Comparative evaluation of warfarin utilisation in two primary healthcare clinics in the Cape Town area

Background: Although warfarin remains the anticoagulant drug of choice in a wide range of patients, its narrow therapeutic window makes patients susceptible to a high risk of bleeding complications or failure to prevent clotting. This has necessitated therapeutic monitoring in warfarinised patients. Factors that could be responsible for the fluctuating responses to warfarin vary from pharmacogenetic to concomitant morbidity, diet and medication. In order to assess the quality of management of warfarin treatment in a local primary-care setting, the current study evaluated warfarin utilisation and monitoring records in two hospitals with different patient groups. Methods: A retrospective study was undertaken in the specialised warfarin clinics at Wesfleur and Gugulethu hospitals (Western Cape, South Africa) covering all warfarin-related therapy records over a 12-month period. Data extracted from the patients' folders included age, gender, race, weight, address, concurrent chronic illnesses, treatment and medication, indication for warfarin and INR history. Results: A total of 119 patients' folders were analysed. Attendance at the clinics reflects the demographics and racial distribution of the host location of the hospitals. While all the patients were maintained above the minimum international normalised ratio (INR) value of 2, about 50% had at least one record of INR above the cut-off value of 3.5. However, over a third of the patients (32.2%) had at least one record of INR greater than 3.5 in Gugulethu Hospital, compared to over half (58.3%) in Wesfleur Hospital. In total, atrial fibrillation was the most common indication for warfarinisation while hypertension was the most common concurrent chronic condition in warfarinised patients. All patients who received quinolone antibiotics had INR values above the cut-off point of 3.5 within the same month of the initiation of antibiotic therapy, suggesting drug-induced warfarin potentiation. Other co-medications, including beta-lactam antibiotics, non-steroidal anti-inflammatory drugs (NSAIDs) and anti-ulcer drugs appeared to alter warfarin responses as measured by recorded INR values. Conclusion: The study found inter-individual variability in the response to warfarin therapy, which cut across racial classifications. It also confirms the possible influence of concomitant morbidity on patient response to anticoagulant therapy.