Analysis of Drug Content and Weight Uniformity for Half-Tablets of 6 Commonly Split Medications

BACKGROUND: Cost savings can be achieved with the practice of tablet splitting. Previous research has shown weight nonuniformity within tablet halves. However, limited research to date has found that the potential dose inaccuracy resulting from splitting tablets does not significantly affect clinical outcomes. OBJECTIVES: To determine the drug content and weight in split half-tablets of 6 commonly split medications using drug assay analysis. METHODS: This study was performed by 2 fourth-year pharmacy students using 30 randomly selected tablets of each of the following 6 medications: warfarin sodium 5 milligrams (mg), simvastatin 80 mg, metoprolol succinate 200 mg, metoprolol tartrate 25 mg, citalopram 40 mg, and lisinopril 40 mg. A randomly selected half of the tablets were split by a single pharmacy student using a tablet cutter, and the remaining tablets were kept whole. Drug content was analyzed for 15 whole tablets and 30 half-tablets for each of the 6 drugs using high performance liquid chromatography, an analytical tool used to identify and quantify substances in solution. Drug content uniformity was assessed by comparing drug content within halftablets with one-half of the drug content mean found for all whole tablets in the sample. Weight uniformity was assessed by comparing half-tablet weights, as determined by a Mettler analytical balance, with one-half of the mean weight for whole tablets in the sample. The percentages by which each whole tablet's or half-tablet's drug content and weight differed from sample mean values were compared with proxy United States Pharmacopeia (USP) specification ranges for drug content (95%-105% for warfarin sodium and 90%-110% for the other 5 drugs). Additionally, these outcomes were compared for nonscored versus scored tablets. The percent relative standard deviation (%RSD, ratio of the standard deviation to the mean), a commonly used measure of the repeatability and precision of assays used to analyze drug content, was also calculated in order to determine whether the drugs met proxy USP specification for %RSD (less than 6% for all drugs studied). RESULTS: A total of 43 of 180 half-tablets (23.9%) differed from sample mean values by a percentage that fell outside of proxy USP specification for drug content; warfarin sodium (11 of 30 half-tablets, 36.7%), simvastatin (3 of 30 half-tablets, 10.0%) metoprolol succinate (10 of 30 half-tablets, 33.3%), metoprolol tartrate (4 of 30 half-tablets, 13.3%), citalopram (5 of 30 half-tablets, 16.7%), and lisinopril (10 of 30 half-tablets, 33.3%). Half-tablets outside of proxy USP specification for weight included warfarin sodium (10 of 30 half-tablets, 33.3%), metoprolol succinate (6 of 30 half-tablets, 20%), and lisinopril (7 of 30 half-tablets, 23.3%). The %RSDs for drug content and weight fell outside of the proxy USP specification for %RSD for metoprolol succinate (drug content = 8.98%, weight = 7.70%) and lisinopril (drug content = 10.41%, weight = 8.13%). Mean percent weight loss after splitting was less than 1% for all drugs except lisinopril, which had an average weight loss of 1.25%. The total numbers of scored (nonscored) tablet halves that fell outside of proxy USP specification were 20 (23) for drug content and 10 (13) for weight. When measuring drug content, the numbers of out-of-range half-tablets for scored (nonscored) drugs were 36 (44) at 95%-105%, 9 (23) at 90%-110%, 0 (10) at 85%-115%, and 0 (1) at 75%-125%. When measuring weight, the numbers of out-of-range halftablets for scored (nonscored) drugs were 28 (38) at 95%-105%, 0 (14) at 90%-110%, 0 (3) at 85%-115%, and 0 (0) at 75%-125%. CONCLUSIONS: Dose variation exceeded a proxy USP specification for more than one-third of sampled half-tablets of warfarin sodium, metoprolol succinate, and lisinopril and appeared to be greater for nonscored tablets as compared with scored tablets. Drug content variation in half-tablets.