Abstract
Aims
Lipid formulations of amphotericin B, rather than conventional amphotericin (c‐amB), are increasingly used despite limited data comparing these preparations in children. Data on the incidence of adverse effects with amphotericin B at standard doses are scarce. This study aimed to compare the adverse effects associated with standard doses of c‐amB and liposomal amphotericin (l‐amB) in children.
Methods
Children admitted to the Royal Children's Hospital Melbourne and treated with c‐amB or l‐amB between January 2010 and September 2013 were included. Clinical and laboratory data were retrospectively extracted from medical records to compare amphotericin‐related infusion reactions, nephrotoxicity (glomerulotoxicity and tubulopathy) and hepatotoxicity.
Results
Seventy‐six children received c‐amB and 39 received l‐amB. Standard drug administration (recommended dose and infusion time) occurred in 74% (56/76) of patients on c‐amB and 85% (33/39) on l‐amB. In these 89 children, infusion‐related reactions were similar for both c‐amB and l‐amB (23% (13/56) vs. 9% (3/33); P = 0.15); none occurred in children aged <90 days. There was no difference in amphotericin‐associated glomerulotoxicity (c‐amB 14% (8/56) vs. l‐amB 21% (7/33); P = 0.40) or in the median maximum potassium requirements (c‐amB 3.1 vs. l‐amB 2.3 mmol kg−1 d−1; P = 0.29). Hepatotoxicity occurred more frequently with l‐amB than c‐amB (83% (24/29) vs. 56% (20/36); P = 0.032).
Conclusions
When appropriately administered, l‐amB was associated with more hepatotoxicity than c‐amB, with no difference in infusion‐related reactions or nephrotoxicity. Differences in adverse effects between the preparations is not as marked in children as reported in adults.
Keywords: adverse effects, Amphotericin‐B, antifungal, liposomal, paediatric
What is Already Known about this Subject
Lipid formulations of amphotericin B may be better tolerated in adults, but are more expensive.
Studies comparing conventional and liposomal amphotericin B in children are scarce.
Dose and infusion time of amphotericin B may influence the incidence of adverse effects, but this has not been adequately investigated.
What this Study Adds
This is the first study to investigate adverse effects in the subset in whom conventional and liposomal amphotericin B were administered at a standard dose and infused rate.
With standard dosing of liposomal amphotericin B, there was no difference in infusion‐related reactions or nephrotoxicity.
A higher incidence of hepatotoxicity was associated with l‐amB therapy; however, this effect did not persist when other hepatotoxic medications were given concomitantly and requires further investigation.
Introduction
The prevalence of invasive fungal infections has increased worldwide due to advances in cancer therapies, increasing numbers of hematopoietic stem cell transplantations and broad‐spectrum antimicrobial use 1, 2, 3. Amphotericin B is a broad‐spectrum polyene antifungal agent, which is used as first‐line treatment for suspected fungal infections in both children and adults. The routine use of lipid formulations of amphotericin B (amphotericin B colloidal dispersion, amphotericin B lipid complex and liposomal amphotericin [l‐amB]) rather than conventional amphotericin B (c‐amB, amphotericin deoxycholate) is now common. This is largely due to studies, predominantly in adults, that show a higher rate of adverse effects, including infusion‐related reactions, nephrotoxicity and hepatotoxicity, associated with the conventional preparation 4, 5. However, the higher cost of lipid formulations of amphotericin B means they may not be available in some settings 6, 7, 8.
A systematic review comparing the use of conventional and lipid formulations of amphotericin B in children highlighted the scarcity of studies in children 9. Theoretically, children are at lower risk of adverse effects from c‐amB than adults as they have increased drug clearance and a smaller volume of distribution resulting in lower plasma levels 10. There are few studies investigating the risk of adverse effects when amphotericin B is administered optimally.
This study aimed to compare the adverse effects (infusion‐related reactions, nephrotoxicity [glomerulotoxicity and tubulopathy] and hepatotoxicity) associated with conventional and liposomal preparations of amphotericin B in children when administered at a correct dose and infused over the recommended duration.
Materials and methods
All patients who received treatment with c‐amB (Fungizone®) or l‐amB (Ambisome®) at the Royal Children's Hospital Melbourne (RCH) over a 45‐month period (January 2010–September 2013) were eligible for inclusion. The RCH is a tertiary paediatric referral hospital with a 34‐bed oncology unit, 24‐bed intensive care unit and 34‐bed neonatal intensive care unit. Only the first episode of treatment with amphotericin B for each patient was included in the analysis. Patients with underlying renal disease before treatment with amphotericin B were excluded. The study was approved by the local Human Research Ethics Committee (No. 33092).
