LOS ANGELES – Carotid stenting and carotid endarterectomy led to very similar health status outcomes a year after surgery, according to the quality-of-life analysis of the CREST trial.
Dr. Joshua M. Stolker
Although patients with carotid artery stenosis randomized to the stenting arm of the study generally had better health status measures at 2 and 4 weeks following their intervention, by 1 year "all between-group differences had entirely resolved" when compared with the patients randomized to carotid endarterectomy, Dr. Joshua M. Stolker said at the International Stroke Conference.
The two treatment alternatives tested in the Carotid Revascularization Endarterectomy vs Stenting Trial (CREST) led to nearly identical long-term outcomes for quality-of-life despite an excess of strokes in the carotid stenting arm, and despite the finding in a new, exploratory analysis of health status outcomes that strokes produced a markedly larger decrement in health status than did myocardial infarctions or cranial nerve palsy. The worse outcomes with stroke probably did not affect the overall outcome of the stented patients because the number of strokes – and especially the number of major strokes – was relatively small in CREST.
Because of "the small absolute difference in stroke rates" between the stented and open-surgery patients and the fact that "most of the difference was in minor strokes, "Overall, there was no difference for the whole population despite the impact of stroke," said Dr. Stolker, a cardiologist at Saint Louis University.
In CREST, stented patients had a 1.8% higher rate of all strokes, compared with the endarterectomy patients, including a 0.5% excess rate of major strokes.
Stroke patients showed statistically significant impairments, compared with the patients who did not have a stroke, in seven of the eight health status outcome subscales measured using the Short Form (SF)-36: physical function, role physical, vitality, pain, social function, role emotional, and mental health. In contrast, after 1 year patients in CREST who had an MI following their surgery had a deficit only in the measure of general health, compared with the CREST patients who did not have an MI. And patients who developed cranial-nerve palsy after their surgery had no 1-year deficits for any health-status measure, compared with the patients who did not develop cranial-nerve palsy.
These results "confirmed what a lot of us already suspected, that periprocedural stroke had a significant impact on 1-year health status across multiple SF-36 outcomes, but periprocedural MI and cranial-nerve palsy did not," Dr. Stolker said.
The greatest impact on health status occurred in patients with major periprocedural strokes, who on average showed significant reductions in all eight subscales of the SF-36 at 1-year follow-up. Patients with minor periprocedural strokes had, on average, significant 1-year drops in "three or four" SF-36 subscales, compared with CREST patients who did not have strokes, he said in an interview.
CREST randomized more than 2,500 patients with symptomatic or substantial asymptomatic carotid disease to either endarterectomy or carotid stenting during 2000-2008. The study’s primary end point results – the periprocedural rate of stroke, MI, or death plus the 4-year rate of ipsilateral stroke – showed no significant between the two study arms, results reported last year (N. Engl. J. Med. 2010;363:11-23). Dr. Stolker’s new report focused on a prespecified analysis of health status outcomes.
In addition to the eight subscales of the SF-36, the researchers also measured changes in health status by six parameters measures by modified Likert scales: difficulty walking, difficulty eating and swallowing, difficulty driving, headaches, neck pain, and leg pain.
The new analysis showed that at 2 weeks after the interventions, patients who had undergone carotid stenting had, on average, significantly better health-status scores for five of the eight SF-36 subscales. But at 4 weeks after treatment, this fell to significantly better scores for stenting in three subscales: physical function, social function, and role physical. One year after the intervention, both the stented and open surgery arms showed, on average, essentially identical scores for all eight subscales.
The six additional parameters scored on a modified Likert scale showed a roughly similar pattern. After 2 weeks, the stented patients had significantly better scores for difficulty eating and swallowing, difficulty driving, headache, and neck pain, while the endarterectomy patients had a significant edge for difficulty walking and leg pain. These between-group differences began diminishing at 4 weeks and, at 1 year after the intervention, neither procedure had a significant difference for any of the six measures.
CREST was primarily funded by the National Institutes of Health. It also received supplemental support from Abbott Vascular Solutions, the company that markets the carotid stent and embolic protection device used in the study. Dr. Stolker said that he has been on the speakers bureau of AstraZeneca and Pfizer.