Abstract
Primary stabbing headache (PSH) is a short-lasting but troublesome headache disorder which has been known for several decades. We surveyed and registered consecutive patients with PSH in a headache clinic in Taiwan. A total of 80 patients (24 M/56 F, 53.2 ± 16.2 years) were enrolled in our study. Migraine was reported in 20 (25%) patients and was less common in those with PSH onset at >50 years than those with onset at <50 years (14% vs. 38%, P = 0.02). The headache was unilateral in 59% of the patients and always in a fixed area in 36%. The head pain frequently involved extratrigeminal regions (70%) and in 30 patients (38%) was accompanied by jolts, i.e. head or body movements. Indomethacin was effective (74%) in patients who received it. Our study showed primary stabbing headache was a common and easily treated headache disorder in headache clinic. However, 70% of our patients could not fulfil criterion A ‘exclusively or predominantly in the distribution of the first division of the trigeminal nerve’ and 15% could not fulfil criterion C ‘no accompanying symptoms’ of the International Classification of Headache Disorders-II criteria proposed for PSH.
Keywords
Introduction
Primary stabbing headache (PSH), first described in 1964 as ‘ophthalmodynia periodica’ (1), consists of brief, sharp, jabbing pains. PSH has also been known by many names, such as ‘ice pick-like pains’, ‘sharp short-lived head pains’, ‘needle-in-the-eye syndrome’ and ‘jabs and jolts syndrome’ (2–5). The International Classification of Headache Disorders, 2nd edition (ICHD-II) has proposed diagnostic criteria for PSH (code 4.1) (Table 1) (6), which require that head pain occurs exclusively or predominantly in the distribution of the first division of the trigeminal nerve. However, some studies have shown that PSH can occur outside the trigeminal region in up to 22–55% patients (2–4, 7–9), and even in the extracephalic region (10, 11). Although most attacks of PSH last ≤3 s (5), its duration has been reported as from a fraction of second to 120 s (11, 12). It is therefore important to field test the ICHD-II criteria.
Diagnostic criteria of primary stabbing headache (code 4.1) in the International Classification of Headache Disorders, 2nd edition and the percentages of study patients who fulfilled each criterion
PSH is generally thought to be a primary headache syndrome, onset at young age, predominantly in women, commonly associated with migraine or cluster headache and responsive to indomethacin (13–15). A recent population-based study has shown that PSH is common in the general population (7, 11). However, most people have attacks of low frequency (11), so they seldom visit the doctor. Patients who visit the headache clinic might be different from those surveyed in the community, but there are few large-scale clinic-based studies of PSH. Moreover, the bulk of current knowledge and understanding of PSH comes from western-based studies. The clinical characteristics of PSH in other regions of the world are ill understood.
To explore the clinical picture of PSH in Taiwanese and field test the ICHD-II criteria, we surveyed and registered consecutive patients with PSH in a headache clinic in Taiwan.
Subjects and methods
We consecutively recruited non-veterans patients with episodic short-duration headache who visited the headache clinic of Taipei-Veterans General Hospital (VGH). Patients were recruited who complained ultrashort headache attacks (stabs). The stabs might be single or of multifocus localization and the frequency could be single or a series of attacks. Contrary to the ICHD-II criteria (6), the location of pain was not necessarily confined to the first division of the trigeminal nerve. Possible differential diagnoses were also excluded, such as trigeminal neuralgia, occipital neuralgia, cervicogenic headache and other short-lasting headaches (2). Other concomitant headache types, such as migraine, cluster headache and tension-type headache were also diagnosed in each patient. This study protocol was approved by the Institution Review Board of VGH. The reason for excluding veterans was because they differed in basic demographics from the other patients (see below).
VGH is a 2198-bed medical centre serving both veterans and non-veteran citizens. Most of the patients come from northern Taiwan. In Taiwan, the National Health Insurance programme was launched in 1995. It is a government-run, single-payer system. More than 96% of Taiwan's population was enrolled and >90% of Taiwan's healthcare providers were included. Unlike the US, British or Canadian systems, the referral system was not adopted in Taiwan and patients have the complete freedom to choose hospitals and physicians, i.e. they can seek care at tertiary care institutions, regardless of the nature or severity of their illness. Therefore, almost all patients in our headache clinic were self-referred (16). The headache clinic of Taipei-VGH has been operating since 1997.
A physician (K-H.K.) interviewed all patients based on a structured questionnaire designed for PSH. Information obtained included demographics and headache profiles: age at onset, headache duration, intensity, frequency, location of the attacks, quality of the pain, precipitating factors and accompanying symptoms.
The locations of stabs was determined by the cutaneous projections of the cranial bones (frontal, temporal, parietal and occipital bones) and classified into anterior and posterior region by an imaginary line from the vertex, the highest point on the skull in the sagittal plane (3). Trigeminal areas included orbital, frontal, anterior temporal, anterior parietal and vertex areas. Extratrigeminal areas included auricular, posterior parietal, occipital and nuchal areas.
