Invasive fungal infections are associated with significant morbidity and mortality. In the United States, Candida species are the most common fungal pathogens causing invasive fungal infections and are the 4th most common cause of hospital-associated bloodstream infections.1 children with candidemia and invasive candidiasis, but can be associated with significant drug-related toxicity (DRT).
In adult studies, the incidence of AmB-related nephrotoxicity (NT) ranged from 16% to 80% and infusion-related toxicity (INFRT) occurred in up to 73.6% of patients.2–6 children with fever and neutropenia.7–9 children .10 children receiving AmB colloidal dispersion.7 , 11 , 12 children . Children tend to tolerate AmB formulations better than adults because the elimination half-life of AmB varies inversely with age, and the volume of distribution in children is smaller than that in adults; therefore, children have improved renal clearance of AmB.13–15 7 children , specifically in patients with fever and neutropenia, have shown that the treatment efficacy of echinocandins and AmB is similar.7 , 16–18 children , and it is difficult to predict which patients will develop DRT.
Risk factors associated with NT in adults have included male sex, obesity, preexisting renal disease, concomitant use of nephrotoxic drugs and cumulative dose of AmB; however, substantial patient heterogeneity has been observed in these studies.4 , 19–21 19 19 children are limited.
It is important to identify risk factors for DRT with AmB in children to enable clinicians to determine the optimal empirical antifungal therapy and prevent DRT if the risk factors are modifiable. Therefore, the primary aim of our study was to determine the incidence and risk factors for the most common AmB DRT in children , namely NT, INFRT and hypokalemia. The secondary aim was to evaluate whether the incidence of DRT differed in patients who received conventional AmB (C-AmB) versus lipid formulations of AmB.
MATERIALS AND METHODS
Study Design and Setting
We conducted a retrospective case-control study from January 2003 through December 2009 at Texas Children ’s Hospital, Houston, TX, a 639-registered-bed tertiary-care center. Patients aged 6 months (full term) to ≤18 years with bloodstream infection caused by Candida species were identified from the Pediatric Infectious Diseases database, international classification of diseases, 9th revision billing codes and the Microbiology Laboratory Database. Preterm infants who were ≤36 weeks gestation at birth were included if they were ≥2 years of age. Neonates, including preterm infants who were ≤36 weeks and <2 years of age and children <6 months of age, were excluded from the study because their pathophysiologic basis for invasive fungal infections may differ from older children . Patients who received at least 1 dose of AmB for the treatment of candidemia were included in the study. This study was approved by the Baylor College of Medicine Institutional Review Board.
The choice of AmB agents used was based on instituitional guidelines. In general, immunocompromised patients with fever requiring treatment for invasive fungal diseases, patients with baseline renal insufficiency who developed NT while receiving C-AmB, despite volume expansion and sodium loading, and those with INFRT on C-AmB, received lipid formulations. For C-AmB, subjects generally received 0.5 mg/kg /dose on the first day and then 1 mg/kg/dose thereafter. For lipid formulations of amphotericin, subjects received full doses (3–5 mg/kg/d) starting on day 1 of treatment.
Case Definitions
Cases were defined as patients with candidemia who developed DRT with at least 1 dose of AmB. Controls were unmatched and were defined as patients with candidemia who did not develop DRT with AmB. A case of candidemia was defined as ≥1 positive blood culture obtained by a peripheral venipuncture or through an indwelling central venous catheter that yielded growth of any Candida species. NT was evaluated by the pediatric Risk, Injury, Failure, Loss, End-stage Kidney disease (pRIFLE) criteria22 23
Data Collection
Information regarding patient demographics, underlying medical conditions and surgical procedures were recorded from medical charts. Risk factors evaluated included: receipt of nephrotoxic medications within 30 days before or concomitant with AmB; total parenteral nutrition, blood transfusion, stay in the intensive care unit, mechanical ventilation, use of vasopressors and neutropenia (absolute neutrophil count <500/mm3 ) if present within 2 weeks before the onset of candidemia. The use of acetaminophen, diphenhydramine, meperidine and prednisone as premedications or administered secondary to AmB INFRT was documented. The exact timing or dosage of premedications was not documented. The use of normal saline (NS) bolus before AmB infusion and intravenous fluids while receiving AmB was recorded. If AmB was discontinued (including changes from C-AmB to the lipid formulations) secondary to DRT, the reason for discontinuation was noted. The duration and cumulative dose of AmB was obtained from pharmacy records.
