Cardiac cephalalgia: a narrative review and ICHD-3 criteria evaluation

We found a total of 88 cases of cardiac cephalalgia, 58 from case reports [7‐53] (a complete description of these cases can be seen in supplementary material) and 30 patients from a recent observational study [54]. Among case reports, mean age at onset was 62.3 years (± 13.1) and most patients were males (n = 36; 62.1%). These data are in line with the study from Xu et al. that reported a mean age of 64.6 years (± 11.9) and a predominance of male patients (53.3%) [54]. History of cardiovascular risk factors was common. It was present in 41 (70.7%) of the case reports and in all the patients from the observational study [54]. History of headache was reported only in 5 patients (8.6%). Nevertheless, it was not specified in most case reports and was not assessed in the observational study [54] (Table 3).

Regarding headache features, location of pain was variable and occipital pain was the most frequently reported among previous case reports (n = 16, 27,6%) [7‐53]. Nevertheless, pain located in the frontotemporal region was the most frequent in the study by Xu et al. (n = 14, 46.6%) [54]. Bilateral headache was reported in 24 patients (41.4%) and unilateral headache in 7 (12.1%) of the case reports (Table 3). Headache quality was also very variable. Among case reports, throbbing (n = 8, 13.8%) and pressing pain (n = 7, 12.1%) were the most frequent but in one-half of the cases headache quality was not specified. Xu et al. found that pulsating quality was the most frequent (33.3%) followed by dull (16.7%) and stuffy pain (16.7%) [54] (Fig. 2).

In total, more than half of the 88 cases reported severe intensity (n = 51, 57.9%), moderate intensity was described by 15 cases (17%), mild intensity by 6 (6.8%) and variable intensity by 3 (3.4%). In 13 patients (14.7%) headache intensity was not specified. Headache was accompanied by nausea in 28 cases (31.8%), was not accompanied by photophobia or phonophobia in 49 patients (55.7%) and was aggravated or induced by exertion in 49 (55.7%). It is also interesting that two cases reported tearing and one case reported rhinorrhoea [54].

In 21 patients from the case reports headache was the only symptom of the myocardial ischemia [7‐53]. Nevertheless, in line with the results from the study from Xu et al. [54], most patients presented other cardiac symptoms including typical angina manifestations (Table 4). 23 patients underwent stress test, which was positive in most cases and in 17 patients headache was reproduced during the stress test (Table 4). Coronary angiography was performed in 78 patients (88.6%) and most patients presented severe coronary disease. Single-vessel disease was reported in 21 cases (23.9%), double-vessel disease in 15 cases (17%), triple-vessel or multivessel disease in 39 cases (44.3%), vasospasm was found in one case [35] and two cases presented normal coronary arteries [30, 35]. Although the affected vascular territory was variable, the left anterior descending artery (n = 27, 30.7%) and the right coronary artery (n = 26, 29.5%) were the most frequently affected (Table 4). 22 patients (42.3%) were treated by percutaneous transluminal coronary angioplasty, 12 cases (23.1%) required coronary artery bypass and 12 patients (23.1%) received medical treatment. Headache improved or resolved in all cases after treatment of the myocardial ischemia. Although the follow-up period was variable, 37 patients (42%) experienced complete resolution, 33 (37.5%) improved, five patients (5.7%) experienced headache recurrence, eight patients (9.1%) died in the acute phase or during the follow-up period and in five cases (5.7%) outcome was not specified. Four of the patients with headache recurrence experienced simultaneous myocardial ischemia recurrence [11, 15, 18, 42].

Criterion B ‘Acute myocardial ischemia has been demonstrated’ was fulfilled by 87 cases (98.9%). In one case of headache with and without exertion relieved by nitrates myocardial ischemia was not demonstrated because patient died suddenly without electrocardiogram and cardiac enzymes determination [25] (Table 5).

To evaluate criterion C1 `Headache has developed in temporal relation to the onset of acute myocardial ischaemia’ it is necessary to observe the temporal relationship of headache and cardiac symptoms but it can only be confirmed if we observe the onset of headache during a stress test. We considered that 34 patients from the case reports met this criterion. 17 patients reported headache onset in relation to onset of cardiac symptoms and in 17 cases stress tests confirmed the temporal relationship between myocardial ischemia and headache. All 30 patients from the study from Xu et al. met this criterion [54] (Table 5).

Regarding criterion C2 most patients (93.2%) fulfilled criterion C2b ‘Headache has significantly improved or resolved in parallel with improvement in or resolution of the myocardial ischaemia’. However, evaluate criterion C2a is difficult because it is hard to determine the worsening of myocardial ischemia. For this reason, we considered that only three cases who presented worsening of headache in relation to confirmed worsening of the myocardial ischemia during the coronarography met criterion C2a [13, 14, 52]. Xu et al. reported that all 30 patients met this criterion [54] (Table 5).

