MSS15: COMPARISON OF OUTCOMES BETWEEN THE NATIONAL SURGICAL QUALITY IMPROVEMENT PROGRAM (NSQIP) AND AN EMERGENCY GENERAL SURGERY REGISTRY AT A MILITARY MEDICAL CENTER
Robert W Despain, MD1, William J Parker, MD1, Angela T Kindvall, BSN2, Peter A Learn, MD1, Eric A Elster, MD1, Elliot M Jessie, MD1, Carlos J Rodriguez, DO3, Matthew J Bradley, MD1; 1Department of Surgery, Uniformed Services University of Health Sciences and Walter Reed National Military Medical Center, 2Department of Surgery, Walter Reed National Military Medical Center, 3John Peter Smith Hospital, Fort Worth, TX
Objective: The National Surgical Quality Improvement Program (NSQIP) has become the gold standard for assessing practice improvement in surgery. At our tertiary military hospital, we implemented an emergency general surgery (EGS) registry to actively track and review 100% of EGS cases. NSQIP abstracts 20% of cases in our institution. We compared our EGS registry to the NSQIP data. We hypothesize NSQIP may not fully represent EGS cases and outcomes.
Methods: A formal EGS Process Improvement Program, complete with a full registry, was implemented in 2016. From 2016 to 2018, the four most common EGS procedures were queried in the registry which included appendectomy, cholecystectomy, surgery for small bowel obstruction (SBO), and hernia repair. The CPT codes for these cases were used to identify cases within the NSQIP database and the EGS cases were identified from that search. Case volume and outcomes were compared between the EGS registry and the cases abstracted by NSQIP.
Results: In 2016, the EGS registry identified 128 cases with 15 (12%) patients having a complication. 53% (8/15) were associated with complications from surgery for SBO. NSQIP abstracted 14% (18/128) of EGS cases with 17% (3/18) patients having a complication, all related to cholecystectomies. These three complications were not considered adverse events within the EGS registry. Two of the complications were a hospital readmission for a medical problem unrelated to cholecystectomy and the third was a negative ERCP for a possible retained common bile duct stone. In 2017, the EGS registry identified 93 cases with 11 (12%) patients having a complication. The majority (7/11, 64%) of complications were associated with surgery for SBO. NSQIP abstracted 22% (20/93) of EGS cases with zero patients having a complication. In 2018, the EGS registry identified 79 cases, with nine (11%) patients having a complication. 67% (6/9) of complications were related to cholecystectomies. NSQIP abstracted 15% (12/79) of EGS cases with zero patients having a complication. Within the three years, one patient was abstracted by NSQIP as not having an adverse event, however, within the EGS registry this patient was found to have an unplanned ICU admission for diabetic ketoacidosis following laparoscopic appendectomy.
Conclusion: NSQIP did not adequately capture EGS outcomes. In two of the three years, NSQIP did not identify a single complication from EGS cases. Resources should be allocated to improve the NSQIP rate of abstraction. Alternatively, institutions should implement an EGS Process Improvement Program complete with a registry.