Developing World's Varied Heart-Failure Mortality Rates Mostly a Mystery: INTER-CHF

May 10, 2017

PARIS, FRANCE — One-year mortality from heart failure varies widely across different regions of the world that include predominantly low- to middle-income developing countries, and there are also telling variations in HF etiology and comorbidities, according to findings from a groundbreaking longitudinal registry study[1].

On average for the 5823-patient International Congestive Heart Failure (INTER-CHF) registry cohort, 1-year mortality averaged 16.5% but ranged from a high of 34% to a low of 9% and "was highest in Africa and India, followed by Southeast Asia, and lowest in China, South America, and the Middle East," observed Dr Hisham Dokainish (Population Health Research Institute, McMaster University, Hamilton, ON).

The observed geographic differences in mortality persisted after adjustment for a lot of possible clinical and socioeconomic risk factors, with more than half the variation left unexplained by the analysis, he said when presenting the study here at the European Society of Cardiology (ESC) Heart Failure 2017. The study was published online at about the same time in Lancet Global Health.

Based on that, he said, "We hypothesize that regional variations in death in heart-failure patients may be influenced by unmeasured variables in this study, and in most studies to be frank, including healthcare infrastructure, quality and access, or environmental and genetic factors."

Adjusted Hazard Ratio* (95% CI) for 1-Year All-Cause Mortality in Patients with Heart Failure, by World Region
Region HR (95% CI)
Africa 3.8 (2.6–5.5)
India 2.9 (1.9–4.3)
Southeast Asia 2.6 (1.7–3.9)
All HRs P<0.0001 vs the index, South America (Argentina, Chile, Colombia, and Ecuador). HR not significant for the Middle East (Egypt, Qatar, and Saudi Arabia) or China. Africa=Mozambique, Nigeria, South Africa, Sudan, and Uganda; Southeast Asia=Malaysia and the Philippines

Two thirds of the cohort were clinic outpatients and the rest were hospitalized; patients with comorbidities that were life-threatening in the short term were excluded. Noting that registry studies are often limited from not including a broadly representative populations, Dokainish said that INTER-CHF made a point of recruiting from a variety of settings and then following the patients carefully.

Indeed, observed Prof Carolyn Lam Su Ping (SingHealth Duke-National University of Singapore) as assigned discussant for the study presentation, "Before this study, we did not have a systematic evaluation of real-world heart failure in the lower- and middle-income regions that used standardized enrollment criteria and a uniform protocol; that included a mix of rural and urban centers, inpatients and outpatients, and specialists and primary-care centers; and that included longitudinal follow-up, with an impressive 98% follow-up rate."

Among the noteworthy baseline findings: Many of the patients were young. "The patients in Africa, India, and Southeast Asia were in their 50s, a decade younger than those in South America and China," she said; and, going outside the current data, they were a decade younger than in Europe.

"Also very striking are the regional patterns that I think have public-health, mental-healthcare, and clinical-trial-design implications." The primary HF etiology in Africa was hypertension, for example, and it was ischemic heart disease in India and Southeast Asia. "And then in the Middle East, there was a striking prevalence of diabetes of 57% in patients, who also had the largest BMI of 30," Lam said.

"But most important, and sadly, really, is that the youngest patients were the most likely to die. Astoundingly, more than one in three African patients were dead at 1 year. Almost one in four Indian patients appeared to die at 1 year."

Relative Contributions of Risk Factor Categories to Overall Mortality Risk at 1 Year

End-Point Category Contribution (%)
Clinical variables, region, medications, demographics, or socioeconomic factors 46
Clinical variables only 25
Region only 16
Medications only 1.1
Demographics only 0.7
Socioeconomic factors only 0.7

Of the 858 deaths at 1 year, according to Dokainish, 46% were from cardiac causes, 16% were noncardiac, and 38% were from unknown causes. Most deaths were from unknown causes in Africa, he said, where the patients are more far more likely to die at home than in the hospital.

Independent predictors of 1-year mortality included NYHA class 3-4 heart failure, prior HF admission, valve disease by echocardiography, increased systolic blood pressure, BMI, chronic kidney disease, and chronic obstructive pulmonary disease.

Independent predictors of reduced 1-year mortality included use of ACE inhibitors or angiotensin-receptor blockers (ARBs) at baseline and treatment with digoxin; both were significant at 0.8 (95% CI 0.7–0.9). There was no significant association with baseline beta-blocker use.

Interestingly, Dokainish said, socioeconomic variables such as illiteracy, residing in a rural location, and lack of health insurance were significantly predictive in unadjusted analysis, "but they seemed to interact with region" to become nonsignificant in multivariate analysis.

In fact, only 46% of the variability in 1-year mortality in the total cohort could be accounted for by patient clinical features, their medications, region, demographics, or socioeconomic factors.

Mortality and Baseline Characteristics for 6 World Regions in INTER-CHF

End points Overall (N=5823) Africa (N=1294) India (N=858) Southeast Asia (N=811) Middle East (N=1000) China (N=991) South America (N=869)
1-year mortality (%) 16.5 34 23 15 9 7 9
BMI (kg/m2) 26 26 23 26 30 24 29
Diabetes (%) 29 17 26 41 57 19 21
Ischemic heart disease (%) 39 20 46 56 50 45 25
Hypertensive heart disease (%) 17 35 14 15 10 14 21
Idiopathic dilated cardiomyopathy(%) 12 14 11 3 18 15 15
BMI=body mass index

INTER-CHF was funded by Novartis. Dokainish disclosed that he has no relevant financial relatinships. Disclosures for the coauthors are listed in the article. "Novartis staff provided assistance and support for the study."

Follow Steve Stiles on Twitter: @SteveStiles2. For more from, follow us on Twitter and Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.