J. clin. Path. (1964), 17, 492

Bacteriological complications from the use ofurethral instruments: Principles of prevention'

J. P. MITCHELL AND W. A. GILLESPIE

From the Departments of Urology and Bacteriology,United Bristol Hospitals and University of Bristol

sYNoPsis Guidance by bacteriologists and changes in theatre and ward disciplines have greatlyimproved the results of urological surgery in recent years. The most noteworthy successes have beenin treatment of chronic retention with renal failure, and in prevention of the bacteraemia whichmay occur following removal of the post-operative catheter or the passage of an instrument todilate a stricture. It is hoped that safer urethral instrumentation may also reduce the incidence ofchronic pyeloniephritis.

PART I: THE PROBLEMS AS SEEN BY THE SURGEON

Many papers have been written about urinary tractinfection, but the risks of contamination by instru-ments passed along the urethra have received littleattention until recently. The immediate acutesymptoms of instrumentally-induced infection in themale are easily recognized, for example, acuteprostatitis and epididymitis. Though many infectionsclear up without causing great harm, the con-sequences may be serious as the following briefhistory illustrates.

A patient had a transurethral resection of the prostateand bladder neck performed some years ago. Folowingthe operation, the urine was infected with Proteus, whichwas not eradicated. However, because pus had dis-appeared from his urine, it was assumed that his naturalresistance had overcome the infection and there was noneed to follow him any longer. Within nine months ofhis original operation, bilateral renal calculi had developed(Fig. 1). Proteus is a potential stone former, with apredilection for the urinary tract, but fortunately it isunusual for a stone to be produced so quickly.

Until recent years, some infection followingprostatectomy was regarded as inevitable. In fact,some of the older surgeons practising urology used tosay that a 'touch of infection' of the urinary tractwas an advantage. This impression probably arosefrom the results of treating patients with chronicretention of urine. A large volume of uninfectedurine remaining in the bladder after micturition'Based on two papers read at the April meeting of the Association ofClinical Pathologists.

always carries a serious risk, because of the possibleassociated hydronephrosis and renal failure. Thekidneys had so far withstood the back pressure, butnot infection. Surgery, or eventhesimplestprocedure,such as the passage of a catheter, was sufficient tointroduce infection, adding a further burden to thealready damaged kidneys, with the result that the

FIG. 1. Bilateral renal calculi after Proteus infection2 hadpersistedfor only nine months.

492

copyright. on D

ecember 4, 2020 by guest. P

rotected byhttp://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.17.5.492 on 1 Septem

ber 1964. Dow

nloaded from

Bacteriological complicationsfrom the use ofurethral instruments: Principles ofprevention

patients went downhill and many died fromascending urinary infection and acute pyelonephritis.The patient who came for treatment of chronicretention, having already survived an infection, wasin a stronger position. Drainage of the bladderproduced nothing but benefit because the kidneyshad already withstood both back pressure andurinary infection. The catheter relieved the pressureand gave such a patient the chance to combat theinfection. It was for this reason that the oldersurgeons gained the impression that 'a touch ofinfection' was an advantage. This stage has beenpassed, as nowadays we can often avoid infectionaltogether. Our aim now is to drain the uninfectedchronic retention, taking meticulous care to preservea sterile urine until the kidneys recover.Another example of infection of the urinary tract,

which in the past has been accepted as an inevitablehazard, was in regular attenders at urethral strictureclinics. These men were accustomed to having a

rigor during the evening following dilatation of thestricture. By the next morning they usually felt fit toreturn to work. Scott (1929) and Barrington andWright (1930) showed that the attacks were due tobacteraemia. Mitchell, Slade, and Linton (1962)confirmed that these patients were suffering frombacteraemia and described methods of introducingdisinfectants into the urethra and bladder (seepart II); their investigations were stimulated by atragic episode which befell one patient with a

traumatic stricture who, as an out-patient, had beentreated periodically by dilatation. On returninghome after a difficult dilatation he had a rigor farworse than any previous one, but unfortunately didnot call his doctor for 48 hours. By this time a

diagnosis of acute coronary occlusion was made,and the patient was admitted to a medical ward.He proved, in fact, to have a fulminating septic-aemia caused by a coliform bacillus. He lost one eyefrom panophthalmitis, had several metastaticabscesses, and a spondylitis which took 18 monthsto settle. A tragedy such as this must be very rare,but it emphasizes the need to investigate the rigorsand if possible prevent them.Another example of a similar nature is the so-