Statistical Analysis
Statistical analyses were performed with STATA version 11.0 (STATA Corporation, College Station, TX). Baseline characteristics, comorbidities and risk factors were compared between the groups (those with AmB DRT versus those without AmB DRT) by univariate analyses. Continuous variables were analyzed by using the paired sample Wilcoxon rank sum test. Categorical variables were analyzed using the χ2 t test. Factors associated with NT, hypokalemia and INFRT also were evaluated by univariate analyses. All variables with P < 0.2 were included in the multivariate logistic regression model to predict independent risk factors associated with each of the above categories. A 2-tailed P ≤ 0.05 was considered significant for all statistical tests.
RESULTS
A total of 223 episodes of candidemia occurred in 179 patients. AmB was administered in 172 (77%) episodes in 138 patients. C-AmB was administered in 65 (38%) episodes, whereas ABLC and L-AmB were given in 96 (55%) and 11 (6.4%) episodes, respectively. When the first episode of AmB use was analyzed separately (n = 138), DRT occurred in 83% (n = 114). Among the first episodes, ABLC was most commonly used (n = 75, 54.4 %) followed by C-AmB (n = 56, 40.6%) and L-AmB (n = 7, 5%).
Patient Characteristics
Complete demographic information was available for all patients. The median age of the patients was 43 months. There was no difference in age, sex, ethnicity or underlying diagnoses among patients with and without AmB DRT (Table, Supplemental Digital Content 1, https://links.lww.com/INF/B232
Evaluation of NT
NT occurred in 45% (n = 62) of patients. As per the pRIFLE criteria, 9% (n = 12), 16% (n = 22) and 20% (n = 28) of patients were in the Failure, Injury and Risk categories, respectively. The median baseline Crcl in patients with NT was higher than that in patients without NT (P = 0.004). There was no difference in demographics among patients with and without NT from AmB products. The use of immunosuppressants before receipt of AmB was associated with a lower likelihood of NT (P = 0.02). The median lowest Crcl in patients with NT was 1.3 times lower than that in patients without NT (P < 0.0001). In patients with NT, the mean and median day to reach the lowest Crcl was 7.9 (+/− 5) and 6 (range, 2–22) days, respectively. Crcl 1 week posttherapy was not different between patients with and without AmB DRT (Table, Supplemental Digital Content 2, https://links.lww.com/INF/B233
When C-AmB was compared with ABLC, the former was more likely to cause NT (55.3%, n = 31 versus 37.3%, n = 28; P = 0.05). When L-AmB (42.8%, n = 3) was compared with C-AmB or ABLC, no statistical difference in NT was observed (P = 0.69 and P = 0.79, respectively).
Three patients required hemodialysis secondary to renal failure while receiving AmB products; these cases were thought to be multifactorial and not related to AmB use alone. Two of these patients were receiving ABLC and 1 patient, C-AmB. Two patients died due to causes unrelated to candidemia. Five patients were changed from C- AmB to ABLC, 2 patients from C-AmB to L-AmB and 1 patient from C-AmB to fluconazole secondary to NT.
Evaluation of INFRT
Thirty-one percent (n = 41) of patients developed INFRT. The most common manifestation was chills and rigors (80%, n = 33) followed by fever (31.7%, n = 13), hypotension (9.7%, n = 4), hypertension (4.8%, n = 2), bronchospasm (4.8%, n = 2), rash (2.4%, n = 1) and skin necrosis (2.4%, n = 1). Seventy-two (57.6%) of 125 patients with available data received premedications. Of the 53 patients who did not receive premedications, 49% (n = 26) received medication after INFRT occurred. Acetaminophen was used in 79% (n = 91), diphenhydramine in 69% (n = 79), meperidine in 33% (n = 37) and hydrocortisone in 12% (n = 15) of patients (Table 1 ).
TABLE 1: Univariate and Multivariate Analysis of Risk Factors for Amphotericin B Infusion-related Toxicity
The receipt of premedications was independently associated with a lower likelihood of INFRT (P ≤ 0.0001). There was no difference in receipt of premedications in patients who received C-AmB versus lipid formulations (P = 0.96). When adjusted for premedication use, ABLC (n = 8, 67%) was more likely to cause INFRT than C-AmB (n = 4, 33%). L-AmB was not associated with INFRT when premedications were used (n = 3). Two patients had INFRT with L-AmB when premedications were not used.
