Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Garey, K. W. Articles by Davis, B. R. Search for Related Content PubMed PubMed Citation Articles by Garey, K. W. Articles by Davis, B. R.

 Previous Article  |  Next Article 

Antimicrobial Agents and Chemotherapy, February 2008, p. 446-451, Vol. 52, No. 2
0066-4804/08/$08.00+0     doi:10.1128/AAC.00495-07
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

Interrupted Time Series Analysis of Vancomycin Compared to Cefuroxime for Surgical Prophylaxis in Patients Undergoing Cardiac Surgery

K. W. Garey,^1,2* D. Lai,³ T. K. Dao-Tran,² L. O. Gentry,² L. Y. Hwang,³ and B. R. Davis³

University of Houston College of Pharmacy,¹ St. Luke's Episcopal Hospital,² University of Texas School of Public Health, Houston Texas³

Received 12 April 2007/ Returned for modification 23 May 2007/ Accepted 6 November 2007

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The increased incidence of methicillin-resistant Staphylococcus aureus (MRSA), the emergence of community-acquired MRSA, and the continued high incidence of methicillin-resistant Staphylococcus epidermidis have required that certain institutions choose vancomycin for surgical prophylaxis. However, the data supporting the use of vancomycin for surgical prophylaxis are controversial. The purpose of this project was to assess the effect of the change from cefuroxime to vancomycin for surgical site infection (SSI) rates in patients undergoing coronary artery bypass graft (CABG) surgery. The monthly rates of SSIs from 2001 to 2005 were analyzed before and after a change from cefuroxime to vancomycin antibiotic prophylaxis in patients undergoing CABG by using an interrupted time series analysis. Patients who underwent cardiac valve replacement surgery and who had received vancomycin during the entire study period were used as a comparator group. A total of 6,465 patients underwent CABG surgery (n = 4,239) or valve replacement surgery (n = 2,226) during the study period. On average, the monthly SSI incidence rate in patients undergoing CABG surgery decreased by 2.1 cases per 100 surgeries after the switch from cefuroxime to vancomycin (P = 0.042) when patients undergoing valve replacement were used as a comparator group. The change in SSI rates was associated with a decrease in the incidence of infections caused by coagulase-negative Staphylococcus and MRSA isolates, with little change in the incidence of SSIs due to other gram-positive organisms or gram-negative organisms. In institutions with a high incidence of methicillin-resistant Staphylococcus species, this study provides evidence for the clinical efficacy of vancomycin prophylaxis for the prevention of postoperative SSIs in patients undergoing CABG surgery.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Surgical site infections (SSIs) are the second most common type of nosocomial infection (2, 6). Antibiotic prophylaxis is routinely given to patients to prevent postoperative SSIs. For cardiothoracic surgery, cefazolin, cefuroxime, and cefamandole are the most commonly recommended antibiotics due to their activities against the most commonly isolated pathogens, such as Staphylococcus aureus and Staphylococcus epidermidis; favorable safety profiles; and low costs. Recently, the increased incidence of methicillin-resistant S. aureus (MRSA), the emergence of community-acquired MRSA, and the continued high incidence of methicillin-resistant S. epidermidis have required that certain institutions choose alternative antibiotics for surgical prophylaxis (21, 23). Although vancomycin is not routinely recommended due to the possible emergence of vancomycin-resistant Enterococcus (VRE) or vancomycin-resistant S. aureus, vancomycin is recommended for use by institutions with high incidences of infections caused by methicillin-resistant Staphylococcus species (1, 6, 7, 14).

