By Michael H. Crawford, MD, Editor
SYNOPSIS: A prespecified subgroup analysis of diabetic patients in the POST-PCI study, which randomized patients post-percutaneous coronary intervention to routine stress testing vs. standard care at one year and followed for two years, has shown that adverse cardiac outcomes and death rates were not improved by routine stress testing.
SOURCE: Kim H, Kang DY, Lee J, et al. Routine stress testing in diabetic patients after percutaneous coronary intervention: The POST-PCI trial. Eur Heart J 2023; Nov 2:ehad722. doi: 10.1093/eurheartj/ehad722. [Online ahead of print].
Patients with diabetes are at higher risk of ischemic cardiovascular events and mortality than non-diabetic patients after percutaneous coronary interventions (PCI), but the ideal surveillance strategy after complex PCI in diabetic patients is unclear. Thus, the Pragmatic Trial Comparing Symptom-Oriented versus Routine Stress Testing in High-Risk Patients Undergoing PCI (POST-PCI) study was conducted, and randomization was stratified by the presence or absence of diabetes. POST-PCI was conducted in 11 hospitals in South Korea from 2017 to 2019 and included 1,706 patients with high-risk anatomical or clinical features who had undergone PCI. The participants were randomized to routine functional testing at one year or standard care where functional testing was done only if clinically indicated.
Functional testing consisted of exercise nuclear perfusion imaging or echocardiography; electro-cardiogram alone was discouraged. The primary outcome was a composite of major cardiovascular events (death, myocardial infarction [MI], or hospitalization for unstable angina) after two years of follow-up and was adjudicated by an independent committee. Several secondary clinical outcomes were assessed as well. Diabetes was present in 39% of the POST-PCI trial patients, of whom 11% were on insulin. The 660 patients with diabetes were older, more likely to be women, and had more complex coronary disease and more comorbidities.
In the routine functional testing group of diabetic patients (n = 289), 90% underwent stress testing, as did 6% of the standard care group. Similar proportions of the non-diabetic patients (n = 1,046) underwent stress testing. Diabetic patients had a higher two-year risk of the primary endpoint compared to the non-diabetics (7.3% vs. 4.8%; hazard ratio [HR], 1.52; 95% confidence interval [CI], 1.02-2.27, P = 0.039). However, the risk of the primary endpoint was similar between the routine function testing and standard care in the diabetic patients (7.1% vs. 7.5%, P = 0.82) and the non-diabetics (4.6% vs. 5.1%, P = 0.68). These results were similar in the insulin-dependent patients.
Among the secondary outcomes, the incidences of invasive coronary angiography and repeat revascularization were higher in the routine functional testing group compared to the standard care group irrespective of diabetes status. The authors concluded that patients with diabetes, despite being at higher risk of adverse cardiovascular events, did not benefit from routine stress testing at one year compared to standard care alone.
COMMENTARY
Previous studies of routine stress testing post-PCI are observational and largely negative. Most guidelines give this practice a IIb indication. The results of POST-PCI suggest that a class III indication may be more appropriate. However, studies have shown that residual ischemia is common in diabetic patients post-PCI, and they are more likely to experience adverse cardiovascular events. Thus, this prespecified subgroup analysis of POST-PCI is of interest. It has shown that diabetic patients had about a 50% higher risk of the primary outcome, but that the risk at two years was similar in those routinely stress tested after one year and those not tested. Also, repeat revascularization was more frequent in those routinely stress tested irrespective of diabetes status, but testing was not associated with a decrease in major cardiac events. Standard care permitted a symptom-oriented surveillance strategy, which was shown to be safe and effective compared to routine testing. This result is not particularly surprising, since routine testing rarely is better than a more nuanced approach that relies on physician judgment.
This subgroup analysis of POST-PCI has several weaknesses. There is only a one-year follow-up after stress testing. A longer follow-up period may have detected benefits from routine testing at one year. Also, routine stress testing earlier or later than one year after PCI could be of value. There was a low number of events, which precluded subpopulation statistical analyses, such as the event rates in the insulin-dependent patients. In addition, there was no differentiation of type 1 or 2 diabetes; although, given the small number on insulin, most of the patients probably were type 2. There was no information on the adequacy of diabetes control, which could have influenced the results. Finally, this was an Asian population that had a high rate of cigarette smoking (> 25%), so the results may not be applicable to other populations with different habits.
I view this as another study supporting physician decision-making over care dictums. I believe there is very little objective support for most dictums, since there are nuances to patient care that can never be completely captured. Thus, I favor guidelines that are more flexible and allow for individual physician and patient shared decisions. Perhaps it is time to retire class I, II, and III.
