A Large Retrospective Assessment of Voriconazole Exposure in Patients Treated with Extracorporeal Membrane Oxygenation

Design

Retrospective, multicenter, observational, cohort study

N= 69

Objective

To assess the effect of extracorporeal membrane oxygenation (ECMO) on voriconazole systemic exposure in order to evaluate the need for dose adjustments

Study Groups

Voriconazole and ECMO (N= 69)

Inclusion Criteria

Adult patients hospitalized in the intensive care unit; treated with voriconazole and simultaneously received ECMO support during at least a part of the antifungal treatment; had at least one tough concentration of voriconazole taken during ECMO

Exclusion Criteria

Non-trough samples

Methods

This was a retrospective study from 8 different European institutions of patients who had voriconazole levels drawn during ECMO. There was no restriction on voriconazole indication or dose. A control included voriconazole concentrations in the same patients before or after ECMO.

Duration

Outcome Measures

Baseline Characteristics

All patients (N= 69)

Age, years (IQR)

Male

Length of ICU stay, days (range)

18 (14-23)

ICU mortality

Type of ECMO

VV-ECMO

VA-ECMO

Both

74%

10%

16%

32%

ECMO duration, days (IQR)

Results

Non-ECMO

Median trough concentration, mg/L (IQR)

Previous daily voriconazole dose, mg/kg (IQR)

2.4 (1.2-4.7)

9.2 (6.7-10.9)

2.5 (1.4-3.9)

8.1 (6.5-11.1)

0.58

0.76

Subtherapeutic trough (<2 mg/L)

Therapeutic trough (2-5.5 mg/L)

Supratheraputic trough (>5.5 mg/L)

Previous daily voriconazole dose, mg/kg (IQR)

56%

29%

15%

8.33 (6.6-10.9)

39%

52%

10%

8.0 (6.4-10.7)

0.80

0.37

0.40

0.84

Inter- and intrasubject variability (% coefficient of variation) in the voriconazole trough concentration (corrected for the dose) were 47% and 78% on sampling days under ECMO and 46% and 60% for non-ECMO sampling.

Adverse Events

Not studied

Study Author Conclusions

No significant ECMO effect was observed on voriconazole exposure. A large proportion of patients had voriconazole subtherapeutic concentrations.

InpharmD Researcher Critique

This study, while robust and including centers from 4 countries, is limited by its retrospective design. There was a low number of samples collected during the first three days of ECMO or after circuit change; the highest influence on voriconazole trough levels is expected during the first three days due to the absence of saturation at that point. The study did not report if the non-ECMO values were taken before or after EMCO, nor was the timeframe reported.

Impact of Extracorporeal Membrane Oxygenation on Voriconazole Plasma Concentrations

Design

Single-center, retrospective, observational cohort study

N= 132

Objective

To assess the impact of extracorporeal membrane oxygenation (ECMO) on voriconazole plasma exposure

Study Groups

ECMO group (n= 66)

Non-ECMO group (n= 66)

Inclusion Criteria

Critically ill adults with or without ECMO support receiving intravenous (IV) voriconazole therapy for Aspergillus infection and at least one voriconazole minimum blood plasma concentration (Cmin) lab result during treatment

Exclusion Criteria

Not discussed

Methods

Adult Chinese and Japanese patients who received ECMO support during at least part of voriconazole treatment and had available Cmin during ECMO were assigned to the ECMO group. Those who did not receive ECMO support during treatment, or voriconazole therapy and its therapeutic drug monitoring (TDM) were completed during non-ECMO periods were assigned as the non-ECMO group. Dosing of voriconazole was 6 mg/kg initial loading dose and every 12 hours on day 1, then 4 mg/kg every 12 hours for maintenance. The first TDM results of voriconazole were collected, and the prevalence of subtherapeutic concentrations was analyzed. TDM was performed 5-7 days after administration and a half-hour prior to subsequent administration under steady state conditions. Doses could be adjusted based on clinical reactions and TDM results. Lower limit of Cmin was defined as >2 mg/L. Spearman correlation coefficients were used to evaluate the relationship between initial voriconazole levels and continuious relevant variables. 

