When There Is No Trough: Use and Outcomes of Continuous-Infusion Vancomycin at a Free-Standing Children's Hospital

Abstract

OBJECTIVE

There is minimal published literature regarding the use of continuous-infusion vancomycin (CIV) in children. The objective of this study was to describe the use, dosing requirements, and outcomes of CIV at a free-standing children's hospital.

METHODS

This is a retrospective review of patients who received CIV while admitted to Nationwide Children's Hospital between July 1, 2010, and June 30, 2020. The total daily dose (TDD) of vancomycin required to attain a target serum vancomycin concentration (SVC) was compared between CIV and intermittent-infusion vancomycin (IIV) administration regimens. Safety outcomes and treatment failure were also explored.

RESULTS

Fourteen patients (77% male) with a median age of 7 years (IQR = 1, 10 years) were included. Most patients (71%) were started on CIV in anticipation of outpatient parenteral antimicrobial therapy. The median TDD required to achieve a target SVC was higher with IIV compared with CIV (82.4 mg/kg/day vs 50.5 mg/kg/day; p = 0.02). Despite higher TDD with IIV, median SVC with IIV was similar to SVC with CIV (16.6 mg/L vs 17.6 mg/L; p = 2.00). There were no safety concerns or therapeutic failures identified with CIV.

CONCLUSIONS

Continuous-infusion vancomycin was a well-tolerated and effective alternative to IIV for the patients included in this study. The TDD of vancomycin required to achieve a target SVC was lower in patients receiving CIV compared with those receiving IIV.

Introduction

Vancomycin is a mainstay in the treatment of severe infections caused by Gram-positive organisms, including methicillin-resistant Staphylococcus aureus (MRSA), coagulase-negative staphylococci, and beta-lactam–resistant Streptococcus species. Therapeutic drug monitoring is usually performed with vancomycin courses because of the drug's narrow therapeutic index. Recent consensus guidelines¹ on therapeutic monitoring of vancomycin for serious MRSA infections (2020 ASHP-IDSA Vancomycin Guideline) recommend area under the concentration-time curve (AUC) monitoring, with a target AUC to minimum inhibitory concentration (MIC) ratio of 400 to 600 for treatment of suspected or confirmed serious infections caused by methicillin-resistant Staphylococcus aureus (MRSA). This recommendation represents a shift away from trough serum vancomycin concentration (SVC) monitoring, which has historically been used as a surrogate for vancomycin AUC exposure based largely on adult data.

Vancomycin is usually administered as an intermittent infusion given every 6 to 8 hours in children and every 8 to 12 hours in adults with normal renal function. The increased frequency of administration in pediatric patients is due to enhanced vancomycin clearance that decreases with age.² In adult patients, continuous-infusion vancomycin (CIV) administration has shown similar outcomes and rates of nephrotoxicity compared with intermittent-infusion vancomycin (IIV) in adults who are critically ill or receiving outpatient parenteral antimicrobial therapy (OPAT).¹ Several studies evaluating CIV use in neonates have been published and were comprehensively reviewed by H Girand in a separate publication.³ Unfortunately, there is very little published literature on CIV use in children greater than 30 days of age, and the complex physiological changes that occur during the neonatal period significantly limit extrapolation of neonatal data to children.

Retrospective studies^4,5 have found that a CIV total daily dose (TDD) of 40 to 55 mg/kg/day may be needed to achieve steady-state SVCs of 15 to 20 mg/L in non–critically ill pediatric patients. These doses are larger than the 30 to 40 mg/kg/day recommended in the 2020 ASHP-IDSA Vancomycin Guideline¹ to target steady-state SVCs of 20 to 25 mg/L for initial CIV dosing in adult patients. The limited information available on CIV dosing and outcomes in pediatric patients led to the undertaking of this retrospective study. The primary objective of this study was to describe the use, dosing requirements, and outcomes of CIV at a free-standing children's hospital.

Materials and Methods

Patients who received CIV while admitted to Nationwide Children's Hospital between July 1, 2010, and June 30, 2020, were identified via query of the electronic medical record (EMR) and were reviewed for inclusion in the analysis. Patients were included if they were between 4 weeks and 18 years of age and had at least one appropriately drawn SVC within the target range assigned by the primary team while receiving CIV. Target ranges were determined by the primary team based on vancomycin indication and targeted pathogen and ranged between 10 and 20 mg/L.

The use, dosing requirements, and outcomes of CIV were described through TDD comparisons between IIV and CIV regimens, evaluation of safety outcomes, and reporting of treatment failures. The TDD required to achieve target SVC was compared between IIV and CIV regimens. Safety outcomes included acute kidney injury, defined as a rise in serum creatinine of greater than or equal to 1.5 times the baseline while receiving CIV, and hypersensitivity or infusion reactions that were documented in the EMR. Treatment failure was defined as persistent positive culture for greater than or equal to 7 days, recurrence of infection within 30 days of the end of CIV, or 30-day all-cause mortality.⁶ Data related to baseline demographics, culture results, isolated pathogens, vancomycin dose, SVCs, and renal function were collected.

