This is an observational descriptive study with a series of cases design. The study adhered to the Strengthening the Reporting for Observational Studies in Epidemiology (STROBE) guidance (13).
The study was done in the Headache Unit of Hospital Clínico Universitario, Valladolid, Spain, a third level public hospital with a reference population of 261,000 inhabitants that serves as headache reference center for a 2,700,000 population. Referrals can be done directly from primary care, from specialized care or from general neurology consults. The waiting time between the referral and the evaluation ranged between 2 and 12 weeks during the study period.
Study outcomes
The primary outcome was the 50% responder rate of the preventive treatments, defined as the proportion of patients that present a 50% reduction in the number of headache days per month, evaluated between weeks 8-12 after the onset of treatment, compared with the month prior to the treatment use. This outcome was adapted from the International Headache Society (IHS) guidelines for controlled trials of preventive treatment (14).
The secondary outcomes included: 1) to describe the proportion of patients that required acute or preventive treatment; 2) to describe the 30% and 75% responder rates of the preventive treatments, evaluated between weeks 8-12 (14); 3) to describe the response to the acute treatments, according to the IHS guidelines (15).
Eligibility criteria
Patients were included if they 1) had a confirmed COVID-19 diagnosis, according to a real-time positive polymerase chain reaction test from a nasal or oropharyngeal swab sample and/or a positive result of IgM serum antibody testing in a patient with typical clinical symptoms; 2) presented headache during the course of the disease; 3) had been referred to the Headache Unit due to headache. Cases were excluded if they 1) had any neurological or psychiatric condition resulting in cognitive impairment that could difficult the description of the headache or the effectiveness of the treatments; 2) had language or speech disorders, or insufficient performance of Spanish language; 3) expired during the follow-up.
Recruitment, sampling, and sample size
The recruitment was based in a non-probabilistic approach. All consecutive patients that were referred to the Headache Unit were screened for eligibility. A sample of 100 patients was considered representative, with no formal sample size calculation due to the prior absence of evidence regarding acute or preventive treatment of COVID-19 related headache.
Study period
Patients that had been infected between March 1, 2020, and January 31, 2022, were evaluated. To ensure a minimum follow-up of 3 months after the evaluation, patients were evaluated until April 30, 2022.
Intervention
In the first visit, an in-person evaluation by a headache expert was done in all cases. In the follow-up visits, either telephone or in-person visits were done, depending to the responsible physician judgment. A standardized questionnaire, based in previous studies (2, 4, 16), was administered. Demographic variables included age and sex. Clinical variables addressed family history of headache, prior history of headache, comorbidities (depression, anxiety, insomnia, or other chronical painful syndromes), and vaccination status versus COVID-19. Concerning the COVID-19 infection, the time elapsed between the infection and the evaluation was assessed, and the presence of fever, anosmia, or pneumonia as a COVID-19 manifestations. The evaluation of the COVID-19 related headache included location (hemicranial, holocranial), topography (frontal, temporal, parietal, occipital, periocular, facial, cervical, or vertex), quality of pain (pressing, throbbing, stabbing, electric, burning), intensity of the headache in a numerical rating scale (0: no pain, 10: worst possible pain), associated symptoms (photophobia, phonophobia, osmophobia, nausea, vomiting, cranial autonomic symptoms), worsening of the headache by physical activity, frequency of headache (number of headache days per month, number of acute medication days per month). Patients were allowed to report several locations of the headache and multiple locations of the pain. A headache diary was given to all patients to depict the frequency, intensity and response to treatment was given to evaluate the treatment response. Treatments that had been employed prior to the referral and those therapies that were prescribed in specialized care were evaluated.
Acute medication
The evaluation included the number of types of employed drugs and the clinical response. To evaluate the response to the acute medications, pain freedom at 2 hours was employed (15). Since patients reported the results following multiple uses of each drug, the response to each drug was classified, as per patient and headache diary criteria, based on the proportion of uses where subjects become pain free at 2 hours, stratified into the following ranges: 0-30%, 31-50%, 51-75% and >75% of uses.
Preventive medication
The number and type of preventive medications were gathered. To describe the clinical response, the responder rate was estimated, defined as the percent reduction from baseline in the number of headache days per month between weeks 8-12, compared with the month prior to the use of the treatment. The 30%, 50% and 75% responder rates were estimated per each treatment.
Ethics
The study was approved by the Valladolid East Ethics Review Board (PI 21-2499-TFG) and patients gave their consent prior to any study intervention. The study was done in accordance with the Declaration of Helsinki principles.
Statistical analysis
Since the sample size was 100, qualitative and ordinal variables are described as percentages. When sub-groups or categories with missing data are reported, fractions are reported, being the denominator the total number of cases, along with the percentage. Quantitative variables are reported as mean and standard deviation (SD) or median and inter-quartile range (IQR), depending on the type of distribution. Normality was assessed with the Kolmogorov-Smirnov test.
Availability of data
The anonymized datasets supporting the conclusions of this study are available for other researchers upon request to the corresponding author.