Surgical site infections are a common cause of morbidity and mortality, occurring in up to 25% of patients undergoing abdominal surgery.1 They increase hospital stay and can cost up to $10 billion annually.2 The true incidence of wound infections is likely underestimated as a result of delayed infections after hospital discharge. Cesarean delivery is the most common major surgery in the United States3 and contributes significantly to this burden.4 Cesarean delivery rates have increased 60% since 19965 and currently account for approximately one third of all deliveries.
Precesarean prophylactic antibiotics reduce the incidence of surgical site infection.4,6 The antibiotic must be active against common surgical site flora and given in time to reach adequate tissue concentrations, which must be maintained for the duration of the surgery. Pregnant women have an increased volume of distribution and glomerular filtration rates, which affects drug concentration in different body compartments. Obese pregnant women have an increased volume of distribution compared with lean women, which further affects drug delivery and increases risk for surgical site infection, even with antibiotic prophylaxis.1
Several national societies7 including the American College of Obstetricians and Gynecologists currently recommend preincision dosing with a single antibiotic with increased dosing for obese women.8 There is, however, a paucity of data supporting modified dosing for obese women. Therefore, our objective was to compare adipose tissue cefazolin concentrations of obese women undergoing cesarean delivery who were randomized to receive 2 g or 3 g of cefazolin prophylaxis with the hypothesis that 3 g would attain significantly higher adipose tissue cefazolin concentration.
MATERIALS AND METHODS
This study was conducted at a single institution, Women & Infants Hospital in Providence, Rhode Island, with institutional board approval (Protocol # 13-0004). Inclusion criteria included a body mass index (BMI, calculated as weight (kg)/[height (m)]2) of 30 or more at the first prenatal visit, gestational age of at least 37 weeks, and singleton gestation. Patients were not eligible if they had pregestational diabetes mellitus and renal disease (because of the potential for microvascular disease, which can alter the pharmacokinetics of cefazolin), a known cephalosporin allergy or a severe penicillin allergy, suspected chorioamnionitis, exposure to antibiotics in the 7 days preceding the cesarean delivery, or an emergent indication for cesarean delivery. Renal disease was defined as a known pre-existing condition or a first-trimester creatinine level greater than 0.7 mg/dL.
After written informed consent was obtained, eligible participants were randomly allocated to receive either 2 or 3 g of cefazolin in 100 mL of normal saline administered as an intravenous bolus no more than 60 minutes before the start of the surgery. Permuted block randomization with varying block sizes was used to stratify randomization by patient BMI: 30–39.9 and greater than 40. The randomization list was managed and maintained by the pharmacy staff. All other investigators, study participants, operating room staff, and anesthesiologists were blinded to group allocation.
Intraoperatively, two adipose tissue samples were obtained. The first sample was collected after skin incision but before fascial incision. The second sample was obtained after fascial closure, before skin closure. A secondary, optional part of the study was the measurement of serum concentrations of cefazolin throughout the cesarean delivery. With consent, a second intravenous catheter was placed before the start of the surgery for this purpose. Three blood samples were taken from this intravenous catheter throughout the surgery for serum analysis: the first was taken at the end of antibiotic administration and the second and third samples were obtained at the times of the adipose samples.
Blood samples were allowed to clot for 10 minutes and then centrifuged at 3,200 rpm for 10 minutes. The serum was aliquoted and then stored at −80°C. The adipose tissue was weighed, rinsed, reweighed, and stored at −80°C. Cefazolin concentrations in serum and tissue were measured using a validated high-pressure liquid chromatography method.9 The serum assay was linear over a range of 0.5–50 micrograms/mL (R2=0.999). Intraday (n=10) coefficients of variation for the low (1 microgram/mL) and high (40 micrograms/mL) serum quality control samples were 3.6% and 3.4%, respectively. Before analysis, tissue samples were homogenized in normal saline and cefazolin was extracted as per the method utilized for serum processing. Cefazolin tissue samples were run with a cross-matrix validation on the serum standard curve. Interday (n=8) coefficients of variation for the low and high tissue quality control samples were 5.9% and 3.7%, respectively. Intraday (n=10) coefficients of variation for the low and high tissue quality control samples were 4.2% and 4.4%, respectively.
Data were abstracted from the medical record of all study participants during the hospitalization and up to 6 weeks postpartum to assess for any infection morbidities including fever, cellulitis, wound infection, endometritis, urinary tract infection, or abscess. Participants were contacted at 6–8 weeks after delivery and a standardized transcript of eight questions was asked to determine whether there were any infectious morbidities not captured by medical record abstraction. The questions inquired about emergency department visits, hospital admissions, fevers, wound complications such as opening of the incision and redness around the incision, and antibiotic use.
