Cefazolin Dosing for Surgical Prophylaxisin Morbidly Obese Patients

Vanessa P. Ho,1 David P. Nicolau,2 Gregory F. Dakin,1 Alfons Pomp,1 Barrie S. Rich,1

Christopher W. Towe,1 and Philip S. Barie1,3

Abstract

Background: Cefazolin is used commonly to prevent surgical site infection (SSI) after operations on patients withmorbid obesity (MO), but specific dosing guidelines are lacking. We hypothesized that cefazolin 2 g given byintravenous (IV) push over 5 min (IVP) or infusion over 30 min (INF) would suffice for SSI prophylaxis in MO(body mass index [BMI] 40–50 kg/m2), and cefazolin 3 g would be sufficient in patients with super-morbidobesity (SMO) (BMI > 50 kg/m2).Methods: Twenty-five patients undergoing elective surgical procedures were given a single dose of cefazolin:Ten with MO received 2 g via IVP (MO2-IVP), five with MO received 2 g via 30-min infusion (MO2-INF), fivewith SMO received 2 g via infusion (SMO2-INF), and five with SMO received 3 g via infusion (SMO3-INF).Serum cefazolin concentrations were measured 5, 30, 120, and 360 min after initiation of the dose. The half-life ofthe drug was calculated for each patient, as was the time the concentration was above the minimum inhibitoryfree concentration (fT > MIC) using 8 mcg/mL as the breakpoint. The protective duration of each cefazolin dosewas assessed using the pharmacodynamic target for fT > MIC of 70%.Results: The mean cefazolin concentrations after 30 min were similar in all groups; the mean concentrations at120 and 360 min were 67.1–84.8 mcg/mL and 22.9–40.8 mcg/mL, respectively. The half-life ranged from 2.3 to3.6 h and was unaffected by administration method. The protective duration was 5.1 h for MO2-IVP, 4.8 h forMO2-INF, 5.8 h for SMO2-INF, and 6.8 h for SMO3-INF.Conclusions: A single 2-g dose of cefazolin appears to provide antibiotic exposures sufficient for most commongeneral surgical procedures of < 5-h duration, regardless of BMI.

Obesity rates have been increasing in the United Statesover the last several decades. Recent estimates are that

66% of American adults are overweight or obese [1]. Globalestimates of obesity also indicate dramatic increases [2], suchthat obesity was declared a global epidemic in 1997 by theWorld Health Organization [3]. As a consequence, bariatricsurgery procedures have become more common, from 6,868operations performed in the U.S. in 1996 [4] to 220,000 in 2008[5]. The number of non-bariatric operations in patients withmorbid obesity (MO) also is increasing.

Obesity is postulated to be a risk factor for surgical siteinfection (SSI) after a variety of surgical procedures [6-8].Clinicians should be aware of alternate antibiotic dosingregimens to provide adequate prophylaxis for this popula-

tion. However, the pharmacokinetics of antibiotic dosing inpatients with MO is complicated, and studies are few innumber. Antibiotics that require therapeutic monitoring, suchas aminoglycosides and vancomycin, have been studied mostthoroughly [9,10]. Dosing is drug-specific because of thevariability in a variety of parameters, including body com-position, affinity for plasma proteins, and drug lipophilicity[11,12]. Obese patients additionally have variations in bloodflow to adipose tissue, cytochrome P450 activity, and glo-merular filtration rate, all of which can affect pharmacody-namics [13]. Because of the relative unpredictability ofpharmacokinetics in obese individuals, doses are best esti-mated on the basis of specific studies for individual drugscarried out in obese individuals. For example, vancomycin

Departments of 1Surgery and 3Public Health, Weill Cornell Medical College, New York, New York.2The Center for Anti-Infective Research and Development, Hartford Hospital, Hartford, Connecticut.Presented at the Thirtieth Annual Meeting of the Surgical Infection Society, Las Vegas, Nevada, April 19, 2010. VPH was the recipient of

the Surgical Infection Society Foundation/Wyeth Fellowship in Clinical Evaluative Science.

