N. Pettit2, M. Leiber1, E. Buggie1, M. S. O’Mara1,2 1Grant Medical Center,Trauma And Acute Care Surgery,Columbus, OHIO, USA 2Ohio University Heritage College Of Medicine,Columbus, OHIO, USA
Introduction:
Few studies exist comparing percutaneous bedside tracheostomy (PBT) versus the open surgical tracheostomy (OST) procedure. In this study we examined patients that underwent one of these procedures and we hypothesized that those patients undergoing PBT may have more post-operative complications but decreased costs.
Methods:
579 patients that underwent either PBT or OST from 2008-2014 were retrospectively evaluated. Outcomes were post-operative complications, mortality, length of stay (LOS), and charges. Patients were divided into groups by tracheostomy technique (open or percutaneous) and location (ICU or OR): open-OR, percutaneous-ICU, percutaneous-OR. Variables examined and controlled for in the study were admitting service (trauma vs. medicine), age, body mass index (BMI), and the Charlson Comorbidity Index (CCI).
Results:
Using any-post-op complication as our first outcome there was a significant difference between the 3 tracheostomy groups (p = 0.001). No significant difference existed between open-OR and perc-OR (8.7% vs 7.4% complication rate, respectively, p = 0.70), however significant differences were found between perc-ICU vs perc-OR (19.1% vs 7.4%, odds ratio 3.6, p = 0.028) and between perc-ICU vs open-OR (19.1% vs 8.7%, odds ratio 2.5, p = 0.005). A logistic regression model determined that tracheostomy group (p = 0.011) and service (medicine vs. trauma, p = 0.017) were significant predictors of any post-operative complication. Medicine cases were 2.3 times more likely to have post-op complications than trauma cases (95% CI for odds ratio 1.2 – 4.6). While the CCI is significantly different between the medicine and trauma patients (mean of 3.2 vs. 0.9, respectively, p < 0.0001), CCI was not a predictor of post-op complications. Mortality was not significantly different between the various procedures, and only age was a significant predictor of mortality (p = 0.0001). The service and age were significant predictors of LOS (p = 0.006 and p = 0.002, respectively), with a mean LOS of 27.1 and 22.8 days for trauma and medicine, respectively. Lastly, there are no significant differences in charges between the different procedures, and only the days to procedure was a predictor of the total charge to the patient (p < 0.0001).
Conclusion:
Those patients that had a tracheostomy in the OR, whether perc or open, had less complications than those done in the ICU, and without an identifiable difference in the total charge. Patients with tracheostomies placed by a percutaneous method in the ICU were 3.6 times more likely to have post-op complications than were patients with the same procedure done in the operating room. Furthermore, those patients admitted to medicine, also had more complications with a shorter length of stay. Understanding these results should lead the surgeon to performing tracheostomies in higher risk patients in the operating room, thereby decreasing complications without increasing overall charges.