http://informahealthcare.com/jmf ISSN: 1476-7058 (print), 1476-4954 (electronic) J Matern Fetal Neonatal Med, 2014; 27(14): 1485–1490 ! 2014 Informa UK Ltd. DOI: 10.3109/14767058.2013.860522
Nephrotoxic medication exposure in very low birth weight infants Erika T. Rhone1, J. Bryan Carmody2, Jonathan R. Swanson3, and Jennifer R. Charlton1 1
Department of Pediatrics, Division of Nephrology, University of Virginia, Charlottesville, VA, USA, 2Department of Pediatrics, Division of Nephrology, Eastern Virginia Medical School, Norfolk, VA, USA, and 3Department of Pediatrics, Division of Neonatology, University of Virginia, Charlottesville, VA, USA Abstract
Objective: To quantify exposure to potentially nephrotoxic medications among very low birth weight (VLBW) infants and determine the relationship of nephrotoxic medication exposure to acute kidney injury (AKI) in this vulnerable population. Methods: We reviewed 107 VLBW infants who survived to discharge from April 2011 to March 2012 and measured exposure to the following nephrotoxic medications: acyclovir, amikacin, amphotericin B, gentamicin, ibuprofen, indomethacin, iohexol, tobramycin and vancomycin. Acute kidney injury was determined by the KDIGO guidelines. Results: Exposure to 1 nephrotoxic medication occurred in 87% of infants. The most common exposures were gentamicin (86%), indomethacin (43%) and vancomycin (25%). There was an inverse linear relationship between birth weight and nephrotoxic medications received per day (R2 ¼ 0.169, p50.001). Infants with AKI received more nephrotoxic medications per day than those who did not (0.24 versus 0.15; p ¼ 0.003). Conclusions: VLBW infants are frequently exposed to nephrotoxic medications, receiving approximately two weeks of nephrotoxic medications before discharge or 1 for every 6 d of hospitalization. The greatest exposure occurred among the smallest, most immature infants and those who experienced AKI.
Acute kidney injury, aminoglycosides, neonatal intensive care, non-steroidal anti-inflammatory agents, premature infant
Introduction Medications with nephrotoxic properties – such as aminoglycoside antibiotics and non-steroidal anti-inflammatory drugs (NSAIDs) – are commonly administered to infants in the neonatal intensive care unit (NICU). Nephrotoxic medications are a well-known cause of acute kidney injury (AKI)  in both pediatric patients in general [2,3] and preterm infants in particular [4,5]. AKI is now recognized as an independent risk factor for mortality in very low birth weight (VLBW) infants [6,7] and increases the long-term risk for chronic kidney disease [8–11]. Because administration of nephrotoxic medications may be a modifiable factor in the development of AKI, data describing the prevalence of their use within this vulnerable population are needed. Beyond the risk of AKI, nephrotoxic medications may confer additional risks for premature infants. Nephrogenesis is not complete until 36 weeks of gestational age , meaning that infants born prematurely must continue to form nephrons in an ex utero environment. There is reason to believe that this process may be disrupted by nephrotoxic medications. For instance, in animal studies, NSAIDs reduce the size of Address for correspondence: Jennifer R. Charlton, MD, MSc, Department of Pediatrics, Division of Nephrology, University of Virginia, P.O. Box 800386, Charlottesville, VA 22908, USA. Tel: (434) 924-2096. Fax: (434) 924-5505. E-mail: [email protected]
History Received 25 September 2013 Accepted 26 October 2013 Published online 28 November 2013
glomeruli [13,14], while gentamicin disrupts nephron development in vitro [15,16]. In spite of these concerns, the extent of exposure to nephrotoxic medications in VLBW infants remains undescribed. We therefore sought to define the prevalence and extent of exposure to these medications within a level IV NICU population and establish the relationship between such exposures and AKI.