SURGICAL INFECTIONSVolume 13, Number 1, 2012ª Mary Ann Liebert, Inc.DOI: 10.1089/sur.2010.097

33

should be given on the basis of actual body weight, withdosage adjustments based on serum concentrations [14],whereas aminoglycoside dosing requires calculation of ad-justed body weight via a correction factor [15].

Optimal dosing regimens for beta-lactam antibiotics havebeen less studied in this population, despite the fact that first-generation cephalosporins remain the drugs of choice forsurgical prophylaxis. The efficacy of these antibiotics dependson the time the free drug concentrations are above the mini-mum inhibitory concentration (MIC) (fT > MIC). In general,beta-lactam agents are efficacious when administered suchthat their fT > MIC exceeds 60–70% of the dosing interval [16].

The purpose of this study was to determine an optimaldosing regimen for cefazolin as a prophylactic antibiotic insurgery in patients with MO. We hypothesized that cefazolin2 g, given either via intravenous (IV) push over 5 min (IVP) oras a 30-min IV infusion (INF), would suffice for SSI prophy-laxis, but that a larger dose of cefazolin (3 g) might be neces-sary in patients with super-morbid obesity (SMO).

Patients and Methods

The study protocol was reviewed and approved by theCommittee on Human Rights Research of Weill CornellMedical College. Patients undergoing surgical procedures forMO (including Roux-en-Y gastric bypass, biliopancreatic di-version, sleeve gastrectomy, and gastric banding) were en-rolled according to institutional informed consent guidelines.Twenty-five obese patients scheduled for bariatric surgerywere weighed, and the body mass index (BMI) was calcu-lated. All patients received a single dose of cefazolin. Patientswith BMI > 40 kg/m2 were recruited for the study in fourgroups: Group 1 = ten patients with BMI 40–50 kg/m2 re-ceived cefazolin 2 g via IVP (MO2-IVP); Group 2 = five pa-tients with BMI 40–50 kg/m2 received cefazolin 2 g via INF(MO2-INF); Group 3 = five patients with BMI > 50 kg/m2 re-ceived 2 g cefazolin via INF (SMO2-INF); and Group 4 = fivepatients with BMI > 50 kg/m2 received 3 g of cefazolin viaINF (SMO3-INF). Alanine and aspartate aminotransferasewere assayed, as hepatic disease can affect the half-life andprotein binding of cefazolin [17]. Patients with allergy topenicillins or cephalosporins, current pregnancy or lactation,chronic kidney disease, or chronic hepatic insufficiency were

excluded. Patients were assigned consecutively to groups af-ter determination of BMI.

Blood samples were taken at 5, 30, 120, and 360 minafter dose initiation. The blood was allowed to clot for10 min and centrifuged at 3,200 rpm for 10 min to separatethe serum, which was stored immediately at - 80�C untilanalysis. Cefazolin concentrations were determined usinga previously validated high-performance liquid chroma-tography method at the Center for Anti-Infective Researchand Development (Hartford, CT) [18]. The assay was linearover a range of 0.5 to 50 mcg/mL (R2 = 0.997). Intra-daycoefficients of variation for the low (1 mcg/mL) and high(40 mcg/mL) quality-control samples were 3.6% and 3.4%,respectively. Inter-day coefficients of variation were 4.9%and 3.7%, respectively.

Patient group characteristics were analyzed using mea-sures of central tendency and the Kruskal-Wallis one-wayanalysis of variance. The elimination rate constant (ke) wascalculated using a linear regression of the terminal portion ofthe concentration time profile, and the half-life was deter-mined by 0.693/ke. All calculations were performed usingSTATA version 11.0 (Stata Corp., College Station, TX). ThefT > MIC was determined using a protein binding value of85%. The fT > MIC of 8 mcg/mL was determined for eachpatient, as was the protective duration of each regimen usingthe pharmacodynamic target of fT > MIC of 70%.

As secondary analyses, correlation and univariable re-gression were performed on the 20 patients who received 2 gof cefazolin to determine if weight, BMI, or age was associatedwith the cefazolin concentration at t = 30 min (peak) andt = 360 min (trough).