Duration

August 2017 to December 2021

Outcome Measures

To determine prevalence of subtherapeutic voriconazole

Comparison of clinical characteristics between study groups

Age, years

Female

Acute physiology and chronic health evaluation II (APACHE II)

Sequential organ failure assessment (SOFA)

0.000

Serum Creatinine (umol/L)

Continuous renal replacement therapy

Previous daily dose, mg/kg

0.000

0.001

0.000

0.004

Concomitant medications

Proton pump inhibitors

Glucocorticoids

64

43

40

21

0.000

0.000 

RESULTS

Univariate analysis for different voriconazole trough concentration ranges†

53.2

Acute physiology and chronic health evaluation II (APACHE II)

Albumin, g/L

Serum creatinine, mcmol/L

Concomitant medications

Proton pump inhibitors

Glucocorticoid

89.7%

74.4%

74.2%

37.6%

0.046

0.000

Continuous renal replacement therapy (CRRT)

ECMO

Findings on blood sampling day: Most ECMO patients (93.9%) used veno-venous (VV-ECMO), with remaining useing venoarterial (VA-ECMO). Additionally, the mean ECMO blood flow rate was 3.77 L/min, median duration of 5 days, 31.8% concurrently received renal replacement therapy, and 8 patients who changed membrane oxygenators. For all patients: daily maintenance doses before sampling was mostly 400 mg (93.9%) ranging from 200 to 800 mg. Mean daily maintenance dose per kg was 6.3 ± 1.4 mg/kg; median Cmin was 3.6 mg/L with 29.5% subtherapeutic

†Stastically significant findings 

Multivariant linear regression analysis of voriconazole trough concentration with significant results: increased concentrations: SOFA (p= 0.027), dose before sampling (0.014) and reduced concentrations: concomitant glucocorticoids (0.002), ECMO (p= 0.007)

Binary logistic regression analysis for subtherapeutic trough concentration (<2 mg/L) with significcant results: ECMO Odds Ratio (OR) 7.783; p= 0.012

Adverse Events

Not assessed.

Study Author Conclusions

Our findings showed that ECMO is a significant covariable affecting voriconazole exposure. SOFA score and daily dose were associated with increased voriconazole concentrations, while use of ECMO and coadministration with glucocorticoids were associated with reduced voriconazole concentrations.

InpharmD Researcher Critique

Limitations include lack of accountability of CYP2C19 genotype (given high prevalence in Asian population) and study design (single center and retrospective nature). This study evaluated TDM only after initial dosing, and did not assess or discuss monitoring following subsequent voriconazole doses. 

Voriconazole Pharmacokinetics in Critically Ill Patients and Extracorporeal Membrane Oxygenation Support

Design

Single-center, retrospective comparative case-control study

N= 53 samples in 24 patients

Objective

To compare voriconazole pharmacokinetics (PK) in critically ill patients with and without extracorporeal membrane oxygenation (ECMO) support and to determine optimal voriconazole dosage required to achieve the target plasma concentration

Study Groups

ECMO group (n= 27 samples in 9 patients)

Non-ECMO (n= 26 samples in 15 patients)

Inclusion Criteria

Adult intensive care patients treated with empiric or treatment dose voriconazole, adminstered via oral, nasogastric or intravenous routes, and with at least one voriconazole trough plasma concentration (Cmin) determination

Exclusion Criteria

Voriconazole plasma concentrations drawn outside the 15 to 60 minutes window, including post-dose plasma concentrations

Methods

First plasma concentrations were collected at steady state and 5-7 days after treatment initiation. Cmin was collected 15-50 minutes prior to next dose. Adequate Cmin was defined as 1.0 and 5.5 mg/L. Dose was based on actual body weight (ABW) with body mass index (BMI) <30 kg/m², or calculated ABW with 0.4 factor in those with BMI ≥30 kg/m². Analyses were completed in the ECMO group to differentiate voriconazole Cmin determinations before and after TDM and dose optimization.

