Etiology and epidemiology of persistent diarrhea in northeastern Brazil: a hospital-based, prospective, case-control study.

With the improved therapy for acute diarrhea, persistent diarrhea (> 14 days) is emerging as a major problem in developing countries. However, the etiologies and pathogenesis of persistent diarrhea remain poorly understood. We conducted a prospective case-control study in children < 3 years old presenting to the hospital with persistent diarrhea in Fortaleza, Brazil. Over the study period (August 1988 to March 1991), 56 children seen with persistent diarrhea, 52 children seen with acute diarrhea, and 42 controls attending the same hospital/clinic for illnesses other than diarrhea were enrolled. A potential pathogen was found in 91% of children with persistent diarrhea and 90% of those with acute diarrhea versus 45% of controls (both p's < 0.01). Thirty-four percent of persistent (19/56) and 38% of acute (20/52) diarrhea cases versus 2% (1/42) of controls (both p's < 0.01) had multiple pathogens. Enteroaggregative Escherichia coli (EAggEC) were found in 68% (38/56) of children with persistent diarrhea versus 31% (13/42) of controls (p < 0.01) and in 46% (24/52) of those with acute diarrhea. Furthermore, when the EAggEC were subdivided into aggregative adherence (AA) gene probe positive (18/56; 32%) and negative (20/56; 36%), both subgroups were still significantly different from controls [6/42 (14%) and 7/42 (17%), respectively; both p's < 0.05].(ABSTRACT TRUNCATED AT 250 WORDS)

Community-acquired Legionnaires' disease.

Legionellae are ubiquitous aquatic organisms. They are unique among the agents commonly responsible for bacterial pneumonia in humans in that they are not part of the normal human flora but are acquired from environmental sources. Prospective studies have shown that legionellae consistently rank among the top three bacteria as etiologic agents of community-acquired pneumonia. The clinical presentation of Legionnaires' disease is not distinguishable from that of other bacterial pneumonias. Culture of respiratory secretions using selective media, combined with one or more rapid diagnostic methods (direct fluorescent antibody staining, radiolabelled DNA probe, or urinary antigen detection) provides a specific diagnosis in the vast majority of cases. Sporadic cases have been linked to legionella colonization of water systems in homes and the work setting. Antibiotics commonly used in the therapy of community-acquired pneumonias, such as beta-lactam agents, are ineffective. Specific therapy with erythromycin reduces mortality to less than 10%.

Infections caused by the Pittsburgh pneumonia agent.

Of the Legionellaceae family, Pittsburgh pneumonia agent (Tatlockia micdadei, Legionella micdadei) is second only to Legionella pneumophila in causing human pneumonia. In nosocomial infection, the patients tend to be immunosuppressed. The clinical presentation is nonspecific, although in immunosuppressed hosts the presentation may mimic that of pulmonary embolus (pleuritic chest pain, nonproductive cough, pleural-based densities on chest rontgenogram). The reservoir for the organism is water, and prevention of nosocomial infections can be accomplished by disinfection of the water supply. Diagnosis is best established by isolation of the organism from respiratory secretions by using selective, dye-containing buffered charcoal-yeast extract agar. The organisms can be acid-fast when clinical specimens are stained. Erythromycin is the antibiotic of choice, although tetracyclines, trimethoprim-sulfamethoxazole, and rifampin have also proved to be efficacious.

Legionella species as hospital-acquired respiratory pathogens.

Nosocomial pneumonia caused by legionella is an increasingly recognized entity. Legionella sp responsible for documented nosocomial disease include Legionella pneumophila, Tatlockia micdadei, Legionella bozemanii, Legionella dumoffi and Legionella oakridgensis. The clinical presentation is nonspecific although diarrhea occurs frequently. Hyponatremia occurs significantly more frequently in legionnaires' disease than pneumonias caused by other agents. Chest roentgenographic findings are nonspecific, although cavitation can be seen in immunosuppressed patients. Laboratory methods require the use of direct fluorescent antibody (DFA) stains, culture using selective media, serologic testing, and detection of antigen in urine. The DFA test is not sensitive; however, it does correlate with the severity of disease. Culture from sputa is now feasible. Bronchoalveolar lavage is a promising technique for obtaining specimens. The ideal specimen for culture is that obtained by transtracheal aspiration, which bypasses oropharyngeal contamination. Combination therapy of erythromycin and rifampin is recommended for selected patients. Because the source of the organism is the hospital water distribution system, we recommend routine environmental surveillance, especially in hospitals in which organ transplants are performed. The role of cooling towers as a vector for dissemination of the organism is disputed. Disinfection of the water supply can be accomplished by using heat eradication. Chlorination has generally proven unsatisfactory because of organism persistence as well as corrosive damage to the plumbing system from the chlorine. Both physician awareness and availability of specialized laboratory testing are necessary for the detection of cases.