Results

Sixteen females and nine males were recruited for thestudy; by happenstance, the MO2-INF group included onlyfemales. Operations consisted of 12 Roux-en-Y gastric bypassprocedures, seven placements of adjustable gastric band de-vices, five sleeve gastrectomies, and one biliopancreatic di-version with duodenal switch. Patient characteristics,including age, weight, BMI, and measured creatinine con-centration, are presented in Table 1. The mean BMIs were44.1 – 3.3 (standard deviation [SD]) in the MO2-IVP group,

Table 1. Patient Characteristicsa

Group 1(MO2-IVP)

Group 2(MO2-INF)

Group 3(SMO2-INF)

Group 4(SMO3-INF)

Pvalueb

N 10 5 5 5Sex (M:F) 6:4 0:5 1:4 2:3Weight (kg) 128.4 – 17.9 114.6 – 16.4 164.2 – 23.0 156.0 – 20.3 <0.01Body mass index (kg/m2) 44.1 – 3.3 43.7 – 3.4 55.7 – 8.8 55.2 – 2.9 <0.01Age (years) 37.9 – 11.9 40.2 – 13.1 33.6 – 13.3 52.8 – 15.8 0.22Serum creatinine (mg/dL) 0.80 – 0.19 0.64 – 0.15 0.80 – 0.14 0.78 – 0.15 0.35Serum albumin (g/dL) 3.8 – 0.4 3.6 – 0.2 3.9 – 0.2 3.7 – 0.1 0.28Serum alanine aminotransferase (U/L) 39.7 – 12.0 22.6 – 5.5 26.8 – 8.1 24.0 – 11.8 0.06Serum aspartate aminotransferase] (U/L) 25.6 – 4.5 21.0 – 2.8 23.0 – 5.3 19.2 – 4.1 0.06

aValues presented as mean – standard deviation.bKruskal-Wallis analysis of variance.INF = intravenous infusion over 30 min; IVP = intravenous push over 5 min; MO = morbid obesity; SMO = super-morbid obesity.

34 HO ET AL.

43.7 – 3.4 in the MO2-INF group, 55.7 – 8.8 in the SMO2-INFgroup, and 55.2 – 2.9 in the SMO3-INF group; p < 0.01. Therewere no significant differences in age, serum creatinine con-centration, or serum albumin concentration among thegroups. Whereas there was a trend toward a difference amongthe groups in alanine aminotransferase and aspartate ami-notransferase concentrations (p = 0.06 for both), all concen-trations were well within normal limits.

The mean total cefazolin concentrations in each patientgroup (mcg/mL) are presented in Table 2. After t = 30, meanconcentrations were similar in all groups; p = 0.64 for t = 120and p = 0.51 for t = 360. The administration method did notsignificantly affect the half-life of cefazolin, which rangedfrom 2.3–3.6 hours. As displayed in Table 2, the fT > MIC at anMIC of 8 mcg/mL ranged from 3.4–4.8 h, whereas the pro-tective duration ranged from 4.8–6.8 h in these MO and SMOpatients.

The mean free cefazolin concentration time profiles of theMO and SMO populations are presented in Figure 1. In-travenous push administration gave the highest peak meanconcentration, which was documented immediately after

administration. However, cefazolin was eliminated at a sim-ilar rate in groups, as evidenced by the similar half-lifes ineach group. Even though the administration method led todifferent concentrations initially, the differences diminishedover time such that the fT > MIC was similar in all groups.

In the secondary analyses, determination of the correlationcoefficients and univariable regression were performed on the20 patients who received 2 g of cefazolin to determine ifweight, BMI, or age was associated with cefazolin concen-tration at t = 30 and 360 min. The correlation coefficient att = 30 min was 0.08 for weight, 0.13 for BMI, and 0.20 for age;at t = 360 min, the correlation coefficients were 0.29, 0.42, and- 0.16, respectively. Total body weight, BMI, and age thuswere not correlated with cefazolin concentrations and werenot significantly associated by linear regression at either timepoint (Table 3), indicating that dosing based on weight asopposed to BMI is moot.

