Prevalence of cluster headache in the republic of georgia: results of a population-based study and methodological considerations

Prevalence of cluster headache in the Republic of Georgia:results of a population-based study and methodologicalconsiderations Z Katsarava1, A Dzagnidze2, M Kukava2, E Mirvelashvili3, M Djibuti3,4, M Janelidze2, R Jensen5, LJ
Stovner6 & TJ Steiner7, on behalf of the Global Campaign to Reduce the Burden of Headache
Worldwide and The Russian Linguistic Subcommittee of the International Headache Society
1Department of Neurology, University of Essen, Essen, Germany, 2Department of Neurology and 3School of Public Health, Tbilisi Medical
University, 4NGO, ‘Partnership in Research and Action for Health’, Tbilisi, Georgia, 5Danish Headache Centre, Glostrup University,
Copenhagen, Denmark, 6Norwegian National Headache Centre, Department of Neuroscience, Norwegian University of Science and Technology,
and St Olavs Hospital, Trondheim, Norway, and 7Division of Neuroscience and Mental Health, Imperial College London, London, UK
Katsarava Z, Dzagnidze A, Kukava M, Mirvelashvili E, Djibuti M, Janelidze M,Jensen R, Stovner LJ, Steiner TJ, on behalf of the Global Campaign to Reduce theBurden of Headache Worldwide and The Russian Linguistic Subcommittee of theInternational Headache Society. Prevalence of cluster headache in the Republic ofGeorgia: results of a population-based study and methodological considerations.
Cephalalgia 2009. London. ISSN 0333-1024 We present a study of the general-population prevalence of cluster headache inthe Republic of Georgia and discuss the advantages and challenges of differentmethodological approaches. In a community-based survey, specially trainedmedical residents visited 500 adjacent households in the capital city, Tbilisi, and300 households in the eastern rural area of Kakheti. They interviewed all(n = 1145) biologically unrelated adult occupants using a previously validatedquestionnaire. The household responses rates were 92% in Tbilisi and 100% inKakheti. The survey identified 32 persons with possible cluster headache, whowere then personally interviewed by one of two headache-experienced neurolo-gists. Cluster headache was confirmed in one subject. The prevalence of clusterheadache was therefore estimated to be 87/100 000 (95% confidence interval < 258/100 000). We used a conservative approach, which has an obvious advan-tage of high-quality data collection, but is very demanding of manpower andtime. ᮀCluster headache, prevalence, epidemiology Zaza Katsarava, Department of Neurology, University of Essen, Hufelandstrasse 55,45122 Essen, Germany. Tel. + 49-201-723-2467, fax + 49-201-723-5919, e-mail23 June 2008, accepted 2 December 2008 Europe. No data from the eastern European or Introduction
post-Soviet countries are available yet.
Cluster headache (CH) is a strictly unilateral, severe The Republic of Georgia is a Eurasian country in or very severe retro-orbital or temporal headache the Caucasus, located on the east coast of the Black lasting 15–180 min and accompanied by ipsilateral Sea and bordered on the north by Russia, on the cranial autonomic symptoms and a sense of rest- south by Turkey and Armenia and on the east by lessness or agitation (1). This set of symptoms Azerbaijan. Its current population is 4.4 million (2).
makes the condition easily recognizable but, Between 1921 and 1991, Georgia was part of the because it is relatively uncommon, it is method- Soviet Union; it declared independence after the ologically challenging to measure its prevalence in collapse of USSR. Georgia has many socio-economic general population samples. In fact, few studies and cultural similarities with other, mostly Euro- exist, and all so far have been done in Western pean post-Soviet countries (Armenia, Azerbaijan, Blackwell Publishing Ltd Cephalalgia, 2009 the European part of Russia, Byelorus, Ukraine and Moldova). We present a study on the prevalence ofCH in Georgia, which is part of an epidemiological We interviewed 1701 respondents in 500 house- survey on the prevalence of primary headaches in holds in Tbilisi and 560 in 300 households in Kakheti. The household response rates were 92% inTbilisi (38 households refused contact with theinterviewers) and 100% in Kakheti. The target population consisted of 1145 biologically unrelated The study protocol was approved by the Georgian adults (e.g. cohabiting couples, but not the blood relatives of either partner). The mean age of the The methodology of the study and the validation target population was 45.4 Ϯ 12 years, 60% of of the questionnaire have been reported previously (3, 4). The study was performed in the capital city,Tbilisi, and in a rural area of Kakheti in the east part of Georgia. We used the ‘cold-calling’ method ofdoor-to-door survey (calling unannounced). Prior to We identified 32 subjects with or having had pos- the study, medical residents were specifically trained sible CH, six of them male, with mean age 39 Ϯ 12 by headache-experienced neurologists (Z.K., A.D., years. All were invited to a neurological consulta- M.K.) to identify migraine, tension-type headache tion to be re-interviewed and examined.
