Materials and Methods
This prospective cohort study was conducted at the University of California, Irvine and Miller Children’s and Women’s Hospital, Long Beach Memorial. The protocol was reviewed and approved by the local Institutional Review Board, and all participants provided informed written consent before enrollment.
Inclusion criteria were identical to that of our pilot study. All obese women who underwent scheduled cesarean delivery at term, >37 weeks’ gestation, were screened for enrollment. Exclusion criteria included any chronic medical comorbidity that potentially could affect tissue perfusion, thereby influencing pharmacokinetics; these included chronic hypertension, pregestational diabetes mellitus, and collagen vascular diseases including systemic lupus erythematosus. Other exclusion criteria included patient allergy to cephalosporins, exposure to antibiotics within the 7 days preceding cesarean delivery, multiple gestations, need for emergent delivery, or suspected preexisting infection.
Using accepted World Health Organization classifications for body mass index (BMI) we divided women into 2 categories: obese (BMI, 30-39.9 kg/m 2 ) and extremely obese (BMI, >40 kg/m 2 ). Maternal BMI was calculated with the use of recorded height and weight at time of admission for cesarean delivery and calculated with the following formula: weight (kg)/height (m) 2 . We did not recruit normal and overweight women (BMI, <30 kg/m 2 ), given that, in our historical cohort, all of the women met acceptable tissue concentrations after the standard prophylactic 2-g dose.
Three grams of cefazolin was given parenterally 30-60 minutes before skin incision. A primary study nurse and/or the lead research physician was involved in the timing and collection of all serum and adipose samples. Cefazolin was given as a slow intravenous push over an average of 3-5 minutes. Immediately after skin incision and before fascial incision, the initial adipose sample was collected. Serum collection was coordinated to occur at time of skin incision. The second adipose sample was collected after fascial closure and just before skin closure. Both adipose samples were collected from just below the level of the subcutaneous tissue.
Adipose samples were immediately blotted dry and placed in labeled vials on ice. Maternal blood was allowed to clot and centrifuged for 10 minutes at 3200 rpm. Serum then was removed selectively and, along with the 2 adipose samples, was placed in a –80°C freezer until the time of analysis.
Cefazolin concentrations were determined with the use of a validated high-performance liquid chromatography assay, as previously reported, at the Center for Anti-Infective Research and Development, Hartford Hospital. The serum assay was linear over a range of 0.5–50 μg/m: ( R 2 = 0.998). Intraday (n = 10) coefficients of variation for the low- (1 μg/mL) and high-quality (40 μg/mL) control samples were both <4%. Interday (n = 4) coefficients of variation were 6.9% and 2.8%, respectively. All tissue samples were processed with 1 part tissue and 4 parts saline solution. The tissue assay was linear over a range of 0.5–50 μg/g ( R 2 = 0.996). Intraday (n = 10) coefficients of variation for the low- (1 μg/g) and high-quality (40 μg/g) control samples were 4.2% and 4.4%, respectively. Interday (n = 8) coefficients of variation were 5.9% and 3.7%, respectively.
Criteria for meeting acceptable serum and tissue concentrations were based on the Clinical and Laboratory Standards Institute published susceptibilities of cefazolin. Minimal inhibitory concentrations (MICs) are now set at 8 μg/mL, based on resistance data for common Enterobacteriaceae organisms.
Calculations for an a priori power analysis were based on the results of our historic control subjects. Seventy percent of historic obese subjects obtained tissue concentrations of >4 μg/mL, which is the previous accepted breakpoint for cefazolin at the time of power calculation. Assuming that 95% of obese subjects would attain an MIC of >4 μg/mL with an increased cefazolin dose of 3 g, with an alpha of .05 and a power of 0.8, it was estimated that 12 subjects in each BMI category would be needed to detect a significant difference in adipose cefazolin concentrations after a 3-g prophylactic dose. We planned to enroll 14 subjects in each group to account for potential sample storage, processing, or plating complications that could lead to a “no” result.
Statistical analyses were performed with the statistical program R (version 3.03; R Foundation for Statistical Computing, Vienna, Austria, 2014). All tests were conducted at the .05 significance level. The t -test on the regression coefficient for the group indicator was used to test for differences in means. Categoric variables were evaluated with the Fisher exact test of association. Linear regressions for log-transformed start adipose concentrations were calculated by dose-specific group.
