Morbidity and mortality associated with arterial surgery site infections by resistant microorganisms

Background: Surgical site infection is a severe complication of peripheral vascular surgery with high morbidity and mortality rates. Objective: To evaluate the morbidity and mortality of infections of peripheral artery surgery sites caused by resistant microorganisms. Methods: This was a prospective study of a cohort of patients who underwent peripheral artery revascularization procedures and developed surgical site infections between March 2007 and March 2011. Results: Mean age was 63.7 years; males accounted for 64.3% of all cases. The overall prevalence of bacterial resistance to antimicrobials was 65.7%. The most common microorganism identified was Staphylococcus aureus (30%). Comparison of the demographic and surgical characteristics of both subsets (resistant versus non-resistant) detected a significant difference in length of preoperative hospital stay (9.3 days vs. 3.7 days). The subset of patients with infections by resistant microorganisms had higher rates of reoperation, lower numbers of limb amputations and lower mortality, but the differences compared to the subset without resistant infections were not significant. Long-term survival was similar. Conclusions: This study detected no statistically significant differences in morbidity or mortality between subsets with surgical wound infections caused by resistant and not-resistant microorganisms.


INTRODUCTION
Surgical site infection is a severe complication of peripheral vascular surgery.Surgical site infection occurs in 0.9-22% of arterial vascular surgeries, and recent series have reported rates of around 3%. 1,2 Deep infection with involvement of the wall of the operated artery or the implanted graft occurs in 1.2-13% of arterial surgeries. 2,3França et al. reported a 4.6% incidence of synthetic graft infection in peripheral vascular surgery. 2 This type of infection is more frequent in emergency surgeries and inguinal dissections and when vascular grafts are used in the femoral region or via a subcutaneous route.6][7] Later infections are often caused by Coagulase-negative Staphylococcus.Gram-negative bacterial infections are very destructive and are often caused by Pseudomonas aeruginosa. 8,9he recent occurrence of resistant strains has raised new concerns about antimicrobial prophylaxis and treatment and morbidity and mortality rates.Methicillin-resistant Staphylococcus aureus (MRSA) is currently the most frequently isolated pathogen in complex and severe infections related to vascular surgery. 6Prolonged prophylaxis and indiscriminate use of antimicrobials contributed to the development of bacterial resistance. 7,10In turn, the increased incidence of multi-resistant pathogens in vascular surgeries contributes to increased morbidity and mortality rates. 6,9,11,12However, some authors found that no increase in morbidity 13 or mortality was associated with MRSA infections. 14his study was conducted with the aim of determining the impact of bacterial resistance to antibiotics on morbidity and mortality from surgical site infections after peripheral artery surgery.

METHODS
This was a cohort study of patients who underwent peripheral artery revascularization surgery from March 2007 to March 2011 and who developed surgical site infections during the postoperative period.Sample selection was consecutive.Demographic data, comorbid conditions, surgical details, microbiology and characteristics of wound infection and hospital stay were reviewed.Presence or absence of bacterial resistance to antibiotics was used to stratify the sample.Data on readmissions and repeat operations during the follow-up period were also recorded.The primary outcome was mortality; secondary outcomes were reoperation, graft excision and lower limb amputation.Patients who were subjected to limb revascularization procedures had critical limb ischemia.This study was approved by the ISCMPA ethics committee.
Bacterioscopic examination, cultures and antimicrobial susceptibility tests were conducted for all cases.The material collected comprised surgical site secretions, tissues or graft fragments (if affected) with clinical diagnoses of infection.All material was collected by the same researcher and analyzed by the same laboratory, following Clinical and Laboratory Standards Institute procedures. 15Polymicrobial samples and samples with negative culture results were excluded from the study.Patients with negative cultures were also excluded from the study.All patients in this study received a prophylactic antibiotic prior to skin incision (cephalosporin).
Bacterial resistance to antimicrobials was defined as follows: Staphylococcus aureus -resistance to oxacillin (all beta-lactams), Coagulase-negative Staphylococcus -resistance to oxacillin (all betalactams), Klebsiella pneumoniae, Enterobacter, Escherichia coli and Proteus mirabilis -resistance to amikacin, ceftriaxone, cefepime or ciprofloxacin, Pseudomonas aeruginosa -resistance to imipenem or only sensitivity to imipenem but resistant to other antimicrobials, Enterococcus feacalis -resistance to vancomycin or teicoplanin, and Streptococcus viridans -resistance to penicillin.
Comparisons of the results for quantitative variables were performed using Student's t test and the Mann Whitney test.Categorical variables were compared using chi-square tests and Fisher's exact test.Symmetry of data distribution was evaluated using the Kolmogorov-Smirnov test (P<.10).For survival analysis, the Kaplan-Meier method was used to identify major factors associated with mortality and then we constructed a Cox multiple regression model (Wald method) with death as the dependent variable.Variables with significant or borderline (P<.10) differences according to the bivariate analysis were included in this model.Statistical tests with a two-tailed p-value < .05were considered significant.Analyses were performed using SPSS 17 (Statistical Package for Social Sciences) from SPSS Inc, Chicago, IL.

Baseline characteristics
During the study period, seven hundred patients were treated at our department using peripheral artery surgery.Twenty patients with polymicrobial and negative cultures were excluded from the study.Data from seventy patients who developed surgical site infection after peripheral artery surgery were analyzed during the study, stratified into two subsets: those infected by resistant bacteria and those infected by non-resistant bacteria.The patients' mean age was 63.7 years, and males predominated (64.3%).Patients' demographic data and differences between subsets are summarized in Table 1.No significant differences in demographic characteristics between the two subsets were detected.Mean preoperative hospital stay was significantly longer for patients with infection due to resistant microorganisms (P=0.001).No significant differences between groups were detected for the variables use of therapeutic antimicrobials, graft implantation, type of surgery or Szilagyi 16 classification.The principal surgical characteristics are shown in Table 1.