Duration

November 2012 to February 2022

Outcome Measures

Voriconazole Cmin levels and the impact of various variables, including ECMO support and albumin levels on PK parameters

Baseline Characteristics

Non-ECMO (n= 26)

ECMO (n= 27)

Age, years

Female, %

Actual weight, kg

Treatment days

Number of Cmin determinations per patient

Drug parameters†

Cmin, mg/L

Standardized doses, mg/kg/12 h

Cmin/daily dose/kg, (mg*kg/L*day)

Number of samples

Infra-therapeutic (Cmin <1)

Therapeutic (Cmin 1-5)

Supratherapeutic (Cmin >5)

Amiodarone

3.62

3.53

0.98

1

16

9

5

0.38

4.50

0.12

16 

10

1

0

<0.001

0.007

<0.001 

<0.001

<0.001

<0.001

0.017

Analytical variables

Renal clearance, CDK-EPI mL/min

Gamma-glutamyl transferase, U/L

Albumin, g/L

99.98

97.8

27.50

147.6

614

30 

<0.001

0.001

0.020

Differences between optimised and non-optimised voriconazole samples in the ECMO group

Non-optimized ECMO (n= 16)

Drug parameters

Standardised doses, mg/kg/12 h

Cmin, mg/L

Number of samples

Infra-therapeutic (Cmin <1)

Therapeutic (Cmin 1-5)

Supratherapeutic (Cmin >5)

3.48

0.30

14

1

1

6.44

3.28

2

9

0

<0.001

0.002

<0.001

<0.001

<0.001

Effect of variables on voriconazole Cmin using the generalised estimating equations (GEE) model

Outcomes

Albumin, g/L

ECMO support

-0.023

-0.668

-0.046–-0.001

-0.978–-0.358

0.044

<0.001

Adverse Events

Not studied.

Study Author Conclusions

Voriconazole dose required to attain the objective PKs in ECMO patients is higher than that recommended elsewhere. We observed a maintenance dose of 6 mg/kg BID voriconazole combined with TDM is required to achieve therapeutically effective and safe plasma levels, with a lower incidence of adverse events. Our study confirms the importance of monitoring albumin levels and adjusting the voriconazole Cmin obtained to prevent toxicity.

InpharmD Researcher Critique

This is limited due to: small sample size, non-randomized, retrospective design with increases risk of confounding factors and prevents causality conculsion, did not include CYP219 genotype testing which could contribute to drug metabolism, and possible selection bias due to lack of matching groups.

Fluctuating Voriconazole Concentrations during Extracorporeal Membrane Oxygenation

Design

Case presentation

A 19-year-old Caucasian female, weighing 65 kg and with no past medical history, was transferred to the ICU for acute respiratory distress syndrome (ARDS) secondary to a freshwater drowning event. Upon admission, she had significant lab findings, including elevated lactic acid levels, low arterial pH, bicarbonate levels, and high D-dimer concentration. She also had bilateral patchy opacities on her chest X-ray, indicating acute respiratory distress syndrome (ARDS), and was placed on femoral venoarterial (VA) ECMO due to developing distributive shock and circulatory failure. 

During her hospital stay, she was suspected of having aspiration pneumonia and was treated with broad-spectrum anti-infectives. A mold identified as Scedosporium apiospermum was found in her respiratory cultures, leading to the administration of intravenous voriconazole 200 mg BID. The dose was increased to 6 mg/kg (380 mg) BID for two doses followed by 4 mg/kg (250 mg) BID to target a trough concentration of 1.5-4 mcg/mL. Despite the treatment, her voriconazole trough concentrations remained subtherapeutic after eight days of therapy. Infectious diseases consult recommended administering voriconazole after the oxygenator in hopes of decreasing drug binding and added micafungin therapy for double coverage.

The patient was an intermediate metabolizer per genotype testing for the CYP2C19 enzyme. Voriconazole dosing was adjusted multiple times due to varying serum concentrations, at one point becoming supratherapeutic after changing the ECMO circuit and discontinuing a potential drug interaction with pantoprazole.

After 38 days, a therapeutic voriconazole concentration was achieved at a dose of 5.5 mg/kg (350 mg) BID. The patient underwent routine oxygenator exchanges and eventually was switched to a bi-caval venovenous (VV) ECMO cannula. Two months post-admission, she was decannulated and discharged on oral voriconazole, with an expected 6-month duration of therapy and weekly trough level monitoring.

Two weeks after discharge, a follow-up showed a supratherapeutic voriconazole concentration, so her dose was reduced from 500 mg BID to 400 mg BID. The patient experienced no significant side effects typically associated with high voriconazole levels. Eight weeks post-discharge, her voriconazole was switched to posaconazole due to another supratherapeutic concentration, with further follow-up planned at an outside clinic.

