Significance of consecutive international normalized ratio (INR) outcomes using statistical control rules in long-term anticoagulated patients. Optimization of laboratory monitoring and interpretation of borderline measurements.

摘要

Analysis of serial measurements is needed to elaborate observations in all fields of medical research. In the present study, retrospective data of anticoagulated patients were used to calculate a mean of observations and control limits, X+/-1sigma and X+/-2sigma (mean+/-standard deviation). During 18 months of coagulation monitoring, 45 patients without major oral anticoagulant therapy complications, with more international normalized ratio (INR) determinations in the therapeutic range and with a normal distribution of INR values according to the Kurtosis coefficient, were selected. The cumulative distribution functions allowed us to obtain critical limits of INR with a cumulative probability (p). Control limits, calculated for a therapeutic control chart, indicated through different control rules, 1S2 or 2S1, an alarm signal to analyze the cause of INR outside the therapeutic range. Our investigation suggested that for results at the level of the therapeutic control limits, we needed at least two consecutive INR results to detect a significant over- or under-anticoagulation. The therapeutic control chart method should be a useful means in clinical practice for evaluating the statistical significance of consecutive and borderline INR outcomes. Analytical improvements and control rules applied to laboratory monitoring may help optimize drug dose administration.