Study of Adherence to Antihypertensive Drugs and Statins by Hair Analysis

The problem of adherence to long-term therapies has been known for years and occurs substantially in all countries, although with greater severity in developing countries.There is a positive correlation between poor adherence and mortality, as demonstrated by Simpson et al. with their analysis of data from 46,847 patients: the risk of mortality is halved for patients who take their medications appropriately compared to non-adherent patients (Simpson, 2016). It is reasonable to assume that medication non-adherence will increase as life expectancy increases, resulting in more chronic conditions to be treated.

The main areas in which monitoring adherence to therapy can play a crucial role, according to the aforementioned World Health Organization (WHO) document include primarily cardiovascular disease, The data on the use of drugs in the Italian population show dramatically low adherence to the prescribed therapy, with data of persistence in therapy at one year of 53% for antihypertensive drugs and 47% for lipid lowering drugs (OSMED). There is urgent need to promote cognitive investigations on the causes of abandonment of therapy and to have appropriate tools to assess the real adherence to therapy in the individual patient.

These two objectives can be pursued through longitudinal studies: retrospective studies related to patients for whom there is data collection and prospective studies in which the relationship between the treatment schedule, clinical characteristics of patients and medical intervention can be subjected to experimental verification. For these studies it is necessary to use analysis tools suitable to evaluate, over time, the real exposure of patients to the drugs prescribed. The analysis of pharmaceuticals (and their metabolites) in keratinized tissues, such as hair, appears to meet many needs for retrospective analytical investigation. The known and relatively constant speed of hair growth, considering that the xenobiotics incorporated at the root level remain virtually unmodified for a long time, allows to chronologically place the intake of substances in relation to their distribution along the hair shaft (see: Guidelines for Testing Drugs under International Control in Hair Sweat and Oral Fluid, UNODC, Vienna, 2014). The present investigation is a prospective open study in which 200 patients followed for hypertension at the centres of the Italian Society of Hypertension (SIIA) willing to take part will be evaluated on the basis of their clinical characteristics, the presence of comorbidities and their therapeutic scheme will be subjected to hair sampling for the analysis of drugs of interest and their possible metabolites in this matrix. The use of the prescribed drugs will be recorded along with the relevant clinical parameters and preliminarily evaluated by application of a questionnaire. At the first visit, patients who have been prescribed continuously in the 3 months preceding the study the same active ingredients will be invited to participate in the study, which involves a hair sample (about 100 mg cut from the skin, recommended length 3 cm). They will then be prescribed the same drugs according to the same scheme if the blood pressure values are within the recommended limits. When more than one antihypertensive drug is present in the therapeutic scheme, they will be prescribed as a single pill, as recommended (Hypertension Guidelines 2018, Primary Prevention 2021). For patients with uncontrolled blood pressure, the therapeutic scheme will be modified with the inclusion of other drugs and using pills with drug combinations when possible. Patients treated with statins will continue the treatment according to clinical indication using, when possible,pills containing associations with antihypertensive drugs. The same procedure will apply in case a statin is included in the therapy for clinical indication. Patients will be given two questionnaires: BMQ (Brief Medication Questionnaire); MUAH (Maastricht Utrecht Adherence in Hypertension questionnaire); MMAS (Morisky Medication Adherence Scale). This will allow a comparison of the data obtained from the questionnaires with the analytical data from the hair analysis.Patients will be examined and re-tested 3-4 months after the first visit. The use of prescribed medications will be recorded along with clinical parameters of interest and preliminarily evaluated by application of questionnaires.

At each of the two visits, the investigating physician will measure blood pressure (average of 3 measurements) at the first and second visit, using a semi-automatic device according to the European Society of Hypertension guidelines (ESH-ESC guidelines 2018): in the sitting position (measured at both arms at the first visit) and orthostatic position (3 measurements after 1 and 3 minutes) and heart rate. Ambulatory blood pressure monitoring (24-hour) will be performed within a short time interval from the first visit and the second visit.Total cholesterol, LDL and HDL cholesterol, triglyceridemia, glycemia, uricemia, creatinine, sodium, potassium, and complete blood cells count, recently performed by the patient according to clinical indications, will also be recorded.

The investigating physician will be able to acquire information about the patient, through information requested directly from the family doctor as per clinical practice.

Data management and procedures to ensure data confidentiality: Patients data will remain protected in compliance with European Union data protection regulations (EU General Data Protection Regulation 679/2016).

The collected data will be anonymized by assigning an identification code for each patient that will be reported in the data collection form. The attribution of the code and the link to the patient's name will be in an electronic card whose access is limited only to investigators at the site with protected access. To minimize the risk of unauthorized access, the hardware in which patient data will be recorded will be placed in a restricted access area accessible only to authorized personnel who need to retrieve information for professional purposes. The electronic database will be protected through the use of a username and password that will be changed periodically.

The samples will be stored for 15 years at the Institute of Forensic Medicine of the University of Verona.

The data obtained will be owned by the investigators and co-experimenters (including recruitment centers).

Analytical Methods: Hair extraction ad ultra high performance liquid chromatography (UHPLC)- triple quadrupole mass spectrometry (QQQ/MS) analysis will be performed in Verona (Prof. Franco Tagliaro, Department of Forensic Medicine). Mode of expression of results: concentrations of active ingredient and any metabolites in the keratin matrix are expressed in ng or fractions of ng/mg.

Sample size calculation: The proposed study is aimed at defining the feasibility and the necessary sample size - thus defining the objectives of an intervention study aimed at achieving a high degree of adherence to drug therapy and, consequently, to have an impact on clinical events in the population actively followed compared to the historical data and the reference population. Since this is a pilot study has not been performed an a priori calculation of sample size, but it is expected to enroll about 200 subjects in 6 months of enrollment.

Study duration: Patient Enrollment Duration: 6 months Patient Involvement: 3 months Total Study Duration: 12 months Data analysis: Data will be collected anonymously using a progressive code that can also identify the different centers. Continuous variables will be summarized through mean and standard deviation if normally distributed, and through median and interquartile range if asymmetrically distributed, while for categorical variables absolute or relative frequencies will be used. Significance of differences will be assessed by Student's t test (or the Mann-Whitney test for nonparametric analyses) for quantitative variables and by Fisher's exact test or χ2 test for categorical variables (α=0.05, two-tailed). Multivariable logistic regression models for dichotomous outcomes (adherence=no/yes) and multiple linear regression models for continuous outcomes (concentrations of/drugs in the hair) will be used to study the different factors associated with correct or non-adherence to therapy.

Multivariable analyses will aim to assess any correlation between adherence and clinical/demographic variables in an expository manner. Subsequently, it will be possible to evaluate in which classes of drugs there is a higher/lower adherence.