monitoring is a patient safety strategy.

of these monitors in providing con-sistently accurate information aboutthe cerebral state is dependent onmany factors: the suitability of thetarget area of the brain monitored toaccurately reflect cerebral state, andthe medications and combinations ofmedications administered. It alsodepends on the algorithms devel-oped to present the informationmonitored in a clinically useful form,and the suppression of extraneousnoise and other confounding factors.Many studies have analyzed theeffectiveness of using cerebral func-tion monitors not only to guidedepth of sedation, but also to pre-vent unintended awareness in theparalyzed patient. The real value ofthese monitors may well be inimproving patient safety by detect-ing cerebral insults early, allowingappropriate intervention to be made,and predicting outcome.

The use of a multimodal approachto cerebral monitoring may provide amore comprehensive approach todetecting potential cerebral compro-mise and monitoring the effects ofcorrective actions.

Consciousness

The conscious state is difficult todefine; we know it when we see it butwe would probably all have a differ-ent definition. It is a biological phe-nomenon and consists of qualitativesubjective states of perceiving, feel-ing, thinking, information processingand awareness. Where in the braindoes consciousness exist? Is it a corti-cal function or does it arise fromdeeper areas in the brain and is relat-ed to corticothalamic-thalamocorti-cal reverberations? Everyday con-sciousness is altered by natural sleep,medications such as sedatives andanesthetics, and the effects of self-indulgence. Does how deeplyunconscious a person becomesaffect how well they recover? A grow-ing body of outcome data suggestthat deep sedation and anesthesiamay be associated with sequelaesuch as cognitive deficit and, in anolder patient, increased mortality atone year.



The Role of Brain Monitoring in the ICU

Monitoring brain function is a vitalcomponent for managing the well-being of the patient undergoinganesthesia and sedation therapy.With most sedation and anestheticagents, the transition from lightsedation to deep sedation and thenthrough general anesthesia to “burstsuppression” or coma may occur veryrapidly. Whether this is a linear pro-gression or is a series of definite stepsin the transition from consciousnessto unconsciousness is the focus ofsignificant research.

In 1937 Guedel described fourstages of anesthesia in patientsundergoing open ether induction.Ninety years earlier Snow haddescribed five degrees of narcotismthat followed his close observation ofmental state, respiratory pattern andpulse characteristics. Johns recentlydescribed a cascade of six neuro-physiologic steps that occurred dur-ing induction and recovery fromanesthesia. The concept that depthof anesthesia and sedation can bemeasured on a linear 0-100 scalederived from the cortical activity ofthe brain may not reflect the actualprocess. The development of sensorsthat can probe the deeper levels ofthe brain may be necessary beforethe state of consciousness can beaccurately monitored.

Sedation Monitoring

The consequences of poorly man-aged sedation have long beenknown, yet tight control of sedationis still not universal in critical careunits. Virtually every patient admit-ted into the intensive care unit (ICU)is administered sedation therapy. Theprecise control of the depth of seda-

tion is often not well managed.Patients often are over- or under-sedated with an accompanyingincrease in morbidity, mortality andcost. The effective management ofpain, anxiety and sleep (hypnosis),together with maintenance of anoptimal level of comfort and safety,are the major aims of a sedation ther-apy regimen. Patient-specific care isevolving to the point where titrat-able, target-specific medications canbe administered that reduce sideeffects, modulate the stress andinflammatory responses and reduceICU stay. Effective pain managementwith minimal side effects is the foundation of a well-controlled seda-tion program.

In 1974, the concept of controllingsedation in the ICU and the first seda-tion scoring system, the RamsaySedation Scale,1 were developed. Thesame level of intense managementprovided for all other organ systemswas proposed for brain function.Studies have since shown that opti-mizing sedation managementimproves patient outcomes, shortensICU length of stay, and significantlyreduces costs.

Both subjective and objectivetools are available to aid in theassessment, re-assessment, andadjustment of therapy, to help avoidover- or under-sedation and theirconsequences, and to manage thesepatients more precisely. Sedationscales are subjective tools for bed-side evaluation of patients, whilebrain cortical activity monitors pro-vide objective monitoring of theresponse to sedation therapy andhelp achieve an optimum level of

consciousness.

Critical care physicians must con-tinually reassess and redefine thesedation strategy for their patients asthe clinical state changes. Havingprotocols for administering sedation,analgesic, anxiolytic, antideliriumand paralytic medications, and seda-tion monitoring tools, is important.

Drugs used for sedation includeanalgesics, sedatives, and neurolep-tics All of these agents have adverseconsequences associated with theiruse, such as respiratory depression,abuse potential, lack of orientationand cooperation, delirium, hypoten-sion, and tolerance. Other effectsinclude hyperlipidemia with propo-fol, constipation with opioids, anddelirium with the benzodiazepines. Anew agent in ICU sedation isdexmedetomidine, an alpha-2adrenoagonist that acts on the locusceruleus/norepinephrine axis ratherthan on the GABA receptor, and pro-vides non-REM sleep, sedation, andanxiolysis without respiratorydepression and with an opioid-spar-ing capacity. This drug offers a moreinnovative approach to controlled,co-operative, comfortable sedation.

Protocols should be developedthat enable a patient to be comfort-able, pain free, and safe, and shortenthe ICU length of stay. Poor controlof sedation is no longer acceptable asa standard of care in the ICU.

Brain Function Monitors

The measurement of brain corticalelectrical activity was first describedin humans by Hans Berger in 1929.Within 10 years the effect of anes-thetic agents on the electroen-


The Role of Brain Monitoring in the ICU

cephalogram (EEG) was noted, butthe clinical utility of this instrumentfor monitoring depth of sedation andcerebral well being during surgicalprocedures was not developed until1969. Maynard constructed a two-channel cerebral function monitorthat could record cerebral electricalactivity as a continuous power strip.Investigators were able to use thisdevice to evaluate cognitive out-come after cerebral insult during car-diopulmonary bypass procedures.The utility of the cerebral functionmonitor as a diagnostic and predic-tive clinical tool was developed.

The EEG opens a window on thecortical brain electrical activity,which is a sensitive marker of brainischemia and hypoxia. The EEG mayalso indicate the depth of pharmaco-logical sedation by recording thetransition to low-frequency, high-voltage activity. One of the limita-tions of EEG monitors is that they usea frontal array only and this change is

heterogeneous across the cerebralcortex.

It is important to be able to assessthe depth of sedation in the ICUpatient as objectively as possible toconsistently attain the desired level.There are two types of monitors thatexamine electrical activity: those thatassess cortical activity as a raw signal,a power spectral array or an entropyanalysis; and those that assess theresponsiveness of the brain toevoked potentials. The most com-monly used brain monitors that usethe latter technology are theBispectral Index (Aspect MedicalSystem, Newton, MA), the PatientState Index (Hospira, Lake Forest, IL),Auditory Evoked Potentials (CareFusion San Diego, CA) and Entropy (GE HealthCare, Fairfeld, CT). Other companies with similar technologies are developing new products of the same type.

Another approach to monitoringcerebral well-being is the cerebral

oximeter (Somanetics Corp. Troy, MI).This device uses dual-wavelength,near-infrared spectroscopy to meas-ure regional cerebral oxygen satura-tion. It is ideally suited to indicatechange in the balance of oxygen sup-ply and demand. Low oxygen satura-tion found intraoperatively in cardiacsurgery patients is associated withincreased time on mechanical venti-lation, time in ICU and length of stayin the hospital postoperatively.

Summary

Technology that improves theassessment of cerebral well-being isgetting better. As regular use of brainfunction monitors increases, this willstimulate the development of thistechnology to the degree of sophisti-cation that we have reached with car-diac monitoring. It is incumbent onall of us to continually evaluate theart and science of brain functionmonitoring and consider where it fitsin our practices.

Reference1. Ramsay AM, Sevege TM, Simpson BR, et al.

Controlled sedation with alphaxalone-alphadolone. Br Med J 1974;2:656-9.



PROCEEDINGSPROCEEDINGS

Anxiety, agitation, and pain areproblems frequently encountered incritically ill patients. Increasingly,delirium also has been recognized asa significant complication of criticalillness.1 Consensus guidelines werepublished by the Society of CriticalCare Medicine in 1995 and revised in2002 in an attempt tooptimize managementof these problems.2,3

Non-pharmacologicstrategies and anal-gesics form the founda-tion of these guidelines(see Figure).2,4 In mostinstances, however,additional sedativetherapy is needed incombination with anal-

Key Points

• Sedatives are often used in critically ill patients to manage anxiety, agitation, and delirium.

• No agent being used clinically today has all the characteristics of an ideal sedative.

• Some of the characteristics of an ideal sedative are shared by the sedatives used most commonly, such as benzodiazepines, propofol, dexmedetomidine, and haloperidol.

• These four sedatives differ with regard to mechanism of action, clinical pharmacology, pharmacokinetics, pharmacodynamics, and comparative side effect profiles.

gesics to provide comfort and facili-tate care of the critically ill patient.Characteristics of the ideal sedativein this setting have been outlinedpreviously (see Table 1).5

Unfortunately, no sedative is current-ly available that possesses all of theseattributes. Nevertheless, considera-

tion of these characteristics providesan excellent framework in which tocompare the clinical pharmacologyof those agents routinely used inmanaging anxiety, agitation, anddelirium in the critically ill. The seda-tives that will be outlined in this arti-cle include the benzodiazepines,propofol, haloperidol, anddexmedetomidine.

Clinical Pharmacology of Sedatives in the Critically Ill

Bradley A. Boucher, PharmD, FCCP, FCCM, Professor of Pharmacy, University of Tennessee Health Science Center, Memphis, TN

Table 1. Characteristics of the Ideal Sedative in Critically Ill Patients.4,15

• Rapid onset of action

• Rapid recovery after

discontinuation

• No active metabolites

• Easy to titrate to desired level

of sedation

• Easy to administer

• Few adverse effects

• Lack of tolerance

• Lack of withdrawal symptoms

• Minimum drug interactions.

• Metabolism and/or elimination

not dependent on hepatic or

renal function

• Inexpensive

Nonpharmacologic Interventions

Analgesia

Sedation

Amnesia

Figure. Achieving Optimal Patient Comfortin the ICU 2,4


BenzodiazepinesMechanism of Action andPharmacologic Effects

The effects of benzodiazepines aremediated through binding to a specific site on the γ-aminobutyricacid (GABA) receptor.6 Stimulation ofthe benzodiazepine-GABA receptorcomplex produces general inhibitionof neuronal impulses and blocks theencoding of new information andunpleasant experiences within thelimbic system.2,6 The effect is a con-centration-dependent anxiolysis andanterograde amnesia at lower dosesand sedation-hypnosis at higherdoses.7 The benzodiazepines alsopossess muscle relaxant and anticon-vulsant pharmacologic propertiesthat are useful in selected critically ill patients.

Pharmacokinetics/Pharmacodynamics

The three benzodiazepines usedclinically for sedation in the criticallyill patient are diazepam, midazolam,and lorazepam.The pharmacoki-netic properties ofthese three drugsare outlined inTable 2. Variableeffects betweenthese agents suchas onset of actionand duration ofaction can be read-ily explained by dif-ferences in theirp h y s i o c h e m i c a lproperties. Forexample, diazepamis highly lipophilic,

resulting in rapid distribution intothe brain as well a relatively shortduration of action upon redistribu-tion into peripheral tissues. In con-trast, lorazepam is the least lipophilicof the three benzodiazepines in clini-cal use. Thus, it crosses the blood-brain barrier more slowly comparedwith diazepam and midazolam,which results in a delayed onset ofaction yet produces a much longerduration of action.

Another major distinguishingcharacteristic between the agents isthe formation of active metabolitesfor diazepam and midazolam, whilelorazepam does not have any activemetabolites. The presence of activemetabolites can dramatically affectthe pharmacodynamic effects oflong-term dosing of diazepam andmidazolam, compared with single-dose effects. For example, whilemidazolam has a predictable phar-macokinetic profile when given as abolus dose or short-term infusion,prolonged over-sedation has been

observed in patients receiving long-term infusions or those with hepaticdysfunction.8,9 Furthermore, since theactive metabolites for diazepam andmidazolam are eliminated by the kid-neys, the potential for accumulationof the metabolite and prolongationof effect should be considered in crit-ically ill patients with renal insuffi-ciency. Protein-binding alterationscan also alter the effect of theseagents, since all are relatively highlyprotein bound.

Side-Effect Profile

The benzodiazepines as a class area relatively safe group of drugs.6

However, the potential for respiratorydepression and hypotension doesexist, especially when used in combi-nation with opiate analgesics. Acuteand chronic tolerance has beendescribed with benzodiazepines use,as well as paradoxical responses.10

This latter phenomenon is particular-ly common in elderly patients.

Diazepam is metabolized by the

Table 2. Pharmacokinetics of Sedatives Used Routinely in Critically Ill Patients2,7,11

Drug Onset of Half-life Protein Metabolism Active RouteAction (hours) Binding Metabolites of Elimination(minutes) (%)

Diazepam 2-5 20-120 99 Hepatic Yes Parent: liveroxidation Active metabolite:

renal

Lorazepam 5-20 8-15 91 Hepatic conjugation None Parent: liver

Midazolam 2-5 3-11 95 Hepatic oxidation Yes Parent: liverActive metabolite:renal

Propofol 1-2 26-32 98 Hepatic conjugation None Liver

Dexmedetomidine 5-10 2-7.5 94 Hepatic oxidation None Liver

Haloperidol 3-20 18-54 92 Hepatic oxidation Yes Liver




cytochrome P450 CYP2C19 subfami-ly in the liver.6 The potential forgenetic polymorphisms and interac-tions with other drugs metabolizedby this pathway (e.g., amiodarone,fluconazole, omeprazole, valproicacid) can make the response todiazepam more unpredictable com-pared with alternative sedatives.6

Similarly, midazolam is metabolizedby the CYP3A4 enzyme subfamily. Assuch, the potential for drug interac-tions with inhibitors of this isoen-zyme, such as macrolide antibiotics,diltiazem, cimetidine, propofol, vera-pamil, and fluconazole, can result inaccumulation of the parent com-pound.6

Lorazepam is formulated in poly-ethylene glycol and propylene gly-col.6 While small, usual dosages oflorazepam result in inconsequentialamounts of these excipients beingadministered, long-term infusionsand higher-than-normal dosageshave been associated with lactic aci-dosis, hyperosmolar coma, andnephrotoxicity.2,7 Precipitation oflorazepam can also occur if not dilut-ed according to the manufacturer’srecommendation.7

PropofolMechanism of Action andPharmacologic Effects

Propofol’s pharmacologic effectsare not entirely known, although arelikely mediated through activation ofthe GABAA receptor at a site distinctfrom the benzodiazepines.11

Clinically, propofol produces anxioly-sis and hypnosis, as well as havingbeneficial effects on intracranial pres-sure in patients with traumatic brain

injury.12 It has essentially no analgesicproperties and less amnestic effectthan the benzodiazepines.5,11


As a consequence of its highlipophilicity, propofol has a veryrapid onset of action and a shortduration of action. While an attrac-tive option when rapid awakening isdesired, the wake-up time may besignificantly prolonged in patientsreceiving infusions for several days orlonger. Since the effect of propofol islargely affected by its distributioninto fatty tissue, its elimination isminimally affected by hepatic or liverdysfunction, which is an attractivefeature in critically ill patients withco-morbidities.

