The Risk Analysis Index Demonstrates Superior Discriminative Ability in Predicting Extended Length of Stay in Pituitary Adenoma Resection Patients When Compared to the 5-Point Modified Frailty Index

Author Type(s)


Document Type


Publication Date


Journal Title

World Neurosurgery: X


OBJECTIVE: To compare the predictive abilities of two frailty indices on post-operative morbidity and mortality in patients undergoing pituitary adenoma resection.

METHODS: The National Surgical Quality Improvement Program (NSQIP) database was used to retrospectively collect data for patients undergoing pituitary adenoma resection between 2015-2019. To compare the predictive abilities of two of the most common frailty indices, the 5-point modified frailty index (mFI-5) and the risk analysis index (RAI), receiver operating curve analysis (ROC) and area under the curve (AUC)/Cstatistic were used.

RESULTS: In our cohort of 1,454 patients, the RAI demonstrated superior discriminative ability to the mFI-5 in predicting extended length of stay (C-statistic 0.59, 95% CI 0.56-0.62 vs. C-statistic 0.51, 95% CI: 0.48-0.54, p = 0.0002). The RAI only descriptively appeared superior to mFI-5 in determining mortality (C-statistic 0.89, 95% CI 0.74-0.99 vs. Cstatistic 0.63, 95% CI 0.61-0.66, p=0.11), and NHD (C-statistic 0.68, 95% CI 0.60-0.76 vs. C-statistic 0.60, 95% CI: 0.57-0.62, p=0.15).

CONCLUSIONS: Pituitary adenomas account for one of the most common brain tumors in the general population, with resection being the preferred treatment for patients with most hormone producing tumors or those causing compressive symptoms. Although pituitary adenoma resection is generally safe, patients who experience post-operative complications frequently share similar pre-operative characteristics and comorbidities. Therefore, appropriate pre-operative risk stratification is imperative for adequate patient counseling and informed consent in these patients. Here we present the first known report showing the superior discriminatory ability of the RAI in predicting eLOS when compared to the mFI-5.