(TTH) and CH. The residents visited adjacent house-holds in pre-defined districts in Tbilisi and villages in Kakheti and asked all biologically unrelatedadults (> 18 years old) in the households to undergo One respondent with definite CH was identified: a a structured, questionnaire-based headache inter- 32-year-old man who had experienced his first view, which aimed not to diagnose CH but to screen cluster attack at the age of 27. Thereafter, he had for possible cases according to these criteria: suffered one or two cluster bouts per year of 14–16weeks’ duration. These were characterized by A: recurrent headache that was at least two of the recurrent attacks, each lasting 45–60 min, of strictly following: (i) severe or very severe; (ii) strictly right-sided retro-orbital stabbing headache accom- unilateral; (iii) retro-orbital or temporal; and panied by lacrimation, eyelid oedema and rhinor- B: at least one of the following accompanying rhoea. He used subcutaneous sumatriptan for acute symptoms or signs: (i) ipsilateral conjunctival treatment but had had no preventative medication.
injection or tearing; (iii) ipsilateral nasal con- The lifetime prevalence of CH was estimated junction or rhinorrhoea; (iii) ipsilateral eyelid therefore to be 87 per 100 000 (upper 95% CI limit oedema; (iv) ipsilateral forehead or facial sweat- ing; (v) ipsilateral miosis or ptosis; and (vi) asense of restlessness or agitation.
Respondents fulfilling these criteria, either for current or past headache, were later re-interviewed In 31 respondents the suspected diagnosis of CH and examined in person by one of two headache was rejected. Sixteen had migraine, eight had TTH, experienced neurologists (A.D. and M.K.). CH was five had a combination of migraine and TTH, one diagnosed by International Classification of Head- had acute frontal sinusitis and one had a brain ache Disorders, 2nd edn criteria (1). Symptomatic tumour. All those with migraine had at least one headaches were ruled out by clinical examination cluster-like autonomic symptom which led to the and, where necessary, cranial computed tomogra- suspicion of CH. No obvious reason existed for misdiagnosis of CH in the eight respondents whohad TTH.
Discussion
Crude prevalence of CH was expressed as thenumber of cases per 100 000 inhabitants. The 95% To the best of our knowledge, this is the first study confidence interval (CI) was calculated by the to provide data on the prevalence of CH in an method of Bortz (5). The level of significance was eastern European country. In a general population sample of 1145 people in the Republic of Georgia, Blackwell Publishing Ltd Cephalalgia, 2009 Prevalence of cluster headache in Georgia we identified one case of CH, past or present. This rate of CH of 326 per 100 000 (95% CI 120, 720 per corresponds to a lifetime prevalence of 87 per 100 000), a figure not compatible with ours. In 100 000 with a 95% CI upper limit of 258 per Denmark, a sample of 1000 inhabitants of Copen- 100 000. The power of the study was not enough to hagen was surveyed for primary headaches by estimate both borders of the 95% CI.
face-to-face interview and neurological examination Studying the epidemiology of CH is challenging (7). One case of CH was identified in this study, a for two key reasons. First, the prevalence of CH is finding similar to ours and again not statistically low. Second, its diagnosis can be unreliable in a compatible with the prevalence recorded in Vågå.
non-clinical setting. The clinical features of CH may Studies like this are very costly, and therefore hard be striking when they occur together, but many of to replicate in Western Europe and North America, them—stabbing unilateral headache and each of the where human resource costs are high. However, in typical autonomic symptoms—may also feature not developing countries it may still be possible to only in other trigeminal autonomic cephalalgias perform large-scale epidemiological studies based (which are rare) but also in migraine. Our question- on face-to-face interviews, at least if they can be naire was sensitive rather than specific for CH, performed by lay persons or medical residents, which it needed to be in order not to miss the very because of the low income rates. Our Georgian few expected cases in a sample of just over 1000.
study is an example of this. Another is a These two factors run counter to each other. On the population-based study in Ethiopia, carried out by one hand, ideally the entire sample should be inter- trained lay persons, which estimated a prevalence viewed face-to-face by a headache specialist; on the other, the size of the sample needs to be large. The In order to make a study feasible, the German ideal is therefore heavily demanding of manpower Headache Consortium first screened a population of 3336 people by self-administered questionnaire, An alternative is a two-phase approach with a seeking possible cases of CH. The 182 people sus- screening procedure and a subsequent validating pected to have CH were then examined in person by neurological consultation for all screen-positive headache specialists. The diagnosis was confirmed cases. The screening can be by self-administered in only four cases, giving a 1-year prevalence of CH questionnaire, telephone interview or face-to-face of 119/100 000 (95% CI 3, 238/100 000) (9). This interviews by trained lay or medical personnel, finding again is in keeping with ours and not with which are probably increasingly methodologically the Vågå study. Another German study invited a sound in that order. We used the last. Even so, this random sample of 1312 inhabitants of the City of approach is less robust than the ideal: it saves Dortmund to a first screening interview, among resources but, if a single case is overlooked in a whom 33 reported attacks of unilateral headache sample of manageable size, the estimated preva- lasting < 6 h with at least one of the typical auto- nomic symptoms of CH. All of these were then A second alternative offers itself in countries examined by a headache expert. Two cases of CH where general practitioners’ lists cover almost the were identified, a prevalence rate of 150 per 100 000 entire population: data from general practices may (upper 95% confidence limit 360 per 100 000) (10).
be representative. Against this alternative is the low An Italian group screened a sample of 10 071 accuracy and limited recorded detail of headache patients registered in the lists of seven Parma-based diagnoses made by general practitioners. The final general practitioners using a self-administered ques- possibility is to investigate a population sample tionnaire. Of the 111 suspected cases of CH, the with a higher than average risk (e.g. young men) diagnosis was confirmed in 21, yielding a preva- but, obviously, great caution will be needed in lence estimate of 279 per 100 000 (95% CI 173, 427) drawing conclusions about the wider population.