Study Author Conclusions

These findings demonstrate how voriconazole drug concentrations can be unpredictable when administered during ECMO and the importance of close monitoring of drug concentrations. During prolonged ECMO, there were significant variations in voriconazole concentrations, thought to be related to drug loss attributed to drug binding to the oxygenator and circuit tubing. This patient's pharmacokinetics were also complicated by the duration it took to obtain a therapeutic voriconazole concentration; the range of doses administered, which required two times the recommended maintenance dose; and overall poor therapeutic trough attainment. More studies are needed to provide sufficient guidance on administering voriconazole in critically ill patients receiving ECMO.

Voriconazole Sequestration During Extracorporeal Membrane Oxygenation for Invasive Lung Aspergillosis: A Case Report

Design

Case presentation

A 52-year-old Caucasian man (90 kg), an excavation worker who also trimmed trees without protective gear, was admitted to a Montreal hospital with dyspnea, fever, and headaches. Despite appearing unwell and febrile, his oxygen saturation was normal initially. He received ceftriaxone and doxycycline for presumed community-acquired pneumonia, but his condition worsened, leading to progressive hypoxemia. CT scans showed severe bilateral reticulonodular and ground-glass infiltrates, suggestive of atypical pneumonia or hypersensitivity pneumonitis. Treatment was escalated to include meropenem, liposomal amphotericin B, and methylprednisolone without significant improvement, prompting his transfer to a tertiary care ICU for deteriorating hypoxemic respiratory failure and eventual endotracheal intubation.

A video-assisted thoracoscopic lung biopsy and bronchoalveolar lavage revealed abundant Aspergillus fumigatus, leading to a switch to voriconazole (6 mg/kg IV BID for two doses, then 4 mg/kg IV BID) and caspofungin (70 mg on day 1, then 50 mg/d). ECMO support was initiated for severe acute respiratory distress syndrome on hospital day 15.

Two additional doses of voriconazole 6 mg/kg were administered empirically in an attempt to compensate for expected sequestration in the ECMO circuit, and the caspofungin dose was increased to 70 mg/day. On day 18, the first voriconazole trough was 5.8 mcg/mL, so it was continued at 4 mg/kg BID. However, the next trough on day 22 was very low at 0.8 mcg/mL. The dose of voriconazole was increased to 4.5 mg/kg twice daily on day 23. Afterward, voriconazole trough levels remained subtherapeutic (<1 mcg/mL) despite additional loading doses and dose increases.

Despite adjustments to antifungal therapy, voriconazole levels remained sub-therapeutic, and liposomal amphotericin B 5 mg/kg/d was added. Therapeutic voriconazole levels were eventually achieved on day 34. No drug interaction or rapid metabolism of voriconazole was detected. The patient's renal function remained stable without the need for renal replacement therapy.

Despite extensive treatment, including increased doses of voriconazole and ECMO support, the patient developed a Burkholderia cepacia superinfection and bacteremia, alongside persistent Aspergillosis. After 21 days on ECMO without clinical improvement, active therapy was withdrawn, and the patient passed away due to the complications of the infections and underlying conditions.

Study Author Conclusions

The case highlights the challenges in managing antifungal therapy in patients on ECMO, including the potential need for higher doses and the impact of circuit components on drug levels. Voriconazole exhibits interpersonal variability, but this case was complicated by persistently low trough levels to losses in the ECMO circuit. More research is needed to guide dosing adjustments and optimize outcomes for ECMO patients with fungal infections.

Therapeutic trough levels of voriconazole cannot be guaranteed with standard dosing in patients undergoing ECMO, and much higher doses may be necessary. Empirical use of higher doses and/or combination therapy may be reasonable, while frequent therapeutic drug monitoring is mandatory. Membrane changes may cause sudden drops in concentrations and it may take several days before attaining nadir voriconazole concentrations.

