Renal dose adjustment is commonly operationalized by assigning a patient to a kidney-function category and applying a corresponding dosing recommendation. This approach is practical when renal function is relatively stable, but it becomes increasingly fragile when estimated filtration, acute kidney injury trajectories, extracorporeal clearance, measured drug exposure, and clinical response evolve on different timescales or point toward different dosing decisions. This Current Opinion argues that renal dosing under dynamic kidney function should be treated as a recurrent reconciliation problem rather than a one-time classification task. Estimated glomerular filtration rate and creatinine clearance remain important inputs, but neither directly represents total drug clearance, and both may become misleading when creatinine is non-steady, biomarker determinants are altered, augmented renal clearance emerges, or dialysis contributes variable extracorporeal removal. Drug concentrations can provide a more proximal exposure signal for selected therapies, yet their interpretation depends on sampling, pharmacokinetic targets, assay availability, model assumptions, and clinical context. Likewise, efficacy and toxicity should trigger reassessment without being mistaken for direct measures of renal elimination. We therefore propose a dynamic renal-dose reconciliation approach in which kidney estimates, trajectory, dialysis conditions, drug exposure, and observed response are repeatedly interpreted according to their temporal relevance, drug specificity, directness to exposure, and reliability. The model is conceptual rather than validated, and important boundaries include drugs without informative therapeutic drug monitoring, rapidly changing physiology, heterogeneous dialysis prescriptions, nonrenal clearance, uncertain pharmacodynamic targets, and settings with limited measurement infrastructure.
