Orthopedics

Materials and Methods

A retrospective review was performed of 645 consecutive elective hand surgery patients from 3 surgical centers in the Greater Philadelphia region between 2007 and 2010. Full Institutional Review Board approval was obtained before data collection. Patient records were collected according to International Classification of Diseases-9 codes for the most common soft tissue elective hand surgeries during this time, which included carpal tunnel release, mass excision, trigger finger release, and first dorsal compartment release. Mass excision included resection of the following lesions: ganglion cyst, mucous cyst, epidermal inclusion cyst, lipoma, glomus tumor, giant cell tumor of tendon sheath, neuroma, and granuloma. Soft tissue procedures concomitantly performed with a joint or bone procedure or any procedure involving an implant were excluded; nerve and tendon repairs and reconstructions were also excluded because these operations are routinely administered prophylaxis. Any surgery involving incision into a prior operative site (reoperations) were excluded to prevent confounding. Charts with incomplete medical records were excluded in the final statistical analysis (n=45).

Patient demographic data and known risk factors for infection were recorded. The use or exclusion of an antibiotic agent, the type of antibiotic agent, the type of procedure, and the development of subsequent infection were additionally noted. All patients were prescrubbed with a 4% chlorhexadine gluconate sponge diluted in 500 mL of sterile water. Surgical sites were then dried with sterile towels and draped using a sterile technique. Cefazolin (n=187) was considered the first-line agent if prophylaxis was chosen; if cephalosporins were contraindicated, vancomycin (n=5) or clindamycin (n=20) was selected. Antibiotics were administered within 1 hour of incision; however, they were given at the discretion of the attending hand surgeon. All surgical wounds were irrigated with 0.9% saline prior to closure, approximated with nylon sutures, and covered with nonadherent sterile dressings. Absorbable sutures were not used to close wounds.

Patients were followed up at 2- and 4-week intervals after the index operation. Because surgical wound infection is most likely to present during this time,^8,16 the search was limited to all records within the 30-day postoperative period. Office visit, inpatient, and emergency room records were reviewed for the presence of superficial or deep infection. Superficial and deep infections were defined by Centers for Disease Control criteria (Table 2).⁸ Superficial infections were treated with intravenous or oral antibiotic therapy, and deep infections were surgically drained or debrided.

Table 2: Centers for Disease Control Criteria for Surgical Site Infection

Statistical hypothesis testing was performed via Student’s t test for age and Fisher’s exact test for all other variables. Statistical significance was defined as P<.05.

Results

Complete operative and follow-up records were available for 600 patients. Forty-five charts did not have complete follow-up records and were excluded. The overall infection rate was 0.66%. The patients were separated into groups based on the administration of a prophylactic antibiotic agent (Table 3). Antibiotic prophylaxis was administered in 212 patients, and 388 patients did not receive prophylaxis. Mean patient age was 54 years (range, 16–95 years). Sex, diabetes mellitus, and smoking status were not significantly different between the groups. Superficial infection was found in 0.77% of patients who did not receive prophylaxis and 0.47% in those who did receive it; this difference was not statistically significant. No deep infections were identified in either group.

Table 3: Overall Comparisons of Antibiotic Prophylaxis

When sorting the frequency of superficial infection as a function of procedure, infections were found in 1% of carpal tunnel releases and 0.58% of trigger finger releases. No infections occurred among the first dorsal compartment releases or mass excisions (Table 4). Demographics of patients who became infected are listed in Table 5; 3 of the 4 patients did not receive antibiotics, 1 was a diabetic, and none were smokers.

Table 4: Infection Rates by Procedure

Table 5: Cases of Surgical Site Infection

Discussion

Numerous reports have evaluated the value of antibiotic prophylaxis in orthopedic surgery, but most of these studies have only included elective hand cases as a subset in the study design.^2,6,10–15 Thus, because a paucity of data exists, the authors of the current study aimed to expand the evidence on this topic with a specific focus on elective soft tissue hand surgery using postoperative infection as a primary outcome measure.

This study was limited by the retrospective design, which reduced the numbers of available complete records; as a result, many patients were excluded and some infection cases may have been treated at alternative centers. Second, the antibiotics were prescribed nonrandomly at the discretion of the attending hand surgeon. Third, the rates of infection were so low that individual risk factors could not be assessed; a multivariate analysis was attempted but could not be constructed with only 4 infections. Last, the potential for type II errors existed because the low rates of infection would require thousands of patients to achieve the standard 80% power. Defining significance in hypothesis testing for studies with close, low percentages is difficult; several authors have commented on this quandary by saying, “the efficacy of such prophylaxis in surgical procedures that have a low infection rate is nearly impossible to prove.”⁷

Whether ideal statistical testing in this setting can be achieved, the current study demonstrates a tangible clinical significance: The frequency of infection following soft tissue hand surgery is low. The overall rate was 0.6%; all of these were considered superficial, and none required surgical intervention. No statistically significant differences were detected in infection rates between patients who received antibiotic prophylaxis and those who did not. These findings are consistent with previously published reports. Harness et al⁸ reported a double-blinded, randomized, controlled trial of 2336 carpal tunnel releases with an overall infection rate of 0.36%; they did not find that infection rates were significantly higher when bypassing antibiotics. Similarly, Hanssen et al⁷ retrospectively reviewed 3620 carpal tunnel releases and found an infection rate of 0.47%; 15 of the 3620 patients were administered antibiotics preoperatively. Platt and Page¹³ prospectively evaluated 249 elective and emergency hand surgery cases for risk factors associated with postoperative infection. In 112 elective cases, infection rates were 8% and 12% for those with and without prophylaxis, respectively; this difference was not statistically significant. However, shortcomings of the study were low sample size, an unexplained high infection rate, and a nonrandomized design.¹³

Despite the shortcomings in the literature, most reports have demonstrated a low rate of infection following elective hand surgery, and the clinical significance of these rates becomes more relevant when juxtaposed against rates of complications of antibiotics. The prevalence of drug-resistant hand infections is increasing in communities nationwide, and bacterial resistance to antibiotics has been reported after short prophylactic dosing.^14,17,18 Furthermore, the incidence of postoperative pseudomembranous colitis has been reported to be as high as 6%.¹⁸ A risk-to-benefit analysis in this regard would provide an interesting implication for future trials because rates of postoperative infection and antibiotic complications in a hand surgery population have yet to be examined concurrently.

Notwithstanding that the literature has yet to define specific indications for prophylaxis in the hand population, surgical centers must have increasing awareness of this issue because physician reimbursements are now potentially affected by compliance with the SCIP measures.⁵ Furthermore, as insurance companies increasingly threaten to withhold reimbursements for complications such as surgical site infections, additional data will be required to guide these policies.

Conclusion

The role of antibiotic prophylaxis in hand surgery remains controversial. Some authors have suggested that this controversy exists because of convention, varying definitions of superficial and deep infections, lack of randomized, controlled trials, and difficulty in statistical testing. The current study suggests that antibiotics added neither clinical nor statistical protection from surgical site infection for elective soft tissue hand surgery. However, a large randomized study or meta-analysis is needed to form more definitive guidelines.