Key findings from the Atahualpa Project: what should we learn?

KEYWORDS:

• Atahualpa Project
• population-based studies
• rural setting
• Stroke
• cerebral small vessel disease
• sleep disorders
• cognitive performance
• Ecuador

1. Introduction

Owing to increased life expectancy and changes in lifestyles, the burden of noncommunicable neurological diseases is on the rise in rural areas of developing countries, to the point of being considered the new epidemics in these regions. However, knowledge on the impact of these conditions in rural setting is limited because the practice of population-based studies is complicated due to several cross-cultural factors. For example, the common practice of telephone interviews in population-based studies conducted in developed countries is next to impossible in rural areas, where face-to-face interviews are mandatory to get correct information. In addition, reliable prevalence rates of conditions associated with stigma can better be assessed if acquainted field personnel conduct the interviews [1]. Moreover, accurate diagnosis of most noncommunicable neurological diseases requires the use of sophisticated technology that is not available in these areas.

2. Study population

Atahualpa is representative of rural villages of Coastal Ecuador and was selected because it achieves several requisites to be considered an optimal setting for the practice of epidemiological studies. Atahualpa has a low index of migration rate (less than 30 adults leave the village every year). Inhabitants are homogeneous regarding race/ethnicity, lifestyles, and diet, which is rich in oily fish. Atahualpa residents are natives with little evidence of cross-breeding (Amerindians). Almost all men belong to the blue-collar class and most women are homemakers. These consistencies reduce the risk of unexpected confounders at the time of data analyses.

3. Study design

The Atahualpa Project is a population-based prospective cohort study designed to reduce the increasing burden of noncommunicable neurological and cardiovascular diseases in rural Ecuador [2]. Before the study, our executive committee met with community leaders to explain the aims of the study and to learn about the culture and traditions of villagers. Field personnel were trained to assure uniformity in data collection. With the aid of satellite view (Google Earth, Google Inc., Mountain View, CA), an updated sketch of the village was performed to identify blocks and houses.

The study initially followed a three-phase cross-sectional design. During Phase I, Atahualpa residents were identified by means of a door-to-door survey and interviewed with validated questionnaires to identify those with suspected neurological disorders of interest. In Phase II, neurologists examined suspected cases as well as random samples of negative individuals, to assess reliability of field instruments and to determine prevalence of the investigated conditions. In Phase III, patients with confirmed clinical diagnosis underwent complementary tests. Thereafter, incoming Atahualpa residents have been recruited and followed-up by yearly door-to-door surveys and other overlapping sources to estimate incidence of the disorders of interest.

The Atahualpa Project includes a neuroimaging sub-study, which consists in the practice of brain MRI and MRA of intracranial vessels to all consenting participants aged ≥60 years, as well as to those presenting with specific neurological complains irrespective of age. Likewise, all participants aged ≥40 years have been invited for the practice of a head CT. Neuroimaging exams have been performed with the use of a Philips Intera 1.5-T MRI machine, and a Philips Brilliance 64 CT scanner, following predefined protocols. All exams have been independently reviewed by a neurologist and a neuroradiologist, with adequate kappa coefficients for inter-rater agreement for most lesions of interest. MRIs and MRAs were performed to assess stroke prevalence (over or silent), incidence, as well to better understand pathogenetic mechanisms underlying stroke and cognitive decline in the population, including the burden of silent neuroimaging signatures of cerebral small vessel disease (SVD). In contrast, CTs were performed to assess causes related to the presence of epilepsy, prevalence and correlates of carotid siphon calcifications, and other conditions such as pineal gland calcifications.

4. Cardiovascular health status

A major aim of the study has been the assessment of cardiovascular risk factors. For this, we used the seven health metrics proposed by the American Heart Association to assess the cardiovascular health (CVH) status, including smoking status, physical activity, diet, the body mass index, blood pressure, fasting glucose, and total cholesterol blood levels [3]. In a preliminary report including 642 stroke-free individuals aged ≥40 years, 13 (2%) persons had an ideal CVH status, 173 (28%) had an intermediate CVH status, and the remaining 430 (70%) had a poor CVH status. Among persons with poor CVH status, the poorest CVH metrics were blood pressure, fasting glucose, and the body mass index. In contrast, smoking status, diet, physical activity, and total cholesterol were satisfactory, with values in the poor range found in less than 10% of persons [4]. Then, we compared our results with those obtained in a population of Hispanics enrolled in the Northern Manhattan Study and found that Atahualpa residents scored better than Hispanics for all metrics with the exception of fasting glucose levels [5]. Potential reasons for these differences include the lower prevalence of highly processed foods in Atahualpa and the fact that most of their residents engage in regular physical activities. Also, most of them do not smoke and few drink soft drinks on a daily basis.

