Predictors of Perioperative Outcomes in Bariatric Surgery: The Impact of Obesity Severity, Obstructive Sleep Apnea, and Comorbidities on Hospital Length of Stay and Complication Risk

Abstract Introduction: Bariatric surgery is the most effective obesity treatment. This study aims to assess patient characteristics and perioperative outcomes in bariatric surgery to enhance postoperative care. Methods Used: We conducted a retrospective chart review among patients who underwent bariatric surgery at a tertiary care hospital in NJ, USA. Results: This study included 119 patients who underwent bariatric surgery, with participants having a mean age of 40.07 years (SD = 11.31), 84.87% were females, and a mean BMI of 39.72 kg/m2 (SD = 7.63). The average length of stay was 2.08 days (SD = 1.27). 57.14% of individuals had a history of OSA, 36.13% had hypertension, and 26.05% had diabetes mellitus. Pearson correlation analysis revealed a weak, non-significant positive correlation between OSA history and length of stay (LOS) (r = 0.13, p = 0.15). However, a weak but statistically significant positive correlation was observed between OSA history and BMI (r = 0.18, p = 0.04), indicating that individuals with OSA tend to have a slightly higher BMI. Logistic regression models confirmed that diabetes (odds ratio [OR] = 3.51, p = 0.014) significantly increases the odds of having a history of OSA among bariatric surgery patients.Postoperative complications were minor, with anemia in 4 (3.3%) patients, mesenteric vein thrombosis in 3 (2.5%), and escalation of care to ICU in 3 (2.5%) patients. Logistic regression predicting postoperative complications in bariatric surgery patients, none of the factors—age (OR = 0.995, p = 0.825), BMI (OR = 1.005, p = 0.897), length of stay (OR = 0.969, p = 0.877), obesity class (OR = 0.826, p = 0.565), or comorbidities such as OSA history (OR = 1.014, p = 0.977) and diabetes (OR = 1.376, p = 0.578)—showed any statistically significant association. In predicting hospital LOS, obesity class was the only statistically significant factor (OR = 2.72, p = 0.005), indicating that individuals in higher obesity classes are more likely to experience longer stays. Conclusion: Obesity class was a strong predictor of longer hospital stays, underscoring the importance of obesity severity in surgical recovery. This study also implies that unmeasured factors may better predict complications in this bariatric surgery population or that complications are relatively rare and thus, harder to predict with the available variables. Further research is needed to better understand the complex relationships between these factors and their role in bariatric surgery outcomes.