e

astpnotsdpa4r4ntstautc

e

p

gU

RA

The Journal of Emergency Medicine, Vol. 27, No. 3, pp. 225–231, 2004Copyright © 2004 Elsevier Inc.

Printed in the USA. All rights reserved0736-4679/04 $–see front matter

doi:10.1016/j.jemermed.2004.04.009

OriginalContributions

IMPACT OF CT SCAN IN PATIENTS WITH FIRST EPISODE OF SUSPECTEDNEPHROLITHIASIS

Michael Ha, MD* and Russell D. MacDonald, MD, MPH, CCFP, FRCPC†‡

*Division of Emergency Medicine, Department of Family Medicine, University of Manitoba, Winnipeg, Manitoba, Canada, †Division ofEmergency Medicine, Department of Medicine, University of Toronto, Toronto, Ontario, Canada, and ‡Ontario Air Ambulance Base

Hospital Program, Division of Prehospital Care, Sunnybrook and Women’s College Health Sciences Centre, Toronto, Ontario, CanadaReprint Address: Dr. Russell D. MacDonald, MD, MPH, CCFP, FRCPC, Medical Director, Central Region, Ontario Air Ambulance Base

Hospital Program, 1120 Finch Avenue West, Suite 405, Toronto, Ontario M3J 3H7, Canada

yT)li-

-f

s-a-nte,,y,inig-d

f

der-n

H osiso withs diesi rami andth (10–4l etro-s enalc ddi-t a-t ofC naryd yeare pec-t withc

xam-i tingt tedn r thep canw ctedr

mer-gton,

Abstract—This prospective observational outcome studssessed the impact of helical computed tomography (Ccan in patients with a first episode of suspected nephrohiasis. Before CT scanning, Emergency Physicians comleted a questionnaire, including diagnostic certainty oephrolithiasis and anticipated patient disposition. Primaryutcome measure was the comparison of physician diagno

ic certainty and CT scan results. Secondary outcome meures included alternate diagnoses and changes in patieisposition after CT scan. Four categories grouped thre-CT diagnostic certainty: 0–49%, 50–74%, 75–90%nd 90–100%. The CT scan found urinary calculi in 28.6%5.7%, 74.2%, and 80.5% of patients in each categorespectively. CT scanning revealed alternate diagnoses0 cases (33.1%). Of these, 19 (47.5%) included other sificant pathology. Before CT scanning, physicians planne

o discharge 115 patients and admit six patients. After CTcanning, six of the former group were admitted, and five ohe latter group were discharged. Patients presenting with

first episode of clinically suspected nephrolithiasis shoulndergo CT scanning because it enhances diagnostic c

ainty by identifying alternate diagnoses not suspected olinical grounds alone. © 2004 Elsevier Inc.

Keywords—kidney calculi; helical computed tomogra-hy; diagnosis; CT scanning; nephrolithiasis

This study was presented at the American College of Eency Physicians Scientific Assembly, Seattle, WashinSA, October 2002.

ECEIVED: 21 March 2003; FINAL SUBMISSION RECEIVED: 3

CCEPTED: 8 April 2004

225

INTRODUCTION

istorically, the Emergency Department (ED) diagnf nephrolithiasis was based on clinical presentationelective use of intravenous pyelography (IVP). Stundicate that helical computed tomography (CT) urogs superior to IVP in the diagnosis of nephrolithiasishe investigation of urinary calculi (1–5). In addition,elical CT scan demonstrates a large percentage5%) of nonurinary calculi diagnosis (1–4,6–8). The

argest series of CT scan for suspected renal colic rpectively reviewed 1000 consecutive CT scans for rolic and found 10% of cases with alternative or aional diagnosis (1). A review of 100 consecutive pients in a follow-up study 1 year after the introductionT scanning for suspected renal colic found extrauriiagnoses in 45% of patients compared to 16% 1arlier (2). However, most of these studies are retros

ive, examining all patients who presented to the EDlinically suspected nephrolithiasis.

