Abstract
Introduction
Headache is a hallmark symptom of chronic mountain sickness (CMS), a high altitude disease thought to be induced by excessive erythrocytosis (EE) in highlanders. Nevertheless, headache characteristics related to permanent high altitude residence remain understudied, as does the association between headache occurrence and hemoglobin concentration ([Hb]) in highlanders.
Methods
A pilot cross-sectional study was conducted in La Rinconada, Peru (altitude 5100 m), over a 2-y period. During all comers’ medical consultations, highlanders with no known history of chronic medical conditions (except EE/CMS) were questioned regarding headache symptoms. Headache characteristics were collected as well as clinical data and [Hb].
Results
Headache prevalence was 61% (200 of 330 highlanders). Clinical characteristics of headache attacks (ie, location, onset, intensity, quality, and duration) were similar between highlanders with and without EE. In multivariate analysis, headache occurrence was associated with lower oxygen saturation (P<0.001), higher body mass index (P<0.001), and lower self-reported sleep duration (P<0.001) but not with [Hb] or sex.
Conclusions
The absence of a statistically significant association between headache occurrence and [Hb] questions the use of headache as a symptom reflecting EE in CMS highlanders.
Introduction
High altitude headache is a well-defined symptom induced by acute hypoxic hypobaric exposure in lowlanders ascending to high altitude (ie, >2500 m), likely induced by trigeminovascular activation.1–3 In this environment, the occurrence of headache is also a cornerstone criterion for acute mountain sickness diagnosis, a frequent syndrome occurring in unacclimated travelers to high altitude areas, especially after a rapid ascent. 2 However, the current definition of high altitude headache, including headache resolution following descent, is inaccurate for the estimated 80 million highlanders permanently living above 2500 m.1,2 Few studies have been carried out to study headache prevalence, characteristics, mechanisms, or associated factors in high altitude populations, where chronic hypoxic exposure may trigger headache.3–7 In high altitude dwellers, headache is one of the cardinal symptoms defining chronic mountain sickness (CMS), a specific high altitude condition mostly affecting Andean highlanders and characterized by excessive erythrocytosis (EE) and a large range of clinical signs or symptoms. 2 In CMS, clinical manifestations are assumed to be the consequence of an exaggerated increase in hemoglobin concentration ([Hb]) leading to high blood viscosity, although the exact relationship and underlying mechanisms between EE and symptoms are not totally understood. 2 Thus, the association between headache symptom and EE has been challenged in cross-sectional studies conducted by us and others in CMS populations.8,9 Thus, in a cross-sectional preliminary study, we aimed to 1) describe the semiologic characteristics of headache in high altitude dwellers and 2) clarify the association between headache occurrence and the degree of hypoxemia and [Hb] in highlanders permanently living in La Rinconada, Peru (altitude 5100 m), the highest city in the world. 10
Methods
We conducted a pilot cross-sectional study in La Rinconada, a gold-mining city located at an altitude of 5100 m in southeastern Peru, considered the highest permanent settlement in the world. 10 After providing informed consent, volunteer highlanders were recruited during free preventive medical campaigns organized in the miner population center of La Rinconada in coordination with the mining cooperative over a 2-y period. The inclusion criterion for the study was residence at 5100 m. Exclusion criteria included chronic cardiovascular, respiratory, neurologic, metabolic, or renal disease; regular use of medication; and no newly diagnosed or suspected medical condition at the time of clinical evaluation—apart from high altitude–related conditions such as EE or CMS. The study was approved by the Ethics Committee of the Universidad de San Martin de Porres with FWA International Registry for the Protection of Human Subjects (No. 00015320, IRB No. 00003251).
