A Matched Comparison of Patient Outcomes in Emergency General Surgery Conditions: Understanding Variability in...

Emergency general surgery (EGS) is a subset of acute care surgery that can be managed surgically and conservatively. Various factors influence decisions regarding operative or nonoperative management. Our study aimed to identify EGS patients who underwent surgical intervention and compare their outcomes to those who underwent nonoperative management.

Data from patients aged ≥ 18 years with primary index admission and EGS conditions defined by the American Association for the Surgery of Trauma from Pakistan's first Joint Commission International Accredited Center were analyzed from 2010 to 2019. The primary exposure was surgical intervention. Differences in inpatient mortality, complications, and length of stay (LOS) were compared using logistic and generalized-linear models after coarsened exact matching.

Records from 32,280 primary index admissions showed a higher number of younger patients (mean 47.83 vs. 52.40 years) and no preexisting conditions (60.22% vs. 42.30%) in the operated group compared to the nonoperated group. There were relatively higher uninsured individuals in the nonoperated group compared to the operated group (84.36% vs. 74.22%), respectively. Risk-adjusted differences in outcomes showed higher odds of complication (AOR 1.34 and 95% CI 1.20 and 1.48) and prolonged LOS (β 0.78 and 95% CI 0.65 and 0.91) in operated patients. The risk-adjusted observed/expected rates showed lower inpatient mortality rates in operated patients across all EGS diagnoses.

The results showed that patients who underwent surgery had lower risk-adjusted mortality even though they had more complications across all EGS diagnoses, which highlights the urgent need to improve surgical access in developing countries due to higher uninsured individuals in the nonoperated group. Also, the findings stress the need for risk stratification and further studies to mitigate risks and optimize patient recovery based on patient-level factors.