Abstract
Keywords
Background
Primary headache associated with sexual activity (primary HSA) is described as at least two episodes of headache precipitated by sexual activity only and the pain in the head increases in intensity with increasing sexual arousal, and/or it has abrupt explosive intensity just before or with orgasm, which lasts from one minute to 24 hours with severe intensity and/or up to 72 hours with mild intensity in the absence of any intracranial disorders based on the third edition of the International Classification of Headache Disorders (ICHD-3) criteria (1). Primary thunderclap headache is described as severe head pain with abrupt onset reaching maximal intensity in less than one minute and lasting for at least five minutes, after all organic causes have been demonstrably excluded (1). In a large systematic review in 2014 including many previously published case reports, case series and cohort studies, 459 patients who presented with sudden and severe headache were diagnosed with idiopathic thunderclap headache, among which the first two etiologies were primary thunderclap headache (n = 213, 46%) and primary HSA (n = 180, 39%) (2). Hence, the disease nature of these two diseases might be similar and deserve further investigation and illustration. This review aims to update the latest information on primary HSA and primary thunderclap headache.
Primary headache associated with sexual activity
Introduction
Primary HSA can be traced back to the era of classical Greece; at that time, Hippocrates described a headache that may result from ‘immoderate venery’ (3). A benign form of headache during sexual activity did not attract attention until the 1970s (3–5). The first systematically investigated report based on 21 patients with headache triggered by sexual activity conducted by Lance identified two different subtypes of headaches (3). One is described as a neck and jaw muscle contraction headache developing when sexual excitement is increasing; the other is described as an explosive headache occurring at the time of orgasm (3). Additionally, Paulson and Klawans reported a third type of headache following lumbar puncture, probably related to a dural tear forming during sexual intercourse and leading to low cerebrospinal fluid (CSF) pressure (5). By 1986, Johns summarized all 110 published cases with headache triggered by sexual activity, and the most common type among these three types is the explosive type of headache (69%) that occurs at or shortly before orgasm (6). In sum, in the ICHD-1 (1988),
The third edition of the International Classification of Headache Disorders (ICHD-3) Criteria for
Epidemiology
The incidence of primary HSA in the general population is difficult to determine and is probably underdiagnosed and underreported since it is usually embarrassing for people with headache triggered by sexual activity to seek medical care. The lifetime prevalence of headache triggered by sexual activity is estimated to be 1–1.6% in two population-based studies (11,12). There is a male preponderance, with male to female ratios of approximately 2:1 to 4:1, and it typically first occurs in their mid-thirties to mid-forties (10,13–15). Intercourse (83–95%) is the most frequently reported sexual act causing headache, followed by masturbation (14–35%) (3,10,16). In a German cohort, some patients (8.4%) had headache under a specific sexual position of intercourse, including 3.9% in the kneeling position, 3.9% in the supine position and 2.0% in fellatio (10).
Clinical features
Primary HSA typically presents a dull, progressive headache, or an abrupt exploding headache, or a combination of the above two headache types and the most common manifestation of primary HSA is probably an acute severe headache during orgasm (3,10,15,17). Primary HSA is predominantly bilateral (61–80%), occipital (58–80%), and with explosive (61–80%) and throbbing (50–90%) sensation (1,10,13,15). The headache lasts several minutes to 24 hours in most patients (13,15), and may be associated with nausea, vomiting, phonophobia, or photophobia, but without autonomic or vegetative symptoms (1,10,13,15). Some patients may have comorbidities of hypertension, migraine, or tension-type headache (10,13,15).
