[Clinical efficacy and safety of vancomycin compared with linezolid for the treatment of neonatal gram-positive bacterial sepsis].

OBJECTIVE

To evaluate the clinical efficacy and safety of vancomycin and linezolid for the treatment of gram-positive neonatal bacterial sepsis.

METHOD

The data of neonates diagnosed as gram-positive bacterial sepsis in neonatology department of Suzhou Municipal Hospital from June 2009 to December 2015 were retrospectively collected. These neonates were divided into vancomycin group and linezolid group. Propensity score matching (PSM) on baseline variables was used to balance the two groups by identifying a comparable group of neonates who received vancomycin and linezolid therapy. Clinical and microbiologic success rates were compared by chi-square test, and changes of laboratory parameters before and after treatment at the end of treatment were then directly compared by rank-sum test between the matched groups. In vancomycin group, correlation between trough concentration of vancomycin and clinical efficacy were evaluated.

RESULT

Totally 108 and 209 cases were respectively selected in vancomycin and linezolid groups; 108 cases with well-matched baseline characteristics were included in matched linezolid group. The clinical success rates of vancomycin vs. linezolid therapy were 86.1%(93/108) and 88.9%(96/108)(P=0.681), and the microbiologic success rates were 91.7%(99/108) and 93.5%(101/108)(P=0.795). The average trough concentration of vancomycin was (12±8) mg/L. The rate of reaching the high trough concentration standard was only 33.3%(36/108). In 50 (46.3%) cases the dose was adjusted according to the initial concentration data. Compared to less than 10 mg/L, the clinical efficacy of trough concentration in 10-20 mg/L was much higher(93.9%(46/49) vs. 78.6%(33/42), P=0.031). Total bilirubin and platelet count had significant difference between the two-matched groups (34.1(14.9, 91.0)μmol/L vs. 53.0(27.0, 121.6)μmol/L, P=0.034; 301.0(198.8, 416.0)×10(9)/L vs. 195.5(94.0, 283.2)×10(9)/L, P=0.000). The incidence of linezolid related thrombocytopenia was 13.4%(28 cases).

CONCLUSION

The clinical effect of vancomycin and linezolid on gram-positive bacterial sepsis in neonates is comparable. The rate of trough concentration of vancomycin reaching the high trough concentration standard is low, and the clinical efficacy is related to trough concentration. Linezolid have an effect on bilirubin and platelet count, the risk of thrombocytopenia should be monitored closely during linezolid treatment.