Pouya Tahsili-Fahadan, MD

Dasenbrock HH, Angriman F, Smith TR, Gormley WB, Frerichs KU, Aziz-Sultan MA, et al. Readmission After Aneurysmal Subarachnoid Hemorrhage: A Nationwide Readmission Database Analysis. Stroke. 2017

Readmission (within a pre-defined period of time from discharge) is frequently measured and reported as a quality measure for care provided by physicians and hospitals. However, it is debatable whether this measure is an appropriate quality metric for various indications and etiologies of the index hospitalization. Dasenbrock et al. investigated this question by analyzing the Nationwide Readmission Database (NRD) for readmission after aneurysmal subarachnoid hemorrhage (SAH).

Data from this longitudinal administrative database within 21 states were extracted for 3806 non-elective adult patients admitted for treatment of aneurysmal subarachnoid or intracerebral hemorrhage and discharged alive in 2013. Mortality during the index hospitalization and readmission were 11% and 1.7%, respectively, and about two thirds of survivors were discharged home. The median cost of the index and readmission hospitalizations were $266,304 and $45,091, respectively, and readmission was associated with increased total costs. Within the next 30 days from discharge, 10.2% of patients were readmitted with 34.4%, 65.6%, and 82.4% of readmissions within 1, 2, and 3 weeks from discharge, respectively. As expected, patients who were readmitted had higher SAH severity scale, higher incidence of cerebral edema, and complications during their index hospitalization, and were more likely to undergo tracheostomy or gastrostomy, and less likely to be discharged home. Treatment modality (clipping versus coiling) was not associated with increased rate of readmission. Independent predictors for readmission, however, were identified as comorbidity score equal or more than 3, higher SAH severity, and discharge destination other than home; the more predictors, the higher chance of readmission. Of note, high-volume institutions had lower risk of readmission and mortality. The most common reasons for readmission included hydrocephalus, other neurological complications, infections, and thromboembolic events. Neurosurgical procedures and surgeries were among the most common operations performed after readmission. Importantly, hydrocephalus during index hospitalization was associated with increased risk of readmission for hydrocephalus.

The findings of this large nationwide study are important and suggest that using readmission as a quality metric may not be appropriate in this population. Although the readmissions rate after SAH is high, it was not associated with high mortality and morbidity and rarely related to new ischemic or hemorrhagic stroke in this population. Importantly, the factors independently associated with readmission included baseline comorbidities and SAH severity. Also, aside from hydrocephalus, readmissions were not a result of the complications during index admission. In regards to hydrocephalus, emergence of delayed hydrocephalus and its complications are well known after SAH and not reflective of poor decisions at the time of initial discharge. This report further emphasizes the importance of selecting appropriate outcome and quality metrics relevant to specific populations and conditions, rather than a universal approach to quality assessment during hospitalization.