called 'catheter fever', when a patient, following an

uneventful prostatectomy, develops slight pyrexiawithin a few hours of the removal of the catheter.In the days when most prostatectomies were

followed by infection, this was a remarkably commonoccurrence. It was shown that blood culture takenimmediately after removal of the catheter was

positive in almost half the patients with infectedurine but not in patients with sterile urine (Slade,1958). The urethral mucosa had become hyperaemicdue to the presence of the catheter, and after the

catheter was removed, the inflamed mucosa allowedorganisms to pass into the blood stream. Thisbacteraemia often seemed to occur with the firstact of micturition, and was found to be aggravatedif the patient had the additional pelvic disturbance ofa bowel action at the same time.

Catheter-induced bacteraemia, though oftentransient and symptomless, not infrequently causesbrief pyrexia and rigor. Occasionally the consequenceis much more serious and has been aptly describedas bacteraemic shock (Waisbren, 1951). Theresponsible organism is usually a Gram-negativebacillus. The clinical picture is usually a febrile toxiccondition, but it can also present with little or nofever in a very ill patient whose sudden fall of bloodpressure suggests a cardiovascular accident. In fact,it is these cases of hypotensive febrile bacteraemiawhich require the most urgent and effective treatmentand the improvement can then be equally rapid anddramatic. The sudden collapse of a urologicalpatient should warn the clinician that the cause maybe bacteraemia. The most important step in treat-ment is to inject one or more antibiotics which willbe likely to deal with prevalent cross-infectingorganisms (British Medical Journal, 1964).The danger of catheter-induced infection in

females has not been appreciated until recently.In the male, catheterization was known to carrysome danger, because of the epididymitis andprostatitis which sometimes followed it. In thefemale, on the other hand, acute immediatesymptoms rarely followed the passage of a catheterand therefore its dangers escaped notice. Not onlywere women catheterized to obtain urine for thelaboratory, but also it was the practice in labourwards to pass a catheter just before delivery to makesure that the bladder was empty. Gynaecologistshave used catheters repeatedly; in fact, catheteriz-ation was almost an essential part of a pelvic repair.Many gynaecological patients were either catheter-ized repeatedly after operation or had an indwellingcatheter for several days. Specimens of urine fromsuch patients with open-draining indwelling cathetersshowed that 97% became infected (Gillespie,Lennon, Linton, and Slade, 1964).We know that in women a urinary infection

sometimes appears for the first time spontaneously.Furthermore, if the urinary infection produces anysymptoms and the patient is then treated withantibiotics, it usually clears up. But some infectionsin women do not respond to treatment or recur soonafter the antibiotic is withdrawn. Presumably thereis a risk that infection in the lower urinary tract,whatever its origin, may ascend to the kidney. Howoften, then, may pyelonephritis have resulted fromindiscriminate catheretization of a woman either to

493

copyright. on D

ecember 4, 2020 by guest. P

rotected byhttp://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.17.5.492 on 1 Septem

ber 1964. Dow

nloaded from

J. P. Mitchell and W. A. Gillespie

collect a specimen of urine, or as a preliminary tochildbirth?When it was realized that the infections which so

commonly followed operation and catheterizationwere sometimes harmful, methods of avoiding themwere introduced. These were the avoidance ofunnecessary catheterization, efficient sterilization ofinstruments, disinfection of the urethra at the timeof operation or catheterization, and maintenance ofsterility during the period of catheter drainage.These methods are dealt with in part II but it isappropriate at this point to emphasize the peculiardifficulties of sterilizing cytoscopes. Everyone wouldagree that disinfection by heat is most reliable butalthough the cement of lens mountings can bemadetowithstand temperatures of 80°C. or 100°C., there arestill certain telescopes such as those used in panendo-scopes and in children's cystoscopes in which themore heat-resistant cements cannot be used. Thefine insulation in the cystoscope sheath carrying thecable to the cystoscope lamp is also liable to perishwith repeated heat changes. Finally, the air trappedinside the telescope will expand and contract withtemperature changes and put an additional strain onthe lens mountings. Autoclaving a cystoscope istherefore unsafe, and it seems doubtful whether anyperfect method of sterilizing a cystoscope has yetbeen devised, though the best is some form ofpasteurization as described by Francis (1959).The speed of sterilization is also important. So

