This Article Abstract Full Text (PDF) Free Letters to the Editor: Submit a response Alert me when this article is cited Alert me when Letters to the Editor are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My File Cabinet Download to citation manager Rights and Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Sign up for keyword alerts Google Scholar Articles by KLEINERT, J. M. Articles by FIRRELL, J. C. Search for Related Content PubMed PubMed Citation Articles by KLEINERT, J. M. Articles by FIRRELL, J. C. Social Bookmarking                   What's this?

Postoperative Infection in a Double-Occupancy Operating Room. A Prospective Study of Two Thousand Four Hundred and Fifty-eight Procedures on the Extremities*

JAMES M. KLEINERT, M.D., JUDY HOFFMANN, R.N., GOLDIE MILLER CRAIN, , CLAUS FALCK LARSEN, M.D., PH.D., L. JANE GOLDSMITH, PH.D.¶ and JOHN C. FIRRELL, PH.D., LOUISVILLE, KENTUCKY

Investigation performed at the Christine M. Kleinert Institute for Hand and Micro Surgery and Jewish Hospital, Louisville

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
The purposes of this study were to determine the rate of infection associated with elective outpatient operations on an extremity, performed in a double-occupancy operating room (one operating room designed to accommodate two separate operating teams), and to determine which factors influenced this rate. We evaluated the records of 2458 consecutive patients who had had such a procedure, performed by one of nine surgeons during a two and one-half-year period, and in whom the operative wound had been classified as clean (without a drain) or clean-contaminated (with a drain). The information regarding the factors associated with the operation and the operating-room environment was recorded for each patient at the time of the operation. Each wound was inspected periodically in the attending surgeon's office for at least thirty days postoperatively. Using definitions established by the Centers for Disease Control, the attending surgeon determined the presence of infection primarily by judging whether there was purulent drainage or whether erythema or swelling at the operative site was beyond that expected from the procedure. Of the 2458 patients, thirty-seven (1.5 per cent; 95 per cent confidence interval, 1.1 to 2.1 per cent) had infection of the operative wound. Only eight patients (0.3 per cent) had deep infection, with seven of the infections necessitating a reoperation. Infection developed in thirty of the 2311 clean wounds, a rate of 1.3 per cent (95 per cent confidence interval, 0.9 to 1.8 per cent), and in seven of the 147 clean-contaminated wounds, a rate of 4.8 per cent (95 per cent confidence interval, 2.3 to 9.5 per cent) (p = 0.001). No cross-contamination occurred between patients who had infection. The rate of infection was not related to the number of patients who were operated on in the same room at the same time. Logistic regression analysis, used to account for confounding factors, demonstrated a significant association between the classification of the wound (use of a drain) and a higher rate of infection (p = 0.006) as well as between the instillation of a topical steroid solution and a lower rate of infection (p = 0.04). It also demonstrated a significant difference, with respect to the rate of infection, among individual surgeons (p = 0.02).

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Postoperative infection of the wound is an important factor in the outcome of an operative procedure. The rates of infection have been determined for a number of procedures, but most of these rates have been compiled from data on hospital inpatients^{6,7,9,10,12,13,15,17,18}. With the trend toward outpatient operations, the responsibility of monitoring for infection of the wound is assumed by the attending surgeon during follow-up office visits. Whereas infection rates in hospitals are considered routinely during quality-assurance reviews, this has not been the case in most outpatient clinics. In particular, there has been no over-all report, to our knowledge, on infection associated with operative procedures on the hand or an extremity.

In addition to performing a basic quality-assurance review, we wanted to evaluate the rate of infection associated with the use of our particular operating-room environment. A double-occupancy operating room (one operating room designed to accommodate two separate operating teams) has been used for operations on an extremity at our facility since 1960, for more efficient utilization of resources. It was our impression that no cross-infections were occurring, yet no comprehensive study had been performed to evaluate the safety of this environment or to assess the associated over-all rate of infection.

The aims of this study were (1) to determine the rate of infection associated with elective outpatient operations on an extremity, (2) to determine whether cross-contamination occurred with use of the double-occupancy operating room, (3) to compare the rate of infection in our study with previously reported over-all postoperative rates and with rates for specialty-specific operative procedures, and (4) to identify specific factors associated with a higher rate of infection.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
This collaborative study included patients who had an operative procedure performed on an extremity in a double-occupancy operating room between March 1, 1990, and September 30, 1992. The participants included nine hand surgeons as well as the operative nursing staff at the Outpatient Care Center, Jewish Hospital, Louisville, and the nursing and research staff at Kleinert, Kutz, and Associates Hand Care Center, Louisville, Kentucky.

