Patient, operative, and hospital factors were found to be significant predictors of the risk of surgical site infection in patients who underwent open lower extremity bypass procedures, according to the results of a retrospective data analysis.
The study assessed the outcomes of 3,033 patients who underwent elective or urgent open LEB procedures between January 2012 and June 2015 using data from the Blue Cross Blue Shield of Michigan Cardiovascular Consortium Vascular Intervention Collaborative (BMC2 VIC), a statewide cardiovascular consortium of 35 hospitals, according to Frank M. Davis, MD, and his colleagues at the University of Michigan, Ann Arbor.
Demographic information, medical history, laboratory test results before and after the procedure, procedural indication, procedural urgency, technical details of procedures, and associated complications were assessed for each patient. Women comprised 31% of patients, the average patient age was 66 years, and 83% of the population was white (J Vasc Surg. 2017 Jun;65[6]:1769-78).
Among all of the patients treated, 320 developed SSIs and 2,713 did not. The procedural indications included one or more of the following: claudication (72%), rest pain (50.5%), ulcer/gangrene (32.4%), or acute limb ischemia (15.1%). Antibiotics were appropriately administered to 97% of the patients, according to the researchers, “demonstrating high compliance across the BMC2 VIC.”
- Patient factors: As indicated by previous studies, obesity (odds ratio, 1.78), dialysis dependence (OR, 4.33), and hypertension (OR, 4.29) conferred a significant increased risk of SSI after LEB, according to Dr. Davis and his colleagues. In addition, however, they found that previous vascular surgery (OR, 1.57), previous percutaneous coronary intervention (OR, 1.47), use of antiplatelet medication (OR, 4.29), and low Peripheral Artery Questionnaire symptom severity (OR, 1.48) were significant independent predictors of SSI.
- Operative factors: Prolonged procedural length (OR, 2.95), iodine-only antiseptic skin preparation (OR, 1.73), and high peak intraoperative glucose (defined as a peak glucose greater than 180 mg/dL; OR, 1.99) were significant independent predictors of SSI. However, concomitant stent placement was found to be significantly predictive (OR, .38), “perhaps due to improvement in regional and subcutaneous vascular flow after the intervention,” the researchers suggested.
- Hospital factors: Larger overall hospital size (OR, 2.22) and major teaching center (OR, 1.66) were associated with increased risk of SSI. “Interestingly, we did not find an association with SSI and the hospital annual volume or the hospital urgent/emergent procedure rate,” the researchers added.
SSIs were not found to be significantly associated with a difference in 30-day mortality. However, they were significantly associated with an increased rate of several postoperative morbidities, including transfusion, lymph leak, major amputation, and open surgical bypass revision at or within 30 days of the index operation, according to Dr. Davis and his colleagues.
“Although some factors, such as patients comorbidities, are not modifiable, others represent areas for quality improvement in at-risk patients,” the researchers indicated. “Diligence should be devoted to decreasing operative length, controlling intraoperative glucose levels, and avoiding iodine-only skin preparation to decrease the rate of SSIs and its numerous associate morbidities in vascular surgery patients.”
In discussing the issue of antiplatelet medication being an indicator of increased risk, the authors pointed out that it was a hitherto unreported factor in the vascular literature, and of concern because, “as expected, a high percentage of patients (78.7%) were taking antiplatelet medication at the time of their LEB.”
Because the association of antiplatelet medication with SSIs was independent of the need for operative transfusion or the need for repeat intervention, the researchers speculated that “all antiplatelet agents have the theoretical potential to diminish activation-dependent platelet immune functions.” They referred to previous studies showing that clopidogrel was associated with significantly higher clinical rates of infection, particularly pneumonia.
Limitations cited for the study were the retrospective nature of the database analysis, the possibility of confounders not assessed in the data, and the fact that outcomes were limited to 30-day events, which would not take into account longer-term graft failure or mortality.
The authors reported having no conflicts of interest with regard to the study.