Microbiology
The most frequent microorganism isolated from surgical site infections was Staphylococcus aureus, present in 30% of all cases and the most prevalent bacteria in both subsets.Overall prevalence of bacterial resistance to antimicrobials was 65.7%.Infections due to Coagulase-negative Staphylococcus and Escherichia coli were significantly more common (P=0.001) in the group with non-resistant bacteria; whereas infection due to Enterobacter was statistically more common (P=0.001) in the group with resistant bacteria.Table 2 lists all of the microbiological analysis results.

Morbidity and mortality
The group of patients with resistant surgical site infections exhibited higher rates of repeat surgery, but lower rates of limb amputation and mortality, although the differences were not significant.These data are summarized in Table 3. Multivariate analysis (logistic regression) of parameters that were statistically significant in bivariate analyses did not detect significant associations with death (P>0.05).
One-year survival rates for the non-resistant and resistant subsets were 87.9% and 79.7% respectively and 5-year survival rates were 65.9% and 47.6% respectively.Analysis of Kaplan-Meier curves did not detect significant differences between the subsets with and without resistant bacteria (Mantel-Cox log rank test = 0.206, P=0.652).There were therefore no statistically significant differences between mean time of survival estimates for the subsets with antimicrobial resistant (42.7 months ± 3.2 days) and non-resistant (48.7 months ± 6.6 days) pathogens (Figure 1).

DISCUSSION
Resistant bacterial strains raise new concerns about antimicrobial prophylaxis and treatment and morbidity and mortality rates within the context of peripheral vascular surgery. 17The importance of bacteria that are resistant to antibiotics lies in the fact that they have become much more prevalent in hospitals, 18,19 reducing the efficacy of antimicrobial treatments, 20 disseminating resistance, 21 increasing medical costs 22 and raising morbidity and mortality rates among surgery patients. 6,9,11,12The main factors involved in development of surgical site infections are graft implantation, local complications, age, comorbidities, [23][24][25][26][27][28] grafts in extra-anatomical positions, prolonged preoperative hospital stays, emergency surgery, extended periods in surgery, concomitant gastrointestinal surgery and arterial reoperation. 3,29In this study, patients had a high mean age, were predominantly male and all had comorbidities, most of which were multiple, which are typical characteristics of patients undergoing peripheral artery surgery.Patients who developed infections caused by resistant microorganisms had longer preoperative hospital stays, which the literature shows can be related to increased infection rates and the prevalence of resistant bacteria. 30One explanation for this long preoperative stay could be difficulties with availability of hospital beds and the need to conduct extensive investigations before the revascularization procedure.Other baseline characteristics were similar for both subsets in this study.
Many authors agree that the increased incidence of multi-resistant pathogens and more aggressive strains in vascular surgery infections contributes to high rates of morbidity and mortality, 6,9,11,12 with mortality rates of up to 75% reported. 6,9High rates of amputation (40%) and prosthesis excision (11.5%) and longer duration hospital stays are reported when patients are develop surgical site infections that have resistance. 12Cowie et al. showed that MRSA infections in patients undergoing vascular procedures increased the risk of mortality and clinical complications, including the need for hospitalization in the intensive care unit, repeated hospitalizations and interventions, and longer hospitalization. 11 study conducted by Taylor et al. showed that infections caused by Staphylococcus aureus resistant to beta-lactam antibiotics are associated with high morbidity rates, longer hospital stays (29.6 days vs. 22.7 days), higher rates of lower limb amputation (40%) and a greater probability that an infected graft will have to be removed (7.7%). 12However, some authors found no increase in morbidity 13 or mortality associated with MRSA infection. 14n our study, patients with resistant surgical wound infections had higher rates of repeat surgeries, and lower limb amputation and mortality; but these results were not statistically significant.On the other hand, the group with non-resistant infections had a high rate of graft excision, which was probably associated with early and aggressive treatment.Multivariate analysis showed that mortality was not associated with bacterial resistance, after adjustment of baseline factors.Patients with non-resistant infections had higher survival rates after one-year and five-year follow-up, but this result also failed to attain significance.Long-term survival was statistically similar for both subsets.This finding could be due to the multiple comorbidities, advanced diseases and high cardiovascular risk levels of these patients.These results shows that vascular surgery patients with resistant surgical site infections may suffer worse outcomes, especially over the short-term follow-up period, irrespective of correct management and care, with high rates of complications and death.
Study limitations include the sample size, lack of uniformity in the infection samples collected, exclusion of cases with polymicrobial infections and with negative cultures, and the sample drawn from a specific and limited population, consisting of patients with multiple advanced diseases who face barriers to access of appropriate medical support.It is probable that statistical analysis of a larger sample could detect more significant results.
This study detected no statistical differences in morbidity or mortality between a subset of patients with surgical wound infections caused by resistant microorganisms and another with non-resistant infections.Patient follow-up revealed a trend towards poorer outcomes among patients with resistant infections, but over the long term survival was similar in both subsets.

Figure 1 .
Figure 1.Survival curve (Kaplan-Meier) for patients with resistant infections and non-resistant infections.

Table 1 .
Demographic and surgical characteristics of patients with surgery wound infections.
Me: Median, concentration of 50% of the sample with values less than or equal to the median.

Table 2 .
Distribution of organisms isolated from surgical wound infections in vascular patients.

Table 3 .
Analysis of primary and secondary outcomes (mortality and morbidity) for patients with resistant and non-resistant bacterial infections.