Readmissions High After Endovascular Therapy for Stroke

December 17, 2018

Patients undergoing endovascular therapy for acute ischemic stroke have a high rate of 30-day non-elective hospital readmission, although no higher than those receiving thrombolysis alone, a large cohort study suggests.

"As endovascular therapy for stroke has now shown benefits in several randomized trials its use is going to grow. We wanted to look at real-world data on readmissions to see if this showed any unexpected hazards," senior author Deepak Bhatt, MD, Brigham and Women's Hospital, Boston, Massachusetts told Medscape Medical News.

"Our key message is that we did not find any hazard of this invasive procedure in terms of excess admissions — that is reassuring," he added.

The study was published online December 10 in JACC: Cardiovascular Interventions.  

However, the author of an accompanying editorial, Salvador Cruz-Flores, MD, Texas Tech University Health Sciences Center, El Paso, suggests that the data could raise questions about the effectiveness of endovascular therapy in the real world.

"While on one hand, you could say that the data are reassuring in that they do not show an increase in readmissions with this invasive procedure, but surely as the randomized trials have shown a clinical benefit with endovascular therapy, you would expect readmission rates to be lower," he commented to Medscape Medical News.

"Readmissions because of infections, cardiac causes, and recurrent strokes occur more frequently after severe strokes," he added. "We can't say anything for sure from this administrative data as it doesn't provide clinical information on stroke severity and the neurologic deficit on discharge, but I do think these data raise the question of whether the benefit of endovascular therapy seen in randomized trials is translated into the real world." 

Bhatt responded: “We were not looking at acute outcomes in this study. The randomized trials have already done that and shown benefits for endovascular therapy in selected groups of patients."

"But this is an invasive treatment and we wanted to look in the real world to see if that might be associated with an increase in complications producing an increase in readmissions,” he said. “But after propensity matching to adjust for the severity of illness in the patients we didn't see any hazard of endovascular therapy in this respect. I would say that this is more supportive data for endovascular therapy for stroke and no reason to restrict its role."

"The majority of readmissions in these patients are from infection or cardiac causes which would not have been influenced by which stroke therapy was used," Bhatt added. "They would have happened anyway, and readmission from recurrent stroke or transient ischemic attack [TIA] (which only accounted for 14% of readmissions) were lower in the endovascular group. That fits in with randomized data."

Better Secondary Prevention Needed

But what is more worrying, Bhatt said, is the finding that 12% (one in eight) patients had a non-elective readmission within 30 days, with the most common cause being infection, cardiac causes, and recurrent stroke/TIA. 

"This emphasizes the need for further optimization of secondary preventive measures and incorporation of comprehensive multidisciplinary treatment during presentation and transitions in care for acute ischemic stroke," the authors say. "Arranging for early post-discharge follow-up, especially for those at increased risk such as older patients, may help mitigate this risk," they add.

For the study, Bhatt and colleagues analyzed data from the 2013 to 2014 Nationwide Readmissions Database, which represents about 50% of total hospitalizations in the United States and is the largest national database to examine readmission patterns.

Results showed that among 2,055,365 hospitalizations for acute ischemic stroke with survival to discharge, 70,046 (3.4%) received any recanalization therapy (endovascular therapy and/or thrombolysis). Of these, 10,795 (0.5%) underwent endovascular therapy with or without thrombolysis.

Among those who underwent endovascular therapy, 12.4% were readmitted within 30 days (median 9 days) after discharge. In a propensity score-matched analysis, compared with thrombolysis alone, endovascular therapy had similar odds of 30-day readmissions (12.4% vs 12.6%; hazard ratio, 0.98; 95% CI, 0.91 - 1.05).

Multivariate analysis identified the following patient-related characteristics as independent predictors of 30-day readmission after endovascular therapy: Medicare or Medicaid payer (compared with private insurance), diabetes, coagulopathy, gastrostomy tube placement, acute myocardial infarction, atrial fibrillation, and heart failure during the index hospitalization. Thrombolysis co-administration with endovascular therapy was not an independent predictor of 30-day readmission.

The most frequent reasons for readmission in those who underwent endovascular therapy were infections (17.2%), cardiac conditions (17.0%), and recurrent stroke/TIA (14.8%). Bleeding causes were less frequent (2.7%) as were procedural complications (7.9%). By contrast, recurrent stroke/TIA (23.4%) was the most common reason for 30-day readmissions in those receiving thrombolysis alone.

The rate of recurrent stroke/TIA was lower with endovascular compared with thrombolysis alone (absolute difference, 8.6%; 95% CI, 6.35 - 10.69).  Recurrent ischemic stroke was the most common reason for readmission of recurrent cerebrovascular events in both groups, and the incidence of hemorrhagic stroke was similar (2.5%).

In his editorial, Cruz-Flores points out that the risk of readmissions found in this study is unchanged compared with previous studies and shows a high rate of readmissions related to stroke and TIA recurrence.  

"These findings at face value raise concerns about the effectiveness of endovascular therapy and thrombolysis limiting disability in practice, the effectiveness of secondary prevention, and the effectiveness and efficiency of systems of care, particularly at the transitions," he writes.

But he notes that because of the limitations of administrative datasets no clear conclusion can be drawn and further research is necessary.   

JACC Cardiovasc Interv. 2018;11:2414-2424, 2425-2426. Abstract, Editorial 

For more Medscape Neurology news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: