History of Migraine Predicts Headache at High Altitude

Abstract

Davis, Christopher, Elaine Reno, Edward Maa, and Robert Roach. History of Migraine Predicts Headache at High Altitude. High Alt Med Biol. 17:300–304, 2016.— Objective: To characterize the spectrum of headaches and their association with migraine history within a population of recreational hikers above 4300 m. Methods: Using a cross-sectional survey design, a convenience sample of 667 hikers participated in a written survey after descent from Mount Gray/Torreys (4349 m). Headaches were characterized as migraine, high altitude headache (HAH), and/or acute mountain sickness (AMS) using International Headache Society Lake Louise AMS scoring criteria. A univariate odds ratio was calculated to determine whether a history of migraine increased the risk of migrainous headache. Multivariate logistic regression was used to assess whether a priori identified risk factors such as age, sex, recent alcohol consumption, home elevation, and self-reported fluid intake and whether summit success increased the risk of any headache at altitude. Results: Sixty percent of hikers were male with an age range of 17–62 years. Eighty percent reached the summit of Mount Grays/Torreys (4349 m). Seventy-nine percent of participants resided in Colorado; the median elevation of residence for subjects was 1697 m (interquartile range 1557–1765 m). HAH occurred in 39% of hikers, while AMS occurred in 26% of hikers. A history of migraine was associated with increased risk of any headache at altitude (OR: 2.49, 95% CI: 1.62–3.65) and was strongly associated with the development of migrainous headache while at altitude (OR: 14.05, 95% CI: 5.49–35.93). Conclusions: A history of migraine is a risk factor for the development of headache at altitude and is strongly associated with the development of migrainous headache.

Introduction

Headache is the shared clinical end point for several high altitude disorders. Three headache syndromes are commonly recognized at high altitude as follows: acute mountain sickness (AMS), high altitude headache (HAH), and migraine triggered by high altitude (Serrano-Duenas, 2007). While headache is the cardinal symptom of AMS, HAH is recognized as a distinct clinical entity (Olesen and Steiner, 2004). A recent study observed that 52% of individuals with HAH did not have AMS as defined by the Lake Louis criteria (Silber et al., 2003). Classification of these disorders is not merely academic. Misidentifying the etiology of headache at high altitude may lead to clinical errors with consequences (Karle and Auerbach, 2014). Unfortunately, overlap in the classification of these headache disorders can lead to diagnostic uncertainty.

The aim of this study was to classify the spectrum of headache disorders occurring in a cohort of hikers above 4300 m using criteria outlined by the International Headache Society and the Lake Louise AMS score (Roach et al., 1993).

A secondary aim was to describe the epidemiological risk factors for development of headache at high altitude. High altitude has previously been associated with a higher prevalence of migraine (Arregui et al., 1991). Our study specifically explored the link between migraine history and the development of migrainous headache while at altitude. A better understanding of the link between these disorders could help improve classification and highlight shared pathophysiology.

Methods

A convenience sample of 667 hikers participated in a written survey after attempting to hike Mount Grays/Torreys, Summit County, Colorado. Subjects were enrolled in August of 2008. On the morning of the survey, most subjects drove a private vehicle from the Denver area (1610 m) to the trailhead of Mount Gray/Torreys at 3438 m and then summited at an altitude of 4349 m. Surveys were completed at the main Grays/Torreys trailhead, after descent from the summit. The duration of exposure to altitude greater than 3000 m was less than 12 hours before survey administration. Written consent was obtained before survey administration. Inclusion criteria were age over 17 years, English literacy, and a summit attempt that day of Mount Grays/Torreys. Survey questions included demographic characteristics such as age or sex, historical characteristics of previously experienced headaches, Lake Louise AMS scores, home altitude, summit success, recent alcohol consumption, and hike length, along with self-reported hydration and physical fitness. Reported headache characteristics were not exclusive of one another.

