Liposomal amphotericin B-associated electrolyte abnormalities among adult patients in a tertiary hospital Authors: Najwa Al Himali, Hanan Al Habsi, Marwa Al Harthi, Shaikha Al Ghailani, Dana Al Ghailani, Leena Kashoub, Aly M. Abdelrahman Published: 2026 ABSTRACT Highlights•Treatment with liposomal amphotericin B (L-AMB) was associated with a significant decline in serum potassium over the first week of therapy.•Treatment with L-AMB was associated with a significant rise in sodium over the first week of therapy.•Higher dose/kg of L-AMB and pre-infusion hydration independently predicted hypokalemia. Introduction Amphotericin B belongs to the polyene class of antifungals. It exerts its antifungal activity through high-affinity interactions with ergosterol in the fungal plasma membrane. These interactions can produce membrane permeabilizing assemblies and ion leakage, while additional evidence supports an ergosterol extraction mechanism involving an extramembranous, sponge or lattice-like amphotericin B sterol structure. It is used to treat a wide range of potentially life-threatening fungal infections, such as aspergillosis, visceral and cutaneous leishmaniasis, mucormycosis, and infections by other molds, including Fusarium spp. It is also a US Food and Drug Administration-approved antifungal agent for invasive candidiasis. It is effective against the majority of Candida species, including Candida albicans, Candida krusei, Candida tropicalis, and Candida parapsilosis. In addition, it is indicated for opportunistic fungal infections in immunocompromised patients, such as patients with human immunodeficiency virus, and as empiric treatment in a persistently febrile neutropenic host [[1], [2], [3]]. Liposomal amphotericin B (L-AMB) is a preferred formulation of amphotericin B compared with other traditional formulations. This is due to the reduced risk of nephrotoxicity. However, this formulation can cause significant electrolyte abnormalities, particularly hypokalemia, that may lead to various complications of clinical significance, especially in vulnerable populations with existing comorbidities [1,[4], [5], [6]]. The reported incidence of different electrolyte abnormalities varies among various studies. For example, the reported incidence of hypokalemia among hematology patients was higher, reaching 76.6% compared with 33-45% in other patients in other studies. Also, the predictors for the development of different electrolyte abnormalities are unclear. Preexisting renal dysfunction, concomitant nephrotoxic medications, and female gender are believed to contribute to the risk of hypokalemia with L-AMB. Other electrolyte data are limited [1,[4], [5], [6]]. It is crucial to understand the potential electrolyte abnormalities and their risk factors for better management during L-AMB use. In addition, patients receiving L-AMB must be monitored during therapy, and early supplementation must be considered to alleviate the risks. This study aimed to evaluate the possible electrolyte imbalances that can occur with the use of L-AMB among adult patients admitted to Sultan Qaboos University Hospital (SQUH). Methods Study design and setting This is a retrospective observational study of patients who received L-AMB (AmBisome®, Gilead Sciences) during their admission at SQUH. Inclusion/exclusion criteria Inclusion criteria:1.Patients who received L-AMB in the past 10 years.2.Adult patients (≥18 years).3.Patients who received at least 1 week of L-AMB. 1.Patients who received L-AMB in the past 10 years. Patients who received L-AMB in the past 10 years. 2.Adult patients (≥18 years). Adult patients (≥18 years). 3.Patients who received at least 1 week of L-AMB. Patients who received at least 1 week of L-AMB. Exclusion criteria:1.Patients with a history of acute kidney injury (AKI) (creatinine level increased by 0.3 mg/dL or more within 48 h, or creatinine level increased by 1.5 times or more than baseline in the previous 7 days, or urine volume <0.5 mL/kg/h for 6 hours)2.Patients with a history of chronic kidney disease (baseline estimated glomerular filtration rate [eGFR] <60 mL/min/1.73 m²).3.Patients receiving hemodialysis or peritoneal dialysis. 