Diagnosis
When encountering patients with headache triggered by sexual activity, history taking and neurologic examinations should first focus on life-threatening conditions, especially subarachnoid hemorrhage (SAH). In a large systematic review, subjective neck pain (a positive likelihood ratio = 4.1) and objective neck stiffness (a positive likelihood ratio = 6.6) were most strongly associated with SAH (18). The Ottawa SAH Rule (If ≥1 of the following high-risk variables are present: age ≥40 years; neck pain or stiffness; witnessed loss of consciousness; onset during exertion; thunderclap headache; limited neck flexion on examination.) is proposed for patients with a normal neurologic examination for identifying SAH with a very high sensitivity (100%) (19). In addition, a CSF study is mandatory to exclude SAH, central nervous system infection or inflammation (20), but it is highly technique dependent, and the traumatic tap rate has been reported to be 15% to 20%, which makes test interpretation difficult (21). The American College of Emergency Physicians (ACEP) Board suggests that patients with SAH who have initial negative findings of noncontrast brain computed tomography (CT) are preferably diagnosed with CT angiography (CTA) over lumbar puncture (21).
If available, brain magnetic resonance imaging (MRI)/magnetic resonance angiography (MRA) should be performed on patients with HSA at first (13). The MRI protocol (22–24) preferably consists of T1-weighted imaging, T2-weighted imaging, and contrast-enhanced T1-weighted imaging sequences to exclude structural lesions; diffusion weighted imaging (DWI) and apparent diffusion coefficient (ADC) mapping to exclude ischemic stroke or posterior reversible encephalopathy syndrome; gradient echo (GRE) and/or susceptibility weighted imaging (SWI) to exclude subtle intracerebral hemorrhage or SAH; fluid-attenuated inversion recovery (FLAIR) to evaluate white matter hyperintensity lesions or subtle SAH; contrast-enhanced FLAIR imaging to evaluate disruption of the blood-brain barrier (BBB); three-dimensional time-of-flight (TOF) MRA to assess segmental cerebral arterial vasoconstriction or other vascular abnormalities such as aneurysm or arterial dissection; and magnetic resonance venography to exclude cerebral venous thrombosis (CVT) or high-grade venous sinus stenosis (22–26).
However, the distinction between primary HSA and secondary causes, which are attributed to other disorders, especially reversible cerebral vasoconstriction syndrome (RCVS), i.e., a disorder of recurrent thunderclap headaches and reversible cerebral vasoconstrictions (Table 2) (20), is difficult and is a great challenge to clinicians. In 2010, a prospective study reported that RCVS was the most common secondary cause for headache triggered by sexual activity (90%) and it was clinically indistinguishable from primary HSA in terms of headache characteristics (15). Probable RCVS is a term used to describe a condition whereby RCVS is highly suspected, but cerebral angiography is normal based on the ICHD-3 criteria (Table 3) (1). Since primary HSA and probable RCVS share both similar clinical manifestations and normal neuroimagings, these two disorders are hard to differentiate. The main differentiation of the diagnosis between primary HSA and probable RCVS may be the headache duration of up to one month defined by the ICHD-3 criterion to diagnose probable RCVS (1). However, the one-month rule is debatable and needs further research since a minority of patients with probable RCVS had headache duration more than one month (27), and many historically-reported patients with primary HSA had headaches lasting for less than one month (10,14). From a clinical perspective, physicians might need a more operational definition of these two disorders since it is impractical to wait for one month to diagnose and to treat the patients. A prospective study that used existence of sexual activity as the only trigger or accompanied with additional triggers to classify patients to primary HSA and probable RCVS might be a more practical way (13). A composite clinic-radiologic algorithm (Figure 1) is proposed to differentiate and classify headache triggered by sexual activity, i.e., primary HSA, RCVS, probable RCVS, primary thunderclap headache and other secondary causes (13). Of course, this should be scrutinized and field-tested by better-designed studies in the future.
The third edition of the International Classification of Headache Disorders (ICHD-3) Criteria for
The third edition of the International Classification of Headache Disorders (ICHD-3) Criteria for

Proposed flowchart of the diagnosis of headache triggered by sexual activity by the authors but not totally complying with the ICHD-3 criteria. Adapted and modified from Lin et al. (13).