To date there has been no prospective study to ene the impact of helical CT scan on patients preseno the ED with a first episode of clinically suspecephrolithiasis. Previous studies are unclear whetheatients with alternate diagnosis identified by CT sere patients with their first presentation of suspe

enal colic or those with recurrent attacks.

h 2004;

Marc

tcasoca

TsacbpspscsgonPf

ttnEosrbc

iiicRprcwPianfa

dddb

sbopsnp

isshrss

tsan

dmsSsrdv

TrOT1

T

PHTOAC

226 M. Ha and R. D. MacDonald

We believe the Emergency Physician’s (EP) diagnos-ic certainty for nephrolithiasis in subjects with their firstlinical presentation of clinically suspected nephrolithi-sis is enhanced using helical CT scan. We studied allubjects presenting to the ED with their first presentationf unilateral flank pain where nephrolithiasis was clini-ally probable to determine the certainty of clinical di-gnosis and assess the impact of helical CT scan.

MATERIALS AND METHODS

his prospective, observational study enrolled all eligibleubjects who presented to the ED of a university-affili-ted, tertiary care teaching hospital with a first episode oflinically suspected nephrolithiasis. Subjects were eligi-le if, in the opinion of the EP or physician designate, theatient’s clinical diagnosis was nephrolithiasis. Exclu-ion criteria included subjects under 18 years of age;regnancy; cognitive impairment; presence of bilateralymptoms; traumatic injury; past diagnosis of urinaryalculi (history of nephrolithiasis, noninvasive or inva-ive intervention related to nephrolithiasis); history ofenitourinary anatomic abnormality; transfer from an-ther institution to a consultant service for a genitouri-ary problem, or contraindication to helical CT scan.atients were enrolled in a consecutive 10-month periodrom April 2001 to February 2002.

The EP, triage nurse, or designate identified all po-entially eligible subjects. The physician determined ifhe subject met the study inclusion criteria and ensuredo exclusion criteria were present. The attending EP ormergency Medicine resident, under direct supervisionf the attending physician, completed the data collectionheets. Investigators routinely surveyed the patient careecords, patient registration logbooks, ED discharge log-ooks, and diagnostic imaging logbooks to ensure noases were missed.

The physician recorded the subject’s demographicnformation, physician’s clinical findings, and laboratorynvestigations on the data collection sheet. Before imag-ng studies, the attending physician also recorded theertainty of diagnosis and anticipated patient disposition.esidents could complete the certainty of diagnosis andatient disposition sections on the data form only aftereviewing the cases with the attending physician. Physi-ian pre-CT scan diagnostic certainty of nephrolithiasisas determined using a 10-cm linear visual analog scale.atients then underwent unenhanced helical CT scan to

mage the genitourinary system. The radiologist recordedn interpretation of the CT scan on the data sheet, andotified the EP of the CT scan result. The radiologistorwarded the data collection sheet to the investigators,

nd did not return it to the EP. The EP had complete e

iscretion regarding patient management and dispositionecisions. Investigators obtained discharge diagnosis andisposition information by review of ED discharge log-ooks and medical records maintained by the ED.

The hospital’s research ethics board approved thetudy. The board waived the need for patient consentecause it considered helical CT scanning in the settingf suspected nephrolithiasis to be a standard of care. Thearticipating physicians consented to participating in thetudy during an introductory meeting with study person-el, and by agreeing to enroll patients and complete eachatient survey.

The study institution performs helical CT scans dur-ng weekdays only. The physician could discharge studyubjects from the ED before CT scanning when the CTcanner was not in operation. These subjects underwentelical CT scanning the following morning, with scanesults reviewed by the EP on duty at that time. Studyubjects enrolled on weekends underwent helical CTcanning the following weekday.

The primary outcome measure was the comparison ofhe EP’s diagnostic certainty and the results of the CTcan. Secondary outcome measures include alternate di-gnoses and changes in patient disposition after CT scan-ing.

