Treatment of respiratory infections in the patient at risk.

Therapeutic intervention of varying potency and toxicity is now available for bacterial, fungal, parasitic, and viral pneumonia, but the problem of establishing the precise origin to evaluate treatment protocols remains. Antibacterial agents reduce postinfluenzal morbidity and mortality. Earlier appropriate treatment of gram-positive coccal infections has reduced mortality to less than 10 percent unless bacteremia is present. Gram-negative bacillary pneumonia remains a major problem and response rates do not exceed 65 percent except in selected series in which two effective antimicrobial agents have been used. Nevertheless, the superiority of two agents has not convincingly been demonstrated in any retrospective series or prospective randomized trial; more efficacy data in patients with gram-negative pneumonia treated with the newer beta-lactam agents would be useful. Little information is available on the laboratory correlates of successful treatment of gram-negative pneumonia, such as sputum antibiotic levels. Moreover, aerosolization of antimicrobial agents has been used with reported success by some investigators, but this therapeutic approach remains controversial. Pneumonia or lung abscess due to anaerobes may require increasingly larger doses of penicillin or alternative antianaerobe agents. Parasitic pneumonias such as those due to Pneumocystis carinii have responded well to trimethoprim-sulfamethoxazole. Acyclovir appears effective against some herpes viruses (simplex and varicella zoster), but it has failed to affect cytomegalovirus pneumonia even when combined with interferon. Opportunistic fungal pneumonias are poorly treated with all available agents and usually do not respond unless there is amelioration of predisposing factors or improvement in underlying disease. The latter statement appears relevant in most patients with severely compromised host defenses and pulmonary infection.