Discussion

Concerns in the political and regulatory arenas about in-creasing healthcare costs have led to the development ofmeasures intended to limit cost and improve quality. Onesuch measure, devised by the Centers for Medicare andMedicaid Services (CMS), specifically targets SSI after bar-iatric surgery as a ‘‘hospital-acquired condition.’’ The impli-cation of this designation is that in 2009, CMS stoppedreimbursements to hospitals for treatment of SSI in inpatientsafter bariatric surgery [19]. Therefore, the development of SSIin bariatric patients has not only clinical implications forpatients but also economic implications for hospitals andphysicians.

The SSI rate after bariatric surgery has been reported to bebetween 1.3% and 22.0%, with wide variation secondary toevolving practices and non-standardized reporting [20-22].Recent data from a multi-institutional prospective trial ex-amining staple height as a function of leak rate after laparo-scopic gastric bypass surgery reported the overall superficialincisional SSI rate to be 7.9% [23]. Morbid obesity is itself amajor risk factor for SSI; a 1989 study noted a 16.5% incidencein patients with MO undergoing clean-contaminated surgery,compared with 2.5% in a non-obese group [20]. Laparoscopyis associated with lower rates of SSI, with a significant dif-ferential in SSI rates from as high as 10.5% after open opera-tions to 1.3% after laparoscopic procedures [22]. In addition, avariety of other factors have been associated with higher

Table 2. Cefazolin Concentrations (Total) and fT > MIC Exposure

Group 1 (MO2-IVP) Group 2 (MO2-INF) Group 3 (SMO2-INF) Group 4 (SMO3-INF)

Mean – SD (mcg/mL)5 min 309.7 – 147.0 – – – – – –30 min 140.6 – 62.9 165.1 – 38.5 197.7 – 13.8 182.1 – 57.5120 min 74.0 – 20.1 67.1 – 18.0 91.2 – 4.0 84.8 – 29.6360 min 22.9 – 9.0 26.8 – 17.2 30.3 – 6.2 40.8 – 25.8

Half-life (h) 2.3 – 0.4 2.9 – 1.1 2.7 – 1.2 3.6 – 1.6fT > MIC (ha) 3.6 3.4 4.1 4.8Protective duration (hb) 5.1 4.8 5.8 6.8

aFor MIC 8 mcg/mL.bCalculated using fT > MIC = 0.70, where MIC = 8 mcg/mL.INF = intravenous infusion over 30 min; IVP = intravenous push over 5 min; MIC = minimum inhibitory concentration; MO = morbid

obesity; SD = standard deviation; SMO = super-morbid obesity.

FIG. 1. Mean free cefazolin concentrations (mcg/mL) overtime for four patient groups. MO2-IVP = Group 1, Body massindex (BMI) 40–50 kg/m2; received cefazolin 2 g via intra-venous (IV) push; MO2-INF = Group 2, BMI 40–50 kg/m2;received cefazolin 2 g via 30-min IV infusion; SMO2-INF =Group 3, BMI > 50 kg/m2; received cefazolin 2 g via 30-minIV infusion; SMO3-INF = Group 4, BMI > 50 kg/m2; receivedcefazolin 3 g via 30-min IV infusion.

CEFAZOLIN DOSING IN MORBID OBESITY 35

infection rates. One study of laparoscopic gastric bypassnoted a decrease in SSI rate from 22% to 1.5% after institutionof routine bowel preparation and systematic use of a woundprotector for removal of the stapler [24].

Administration of antibiotics and the dose administeredalso are key components of prevention of SSI in MO. In the1980s, two studies reported use of cefazolin in MO popula-tions. An early study of open procedures showed a reductionin SSI incidence from 21% to 4% when a prophylaxis regimen of1 g cefazolin was used compared with placebo [21]; a laterstudy showed a similar effect when cefazolin 2 g was usedinstead of 1 g, with a decrease from 16.5% to 5.6% [20]. A morerecent study evaluated serum and tissue concentrations ofcefazolin in three groups of patients: BMI 40–49 kg/m2,50–59 kg/m2, and ‡ 60 kg/m2. All three groups received 2 gfor prophylaxis, with a scheduled second dose at 3 h. Bloodsampling was performed relative to the time of incision, asopposed to drug administration time. Serum concentrationsmeasured immediately before the second dose noted that41.1% of the samples from the BMI 40–49 kg/m2 group wereabove the therapeutic threshold, defined as 32 mcg/mL; 18.2%of samples from the 50–59 kg/m2 group and no samplesfrom the ‡ 60 kg/m2 group were above the cut point [25], butcefazolin protein binding was not considered in the analysis.