Results
We enrolled 30 women in this prospective study between August 2013 and January 2014. Two women were excluded from the study before collection of all tissue and serum samples because of inability to achieve skin incision in the required time of 30-60 minutes after dosing of cefazolin. Of the remaining 28 participants, there were 14 women in the obese group with an average BMI of 33.8 kg/m 2 and 14 women in the extremely obese group with an average BMI of 45.0 kg/m 2 . Additional demographic characteristics are outlined in Table 1 . All major race/ethnicities were represented in both cohorts, and mean BMIs were comparable, differing by ≤0.5 BMI points between the current and historic cohorts.
| Variable | Body mass index, kg/m 2 | P value | ||
|---|---|---|---|---|
| <30: 2 g cefazolin (n = 10) | 30-40/>40: 2 g cefazolin (n = 19) | 30-40/>40: 3 g cefazolin (n = 29) | ||
| Maternal characteristics | ||||
| Age, y b | 28.0 (23.0–32.0) | 30.0 (25.5–33.5) | 31.5 (26.5–36.5) | .60 |
| Race-ethnicity, n (%) | .69 | |||
| Asian/Pacific Islander | 1 (10) | 1 (5.26) | 2 (7.14) | |
| African American | 3 (30) | 3 (15.79) | 3 (10.71) | |
| Latina | 3 (30) | 9 (47.37) | 17 (60.71) | |
| White | 3 (30) | 6 (31.58) | 5 (17.86) | |
| Other | 0 | 0 | 1 (3.57) | |
| Gestational age, d b | 273 (271–274) | 273 (272–274) | 275 (274–279.5) | < .001 |
| Body mass index category | ||||
| Height, cm b | 157.3 (152.0–162.0) | 161.0 (160.0–167.5) | 163.8 (160.0–170.2) | .60 |
| Weight, kg b | 67.0 (64.0–68.0) | 104.0 (86.5–120.5) | 105.6 (87.2–120.0) | .80 |
| <30 kg/m 2 c | 26.7 ± 1.29 | — | — | |
| 30-40 kg/m 2 c | — | 34.12 ± 2.65 | 33.82 ± 2.92 | .79 |
| >40 kg/m 2 c | — | 44.48 ± 4.5 | 45.03 ± 3.77 | .76 |
| Cesarean delivery variables | ||||
| Repeat cesarean delivery, n (%) | 8 (80) | 17 (89.5) | 25 (89.3) | .98 |
| Preoperative hemoglobin, g/dL b | 12.2 (11.8–12.7) | 12.0 (11.7–12.5) | 12.0 (11.4–12.6) | .86 |
| Blood loss, mL b | 700 (700–800) | 800 (700–800) | 702 (612–829) | .10 |
| Total operative time, min b | 61.5 (49.0–71.0) | 52.0 (45.0–66.0) | 33.5 (25.0–53.0) | .001 |
| Time from antibiotic infusion completion to first adipose sample, min b | 31.5 (30.0–33.0) | 35.0 (30.5–47.0) | 41.0 (32.5–47.0) | .27 |
| Time from antibiotic infusion completion to second adipose sample, min b | 93.0 (81.0–102.0) | 92.0 (85.0–103.5) | 72.0 (63.0–86.5) | .005 |
| Time from antibiotic infusion completion to serum sample, min b | 82.0 (65.0–109.0) | 92.0 (82.5–96.5) | 42.5 (32.5–48.5) | < .001 |
a All 2-g data were reported previously by Pevzner et al in phase 1 of the investigation
b Data are reported as median (interquartile range) for continuous variables
With the use of the new established MIC of 8 μg/mL as a reflection of adequate tissue antibiotic concentration, 80% of initial adipose tissue samples from women with a BMI of 20-30 kg/m 2 reached appropriate concentrations after 2 g of cefazolin, with a median concentration of 8.7 μg/g. Only 20% of initial adipose samples from women with a BMI of 30-40 kg/m 2 reached the acceptable MIC after 2 g; 100% of the women exceeded it after 3 g, with median concentrations of 6.5 and 22.4 μg/g respectively ( P < .001). No subjects in the BMI of >40 kg/m 2 cohort met acceptable MIC concentrations within initial adipose samples after 2 g; 71.4% of them achieved this concentration after 3 g, with median concentrations of 4.7 and 9.6 μg/g, respectively ( P = .002). Individual concentration data by BMI and tissue sample are presented in Table 2 and Figure 1 . Data by weight, in 20-kg increments, are presented in Table 3 and Figure 2 . All subjects achieved appropriate MIC values within serum samples.
