How slow is too slow? Correlation of operative time to complications: an analysis from the Tennessee Surgical Quality Collaborative

Abstract

Background: The Tennessee Surgical Quality Collaborative analyzes NSQIP data from 21 participating hospitals. The Tennessee Surgical Quality Collaborative has reduced surgical complications, but causative factors are unclear. We sought to correlate surgical duration with complications to reveal mitigating strategies.

Study design: Risk-adjusted Tennessee Surgical Quality Collaborative data on 104,632 general and vascular cases had a standard duration for 35 procedures (eg, breast, colectomy) calculated and NSQIP outcomes complication rates recorded. We derived a marginal time risk for each extra hour of operative time and reported per 1,000 cases.

Results: Procedures taking <95% upper confidence standard time limit (n = 99,741) were deemed "not long" and had significantly fewer urinary tract infections, organ-space surgical site infection, sepsis/septic shock, prolonged intubation, and pneumonia. "Long" cases had increased rates of these complications and also deep venous thrombosis, deep incisional infection, and wound disruption. Per 1,000 cases, there were 116 occurrences per operating room hour. Surgical site infections occurred in 14.4/1,000 cases per hour; risk started at 42 minutes of operative time. Death, pneumonia, and prolonged intubation saw their risks begin before the operation. The highest marginal time risk was for sepsis, occurring 16.6 times per additional hour of operative time over standard. Studying only the 25,146 clean procedures, a significant correlation (p < 0.001) to operation duration persisted, despite an occurrence incidence of 4.5%.

Conclusions: Duration of operation correlates with complications and time longer than a statewide established standard carries higher risk. To reduce risk of complications, these data support expeditious surgical technique and preoperative pulmonary training, and offer accurate outcomes assessment for patient counseling based on case duration. These data can be used directly to counsel individual surgeons to improve outcomes.