Investigation of a Cluster of Pediatric Cardiac [PC] Surgical Site Infections [SSI] Leads to Discovery of Significant Difference in PC SSI Rates by Specific Procedure [SP]

ISSUE: Pediatric cardiac surgery is becoming increasingly complicated and involving more compromised children. In our 160 bed Childrenʹs Center, we perform 200 PC surgeries annually. For the past 2 years, we have tracked SSI rates among high-risk PC patients. SSI are defined using NNIS criteria. Because no national benchmarks for PC SSI rates exist, we compare our SSI rate to those in the past. In July 2005, we investigated a cluster of PC SSI which did not increase our mean infection rate [5%]. The cluster was identified because patients had their surgery during the same week. PROJECT: An outbreak investigation was undertaken and included a literature search, a line list that encompassed potential risk factors [type of hardware, underlying diagnosis, age, length of surgery] and potential commonalities [infecting organism, operating room (OR) number] for the cluster cases, and direct observation in the OR. In addition, PC cases for the past 18 months were re-analyzed. As part of the re-analysis, we examined our SSI rates by SP as defined by NNIS ICD9 procedure codes. RESULTS: We were unable to detect any commonalities that would account for the SSI cluster. There were no digressions from best practices noted during our OR observation. However, the analysis of our infection rates by SP using the NNIS ICD9 codes revealed that our infection rate for children having cardiac vascular surgery [CVS] was significantly higher [10.4%] compared to those having surgery on the valves or septum of the heart [2.6%] [P = 0.03]. LESSONS LEARNED: Even though our PC SSI rate [5.49%] for the first 6 months of 2005 was comparable to our cummulative mean infection rate [5%], looking at the data in a different way revealed a problem we were not aware of. Now that we are aware of the increased rate in our CVS patients, we can work with our Cardiac Surgery Team to develop appropriate interventions. This may suggest that risk stratification methods for children are different and needs further validation. As we strive to achieve zero tolerance for SSI, we review data in a variety of ways to assure that potential risk factors are not overlooked.