The primary outcome for the study was the opening adipose tissue concentration of cefazolin. We determined the sample size for this study a priori based on prior data from a similar population. In the 2-g group, we estimated that the mean cefazolin concentration would be 6.37 micrograms/g (standard deviation±2.3).10 To detect a 3-microgram/g difference in this concentration between the 2-g and 3-g groups (two-tailed α=0.05, β=0.2) required 29 participants per group. Data analysis was performed using SAS 9.2. Categorical variables were compared by χ2 or Fisher exact test and continuous variables were compared by t test or nonparametric Wilcoxon rank-sum test for data that were not normally distributed. P values were two sided and P<.05 is considered significant.
Of the 649 patients who were screened for eligibility, 454 did not meet inclusion criteria, 32 declined participation, and 105 were not enrolled for other reasons (Fig. 1). From April 2013 to July 2014, 58 patients were randomized. As a result of a protocol violation, one patient who was randomized did not receive the study drug nor were any of her specimens collected for analysis. Therefore, 57 patients received the intervention, and it is their data that form the basis for this report. All charts were reviewed for readmissions and complications in the 6 postoperative weeks. Phone interviews were accomplished in 32% (18/57) of the patients. There were similar follow-up rates in the 2-g and 3-g groups (29% compared with 35%, P=.63).
We noted similar demographic (Table 1), preoperative, and intraoperative characteristics between the two groups (Table 2). Median cefazolin adipose tissue and serum concentrations are summarized in Table 3. There was no statistically significant difference in the primary outcome, median opening adipose tissue cefazolin concentrations, between the 2-g and 3-g groups (9.4 compared with 11.7 micrograms/g, P=.12). Similarly, there was no statistically significant difference in the median closing tissue concentrations (8.4 compared with 8.7 micrograms/g, P=.31). A two-way analysis of variance was performed and there was no effect modification by BMI group. There was a large variation around the median concentrations (Table 3); however, this was expected with fixed dosing (rather than mg/kg) and is the result of individual pharmacokinetic variability of the patients because each has different transfer constants of the drug, such as body weight, hydration status, and differences in renal function.
The generally accepted revised minimally inhibitory concentration (MIC) for Staphylococcus species is 8 micrograms/g and 2 micrograms/g for Enterobacteriaceae.11 We found no statistically significant difference in the percentage greater than the MIC for Staphylococcus species in the 2-g compared with the 3-g group in the opening concentration (60.7% compared with 72.4%, P=.35) or in the closing concentration (50.0% compared with 60.7%, P=.42). All samples were above 2 micrograms/g, the MIC for Enterobacteriaceae. There were 20 participants who consented to the serum arm of the study, nine in the 2-g group and 11 in the 3-g group. The initial serum concentrations of cefazolin were significantly higher in the 3-g group (113.1 compared with 208.7 micrograms/g, P=.004); however, there were no differences in serum concentration at the time of the adipose tissue collections. The demographic characteristics of the group that participated in the serum part of this study were similar to the overall cohort. Serum participants were also similar to their respective groups (2 g compared with 3 g).
We then analyzed each BMI stratification group separately. There were no statistically significant differences in median adipose tissue concentration in the opening or the closing sample in the BMI 30–39.9 group or in the BMI 40 or greater group (Table 4).
Finally, there was no statistically significant difference in infection morbidities between the two groups. There was one wound complication in the 2-g group (wound hematoma) and two complications in the 3-g group (wound infection and superficial wound dehiscence).
As opposed to 2 g, 3 g of preoperative cefazolin in obese women undergoing cesarean delivery did not result in either higher adipose tissue concentrations or increase the proportion who achieved the MIC for Staphylococcus species for cefazolin in adipose tissue.
This study has several strengths. First, it was a double-blinded and randomized study. Second, we measured cefazolin adipose tissue concentration both at surgery start and surgery end. Third, we achieved our a priori sample size of 58 participants and had outcome data on 98% (57/58) of them. Finally, we used high-pressure liquid chromatography assays, which have been previously validated and have low interday and intraday coefficients of variation for both serum and tissue.
The study has several limitations. It was designed to examine antibiotic tissue concentrations and the sample size was inadequate to detect differences in clinical and infection outcomes. With a larger sample size we may have been able to detect smaller differences in the proportion of each group that was above the MIC for Staphylococcus species. We performed a post hoc power analysis and estimate that a sample size of 542 would be necessary to detect an 11% difference (the actual difference we observed in our study) in the proportion who achieves this MIC. We also had lower than anticipated phone follow-up (32%), which limited our ability to detect late postoperative complications managed on an outpatient basis or at another institution. The study was not powered to detect any difference in infection outcomes.
