A propensity-matched cohort study of vancomycin-associated nephrotoxicity in neonates

Con­stance et al, Arch Dis Child Fetal Neona­tal Ed. 2015 Sep 23. 

BACKGROUND:

The inci­dence of nephro­tox­i­c­i­ty among van­comycin-treat­ed neonates has been report­ed to range from 2% to 20%. These wide­ly vary­ing esti­mates have led to con­fu­sion and con­tro­ver­sy regard­ing the safe­ty of van­comycin among neonates.

OBJECTIVE:

Eval­u­ate the inci­dence of nephro­tox­i­c­i­ty among neonates receiv­ing van­comycin con­comi­tant­ly with gen­tam­icin.

DESIGN:

Ret­ro­spec­tive obser­va­tion­al cohort study using propen­si­ty score match­ing to pro­vide covari­ate bal­ance between neonates who did or did not receive van­comycin based on fac­tors known to be relat­ed to the devel­op­ment of renal dys­func­tion.

SETTING:

Hos­pi­tals (n=22) through­out the Inter­moun­tain West, includ­ing a qua­ter­nary care children’s hos­pi­tal.

PATIENTS:

Neonates ≤44 post­men­stru­al weeks (medi­an ges­ta­tion­al age: 31 (IQR 28–36) weeks) receiv­ing intra­venous gen­tam­icin with or with­out expo­sure to van­comycin from Jan­u­ary 2006 to Decem­ber 2012.

RESULTS:

The final cohort was com­prised of 1066 neonates (533 receiv­ing van­comycin and gen­tam­icin vs 533 receiv­ing gen­tam­icin). In a propen­si­ty score-matched cohort that was well bal­anced across 16 covari­ates, AKI was not asso­ci­at­ed with van­comycin use (16 neonates receiv­ing van­comycin vs 7 con­trols expe­ri­enced AKI; OR 1.5; 95% CI 0.6 to 4.0). How­ev­er, the pres­ence of a patent duc­tus arte­rio­sus, con­comi­tant non-steroidal anti-inflam­ma­to­ry drug use, ≥1 pos­i­tive blood cul­tures, low birth weight and high­er sever­i­ty of ill­ness and risk of mor­tal­i­ty scores were asso­ci­at­ed with an increased risk of nephro­tox­i­c­i­ty.

CONCLUSIONS:

These results cor­rob­o­rate sev­er­al ear­li­er reports and much anec­do­tal evi­dence describ­ing the infre­quent occur­rence of nephro­tox­i­c­i­ty in neonates receiv­ing con­comi­tant van­comycin and gen­tam­icin.