Headaches experienced on the day of the survey were characterized using International Headache Society diagnostic criteria for HAH and migraine (Table 1), with one exception (Headache Classification Committee of the International Headache Society, 2013). As the surveys were administered at an altitude of 3438 m, data were not captured regarding resolution of headache symptoms within 8 hours upon descent below 2500 m. Surveys were also scored using the Lake Louise Criteria to assess for AMS. However, as the majority of subjects slept at intermediate altitude (1610 m) on the evening before survey administration, data on sleep quality were not collected. To determine whether a history of migraine was present, self-reported historical headaches were also classified based on IHS criteria (Headache Classification Committee of the International Headache Society, 2013). To qualitatively assess for overlap of headache criteria, all headaches were then organized using a nonexclusive categorization strategy (i.e., a given headache could be categorized as both AMS and migraine if the headache met criteria for both). A univariate odds ratio was calculated to determine whether a history of migraine increased the risk of migrainous headache. A chi-squared test was performed to compare Lake Louise AMS scores (LLS) for those reporting pulsatile headache versus those with dull or aching headache. A multivariate logistic regression model was used to assess whether a priori identified risk factors such as age, sex, recent alcohol consumption, home elevated, self-reported fluid intake, and summit success increased the risk of headache at altitude. Statistical analysis was done using JMP Pro 11.1.1 and R version 2.15.3. The Colorado Multiple Institutional Review Board (Protocol # 08-0320) granted approval for this study.

Table 1.

International Headache Society Diagnostic Criteria for Migraine and High Altitude Headache

High altitude headache
A. Headache with at least two of the following characteristics and fulfilling criteria B, C, and D:
1. Bilateral
2. Frontal or frontotemporal
3. Dull or pressing quality
4. Mild or moderate intensity
5. Aggravated by exertion, movement, straining, coughing, or bending
B. Ascent to altitude above 2500 m
C. Headache develops within 24 hours after ascent
D. Headache resolves within 8 hours after descent^a
Migraine (without aura)
A. At least five attacks fulfilling criteria B–D
B. Headache attacks lasting 4–72 hours (untreated or unsuccessfully treated)
C. Headache has at least two of the following characteristics:
1. Unilateral location
2. Pulsing or throbbing quality
3. Moderate or severe pain intensity
4. Aggravation by, or causing avoidance of routine physical activity
D. During headache at least one of the following:
1. Nausea and/or vomiting
2. Photophobia and phonophobia
E. Not attributed to another disorder

For this study, data not collected on resolution of headache.

Results

Approximately, 800 individuals were approached to participate in this study. Six hundred sixty-seven subjects completed the survey resulting in an estimated capture rate of 80%. Eighty-two surveys were discarded due to incomplete reporting. Sixty-two percent of subjects were male. The age range was 17–62 years old. Eighty-eight percent of hikers successfully summited Mount Grays/Torreys (4349 m). Average total hike time was more than 6 hours (367 ± SD 82.9 minutes) with an approximate ascent rate of 250 m/hr. Seventy-nine percent of survey participants resided in Colorado; the median elevation of residence for subjects was 1697 m (interquartile range 1557–1765 m). On the morning of the survey, 73% of participants drove from the Denver area (1610 m). The remaining participants slept the evening before the study in the Summit County area at an average elevation of 2740 m.

Forty-one percent of those surveyed experienced some type of headache (242 of 582). The aggregated headache characteristics for this cohort are summarized in Table 2. Ninety-eight percent of survey participants who experienced headache broadly met criteria for HAH based on International Headache Society guidelines (238 of 242). Ten percent (23 of 242) of headaches could be classified as migraine without aura, while 7% of headaches met criteria for both HAH and migraine (19 of 260). Twenty-six percent of participants met criteria for AMS based on Lake Louise Scoring (155 of 647). The median LLS was 4 (interquartile range of 3–5). There was no difference in LLS scores between those individuals with pulsatile versus dull/aching headache (p = 0.43).

Table 2.

