Controversies in Reduction Mammoplasty: Being A “Clean” Operation,Does It Mandate Antibiotic Prophylaxis?

Patients and Methods

This study was conducted at the Hospital Agamenon Magalhães in Recife, Brazil, with the approval of the Research Ethics Committee (CAAE 33103614.9.0000.5197). All patients provided informed consent. The study was randomized, double-blinded, and interventional and was registered in the Brazilian Registry of Clinical Trials (Rebec). The patients were older than 18 years and had been approved for bilateral reduction mammoplasty by the medical board of the hospital. They had a body mass index (BMI) below 28 kg/m², did not smoke or have diabetes mellitus, did not have breast implants, and had not received antibiotics in the 15 days preceding the mammoplasty.

Two groups were created by a computerized randomization table generated by random.org, with 75 patients in each group. Group 1 received prophylactic antibiotic (cefazolin 2 g in 100 mL of 0.9% saline intravenously 30 min before the start of surgery and reapplication of 1 g every 3 h until the end of surgery. Group 2 (control subjects) received no antibiotic.

Routine pre-operative tests were blood studies (hemoglobin >12 g/dL, glycemia <100 mg/dL), cardiac evaluation, and mammary imaging (ultrasonography/mammography). The patients received a regular soap shower, and the skin was prepared in the operating room with a chlorhexidine 2% solution and operated on by the same surgical team. The procedure was performed with the patient under general anesthesia and consisted predominantly of operation by the technique described by Pitanguy [9] (scar inverted T) with approximately 24 h of hospitalization. The operating room temperature was set to 21°C. No air warming device was available to help control body temperature. Sutures made of nylon were inserted in two planes (glandular and subcutaneous) along with polyglecaprone intradermal suture. Occlusive dressings were placed with microporous tape.

Patients were followed up by a member of the research team who did not know to which group the patient had been assigned. Dressing change were performed on the third and seventh post-operative days (PODs), and outpatient revaluations were conducted weekly until the 30th POD for identification of SSI, which was characterized, according to the CDC standards [10], by pain, heat, swelling, redness, purulent discharge from the incision, or isolation of microorganisms in fluids or tissues obtained aseptically.

Patient information was collected, namely age, co-morbidities, degree of ptosis and breast enlargement according to the Regnault classification [11], surgical technique, duration of surgery, resected breast weight, post-operative complications, and any infection with its treatment, and the outcome of the case. Data were analyzed descriptively by absolute and percentage frequencies for categorical variables. The measures were expressed as mean, standard deviation, and median for numerical variables and analyzed inferentially by the Mann-Whitney test because of the lack of data normality. Normality of the data had been checked by performing the Shapiro-Wilk test to compare the patients who had infection with those who did not. The Pearson χ² test was used to verify the presence of a significant association between the presence of infection and categorical variables. In the event that the conditions for using the χ² test were not present, the Fisher exact test was used.

The margin of error used in the statistical tests was 5%. Data were tabulated in a Microsoft Excel spreadsheet, and the software used for statistical calculations was SPSS version 21.

Results

Five patients were excluded from the analysis because of antibiotic use post-operatively for dehiscence of the incision (n = 2), hematoma (n = 1), tonsillitis (n = 1), or areola papillary complex necrosis (n = 1). No SSIs were found in these patients.

There was no significant difference in the age, ptosis, hypertrophy, surgery duration, or resected volume in the groups with or without antibiotic prophylaxis (Table 1). There were 13 cases of SSI, 3 (4.1%) in Group 1 and 10 (13.9%) in Group 2 (p = 0.039). In such cases, the diagnosis was made in the first and second week in 12 patients; only one case of SSI was diagnosed in the third week. All of the secretion cultures in these cases were positive for Staphylococcus aureus. Patients were treated with secretion drainage, antibiotic therapy with cephalexin for seven d, and daily dressing changes.

There were more cases of SSI in the nine patients with resections of greater than 1201g, but the difference from the cases with lesser resections was not statistically significant. Age, BMI, and surgical technique were not decisive for the occurrence of infection. Patients in Group 1 with SSIs had a longer surgery duration and less infection than Group 2 (Table 2).

Discussion

Breast reduction surgery is classified as a clean operation by the CDC definition because there is no prior infection; no penetration of the respiratory, gastrointestinal, or urinary tract; and primary suture closure. Therefore, the operation should not require antibiotic prophylaxis [12]. However, in our study, the incidence of SSI in patients who did not receive prophylaxis was 9%, well above the arbitrary value for clean surgery (<3.4%) but consistent with the findiings of Veiga Filho et al. (8%) [7], Ahmadi et al. (24%) [13], and Hardwicke et al. [3]. The last-mentioned article, published in 2013, found an overall SSI rate of 11.9% in the studies reviewed. The surgical conditions, such as skin preparation, room temperature, and staff expertise, were similar when these conditions were described. Despite the shorter surgery time, the patients not receiving antibiotic had a statiscally significantly higher infection rate (p = 0.049).

The use of a cephalosporin (cefazolin) was chosen for prophylaxis in view of the organisms usually found in breast surgery, the most common being S. aureus [3,12]. Its minimum inhibitory concentration is 8 mcg/mL and with bolus administration, ensures double the minimum inhibitory concentration in the serum until the third hour, when repeat administration is advisable in view of the drug's half-life [14]. The benefits of administration as a continuous infusion remain unclear.

The infection rates in this study (Group 1 4.1% and Group 2 13.9%) demonstrate the value of antibiotic prophylaxis (p = 0.039), with the number of patients to be treated to reduce the risk of one infection being 10.2. The comparison with the literature shows a dearth of randomized controlled trials, inadequate description of the exclusion criteria and withdrawals, as well as failure of the masking mechanism (double-blinding) according to the Jadad criteria for reliability [15]. The diversity of antibiotic regimens used, which include amoxicillin, azithromycin, levofloxacin, and clindamycin, as well as non-systematic administration, with research being done before, during, and as many as seven d after surgery [1, 3, 6, 7, 13, 16, 17], also significantly affect the comparisons.

Every effort to reduce the risk of SSIs is valid in plastic surgery, as well as the avoiding the need for hospitalization and the costs of SSI (estimated to be an additional U.S.$4,091 per patient in 2004 by Platt and Olsen [5, 18], Many times, this complication has a decisive influence on the outcome of the scar and surgery as a whole. However, one cannot place all responsibility for changes in the incision, which can range from small necrosis (mainly at the confluence of the inverted T) to dehiscence and consequent unsightly scars, on prophylactic antibiotics.

Conclusion

Reduction mammoplasty had a 13.9% incidence of SSI in this study, much higher than would be expected from an operation categorized as clean. In our study, antibiotic prophylaxis produced a statistically significant reduction in SSIs, to 4.1%.