Antifungal therapy was changed in 10 patients secondary to INFRT. Four patients initially receiving ABLC were changed to fluconazole (n = 3) or micafungin (n = 1). Five patients on C-AmB were changed to ABLC (n = 2), fluconazole (n = 2) and L-AmB (n = 1), and 1 patient on L-AmB was changed to fluconazole.
Evaluation of hypokalemia
Data on serum potassium concentrations could be obtained in 133 patients. Forty-seven percent (n = 62) had hypokalemia, of which 15% (n = 20) had grade 3 and 32% (n = 42) had grade 2 abnormalities. In the multivariate analysis, neutropenia and prior hypokalemia were independently associated with hypokalemia. The incidence of hypokalemia was similar in patients who received C-AmB (44.4%, n = 24) and ABLC (52.8%, n = 38; P = 0.35) and less in the L-AmB group (n = 0; P = 0.04; Table 2 ).
TABLE 2: Univariate and Multivariate Analysis of Risk Factors for Amphotericin B-induced Hypokalemia
Twelve patients required a change in their antifungal therapy secondary to hypokalemia. Nine of these patients were receiving C-AmB (5 were changed to ABLC, 1 to L-AmB, 3 to fluconazole) and 3 were receiving ABLC (who were changed to either fluconazole or L-AmB).
DRT in Recurrent candidemia
Patients with recurrent candidemia were not more likely to develop NT, INFRT or hypokalemia in subsequent episodes compared with their first episodes when AmB products were used (data not shown).
DISCUSSION
Our data suggest that AmB DRT is common in children with candidemia. NT occurred in 45% of patients, although renal failure occurred in 9%. INFRT was observed in 31% and hypokalemia in 47% of the patients. Although multiple studies have evaluated the risk factors for NT in adults, few studies have been performed in children . This is the largest study in children to determine the incidence and risk factors for AmB DRT.
NT in adults has been shown to be associated with male sex,24 25 3 , 25 children , cyclosporine, nystatin and ciprofloxacin have been associated with NT when used concomitantly with AmB.4 children with higher baseline Crcl are at risk of developing NT, which is consistent with adult data,25 , 26 27
Our rate of NT is higher than that previously reported in children , which may be related to our definition of NT using pRIFLE criteria. Studies performed in children have defined NT as ≥100% rise in creatinine, ≥50% reduction in Crcl or an increase of at least 1 mg/dL in serum creatinine from baseline.8 , 9 , 28 children at risk of developing NT and provide an opportunity for earlier intervention. We found that the Crcl 1 week posttherapy with AmB was not different in children who developed NT and those who did not, indicating that AmB NT is quickly reversible, likely due to the more rapid renal clearance of AmB in children compared with adults.
It is not surprising, and is consistent with other studies, that NT is less likely in patients receiving ABLC than C-AmB. We are unable to comment on L-AmB–related NT due to the small number of patients who received L-AmB in our study. From late 2008 to early 2009, L-AmB use in our hospital increased significantly, likely related to the increased complexity of our patient population including the increase in the number of immunocompromised children in addition to a hospital pharmacy formulary change from ABLC to L-AmB; thus; it would be interesting to determine the incidence and risk factors for DRT with L-AmB with a larger sample size.
The incidence of AmB-related INFRT and its risk factors are largely unknown in children . Fever and chills have been reported in 44–54% of adults with C-AmB and 17–18% with L-AmB.6 children have varied significantly, ranging from 1–2% to 95%.6 , 8 16 children has not been systematically studied or reported. More than half of our patients received premedications. A study performed in adults showed no effect of premedications on the prevention of AmB-related INFRT and recommended against the routine use of preventive premedications.29 30 31 32 children .
Hypokalemia is a common side effect of AmB infusion secondary to renal salt wasting, more commonly associated with C-AmB versus lipid formulations of AmB. A significantly large number of patients (47%) had hypokalemia in our study, with grade 3 hypokalemia occurring in 15%. Neutropenia and prior potassium imbalance were independently associated with hypokalemia. Electrolyte abnormalities (hypokalemia [up to 48%], hyponatremia [up to 68%] and hypomagnesemia [54%]) have been shown to be common in patients with fever and neutropenia, though the mechanism is unknown.33
There are several limitations to our study. Due its retrospective nature, few patients had follow-up serum creatinine values after the cessation of AmB, which may have led to misclassification bias in identifying patients with renal insufficiency. A temporal relation of AmB and NT could not be ascertained with certainty with a retrospective study because some cases of NT may reflect evolving candidemia as well. We also did not include the concurrent use of diuretics and other nephrotoxic drugs such as nonsteroidal antiinflammatory agents, which might have contributed significantly to NT and electrolyte abnormalities. We did not evaluate for Loss and End-stage Kidney Disease, and therefore, cannot comment on the long-term effects of AmB use in children . In addition, the accuracy of the data relied on adequate documentation by the managing physician, especially those related to symptoms of INFRT. The infusion rates of AmB and the timing of administration of premedications were not documented. Additionally, the use of electrolyte supplements or electrolyte supplementation through total parenteral nutrition was not accounted for, which may have confounded the results of hypokalemia. Furthermore, a causal relationship between AmB and toxicity cannot be made with certainty without a prospective study.