Two smaller studies have looked at cefazolin adipose tissue concentrations. The first was an observational study, which included 19 obese women undergoing cesarean delivery and demonstrated that adipose tissue concentrations did not reach the MIC in women with BMIs of 30–39.9 20% of the time and in women with BMIs of 40 or greater 45% of the time.10 Direct comparison with our study is limited by the MIC threshold of this study, which used 4 micrograms/g compared with the current standard of 8 micrograms/g and a different methodology (microbiologic plate assay and zones of inhibition measurements). Stitely et al,12 in a randomized but unblinded study, compared 2 g compared with 4 g of cefazolin prophylaxis in 20 pregnant women undergoing cesarean delivery who had BMIs of 35 or greater. They found significantly higher concentration in mean adipose tissue concentration in the 4-g group compared with the 2-g group at incision (40.11±24.10 compared with 18.36±6.68, P<.001) and at closure (34.89±17.42 compared with 21.73±16.02, P=.044). Again, comparison with our study is difficult because the 2-g group achieved mean adipose tissue concentrations twice what we found and, like Pevzner et al, do not report the proportion that attained an 8-microgram/g threshold.
Another important aspect of pharmacokinetic studies is the measurement of serum levels of the specified drug. As is evidenced in our study, the concentrations of cefazolin in the serum attained a statistically significant difference in the initial samples between the 2-g and 3-g groups. Stitely et al12 also found a statistically significant difference in the plasma concentration of cefazolin in the 2-g compared with the 4-g group at both incision and closure. Given that the varying dosing schedules affect serum concentrations and do not uniformly affect adipose tissue levels, it is important to look at both body compartments when possible.
In conclusion, our findings do not support the use of 3 g of cefazolin in obese women undergoing cesarean delivery although this dose is recommended by the American Society of Health-System Pharmacists, Infectious Disease Society of America, Surgical Infection Society, and Society for Healthcare Epidemiology of America in women weighing more than 120 kg.7 In our BMI 40 or greater group, the group most likely to have weighed greater than 120 kg, the cefazolin concentrations were very similar (Table 4). Although we did not have any adverse outcomes to the 3-g dose, there are limited data on this dose in pregnant women, and the potential for harm exists without adequate studies. Given the results of this study, alternate prophylactic regimens should continue to be explored but the current recommendations by the American College of Obstetricians and Gynecologists maintained.
1. Toma O, Suntrup P, Stefanescu A, London A, Mutch M, Kharasch E. Pharmacokinetics and tissue penetration of cefoxitin in obesity: implications for risk of surgical site infection. Anesth Analg 2011;113:730–7.
2. Anderson DJ, Kaye KS, Classen D, Arias KM, Podgorny K, Burstin H, et al.. Strategies to prevent surgical site infections in acute care hospitals. Infect Control Hosp Epidemiol 2008;29(suppl 1):S51–61.
3. DeFrances CJ, Cullen KA, Kozak LJ. National Hospital Discharge Survey: 2005 annual summary with detailed diagnosis and procedure data. Vital Health Stat 13 2007;1–209.
4. Tita AT, Rouse DJ, Blackwell S, Saade GR, Spong CY, Andrews WW. Emerging concepts in antibiotic prophylaxis for cesarean delivery: a systematic review. Obstet Gynecol 2009;113:675–82.
5. Martin JA, Hamilton BE, Ventura SJ, Osterman MJ, Mathews TJ. Births: final data for 2011. Natl Vital Stat Rep 2013;62:1–69, 72.
6. Dinsmoor MJ, Gilbert S, Landon MB, Rouse DJ, Spong CY, Varner MW, et al.. Perioperative antibiotic prophylaxis for nonlaboring cesarean delivery. Obstet Gynecol 2009;114:752–6.
7. Bratzler DW, Dellinger EP, Olsen KM, Perl TM, Auwaerter PG, Bolon MK, et al.. Clinical practice guidelines for antimicrobial prophylaxis in surgery. Surg Infect (Larchmt) 2013;14:73–156.
8. Use of prophylactic antibiotics in labor and delivery. Practice Bulletin No. 120. American College of Obstetricians and Gynecologists. Obstet Gynecol 2011;117:1472–83.
9. Dudley MN, Nightingale CH, Drezner AD, Low HB, Quintiliani R. Comparative penetration of cefonicid and cefazolin into the atrial appendage and pericardial fluid of patients undergoing open-heart surgery. Antimicrob Agents Chemother 1984;26:347–50.
10. Pevzner L, Swank M, Krepel C, Wing DA, Chan K, Edmiston CE Jr. Effects of maternal obesity on tissue concentrations of prophylactic cefazolin during cesarean delivery. Obstet Gynecol 2011;117:877–82.
11. CLSI. Performance standards for antimicrobial susceptibility testing; twenty-second informational supplement. CLSI document M100-S22. Wayne (PA): Clinical and Laboratory Standards Institute; 2012.
12. Stitely M, Sweet M, Slain D, Alons L, Holls W, Hochberg C, et al.. Plasma and tissue cefazolin concentrations in obese patients undergoing cesarean delivery and receiving differing pre-operative doses of drug. Surg Infect (Larchmt) 2013;14:455–9.