Decreasing Voriconazole Requirement in a Patient After Extracorporeal Membrane Oxygenation Discontinuation: A Case Report

Design

Case presentation

A 32-year-old Caucasian male (62 kg) with a history of medullary cystic kidney disease and a renal transplant 10 months prior was admitted to an ICU for acute hypoxic respiratory failure. He was a former smoker, had no known drug allergies, and his medication regimen included tacrolimus, leflunomide, prednisone, aspirin, omeprazole, and medical marijuana. Initially treated for pneumonia with vancomycin, cefepime, azithromycin, and SMX-TMP (sulfamethoxazole-trimethoprim) at an outside hospital, his condition progressed despite treatment. Imaging revealed chronic interstitial changes and diffuse ground-glass opacities, and cultures and panels identified Moraxella catarrhalis, human rhinovirus/enterovirus, human metapneumovirus, and respiratory syncytial virus in which ampicillin-sulbactam and ribavirin was added. Due to elevated serum 1-3 beta-D-glucan levels and worsening renal function, his treatment was adjusted to include clindamycin, primaquine, and later micafungin. 

Upon transfer to another center for further care, his medication was adjusted due to leukopenia and suspicion of severe Pneumocystis jirovecii pneumonia. His oxygen requirements increased, leading to intubation and initiation of venovenous extracorporeal membrane oxygenation (VV-ECMO). After a positive bronchoalveolar lavage (BAL) Aspergillus galactomannan antigen test, IV voriconazole (5.8 mg/kg x2 doses followed by 3.9 mg/kg BID) was started for suspected Aspergillosis. The patient was on ECMO for 33 days, during which his voriconazole levels were monitored and adjusted. Therapeutic levels were eventually achieved after the dose was increased to 11.3 mg/kg/dose q12h. 

His condition eventually improved, allowing for decannulation and a subsequent empiric reduction in voriconazole dosage from 11.3 mg/kg BID to 3.2 mg/kg BID. Voriconazole level post-decannulation was 1.0 mcg/mL and increased to 3.6 mcg/mL following a dose increase to 7.3 mg/kg BID.

He was discharged to an inpatient rehabilitation unit after 45 days, with adjustments to his immunosuppressive therapy, tapering of prednisone, and a 1:1 conversion to PO voriconazole. Liver function tests, which had fluctuated during ECMO, normalized post-ECMO, and tacrolimus was restarted and adjusted based on the patient's condition. The medications administered during or post ECMO were not expected to cause changes to the voriconazole concentrations and the CYP2C19 genotype and CRP levels were unknown 

Study Author Conclusions

This case highlights the complexity of managing acute respiratory failure in a renal transplant patient, involving a multidisciplinary approach and careful adjustment of medications due to interactions and the patient's changing clinical status. Patients may experience large alterations in pharmacokinetics while on ECMO therapy and the ideal dose adjustment remains unknown. Further evidence is needed to guide dose modifications for voriconazole therapy in patients transitioning on or off of ECMO therapy. These results may not be applicable to other patients due to differences in ECMO circuitry, patient genotype status, and clinical status.

Therapeutic Drug Monitoring of Voriconazole in a Child With Invasive Aspergillosis Requiring Extracorporeal Membrane Oxygenation

Design

Case presentation

A 5-year-old boy with a history of pre-B-cell acute lymphatic leukemia (ALL) and treated according to the Dutch Childhood Oncology Group ALL-10 protocol two years prior, was hospitalized for acute respiratory failure due to varicella zoster pneumonia during the consolidation phase of his ALL treatment, which included methotrexate and 6-mercaptopurine. He presented with pancytopenia and required intensive artificial ventilation. His condition worsened, leading to the initiation of venovenous extracorporeal membrane oxygenation (ECMO) therapy and continuous venovenous hemofiltration (CVVH) due to fluid retention from a systemic inflammatory response syndrome.

During ECMO, the boy developed a complication from an Aspergillus fumigatus infection, which was susceptible to voriconazole. Monitoring trough plasma concentrations of voriconazole was recommended to ensure efficacy against the fungal infection. Initially, the voriconazole dosage was started at 6.7 mg/kg BID and increased based on plasma concentration to achieve the target trough of >1 mg/L. On day 5 of voriconazole therapy, trough levels were 0.7 mg/L, prompting a dose increase to 10 mg/kg BID. The dose was further increased to 14 mg/kg BID to achieve adequate exposure after therapeutic drug monitoring consultation.

The increase to 14 mg/kg BID over a 4-hour infusion produced adequate voriconazole concentrations from hospital days 3 to 7 (mean trough, 4.22 mg/L). No further increase of voriconazole was considered necessary, and therapy was continued. 