1.Patients with a history of acute kidney injury (AKI) (creatinine level increased by 0.3 mg/dL or more within 48 h, or creatinine level increased by 1.5 times or more than baseline in the previous 7 days, or urine volume <0.5 mL/kg/h for 6 hours) Patients with a history of acute kidney injury (AKI) (creatinine level increased by 0.3 mg/dL or more within 48 h, or creatinine level increased by 1.5 times or more than baseline in the previous 7 days, or urine volume <0.5 mL/kg/h for 6 hours) 2.Patients with a history of chronic kidney disease (baseline estimated glomerular filtration rate [eGFR] <60 mL/min/1.73 m²). Patients with a history of chronic kidney disease (baseline estimated glomerular filtration rate [eGFR] <60 mL/min/1.73 m²). 3.Patients receiving hemodialysis or peritoneal dialysis. Patients receiving hemodialysis or peritoneal dialysis. Data collection Data were extracted from patients’ medical records (SQUH TrakCare system), including the following:•Patient demographics, history, and comorbidities (hypertension, diabetes, cancer, liver disease, cardiovascular disease).•Details related to L-AMB (dose per kg, indication, duration).•Electrolyte levels (potassium [K], magnesium [Mg], calcium [Ca], and sodium [Na]) at day 0 (baseline), day 3, day 5, and day 7.•Any co-administered medications known to be nephrotoxic or to cause electrolyte imbalance.•Kidney function tests, including urea, creatinine, and eGFR.•Hydration status before therapy (defined by any administered fluid before therapy and type/volume of fluid). •Patient demographics, history, and comorbidities (hypertension, diabetes, cancer, liver disease, cardiovascular disease). Patient demographics, history, and comorbidities (hypertension, diabetes, cancer, liver disease, cardiovascular disease). •Details related to L-AMB (dose per kg, indication, duration). Details related to L-AMB (dose per kg, indication, duration). •Electrolyte levels (potassium [K], magnesium [Mg], calcium [Ca], and sodium [Na]) at day 0 (baseline), day 3, day 5, and day 7. Electrolyte levels (potassium [K], magnesium [Mg], calcium [Ca], and sodium [Na]) at day 0 (baseline), day 3, day 5, and day 7. •Any co-administered medications known to be nephrotoxic or to cause electrolyte imbalance. Any co-administered medications known to be nephrotoxic or to cause electrolyte imbalance. •Kidney function tests, including urea, creatinine, and eGFR. Kidney function tests, including urea, creatinine, and eGFR. •Hydration status before therapy (defined by any administered fluid before therapy and type/volume of fluid). Hydration status before therapy (defined by any administered fluid before therapy and type/volume of fluid). Electrolyte abnormalities on days 3, 5, and 7 were additionally graded according to the Common Terminology Criteria for Adverse Events, version 6.0 (CTCAE v6.0), using laboratory-value-only criteria. Because CTCAE Grade 1 and Grade 2 hypokalemia occupy the same laboratory range (<3.5-3.0 mmol/l) and are distinguished only by symptom status, which could not be reliably determined retrospectively, these two grades were combined; Grade 3 and Grade 4 thresholds for K, Na, and Ca are defined by laboratory value alone and were graded without modification. Hospital electrolyte reference ranges *Normal K level: 3.5-5.1 mmol/l*Normal Mg level: 0.66-1.07 mmol/l*Normal Ca level: 2.15-2.55 mmol/l*Normal Na level: 135-145 mmol/l *Normal K level: 3.5-5.1 mmol/l *Normal K level: 3.5-5.1 mmol/l *Normal Mg level: 0.66-1.07 mmol/l *Normal Mg level: 0.66-1.07 mmol/l *Normal Ca level: 2.15-2.55 mmol/l *Normal Ca level: 2.15-2.55 mmol/l *Normal Na level: 135-145 mmol/l *Normal Na level: 135-145 mmol/l Statistical analysis Descriptive statistics were used to describe the data. Categorical variables were presented as percentages, and differences were analyzed using the chi-square test and Fisher’s exact test for expected cell counts <5. Continuous variables were presented as medians. To compare electrolyte levels at various intervals of L-AMB with baseline electrolyte levels, the