American Journal of Infection Control

Volume 18 Number 5 October 1990

ARTICLES

Effect of surgeon’s diagnosis on surgical wound infection rates Geoffrey Taylor, MD, FRCPC Margaret McKenzie, RN, BScN Teresa Kirkland, RN, BScN Rhoda W/ens, RN, BScN Edmonton, Alberta, Canada

To determine the impact of a surgeon’s diagnosis of surgical wound infections on infection rates, during a 6-month period we prospectively examined patients undergoing surgical wound surveillance for any of four services (orthopedic surgery, general surgery, neurosurgery, or cardiovascular surgery). Criteria were judged as standardized if the infection control practitioner observed pus, redness, or drainage associated with positive culture or if a diagnosis of deep-seated infection was made. Surgeon’s diagnosis was judged as a nonstandardized criterion. Using the Centers for Disease Control’s criteria, we identified 113 surgical wound infections in 3024 patients undergoing surgical procedures in the four services. Of these, 95 (84%) met objective criteria (pus observed in 53%; drainage, redness, and positive culture in 20%; and deep-seated infection in 11%). In 18 patients (16%), the nonstandardized criterion alone was used for diagnosis. There was wide variation in use of the nonstandardized criterion, ranging from 5% of orthopedic infections to 21% of cardiovascular surgery infections and 40% of neurosurgical infections. For individual surgeons with at least one wound infection, the range of surgeon’s diagnosis was up to 67%. We conclude that a surgeon’s diagnosis can have .a major impact on surgical wound infection rates; this impact is not borne equally among surgical services or individual surgeons. (AM J INFECT CONTROL 1990;18:295-9)

Several studies have demonstrated that sur- gical wound infection rates can be reduced by reporting their specific infection rates to sur-

From the Infection Control Unit, University of Alberta Hospitals.

Reprint requests: G. D. Taylor, MD, FRCPC, 2E3.11, Walter Mackenzie Center, University of Alberta Hospitals, Edmonton, Alberta, Canada T6G 287. Presented at the Sixteenth Annual Conference of the Associ- ation for Practitioners in Infection Control, Reno, Nev., May 21- 26, 1989.

17146117544

geons.lm3 Hospital infection control programs are being urged to adopt surgeon-specific sur- gical wound surveillance. In the most recent version of their infection control guidelines, the Centers for Disease Control (CDC) has made such surveillance a category II (moderately rec- ommended) standard.4 In a recent article Haley5 suggested that infection control pro- grams that do not adopt this form of surveil- lance risk being eclipsed by surgeons who will do it themselves.

There has, however, been little discussion

295

296 Taylor et al. American Journal of

INFECTION CONTROL

Table 1. Surgical wound infections by criteria for diagnosis and surgical service

No. of infectlons (%)

Neurosurgery General surgery

Csrdiovssculsr Orthopedic surgery surgery Total

PUS 2 (40) 32 (60) 14 (41) 12 (57) 60 (53) Drainage, positive 1 cm 8 (15) 9 (26) 5 (24) 23 (20)

culture Deep 0 (0) 5 (9) 4 (12) 3 (14) 12 (11)

Surgeon’s diagnosis TOTAL

18 (16) ii3

Table 2. Clean wound infection rates by alternative diagnostic criteria

Infection rate (%)

Criteria l-4 Criteria 1-3 Criteria 1 and 2 Criterion 1

Neurosurgery General surgery Cardiovascular surgery Orthopedic surgery All services

5/81 (6.2) 3181 (3.7) 3181 (3.7) 2181 (2.5) 7/484 (1.4) 5/484 (1 .O) 41484 (0.8) 3/484 (0.6)

31/471 (6.6) 231471 (4.9) 20/471 (4.2) 11/471 (2.3) 101832 (1.2) 9/832 (1 .l) 7/832 (0.8) 61832 (0.7) 5311868 (2.8) 40/1868 (2.1) 34/1868 (1.8) 22/1868 (1.2)

concerning the actual definition of surgical wound infection. Infection control programs starting this type of surveillance will be tempted to adopt the CDC’s definitions6 if they have not already done so during their ongoing surveillance program. For surgical wound in- fection this definition permits a surgeon’s di- agnosis of infection alone to be accepted. On the other hand, most published studies on surgical wound infection do not accept the surgeon’s diagnosis alone. Using the CDC’s def- inition, we have been conducting surgeon- specific surgical wound surveillance for the last 3 years, We wished to determine the impact of the surgeon’s diagnosis on infection rates and therefore prospectively analyzed the criteria for diagnosis of these infections during a 6-month period.

