Evaluation of Continuous Infusion Vancomycin in a Pediatric Hematology/Oncology Population

Abstract

Background:

Continuous infusion vancomycin (CIV) may benefit children who are unable to achieve therapeutic concentrations with intermittent vancomycin dosing and may facilitate outpatient administration by alleviating the burden of frequent dosing intervals. Previous studies have used variable dosing regimens and steady-state concentration goals. The purpose of this study was to evaluate the total daily dose (TDD) of CIV required to achieve therapeutic steady-state concentrations of 15 – 25 μg/mL in pediatric hematology/oncology patients.

Methods:

A single-center retrospective study was performed for patients treated with CIV from January 2017 to June 2019. The primary outcome was the TDD required to achieve therapeutic steady-state concentrations on CIV. Secondary outcomes included time to reach therapeutic steady-state concentrations, CIV indications, and adverse events associated with CIV.

Results:

Data were collected for 71 courses of CIV in 60 patients. Median patient age was four years (range 0.4 – 20 years). The median TDD required to achieve initial therapeutic concentrations was 50.3 mg/kg/day (IQR 38.8, 59.2) and was further divided into age-based cohorts. TDD in mg/kg was significantly lower in the older cohort (P < 0.001), but there was no statistically significant difference between age-based cohorts with TDD in mg/m² (P = 0.97). Median time to achieve first therapeutic concentration was 19.3 hours (range 8.6 – 72.3 hours). The most common indication for CIV was ease of outpatient administration (69.0%). Acute kidney injury incidence was minimal (4.2%).

Conclusions:

CIV is associated with rapid attainment of target concentrations in pediatric hematology/oncology patients and is safe and well tolerated.

INTRODUCTION

Vancomycin is a frequently used intravenous glycopeptide antibiotic selected for treatment of severe Gram-positive bacterial infections, including those caused by methicillin-resistant Staphylococcus aureus (MRSA).¹ It exhibits time-dependent bactericidal activity through inhibition of bacterial cell wall synthesis. Thus, maintaining vancomycin blood concentrations above the minimum inhibitory concentration (MIC) of the bacterial organism for a longer period of time appears to increase its antimicrobial killing and augment efficacy.^1-2 However, vancomycin carries a risk of nephrotoxicity that increases with higher drug exposure, prolonged treatment duration, and concomitant nephrotoxins. Therapeutic drug monitoring is recommended to balance the risks of therapy with the need to maximize therapeutic effect for optimal clinical outcomes.

Current published guidelines recommend administration of vancomycin at doses of 60 – 80 mg/kg/day divided into three or four daily doses (i.e., every 6 to 8 hours) for children aged 3 months to less than 12 years in order to achieve an area under the concentration-time curve to the minimum inhibitory concentration (AUC/MIC) ratio of 400 – 600 mg·h/L for severe MRSA infections.¹ AUC-guided monitoring is recommended for serious MRSA infections (e.g., bacteremia, pneumonia, meningitis, osteomyelitis, infective endocarditis, and sepsis) in both adults and children.¹ However, the ideal pharmacokinetic/pharmacodynamic target for other systemic infections or additional Gram-positive organisms is not well defined. Trough concentrations are commonly used for monitoring as a surrogate for AUC with goal trough concentrations ranging from 10 – 20 μg/mL.^3-5 However, pediatric patients can experience difficulty achieving these therapeutic trough concentrations with intermittent intravenous vancomycin (IIV) dosing and have poor correlation between trough concentration and AUC values.^6-8 In addition, pediatric patients with malignancy have been shown to have increased vancomycin clearance compared to the general pediatric population.^9-11 Inability to achieve therapeutic concentrations increases the risk for clinical treatment failure.⁴ Pediatric oncology patients in particular require rapid attainment of therapeutic vancomycin concentrations due to their high risk of serious bacterial infections via compromised immune systems from use of myelosuppressive chemotherapeutic agents in addition to the presence of indwelling intravenous catheters.¹²

Continuous infusion vancomycin (CIV) has been studied in adults in a wide variety of patient populations. This dosing approach has shown reduced time to therapeutic vancomycin concentrations without increasing the risk of nephrotoxicity with steady-state concentrations less than 25 μg/mL.^13-17 Although current guidelines do not routinely recommend the administration of vancomycin as a continuous infusion due to limited outcomes data, CIV may benefit children who are unable to reach therapeutic concentrations with IIV dosing and facilitate outpatient administration.^1,4,5,18 There is a current lack of consensus on CIV dosing in pediatric patients, and previous studies in children have used heterogenous dosing regimens and ranges of desired steady-state concentrations.^3,5,12,18,19

This retrospective cohort study aimed to evaluate the total daily dose (TDD) of CIV required to achieve therapeutic concentrations in a pediatric hematology/oncology population.