Logistics of a Double-Occupancy Room for Operations on the Hand
The double-occupancy operating room is virtually two separate operating rooms within one physical location. The preoperative preparation of the patient, the preparation of the wound, the operative draping, and the operative instruments are the same as those for operative procedures to be done in a single-occupancy operating room. No instruments are shared. The anesthesia personnel are responsible for both patients. The responsibilities of the circulation nurse are divided between two individuals. One nurse handles the needs related to the procedure and to the patient and ensures that sterile fields are maintained. The other nurse manages the flow of patients, with the focus on ensuring that the next patient is ready when the current patient is ready to leave the operating room; relays messages to the staff; schedules emergency procedures; and handles documentation. Once an operation has been completed, the operating room and floor are cleaned around the vacated area, which also is standard procedure in single-occupancy operating rooms. Positive-pressure airflow with twenty exchanges per hour is standard.

Patients
During the two and one-half years of the investigation, 2458 patients were entered into the study. The average age of the 1124 male patients was thirty-eight years (range, one to eighty-nine years), and the average age of the 1334 female patients was forty-three years (range, one to ninety-one years).

Each operative wound was classified as either clean (a non-traumatic, uninfected wound that was not inflamed and was not associated with a break in sterile technique) or clean-contaminated (a wound in which a drain had been inserted at the discretion of the surgeon)¹.

The operative procedures ranged from the simple removal of a pin to extensive nerve-grafting. The study did not include patients who had an emergency procedure, but it did include those who had an elective outpatient procedure in the same room and at the same time as an emergency procedure. As a matter of policy, no operation involving an obviously infected wound was performed at the same time as an elective operation that was expected to involve a clean or clean-contaminated wound. All patients who had infection at the time of an elective procedure were excluded from the study.

Of the 2458 patients, ninety-three (3.8 per cent) had an operation in the double-occupancy room without another patient in the room during the procedure and 2365 (96.2 per cent) had at least one other patient in the same room during a portion of the procedure. Only one other patient was treated in the same operating room during the time of the procedure for 1666 patients, whereas two patients or more were treated, at different times, during the procedure for 699 patients.

Collection of the Data
Information regarding the operating-room environment and the factors associated with the operation was documented for each patient at the time of the operation. Associated factors included the name of the surgeon; the type and duration of the procedure; the classification of the wound as either clean (no drains used) or clean-contaminated (drains used)¹; whether the procedure involved the use of implants or Kirschner wires, systemic administration of antibiotics at the time of the operation, irrigation of the wound with Ringer solution (with or without antibiotics), instillation of a topical steroid solution (betamethasone sodium) into the wound, or application of an antibiotic ointment (containing bacitracin zinc, neomycin sulphate, and polymyxin B) to the wound after closure; and the anesthesia class⁸. The anesthesia class was defined according to the system of the American Society of Anesthesiologists. Class 1 indicates that "the pathologic process for which operation is to be performed is localized and does not entail a systemic disturbance"; class 2, that there is "mild to moderate systemic disturbance caused either by the condition to be treated surgically or by pathophysiologic processes"; class 3, that there is "severe systemic disturbance or disease from whatever cause, even though it may not be possible to define the degree of disability with finality"; and class 4, that there are "severe systemic disorders that are already life-threatening, not always correctable by operation."⁸

The preferences of the surgeons differed with regard to systemic administration of antibiotics, irrigation of the wound (with or without antibiotics in the solution), instillation of a topical steroid solution, and application of an antibiotic ointment. These practices were not controlled or randomized but instead were the surgeon's choice, as the objective of the study was to document existing practices. The names of other patients who were operated on at the same time were noted on each patient's record, so that we could determine whether cross-contamination occurred or whether the rate of infection was influenced by the presence of other patients in the operating room.

The patients were evaluated periodically for a minimum of thirty days postoperatively. The attending surgeon examined the wound and, with use of the definitions established by the Centers for Disease Control¹¹, determined whether infection was present primarily on the basis of whether there was purulent drainage or whether wound erythema or swelling was beyond that anticipated for the type of procedure. It was also noted whether systemic antibiotic therapy was prescribed at any time during the thirty-day period. Ninety-five patients were contacted by telephone because follow-up data were missing or because the patient had not returned for an evaluation during the initial thirty days after the procedure. To ascertain if infection was present, a registered nurse asked these patients if there had been unusual redness, swelling, or increased temperature at the operative site, or if there had been any drainage from the wound. In addition, patients in whom an implant had been inserted were contacted one year after the procedure to determine whether infection had developed.