Methods

Surgical wound surveillance is carried out by infection control practitioners, (ICPs) who di- rectly observe the surgical wound as part of the process of determining the presence or absence of infection. A patient’s incisions are examined by an ICP beginning 48 hours postoperatively and every 2 days thereafter until discharge, or

for 2 weeks, whichever comes first. Cultures may be taken, if necessary, without a doctor’s order by the ICP during the observation process or by staff nurses during dressing changes. Dur- ing a 6-month period beginning in June 1988, all patients in four surgical services of our hos- pital (general surgery, neurosurgery, orthope- dics, and cardiovascular surgery) undergoing a surgical procedure in which an incision of the skin was made were prospectively examined.

A diagnosis of surgical wound infection was made if 1. Pus was observed issuing from the wound by

the ICP 2. The ICP observed erythema and serosan-

guinous discharge, and a wound culture was positive for a known pathogen (ex- cluding common skin contaminants such as coagulase-negative staphylococci, Propioni-

bacterium sp. or Corynebacterium sp.) 3. A diagnosis of a deep-seated surgically re-

lated infection was made (e.g., intraabdom- inal abscess, osteomyelitis) in the presence of a noninfected incision

4. Surgeon’s diagnosis of wound infection in the presence of some incisional abnormali- ties that by themselves were insufficient to

Volume 18 Number 5

October 1990 Effect of surgeon’s diagnosis on infection rates 297

Table 3. Overall wound infection rates by alternative diagnostic criteria

Infection rate (%)

Neurosurgery General surgery Cardiovascular surgery Orthopedic surgery All services

Crlbrla 14 Critarla l-3 Criteria 1 and 2 Criterion 1

5/89 (5.6) 389 (3.4) 3/89 (3.4) 2/89 (2.2) 53/1361 (3.9) 4511361 (3.3) 40/1361 (2.9) 32/1361 (2.3) 34/565 (6.0) 27/565 (4.8) 23/565 (4.0) 14/565 (2.5) 2111009 (2.1) 20/1009 (2.0) 1711009 (1.7) 12/1009 (1.2)

113/3024 (3.7) 95/3024 (3.1) 82/3024 (2.7) 60/3024 (2.0)

make a diagnosis under criteria I or 2. A surgeon’s diagnosis alone could be made by a. A statement to that effect in the progress

notes b. Prescription by the surgeon of topical

therapy (e.g., removal of sutures and sa- line solution soaks to the incision)

c. Prescription of a course of antibiotics in the postoperative period beyond the pro- phylaxis period

Infection rates by service and surgical wound class were calculated by alternative diagnostic criteria for the 6-month period. Surgical wounds were classified on the clean-dirty scale by standard methods.7

RESULTS

Three thousand twenty-four patients were ex- amined (1361 general surgery, 89 neurosurgery, 565 cardiovascular surgery, and 1009 orthope- dics). With the use of CDC criteria, 116 surgical wound infections were diagnosed. Table 1 clas- sifies these infections by criteria for diagnosis. Ten infections (9%) were diagnosed by the ICP without evidence of an accompanying doctor’s diagnosis by the time of discharge. On the other hand, 18 infections (16%) overall were diag- nosed solely by the surgeons. Of the 18 patients, 10 had a notation by the surgeon in the chart, one was prescribed topical therapy, and seven were prescribed a course of antibiotics in as- sociation with minor wound abnormalities.

Tables 2 and 3 calculate the clean and overall wound infection rates on the basis of alternative definitions of infection for each surgical service. For those surgeons with patients with at least one wound infection during the study period, Table 4 shows the proportion of diagnoses made by the surgeon alone.

