Evaluation of Voriconazole and Posaconazole Dosing in Patients With Thermal Burn Injuries

Abstract

Fungal infections are a recognized cause of increased morbidity and mortality in thermal burn patients. Adequate treatment regimens remain a challenge due to unpredictable pharmacokinetic/pharmacodynamic changes caused by a hypermetabolic state and individual patient factors. A retrospective evaluation of adult thermal burn patients from April 2014 to April 2020 was conducted to assess voriconazole and posaconazole antifungal dosing regimens. The primary outcome was the incidence of attaining a therapeutic steady-state trough level on the patient’s initial voriconazole or posaconazole regimen. Of the 33 patients analyzed, 26 (78.8%) patients achieved a therapeutic level during azole therapy. However, only 11 (33.3%) patients achieved a therapeutic level on their first azole regimen. The median time to therapeutic level was 8.0 + 21.8 days from the start of azole therapy. Optimal dosing strategies for azole therapy in patients with thermal burns remain undefined. Further assessment is needed to delineate patient-specific factors that can contribute to subtherapeutic azole levels in thermal burn patients and the overall clinical impact of population-specific dosing regimens.

Fungal infections are a known, independent cause of increased morbidity and mortality in thermal burn patients.¹ Available literature varies on the prevalence in this patient population (6–26%), but it is likely an underreported complication of burn injury due to limitations of fungal microbiological testing.² Practitioners may attempt to treat a suspected cutaneous fungal infection with topical agents. However, the risk of delaying systemic treatment while awaiting microbiological confirmation may increase the likelihood of invasive fungal infections, with Aspergillus species being of particular concern.^1,3 Without an identified fungal species, agent selection is driven by suspected pathogen(s). Other challenges in treating fungal infections in this population include alterations of drug properties due to the body’s response to thermal injury and lack of data for appropriate dosing and monitoring practices in burn patients.⁴

In the 48 to 72 hours following an acute burn injury, the ebb phase of the stress–metabolic response works to combat acute fluid loss and preserve end-organ perfusion. Alternatively, the flow phase (48 hours to years postinjury) is characterized by hyperdynamic circulation and systemic hypermetabolism as the body responds to inflammatory mediators. Hemodynamic stability is critical for patient survival in the ebb phase; however, the majority of a patient’s treatment course and complications will occur during the hypermetabolic flow phase. As the body responds to the damaged cutaneous tissue, physiologic changes lead to alterations in absorption, distribution, metabolism, and excretion processes which can alter drug pharmacokinetics (PK) properties. The most impactful changes include decreased muscle mass affecting serum creatinine, decreased albumin, changes in the volume of distribution due to frequent changes in fluid status, and increased renal clearance and hepatic metabolism.^2,4 In addition to metabolic changes, thermal burns cause disruptions in innate immunity and increase systemic exposure to common human flora and environmental microbes. This, along with the prolonged hospitalization required for many burn patients, increases their risk of exposure to hospital-acquired infections, extended systemic antibiotic courses, and multidrug-resistant organisms.^5,6 There have been numerous articles published on the specific PK/pharmacodynamics (PD) parameters and dosing recommendations in thermal burn patients for a variety of antibiotics, including cefepime, ceftazidime, imipenem, meropenem, daptomycin, tobramycin, linezolid, and colistin.^7–13 When it comes to antifungals, however, data are limited to fluconazole. One study assessing fluconazole use in burn patients found that the half-life of fluconazole was 13% shorter and the clearance was 30% greater compared to nonburn patients.¹⁴ A separate population PK analysis found that plasma concentrations of fluconazole are predicted to be lower in burn patients due to changes in their clearance and volume of distribution.¹⁵

Due to the lack of data on interpatient PK variability, current recommendations for other azole antifungals, such as voriconazole and posaconazole, include giving standard doses and monitoring drug levels.^2,16 The aim of this study is to evaluate if patients with thermal burn injuries achieve therapeutic drug levels with traditional dosing strategies for azole antifungals, specifically voriconazole and posaconazole.