All data were entered into a commercially availableatabase (Access 2000™, Microsoft Corporation, Red-ond, WA) and analyzed using a commercially available

preadsheet (Excel 2000™, Microsoft Corporation).ubjective and objective clinical findings, imaging re-ults, diagnoses, and patient dispositions were catego-ized as discrete, unordered variables. Physician pre-CTiagnostic certainties were categorized as a continuousariable. Descriptive statistics summarized the data.

RESULTS

hree hundred sixty-five patients with suspected neph-olithiasis presented to the ED during the study period.f these, 206 (56.4%) were excluded from the study.able 1 lists the reasons for exclusion. The remaining59 patients met the study’s inclusion criteria and were

able 1. Reasons for Exclusion

Reason N

rior diagnosis of urinary calculi 183istory of genitourinary abnormality 12ransfer from other institution 6ther 3ge less than 18 years 1ontraindication to CT scan 1

ligible for enrollment. The EPs enrolled 132 (83.0%) of

tedsh

1T5

cwdo

ica(1T

tpap0c

ictasqTpFc

tma

T

579

* tone”

T

CT scan in nephrolithiasis 227

he 159 eligible patients. The remaining 27 (17%) of theligible subjects were not enrolled because the physicianid not identify them as potential candidates for thetudy, did not complete the patient enrollment form, or aelical CT scan was not performed.

The EPs completed the questionnaire for 129 of the32 enrolled patients, and 121 underwent CT scanning.he mean patient age was 47 years (range 19–90), and8.7% were men.

Six patients did not undergo CT scanning due to physi-ian preference for intravenous pyelography. Two patientsho presented when the CT scanner was not in operationid not return for their CT scan. The study results are basedn the 121 patients who underwent CT scanning.

Physician pre-CT diagnostic certainties were groupednto four categories to reflect different degrees of physi-ian confidence in the clinical diagnosis of nephrolithi-sis: 0–49% (low), 50–74% (intermediate), 75–90%high) and 90–100% (very high) confidence. There were4, 35, 31, and 41 patients in each category, respectively.able 2 compares the pre-CT diagnostic certainty with

able 2. Pretest Diagnostic Certainty and CT Scan Results

CT scan results

Pretest diagnosticcertainty Stone No sto

0–49% 4 100–74% 16 195–89% 23 80–100% 33 8

Other significant pathology identified in either “stone” or “no s

able 3. Other Significant Pathology Identified by CT Scan

Pretest diagnosticcertainty (%) Nephrolithiasis

20 No M20 No R39 No R42 No E50 No R50 Yes M66 No L70 No S79 Yes R88 No M88 Yes P89 No L91 Yes R95 Yes R96 Yes R96 No R98 No L98 No L

100 Yes S

he results of CT scanning, and includes the number ofatients where the CT scan also identified significantlternate pathology. The CT scan results indicate therobability of a correct diagnosis by the physician is.29, 0.46, 0.74, and 0.81 for each pre-CT diagnosticertainty category, respectively.

After CT scanning, an alternate diagnosis was maden 40 (33.1%) study subjects. Of these, 19 (47.5%) in-luded significant pathology, including 7 (17.5%) pa-ients in whom the CT scan revealed nephrolithiasis inddition to the significant pathology. The study definedignificant pathology as a finding on CT scan that re-uired further investigation, treatment, or consultation.able 3 lists the significant alternate diagnoses and thehysician pre-CT diagnostic certainty of nephrolithiasis.igures 1 and 2 illustrate examples of CT findings indi-ating nephrolithiasis or other significant pathology.

Before CT scanning, physicians determined 115 pa-ients would be discharged and six patients would be ad-itted. After CT scanning, six of the former group were

dmitted, and five of the latter group were discharged.

Probability ofcorrect diagnosis

Other significantpathology found*

0.29 40.46 40.74 40.81 7

group.