Side-Effect Profile

Hypotension is one of the majorside effects associated with propofoluse. Respiratory depression also canbe observed when combined withthe benzodiazepines or opiates.5,7

Propofol is water insoluble and for-mulated as an oil-in-water emulsionthat provides 1.1 kcal/mL from fat.5

As such, hypertriglyceridemia is apotential effect in critically ill patientsreceiving relatively high doses ofpropofol. In patients receiving propo-fol the additional caloric intakeshould also be considered as part ofthe overall nutritional supplementa-tion to avoid excessive caloric intake.2

Other potential problems related toits lipid formulation include infec-tious complications and drug incom-patibilities.5

A potentially life-threateningcomplication associated with exces-sively high doses of propofol isknown at the “Proprofol InfusionSyndrome.”13 This syndrome is char-acterized by metabolic acidosis,bradycardia, hepatomegaly, hyper-lipidemia, dysthythmias, rhabdomy-olysis and cardiac arrest.13 While orig-inally reported in pediatric patients, ithas subsequently been reported inadults as well.13

DexmedetomidineMechanism of Action andPharmacologic Effects

Dexmedetomidine is a potent α2-adrenoreceptor agonist that possess-es similar pharmacologic propertiesto the prototypical antihypertensivedrug from this class of drugs, namely,clonidine. Activation of the postsy-naptic α2-adrenoreceptor in the cen-tral nervous system results in inhibi-tion of norepinephrine release presy-naptically and sympatholysis.7 Theresulting pharmacologic effectsobserved with dexmedetomidine aresedation and analgesia.


Dexmedetomidine is rapidly dis-tributed upon administration with ahalf-life of approximately 2 hours.14

Based on its linear kinetic profile,dose titration to the desired sedativeeffect is relatively straightforward.7,14

While dosage reductions are warranted in patients with hepaticdysfunction, no dosage adjustmentis needed in patients with renal insufficiency.



Side-Effect Profile

Hypotension and bradycardiaresulting from the sympatholyticeffects of dexmedetomidine are twoof the more significant side effectsassociated with the use of this seda-tive in the critically ill patient. Onemajor advantage of dexmedetomi-dine is the relative lack of respiratorydepression with its use. The ability toeasily arouse patients receivingdexmedetomidine from a sedatedstate (“cooperative sedation”) is alsodeemed to be a distinct advantage inselected critically ill patients.7,14

HaloperidolMechanism of Action andPharmacologic Effects

Haloperidol is a butyrophenoneneuroleptic that antagonizesdopamine-mediated neurotransmis-sion in the basal ganglia of the brain.Pharmacologically, haloperidol hasthe effect of reversing or diminishinghallucinations, delusions or unstruc-tured thought patterns. Thus,haloperidol has utility in treatingdelirium in the critically ill patient.


Haloperidol’s onset of actionoccurs within minutes although itsduration of activity is relatively long.

Dosing often requires repeated intra-venous injections until the desiredoutcome is observed.

Side-Effect Profile

Haloperidol can produced seriousdysrrhythmias (e.g., QTc prolonga-tion, torsade de pointes) especiallywith large daily doses.Extrapyramidal side effects have alsobeen associated with its use.Hypotension can also occur due toα-adrenergic blocker properties. Arare condition known as the neu-roleptic malignant syndrome candevelop in patients receivinghaloperidol.

Summary

Use of sedatives in the critically illis an important yet highly challeng-ing task for clinicians. The agentsmost commonly employed for treat-ment of anxiety and agitation, andminimization of stress in the inten-sive care setting are the benozodi-azepines, propofol, and more recent-ly, dexmedetomidine. Haloperidol isthe drug of choice for the treatmentof delirium. A firm understanding ofthe clinical pharmacology of theseagents is imperative for their optimaluse in the critical care setting.

References 1. Ely EW, Siegel MD, Inouye SK. Delirium in the

intensive care unit: an under-recognized syndrome of organ dysfunction. Semin Respir Crit Care Med2001;22:115-26.

2. Jacobi J, Fraser GL, Coursin DB, et al. Clinical practice guidelines for the sustained use of sedatives and analgesics in the critically ill adult. Crit Care Med 2002;30:119-41.

3. Shapiro BA, Warren J, Egol AB, et al. Practice parameters for intravenous analgesia andsedation for adult patients in the intensive care unit: an executive summary. Society of Critical Care Medicine. Crit Care Med 1995;23:1596-1600.

4. Gehlbach BK, Kress JP. Sedation in the intensive care unit. Curr Opin Crit Care 2002;8:290-8.

5. Angelini G, Ketzler JT, Coursin DB. Use of propofoland other nonbenzodiazepine sedatives in the intensive care unit. Crit Care Clin 2001;17:863-80.

6. Young CC, Prielipp RC. Benzodiazepines in the intensive care unit. Crit Care Clin 2001;17:843-62.

7. Skaar DJ, Weinert CR. Sedatives and Hypnotics. In:Fink MP, Abraham E, Vincent JL, et al., eds. Textbook of Critical Care. Philadelphia: Elsevier Saunders; 2005:1715-24.

8. Barr J, Donner A. Optimal intravenous dosing strategies for sedatives and analgesics in the intensive care unit. Crit Care Clin 1995;11:827-47.

9. Power BM, Forbes AM, van Heerden PV, et al. Pharmacokinetics of drugs used in critically ill adults. Clin Pharmacokinet 1998;34:25-56.

10.Shafer A. Complications of sedation with midazolam in the intensive care unit and a comparison with other sedative regimens. Crit Care Med 1998;26:947-56.

11.Wagner BK, O'Hara DA. Pharmacokinetics and pharmacodynamics of sedatives and analgesics in the treatment of agitated critically ill patients. Clin Pharmacokinet 1997;33:426-53.

12.Kelly DF, Goodale DB, Williams J, et al. Propofol in the treatment of moderate and severe head injury: a randomized, prospective double-blindedpilot trial. J Neurosurg 1999;90:1042-52.

13.Kang TM. Propofol infusion syndrome in critically ill patients. Ann Pharmacother 2002;36:1453-6.

14.Maze M, Scarfini C, Cavaliere F. New agents for sedation in the intensive care unit. Crit Care Clin2001;17:881-97.

15.Ostermann ME, Keenan SP, Seiferling RA, et al.Sedation in the intensive care unit: a systematicreview. JAMA 2000;283:1451-9.



PROCEEDINGS

Guidelines for the treatment of avariety of conditions have been pro-posed to maximize the effectivenessof patient care, while minimizing therisks. The development and imple-mentation of clinical practice guide-lines require a multi-professional,team effort. Education and ongoingreinforcement or intervention maybe needed for sustained adherenceto these guidelines. Improved use ofsedation medications has the poten-tial to reduce the duration ofmechanical ventilation, intensivecare unit (ICU) length of stay, the useof neuromuscular-blocking agents(NMBAs), and the cost of sedativeagents.

Key Points

• Guidelines for the sustained use of sedatives and analgesics have been proposed to improve efficacy and safety.

• Improved use of sedation has the potential to reduce the duration of mechanical ventilation, intensive care unit length of stay, the use of neuromuscular-blocking agents, and the cost of sedative agents.

• Sedation complications may be specific to the agent utilized (e.g., propofol, lorazepam, and remifentanil).

• Reports vary on the clinical and cost improvement resulting from sedation guidelines implementation.

• To realize the potential benefits of sedation guidelines implementation, clinicians will need to measure the impact of these interventions, andre-evaluate and re-educate as needed to maintain adherence with them.

Guidelines for the sustained use ofsedatives and analgesics were pub-lished in 2002 by the Society ofCritical Care Medicine (SCCM) andthe American Society of Health-Systems Pharmacists (ASHP).1

Important components of the seda-tion guidelines included a focus onoptimal analgesia with systematicassessment of agitation and sedationusing a validated scale. The guide-lines recommend the use of intermit-tent anxiolysis in conjunction withanalgesia as the recommended start-ing point, while recognizing thatsome patients may benefit from con-tinuous-infusion sedation (need for

ongoing or deep sedation). The risksof continuous sedation includeexcessive sedation and prolongationof mechanical ventilation or compli-cations such as ventilator-associatedpneumonia. Deep levels of sedationand amnesia have also been associat-ed with delirium, hallucinations andpost-traumatic stress disorder.2,3

Sedation complications may be spe-cific to the agent utilized.

Propofol

Propofol-related infusion syn-drome is a potentially life-threaten-ing complication associated withhigh doses of propofol. The syn-drome features cardiac failure, rhab-domyolysis, lactic acidosis, and renalfailure, but has not been systemati-cally defined. Vasile and colleagueshave summarized the publishedcases and proposed a mechanism fora propofol-related infusion syndrome(PRIS).4 While they detail only 14cases of PRIS in adults, they common-ly were fatal. The use of high-dosepropofol or prolonged therapy, per-haps in conjunction with cate-cholamines and steroids, may be trig-gering factors that lead to PRIS,although a primary or secondarydefect in mitochondrial metabolismaltering the respiratory-chain com-plex or fatty-acid oxidation may also

SCCM Guidelines for Sedation Management: Development and Use in Clinical Practice

Judith Jacobi, PharmD, FCCM, FCCP, BCPS, Critical Care Pharmacist, Methodist Hospital/Clarian Health, Indianapolis, IN



be a factor.5 Thus, propofol remains apreferred agent for limited dosing inshort-term sedation. Patients receiv-ing high dose or long-term propofolinfusions should have triglycerideconcentrations monitored and bescreened for lactic acidosis.1

Lorazepam

Lorazepam is the recommend agentfor long-term sedation.1 Metabolismvia glucuronidation and the lack ofactive metabolites are desirable char-acteristics. Lorazepam and midazo-lam pharmacokinetics and dynamicswere compared in 24 elderly post-operative males receiving target-controlled infusions with a goal con-centration of 50 ng/mL.6 The infu-sions were titrated to a modifiedRamsay scale without the use ofintravenous bolus doses. At the sametarget level of sedation, lorazepamserum concentrations were approxi-mately half of the midazolam con-

centrations; however, the amnesticeffects were approximately 4:1 forlorazepam to midazolam (Figure 1).The pharmacodynamic effectsassessed were time to awakeningand time to extubation.6 Midazolampatients awakened after 3 ± 2.6 hoursversus 8.7 ± 5.9 hours for lorazepam.Times to extubation were 5.4 ± 2.4hours versus 21.2 ± 15.9 hours formidazolam vs. lorazepam, respec-tively. The longer awakening time forlorazepam may have been the resultof context-sensitive half-times from alonger infusion: 36.9 ± 31 hours forlorazepam versus 15 ± 3 hours formidazolam. The potential impact ofmidazolam metabolite accumulationwas not assessed due to the shortduration of the clinical trial.

In addition to oversedation,another reason to limit lorazepamdoses relates to the propylene glycol(PG) solvent. Two reports have shown

a linear relationship between PG con-centrations and osmol gap.7,8 Thelower-level PG concentration andcorresponding osmol gap that pose arisk for renal failure or acidosis areunknown, but lorazepam dosesabove 10 mg/hr produced a veryhigh PG concentration and osmolgap above 20 (Figure 2). 7,8

Remifentanil

The use of short-acting agents tomaximize analgesia with sedationhas received further study.Remifentanil is an ultra-short-actingopioid-receptor agonist that ismetabolized by blood and tissueesterases with an elimination half-lifeof less than 10 minutes. It may be aparticularly useful agent for neuro-surgical patients requiring periodicawakening for neurologic assess-ment. Two recent clinical trials havedocumented a potential role ofremifentanil in the ICU. Muellejansand colleagues compared remifen-tanil and fentanyl for analgesia of 196cardiac ICU patients requiring up to24 hours of sedation.9 Propofol infu-sion was added if opioid administra-tion exceeded a pre-defined dose.This study demonstrated similar effi-cacy for the two opioids titrated to astandard sedation scale, but morepain in the remifentanil group post-discontinuation.

Remifentanil was also comparedwith morphine in 40 surgical ICUpatients.10 Remifentanil producedoptimal sedation for a greater per-centage of time with less frequentdosing changes. The mean durationof mechanical ventilation (14 vs 18hours), extubation time (17 vs. 73

100

80

60

40

20

00 3 6 9 12 15 18 21 24 27 30 33 36

Hours post infusion

100

80

60

40

20

0

Plazma Benzodiazepine Concentration (ng/mL)10 100 100010.1

Lorazepam Midazolam

Lorazepam Midazolam

a

b

Figure 1. Lorazepam vs. Midazolam6

Amnesia Persistence

Amnestic potency

Midazolam 1Lorazepam 4



minutes) and ICU discharge time (20vs 42 hours) were significantly lowerin the remifentanil group comparedwith the morphine group, respective-ly. Midazolam was available for sup-plemental sedation when the opioiddose exceeded a pre-defined dose.Midazolam was needed by more ofthe patients treated with morphinethan remifentanil. Remifentanil doesappear to have some advantagesover other sedating/analgesicagents, although the risk of tolerancewith ongoing infusion remains to befully defined. Remifentanil has ahigher acquisition cost than otheropioids, although drugs constitute arelatively insignificant component ofthe $1500 cost per day of mechanicalventilation.11

Impact of SedationGuidelines Implementation

Bratebrø and colleagues imple-mented a sedation protocol in1999 ina ten-bed surgical ICU.12 The level ofsedation was defined twice daily andintervention to maintain the targetsedation score (based on the MotorActivity Assessment Scale) reducedventilator time by 2 days and mean

length of stay by 1 day. Mascia andcolleagues also implemented a pro-tocol for the use of sedatives, anal-gesics, and NMBAs.13 They demon-strated a significant reduction inNMBA use and sedation drug costs,despite a higher severity of illness inthe guideline population.

Unfortunately, not every reporthas shown this same benefit follow-ing sedation guideline implementa-tion. MacLaren and colleagues imple-mented a sedation protocol in a 24-bed medical/surgical ICU.14 The pro-tocol patients (n=86) had greater useof NMBA, longer time on mechanicalventilation, and a longer time toextubation, but lower drug expendi-ture per hour and improved sedationquality with a greater percentage oftime at the Ramsay sedation goalcompared with historical controlshaving a similar severity of illness.The impact of this protocol wasgreatest on patients requiring long-term sedation.

Further examination of sedationcosts were reported in a three-monthaudit of 155 general ICU patients.15 Asedation protocol was used and the

most common agents were propofol(88%), alfentanil (68%), morphine(30%), and midazolam (25%). Half ofthe patients had an ICU length of staygreater than two days, and thesepatients accounted for 94% of thedrug expenditure.