METHODS

Study Design

This was a single-center, retrospective, observational analysis of adult patients admitted to an American Burn Association (ABA)-verified burn center between April 4, 2014 and April 3, 2020 with a thermal burn injury. Patients were included if they were ≥18 years of age, were initiated on voriconazole or posaconazole, and had a steady-state trough level drawn during their hospital admission. Patients were excluded if they were taking voriconazole or posaconazole prior to admission, were pregnant, or were incarcerated at the time of admission. At this ABA-verified burn center, patients requiring voriconazole or posaconazole are dosed and monitored based on an institutional guideline developed for the general patient population. The institutional dosing guidelines, as well as dose adjustment recommendations, are summarized in Table 1. Approval was obtained by the institutional review board for human research prior to data collection.

Table 1.

Institutional azole dosing guideline

	Voriconazole		Posaconazole
Loading dose	6 mg/kg IV/PO every 12 hours × 2 doses		300 mg IV/PO every 12 hours × 2 doses
Maintenance dose	4 mg/kg IV/PO every 12 hours		IV: 300 mg daily; OR PO tablet: 300 mg daily; OR PO suspension: 200 mg every 6 to 8 hours
Steady-State Trough Level: Measured 5 to 7 days after initiation
Level descriptor	Level	Dose adjustment	Level	Dose adjustment
Subtherapeutic	<1 μg/mL	50% dose increase	<1000 ng/mL	Attempt to improve absorption; 50% dose increase
Therapeutic	1 to 5 μg/mL	Continue dose	1000 to 3750 ng/mL	Continue dose
Supratherapeutic	>5 μg/mL	25 to 50% dose reduction; if adverse effects, hold therapy and restart with 50% dose reduction when trough <5 ug/mL	>3750 ng/mL	25 to 50% dose reduction

IV, intravenous; PO, by mouth.

Data Collection

A manual chart review was performed and data were compiled in a REDCap^® database (Vanderbilt University, Nashville, TN). A medication and its corresponding dose were considered a new regimen if either the primary agent, route of administration, or mg/kg dose was changed. Loading doses were considered to be part of the initial regimen and not as a separate entity. For the purpose of this study and the desire to delineate the incidence of attaining subtherapeutic levels with standard azole antifungal dosing in thermal burn patients, supratherapeutic levels were documented as “therapeutic.” According to institutional recommendations, levels should be drawn at a steady state approximately 30 minutes prior to the administration of a dose.

Outcomes

The primary outcome was the incidence of attaining a therapeutic steady-state trough level on the patient’s initial voriconazole or posaconazole regimen. Secondary outcomes included the incidence of achieving a therapeutic steady-state trough level for voriconazole or posaconazole, time to therapeutic steady-state trough level, incidence of patients with confirmed fungal infections, and the median daily dose required to reach a therapeutic level. The following exploratory subgroup analyses were also performed to delineate potential patient-specific factors to consider when dosing azole antifungals in thermal burn patients: Total body surface area (TBSA) <20% vs ≥20%, voriconazole vs posaconazole, male vs female, age <35 vs ≥35 years, body mass index <30 vs ≥30 kg/m², intravenous (IV) vs oral (PO) tablets vs oral suspension formulations, receipt of loading dose vs no loading dose, and <3 days from burn injury to azole initiation vs ≥3 from burn injury to azole initiation.

Statistical Analysis

For categorical variables, frequencies and percentages were documented and P values were calculated using chi-squared tests or Fisher’s exact tests. Medians and interquartile ranges (IQRs; Q1–Q3) were also reported, as appropriate. P values of less than .05 were considered to be statistically significant. All statistical analyses were conducted using STATA Statistical Software: Release 16 (StataCorp LLC, College Station, TX).

RESULTS

Patient Characteristics

Overall, 36 patients were identified as meeting inclusion criteria. Three patients were later excluded for lack of documentation of thermal burn in the medical chart (n = 1), nonthermal burn injury (n = 1), and admission to a unit other than the burn intensive care unit (ICU; n = 1). The majority of patients were Caucasian (69.7%), male (84.8%), with a median burn TBSA of 35.0% (IQR 16–53). Baseline patient characteristics are summarized in Table 2. Of the 33 patients included, 21 (63.6%) received initial treatment with voriconazole and 12 (36.4%) received posaconazole. The most commonly used voriconazole formulation was PO tablets, while the most commonly used posaconazole formulation was IV. Of the patients whose first azole antifungal was administered via the oral, nasogastric, or nasojejunal route, 12 (60%) received concurrent acid-suppressive therapy, with 7 (35%) receiving proton pump inhibitors (PPIs) and 5 (25%) receiving histamine-2 receptor antagonists. Twenty-one (63.6%) patients received a loading dose in accordance with institutional dosing guidelines. The median length of time from burn injury to initiation of azole therapy was 8 days (IQR 4.5–31.5). Characteristics of patients’ initial azole regimen are summarized in Table 3.