Methods We conducted a retrospective review of all VLBW infants admitted to the University of Virginia’s level IV NICU from 1 April 2011 to 31 March 2012. All infants with birth weight 1500 g were initially included. Patients who did not survive to discharge were excluded from analysis. Infants who were admitted to the NICU at 42 d of age were also excluded since laboratory data and medication administration records prior to transfer were often incomplete or inaccessible. The Institutional Review Board at the University of Virginia approved the study protocol and waived the need for consent. Data collected from accessing the electronic medical record included demographics, gestational age, birth weight, Apgar scores, lab results (serum creatinine and drug levels), length of stay and mortality. Exposure to any of the following potentially nephrotoxic medications was determined by review of the medication administration record: acyclovir,
E. T. Rhone et al.
J Matern Fetal Neonatal Med, 2014; 27(14): 1485–1490
amikacin, amphotericin B, gentamicin, ibuprofen, indomethacin, iohexol, tobramycin and vancomycin. The indications for antimicrobial courses lasting 472 h were documented, and microbiologic cultures were reviewed for each infant. Nephrotoxic medication exposure was quantified as total nephrotoxic medication days, calculated by summing the number of treatment days for all nephrotoxic medications received. To account for the effect of length of hospital stay, mean nephrotoxic medications per day was calculated as total nephrotoxic medication days divided by length of stay in days. Acute kidney injury was classified using the Kidney Disease: Improving Global Outcomes (KDIGO) system , modified for use in neonates as has been previously proposed . Patients with a rise in serum creatinine of 0.3 mg/dL occurring within 48 h from a previous trough or who had a rise in serum creatinine of 150–199% of a previous trough were classified as stage 1 AKI, while patients who experienced a rise in creatinine of 200–299% of previous trough were classified as stage 2 AKI. Patients with an increase in serum creatinine to 300% of a previous trough, a peak of over 2.5 mg/dL, or the need for renal replacement therapy were classified as stage 3 AKI. Measurements of creatinine were performed in the University of Virginia laboratory using an alkaline picrate method traceable to isotope dilution mass spectrometry. Classification of infants as small (SGA), appropriate, and large for gestational age was performed using previously reported reference standards .
Statistical analysis Continuous variables were compared using the Mann– Whitney U or Student’s t-test and categorical variables were compared using 2 or Fisher’s exact tests as appropriate. Linear regression was used to determine the association between factors of interest and continuous outcomes, while logistic regression was used for binary outcomes. Multivariable models were created to adjust for confounding factors. All statistical analyses were performed using IBM SPSSÕ version 20 (Armonk, NY). A two-sided significance level of 0.05 was set for all tests.
Results Study population One hundred and thirty VLBW infants were admitted during the study period. Of these, 10 were excluded due to admission at42 d old and 13 did not survive to discharge. The remaining 107 infants were analyzed further. Patient demographics are displayed in Table 1. Exposure to nephrotoxic medications Exposure to at least one nephrotoxic medication occurred in 93 (86.9%) infants. The mean number of nephrotoxic medications received was 1.64 (SD 1.08), with a median of 2. Exposure to individual medications is shown in Table 2. Compared to infants who did not receive nephrotoxic medications, patients who received nephrotoxic medications had lower gestational age and birth weight, longer length of stay, and were less likely to be classified as SGA (p50.001). Acute kidney injury
Table 1. Demographic information for the study population. Study population (n ¼ 107)
Characteristic Gestational age (weeks) Birth weight (g) Length of stay (d) Sex Race
Small for gestational age (SGA) AKI Exposure to any nephrotoxin
28.0 (SD 3.0, range 23–37) 1049 (SD 266, range 440–1475) 68.0 (SD 33.8, range 13–186) 52 (48%) male 70 white (65%) 23 black (21%) 11 Hispanic/Latino (10%) 3 other (3%) 23 (21.5%) 28 (26.2%) 93 (86.9%)
AKI occurred in 28 (26.2%) infants. Of these, 21 (75%) reached stage 1 AKI, 6 (21.4%) reached stage 2 and 1 (3.6%) reached stage 3. No infant received dialysis. All infants who experienced AKI received at least one nephrotoxic medication; infants who did not receive nephrotoxic medications did not experience AKI. The 41 discrete episodes of AKI among these 28 patients were analyzed to determine the timing of nephrotoxic medication administration. Overall, 47% of nephrotoxic medications were administered prior to AKI episodes and 53% of nephrotoxic medication exposure occurred after the creatinine had peaked.