MATERIALS AND METHODS

This was a single-center, retrospective cohort study of pediatric and adolescent patients with cancer or non-malignant hematologic disorders treated at St. Jude Children’s Research Hospital in Memphis, TN, USA. This study was reviewed by the Institutional Review Board and determined to be exempt under the revised Common Rule. Patients were evaluated for inclusion if they received CIV at St. Jude from January 1, 2017, to June 30, 2019. In brief, vancomycin orders were queried from pharmaceutical records. The study team focused on orders with dosing frequencies of ‘once daily’ and ‘every 24 hours’ to identify potential cases of CIV. Patients were excluded if no serum vancomycin concentration was obtained while on CIV.

Data were obtained from the electronic medical record and included demographic information, laboratory values such as baseline and peak serum creatinine and cystatin C, concurrent nephrotoxic medications (identified via the Nephrotoxic Injury Negated by Just-in-time Action or NINJA list), microbiological information such as culture identification and susceptibility data, and available vancomycin serum concentrations.²⁰ Clinical data were collected from the start of IIV therapy through the last day of CIV therapy. Vancomycin dosing information was collected, including the total daily dosage and administration interval of IIV prior to switching to CIV, days of IIV prior to switching to CIV, associated trough vancomycin concentrations while receiving IIV, indications for vancomycin use, indications for administering CIV, vancomycin concentrations while receiving CIV (initial, first therapeutic concentration, and time to first therapeutic concentration), CIV dosage associated with therapeutic vancomycin concentration, location of CIV administration (inpatient or outpatient), and total duration of therapy.

CIV courses that were not converted from IIV were preceded by a loading dose of vancomycin (15 – 20 mg/kg). Otherwise, the final intermittent dose of vancomycin was utilized as a loading dose prior to initiation of CIV.

The primary objective of this study was to determine the TDD required to achieve therapeutic concentrations on CIV. A vancomycin concentration range of 15 – 25 μg/mL was chosen as the therapeutic target to correlate with the current recommended AUC/MIC goal of 400-600 mg·h/L. AUC can be quickly calculated for patients on CIV by multiplying the steady-state concentration by 24.¹ Therefore, steady-state concentrations of 15 – 25 μg/mL should correspond with an AUC/MIC range of approximately 360 – 600 mg·h/L, assuming an MIC of 1 mg/L. Secondary outcomes included the percentage of courses that achieved therapeutic initial concentrations, median time to reach therapeutic concentrations, causes of intolerability or premature discontinuation of CIV, and incidence of adverse effects, including acute kidney injury (AKI) and vancomycin infusion reaction. Definition for AKI was consistent with the Kidney Disease: Improving Global Outcomes (KDIGO) guidelines including a rise in serum creatinine (SCr) of ≥ 0.3 mg/dL in 48 hours or increase ≥ 1.5 times baseline.²¹

Descriptive statistical analysis was performed for all variables and presented as proportions for categorical data and medians and ranges for continuous data. All reported p – values were two-sided and considered significant at p < 0.05. Correlations between variables were assessed using Spearman's rank correlation coefficient. Continuous variables were compared using the Wilcoxon-Mann-Whitney test or Kruskal-Wallis test when appropriate. A cluster analysis using the k-means clustering technique was performed for age categorization according to vancomycin dosages that achieved the desired concentration. The silhouette method was applied to decide the appropriate number of clusters. Dosages of vancomycin by cluster and age group were then calculated to determine the most appropriate integer cutoffs of age. Linear regression was utilized to determine therapeutic TDD divided by both body weight and body surface area (BSA) vs age. All data analyses were performed using R 4.0.2.