long as only the standard cystoscope is required, nodelay should be experienced, because these instru-ments can be duplicated in the theatre in sufficientnumbers to allow a reasonable time for sterilization.On the other hand, a urologist starting an endoscopiclist does not know for certain what instruments hewill require beforehand. To sterilize every endoscopicinstrument before starting on a cystoscopy listwould be to subject some of the less commonly usedinstruments to unnecessary trauma. The bestexample is endoscopy in children, when a variety ofsizes of instruments may be called for at short notice.For this reason, a rapid method of sterilizing cysto-scopes is necessary.The obvious answer to this problem is storage of

all endoscopy instruments in a sterile state. In thepast, this has been attempted by the formalincabinet but this has obvious bacteriological short-comings in addition to the conjunctival discomfortof the surgeon and the danger of irritating thepatient's urethra. The formalin-sterilized cystoscopestill smells of formalin however much it is rinsed orwiped. In a nutshell, then, what is required is arapid effective method of sterilization which willnot damage the instruments and some means ofstoring them indefinitely in a sterile state.

PART II: BACTERIOLOGICAL PRINCIPLES IN

THE PREVENTION OF INFECTION

Prevention should be based on knowledge of howbacteria reach the bladder of the urological orgynaecological patient. The organisms may comefrom the patient, by self-infection, or from anotherpatient, by cross-infection (Fig. 2). The mostimportant source of cross-infection is the urine ofanother patient, from which organisms may betransferred by cystoscopes, urine bottles, etc.However, with the exception of the patient withsuprapubic drainage, there is only one final portalof entry which the organisms must pass, the urethra.

Bacteria may enter the bladder at the moment ofinserting the cystoscope or catheter. They maycome from the instrument, if it is not sterile, orfrom the urethra, which is never sterile. If a catheteris left indwelling, it may permit entry of bacteria intwo more ways. First, with open drainage, bacteriamultiplying in urine in the collecting vessel willcontaminate the lower end of the catheter and thencebe carried up to the bladder by reverse currents offluid propelled by rising air bubbles, and perhapsin other ways. These organisms will often be cross-infecting strains. The second route is mainly one ofself-infection, and is the space between the catheterand urethral wall. The short female urethra permitsthe catheter to move when the patient moves. This'extra-catheter' route is probably less important inmen.

These facts suggest two approaches to the problemof preventing infection. One is to exclude bacteriafrom the urethra, and the other is to destroy them assoon as they enter the bladder. For these purposes,local disinfectants are preferable to systemic chemo-therapy with its tendency to breed resistant strains.In Bristol we rely on asepsis and closed bladderdrainage, supplemented by local antisepsis.

VECTORS

'Other patients':

CROSS- Infeoted urine

INFECTION Wounds s

"I Faeaes, etc.

SOURCES

rPatient's own:

SELF- Urethra *I NFECTION Faeces

eta.

CystosoopesCatheters

Urine bottlesBladdersyringes PATIENTS

Nurses' _, URETHRAhands OR

Bedding CATHETER

Baths

Spigots

s etc.

FIG. 2. Sources and routes of urinary infection.

494

copyright. on D

ecember 4, 2020 by guest. P

rotected byhttp://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.17.5.492 on 1 Septem

ber 1964. Dow

nloaded from

Bacteriological complications from the use of urethral instruments: Principles ofprevention

METHODS OF PREVENTION IN WOMEN With inter-mittent catheterization the risk of infection may bereduced by disinfecting the urethra before passingthe catheter. Figure 3 shows two groups of patientswho were repeatedly catheterized after gynaeco-

logical operations. In the control group, the urethrawas lubricated with a non-disinfectant jelly, and inthe test group, with Hibitane (chlorhexidine)obstetric cream three minutes before each catheterinsertion. The incidence of urinary infection was

moderately reduced. The reduction was entirely inGram-negative bacilli. Infections by Streptococcusfaecalis, which is less sensitive than coliform bacillito chlorhexidine, were unchanged (Gillespie, Lennon,Linton, and Slade, 1962).

Chlorhexidine may also be used by instilling aweak solution into the bladder through the catheter,as described by Paterson, Barr, and Macdonald(1960). We now use both methods.