Wilcoxon signed-rank test and Friedman test were employed. To examine the differences between abnormally distributed continuous variables and categorical variables, the Mann-Whitney test was utilized. Multivariate logistic regression analysis was utilized to examine the factors associated with electrolyte changes in patients receiving L-AMB. An a priori two-tailed level of significance was set at 0.05. Data were analyzed using Stata software version 18.1 (STATA Corporation, College Station, TX, USA). Results Patients’ demographics and clinical characteristics A total of 237 adult patients received L-AMB in the past 10 years at SQUH. Out of the 237 patients, 94 patients met the inclusion criteria. The majority of the patients were male (59/94), and their median age was 38.5 years. Table 1 describes the patients’ demographics and clinical characteristics. Out of 94 patients, 64 patients had cancer as a comorbidity. 44 patients had sepsis.Table 1Patients’ demographics and clinical characteristics.Table 1 dummy alt textMale, number (%)59 (62.8)Age, median (IQR), years38.5 (29-55)Urea, median (IQR), mmol/l4.7 (3-7)Creatinine, median (IQR), µmol/l53.5 (41-69)eGFR, median (IQR), mL/min/1.73 m²112 (90-132)Cardiovascular disease, number (%)5 (5.3)Liver disease, number (%)15 (16.0)Hypertension, number (%)17 (18.1)Diabetes, number (%)16 (17.0)Cancer, number (%)64 (68.0)Diarrhea, number (%)18 (19.1)On total parenteral nutrition, number (%)12 (12.8)On nasogastric tube, number (%)20 (21.3)Vomiting, number (%)22 (23.4)Sepsis, number (%)44 (46.8)Dose/kg, median (IQR), mg/kg5 (4.6-5.3)Indication, number of patients (%)Empiric useAspergillosis treatmentMucormycosisCandidiasisOthers44 (46.8)27 (28.7)13 (13.8)5 (5.3)5 (5.3)Duration, median (IQR), days18 (11-30)Concomitant antifungal agent, number of patients (%)CaspofunginAnidulafunginVoriconazolePosaconazoleIsavuconazole16 (17.0)11 (11.7)4 (4.3)1 (1.1)1 (1.1)Concomitant medications, number of patients (%)Nonsteroidal anti-inflammatory drugsVancomycinPiperacillin-tazobactamSteroidsAcyclovirAminoglycosidesDiureticsFluoroquinolonesInsulinCalcineurin inhibitorsAngiotensin-converting enzyme inhibitorsPlatinum4 (4.3)39 (41.5)22 (23.4)29 (30.9)28 (29.8)9 (9.6)25 (26.6)8 (8.5)22 (23.4)2 (2.1)1 (1.1)0 (0)Allergy to other antifungal agents, number (%)2 (2.1)Abbreviations: eGFR, estimated glomerular filtration rate; IQR, interquartile range. Table 1Patients’ demographics and clinical characteristics.Table 1 dummy alt textMale, number (%)59 (62.8)Age, median (IQR), years38.5 (29-55)Urea, median (IQR), mmol/l4.7 (3-7)Creatinine, median (IQR), µmol/l53.5 (41-69)eGFR, median (IQR), mL/min/1.73 m²112 (90-132)Cardiovascular disease, number (%)5 (5.3)Liver disease, number (%)15 (16.0)Hypertension, number (%)17 (18.1)Diabetes, number (%)16 (17.0)Cancer, number (%)64 (68.0)Diarrhea, number (%)18 (19.1)On total parenteral nutrition, number (%)12 (12.8)On nasogastric tube, number (%)20 (21.3)Vomiting, number (%)22 (23.4)Sepsis, number (%)44 (46.8)Dose/kg, median (IQR), mg/kg5 (4.6-5.3)Indication, number of patients (%)Empiric useAspergillosis treatmentMucormycosisCandidiasisOthers44 (46.8)27 (28.7)13 (13.8)5 (5.3)5 (5.3)Duration, median (IQR), days18 (11-30)Concomitant antifungal agent, number of patients (%)CaspofunginAnidulafunginVoriconazolePosaconazoleIsavuconazole16 (17.0)11 (11.7)4 (4.3)1 (1.1)1 (1.1)Concomitant medications, number of patients (%)Nonsteroidal anti-inflammatory drugsVancomycinPiperacillin-tazobactamSteroidsAcyclovirAminoglycosidesDiureticsFluoroquinolonesInsulinCalcineurin inhibitorsAngiotensin-converting enzyme inhibitorsPlatinum4 (4.3)39 (41.5)22 (23.4)29 (30.9)28 (29.8)9 (9.6)25 (26.6)8 (8.5)22 (23.4)2 (2.1)1 (1.1)0 (0)Allergy to other antifungal agents, number (%)2 (2.1)Abbreviations: eGFR, estimated glomerular filtration rate; IQR, interquartile range. Patients’ demographics and clinical characteristics. Abbreviations: eGFR, estimated glomerular filtration rate; IQR, interquartile range. L-AMB dosing, indication, and concomitant medications Table 1 shows that the main indication for L-AMB was empiric use (44/94), followed by