Analysis of the Data
A rate of infection was calculated for the entire study population as well as for each associated variable. The 95 per cent confidence intervals were calculated with use of the exact Wilson quadratic procedure. The rates were compared with and without the presence of factors, and differences were initially tested for significance with use of the Fisher exact test on all available data. Subsequently, logistic regression analysis¹⁶ was performed. Two surgeons had performed a procedure on only seventeen and twenty-four patients each, and these surgeons and patients were not included in the logistic regression analysis because the number of patients was too small to provide sufficient power to register a statistical difference. Therefore, the data for 2417 patients were included in the logistic regression analysis. The records of twenty-two patients were missing some information, leaving the records of 2395 patients available for the final analysis. The variable treatment parameters were first screened by univariate testing and were included in the multivariate logistic regression only if the initial p value was less than 0.25. Two groups were compared—patients who had infection and those who did not—with use of the t test for the continuous variables of operative time and age, the Mann-Whitney U test for the ordinal data of anesthesia class and the number of other patients who had been operated on during the time of the procedure, and the chi-square test for the other categorical data (the presence or absence of the treatment variable of interest).

A stepwise logistic regression analysis was carried out with SPSS for Windows software (SPSS, Chicago, Illinois). As many as fifty iterations were allowed for each step. Variables with the most significant findings were included first. The step program parameters were set to include a factor if p was less than = 0.15 and to exclude a factor if p was greater than = 0.20.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 

Over-All Rate of Wound Infection
Superficial or deep infection of the operative wound developed in thirty-seven of the 2458 patients, a rate of 1.5 per cent (95 per cent confidence interval, 1.1 to 2.1 per cent) (Table I). The infections were identified at a median of eleven days (range, one to thirty days). For two patients, the infection was diagnosed by a local physician.

Outcome for Patients Who Had Infection
Twenty-eight of the thirty-seven patients who had infection were managed with the oral administration of antibiotics, with three patients also being managed with parenteral antibiotic therapy. None of the twenty-eight patients had additional operative intervention. Two patients had the sutures removed earlier than anticipated; one of them also had a drain placed in the wound in the surgeon's office. This patient was considered to have deep infection of the wound, but operative débridement was not necessary.

Seven other patients who had deep infection needed additional operative intervention. Three patients had a simple incision and drainage, one patient needed five incision-and-drainage procedures, and one patient needed six such procedures. The latter patient had ablation of the involved digit at the level of the proximal interphalangeal joint because of chronic osteomyelitis. In the other two patients, the infection resolved after removal of an implant or a tissue expander.

Rates of Infection after the Five Most Common Procedures
The over-all rates of infection after the five most frequently performed operations were 1.9 per cent (ten of 517) after carpal tunnel release, 0.4 per cent (one of 224) after ganglionectomy, 1.9 per cent (four of 206) after operative procedures on a tendon, 1.9 per cent (three of 156) after synovectomy, and 0.7 per cent (one of 140) after trigger-finger release. Deep infection developed after two (0.4 per cent) of the carpal tunnel releases and after one (0.7 per cent) of the trigger-finger releases; no deep infection developed after the other three procedures.

Infection developed after five (4.1 per cent) of the 121 operative procedures that had been done on a lower extremity. We could not detect a significant difference between this rate and the rate associated with operative procedures that had been done on an upper extremity (thirty-two of 2337; 1.4 per cent) (p = 0.1).

Rates of Infection Associated with the Factors of Interest

Age and Gender
The rate of infection for male patients was 1.9 per cent (twenty-one of 1124), and the rate for female patients was 1.2 per cent (sixteen of 1334). Infection developed in fifteen (2.3 per cent) of the 648 male patients who were less than forty years old and in six (1.3 per cent) of the 476 who were forty years old or more. The corresponding rates for the female patients were 0.9 per cent (five of 567) and 1.4 per cent (eleven of 767). Chi-square analysis demonstrated no significant differences between the rates according to gender or according to age (as grouped) for either male or female patients. The factors of gender and age therefore were not included in the logistic regression analysis.

Number of Patients in the Operating Room
The rate of infection was 1.3 per cent (twenty-two of 1666) when one other patient had been treated in the operating room during the time of the procedure, 1.9 per cent (thirteen of 699) when at least two other patients had been treated in the operating room (at different times) during the procedure, and 2.2 per cent (two of ninety-three) when no other patient had been treated in the operating room during the procedure (Table II). With the numbers available, no significant difference could be detected among these rates. There was no trend for a higher rate of infection in association with a greater number of patients treated in the operating room during the time of the procedure. With a p value of 0.9, it was not even considered as a factor in the logistic regression model.