Long-term patients in the ICU arethe optimal targets for cost-reduc-tion strategies, especially those onmechanical ventilation. The sedativesthemselves constitute less than 1%of the total daily cost of ICU care.However, inappropriate use of seda-tives can contribute to ICU costs bydelaying liberation from mechanicalventilation. Dasta and colleaguesevaluated hospital billing data from253 US hospitals.11 Daily costs werecalculated from hospital-specificcharge-to-cost ratios. The meanincremental daily cost of mechanicalventilation also was calculated. Costswere greatest in the first three days,but then stabilized such that costs forpatients on mechanical ventilationwere an average of $1522 higher peradditional day of ventilation. While aformal evaluation has not been per-formed, these data suggest that ashort-acting sedative may have aneconomic advantage over agentsthat contribute to delayed awaken-ing and extubation.

Summary

A growing body of literature chal-lenges clinicians to incorporate prac-tice guidelines into daily patient care.In order to achieve potential benefitssuch as reduction in duration ofmechanical ventilation, ICU length ofstay, the use of NMBAs and cost ofsedative agents, clinicians will

80

70

60

50

40

30

20

10

00 100 200 300 400

Figure 2. Propylene Glycol Level Correlates with Osmol Gap 7,8


need to measure the impact of theseinterventions, re-evaluating and re-educating as needed to maintainadherence.

References 1. Jacobi J, Fraser GL, Coursin DB, et al. Clinical

practice guidelines for the sustained use of sedatives and analgesics in the critically ill adult. Crit Care Med 2002;30:119-41.

2. Ely EW, Shintani A, Truman B, et al. Delirium as a predictor of mortality in mechanically ventilated patients in the intensive care unit. JAMA2004;291:1753-62.

3. Jones C, Griffiths RD, Humphris G, Skirrow PM. Memory, delusions, and the development of acuteposttraumatic stress disorder-related symptoms after intensive care. Crit Care Med 2001;29:573-80.

4. Vasile B, Rasule F, Candiani A, et al. The pathophys-iology of propofol infuseion syndrome: a simple name for a complex syndrome. Intensive Care Med2003;29:1417-25.

5. Farag E, DeBoer G, Cohen BH, et al. Metabolicacidosis due to propofol infusion. Anesthesiology2005;102: 697-8.

6. Barr J, Zomorodi K, Bertaccini EJ, et al. A double-blind comparison of i.v. lorazepam versusmidazolam for sedation of ICU patients via a pharmacologic model. Anesthesiology 2001;95:286-98.

7. Yahwak JA, Riker RR, Fraser GL, et al. Anobservational pilot study of osmolar gap monitoring to detect propylene glycol toxicity in acritical care population: a rospective series. [abstr]Chest 2003;124:178S.

8. Arroliga AC, Shhabi N, McCarthy K, et al. Relationship of continuous infusion lorazepam to serum propylene glycol concentration in criticallyill adults. Crit Care Med 2004;32:1709-14.

9. Muellejans B, López A, Cross MH, et al. Remifentanil versus fentanyl for analgesia based sedation to provide patient comfort in the intensive care unit: a sandomized, double-blind controlled trial. Critical Care 2004; 8:R1-R11.

10.Dahaba AA, Grabner T, Rehak PH, et al. Remifentanil versus morphine analgesia and sedation for mechanically ventilated critically ill patients. Anesthesiology 2004; 101:640-6.

11.Dasta JF, McLaughlin TP, Mody SH, et al. Dailycost of an intensive care unit day: The contributionof mechanical ventilation. Crit Care Med2004;33:1266-71.

12.Brattebrø G, Hofoss D, Flaatten H, et al. Effect of a scoring system and protocol for sedation on duration of patient's need for ventilator support ina surgical intensive care unit. BMJ 2002;324:1386-9.

13.Mascia MF, Koch M, Medicis JJ. Pharmaco-economic impact of rational use guidelines on the provision of analgesia, sedation, and neuromuscular blockade in critical care. Crit Care Med 2000;28:2300-6.

14.MacLaren R, Plamondon JM, Ramsay KB, et al. A prospective evaluation of empiric versus protocol-based sedation and analgesia. Pharmacotherapy2000;20:662-72.

15.Al-Haddad M, Hayward I, Walsh TS. A prospective audit of cost of sedation, analgesia and neuromuscular blockade in a large British ICU. Anaesthesia 2004;59:1121-5.




PROCEEDINGS

Sedation of critically ill patientshas become an important issuebecause patients can be maintainedfor long periods of time on complexlife support systems. These systemsare often uncomfortable, frequentlypainful, and may require almost com-

Key Points

• Sedation for the treatment of agitation has become an important issue because critically ill patients are being maintained for long periods of time on complex life support systems.

• Successful management of sedation involves complex nursing issues that need to be recognized and addressed to provide effective treatment of agitation and ensure patient comfort and safety.

• Patient-related factors that need to be considered for sedation include the complexity of concommitant disease states, rapidly changing hemodynamic status, and altered pharmacokinetics.

• Challenges in sedation management include differentiating agitation from other conditions such as pain, standardizing the assessment of agitation, and obtaining multidisciplinary agreement about treatment interventions and priorities.

• In addition to bedside decision-making issues, other factors include staff attitudes and perceptions of comfort, family member's perception of agitation, nurses' workload, staffing ratios, and nurses' experience and ability to multi-task amidst rapidly changing conditions.

• New approaches to nursing education and evolving technologies have the potential to offset increased bedside practitioner workload, improve communication, and help address the nursing issues related to the sedation management.

plete patient immobility. Anxiety andagitation are common in the inten-sive care unit (ICU). Despite the fre-quency of these conditions in theacutely ill patient, the bedside imple-mentation of a multidisciplinarytreatment plan to address them

remains challenging. There are com-plex nursing issues related to manag-ing sedation that need to be recog-nized and addressed to provideeffective treatment of agitation andensure patient comfort and safety.These issues include patient-specificvariables, factors affecting bedsidestaff, and interfering forces.

Patient-related Factors

Agitation is a continuum of con-tinuously changing physiologicstates with varying patient behaviorsand responses that depend on theseverity and complexity of their con-ditions. ICU patients typically havecomplex disease states with rapidlychanging hemodynamic status, sothat their agitation and requirementsfor treatment fluctuate over time. Apatient's underlying condition andunpredictable hemodynamic statusoften trigger multi-system organderangements that alter the pharma-cokinetics of medications that areadministered. A single dose of med-ication may be administered, whensuddenly the patient’s blood pres-sure falls, the ventilator starts toalarm, and the nurse has to immedi-ately decide which syringe to grab,which pump to grab, which monitorto assess first, and perhaps other

Nursing Issues Related to Sedation ManagementAnne Pohlman RN, MSN, CCRN, Critical Care Clinical Research, University of Chicago, Chicago, IL


Nursing Issues Related to Sedation Management

responses – and document every-thing. Constantly changing variablesmean that bedside clinicians need toreassess and redefine the goals oftherapy frequently.

ICU patients often have multipleorgan derangements. For example, ageneral care unit patient may havehad some degree of respiratory fail-ure, been intubated, and be trans-ferred to the ICU. Upon arrival in theICU, this patient may have no bloodpressure, no intravenous (IV) access,and need four to five medications, allof which should be administeredsimultaneously. Two or three patientsmay arrive in the ICU with similarproblems. Many ICUs now have 16 to24 beds. Keeping all the necessarypatient information current and doc-umented, and performing all inter-ventions safely in the midst of familymembers calling and other interrup-tions is challenging.

To optimize care in this hectic set-ting, the nurse needs to recognizethe effects and side effects of variouspharmacologic interventions, jugglethe administration of multiple drugs,and decide about priorities duringthe bedside assessments to deter-mine the outcome goals for thepatient. This is no simple task, evenfor an experienced nurse.

Staffing Challenges

Challenges for bedside staff insedation management include differ-entiating agitation from other condi-tions such as pain, standardizing theassessment of agitation, and obtain-ing multidisciplinary agreementabout treatment interventions andpriorities.

The signs and symptoms of agita-tion are fairly obvious. Descriptiveterms commonly used include rest-lessness; thrashing around in bed;pulling at lines, tubes, and restraints;over-breathing the ventilator; anddysynchrony with the current venti-lator settings. Vital-sign abnormali-ties such as tachycardia, tachypnea,and hypertension are common in anagitated patient. The need for assess-ment scales to move from the subjec-tive terms associated with agitationto a standardized scale that can beused to manage sedation has beendiscussed frequently in the literature.The ideal scale should be simple toapply with clearly described gradechanges between levels to allow cli-nicians to titrate sedation to thepatient's condition. Although manyscales and tools are available to mon-itor the degree of agitation, their util-ity to bedside staff in implementingtreatment and intervention plans isnot well understood.

Guidelines and standards of carehave been developed to assist bedside staff with the challenges ofsedative administration. However,follow-up studies have shown thatimplementation of these guidelinesin bedside care has been less thansatisfactory.

It is therefore important to consid-er the challenges facing nurses in theimplementation of such guidelines.

A recent study showed that onlyabout 20% of hospitals were able toimplement daily wake-up assess-ments. Possible reasons for thisinclude assessments not being per-formed, not being documented, or

not being identified in the electronicdatabases.

Even the question of when to perform an assessment can presentmajor challenges. Should it be rightbefore the patient is suctioned, whilesuctioning the patient, or five min-utes after suctioning the patient?Which score is the one being soughtfor the assessment? And which scorerequires the nurse to intervene?

If a patient is asleep, calm andsedate when the nurse enters theroom, that is an ideal sedation score.When a patient is repositioned in bedbefore being suctioned, how longshould the nurse wait for the seda-tion score to return to the previouslevel before doing an intervention,administering a drug, or discontinu-ing a drug? Nurses are willing to dothe different sedation assessmentsand to titrate interventions based onthe scores, but they are not clearwhat exactly is expected of them atbedside, every minute, hour or day.For what portion of a 12-hour periodare they supposed to maintain apatient at a particular assessmentscore? Is 70% of a 12-hour shiftacceptable? Or should the goal besomething else? Despite the short-comings these questions suggest,protocol-driven intervention plansusing agitation scales have beenshown to improve patient outcomessuch as duration of mechanical venti-lation and length of stay in the ICU.

Other Factors

A few studies have addressed bed-side decision-making issues. Otherfactors affecting sedative administra-tion include staff attitudes and per-


Nursing Issues Related to Sedation Management

ceptions of comfort, family member'sperception of agitation, nurses' work-load, and staffing ratios.

An ICU nurse often has to do sev-eral things at once during rapidlychanging situations. The number ofmore senior experienced nurses whoare thought of as being the mainstayin the ICU is dwindling. New, lessexperienced nurses often do nothave the ability to do several thingsat once and still effectively use vari-ous protocols, assessment tools, andperform complex multidisciplinaryinterventions simultaneously.Training ICU bedside staff and main-taining experienced staff members atthe bedside is a major issue.

Addressing Nursing Issues

Critical-care educators need torecognize the many complex issuesassociated with agitation, teach bed-side staff the critical decision-makingskills necessary to optimally manage

agitation, and facilitate the imple-mentation of evidence-based prac-tice strategies in bedside care. Thenewer type of assessments and inter-ventions require a change in mindsetand practice that does not happenovernight. Making these changes willrequire buy-in from nursing leader-ship management, and educators.

The multidisciplinary critical carecommunity has made sedation man-agement a priority in improvingpatient outcomes. National organiza-tions including American Associationof Critical Care Nurses, Society ofCritical Care Medicine, and the JointCommission on Accreditation ofHealthcare Organizations have allrecognized the need for standards ofcare for patients requiring sedation.These standards often overlap withother important standards such asthose for pain management and theuse of restraints. For the bedsidenurse, the challenge is to incorporateall the required standards into a

patient-specific treatment plan thatalso includes strategies for efficientdocumentation of care. Electroniccharting instruments, infusionpumps with real-time data acquisition systems, and medicationadministration logs may helpimprove efficiency.

Summary

Establishing a multidisciplinarystandard of care for assessing, treat-ing, and monitoring agitation in theICU is imperative to optimize patientmanagement and improve out-comes. All disciplines involved in theprocess need to recognize the nurs-ing challenges and find new ways tomake the bedside staff's work easier.Development of new technology tooffset bedside practitioner workloadand improve communication mayhelp to address the challenging nurs-ing issues related to the sedation ofcritically ill patients.



PROCEEDINGS

Daily Interruption of Sedation in MechanicallyVentilated Patients: The ICU Sedation Vacation

John P. Kress, MD, Assistant Professor of Medicine, Department of Medicine, Section of Pulmonary and Critical Care, University of Chicago, Chicago, IL

Mechanically ventilated patientsfrequently suffer symptoms of sub-jective respiratory distress, and themajority receive analgesics and/orsedatives while they are undergoingmechanical ventilation. As experi-ence with intensive care unit (ICU)patient outcomes has grown, ourunderstanding of the complex phar-macology of sedatives and anal-gesics has increased. This has led toan awareness of the need for evi-dence-based protocols to guide theadministration of sedatives to ICUpatients.

Strategies for Delivery ofSedatives in MechanicallyVentilated Patients

Since no single drug can achieveall of the desired effects for sedationand pain control, using a combina-

Key Points

• Sedation and pain control are important components of the treatment of mechanically ventilated patients.

• Directing treatment to specific and individualized goals will better assure that patient needs are met.

• Drug administration strategies based on principles of sedative pharmacology in critical illness should be used.

• Daily interruption of continuous sedative infusions can reduce many of the complications of sedation, including duration of mechanical ventilation and intensive care unit length of stay.

tion of drugs, each titrated to specificendpoints, may be a more effectivestrategy. This approach can allowlower doses of individual drugs andreduce problems associated withdrug accumulation. Both continuousinfusion and intermittent boluses forintravenous drug administrationhave been advocated. Intermittentbolus administration of sedatives andanalgesics may lead to fluctuations inlevel of sedation and increasedemands on nursing time, whichcould take time away from otherpatient care issues. The perceivedbenefits of continuous sedative infu-sions include a more consistent levelof sedation with better patient com-fort. The convenience of continuousinfusion for both patients and care-givers is likely the greatest reason forits popularity.

Strategies for sedation and anal-gesia in critically ill patients shouldideally be based on pharmacokineticand pharmacodynamic principles.Unfortunately, ICU patients oftenexhibit unpredictable responses tomedications so that precise guide-lines for drug administration for allpatients are not possible. For exam-ple, when "short-acting" benzodi-azepines such as midazolam andlorazepam are administered, theyaccumulate in tissue stores, whichprolongs their clinical effect. Otherfactors that change the pharmaco-logic behavior of sedatives and anal-gesics include altered hepatic and/orrenal function, polypharmacy withcomplex drug-drug interactions,altered protein binding, and circula-tory instability. Therefore, titration ofsedatives and analgesics against dis-cernable clinical endpoints is themost common approach.