Although the incidence of DRT was high among AmB recipients, most DRT was reversible and not life-threatening. Furthermore, patients who developed DRT in 1 episode were not more likely to develop similar DRT in subsequent episodes, which suggests that some risk factors are modifiable. Using the pRIFLE criteria might identify the at-risk patients earlier. Routine use of electrolyte supplementation and premedications may reduce the incidence of INFRT and hypokalemia in children . Only a few of our patients received NS boluses before AmB infusion; therefore, we were not able to evaluate this intervention in terms of its ability to decrease the incidence of AmB DRT. Prospective studies evaluating the routine use of premedications, electrolyte supplementation and NS bolus before AmB infusion are needed to evaluate the effect of these measures on AmB-related DRT in children . This need is especially relevant for children receiving C-AmB as the drug of choice for urinary tract and central nervous system candidiasis.
Although most of the literature in children and adults about AmB toxicity has been focused on patients with febrile neutropenia, our study population was heterogeneous with varied underlying diagnoses and included both immunocompromised and immunocompetent patients. Although there are some studies in children about risk factors for NT with AmB, this is the first study, to our knowledge, that has sought to evaluate overall risk factors for AmB DRT, including INFRT and hypokalemia in addition to NT. Host and pharmacogenomic factors, in addition, probably increase susceptibility to AmB DRT, and future studies are needed to ascertain that possibility.
REFERENCES
1. Centers for Disease Control and Prevention. . National Nosocomial Infections Surveillance (NNIS) system report, data summary from January 1992 to June 2001. Am J Infect Control.. 2001;29:404–421
2. Pathak A, Pien FD, Carvalho L. Amphotericin B use in a community hospital, with special emphasis on side effects. Clin Infect Dis. 1998;26:334–338
3. Luber AD, Maa L, Lam M, et al. Risk factors for amphotericin B-induced nephrotoxicity. J Antimicrob Chemother. 1999;43:267–271
4. Goldman RD, Ong M, Wolpin J, et al. Pharmacological risk factors for amphotericin B nephrotoxicity in
children . J Clin Pharmacol. 2007;47:1049–1054
5. Clements JS Jr, Peacock JE Jr. Amphotericin B revisited: reassessment of toxicity. Am J Med. 1990;88(5N):22N–27N
6. Walsh TJ, Finberg RW, Arndt C, et al. Liposomal amphotericin B for empirical therapy in patients with persistent fever and neutropenia. National Institute of Allergy and Infectious Diseases Mycoses Study Group. N Engl J Med. 1999;340:764–771
7. Maertens JA, Madero L, Reilly AF, et al.Caspofungin Pediatric Study Group. A randomized, double-blind, multicenter study of caspofungin versus liposomal amphotericin B for empiric antifungal therapy in pediatric patients with persistent fever and neutropenia. Pediatr Infect Dis J. 2010;29:415–420
8. Prentice HG, Hann IM, Herbrecht R, et al. A randomized comparison of liposomal versus conventional amphotericin B for the treatment of pyrexia of unknown origin in neutropenic patients. Br J Haematol. 1997;98:711–718
9. Sandler ES, Mustafa MM, Tkaczewski I, et al. Use of amphotericin B colloidal dispersion in
children . J Pediatr Hematol Oncol. 2000;22:242–246
10. Wiley JM, Seibel NL, Walsh TJ. Efficacy and safety of amphotericin B lipid complex in 548
children and adolescents with invasive fungal infections. Pediatr Infect Dis J. 2005;24:167–174
11. Roden MM, Nelson LD, Knudsen TA, et al. Triad of acute infusion-related reactions associated with liposomal amphotericin B: analysis of clinical and epidemiological characteristics. Clin Infect Dis. 2003;36:1213–1220
12. Johnson MD, Drew RH, Perfect JR. Chest discomfort associated with liposomal amphotericin B: report of three cases and review of the literature. Pharmacotherapy. 1998;18:1053–1061
