Hair testing as a marker of adherence to headache treatments

Background Medication adherence has a critical importance in the therapy of primary headaches, both to gain pain relief and improving patients’ quality of life. Nowadays, hair is a fundamental biological specimen, alternative to the usual samples like blood and urine, to monitor adherence to pharmacological treatments: it is uninvasive and it can detect many drugs simultaneously with a detection window of many months. Therefore, in order to understandt if hair analysis should be useful to document headache patients’ adherence to chronic treatments, we analysed the detection rate of 23 therapeutic drugs in the hair of headache patients, the degree of agreement between self-reported drug and type of drug found in hair, whether the concentrations measured in hair reflected the drug doses that the patients had reported to have taken in the previous 3-month period. Methods 93 headache patients (mean age ± SD: 48.13 ± 12.03 years; females 94%) were enrolled. All these patients declared a chronic assumption, for at least 3 months, for headache prophylaxis or for their psychiatric comorbidities, with at least one of the following 23 drugs: alprazolam clonazepam, delorazepam diazepam, flurazepam, lorazepam, lormetazepam, triazolam, zolpidem, amitriptyline, clomipramine, citalopram, fluoxetine, paroxetine, sertraline, duloxetine, venlafaxine, mirtazapine, trazodone, levomepromazine, levosulpiride, quetiapine e topiramate. This study was approved by the Provincial Ethical Committee of Modena and every patient gave a free-informed consent to participate the study. A detailed pharmacological history and hair sample (7 mm diameter and 4 cm of lenght) was collected for each patient. Hair samples were analysed by liquid chromatography-electrospray tandem mass spectrometry (LC-MS/MS) by a method that we developed. Statistical analysis was performed with the StataIC 13 software. Results All the previous 23 drugs were detected in hair samples of patients who declared them. In particular, 82% of the declared drugs were found in the 100% of samples. Negative results for the declared drugs were only 9. The agreement between type of self-reported drug and type of drug found in hair was excellent for all drugs (P<0.01, Cohen’s kappa). The relationship between the last 3 months cumulative dose and hair concentration was statistically significantly (P<0.05, linear regression analysis) for delorazepam, lorazepam, amitriptyline, citalopram, duloxetine, venlafaxine and for its metabolite, desvenlafaxine. Conclusion Hair appears to be a unique matrix to monitor chronic drug use in headache patients. Drugs hair concentrations are a reliable marker of adherence to headache pharmacological treatments.