Pathophysiology
The pathophysiology and the disease nature of primary HSA remain unclear. The human sexual response is characterized by a hypersympathetic response since norepinephrine increases in both plasma and CSF during sexual arousal and orgasm (28), which might account for one of the underlying pathogeneses of headache triggered by sexual activity. A study of cerebral hemodynamics showed that patients with primary HSA had both a greater increase of cerebral blood flow velocity and a lower decrease of the pulsatility index of the middle cerebral artery after administration of acetazolamide in comparison with healthy subjects, which demonstrated an impairment of metabolic component of the cerebrovascular vasodilator reserve in primary HSA (29).
Treatment and prognosis
Preemptive therapy with indomethacin 25–50 mg 30–60 minutes before intercourse can be given to patients with primary HSA (13), with a good response (90%) reported by a German study (30). Other prophylactic treatments may be tried, such as beta-blockers (propranolol, metoprolol or bisoprolol) (30), topiramate (31) or triptans (32), but the response is quite variable across different studies. Nimodipine, a dihydropyridinic calcium channel blocker, can be used in RCVS (26,33–35), which also showed a good response rate (85.7%) in patients with primary HSA in a study (13), but the evidence from randomized controlled trials is still lacking. Other calcium channel blocker, diltiazem, has been reported to be a successful response in a case report (36).
Most patients with primary HSA have a self-limited course, but studies have shown that up to 40% run a chronic course for more than a year (1). Thus, specific instructions may be needed for these patients to cope with future attacks, such as maintaining a passive sexual role (37), since the headache would be aggravated as sexual excitement escalates (3) or keeping the neck lower than the trunk (17).
Primary thunderclap headache
Introduction
Thunderclap headache was first coined by Day and Raskin in the 1980s to describe a patient who developed an abrupt and severe headache that they considered to result from an unruptured aneurysm (38). In addition to primary HSA (13), numerous etiologies of thunderclap headache have been identified (39–41), including SAH (42,43), RCVS (26), cervicocephalic artery dissection (44), CVT (45), cerebral infarction (43,46), intracerebral hemorrhage (43), meningitis (43), spontaneous intracranial hypotension (47), acute hypertensive crisis (48), pituitary apoplexy (49), aqueduct stenosis (50), giant cell arteritis (51), etc. Based on the ICHD-3 (2018) criteria for
The third edition of the International Classification of Headache Disorders (ICHD-3) Criteria for
While thunderclap headache is the main clinical manifestations of RCVS, whether vasoconstriction is a cause, a consequence or epiphenomenon of thunderclap headache remains unclear. Because the thunderclap headaches in RCVS usually remit at the time when vasoconstriction reaches its maximum severity (24) vasoconstriction is unlikely the direct cause of thunderclap headache. In contrast, dilatation of distal arterioles or meningeal collaterals that abruptly stretch perivascular nerve fibers may be a potential cause of thunderclap headache. However, there is lack of direct evidence to support this speculation. RCVS often cannot be diagnosed initially, since an angiogram performed in the early stage of RCVS can be normal (22,33,55,56) or is further considered as probable RCVS based on the ICHD-3 criteria (1,27). Studies have also suggested that even without significant angiographic vasoconstrictions, recurrent thunderclap headaches were almost diagnostic of RCVS (26,57–59).
Epidemiology
The lifetime prevalence of primary thunderclap headache in the general population was 0.3% in the Bruneck study (60). A population-based Swedish study showed that the incidence of thunderclap headache caused by non-SAH etiologies was 38/100,000 persons per year (43). However, with fast-growing diagnostic tools, many more secondary etiologies of thunderclap headache can be identified, and the incidence of primary thunderclap headache may decrease.
Diagnosis
Since primary thunderclap headache should be a diagnosis of last resort or by exclusion (1), exhaustive evaluations need to be performed. A comprehensive meta-analysis in 2016 demonstrated that in 8907 patients with thunderclap headache, a noncontrast brain CT performed within six hours was sufficient to rule out SAH, with a sensitivity of 98.7%, specificity of 99.9%, and a negative likelihood ratio of 0.010 (61). As a result, in the 2019 American College of Radiology (ACR) Appropriateness Criteria for
However, many other secondary causes of thunderclap headache may be obscured when using noncontrast brain CT only. For example, in a large systematic review comprising 2345 patients who presented with acute headache, a great proportion of patients had etiologies different from SAH (n = 234, 10.0%), such as RCVS (n = 234, 10.0%), cervicocephalic arterial dissection (n = 31, 1.3%), CVT (n=19, 0.8%), etc (2). Under such circumstances, the aforementioned MRI protocol in the investigation of primary HSA can be also applied in these patients.
