AbstractWhat is known and objective:
Vancomycin is still the first-line treatment for resistant gram-positive infections, particularly for methicillin-resistantStaphylococcus aureus(MRSA) infections. The vancomycin treatment guideline is based on the association between vancomycin trough concentration and clinical outcome. We here present a retrospective analysis of whether the trough level of vancomycin is associated with clinical outcome in Chinese patients with gram-positive infections.Methods:
Clinical data were collected retrospectively from patients under vancomycin therapeutic drug monitoring in Huashan Hospital from March 2004 to September 2014.Results and discussion:
A total of 148 inpatients with gram-positive infection were identified and data on their corresponding vancomycin serum trough concentration retrieved. A total of 113 strains of gram-positive bacteria were isolated from 111 patients, including 90 strains of MRSA. Vancomycin was used for 11 to 13 days on average. The overall bacterial eradication rate was 67·3% (76/113), including 61·1% (55/90) for MRSA and 91·3% (21/23) forEnterococcus. Multivariate logistic model analysis showed that vancomycin trough concentration was not associated with clinical outcome (OR: 1·0; 95% CI: 0·92, 1·08,P= 0·9613). The incidence of adverse drug reactions was low and not related to vancomycin trough concentration.What is new and conclusion:
This retrospective analysis failed to demonstrate an association between vancomycin trough concentration and the clinical and microbiological response. Prospective controlled studies are necessary to further establish the need for the higher trough concentrations normally cited for clinical efficacy.
Association between vancomycin trough concentration, baseline creatinine clearance and vancomycin daily dose. Vancomycin daily dose was positively associated with vancomycin trough concentration (slope of logistic regression = 3.56, P < 0.0001), while creatinine clearance was negatively associated with vancomycin trough concentration (slope of logistic regression = −0.03, P = 0.0152). However, the association between vancomycin trough concentration and clinical or microbiological response was not significant in patients with gram-positive infection.