Anesthesia Class
The records for two patients did not contain information regarding the anesthesia class⁸. The differences in the rates of infection associated with the classes were small (Table II). The rate was 1.1 per cent (twelve of 1126) in association with anesthesia class 1, 1.9 per cent (twenty-one of 1082) in association with class 2, and 1.3 per cent (three of 224) in association with class 3. The rate associated with class 4 (4.2 per cent; one of twenty-four) was slightly higher, but with the small number of patients for study this difference could not be shown to be significant. According to the Mann-Whitney U test, the p value was low enough to be included in the logistic regression model. However, with the numbers available, we could not show the anesthesia class to be an important predictor of infection in this study.

Surgeon
The rates of infection varied among the nine surgeons (range, 0 [0 of 234] to 2.7 per cent [six of 223]), and comparisons of the rates for the seven who had performed more than twenty-four procedures revealed a significant difference among them (p = 0.003) (Table II). Each surgeon performed a variety of operations and also often used different intraoperative tactics, such as the instillation of a topical steroid solution, the systemic administration of antibiotics, and the application of an antibiotic ointment (Table III). Logistic regression analysis was used to determine if the surgeon-specific factor influenced the rate of infection.

Duration of the Operative Procedure
The median operative time was sixty minutes (range, five to 387 minutes) for the thirty-seven operations after which infection developed and forty-five minutes (range, two to 588 minutes) for the remaining 2421 operations. The over-all median operative time was forty-five minutes. Infection developed after ten (0.8 per cent) of the 1268 procedures that had lasted for forty-five minutes or less and after twenty-seven (2.3 per cent) of the 1190 procedures that had lasted for more than forty-five minutes. This difference was significant (p = 0.002), and the factor of operative time was included in the logistic regression analysis.

Use of an Implant
Of the thirty-seven patients in whom infection developed, three (8 per cent) had an implant. Thus, the rate of infection for the patients in whom an implant had been inserted was 2.5 per cent (three of 119). With the numbers available, we could not determine this rate to be significantly different (p = 0.36) from the rate for the remaining patients (1.5 per cent; thirty-four of 2339) (Table II), and it was therefore not considered a factor in the subsequent logistic regression analysis. The rate of deep infection associated with the use of an implant was 1.7 per cent (two of 119). The implant was removed if the patient did not respond to treatment with antibiotics. No additional infections were found during the one-year follow-up of these patients.

Use of Kirschner Wires
Kirschner wires had been used in eight (22 per cent) of the thirty-seven patients in whom infection developed. Thus, the rate of infection for the patients in whom such wires had been used was 2.9 per cent (eight of 276). This rate was not significantly different (p = 0.08) from the rate for the remaining patients (1.3 per cent; twenty-nine of 2182) (Table II). The patients in whom infection developed were managed with the oral administration of antibiotics, and the wires were not removed until the fracture was considered to be healed. The use of Kirschner wires was entered as a factor into the logistic regression analysis because the initial p value of 0.07 was less than the required 0.25. However, we could not determine this factor to be an important predictor of infection.

Classification of the Wound (with or without a Drain)
Postoperative infection developed in thirty (1.3 per cent) of the 2311 clean wounds (those that did not have a drain) and in seven (4.8 per cent) of the 147 clean-contaminated wounds (those that had a drain). This difference was significant (p = 0.001). Classification of the wound was entered as a factor into the logistic regression analysis and was found to be a significant contributing factor to infection.

Application of an Antibiotic Ointment
One operative record had no information regarding the use of an antibiotic ointment. Infection developed in thirty-four (2 per cent) of the 1741 wounds that had been treated with an antibiotic ointment and in three (0.4 per cent) of the 716 that had not been so treated. This difference was significant (p = 0.002) (Table II). The application of an antibiotic ointment thus appeared to be a negative predictor of infection. However, when it was included as a factor in the logistic regression analysis with all of the other factors likely to influence infection, it appeared to have been confounded with another factor and was not found to be a significant contributor to the rate of infection.

Systemic Administration of Antibiotics
The records for three patients did not contain information regarding the systemic administration of antibiotics at the time of the operation. Infection developed in fifteen (2.6 per cent) of the 571 patients who had received systemic antibiotic therapy and in twenty-two (1.2 per cent) of the 1884 patients who had not received such therapy. This difference was significant (p = 0.002). However, as with application of an antibiotic ointment, logistic regression analysis showed that the systemic administration of antibiotics was not a predictor of infection.