In addition to dose, the optimal regimen for cefazolinshould specify the infusion method and parameters for re-dosing. Our data indicate that cefazolin is appropriate pro-phylaxis for MO and SMO against pathogens with expectedMICs < 8 mcg/mL, the typical pathogens causing SSIs afterelective surgery. In the 2-g groups, an infusion over 30 min ledto similar concentrations and durations as administration viaa 5-min IV push, the latter being ideal for convenience ofadministration in the immediate pre-operative setting. ForMO and SMO, cefazolin 2 g can be administered safely by IVpush or by 30-min infusion. As administered, the 2-g or 3-gdoses will provide concentrations above the MIC of 8 mcg/mLfor at least 3.4 h. Using a protective duration of fT > MIC of70%, the time to redosing would be 5 h. Although the 3-g doseof cefazolin is safe for patients with BMI > 50 kg/m2, this dosedoes not appear necessary, as a 2-g dose provided sufficientexposure. Our data suggest that a single 2-g dose of cefazolinis likely to be sufficient for most common general surgicalprocedures of < 5 h duration, regardless of body mass. In ouranalysis, neither patient age, weight, nor BMI was associatedsignificantly with peak or trough cefazolin concentrations.

This study has several limitations. The sample size wassmall and consisted only of healthy patients undergoingelective surgery. These data thus cannot be extrapolated tocritically ill populations or patients with renal dysfunction.

There also was no control group of patients with normal BMI,but the pharmacokinetics of normal-weight individuals havebeen described for cefazolin. The small sample size did notallow determination of the relation between serum cefazolinconcentration and clinical outcomes. However, the goal of thisstudy was to perform a pharmacokinetic study on a morbidlyobese population, and the sample size is adequate for thispurpose. Additionally, clinicians should consider the distri-bution of the antibiotic to the peripheral tissues for maximumefficacy. Future studies should look more closely at the timingof tissue distribution and concentrations to determine thepotential for efficacy.

In 2010, the Clinical and Laboratory Standards Institute(CSLI) revised the MIC breakpoints for treatment of En-terobacteriaceae infections with multiple cephalosporins andaztreonam, including cefazolin. For cefazolin, an MIC‡ 4 mcg/mL is now considered resistant, MIC 2 mcg/mL isconsidered intermediate, and only MIC £ 1 mcg/mL indicatessusceptibility [26]. These breakpoints were revised to reflectcontemporary gram-negative pathogens, accounting for in-creasing occurrence of resistance mechanisms such as ex-tended-spectrum beta-lactamases and AmpC-type enzymes.In light of these new resistance breakpoints, interpreted for anMIC £ 1 mcg/mL, the dosing interval for all four groupscould be extended to > 7 h to maintain fT > MIC at 100%. Al-though clinicians need to consider that gram-negative path-ogens are showing increasing resistance to cefazolin in thecontext of surgical prophylaxis, these pathogens are infre-quent causes of infection. When considering methicillin-sus-ceptible Staphylococcus aureus, the majority of MICs forcefazolin are £ 2 mcg/mL; thus, using the prior breakpoint of£ 8 mcg/mL with the 2-g prophylaxis regimen utilized in thisstudy provides a high probability that adequate exposureswill be achieved for procedures lasting £ 5 h.