Headache Characteristics

	Headache characteristics N = 242
Severity
Mild	169 (70.0%)
Moderate	71 (29.2%)
Severe	2 (0.8%)
Characteristics of pain
Aching/Dull	155 (64.0%)
Pulsing/throbbing	48 (19.8%)
Tight	37 (15.3%)
Burning	2 (0.8%)
Location
Diffuse	43 (17.8%)
Bilateral	180 (74.4%)
Unilateral	32 (13.2%)
Temporal	21 (8.7%)
Frontal	105 (43.4%)
Occipital	63 (26.0%)
Ocular	70 (28.9%)
Associated symptoms
Photophobia	35 (14.5%)
Phonophobia	20 (8.3%)
Nausea	53 (21.9%)
Vomiting	3 (1.2%)
Tearing	19 (7.9%)
Runny nose	103 (42.6%)
Visual changes	33 (13.6%)
Activity exacerbates
No	152 (62.8%)
Yes	90 (37.2%)
Onset of HA
[Time zero = start of hike]
<1 hour	40 (16.5%)
2–3 hours	115 (47.5%)
4–5 hours	57 (23.6%)
>5 hours	30 (12.4%)

HA, headache.

Table 3 summarizes the headaches in this study cohort and highlights the overlap between HAH, AMS, and migraine. Twenty percent of participants were classified as having a history of migraine (N = 119). A history of migraine was strongly associated with development of migrainous headache when summiting Mount Grays/Torreys (OR: 14.05, 95% CI: 5.49–35.93).

Table 3.

Contingency Table of Headache Syndromes at Altitude

	Any HA	HAH	HAH w/o AMS	AMS	Migrainous HA
Any HA	242	238	83	155	23
HAH	238	—	83	155	17
HAH w/o AMS	83	83	—	0	1
AMS	155	155	0	—	23
Migrainous HA	23	17	1	23	—

AMS, acute mountain sickness; HAH, high altitude headache.

Risk factors for the development of any headache at high altitude estimated by logistic regression are summarized in Table 4. A history of migraine was associated with increased risk of any headache at altitude (OR: 2.49, 95% CI: 1.62–3.65). While past headache frequency was associated with an increased risk of headache, this risk factor was highly correlated with history of migraine and so was dropped from the final logistic regression model (Spearman's ρ = 0.708). When partitioned by age, individuals aged 35–50 years old were less likely to experience HAH compared to the younger cohort (OR: 0.53, 95% CI: 0.33–0.84), while individuals more than 51 years old were even less likely to experience HAH (OR: 0.46, 95% CI: 0.22–0.98). Residing at an elevation of 1000 meters or more (472 of 582) was associated with less HAH (OR: 0.59, 95% CI: 0.41–0.92). Sex, self-reported fitness and hydration, and recent alcohol consumption were not associated with increased risk of headache.

Table 4.

Risk Factors for Development of Headache at High Altitude Modeled by Logistic Regression

	Odds ratio
History of migraine
Yes	2.49 (95% CI: 1.62–3.65)
No
Age
17–35
36–50	0.53 (95% CI: 0.33–0.84)
51+	0.46 (95% CI: 0.22–0.98)
Number of alcohol beverages in last 24 hours
None	1.09 (95% CI: 0.74–1.60)
1 or more
Exercise Frequency
Greater than weekly
Between weekly and monthly	1.34 (95% CI: 0.64–2.01)
Less than 2–3/month	5.48 (95% CI 1.03–29.05)
Sex
Male
Female	0.78 (95% CI: 0.52–1.17)
Successfully summited?
No
Yes	1.21 (95% CI: 0.61–2.39)
Hydration in last 24 hours
1.5 L or less
Greater than 1.5 L	1.13 (95% CI: 0.74–1.70)
Home elevation
1000 m or less
Greater than 1000 m	0.59 (95% CI: 0.41–0.92)