Irrigation of the Wound
Infection developed in twenty-one (1.3 per cent) of the 1576 wounds that had been irrigated with Ringer solution (with or without antibiotics) and in sixteen (1.8 per cent) of the 882 that had not been so irrigated (Table II). We could not show this difference to be significant (p = 0.34). Twelve records did not contain information regarding the type of irrigation solution that had been used (whether or not it contained antibiotics). Infection developed in four (2.2 per cent) of the 179 wounds that had been irrigated with a solution that contained antibiotics and in seventeen (1.2 per cent) of the 1385 wounds that had been irrigated with a solution that did not contain antibiotics (Table II). With the numbers available, this difference could not be shown to be significant (p = 0.32). Irrigation of the wound was not included as a factor in the logistic regression analysis.

Instillation of a Topical Steroid Solution
Twenty-three operative records contained no information regarding the instillation of a topical steroid solution (betamethasone). Infection developed in four (0.6 per cent) of the 662 wounds that had been treated with the steroid solution and in thirty-three (1.9 per cent) of the 1773 wounds that had not been so treated (Table II). This difference was significant (p = 0.02). Once the confounding factors were considered in the logistic regression analysis, the instillation of a topical steroid solution still apparently influenced the rate of infection. It should be noted that the steroid solution had not been instilled into any of the wounds in which deep infection subsequently developed.

Stepwise Logistic Regression Analysis
The stepwise logistic regression procedure was used to determine which of the nine factors best predicted postoperative infection. These factors included the patient's gender, the anesthesia class⁸, the surgeon, the operative time, the use of Kirschner wires, the classification of the wound (use of a drain), the application of an antibiotic ointment, the systemic administration of antibiotics, and the instillation of a topical steroid solution (Table IV). The regression was performed with use of -to-enter of 0.15 and -to-remove of 0.20 in order to admit a large number of candidate predictors.

The final significant predictive factors were (1) the classification of the wound (use of a drain), which had a logistic regression coefficient of 0.9943, a standard error of 0.4442, a new p value of 0.006 (as determined by chi-square analysis after the factor was entered into the model), and an odds ratio of 2.7029; (2) the instillation of a topical steroid solution, which had a logistic regression coefficient of -1.5923, a standard error of 0.5750, a new p value of 0.04, and an odds ratio of 0.2034; and (3) the surgeon, which had a new p value of 0.02 (the logistic regression coefficient, standard error, and odds ratio were not applicable). These three factors were the dominant statistically important factors once all confounding factors had been considered. The odds ratio suggested that the odds of infection developing were 2.7 times greater if a drain had been used than if it had not. The instillation of a topical steroid solution reduced the rate of infection with an odds ratio of 0.2, but the statistical strength was not as strong as that for the use of a drain.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
The number of outpatient operations performed on the hand has been increasing in recent years. To improve the utilization of our operating-room facilities, we designed one room to accommodate two separate operating teams at the same time. This unconventional method warranted effective surveillance to assess the rate of postoperative infection that was associated with the use of this room. The present study showed that no cross-contamination occurred in 2365 patients who had been operated on in a room in which at least one other patient had been treated during the time of the procedure. It has been suggested that the prevalence of infection is related to the amount of activity in an operating room¹⁹. This was not evident in our study, as the infection rate for the patients who had been operated on with another patient in the room was not significantly higher than that for the patients who had been operated on with no other patient in the room. This finding supports the use of a double-occupancy operating room for elective operations on hands without infection. A double-occupancy operating room allows the optimum use of available equipment, personnel, and space.

The over-all rate of infection of the wound after an outpatient operation on an extremity was 1.5 per cent (thirty-seven of 2458). To our knowledge, the only comparable study of infections of the wound after similar operations was by Hanssen et al., who reported a rate of deep infection of 0.47 per cent (seventeen of 3620) after release of the median nerve in the carpal canal. The rate of deep infection associated with this operative procedure in our study was slightly lower (0.39 per cent; two of 517). Our over-all results appear to be comparable with the findings in other reports on less specific orthopaedic and plastic surgery procedures^6,7,21.