RESULTS

Seventy-one courses of CIV were identified in 60 patients (48% female), and no courses met exclusion criteria. CIV was primarily used in young patients with a median age of four years (range 0.4 – 20 years). CIV was most administered to patients with an underlying solid or primary central nervous system tumor, and 25.4% of the CIV courses given were in patients who had received a bone marrow transplant prior to initiation of CIV (Table 1).

Table 1:

Demographics and Baseline Characteristics

		All Courses (n=71)	CIV Courses that Achieved a Therapeutic Concentration (n=54)
Age; median (range)		4.5 years (0.4 – 20 years)	4.9 years (0.4 – 15 years)
Gender; n (%)	Male	37 (52.1)	26 (48.1)
Race; n (%)	White Black Other	55 (77.5) 12 (16.9) 4 (5.6)	43 (79.6) 8 (14.8) 3 (5.6)
Primary Diagnosis; n (%)	Leukemia/Lymphoma Solid Tumor/Neuro-Onc Non-malignant Hematology	31 (43.7) 37 (52.1) 3 (4.2)	23 (42.6) 29 (53.7) 2 (3.7)
Prior Transplant; n (%)		18 (25.4)	14 (25.9)
Switched from Intermittent Vancomycin; n (%)		67 (94.4%)	54 (100%)

Overall, the median TDD required to achieve a therapeutic vancomycin concentration was 50.3 mg/kg/day [interquartile range (IQR) 38.8, 59.2]. The traditional weight-based (mg/kg) dosing approach (Figure 1a) was found to be inversely associated with age (p < 0.001). However, the therapeutic TDD when scaled to BSA (Figure 1c) did not show a statistically significant difference in therapeutic TDD based on age (p = 0.97). Using age ranges identified by cluster analysis, the median therapeutic TDD was then further stratified into individual age-based cohorts of < 8 years and ≥ 8 years (Figure 1b and 1d). The median therapeutic TDD of CIV in mg/kg was inversely proportional to age with the younger cohort of patients requiring significantly higher doses of 54.5 mg/kg/day [IQR 49.4, 60.8 mg/kg/day] vs 35.7 mg/kg/day in the older cohort [IQR 34.8, 39.0 mg/kg/day] (p < 0.001) (Figure 1b). Median therapeutic TDD of CIV using a BSA-based approach was not significantly different at 1251.2 mg/m²/day in the younger cohort [IQR 1166.9, 1344.9 mg/m²/day] and 1031.7 mg/m²/day [IQR 905.1, 1307.3 mg/m²/day] in patients ≥ 8 years of age (p = 0.15) (Figure 1d).

Figure 1.

(A) Scatter plots demonstrating the correlation of age with TDD of CIV divided by body weight (mg/kg/day) and (C) body surface area (mg/m²/day) in courses that achieved a therapeutic steady-state concentration. Linear regression lines (solid blue) have been fitted to the data points. (B) Boxplots comparing the TDD of CIV divided by body weight (mg/kg/day) and (D) body surface area (mg/m²/day) between two age-based cohorts in courses that achieved a therapeutic steady-state concentration. The top and bottom of the boxplots show the 25^th and 75^th percentile and the dark band shows the median. The dots represent outliers. CIV = continuous infusion vancomycin; TDD = total daily dose

Fifty-four courses of CIV (76.1%) achieved a concentration within the goal range. Many of these patients (65%) achieved concentrations on the lower end of the defined range of 15 – 20 μg/mL, while the remainder (35%) fell within the higher end of the defined range of 20.1 – 25 μg/mL. Of these, 74.1% achieved therapeutic concentrations with the first concentration obtained. Time to initial therapeutic concentration was rapid at a median of 19.3 hours (range 8.6 - 72.3 hours). The remaining 17 courses (23.9%) did not achieve a serum concentration within the desired range with 47% of these patients experiencing supratherapeutic concentrations (> 25 μg/mL) and 53% having subtherapeutic concentrations (< 15 μg/mL). The most common reasons identified were a switch to alternative therapy (n = 6), subsequent completion of vancomycin within 72 hours (n = 5), and concentrations below the target range (lowest result of 14.4 μg/mL) for which no dose adjustments were made based on clinical judgment (n = 3). Patients treated solely in the inpatient setting (n = 22) had a median CIV therapy duration of six days (range 1 – 37 days), and patients who transitioned to the outpatient setting (n = 49) had a median CIV therapy duration of seven days (range 1 – 55 days).