The data supporting the use of vancomycin for surgical prophylaxis are controversial. A meta-analysis of seven randomized trials found the activity of vancomycin to be comparable to the activities of cephalosporins (4). However, many of those studies were performed more than a decade ago in hospitals with a low incidence of methicillin-resistant Staphylococcus species. A single-center, randomized trial of 885 patients demonstrated similar infection rates, although patients given vancomycin were more likely to be infected with methicillin-sensitive Staphylococcus species and patients given cefazolin were more likely to be infected with methicillin-resistant Staphylococcus species (10). Finally, a change from cefazolin to vancomycin with rifampin was associated with a 50% decrease in SSIs in Australian patients undergoing coronary artery bypass graft (CABG) surgery (24). On 1 October 2002, vancomycin replaced cefuroxime as surgical prophylaxis for all patients undergoing CABG surgery at St. Luke's Episcopal Hospital (Houston, TX) due to the high rates of surgical site infections caused by methicillin-resistant Staphylococcus species and MRSA infection rates greater than 60% hospitalwide. Vancomycin had already replaced cefuroxime 2 years earlier in patients undergoing cardiac valve replacement surgery, due to the similar rates of cefuroxime-resistant organisms in this patient population. With a significant potential to decrease SSI rates, the purpose of this study was to assess the effect of the change from cefuroxime to vancomycin on SSI rates in patients undergoing CABG surgery.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Setting. This study was conducted at St. Luke's Episcopal Hospital, a 664-bed adult tertiary-care hospital, and was approved by the hospital's institutional review board. With the exception of the choice of antibiotic prophylaxis, other preoperative procedures remained constant during the study period. Elective surgery patients were generally admitted to a same-day admission unit in the hospital for initial processing. In this unit, they were required to shower with chlorhexidine soap and undergo preoperative blood work. From this unit, they were transferred to a preoperative holding area for final surgical instructions before they were transferred to the surgical suite.

Surgical prophylaxis protocol. Prior to October 2002, patients undergoing CABG surgery were given cefuroxime at 1.5 g before surgery for surgical prophylaxis and two doses postoperatively. Starting on 1 October 2002, the patients were given vancomycin at 1 g before surgery and two doses postoperatively for surgical prophylaxis. Two years earlier, vancomycin had replaced cefuroxime for patients undergoing cardiac valve replacement surgery. Patients undergoing valve replacement surgery were given vancomycin with the same schedule as patients undergoing CABG surgery (6). Surgical prophylaxis was discontinued after 24 h postoperatively for all cardiac surgeries. Patients do not receive intraoperative dosing of vancomycin. The time of antibiotic administration in relation to the surgery start time was not routinely collected prior to 2005.

Diagnosis of SSIs. All patients who underwent CABG or valve replacement surgery between February 2001 and August 2005 were prospectively monitored for the development of SSIs. Patients were excluded if surgery was due to an infection-related diagnosis, such as endocarditis. Patients were monitored for 30 days postoperatively for the development of SSIs of the sternum or donor site by trained infection control personnel or infectious disease physicians as described previously (11, 13). Briefly, infection occurs within 30 days after surgery and can include the following types of infections: (i) superficial incisional infections (infections above the sternum with no bony involvement), (ii) deep incisional infections (infections involving the sternum), and (iii) organ/space infections (site-specific infections, such as mediastinitis). Patients with SSIs must have positive cultures of mediastinal or leg donor site fluid or tissue specimens; evidence of infection during surgical reexploration; or either fever, chest pain, and sternal instability and at least one of the following: purulent drainage from the mediastinal area, positive blood culture results, or positive results of cultures of drainage fluid samples. Microbiologic cultures, admissions reports, and surgical schedules were scanned daily for possible SSIs. A diagnosis of an SSI was confirmed by medical chart review, patient examination, or discussion with the primary surgeon. Identification and susceptibility testing of the pathogens from SSIs were performed by the hospital's Clinical Microbiology Laboratory by using automated microbiologic techniques (Vitek), according to the guidelines published by the Clinical Laboratory Standards Institute (formerly NCCLS).

The rates of SSIs were calculated by dividing the number of cases of SSIs per month by the total number of surgeries for CABG and valve replacement per month. The SSI rates for specific microorganisms isolated in patients who underwent CABG surgery and who were given cefuroxime or vancomycin were calculated by categorizing pathogens as coagulase-negative Staphylococcus, MRSA, other gram-positive organisms, or gram-negative organisms. Pathogens from patients with polymicrobial infections were counted only once for each group. Pathogen-specific infection rates were calculated by dividing the number of cases of SSIs per month for each pathogen group by the monthly number of surgeries, as described above.

