An audit of sodium valproate assays

Several authors have questioned the usefulness of serum valproic acid analysis in the routine management of patients with epilepsy. This audit assesses the current practice in a Scottish NHS Trust laboratory for requesting sodium valproate assays and the impact of an intervention on the number and “appropriateness” of the assays. A literature search suggested that assays are only of benefit in patients who are poorly controlled despite optimal dosage, where there is suspected toxicity or where compliance is questioned. A retrospective data collection covering a two‐year period was performed, followed by a 12‐month prospective collection. All sodium valproate assays requested were studied and the reason for requesting them, the assay result and the dose recorded. Throughout both periods the number of other assays was also recorded. Sodium valproate assays were considered “appropriate” if the reason they were requested was poor control, suspected toxicity or doubts about compliance. An arbitrary dose of 2000mg/day was chosen as optimal dosage. Between the two data collection periods an intervention was introduced, namely, the use of two new interpretation comments which reflected the results of the literature search and were used on most of the sodium valproate assays reported. Following the intervention there was a significant fall in the number of valproate assays performed ( P = 0.046), an increase in the number of assay results outside the “target range” ( P <0.001), a change in the distribution of reasons for the requests ( P = 0.030) and an increase in the number of “appropriate” assays ( P <0.001).