Data Collection
During the medical consultation, the 7 clinical signs or symptoms of CMS (ie, breathlessness and/or palpitations, sleep disturbance, cyanosis, veinous dilatation, paresthesia, headache, and tinnitus) were collected and scored according to the Qinghai CMS score (ranging from 0=no symptom to 3=severe symptom) and reported further as the CMS clinical score.2,8 Hemoglobin was measured from a capillary blood sample using a HemoCue system (Hb 201+; HemoCue AB, Ängelholm, Sweden), and EE was defined as a [Hb]≥21 g·dL−1 for males and 19 g·dL−1 for females based on the current international consensus. 11 CMS diagnosis was retained when the total Qinghai CMS score was >5, including the presence of EE (adding 3 points to the CMS clinical score). 11 Peripheral oxygen saturation (SpO2) was measured with a finger sensor (Nellcor OxiMax N-65, Tyco Healthcare, Medtronic, Parkway, MN) after a 5-min rest period in a sitting position. Arterial blood pressure was measured manually using a sphygmomanometer. The semiologic characteristic of the headache (including location, intensity, quality, and attack duration) and demographic data of the participants were recorded during the medical consultation by the in-charge physician using a semistructured case report form. Using a previously described hierarchical diagnosis approach based on the International Classification of Headache Disorders (ICHD-3), the reported characteristics were used to classify the observed phenotypes as definite/probable migraine or tension-type headache (TTH) phenotypes.1,7,12
Statistical Analyses
Owing to the exploratory nature of this observational study, no sample size estimation was performed. Categorical data were expressed as absolute counts and percentages and compared using the Pearson χ2 test or Fisher's exact test as appropriate. Almost all continuous data exhibited skewed distributions and thus were expressed as median (25th–75th percentiles) and compared using the Mann-Whitney U test. A univariate analysis was first performed between highlanders who reported headache and those who did not. Then, to identify independent predictors of headache, we performed a multivariate logistic regression analysis using a backward stepwise selection of the variables; variables with a P value of <0.20 in the univariate analysis and/or those known as associated with headache at high altitude were initially entered into the model. Potential multicollinearity was ruled out using variance inflation factors computation. Headache characteristics and subsequent headache phenotype using a diagnostic algorithm were presented with alluvial plots. All statistical analyses were performed using the R software (version 4.2.2; R Foundation for Statistical Computing, Vienna, Austria).
Results
Among the 330 highlanders permanently living at 5100 m and included in this study, 61% (200 of 330) reported suffering from headache. Almost all participants (97%; 319 of 330) were natives from high altitude areas (≥2500 m). The main semiologic characteristics of headache attacks according to the EE status of participants are presented in Figure 1. Onset of the attacks was reported as progressive for 84% of the cephalalgic participants and as sudden for the others. No statistical difference was highlighted in term of location, intensity, duration, quality, and onset of the headache attacks between highlanders with and without EE (Figure 1; all P values >0.2). Associated symptoms (eg, nausea or scotoma) were reported in only 11 cephalalgic highlanders (6%). Headache attacks mostly could be classified as probable TTH (81% of the cephalalgic highlanders) and then as probable migraine (19%), without any significant distribution difference between highlanders with and without EE (Figure 1; P=0.23).
Alluvial plots showing the semiological characteristics of headache attacks reported by the highlanders with (A) and without (B) excessive erythrocytosis and classified using a diagnostic algorithm approach as probable migraine or probable tension-type headache. 12
Demographic and clinical characteristics of both cephalalgic and noncephalalgic highlanders are shown in Table 1, as well as univariate comparison between them. Briefly, cephalalgic highlanders were more likely to be male, to have a higher body mass index (BMI), to be more hypoxemic, to have higher diastolic arterial blood pressure, and to report lower sleep duration. The proportion of highlanders with self-declared sleep disturbance (corresponding to a subscore of 1–3 for the corresponding item in the Qinghai score) was similar between cephalalgic and noncephalalgic highlanders (27% vs 22%, respectively; P=0.34). Median hemoglobin level and proportion of participants suffering from EE were similar between the 2 subgroups of highlanders. By definition (because headache is one symptom used to diagnose CMS), CMS clinical score and proportion of CMS highlanders were significantly higher in the cephalalgic group of participants. In multivariate analysis, only BMI (odds ratio [OR] = 1.14; 95% CI, 1.05–1.24; P<0.001), self-reported sleep duration (OR = 0.73; 95% CI, 0.61–0.87; P<0.001), and SpO2 (OR = 0.88; 95% CI, 0.83–0.93; P<0.001) remained significantly associated with headache occurrence; higher BMI, lower sleep duration, and lower SpO2, respectively, were associated with an increased risk of headache.
Characteristics of the highlanders from La Rinconada included in the study, according to the presence or the absence of headache.
BMI, body mass index; SpO2, peripheral oxygen saturation; ABP, arterial blood pressure; CMS, chronic mountain sickness
Notes: Data are expressed as median (25th–75th percentiles) or number (%). Excessive erythrocytosis was defined as a hemoglobin concentration ≥21 g·dL−1 in males and 19 g·dL−1 in females; chronic mountain sickness was defined as a total Qinghai CMS score of ≥6, including the presence of an excessive erythrocytosis, according to the last international consensus. 11
a Comparison between highlanders with and without headache.
b Factors included in the multivariate analysis.