Some investigators have assessed the rate of infection according to risk factors such as the severity of the clinical problem, the condition of the patient, or the duration of the operative procedure^7,9,13,15. Although there was a trend for an increased rate of infection in association with each of those factors in our study, the logistic regression analysis did not support the influence of the combination of factors as reported elsewhere⁷. Of the nine factors analyzed as possible predictors of infection, only three proved to be significant: the instillation of a topical steroid solution (betamethasone), which was associated with a lower rate of infection (p = 0.04); the particular surgeon who had performed the operation (p = 0.02); and the use of a drain, which was associated with a higher rate of infection (p = 0.006). Hanssen et al. reported a greater likelihood of infection after the use of a drain or when the operative time had been longer than thirty-five minutes.

Obviously, many factors may influence the rate of infection, not the least of which is whether the infections are accurately diagnosed. Surgeons may tend to underdiagnose infections in patients in whom they have performed the operation. Patients who did not have postoperative infection were given antibiotics very rarely, for such reasons as a foreign-body reaction or the presence of diabetes.

Another major influence on the rate of infection is the effect of known surveillance^5,6. The results of this study were not disclosed to the surgeons before the study had been completed, as knowledge of the findings may have changed a surgeon's practice and thus the rate of infection after that surgeon's operations. No interim reports were given to the participants in the study, except at approximately three months before the ending date of the study. At that time, the surgeons were told that the over-all rate of infection was 1.5 per cent, which was the final result as well, but differences among the rates were not disclosed.

The operative procedures that were performed in this study were done for a wide spectrum of clinical diagnoses. This may have introduced some variability in the rate of infection related to the duration of operative exposure, operative tactics, and necessary equipment. Other than determining the rates associated with the five most common procedures, we did not analyze the rate according to procedure, except as it related to use of a drain or to whether implants or Kirschner wires had been used.

Previous studies have documented that the duration of the operative procedure is a risk factor for postoperative infection of the wound^2,5-7,20. In the present study, the risk of infection seemed to be increased after operations that had lasted for more than forty-five minutes. In addition, a higher rate of infection was associated with surgeons who, on the average, took longer to perform operations. Despite this agreement with previously reported findings, there was no significant relationship between the operative time and the rate of infection once the influence of confounding factors had been eliminated through logistic regression analysis. Our failure to find a relationship may be due to the relatively short duration of the procedures in the current study compared with that for other operative specialties.

Five hundred and seventy-one (23 per cent) of the patients in our study were managed prophylactically with systemic antibiotic therapy at the time of the operation. Because of selection bias (the surgeon prescribing antibiotics for a patient who is at greater risk for infection), it is not meaningful to compare the rates of infection between the patients who were and those who were not managed with systemic antibiotic therapy. This effect has been identified in previous reports^3,6. However, if the association between systemic antibiotic therapy and a higher rate of infection in the present study was real, it might be partially explained by the fact that, at the time of the study, standard protocol dictated that, when antibiotics were given, they were administered only during or after the operation. It has since been reported that systemic antibiotic therapy is effective only when it is given before the operation⁴. Thus, it is possible that antibiotics were prescribed for patients who were at greater risk for infection, or when there had been a minor breakdown in sterile technique, and that the antibiotics were less effective because they were administered during or after the operation. In any event, the systemic administration of antibiotics was not considered to be an important factor, as determined by logistic regression analysis.

The lack of influence of irrigation of the wound with Ringer solution (with or without antibiotics) suggests that irrigation may not be especially useful for the prevention of infection. Indeed, the use of antibiotics in the irrigation solution for patients who had a clean or clean-contaminated wound may be questioned. Similarly, the application of an antibiotic ointment to help to prevent infection was not seen as particularly useful in these patients.

The positive influence of instillation of a topical steroid solution (betamethasone) was noteworthy. Not all surgeons routinely employed this solution; the rate of usage ranged from 0.9 per cent (two of 233 procedures) to 65 per cent (144 of 221 procedures). The positive influence was even more surprising because the patients managed by the surgeon who used it the least also had the lowest rate of infection (Table III). For the use of this solution to be statistically related to the rate of infection, as determined by logistic regression analysis, it had to be relatively independent of the other factors of interest. It is apparent that any factor related to the operative technique was independent of the use of the steroid solution. A topical steroid solution is used occasionally by some surgeons in our practice because two surgeons in our group had noted that it reduced postoperative swelling and pain. The solution is applied directly to the exposed tissue, before the tourniquet is removed. Once hemostasis has been achieved, the wound is closed; thus, the tissue is not left flooded with the steroid solution. The findings of our study support the use of this topical steroid solution, in contrast with the conclusion of Hanssen et al. However, in that study, all of the patients in whom infection developed after the use of a topical steroid solution had had a synovectomy and placement of a drain¹⁴. In our study, the use of a drain was associated with a higher rate of infection. Thus, the infections in the study of Hanssen et al. may have been related to the use of a drain rather than to the use of a topical steroid solution. Additional studies are needed to determine whether a topical steroid solution is beneficial or detrimental to the healing process in this population of patients.