(11). A study in San Marino reviewed the past 15 Fewer than 10 studies of the prevalence of CH are years’ medical records of neurological, ophthalmo- reported in the world literature, and these have logical and ear, nose and throat practices in the used several approaches. The most exacting meth- whole republic (12). Suspected cases of CH were odology was employed in the Vågå study in examined by neurologists, with the diagnosis con- Norway, in which the lead author, with great com- firmed in 15 people. The prevalence rate was esti- mitment, personally interviewed and examined the mated at 69 per 100 000, but it was not possible in entire study population in order to determine the this study to know precisely what the denominator prevalence of all rare unilateral headaches (6). This was. A recent Swedish twin register study has resulted in the highest ever reported prevalence provided an estimate of 151 per 100 000 (13).
Blackwell Publishing Ltd Cephalalgia, 2009 These studies to some extent demonstrate the of Headache Disorders, 2nd edition. Cephalalgia 2004; 24 advantages and challenges of different approaches, which were obviously chosen by authors in con- 2 Statistic yearbook of Georgia. Tbilisi, GA: National sideration of prevailing circumstances. There are 3 Katsarava Z, Kukava M, Mirvelashvili E, Tavadze A, several important general issues. Studies must be Dzagnidze A, Djibuti M, Steiner TJ. A pilot methodologi- population-based in order to achieve results repre- cal validation study for a population-based survey of the sentative of the population. Whereas it is optimal to prevalences of migraine, tension-type headache and conduct expert examinations of the entire sample, chronic daily headache in the country of Georgia. J Head- this is not possible in most situations and, when attempted, is likely to result in a reduced sample 4 Kukava M, Dzagnidze A, Janelidze M, Mirvelashvili E, Djibuti M, Fritsche G et al. Validation of a Georgian size. We believe that self-administered question- language headache questionnaire in a population-based naires (even if well validated) cannot assure diag- sample. J Headache Pain 2007; 8:321–4.
nostic accuracy for CH. In the German Headache 5 Bortz J. Statistik für Human- und Sozialwissenschaftler, 6.
Consortium study, the diagnosis of CH was found Aufl. ed. Heidelberg: Springer Verlag 2005.
to be correct in only four of 182 putative cases (9), 6 Sjaastad O, Bakketeig LS. Cluster headache prevalence.
a specificity of 2.2%. In our Georgian study, even Vaga study of headache epidemiology. Cephalalgia 2003; trained medical residents misdiagnosed 31 cases, 7 Rasmussen BK, Jensen R, Schroll M, Olesen J. Epidemi- among which, and especially striking, were eight ology of headache in a general population—a prevalence people with pure TTH. We therefore conclude that study. J Clin Epidemiol 1991; 44:1147–57.
suspected cases of CH must be confirmed by clini- 8 Tekle Haimanot R, Seraw B, Forsgren L, Ekbom K, cal review by doctors with expertise in headache Ekstedt J. Migraine, chronic tension-type headache, and cluster headache in an Ethiopian rural community.
Finally, we stress the strengths and challenges of our study. The methodology assured high quality of 9 Katsarava Z, Obermann M, Yoon MS, Dommes P, Kuznetsova J, Weimar C, Diener HC. Prevalence data collection. We studied a capital city as well as a of cluster headache in a population-based sample in rural area of the country. We achieved very high Germany. Cephalalgia 2007; 27:1014–19.
response rates of > 92%. All subjects with possi- 10 Evers S, Fischera M, May A, Berger K. Prevalence of ble CH were reviewed personally by headache- cluster headache in Germany: results of the epidemio- experienced neurologists. The main limitation of the logical DMKG study. J Neurol Neurosurg Psychiatry study was that the population sample was too small to allow an accurate estimate of a rare disorder like 11 Torelli P, Beghi E, Manzoni GC. Cluster headache preva- lence in the Italian general population. Neurology 2005; CH. Ideally, it should have been three to four times as large to have sufficient statistical power.
12 D’Alessandro R, Gamberini G, Benassi G, Morganti G, Cortelli P, Lugaresi E. Cluster headache in the Republicof San Marino. Cephalalgia 1986; 6:159–62.
References
13 Ekbom K, Svensson DA, Pedersen NL, Waldenlind E.
Lifetime prevalence and concordance risk of cluster head- 1 Headache Classification Subcommittee of the Interna- ache in the Swedish twin population. Neurology 2006; tional Headache Society. The International Classification Blackwell Publishing Ltd Cephalalgia, 2009

Source: http://www.pain.org.ge/upload/pdf/epidemiology_Georgia_cluster_headache.pdf