Data analysis. An interrupted time series design with a comparator group was used to assess the change in SSI rates in patients undergoing CABG surgery after the change in antibiotic prophylaxis from cefuroxime to vancomycin (3, 5, 12, 18, 26, 27). Patients undergoing valve replacement surgery were used as the comparator series to assess the changes in SSI rates independent of the intervention. Serial autocorrelations of the SSI rates among patients undergoing CABG surgery and valve replacement were assessed with an ARIMA model by using a Box-Jenkins-Tiao strategy (12). Briefly, the monthly rates of postoperative surgical site infections were plotted over time for patients undergoing CABG (case group) or valve replacement (comparator group) surgery. Patients undergoing both valve replacement and CABG surgery were categorized as undergoing valve replacement surgery. The graphs were visually inspected to assess the trends or the nonstationarity of the data. A lag plot was created by using the sample autocorrelation coefficients, r_ks, plotted against the corresponding lags. The time series was determined to be uncorrelated if the sampling distribution for r_k was within the limits ± 2/square root of n (where n is the number of patients) to assess significant departures from zero, which were inspected by visual examination of the correlogram and the partial correlogram for significant spikes in the plot. The data were considered to not have an autocorrelation if no significant spikes in the correlogram or the partial correlogram were observed. Finally, the Q statistic was computed to test analytically whether the autocorrelation function was different from a white noise process (9). If the null hypothesis of white noise is rejected, then the residuals were further modeled. However, the residuals were found not to be significantly different than white noise.

Evaluation of the intervention. As no serial autocorrelation was detected in the data, a segmented regression analysis of the interrupted time series with a comparator group was used to assess the effect of the intervention (26). Common segmented regression models fit a least-squares regression line to each segment of the independent variable (time [t]) and thus assumes a linear relationship between time and the outcome within each segment. The effect of the intervention was assessed with and without the use of the comparator group. In the first analysis, the following model was used: Y_t = B₀ + B₁ (time_{t = 0, 1, 2, ..., 54}) + B₂ (intervention_t) + B₃ (time after intervention_{t = 0, 0, 0 ..., 1, 2, 3, ..., 35}) + e_t, where Y_t was the SSI rate in patients undergoing CABG surgery per month t, time is a continuous variable indicating time (in months) at time t from the start (t = 0 month) until the end (t = 54 months) of the observation period, intervention is an indicator variable for time t occurring before (t = 0 month) or after (t = 1 month) the change to vancomycin prophylaxis, and e_t is the error term at time t. The time after the intervention (months) is a continuous variable that counts the number of months after the intervention at time t, coded time 0 before the cap and (time–19) after the cap. Nonsignificant variables were removed from the model by using a backward elimination strategy to achieve the most parsimonious model. The Durbin-Watson statistic was calculated to test for a serial autocorrelation of the error terms in the regression models.