Table 2. Exposure to individual nephrotoxins among VLBW infants. Drug
Amphotericin Gentamicin Indomethacin Tobramycin Vancomycin
4 (4.3%) 92 (86.0%) 46 (50.0%) 3 (3.3%) 27 (29.3%)
Mean (SD) Days Cumulative Days Cumulative Days Cumulative Days Cumulative Days Cumulative Days Cumulative
dose dose dose dose dose dose
9.3 d 486.7 mg/kg 6d 6 mg/kg 10.3 d 32.4 mg/kg 4.7 d 1.0 mg/kg 5.7 d 22.3 mg/kg 8.2 d 190.6 mg/kg
(11.0) (637.9) (3.3) (3.3) (8.9) (31.3) (2.3) (0.76) (2.3) (13.6) (6.3) (153.6)
4d 180 mg/kg 6d 6 mg/kg 8d 20 mg/kg 3.5 d 0.7 mg/kg 7d 24 mg/kg 7d 135 mg/kg
2–22 d 60–1220 mg/kg 2–10 d 2–10 mg/kg 1–39 d 2–163 mg/kg 2–10 d 0.3–2.6 mg/kg 3–7 d 8–35 mg/kg 2–21 d 10–498 mg/kg
1 (33.3%) 2 (50.0%) 27 (29.3%) 22 (47.8%) 1 (33.3%) 12 (44.4%)
Nephrotoxins and VLBW infants
Total nephrotoxic medication days For the entire dataset, mean total nephrotoxin days was 13.6 (SD 14.5) with median 8 d (IQR 3–21). For infants who experienced AKI, the mean was 23.9 d (SD 16.8) with median 20.5 d (IQR 13.25–36.25). Infants who experienced AKI had greater total nephrotoxic medication days than infants who did not (23.9 d versus 9.9 d; p50.001, Mann– Whitney U-test). In logistic regression models, total nephrotoxic medication days were associated with occurrence of AKI (p50.001). However, in multivariate models including birth weight or gestational age, the association with total nephrotoxic medications days was non-significant. For infants who experienced AKI, there was a linear association between total nephrotoxic medication days and peak creatinine (mg/dL) with AKI ( ¼ 0.006, 95% CI ¼ 0.001–0.011; R2 ¼ 0.207; p ¼ 0.015). This association persisted in multivariable models accounting for birth weight or gestational age. Mean total nephrotoxic medication days varied by birth weight category (p50.001, one-way ANOVA) (Figure 1). Using linear regression, there was an inverse relationship between birth weight (g) and total nephrotoxin days ( ¼ 0.035, 95% CI ¼ 0.043 to 0.027; R2 ¼ 0.405; p50.001). Mean nephrotoxic medications per day In the entire dataset, mean nephrotoxic medications per day was 0.17 (SD 0.15), with median 0.15 (IQR 0.06–0.24). Infants who experienced AKI had higher mean nephrotoxic medications per day than infants who did not (0.24 versus 0.15; p ¼ 0.003, Mann–Whitney U-test). There was an inverse linear relationship between birth weight and mean nephrotoxic medications received per day ( ¼ 0.231, R2 ¼ 0.169, p50.001) (Figure 2). In logistic regression models, nephrotoxic medications per day was associated with occurrence of AKI (p ¼ 0.006). However, in multivariate models including birth weight or gestational age, Figure 1. Mean total nephrotoxic medication days varied by birth weight category, with smaller infants receiving more days of nephrotoxic medication (p50.001, one-way ANOVA).
the association with nephrotoxic medications per day was non-significant. Among infants who experienced AKI, there was a linear association between mean nephrotoxic medications per day and peak creatinine (mg/dL) with AKI ( ¼ 0.821, 95% CI ¼ 0.360–1.283; R2 ¼ 0.340; p ¼ 0.001) (Figure 3). This persisted after accounting for the effect of birth weight and gestational age ( ¼ 0.765, 95% CI ¼ 0.285–1.244; R2 ¼ 0.378; p ¼ 0.003). Drug level monitoring Fifteen percent (14/92) of infants who received gentamicin and 7% (2/27) of infants who received vancomycin had an elevated trough level of these medications at least once during their hospital course. There was no association between the occurrence of AKI and elevated medication trough levels.