Indwelling catheters nearly always cause infectionif open drainage is used. In a series of gynaecologicalpatients who were treated with open-drainageindwelling catheters for four days after operation,and subsequently were catheterized intermittently tomeasure residual urine, 97% developed urinaryinfection. Infection usually began during the periodof indwelling catheter drainage. The causativepathogens ascended to the bladder both inside andoutside the catheter. The few patients who survivedthis stage without infection became infected whenintermittently catheterized afterwards. However, sincean indwelling catheter is much pleasanter forpatients and nurses than frequent intermittentcatheterization throughout the post-operativeperiod,

80

70 -

60 -

50 -

#AI-

J .40I.-

30 -

20 -

10

NOT INFECTED24% 58%

INFECTED BYGRAM NEGATIVE

BACILLI5 I/%- 1 8%/

INFECTED BYStr foeco/is

ALONE25% 24%

DISI

FIG. 3. Prevention of urinary infection by urethraltreatment with chlorhexidine.

three modifications were instituted in order to makeit safer. (1) The risk of infection at the moment ofinserting a catheter was reduced by disinfecting theurethra before insertion and the bladder immediatelyafterwards, by the methods already described.(2) Closed drainage into vessels containing disinfect-ant was employed. (3) The indwelling Foley catheterwas modified by a device which prevented entry oforganisms between the catheter and urethral wall.This was a collar of polyurethane plastic foam,tightly threaded on to the catheter and pushed upagainst the external urethral meatus as soon as thecatheter was inserted and its retaining bulb inflated.The sponge collar was smeared twice daily withchlorhexidine cream so as to maintain a disinfectantbarrier at the urethral meatus. Each of thesemodifications brought some improvement, and whenthey were used together for two years, the post-operative urinary infection rate averaged 13%compared with the previous rate of 970% (Gillespieet al., 1964).Hannah (1963) obtained a similar improvement

by using different methods. It is therefore clear thatpost-operative infection is generally preventable,even when indwelling catheters are used.The change from open to closed drainage was

followed by an interesting change in the predominantinfecting bacteria. Infections during open drainagewere caused by a variety of organisms, includingProteus and Pseudomonas, but the commonest wereKlebsiellae, many of which were antibiotic-resistant.This finding was in line with results of other workers,some of whom have suggested that Klebsiellaedevelop drug resistance more easily than Escherichiae(Dutton and Ralston, 1957; Talbot, Cunliffe, andGower, 1957; Lattimer, Seneca, Zinsser, and Troc,1959). It was therefore interesting to find that whenclosed drainage was substituted for open drainage,thus reducing cross-infection, though still allowingself-infection by the patients' own flora, Klebsiellaebecame rare, and Escherichiae common (Gillespieet al., 1964). Escherichiae also predominated inother circumstances where cross-infection wasunlikely, for example, in infection acquired outsidehospital and in infections following intermittentcatheterization.

METHODS OF PREVENTION IN MEN The value ofsterile drainage into a closed vessel containingdisinfectant was demonstrated some years ago inmen with acute retention of urine whose bladderswere drained for short periods (Miller, Linton,Gillespie, Slade, and Mitchell, 1960b). Recently wehave investigated the use of carefully supervisedclosed drainage in 32 men whose bladders weredrained for much longer periods (Fig. 4). Some

495

D

copyright. on D

ecember 4, 2020 by guest. P

rotected byhttp://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.17.5.492 on 1 Septem

ber 1964. Dow

nloaded from

J. P. Mitchell and W. A. Gillespie

INFECTION

-n ANTIBIOTIC TREATMENT

FIG. 4. Urinaryinfections in mentreated withclosed drainage.

10DAYS OF DRAINAGE

patients were given antibiotics for respiratoryinfection, but this did not greatly influence theresults. The drainage treatment was closely super-vised to ensure that asepsis was maintained,especially when changing bottles. Many patientswere kept free from infection for several weeks.When infection did occur, it was often attributable toelementary faults in technique. Previous experienceshowed that with open drainage all these patientswould have been infected within a few days, but itwas not considered justifiable to include a controlgroup with this series.More complicated methods were needed to prevent

infection after prostatectomy by the transurethraland retropubic methods. In addition to closeddrainage, the urethra was disinfected before opera-tion, disinfection of cystoscopes was improved, anda device for irrigating the bladder without interrupt-ing closed drainage was adopted (Miller, Gillespie,Linton, Slade, and Mitchell, 1958). The results oftrials carried out in two hospitals between 1956 and1959 showed a reduction of the infection rate from anaverage of 82% to 11 % (Miller, Gillespie, Linton,Slade, and Mitchell, 1960a). The same methodshave been in routine use since 1959, but are lessclosely supervised by the laboratory. Results arechecked by quantitative cultures of two specimensof urine from each patient, the first obtained atoperation and the second five to seven days later.Although some specimens have been missed,sufficient pairs have been cultured to allow theinfection rates in one of the hospitals to be calculatedeach year since 1959 (Table). The results were notquite as good as during the period of close labora-tory supervision but remained much better thanbefore the new methods were adopted.