In the second analysis, the following model was used: Y_t = B₀ + B₁ (time_{t = 0, 1, 2, ..., 54}) + B₂ (intervention_t) + B₃ (time after intervention_{t = 0,0,0 ..., 1,2,3, ..., 35}) + B₄ (surgery type) + e_t. For the second analysis, the Y_t values were the SSI rates in patients undergoing CABG and valve replacement surgery, and surgery type was a dummy variable indicating the type of surgery for each patient. The other variables were as described above. The Durbin-Watson statistic was calculated to test for the serial autocorrelation of the error terms in the regression models. The statistical package SAS version 9.1 (SAS Institute, Cary, NC) was used for all analyses. A P value of <0.05 was considered significant.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Six thousand four hundred sixty-five patients underwent CABG (n = 4,239) or valve replacement (n = 2,226) surgery during the study period. The demographics of the patients in the CABG surgery and comparator groups are shown in Table 1. Compared to the patients undergoing valve replacement surgery, patients undergoing CABG surgery were generally nonwhite and male, had a National Nosocomial Infection Surveillance (NNIS) score of 1, and were more likely to have undergone elective surgery (P < 0.001 for each comparison). Patient demographics, the severity-of-illness measure, and surgery type did not change significantly over the course of the study period for either group. Forty-two percent of the patients undergoing valve replacement surgery underwent concomitant CABG surgery. The average number of surgeries per month was 60 ± 26, and the average rates of SSIs were 5.5 ± 3.8 infections per 100 surgeries. The average SSI rates were 6.8 ± 3.3 infections per 100 surgeries for patients undergoing CABG surgery and 4.2 ± 3.7 infections per 100 surgeries for patients undergoing valve replacement surgery (Fig. 1). Five hundred fifty-nine isolates were recovered from patients with SSIs. The pathogens included coagulase-negative Staphylococcus (41%), methicillin-susceptible S. aureus (15%), Pseudomonas aeruginosa (5.5%), MRSA (5.2%), Enterococcus species (4.1%), Proteus mirabilis (3.9%), Escherichia coli (3.9%), Klebsiella pneumoniae (3.6%), Enterobacter aerogenes (3.6%), Enterobacter cloacae (2.5%), Serratia marcescens (2.0%), and beta Streptococcus group B (1.3%). Other isolates (<1%) included Enterobacter species, Prevotella species, Salmonella species, Stenotrophomonas maltophilia, group A Streptococcus, Acinetobacter species, Citrobacter freundii, Klebsiella oxytoca, Morganella morganii, Proteus species, and Citrobacter koseri. For patients undergoing CABG surgery, 85% of the patients had isolates recovered from the sternum only, 12% of patients had isolates from their donor site only, and 3% had isolates recovered from the sternum and the donor site. During the time period of the study, greater than 85% of the coagulase-negative staphylococci were resistant to methicillin.

View this table: [in this window] [in a new window]

TABLE 1. Baseline characteristics of study patients

View larger version (34K): [in this window] [in a new window]

FIG. 1. Rates of surgical site infections in patients undergoing cardiac surgery, 2000 to 2005. On 1 October 2002, vancomycin replaced cefuroxime for surgical prophylaxis for patients undergoing CABG surgery. Patients undergoing valve replacement received vancomycin during the entire study period.

Effect of change in antibiotic prophylaxis in patients undergoing CABG surgery. No significant departure from zero was observed in the lag plot, and no significant spikes in the correlogram or the partial correlogram for SSI rates were observed for patients undergoing CABG surgery. The chi-square Q statistic used in the test for lack of fit, computed by using the Ljung-Box formula, was nonsignificant, and it was concluded that it was not necessary to correct for autocorrelation (15). The full segmented regression model showed no baseline trend and no change in the trend after the intervention. These variables were removed from the model. The most parsimonious model contained the intercept, B₀, and the effect of the intervention, B₂. On average, the monthly SSI incidence rate in patients undergoing CABG surgery decreased by 2.1 cases per 100 surgeries after the switch from cefuroxime to vancomycin (P = 0.032). The Durbin-Watson statistic was 2.405 (P = 0.92 for the hypothesis of a positive autocorrelation and P = 0.085 for the hypothesis of a negative autocorrelation).

Effect of change in antibiotic prophylaxis in patients undergoing CABG surgery by using patients undergoing valve replacement surgery as a comparator group. No significant departure from zero for estimates of r_k was observed in the lag plot, and no significant spikes in the correlogram or the partial correlogram for SSI rates were observed for the valve replacement patients. The chi-square Q statistic used in the test for lack of fit, computed by using the Ljung-Box formula, was nonsignificant, and it was concluded that it was not necessary to correct for autocorrelation (15). The full segmented regression model showed no baseline trend, no change in the trend after the intervention, and no effect of the intervention on the SSI rates for the comparator group (P > 0.05 for all parameters). The Durbin-Watson statistic was 1.902 (P = 0.22 for the hypothesis of a positive autocorrelation and P = 0.78 for the hypothesis of a negative autocorrelation). There was no statistically significant correlation between the SSI rates in patients undergoing CABG surgery and patients undergoing valve replacement surgery (r = 0.26; P = 0.055).

