RA Cohort Links Heart Failure to 10-Year CV Burden

Key Takeaways

• In this Martinique rheumatoid arthritis cohort, the broader hospitalized cardiovascular endpoint yielded a higher 10-year burden estimate than the sensitivity analysis restricted to 3-point MACE.
• Heart failure was the leading hospitalized cardiovascular presentation, and reviewed heart failure cases were largely non-ischemic.
• Baseline cardiometabolic comorbidity was common in this cohort, while smoking exposure was uncommon.
• Patients with hospitalized cardiovascular events were older at rheumatoid arthritis diagnosis and more often had diabetes, but the small event count limited independent-association claims.

Cardiovascular manifestations of rheumatoid arthritis (RA) in Afro-Caribbean patients remain less well described than those in the European and North American cohorts that shape much of the field’s risk discussion. In Martinique, RA coexists with a distinct risk-factor pattern marked by common cardiometabolic comorbidity and little smoking exposure, so this cohort adds Caribbean context rather than a directly comparable US estimate. Investigators therefore examined which cardiovascular events led to hospitalization during the first decade after RA diagnosis in this setting.

Investigators conducted a retrospective hospital-based analysis at the University Hospital of Martinique (CHUM) in the French West Indies, where acute cardiovascular care is centralized. They included 205 adults with specialist-confirmed RA diagnosed between 2005 and 2015, identified through Programme de médicalisation des systèmes d'information (PMSI) screening with ICD-10 codes M05 and M06 and confirmed by 2 rheumatologists through full-record review; patients with a hospitalized cardiovascular event before RA diagnosis were excluded.

The primary endpoint was first hospitalized cardiovascular event within 10 years after RA diagnosis, including acute heart failure, stroke, myocardial infarction, unstable angina, and cardiovascular death. Investigators also performed a sensitivity analysis restricted to 3-point major adverse cardiovascular events (MACE), used a Fine and Gray competing-risk approach, and ascertained outcomes through discharge-code screening plus systematic review of complete records.

Over 10 years, the cumulative incidence of hospitalized cardiovascular events was 8.3% (95% CI 4.5-12.1), compared with 4% (95% CI 1.3-6.8) when the endpoint was restricted to 3-point MACE. Events accumulated progressively over follow-up, with most occurring after the first several years of RA evolution.

Heart failure accounted for 8 of 17 hospitalized cardiovascular events (47.1%) and was the leading subtype. Reviewed heart failure cases were heterogeneous and largely non-ischemic. Five of eight patients with HF had no pre-existing cardiovascular disease, suggesting incident HF in most cases, although the authors cautioned that classification was limited by retrospective data availability. Phenotypes spanned preserved, mildly reduced, and reduced ejection fraction, and confirmed transthyretin amyloid cardiomyopathy appeared in a minority of cases.

Exploratory comparisons showed that patients with hospitalized cardiovascular events were older at RA diagnosis and more often had diabetes, with diabetes present in 76.5% versus 25.7%, p < 0.001. Patients with events also had a lower mean number of bDMARD/tsDMARD exposures, although the small number of events precluded multivariable analysis and the authors cautioned against interpreting this as an independent association.

Interpretation is limited by the retrospective, single-center, hospital-based design and by the absence of an age- and sex-matched general-population control cohort, so these findings do not quantify excess cardiovascular risk attributable to RA or provide a population-wide incidence estimate. The small event count precluded multivariable analysis, making subgroup differences descriptive only. RA disease activity was not consistently available over time, and some echocardiographic and renal details used to characterize heart failure came from follow-up rather than the index hospitalization. Privately managed milder cases and nonhospitalized, out-of-island, or nonadmitted events also may not have been fully captured. The broader hospitalized cardiovascular endpoint and the narrower 3-point MACE definition therefore describe materially different views of burden.

The authors concluded that hospitalized cardiovascular burden was substantial in this Martinique RA cohort and that heart failure, rather than only atherothrombotic events, helped define the cardiovascular phenotype. They also noted that the pattern was comparable to other RA cohorts despite the cohort’s distinct baseline risk-factor mix. These data add descriptive evidence from an underrepresented Caribbean setting without establishing a causal comparison or a directly generalizable population estimate.

Clinician Questions

How were hospitalized cardiovascular events defined in the Martinique rheumatoid arthritis cohort?

The main endpoint was first hospitalized cardiovascular event within 10 years after RA diagnosis and included acute heart failure, stroke, myocardial infarction, unstable angina, and cardiovascular death. Investigators also performed a sensitivity analysis using a narrower 3-point MACE definition and treated death before cardiovascular hospitalization as a competing event in the Fine and Gray analysis.

What did chart review show about heart failure phenotype in rheumatoid arthritis patients hospitalized in Martinique?

Chart review found no identified ischemic etiology among the reviewed heart failure cases and showed preserved, mildly reduced, and reduced ejection fraction phenotypes. Most cases appeared incident because pre-existing cardiovascular disease was usually absent, confirmed transthyretin amyloid cardiomyopathy was identified in 2 cases, and some echocardiographic and renal details came from follow-up assessments rather than the index hospitalization.

Why can’t this Martinique rheumatoid arthritis cohort be used to estimate excess cardiovascular risk from rheumatoid arthritis itself?

This cohort had no age- and sex-matched general-population control group, used a retrospective single-center hospital-based design, and included too few hospitalized cardiovascular events for multivariable analysis. It therefore describes hospitalized cardiovascular burden within the Martinique RA cohort rather than providing a causal estimate or a population-wide measure of excess cardiovascular risk attributable to RA.

Recommended Reading

• For more on cardiovascular risk in RA: Cardiac Rehabilitation Lowers Risk in Rheumatoid Arthritis with Hypertension