prevalence - epidemiology - cluster headache - family health strategy
prevalência - epidemiologia - cefaleia histamínica - estratégia saúde da família
In the group of primary headaches, there is an important category of pain with autonomic
trigeminal involvement known as cluster headache (CH), which has the clinical distinction
of being one of the most painful headaches[1],[2]. Despite presenting with characteristic signs and symptoms, it is necessary to make
a differential diagnosis, particularly between migraine, chronic paroxysmal hemicrania
and trigeminal neuralgia, to enable the diagnosis and treatment of CH with confidence.
Cluster headache is a clinical entity with diagnostic criteria that are highly specific
and sensitive[3], like those of other primary headaches. Among the clinical particularities of CH,
we highlight a unilateral headache of severe intensity, occurring in short attacks
(30-180 minutes), and concomitant autonomic symptoms [e.g. tearing (84–91%), eyelid
ptosis (57–74%), nasal congestion (48–72%) and/or rhinorrhea (43–72%)], in the patients[4]. Most patients have a circannual and circadian periodicity[2],[4],[5]. Attacks tend to occur between once every other day and eight times a day, generally
lasting from 4–12 weeks, followed by a pain-free period of one to two years, with
reports of remission of up to 20 years. According to one prospective clinical study,
the mean maximum duration of a crisis was 159 minutes, while the mean duration was
72 minutes among their patients[1].
A review by Almeida[1]of the prevalence of CH, refers to studies in the following populations: 0.1% in
Denmark; 0.08% in women and 0.4% in men in USA; 0.09% in Sweden and 0.07% in the Republic
of San Marino.
According to a meta-analysis carried out by Fischera and coworkers[6]in 2004, the one-year prevalence was 0.54%. However, the authors emphasized the difficulty
in establishing the prevalence, due to the differences between studies when it came
to methodology, population and diagnostic criteria, which hindered comparisons between
the data[6].
The objective of this study was to determine the prevalence of CH in the population
of Barbacena, a medium-size city in the State of Minas Gerais, Brazil, and the effectiveness
of the Brazilian health system and its employees in identifying CH patients using
a questionnaire.
METHOD
Study outline
This was an observational cross-sectional study, which occurred from June 2015 to
June 2016, that sought to determine the lifetime prevalence of CH in the population
of Barbacena, Brazil. The age, sex, marital status, education, income, and profession
of CH patients were also evaluated. All patients meeting the two criteria in the questionnaire—strictly
unilateral headache with tearing—were also evaluated by a neurologist specializing
in headache.
Population
The population in the city of Barbacena, Brazil is 126,284, and covers an area of
759,186 km[2]
[7]. The study was carried out in this city for its convenience for most of the researchers
who live there, and for having demographic characteristics that are similar to those
found in other cities of similar size in Brazil.
The Family Health Strategy (FHS) Program covers 84,610 inhabitants, which is 67% of
the population of Barbacena. In this study, 36,145 of the inhabitants were included.
In order to identify patients with CH in the study population, we carried out a screening,
followed by the application of a questionnaire by 181 health agents distributed among
the 28 health posts belonging to the FHS network.
In Brazil, the FHS acts as part of a governmental strategy to reorganize basic health
care. It is based on the precepts of Sistema Único de Saúde - SUS (Brazil’s public health system), the purpose of which is to improve the effectiveness
and impact of basic health care on the health of people and communities. The FHS multidisciplinary
team is made up of a number of professionals, including a general practitioner, also
called a family health specialist or family and community physician; a general nurse,
also called a family health specialist; a nursing assistant, and community health
agents who were responsible for the application of the questionnaire.
Inclusion and exclusion criteria for the study
The following were adopted as inclusion criteria:
-
Patients who agreed to participate in the research;
-
Patients aged 18 years or older; and
-
Patients reporting some type of headache with autonomic symptoms.
The following were adopted as exclusion criteria:
Research tools
For the initial population screening and the control of the number of people approached
for the project, a list was built containing the following information: the total
number of selected individuals during the initial screening, the number of people
with headache, and the number of patients with tearing. This list took tearing into
account as it is the most prevalent sign associated with CH[4].
Individuals who reported headache and tearing answered an informal-language questionnaire
containing objective questions (yes or no) that were based on the International Classification
of Headache Disorders (ICHD) criteria, to aid the health agents in the identification
of possible CH patients[8]. Those patients who did not present with tearing, but presented with other clinical
characteristics of CH, were also evaluated by a neurologist.
In order to formulate the aforementioned questionnaire and establish a diagnosis of
CH, we used the criteria defined by the ICHD[9]:
-
At least five attacks fulfilling the criteria from B to D.
-
Severe or very severe, unilateral, supraorbital and/or temporal pain lasting 15–180
minutes (when untreated).