Discussion

Up to 80% of individuals experience some type of headache with ascent to high altitude (Wilson et al., 2009). In this cohort, while individuals experienced a relatively short exposure to high altitude, 41% of participants experienced some type of headache. Three types of headache disorders commonly occur in this environment as follows: AMS, HAH, and migraine precipitated by ascent to high altitude (Serrano-Duenas, 2007). However, less common headache syndromes are also recognized at altitude, especially during exercise. These include benign exertional headache and exercise-induced migraine. The incidence of these disorders at high altitude is not currently known. The diagnostic landscape is further confounded by dehydration and hangover, which are common mimics for AMS. A complex classification system, along with overlapping diagnostic criteria, can be frustrating to both clinicians and researchers alike. Misidentifying a headache as migrainous, rather than AMS, can lead to costly and risky evacuation (Karle and Auerbach, 2014). Mistaken diagnosis could also lead a clinician to choose an ineffective treatment strategy (Utiger et al., 2002). Correctly classifying headache disorders at altitude also has important research implications. Improving the accuracy of classification could reduce potential confounding and could help establish shared or discrete pathophysiology.

The author's first goal was to measure the incidence of HAH, AMS, and migraine upon ascent to high altitude. The authors used the Lake Louise Score to grade AMS and current IHS criteria to classify HAH and migraine (without aura). Discriminating between these headache syndromes is largely reliant on clinical descriptors. Using current headache classification methods, these descriptors largely overlap. Table 3 is a contingency table that summarizes the spectrum of headaches observed in this study. Overall, the vast majority of headaches could be classified as HAH (238 of 242). All migrainous headaches also met criteria for AMS (23 of 23), with headache and nausea accounting for the preponderance of symptoms. Most migrainous headaches could also be classified as HAH (19 of 23). Given the lack of a gold standard to differentiate between these headache syndromes, the authors chose to describe these headaches qualitatively. Ultimately, Table 3 highlights why further effort is warranted to refine these diagnostic criteria to improve both accuracy and specificity.

The vast majority of headaches in this cohort could be classified as HAH. However, there are conflicting data on the pain quality of HAH. Silber described HAH as “dull or pressure-like” in a cohort of mountaineers ascending above 5100 meters (Silber et al., 2003). This clinical description is reflected in current diagnostic criteria for HAH (Table 1). Alternatively, Serrano-Duenas remarked that the majority of HAHs were pulsatile in nature, typically a clinical descriptor for migrainous headache (Serrano-Duenas, 2005). Both studies described headaches that were bilateral and/or global, and worse with exertion. In this study, most headaches were described as mild, bilateral, and dull or aching in nature (Table 2). While exacerbation of headache by exercise was common in this cohort (37%), it was not reported in a majority of participants.

High altitude has been associated with risk of migraine, or at least migrainous headache, but controlled studies are lacking. Arregui et al. noted a higher than normal prevalence of migraine in a population of high altitude residents of South America (Arregui et al., 1991), while Watson described a case report of migraine induced by travel to 1800 meters in Colorado (Watson et al., 2011). The prevalence of migraine in the United States is estimated at 15%–20% (Smitherman et al., 2013). Data regarding the prevalence of migraine specific to Colorado are not available. In this cohort, 20% of individuals met criteria for a history of migraine. Our data showed a strong association between history of migraine and development of migrainous headache (OR: 14.05, 95% CI: 5.49–35.93). Whether these migrainous headaches were true migraines, or in actuality AMS, remains unclear. In this cohort, certainly several migraine triggers were possible. Anecdotally (and from personal experience), hikers on Colorado's 14Kers (Any of the 53 mountain peaks in Colorado above 14,000 ft) arise early, drive to a high-elevation trailhead, and hike rapidly above 4000 m. Therefore hypoxia, physical exertion, sleep deprivation, and dehydration were all possible triggers for migraine headache in this study.

Recently, Broessner described a subgroup of individuals without migraine history who experienced a migrainous headache with exposure to hypoxia (Broessner et al., 2016). Given the potential overlap in clinical criteria for migraine and HAH, and the observation that migrainous headache can occur in those individuals without a migraine history at simulated altitude, this raises several important questions. Should the criteria for HAH be changed to reflect this migrainous subpopulation with pulsatile headache? Or is this migrainous subpopulation a separate clinical entity from the dull aching headache described by IHS criteria? In 2002, Schneider reported in a brief abstract that milder AMS has a predominantly tension-type headache, while more severe AMS mimics migraine (Abstracts of the World Congress on Mountain Medicine, 2002). However, we found no difference in AMS scores in those reporting pulsatile vs dull/aching headaches (p = 0.43)