Table 4. Surgeon-specific wound infections by diagnostic method

Wound infections % Surgeon’s

Surgeon With surgeon’s Without surgeon’s diagmsis No. dlagnosls diagnosis alone

1 3 1 67 2 2 2 0 3 8 7 12.5 4 2 2 0 5 4 4 0 6 12 9 25 7 6 5 17 8 4 2 50 9 3 3 0

10 6 5 17 11 7 7 0 12 1 1 0 13 13 12 8 14 7 5 29 15 4 2 50 16 4 4 0 17 6 4 33 18 4 4 0 19 2 2 0 20 1 1 0 21 2 1 50 22 2 2 0 23 2 2 0 24 4 4 0 25 4 4 0

DISCUSSION

Whenever comparisons are made between studies on surgical wound infection, it is evi- dently critically important that the definition of infection, the means by which the data was collected, and the duration of follow-up be com- pared. Unfortunately, there is very little uni- formity on any of these points in the published literature. Some studies do not state or have unclear definitions of infection.3,8*9 Some stud-

298 Taylor et al. American Journal of

INFECTION CONTROL

ies consider only superficial infections,‘*“-‘* whereas others count both superficial and deep.13-16 For superficial infections some studies allow pus discharging from the incision only whereas others also allow some form of wound cellulitis.17-19 Occasional studies accept sur- geon’s diagnosis, however that may be defined, in association with wound abnormalities,18*19 but we could find none that accepted surgeon’s diagnosis alone.

The CDC’s definition of nosocomial infection as updated in 1988 is often used by infection control programs. For wound infection this def- inition allows both superficial and deep infec- tions to be considered and also accepts “sur- geon’s or attending physician’s diagnosis of in- fection,” although it does not state how this diagnosis is determined. Our study has shown that accepting doctor’s diagnosis can have a sig- nificant impact on overall infection rates: 16% of all wound infections, and 25% (13/53) of clean wound infections were diagnosed by the surgeon’s clinical impression alone. Further- more, this impact was not equally borne among surgical divisions or individual surgeons: only 4% of orthopedic infections but 30% of cardio- vascular and 40% of neurosurgical infections were diagnosed only by the surgeon. For indi- vidual surgeons the range was even greater: up to 67%. Some surgeons apparently are more ag- gressive than others in making an early diag- nosis of surgical wound infection, possibly as a result of training, experience, or personal fac- tors. Although the numbers are small, it is in- teresting to speculate that the tendency to make an early diagnosis of wound infection in clean procedures, particularly in cardiovascular and neurosurgery, may be a reflection of the disas- trous consequences of infection in those ser- vices.

It is possible that some wounds with mi- nor abnormalities, if not aggressively treated, would have progressed to infection as defined by more objective criteria. However, possibly many would have resolved without treatment. Without a randomized study there is no way to determine which or what proportion might do so. As we have shown, to count all wounds with minor abnormalities that are aggressively man-

aged as infection will greatly increase the in- stitution’s or individual surgeon’s wound infec- tion rate.

Infection control programs are likely at- tracted to surgical wound surveillance by the evidence that overall wound infection rates can be reduced by this technique. The mechanism by which surgeon-specific surgical wound sur- veillance reduces wound infection rates is un- known.

It may be that allowing surgeons to compare their rates with those of their colleagues within the same institution permits those whose rates are high to take steps necessary to bring their practice in line with others. In this setting there is no need to refer to rates outside the institu- tion. However, other groups within the hospital find surgical surveillance data of value. Hos- pital legal departments may find the data useful in combating surgical infection litigation. Hos- pital administrators or surgical program direc- tors may find the data useful in submissions to accrediting agencies. In these cases comparing local data with those published elsewhere will be necessary. In our experience during the last 3 years, the most common question asked by surgical department directors, infection control committee members, and hospital administra- tors concerning our wound surveillance pro- gram is how local data compares to those else- where. Further, even if only intrahospital com- parisons are made, as we have shown, there are marked differences in the tendency of surgeons to make a diagnosis of wound infection when “standardized” criteria do not support that di- agnosis. Consequently, surgeon’s diagnosis be- comes a confounding variable when compari- sons of rates among surgeons are made. Unless infection control programs continuously mon- itor for this effect, one explanation for a higher surgical infection rate would be a greater ten- dency by that surgeon to make a diagnosis of wound infection. Finally, there is the problem of determining when a surgeon has made a di- agnosis of superficial wound infection. In our experience prescription of antibiotics in the postoperative period is common and the indi- cation for such treatment is not always obvious. Documentation in the chart is often absent.

Volume If3 Number 5

October 1990

Likewise, prescription of local treatments, such as removal of sutures and/or saline solution soaks, may be undertaken.

