Peripheral Matters: High Rate of Unplanned Admissions and Predictors Identified in Critical Limb Ischemia
The majority of readmissions for critical limb ischemia (CLI) at 30 days and six months are unplanned, and these are associated with a higher subsequent mortality rate, according to a study published on March 6 in the Journal of the American College of Cardiology.
The study, led by Shikhar Agarwal, MD, MPH, et al., retrospectively analyzed data for the period of 2009-2013 from the states of Florida, New York and California, which provide data on repeat admissions, obtained from the State Inpatient Database. A total of 284,189 admissions from 212,241 patients for the principal diagnosis of CLI were included in the study.
At 30 days and six months, the rate of all-cause readmissions was 27.1 percent and 56.6 percent, while the rate of unplanned readmissions was 23.6 percent and 47.7 percent, respectively. The most frequent causes of unplanned readmission were miscellaneous, primary CLI-related causes, post-procedure complications, septicemia and diabetes-related non-vascular causes.
Predictors of unplanned readmissions at six months were age, female sex, black or Hispanic race, prior amputation, a higher Charlson comorbidity index, need for rehabilitation facility or home health care services, as well as type of insurance (lower rate with private insurance). Only 33.6 percent of primary CLI admissions were discharged to home.
The rate of in-hospital mortality was 5.0 percent and 1.8 percent in the unplanned and planned admission groups at 30 days, and these rates were 4.3 percent and 1.5 percent at six months (p < 0.001 for both). The overall in-hospital mortality rate was 2.3 percent among all primary CLI admissions. Age, female sex, revascularization or major amputation during hospitalization, and a higher Charlson comorbidity index were predictors of in-hospital mortality.
Notably, in patients over the age of 80, in-hospital mortality was higher for those with private insurance compared with Medicare or Medicaid/no insurance. Revascularization or major amputation was more frequent during planned than unplanned readmissions.
Most of the study patients had Medicare insurance (71 percent), with a mean age of 68, and most were men and white (59 percent each). The Charlson comorbidity index was 0-2 in 36 percent and 5+ in 26 percent of the patients.
Already an important performance metric, readmission rates will be tied to reimbursement-based improvement incentives by Medicare and Medicaid, requiring a better understanding of the drivers of planned and unplanned admissions in this population, the authors state. Readmission rates have been traditionally high after vascular procedures, with a constant rate of CLI admissions over the last decade, despite an increase in endovascular over surgical procedures. The authors also note there are scarce data on readmission rates in CLI beyond 30 days.
The investigators also examined the impact of the type of hospital, length of stay, and travel time, showing for what they believe is the first time that a shorter travel time was associated with a higher readmission rate.
The authors state their findings have implications in relation to managing patient discharge and for policymakers in relation to payment reform based on length of stay or readmissions. In an accompanying editorial, Mehdi H. Shishehbor, DO, MPH, PhD, FACC, and Herbert D. Aronow, MD, MPH, FACC, comment that most of the independent predictors of readmission are not modifiable, and that although “revascularization may be the most easily modifiable, a significant proportion of patients with CLI undergo major amputation” in the U.S. without undergoing a vascular evaluation in the preceding year. More robust data are needed, they add, to understand the factors that can reduce readmissions to inform clinical and policy decision making.
Clinical Topics: Diabetes and Cardiometabolic Disease
Keywords: Cardiology Interventions, ACC Publications, Diabetes Mellitus, Lipoatrophic, Comorbidity, Patient Discharge, Patient Readmission, Retrospective Studies, Sepsis
< Back to Listings