Bacteremia was the most common indication for vancomycin therapy. Positive culture results were obtained in 81.7% of courses. Staphylococcus epidermidis was the most common organism identified in 60.3% of the courses with available culture results. Other common indications for the use of vancomycin therapy included skin and soft tissue infections, prosthetic device infections, and meningitis. Two patients (2.8%) were treated for MRSA bacteremia. Additional information is provided in Table, Supplemental Digital Content 1.

The main reasons identified for the transition from IIV to CIV were ease of outpatient use (69%) and difficulty achieving therapeutic concentrations with intermittent dosing (22.5%). The median trough concentration on IIV prior to transitioning to CIV was 5.5 μg/mL for patients that had difficulty obtaining therapeutic concentrations. No courses of CIV were discontinued early because of intolerability. AKI occurred during three (4.2%) courses of CIV. Two of the three patients that experienced AKI with CIV also received multiple concurrent nephrotoxic medications, including tacrolimus and furosemide. Vancomycin concentrations reached values of 15.0, 21.4, and 25.0 μg/mL in the three patients who experienced AKI. One patient had previously developed AKI on IIV prior to CIV initiation and completed a treatment course with CIV for septic thrombophlebitis with slow resolution of pre-existing AKI. No episodes of flushing or other symptoms of vancomycin infusion reaction were documented.

DISCUSSION

The present retrospective study reports on a single institution’s experience with CIV administration in pediatric and adolescent hematology and oncology patients. Vancomycin disposition in children is highly variable based on age, volume status, renal function, and coexisting conditions such as sepsis and malignancy.^22-24 The attainment of vancomycin concentrations that are commonly considered therapeutic can be challenging in the pediatric oncology population due to suspected augmented renal clearance.^{25, 26} The findings of this study provide additional insights into the complexity of vancomycin dosing in pediatric hematology and oncology patients and proposes a simplified dosing strategy based on body surface area.

CIV offers many benefits for pediatric patients, including the ability to quickly achieve therapeutic steady-state concentrations, the opportunity to provide less burdensome dosing for outpatient administration, and simplification of therapeutic drug monitoring. IIV dosing requires either troughs to be utilized as a surrogate marker for AUC or multiple concentrations to be drawn at specific time points to estimate the AUC. These can be especially challenging to time appropriately in pediatric patients receiving vancomycin every six hours. Conversely, CIV offers the benefit of a single concentration that can be drawn at any time of day concurrently with other scheduled laboratory blood draws to easily calculate AUC. Vancomycin can often require two weeks of therapy or longer depending on its indication.¹ The ability to transition to outpatient use of vancomycin has the potential to reduce healthcare costs associated with prolonged inpatient stays for antibiotic therapy. Administration of CIV has certain limitations, including the potential for incompatibility with other intravenous medications that may be required, the need for a central venous catheter for outpatient CIV administration, and the possibility of blood sample contamination in cases where only single IV access is available. To ensure that interruptions in the infusion are minimized, it is imperative to account for intravenous medication incompatibilities when administering continuous vancomycin to patients with restricted intravenous access. In patients with a solitary central access site, we recommend halting the vancomycin infusion for a period of five minutes, flushing the central line with normal saline, and drawing a small amount of blood to waste prior to collecting the vancomycin sample. This has been found to be a reliable method of preventing sample contamination.

Few papers have been published using CIV within the pediatric population, and even fewer within the pediatric hematology/oncology patient population. A randomized controlled trial conducted in infants compared initial IIV to CIV and demonstrated that CIV led to more rapid attainment of target vancomycin concentrations and required a lower total daily dosage. Additionally, there was no significant difference in the rate of drug-related adverse events between the two methods. These findings suggest that CIV infusion of vancomycin may offer clinical advantages over initial IIV infusion in infants.²⁷ A retrospective cohort study at Children’s Hospital Colorado described their experience using their institutional dosing guideline for CIV. Patients were stratified according to age (< 2 years, 2 – 8 years, or > 8 years) and goal serum vancomycin concentration (10 – 15 μg/mL or 15 – 20 μg/mL). This dosing strategy displayed variable effects of 41 – 82% attainment of therapeutic concentrations depending on stratification by age and goal concentration.¹⁸ Of note, patients in the 10 – 15 μg/mL cohort were unlikely to achieve therapeutic AUC/MIC ratios recommended by current guidelines. The selection of the target concentration range in this report was based on a comprehensive review of suggested AUC/MIC targets for severe MRSA infections, published concentration target ranges, and documented safety experiences in pediatric patients at vancomycin steady-state concentrations below 25 μg/mL for CIV.^1,12,19 Microbiological or other clinical data did not alter the study-defined therapeutic range, although in clinical practice this may need to be considered.

