Headache patients in the emergency department of a Greek tertiary care hospital

In our survey, headache patients account for 1.3% of all adult patients presenting to the ED during the study period. This percentage is concordant to pertinent data from the comparable studies that report a headache incidence of 0.4–4.5% among ED patients [6‐13]. Headache was the leading complain in 15.5% of all patients assessed by the ED neurologist. This percentage can also be compared with the results (12.7%) of an Italian study in 2005 [7]. ‘Headache’ was the second most frequent reason—after the symptom of ‘dizziness-vertigo’ (17%)—for which patients were examined by a neurologist in the ED of our hospital.

In contrast to other studies on headache incidence among ED patients, the most frequent diagnosis for primary headaches was the TTH with 33.6% and not migraine (with 15.4%), which was diagnosed less often than secondary headache and headache NOS. In all other reports, the most frequent ED diagnosis for primary headache was migraine [6, 14‐18]. In a nationwide study covering the period between 1992 and 2001 on headache management in US EDs, Goldstein et al. [8] reported that the diagnosis of migraine was made in 63.5% of patients who attended the ED suffering from headache. This percentage is four times higher than in our sample. But it has to be taken into account that in other studies, especially in the USA, the majority of the ED headache patients were seen by the ED physician and not by a board certified neurologist [6‐9, 11, 13, 14, 17‐19]. This difference in the management of headache patients does also lead to differences in treatment practice: In our hospital’s ED, opioids are rarely used for acute headache treatment, in contrast with US studies [13, 19]. It also may be argued that migraine is often under-diagnosed and under-treated in the ED [17, 18], or that the ICHD-2 criteria for migraine are underused in the ED, with the result that many patients leave without a clear discharge diagnosis and medications [13]. However, we are convinced that in our study the main reason for the discrepancy in the incidence of migraine diagnosis is the fact that almost one-third of our patients had been suffering from headache for more than a week before attendance. This time interval rules out uncomplicated migraine attacks that may be observed more often in patient samples with smaller referral intervals. Our explanation is corroborated by the fact that headache patients in the Greek NHS usually have to wait for months for an appointment in a specialized Headache Outpatient Department and often try to speed up their management presenting to the busy ED rather than waiting for a regular appointment at a Headache Centre. This goes in line with the finding that 33.3% of the patients suffering from headache for more than a week used a very urgent mean of transport (EMS-ambulance), in order to present to the ED. Thus, recurrent headaches, mainly (chronic or episodic) TTH, make up the great part of the neurologists workload at the ED, and not patients suffering either from ‘first or worst’ headache or migraine attacks. It is of note that a study from another Mediterranean country (Italy) shows similar conclusion, suggesting that Italian EDs are used instead of a visit to the general practitioner [16]. The assumption that chronic headache patients overuse Greek ED services is confirmed by the fact that 40% of headache patients who attended the ED were neither given any acute analgesic medication nor IV fluids by the on-call neurologist, a percentage lower than reported from a recent US study [13] and from the Italian study [16]. Most of our patients received a prescription to start medical treatment with non-steroidal anti-inflammatory analgesics drugs (never narcotics) home or were informed to consult a specialized Headache Centre.

In our study, the ratio of female–male headache patients was approximately 2:1. This goes in line with other studies reporting a female–male sex ratio between 2:1 [6, 8, 13] and 3:1 [9, 14, 15]. Even though the overall VAS pain score was high (almost 50% of patients rated their headache with a score higher or equal to 7), female patients rated their headache intensity with a higher score than male. Therefore, they received more often analgesic treatment from the on-call neurologist. Thus, female headache patients were ‘more frequent and younger’ with ‘more headache’ resulting in ‘more treatment’.

As most primary headaches are observed with equal frequency among the genders, it does not astonish that, with the exception of cluster headache (90% male patients) there was no statistically significant sex difference in the incidence of primary headaches in our sample.

50.2% of patients were diagnosed to suffer from primary headache and the rest from secondary (22.1%) or headache NOS. The high percentage (27.7%) of patients suffering from headache NOS in our study may be considered high, as in other reports, the incidence of primary headaches in the EDs reaches 64 [19], 80 [11] and 81.2% [20] and headache NOS was diagnosed in 14.9 [14] or 26% [6] of all headache patients only. However, other studies report a higher incidence of headache NOS, ranges from 42% [15, 16] to 59% [17]. Even in well structured reports from the US, a relevant portion of the primary headaches could not be classified, and therefore, 20% [19] to 36% [13] of the acute patients were not given a diagnosis in the ED. Comparable data were reported from an Italian study [7] with more than one-third of headache patients classified as headache NOS (Table 3). In our study, this percentage of headache NOS diagnosis increased to 33.3% among patients who were transferred by an EMS-ambulance. Given the high number of patients (n = 5,491) referred to the ED neurologist for assessment during the study period, the ED neurologist’s first concern is to recognize headache secondary to an acute neurological disease, and his second concern should be to give specific headache treatment (e.g. for migraine). The exact diagnosis of primary headache subtypes is difficult to be made in ED settings and is a typical task for a Headache Center. On the ED level, it may be sufficient to distinguish primary from secondary headache, especially benign from possibly life-threatening causes of acute headache, and to give an acute pharmacological treatment. Furthermore, Bø et al. [21] demonstrated that this distinction must not be based on clinical features alone, and that all unclear cases have to be admitted to a neurological ward for further evaluation, including neuroimaging, where appropriate.

From our data, it is obvious that headache in the ED is a symptom that concerns mainly patients younger than 50 years. This finding is concordant with other respective reports [6, 8, 12, 14]. It is of note that patients younger than 50 years rated their headaches with statistically significant higher VAS scores than older patients. This higher subjective pain intensity is the reason why patients under 50 years of age more easily present to attend the ED. In our study, the mean age of headache patients was approximately 40 years. A lower mean age is reported from other studies that—in contrast to our survey—also included children. Apparently, headache diagnosis was more easy in patients younger than 50 years, as headache NOS was diagnosed in one quarter of these patients, in contrast to older patients, where headache NOS was diagnosed in 1/3 of the patients. Ninety percentages of cluster headaches patients were younger than 50 years, and migraine was also more often diagnosed among these patients. In contrast to the nationwide US survey that showed that the patients older than 50 years had a fourfold risk to have a headache secondary to an underlying disease [8], in our study this risk was not statistically significant from the risk in older patients, although our admission rate for headache patients over 50 years of age (19.3%) was higher than reported from the US study.

An overall 17% of headache patients presenting to the ED were admitted to a ward for further examinations, a value comparable to other studies [12, 16]. Of the 60 patients who were transferred by an ambulance, only 30% were admitted to a ward, even though the diagnosis of headache NOS was more often and these patients rated their headache pain intensity with a higher score compared to patients who used private transportation means. In a study by Nemer et al. 12.2% of headache patients examined in the ED had been transported by an ambulance (a percentage higher than the 7.1% observed in our study). These patients had a 18.5-fold risk to have a serious reason for their headache (such as meningitis, intracranial haemorrhage or tumour of the central nervous system), compared to ED headache patients that had used private transportation means [22]. We are convinced that in our study, chronic headache patients show a tendency to misuse EMS ambulances.