Using a Modified Frailty Index to Predict Mortality Risk after Tracheostomy

Abstract

Objectives:

Evaluate the utility of a modified frailty index to predict post-operative mortality in patients undergoing tracheostomy.

Methods:

A retrospective chart review was conducted including all patients who underwent tracheostomy between 2007 and 2012. A modified frailty index consisting of 11 items based on the Revised Minimum Data Set Mortality Rating Index was retrospectively applied using the patients’ status immediately prior to tracheostomy. The resultant 6-month predicted mortality was compared to the Veterans Health Administration Surgical Quality Improvement Program’s (VASQIP) 30-day predicted mortality as well as actual mortality.

Results:

A total of 100 patients underwent tracheostomy during the study period. No patients were excluded. Sixty-nine patients died within the study period. The 1, 6, and 12-month mortality rates after tracheostomy were 25%, 43%, and 59% respectively. The predicted 30-day mortality risk was 9.1% for non-survivors compared to 5.2% for survivors (p = 0.05). The predicted 6-month mortality risk was 40.5% for non-survivors compared to 25.4% for survivors (p = 0.0003). The VASQIP calculator and modified frailty index both differentiated mortality risks between patients who survived less than 6 months versus greater than 6 months (p = 0.004 and 0.001, respectively). However, neither the VASQIP or modified frailty index differentiated mortality risks for head and neck cancer patients who survived less than 6 months versus greater than 6 months (p = 0.46 and 0.13, respectively).

Conclusions:

The modified frailty index identifies non-cancer patients at high risk of post-operative mortality after tracheostomy but does not identify high-risk patients with head and neck cancer.