13. Koren G, Lau A, Klein J, et al. Pharmacokinetics and adverse effects of amphotericin B in infants and
children . J Pediatr. 1988;113:559–563
14. Starke JR, Mason EO Jr, Kramer WG, et al. Pharmacokinetics of amphotericin B in infants and
children . J Infect Dis. 1987;155:766–774
15. Benson JM, Nahata MC. Pharmacokinetics of amphotericin B in
children . Antimicrob Agents Chemother. 1989;33:1989–1993
16. Walsh TJ, Teppler H, Donowitz GR, et al. Caspofungin versus liposomal amphotericin B for empirical antifungal therapy in patients with persistent fever and neutropenia. N Engl J Med. 2004;351:1391–1402
17. Blau IW, Fauser AA. Review of comparative studies between conventional and liposomal amphotericin B (Ambisome) in neutropenic patients with fever of unknown origin and patients with systemic mycosis. Mycoses. 2000;43:325–332
18. Blyth CC, Hale K, Palasanthiran P, et al. Antifungal therapy in infants and
children with proven, probable or suspected invasive fungal infections. Cochrane Database Syst Rev. 2010:CD006343
19. Goldman RD, Koren G. Amphotericin B nephrotoxicity in
children . J Pediatr Hematol Oncol. 2004;26:421–426
20. Wingard JR, White MH, Anaissie E, et al.L Amph/ABLC Collaborative Study Group. A randomized, double-blind comparative trial evaluating the safety of liposomal amphotericin B versus amphotericin B lipid complex in the empirical treatment of febrile neutropenia. L Amph/ABLC Collaborative Study Group. Clin Infect Dis. 2000;31:1155–1163
21. Harbarth S, Pestotnik SL, Lloyd JF, et al. The epidemiology of nephrotoxicity associated with conventional amphotericin B therapy. Am J Med. 2001;111:528–534
22. Zappitelli M, Moffett BS, Hyder A, et al. Acute kidney injury in non-critically ill
children treated with aminoglycoside antibiotics in a tertiary healthcare centre: a retrospective cohort study. Nephrol Dial Transplant. 2011;26:144–150
23. Trotti A, Byhardt R, Stetz J, et al. Common toxicity criteria: version 2.0. an improved reference for grading the acute effects of cancer treatment: impact on radiotherapy. Int J Radiat Oncol Biol Phys. 2000;47:13–47
24. Bates DW, Su L, Yu DT, et al. Correlates of acute renal failure in patients receiving parenteral amphotericin B. Kidney Int. 2001;60:1452–1459
25. Fisher MA, Talbot GH, Maislin G, et al. Risk factors for Amphotericin B-associated nephrotoxicity. Am J Med. 1989;87:547–552
26. Alexander BD, Wingard JR. Study of renal safety in amphotericin B lipid complex-treated patients. Clin Infect Dis. 2005;40(suppl 6):S414–S421
27. Branch RA, Jackson EK, Jacqz E, et al. Amphotericin-B nephrotoxicity in humans decreased by sodium supplements with coadministration of ticarcillin or intravenous saline. Klin Wochenschr. 1987;65:500–506
28. White MH, Bowden RA, Sandler ES, et al. Randomized, double-blind clinical trial of amphotericin B colloidal dispersion vs. amphotericin B in the empirical treatment of fever and neutropenia. Clin Infect Dis. 1998;27:296–302
29. Goodwin SD, Cleary JD, Walawander CA, et al. Pretreatment regimens for adverse events related to infusion of amphotericin B. Clin Infect Dis. 1995;20:755–761
30. Paterson DL, David K, Mrsic M, et al.PRoACT Investigators. Pre-medication practices and incidence of infusion-related reactions in patients receiving AMPHOTEC: data from the Patient Registry of Amphotericin B Cholesteryl Sulfate Complex for Injection Clinical Tolerability (PRoACT) registry. J Antimicrob Chemother. 2008;62:1392–1400
31. Tynes BS, Utz JP, Bennett JE, et al. Reducing amphotericin B reactions. A double-blind study. Am Rev Respir Dis. 1963;87:264–268
32. Burke D, Lal R, Finkel KW, et al. Acute amphotericin B overdose. Ann Pharmacother. 2006;40:2254–2259
33. Shaikh AJ, Bawany SA, Masood N, et al. Incidence and impact of baseline electrolyte abnormalities in patients admitted with chemotherapy induced febrile neutropenia. J Cancer. 2011;2:62–66
