Early-onset severe voriconazole-induced liver injury with progressive cholestasis in pulmonary aspergillosis: A case report

Abstract

Background

Voriconazole is the recommended first-line treatment for invasive pulmonary Aspergillosis. However, its nonlinear pharmacokinetics and marked interindividual variability increase the risk of drug-induced liver injury, particularly in older patients with multiple comorbidities.

Case presentation

A 76-year-old Asian man with chronic obstructive pulmonary disease, type 2 diabetes mellitus, hypertension, polycythemia, and glucocorticoid-induced adrenal insufficiency was admitted with an infective exacerbation of COPD. Sputum culture identified Aspergillus fumigatus, and oral voriconazole (200 mg twice daily) was initiated after initial itraconazole therapy.Four days after starting voriconazole, the patient developed jaundice and severe liver injury, with alanine aminotransferase of 1370 U/L, aspartate aminotransferase of 389 U/L, gamma-glutamyl transferase of 1514 U/L, and total bilirubin of 90.3 μmol/L. Voriconazole was discontinued immediately. Viral, autoimmune, and mechanical causes of liver injury were excluded. Despite drug withdrawal and supportive therapy, cholestatic liver injury continued to progress, with peak total bilirubin of 592 μmol/L. Because of persistent clinical deterioration, therapeutic plasma exchange was performed. During subsequent follow-up, jaundice gradually improved, and liver biochemical parameters showed an overall recovery trend. Coagulation, renal function, and neurological status remained preserved throughout hospitalization.

Conclusions

This case highlights early-onset severe voriconazole-induced liver injury with subsequent progressive cholestasis in an older patient with multiple comorbidities. Liver injury may continue to worsen despite prompt drug discontinuation. Careful monitoring of liver function during the first 1-2 weeks of voriconazole therapy and therapeutic drug monitoring, when available, may facilitate earlier recognition of hepatotoxicity and improve patient safety.

1. Introduction

Voriconazole is a second-generation triazole antifungal agent with broad-spectrum activity and remains the recommended first-line therapy for invasive aspergillosis in current international clinical guidelines because of its proven efficacy and favorable antifungal spectrum [1] However, its clinical use is complicated by nonlinear pharmacokinetics, a narrow therapeutic window, and substantial interindividual variability in systemic exposure. Because voriconazole is predominantly metabolized by hepatic cytochrome P450 enzymes, particularly CYP2C19 serum drug concentrations may be markedly influenced by genetic polymorphisms, age, systemic inflammation, liver function, and drug-drug interactions [2,3].

Drug-induced liver injury (DILI) is one of the most clinically important adverse effects associated with voriconazole. In a large retrospective cohort, Lou et al. (2025) reported an overall incidence of voriconazole-associated liver injury of approximately 11%, with hepatocellular and cholestatic injury accounting for the majority of cases [4]. Likewise, Nandan et al. (2025) identified abnormal liver function tests as one of the most frequent adverse events during voriconazole therapy. Most patients experience mild and reversible biochemical abnormalities after dose adjustment or drug discontinuation [5]. However, reports describing rapidly progressive cholestatic liver injury with marked hyperbilirubinemia and jaundice remain limited in the published literature [1,4].

Older patients with chronic respiratory diseases may be particularly susceptible to voriconazole-induced hepatotoxicity because of age-related reductions in hepatic metabolic capacity, multiple comorbidities, polypharmacy, and systemic inflammation. Consistent with this, Xiao et al. (2023) identified advanced age and several clinical factors as independent predictors of voriconazole-related liver injury, while Feng et al. (2026) developed a predictive model to facilitate early identification of patients at increased risk [6,7].

We report an unusual case of early-onset severe voriconazole-induced liver injury that developed only four days after treatment initiation in a 76-year-old man with chronic obstructive pulmonary disease following respiratory isolation of Aspergillus fumigatus. Despite immediate discontinuation of voriconazole and conventional supportive therapy, liver injury progressed from an initial hepatocellular pattern to severe cholestasis with profound hyperbilirubinemia, ultimately necessitating therapeutic plasma exchange. This case highlights the importance of early liver function monitoring, recognition of delayed cholestatic progression despite drug withdrawal, and individualized management of severe voriconazole-associated hepatotoxicity.