The parameter estimates, standard errors, and P values from the full and most parsimonious segmented regression models that included the comparator group are shown in Table 2. The full segmented regression model showed no baseline trend and no trend change after the intervention. These variables were removed from the model. The most parsimonious model contained the intercept, B₀; the effect of the intervention, B₂; and the surgery type, B₄. On average, the monthly SSI incidence rate in patients undergoing CABG surgery decreased by 2.1 cases per 100 surgeries after the switch from cefuroxime to vancomycin (P = 0.042) when patients undergoing valve replacement were used as a comparator group. The Durbin-Watson statistic was 1.687 (P = 0.060 for the hypothesis of a positive autocorrelation and P = 0.9405 for the hypothesis of a negative autocorrelation). The residuals were randomly scattered over time, indicating no autocorrelation.

View this table: [in this window] [in a new window]

TABLE 2. Parameter estimates, standard errors, and P values from the full and most parsimonious segmented regression models predicting mean SSI rates in patients undergoing CABG surgery over time by using a comparator group of patients undergoing valve replacement surgery to assess changes that occurred independently of the intervention

Microorganisms isolated according to antibiotic prophylaxis. The change in SSI rates was associated with a decrease in the incidence of infections caused by coagulase-negative Staphylococcus and MRSA, with little change in SSIs due to other gram-positive organisms or gram-negative organism (Table 3).

View this table: [in this window] [in a new window]

TABLE 3. Microorganisms isolated according to antibiotic prophylaxis from patients undergoing CABG surgery

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In August 2002, the Centers for Medicare and Medicaid Services and the Centers for Disease Control and Prevention implemented the National Surgical Infection Prevention (SIP) project (6). The goal of the SIP project was to decrease the morbidity and mortality associated with perioperative SSIs by promoting the appropriate selection and timing of use of prophylactic antibiotics. In 2004, recommendations for antimicrobial prophylaxis were published by the SIP project guidelines workgroup. This document and previous guidelines recommended cefazolin, cefuroxime, or cefamandole as the preferred antibiotics for surgical prophylaxis. Vancomycin was recommended for use only for patients with a beta-lactam allergy or in institutions with a high frequency of methicillin-resistant Staphylococcus species. This recommendation was partly based on the fact that there was no evidence that the routine use of vancomycin in hospitals with high rates of methicillin-resistant Staphylococcus species would result in fewer cases of SSIs compared to the number of cases that would occur with the routine use of cephalosporins. This was supported by a randomized clinical trial by Finkelstein et al. (10) of 885 cardiac surgery patients at a hospital with a high rate of MRSA infections (5.6 MRSA infection or colonization cases per 100 hospital admissions) given vancomycin or cefazolin antibiotic prophylaxis. The SSI rates were similar in patients given vancomycin (9.5%) or cefazolin (9.0%) (P = 0.8). However, SSIs due to methicillin-resistant Staphylococcus species were more common in the cefazolin group than in the vancomycin group (P = 0.04). As MRSA infections have been shown to be independently associated with an increased risk for hospital mortality, clinicians have argued for the inclusion of vancomycin as a first-line option for antimicrobial prophylaxis (8, 19). However, data supporting this recommendation are lacking.