Does this constitute a diagnosis of wound in- fection? Unless one directly approaches the sur- geon concerned, which is not always easily ac- complished and is often time consuming, ques- tions often exist as to what diagnosis the surgeon has in mind.

For these reasons in our view it is preferable not to allow a surgeon’s diagnosis alone for di- agnosis of superficial wound infection. The pro- cess of data collection is simplified, and sur- geon-surgeon comparisons are easier, as are comparisons, if necessary, with other institu- tions .

We discussed our findings with surgical di- visional directors, the departmental chairman, individual surgeons, and the infection control committee. Some surgeons acknowledged a tendency to manage aggressively minor wound abnormalities in the hope of preventing more serious, deep-seated infections. They believed this tendency was especially likely in the setting where infections, if they were to occur, would be disastrous (e.g., cardiac surgery or surgery involving implantation of prosthetic devices). The surgeons strongly supported the surveil- lance program; however, they wished to have data that permitted surgeon-surgeon or insti- tution-institution comparisons with as few con- founding variables as possible. Consequently, our proposal to discontinue recognition, for sta- tistical purposes, of surgeon’s diagnosis alone was unanimously endorsed.

References 1. Cruse P, Foord R. The epidemiology of wound infection.

Surg Clin North Am 1980;60:27-40. 2. Condon RE, Schulte WJ, Malangoni MA, Anderson-

Teschendorf MJ. Effectiveness of a surgical wound sur- veillance program. Arch Surg 1983;118:303-7.

3. Haley RW, Culver DH, White JW, et al. The efficacy of infection surveillance and control programs in pre-

Effect of surgeon’s diagnosis on infection rates 299

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

venting nosocomial infections in U.S. hospitals. Am J Epidemiol 1985; 121: 182-204. Gamer JS. Guideline for prevention of surgical wound infections. Washington DC: US Department of Health and Human Services, 1985. Haley RW. Who will generate surgeon-specific rates? The gauntlet is down. Infect Control Hosp Epidemiol 1988;9:475-6. Gamer JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definitions for nosocomial infections, 1988. AM J INFECT CONTROL 1988; 16; 128-40. Altemeier WA, Burkert F, Pruitt BA, Sandusky WR, eds. Manual on control of infection in surgical patients. 2nd ed. Philadelphia: JB Lippincott, 1984. Edwards LD. The epidemiology of 2056 remote site in- fections and 1966 surgical wound infections occurring in 1865 patients. Ann Surg 1976;184:758-66. Cerrier ED, Bossart KJ, Heer FW. Reduction of infec- tion rates in abdominal incisions by delayed wound closure techniques. Am J Surg 1979;138:22-8. Alexander JW, Fischer JE, Boyajian M, Palmquist J, Morris MJ. The influence of hair removal methods on wound infections. Arch Surg 1983;118:347-52.

Farber BF, Kaiser DL, Wenzel RP. Relation between surgical volume and incidence of postoperative wound infection. N Engl J Med 1981;305:200-4. Green VW, Wenzel RP. Postoperative wound infection: a controlled study of increased duration of hospital stay and direct cost of hospitalization. Ann Surg 1977; 185:264-8. Nagachinta T, Stephens M, Reitz B, Polk BF. Risk fac- tors for surgical wound infection following cardiac sur- gery. J Infect Dis 1987;156:967-73. Kaiser AB, Petracek MR, Lea JW, et al. Efficacy of ce- fazolin, cefamandole and gentamicin as prophylactic agents in cardiac surgery. Ann Surg 1987;206:791-8. Simchen E, Shapiro M, Marin G, et al. Risk factors for postoperative wound infection in cardiac surgery pa- tients. Infect Control 1983;4:215-20. Slama TG, Sklar SJ, Misinski J, Fess SW. Randomized comparison of cefamandole, cefazolin and cefuroxime in open heart surgery. Antimicrob Agents Chemother 1986;29:744-7. Olson M, O’Connor M, Schwartz ML. Surgical wound infections: a S-year prospective study of 20,193 wounds at the Minneapolis VA Medical Center. Ann Surg 1984;199:253-9. Shapiro M, Munoz A, Tager IB, et al. Risk factors for infection at the operative site after abdominal or vag- inal hysterectomy. N Engl J Med 1982;307:1661-6. Ehrenkranz NJ. Surgical wound infection occurrence in clean operations: risk stratification for interhospital comparisons. Am J Med 1981;70:909-14.