The boy's liver enzymes were monitored and remained within acceptable limits, and his renal function stayed stable throughout the treatment. However, due to full sedation, assessments for ocular or central nervous system toxicity from voriconazole were not possible. Despite achieving adequate voriconazole concentrations, the patient's condition deteriorated, and he died after 19 days on ECMO due to a combination of disturbed coagulation, leading to a massive hemothorax, and invasive aspergillosis. Autopsy was not performed, but it was suggested that the combination of diffuse intravascular coagulation and invasive pulmonary aspergillosis contributed to the fatal complication.

Study Author Conclusions

The patient presented in this report demonstrated a slower clearance rate compared with pediatric reference values. This may be caused by the higher dose (due to saturation of enzymatic processes), but no conclusions can be drawn on the basis of these findings because it is within the distribution of the reference population. ECMO might have led to an increased volume of distribution, causing lower plasma concentrations.

Irrespective of the ECMO treatment, concentrations of voriconazole show a wide interindividual variation caused by many factors such as disease status, altered clearance, changes in volume of distribution, etc, and should therefore be monitored.

Pharmacokinetics of Caspofungin and Voriconazole in Critically Ill Patients During Extracorporeal Membrane Oxygenation

Design

Case series

Case 1

A 41-year-old male weighing 80 kg, with a history of pancreatitis, experienced a relapse and was admitted to the hospital. He initially received piperacillin/tazobactam, but on day 27, he developed septic shock and was transferred to the ICU. His treatment was changed to meropenem, amikacin, and fluconazole, and he was put on mechanical ventilation. Due to inadequate respiratory function, venovenous ECMO was initiated on day 29. Blood cultures identified Candida albicans, but despite ongoing treatment, the patient's condition did not improve. Thus, fluconazole was replaced with caspofungin (with a loading dose of 70 mg, followed by a daily maintenance dose of 70 mg).

On day 32, caspofungin was discontinued after it was found that C. albicans was susceptible to fluconazole. The patient's condition improved, allowing for the discontinuation of ECMO on day 33. Throughout caspofungin therapy, his laboratory values remained stable, with a slight decrease in creatinine clearance (average 66 mL/min), elevated liver enzymes (bilirubin 2.32 mg/dL, alanine aminotransferase 40.5 U/L), and a moderate decrease in albumin levels (average 27 g/L).

Case 2

A 17-year-old male weighing 65 kg with leukemia was admitted due to toxic megacolon and shock. Upon admission, he was intubated, and antibiotic treatment commenced. By day 7, he was suspected of having invasive pulmonary and cerebral aspergillosis based on clinical signs, cultures, and antigen titers.

Treatment with voriconazole was initiated on day 8, starting with a loading dose of 400 mg twice daily followed by a maintenance dose of 280 mg twice daily, both intravenously. Despite maximal ventilatory support, his respiratory function worsened, leading to the start of venovenous ECMO on day 14. The voriconazole dose was increased to 400 mg to compensate for expected losses in the ECMO circuit, aiming to maintain adequate drug levels in plasma and CSF. The patient's renal function stayed stable (mean CLCR, 92 mL/min), while his bilirubin levels rose from an average of 3.36 mg/dL before ECMO to 6.94 mg/dL during ECMO treatment.

Pharmacokinetic Parameters

Study Author Conclusions

Despite expectations of altered drug levels due to ECMO, caspofungin levels remained within the normal range, suggesting minimal sequestration by the ECMO circuit. However, voriconazole levels were significantly higher than the therapeutic range, likely due to impaired metabolic capacity and potential partial sequestration by the ECMO circuit. This indicates the need for careful monitoring and possibly adjusting dosages of these drugs in ECMO patients to achieve therapeutic levels.

Potential voriconazole and caspofungin sequestration during extracorporeal membrane oxygenation

Design

Case presentation

A 31-year-old woman, weighing 50 kg, was admitted to the ICU for cardiogenic shock due to fulminant myocarditis. When her condition worsened, she was placed on ECMO support and also received continuous venovenous hemodialysis. Despite initial antifungal therapy with voriconazole (6 mg/kg BID day 1, then 4 mg/kg BID) and caspofungin (70 mg day 1, then 50 mg/day), her condition didn't improve, showing persistent Aspergillus fumigatus in cultures and a high BAL (Bronchoalveolar Lavage) galactomannan index, indicating a severe fungal infection. Due to treatment failure, voriconazole was replaced with an antifungal combination of IV liposomal amphotericin B, flucytosine, and continued caspofungin. Nebulized liposomal amphotericin B (3 mg/kg/ day) was also administered.