Discussion While nephrotoxic medications are commonly used in the NICU, the prevalence and extent of such postnatal exposure among at-risk VLBW infants has never been described. This is the first study to not only determine the most frequent types of nephrotoxic exposures in this population, but also to quantify exposure by calculating cumulative exposure to nephrotoxic medication. We have shown that exposure to nephrotoxic medications in a regional tertiary care NICU is nearly universal, with 86.9% of VLBW infants receiving at least one nephrotoxic medication prior to discharge. The duration of exposure was substantial, with the average VLBW infant receiving 13.6 total days of nephrotoxic medications, or nearly 2 weeks of therapy. After accounting for length of stay, the mean nephrotoxic medications received per day was 0.17, or one nephrotoxic medication received for every 6 d spent in the NICU. These results are remarkably similar to those found in a large single center study of non-critically ill hospitalized pediatric patients, in which 86% of patients received at least one nephrotoxic medication with a median of 9 d of exposure
E. T. Rhone et al.
J Matern Fetal Neonatal Med, 2014; 27(14): 1485–1490
Figure 2. Inverse linear association between birth weight in grams and nephrotoxic medications received per day ( ¼ 0.231, R2 ¼ 0.169, p50.001).
Figure 3. Association between nephrotoxic medications per day and peak creatinine with AKI for the 27 infants who experienced AKI ( ¼ 0.821, R2 ¼ 0.340; p ¼ 0.001).
. However, there is a cause for concern that the use of nephrotoxic medications in the VLBW population may be associated with greater long-term risks than among other patient populations. In the United States, VLBW infants compose 1.5% of all the births . Because nearly 90% of these infants now survive to hospital discharge , there is an increasing clinical focus on avoiding morbidity and improving long-term outcomes. Unfortunately, both low birth weight and premature infants who survive the NICU do so with a decreased number of nephrons, which places them at a higher long-term risk for developing chronic kidney disease than the general population . While infants with intrauterine growth restriction are born with fewer functioning nephrons , premature infants have not completed nephrogenesis at the time of birth and must continue nephrogenesis in the ex utero environment. In either case, there is reasonable concern that treatment with nephrotoxic medications could cause a further decrease in nephron number. Experimental data have shown a clear association between nephrotoxic medication exposure and impaired nephrogenesis. NSAID exposure disrupts glomerulogenesis in mice and rats, reducing the size of glomeruli [13,14], while premature baboons exposed to ibuprofen have a smaller nephrogenic zone, potentially indicating cessation of nephrogenesis . Gentamicin affects nephron development in in vitro studies of metanephric organ culture resulting in reduced ureteric bud branching and glomerular number [15,16], while rats exposed to gentamicin prenatally have 20% fewer glomeruli and develop glomerulosclerosis as they age . These studies may have particular relevance in our population, where the most common nephrotoxic medication exposures were gentamicin and indomethacin. Forty-two percent of the study population received both of these medications, and cumulative doses were often large, with total doses of up to 163 mg/kg (39 d) of gentamicin and 2.6 mg/kg (10 d) of indomethacin. Exposure to nephrotoxic medications was not equally distributed among VLBW infants: the highest exposures were clustered among the infants with the lowest birth weights and gestational ages, many of whom also experienced AKI. The implications of this may be significant. For example, infants with birth weight 5500 g received an average of 45 total nephrotoxic medication days, or approximately 6 weeks. Data from autopsy studies suggests that postnatal nephrogenesis ceases after 40 d , meaning that nephrotoxic medication exposure could occur throughout this critical period. Retrospectively, we could not distinguish AKI caused by nephrotoxic medication exposure from the condition for which the nephrotoxic medications were given. Yet even in a prospective trial, such a distinction might not be able to be reliably made. For instance, an infant with sepsis or a patent ductus arteriosus may have kidneys that are ‘‘primed’’ for AKI with the addition of a nephrotoxic medication due to preexisting hypoperfusion or increased oxidative stress. Our intention was not to assess whether nephrotoxic medication administration in VLBW infants independently causes AKI, but rather to make a risk assessment, and believe that it is noteworthy that nephrotoxic medication administration and
Nephrotoxins and VLBW infants
