Headache as a Cardinal Symptom of Coronavirus Disease 2019

A Cross-Sectional Study

Javier A. Membrilla, MD; Íñigo de Lorenzo, MD; María Sastre, MD; Javier Díaz de Terán, MD


Headache. 2020;60(10):2176-2191. 

In This Article


To the best of our knowledge, this is the first study to specifically evaluate the semiological features of headache in patients with COVID-19 visiting the emergency department. The prevalence of headache in COVID-19 in our sample is similar to that reported for health workers tested for SARS-CoV-2 in a previous study.[25] One recently published meta-analysis of 3062 COVID-19 patients reported a much lower headache prevalence.[22] This difference is probably because both our study and the health workers' survey specifically addressed headaches. Presumably, this symptom is underrepresented in most cases, as other serious clinical scenarios may overshadow it. Our study was conducted in the emergency department, which highlights the recognition of headache as a cardinal and frequent symptom of acute COVID-19.

The ICHD-III diagnostic criteria for acute headache attributed to systemic viral infection points out that headache may be of any duration, if evidence of causation is demonstrated by certain clinical aspects, including diffuse pain and moderate or severe intensity. In our sample, only a minority of patients reported diffuse pain. Most of them could identify a certain pain location in relation to the cranial site, most notably frontal, followed by parietal. Most patients in our sample experienced moderate-to-severe headache, which was a constant symptom in patients with migraine. In the only case report of meningoencephalitis caused by SARS-CoV-2 with microbiological confirmation in CSF, the patient presented with headache among the prodromal symptoms that were followed by altered level of consciousness and seizures, but the semiological characteristics of this headache were not reported.[21] None of our patients presented neck stiffness, altered level of consciousness, seizures, or other symptoms that raised the suspicion of a CNS infection. In that case, they would have been classified higher than 4 in the Manchester Triage System, and therefore they would not have met the inclusion criteria. In our sample, some semiological features related to intracranial hypertension were not uncommon (most notably, worsening with cough but no other Valsalva maneuver). Nevertheless, worsening with every Valsalva maneuver and on lying down are much infrequent. As none of the subjects in our sample underwent CSF analysis, we cannot rule out that COVID-19-related headache may be caused by SARS-CoV-2 meningeal involvement in some cases.

We found that the presence of migraine-like symptoms such as photophobia and phonophobia, and aggravation by physical activity was frequent, in both migraine and non-migraine patients. This fact could be partly explained by convergent pathogenic mechanisms between migraine and COVID-19-related headaches. Proinflammatory cytokines, including interleukin (IL) 1β, IL-6, IL-8, and tumor necrosis factor α, have been implicated in migraine pain and are also released in the immune reaction against influenza viruses, rhinoviruses, CoVs, and other pathogens.[24,33] Although there is no solid evidence, the immune response to SARS-CoV-2 is a probable mechanism in the pathophysiology of COVID-19-related headache, with or without meningeal inflammation. However, we did not find a statistically significant correlation between laboratory parameters and intensity of pain or duration of headache episodes. An assessment of how COVID-19 affects patients with a previous headache disorder is needed for optimal management. Almost all the patients in our sample with a prior history of headache disorders had pain that was different from the usual. Adequate education on COVID-19 for patients attending headache clinics should highlight that COVID-19 can start with a headache than is different from that usually experienced, before other COVID-19-related symptoms. We emphasize the importance of recognizing this symptom to establish an early diagnosis and preventive measures.

We found that patients with migraine tend to have longer and more intense episodes of COVID-19 headaches. Increased nociceptive processing in the trigeminal cervical complex, especially in the context of peripheral sensitization of the trigeminal nerve that may occur after sustained migraine attacks, can lead to the development of central sensitization and lower pain thresholds in patients with migraine and perceptual responses being exaggerated, prolonged and of wider spread.[34–36] Taking this into account, it is not surprising that migraine individuals experience more prolonged and severe secondary cephalalgia, including COVID-19-related headaches, than those without this condition. Our results showed that COVID-19 in patients with migraine tends to debut with headache as the first symptom when compared with non-migraine individuals. The same pathophysiologic model may explain this finding as well, as a sensitized trigeminal cervical complex, which might react earlier to the inflammatory response of COVID-19, leading to headaches.

Many of the features reported by the headache specialist who described his own experience with headache during COVID-19 infection align with our most frequent findings, such as the quality of pain being pressing and aggravated by coughing.[26] Our patients did not report a multiphasic course with headache changing its semiology, but our study is not appropriate to exclude this possibility owing to its cross-sectional design.

Our study has several limitations, such as the small sample size, the cross-sectional design, the fact that multiple questionnaire items relied on patient memory and subjective perception of symptoms (such as identifying triggers) and the limited number of patients suffering from migraine. Besides, slightly less than half of the patients showed microbiological confirmation of SARS-CoV-2. We used the WHO guidelines definitions to define a probable COVID-19 case, but little can be said about the reliability of this classification. Some patients presented with typical COVID-19 pneumonia with negative PCR testing (most probably a false-negative result). No microbiological determination was performed on the remaining patients, but they had a compatible clinical syndrome and close risk contacts. The contagious capacity of SARS-CoV-2 and the epidemiological context in Spain makes COVID-19 infection the most likely diagnosis. However, patients with confirmed COVID-19 showed a statistically significant incidence of anosmia, pneumonia and necessity for hospital admission than those with probable COVID-19. Studies investigating headache in a large sample of confirmed COVID-19 patients are needed. Headache variables did not present a statistically significant difference (apart from headache at onset, which was lower in the probable COVID-19 group). Patients visiting the emergency department were included; thus, this sample may not be fully representative of headache attributed to COVID-19 in other clinical settings. Many comparisons were required to study evaluate this study hypotheses; therefore, type I error is probably greater than desirable given the number of statistical interferences. These patients did not undergo the full battery of tests to exclude all complications of COVID-19 that may cause headache (CNS infection, stroke, cerebral venous thrombosis, etc.). However, none of them presented with altered level of consciousness, neurological focal signs, or other clinical data that raised suspicion about these conditions. Lastly, this study has inherent biases associated with convenience sampling: our findings cannot be inferred to the complete spectrum of the COVID-19 disease, as we excluded patients who required emergent diagnostic-therapeutic intervention at the time of consulting. Nevertheless, we decided that this method of sampling was the most appropriate for recruiting patients in the emergency department in the context of the epidemiological crisis at the time the study was conducted.

Among the strengths of the study, the questionnaire, used according to the recommendations of the International Headache Society, was designed by a headache specialist neurologist, which made it easier for emergency physicians to describe the headaches accurately. The study was carried out in a public hospital that had confirmed more than 4000 people with COVID-19, and has become a reference for patients with this infection. For all these reasons, our research has clinical implications in the fight against the pandemic and places headache as a symptom to be considered primarily among patients with COVID-19.

Furthermore, well-designed prospective studies are needed to enhance our knowledge on headaches and COVID-19 for better management of the patients.