5. Stroke prevalence and incidence

A systematic review published in 2003 showed that stroke was less common in South America than in developed countries, but the distribution of stroke subtypes was quite different, with higher relative prevalence rates of cerebral SVD and intracranial hemorrhages, and lower prevalence of extracranial atherosclerosis and cardiogenic brain embolism than in the United States and European countries [6]. However, stroke burden is on the rise in rural regions of South America. We evaluated prevalence, pattern of subtypes, and pathogenetic mechanisms underlying stroke in Atahualpa. Accordingly, suspected cases were detected by a door-to-door survey. Then, neurologists evaluated suspected cases and randomly selected negative individuals, and confirmed patients underwent complementary exams. Overt strokes were diagnosed in patients who had experienced a rapidly developing event characterized by clinical signs of focal or global disturbance of cerebral function, lasting ≥24 h, with no apparent cause other than vascular. In such study, stroke prevalence was 31.15‰ in persons aged ≥40 years, which increased with age [7]. Hypertensive arteriolopathy was the most likely mechanism underlying strokes (55% patients) and 15% had intracranial hemorrhages. Extracranial atherosclerotic lesions or cardiac sources of emboli were not found in any case. Thereafter, we conducted a cohort study. For this, first-ever strokes occurring over 4 years were identified. Follow-up was achieved in 718 (89%) of 807 stroke-free individuals enrolled in the Atahualpa Project. Stroke incidence rate was 2.97 per 100 person-years of follow-up, which increased to 4.77 when only individuals aged ≥57 years were considered. Poisson regression models, adjusted for relevant confounders, showed that high blood pressure (IRR: 5.24) and severe edentulism (IRR: 5.06) were the factors independently increasing stroke incidence [8]. In a separate study, we assessed the prevalence of atrial fibrillation (AF) in community-dwelling older adults living in Atahualpa by means of 24-h Holter monitoring [9]. We found only 7 cases among 298 participants (crude prevalence: 2.3%, 95% CI: 1.1–4.8%). In only one of these cases, AF was associated with a silent lacunar infarction. Probable causes explaining this low prevalence of AF in Atahualpa included the low stature of the population (mean height: 147.9 ± 8.9 cm) and the high rate of oily fish intake (mean serving per week 8 ± 4). Both conditions have been associated with low AF prevalence in different populations.

6. Silent markers of cerebral SVD

The prevalence of silent markers of SVD in Atahualpa residents aged ≥60 years is similar to that reported from industrialized nations. According to our studies, 24% had moderate-to-severe white matter hyperintensities (WMHs) of presumed vascular origin, 13% had silent lacunar infarctions, and 9% had deep cerebral microbleeds [10]. We found that correlates of WMH are not the same as those of lacunar infarctions. Moderate-to-severe WMHs were associated with vitamin D deficiency [11], hypertensive retinopathy [12], poor sleep quality [13], and the apnea–hypopnea index [14], while lacunar infarctions were associated with peripheral artery disease [15], and the presence of high calcium content in the carotid siphon [16]. Such differences suggest that a sizable proportion of lacunar infarcts in our population is not related to SVD but to other causes, such as artery-to-artery embolism, reinforcing the old concept of the ‘fallacy of the lacunar hypothesis.’

7. Studies on sleep disorders

Some particularities make Atahualpa unique for assessing specific correlates of sleep disorders that could not be investigated even in areas where sophisticated technology is available. For example, the effects of sun exposure on the quality of sleep should better be assessed in tropical regions, where individuals are exposed to 12 daily hours of sunlight all over the year. In addition, shift working is limited and nighttime light pollution is scarce, providing a better scenario for studying sleep-related symptoms. From the many field instruments available for the detection of individuals with sleep disorders, we choose validated Spanish versions of the Pittsburgh sleep quality index, the Epworth sleepiness scale, and the Berlin questionnaire. In addition, a sleep unit, working under all the required standards as set forth by the American Academy of Sleep Medicine, was constructed for evaluating sleep-disordered breathing.