2. Case presentation

A 76-year-old man was admitted to the Institute of Respiratory Medicine, Bach Mai Hospital, because of progressively worsening dyspnea, increased production of purulent sputum, and a low-grade fever (approximately 38°C) for three days. He had been diagnosed with chronic obstructive pulmonary disease (COPD) four years earlier and was receiving maintenance inhaled therapy and long-term home oxygen therapy. His medical history included essential hypertension, type 2 diabetes mellitus, polycythemia, and glucocorticoid-induced adrenal insufficiency secondary to prolonged systemic corticosteroid exposure. He was a former long-term smoker of cigarettes and traditional water-pipe tobacco.

The initial diagnosis was an infective exacerbation of COPD complicated by acute respiratory failure. He received noninvasive ventilation using bilevel positive airway pressure (BiPAP) in combination with medications including piperacillin/tazobactam, levofloxacin, methylprednisolone, and nebulized ipratropium/salbutamol. Chest computed tomography demonstrated bronchial wall thickening, scattered ground-glass opacities, tree-in-bud nodules, bilateral upper-lobe fibrotic calcified sequelae, and diffuse emphysema (Fig. 1).

Fig. 1.

Chest CT showing scattered tree-in-bud nodules and ground-glass opacities suggestive of bronchiolitis, with bilateral upper-lobe calcified fibrotic sequelae and diffuse emphysema.

3. Diagnostic work-up for probable invasive pulmonary aspergillosis

On the first hospital day, direct microscopic examination of sputum demonstrated septate, acute-angle branching fungal hyphae morphologically consistent with Aspergillus species. Repeated fungal cultures subsequently yielded Aspergillus fumigatus. Antifungal susceptibility testing demonstrated susceptibility to voriconazole (MIC 0.5 μg/mL), with wild-type susceptibility to itraconazole (MIC 0.06 μg/mL) and amphotericin B (MIC 2 μg/mL).

Repeated sputum bacterial cultures also isolated carbapenem-resistant Acinetobacter baumannii (CRAB). Antimicrobial susceptibility testing demonstrated resistance to ampicillin/sulbactam, piperacillin/tazobactam, ceftazidime, cefepime, imipenem, meropenem, gentamicin, amikacin, ciprofloxacin, levofloxacin, and trimethoprim-sulfamethoxazole. Minocycline showed intermediate susceptibility, whereas colistin demonstrated an MIC of 0.5 μg/mL.

Extensive investigations were performed to exclude alternative causes of liver disease and other underlying disorders. Serologic testing for viral hepatitis was negative, including hepatitis A virus IgM, hepatitis B surface antigen, hepatitis C antibody, and hepatitis E virus IgM. Additional viral studies, including hepatitis B core antibody and herpes simplex virus (HSV)-1/2 IgM, were also negative.

Autoimmune screening was entirely negative, including antinuclear antibody (ANA), anti-double-stranded DNA (anti-dsDNA), anti-liver kidney microsomal type 1 antibody (anti-LKM1), antimitochondrial antibody M2 (AMA-M2), anti-asialoglycoprotein receptor antibody (ASGPR), myeloperoxidase antineutrophil cytoplasmic antibody (MPO/pANCA), and proteinase-3 antineutrophil cytoplasmic antibody (PR3/cANCA).

Serological tests for parasitic infections, including Strongyloides stercoralis, Echinococcus granulosus, Schistosoma spp., Cysticercus cellulosae, Toxocara spp., and Fasciola spp., were all negative.

Given the patient's severe COPD requiring prolonged corticosteroid therapy, compatible chest CT findings, direct microscopic evidence of septate fungal hyphae, and repeated isolation of A. fumigatus from respiratory specimens, the diagnosis of probable invasive pulmonary aspergillosis (IPA) was established according to the Bulpa diagnostic criteria for COPD-associated IPA.

The patient was initially treated with oral itraconazole (200 mg/day). Baseline liver biochemistry before voriconazole therapy was unremarkable. On March 26, serum aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were 21 U/L and 29 U/L, respectively. On April 1, immediately before initiation of voriconazole, AST was 40 U/L and ALT was 87 U/L. Based on antifungal susceptibility testing and current guideline recommendations, itraconazole was replaced with oral voriconazole at 200 mg twice daily (400 mg/day).

