A novel frailty scale was found to accurately predict adverse events in patients undergoing percutaneous left atrial appendage occlusion (LAAO), with patients classified as frail at greater risk of in-hospital complications and 45-day mortality, according to a study published in the Journal of Interventional Cardiac Electrophysiology.

Douglas Darden, MD, et al., included 57,728 patients (mean age 76.2 years, 41.2% women, 68.1% implanted with Watchman FLX) from the ACC’s LAAO Registry between 2016 and 2021. Patients were evaluated using a five-item scale including the following elements: hemoglobin <13.0 g/dL in men, <12.0 g/dL in women; creatinine ≥1.2 mg/dL; albumin <3.5 g/dL; body mass index <20 kg/m2; and increased fall risk. The authors then categorized patients as either non-frail (0 points), pre-frail (1-3 points) or frail (4-5 points).

Results showed that 76.8% of patients analyzed were pre-frail and 13.3% were frail. These patients exhibited a higher burden of comorbidities and more disability based on the Modified Rankin Scale when compared with those deemed non-frail. Frail patients had a higher risk of in-hospital major complication (odds ratio [OR], 1.29, 95% CI, 1.02-1.62, p=0.01), any complication (OR, 1.29, 95% CI, 1.09-1.52, p=0.0005), and death (OR, 5.79, 95% CI, 1.75-19.17, p=0.001) when compared with non-frail patients. There was no difference in risk of adverse events when comparing pre-frail and non-frail patients.

At 45-days post procedure, no difference was seen in risk of complications between frail and non-frail patients, but rate of mortality was much higher (OR, 3.01, 95% CI, 1.97-4.85, p<0.0001).

“While this frailty score identifies high-risk populations, it has not been validated in separate registries or compared to other validated frailty scores as we are limited to available variables within the robust LAAO Registry,” write the study authors. “Our results are strengthened by its high predictive ability and correlation with increasing levels of disability…although it will need to be validated in other datasets and prospective studies.”