Descriptive analysis was performed using frequency counts, percentages, medians, and IQRs, as appropriate. The Mann-Whitney U-test was used to compare TDD and SVC while on IIV and CIV. A p value of less than or equal to 0.05 was considered statistically significant.

Results

A total of 15 patients were reviewed for inclusion. One patient was excluded because a target SVC was not achieved prior to CIV discontinuation, leaving a total of 14 patients meeting the inclusion criteria for this study. Most patients were male (77%), with a median age of 7 years (IQR, 1–10). The most frequent indication for vancomycin was bacteremia (36%), followed by central nervous system infection (29%). Four patients (29%) had health care–associated infections related to central venous catheters or implanted devices. The most common isolated pathogen was MRSA (43%). All patients had vancomycin therapy initiated as IIV and were transitioned to CIV for one of 2 reasons: 11 patients (79%) were transitioned to CIV for convenience of administration in an outpatient setting, while 3 patients (21%) were transitioned to CIV because of trough SVCs that were persistently below their assigned target range on IIV despite dose escalation. Trough SVCs for monitoring of IIV were all obtained at steady state within 60 minutes of the next dose. Steady-state SVCs obtained during CIV were all collected at least 16 hours after initiation of CIV or dose change. The target SVC range determined by the primary team was either 10 to 15 mg/L (21%) or 15 to 20 mg/L (79%). The same target SVC range was used for both IIV and CIV regimens for all patients. Additional baseline characteristics are detailed in Table 1.

Table 1.

Patient Characteristics ^*

Parameter	Value
Age, median (IQR), yr	7 (1–10)
Male sex, n (%)	11 (77)
Weight, median (IQR), kg	23 (11–37)
Baseline serum creatinine, median (IQR), mg/dL	0.34 (0.22–0.58)
Vancomycin indication, n (%)
Bacteremia	5 (36)
Central nervous system infection	4 (29)
Infectious endocarditis	2 (14)
Skin and soft tissue infection	1 (7)
Osteomyelitis	1 (7)
No source identified	1 (7)
Isolated pathogen
MRSA	6 (43)
Coagulase-negative staphylococci	5 (36)
Brevibacterium species	1 (7)
No pathogen isolated	2 (14)
Target SVC range
15–20 mg/L	11 (79)
10–15 mg/L	3 (21)
Reason for CIV initiation
Outpatient administration convenience	11 (79)
Persistently low trough SVCs with IIV	3 (21)

N=14; IIV, intermittent-infusion vancomycin; MRSA, methicillin-resistant Staphylococcus aureus; SVC, serum vancomycin concentration

^* All values are n (%), unless otherwise indicated.

Ten patients achieved target SVCs on both IIV and CIV. Among these patients, the difference in median TDD (in milligrams per kilogram per day) required to achieve target SVC was found to be significantly lower when patients received CIV compared with IIV (50.5 mg/kg/day vs 82.4 mg/kg/day, respectively; p = 0.02). The median TDD of vancomycin required to achieve target SVC with CIV administration was reduced by 39% compared with the TDD of IIV. The median SVC was similar between CIV and IIV regimens (17.6 mg/L vs 16.6 mg/L, respectively; p = 2.00), despite the lower TDD seen with CIV (see Table 2).

Table 2.

Comparison of Total Daily Dose Required to Achieve Target Serum Vancomycin Concentrations With Different Infusion Methods

	IIV	CIV	p value
TDD (IQR), mg/kg/day	82.4 (58.5, 103.4)	50.5 (34.1, 66.5)	<0.02
Percent reduction in TDD		38.7
Serum vancomycin concentration (IQR), mg/L	16.6 (15.1, 17.9)	17.6 (16.1, 19.2)	2.00

n=10; Data presented as median and IQR; CIV, continuous-infusion vancomycin; IIV, intermittent-infusion vancomycin; TDD, total daily dose; SVCs for IIV were all obtained as trough concentrations within 60 minutes of the next dose; SVCs for CIV were all obtained as random steady-state concentrations at least 16 hours after initiation or dose change

Three patients had trough SVCs that were persistently below their assigned target range with IIV before transition to CIV. Two of these patients achieved a target SVC on a lower TDD of CIV (25% and 31% reductions in TDD), and the third had a doubling of SVC when converted to CIV, with a 5% increase in TDD. No patient experienced treatment failure with CIV, and there were no incidences of infusion reaction, hypersensitivity reaction, or acute kidney injury identified.

Discussion

Continuous-infusion vancomycin was a well-tolerated and effective alternative to IIV for the patients included in our study. The majority of our patients were transitioned to CIV in preparation for OPAT because the once-daily infusion change that occurs with CIV is generally more convenient in the outpatient setting compared with multiple-daily infusions. Continuous-infusion vancomycin for OPAT has been described in the adult literature^7–9 with similar outcomes and a trend toward lower rates of nephrotoxicity compared with IIV. Given our small sample size, we were unable to confirm these findings in children.