4. Development of voriconazole-induced liver injury

Four days after initiation of voriconazole, the patient developed progressive fatigue, marked scleral icterus, and jaundice. Laboratory evaluation revealed ALT 1370 U/L, AST 389 U/L, gamma-glutamyl transferase (GGT) 1514 U/L, alkaline phosphatase (ALP) 156 U/L, total bilirubin 90.3 μmol/L, and direct bilirubin 54.3 μmol/L. The R-value (ratio of ALT to ALP) was used to differentiate between hepatocellular and cholestatic patterns of liver injury. It is calculated using the formula:

$$R=\left(\frac{ALT}{ULNofALT}\right)÷\left(\frac{ALP}{ULNofALP}\right)$$

where ULN is the upper limit of normal.

Based on the R-value, an R ≤ 2 indicates a cholestatic pattern, R > 2 to < 5 suggests a mixed pattern and R ≥ 5 indicates hepatocellular injury. In this patient, the R-value was consistent with a cholestatic pattern of liver injury [8]. Using an upper limit of normal of 45 U/L for ALT and 129 U/L for ALP, the calculated R ratio was approximately 25, indicating an initial hepatocellular pattern of drug-induced liver injury. Voriconazole and levofloxacin were immediately discontinued, and the patient received intravenous N-acetylcysteine together with hepatoprotective therapy (silymarin).

Magnetic resonance imaging with magnetic resonance cholangiopancreatography demonstrated normal liver morphology without focal lesions. The intrahepatic and extrahepatic bile ducts were not dilated, and no choledocholithiasis, biliary stricture, biliary tract obstruction, or hepatic vascular abnormality was identified. The pancreas and spleen appeared normal, and only incidental bilateral simple renal cysts were observed. These findings effectively excluded obstructive hepatobiliary disease as the cause of jaundice.

The diagnosis of DILI was further supported by the Roussel Uclaf Causality Assessment Method (RUCAM), which yielded a score of 8, indicating a probable association with voriconazole. RUCAM is a validated and widely used tool for assessing drug-induced liver injury (DILI) and herb-induced liver injury (HILI) [9].

5. Progressive cholestatic liver injury after drug withdrawal

Despite prompt discontinuation of voriconazole, liver injury continued to progress. Although serum aminotransferase concentrations gradually declined, cholestatic parameters worsened progressively. Five days after drug withdrawal, AST had decreased to 125 U/L and ALT to 790 U/L, whereas GGT increased to 2091 U/L, ALP to 236 U/L, total bilirubin to 154.3 μmol/L, and direct bilirubin to 104.7 μmol/L.

By April 13, total bilirubin had increased to 382.7 μmol/L and direct bilirubin to 362.0 μmol/L. On April 14, the patient reached peak biochemical abnormalities, including total bilirubin 592 μmol/L, direct bilirubin 397 μmol/L, GGT 2713 U/L, ALP 313 U/L, AST 313 U/L, ALT 730 U/L, and blood ammonia 70 μmol/L.

Despite profound hyperbilirubinemia and severe biochemical liver injury, the patient remained mentally alert without evidence of hepatic encephalopathy. Hepatic synthetic function was preserved, with a prothrombin time of 104% and an international normalized ratio (INR) of 0.97. Accordingly, the clinical presentation was consistent with severe acute liver injury rather than acute liver failure.

Because of progressive cholestatic liver injury despite prompt withdrawal of the suspected offending drug and supportive medical therapy, the patient was transferred to the Poison Control Center for further management.

5.1. Therapeutic intervention

The patient underwent therapeutic plasma exchange. The decision was made in the context of severe and progressively worsening hyperbilirubinemia, persistent liver biochemical abnormalities, and concern regarding further clinical deterioration.

Following two plasma-exchange sessions, the biochemical abnormalities began to improve. Aspartate aminotransferase decreased to 36 U/L and alanine aminotransferase to 162 U/L. Total bilirubin decreased from 592 μmol/L to 363.5 μmol/L, while direct bilirubin decreased to 315.1 μmol/L.

