Falsely elevated vancomycin plasma concentrations sampled from central venous implantable catheters (portacaths)

Cefepime therapeutic drug monitoring: Evaluation of agreement between peripheral and central venous blood sampling

BACKGROUND

Therapeutic drug monitoring for cefepime is increasingly being performed because of the potential relation between exposure and neurotoxicity. An in vitro pilot study suggested significant carryover of cefepime from central venous catheters when blood sampling is carried out via the same catheter used for administration of cefepime. Therefore, the aim of this study was to evaluate carryover of cefepime in a real-life clinical setting.

METHODS

A prospective single-center study was conducted at the hematology department of the University Hospitals Leuven. Patients treated with cefepime, and having a central venous catheter, were included. Cefepime trough samples were taken simultaneously via the central venous catheter and peripheral venepuncture.

RESULTS

Twenty-four patients were included in this study, resulting in 28, 11 and 5 paired samples for tunnelled catheters, implantable port catheters and peripherally inserted central catheters, respectively. No statistically nor clinically significant difference was found between cefepime concentrations measured in centrally versus peripherally obtained blood samples, overall and for all three types of central venous catheters separately. Of note, four paired samples showed a difference larger than 10%, with lower central concentrations probably reflecting a dilution error.

CONCLUSION

There was no significant carryover of cefepime from long-term central venous catheters. Cefepime samples can be drawn reliably via the central venous catheter, after flushing and discarding the first blood sample. Although, flushing and discard volumes should be standardized to avoid potential dilution errors.

An investigation of enzymatic creatinine interference in a patient receiving dopamine and dobutamine

BACKGROUND

A laboratory investigation was initiated after a renal failure patient had a 2.18 mg/dL decrease in serum creatinine, which was not explained through medical intervention. The investigation revealed specimens providing questionably low results had been collected from a peripherally inserted central catheter (PICC) line.

METHODS

Patient specimens and serum pools were analyzed by the Siemens Vista enzymatic creatinine measurement procedure. A simulation of the patient's infusion protocol examined potential PICC line carryover and specimen collection technique.

RESULTS

A simultaneously collected specimen set, arterial line and PICC line, yielded a difference of 1.86 mg/dL. Infusion and collection simulation studies suggested the most likely scenario was the infusion pump was not shut off while the specimen collection occurred and contaminated the specimen.

CONCLUSION

Providers should be aware of erroneously low creatinine results when administering catecholamine drugs and collecting specimens through the same catheter. The mechanism of specimen contamination is consistent with a siphoning effect from one lumen to the other during collection with the infusion pumps still running.

Invasive fungal infections in high-risk patients: report from TIMM-8 2017

Trends in Medical Mycology (TIMM) is the biennial meeting of the Infectious Disease Group of the European Organisation for Research and Treatment of Cancer (EORTC) and the European Confederation of Medical Mycology (ECMM). It brings together clinicians and researchers from across the world to share the latest R&Ds in medical mycology. Despite advances in treatment, invasive fungal infections remain a major cause of morbidity and mortality in certain high-risk groups of patients, particularly in immunocompromised patients, including those undergoing solid organ transplantation and those with acute leukemia. The challenges for clinicians are now compounded by the rapid development of multidrug resistance. The latest data and approaches to identifying patients at high risk for invasive fungal infections, ensuring early diagnosis and achieving effective treatment, including when and how to use therapeutic drug monitoring with azoles, were shared with >1000 clinicians and researchers from around the world attending the eighth TIMM, held in Belgrade, Serbia, in October 2017 (TIMM-8 2017).

Therapeutic antibiotic serum concentrations by two blood collection methods within the pediatric patient: A comparative effectiveness trial

Repeated venipunctures and fingersticks to confirm serum drug concentrations cause pain and dissatisfaction for pediatric patients and their families. In many organizations, the standard of care to obtain therapeutic serum drug concentrations by peripheral venipuncture or capillary fingerstick, even when the patient has an existing peripheral intravenous catheter (PIV) or central venous catheter (CVC). The primary objective of this study was to assess agreement between serum tobramycin/vancomycin concentrations collected from a CVC or PIV, versus venipuncture or fingerstick. Among hospitalized pediatric patients (age 3 months to 22 years), 36 paired blood samples were collected. Serum trough vancomycin and random tobramycin concentrations were compared between peripheral intravenous or CVC samples, and venipuncture or fingerstick samples within the same patient. A strict sampling protocol for obtaining the samples was followed, that included collection of the CVC/PIV sample before the venipuncture or fingerstick, less than 2 min between collections of samples from the different sites, and a strict volume-based flush and waste protocol. Concordant correlation coefficients demonstrated substantial agreement between CVC/PIV and venipuncture/fingerstick concentrations for vancomycin (n = 17) and tobramycin (n = 19) (Rc = 0.982 for both). Bland-Altman analyses demonstrated good overall between-method agreement within subjects and minimal bias. Consequently, using a lumen volume-based flush and waste protocol, children with indwelling catheters may not require additional venipunctures and/or fingersticks for confirmation of drug concentrations while hospitalized, improving the quality of care and patient satisfaction.

