The EUCLIDian perspective of global peripheral artery disease (PAD)

It is estimated that more than 200 million people worldwide are affected by peripheral artery disease (PAD), with a sharp increase in prevalence in this century. ¹ Although the majority of patients are asymptomatic, those with symptoms require proportionally increased medical attention, which can be a significant public health concern and economic burden.

In this issue, Norgren and co-investigators of the EUCLID (Examining Use of tiCagreLor In peripheral artery Disease) trial – an international study having recruited patients with PAD from 28 countries in four global regions – highlighted the disparities of the participants according to their geographical localization. ² Because of the lack of difference in outcomes between the two arms of the study (i.e., ticagrelor vs clopidogrel), the authors were able to provide the overall patients’ characteristics and results, independent from the randomized allocation of enrollees. The authors report regional differences for the inclusion criteria, risk factors, associated therapies, and outcomes. Of note, EUCLID is one of the largest trials ever performed within a population of patients with PAD, and from this standpoint, as well as the expanse of its global recruitment, the results presented here are a unique opportunity to address the regional disparities in a PAD clinical trial population.

In brief, the comparison of data from patients enrolled in North and Central/South (C/S) America, Asia, and Western/Eastern Europe reveal some significant differences in demographic characteristics (e.g., older participants with lower body weight in Asia), risk factors (e.g., more diabetes among participants in C/S America, more hypertension and hyperlipidemia in North America), and inclusion criteria (more asymptomatic patients with history of revascularization enrolled in Asia, and more past amputees in C/S America). ² There were also different patterns of comorbidities (higher rates of prior stroke in Asia and prior myocardial infarctions in North America) and use of cardiovascular protective drugs (high rates of statin therapy in North America and Europe and less ACE-inhibitors (ACEI) and aspirin in Asia). A high rate of ACEI intolerance due to cough has been reported among Asians, which could account for the lower usage rate of this treatment and higher usage rate of angiotensin receptor blockers. ³ Of note is the high usage of cilostazol in Asia (first marketed in Japan) and C/S America, presumably for the relief of intermittent claudication, but which may have been used for its antiplatelet effect.^4,5 Some of these findings are in line with data published from the international REACH registry, including 8322 participants with PAD, ⁶ with the highest proportions of patients with PAD and diabetes in C/S America (59% in REACH, 57% in EUCLID), hyperlipidemia/dyslipidemia in North America (81% in REACH, 92% in EUCLID), hypertension in North America (86% in both REACH and EUCLID), the highest rates of statin use in North America (77% in REACH, 84% in EUCLID), and the lowest rates of aspirin use (55% in REACH and 56% in EUCLID) and ACEI use in Asia (30.5% in REACH and 12.8% in EUCLID).

Regarding trial outcomes, the total and cardiovascular mortality in EUCLID was the highest among patients in C/S America, despite younger age at enrollment, whereas fatal plus nonfatal cardiovascular event rates were the highest in North America. ² Overall, when dividing the European countries into two regions, those enrolled in Western Europe had similar profiles and outcomes as those in North America. Regarding the limb outcomes, the probability of acute limb ischemia and lower-extremity revascularization was higher in North America and Europe, whereas amputations were more frequent in C/S America. Further adjustments to available baseline characteristic differences had minor effects on regional differences in outcomes.

The reasons for these differences among global regions are definitely multifactorial and include, nonexclusively, the genetic background, the prevalence of risk factors in each region, the knowledge about PAD in the population and among health care providers, the possibilities of (and affordability to) access to medical care, the availability of pharmacological and interventional means to manage the patients, and patient response to treatment.^7,8 Similar differences are found in international studies regarding patients’ profile, management, and outcome in those with other cardiovascular conditions, including coronary artery disease.^9,10 However, one should bear in mind that these variations can hardly reflect perfectly the actual discrepancies in profiles and outcomes of patients with PAD worldwide. As, for example, the higher risk for major adverse cardiovascular events in North America found in EUCLID is not in line with data from the REACH registry, an observational study where no intervention was proposed. ⁶ As a trial, even with inclusion criteria as liberal as EUCLID, this study presents with inherent selection bias related to multiple reasons why an investigator offers to enroll (or not) a patient to such a trial, and why that patient accepts or declines the enrollment, which can also vary from one country to another. Regarding the patients’ consent, socioeconomic factors are of paramount importance. For instance, it is well known that in countries without full health coverage, patients with poor levels of insurance might be keen to accept such a trial that insures medical surveillance, or even health care, during the study period.^11,12 These socioeconomic factors are also known to affect the prognosis of patients with PAD, with poorer outcomes among those of low economic status.^13,14 As an illustration, the odds for hospitalization for PAD in a US cohort more than doubled in those with low incomes or low educational level. ¹⁴

How are these differences important when conducting a trial on PAD? Essentially, large trials are often needed nowadays to demonstrate statistically significant differences between different therapeutic strategies, and as the incremental value of a new strategy versus the classical one is often of small magnitude in terms of absolute risk reduction, so the collaboration of investigators across the borders and oceans are of absolute necessity to ensure an adequate sample size. Furthermore, the quality of a trial and its results depend in part on its ubiquity (i.e., generalization of the results worldwide), despite differences in patients’ profiles and clinical practices. The results of such trials are often quite homogeneous in different global areas, although some differences can be observed, without reaching significant interaction. Any significant interaction between the results of a trial and one geographic region is extremely unusual and potentially disturbing. One famous example is the TOPCAT trial assessing the efficacy of spironolactone in the management of heart failure with preserved ejection fraction, with a nearly fourfold difference in the composite event rates between Russia and Georgia on the one hand, and North and South America on the other, which were related to substantial differences in the profiles of recruited patients, and leading ultimately to inconclusive results in practice. ¹⁵ Similarly, a post hoc analysis of the NAVIGATOR trial revealed significant geographic differences in the risk of incident diabetes mellitus and cardiovascular outcomes for patients with impaired glucose tolerance. ¹⁶ After adjustment, compared with North American participants, Europeans had a lower risk of developing diabetes mellitus and Latin Americans had a higher risk of cardiovascular death and the composite cardiovascular outcome of cardiovascular mortality, nonfatal myocardial infarction, or nonfatal stroke. Such findings, especially when validated in future studies, may form the basis of specific recommendations for the use of therapy, in this case angiotensin-receptor blocker agents, in specific patient populations in regions with a high risk of an outcome of interest.

The global regional variations reported in EUCLID require verification in other large international trials of PAD (e.g., COMPASS and VOYAGER trials) before drawing definite conclusions. More specifically, higher rates of outcomes of interest in patients with PAD in some regions can affect the sample size calculations for future trials, and plans to reach the expected targets.