aspergillosis treatment (27/94). It was mostly prescribed as 5 mg/kg daily for 18 days. 41.5% and 30.9% of the patients were on concomitant vancomycin and steroids. L-AMB was the first option to treat the fungal infections in some patients because 2.1% of them had allergies to other antifungal agents. Hydration during L-AMB therapy The majority of the patients were hydrated before L-AMB therapy (70/94). The most common hydration fluids were 0.9% normal saline and dextrose. Out of the 94 patients, 47 received 0.9% normal saline during L-AMB therapy (Table 2).Table 2Hydration during L-AMB therapy.Table 2 dummy alt textPre-therapy hydration, number (%)70 (74.5)Not hydrated, number (%)24 (25.5)Volume of hydration, median (IQR), mL500 (480-1920)Hydration with:1. 0.9% normal saline2. Dextrose3. Normal saline + dextrose4. Ringer lactate5. 0.45% normal salineBefore L-AMB therapy509920During therapy4725541Abbreviation: IQR, interquartile range; L-AMB, liposomal amphotericin B. Table 2Hydration during L-AMB therapy.Table 2 dummy alt textPre-therapy hydration, number (%)70 (74.5)Not hydrated, number (%)24 (25.5)Volume of hydration, median (IQR), mL500 (480-1920)Hydration with:1. 0.9% normal saline2. Dextrose3. Normal saline + dextrose4. Ringer lactate5. 0.45% normal salineBefore L-AMB therapy509920During therapy4725541Abbreviation: IQR, interquartile range; L-AMB, liposomal amphotericin B. Hydration during L-AMB therapy. Abbreviation: IQR, interquartile range; L-AMB, liposomal amphotericin B. Baseline electrolytes (K, Mg, Ca, Na) during L-AMB therapy and on days 3, 5, and 7 Out of the 94 patients, 67, 67, 70, and 57 patients had normal baseline K, Mg, Ca, and Na, respectively (Figure 1). As shown in Table 3, the median levels of K, Mg, Ca, and Na were 3.9, 0.77, 2.29, and 137 mmol/l, respectively.Figure 1Electrolyte levels following L-AMB. L-AMB, liposomal amphotericin B.Figure 1 dummy alt textTable 3Electrolytes before and during L-AMB treatment.Table 3 dummy alt textDay 0Day 3Day 5Day 7Friedman Q (df)P-valuePotassium, median (IQR), mmol/l3.9(3.4-4.3)3.5(3.0-4.0)3.4(3.0-4.0)3.3(3.0-3.8)23.5256 (3)<0.0001Magnesium, median (IQR), mmol/l0.77(0.67-0.85)0.75(0.65-0.84)0.77(0.69-0.88)0.77(0.67-0.86)4.4865(3)0.2135Calcium, median (IQR), mmol/l2.29(2.15-2.35)2.23(2.15-2.32)2.28(2.19-2.35)2.30(2.20-2.37)12.1047 (3)0.0070Sodium, median (IQR), mmol/l137(134-140)138(135-140)139(135-143)140(137-144)25.6730 (3)<0.0001Abbreviations: df, degrees of freedom; IQR, interquartile range; L-AMB, liposomal amphotericin B. Electrolyte levels following L-AMB. L-AMB, liposomal amphotericin B. Table 3Electrolytes before and during L-AMB treatment.Table 3 dummy alt textDay 0Day 3Day 5Day 7Friedman Q (df)P-valuePotassium, median (IQR), mmol/l3.9(3.4-4.3)3.5(3.0-4.0)3.4(3.0-4.0)3.3(3.0-3.8)23.5256 (3)<0.0001Magnesium, median (IQR), mmol/l0.77(0.67-0.85)0.75(0.65-0.84)0.77(0.69-0.88)0.77(0.67-0.86)4.4865(3)0.2135Calcium, median (IQR), mmol/l2.29(2.15-2.35)2.23(2.15-2.32)2.28(2.19-2.35)2.30(2.20-2.37)12.1047 (3)0.0070Sodium, median (IQR), mmol/l137(134-140)138(135-140)139(135-143)140(137-144)25.6730 (3)<0.0001Abbreviations: df, degrees of freedom; IQR, interquartile range; L-AMB, liposomal amphotericin B. Electrolytes before and during L-AMB treatment. Abbreviations: df, degrees of freedom; IQR, interquartile range; L-AMB, liposomal amphotericin B. A Friedman test was conducted to evaluate changes in electrolyte levels across multiple time points post-exposure. Results indicated significant differences for K (χ²(3) = 23.53, P < 0.001), Na (χ²(3) = 25.67, P < 0.001), and Ca (χ²(3) = 12.10, P = 0.007), but not for Mg (χ²(3) = 4.49, P = 0.214) (Table 3). L-AMB therapy was associated with a significant decline in K levels beginning at day 3, with median K decreasing from 3.9 mmol/l at baseline to 3.5 mmol/l at day 3, 3.4 mmol/l at day 5, and 3.3 mmol/l at day 7. Na levels showed a gradual increase over time (Table 3). When graded according to CTCAE v6.0, grade 3-4 hypokalemia was frequent and persisted throughout therapy, affecting 23.7% of patients at day 3, 20.7% at day 5, and 22.0% at day 7. Grade 3-4 hypernatremia