Further complicating the adminis-tration of sedatives in the ICU is thedramatic differences in the levels ofsedation desired. Since over-sedatedpatients are easier to manage thanunder-sedated patients, cliniciansmay heavily sedate agitated patients.In the initial stages of critical illness,this is appropriate; however, main-


Daily Interruption of Sedation in Mechanically Ventilated Patients:The ICU Sedation Vacation

taining deep levels of sedation afterpatients are stabilized on mechanicalventilation can lead to the problemsof prolonged sedation describedabove.

A daily assessment for organ fail-ures should be routine for every criti-cally ill patient. Assessment of end-organ perfusion and function is par-ticularly important during the resus-citative phases of ICU care. Mentalstatus examination is an importantgauge of brain perfusion. Since braininjury is a devastating complicationof critical illness, acute cerebral dys-function must be detected quicklyand corrected, if possible, before per-manent injury takes place. Sedativesmay interfere with neurologicalassessment.

Daily Interruptions ofContinuous SedativeInfusions

Using a protocol for sedation hasbeen shown to alleviate many prob-lems. It can reduce the duration ofmechanical ventilation, ICU and hos-pital length of stay, and the need fortracheostomy. Protocols assure ade-quate analgesia and sedationthrough frequent assessments ofpatient needs and goal-directedtitration of analgesics and sedatives.A routine protocol of daily interrup-tion of continuous sedative infusionscan reduce many of the complica-tions of sedation, including durationof mechanical ventilation and ICUlength of stay. This strategy allowspatients to spend some of their ICUtime awake and interactive, poten-tially reducing the amount of seda-tive and opiate given, and reducingthe need for diagnostic studies (e.g.,

brain CT scan) to evaluate unex-plained alterations in mental status.Sedation protocols may allow thedepth of sedation to be decreasedwithout compromising the statedgoals of sedation.

At first, the thought of decreasingor stopping sedatives in a critically illpatient who has been agitated maybe unsettling. Clinicians may aggres-sively sedate patients early in theirICU course and then maintain thesame level of deep sedation indefi-nitely. A daily break from sedativescan eliminate the tendency to “lockin” to a high sedative infusion ratethat, while appropriate early in ICUcare, may be unnecessary on subse-quent days. When sedative infusionsare decreased or stopped, tissuestores can redistribute drug back intothe circulation. Interruption of seda-tive infusions may lead to abruptawakening and agitation. This mustbe anticipated by the ICU team toavoid complications such as patientself-extubation. If excessive agitationis noted, sedatives should beresumed. Though attempts at awakening and communication mayfail on a given day, this does not portend inevitable failure on subse-quent days.

When awakening patients fromsedation, for some the ideal may beto reach the brink of consciousnesswithout precipitating excessive agi-tation. Once objective signs of con-sciousness are demonstrated, restart-ing sedatives as needed is reasonable.Restarting the sedative infusion athalf of the previous dose also is rea-sonable. Adjustments from this start-ing point can be individualized topatient needs.

Reducing Complications

Studies of long-term conse-quences of recovery from respiratoryfailure and sedation are limited.Available data suggest that post-ICUdepression is common in patientswho require mechanical ventilationduring critical illness. Post-traumaticstress disorder (PTSD) followingrecovery has also been reported.Some data suggest that lack ofpatient awareness about beingsedated and/or their underlying ill-ness is associated with developmentof PTSD, and that preservation of thisawareness during mechanical venti-lation may reduce this problem.

Previously, we reported how dailyinterruption of sedative infusions canreduce many of the complications ofsedation in the ICU setting.1 Patientswho received either midazolam andmorphine or propofol and morphineby continuous infusion were evaluat-ed. The patients were managed bythe primary ICU team and random-ized to a group where there was adaily scheduled interruption of thesedative and opiate infusions or to acontrol group without a mandatorydaily interruption.

In the interruption group, theduration of mechanical ventilationwas reduced by 2.5 days and ICUlength of stay was reduced by 3.5days compared to the control groupwithout interruption of sedative infu-sions. A significant reduction in diag-nostic studies to investigate unex-plained alterations in mental statuswas also noted. In the control group,27% of patients underwent brain CT,brain MRI or lumbar puncture toinvestigate causes of mental status


Daily Interruption of Sedation in Mechanically Ventilated Patients: The ICU Sedation Vacation

changes; in the sedative-interruptiongroup only 9% of patients underwentsuch studies. When diagnostic testswere done in these patients, theresults were more often conclusive inthe study compared to the controlgroup. Only 25% of the patients inthe control group had a test resultthat explained their mental statuschanges, compared with 50% of thepatients in the sedative-interruptiongroup. The patients in the sedative-interruption group averaged 86% oftheir ICU days awake and able to fol-low commands, compared with thecontrol group, who averaged only 9%of their ICU days awake and able tofollow commands. The amount ofmidazolam and morphine adminis-tered was also significantly reducedin the sedative-interruption group.

The strategy of daily sedativeinterruption allowed a focuseddownward titration of sedative infu-sion rates over time, streamliningadministration of these drugs andminimizing the tendency for accu-mulation. Based on this study, webelieve that one approach to opti-mizing patient outcome is a dailyinterruption of sedative infusions topermit physician and patient com-munication and to facilitate the phys-ical examination.

While it is clear that sedation pro-tocols can improve patient out-comes, it is important to recognizethat a treatment protocol is only a

guide and cannot address every clin-ical situation. Limitations of studiesof the use of protocols include poten-tial lack of generalizability (e.g., a sin-gle academic medical center may bedifferent than a community setting),lack of blinding, lack of reporting of“other” outcomes such as long-termfollow-up, and uncertainty aboutlevel of compliance needed to assuredesired outcomes. A multidiscipli-nary, cooperative approach is neces-sary to assure compliance and suc-cessful implementation of protocols.

Summary and Conclusion

Sedation and pain control areimportant components of the treat-ment of mechanically ventilatedpatients. Directing treatment to spe-cific and individualized goals will bet-ter assure that patient needs are met.All currently available agents for usein mechanically ventilated patientshave limitations, and complicationsrelated to the use of these agents arecommon. Rather than seeking anideal drug, drug administrationstrategies based on principles ofsedative pharmacology in critical ill-ness should be used. When thesedrugs are given to individualpatients, establishing specific goalswill allow the use of rational adminis-tration strategies, which should leadto improvement in both short- andlong-term outcomes.

References1. Kress JP, Pohlman A, O'Connor MF, et al.: Daily

interruption of sedative infusions in critically ill patients undergoing mechanical ventilation. N Engl J Med 2000;18;342(20):1471-7.


PROCEEDINGS

IV Medication SafetySystems: CQI Data

In October 2002 the St.Joseph’s/Candler Health System (SJC)in Savannah, Georgia implemented amodular “smart” intravenous (IV)infusion systema with safety softwarethat incorporates a pre-defined set ofdosing limits for infused medicationscommonly used in our hospitals.These limits are collectively known asGuardrails® limits and are designedto alert clinicians if a specific pro-

Key Points

• Smart pump technology provides comprehensive continuous quality improvement (CQI) data that can identify medication safety issues for drugs given by the IV route.

• Although clinical practice guidelines for intensive care unit (ICU) sedation have been developed by the Society of Critical Care Medicine, in many settings they may not be consistently used.

• Propofol is a widely used, highly effective IV sedative for ventilator-assisted patients in the ICU.

• Our CQI data demonstrate overuse of propofol in these patients that maybe associated with iatrogenic consequences not previously recognized.

• Preliminary analysis of outcomes and other clinical variables suggest that reducing the variability and doses of propofol has improved patient care and reduced treatment costs in patients with mechanical ventilation in the ICU.

grammed dose of a medication isoutside a minimum effective or max-imum safe dose for the drug. Infusioncannot begin until the alert isaddressed. The intent of theGuardrails® limits is to prevent med-ication errors. A total of 525 smartinfusion systems were installed in ourtwo tertiary-care referral hospitals.Each instance of a programmed doseless than or greater than theGuardrails® limits is electronicallyrecorded as an “event” for continuousquality improvement (CQI) purposes.

SJC implemented smart IV infu-sion technology to improve the safe-ty of medications with the highestpotential to cause harm. Since imple-mentation, event data have beenaccumulated in the computer “brain”for each device. All events are collect-ed and saved by the system for futureCQI analysis. These data have beensynthesized for analysis of our initialexperience with the safety softwareat SJC. The data indicate that mostalerts were warnings of the possibili-ty of drug overdose. The data alsoindicate that 7.2% of events resultedin the nurse canceling the adminis-tration process or resetting the pump(averted errors). These warningsinvolved multiple medications,including some of those identified bythe USP from their MEDMARX® data-base as being associated with thehighest liability for harm.1

Smart IV infusion technology atSJC has provided a way to documentactual drug administration activity atthe bedside. The system allows us toidentify and prevent medicationerrors, record pump programmingsteps that occurred before an inter-vention, record all actions taken bythe nurse even if he/she continuedthe administration process after hav-

ICU Sedation with Propofol:Measuring and Reducing Variation

Ray R. Maddox, PharmD, Director, Clinical Pharmacy, Research and Pulmonary Medicine, St. Joseph's/Candler Health System, Savannah, GA


ICU Sedation with Propofol: Measuring and Reducing Variation

ing been “warned” that a pump pro-gramming instruction was outside ofthe guardrails limits, and implementcorrective actions to reduce thepotential for future medicationerrors.

Maintaining optimal comfort andsafety for mechanically ventilatedcritically ill patients is an importantgoal. The Society of Critical CareMedicine (SCCM) has developed clin-ical practice guidelines to assist clini-cians in the use of sedatives and anal-gesics for these patients (Table 1).2

Indications for sedative agents arenot well defined; nonetheless, theyare commonly used adjuncts for thetreatment of anxiety and agitation inthe ICU.2 Several drugs, including thebenzodiazepines (diazepam, loraze-pam, midazolam), propofol, and cen-tral-agonists (clonidine, dexmedeto-midine), are used to sedate themechanically ventilated patient.

Propofol Overuse

Among the commonly used agents,propofol (2,6-diisopropylphenol) hasbeen favorably compared to the ben-zodiazepines, particularly midazo-lam, because it reduces the timeneeded for recovery of spontaneousrespiration and successful weaningof patients from the respirator (Table2).3-5 It has a rapid onset and shortduration of sedation once discontin-ued.6 Propofol is available as an emul-sion that provides 1.1 kcal/mL fromfat and is an important caloric sourcefor patients receiving a continuousinfusion.7

Propofol is widely used in manyICU settings and is a convenient andefficient pharmacologic agent for

sedation and ventilator managementbecause of its rapid onset and cessa-tion of action. ICU nurses and physi-cians can titrate patients to a desiredlevel of sedation with minimal con-cern for the potential of overdose ornegative effects.

Serious adverse events can occurwith propofol and include a numberof metabolic,8-11 neurologic,12-13 car-diac,14-16 infectious,17 pulmonary18 and

local reactions.7 Propofol can alsocause anaphylactoid type reac-tions,19-20 and tachyphylaxis21 hasbeen documented.

In spite of the availability of clini-cal practice guidelines for the use ofsedatives including propofol in thecritically ill adult patient, our CQI datashow that clinicians may administerthis drug with insufficient concern forthe quantity of drug infused per unit

Table 1. ICU Sedation Guidelines2

• A sedation goal or endpoint should be established and regularly redefined for each patient; regular assessment and response to therapy should be systematically documented.

• The use of a validated sedation assessment scale is recommended.

• The titration of the sedative dose to a defined endpoint isrecommended with systematic tapering of the dose ordaily interruption with retitration to minimize prolongedsedative effects.

• The use of sedation guidelines, an algorithm, or a protocol is recommended.

Table 2. Propofol as an ICU Sedative3-5

• Highly lipophilic; extensively distributed in tissue; fat-solubilized emulsion

• Vd ≈ 60 L/kg

• Rapid action: 1-2 min; short duration after discontinuation

• Tri-phasic elimination process - t½ (α=1-8 min; β=1.5-12.4 hr; δ=26-32 hr) after infusion

• Serum conc for sedation: 0.14 - 1.90 mcg/mL constant infusion

• Dosing is weight dependent: 5-80 μg/kg/min range

• Tachyphylaxis can occur

• Expensive ( ≈ $300/day/70-kg pt)



of time and for the potential occur-rence of adverse events. Physiciansroutinely order this medication byindicating “Diprivan Drip,” “TitrateDiprivan,” or “Sedate with Diprivan.”In a 9-month period, the averagedose of propofol was 100mcg/kg/min for patients given con-tinuous infusions. The recommendedmaximum rate of infusion for thisindication is 80 mcg/kg/min whengiven for sedation of mechanicallyventilated critically ill adult patients.2

Data from a national cohort of 52hospitals where “smart” infusiondevices are used suggest that ourinstitution is not unique. Propofol isgiven using frequent bolus doses andadministered at rates that exceed theupper limits of recommended dosingby the SCCM.22 CareFusionPharmacy Services manages morethan 180 acute care hospital pharma-cies throughout the nation. Propofolranks fifth among all purchases ofpharmaceuticals in these hospitalsdespite the fact that it is available asa generic product. This would sug-gest that this drug is being used inlarge quantities in many institu-tions.22 It currently represents 81% ofthe expenditures for all IV sedativesused in these facilities. 23

Best Practice Improvements

In order to address the overuse ofpropofol at SJC a standardized ICUsedation order set was developedwhich incorporates SCCM good clini-cal practice guidelines for sedatives,analgesics, and neuromuscularblocking agents used in ventilator-assisted patients. The use of thisorder set and the accompanying

sedation management protocol isrequired at our institution. The proto-col and order set requires physiciansto prescribe a specific level of desiredsedation for each patient dependentupon the physical condition and dis-ease entity being treated. Nursesthen dose propofol and/or othersedatives to maintain that level ofsedation using the Motor ActivityAssessment Scale (MAAS) sedationscore.24 Patients are given scheduled“sedation vacations” as appropriateto assess mental status.

Results

Data from our hospitals indicatethat this change in process has dra-matically improved how propofol isbeing used in our ICUs. Propofol dos-ing alerts from the infusion pumpshave been reduced by greater than50%; the number of reprogrammeddoses after alerts, which indicateaverted medication errors, has risensharply. The number of bolus dosesof propofol has been almost elimi-

nated. Preliminary data from ananalysis of ventilated patients treatedwith propofol in the ICU before andafter the implementation of thesedation protocol and order set isdescribed in Table 3. Reductions inthe total cost of drug, patient days onventilators, and ventilator-associatedpneumonia appear to be significant;however, further data collection andanalysis is underway. Questionsbeing addressed in our analysisinclude: Are there differences in out-comes among patients being treatedwith/without sedation guidelines inplace? Are there differences in thenumber and type of guardrails alerts?And are there differences in theamount and cost of propofol admin-istered?

