106 - Association between Overlapping High-Risk Nephrotoxic Antimicrobial Exposure in the NICU and Future Kidney Function in Adolescents Born Preterm with Very Low Birth Weight

Poster Number: 106
Publication Number: 106.125

Andrew Schiff, Wake Forest Baptist Health - Brenner Children's Hospital, Winston-Salem, NC, United States; Andrew M. South, Wake Forest School of Medicine of Wake Forest Baptist Medical Center, Winston Salem, NC, United States; Lisa k. Washburn, Wake Forest School of Medicine of Wake Forest Baptist Medical Center, Winston-Salem, NC, United States

Background: We previously observed a cumulative dose-response relationship between exposure to high-risk nephrotoxic antimicrobials, especially vancomycin, in the neonatal intensive care unit (NICU) and worse kidney function in adolescents born preterm with very low birth weight (VLBW). However, it is unknown whether simultaneous exposure to high-risk antimicrobials further increases this risk.

Objective: Determine the association between simultaneous exposure to high-risk antimicrobials in the NICU and adolescent kidney function.

Design/Methods: Prospective cohort of persons born preterm with VLBW as singletons from 1/1/1992 to 6/30/1996 at a single center and assessed at age 14 years with blood or urine collected. Participants with congenital anomalies, significant health conditions, or missing exposure data were excluded. We recorded presence of any exposure to ≥1 high-risk antimicrobial (vancomycin, gentamicin, tobramycin, and/or amphotericin B) in the NICU and whether participants received ≥2 drugs simultaneously (“overlap”) as a 3-level ordinal variable: unexposed (referent), exposed-no overlap, or exposed-overlap. Serum creatinine and first-morning urine albumin and creatinine were measured; we estimated glomerular filtration rate (eGFR) and calculated albumin/creatinine ratio (ACR), defining albuminuria as ACR >30. We estimated associations with inverse probability censoring weighted, multivariable generalized linear models adjusted for gestational age, birth weight, and birth weight z-score.

Results: Of 161 participants, 55% were female and 41% were Black; 78% had exposure-no overlap and 4% had exposure-overlap (Table 1). Median eGFR was 124.9 ml/min/1.73 m2 [IQR 109.7, 140.8] (n=123) and median ACR 5.3 [3.0, 11.4] (n=133); 7% had albuminuria. Compared to unexposed, exposure to high-risk antimicrobials with or without overlap was not associated with serum creatinine, eGFR, ACR, or albuminuria on adjusted analysis (Table 2).Conclusion(s): Compared to unexposed participants, we did not observe an association among overlapping or non-overlapping exposure to high-risk nephrotoxic antimicrobials in the NICU with worse kidney function in adolescents born preterm with VLBW. To further elucidate long-term kidney risk of exposure to nephrotoxic antimicrobials in the NICU, ongoing analyses include investigating overlapping exposure to both high-risk and lower-risk medications and cumulative duration of overlapping exposure.
Andrew Schiff_CV_070421.pdf
Table 2: Association between overlapping exposure to high-risk nephrotoxic antimicrobials in the neonatal intensive care unit and later kidney function in adolescents born preterm with very low birth weightGeneralized linear models with linear regression (normal distribution, identity link) or Poisson regression (Poisson distribution, log link). Referent group those unexposed to high-risk antimicrobials.
aMinimally sufficient adjustment set included gestational age at birth, birth weight, and birth weight z-score identified by directed acyclic graphs. All models inverse probability censoring weighted for the exposure, race, and antenatal corticosteroids to account for potential selection bias in the excluded participants who did not have exposure data or provide blood or urine samples.
eGFR, estimated glomerular filtration rate; ACR, urine albumin/creatinine ratio; ln, natural-log.