OBJECTIVE We conducted population anticoagulant pharmacodynamic analysis for patients administered the low-molecular weight heparin tinzaparin. METHODS Data from 425 patients (2,631 observations) who participated in 2 Phase III clinical studies were utilized in an analysis based on a pharmacodynamic structural model of anti-Xa activity using non-linear mixed effects modeling techniques. Anti-Xa activity from patients participating in a multicenter, randomized, double-blind clinical trial comparing intravenous once daily subcutaneous tinzaparin (175 IU/kg) with heparin for the treatment of deep vein thrombosis (DVT) was first examined using a 2-compartment model with first-order absorption and endogenous anti-Xa activity. Covariates included renal function, body weight, age, gender, race, obesity, concomitant administration of warfarin, and diabetes. RESULTS The population estimates and 90% confidence intervals (CI) for oral clearance (CL) and apparent volume of distribution of the plasma compartment (Vc) were 0.0176 l/h/kg (CI = 0.012-0.023) and 0.098 l/kg (CI = 0.088 - 0.109), respectively. The elimination half-life was 3.9 h (CI = 2.5-5.2). These estimates are similar to findings in healthy volunteers. The inter-patient variability in clearance was related to plasma creatinine and percent above ideal body weight. Clearance decreased by 22% for patients with severe renal function impairment (creatinine clearance < 30 ml/min), and by about 25% in obese patients (BMI > 30 kg/m2). CONCLUSIONS Changes of these magnitudes were not clinically important in pooled clinical trial safety and efficacy analyses. Body weight was not a significant covariate in the model supporting the observations in earlier well-defined trials, where tinzaparin was dosed on a weight basis (lU/kg). Clearance was not influenced by age, race or gender. The same model was applied to data obtained from a prospective, randomized, double-blind clinical trial comparing tinzaparin (4,500 IU) to enoxaparin (40 mg) once daily in patients undergoing total hip replacement. Model parameters were similar to those previously obtained supporting the extension of these results across dose and indication. Population analysis in patients with disease and heterogeneity indicated similar pharmacodynamics as in volunteers, supporting weight-based dosing and identified the dependence of clearance on obesity and severe renal function, although the magnitude of these effects are probably not clinically significant.