Other significant pathology

eric and retroperitoneal lymphadenopathy, possible lymphomadrenal adenomavarian neoplasm with liver massd prostatedrenal mass

liver lesions, possibly metastatical cell carcinoma, abdominal aortic aneurysmegaly, cause not yet determined

drenal adenomaeric lymphadenopathy, possible lymphomae ulcerative colitisiureteric mass with hydronephrosis, liver metastasisnal cysts, possible malignancy

drenal adenomanal mass, possible malignancycal lymphadenopathy, possible metastatic testicular cancerrian cancer

in right hepatic lobeegaly, cause not yet determined

ne

esentight aight onlargeight aultiple

eft renplenomight aesentossibleft peright reight aight reetroceeft ovaesionplenom

Frc

228 M. Ha and R. D. MacDonald

igure 1. (a) CT scan of a 41-year-old male patient with high pre-CT diagnostic certainty of nephrolithiasis demonstrating mildight-sided hydroureter (arrow) due to nephrolithiasis. (b) CT scan of a 41-year-old male patient with high pre-CT diagnosticertainty of nephrolithiasis demonstrating nephrolithiasis (arrow) at proximal right ureterovesical junction.

Tip

nctio

Fl(swl

CT scan in nephrolithiasis 229

DISCUSSION

his study illustrates the importance of helical CT scann patients with their first presentation of unilateral flankain where nephrolithiasis is the suspected clinical diag-

igure 2. (a) CT scan of 62-year-old male patient with higeft-sided hydronephrosis (white arrow), clot in left renal collethin white arrow). (b) CT scan of a 62-year-old male patientrating obstruction of left ureter due to soft tissue mass (mith high pre-CT diagnostic certainty of nephrolithiasis dem

iver lesion due to metastatic disease (black arrow).

osis. Helical CT scan can enhance physician diagnosticertainty and patient disposition. It also identifies pa-ients with pathology that mimics nephrolithiasis on clin-cal presentation. This study is similar to others in dem-nstrating that CT scan establishes a significant

-CT diagnostic certainty of nephrolithiasis demonstratingsystem (black arrow), and possible metastatic lesion in liverhigh pre-CT diagnostic certainty of nephrolithiasis demon-ncy; white arrow). (c) CT scan of a 62-year-old male patientting left-sided hydronephrosis (white arrow) and a calcified

h precting

t withalignaonstra

pc

teplpswi

aCdtflCIaa9a(uedw

nHnicii

rfiEadnp

4hih

ctasep

scitmtctsdl(cvfIfmIditiAdadd

tp

T

FFHU

*

230 M. Ha and R. D. MacDonald

ercentage of alternate diagnoses in patients with clini-ally suspected nephrolithiasis (1–4,6–8).

This study differs from others because it is a prospec-ive investigation of patients presenting with their firstpisode of clinically suspected nephrolithiasis. Theseatients are likely to be less differentiated and moreikely to have an alternate diagnosis than those with arevious history of nephrolithiasis. In the ED, it is thisubset of patients with undifferentiated flank pain inhom helical CT scan would have the most significant

mpact on diagnosis and management.Ureteral calculi are treated on the basis of size, location,

nd composition. Previous studies demonstrate that helicalT scan has a high sensitivity, specificity, and accuracy inetecting ureteral calculi (1,3,5,7). Miller et al. prospec-ively compared CT scan to IVP in the evaluation of acuteank pain where all patients had CT scan followed by IVP;T scan was found to be 96% sensitive and 100% specific;

VP was 87% sensitive and 94% specific (5). Nachmann etl. retrospectively evaluated 412 consecutive patients withcute flank pain using CT scan and found a sensitivity of7%, specificity of 92%, a positive predictive value of 88%nd a negative predictive value of 98% at stone detection7). CT scan reveals the presence, size, and location ofreteral calculi with greater accuracy than intravenous py-lography (1,3,5). Helical CT scan also has the advantage ofiagnosing nephrolithiasis when the stone has passed,hich IVP can miss (1,8).Emergency physicians are often required to make diag-

oses on clinical grounds after limited or no investigations.istorically, they have relied on clinical findings and uri-alysis to make the diagnosis of nephrolithiaisis. Table 4llustrates the sensitivity and specificity of some commonlinical features used by physicians in our study. The resultsndicate these features lack sensitivity, specificity, or both,n establishing a diagnosis of nephrolithiasis.