was less common but increased over the course of treatment, from 0% at day 3 to 2.2% at day 5 and 4.4% at day 7, including one patient (1.1%) who developed Grade 4 hypernatremia (Na >160 mmol/l) by day 7. Grade 3 hypocalcemia occurred in a single patient (1.3%) at day 3, with no Grade 4 hypocalcemia observed at any time point (Table 4).Table 4Electrolyte grade (CTCAE v6.0 criterion).Table 4 dummy alt textElectrolyte grade (CTCAE v6.0 criterion, mmol/l)Day 3, n (%)Day 5, n (%)Day 7, n (%)HypokalemiaGrade 1-2 hypokalemia (<3.5 to 3.0)21 (22.6%)31 (33.7%)30 (33.0%)Grade 3 hypokalemia (<3.0 to 2.5)21 (22.6%)17 (18.5%)15 (16.5%)Grade 4 hypokalemia (<2.5)1 (1.1%)2 (2.2%)5 (5.5%)HypernatremiaGrade 1 hypernatremia (>145 to 150)5 (5.4%)7 (7.7%)9 (9.9%)Grade 2 hypernatremia (>150 to 155)5 (5.4%)4 (4.4%)3 (3.3%)Grade 3 hypernatremia (>155 to 160)0 (0%)2 (2.2%)3 (3.3%)Grade 4 hypernatremia (>160)0 (0%)0 (0%)1 (1.1%)HypocalcemiaGrade 1 hypocalcemia (<2.15 to 2.00)14 (18.7%)9 (12.3%)9 (12.0%)Grade 2 hypocalcemia (<2.00 to 1.75)1 (1.3%)1 (1.4%)3 (4.0%)Grade 3 hypocalcemia (<1.75 to 1.50)1 (1.3%)0 (0%)0 (0%)Grade 4 hypocalcemia (<1.50)0 (0%)0 (0%)0 (0%)Abbreviation: CTCAE, Common Terminology Criteria for Adverse Events. Table 4Electrolyte grade (CTCAE v6.0 criterion).Table 4 dummy alt textElectrolyte grade (CTCAE v6.0 criterion, mmol/l)Day 3, n (%)Day 5, n (%)Day 7, n (%)HypokalemiaGrade 1-2 hypokalemia (<3.5 to 3.0)21 (22.6%)31 (33.7%)30 (33.0%)Grade 3 hypokalemia (<3.0 to 2.5)21 (22.6%)17 (18.5%)15 (16.5%)Grade 4 hypokalemia (<2.5)1 (1.1%)2 (2.2%)5 (5.5%)HypernatremiaGrade 1 hypernatremia (>145 to 150)5 (5.4%)7 (7.7%)9 (9.9%)Grade 2 hypernatremia (>150 to 155)5 (5.4%)4 (4.4%)3 (3.3%)Grade 3 hypernatremia (>155 to 160)0 (0%)2 (2.2%)3 (3.3%)Grade 4 hypernatremia (>160)0 (0%)0 (0%)1 (1.1%)HypocalcemiaGrade 1 hypocalcemia (<2.15 to 2.00)14 (18.7%)9 (12.3%)9 (12.0%)Grade 2 hypocalcemia (<2.00 to 1.75)1 (1.3%)1 (1.4%)3 (4.0%)Grade 3 hypocalcemia (<1.75 to 1.50)1 (1.3%)0 (0%)0 (0%)Grade 4 hypocalcemia (<1.50)0 (0%)0 (0%)0 (0%)Abbreviation: CTCAE, Common Terminology Criteria for Adverse Events. Electrolyte grade (CTCAE v6.0 criterion). Abbreviation: CTCAE, Common Terminology Criteria for Adverse Events. Patients with cancer had significantly lower baseline K and Ca compared with patients without cancer. The median baseline K was 3.7 mmol/l (interquartile range [IQR], 3.3-4.2) in patients with cancer vs 4.3 mmol/l (IQR 3.9-4.7) in patients without cancer (P = 0.001). Median baseline Ca was 2.26 mmol/l (IQR 2.15-2.33) vs 2.34 mmol/l (IQR 2.25-2.37) (P = 0.02). Median baseline magnesium was 0.76 mmol/l (IQR 0.64-0.85) vs 0.80 mmol/l (IQR 0.72-0.91) (P = 0.086). Median baseline sodium was 137 mmol/l (IQR 134-141) vs 136 mmol/l (IQR 133-138) (P = 0.21). There were no significant differences in baseline Mg or Na between the two groups. There was no significant drop in K levels in patients receiving nephrotoxic medications such as acyclovir, vancomycin, and nonsteroidal anti-inflammatory drugs (Table 5).Table 5Concomitant acyclovir (AC), vancomycin (VAN), or nonsteroidal anti-inflammatory drugs (NS) and potassium levels during L-AMB.Table 5 dummy alt textPotassium level, median (mmol/l)ACNo ACP-valueVANNo VANP-valueNSNo NSP-valueDay 03.654.050.083.903.900.394.403.900.09Day 33.503.550.773.403.600.303.853.500.48Day 53.203.500.413.303.450.253.153.400.68Day 73.303.400.983.503.300.473.003.400.29Abbreviations: AC, acyclovir; L-AMB, liposomal amphotericin B; NS, nonsteroidal anti-inflammatory drugs; VAN, vancomycin. Table 5Concomitant acyclovir (AC), vancomycin (VAN), or nonsteroidal anti-inflammatory drugs (NS) and potassium levels during L-AMB.Table 5 dummy alt textPotassium level, median (mmol/l)ACNo ACP-valueVANNo VANP-valueNSNo NSP-valueDay 03.654.050.083.903.900.394.403.900.09Day 33.503.550.773.403.600.303.853.500.48Day 53.203.500.413.303.450.253.153.400.68Day 73.303.400.983.503.300.473.003.400.29Abbreviations: AC, acyclovir; L-AMB, liposomal amphotericin B; NS, nonsteroidal anti-inflammatory drugs; VAN, vancomycin. Concomitant acyclovir (AC), vancomycin (VAN), or nonsteroidal anti-inflammatory