Assessment of the Measurement Error in Cyclosporine Levels Drawn Between Peripheral and Central Sources

OBJECTIVES

Cyclosporine is often monitored by drug levels drawn through central venous catheters (CVCs), which may be falsely elevated due to reversible drug adsorption onto the catheter. Therefore, we assessed the correlation between cyclosporine levels drawn peripherally and through CVCs.

METHODS

Bone marrow transplantation patients had a weekly collection of both peripheral and CVC draws from dual-lumen catheters simultaneously to assess cyclosporine levels after research ethics approval. Our primary outcome was the proportion of paired samples that were incongruent-defined as the mean of the CVC level being greater than 2 standard deviations from the peripheral level mean.

RESULTS

After approaching 27 eligible patients, 20 patients (77.8%) provided samples. Of 53 paired samples, seven were incongruent (13.2%). Peripheral and CVC levels correlated (r = 0.91) and agreed well.

CONCLUSION

Despite potential for preanalytical error due to adsorption, cyclosporine infusion and monitoring via CVCs produce results similar to monitoring via peripheral blood draws.

High-Performance Liquid Chromatography Method for Rich Pharmacokinetic Sampling Schemes in Translational Rat Toxicity Models With Vancomycin

A translational need exists to understand and predict vancomycin‐induced kidney toxicity. We describe: (i) a vancomycin high‐performance liquid chromatography (HPLC) method for rat plasma and kidney tissue homogenate; (ii) a rat pharmacokinetic (PK) study to demonstrate utility; and (iii) a catheter retention study to enable future preclinical studies. Rat plasma and pup kidney tissue homogenate were analyzed via HPLC for vancomycin concentrations ranging from 3–75 and 15.1–75.5 μg/mL, respectively, using a Kinetex Biphenyl column and gradient elution of water with 0.1% formic acid: acetonitrile (70:30 v/v). Sprague‐Dawley rats (n = 10) receiving 150 mg/kg of vancomycin intraperitoneally had plasma sampled for PK. Finally, a catheter retention study was performed on polyurethane catheters to assess adsorption. Precision was <6.1% for all intra‐assay and interassay HPLC measurements, with >96.3% analyte recovery. A two‐compartment model fit the data well, facilitating PK exposure estimates. Finally, vancomycin was heterogeneously retained by polyurethane catheters.

Unexplained increases in serum vancomycin concentration in a morbidly obese patient

INTRODUCTION

To report a case of increases in vancomycin concentrations without additional vancomycin doses being given.

CASE STUDY

A 64 year-old morbidly obese female received three total doses of vancomycin for surgical prophylaxis and for ventilator-associated pneumonia. Subsequent vancomycin concentrations from the patient's central venous catheter (CVC) demonstrated increasing drug levels from 27.1 to 45.9mcg/mL despite no additional vancomycin being given and proper line flushing prior to sample collection. There is no clear explanation for the increase in the patient's vancomycin concentration. Drug leaching from the CVC, enterohepatic recycling, drug redistribution from adipose or other tissues, and assay cross-reactivity with other medications are all potential explanations for the increased vancomycin concentrations.

CONCLUSION

This case report describes an unexplained increase in vancomycin concentrations and reinforces both the fallibility of laboratory testing and that unusual circumstances do occur. Several potential causes are hypothesised with CVC drug leaching being the most likely. Nurses and other healthcare providers with similar scenarios should consider a peripheral blood sample to rule out the potential for CVC drug leaching as a possible explanation.

A Comparison of Antibiotic Serum Concentrations Drawn Simultaneously from Peripherally Inserted Central Catheters and Peripheral Veins in Children with Respiratory Infection

The purpose of the study was to evaluate serum concentration of antibiotics drawn from a peripherally inserted central catheter (PICC) compared with a peripheral venipuncture. This prospective comparative study included patients with ages 1month to 21years admitted with a respiratory infection requiring IV vancomycin or IV tobramycin via a newly placed PICC. The difference between the antibiotic levels from the venipuncture and PICC samples was statistically significant for both the peak and trough levels. However, the difference in values was not enough to impact antibiotic dosing and therefore was not clinically significant.

Sources of preanalytical error in pharmacokinetic analyses - focus on intravenous drug administration and collection of blood samples

INTRODUCTION

Pharmacokinetic (PK) studies for long-established drugs are generally performed outside the well-standardized settings of pharmaceutical industry trials. Instead, such studies are usually performed within daily clinical practice of hospitals.

AREAS COVERED

This article describes aspects of intravenous (i.v.) drug administration and blood sampling that contribute to potential sources of preanalytical errors for PK investigations. Parameters that bias determination of start and end time of i.v. infusions, as well as consistent rate of drug delivery, are discussed. Causes for drug loss in the infusion device, including adsorption and insufficient flushing, are outlined. The advantages and disadvantages of different blood sampling techniques are reviewed, with an emphasis on pediatric studies.