Conclusions

Propofol is an important, potent,and expensive ICU sedative with aninherent potential for systemic,metabolic, neurologic and cardiovas-cular toxicity. CQI data from smart IV

7.05.8VAP Rate *

5.5%8.2%% with VAP

7.914.0Ave Vent Days

1,2832,912Total Vent Days

$3,990$8,531• Cost per pt

$650,399$1,774,395• Total Cost

Propofol Use

Post (n=163)Pre (n=208)

*VAP Rate = (VAP infections/vent days)*1000

Table 3. Changes after ICU SedationProtocol Implemented



infusion safety systems indicate thatit frequently is dosed indiscriminate-ly in many hospitals. Sedation shouldbe more closely monitored by nursesand physicians with specific dosagetitration using good clinical practiceguidelines. CQI data from smart IVinfusion systems can help to identifyunrecognized problems with med-ication management that might becontributing to iatrogenic patientmorbidity and mortality.

References1. Williams C, Maddox RR. Implementation of an i.v.

medication safety system. Am J Health-Syst Pharm. 2005; 62:530-6.

2. Jacobi J, Fraser GL, Coursin DB, et al. Clinical practice guidelines for the sustained use of sedatives and analgesics in the critically ill adult. Crit Care Med 2002; 30:119-41.

3. Carrasco G, Molina R, Costa J, et al. Propofol versus midazolam in short-, medium-, and long-term sedation of critically ill patients: A cost-benefit analysis. Chest 1993; 103:557-64.

4. Kress JP, O'Conner MF, Pohlman AS, et al. Sedation of critically ill patients during mechanical ventilation: A comparison of propofoland midazolam. Am J Respir Crit Care Med 1996; 153:1012-8.

5. McCollam JS, O'Neil MG, Norcross ED, et al. Continuous infusions of lorazepam, midazolam, and propofol for sedation of the critically ill surgery trauma patient: a prospective, randomized comparison. Crit Care Med 1999; 27:2454-8.

6. Bailie GR, Cockshott ID, Douglas EJ, et al.Pharmacokinetics of propofol during and after long term continuous infusion for maintenance of sedation in ICU patients. Br J Anaesth 1992; 68:486-91.

7. Product Information. Drug Facts and Comparisons 2004, pp 990-4.

8. Valente JF, Anderson GL, Branson, RD, et al. Disadvantages of prolonged propofol sedation inthe critical care unit. Crit Care Med 1994; 22:710-2.

9. Eddleston JM, Shelly MP. The effect on serum lipidconcentrations of a prolonged infusion of propofol-hypertrigliceridaemia associated with propofol administration. Intensive Care Med 1991;7:424-5.

10.Gottardis M, Khunl-Brady KS, Koller W, et al. Effect of prolonged sedation with propofol on serum triglyceride and cholesterol concentrations. Br J Anaesth 1989; 62:393-6.

11.Stetlow EB, Johari VP, Smith SA, et al. Propofol-associated Rhabdomyolysis with cardiac involvement in adults: chemical and anatomicFindings. Clin Hem 2000;46(4):577-81.

12.Collier C, Kelly K. Propofol and convulsions, the evidence mounts. Anaesth Intensive Care 1991; 19:573-5.

13.McHugh P. Acute choreoathetoid reaction to propofol. Anesthesia 1991; 46:425.

14.Munoz R, Goldberg ME, Cantillo J, et al. Perioperative arrhythmias with a propofol-based anesthetic. J Clin Anesth 1991; 3:149-52.

15.Perrier ND, Baerga-Valera Y, Murray MJ. Deathrelated to propofol use in an adult patient. CritCare Med 2000;28:3071-4.

16.Parke TJ, Stevens JE, Rice ASC, et al. Metabolic aci-dosis and fatal myocardial failure after propofol infusion in children: five case reports. Br Med J1992; 305:613-6.

17.Bennet SN, McNeil MM, Bland LA, et al.Postoperative infections traced to contamination of an intravenous anesthetic, propofol, N Eng J Med 1995; 333:147-54.

18.El-Ebiary M, Torres A, Ramirez J, et al. Lipid deposition during long-term infusion of propofol,Crit Care Med 1995; 23:1928-30.

19.Laxenaire MC, Mata-Bermejo E, Moneret-Vautrin DA, et al. Life-threatening anaphylactoid reactions to propofol (Diprivan). Anesthesiology1992; 77:275-80.

20.DeLeon-Casasola OA, Weiss A, Lema MS. Anaphylaxis due to propofol. Anesthesiology1992; 77:384-6.

21.Deer TR, Rich GF. Propofol tolerance in a pediatricpatient. Anesthesiology 1992; 77:828-9.

22.Crass R. CareFusion CQI Consulting Services, CareFusion Hospital Pooled Data, 2005.

23.Ham C. CareFusion Pharmacy Services - CAT2

data on file, 2005.

24.Devlin JW, Boleski G, Mlynarek M, et al. Motor Activity Assessment Scale: a valid and reliable sedation scale for use with mechanically ventilated patients in an adult surgical intensive care unit. Crit Care Med 1999; 27:1271-5.



PROCEEDINGS

The Problem of Sedation

Agitation interferes with thera-peutic procedures and compromisessafety of the patient and medicalstaff. Most sedation in intensive careis given to reduce agitation.1

However, the use of drugs to controlagitation may also compromise care.

The administration of medicationsfor sedation is one of the most arbi-trarily applied therapies in the inten-sive care unit (ICU), but there is littleagreement on any particular method.As a result, practices and drug selec-tion vary widely. The assessment ofagitation is highly subjective and

Key Points

• If not used carefully, medications being used for sedation can result in cycling between agitated and over-sedated states, over-sedation, and increased length of stay.

• If agitation is effectively controlled, over-sedation can be avoided.

• Consistent delivery of medications to control sedation provide better control of agitation compared with titration using subjective measures of patient response.

• Use of agitation level, not sedation level, as endpoint is a better approach to sedation management.

• The use of computer-assisted technology to improve titration of sedation based on changes in agitation level has been successful.

dependent upon intuition and theexperience of staff. Frequently thereis concomitant use of many differentdrugs. Poor control of sedation canresult in cycling between an agitatedand sedated state, over-sedation,and increased length of stay.1

To understand sedation manage-ment, it is necessary to understandthe process of managing agitation.Recent literature has focused on clin-ical practice improvements throughuse of sedation-agitation scores1,2 andinterruption of continuous infusions.

These approaches do not addressthe fundamental problem of agita-tion control, which is a drug-dosing

control problem best addressed byaccurate measurement and under-standing of the underlying dynamics.

The inability to effectively controlagitation is an important factor asso-ciated with over-sedation. If agitationis controlled, over-sedation can beavoided. This eliminates the need forsubjective sedation-agitation scoresor interruption of sedation treat-ment.

Development of a Semi-automated Closed-loopSedation Control System

A sedation management algo-rithm was developed that providedfor simultaneous assessments ofsedation and agitation.2 Publishedscoring systems place agitation andsedation at either end of a linearscale.3,4 Critically ill, agitated patients,however, nearly always have areduced level of consciousness that isindistinguishable from the effects ofsedation. It is therefore impossible toselect a single point on a linear scalethat defines a patient's agitation-sedation state. This problem can beresolved by splitting the RikerSedation-Agitation Scale4 into sepa-rate sedation and agitation scales(Table 1).

Automation in Critical Care: Future Directions in Sedation Delivery

Geoffrey M. Shaw1, J. Geoffrey Chase2, Christopher E. Hann2, Richard Dove3, Kathryn Greenfield3

1 Dept of Intensive Care Medicine, Christchurch School of Medicine and Health Sciences, Christchurch, New Zealand; 2

Dept of Mechanical Engineering, Centre for Bio-Engineering, University of Canterbury, Christchurch, New Zealand; 3

Dept of Medical Physics and Bio-Engineering, Canterbury District Health Board, Christchurch, New Zealand



Consistent delivery of sedationmedications to all patients providesbetter control of agitation than titra-tion of a number of drugs usingwidely variable subjective measuresof patient response. Since there areno objective measures of agitation, afixed mixture of drugs, such as mor-phine (1.0 mg/mL) and midazolam(0.5 mg/mL), and age-adjusted bolussizes can be used for consistent seda-tion delivery.

Boluses of a fixed mixture of seda-tion drugs are given to reduce agita-tion using a Graseby 3500 syringepump. The background infusion ratecan be set as a proportion of the totalamount of the drug combinationgiven over the preceding 4 hours tocontrol agitation. Only a portion ofthe background sedation is carriedthrough to the next time period. Aslong as agitation is present, a back-ground infusion is maintained. Asthe agitation diminishes, less seda-tion is required and the backgroundrate is reduced accordingly.

The intention is to use agitationlevel, not sedation level, as the pri-mary endpoint in sedation manage-ment. When mild agitation cannot

be tolerated (e.g., in severe headtrauma), thresholds for treating agi-tation are reduced and an increasedlevel of sedation is tolerated. In mostcases, sedation is the useful sideeffect of treating agitation.

From time to time additional seda-tion is needed to provide anaesthesiato carry out uncomfortable or painfulprocedures. This differs from usingthe same drugs to reduce agitation.Feedback control of anaesthesiarequires measurements of sedationand analgesia, whereas control ofagitation requires measurements of

agitation response. Each therapy hasa different clinical end point; hence,this additional sedation does notinfluence the background rate.

This algorithm has been success-fully trialed using a paper flow chart.In virtually all cases, nursing staffreadily titrated sedation to an accept-able and safe level of agitation.

Computerized SedationDelivery Interface: TheInfuse-Ritea

A modification of the algorithmallowing hourly adjustments of thebackground infusion was incorporat-ed into a simple user interface con-nected to the serial port of a Graseby3500 syringe pump7 (Figure 1).Patient information (includingweight for children less than 14years) is uploaded to the Infuse-Riteusing a personal computer (PC).Syringe labels are printed with thepatient's name, hospital number andappropriate doses of morphine andmidazolam that are weight-adjustedfor children. The Infuse-Rite is then

Table 1. Modified Riker Sedation-Agitation Scale4

Sedation scale Agitation scale

-3 Unresponsive +3 Dangerously agitated

-2 Responsive to tracheal suction or pain +2 Moderately agitated

-1 Responsive to voice +1 Mildly agitated

0 Calm and cooperative (or sleeping)

Figure 1. Infuse-Ritea

Infuse-Rite mounted on Graseby 3500 syringe pump (left); the storage rack with PC used to input and retrieve data (right)



clipped onto the syringe pump andconnected to its serial port. Nursingstaff select “Infusion bolus” to controlagitation or “Procedural bolus” toincrease sedation for uncomfortableprocedures. Each hour the previoushour's drug dosage is displayed,which is recorded by nursing staff ona 24-hour chart. When sedation isceased, the Infuse-Rite is reconnect-ed to the PC and its drug administra-tion data is downloaded. The drugdelivery record is printed for thepatient's file and stored for auditing.

The Infuse-Rite provides a simpleuser interface allowing titration ofsedation when there are changes inagitation level. Its success relates toease of use, consistency in terms ofsimple input rules, and algorithmical-ly determined outputs.

The rules are easily understoodand the background infusion is con-tinuously reduced in the absence ofagitation, minimizing over-sedation.The use of agitation levels as clinicalendpoints obviates the need forcombined sedation-agitation scales.

Results

This approach has been success-ful, with nursing staff reporting highlevels of satisfaction with regard toagitation-sedation control, patientand personal safety, and efficient useof time. Thirty nurses from theChristchurch Hospital ICU with morethan 1 year’s experience were askedto participate in a survey comparingthe computerized system with theirprevious experiences.6 The semi-automated sedation delivery systemwas rated higher in all areas: pain anddiscomfort, sedation control, agita-

tion control, time efficiency, ease ofadministration, patient safety, per-sonal safety, and legal safety.6

Data from patients admitted toChristchurch Hospital ICU betweenMarch 2002 and October 2005 whowere sedated using the Infuse-Ritewere available for analysis. A total of1070 patients who received sedationfor more than 12 hours were includ-ed. Sedation doses are presented asmilligrams of morphine. Midazolamconsumption is half this value andomitted for clarity. Our data contain120,141 boluses and infusion-ratechanges to control agitation andnearly 36,000 additional boluses forprocedures over 68,694 hours (7years, 10 months). More than 156,000doses or infusion adjustments weremade without error. The Infuse-Ritewas also highly adaptable and ableto meet a wide range of demands.For example, one patient received 26times more than the average dose(Table 2).

It is often felt that patients receivegreater amounts of sedation at nightto allow for sleep and rest, hence theearly morning cessation of sedation

as described by Kress and col-leagues.6 Our data show a significantreduction in drug administrationaround noon compared with mid-night (p= 0.0133).

We hypothesized that large varia-tions in the cycling period of seda-tion delivery are an indicator ofinconsistent sedation delivery andpoor control. Specifically, large varia-tions in the time period betweenpeaks and troughs of sedation deliv-ery are seen to indicate variability inconsistency of care and hence poorcontrol of agitation. The log of themaximum value of the peak-to-trough period is thus correlated withsedation duration, indicating thatgreater variability and poorer controlcontribute to prolongation of seda-tion time (Figure 2).

The linear relationship on the logscale indicates an exponential rela-tionship between variability in deliv-ery and sedation duration. This resulthighlights the potential importanceof controlling this metric in any seda-tion delivery protocol. Thus, thisresult would indicate that sedationinterruption routines are effective

Table 2. Administration of morphine via Infuse-Rite

Sedation method Mean delivery Max Min Hourly rate (mg) (mg) (mg) (mg/h)

Procedural boluses 36.2 950.0 0.0 0.6

Sedation background 136.8 3795.6 0.3 2.1

Sedation boluses 88.2 2103.0 0.5 1.4

Total sedation 261.2 6848.6 0.8 4.1



primarily because they reduce deliv-ery in situations of poor control.

Conclusions

Computerized sedation interfaceeliminates drug administration errorand provides greater consistency andcontrol of patient agitation. TheInfuse-Rite algorithm with its goal ofconsistently weaning drug dose inthe absence of patient agitation aimsto minimize drug dose and over-sedation.

The use of the semi-automatedInfuse-Rite controller still has limita-tions. In particular, assessment of agi-tation by clinicians is subjective andprone to error, while the controller'soutput uses fixed drug ratios and

parameters that almost certainly donot suit every patient. It also pro-duces a slowly variable infusion ratewith a fixed rate of variation, in spiteof the fact that observed agitationdynamics are rapid and highly vari-able. The interface does provide highquality data and new insights intothe dynamics of patient agitation.

Research should now focus on val-idating innovative and physiological-ly-based methods of optimizingsedation. This semi-automated con-troller provides consistency of seda-tion delivery, albeit with subjectiveand inconsistent inputs. In the futurean ideal sedation controller wouldimmediately respond to objectivechanges in agitation with timely,

appropriate boluses of sedation. Thiswould provide a true, objective andphysiologically-based feedback con-trol system.

