The presence of Mycobacterium avium complex (MAC) in stool specimens may be a predictor of disseminated MAC infection, yet the methods for decontaminating stools have not been evaluated for their usefulness in recovering MAC organisms. In the present study, four decontamination methods commonly used to recover acid-fast bacteria from respiratory specimens were compared for their utility in recovering MAC from stool specimens. Ten strains of MAC were used at a level of 10(4) to 10(6) CFU to seed the stool specimens. Specimens were divided into four portions and were decontaminated by using the following treatments: (i) N-acetyl-L-cysteine-sodium hydroxide (NALC-NaOH), (ii) cetylpyridinium chloride-sodium chloride (CPC-NaCl), (iii) oxalic acid, or (iv) benzalkonium chloride-trisodium phosphate (BC-TSP). The specimens were then plated onto a total of five pieces of selective and nonselective egg- and agar-based media. The oxalic acid method yielded the greatest number of MAC CFU from seeded stool samples; this was followed by NALC-NaOH, BC-TSP, and CPC-NaCl. The difference between the oxalic acid method and each of the other methods was statistically significant (analysis of variance at the 95% significance level). Although more MAC CFU was recovered from seeded stool samples by using oxalic acid than NALC-NaOH, no difference in culture positivity rates was observed when the two methods were used to test 368 clinical stool specimens processed with either oxalic acid (164 specimens) or NALC-NaOH (204 specimens) (P = 0.07) or 67 specimens processed by both methods (P = 0.77). The oxalic acid and NALC-NaOH decontamination methods both appear to be useful for the recovery of MAC organisms from stool specimens.