Predicting outcomes of dual antihypertensive therapy

Abstract

Objective:to establish predictors of failure of dual antihypertensive therapy in patients with arterial hypertension (AH) of high and very high cardiovascular risk, to create a predicting model for the negative outcome of dual antihypertensive therapy. Materials and methods: The study included 88 patients with uncontrolled hypertension who did not receive basic antihypertensive therapy, mean age 58.0±1.52 years. The examination was carried out on the basis of the current regulatory documents regulating the examination at the stationary stage. Additionally, the level of highly sensitive C-reactive protein and blood uric acid was determined. 24-hour blood pressure monitoring (ABPM) and electrocardiogram (ECG), respiratory polygraphy of sleep, and computerized  capillaroscopy  were performed. The search for predictors and the creation of a predictive model were carried out using  the binary logistic  regression method. Results: the effectiveness of dual drug therapy was 33.0%. Prognostic markers associated with a negative outcome of treatment were identified: interventricular septal thickness (IVS) (OR 3.44; p=0.009), linear density of the capillary network (OR 4.65, p<0.001), area density of the capillary network (OR 3, 98; p<0.001); remodeling  coefficient (OR 8.40; p=0.003), mean nighttime systolic blood pressure (SBP) (OR 1.94; p=0.014), mean daily SBP (OR 2.86; p=0.014), SDNNi (OR 2 .86; p=0.007). The final model included the IVS, areal density of the capillary network, and the average daily SBP. Model characteristics: regression coefficient 25.74; p<0.001; p Wald 0.049; OR 9.1 (95% CI 3.12; 26.82). The sensitivity of the method was 83.1%, the specificity was 69.0%. The area under the ROC curve was 0.805±0.05 (95% CI 0.707; 0.882). The cut-off point corresponded to the calculated probability value of 0.599. Conclusions: the study demonstrated the low efficacy of dual antihypertensive therapy. Given the lack of approved algorithms for predicting outcomes of drug treatment in patients with uncontrolled hypertension, the developed model is of clinical interest and may be useful in achieving better BP control. To improve the quality of the model in order to increase the sensitivity and specificity of the method, further study on larger samples is required.