| Variable | BMI | ||||||
|---|---|---|---|---|---|---|---|
| 20-30 kg/m 2 | 30-40 kg/m 2 | P value | >40 kg/m 2 | P value | |||
| 2 g cefazolin (n = 10) | 2 g cefazolin (n = 10) | 3 g cefazolin (n = 14) | 2 g cefazolin (n = 9) | 3 g cefazolin (n = 14) | |||
| Initial adipose concentration, μg/g b | 8.7 (8.4–11.2) | 6.5 (4.2–7.2) | 22.4 (20.3–34.4) | < .001 | 4.7 (3.1– 5.0) | 9.6 (7.6–15.8) | .002 |
| Percentage of samples with MIC >8 mcg/mL, % c | 80 | 20 | 100 | < .001 | 0 | 71.4 | .003 |
| Closure adipose concentration, μg/g b | 7.5 (5.0–10.1) | 5.1 (4.0–9.3) | 24.8 (18.9–34.1) | < .001 | 4.3 (3.5–6.2) | 7.1 (6.2–9.8) | < .001 |
| Percentage of samples with MIC >8, μg/mL, % c | 50 | 40 | 92.9 | < .001 | 0 | 42.9 | < .001 |
| Serum concentration, μg/mL b | 57.2 (48.6–66.3) | 42.6 (27.3–64.8) | 166.7 (143.8–185.7) | < .001 | 39.7 (35.5–71.5) | 176.5 (127.9–192.6) | < .001 |
a All 2-g data were reported previously by Pevzner et al in phase 1 of the investigation
b Data are reported as median (interquartile range)
c As defined by The Clinical and Laboratory Standards Institute, 8 μg/mL serves as the acceptable concentration for inhibition of Enterobacteriaceae species.
| Variable | Weight | |||||||
|---|---|---|---|---|---|---|---|---|
| <80 kg | 80-100 kg | 100-120 kg | >120 kg | |||||
| 2 g cefazolin (n = 11) | 3 g cefazolin (n = 2) | 2 g cefazolin (n = 8) | 3 g cefazolin (n = 9) | 2 g cefazolin (n = 5) | 3 g cefazolin (n = 10) | 2 g cefazolin (n = 5) | 3g cefazolin (n = 7) | |
| Initial adipose concentration, μg/g b | 8.7 (7.5–11.2) | 44.5 (18.6–70.4) | 6.2 (4.5–6.9) | 21.9 (13.3–26.6) | 4.2 (3.4–5.0) | 19.2 (9.9–24.8) | 4.9 (3.1–5.8) | 7.6 (6.9–11.3) |
| Percentage of samples with MIC >8 mcg/mL, % c | 72.7 | 100 | 12.5 | 100 | 0 | 100 | 0 | 42.9 |
| Closure adipose concentration, μg/g b | 8.4 (5.0–10.1) | 37.5 (27.4–47.6) | 5.9 (4.5–10.3) | 18.9 (10.4–25.5) | 3.8 (3.5–4.3) | 10.8 (6.5–24.1) | 5.0 (3.5–6.2) | 6.5 (5.8–8.3) |
| Percentage of samples with MIC >8 μg/mL, % c | 54.5 | 100 | 37.5 | 88.9 | 0 | 60 | 0 | 28.6 |
| Serum concentration, μg/mL b | 57.9 (48.9–67.8) | 77.2 (4.8–149.6) | 42.6 (31.4–55.9) | 185.1 (151.2–186.3) | 41.1 (37.6–73.1) | 183.4 (179.1–193.1) | 39.7 (30.2–48.1) | 127.9 (104.0–157.4) |
a All 2-g data were reported previously by Pevzner et al in phase 1 of the investigation
b Data are reported as median (interquartile range)
c As defined by The Clinical and Laboratory Standards Institute, 8 μg/mL serves as the acceptable concentration for inhibition of Enterobacteriaceae species.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