AKI were so strongly associated. Whether increased nephrotoxic medication administration actually caused AKI or simply occurred disproportionately in infants that had already experienced AKI for some other reason may not change the long-term implications for these infants. In this study, all nephrotoxic medications were treated equally for statistical purposes. In reality, the nephrotoxic medications we studied may have differing toxicities and likely confer different short- and long-term risks. However, we believe that calculating total nephrotoxic medication days provides useful information by providing a total estimate of renal risk. Every day that a practitioner prescribes a potential nephrotoxin, there is an element of risk involved; whether the potential benefit justifies that risk must be determined on a case-by-case basis. Recognizing this risk has important clinical implications. The use of an electronic medical record to identify patients at high risk for nephrotoxic medication-associated AKI was associated with a reduction in AKI frequency per 100 exposure days , and multidisciplinary quality improvement processes to reduce the use and mitigate the risk of nephrotoxic medications  have also been proposed . Although our study was not designed to evaluate the clinical appropriateness of each nephrotoxic medication administered, our data suggest that a substantial portion of nephrotoxic medication administration may be avoidable, particularly with regard to antimicrobial agents. In our study population, five of the six most commonly administered nephrotoxic medications were antimicrobials, and 52.3% (56/107) of subjects received at least one course of antimicrobials lasting 472 h. In many cases, the indications for antimicrobials were clear-cut: nearly 16% of VLBW infants studied had a positive blood or urine culture during their hospitalization. However, empiric antibiotic use – even for prolonged treatment courses – was common, and 40.2% (43/107) received at least one course of empiric antimicrobials lasting 472 h for presumed septicemia without positive culture results or focal signs of infection. Some of these patients may have had culture-negative sepsis, yet it is plausible that many of these patients could have safely avoided treatment with antimicrobials in general or nephrotoxic antimicrobials in particular. Inappropriate antibiotic use in the NICU is known to be common: in one study, 4% of antibiotic courses were initiated inappropriately, and 25% of all appropriately-initiated antibiotic courses were inappropriately continued beyond 72 h . It seems likely that some of these nephrotoxic medication exposures could have been avoided, and that population exposure to nephrotoxic medications could be reduced with improved antimicrobial stewardship. In conclusion, exposure to potentially nephrotoxic medications in VLBW infants is common and clusters in a small group of high-risk infants. Almost 90% of all NICU survivors received at least one nephrotoxic medication, with the greatest nephrotoxic medication exposure occurring in the smallest, most immature infants and those who experienced AKI. Better antimicrobial stewardship may reduce exposure to nephrotoxic medication in this population. Further study of the acute and long-term effects of nephrotoxic medications in this population is warranted.
E. T. Rhone et al.
Declaration of interest The authors report no declarations of interest.
References 1. Pannu N, Nadim MK. An overview of drug-induced acute kidney injury. Crit Care Med 2008;36:S216–23. 2. Hui-Stickle S, Brewer ED, Goldstein SL. Pediatric ARF epidemiology at a tertiary care center from 1999 to 2001. Am J Kidney Dis 2005;45:96–101. 3. Misurac JM, Knoderer CA, Leiser JD, et al. Nonsteroidal antiinflammatory drugs are an important cause of acute kidney injury in children. J Pediatr 2013;162:1153,9, 1159.e1. 4. Cataldi L, Leone R, Moretti U, et al. Potential risk factors for the development of acute renal failure in preterm newborn infants: a case-control study. Arch Dis Child Fetal Neonatal Ed 2005;90: F514–19. 5. Cuzzolin L, Fanos V, Pinna B, et al. Postnatal renal function in preterm newborns: A role of diseases, drugs and therapeutic interventions. Pediatr Nephrol 2006;21:931–8. 6. Askenazi DJ, Griffin R, McGwin G, et al. Acute kidney injury is independently associated with mortality in very low birthweight infants: a matched case-control analysis. Pediatr Nephrol 2009;24: 991–7. 7. Koralkar R, Ambalavanan N, Levitan EB, et al. Acute kidney injury reduces survival in very low birth weight infants. Pediatr Res 2011; 69:354–8. 8. Viaud M, Llanas B, Harambat J. Renal outcome in long-term survivors from severe acute kidney injury in childhood. Pediatr Nephrol 2012;27:151,2; author reply 153. 9. Askenazi DJ, Feig DI, Graham NM, et al. 3-5 year longitudinal follow-up of pediatric patients after acute renal failure. Kidney Int 2006;69:184–9. 10. Abitbol CL, Bauer CR, Montane B, et al. Long-term follow-up of extremely low birth weight infants with neonatal renal failure. Pediatr Nephrol 2003;18:887–93. 11. Zaffanello M, Bassareo PP, Cataldi L, et al. Long-term effects of neonatal drugs on the kidney. J Matern Fetal Neonatal Med 2010; 23:87–9. 12. Osathanondh V, Potter E. Development of human kidney as shown by microdissection. III. Formation and interrelationship of collecting tubules and nephrons. Arch Pathol 1963;76:290–302. 13. Akinola O, Noronha C, Oremosu A, et al. The effect of the cyclooxygenase blockers, ibuprofen on the development of glomeruli in sprague-dawley rats. Niger Postgrad Med J 2003;10: 46–50.