Blood levels of the antifungals were monitored, demonstrating complex pharmacokinetics influenced by ECMO, which can affect drug distribution and clearance. Notably, voriconazole levels were very low or undetectable, suggesting significant loss or binding in the ECMO circuit despite high dosage. This observation is consistent with other studies indicating ECMO can significantly alter the pharmacokinetics of drugs, including antifungals, due to factors like molecular size, degree of ionization, and lipophilicity. The patient eventually improved, with ECMO being discontinued 21 days after admission and negative BAL cultures 16 days after starting the three-drug regimen. She was discharged from the ICU on day 53.

Study Author Conclusions

The case underscores the challenges of managing drug levels in patients on ECMO, particularly highlighting the unpredictable pharmacokinetics of antifungals like voriconazole and caspofungin in such settings. The case suggests the importance of monitoring antifungal blood levels in ECMO patients and considers using liposomal amphotericin B for invasive fungal infections due to its more predictable therapeutic range. Further studies on the effects of ECMO on drug levels are advocated.

The pharmacokinetic challenge of treating invasive aspergillosis complicating severe influenzae assisted by extracorporeal membrane oxygenation

Design

Case presentation

A 57-year-old man with severe influenza developed invasive aspergillosis (IA) and was treated with intravenous (IV) voriconazole in the intensive care unit (ICU), followed by therapeutic drug monitoring to maintain plasma concentration within the target range (residual concentration between 2 and 6 mg/L). Despite treatment, he required veno-venous ECMO support due to refractory respiratory acidosis, which resulted in frequent voriconazole plasma concentration drops after each ECMO membrane change. To address high pharmacokinetic variability and underdosing of voriconazole, liposomal amphotericin B was added to the treatment from week 2. However, the patient's IA remained refractory, with persistent lung infiltrates and Aspergillus detection, leading to treatment failure and eventual withdrawal of care after 5 months. The patient's total ICU stay lasted 5 months, with 4 months under ECMO support. Intense voriconazole sequestration by ECMO new membranes was identified as a significant challenge in achieving effective drug levels.

Study Author Conclusions

Given the high number of IA-complicating influenza and the increasing development of the ECMO technique for refractory acute respiratory failure, the described situation might become frequent. Data regarding the pharmacokinetics and efficacy of other antifungal drugs, such as isavuconazole, during ECMO therapy are urgently needed.

Voriconazole pharmacokinetics in a critically ill patient during extracorporeal membrane oxygenation

Design

Case presentation

A 24-year-old man with acute mixed leukemia was admitted with recurrent fever and respiratory symptoms, later developing severe acute respiratory distress syndrome (ARDS) requiring ECMO therapy. Despite various treatments, including antifungals like voriconazole (VRC), his condition worsened, leading to the introduction of ECMO on day 8. VRC was continued to be administered at the dose of 200 mg every 12 h, and therapeutic drug monitoring was performed.

Two pharmacokinetic (PK) profiles were drawn within two dosing intervals during day 8–day 9 (i.e., before and after ECMO). While the trough levels (both C0 and C12) with ECMO showed slightly lower than those without ECMO (12.58 and 12.84 vs. 14.02 µg/mL), the peak levels and the area under the concentration-time curve from 0 h to 6 h (AUC0–6) were comparable (16.36 vs. 16.06 µg/mL and 90.78 vs. 91.45 µg*h/mL, respectively), suggesting minimal impact of ECMO on VRC plasma exposure.

Study Author Conclusions

Previous research offers mixed findings on ECMO's effect on VRC levels, with some suggesting significant sequestration by ECMO circuits and others reporting negligible impact. This case contributes to the ongoing investigation into how ECMO therapy affects drug pharmacokinetics, specifically VRC, in critically ill patients. The circuit factors including the type of membrane need to be taken into account to further identify the effects of ECMO on the PK of VRC and it should be cautious to increase VRC dose during ECMO therapy before obtaining more evidence.