CIV dosing in our patient population was shown to be inversely correlated with age utilizing the traditional mg/kg approach. This is consistent with dosing practices required for IIV in pediatric patients. Interestingly, dosing scaled to BSA in our cohort of patients provided more consistent vancomycin exposure across a wide age range. This may allow for a more unified approach to dosing in children, similar to the approach to dosing chemotherapy. Sawrey et al. compared a body weight-based intermittent dosing regimen to a BSA-based intermittent dosing regimen in pediatric patients and found a BSA-dosing approach to provide higher initial vancomycin troughs.²⁸ Rapid attainment of therapeutic concentrations has the potential to reduce treatment failure for severe infections and reduce time spent with suboptimal dosing.

Of note, lower doses of CIV were required in this study to achieve therapeutic concentrations compared to initial IIV dosing of 60 – 80 mg/kg/day recommended by vancomycin dosing guidelines.¹ These findings are consistent with other recently published institutional cohort studies that also displayed lower TDD associated with continuous administration versus intermittent dosing.^29-31 Overall, lower TDD may be beneficial as higher daily doses of vancomycin have been shown to be associated with increased risk of nephrotoxicity.²⁹ Nephrotoxicity is a significant historical concern when using vancomycin, which requires vigilant monitoring of renal function throughout therapy.^2,13-17 However, lower variability in serum concentrations with use of CIV may mitigate this risk by providing more consistent drug exposure. Literature from both adult and pediatric patients has not shown an increased incidence of nephrotoxicity using CIV.³⁰ Consistent with these previously published studies, utilization of CIV did not elevate this risk in our patient population. An additional benefit of CIV is the ability to provide additional hydration with the CIV infusion itself, eliminating the need for an additional bag of IV fluids. The majority of CIV courses (n = 41; 69%) were administered concurrently with IV hydration fluids added to the CIV bag to prevent development of nephrotoxicity.

Limitations of this study include its retrospective nature and small sample size. TDD information was based on the first therapeutic concentration obtained. Although the majority of patients achieved therapeutic concentrations on CIV, doses were often later modified before course completion based on subsequent concentration values that were not evaluated in the scope of this retrospective study. Generalization of dosing found in this study to all patients should be done with caution due to small sample sizes in the infant and adolescent population in particular. Neonatal patients display differences in vancomycin clearance and other pharmacokinetic characteristics based on gestational age and are not represented in our cohort to guide dosing for this unique patient population.

The optimal dosing strategy for CIV in pediatrics is yet to be determined, as evidence is thus far limited to retrospective single-center cohorts. Future trials should feature a randomized comparison of a continuous infusion cohort with an intermittent infusion cohort to assess treatment outcomes and validate a vancomycin dosing strategy in pediatric patients. A BSA-based approach is especially interesting to evaluate, given our findings display more consistent TDD required to achieve a therapeutic concentration compared to traditional mg/kg dosing. Our findings in this retrospective study suggest that CIV is both an attractive and well-tolerated option for treatment of Gram-positive infections in pediatric patients. Therapeutic target serum concentrations are quickly obtained with CIV without the burden of frequent administration, especially in the outpatient setting. Infusion reactions were not shown to be a concern, and nephrotoxicity was minimal. To our knowledge, this is the first study to investigate a BSA-based dosing strategy for CIV in pediatric patients. Our future institutional approach to dosing CIV will likely seek to validate a BSA-based approach based on these findings.

Supplementary Material

Supplemental Digital Content (Including Legend)

Supplemental Digital Content 1. Antimicrobial Identification of Organisms and Indications for Use of Vancomycin and Initiation of Continuous Infusion [table]