Strategies beyond conventional prophylaxis to prevent SSIhave been postulated. Both randomized and non-randomizedstudies suggest that continuous infusion of beta-lactam anti-biotics increases the fT > MIC and allows more consistent at-tainment of therapeutic concentrations [27, 28]. However,these trials were performed in specific clinical settings such ascritically ill patients or those with chronic obstructive pul-monary disorders [29, 30]. Whether this administrationmethod is effective for prophylaxis has yet to be elucidated[31]. Alternatively, some success has been noted in MO uti-lizing an infusion of kanamycin into the subcutaneous spaceof the incision; in one study, only six of 837 patients under-going open abdominal bariatric surgery (0.72%) developeddeep incisional SSIs, a rate much lower than other publishedSSI rates in MO [32]. Prevention of SSI in MO and SMO will

Table 3. Univariable Linear Regression

Variable [Cefazolin], mcg/mL t = 30 min [Cefazolin], mcg/mL t = 360 min

Age b coefficient - 0.11 0.03(95% CI) ( - 0.44–0.22) ( - 0.04–0.11)

Body mass index b coefficient 0.15 0.11(95% CI) ( - 0.41–0.72) ( - 0.01–0.22)

Weight b coefficient 0.02 0.02(95% CI) ( - 0.13–0.18) ( - 0.02–0.06)

CI = confidence interval.

36 HO ET AL.

require vigilant, simultaneous use of multiple strategies, in-cluding appropriate antibiotic administration, antisepsistechniques, careful tissue handling, and management of in-dividual patient risk factors.

Acknowledgments

The authors thank Christina Sutherland of the Center forAnti-Infective Research and Development for her assistancewith determination of cefazolin concentrations.

Author Disclosure Statement

No conflicting financial interests exist.

References

1. Wang Y, Beydoun MA. The obesity epidemic in the UnitedStates—gender, age, socioeconomic, racial/ethnic, and geo-graphic characteristics: A systematic review and meta-regression analysis. Epidemiol Rev 2007:6–28.

2. Caballero B. The global epidemic of obesity: An overview.Epidemiol Rev 2007;29:1–5.

3. Obesity: Preventing and managing the global epidemic:Report of a WHO Consultation Series WHO Technical Re-port Series 894. Geneva. World Health Organization. 2000.

4. Livingston EH. Procedure incidence and in-hospital com-plication rates of bariatric surgery in the United States. Am JSurg 2004;188:105–110.

5. Buchwald H, Oien DM. Metabolic/bariatric surgery world-wide 2008. Obes Surg 2009;19:1605–1611.

6. Dindo D, Muller MK, Weber M, et al. Obesity in generalelective surgery. Lancet 2003;361:2032–2035.

7. Lillenfeld DE, Vlahov D, Tenney JH, et al. Obesity and di-abetes as risk factors for postoperative wound infectionsafter cardiac surgery. Am J Infect Control 1988;16:3–6.

8. Pikarsky AJ, Saida Y, Yamaguchi T, et al. Is obesity a high-risk factor for laparoscopic colorectal surgery? Surg Endosc2002;16:855–858.

9. Pai MP, Bearden DT. Antimicrobial dosing considerations inobese adult patients. Pharmacother 2007;27:1081–1091.

10. Bearden DT, Rodvold KA. Dosage adjustments for antibac-terials in obese patients: Applying clinical pharmacokinetics.Clin Pharmacokinet 2000;38:415–426.

11. Cheymol G. Effects of obesity on pharmacokinetics: Implica-tions for drug therapy. Clin Pharmacokinet 2000;39:215–231.

12. Abernethy DR, Greenblatt DJ. Pharmacokinetics of drugs inobesity. Clin Pharmacokinet 1982;7:108–124.

13. Pieracci FM, Barie PS, Pomp A. Critical care of the bariatricpatient. Crit Care Med 2006;34:1796–1804.

14. Rybak MJ, Lomaestro BM, Rotschafer JC, et al. Vancomycintherapeutic guidelines: A summary of consensus recom-mendations from the Infectious Diseases Society of America,the American Society of Health-System Pharmacists, and theSociety of Infectious Diseases Pharmacists. Clin Infect Dis2009;49:325–327.

15. Traynor AM, Nafziger AN, Bertino JS Jr. Aminoglycosidedosing weight correction factors for patients of various bodysizes. Antimicrob Agents Chemother 1995;39:545–548.