The following data were extracted retrospectively from detailed review of medical records: patient demographics, underlying diagnoses, indication for treatment, adverse reactions, concomitant nephrotoxic drugs, antifungal treatment regimen and reason for discontinuation. All documented symptoms and signs of an infusion‐related reaction were also collected.
Results of relevant microbiological and biochemical investigations, including liver function tests at baseline and end of treatment, were also recorded. For assessment of glomerulotoxicity, baseline and peak creatinine levels as well as numbers of days to normalization of creatinine were collected. Serum creatinine is determined by enzymatic method at our laboratory. Glomerulotoxicity was defined as doubling of the baseline serum creatinine level. For comparison of the severity of hypokalaemia between the two formulations, the maximum daily requirement of potassium (mmol kg−1 d−1) was also collected. Hepatotoxicity was defined as per the common terminology criteria for adverse events (CTCAE) 11.
Adverse effects of amphotericin B are influenced by both dose and the rate of administration. Therefore, to exclude cases of toxicity due to prescription or administration errors, a separate analysis was done that included only the subset of patients who had ‘standard amphotericin B administration’ (defined as the manufacturer's recommended dosage infused over an appropriate duration, i.e. c‐amB 0.5–1.5 mg kg−1 infused over at least 4 h 12 and l‐amB 3–5 mg kg−1 infused over at least 1 h 13).
Fisher's exact test was used to compare the proportion of patients in each treatment arm who had adverse effects (infusion‐related reactions, nephrotoxicity, hepatotoxicity) or had additional therapies (i.e., concomitant nephrotoxic drugs or supplementary potassium). The median of the maximum potassium requirement was compared between those who received c‐amB and l‐amB using Bonnett–Price confidence intervals and the distribution of maximum potassium requirements in these two groups was compared using the Mann–Whitney test. We investigated whether the effect of amphotericin formulation on hepatotoxicity was modified by the use of concomitant drugs (dichotomized as no drugs and one or more agents). This relationship was adjusted for age (dichotomized as 90 days or older and less than 90 days).
Results
A total of 115 patients received amphotericin B during the study period. The majority of patients were male (72/115, 62.6%). The median age at time of treatment was 5.7 years (range 5 days–19.3 years). At the start of treatment, 19/115 (16.5%) patients were aged less than 90 days. Of these infants, the most common underlying conditions were prematurity (8/19, 42.1%), cardiac disease (6/19, 31.6%), respiratory disease (6/19, 31/6%) and necrotizing enterocolitis (6/19, 31.6%). Among the 96 children aged 90 days or more, the most common underlying conditions were haematological malignancy (50/96, 52.1%), hematopoietic stem cell transplant (20/96, 20.8%) and gastrointestinal disease (10/96, 10.4%) (Table 1). There were six patients who ceased treatment due to death or palliation. Invasive fungal disease was a cause of death in two of these patients (both received l‐amB).
Table 1.
Underlying diagnoses and indication for amphotericin B treatment
| c‐amB (n = 76) | l‐amB (n = 39) | |
|---|---|---|
| n (%) | n (%) | |
| Underlying diagnosis a age < 90 days (n = 19) | 9/76 (11.8) | 10/39 (25.6) |
| Prematurity | 3/8 (37.5) | 5/8 (62.5) |
| Cardiac disease | 1/7 (14.3) | 6/7 (85.7) |
| Respiratory disease | 2/6 (33.3) | 4/6 (66.7) |
| Necrotising enterocolitis | 2/6 (33.3) | 4/6 (66.7) |
| Congenital diaphragmatic hernia | 3/4 (75.0) | 1/4 (25.0) |
| Immunodeficiency | 1/1 (100.0) | 0 |
| Other | 3/7 (42.9) | 4/7 (57.1) |
| Underlying diagnosis a age ≥ 90 days (n = 96) | 67/76 (88.2) | 29/39 (74.4) |
| Haematological malignancy | 41/50 (82.0) | 9/50 (18.0) |
| Haemopoietic stem cell transplant | 7/20 (35.0) | 13/20 (65.0) |
| Gastrointestinal disease | 5/10 (50.0) | 5/10 (50.0) |
| Immunodeficiency | 2/7 (28.6) | 5/7 (71.4) |
| Haematological disease | 3/6 (50.0) | 3/6 (50.0) |
| Solid organ transplant | 3/6 (50.0) | 3/6 (50.0) |
| Cardiac disease | 2/6 (33.3) | 4/6 (66.7) |
| Solid organ tumour | 4/4 (100.0) | 0 |
| Respiratory disease | 3/3 (100.0) | 0 |
| Gastrointestinal surgery | 2/2 (100.0) | 0 |
| Neurological disease | 2/2 (100.0) | 0 |
| Other | 4/7 (57.1) | 3/7 (42.9) |
| Treatment indication b | n / 76 (%) | n /39 (%) |
| Proven IFI c | 23 (30.3) | 10 (25.6) |
| Probable IFI c | 8 (10.5) | 6 (15.4) |
| Possible IFI c | 20 (26.3) | 9 (23.1) |
| Other empiric indication d | 24 (31.6) | 13 (33.3) |
| Prophylaxis | 1 (1.3) | 3 (7.7) |
IFI, invasive fungal infection.