J Matern Fetal Neonatal Med, 2014; 27(14): 1485–1490
14. Komhoff M, Wang JL, Cheng HF, et al. Cyclooxygenase2-selective inhibitors impair glomerulogenesis and renal cortical development. Kidney Int 2000;57:414–22. 15. Gilbert T, Gaonach S, Moreau E, Merlet-Benichou C. Defect of nephrogenesis induced by gentamicin in rat metanephric organ culture. Lab Invest 1994;70:656–66. 16. Gilbert T, Cibert C, Moreau E, et al. Early defect in branching morphogenesis of the ureteric bud in induced nephron deficit. Kidney Int 1996;50:783–95. 17. Kidney disease: Improving global outcomes (KDIGO) acute kidney injury work group. KDIGO clinical practice guidelinefor acute kidney injury. Kidney Inter Suppl 2012;2:1–138. 18. Jetton JG, Askenazi DJ. Update on acute kidney injury in the neonate. Curr Opin Pediatr 2012;24:191–6. 19. Alexander GR, Himes JH, Kaufman RB, et al. A United States national reference for fetal growth. Obstet Gynecol 1996;87:163–8. 20. Moffett BS, Goldstein SL. Acute kidney injury and increasing nephrotoxic-medication exposure in noncritically-ill children. Clin J Am Soc Nephrol 2011;6:856–63. 21. Eichenwald EC, Stark AR. Management and outcomes of very low birth weight. N Engl J Med 2008;358:1700–11. 22. Horbar JD, Carpenter JH, Badger GJ, et al. Mortality and neonatal morbidity among infants 501 to 1500 grams from 2000 to 2009. Pediatrics 2012;129:1019–26. 23. White SL, Perkovic V, Cass A, et al. Is low birth weight an antecedent of CKD in later life? A systematic review of observational studies. Am J Kidney Dis 2009;54:248–61. 24. Hinchliffe SA, Lynch MR, Sargent PH, et al. The effect of intrauterine growth retardation on the development of renal nephrons. Br J Obstet Gynaecol 1992;99:296–301. 25. Sutherland MR, Yoder BA, McCurnin D, et al. Effects of ibuprofen treatment on the developing preterm baboon kidney. Am J Physiol Renal Physiol 2012;302:F1286–92. 26. Gilbert T, Lelievre-Pegorier M, Merlet-Benichou C. Long-term effects of mild oligonephronia induced in utero by gentamicin in the rat. Pediatr Res 1991;30:450–6. 27. Rodriguez MM, Gomez A, Abitbol C, et al. Comparative renal histomorphometry: a case study of oligonephropathy of prematurity. Pediatr Nephrol 2005;20:945–9. 28. Goldstein SL, Kirkendall E, Nguyen H, et al. Electronic health record identification of nephrotoxin exposure and associated acute kidney injury. Pediatrics 2013;132:e756–67. 29. Zappitelli M, Selewski D, Askenazi D. Nephrotoxic medication exposure and acute kidney injury in neonates. NeoRev 2012;13: e420–7. 30. Patel SJ, Oshodi A, Prasad P, et al. Antibiotic use in neonatal intensive care units and adherence with centers for disease control and prevention 12 step campaign to prevent antimicrobial resistance. Pediatr Infect Dis J 2009;28:1047–51.
Copyright of Journal of Maternal-Fetal & Neonatal Medicine is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.