LAUSR

We read with great interest the recently published article by Koo et al. This retrospective cohort study used the Hospital Frailty Risk Score (HFRS) to measure ‘frailty’ and its association with complication rates and healthcare resource utilization in patients who underwent endovascular treatment of ruptured intracranial aneurysms. Koo et al used ICD10CM codes to identify patients, who were then categorized into frailty groups: low (HFRS <5), intermediate (HFRS 5–15), and high (HFRS >15). The study concluded that greater frailty, as defined by the HFRS, was associated with increased postoperative complications, length of hospital stay, total healthcare cost, and nonroutine discharge disposition. We would appreciate clarification from the authors regarding the following concerns. The HFRS is a risk score derived from >1000 ICD10CM codes overrepresented in a population of hospitalized older adults (≥75 years of age). One of the most concerning aspects of the HFRS design is the excessive number of points awarded to acute conditions and/ or measures of disease severity, including acute neurological deficits (table 1). This is especially concerning in the neurointerventional setting, as a large proportion of patients are acutely ill with temporary neurological deficits before eventual recovery. For example, the HFRS captures diagnoses for acute neurosurgical deficits such as gait disturbance, altered mental status, somnolence/stupor/ coma, and incontinence. Furthermore, the HFRS likely directly incorporates postoperative sequelae, as the National Inpatient Sample (NIS) is a database without temporality limits to distinguish preoperative versus postoperative conditions. The possibility of overestimating patient frailty and excluding appropriate surgical candidates is very plausible with the NIS, and this would have profoundly negative consequences. Furthermore, Gilbert et al acknowledge the poor discriminatory accuracy of HFRS Letter