Abstract Objective This study aimed to improve age-independent risk stratification for patients undergoing endoscopic transnasal transsphenoidal (TNTS) approach to pituitary mass resection by investigating the associations between frailty, American Society of Anesthesiologists (ASA), and comorbidity scores with severe complications following TNTS. Design This study is a retrospective review. Setting This review was conducted utilizing the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) database. Participants A total of 680 cases of TNTS identified from 2010 to 2013 were included in this study. Main Outcome Measures The modified frailty index (mFI) was calculated to quantify frailty. ASA and Charlson's comorbidity index (CCI) scores were obtained as physiologic status and comorbidity-based prognostic markers. Severe complications were separated into intensive care unit (ICU)-level complications, defined by Clavien–Dindo grade IV (CDIV) criteria, and mortality. Results Overall, 24 CDIV complications (3.5%) and 6 deaths (0.9%) were recorded. Scores for mFI (p = 0.01, R 2 = 0.97) and ASA (p = 0.04., R 2 = 0.87) were significantly correlated with CDIV complications. ASA scores were significantly correlated with mortality (p = 0.03, R 2 = 0.87), as well as independently associated with CDIV complication by multivariable regression models (odds ratio [OR] = 2.96, 95% confidence interval [CI]: 1.35–6.83, p < 0.01), while mFI was not. CCI was not significantly associated with CDIV complications or mortality. A multivariable regression model incorporating ASA had a lower Akaike's Information Criteria (AIC; 188.55) than a model incorporating mFI (195.99). Conclusion Frailty and physiologic status, as measured by mFI and ASA scores respectively, both correlate with ICU-level complications after TNTS. ASA scores demonstrate greater clinical utility than mFI scores; however, as they are more easily generated, uniquely correlated with mortality and independently associated with ICU-level complication risk on multivariable regression analysis.