In the current study, the monthly SSI incidence rate in patients undergoing CABG surgery decreased on average by 2.1 cases per 100 surgeries after the switch from cefuroxime to vancomycin (P = 0.042), when valve replacement patients were used as a comparator group and by use of an interrupted time series analysis. The decreased SSI rates were due to a decrease in the incidence of pathogens with in vitro susceptibilities to vancomycin but not cephalosporins, namely, coagulase-negative Staphylococcus and MRSA. There was little change in the incidence of SSIs due to other gram-positive or gram-negative organisms. Similar to the study of Finkelstein et al. (10) described above, we observed a decrease in the incidence of pathogens with in vitro susceptibilities to vancomycin and resistance to cephalosporins. In contrast to the study of Finkelstein et al. (10), the change to vancomycin was not associated with an increase in the incidence of methicillin-sensitive Staphylococcus, and overall, we observed a decrease in SSI rates. Why these differences were observed is uncertain, although the patient populations were slightly different in the two studies. Approximately 20% of the patients in the study of Finkelstein et al. (10) underwent non-CABG cardiac surgery, whereas all patients in the current study underwent CABG surgery. We have previously shown that the timing of vancomycin administration in relation to the surgery start time is an important variable for the prevention of SSIs (11). We did not have complete timing data for the time period of this study and thus were not able to evaluate this variable in our study. From our previous study, the timing of drug administration was similar in patients undergoing CABG surgery and patients undergoing valve replacement surgery. Thus, any risk of SSIs due to the inappropriate timing of vancomycin administration would be similar between the two groups. A trend for more serious infections was observed in the study of Finkelstein et al. (10), and the unblinded nature of the study was mentioned as an important limitation. The results from our study are supported by the findings of a previous double-blind study by Maki et al. (17) of 320 adults undergoing cardiac or vascular surgeries that demonstrated the superiority of vancomycin to cefazolin for antibiotic prophylaxis for the prevention of thoracic SSIs other than mediastinitis.

The strengths of this study include a large sample size (>6,400 patients) over an extended period of time (4 years) and an analysis that used a comparator group to evaluate other factors that may have influenced changes in SSI rates. As with all studies, certain limitations exist. We were not able to evaluate other important variables, such as the rate of compliance with the timing of antibiotic administration and the type of SSI, due to a lack of data for these variables. However, there were no data to suggest that these important variables changed over the time period of this study. Segmented regression analysis assumes a linear trend in the outcome with each segment, and it is unknown whether the change in antibiotic prophylaxis may have caused a nonlinear effect. The comparator group was different from the group of patients undergoing CABG surgery in certain baseline characteristics, such as the gender and the race distribution, the average NNIS score, and the surgery type; but these baseline characteristics did not change over the time period of the study for either group. Although the use of a comparator group allowed us to assess changes in SSI rates that were independent of the intervention, other unknown variables may still have influenced the change in SSI rates. The surgical procedure for patients undergoing valve replacement surgery is not identical to that for patients undergoing CABG surgery, which could potentially limit the utility of the comparator group. We were also not able to assess changes in donor site infections due to low incidence rates in both groups. Even with over 6,400 patients, this study did not have an adequate power to perform an organism-specific statistical analysis. Finally, it should be noted that the first-line use of vancomycin should be reserved only for institutions with high rates of SSIs due to methicillin-resistant Staphylococcus species. In a previous study of 100 patients undergoing cardiac surgery, we noted a 4% incidence of VRE rectal colonization in patients given vancomycin prophylaxis (14). Infections due to VRE are associated with considerable morbidity, and the transfer of the vanA Enterococcus resistance gene to S. aureus, with the associated S. aureus resistance to vancomycin, makes the prevention of VRE infections of upmost clinical importance (22, 25). It is also possible that vancomycin prophylaxis may cause a decreased susceptibility of S. aureus to vancomycin. Patients infected with S. aureus strains with reduced susceptibility to vancomycin have been shown to have poor outcomes, and further research will be necessary to test whether vancomycin prophylaxis may exacerbate this problem (16, 20). However, in institutions with a high incidence of methicillin-resistant Staphylococcus species, this study provides evidence for the clinical efficacy of vancomycin prophylaxis for the prevention of postoperative SSIs.

Conclusion. In a study of 6,465 patients undergoing cardiac surgery, the average monthly SSI incidence rate in patients undergoing CABG surgery decreased by 2.1 cases per 100 surgeries after the switch from cefuroxime to vancomycin (P = 0.042) by using patients undergoing valve replacement surgery as a comparator group. The change in antibiotic prophylaxis was associated with a decrease in SSIs due to coagulase-negative Staphylococcus species and MRSA, with little change in the rates of SSIs due to gram-negative or other gram-positive species.

