Methods: This retrospective study was conducted at two tertiary care centers within the Texas Medical Center: Ben Taub Hospital (BTH) and CHI Baylor St. Luke’s Medical Center (BSLMC). Patients with pathology samples sent to the University of Washington for MAC PCR between 2012 and 2015 were identified. Medical records were reviewed and correlated to clinical, pathology and microbiology data.
- BTH – 1/4 cases had positive PCR for MAC from a paraffin-embedded retroperitoneal lymph node with acid-fast bacilli (AFB) seen on pathology; the PCR result enabled clinicians to narrow the treatment regimen before the culture ultimately turned positive for MAC. In 2/3 cases with negative PCR, culture was ultimately positive for MAC.
- BSLMC – In 2/20 cases, PCR was positive for MAC and enabled the treating physicians to avoid anti-tuberculous therapy. Both positive cases were from lung specimens that were paraffin-embedded and had AFB visualized on pathology. Culture was sent on only one of these samples and was positive. Of the remaining 18 samples negative by PCR, 13 had a negative AFB culture and 2 were positive for another mycobacterial species; in 3 cases culture was not sent.
Conclusion: Overall, the sensitivity of MAC PCR from a paraffin-embedded pathology sample is low even with visualized AFB; however a positive result is still helpful to guide treatment, particularly when culture data is either delayed or unavailable.
C. Stager, None
T. Lasco, None
S. R. Rose, None