b B y repeating Etests on 36 blood isolates of methicillin-resistant Staphylococcus aureus (MRSA) over 9 months, we explored the effects of isolate storage on vancomycin and daptomycin MICs. We identified overall declines in vancomycin and daptomycin MICs taken from the same isolates at 3-month intervals (P Ͻ 0.001). Declines of Ն1 doubling dilution were observed in 75% and 67% of isolates for vancomycin and daptomycin MICs, respectively. Effects of storage should be considered in evidence of MIC "creep."Decreasing effectiveness of vancomycin in the treatment of methicillin-resistant Staphylococcus aureus (MRSA) infections has been linked to increasing vancomycin MIC within the susceptibility range (Յ2 mg/liter) (5). Subinhibitory vancomycin exposure may also be a risk factor for daptomycin heterogenous susceptibility (18,22,28). However, evidence of MIC "creep" is mixed (24), and the implications for treatment decisions of a vancomycin MIC that is high but still within the susceptibility range are uncertain (5). We previously demonstrated that detection of vancomycin and daptomycin MIC creep may be method dependent (7). In this study, we further explored the hypothesis that isolate storage may explain divergent results in investigations of susceptibility trends.A prospective repeated-measure design involved susceptibility testing of MRSA bloodstream isolates (BSI) at the time of isolation and at 3-month intervals. A 9-month follow-up period was chosen on the basis of previous data suggesting that changes in susceptibility occurred within this time. Our sample included all nonreplicate MRSA BSI identified in adult, nonobstetric patients at Aberdeen Royal Infirmary (Scotland) between January and March 2011. Susceptibility testing at the time of isolation was by Etest performed in accordance with the manufacturer's guidelines on Mueller-Hinton agar (Oxoid), and results were read blind in duplicate. Each isolate was then stored in a Cryobank storage container (Mast Diagnostics), which contained glycerol, peptones, sucrose, and saline, at Ϫ70°C. Isolates were recovered from storage at 3, 6, and 9 months and subcultured twice prior to repeat Etests. Assessors were blinded to previous MIC results.Interobserver agreement was assessed by weighted Cohen's kappa (). Friedman tests were used to test for overall difference in MICs across repeated readings, with post hoc analyses by Wilcoxon signed-rank tests with Bonferroni adjustment (␣ ϭ 0.0083). Variation in trends was investigated by independent-sample t test to assess differences in means of slopes for isolates grouped by "high" (Ͼmedian) or "low" (Ͻmedian) baseline MICs. All analyses were repeated with MICs from Etest converted to doubling dilutions by rounding up intermediate increments.To evaluate the potential for systematic bias arising from differences in storage of isolates before susceptibility testing in stud-