As mentioned in

Proposed flowchart of the diagnosis of thunderclap headache by the authors but not totally complying with the ICHD-3 criteria.
Pathophysiology
The nature of primary thunderclap headache remains unknown. The sympathetic system around the vessels has been proposed to be responsible for the abrupt onset of both headache and vasospasm in patients with thunderclap headache (64). Vasospasm, which refers to a reversible process of a localized smooth muscle contraction of the blood vessel walls resulting in segmental luminal narrowing, can be an expression of high sympathetic tone (65–67). However, vasospasm only appears in RCVS but not in primary thunderclap headache. BBB breakdown may serve as a marker for understanding the pathophysiology and assisting in the diagnosis of RCVS (68,69). In a prospective cohort study with 72 consecutive patients with thunderclap headache, BBB breakdown was documented not only in 29 patients (69%) with definite RCVS but also in three patients (12.5%) with primary thunderclap headache (56). In addition, oral nimodipine demonstrated a good response not only in RCVS but also in all nine patients with primary TCH (100%), in whom brain MRA showed no vasospasm in our study (35). Nimodipine is effective for patients with and without MRA vasoconstriction (25). Therefore, these studies showed that the underlying mechanisms specific to primary thunderclap headache might be in part similar to those of RCVS.
Treatment and prognosis
The management of thunderclap headache depends on the underlying etiology. Primary thunderclap headache has a relatively benign course even without specific management (70). Two prospective studies involving 89 patients with thunderclap headache after excluding SAH who were followed up for 1.7 to 3.3 years demonstrated a very good outcome (71,72). Due to its self-limited course, some authors suggest that no treatment is needed (73). In contrast, as the diagnosis of primary thunderclap headache or probable RCVS may be ambiguous initially, oral nimodipine 30 to 60 mg every four hours can be tried in these patients to prevent further attacks of thunderclap headache (35).
Conclusion
All patients with primary HSA or primary thunderclap headache need to be investigated for an underlying secondary cause both expeditiously and exhaustively through thorough neuroimaging to avoid misdiagnosis. The disease nature of primary HSA or primary thunderclap headache remains unclear. Prior studies suggest the mechanism of RCVS is dysfunctional regulation of vascular tone that leads to vasoconstriction (20), which may result from sympathetic hyperactivity (68) or release of inflammatory cytokines (74). One study of cerebral hemodynamics showed that the breath-holding index, a measure of cerebral endothelium-dependent vasodilation, was significantly lower in patients with RCVS than in healthy subjects (75). Another study of cerebral hemodynamics also showed that patients with primary HSA had an impairment of vasodilation (29). These latter two studies demonstrated an impairment of the cerebrovascular reactivity of primary HSA and RCVS, which might suggest in part an overlapping mechanism between these two disorders. Most primary HSA and primary thunderclap headache are self-limited diseases and the prognoses are good, except that some patients with primary HSA may have a prolonged course. As a result, treatment may not be necessary for all patients. However, nimodipine shows a good response in both primary HSA and primary thunderclap headache, and indomethacin may be used as a preemptive treatment for primary HSA.
Clinical implications
Primary headache associated with sexual activity, which means a headache precipitated by sexual activity, mostly manifests as a thunderclap headache. A comprehensive neuroimaging study is needed for distinguishing primary headache associated with sexual activity or primary thunderclap headache from secondary causes. Primary headache associated with sexual activity and primary thunderclap headache are self-limited diseases and the prognoses are good, except that some patients with primary headache associated with sexual activity may have a prolonged course.