References1. Jacobi J. Clinical practice guidelines for the

sustained use of sedatives and analgesics in the critically ill adult. Crit.Care Med 2002;30:119-41.

2. Rudge AD, Chase JG, Shaw GM, et al. Impact of control on agitation-sedation dynamics. Control Eng Prac 2005;13:1139-49.

3. Sessler CN, Gosnell MS, Grap MJ, et al. The Richmond Agitation-Sedation Scale: validity and reliability in adult intensive care unit patients. Am J Respir Crit Care Med 2002;166:1338-44.

4. Riker RR, Picard JT, Fraser GL. Prospective evaluation of the Sedation-Agitation Scale for adult critically ill patients. Crit Care Med 1999;27:1325-9.

5. Kress JP, Pohlman AS, O'Connor MF, et al. Daily interruption of sedative infusions in critically ill patients undergoing mechanical ventilation. N Engl J Med 2000;342:1471-7.

6. Shaw GM, Chase JG, Rudge AD, et al. Rethinking sedation and agitation management in critical illness. Crit Care Resusc 2003; 5:198-206.

7. Greenfield KM, Dove RA, Shaw GM. Optimisation of sedation therapy within an intensive care setting. Proceedings. Engineering and Physical Sciences in Medicine Conference, October 2001, Fremantle, Australia. [Abstract].

a. The InfuseRite was manufactured by the Department of Medical Physics and Bioengineering at Christchurch Hospital. (Richard Dove and Kathryn Greenfield).

Log (duration of sedation) (days)

Figure 2. Comparison of the maximum variation of patient drug dosing cycle period(y-axis) and the duration of sedation (x-axis) for each of the 1070 patients.The naturallog scale and resulting linear relationship indicates the linear exponential relation-ship between these variables. As the peak variability of sedation delivery cyclingincreases, hence loss of agitation control, sedation times increase.

Figure 2. Comparison of maximum variation of patient drug dosingcycle period (loss of agitation control) with duration of sedation



PROCEEDINGS

Target-controlled Infusion: IntroductionMichel MRF Struys, MD, PhD; Professor in Anesthesia and Research Coordinator;

Department of Anesthesia; University Gent; Gent, Belgium

Target-controlled infusion (TCI) iswidely used in Europe and potential-ly could be used in the United States,if cleared to market by the Food andDrug Administration. TCI is based onthe use of pharmacokinetic model-ing and calculations to control infu-sion to achieve a predicted concen-tration of drug. An understanding ofthe drug dose-response relation isessential to understanding TCI.

Drug Dose-responseRelation

The dose-response relation can bedivided into three parts: the relationbetween administered dose andplasma concentration (the pharma-cokinetic phase), the relationbetween effector-organ concentra-tion and clinical effect (the pharma-codynamic phase) and the couplingbetween pharmacokinetics andpharmcodynamics. The ultimate goal

when administering a particular doseof a drug is to obtain the desired clin-ical effect, for which a specific thera-peutic concentration of the drug atthe site of action (e.g., the receptorsite or effect site) is necessary.

A drug interacts with a receptorand the pharmacological effectresults. For most drugs, this effect-site concentration is not the plasmaconcentration, and the effect-siteconcentration is not measurable. Byunderstanding the relation betweenthe therapeutic plasma concentra-tion and the clinical effect, however,it is possible to select a target con-centration. This target concentrationis defined as the plasma concentra-tion that, on the basis of availableinformation, is most likely to yield thedesired effect.1 Figure 1 shows therelationship between administereddose and resulting effect-intensity ofa drug. Pharmacokinetic factors suchas distribution, metabolism, and/orexcretion determine the relationshipbetween drug dose and effect-sitedrug concentration.

The above is also true for anes-thetic agents: the drug-effect site isnot the plasma,2 and the concentra-tion of the drug at the site of action isnot measurable The apparent rate of

Key Points

• Measurement of the plasma level of a drug is often used to individualize drug therapy, but drug concentration at the site of action is more important in determining patient response.

• It is very difficult to measure the concentration of a drug at the site of action (effect site) compared to the plasma and is not practical in the clinical setting.

• The apparent rate of drug flow into and from the site of action can be characterized by the time course of the drug effect.

• Pharmacokinetic modeling is one way to predict the concentration of a drug at the effect site, so that measurement of the concentration of the drug at that site is not necessary.

• Target-controlled infusion (TCI) technology uses pharmacokinetic modeling and calculations to predict drug concentration at the site of action and can rapidly achieve and maintain the desired predicted concentration.

• Clinicians can use TCI to improve the control of drug concentration at the site of action and clinical response.


Target-controlled Infusion: Introduction

drug flow into and from the site ofaction can be characterized by thetime course of the drug effect.3 Theconcentration at the site of action iscalled “the effect-site concentration,”and the corresponding compartmentis called “the effect-site compart-ment.” The effect-site concentrationis characterized by a temporal disso-ciation (or delay) compared with theplasma concentration. By knowingthe effect-site concentration, thistemporal problem can be overcome.3

Ultimately, the effect-site concentra-tions can be the same as the plasmaconcentration when equilibrated.

The effect-site concentration par-allels the clinical effect of the drug.This means that the effect can be cor-related with an effect-site concentra-tion, which allows for a quantitativeapproach.

Many intravenous (IV) drugs areadministered on the basis of dose perkilogram of body weight by means of

manually controlled infusion pumps.Measurement of either the plasma orthe effect-site concentration of an IVdrug is not currently possible. Withthe development of short-actingdrugs, advanced computer technolo-gy, and a better understanding ofpatients' individual pharmacokinet-ics and pharmacodynamics, it has

been possible to develop accurate,computer-controlled, IV drug deliv-ery systems for clinical use. Becauseof its direct relation to clinical effect,the effect-site concentration can beused to quantify the Ce50 (clinicaleffect is seen in 50 % of the patientpopulation) or Ce95 (clinical effect isseen in 95 % of the patient popula-tion). For examples, see Figure 2.4

Target-controlled Infusion

Based on the pioneering work ofKruger-Thiemer5 and Schwilden etal.,6 TCI techniques use pharmacoki-netic modeling and calculations topredict a set concentration in one ofthe pharmacokinetic compartments.TCI devices rapidly achieve and main-tain a desired predicted concentra-tion in the specific compartment. Thespecific target concentration is setafter having evaluated the patient'sclinical signs. When a certain concen-tration in the plasma compartment istargeted, this is called “open-loopplasma-controlled TCI.” When a cer-tain concentration at the effect com-

Figure 1. Schematic Representation: Pharmacokineticand Dynamic Processes1

DISTRIBUTION EXCRETION

METABOLISM

DOSE

PLASMA CONCENTRATION

BIOPHASE CONCENTRATION

DRUG-RECEPTOR INTERACTION

EFFECTUATION

EFFECT

Plasma alfentanil (ng/mL)

Adequate anesthesia

skin closure

skin incision

intubation

EEG

0 200 400 600 800 10000

100

50

Figure 2. Adequate Anesthesia4


Target-controlled Infusion: Introduction

partment is targeted, this is called“open-loop effect-site controlled TCI.”

The use of TCI enables the clinicianto continuously control the drug con-centration in the plasma or at theeffect-site, and to administer anes-thetics according to their pharmaco-kinetic profile, without having to cal-culate the compartmental drug con-centration on the basis of highlycomplex polyexponential functions.Instead of measuring the plasma oreffect-site concentration (which isimpossible in vivo), the computer-controlled infusion device predictsthe drug concentration, which allowsan anesthetist to dose an IV drug to atarget plasma or effect-site concen-tration, instead of on the basis ofdose per kilogram of body weight.

TCI infusion pumps can be usedoptimally only when three elementshave been carefully established. First,the pharmacodynamic-pharmacoki-netic model that controls the pumphas to work accurately. Second, thepharmacokinetic parameters of aparticular drug entered into the com-puter model have to match the phar-macokinetics of the patient. Third,the pharmacodynamics of theadministered drug have to be welldefined, in order for the anesthetistto attain the plasma concentrationrequired for the desired effect.

Summary

The safe and effective use of rap-idly acting drugs with a short dura-tion of action will benefit from newtechnology that can target infusion

rates based on pharmacokinetic andpharmacodynamic characteristics ofindividual patients. TCI is such a tech-nology, and has the promise ofimproving the use of anesthetics andanalgesics if cleared for market use inthe clinical setting by the FDA.

References1. Danhof M. In: Breimer DD, Crommelin DJC,

Midha KK, Eds. Topics in Pharmaceutical Science. Noordwijk, Amsterdam Med. Press BV, 1989:573-86.

2. Shafer SL. Towards optimal intravenous dosing strageties. Sem Anesth. 1993;12:222-34.

3. Schneider TW, Minto CF, Stanski DR. The effect compartment concept in pharmacodynamic modeling. Anaesth Phramacol Rev 1994;2:204-13.

4. Ausems ME, Hug CC, Jr., Stanski DR, et al.Plasma concentrations of alfentanil required to supplement nitrous oxide anesthesia for general surgery. Anesth 1986:65;362-73.

5. Kruger-Theimer E. Continuous intravenousinfusion and multicompartment accumulation. Eur J Pharmacol 1968;4:317-24.

6. Schwilden H. A general method for calculating the dosage scheme in linear pharmacokinetics. Eur J Clin Pharmacol 1981;20:397-86.


PROCEEDINGS

Earlier generations of target-con-trolled infusion (TCI) systems provid-ed the capability for clinicians toadminister and adjust steady-state“target” blood concentrations of adrug. Since then it has been recog-nized that there is a time delay beforeequilibration occurs between theconcentration in the blood and at thesite of action (the “effect-site”) of thesedative and anesthetic agents.Objective measures of drug effect,such as parameters derived from anelectroencephalogram (EEG), pro-vide indirect assessments of thetime-course of blood-effect-siteequilibration, and allow the calcula-tion of a blood–effect-site equilibra-tion rate constant. By using this rateconstant and iterative mathematical

processes, newer generations of TCIsystems have been able to imple-ment effect-site-concentration tar-geting.

This (blood- or effect-site-concen-tration targeting) does not mean thatTCI systems offer a sedation or anes-thetic on-off switch (i.e., the clinicianselects an “appropriate” target con-centration and the pump does therest). One reason is that these sys-tems rely on pharmacokinetic mod-els to calculate infusion rates andestimate blood and effect-site con-centrations. The estimated concen-trations may not be accurate,because there is considerable vari-ability in the way that patients dis-tribute and metabolize drugs. Moreimportantly, there is a very large vari-

ability in pharmacodynamic sensitiv-ity among patients. An inadequateblood concentration for one patientmay be excessive for another.Titration of dose (estimated bloodconcentration) to clinical effect isalways required. Therefore, there areseveral compelling reasons to recom-mend the use of TCI systems for pro-cedural and ICU sedation.

Ease of Use

Anesthesiologists new to TCI tech-nology find it easy to use, report thatthe systems make it easy to adjustthe depth of anesthesia, and soondevelop a preference for TCI systemsover manual control of infusionrates.1 This is probably the major rea-son why in the United Kingdom andmost of Europe, the majority of anes-thetists administering total intra-venous anaesthesia use a TCI system.

TCI use in the ICU environment isnot as widespread as in the operatingroom. McMurray and colleaguesreported the outcome of a multi-cen-tre study of the use of propofol TCIsystems for ICU sedation in 122patients.2 In the vast majority ofpatients the clinicians reported thatease of control of sedation was“good” or “adequate.”

Target-controlled Infusions for ICU andProcedural Sedation

Anthony Absalom, MBChB, MD, FRCA ILTM, Anaesthetic Department, Addenbrookes Hospital, Cambridge, UK

Key Points

• Clinicians new to target-controlled infusion (TCI) technology find it easy to use and soon develop a preference to it compared to manual control of infusion rates.

• When manual infusion regimens are used, there is a lengthy delay before a change in infusion rate results in a significant change in blood concentra-tion and an even longer delay in effect-site concentration.

• With TCI systems, a proportional change in blood or effect-site concentra-tion can be rapidly and accurately implemented.

• TCI systems enable both patient controlled analgesia and sedation to be accomplished safety.



Target-controlled Infusions for ICU and Procedural Sedation

Steady-state BloodConcentrations, Assessmentof Treatment Efficacy

By definition, TCI systems providesteady-state blood (or effect-site)concentrations. This makes possiblerapid and rational assessment of effi-cacy of the administered dose (targetconcentration), particularly wheneffect-site targeting is used. By con-trast, assessment of treatment effica-cy is more difficult when manuallycontrolled infusions are used. Whencurrent agents are administered at afixed infusion rate, there is a long(and variable) delay before anythingapproaching a steady-state situationis reached (Figure 1). For example, iffentanyl is infused at a fixed rate, theblood concentration will rise steeplyfor 24 hours and only reach steadystate at 48 hours. A single, fixed infu-sion rate may provide inadequatesedation for a time, adequate seda-tion for a subsequent period, andfinally excessive sedation.

How and when should a clinicianassess the adequacy of a dose orguess an appropriate dose or infu-sion rate for an individual patient? Toproperly assess the efficacy of achange in infusion rate, the clinicianshould wait until a new steady stateis reached, but for many sedative,hypnotic and analgesic agents this isnot practical, because of the longdelays before new steady states arereached. If changes are made beforesteady-state concentrations arereached, then over- and under-shooting of the blood concentrationsare likely.

Speed and Accuracy of Titration

In the ICU and during proceduralsedation, rapid changes (increases ordecreases) in blood and effect-siteconcentrations of sedative and anal-gesics are often required. For exam-ple, increases might be required fortracheal toilette, removal of surgicaldrains or physiotherapy, while

decreases might be necessary aftersuch procedures, or in a patient whohas become apnoeic or developedcardio-vascular instability as a resultof excessive administration.

When manual infusion regimensare used with current agents, there isa lengthy delay before a change ininfusion rate results in a significantchange in blood concentration(Figure 2) and an even longer delaybefore this results in a significantchange in effect-site concentration.Moreover, during the first few hoursof infusion of agents such as propofolor fentanyl (when blood concentra-tions are rising steeply), the bloodconcentration may continue toincrease despite small decreases inthe infusion rate.

When rapid increases are required,most clinicians using manually con-trolled infusions administer a bolusdose. The size of a bolus dose is diffi-cult to judge - if it is too small, theeffects will be inadequate; if it is toolarge, then the resulting largeincrease in blood and effect-site con-centrations may cause seriousadverse effects. At commonly usedconcentrations of current sedativesand analgesics, fairly small bolusdoses generally are required.