5.2. Follow-up and outcomes

The patient's general condition gradually improved. He remained alert and fully oriented and developed no clinical evidence of hepatic encephalopathy. Jaundice and scleral icterus progressively decreased. He reported no abdominal pain, and bowel movements and urine output remained normal.

His respiratory condition also stabilized. Dyspnea resolved, respiratory rate was approximately 18 breaths/min, and peripheral oxygen saturation was maintained at 95–97% while breathing room air.

Renal function remained normal, with serum creatinine of 62 μmol/L and urea of 4.2 mmol/L. Coagulation parameters remained within the normal range. The temporal changes in the patient's liver biochemical parameters are summarized in Table 1, Fig. 2.

Table 1.

Clinical timeline, liver biochemical changes, and management.

Timeline	Clinical events	Key liver biochemistry	Management
Admission	Hospitalized for an acute exacerbation of COPD. Sputum microscopy revealed filamentous fungi.	AST 21 U/L, ALT 29 U/L	Antibiotic: piperacillin/tazobactam, Levofloxacin Bronchodilator nebulization and non-invasive ventilation (BiPAP).
Day 1	Aspergillus fumigatus identified by sputum culture.	AST 40 U/L, ALT 87 U/L	Oral voriconazole initiated at 200 mg twice daily (400 mg/day).
Day 4	Progressive fatigue, jaundice, and scleral icterus developed.	AST 389 U/L, ALT 1370 U/L, GGT 1514 U/L, ALP 156 U/L, total bilirubin 90.3 μmol/L, direct bilirubin 54.3 μmol/L	Voriconazole and levofloxacin discontinued immediately. Intravenous acetylcysteine and silymarin initiated.
Day 9	Cholestasis progressed despite discontinuation of voriconazole.	AST 125 U/L, ALT 790 U/L, GGT 2091 U/L, ALP 236 U/L, total bilirubin 154.3 μmol/L, direct bilirubin 104.7 μmol/L	Continued supportive treatment.
Day 13	Progressive hyperbilirubinemia despite supportive therapy.	AST 151 U/L, ALT 653 U/L, GGT 2590 U/L, total bilirubin 382.7 μmol/L, direct bilirubin 362.0 μmol/L	Transferred to the Poison Control Center for advanced management.
Day 14	Peak liver injury with severe cholestasis.	AST 313 U/L, ALT 730 U/L, GGT 2713 U/L, ALP 313 U/L, total bilirubin 592.0 μmol/L, direct bilirubin 397.0 μmol/L	Therapeutic plasma exchange (PEX) performed.
Day 19	Marked clinical improvement with resolution of jaundice.	AST 36 U/L, ALT 162 U/L, total bilirubin 363.5 μmol/L, direct bilirubin 315.1 μmol/L	Clinical recovery following two sessions of therapeutic plasma exchange.

Abbreviations: ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; BiPAP, bilevel positive airway pressure; COPD, chronic obstructive pulmonary disease; GGT, gamma-glutamyl transferase; PEX, therapeutic plasma exchange.

Fig. 2.

Temporal changes in liver biochemical parameters during the clinical course.

6. Discussion

Voriconazole-induced liver injury is well recognized; however, the clinical presentation and severity vary considerably among patients. The present case is noteworthy for three reasons. First, liver injury developed only four days after initiation of standard-dose voriconazole, indicating a very early onset of hepatotoxicity. Second, although voriconazole was discontinued immediately after the onset of liver injury, the biochemical pattern evolved from marked hepatocellular injury to progressive cholestasis with profound hyperbilirubinemia. Third, because of persistent worsening despite drug withdrawal and supportive therapy, therapeutic plasma exchange (PEX) was performed as an individualized rescue intervention.

Although the exact mechanism of voriconazole-induced hepatotoxicity remains unclear, oxidative stress is considered a key driver. Excessive production of reactive oxygen species (ROS) during drug metabolism leads to hepatocyte injury and apoptosis. This damage triggers the release of damage-associated molecular patterns, activating innate immune responses and inflammation. A positive feedback loop between oxidative stress and inflammation ultimately results in hepatocellular dysfunction and clinical liver injury.