In our headache clinic, headache specialists usually treat patients with PSH empirically with either indomethacin (25 mg bid or tid) or rofecoxib (50 mg per day) based on the specialists' judgement. Of note, rofecoxib was not withdrawn from the market during the study. The treatment response was recorded after 2 weeks of treatment as: complete remission, partial response (>50% decrease of jabs) and poor response.
Statistics
SPSS Version 11.0 for Windows (SPSS Inc., Chicago, IL, USA) was used for statistical analysis. For continuous measures, Student's t-test was used to test the difference between two groups. For categorical data, the χ2 test was used. All calculated P-values were two-tailed and statistical significance was defined as P < 0.05.
Results
Subjects
During the study period, 872 patients with headache visited the headache clinic and 115 (13%) were diagnosed with PSH. After excluding 35 veterans patients (all male, mean age 74.7 ± 7.3 years, range 60–90), 80 patients (70%) (56 F/24 M, female:male ratio 2.3, mean age 53.2 ± 16.2 years, range 13–83) participated. The mean age of onset was 47.9 ± 17.3 years (range 13–83) and did not differ between men and women. Sex and age of onset distribution are shown in Fig. 1. In women, the highest age of onset occurred in the range of 51–60, followed by 61–70. They accounted for 50% of female patients with PSH. In men there was no wide variation of age of onset.
The distribution of ages of onset in patients with primary stabbing headache between males and females. ▪, Male; □, female.
Severe pain (N = 22, 28%) and fear of stroke (N = 14, 18%) or malignancy (N = 14, 18%) were the three main reasons for physician consultation. Twenty patients (25%) had a past history of migraine headache and patients with PSH onset at <50 years old had a higher proportion of migraine headache compared with those with older age of onset (38% vs. 14%, P = 0.02). None of them had cluster headache. Four patients (5%) had had computed tomography of brain to rule out a structural lesion.
Headache profiles
Headache quality, duration, frequency and intensity
The interval between the onset of the latest PSH and the first visit our clinic ranged widely from 1 day to 20 years (median 14 days). Forty-six patients (56%) had had previous experience of such a headache. The pain was stabbing in nature in all patients. The duration of a single jab was <3 s in 64 patients (80%), 3–5 s in 13 patients (16%) and 6–60 s in three patients (4%). Mild intensity of headache was reported in 18 (23%), moderate in 33 (41%) and severe in 29 (36%) patients. Twenty-three (29%) patients had experienced being awakened by this headache. Sixty-seven (83%) had more than one jab per day. Of these, 21 (26%) had <10 jabs/day, 44 (55%) had an average of 10–99 jabs/day and two (3%) had >100/day. Thirty patients (38%) had a pattern of sporadic single stabs, 24 (30%) a pattern of a series of stabs (volleys of jabs) and 26 (32%) had both patterns.
Stab location
The jabs were unilateral in 45 patients (56%), either on the right (N = 21) or on the left side (N = 24). The locations of jabs were in one fixed area in 29 patients (36%) and in the reamainder in more than one area. The pain was most frequently localized posteriorly, especially in the occipital and nuchal areas (N = 46, 58%), followed by parietal (N = 39, 49%) (Fig. 2). Based on the nerve distributions, only 24 patients (30%) had jabs located in the trigeminal nerve region.
The location of stabs in patients with primary stabbing headache.
All three of the patients with the longer-lasting jabs (6–60 s) localized their pain over the temporal region and only one of them was unilateral. None of these three patients reported any autonomic signs such as tearing and conjunctival injection during their attacks. Therefore, we could not assign a diagnosis of short-lasting unilateral neuralgiform headache with conjunctival injection and tearing.
Jolts and other accompanying phenomena
Jolts were sudden movements following the jabs and they were reported in 30 patients (38%). These movements included grimacing (n = 25, 31%) and shrugging (n = 20, 25%). Vocalization was presented in 14 patients (18%). An uncomfortable sensation (allodynia) while touching hair/skin was reported in 19 (24%) patients. Fifteen patients had accompanying symptoms, i.e. nausea or vomiting (n = 8, 10%) and photophobia/phonophobia (n = 6, 8%) during attacks. Ten patients (13%) had bodily jabs. Twelve (15%) complained dizziness during the attack. None of the patients had autonomic signs during attacks such as conjunctiva injection, tearing or rhinorrhoea.
Precipitating factors and triggers
Twenty-five patients (31%) reported precipitating factors of their jabs. The three most common factors were poor sleep (N = 18, 23%), weather change (N = 17, 21%) and common cold (N = 16, 20%). Fifteen patients (15%) reported that neck movements and 11 (14%) the Valsalva-like movements such as coughing, stool straining, exertion and sexual behaviour would trigger their PSH attacks. None of the patients reported cutaneous stimulation in the trigeminal region as a trigger for their headaches.