TCI systems can overcome allthese difficulties. They can producealmost step-wise increases in con-centrations, by automatically admin-istering a bolus (a rapid infusion, withthe rate and duration accurately cal-culated), followed by step-wisedecreases in infusion rate. When adecrease in target concentration isrequired, TCI systems produce the

0

0.5

1

1.5

2

2.5

3

3.5

0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48

Time (hours)

Prop

ofol, fen

tany

l con

centratio

n

0

50

100

150

200

250

300

350

400

450

Midazolam

con

centratio

n

Propofol (μg/ml)

Fentanyl (ng/ml)

Midazolam (μg/ml)

Figure 1. Simulation showing predicted concentrations resulting from 48 hours infusions of propofol (200 mg/hr), fentanyl (100 μg/hr) and midazolam (10 mg/hr)

Figure 1. Predicted Concentrations: 48-hr Infusions



most rapid possible fall by switchingoff the infusion until the new targetconcentration is achieved. Althoughit is feasible for a clinician to tem-porarily switch off a manually con-trolled infusion, this is generallyunwise, because if he or she forgetsto restart the infusion, then furtherproblems may arise from inadequatedosing.

With TCI systems, proportionalchanges to the blood or effect-siteconcentration can be rapidly andaccurately implemented. Althoughmeasurement of the blood concen-tration in an individual may reveal alarge difference from the target con-centration, this does not matter. If thetarget concentration is changed, theachieved (measured) concentrationwill change by the same proportion,facilitating rational assessment of theimpact of the change.

Cardiovascular Stability

When compared with manuallycontrolled infusions, TCI systemshave been shown to be associatedwith equivalent or improved cardio-vascular stability in patients under-going general anesthesia.1,3-7 Thechief reason is that when TCI systemsare used for induction, the infusionrates and total doses are lower thanthose used by anesthesiologists (whotend to be impatient and administerboluses at rates far greater than themaximum infusion rates of the TCIsystems). At present there is little evi-dence that TCI systems are associatedwith improved cardiovascular stabili-ty during procedural or ICU sedation.

Patient-controlled TCISystems for Analgesia and Sedation

Studies of patient-controlled-anal-gesia systems have shown that

patients prefer to have control overthe administration of their anal-gesics. In recent years several groupshave developed patient-controlled-sedation systems. With currentlyavailable agents, sedation systemsthat administer a bolus are unlikelyto be successful, because they resultin large fluctuations in blood concen-trations with the potential foradverse outcomes resulting fromtroughs of excessively low concentra-tions and peaks of excessively highconcentrations.

Patient-maintained-analgesia and-sedation systems combine the ben-efits of patient control with TCI tech-nology. With these systems, a patientdemand (using a handset or othermeans) made outside of a lockoutperiod will result in a pre-definedincrease in the target concentration.If no further demands are made, thesystem maintains the blood concen-tration at the current target concen-tration for a pre-set period of time,and then slowly reduces the targetconcentration to maintain safety.

Studies of the use of these sys-tems for analgesia and sedationshow that they are safe and effective,and that patients like using them.8-18

A recently completed study com-pared patient-maintained sedation(PMS) with anaesthesiologist-admin-istered sedation, and showed thatPMS was associated with lesshypotension and oversedation.19

Accuracy of TCI Systems

Accuracy of TCI systems some-times concerns clinicians unfamiliarwith them. The chief benefit of TCIsystems is that they administersteady-state blood concentrations

0

0.5

1

1.5

2

2.5

3 4 5 6 7 8 9

Time (hours)

Con

centratio

n (m

cg/m

l)

Propofol

Fentanyl

Figure 2. Simulation showing the predicted effect of 50% increases in the infusionrates of fentanyl (infused at 100 μg/hr for 6 hours, and at 150 μg/hr thereafter, pre-dicted concentrations based on Shafer model) and propofol (infused at 200 mg/hr for6 hours and at 300 mg/hr thereafter, predicted concentrations based on Marsh model).Note that it takes many hours for the blood concentrations to increase by 50%.

Figure 2: Predicted Effect: 50% Increases in Infusion Rates



rather than exact concentrations.Even if 100% predictive accuracywere achieved, titration to effectwould remain necessary to accom-modate the large inter-individualvariability in pharmacodynamic sen-sitivity among different patients tomost sedative and analgesic agents.

A TCI system administers a druginfusion at rates determined by amodel derived from studies of thepharmacokinetics of a drug in a pop-ulation of patients or subjects. Thus,the concentration shown on the userinterface is only an estimate. Varveland colleagues have proposed a setof standard criteria for assessing thepredictive performance of comput-erised infusion pumps.20 These crite-ria are median performance error(MDPE), a measure of bias or offset;median absolute performance error(MDAPE), a measure of inaccuracy orimprecision; wobble, a measure ofthe intra-individual variability inerrors; and divergence, a measure ofany trend over time in the size andmagnitude of errors. For computer-controlled infusion pumps a MDPE(bias) of 10% to 20 % and a MDAPE of20% to 40 % have been proposed asacceptable,21-22 and most of the com-monly used models perform wellwithin these limits.

Many factors can influence theactual drug concentrations achieved.The models in use were usuallyderived from studies in healthypatients. Critically ill patients suffermany physiological derangementsthat have the potential to alter drugdistribution and metabolism com-pared to healthy patients. For exam-

ple, critically ill patients suffer alter-ations in regional blood flow includ-ing reduced hepatic blood flow, caus-ing impairment of hepatic function.Simulations show that large changesin the compartment volumes pro-duce relatively insignificant changesin model performance, generallyresulting in a short-lived overshootfollowing a target increase. Largedecreases in metabolic and distribu-tion clearance rates, however, cancause significant, sustained errors.

There are few data about the per-formance of current pharmacokineticmodels used in the critically ill.McMurray found that for propofol,clearance was almost normal and thevalues for bias and inaccuracy werevery similar to those found in healthyadults.2 Cavaliere has shown thathypoalbuminemia has little effect onpropofol pharmacokinetics,23 andServin showed that the pharmacoki-netics of propofol (given by infusion)were not markedly altered in patientswith cirrhosis.24 Thus, for propofolthere is little evidence at present tosuggest that model performance inthe critically ill is significantly worsethan in healthy patients. Furtherstudies are needed to assess theaccuracy of the existing pharmacoki-netic models for other agents com-monly administered by TCI such asremifentanil and alfentanil.

Conclusions

TCI systems administer steady-state blood drug concentrations byconstantly adjusting the infusion rate(to take account of the reduced re-distribution loss from the centralcompartment as time goes by). They

enable rapid and precise titration ofdrugs used for sedation, anesthesiaand analgesia, and rational assess-ment of the impact of any changes indose. In healthy patients, the per-formance accuracy for commonlyused drugs and models is acceptable,and current data indicate that theperformance accuracy for pharmaco-kinetic models for propofol is likely tobe similarly acceptable in critically illpatients. TCI systems do not providean “on-off” switch for sedation andanalgesia, but do provide a morerational method of administeringthese drugs.

References1. Russell D, Wilkes MP, Hunter SC, et al. Manual

compared with target-controlled infusion of propofol. Br J Anaesth 1995;75:562-6.

2. McMurray TJ, Johnston JR, Milligan KR, et al. Propofol sedation using 'Diprifusor' target-controlled infusion in adult intensive care unit patients. Anaesth 2004;59:636-41.

3. Struys M, Versichelen L, Rolly G. Influence of pre-anaesthetic medication on target propofol concentration using a 'Diprifusor' TCI system during ambulatory surgery. Anaesth 1998;53(Suppl1):68-71.

4. Barvais L, Rausin I, Glen JB, et al. Administration of propofol by target-controlled infusion in patients undergoing coronary artery surgery. J Cardiothorac Vasc Anesth 1996;10:877-83.

5. Swinhoe CF, Peacock JE, Reilly CS. Evaluation of the accuracy of the 'Diprifusor'. Eur J Anaesth1995;10(Suppl):84.

6. Servin FS, Marchand-Maillet F, Desmonts JM. Influence of analgesic supplementation on the target propofol concentrations for anaesthesia with 'Diprifusor' TCI. Anaesth 1998;53(suppl1):72-6.

7. Richards AL, Orton JK, Gregory MJ: Influence of ventilatory mode on target concentrations required for anaesthesia using a 'Diprifusor' TCI system. Anaesth 1998;53(suppl1):77-81.

8. Irwin MG, Campbell RC, Lun TS, et al. Patient maintained alfentanil target-controlled infusion for analgesia during extracorporeal shock wave lithotripsy. Can J Anaesth 1996;43:919-24.

9. Irwin MG, Thompson N, Kenny GN. Patient-maintained propofol sedation: assessment of a target-controlled infusion system. Anaesth1997;52:525-30.

10.Murdoch JA, Kenny GN. Patient-maintained propofol sedation as premedication in day-case surgery: assessment of a target-controlled system. Br J Anaesth 1999;82:429-31.

11.Murdoch JA, Grant SA, Kenny GN. Safety of patient-maintained propofol sedation using a target- controlled system in healthy volunteers. Br J Anaesth 2000;85:299-301.



12.Henderson F, Absalom AR, Kenny GN. Patient-maintained propofol sedation: a follow up safety study using a modified system in volunteers. Anaesth 2002;57:387-90.

13.Checketts MR, Gilhooly CJ, Kenny GN. Patient-maintained analgesia with target-controlled alfentanil infusion after cardiac surgery: a comparison with morphine PCA. Br J Anaesth1998;80:748-51.

14.Schraag S, Kenny GN, Mohl U, et al. Patient-maintained remifentanil target-controlledinfusion for the transition to early postoperative analgesia. Br J Anaesth 1998;81:365-8.

15.Campbell L, Imrie G, Doherty P, et al. Patient maintained sedation for colonoscopy using a target controlled infusion of propofol. Anaesth2004;59:127-32.

16.Rodrigo M, Irwin M, Tong C, et al. A randomised crossover comparison of patient-controlled sedation and patient-maintained sedation using propofol. Anaesth 2003;58:333-8.

17.Gillham M, Hutchinson R, Carter R, et al. Patient-maintained sedation for ERCP with a target-controlled infusion of propofol: a pilot study. Gastrointest Endosc 2001;54:14-7.

18.Leitch J, Anderson K, Gambhir S, et al. A partially blinded randomised controlled trial of patient-maintained propofol sedation and operator controlled midazolam sedation in third molar extractions. Anaesth 2004;59:853-60.

19.Stonell CA, Leslie K, Absalom AR. Randomised trial of effect-site targeted patient-controlled sedation or anaesthetist-administered sedation for colonoscopy. Anaesth In press.

20.Varvel JR, Donoho DL, Shafer SL. Measuring the predictive performance of computer-controlled infusion pumps. J Pharmacokinet Biopharm 1992;20:63-94.

21.Schüttler J, Kloos S, Schwilden H, et al. Totalintravenous anaesthesia with propofol and alfentanil by computer-assisted infusion. Anaesth1988;43(suppl):2-7.

22.Glass PJA, Jacobs JR, Reves JG. Intravenous drug delivery. In: Miller RD, ed. Anesthesia New York: Churchill Livingstone; 1990:367.

23.Cavaliere F, Conti G, Moscato U. Hypoalbumin-aemia does not impair Diprifusor performanceduring sedation with propofol. Br J Anaesth2005;94:453-8.

24.Servin F, Cockshott ID, Farinotti R, et al. Pharmacokinetics of propofol infusions inpatients with cirrhosis. Br J Anaesth 1990;65(2):177-83.



PROCEEDINGS

Among many complications thatare of concern in the postoperativeperiod, oxygenation assumes a vitallyimportant place in the managementof any postoperative patient. Thisbecomes all the more importantwhen pain relief medications areused to control postoperative pain.Many pain relief medications result inrespiratory depression and conse-quent hypoxia (Figure 1). Hypoxia, ifunrecognized, leads to unexpectedpostoperative disaster with medico-legal implications. Therefore, post-operative monitoring becomesmandatory for patients who arereceiving pain medications that candepress cardio-respiratory systems.

Clinical Determination ofOxygenation

A pulse oximeter is an excellentand reliable device that monitors

oxygenation and detects hypoxia.However, it does not give enoughforewarning to implement correctivemeasures to prevent hypoxic conse-quences. This is easily understoodbased on the relationship betweenarterial oxygen tension (PaO2) andoxygen saturation (SpO2) (Figure 2).

The SpO2 does not begin todecrease from 98-100% until PaO2decreases below 100 mmHg.Decreases in PaO2 below 80 mmHgare associated with dramaticdecreases in SpO2. This is because of‘S’-shaped relationship betweenSpO2 and PaO2. The decreases inSpO2 can be exaggerated with clini-cal conditions associated with a

Pain Medications

Oxygenation

Blood pressure / HR

Cardiovascular Shock Thrombophlebitis

Respiratory Atelectasis Pneumonia Pulmonary embolism Aspiration

Wound Infection Dehiscence Evisceration Delayed healing Hemorrhage Hematoma

Urinary Acute urine retention Urinary tract infection

Functional Weakness Fatigue Functional decline

Neurologic Delirium Stroke

Figure 1. Potential Post-operative Complications

Key Points

• Oxygenation is vitally important in the management of post-operative patients and those being sedated for procedures.

• Pulse oximetry does not give enough forewarning of hypoxia to implement timely corrective action.

• Capnography is a reliable monitor of ventilation that can be used to monitor pain management in post-operative patients and sedation for procedures.

• The use of pulse oximetry and capnograpy should allow patients to be treated with opiate analgesics and sedatives with less risk of undetected catastrophic respiratory events.

Physiology of Oxygen and Carbon DioxideMonitoring

Bhavani Shankar Kodali; MD, Associate Professor, Harvard Medical School,Brigham and Women’s Hospital, Boston, MA, www.capnography.com


Physiology of Oxygen andCarbon Dioxide Monitoring

reduction in functional residualcapacity (FRC) that acts as the oxy-gen reserve buffer in the lungs.Obesity is a classic example, and FRCis known to decrease followingabdominal surgeries even in normal-weight individuals.

The time interval from impendinghypoxia to actual hypoxia isdecreased with decreased FRC, andthis decreased time may not be suffi-cient enough to call and initiate cor-rective measures to prevent hypoxiafrom occurring. Apart from stablehomodynamics, ventilation is a majordeterminant of oxygenation. Hence,it is necessary to monitor ventilationto detect episodes of inadequate ordecreased ventilation that may resultin hypoxia, if uncorrected. This way,one could avoid potentially life-threatening episodes in the postop-erative period, when the monitoringof vital signs is not as frequent as inthe operating room or in the postop-erative anesthesia care unit (PACU).

Capnography has been shown tobe a reliable monitor of ventilation in

the operating room. Its use is beingextended into many areas outside ofoperating rooms to monitor ventila-tion in patients undergoing a varietyof invasive and non-invasive surgicaland medical procedures. These areasinclude intensive care units, emer-gency rooms, interventional radiolo-gy suites, gastroenterology suites,interventional cardiology units, andpre-hospital settings such as air andemergency ambulance services. It isonly a question of time beforecapnography in association withpulse oximetry will be used to moni-tor postoperative patients receivingpostoperative pain medications viaintravenous route.