Criterion C3 is the criterion that describes headache characteristics. Criterion C3a ‘moderate to severe intensity’ was met by 75%, criterion C3b ‘accompanied by nausea’ by 31.8%, criterion C3c ‘not accompanied by photophobia or phonophobia’ by 55.7% and criterion C3d ‘aggravated by exertion’ by 55.7%. Considering these findings, nausea is not a frequent accompanying symptom in cardiac cephalalgia. This finding is in line with the results of the literature review by Chen et al. [6] that found accompanying nausea only in 23% of cases and with the study from Xu et al. [54] that reported that 36.6% of patients fulfilled this criterion.

Finally, among those cases that reported nitrates administration, Criterion C4 ‘Headache is relieved by nitroglycerine or derivatives of it’ was fulfilled by 100% of cases. However, it must be considered that in 36 case reports the use of nitrates was not specified and that in the study by Xen et al. all 12 patients did not receive nitrates [54].

Taking into account the present evaluation and in order to improve ICHD-3 diagnostic criteria, we might suggest removing criterion C3b ‘accompanied by nausea’ given that it was the least commonly fulfilled criterion.

Even though the pathophysiology of cardiac cephalalgia is not known, four hypotheses have been proposed to illustrate its mechanisms [7, 53‐56].

The first hypothesis states that cardiac cephalalgia could be the result of the convergence of afferent autonomic visceral fibers, which carry nociceptive cardiac stimuli, and somatic sensory fibers onto a common pool of spinothalamic tract (STT) and spinoreticular tract (SRT) neurons. Thus, according to this hypothesis, cardiac cephalalgia would be a referred pain produced because the brain interprets that cardiac pain is originated in the somatic structures [57‐59].

Foreman et al. [58] proposed that ‘typical angina’ and ‘atypical angina’ would have different mechanisms. In the first scenario, sympathetic cardiac afferents may converge onto STT and SRT neurons with the somatic stimuli from the chest and upper arm in the upper thoracic and cervical spinal segments (T1-T5 and C5-C6). In contrast, angina associated to referred pain to the neck and jaw might be mediated by cardiac vagal afferents that predominantly terminate in the nucleus tractus solitarius and then project to STT neurons in the C1-C2 segment. Cervical and trigeminal afferents converge in the trigeminocervical complex which is constituted by C1 and C2 segments and the trigeminal nucleus caudalis [60, 61]. Therefore, vagal cardiac afferents could activate trigeminal neurons and produce referred pain in the cervical and trigeminal territory.

The concentration of cardiac afferent autonomic visceral fibers can vary for each cardiac region [62]. Although the anatomy and physiology of this fibers are not yet fully understood, it is thought that a greater concentration of sympathetic cardiac afferent fibers is primarily located in the anterior part of the left ventricle whereas parasympathetic afferent fibers mainly innervate the inferior-posterior wall of the left ventricle [58]. Since the anterior portion of the left ventricle is mainly perfused by the left anterior descending artery, whereas the posterior and inferior walls are principally supplied by the right coronary artery and the circumflex artery, the predominant activation of sympathetic or parasympathetic fibers may be determined by the artery responsible for the myocardial ischemia [58, 63]. Therefore, cardiac cephalalgia might depend on the area of the heart in which myocardial ischemia occurs. Figure 3 shows a representation of the convergence theory of cardiac cephalalgia.

The second theory states that the abrupt decrease of cardiac output during myocardial ischemia would cause a rising of pressure in the left atrium and left ventricle hindering cerebral venous return. Hence, cardiac cephalalgia would be the result of an increased intracranial pressure secondary to a reduction of cerebral venous drainage [7, 53‐56].

The third hypothesis suggests that the release of inflammatory mediators during cardiac ischemia including bradykinin, serotonin, substance P and histamine could cause vasodilation in brain arteries leading to headache [7, 53‐56].

Finally, Wang et al. [46] proposed a fourth hypothesis. They reported a case of cardiac cephalalgia with cortical cerebral hypoperfusion confirmed by the perfusion-weighted images of brain magnetic resonance (MR) imaging and with normal MR angiography during the headache attack. Therefore, they deduced that cardiac cephalalgia might be the result of the vasoconstriction of small intracranial arteries caused by the activation of the sympathetic system by myocardial ischemia. They also proposed that cortical cerebral hypoperfusion may provoke cortical spreading depression which could additionally contribute to headache.

We provide an updated overview on cardiac cephalalgia from the physiopathology to the diagnostic criteria. The major limitation of our review is that the description of headache characteristics of cardiac cephalagia in a high proportion of previous reports were scarce and that there were a large number of terms to describe headache quality. Another limitation is that headache history was rarely reported. Thus, patients with a worsening of a preexisting headache may have been misdiagnosed as cardiac cephalalgia. To further elucidate the characteristics of cardiac cephalalgia and its pathophysiology it would be of great interest to conduct further studies with standardized data collection of patients with demonstrated myocardial ischemia in order to investigate cardiac cephalalgia prevalence and characteristics.