TABLEURINARY INFECTION AFTER TRANSURETHRAL AND

RETROPUBIC PROSTATECTOMY AT BRISTOL ROYAL INFIRMARY

1955-19641Date Special Precautions (see text)

1955-56 None1958-59 Full precautions, closely supervised1960 Full precautions, routine1961 Full precautions, routine1962 Full precautions, routine1963 Full precautions, routine1964 Full precautions, routine(4 months)

Infection Rate

42/57 (74%Y.)12/89 (13 %)5/38 (13%)8/53 (15%)18/92 (20%)17/73 (23%)5/28 (18 %)

'Infections by Staph. albus were excluded (Mitchell, 1964).

Finally, we must consider how to preventbacteraemia which, in patients with infected urine,often follows the dilatation of urethral strictures.The only certain method of prevention would be toavoid urinary infection. But when the urine isalready infected can anything be done, in additionto, or instead of, prescribing systemic chemotherapywith all its disadvantages? Recent studies haveshown the value of disinfecting the urethra withchlorhexidine before dilating the stricture anddisinfecting the bladder immediately afterwardsbefore the patient passes urine. In a controlled trialblood cultures were taken from patients withinfected urine five minutes after dilatation andafter subsequent micturition (Mitchell, Slade, andLinton, 1962). The unprotected patients experienceda very high incidence of bacteraemia, 39% afterdilatation and 75 %after subsequent micturition. Thebacteraemia was transient and usually caused no

symptoms. Urethral disinfection loweredthe incidenceof positive blood cultures after dilatation from 39%to 10%; disinfection of the urethra or bladder, or

32

28-

24-

20in3~-zL 16-31-

12-

8-

4

40 70

496

copyright. on D

ecember 4, 2020 by guest. P

rotected byhttp://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.17.5.492 on 1 Septem

ber 1964. Dow

nloaded from

Bacteriological complications from the use ofurethral instruments: Principles ofprevention 497

both, reduced the post-micturition bacteraemiasfrom 75% to 10%. These findings are furtherevidence of the value of using local antisepsis tosupplement aseptic techniques. The methods areessentially simple, but require attention to details bynurses who understand the underlying principles.Otherwise, the methods will fail. Standards can bestbe maintained by regular courses of instruction fornurses.

REFERENCES

Barrington, F. J. F., and Wright, H. D. (1930). J. Path. Bact., 33,871.British medical Journal (1964). Editorial, 1, 254.Dutton, A. A. C., and Ralston, M. (1957). Lancet, 1, 115.Francis, A. E. (1959). Proc. roy. soc. Med., 52, 998.

Gillespie, W. A., Lennon, G. G., Linton, K. B., and Slade, N. (1962).Brit. med. J., 2, 13.

- (1964). Ibid., in the press.Hannah, W. J. (1963). Canad. med. Ass. J., 88, 803.Lattimer, J. K., Seneca, H., Zinsser, H. H., and Troc, 0. (1959).

J. Amer. med. Ass., 170, 938.Miller, A., Gillespie, W. A., Linton, K. B., Slade, N., and Mitchell,

J. P. (1958). Lancet, 2, 608.-,-,-,, -, (1960a). Ibid., 2, 886.

Linton, K. B., Gillespie, W. A., Slade, N., and Mitchell, J. P.(1960b). Ibid., 1, 310.

Mitchell, J. P., Slade, N., and Linton, K. B. (1962). Brit. J. Urol., 34,454.

Mitchell, R. G. (1964). J. clin. Path., 17, 105.Paterson, M. L., Barr, W., and Macdonald, S. (1960). J. Obstet.

Gynec. Brit. Emp., 67, 394.Scott, W. W. (1929). J. Urol., 21, 527.Slade, N. (1958). Proc. roy. soc. Med., 51, 331.Talbot, J. M., Cunliffe, A. C., and Gower, N. D. (1957). J. clin. Path.,

10, 222.Waisbren, B. A. (1951). Arch. intern. Med., 88, 467.

copyright. on D

ecember 4, 2020 by guest. P

rotected byhttp://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.17.5.492 on 1 Septem

ber 1964. Dow

nloaded from