drugs (NS) and potassium levels during L-AMB. Abbreviations: AC, acyclovir; L-AMB, liposomal amphotericin B; NS, nonsteroidal anti-inflammatory drugs; VAN, vancomycin. The univariate analysis identified several significant associations between medication-related factors and electrolyte abnormalities. Higher dose per kilogram was significantly associated with increased odds of hypokalemia (odds ratio [OR] 2.94, 95% confidence interval [CI] 1.35-6.40, P = 0.006), and patients receiving pre-infusion hydration had approximately six-fold higher odds of hypokalemia (OR 5.83, 95% CI 1.96-17.38, P = 0.002). Diuretics demonstrated a very strong association with hypernatremia (OR 8.26, 95% CI 2.82-24.24, P < 0.001) and were also significantly associated with hypocalcemia (OR 2.87, 95% CI 1.05-7.82, P = 0.039). In the multivariate models, dose/kg remained an independent predictor of hypokalemia (adjusted odds ratio [aOR] 3.80, P = 0.011), and hydration remained independently significant for hypokalemia (aOR 7.63, P = 0.005). Discussion L-AMB remains an important antifungal in high-risk patients because it substantially reduces nephrotoxicity compared with conventional amphotericin B deoxycholate. Nonetheless, clinically important electrolyte disturbances persist. In this retrospective cohort of 94 adult patients treated with L-AMB at SQUH, there were significant changes in serum K, Na, and Ca during the first 7 days of therapy; Mg concentrations did not change significantly. The majority were male patients with cancer with a median age of 38.5 years, many presenting with sepsis and receiving empiric L-AMB therapy at an median dose of 5 mg/kg over 18 days. The high prevalence of cancer (68%) and sepsis (46.8%) among patients aligns with existing literature showing L-AMB use predominantly in immunocompromised or critically ill patients [7]. The most striking finding in this study was the progressive and significant decline in serum K levels following L-AMB initiation. Median K fell from 3.9 mmol/l at baseline to 3.3 mmol/l by day 7 (P < 0.001), consistent with the known mechanism of tubular injury induced by amphotericin B. The drug increases membrane permeability of distal renal tubular epithelial cells, leading to K leakage into the urine and resultant hypokalemia. This tubular effect, while attenuated in the liposomal formulation relative to conventional amphotericin B deoxycholate, remains clinically relevant. The incidence and severity of L-AMB-induced hypokalemia reported in prior studies have been variable, ranging from 12% to over 53%. Usami et al. [8] reported a hypokalemia rate of 53.8% in adult patients, with recovery of potassium levels 2-3 days after treatment completion. Our findings of progressive decline beginning as early as day 3 are consistent with these reports and underscore the importance of proactive electrolyte monitoring from the very early days of therapy [5,8,9]. Persistent hypokalemia may cause convoluted tubule degeneration and subsequent AKI [10]. It is also of clinical significance as it predisposes patients to cardiac arrhythmias and other cardiac events such as ventricular tachycardia and atrial fibrillation. However, a study by Ota et al. [11] found no significant association between K supplementation during L-AMB and the occurrence of AKI. Usually, early K supplementation is recommended in case of a drop in K (before a fall < 2.8 mmol/l) to prevent the development of severe hypokalemia [5], [7], [8], [10], [11]. Importantly, our multivariate analysis identified two independent predictors of hypokalemia: higher L-AMB dose per kilogram (aOR 3.80, P = 0.011) and pre-infusion hydration (aOR 7.63, P = 0.005). The dose-dependent association is biologically plausible, as higher drug exposure would be expected to produce greater tubular injury. This relationship has been described for conventional amphotericin B and, to a varying extent, for liposomal formulations. Notably, some prior pharmacokinetic studies have not demonstrated a strictly dose-dependent relationship with hypokalemia for L-AMB specifically; however, our findings from a clinically