This study demonstrates that clinical diagnosis of neph-olithiasis is not adequate in patients who present with theirrst suspected episode, and helical CT scan can enhance theP’s diagnostic certainty. The study also demonstrates thatlternate diagnoses and changes in patient disposition occurue to results of the CT scan. Finally, even in the setting ofephrolithiasis, some patients may also have coexisting

able 4. Clinical Features Potentially Associated with Nephr

Clinical feature

lank painlank pain on palpation or percussionematuria*nable to find comfortable position

Gross or microscopic.

athology that may require further investigation. Although p

of the 7 patients with concomitant pathology in this studyad pathology anatomically proximate to the renal collect-ng system, it is unlikely the additional pathology wouldave been identified by IVP.

Availability of helical CT scanning is limited in someenters. At the study institution, CT scanning is not rou-inely available at night or on weekends. Despite its limitedvailability, all patients who present with a first episode ofuspected nephrolithiasis should undergo CT scanning tonsure accurate diagnosis in this undifferentiated patientopulation.

This study was not designed to compare helical CTcanning vs. other imaging modalities in terms of relativeost, radiation exposure, evaluation or treatment time. Othernvestigators have demonstrated the average total diagnos-ic imaging costs for CT scanning are comparable to otherodalities (9,10). Patients undergoing helical CT scanning

o investigate suspected nephrolithiasis also spend signifi-antly less total time in the ED (291 vs. 410 min, respec-ively), and the time to definitive diagnosis is significantlyhorter (6.3 vs. 16.8 h, respectively) (11,12). The onlyisadvantage to helical CT scanning is the two- to threefoldarger radiation dose compared with the 3-film IVP8,9,13). Thomson et al. conducted a randomized studyomparing the cost and radiation exposure of CT scannings. IVP, and found the average total diagnostic imaging costor the CT scan group to be A$181.94 and A$175.46 for theVP group; mean radiation dose to diagnosis was 5.00 mSvor CT scan vs. 2.50 mSv for IVP (9). In the same study,ean imaging time was 30 min for CT scan and 75 min for

VP (9). Greenwell et al. reported the effective radiationose of CT scan to be 4.7 mSv vs. 1.5 mSv for IVP, anddentical cost for both IVP and CT scan (8). In comparison,he average radiation dose for upper abdominal X-ray scans 3 mSv, and chest X-ray scan (PA) 0.02 mSv. (14).lthough unenhanced helical CT scanning provides greateriagnostic information, requires less time to definitive di-gnosis, and avoids the risk of intravenous contrast me-ium, physicians should weigh this against the greater ra-iation dose to the patient.

This study has several limitations. The radiologists in-erpreting the helical CT scans were not blinded to theatients’ symptoms or attending physician’s clinical im-

sis

Sensitivity Specificity

.96 (0.89–0.99) 0.04 (0.01–0.15)

.63 (0.52–0.73) 0.33 (0.21–0.47)

.80 (0.69–0.87) 0.41 (0.28–0.56)

.18 (0.11–0.29) 0.65 (0.57–0.76)

olithia

0000

ression. The extent to which this affected the study is not

kbpwtsdhbnit

fiucpaa

AtFatpSf

1

1

1

1

1

CT scan in nephrolithiasis 231

nown. A total of 159 patients were eligible for the study,ut only 121 (76.1%) underwent CT scanning. It is notossible to determine what impact, if any, these 38 patientsould have made on the results. The study examined only

hose patients presenting with a first episode of clinicallyuspected nephrolithiasis. This patient population is lessifferentiated from the population of patients with a prioristory of nephrolithiasis. The results of this study may note applicable to the latter, more prevalent population. Fi-ally, this study did not track patients with other pathologydentified by CT scan to determine the long-term impact ofhese alternate diagnoses.