 
No financial support was obtained for support of this study.

None of the authors reports a potential conflict of interest.

 
* Corresponding author. Mailing address: College of Pharmacy, Texas Medical Center, University of Houston, 1441 Moursund Street, Houston, TX 77030. Phone: (713) 795-8386. Fax: (713) 795-8383. E-mail: kgarey{at}uh.edu

Published ahead of print on 19 November 2007.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Anonymous. 2001. Antimicrobial prophylaxis in surgery. Med. Lett. Drugs Ther. 43:92-97.[Medline]
2
2. Anonymous. 2004. National Nosocomial Infections Surveillance (NNIS) System report, data summary from January 1992 through June 2004, issued October 2004. Am. J. Infect. Control 32:470-485.[CrossRef][Medline]
3
3. Ansari, F., K. Gray, D. Nathwani, G. Phillips, S. Ogston, C. Ramsay, and P. Davey. 2003. Outcomes of an intervention to improve hospital antibiotic prescribing: interrupted time series with segmented regression analysis. J. Antimicrob. Chemother. 52:842-848.[Abstract/Free Full Text]
4
4. Bolon, M. K., M. Morlote, S. G. Weber, B. Koplan, Y. Carmeli, and S. B. Wright. 2004. Glycopeptides are no more effective than beta-lactam agents for prevention of surgical site infection after cardiac surgery: a meta-analysis. Clin. Infect. Dis. 38:1357-1363.[CrossRef][Medline]
5
5. Box, G. E. P., and G. C. Tiao. 1975. Intervention analysis with applications to economic and environmental problems. J. Am. Stat. Assoc. 70:70-92.[CrossRef]
6
6. Bratzler, D. W., and P. M. Houck. 2004. Antimicrobial prophylaxis for surgery: an advisory statement from the National Surgical Infection Prevention Project. Clin. Infect. Dis. 38:1706-1715.[CrossRef][Medline]
7
7. Centers for Disease Control and Prevention. 1997. Interim guidelines for prevention and control of staphylococcal infection associated with reduced susceptibility to vancomycin. Morb. Mortal. Wkly. Rep. 46:626-628, 635.[Medline]
8
8. Chenoweth, C. E., D. D. DePestel, and R. L. Prager. 2005. Are cephalosporins adequate for antimicrobial prophylaxis for cardiac surgery involving implants? Clin. Infect. Dis. 41:122-123.[Medline]
9
9. Diggle, P. J. 2000. Time series. A biostatistical introduction, 5th ed. Oxford University Press, New York, NY.
10
10. Finkelstein, R., G. Rabino, T. Mashiah, Y. Bar-El, Z. Adler, V. Kertzman, O. Cohen, and S. Milo. 2002. Vancomycin versus cefazolin prophylaxis for cardiac surgery in the setting of a high prevalence of methicillin-resistant staphylococcal infections. J. Thorac. Cardiovasc. Surg. 123:326-332.[Abstract/Free Full Text]
11
11. Garey, K. W., T. Dao, H. Chen, P. Amrutkar, N. Kumar, M. Reiter, and L. O. Gentry. 2006. Timing of vancomycin prophylaxis for cardiac surgery patients and the risk of surgical site infections. J. Antimicrob. Chemother. 58:645-650.[Abstract/Free Full Text]
12
12. Harvey, A. 1996. Intervention analysis with control groups. Int. Stat. Rev. 64:313-328.
13
13. Horan, T. C., R. P. Gaynes, W. J. Martone, W. R. Jarvis, and T. G. Emori. 1992. CDC definitions of nosocomial surgical site infections, 1992: a modification of CDC definitions of surgical wound infections. Infect. Control Hosp. Epidemiol. 13:606-608.[Medline]
14
14. Kachroo, S., T. Dao, F. Zabaneh, M. Reiter, M. T. Larocco, L. O. Gentry, and K. W. Garey. 2006. Tolerance of vancomycin for surgical prophylaxis in patients undergoing cardiac surgery and incidence of vancomycin-resistant enterococcus colonization. Ann. Pharmacother. 40:381-385.[Abstract/Free Full Text]
