Quality Assurance when Using Therapeutic Drug Monitoring: A Lamotrigine Split Sample Method to Reflect Comparability

Roy G Beran1* and Sandiya Sathiyaseelan2

Australia

Sydney, Australia, Tel: (02) 9411 4991; Email: research@sheresearch.com patients with epilepsy.

further evaluated to compensate for absolute and/or clinically significant differences.

difference in the levels reported by the two labs was neither significant nor statistically different.

assurance when changing labs.

Introduction

Therapeutic drug monitoring (TDM) reflects the measurement of the level of a given medication, in the blood of patients being treated with that medication, to reflect that sufficient amounts of that medication have been prescribed to achieve a desired therapeutic result. It offers a valuable adjunct to patient care [1] especially in the management of epilepsy which is a chronic condition in which antiepileptic medications offer the best form of intervention to control the seizures in the majority of patients who are prone to seizures. The value of TDM in patients with epilepsy is to determine anti-epileptic medication (AEM) blood levels, referable to a patient’s state of health and wellbeing, at the time of measuring the AEM concentration [2]. Monitoring AEM levels may prevent breakthrough seizures by identifying non- compliance and may identify the offending AEM if there exists toxicity [3, 4]. It follows that the more immediate the result, the better it reflects the effect of the AEM at that time [5]. Blood samples for newer AEM’s, such as Lamotrigine (LTG) (branded product, Lamictal®) which is an antiepileptic medication that acts at the sodium channel to reduce the post synaptic potentiation of epileptic discharges, are not widely or immediately available St Louis, et al. [6]. As a result, some clinicians are uncertain about proper use of AEM blood level determination for newer AEM’s, due to a lack of confidence if levels cannot readily be measured [6].

The laboratory (lab) that is commonly used by the clinic in this study, to assess LTG levels, required 2 weeks to provide results with no allowance for urgency, either in the face of ongoing seizures (when compliance may be in question) or at times of apparent toxicity (when dosages or drug-interactions may be highly relevant). To counter these issues, the clinic identified an alternative lab which could provide results of LTG levels within a 24-hour window, thereby enhancing the capacity to interpret such results for an individual patient.

Studies have demonstrated that results from the same sample, if analysed separately, can yield different results as these can be affected by lab techniques, methods of handling samples, reagents or equipment that is used Abdollahi, et al. [7], Phillips, et al. [8]. Such discrepancies can result in less efficient management and deny optimal patient care by contributing to unnecessary or inappropriate dose adjustments [9, 10].

As a means of quality assurance, the reliability of the lab to which the clinic chooses to send its patients’ samples should be critically assessed. While most reputable labs are accredited by the National Authorities of Testing, Australia (NATA) [11], it is not until there is a direct comparison between labs that the clinician can feel suitably comfortable in accepting results.

The clinic has an established idiosyncratic therapeutic range for LTG at 10-16 mg/L (40-60µmol/L) Patel, et al. [12]. Based on the results from the clinic’s usual lab. Prior to switching labs, quality assurance was needed to ensure that the results of the second lab mirrored those of the first lab, to satisfy physician confidence in interpreting the results.

The aim of this study was to determine if there existed a confounding variation in lab results for LTG between the original lab and the lab providing more immediate results. This was undertaken in the present study using the process of split sampling.

Methods

During June 2017-January 2018 inclusive, 50 blood samples were drawn from patients, attending a private neurology clinic in Sydney, Australia, who were receiving LTG as adjunctive therapy to treat their epilepsy. The samples were split and sent to two labs in Australia with each patient made aware that this approach to quality assurance was being undertaken to ensure that, should the clinic use an alternative laboratory to provide more immediate results which would better reflect the immediacy of the TDM, the results of the test would be comparable to those currently being used. All patients involved in the study gave informed consent to the study and acknowledged that it was designed to accommodate their best interests and ensured that, before changes were made, quality assurance both should and would prevail.

The LTG results reported by the two labs were collated and evaluated in the following three ways: Category Variation The results were categorised in an Excel spreadsheet (Table 1) in accordance with the clinic’s LTG Reference Range (Table 2) to determine if such categorisation varied between the two laboratories. Statistical Significance Statistical Modelling Software, SPSS, was used to determine whether the results reported by the two labs were significantly different. A two-tailed ‘T-test’ sampling procedure was used with the following hypotheses:

Null hypothesis: H0: µ1=µ2 Alternative hypothesis: Ha: µ1≠µ2.

The compared results were further evaluated to compensate for clinical significance. Where results were so close to the three subcategories (namely within 1mg/L), thereby not sufficiently deviating to suggest a change in LTG regimen, this was accepted as a comparable result.

For this analysis, a P value of <0.05 was considered significant for rejection of the null hypothesis.

Clinical Significance

Discussion

It is important to ensure that AEM levels are reported and analysed in a timely fashion to increase clinical relevance. If the clinic has an established lab in which it routinely measures levels, the clinic should ensure that any alternative lab does not report significantly different results for the same test. This equates to quality control and underpins physician confidence in interpreting the results. The results from the current study appear to show that the labs being compared provided comparable AEM levels for LTG; thereby supporting the move to the new lab, without compromising patient safety and wellbeing. The move would ensure that the two-week delay was no longer an issue and that results were reflective of the patient’s current circumstances, thereby removing unnecessary doubt regarding compliance or toxicity when considering potential changes in a patient’s therapy regimen. The minor discrepancy found in seven (14%) samples was insufficient to generate concern to change dosages and reinforced the notion that use of AEM levels is of value as an adjunct to clinical decision making while it does not reflect an absolute indication for change. The discrepancy found in two samples, in which one sample was therapeutic and the other, supra-therapeutic (in both cases) would be grounds for some concern, but, again, if the level was at odds with the patient’s clinical history, a repeat measurement would have resulted.

A limitation of the study is the small sample size, which was restricted to 50 patients, thereby limiting the power to identify absolute difference between labs, when subjected to statistical analysis. The contrary view would be that the concordance between labs was so great (96%), for a test which is considered within the context of clinical decision-making based on patient status, that the results are sufficient to allow the clinician to change labs with a high degree of reliability.

Split sample studies, as presented in this report, offer the clinician a simple and effective means of achieving confidence when changing labs without jeopardising patient safety. This technique has generated a real example of how one can improve the use of TDM in the treatment of people with epilepsy to enhance the relevance and immediacy of the results for decisions in patient care.

Conclusion

TDM is a useful adjunct to patient care if results are processed efficiently. Split-sampling procedures are a viable method of testing whether there is a comparable difference in results before choosing to switch labs. On the basis of these data, it would seem reasonable to use lab 2 as it provides more immediate, and hence more representative, results. The reduction in time (2 weeks versus 24 hours) allows the clinician to have better control of the situation without compromising patient care or relying on flawed data.