The fact that the surgeon had a strong influence on the rate of infection is important. (Again, the operative time initially seemed to be important but was not revealed to be significant by logistic regression analysis. Thus, any effect of operative time may be due to other confounding factors.) The surgeon's influence was not necessarily due to any of the factors that we evaluated. The relevant factors still must be identified. It would be useful to observe techniques more closely and also to consider other variables, such as the usual operative team.

The strong association between the use of a drain and postoperative infection in this study suggests that the use of a drain needs to be re-examined. Open drains, such as those used in the present study, were associated with higher rates of infection in a previous study⁶. The rationale for their use was that the wound heals better if fluid is allowed to drain. Thus, if the drains had not been used, the wounds may not have healed as quickly and the rate of complications may have been higher. In addition, the rationale for the use of the drain may have been the extent of the wound; therefore, the higher rate of infection may be related to the condition of the wound rather than to the presence of the drain. The type and use of drains should be the subject of another study.

In conclusion, we determined that the over-all rate of infection of the wound after an outpatient operation on an extremity was approximately 1.5 per cent (thirty-seven of 2458). The rate of deep infection was even lower (0.3 per cent; eight of 2458). The factors that were most strongly associated with the rate of infection were, in order of importance, the use of a drain and the individual surgeon. The instillation of a topical steroid solution (betamethasone) into the wound was associated with a lower rate of infection. Some commonly held views on the value of irrigating the wound with a solution containing antibiotics or of administering systemic antibiotic therapy are not supported by the findings of the present study. The use of a double-occupancy operating room did not have a negative influence on the rate of infection.

NOTE: The authors thank the nursing staffs of the Jewish Hospital Outpatient Care Center and the Kleinert, Kutz, and Associates Hand Care Center, P.L.L.C.

	Footnotes

 

*Although none of the authors has received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article, benefits have been or will be received, but are directed solely to a research fund, foundation, educational institution, or other non-profit organization with which one or more of the authors is associated. Funds were received in total or partial support of the research or clinical study presented in this article. The funding source was Jewish Hospital, Louisville, Kentucky.

Christine M. Kleinert Institute for Hand and Micro Surgery, One Medical Center Plaza, Suite 850, 225 Abraham Flexner Way, Louisville, Kentucky 40202.

Outpatient Care Center, Jewish Hospital, 217 East Chestnut Street, Louisville, Kentucky 40202.

Section of Hand Surgery, U9 4131, Rigshospitalet, University of Copenhagen, Blegdamsvej 9 2100, Copenhagen, Denmark.

¶University of Louisville, Health Sciences Campus, Louisville, Kentucky 40292.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Altemeier, W. A.: Surgical infections: incisional wounds. In Hospital Infections, pp. 287-306. Edited by J. V. Bennett and P. S. Brachman. Boston, Little, Brown 1979.

2. Bhattacharyya, N., and Kosloske, A. M.: Postoperative wound infection in pediatric surgical patients: a study of 676 infants and children. J. Pediat. Surg., 25: 125-129, 1990.[Medline]

3. Bremmelgaard, A.; Raahave, D.; Beier-Holgersen, R.; Pedersen, J. V.; Andersen, S.; and Sorensen, A. I.: Computer-aided surveillance of surgical infections and identification of risk factors. J. Hosp. Infect., 13: 1-18, 1989.[Medline]

4. Classen, D. C.; Evans, R. S.; Pestotnik, S. L.; Horn, S. D.; Menlove, R. L.; and Burke, J. P.: The timing of prophylactic administration of antibiotics and the risk of surgical-wound infection. New England J. Med., 326: 281-286, 1992.[Abstract]

5. Cruse, P.: Wound infection surveillance. Rev. Infect. Dis., 3: 734-737, 1981.[Medline]

6. Cruse, P. J., and Foord, R.: The epidemiology of wound infection: a 10-year prospective study of 62,939 wounds. Surg. Clin. North America, 60: 27-40, 1980.[Medline]

7. Culver, D. H.; Horan, T. C.; Gaynes, R. P.; Martone, W. J.; Jarvis, W. R.; Emori, T. G.; Banerjee, S. N.; Edwards, J. R.; Tolson, J. S.; Henderson, T. S.; and Hughes, J. M.: Surgical wound infection rates by wound class, operative procedure, and patient risk index. Am. J. Med., 91: 152S-157S, 1991.