About 30% of Atahualpa residents aged ≥40 years have a poor sleep quality. We found a positive effect of dietary oily fish intake on sleep quality [17] and an important age-related effect on the association between frailty and a poor sleep quality [18]. In addition, a significant association between poor sleep quality and the presence of WMH was found, which was probably mediated by the disruption of periventricular fibers connecting the frontal lobes with basal ganglia and hypothalamus [13].

By the use of the Epworth sleepiness scale, one-fourth of the screened population had excessive daytime somnolence, which correlated poorly with several variables investigated. This was probably related to the fact that some questions might induce people to imagine themselves in situations that are not usual to them. The Epworth sleepiness scale and the Berlin questionnaire also correlated poorly with the presence of obstructive sleep apnea as assessed by polysomnography [19].

8. Cognitive function assessment

Assessment of cognitive impairment in rural areas of developing countries is complicated by illiteracy and other factors. The Montreal Cognitive Assessment (MoCA) was used to estimate cognitive performance in Atahualpa residents. However, we had to modify previously recommended cutoffs to avoid overdiagnosis of mild cognitive impairment. After correlating MoCA scores with neuroimaging signatures of cortical and subcortical atrophy, a cutoff of 19–20 points would better define cognitive impaired individuals in rural areas [20]. The total cerebral SVD score can also be used as a reliable predictor of poor cognitive performance, although its predictive power is not better than that of isolated markers of cerebral SVD [21]. We noticed that severe edentulism (used as a surrogate of chronic periodontitis), psychological distress, and the presence of high calcium content in carotid siphons were inversely associated with cognitive performance [22,23]. In contrast, we found a dose-dependent positive relationship between dietary oily fish intake and cognitive performance [24].

9. The search of surrogates for neuroimaging studies

Diagnosis of many neurological diseases requires the use of sophisticated technology, which is not available in rural areas. Efforts should be directed to find portable diagnostic tools that may help identify candidates for neuroimaging screening. In the Atahualpa Project, we have the unique opportunity to test the accuracy of non-expensive tests for identifying candidates for the practice of neuroimaging studies. We have assessed the value of the ankle–brachial index (ABI), the reliability of the neutrophil-to-lymphocyte ratio (NLR), and the accuracy of hypertensive retinopathy to detect individuals with SVD. Individuals with an abnormal ABI have four times of odds of having a silent lacunar infarct than those with a normal ABI [15], a high NLR is highly specific for detecting persons with at least one imaging signature of SVD [25], and individuals with hypertensive retinopathy Grades 2–3 are almost four times more likely to have moderate-to-severe WMH than those with only Grade 1 retinopathy [12]. We are still in the search of readily available biomarkers that allow the identification of apparently healthy persons at risk of suffering a catastrophic cerebrovascular event.

10. Comment

Only well-designed epidemiological studies may provide useful insights on the prevalence and correlates of noncommunicable neurological disorders in remote areas. The Atahualpa Project has found several particularities of these conditions in a rural population, suggesting that their pattern of expression is not the same as in the developed world. Further collaborative studies, using similar protocols, may allow public health authorities the implementation of cost-effective strategies directed to reduce the increasing burden of noncommunicable neurological disorders in underserved populations.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Compliance with ethical standards

The Institutional Review Board of Hospital-Clínica Kennedy, Guayaquil, Ecuador (FWA 00006867) approved the protocol and the written informed consent that individuals must sign before enrollment.

Acknowledgments

The authors would like to acknowledge the work of all the local investigators, technologists, and field personnel involved in the Atahualpa Project, as well as the collaboration of international Institutions that have sent medical students, technicians or physicians for in-site rotations, including the Langone Medical Center, New York University (New York, NY), the Department of Neurology, University of Chicago (Chicago, Il), the Sleep Disorders Center, Mayo Clinic College of Medicine (Jacksonville, Fl), the School of Medicine, Stonybrook University (New York, NY), the Department of Neurology, Loyola University (Chicago, Il), Hospital Medica Sur (Mexico City, Mexico), and the Center for Global Health, Universidad Peruana Cayetano Heredia (Lima, Peru).

Additional information

Funding

This study was supported by an unrestricted grant from Universidad Espiritu Santo – Ecuador, Guayaquil, Ecuador. The sponsor had no role in the design of the study, nor in the collection, analysis, or interpretation of data.