The remaining patients in our study were transitioned to CIV because of trough SVCs that were persistently below their assigned target range while on IIV. All 3 of these patients achieved target steady-state SVCs with CIV. The 2020 ASHP-IDSA Vancomycin Guideline¹ recommends the use of AUC monitoring over trough SVC monitoring for serious MRSA infections. There is evidence^10,11 that children achieve AUC targets with lower trough SVCs compared with adults.

With the move to AUC monitoring we may find that children are able to achieve therapeutic targets with lower TDD, reducing the need to transition to CIV. However, home-going convenience will still be a factor in deciding to transition to CIV.

Patients receiving CIV had a significantly lower overall exposure to vancomycin, as evidenced by lower TDD to achieve target SVCs compared with IIV TDD. While trough SVCs obtained during IIV and steady-state SVCs obtained during CIV are not directly comparable, the patients in our study had the same SVC target regardless of the administration technique. This allowed for a comparison of the TDD needed to achieve target SVCs between IIV and CIV administration. We found that patients achieved a similar SVC while on a lower TDD of vancomycin while receiving CIV vs IIV (median SVC of 17.6 mg/L vs 16.6 mg/L and median TDD of 50.5 mg/kg/day vs 82.4 mg/kg/day, respectively). In critically ill adults, CIV administration has resulted in similar rates of nephrotoxicity when compared with IIV.^12,13 Given our small sample size and the lack of adverse events attributable to vancomycin in our population, we were unable to assess if the lower overall vancomycin exposure with CIV resulted in a decrease in adverse reactions.

Efficacy in treating severe MRSA infections has been associated with attaining an AUC/MIC ratio of 400 to 600, which can be roughly correlated to trough SVCs of 15 to 20 mg/L in adult patients. As such, trough SVC monitoring was previously recommended by the 2009 ASHP-IDSA Vancomycin Guideline,¹⁴ given ease of obtainment and interpretation compared with AUC monitoring. However, the most recent 2020 ASHP-IDSA Vancomycin Guideline recommends the use of AUC monitoring over trough SVC monitoring for serious MRSA infections. In regard to CIV, the guidelines recommend targeting a steady-state SVC of 20 to 25 mg/L, corresponding to an AUC of 480 to 600 mg·hr/L (SVC multiplied by a factor of 24).¹ However, patients in our study had lower target steady-state SVCs on CIV (10–20 mg/L). The median steady-state SVC observed in our study of 17.6 mg/L for patients receiving CIV corresponds to an AUC of 422 mg·hr/L (SVC multiplied by a factor of 24) and is lower than the steady-state target of 20 to 25 mg/L recommended in the 2020 ASHP-IDSA Vancomycin Guideline. However, we did not identify any treatment failures among our patients. Of note, the 2020 Guidelines are specific to severe MRSA infections, which was the vancomycin indication for only 43% of patients included in this study. Lower vancomycin SVC or AUC targets may be acceptable for patients with non-MRSA infections. Hurst and colleagues⁴ and McKamy and colleagues⁵ also described the use of lower target steady-state SVCs for CIV (10–20 and 15–20 mg/L, respectively) in non–critically ill children for both MRSA and non-MRSA infections but did not report incidence of treatment failure. Additional research is needed to determine if lower steady-state SVC targets are acceptable for children receiving CIV.

Our study has several limitations. First, the retrospective data collection required reliance on accurate EMR documentation to identify infusion and hypersensitivity reactions. Furthermore, the small sample size limited our ability to assess whether the trend towards lower rates of nephrotoxicity with CIV found in adult studies was also true in children; although, it is reassuring that no patient in our study experienced an infusion or hypersensitivity reaction, acute kidney injury, or treatment failure while receiving CIV. Finally, the heterogenicity of age, vancomycin indication, isolated pathogens, and target SVCs among our population may limit generalizability.

In conclusion, CIV was well tolerated in the patients included in this study, and the TDD of vancomycin required to achieve a target SVC was significantly reduced when patients were transitioned from IIV to CIV. This study was limited by the retrospective nature of the evaluation, small sample size, and its heterogeneous population. While CIV was a well-tolerated and effective alternative to IIV for the patients included in our study, more research is needed to determine optimal dosing, therapeutic targets, and long-term safety and efficacy of CIV in pediatric patients.

ABBREVIATIONS

AUC

area under the concentration-time curve

CIV

continuous-infusion vancomycin

EMR

electronic medical record

IIV

intermittent-infusion vancomycin

MIC

minimum inhibitory concentration

MRSA

methicillin-resistant Staphylococcus aureus

OPAT

outpatient parenteral antimicrobial therapy

SVC

serum vancomycin concentration

TDD

total daily dose