It is common practice to dilute food products in 0.1% peptone before microbiological analysis. However, this diluent may not be appropriate for detection of injured organisms present in acidic foods. Shelf-stable unclarified apple juice (pH 3.6) was inoculated with approximately 1 x 10(7) CFU/ml of Escherichia coli O157:H7 and held at 23 +/- 2 degrees C (control) or frozen to -20 +/- 2 degrees C for 24 h to induce injury before sampling. Unfrozen or frozen and thawed juice was diluted 1:1 or 1:10 in 0.1% (wt/vol) peptone (pH 6.1) or 0.1 M phosphate buffer (pH 7.2). Juice samples were plated onto tryptic soy agar with 0.1% (wt/vol) sodium pyruvate (TSAP) to measure survival or onto sorbitol MacConkey agar (SMA) to indicate injury. Counts on TSAP or SMA were the same for control samples held in peptone or phosphate buffer for up to 45 min. However, populations of E. coli in frozen and thawed samples declined rapidly upon dilution in 0.1% peptone. Within 20 min, E. coli underwent a >1-log10 CFU/ml reduction in viability as measured on TSAP and a >2-log10 CFU/ml reduction to below the limit of detection (1.6 or 2.3 log10 CFU/ml) on SMA. In contrast, populations of E. coli in frozen and thawed samples diluted in phosphate buffer did not decrease significantly on TSAP and decreased by <0.6 log CFU/ml on SMA during a 45-min holding period. The acidity of apple juice appears to interfere with the recovery of freeze-thaw-injured E. coli O157:H7 during sampling. Using 0.1 M phosphate buffer (pH 7.2) as a diluent results in superior recovery of these organisms on both selective and nonselective plating media.