Overview of Physiology ofCapnography

Carbon dioxide (CO2) is producedin the tissues and diffuses into thevenous blood, which reaches theright side of the heart and reachesthe lungs via pulmonary circulation.Here oxygen (O2) enters the bloodand carbon dioxide (CO2) is eliminat-ed during expiration. CO2 is typically

measured at the mouth, nose, or atthe junction between the endotra-cheal tube and ventilator circuit.

At the end of inspiration, assum-ing that there is no rebreathing, theairway and the lungs are filled withCO2-free gases. Carbon dioxide dif-fuses into the alveoli and equilibrateswith the end-alveolar capillary blood(PaCO2 = PcCO2 = 40 mm Hg). Theactual concentration of CO2 in thealveoli is determined by the extent ofventilation and perfusion into thealveoli (V/Q ratio). The alveoli withhigher ventilation in relation to per-fusion (high V/Q alveoli) have lowerCO2 compared to alveoli with lowV/Q ratio that would have higherCO2. As one moves proximally in therespiratory tract, the concentrationof CO2 decreases gradually to zero.The volume of CO2-free gas is termed“respiratory dead space” and herethere is no exchange of O2 and CO2between the inspired gases and theblood. As the patient exhales, a CO2sensor at the mouth will detect noCO2 as the initial gas sampled will bethe CO2-free gas from the deadspace. As exhalation continues, CO2concentration rises gradually andreaches a peak as the CO2-rich gasesfrom the alveoli make their way tothe CO2 sensing point at the mouth.At the end of exhalation, the CO2concentration decreases to zero(base line) as the patient commencesinhalation of CO2-free gases. Theevolution of CO2 from the alveoli tothe mouth during exhalation, andinhalation of CO2-free gases duringinspiration gives the characteristicshape to the CO2 curve, which isidentical in all humans with healthy

Figure 2. Relationship Between PaO2 and SpO2

Pulse Oximetry



lungs. Any deviation from this identi-cal shape should be investigated todetermine a physiological or a patho-logical cause producing the abnor-mality.

Time Capnograms

A time capnogram can be dividedinto inspiratory (phase 0) and expira-tory segments (Figure 3). The expira-tory segment, similar to a singlebreath nitrogen curve or singlebreath CO2 curve, is divided intophases I, II and III, and occasionally,phase IV, which represents the termi-nal rise in CO2 concentration. Theangle between phase II and phase IIIis the alpha angle. The nearly 90-degree angle between phase III andthe descending limb is the betaangle. For more details on this sec-tion: please refer to www.capnogra-phy.com terminology section.

The CO2 concentration reaches amaximum at the end of exhalation.This maximum concentration iscalled end-tidal carbon dioxide con-centration or tension depending onwhether it is expressed in fractionalconcentration or mm Hg. End-tidalcarbon dioxide reflects CO2 concen-tration of alveoli emptying last. Thenormal values of EtCO2 are around5% or 35-37 mm Hg. The gradientbetween the blood CO2 (PaCO2) andexhaled CO2 (end tidal CO2 orPetCO2) is usually 5-6 mm Hg.PetCO2 can be used to estimatePaCO2 in patients with essentiallynormal lungs.

The three major determinants ofCO2 waveform are cardiac output,ventilation, and CO2 production.Hypermetabolic states such as thyro-

toxicosis or malignant hyperpyrexiaare associated with increased CO2production and therefore with high-er end-tidal CO2.

Capnography SamplingDevices

Capnography not only detectsabnormalities in ventilation perfu-sion, such as those resulting fromCOPD or bronchial asthma, but alsomonitors PaCO2 indirectly. If thesampling of end-tidal gas is ade-

quate, hypercarbia can be detectedfrom capnography monitoring. Inaddition, capnography displaysabnormal respiratory patterns thatmay suggest inadequate ordepressed ventilation. Adequatesampling can be ensured by severalmethods and sampling devices(Figure 4).

The capnograms from good sam-pling devices appear similar to con-ventional capnograms obtained viaendotrachal intubation. However,

Expiration Inspiration

0

betaalpha

III

II

I

Figure 3. Time Capnogram

Figure 4. Capnograph Sampling Device



dilution of expired gases is unavoid-able in certain circumstances. Underthese conditions, a change frombaseline should forewarn of respira-tory depression or at least shouldencourage a clinician to investigatethe change in capnograms.

Summary

An understanding the physiologyof oxygenation and ventilationshould make it apparent that it isquite logical to monitor not only oxy-genation but also ventilation in post-operative patients receiving nar-cotics for pain relief. Capnographyhas been used successfully in gas-troenterology sedation proceduresto detect inadequate respiratoryevents well before the changes inoxygenation. Using combined tech-nology should allow patients torecover safely in the potentially tur-bulent postoperative period, withoutbeing subjected to the risk of unde-tected catastrophic respiratoryevents.

BibliographyAskrog V. Changes in (a-A)CO2 difference and pul-monary artery pressure in anesthetized man. J Appl Physiol 1966;21:1299-305.

Callaham M, Barton C. Prediction of outcome of CPRfrom end-tidal carbon dioxide concentration. Crit Care Med 1990;18:358-62.

Callaham M, Barton C, Mathay M. Effect of epineph-rine on the ability of end-tidal carbon dioxide read-ings to predict initial resuscitation from cardiacarrest. Crit Care Med 1992;20:337-43.

Chase PB, Kern KB, Sanders AB, et al. Effects of grad-ed doses of epinephrine on both non-invasive andinvasive measures of myocardial perfusion andblood flow during cardio-pulmonary resuscitation.Crit care Med 1993;21:413-9.

Gedeon A, Krill P, Kristensen J, et al. Noninvasive car-diac output determined with a new method basedon gas exchange measurements and carbon dioxiderebreathing: A study in animals/pigs. J Clin Monit 1992;8:267-78.

Isserles SA, Breen PH. Can changes in end-tidalPCO2 measure changes in cardiac output? AnesthAnalg 1991;73:808-14.

Jin X, Weil MH, Povoas H, et al. End-tidal carbondioxide as a noninvasive indicator of cardiac indexduring circulatory shock. Crit Care Med 2000;28:2415-9.

Kalenda Z. Capnogram as a guide to the efficacy ofcardiac massage. Resuscitation 1978;6:259-63

Leigh MD, Jones JC, Motley HL. The expired carbondioxide as a continuous guide of the pulmonary andcirculatory systems during anesthesia and surgery. JThoracic Cardiovasc Surg 1961;41:597-610.

Lewis LM, Stothert J, Standeven J, et al : Correlationof end-tidal carbon dioxide to cerebral perfusionduring CPR. Ann Emerg Med 1992;21:1131-4.

Maslow A, Stearns G, Bert A, et al. Monitoring end-tidal carbon dioxide during weaning from cardiopul-monary bypass in patients without significant lungdisease. Anesth Analg 2001;92:306-13.

Ornato JP, Garnett AR, Glauser FL. Relationshipbetween cardiac output and the end-tidal carbondioxide tension. Ann Emerg Med 1990;19:1104-6.

Pianosi P, Hochman J. End-tidal estimates of arterialPCO2 for cardiac output measurements by CO2rebreathing: a study in patients with cystic fibrosisand healthy controls. Pediatr Pulmonol 1996;22:154-60.

Sanders AB, Atlas M, Ewy GA, et al. Expired PCO2 asan index of coronary perfusion pressure. Am J Emerg Med 1985;3:147-9.

Shibutani K, Muraoka M, Shirasaki S, et al. Dochanges in end-tidal PCO2 quantitatively reflectchanges in cardiac output? Anesth Analg1994;79:829-33.

Ward KR, Yealy DM. End-tidal carbon dioxide moni-toring in emergency medicine, part 2: Clinical appli-cations. Acad Emerg Med 1998;5:637-46.

Weil MH, Bisera J, Trevino RP, Rackow EC. Cardiacoutput and end-tidal carbon dioxide. Crit Care Med1985;13:907-9.

White RD, Asplin BR. Out of hospital quantitativemonitoring of end-tidal carbon dioxide pressureduring CPR. Ann Emerg Med 1994;23;25-30.

ATTENDEES

A. R. Absalom, MBChB, MD, FRCA, ILTMAnaesthetic DepartmentAddenbrookes HospitalCambridge, United Kingdom CB2 2QQ

Pam Almandinger, RNClinical Educator, Project ManagerAdventist Medical CenterPortland, OR

Carol Bess, RN, BSN, CRNIInfusion Nurse SpecialistBellin HospitalGreen Bay, WI

Ian H. Black, MD, MAJ, MCAssistant Professor of Anesthesiology, Uniformed Services University of the Health Sciences Director, The TARGIT Center Chief of Anesthesia, US Army Institute of Surgical ResearchFort Sam Houston, TX

Bradley A. Boucher, PharmD, FCCP, FCCMProfessor of PharmacyUniversity of Tennessee Health Science CenterMemphis, TN

Michael R. Cohen, RPh, MS, ScD Institute for Safe Medication Practices Huntingdon Valley, PA

Joseph F. Dasta, MScProfessor of PharmacyThe Ohio State UniversityCollege of PharmacyColumbus, OH

Susan Dempsey, RN, MN, CNSClinical Nurse SpecialistPalomar Pomerado Health SystemSan Diego, CA

John Engelbert, PharmDPharmacy Clinical SupervisorSharp Memorial Hospital San Diego, CA

Lynn Eschenbacher, PharmDMedication Safety OfficerDepartment of PharmacyDuke University HospitalDurham, NC

Peter Glass, MDProfessor and Chairman Department of Anesthesia SUNY Stony Brook Health Sciences Center Stony Brook, New York

Jeffrey B. Gross, MDProfessor of Anesthesiology and PharmacologyUniversity of Connecticut School of MedicineFarmington, CT

Pamela C. HaganChief Programs Officer American Nurses Association Silver Spring, MD

Regina Izu, RN, PHN, MSNM/S Advanced Practice NurseChair, Nurse Practice CouncilScripps Memorial HospitalSan Diego, CA

Judith Jacobi, PharmD, FCCM, FCCP, BCPS Critical Care Pharmacist Methodist Hospital/Clarian Health Indianapolis, IN

Bhavani Shankar Kodali, MD Associate Professor, Anesthesiology Harvard Medical SchoolBrigham and Women’s Hospital Boston, MA

Center for Safety and Clinical Excellence Sixth Conference

Sedation Therapy: Improving Safety and Quality of CareNovember 17-18, 2005

John P. Kress, MDAssistant Professor of MedicineUniversity of ChicagoChicago, IL

Denise Li, RN, MS, CNS, PhD(c)Adjunct Assistant Clinical ProfessorPhysiological NursingUniversity of California, San Francisco San Francisco, CA

Mary E. Lough, RN, MS CNS, CCRN, CNRNCritical Care Clinical Nurse Stanford University Hospital and ClinicsPalo Alto, CA

Ray R. Maddox, PharmDDirector, Clinical Pharmacy,Research & Pulmonary MedicineSt. Joseph’s/Candler Health SystemSavannah, GA

Thomas McCarter, Jr., MDSr. Vice President and Chief Medical OfficerMain Line HealthPhiladelphia, PA

Gladstone C. McDowell II, MDOhio HealthColumbus, OH

Richard E. Moon, MD, FRCPC, FACP, FCCP Professor of Anesthesiology, Associate Professor of Medicine Duke University Medical Center Durham, NC

H. Julius A. Oglesby, RRT/RCPManager, Pulmonary MedicineCandler HospitalSt. Joseph’s/Candler Health SystemSavannah, GA

Frank J. Overdyk, MSEE, MDProfessor of AnesthesiologyMedical University of South CarolinaDepartment of AnesthesiologyCharleston, SC

Julie Painter, RN, MSN, OCN Clinical Nurse Specialist/Adult Nurse Practitioner Community Health Network Indianapolis, IN

Joel Parnes, PharmD, MHAShands JacksonvilleDepartment of PharmacyManager, Central OperationsAssistant ProfessorUniversity of Florida College of PharmacyGainesville, FL

Chris Pasero, RN, MS, FAANPain Management Educator and Clinical ConsultantEl Dorado Hills, CA

James Paul, MDDirector, Acute Pain ServiceDepartment of AnesthesiaHamilton Health SciencesHamilton, ONT

Anne Pohlman, RN MSN CCRNCritical Care Clinical ResearchUniversity of Chicago Chicago, IL

Brenda Truman Pun, RN, MSN, ACNP Coordinating Center Manager Vanderbilt University Medical Center Nashville, TNClinical Faculty, University of North Carolina - School of Nursing Chapel Hill, NC

Kathleen Puntillo, RN, DNSc, FAANProfessor of Nursing and Faculty,Critical Care/Trauma ProgramDepartment of Physiological NursingUniversity of California, San FranciscoSan Francisco, CA

Michael Ramsay, MD (via telephone)Chair, Department of AnesthesiaBaylor Health Care SystemDallas, TX

Philip J. Schneider, MS, FASHPClinical Professor and DirectorLatiolais Leadership ProgramDivision of Pharmacy Practice and AdministrationCollege of PharmacyThe Ohio State UniversityColumbus, OH

Curtis Sessler, MDOrhan Muren Professor of MedicineDivision of Pulmonary and Critical Care MedicineVirginia Commonwealth University Health SystemMedical Director of Critical CareMedical College of Virginia HospitalsRichmond, VA

Dr. Geoff Shaw, MBChB, FANZCA, FJFICMClin Sen Lecturer, Department of Medicine,Christchurch School of Medicine and Health Sciences University of Otago, New ZealandSenior Fellow, Department of Mech EngineeringUniversity of Canterbury, New ZealandIntensive Care Specialist, Department of Intensive CareMedicineChristchurch Hospital, Christchurch, New Zealand

Maria Shepherd Vice President of MarketingOridion CapnographyNeedham, MA

Kathleen StacyPalomar Pomerado Health SystemSan Diego, CA

Robert K. Stoelting, MDPresident, Anesthesia Patient Safety FoundationIndianapolis, IN

Michel MRF Struys, MD, PhDProfessor in Anesthesia and Research CoordinatorStaff AnesthesiologistProfessor in Clinical PharmacologyGhent University HospitalDepartment of AnesthesiaGent, BELGIUM

Tim Vanderveen, MS, PharmDVice PresidentCenter for Safety and ClinicalExcellenceSan Diego, CA

John VanEeckhout, PharmD Vice President, Clinical Services Child Health Corporation of America Shawnee Mission, KS

Vickey Weir, BSN, MPAClinical Quality ManagerStanford University Medical CenterPalo Alto, CA

Margie Whittaker, RN, MN, MS, CCRNNurse ManagerMission Hospital Regional Medical CenterMission Viejo, CA

Carolyn K. Williams, BSPharm Medication Safety SpecialistSt. Joseph's/Candler Health System, Inc.Savannah, GA

SC2602 0311/PDF

Sedation Therapy - CareFusion · pain management and sedation in intensive care patients. This...

Documents

Transcript of Sedation Therapy - CareFusion · pain management and sedation in intensive care patients. This...