Patients could have had more than one underlying diagnosis.
Patients could have had more than one treatment indication.
As defined by the European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative and National Institute of Allergy and Infectious Diseases Mycoses Study Consensus Group 32.
Prolonged fever or sepsis (n = 23); heart transplant recipient whose donor was cryptococcus positive (n = 1).
Of the 115 patients, 76 (66.1%) received c‐amB at a median dose of 1.0 (range 0.5–5.0) mg kg−1 d−1. The remaining 39 patients (33.9%) received l‐amB at a median dose of 3.0 (range 1.0–10.0) mg kg−1 d−1. The median duration of c‐amB treatment was 3 (range 1–370) days, and 6 (range 1–54) days for l‐amB. The median infusion time for patients treated with c‐amB was 6 (range 1–24) hours and the median infusion time for l–amB was 2 (range 0.5–24) hours.
Of the 76 patients who received c‐amB, 74 (97.4%) were treated with a standard dose (0.5–1.5 mg kg−1 d−1) and 56 (73.7%) had this dose infused over the recommended infusion duration of 4 h or longer. Thirty‐five (89.7%) of the 39 patients who received l‐amB were treated with a standard dose (3.0–5.0 mg kg−1 d−1) and 33 (84.6%) had this dose infused over the recommended duration of 1 h or longer (Table 2).
Table 2.
Details of amphotericin dosing
| c‐amB | l‐amB | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Age | <90 days (n = 9) | ≥90 days (n = 67) | <90 days (n = 10) | ≥90 days (n = 29) | |||||
| Dose (mg kg −1 d −1 ) | 0.5 ‡ −1.5 | >1.5 | 0.5–1.5 | >1.5 | <3.0 | 3.0–5.0 | >5.0 | <3.0 | 3.0–5.0 |
| No. (%) patients | 9 (100%) | – | 65 (97%) | 2 (3%) | 1 (10%) | 8 (80%) | 1 (10%) | – | 27 (94%) |
| No. (%) patients with correct infusion time | 8 (89%) | – | 48 (72%) | 1 (1.5%) | Unknown | 8 (80%) | 1 (10%) | – | 25 (86%) |
| Median (range) duration (days) | 0.7 (2–7) | – | 1.0 (1–370) | 1.5 (1–2) | 1 | 4 (1–43) | 5 | – | 7 (1–54) |
In the 89 patients who had standard amphotericin B administration, there was no difference in the rate of infusion‐related reactions in those who received c‐amB compared to those who received l‐amB [13/56 (23.2%) vs. 3/33 (9.1%); P = 0.15]. All infusion‐related reactions occurred in children older than 90 days. The most frequent symptoms of an infusion‐related reaction were rigors (9/16, 56.3%), fever (7/16, 43.8%), tachycardia (5/16, 31.3%) and rash (4/16, 25.0%). Other features included dyspnoea, hypotension, nausea, vomiting, headache, flushing and there were single instances of orbital swelling, formication and hypertension. There was no significant difference in the rate of infusion‐related reactions between those who received an infusion of 1.0 mg kg−1 d−1 c‐amB at the recommended rate (at least 4 h) and those who received it at a faster rate [13/40 (32.5%) vs. 4/12 (33.3%), P = 1.00].
Amphotericin‐related glomerulotoxicity occurred in 8/56 (14.3%) of patients on c‐amB and 7/33 (21.2%) on l‐amB (P = 0.40). All eight patients who received c‐amB and developed toxicity were also receiving one or two concomitant drugs. Three of the seven patients who received l‐amB and developed glomerulotoxicity were taking one or two concomitant nephrotoxic drugs and four patients were taking four or five other nephrotoxic drugs. The mean number of concomitant nephrotoxic drugs received was higher for those on l‐amB than c‐amB (2.5 vs. 2.0). One patient (1/8, 12.5%) on c‐amB and two patients (2/7, 28.6%) on l‐amB had also recently received nephrotoxic chemotherapy. In this subset, 10/15 (66.7%) patients had an underlying haematological malignancy.
