Monitoring vancomycin blood concentrations reduces mortality risk in critically ill patients: a retrospective cohort study using the MIMIC-IV database

Background: The incidence and mortality of severe Gram-positive cocci infections are particularly high in intensive care units (ICUs). Vancomycin remains the treatment of choice for severe infections caused by Gram-positive cocci, particularly methicillin-resistant Staphylococcus aureus (MRSA). Some guidelines recommend therapeutic drug monitoring (TDM) for critically ill patients treated with vancomycin; however, there is currently a lack of evidence to support that TDM improves the mortality rates of these patients. Therefore, we designed this cohort study to compare the impact of monitoring vancomycin blood concentrations on mortality rates in critically ill patients and to provide evidence to support this routine clinical practice. Methods: Data were extracted from the Medical Information Mart for Intensive Care (MIMIC)-IV database for a retrospective cohort analysis of critically ill patients receiving intravenous vancomycin treatment. The primary outcome was the 28 day mortality rate. The propensity score matching (PSM) method was used to match the baseline characteristics between patients in the TDM group and the non-TDM group. The relationship between 28 day mortality and vancomycin TDM in the critically ill cohort was evaluated using Cox proportional hazards regression analysis and Kaplan-Meier survival curves. Validation of the primary outcomes was conducted by comparing the PSM model and the Cox proportional hazards regression model. The robustness of the conclusion was subsequently verified by subgroup and sensitivity analyses. Results: Data for 18,056 critically ill patients who met the study criteria were collected from the MIMIC-IV database. Of these, 7,451 patients had at least one record of vancomycin blood concentration monitoring, which we defined as the TDM group. The TDM group exhibited a 28 day mortality rate of 25.7% (1,912/7,451) compared to 16.2% in the non-TDM group (1,723/10,605). After PSM, 4,264 patients were included in each of the TDM and non-TDM groups, with a 28 day mortality rate of 20.0% (1,022/4,264) in the TDM group and 26.4% (1,126/4,264) in the non-TDM group. Multivariate Cox proportional hazards analysis revealed a significantly lower 28 day mortality risk in the TDM group when compared to the non-TDM group (adjusted hazard ratio [HR]: 0.86; 95% confidence interval [CI]: 0.79, 0.93; p < 0.001). Further PSM analyses (adjusted HR: 0.91; 95% CI: 0.84, 0.99; p = 0.033) confirmed the lower risk of mortality in the TDM group. Kaplan-Meier survival analysis revealed a significantly higher survival rate at 28 days for the TDM group (log-rank test, p < 0.001). Subgroup analysis results indicated that patients with sepsis, septic shock, estimated glomerular filtration rate <= 60 mL/min/1.73 m(2), undergoing renal replacement therapy, using vasoactive drugs, on mechanical ventilation, and those with higher severity scores (Acute Physiology Score III >= 40, Oxford Acute Severity of Illness Score >= 30, Simplified Acute Physiology Score II >= 30) significantly benefited from monitoring vancomycin blood concentrations. The results remained unchanged excluding patients staying in ICU for less than 48 h or those infected with MRSA. Conclusion: This cohort study showed that monitoring vancomycin blood concentrations is associated with a significantly lower 28 day mortality rate in critically ill patients, highlighting the importance of routinely performing vancomycin TDM in these patients.