-
Either or both of the following:
3.1. At least one of the following ipsilateral signs or symptoms:
-
Conjunctival hyperemia and/or tearing;
-
Nasal congestion or rhinorrhea;
-
Eyelid edema;
-
Forehead and facial sweating;
-
Forehead and facial flushing;
-
Sensation of fullness in the ear;
-
Miosis and/or ptosis.
3.2. Sense of agitation or restlessness.
-
The attacks have a frequency of between one every other day and eight per day, for
more than half of the time during the active stage.
-
Not better explained by another ICHD-3 beta diagnosis.
Project execution
At first, we established contact with the city’s Department of Health by presenting
a cover letter for the project and obtaining authorization for the study. After approval,
we carried out preliminary research with a randomly-selected unit of the Family Health
Strategy program to assess the feasibility of the study.
In the first stage of the project, the health agents received basic information on
CH. This training was carried out using multimedia resources, including a presentation
in PowerPoint format, and distribution of a book containing the specific characteristics
of the headache in question. Following training, the agents were given a list containing
the identification of the FHS, name of the agent, area of performance, number of homes
visited by the agent, and number of residents, the presence or absence of headache,
and presence or absence of tearing.
In the second stage of the project, the questionnaires were evaluated, selected and
assessed by the neurologist for the purpose of ruling out possible false-positive
diagnoses, based on the clinical criteria established by the ICHD[9] ([Figure]). The research project was approved by the Ethics in Research Committee (nº 1.102.470).
Figure Flowchart.
RESULTS
Selection of CH patients
In the screening carried out by the professionals of the Family Health Strategy, a
total of 620 questionnaires were completed. The researchers contacted the interviewees
by telephone and subsequently made a brief assessment of the interviewees at the hospital,
according to the clinical criteria established by the ICHD. As a result, only 39 individuals
presented with all the objective clinical conditions of CH patients, and were referred
to a neurologist. After the specialist’s evaluation, only 15 patients received a diagnosis
of CH, representing a prevalence of 0.0414%; or 41.4 cases in 100,000 inhabitants.
The remaining 24 patients were diagnosed with other types of primary headache.
Characteristics of the CH patients
Regarding the 15 individuals diagnosed with CH, there was a predominance of males,
87% (95%CI; 69.5–104), in relation to females, 13% (95%CI; 0–30.5). Regarding age,
most patients were in the 35–45 year old group, followed by the 46-55 year old (13%)
and 56-65 year old (33%) age groups. The weighted average and median of the ages were
48.8 and 42 years, respectively. When we analyzed the marital status of these patients,
there was also a predominance of married individuals (67%), followed by those living
together (27%) and those who were divorced (7%). Eight of the 15 patients (53%) with
CH did not know about the diagnosis of CH.
As regards education, most of the patients had incomplete elementary schooling (40%),
followed by complete elementary education (20%), high school (27%) and college (13%).
Regarding income, the average was 2.2 minimum wages.
DISCUSSION
The composition of the study population was an important feature and one of the difficulties
in performing the research, as epidemiologic studies carried out in a primary health
care center do not always reflect the reality of the prevalence and the impact of
CH in the community. However, considering the quantitative aspect of the composition
of this sample, it is possible to regard it as representative.
Table 1
Sociodemographic characteristics of patients diagnosed with cluster headache.
|
Variables
|
n
|
%
|
|
Gender
|
|
Female
|
2
|
13.33
|
|
Male
|
13
|
86.67
|
|
Age (age group)
|
|
35–45
|
8
|
53.34
|
|
46–55
|
2
|
13.33
|
|
56–65
|
5
|
33.33
|
|
Marital status
|
|
Married
|
10
|
66.67
|
|
Living together
|
4
|
26.67
|
|
Divorced
|
1
|
6.66
|
|
Education
|
|
Incomplete elementary education
|
6
|
40.00
|
|
Complete elementary education
|
3
|
20.00
|
|
High school
|
4
|
26.67
|
|
College
|
2
|
13.33
|
|
Income (minimum wage)
|
|
No income
|
1
|
6.67
|
|
One
|
3
|
20.00
|
|
Two
|
7
|
46.67
|
|
Three
|
2
|
13.33
|
|
Five
|
2
|
13.33
|
The data collection method was by face-to-face structured interviews of adults of
both sexes, a method with important advantages when compared with telephone or electronic
interviews, or the self-completion questionnaire. In addition, the clinical interview
conducted by a headache expert, which was considered as the gold standard for the
performance of this study, enabled correct diagnoses, considering the subtle clinical
differences among primary headaches[10].
Despite its limitations, this study is the first epidemiologic CH study with classical
and accepted methodology carried out in this country.