Recent real-world studies suggest that liver injury occurs in approximately one in ten patients receiving voriconazole, although reported incidences vary depending on the study population and diagnostic criteria [4]. Most cases are characterized by asymptomatic or mild elevations of aminotransferases that resolve after dose reduction or discontinuation of therapy [4,5]. Pharmacovigilance analyses have similarly identified hepatotoxicity as one of the most frequently reported adverse reactions associated with voriconazole [1]. In contrast, clinically apparent cholestatic liver injury with marked jaundice and severe hyperbilirubinemia has been described much less frequently and is largely limited to isolated case reports [10]. Gupta et al. (2025) reported a rare case of voriconazole-induced cholestatic hepatotoxicity in a patient with fungal pneumonia. In their report, cholestatic markers surged rapidly, with both ALP and GGT reaching peak levels of 1209 IU/L and 119 IU/L, respectively, on day 2 of voriconazole therapy. Following the discontinuation of voriconazole on day 4 and an immediate switch to posaconazole, these enzymes progressively declined and fully normalized by day 9 [11]. Our patient therefore represents an uncommon presentation characterized by rapid progression and persistent cholestasis despite early withdrawal of the suspected drug.

An interesting feature of this case is the dynamic evolution of the liver injury phenotype. At presentation, the calculated R ratio was approximately 25, indicating a hepatocellular pattern according to current DILI criteria [9,12]. However, serial biochemical testing demonstrated progressive increases in bilirubin, ALP, and GGTdespite declining aminotransferase concentrations. This evolution suggests that the initial hepatocellular injury was followed by clinically significant cholestatic injury. Similar temporal changes have been reported in DILI caused by azole antifungal agents, emphasizing that normalization of aminotransferases alone should not be interpreted as recovery when cholestatic indices continue to worsen. Consequently, serial monitoring of bilirubin and cholestatic enzymes is as important as monitoring aminotransferases during follow-up.

Several factors may have contributed to the development of severe hepatotoxicity in this patient. Voriconazole exhibits nonlinear pharmacokinetics, and systemic exposure may increase disproportionately even with standard dosing. In addition, its metabolism depends primarily on CYP2C19 making drug exposure susceptible to genetic polymorphisms, age, systemic inflammation, hepatic function, and drug-drug interactions [2,3]. Older age, multiple comorbidities, acute respiratory infection, and polypharmacy have been identified as important clinical risk factors for voriconazole-associated liver injury [6,7]. Although CYP2C19 genotyping and therapeutic drug monitoring were unavailable in this case, excessive voriconazole exposure remains a plausible explanation for the unusually rapid onset of toxicity.

Current international recommendations emphasize immediate discontinuation of the suspected drug as the cornerstone of management for DILI, together with exclusion of competing causes such as viral hepatitis, autoimmune liver disease, and biliary obstruction [12,13]. In our patient, extensive investigations, including viral serology, autoimmune testing, and magnetic resonance imaging, failed to identify an alternative etiology. Nevertheless, liver injury continued to progress after withdrawal of voriconazole, illustrating that biochemical deterioration may persist despite prompt removal of the offending agent.

To manage hepatotoxicity, several interventions were implemented, including N-acetylcysteine (NAC) infusion, ursodeoxycholic acid (UDCA), and plasma exchange (PEX). The efficacy of NAC has been evaluated in a randomized placebo-controlled trial involving patients with non–paracetamol-induced acute liver failure, which included a subgroup with DILI. The study demonstrated significantly higher transplant-free survival in patients receiving NAC compared with placebo, but only in those at early stages (I–II) of hepatic encephalopathy (58 [52%] vs 56 [30%]; p = 0.010) [14]. Therefore, both ACG and EASL recommend NAC administration in the early stages of liver failure. In the present case, NAC administration was also associated with improvement in hepatocellular injury, as reflected by the subsequent clinical and biochemical response [13]. In addition, given the cholestatic features of liver injury, UDCA was administered. In patients with cholestatic DILI, UDCA is frequently used due to its favorable safety profile, although no randomized controlled trials have yet confirmed its efficacy.