Table 2.

Baseline patient characteristics

	All Patients (N = 33)	Voriconazole (n = 21)	Posaconazole (n = 12)
Sex, n (%) male	28 (84.8)	17 (81.0)	11 (91.7)
Age at admission, median (Q1–Q3), years	48 (33.0–60.5)	47 (30.0–63.0)	48.5 (42.8–58.3)
Race, n (%)
Caucasian	23 (69.7)	14 (66.7)	9 (75.0)
African American	9 (27.3)	7 (33.3)	2 (16.7)
Unknown	1 (3.0)	0 (0.0)	1 (8.3)
History of cirrhosis, n (%)	2 (6.1)	0 (0.0)	2 (16.7)
Weight at admission, median (Q1–Q3), kg	92.8 (70.0–103.9)	96 (71.2–106.5)	83.1 (69.0–93.3)
BMI at admission, median (Q1–Q3), kg/m2	28.6 (24.5–34.6)	31.6 (24.4–35.9)	27.2 (24.3–29.0)
TBSA at admission, median (Q1–Q3), %	35 (16.0–53.0)	35 (16.0–55.0)	40.5 (20.5–49.0)

BMI, body mass index; TBSA, total body surface area; Q1, first quartile; Q3, third quartile.

Table 3.

Antifungal regimen characteristics

	All Patients (N = 33)	Voriconazole	Posaconazole
Initial azole received, n (%)	—	21 (63.6)	12 (36.4)
Initial azole formulation received, n (%)
IV	13 (39.4)	5 (23.8)	8 (66.7)
PO tablets	16 (48.5)	14 (66.7)	2 (16.7)
PO suspension	4 (12.1)	2 (9.5)	2 (16.7)
Loading dose administered according to institutional guidelines, n (%)	21 (63.6)	11 (52.4)	10 (83.3)
Length of time from burn injury to initiation of azole therapy, median (Q1–Q3), days	8 (4.5,31.5)	7 (2.0,16.5)	13 (7.0,35.0)
Concurrent enteral feeding, n (%)*	18 (90.0)	15 (93.8)	3 (75.0)

*Only assessed for a patient who received azole antifungal via PO (by mouth), NG (nasogastric), or NJ (nasojejunal) routes.

Overall, 15 (45.5%) patients had a proven fungal infection during their hospitalization. In these 15 patients, 21 positive fungal cultures were identified. The most common identified fungal species are summarized in Figure 1.

Figure 1.

Identified fungal species (n = 21). *Other includes Penicillium (2), Mycelia sterilia (2), mold (2), Paecilomyces lilacinus (2), Verticillium (1), and Acremonium (1).

Therapeutic Drug Monitoring Results

In regard to the primary outcome, a total of 26 patients (78.8%) attained a therapeutic steady-state trough level of voriconazole or posaconazole during their treatment course. Eleven patients (33.3%) attained a therapeutic level on their first azole antifungal regimen. These results are summarized in Table 4. Overall, the median length of time from the initiation of azole antifungals to the achievement of a therapeutic level was 8 days (IQR 4.0–12.0). The median daily dose required to achieve a therapeutic level was 400 mg (IQR 400–768) and 500 mg (IQR 400–600) for voriconazole and posaconazole, respectively. One patient was switched from posaconazole to amphotericin B due to subtherapeutic posaconazole levels. Two supratherapeutic levels were identified, both of which were in patients receiving voriconazole. In one of these patients, aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were elevated at the time of the supratherapeutic level. Importantly, shock liver was suspected in this patient, potentially explaining the elevated AST/ALT. Hallucinations were not identified in any patient with a supratherapeutic level.

Table 4.