The lifetime prevalence of CH was 0.0414%, or 41.4 cases in 100,000 inhabitants in
the city of Barbacena, Brazil. The study closest to the prevalence found in Barbacena
was carried out in Ethiopia in 1993, in a sample of 15,500 individuals, with a prevalence
of 32 cases/100,000 inhabitants[11]. The following studies found a greater prevalence of CH in their populations: in
Sweden between 1935–1958, with a sample of 31,750 showing a lifetime prevalence of
144 cases/100,000 inhabitants[12], and in 1975–1976 with a sample of 9,803 males aged 18 or older with a lifetime
prevalence of 92/100,000[13]; in the Republic of San Marino in 1985, with a sample of 21,792 people and a prevalence
of 69 cases/100,000 inhabitants[14], and in 1999, with a sample of 26,628 people and lifetime prevalence of 70/100,000
inhabitants[15]; in Porto (Portugal) in 1992, with a sample of 2,008 people and lifetime prevalence
of 100 cases/100,000 inhabitants[16]; in Germany in 2004, in Essen with a sample of 3,336 people and prevalence of 119/100,000[17], and in 2005, in Dortmund with a sample of 1,312 people and prevalence of 150/100,000[18]; in Glostrup (Denmark) in 1989, with a sample of 740 people and lifetime prevalence
of 135/100,000[19]; in Parma (Italy) in 2002, with a sample of 7,522 people and lifetime prevalence
of 279 cases/100,000 inhabitants[20]; finally, in Vaga (Norway) in 1995, with a sample of 1,838 people and lifetime prevalence
of 381 cases/100,000 inhabitants[21].
The data in [Table 2] allows for a better comparison of the prevalence found in these studies.
Table 2
List of references, countries, year of publication, population and prevalence found
in the studies.
|
Reference #
|
Country
|
Year
|
Population
|
Prevalence
|
|
11
|
Ethiopia
|
1992-93
|
15,5
|
32/100,000
|
|
Present study
|
Brazil
|
2016
|
36,145
|
41.4 /100,000
|
|
14
|
Rep. of San Marino
|
1985
|
21,792
|
69/100,000
|
|
15
|
Rep. of San Marino
|
1999
|
26,628
|
70/100,000
|
|
13
|
Sweden
|
1975–76
|
9,803
|
92/100,000
|
|
16
|
Portugal
|
1992
|
2,008
|
100/100,000
|
|
17
|
Germany
|
2004
|
3,336
|
119/100,000
|
|
19
|
Denmark
|
1989
|
740
|
135/100,000
|
|
12
|
Sweden
|
1935–58
|
31,75
|
144/100,000
|
|
18
|
Germany
|
2005
|
1,312
|
150/100,000
|
|
20
|
Italy
|
2002–03
|
7,522
|
279/100,000
|
|
21
|
Norway
|
1995
|
1,838
|
381/100,000
|
Rep.: Republic.
Upon assessing the sample of our study regarding sex, we found that 87% were male,
with a male/female ratio of 13:2. The data confirms the literature, according to which
the greater prevalence of CH was in males[5],[6],[22].
As a national reference, a study carried out in 2015[23], using an electronic questionnaire (n = 658) completed by the patients with probable
CH, confirmed the diagnosis of 324 cases in Brazil and established a clinical and
sociodemographic profile of CH patients. The study showed a greater occurrence in
men (283, 73.1%) than in women (104, 26.9%) giving a ratio of 2.7:1; with an average
age of 39.3 years, which reinforces the data found in the present study. Note that
in the 2015 study[23], the entry criterion was “individuals who have accessed a certain website”.
This study had a larger sample than the aforementioned studies. However, due to the
clear difference between the proposed methodologies, the study population and the
diagnostic criteria applied in the studies, there was difficulty in comparing the
results of the prevalence of CH, reaffirming the conclusion by Fischera[6] that establishing prevalence is difficult.
It takes two years, on average, for a patient with CH to receive a correct diagnosis.
In this article, we are evaluated the effectiveness of the Brazilian health system
and its employees in screening the population to identify patients with CH. A substantial
number of the CH patients identified in the present study (> 50%) did not know about
their diagnosis of CH and, therefore, had not received adequate treatment. This is
a fact of utmost importance, indicating that the public health system, using a simple
questionnaire, which identified individuals with strictly unilateral headache and
tearing, was able to give patients with “probable” CH access to a headache specialist,
as they may not have utilized health care services as much as other chronic afflictions
with higher morbidity and mortality rates[24].
Quite possibly, for various reasons, some individuals with CH may not have been identified
in the present study. For example, a small proportion of CH patients do not report
tearing, or present with agitation without autonomic symptomatology. However, this
appears to be infrequent.
In conclusion, this study helped define the prevalence of CH in Brazil, which, although
being considered a rare disorder, has a big impact on the quality of life of individuals,
justifying the importance of the study. The prevalence of CH found in the sample of
36,145 inhabitants in the city of Barbacena/MG was 0.04%, or 41.4 per 100,000 inhabitants.