Of the 89 patients with standard amphotericin B administration, 16/56 (28.6%) on c‐amB and 7/33 (21.2%) had hypokalaemia at the end of amphotericin B treatment. The majority of patients had grade I or II hypokalaemia (14/16, 87.5% c‐amB vs. 5/7, 71.4% l‐amB) and the remainder had grade III hypokalaemia. No patients had grade IV hypokalaemia. Potassium replacement therapy was given in 50/56 (89.3%) of those on c‐amB and 29/33 (87.9%) on l‐amb (P = 1.00). Only one of these patients who was receiving l‐amB required intensive care admission for the management of hypokalaemia. For those patients aged less than 90 days, 7/8 (87.5%) receiving c‐amB required potassium replacement therapy compared to 8/8 (100.0%) receiving l‐amB (P = 1.00). There was no significant difference in the median of the maximum potassium requirements for the c‐amB and l‐amB groups (3.1 vs. 2.3 mmol kg−1 d−1, P = 0.29). The distribution of maximum potassium requirements between the two groups was also similar (P = 0.99).
Sixty‐five patients with standard drug administration had liver function test (LFT) results available at both baseline and end‐of‐treatment. Amphotericin‐related hepatotoxicity was more common with l‐amB than with c‐amB [24/29 (82.8%) vs. 20/36 (55.6%); crude OR 3.8; 1.2–12.3; P = 0.024]. However, the effect was negatively modified by the use of concomitant drugs (Table 3) – the adjusted relative odds of hepatotoxicity among those on l‐amB compared to c‐amB was higher among patients on amphotericin alone (20.3; P = 0.012) than among those receiving one or more concomitant medications (1.6; P = 0.581). Most hepatotoxicity was low grade: 35/44 (79.5%) patients had grade I or II abnormalities for any parameter (i.e. bilirubin, ALT, ALP and GGT). Nineteen (43.2%) patients had high‐grade abnormalities (i.e. grade III or IV). Grade III and IV abnormalities in bilirubin occurred more frequently with l‐amB [5/29 (17.2%); P = 0.014] compared to c‐amB (0/36). Enzyme derangements primarily affected GGT (27/44, 61.4%) and ALT (23/44, 52.3%) for both c‐amB and l‐amB (Table 3). Of note, the majority of these children with hepatotoxicity had at least one enzyme abnormality at baseline [19/20 (95.0%) patients on c‐amB, 21/24 (87.5%) patients on l‐amB].
Table 3.
Amphotericin‐associated hepatotoxicity among patients with standard drug administration
| Measure of hepatotoxicity | All ages | Age < 90 days | Age ≥ 90 days | ||||||
|---|---|---|---|---|---|---|---|---|---|
| c‐amB | l‐amB | P‐value | c‐amB | l‐amB | P‐value | c‐amB | l‐amB | P‐value | |
| n | n | n | n | n | n | ||||
| Alanine aminotransferase (ALT) | |||||||||
| Gr * I + Gr II (>ULN to 5.0× ULN) | 9/36 (25.0%) | 10/29 (34.5%) | 0.426 | 1/6 (16.7%) | 2/8 (25.0%) | 1.000 | 8/30 (26.7%) | 9/21 (42.9%) | 0.247 |
| Gr III + Gr IV (>5.0 to 20× ULN) | 1 (2.8%) | 3 (10.3%) | 0.316 | 0 (0%) | 2 (25.0%) | 0.473 | 0 (0%) | 1 (4.8%) | 0.412 |
| Gamma‐glutamyl transferase (GGT) | |||||||||
| Gr I + Gr II (>ULN to 5.0× ULN) | 6 (16.7%) | 12 (41.4%) | 0.049 | 0 (0%) | 1 (12.5%) | 1.000 | 6 (20.0%) | 11 (52.4%) | 0.033 |
| Gr III + Gr IV (>5.0 to 20× ULN) | 6 (16.7%) | 3 (10.3%) | 0.720 | 0 (0%) | 0 (0%) | . | 6 (20.0%) | 3 (14.3%) | 0.720 |
| Alkaline phosphatase (ALP) | |||||||||
| Gr I + Gr II (>ULN to 3.0× ULN) | 3 (8.3%) | 3 (10.3%) | 1.000 | 0 (0%) | 0 (0%) | . | 3 (10.0%) | 3 (14.3%) | 0.680 |
| Gr III + Gr IV (>3.0 to 10.0× ULN) | 0 (0%) | 0 (0%) | . | 0 (0%) | 0 (0%) | . | 0 (0%) | 0 (0%) | . |
| Bilirubin >ULN | |||||||||
| Gr I + Gr II (>ULN to 3.0× ULN) | 2 (5.6%) | 4 (13.8%) | 0.395 | 1 (16.7%) | 2 (25.0%) | 1.000 | 2 (6.7%) | 2 (9.5%) | 1.000 |
| Gr III + Gr IV (>3.0 to 10.0× ULN) | 0 (0%) | 5 (17.2%) | 0.014 | 0 (0%) | 1 (12.5%) | 1.000 | 0 (0%) | 4 (19.0%) | 0.024 |
Grades (Gr) of hepatotoxicity defined according to common terminology criteria for adverse events (CTCAE)11.