Results: primary and secondary outcomes

	Therapeutic Level, n (%)	P	Therapeutic Level on Initial Regimen, n (%)	P
Patients (N = 33)	26 (78.8)	—	11 (33.3)	—

Subgroup Analyses

Exploratory subgroup analyses results are summarized in Table 5. Of the 7 patients who failed to achieve a therapeutic azole level, 5 (71.4%) patients were initially started on an enteral formulation and 4 (57.1%) of the 7 received only enteral formulations. Additionally, 5 of the 7 patients who did not achieve a therapeutic level had documented overlap with either an acid-suppressing medication or tube feeds. In the subtherapeutic population, 2 (28.6%) of 7 patients did not receive a loading dose for the first or second regimen. Interestingly, 10 (38.5%) of the 26 patients who achieved a therapeutic level did not receive a proper loading dose prior to their first azole regimen. Patients who received voriconazole as their first azole regimen were significantly more likely to obtain a therapeutic level than those who received posaconazole (P = .027). No statistically significant differences were noted in other subgroup analyses.

Table 5.

Results: subgroup analysis

Subgroup	Therapeutic Level, n (%)	P	Therapeutic Level on Initial Regimen, n (%)	P
Sex
Male, (n = 28)	21 (75.0)	.559	9 (32.1)	1.000
Female, (n = 5)	5 (100.0)		2 (40.0)
Age
<35 years, (n = 9)	5 (55.6)	.068	2 (22.2)	.681
≥35 years, (n = 24)	21 (87.5)		9(37.5)
BMI
<30 kg/m2, (n = 20)	16 (80.0)	1.000	5 (25.0)	.270
≥30 kg/m2, (n = 13)	10 (76.9)		6 (46.2)
TBSA
<20%, (n = 9)	7 (77.8)	1.000	3 (33.3)	1.000
≥20%, (n = 24)	19 (79.2)		8 (33.3)
Time from burn injury to azole initiation
≤3 days, (n = 7)	7 (100.0)	.299	3 (42.9)	.661
≥3 days, (n = 26)	19 (73.1)		8 (30.8)
Azole agent
Voriconazole, (n = 21)	17 (81.0)	.686	10 (47.6)	.027
Posaconazole, (n = 12)	9 (75.0)		1 (8.3)
Loading dose administered
Yes, (n = 21)	16 (76.2)	1.000	5 (23.8)	.149
No, (n = 12)	10 (83.3)		6 (50.0)
Azole formulation for the first regimen
IV, (n = 13)	11 (84.6)	.850	3 (23.1)	.108
PO tablets, (n = 16)	12 (75.0)		8 (50.0)
PO suspension, (n = 4)	3 (75.0)		0 (0.0)
Concurrent acid-suppressive therapy*
Yes, (n = 1)	1 (100.0)	1.000	0 (0.0)	—
No, (n = 3)	2 (66.7)		0 (0.0)
Concurrent enteral feeding†
Yes, (n = 18)	14 (77.8)	.502	8 (44.4)	.338
No, (n = 2)	1 (50.0)		0 (0.0)

*Includes only patients treated with posaconazole via the PO (by mouth), NG (nasogastric), NJ (nasojejunal) routes.

^†Includes only patients treated with azole therapy via the PO (by mouth), NG (nasogastric), NJ (nasojejunal) routes. Excludes those with clear documentation of enteral feeds being held prior to azole administration.

DISCUSSION

While 78.8% of patients achieved a therapeutic azole antifungal level during their hospital stay, only 33.3% of patients did so on their first azole regimen. This suggests an opportunity to optimize dosing strategies to improve the likelihood of thermal burn patients reaching a therapeutic level with an initial dosing regimen. In this study, 63.6% of patients received an appropriate loading dose, as recommended by institutional dosing guidelines. Therefore, improved adherence to loading dose recommendations could be an important aspect of treatment optimization at this institution. A proper loading dose was one that abided by the institutional guidelines in both dose and timing with respect to the maintenance regimen. The current dosing regimen for burn ICU patients is the same as other populations started on a therapeutic regimen of an azole antifungal. Voriconazole is loaded with a dose of 6 mg/kg every 12 hours for two doses followed by a maintenance regimen of 4 mg/kg every 12 hours. Posaconazole’s standard load is 300 mg every 12 hours for two doses followed by 300 mg daily. The exception is that the oral suspension of posaconazole is dosed 200 mg every 6 to 8 hours due to poor and unpredictable absorption. Appropriate weight-based dosing was confirmed by recording patient weight at the time closest to azole administration and comparing it to the dose. It is unclear what contributed to the low amount of compliance with institutional guideline-directed loading doses. However, prior to July 2014, it was common to initiate anti-infectives (including voriconazole) on all burn patients with greater than 5% TBSA. Given this medication was part of a center-specific order set prior to July 2014, it is possible that loading doses were considered unnecessary with the initiation of voriconazole.

