Rapidly growing nontuberculous mycobacteria: a new enemy of the cardiac surgeon.

A review of atypical mycobacterial infections complicating cardiac operations is presented. Proven sources of infections at different institutions include contaminated porcine valves and municipal water supply, but the mode of transmission in the great majority of patients remains unclear. There are two principal clinical forms of atypical mycobacterial infections after cardiac operations--endocarditis and sternal osteomyelitis. The latter has characteristics resembling tuberculotic "cold abscess." Specialized laboratory testing is necessary to confirm the diagnosis, and surgeons may have to take the initiative to request special microbiological investigation in cases where infection is clinically suspected but routine cultures are reported as "negative." The prognosis for patients who have any atypical mycobacterial infection after a heart operation is severe. Those infected with the strain chelonei and those whose cardiac chambers were entered during operation fare worse. This dim clinical prognosis may be improved by appropriate and aggressive antibiotic and surgical therapy. Awareness of the urgency of special bacteriological studies is the key to successful management.

Delayed hypersensitivity reactions in patients with Mycobacterium chelonei and Mycobacterium fortuitum infections.

Delayed hypersensitivity reactions to skin test antigens prepared from rapidly growing mycobacteria were measured in patients with postoperative wound infections due to Mycobacterium chelonei and M. fortuitum. Sixteen of 19 patients with M. chelonei infection had more than 10 mm of induration to the M. chelonei purified protein derivative antigen (PPD-CG) and were significantly more likely to react to PPD-CG than patients or hospital personnel who had no evidence of infection. All but one patient had more induration with PPD-CG than with the antigen prepared from M. fortuitum (PPD-F). Three of 4 patients with M. fortuitum infection had greater than or equal to 9mm of induration with PPD-F and were significantly more likely than noninfected patients to react to PPD-F. Three of the 4 patients also had at least as much induration with PPD-F as with PPD-CG. In an outbreak of infections due to M. chelonei, testing with PPD-CG may be useful epidemiologically. Although the data are less clearly defined for PPD-F, testing with this antigen may be of value in an outbreak caused by M. fortuitum.