Given the rapidly progressive hyperbilirubinemia despite supportive treatment, PEX was selected as an individualized rescue therapy. The rationale for PEX is the removal of albumin-bound bilirubin, inflammatory mediators, and potentially toxic circulating substances. However, evidence supporting its use specifically for voriconazole-induced liver injury remains extremely limited, and no current guideline recommends PEX as standard therapy for this condition [12,13]. Furthermore, biochemical parameters in our patient continued to fluctuate after PEX, making it impossible to determine the extent to which subsequent recovery resulted from plasma exchange, delayed resolution after drug withdrawal, or the combined effect of supportive management. Therefore, this case should not be interpreted as evidence of the efficacy of PEX but rather as an example of an individualized therapeutic approach in a patient with progressive cholestatic liver injury.

This case has several important clinical implications. First, severe voriconazole-induced liver injury may occur within only a few days of treatment initiation, even at standard doses. Second, the pattern of liver injury may evolve over time, and progressive cholestasis may develop despite early discontinuation of the offending drug. Finally, in addition to routine monitoring of aminotransferases, serial assessment of bilirubin and cholestatic enzymes should be performed during the first weeks of voriconazole therapy, particularly in older patients with multiple risk factors. Whenever available, therapeutic drug monitoring may facilitate individualized dosing and potentially reduce the risk of severe hepatotoxicity.

Finally, our report has several limitations. Serum voriconazole concentrations and CYP2C19 genotyping were not available, precluding confirmation of excessive drug exposure. In addition, this is a single case report, and therefore no conclusion can be drawn regarding the effectiveness of PEX in voriconazole-induced liver injury. Nevertheless, the detailed temporal relationship between drug exposure, serial biochemical changes, exclusion of competing causes, and clinical follow-up provides valuable insight into an uncommon but clinically important manifestation of voriconazole-associated hepatotoxicity.

7. Conclusion

Voriconazole may cause severe liver injury within the first few days of treatment, particularly in older patients with acute systemic illness, chronic lung disease, multiple comorbidities, and polypharmacy. In this patient, the liver injury initially demonstrated a hepatocellular pattern and subsequently evolved into progressive cholestasis despite prompt voriconazole discontinuation. This course emphasizes that decreasing aminotransferase concentrations do not necessarily indicate resolution when bilirubin and cholestatic indices continue to worsen.

Baseline and frequent early monitoring of liver biochemical parameters should be performed during voriconazole therapy. Therapeutic drug monitoring should be used whenever available to support individualized dosing in high-risk patients. Therapeutic plasma exchange may be considered in selected cases with severe progressive hyperbilirubinemia, although its efficacy in voriconazole-induced liver injury requires further investigation.

CRediT authorship contribution statement

Phan Thanh Thuy: Conceptualization, Data curation, Formal analysis, Methodology, Resources, Software, Validation, Writing – original draft, Writing – review & editing. Hoang Thanh Tam: Data curation, Formal analysis, Software. Nguyen Thi Thu Ha: Data curation, Investigation, Methodology, Writing – original draft. Dang Minh Duc: Data curation, Formal analysis, Investigation, Writing – original draft. Nguyen Hoang Anh: Investigation, Methodology, Supervision, Writing – review & editing. Nguyen Thi Thanh Huyen: Investigation, Methodology. Nguyen Trung Nguyen: Investigation. Phan Thu Phuong: Supervision, Writing – review & editing. Vu Van Giap: Supervision, Writing – review & editing.

Ethics declaration

Written informed consent to take part in the study and to publish the article has been obtained from all participants or their legal representatives. The privacy rights of participants have been observed.

This study was performed in compliance with relevant laws, regulatory frameworks and guidelines where the research took place. This study was conducted in accordance with CARE guidelines (CAse REport Guidelines). Ethics committee approval was not required under relevant laws and institutional guidelines. This case report was conducted in accordance with applicable institutional requirements and the principles of the Declaration of Helsinki. Written informed consent for publication was obtained from the patient.

This research follows the CARE guidelines and the CARE checklist.

Ethics approval

According to the regulations of Bach Mai Hospital, institutional ethics committee approval was not required for the publication of this single-patient case report. The report was conducted in accordance with the principles of the Declaration of Helsinki.

Consent for publication

Written informed consent for publication of this case report and the accompanying clinical information was obtained from the patient. A copy of the written consent is available for review by the Editor-in-Chief of this journal upon reasonable request.

Availability of data and materials

All data generated or analyzed during this study are included in this published article. Additional anonymized data are available from the corresponding author upon reasonable request.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.