For Interventional Trials, Optimal Medical Management May Entail as Much Training as New Techniques
Experience has long been cited as key to better performance of highly technical procedures, but a new analysis shows that institutional experience might be just as important when it comes to optimal medical management. The findings could have implications for clinical trials where emphasis is often placed on making sure that participating investigators have adequate training in complex interventions, but not necessarily on ensuring that they are experts at controlling patient risk factors.
The new observations come from the 451-patient SAMMPRIS trial, which compared aggressive medical management plus percutaneous transluminal angioplasty and stenting with aggressive medical management alone in patients who had severe intracranial stenosis and a recent stroke or TIA. Published in 2011, SAMMPRIS showed that aggressive medical management alone reduces the risk of periprocedural stroke and death compared with stenting, and offers similar protection against recurrent stroke.
For the current study, David Chiu, MD, of Houston Methodist Hospital (Houston, TX), and colleagues looked specifically at high- versus lower-enrolling centers and compared clinical outcomes and risk factor control. Sites that enrolled at least 12 to 27 patients were considered high-enrolling and sites that enrolled 1 to 11 were deemed low-enrolling.
It is generally accepted, Chiu told TCTMD, that experience level matters for surgical and endovascular procedures, but the impact of experience on complex medical management protocols has not been studied in depth.
“In clinical trials, one might expect that study investigators and coordinators at higher volume sites may become more familiar with the study protocol and treatments, which could lead to improved outcomes in patients treated medically,” he and his coauthors write in their paper.
The findings were published in the December 15 issue of Neurology.
The primary targets of medical management were LDL cholesterol and systolic blood pressure, but secondary goals included lowering hemoglobin A1c in diabetics and reducing non-HDL cholesterol, smoking cessation, weight reduction, and increasing levels of moderate exercise.
In the stenting group, enrollment level did not influence the rate of the primary endpoint, which was a composite of any stroke or death within 30 days or ischemic stroke in the territory of the stenotic artery beyond 30 days (P = .93). In addition, there were no differences between high-enrolling and low-enrolling sites in the percentage of patients reaching risk factor targets or in the mean values for various risk factors.
In the medical-management group, however, the primary endpoint occurred less frequently in patients treated at high-enrolling centers (P = .0005), both at 30 days (1.8% vs 7.3%) and 2 years (9.8% vs 20.9%). Moreover, patients at high-enrolling centers were more likely to achieve LDL cholesterol targets (68% vs 48%; P = .005) and had lower mean levels of LDL and non-HDL cholesterol and systolic blood pressure.
When the 2 trial arms were combined, patients treated at high-enrolling centers were more likely to achieve targets for LDL cholesterol and systolic blood pressure, with a similar nonsignificant trend for non-HDL cholesterol.
The findings show that “it does make a difference how hard each of the centers work at achieving optimal risk factor control,” Chiu said. “If this type of result is replicated in other studies I think it justifies training phases for the medical arms of therapeutic trials just like we have training phases for surgical and endovascular arms.”
Philip Meyers, MD, of Columbia University Medical Center (New York, NY), said that he can believe that centers enrolling more patients would be better at managing patients because “risk factor management in patients with symptomatic intracranial atherosclerosis is challenging and requires expert supervision.”
Why high-enrolling centers had improved clinical outcomes is not clear, said Meyers, who was a SAMMPRIS investigator. “High-enrolling centers may have more patient management experience or possibly more personnel including physician extenders who help enforce the treatment plan.”
Why Results Are Better at High-Enrolling Sites
The authors note that the difference in the primary endpoint between high-enrolling and low-enrolling sites in the medical management arm of the trial could be due to unknown confounders, but add that improved control of risk factors likely provides at least a partial explanation for the finding.
However, “the discrepancy between the extent of the lower stroke risk vs the mean reduction in LDL and [systolic blood pressure] levels at high-enrolling sites suggests that other factors may also have contributed to the lower stroke risk at high-enrolling sites,” they say, pointing to antiplatelet therapy compliance as potentially having a role.
Generally, though, the superior outcomes at high-enrolling centers reflect a greater commitment at those sites to optimizing medical management for stroke prevention, Chiu said. It is difficult to get patients to buy into lifestyle changes and medication regimens to manage risk factors, and experienced centers may have teams that are more aggressive about improving compliance, he said.
“We’re finding that [successful
medical management] probably requires a level of training that is on par with
the type of training that we mandate for the interventionalists in a trial like
SAMMPRIS,” Chiu said.
Chiu D, Klucznik RP, Turan TN, et al. Enrollment volume effect on risk factor control and outcomes in the SAMMPRIS trial. Neurology. 2015;85:2090-2097.
- Chiu reports no relevant conflicts of interest.
- Meyers reports having served as president of the Society of NeuroInterventional Surgery.