ULN, upper limit of normal.
There was no significant difference in the underlying diagnoses of those patients who developed hepatotoxicity when amphotericin B was administered at standard doses. Six patients had underlying diagnoses that specifically predisposed to increased hepatotoxicity. These included liver transplant (n = 4), hepatoblastoma and cystic fibrosis (n = 1 for each). Three of these patients received c‐amB and the other three received l‐amB. The median number of hepatotoxic medications was zero (range 0–4) in those patients receiving a standard dose of amphotericin B.
Amphotericin B treatment was discontinued in a higher proportion of patients receiving c‐amB than in those receiving l‐amB [12/56 (21.4%) vs. 6/33 (18.2%); P = 0.79]. Of the patients who discontinued c‐amB treatment due to drug toxicity, 8/12 (66.7%) ceased due to infusion‐related reactions alone, 2/12 (16.7%) due to renal impairment and 2/12 (16.7%) due to a combination of both infusion reactions and renal impairment. In the c‐amB group, two additional patients discontinued treatment due to limited intravenous line space. Of the six patients who discontinued l‐amB treatment, five (83.3%) stopped due to renal impairment and one (16.7%) stopped due to an infusion‐related reaction. No patients discontinued treatment due to hepatotoxicity. Of the 12 patients who discontinued due to other drug‐related side effects (infusion‐related reactions, renal impairment), eight also had hepatotoxicity (n = 4 for both c‐amB and l‐amB). All but one of these patients had abnormal baseline LFTs. Of note, the four patients on l‐amB with hepatotoxicity had only mild liver enzyme derangements, while 3/4 patients on c‐amB had high‐grade hepatotoxicity.
There was no significant difference in discontinuation between those with standard and non‐standard administration.
Discussion
This is one of the largest studies comparing adverse effects associated with liposomal and conventional preparations of amphotericin B in children. Several international guidelines have changed to recommend lipid formulations over c‐amB for paediatric patients beyond the neonatal age group due to potential adverse effects 9, 14, 15, 16. This shift in clinical practice is largely based on evidence from studies in adults. Systematic reviews have highlighted the absence of randomized controlled trials directly comparing these different formulations of amphotericin B in children 6, 9, 17. This limits the ability to draw firm conclusions about the true incidence of toxicity with these preparations in children. Pharmacoeconomic factors continue to drive the imperative to investigate the use of c‐amB in the paediatric population. The cost of lipid formulations ranges from 10 to 230 times higher than c‐amB 6, 8. Cost‐effective use of medications is important in all settings and the unnecessary use of costly preparations represents wasted expenditure of limited drug budgets. The expense of liposomal preparations of amphotericin B has a particular impact on drug accessibility in low socioeconomic countries 6, 8.
The unique pharmacokinetic profile of c‐amB in children is likely to reduce the risk of amphotericin‐related toxicity. In addition, there are various strategies relating to drug administration that can reduce the risk of adverse effects. These include sodium pre‐infusion, routine premedication with an antihistamine and antipyretic, rationalizing concomitant nephrotoxic drugs and administration of c‐amB as an extended or continuous infusion 6, 18, 19, 20. The routine use of these strategies when administering c‐amB is not included in any guideline and was not standard practice at our institution during the study period. Our finding that over one‐quarter of children who received c‐amB did not have ‘standard administration’ reflects the suboptimal use of c‐amB. This issue is also highlighted in a recent Cochrane review comparing lipid formulations to c‐amB in which the authors refrained from drawing conclusions due to the scarcity of studies in which c‐amB was administered under optimal conditions 21. Our study is the first to investigate adverse effects in the subset in whom c‐amB and l‐amB were administered at a standard dose and infusion time.
In our study, nephrotoxicity was defined as a doubling of serum creatinine, which is equivalent to CTCAE grade II creatinine elevation. This definition was chosen to enable comparison with other studies (both adult and paediatric), as it is the most frequently used. However, the relevance of doubling serum creatinine in children is unclear. Notably, the creatinine may double but remain in the ‘normal’ range for age. While many children may tolerate the renal adverse effects of amphotericin B better than adults due to increased nephronal reserve, there are individuals with multiple comorbidities on multiple nephrotoxic drugs for whom a smaller change in creatinine may be clinically significant 9.