Discussion
In this observational study conducted in a population of highlanders living permanently at an altitude of 5100 m and mostly involved in mining activities, we reported a very high prevalence of headache (61%), higher than previously seen in the mining town of Cerro de Pasco, Peru (47%), which stands ∼1000 m lower. 6 Interestingly, we found no statistical relationship between headache symptom and [Hb] or EE status; the clinical (albeit basic) semiologic characteristics of headache attacks collected and the repartition of the probable headache phenotypes identified were similar between highlanders with and without EE (Figure 1). Furthermore, in the multivariate analysis, the only factors remaining associated with headache were the increase in BMI and the decreases in sleep duration and SpO2. In the absence of a statistical relationship between headache and [Hb], as observed previously,8,9 these preliminary results question the specificity of headache as one of the hallmark CMS symptoms, at least in this very high altitude population permanently living above 5000 m and therefore exposed to chronic and very severe hypoxia. In CMS, symptoms are likely to occur following the exaggerated increase in red blood cells, namely EE, and are assumed to be the consequence of a large increase in blood viscosity, although the relationship has not been formally demonstrated.2,13,14 In contrast with this assumption, our findings suggest that in highlanders, the occurrence of headache could be more directly linked to hypoxia induced by high altitude itself rather than to the increased [Hb]. The large and similar part of participants exhibiting a possible TTH phenotype in both subgroups of highlanders is concordant with this hypothesis because the TTH phenotype (ie, primary headache) shares diagnostic criteria with the secondary high altitude headache phenotype (ie, bilateral location and mild or moderate intensity). 1 This hypothesis also was suggested previously by others because headache has been shown to disappear in CMS highlanders soon after moving to sea level, whereas hematocrit remained unchanged.4,5 Nevertheless, our findings differ slightly from those from a cross-sectional study conducted in Cerro de Pasco, Peru (4300 m) in which a migraine pattern of headache was found to be associated with higher [Hb] but similar SpO2 levels compared with highlanders without headache. 6 Moreover, in this study, the prevalence of migraine and tension-type headaches increased with increasing age, whereas in a longitudinal study conducted over a 14-y period in La Rinconada, we did not observe any change in headache subscore over time. 13 Quispe et al 7 also reported in an epidemiologic study an increase in migraine prevalence with altitude in Peru compared with sea level, although the maximal altitude of residence was 3800 m only, and no information was collected regarding duration of high altitude residence, SpO2, [Hb], and EE or CMS status. Finally, further large-scale epidemiologic studies (ie, cross sectional and longitudinal), including physiologic measurements, conducted at different altitude levels are needed to confirm or refute our preliminary observation and to definitively unravel the relationship between the most common CMS symptoms such as headache and EE, the degree of hypoxia, age, and/or duration of high altitude exposure as well as to better characterize the headache phenotype(s) associated with permanent life at high altitude.
We should acknowledge some limitations to this study. The clinical data (notably the semiologic data regarding headache attacks) were relatively basic and did not allow us to classify the headache attacks more precisely than “probable” phenotypes. 1 Moreover, the absence of interventions, such as, for instance, descending to lower altitude, did not allow us to differentiate primary and secondary headache induced by hypoxia. 1 However, despite sharing characteristics with primary headaches, headaches associated with (chronic) hypoxic hypobaric exposure are assumed to be secondary headaches, resulting from trigeminovascular activation.3–5 Nevertheless, further attempts are needed to better characterize headache pattern(s) in very high altitude populations using well-validated and well-structured questionnaires such as the Headache-Attributed Restriction, Disability, Social Handicap and Impaired Participation (HARDSHIP) Questionnaire in larger samples of highlanders, allowing one to assess both headache characteristics and global burden.7,12 Additionally, we cannot rule out that others environmental factors or undiagnosed chronic medical conditions may have contributed to headache occurrence in this high altitude miner population. In particular, the potential contribution of occupational exposure (eg, mercury or carbon monoxide poisoning) needs to be investigated further. 10
In conclusion, our findings, as well as previous epidemiologic findings,8,9 question the relevance of headache as a specific symptom of CMS in highlanders rather than a chronic hypoxia-induced symptom per se but are, by definition and design, insufficient to conclude on any causal relationship. Further epidemiologic, physiologic, and clinical studies are required to determine the specificities of headache attacks encountered by the high altitude highlanders as well as potential occupational confounders and a confirmed causal relationship according to Bradford Hill criteria.
Footnotes
Author Contribution(s)
Data Availability
The datasets supporting the findings of this study are available from the corresponding author on reasonable request.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Ethical Considerations
This study was approved by the Ethics Committee of the Universidad de San Martin de Porres with FWA International Registry for the Protection of Human Subjects (No. 00015320, IRB No. 00003251) and was conducted in accordance with the Declaration of Helsinki.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Samuel Verges was supported by the Grenoble Alpes University Foundation, the “Fonds de dotation AGIR pour les maladies chroniques,” the Air Liquide Foundation, and the French National Research Agency (ANR-12-TECS-0010) in the framework of the “Investissements d’avenir” Program (ANR-15-IDEX-02).