15
15. Ljung, G. M., and G. E. P. Box. 1978. On a measure of lack of fit in time series models. Biometrika 65:297-303.[Abstract/Free Full Text]
16
16. Maclayton, D. O., K. J. Suda, K. A. Coval, C. B. York, and K. W. Garey. 2006. Case-control study of the relationship between MRSA bacteremia with a vancomycin MIC of 2 microg/mL and risk factors, costs, and outcomes in inpatients undergoing hemodialysis. Clin. Ther. 28:1208-1216.[CrossRef][Medline]
17
17. Maki, D. G., M. J. Bohn, S. M. Stolz, G. M. Kroncke, C. W. Acher, and P. D. Myerowitz. 1992. Comparative study of cefazolin, cefamandole, and vancomycin for surgical prophylaxis in cardiac and vascular operations. A double-blind randomized trial. J. Thorac. Cardiovasc. Surg. 104:1423-1434.[Abstract]
18
18. McDowall, D. 1980. Interrupted time series analysis. Sage, Beverly Hills, CA.
19
19. Mekontso-Dessap, A., M. Kirsch, C. Brun-Buisson, and D. Loisance. 2001. Poststernotomy mediastinitis due to Staphylococcus aureus: comparison of methicillin-resistant and methicillin-susceptible cases. Clin. Infect. Dis. 32:877-883.[CrossRef][Medline]
20
20. Moise-Broder, P. A., G. Sakoulas, G. M. Eliopoulos, J. J. Schentag, A. Forrest, and R. C. Moellering, Jr. 2004. Accessory gene regulator group II polymorphism in methicillin-resistant Staphylococcus aureus is predictive of failure of vancomycin therapy. Clin. Infect. Dis. 38:1700-1705.[CrossRef][Medline]
21
21. Moran, G. J., A. Krishnadasan, R. J. Gorwitz, G. E. Fosheim, L. K. McDougal, R. B. Carey, and D. A. Talan. 2006. Methicillin-resistant S. aureus infections among patients in the emergency department. N. Engl. J. Med. 355:666-674.[Abstract/Free Full Text]
22
22. Murray, B. E. 2000. Vancomycin-resistant enterococcal infections. N. Engl. J. Med. 342:710-721.[Free Full Text]
23
23. Rice, L. B. 2006. Antimicrobial resistance in gram-positive bacteria. Am. J. Infect. Control 34:S11-S19.[CrossRef][Medline]
24
24. Spelman, D., G. Harrington, P. Russo, and S. Wesselingh. 2002. Clinical, microbiological, and economic benefit of a change in antibiotic prophylaxis for cardiac surgery. Infect. Control Hosp. Epidemiol. 23:402-404.[CrossRef][Medline]
25
25. Tenover, F. C., L. M. Weigel, P. C. Appelbaum, L. K. McDougal, J. Chaitram, S. McAllister, N. Clark, G. Killgore, C. M. O'Hara, L. Jevitt, J. B. Patel, and B. Bozdogan. 2004. Vancomycin-resistant Staphylococcus aureus isolate from a patient in Pennsylvania. Antimicrob. Agents Chemother. 48:275-280.[Abstract/Free Full Text]
26
26. Wagner, A. K., S. B. Soumerai, F. Zhang, and D. Ross-Degnan. 2002. Segmented regression analysis of interrupted time series studies in medication use research. J. Clin. Pharm. Ther. 27:299-309.[CrossRef][Medline]
27
27. Yaffee, R. A., and M. McGee. 2000. Introduction to time series analysis and forecasting: with applications in SAS and SPSS. Academic Press, Inc., San Diego, CA.

---

Antimicrobial Agents and Chemotherapy, February 2008, p. 446-451, Vol. 52, No. 2
0066-4804/08/$08.00+0     doi:10.1128/AAC.00495-07
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Garey, K. W. Articles by Davis, B. R. Search for Related Content PubMed PubMed Citation Articles by Garey, K. W. Articles by Davis, B. R.