The overall rates of glomerulotoxicity observed in our study are consistent with other studies in which rates between 15% and 58% have been reported 6. A key finding of our study was that when controlled for optimal drug administration, there was no significant difference in the rate of glomerulotoxicity in those treated with l‐amB (21.2%) compared with c‐amB (14.3%). This finding may be attributable to the high number of concomitant nephrotoxic drugs in those who received l‐amB. As no patients in the c‐amB arm received more than two concomitant nephrotoxic drugs, it was not possible to assess the effect of increasing numbers of nephrotoxic drugs on the risk of glomerulotoxicity associated with c‐amB.
The similar incidence of infusion‐related reactions seen with c‐amB and l‐amB contrasts with previous paediatric and adult studies. Although previous reports suggest that infusion‐related reactions occur more frequently with c‐amB, the comparable frequency of infusion‐related reactions between the two amphotericin B formulations in our study could be attributable to the analysis including only those patients with standard drug administration 4, 5, 6, 9, 22. Notably, infusion‐related reactions have not been reported in patients under the age of 90 days. The mechanism of infusion‐related reactions with c‐amB has been attributed to signalling through toll‐like receptor 2 and the innate immune inflammatory cascade. This is attenuated by lipid formulations which instead trigger toll‐like receptor 4 reactions 4. The absence of this phenomenon in the neonatal population may relate to immaturity of the immune system.
Hypokalaemia is a manifestation of distal tubulopathy and is caused by both conventional and lipid formulations. Randomized controlled trials and meta‐analyses have reported no significant difference in the frequency of hypokalaemia between c‐amB and l‐amB 9, 23, 24. However, there is substantial heterogeneity in the definition of hypokalaemia between studies 9, 20. Previously, studies have been limited by using end‐of‐treatment potassium as a marker of treatment‐associated hypokalaemia. However, as current standard of care is to monitor and replace potassium for patients on amphotericin B treatment, this is simply an indication of the adequacy of potassium replacement. The use of maximum daily potassium requirements is a more appropriate outcome measure of amphotericin‐associated tubulopathy.
Studies have investigated the role of concomitant potassium‐sparing diuretics (e.g. amiloride, spironolactone) to prevent amphotericin‐associated hypokalaemia, the results of which are promising 25. To date, there is no standardized approach to potassium supplementation during amphotericin B treatment.
Hepatotoxicity due to amphotericin B treatment is idiosyncratic and may relate to a combination of genetic predisposition and direct cell membrane damage, though this is difficult to clinically differentiate from liver damage due to fungal infection or an underlying comorbidity 26, 27, 28. Amphotericin‐associated hepatotoxicity is usually mild and infrequent 20, 28, 29. No significant difference in the rate of hepatotoxicity between conventional and liposomal formulations in both paediatric and adult studies have been reported (10–17% for c‐amB, 14–18% for l‐amB) 9, 12, 20. By contrast, the patients in this study treated with l‐amB were significantly more likely to experience hepatotoxicity compared with c‐amB (82.8% vs. 55.6%). This effect was not seen when other hepatotoxic drugs were given concomitantly with l‐amB. The interpretation of these results is limited due to the small sample size and will require further investigation in larger studies. However, a higher rate of hepatotoxicity may be explained by the fact that lipid preparations of amphotericin B result in a much higher delivery of drug to the liver and can result in liver injury at high doses, as seen in animal studies and one case–control study 27, 29.
Among the neonatal population, c‐amB remains the most commonly used antifungal agent and this is supported by international guidelines 6, 14, 15. In this study there were no occurrences of glomerulotoxicity among infants aged under 89 days. In young infants, concomitant infusions of sodium protect against nephrotoxicity 6, 17, 30 and studies show a general trend for improved tolerance with decreasing age. This is thought to relate to faster drug clearance 31. However, neonates have marked variability in c‐amB pharmacokinetics, particularly among the very premature, and the pharmacokinetics of l‐amB in neonates are unknown 6, 17, 31. To date, there are no randomized controlled trials comparing the use of c‐amB with lipid formulations in the neonatal population.
Limitations of our study include its retrospective design, heterogeneous population and some missing clinical data for intravenous line space, use of premedications, other potential causes of hepatotoxicity and LFT results. We attempted to take into account the effect of polypharmacy; however, the list of concomitant medications was not exhaustive. Most importantly, our findings are likely to be influenced by the fact that the choice of using c‐amB or l‐amB was determined by the treating clinician. It is possible that patients at real or perceived greater risk of adverse effects from c‐amB were more likely to receive l‐amB, which therefore may have led to an underestimate of the adverse effects of c‐amB.
Although there is a change towards the routine use of lipid formulations of amphotericin B, c‐amB continues to be used worldwide. Given the potential toxicities associated with this formulation, attention needs to be given to optimizing administration by considering infusion time, dosing and the routine use of premedications.