diverse real-world population, which included doses predominantly at 5 mg/kg, reinforce the need for vigilance at higher weight-adjusted doses. The association between pre-infusion hydration and hypokalemia is a somewhat counterintuitive finding that merits careful interpretation. Hydration with normal saline (0.9% sodium chloride) is widely recommended as a nephroprotective strategy during L-AMB therapy and has been shown to reduce the risk of AKI through attenuation of tubuloglomerular feedback and renal vasoconstriction. However, saline loading in the context of preserved tubular flow may paradoxically enhance the delivery of Na to the distal nephron, augmenting the tubular exchange of Na for K and worsening kaliuresis. An alternative explanation is that patients who received pre-infusion hydration may represent those with more severe underlying illness or longer treatment courses, introducing residual confounding. In the AMBITION trial of single high-dose (10 mg/kg) L-AMB for human immunodeficiency virus-associated cryptococcal meningitis, a pre-emptive approach combining hydration with scheduled K (and, where available, Mg) supplementation and structured laboratory monitoring was associated with a low incidence of severe electrolyte disturbance despite high cumulative drug exposure [12]. The 2022 World Health Organization guideline for cryptococcal disease similarly recommends pre-emptive electrolyte supplementation alongside hydration during L-AMB induction therapy. Clinicians should therefore not interpret this finding as a reason to withhold hydration, which remains a cornerstone of L-AMB nephroprotection; rather, pre-hydrated patients may warrant more intensive K monitoring and earlier supplementation [2,[12], [13], [14]]. Concomitant nephrotoxic medications, including vancomycin, acyclovir, and nonsteroidal anti-inflammatory drugs, were not significantly associated with further K decline in our cohort. This may reflect the relatively well-managed clinical environment, the moderate sample size, or the particular population studied. Prior studies have similarly found that, while concomitant nephrotoxins increase the risk of AKI, their independent effect on L-AMB-associated hypokalemia is less clearly established [2,6,13]. Serum Na levels showed a gradual but statistically significant increase over the study period (P < 0.001), rising from a median of 137 mmol/l at baseline to 140 mmol/l by day 7. The rise in sodium is likely multifactorial. Amphotericin B has long been recognized as one of the drug classes most frequently implicated in hypernatremia, secondary to its effect on renal concentrating ability through tubular injury and impaired antidiuretic hormone-mediated water reabsorption. In addition, the routine use of 0.9% normal saline as hydration fluid (received by 47 of 94 patients in this cohort) may have contributed a sodium load that outpaced renal excretory capacity in some patients. Diuretic use, present in 26.6% of patients, was identified as the strongest independent predictor of hypernatremia in our multivariate model (aOR 15.6, P < 0.001), a finding consistent with the known capacity of loop diuretics to promote free water loss in excess of sodium, resulting in a net hypernatremic state [15]. Ca levels showed a significant but nonlinear trend across time points (P = 0.007), with a modest dip at day 3 followed by a partial recovery toward baseline. The clinical significance of this transient decline is uncertain but may reflect early tubular dysfunction impairing calcium reabsorption. Unlike K, Mg levels did not exhibit a statistically significant change across the study period (P = 0.214). This finding contrasts with the extensive literature on conventional amphotericin B deoxycholate, where hypomagnesemia is a frequently reported complication. The liposomal formulation may cause comparatively less tubular disruption with respect to Mg handling. It is also possible that the routine supplementation practices at SQUH, or the relatively short median treatment duration observed in some patients, attenuated detectable changes. Clinicians should