8. Dripps, R. D.; Eckenhoff, J. E.; and Vandam, L. D.: Introduction to Anesthesia. The Principles of Safe Practice. Ed. 7. Philadelphia, W. B. Saunders 1988.

9. Ehrenkranz, N. J.: Surgical wound infection occurrence in clean operations. Risk stratification for interhospital comparison. Am. J. Med., 70: 909-914, 1981.[Medline]

10. Evans, R. S.; Larsen, R. A.; Burke, J.P.; Gardner, R. M.; Meier, F. A.; Jacobson, J. A.; Conti, M. T.; Jacobson, J. T.; and Hulse, R. K.: Computer surveillance of hospital-acquired infections and antibiotic use. J. Am. Med. Assn., 256: 1007-1011, 1986.[Abstract/Free Full Text]

11. Garner, J. S.; Jarvis, W. R.; Emori, T. G.; Horan, T. C.; and Hughes, J. M.: CDC definitions for nosocomial infections. Am. J. Infect. Control, 16: 128-140, 1988.[Medline]

12. Gaynes, R. P.; Culver, D. H.; Emori, T. G.; Horan, T. C.; Banerjee, S. N.; Edwards, J. R.; Jarvis, W. R.; Tolson, J. S.; Henderson, T. S.; Hughes, J. M.; and Martone, W. J.: The National Nosocomial Infections Surveillance System: plans for the 1990s and beyond. Am. J. Med., 91: 116S-120S, 1991.

13. Haley, R. W.: Nosocomial infections in surgical patients: developing valid measures of intrinsic patient risk. Am. J. Med., 91 (Supplement 3B): 145S-151S, 1991.

14. Hanssen, A. D.; Amadio, P. C.; DeSilva, S. P.; and Ilstrup, D. M.: Deep postoperative wound infection after carpal tunnel release. J. Hand Surg., 14A: 869-873, 1989.

15. Hooton, T. M.; Haley, R. W.; Culver, D. H.; White, J. W.; Morgan, W. M.; and Carroll, R. J.: The joint associations of multiple risk factors with the occurrence of nosocomial infection. Am. J. Med., 70: 960-970, 1981.[Medline]

16. Hosmer, D. W., and Lemeshow, S.: Applied Logistic Regression. New York, Wiley, 1989.

17. National Nosocomial Infections Surveillance System: Nosocomial infection rates for interhospital comparison: limitations and possible solutions. Infect. Control and Hosp. Epidemiol, 12: 609-621, 1991.[Medline]

18. Nichols, R. L.: Surgical wound infection. Am. J. Med., 91 (Supplement 3B): 54S-64S, 1991.

19. Ritter, M. A., and Marmion, P.: Exogenous sources and controls of microorganisms in operating room. Orthop. Rev., 16: 224-232, 1987.[Medline]

20. Shapiro, M.; Munoz, A.; Tager, I. B.; Schoenbaum, S. C.; and Polk, B. F.: Risk factor for infection at the operative site after abdominal or vaginal hysterectomy. New England J. Med., 307: 1661-1666, 1982.[Abstract]

21. Taylor, G. J. S.; Bannister, G. C.; and Colder, S.: Perioperative wound infection in elective orthopedic surgery. J. Hosp. Infect., 16: 241-247, 1990.[Medline]

CiteULike

Connotea

Del.icio.us

Facebook

Technorati

Twitter

This article has been cited by other articles:

				L. Prokuski Prophylactic Antibiotics in Orthopaedic Surgery J. Am. Acad. Ortho. Surg., May 1, 2008; 16(5): 283 - 293. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Free Letters to the Editor: Submit a response Alert me when this article is cited Alert me when Letters to the Editor are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My File Cabinet Download to citation manager Rights and Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Sign up for keyword alerts Google Scholar Articles by KLEINERT, J. M. Articles by FIRRELL, J. C. Search for Related Content PubMed PubMed Citation Articles by KLEINERT, J. M. Articles by FIRRELL, J. C. Social Bookmarking                   What's this?

Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online.
Copyright © 1997 by The Journal of Bone and Joint Surgery, Inc. Online ISSN: 1535-1386.
The Journal of Bone and Joint Surgery®, JBJS®, JB&JS®, and eJBJS® are registered in the U.S. Patent and Trademark Office.