Drug–drug and drug–food interactions with enteral feeding could also be contributing to subtherapeutic antifungal levels due to malabsorption. Posaconazole, unlike voriconazole, exhibits acid-dependent absorption. Acid-suppressing agents have been shown to reduce the overall area under the curve and maximum concentration (C_max) of posaconazole suspension by 32% and 46%, respectively.¹⁷ Additionally, in a retrospective analysis of 56 posaconazole levels from 17 patients in a cardiothoracic ICU, concomitant PPI therapy significantly lowered posaconazole levels.¹⁸ Voriconazole, unlike posaconazole, is metabolized by CYP2C19, CYP2C9, and CYP3A4, leading to the potential for significant drug–drug interactions via cytochrome p450 enzymes. No patients in this study received concomitant therapy with a strong p450 inducer; however, close monitoring for drug–drug interactions is warranted with patients receiving voriconazole. The absorption of both posaconazole and voriconazole can also be affected by concomitant enteral feeding.^19,20 Posaconazole suspension is administered with a high-fat meal or acidic liquid, and the tablets should be given with food for best absorption. However, voriconazole must be separated from enteral feeds by at least an hour before and after administration. For 2 of the 18 patients who received concurrent enteral feeding with azole therapy via the PO, nasogastric, or nasojejunal route, there was a clear documentation of tube feeds being held prior to azole antifungal administration. Given the retrospective nature of the study, it was not possible to determine if tube feeds were held prior to azole antifungal administration in any additional patients. Therefore, it is possible that drug–food interactions also contributed to subtherapeutic levels in 5 of the total 7 patients who failed to achieve a therapeutic azole. Five patients were also receiving concomitant prokinetic medications at some point during their hospitalization while being treated with oral formulations. Inability to appropriately absorb therapy could also be a contributing factor with regard to not achieving therapeutic levels.

Voriconazole metabolism can also be affected by CYP2C19 genetic polymorphisms. Nearing completion of this study, a single patient in the burn ICU was initiated on voriconazole for a suspected fungal infection. After an appropriate loading dose, the patient was started on a standard 4 mg/kg IV every 12 hours regimen of voriconazole. On day 6 of therapy, the level was undetectable, so the dose was increased by 50% according to institutional guidelines and another level was checked 5 days later. The level was, again, undetectable. This time the patient’s dose was increased by another 50%, but the team also decided to check for rapid metabolic variants of the patient’s CYP2C19 gene. The patient finally achieved a therapeutic voriconazole level, almost 2 weeks after starting therapy, on the third dosing regimen and the CYP2C19 test identified the patient as an extensive metabolizer. In this scenario, it is easy to understand why the patient required such high doses of voriconazole because multiple factors contributed to an increased rate of metabolism of the medication. However, in a patient who has opposing forces on metabolic processes, it may be hard to predict the resulting effect on therapeutic drug levels. Further studies are needed to determine the utility of pharmacogenomic testing in burn ICU patients, but current recommendations regarding dose escalations are reasonable with such high interpatient variability.^21,22

This study has several limitations including its single-center, retrospective design. Interpretation of the results of this study, particularly those of the subgroup analyses, is also limited by the small sample size despite a study period spanning 6 years. Additionally, varying azole antifungal formulations, evolution of institutional dosing guidelines, frequency and timing of measuring trough levels, and timing of interacting medications or enteral feeding administration could act as confounding variables. While this complicates the analysis and interpretation of the data, it adds to this study’s external validity by representing the real-world utilization of voriconazole and posaconazole in thermal burn patients.

In conclusion, the results of this study suggest that dosing strategies extrapolated from the general population may not be optimal for patients with thermal burn injuries. Patient-specific factors that contribute to a subtherapeutic level have not yet been established and require further investigation.

Conflicts of Interest: The authors have no conflicts of interest to disclose.