nonetheless remain vigilant for hypomagnesemia in longer treatment courses, particularly given that uncorrected hypomagnesemia can worsen refractory hypokalemia by impairing renal K conservation [16]. A particularly notable finding was that patients with cancer had significantly lower baseline K (3.7 vs 4.3 mmol/l, P = 0.001) and Ca (2.26 vs 2.34 mmol/l, P = 0.02) compared with patients without cancer. This is consistent with the well-established vulnerability of patients with cancer to electrolyte dysregulation. Multiple mechanisms contribute to this susceptibility, including malnutrition, malabsorption, and gastrointestinal losses from chemotherapy-related diarrhea and vomiting. Patients with cancer, who comprised 68% of this cohort, therefore represent a clinically important subgroup at compounded risk: preexisting electrolyte depletion may amplify the magnitude of L-AMB-induced hypokalemia and the clinical consequences thereof. These patients should be prioritized for early and aggressive electrolyte monitoring and replacement at the time of L-AMB initiation. Certain concomitant conditions, such as preexisting renal impairment or cardiac issues, may exacerbate these electrolyte abnormalities. Therefore, individualized monitoring protocols should be implemented. These protocols will guide both electrolyte and fluid management among these patients. This study has some limitations. Its retrospective single-center design limits causal inference, and the exclusion of patients with renal impairment or undergoing dialysis, and those who did not complete 7 days of therapy (n = 106) reduces the generalizability and clinical usefulness of the findings and may introduce survival bias, as patients who were sicker or who developed adverse effects earlier may have been disproportionately excluded. The relatively modest sample size limits the power of multivariate analyses, particularly for less common electrolyte endpoints. Additionally, data on K and Mg supplementation administered during L-AMB therapy were not systematically analyzed, which may have attenuated the observed electrolyte changes. The absence of urinary electrolyte data precludes a mechanistic differentiation between renal and extrarenal causes of electrolyte loss. Also, our study did not report any potential clinical consequences related to the electrolyte abnormalities. Finally, while this study adds important real-world data from a Middle Eastern tertiary care center, its generalizability to other healthcare settings may be limited. Conclusion This study confirms that L-AMB therapy is associated with significant and clinically meaningful electrolyte disturbances, most prominently progressive hypokalemia beginning as early as day 3, along with rising Na and transient changes in Ca. Higher dose per kilogram and pre-infusion hydration were independently associated with hypokalemia. Patients with cancer present with lower baseline electrolytes and are at compounded risk. These findings reinforce the need for structured electrolyte monitoring protocols and early supplementation strategies in all patients receiving L-AMB, particularly those with malignancy or receiving higher drug doses. Given the early onset of hypokalemia and the demonstrated efficacy of pre-emptive supplementation protocols in comparable high-dose L-AMB settings, adopting a pre-emptive, rather than reactive, K (and, where feasible, Mg) supplementation strategy alongside structured day 3/5/7 electrolyte monitoring should be considered. Future studies should examine these associations according to the baseline renal function in a larger multicenter cohort. CRediT authorship contribution statement Najwa Al Himali: Writing – review & editing, Writing – original draft, Validation, Software, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Hanan Al Habsi: Methodology. Marwa Al Harthi: Methodology. Shaikha Al Ghailani: Methodology. Dana Al Ghailani: Methodology. Leena Kashoub: Methodology. Aly M. Abdelrahman: Writing – review & editing, Supervision, Conceptualization. Declaration of competing interest The authors have no competing interests to declare.