In conclusion, this retrospective study found no difference in the rate of infusion‐related reactions, glomerulotoxicity or tubulopathy with c‐amB compared with l‐amB. Interestingly, our data showed a higher incidence of hepatotoxicity associated with l‐amB therapy; however, this effect did not persist when other hepatotoxic medications were given concomitantly and requires further investigation. The availability of two preparations of amphotericin B raises issues of cost, access and risk of prescription and administration errors. This study suggests that c‐amB can be safely used in selected patients in the paediatric population and further emphasizes the need for randomized controlled trials directly comparing these two formulations in a paediatric population.
Competing Interests
There are no competing interests to declare.
Andrew, E. C. , Curtis, N. , Coghlan, B. , Cranswick, N. , and Gwee, A. (2018) Adverse effects of amphotericin B in children; a retrospective comparison of conventional and liposomal formulations. Br J Clin Pharmacol, 84: 1006–1012. doi: 10.1111/bcp.13521.
References
- 1. Dzierzanowska‐Fangrat K, Romanowska E, Gryniewicz‐Kwiatkowska O, Migdal M, Witulska K, Ryzko J, et al Candidaemia in a Polish tertiary paediatric hospital, 2000 to 2010. Mycoses 2014; 57: 105–109. [DOI] [PubMed] [Google Scholar]
- 2. Enoch DA, Ludlam HA, Brown NM. Invasive fungal infections: a review of epidemiology and management options. J Med Microbiol 2006; 55: 809–818. [DOI] [PubMed] [Google Scholar]
- 3. Richardson M, Lass‐Florl C. Changing epidemiology of systemic fungal infections. Clin Microbiol Infect 2008; 14 (Suppl. 4): 5–24. [DOI] [PubMed] [Google Scholar]
- 4. Hamill RJ. Amphotericin B formulations: a comparative review of efficacy and toxicity. Drugs 2013; 73: 919–934. [DOI] [PubMed] [Google Scholar]
- 5. Botero Aguirre JP, Restrepo Hamid AM. Amphotericin B deoxycholate versus liposomal amphotericin B: effects on kidney function. Cochrane Database of Syst Rev 2015; 11: CD010481. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Bes DF, Rosanova MT, Sberna N, Arrizurieta E. Deoxycholate amphotericin B and nephrotoxicity in the pediatric setting. Pediatr Infect Dis J 2014; 33: e198–e206. [DOI] [PubMed] [Google Scholar]
- 7. Cagnoni PJ, Walsh TJ, Prendergast MM, Bodensteiner D, Hiemenz S, Greenberg RN, et al Pharmacoeconomic analysis of liposomal amphotericin B versus conventional amphotericin B in the empirical treatment of persistently febrile neutropenic patients. Clin Oncol 2000; 18: 2476–2483. [DOI] [PubMed] [Google Scholar]
- 8. Sundar S, Chakravarty J, Agarwal D, Rai M, Murray HW. Single‐dose liposomal amphotericin B for visceral leishmaniasis in India. N Engl J Med 2010; 362: 504–512. [DOI] [PubMed] [Google Scholar]
- 9. Blyth CC, Hale K, Palasanthiran P, O'Brien T, Bennett MH. Antifungal therapy in infants and children with proven, probable or suspected invasive fungal infections. Cochrane Database Syst Rev 2010; 2: CD006343. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Starke JR, Mason EO Jr, Kramer WG, Kaplan SL. Pharmacokinetics of amphotericin B in infants and children. J Infect Dis 1987; 155: 766–774. [DOI] [PubMed] [Google Scholar]
- 11. National Cancer Institute . Common terminology criteria for adverse events v4.03. US Department of Health and Human Services, 2010. Retrieved from: http://evs.nci.nih.gov/ftp1/CTCAE/CTCAE_4.03_2010-06-14_QuickReference_8.5x11.pdf (last accessed 2 October 2016).
- 12. Australian Medicines Handbook (online). Adelaide: Australian Medicines Handbook Pty Ltd, 2015. Retrieved from: http://amhonline.amh.net.au/ (last accessed 2 October 2016).