Treatment
Seventy-five patients received medication treatment (indomethacin 54, rofecoxib 21) and five patients received no treatment. The mean dosage of indomethacin was 59 ± 16 mg. Ten patients who received medication were lost to follow-up. Among the other 65 patients (46 indomethacin and 19 rofecoxib), 33 (51%) had complete remission and 14 (22%) had partial response. The remission rate (at least 50% improvement) did not differ between patients who received indomethacin or rofecoxib (74% vs. 68%, P = 0.2). Four patients discontinued indomethacin during the 2-week treatment period: one was allergic to indomethacin and three complained of gastrointestinal upset.
Field-testing of the ICHD-II criteria
All our patients fulfilled criteria A, C and E (Table 1), i.e. head pain occurring as a single stab or series of stabs, stabs lasting for up to a few seconds and recurring with irregular frequency from one to many per day, and not attributable to other disorders. Nevertheless, only 30% of our patients had head pain located exclusively or predominantly in the distribution of the first division of the trigeminal nerve, i.e. criterion B. Twelve patients (15%) had accompanying symptoms and could not fulfil criterion D.
Discussion
This is the first study to field test the ICHD-II criteria for PSH. The duration of the stabbing pain was quite brief in most of our patients, of whom only three reported the duration of their stabs as >5 s. This finding is in agreement with previous observations (5, 7, 17). ICHD-II does not define the duration of stabs strictly, so all of our patients fulfilled this criterion. Consistent with previous reports (2–4, 8, 9), our study found that the location of PSH was not confined to the anterior head. The percentage of subjects involving solely the anterior region ranged from 45% to 62% in previous studies (3, 4, 7, 18). In addition to the extratrigeminal location of the jabs, 15% of our patients had accompanying symptoms. Although some previous studies support the lack of accompanying symptoms in PSH patients (7, 19), a study of juvenile PSH also found that 8% of PSH patients had nausea and 15% had photophobia or phonophobia (18). Most of our patients failed to meet the diagnostic criteria of PSH in ICHD-II by these two criteria (6). We recommend further studies to validate the PSH diagnostic criteria of ICHD-II and modify them to reflect better the clinical features of patients.
The female predominance in this study was consistent with previous studies (4, 5, 7). The peak ages of onset in our study were slightly greater than those in a clinic-based study done in Spain (53 vs. 47 years old). Of note, the population-based epidemiology study of PSH done in Vågå, Norway (7, 11) showed the mean age of onset was 28 years; however, only 358 of 627 cases were able to recall a specific age of onset. In this community-based study, most of their stabs were mild, single jabs and only 5% of the PSH patients experienced more than five attacks per day. In contrast, most of our patients had frequent attacks of head pain and most of their stabs were of moderate or severe intensity. These differences might be caused by the different clinical settings and recall bias.
Although the prevalence of migraine among our PSH patients (25%) was lower than that in the Spanish study (37%) (5), both studies support the fact that PSH occurs more often in migraineurs. We also found that the prevalence of migraine was higher among PSH patients with a younger age of onset. The peak age of onset in our study was slightly greater than that in the Spanish study. Therefore, these differences might result in part from different age distributions of the study samples.
Although the mechanism of PSH is unknown, the irritations of the peripheral branches of the trigeminal or other nerves or deficits in central pain control mechanisms have been proposed (3). We found predominant involvement of the occipital and nuchal regions, which are both innervated by the upper cervical nerves (C2-4). The spinal trigeminal nucleus also extends to the upper cervical region. In addition, 24% of our patients also had allodynia and 13% had bodily jabs. Therefore, we support the contention that deficits of the central pain control mechanism might play a role in patients with PSH.
PSH is considered to be one of the indomethacin-responsive headaches (2, 5, 13). Some case reports have also shown treatment efficacy with celecoxib, nifedipine, melatonin and gabapentin (20–23). Although our study was not a formal clinical trial, it provides a direct comparison of indomethacin and rofecoxib. Although the response to rofecoxib was slightly less favourable, the response rate of indomethacin was similar to one previous study in which 75 mg/day indomethacin was used (5). Of note, our study might have underestimated the response rate because we did not increase the indomethacin dosage any higher than 75 mg. For some patients with a low frequency of attacks, the preventive efficacy of mediations might be uncertain because we recorded only the 2-week response.
In summary, PSH is not uncommon in the headache clinic because of its annoying symptoms. The associations between migraine and PSH are related to the age of onset of PSH. The extratrigeminal location of jabs and accompanying phenomena are common in patients with PSH. Modifications of the ICHD-II criteria for PHS are suggested to include these two common findings in these patients.