In this study, a history of migraine was also associated with the development of any headache at high altitude (OR: 2.49, 95% CI: 1.62–3.65). Recent studies have demonstrated a significant association between history of migraine and risk of HAH (Burtscher et al., 2011; Bian et al., 2013). While the vast majority of headaches in this study also qualified as HAH, the overlap with other headache syndromes makes further inference impossible. Nonetheless, it certainly appears that migraineurs are predisposed to development of headache at altitude, regardless of cause.

Age was also an important risk factor for HAH. Older age was associated with a significant decrease in the risk of HAH, an effect that becomes more pronounced over 50 years of age. While Silber noted a similar protective effect of age, neither Burtscher nor Bian found an association between age and risk of HAH (Silber et al., 2003; Burtscher et al., 2011; Bian et al., 2013). On review, age appears to be relatively homogenous among these studies. It is certainly possible that Burtscher and Bian found no association between age and HAH because these studies were performed on populations with a relatively narrow age range. In the Bian study, participant age ranged only from 18 to 45 years. In contrast, nearly 10% of participants in our study were more than 50 years of age, while 27% were between 36 and 50 years of age. We believe that increasing age, particularly when measured over a representative age range, is associated with a lower risk of HAH. In broad terms, the incidence of all primary headache disorders declines with age (Lipton et al., 1993). Whether there is a common protective mechanism related to aging, which is shared among AMS, HAH, and migraine, is worthy of further study.

In this study, self-reported fluid intake, used as a proxy for hydration status, was not a risk factor for HAH. While this is consistent with previous findings, caution is advised due to well-documented recall bias in self-reported recall (Burtscher et al., 2011). Self-reported exercise frequency, used as a proxy for fitness, was not associated with increased risk of HAH. This is consistent with previous observations that fitness is not associated with AMS, but again these results are limited (Richalet, 1988; Milledge et al., 1991). Primary headache disorders at low altitude are typically more common in females (Macgregor et al., 2011), but gender was not associated with an increased risk of HAH in the present study; this is also consistent with previous data (Burtscher, 1999; Burtscher et al., 2011; Bian et al., 2013).

In this cohort, a home residence above 1000 m was also associated with a lower risk of headache at altitude (OR: 0.59, 95% CI: 0.41–0.92). Nearly 80% of the study subjects lived in Colorado, with the majority living above 1500 m (1557–1765 m). Intuitively, these data suggest that acclimatization to the low end of moderate altitude (1500–2500 m) is associated with a lower risk of headache at high altitude, regardless of cause.

Limitations

There are several important limitations to this study. This was a survey of convenience and thus selection bias must be considered. Those with severe headaches may not have chosen to participate. However, given the comparable incidence of migraine and AMS to previous of studies, this appears less likely. However, these errors would be random in nature and, therefore, less likely to lead to systematic bias. In addition, the relatively brief exposure to hypoxia limits the ability of this study to examine and differentiate AMS from other headache syndromes, as rates of AMS typically peak after 24 hours. However, the rates of AMS observed in this study are similar to previous reports. In addition, the survey was not designed to capture or classify other types of headache syndromes such as benign exertional headache or exercise induced migraine. This was an additional reason the authors chose to limit headache classification to a qualitative description. Finally, self-reports of hydration and fitness are subject to recall bias and may not be reliable.

Conclusions

A history of migraine is a risk factor for the development of headache at altitude and is strongly associated with the development of migrainous headache.

Footnotes

Acknowledgments

This study was supported by the Altitude Research Center, Department of Emergency Medicine, University of Colorado Anschutz Medical Campus. The authors thank Dr. Kirsten Nielsen for her assistance with data collection. The authors also thank the University of Colorado Denver Research Consulting Laboratory for statistical assistance.

Author Disclosure Statement

No competing financial interests exist.