EXPERT OPINION

For PK studies that are integrated into the general hospital routine, a variety of potential sources of error exist. Potential pitfalls depend on the specific drug and trial characteristics and they must be anticipated and discussed in advance. Working procedures need to be developed that address the anticipated problems and in detail describe procedures that need compliance between bed and bench.

Comparison of central venous catheter and peripheral vein samples of antibiotics in children with cystic fibrosis

PURPOSE.: To determine if accurate serum antibiotic levels can be obtained from central venous catheters (CVCs) in pediatric patients with cystic fibrosis. DESIGN AND METHODS.: Fifty paired CVC-peripheral vancomycin or tobramycin specimens were collected within 5 min of each other following a 5-ml flush and discard. Specimen samples were randomized by first site drawn. RESULTS.: CVC-peripheral antibiotic levels were highly correlated (r =.97, p <.001), with no statistically significant difference (t = 1.18, p =.25). Bland-Altman plot analysis revealed a bias of.47, with limits of agreement ranging from -4.20 to 6.87. PRACTICE IMPLICATIONS.: Accurate antibiotic concentrations can be obtained from CVCs, reducing pediatric patient trauma and stress.

Minimization of the preanalytical error in plasma samples for pharmacokinetic analyses and therapeutic drug monitoring--using doxorubicin as an example

BACKGROUND

There are many sources of variability in plasma samples drawn for pharmacokinetic analyses or therapeutic drug monitoring. In this article, methods are proposed on how to prevent sample dilution (Part I) and contamination effects (Part II) in plasma samples, using doxorubicin as an example.

METHODS

Experiments were performed in the laboratory setting to identify factors that could influence plasma samples in clinical practice. In part I, it was hypothesized that saline solution left in a catheter could lead to a dilution of samples drawn through this catheter. The impact of 2 different sampling techniques, the "discard method" and the "push-pull method", was examined. In part II, an infusion system was filled with a 1 mg/mL solution of doxorubicin. After rinsing the system with increasing volumes of saline solution, the drug concentration of the fluid left in the system was analyzed. Furthermore, plasma samples were drawn through the drug administration catheter, and the contamination of these samples with doxorubicin left in the catheter was measured.

RESULTS

In part I, a discard volume of plasma equal to 4 dead volumes of the sampling line was necessary to avoid dilution of a sample taken from a port or double-lumen catheter filled with saline solution ("discard method"). Pulling up and down the same volume through the catheter 5 times ("push-pull method") was proved to be an alternative with no need to discard blood. In part II, after rinsing the infusion system with a volume of saline solution corresponding to 4 dead volumes of the system and after discarding a volume of plasma corresponding to 4 sampling line volumes, the doxorubicin contamination in the samples was negligibly small.

CONCLUSIONS

Under the described conditions, the push-pull method delivered the same results as the discard method to prevent sample dilution. To avoid contamination in plasma samples, development of standardized sampling procedures seems to be essential and feasible.

A reliable and safe method of collecting blood samples from implantable central venous catheters for determination of plasma gentamicin concentrations

STUDY OBJECTIVE

To evaluate the extent of agreement between plasma gentamicin concentrations determined from samples collected by using implantable subcutaneous central venous catheters (ports) with the push-pull method and those collected by finger lancet punctures in children with febrile neutropenia.

DESIGN

Prospective, randomized study.

SETTING

University-affiliated, tertiary care hospital.

PATIENTS

Sixty-two children with cancer who had single- or double-lumen ports and who received gentamicin for treatment of febrile neutropenia between February 2008 and October 2009.

INTERVENTION

One blood sample was collected from the port by using the push-pull method at the same time one blood sample was collected by finger lancet puncture for determination of plasma gentamicin concentrations.

MEASUREMENTS AND MAIN RESULTS

Forty-four pairs of samples were available for assessment of agreement, and 43 were available for pharmacokinetic analysis. Agreement between plasma gentamicin concentrations determined from blood samples from ports and finger lancet punctures was assessed by the intraclass correlation coefficient (ICC), Bland-Altman analysis, and comparison of simulated dosage adjustments. Changes in port patency were monitored for 1 week after port sampling. Differences in simulated dosage adjustments calculated by using either the port or finger lancet puncture samples that differed by greater than 20% were considered clinically significant. Agreement between the 44 finger lancet puncture and port sample pairs was excellent (ICC 0.991, 95% confidence interval 0.984-0.995). Port plasma gentamicin concentrations were 4.7% lower than those concentrations determined in blood from finger lancet punctures. The observed limits of agreement ranged from -20.5% to 11%. Differences in dosage adjustments calculated by using port and finger lancet puncture plasma gentamicin concentrations were not clinically significant in 38 (88%) of 43 cases. No changes in port patency were observed in the week after port sampling.

CONCLUSION

The push-pull method of blood sampling is a reliable and safe option for determining plasma gentamicin concentrations in children with ports.