- 13. Lexicomp Online . Pediatric & Neonatal Lexi‐Drugs. Hudson, OH: Lexi‐Comp, 2015. [Google Scholar]
- 14. Groll AH, Castagnola E, Cesaro S, Dalle JH, Engelhard D, Hope W, et al Fourth European Conference on Infections in Leukaemia (ECIL‐4): guidelines for diagnosis, prevention, and treatment of invasive fungal diseases in paediatric patients with cancer or allogeneic haemopoietic stem‐cell transplantation. Lancet Oncol 2014; 15: e327–e340. [DOI] [PubMed] [Google Scholar]
- 15. Hope WW, Castagnola E, Groll AH, Roilides E, Akova M, Arendrup MC, et al ESCMID* guideline for the diagnosis and management of Candida diseases 2012: prevention and management of invasive infections in neonates and children caused by Candida spp. Clin Microbiol Infect 2012; 18 (Suppl. 7): 38–52. [DOI] [PubMed] [Google Scholar]
- 16. Pappas PG, Kauffman CA, Andes D, Benjamin DK Jr, Calandra TF, Edwards JE Jr, et al Clinical practice guidelines for the management of candidiasis: 2009 update by the Infectious Diseases Society of America. Clin Infect Dis 2009; 48: 503–535. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Clerihew L, McGuire W. Antifungal therapy for newborn infants with invasive fungal infection. Cochrane Database Syst Rev 2012; 6: CD003953. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Eriksson U, Seifert B, Schaffner A. Comparison of effects of amphotericin B deoxycholate infused over 4 or 24 hours: randomised controlled trial. BMJ 2001; 322: 579–582. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Falci DR, dos Santos RP, Wirth F, Goldani LZ. Continuous infusion of amphotericin B deoxycholate: an innovative, low‐cost strategy in antifungal treatment. Mycoses 2011; 54: 91–98. [DOI] [PubMed] [Google Scholar]
- 20. Girois SB, Chapuis F, Decullier E, Revol BG. Adverse effects of antifungal therapies in invasive fungal infections: review and meta‐analysis. Eur J Clin Microbiol Infect Dis 2005; 24: 119–130. [DOI] [PubMed] [Google Scholar]
- 21. Johansen HK, Gotzsche PC. Amphotericin B lipid soluble formulations versus amphotericin B in cancer patients with neutropenia. Cochrane Database Syst Rev 2014; 9: CD000969. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Wade RL, Chaudhari P, Natoli JL, Taylor RJ, Nathanson BH, Horn D. Comparison of adverse events and hospital length of stay associated with various amphotericin B formulations: sequential conventional amphotericin b/lipid versus lipid‐only therapy for the treatment of invasive fungal infections in hospitalized patients. PT 2013; 38: 278–287. [PMC free article] [PubMed] [Google Scholar]
- 23. Prentice HG, Hann IM, Herbrecht R, Aoun M, Kvaloy S, Catovsky D, 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. [DOI] [PubMed] [Google Scholar]
- 24. Walsh TJ, Finberg RW, Arndt C, Hiemenz J, Schwartz C, Bodensteiner D, 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. [DOI] [PubMed] [Google Scholar]
- 25. Karimzadeh I, Khalili H, Farsaei S, Dashti‐Khavidaki S, Sagheb MM. Role of diuretics and lipid formulations in the prevention of amphotericin B‐induced nephrotoxicity. Eur J Clin Pharmacol 2013; 69: 1351–1368. [DOI] [PubMed] [Google Scholar]
- 26. US National Library of Medicine . Clinical and research information on drug‐induced liver injury: Amphotericin B. Bethesda, MD: US National Library of Medicine, 2013. [Google Scholar]
- 27. Fischer MA, Winkelmayer WC, Rubin RH, Avorn J. The hepatotoxicity of antifungal medications in bone marrow transplant recipients. Clin Infect Dis 2005; 41: 301–307. [DOI] [PubMed] [Google Scholar]
- 28. Gallis HA, Drew RH, Pickard WW. Amphotericin B: 30 years of clinical experience. Rev Infect Dis 1990; 12: 308–329. [DOI] [PubMed] [Google Scholar]
- 29. Wingard JR, Leather H. Hepatotoxicity associated with antifungal therapy after bone marrow transplantation. Clin Infect Dis 2005; 41: 308–310. [DOI] [PubMed] [Google Scholar]
- 30. Blyth CC, Chen SC, Slavin MA, Serena C, Nguyen Q, Marriott D. Not just little adults: candidemia epidemiology, molecular characterization, and antifungal susceptibility in neonatal and pediatric patients. Pediatrics 2009; 123: 1360–1368. [DOI] [PubMed] [Google Scholar]
- 31. Turkova A, Roilides E, Sharland M. Amphotericin B in neonates: deoxycholate or lipid formulation as first‐line therapy – is there a 'right' choice? Curr Opin Infect Dis 2011; 24: 163–171. [DOI] [PubMed] [Google Scholar]
- 32. De Pauw B, Walsh TJ, Donnelly JP, Stevens DA, Edwards JE, et al Revised definitions of invasive fungal disease from the European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and the National Institute of Allergy and Infectious Diseases Mycoses Study Group (EORTC/MSG) Consensus Group. Clin Infect Dis 2008; 46: 1813–1821. [DOI] [PMC free article] [PubMed] [Google Scholar]