References

Abstracts of the V World Congress on Mountain Medicine and High Altitude Physiology, and the VIII Congress of the Spanish Society of Medicine and Mountain Rescue. Barcelona, Spain, April 18–22, 2002. (2002). High Alt Med Biol, 3:97–140.

Arregui

, Cabrera

, Leon-Velarde

, Paredes

, Viscarra

, and Arbaiza

. (1991). High prevalence of migraine in a high-altitude population. Neurology, 41:1668–1669.

Bian

, Zhang

, Gao

, Li

, Yu

, Liu

, Dong

, Chen

, and Huang

. (2013). Risk factors for high-altitude headache upon acute high-altitude exposure at 3700 m in young Chinese men: A cohort study. J Headache Pain, 14:35.

Broessner

, Rohregger

, Wille

, Lackner

, Ndayisaba

, and Burtscher

. (2016). Hypoxia triggers high-altitude headache with migraine features: A prospective trial. Cephalalgia, 36:765–771.

Burtscher

(1999). High altitude headache: epidemiology, pathophysiology, therapy and prophylaxis [article in German]. Wien Klin Wochenschr, 111:830–836.

Burtscher

, Mairer

, Wille

, and Broessner

. (2011). Risk factors for high-altitude headache in mountaineers. Cephalalgia, 31:706–711.

Headache Classification Committee of the International Headache, Society. (2013). The International Classification of Headache Disorders, 3rd edition (beta version). Cephalalgia, 33:629–808.

Karle

, 3rd, and Auerbach

. (2014). Migraine headache confounding the diagnosis of acute mountain sickness. Wilderness Environ Med, 25:60–68.

Lipton

, Pfeffer

, Newman

, and Solomon

. (1993). Headaches in the elderly. J Pain Symptom Manage, 8:87–97.

10.

Macgregor

, Rosenberg

, and Kurth

. (2011). Sex-related differences in epidemiological and clinic-based headache studies. Headache, 51:843–859.

11.

Milledge

, Beeley

, Broome

, Luff

, Pelling

, and Smith

. (1991). Acute mountain sickness susceptibility, fitness and hypoxic ventilatory response. Eur Respir J, 4:1000–1003.

12.

Olesen

, and Steiner

TJ.

(2004). The International classification of headache disorders, 2nd edn (ICDH-II). J Neurol Neurosurg Psychiatry, 75:808–811.

13.

Richalet

J-P

, Keromes

, Dersch

, Corizzi

, Mehdioui

, Pophillat

, Chardonnet

, Tassery

, Herry

, and Rathat

. (1988). Caractéristiques physiologiques des alpinistes de haute altitude. Sci Sports, 3:89–108.

14.

Roach

, Bartsch

, Hackett

, and Oelz

. (1993). The Lake Louise AMS scoring consensus committee. In: Hypoxia and Molecular Medicine. Sutton

, Houtson

, Coates

, eds. Burlington VT: Queen City Printers. pp. 272–274.

15.

Serrano-Duenas

(2005). High altitude headache. A prospective study of its clinical characteristics. Cephalalgia, 25:1110–1116.

16.

Serrano-Duenas

(2007). High-altitude headache. Expert Rev Neurother, 7:245–248.

17.

Silber

, Sonnenberg

, Collier

, Pollard

, Murdoch

, and Goadsby

. (2003). Clinical features of headache at altitude: A prospective study. Neurology, 60:1167–1171.

18.

Smitherman

, Burch

, Sheikh

, and Loder

. (2013). The prevalence, impact, and treatment of migraine and severe headaches in the United States: A review of statistics from national surveillance studies. Headache, 53:427–436.

19.

Utiger

, Eichenberger

, Bernasch

, Baumgartner

, and Bartsch

. (2002). Transient minor improvement of high altitude headache by sumatriptan. High Alt Med Biol, 3:387–393.

20.

Watson

, Torres-Trejo

, and Gutmann

. (2011). Altitude induced migraine. W V Med J, 107:22–23.

21.

Wilson

, Newman

, and Imray

. (2009). The cerebral effects of ascent to high altitudes. Lancet Neurol, 8:175–191.