16. Craig WA. Interrelationship between pharmacokinetics andpharmacodynamics in determining dosage regimens forbroad-spectrum cephalosporins. Diagn Microbiol Infect Dis1995;22:89–96.

17. Ohashi K, Tsunoo M, Tsuneoka K. Pharmacokinetics andprotein binding of cefazolin and cephalothin in patients withcirrhosis. J Antimicrob Chemother 1986;17:347–351.

18. Dudley MN, Nightingale CH, Drezner AD, et al. Comparativepenetration of cefonicid and cefazolin into the atrial append-age and pericardial fluid of patients undergoing open-heartsurgery. Antimicrob Agents Chemother 1984;26:347–350.

19. Hospital-Acquired Conditions 2010. Available at www.cms.hhs.gov/HospitalAcqCond/06_Hospital-Acquired_Conditions.asp Accessed March 15, 2010.

20. Forse RA, Karam B, MacLean LD, et al. Antibiotic prophy-laxis for surgery in morbidly obese patients. Surgery 1989;106:750–757.

21. Pories WJ, Van Rij AM, Burlingham BT, et al. Prophylacticcefazolin in gastric bypass surgery. Surgery 1981;90:426–432.

22. Nguyen NT, Goldman C, Rosenquist CJ, et al. Laparoscopicversus open gastric bypass: A randomized study of out-comes, quality of life, and costs. Ann Surg 2001;234:279–289.

23. Nguyen NT, Dakin G, Needleman B, et al. Effect of stapleheight on gastrojejunostomy during laparoscopic gastricbypass: A multicenter prospective randomized trial. SurgObes Relat Dis;6:477–482.

24. Schauer PR, Ikramuddin S, Gourash W, et al. Outcomes afterlaparoscopic Roux-en-Y gastric bypass for morbid obesity.Ann Surg 2000;232:515–529.

25. Edmiston CE, Krepel C, Kelly H, et al. Perioperative anti-biotic prophylaxis in the gastric bypass patient: Do weachieve therapeutic levels? Surgery 2004;136:738–747.

26. Clinical and Laboratory Standards Institute. DocumentM100-S20 (2010): Cephalosporin and Aztreonam BreakpointRevisions Fact Sheet. Available at www.clsi.org/Content/NavigationMenu/Committees/Microbiology/AST/CephalosporinandAztreonamBreakpointRevisionFactSheet/CephalosporinAztreonamBreakpointFactSheet.pdf Accessed October5, 2010.

27. Roberts JA, Lipman J. Continuous infusion of beta-lactamsin the ICU—Best way to hit the target. Crit Care Med2008;36:1663–1664.

28. Roberts JA, Paratz J, Paratz E, et al. Continuous infusion ofbeta-lactam antibiotics in severe infections: A review of itsrole. Int J Antimicrob Agents 2007;30:11–18.

29. Roberts JA, Boots R, Rickard CM, et al. Is continuous infu-sion ceftriaxone better than once-a-day dosing in intensivecare? A randomized controlled pilot study. J AntimicrobChemother 2007;59:285–289.

30. Van Zanten ARH, Oudijk M, Nohlmans-Paulssen MKE,et al. Continuous vs. intermittent cefotaxime administrationin patients with chronic obstructive pulmonary disease andrespiratory tract infections: Pharmacokinetics/pharmaco-dynamics, bacterial susceptibility and clinical efficacy. Br JClin Pharmacol 2007;63:100–109.

31. Mouton JW, Vinks AA. Continuous infusion of beta-lactams.Curr Opin Crit Care 2007;13:598–606.

32. Alexander JW, Rahn R, Goodman HR. Prevention of surgicalsite infections by an infusion of topical antibiotics in mor-bidly obese patients. Surg Infect 2009;10:53–57.

Address correspondence to:Dr. Vanessa P. Ho

Department of SurgeryNew York-Presbyterian Hospital

Weill Cornell Medical Center525 East 68th St.

New York, NY 10065

E-mail: vah9004@nyp.org

CEFAZOLIN DOSING IN MORBID OBESITY 37