This study demonstrates that patients presenting with arst episode of clinically suspected nephrolithiasis shouldndergo CT scanning. It enhances physician diagnosticertainty by identifying alternate diagnoses and significantathology not suspected on clinical grounds alone, and,lthough uncommon, can uncover significant pathology inddition to nephrolithiasis.

cknowledgments—This work was supported by a grant fromhe Health Sciences Centre Medical Staff Council Fellowshipund. We thank the Emergency Department physicians, nurses,nd staff at St. Boniface General Hospital, Winnipeg, Mani-oba, Canada for identifying and enrolling patients, and thehysicians and staff in the Department of Diagnostic Imaging,t. Boniface General Hospital, Winnipeg, Manitoba, Canadaor their assistance in carrying out this study.

REFERENCES

1. Katz DS, Scheer M, Lumerman JM, Mellinger BC, Stillman CA,Lane MJ. Alternative or additional diagnoses on unenhanced he-lical computed tomography for suspected renal colic: experiencewith 1000 consecutive examinations. Urology 2000;56:53–7.

2. Chen MYM, Zagoria RJ. Can unenhanced helical computed to-mography replace excretory urography for evaluation of patientswith acute urinary tract colic? J Emerg Med 1999;17:299–303.

3. Boulay I, Holtz P, Foley WD, White B, Begun FP. Ureteral calculi:diagnostic efficacy of helical CT and implications for treatment ofpatients. AJR Am J Roentgenol 1999;172:1485–90.

4. Longo J, Akbar SA, Schaff T, et al. A prospective comparativestudy of non-contrast helical computed tomography and intrave-nous urogram for the assessment of renal colic. Emerg Radiol2001;8(5):285–92.

5. Miller OF, Rineer SK, Reichard SR, et al. Prospective comparisonof unenhanced spiral computed tomography and intravenous uro-gram in the evaluation of acute flank pain. Urology 1998;52:982–7.

6. Chen MYM, Zagoria RJ, Saunders HS, Dyer RB. Trends in the useof unenhanced helical CT for acute urinary colic. AJR Am JRoentgenol 1999;173:1447–50.

7. Nachmann MM, Harkaway RC, Summerton SL, et al. Helical CTscanning: the primary imaging modality for acute flank pain. Am JEmerg Med 2000;18:649–52.

8. Greenwell TJ, Woodhams S, Denton ER, MacKenzie A, RankinSC, Popert R. One year’s clinical experience with unenhancedspiral computed tomography for the assessment of acute loin painsuggestive of renal colic. BJU Int 2000;85:632–6.

9. Thomson JM, Glocer J, Abbott C, Maling TM, Mark S. Computedtomography versus intravenous urography in diagnosis of acuteflank pain from urolithiasis: a randomized study comparing costsand radiation dose. Australas Radiol 2001;45:291–7.

0. Grisi G, Stacul F, Cuttin R, Rimondini A, Meduri S, Dalla PalmaL. Cost analysis of different protocols for imaging a patient withacute flank pain. Eur Radiol 2000;10:1620–7.

1. Rekant EM, Gibert CL, Counselman FL. Emergency departmenttime for evaluation of patients discharged with a diagnosis of renalcolic: unenhanced helical computed tomography versus intrave-nous urography. J Emerg Med 2001;21:371–4.

2. Patel M, Han SS, Vaux K, Saalfeld J, Alexander JH. A protocol ofearly spiral computed tomography for the detection of stones inpatients with renal colic has reduced the time to diagnosis andoverall management costs. Aust N Z J Surg 2000;70:39–42.

3. Denton ER, Mackenzie A, Greenwell T, Popert R, Rankin SC.Unenhanced helical CT for renal colic—is the radiation dosejustifiable? Clin Radiol 1999;54:444–7.

4. U.S. Food and Drug Administration Department of Health and Hu-man Services. Whole body scanning, using computed tomography.Available at: http://www.fda.gov/cdrh/ct/risks.html#dose. AccessedFebruary 3, 2004.