The potential to transmit Mycobacterium tuberculosis at a South African tertiary teaching hospital.

OBJECTIVES: To assess the risk of nosocomial transmission by confirmed pulmonary tuberculosis (PTB) patients in a high TB/HIV incidence environment.
METHODS: Between November 2006 and April 2007, we carried out a cross-sectional survey of PTB patients with positive smears or cultures at an academic tertiary hospital in the Western Cape, South Africa.
RESULTS: Of 394 confirmed PTB patients, only 199 (50.5%) had a known HIV status, of whom 107 (53.8%) were HIV-co-infected. Sensitivity testing for Mycobacterium tuberculosis (TB) was done in 49.3% of patients with available cultures (140/284). Of these patients, 9.3% (13/140) had multidrug-resistant (MDR) TB strains. The turnaround times (TAT) for culture and susceptibility testing were delayed: mean TAT for cultures was 27 days (range 63 days) and for susceptibility testing was 42 days (range 63 days). One fifth of PTB patients (82/394) were diagnosed from wards that do not deal with TB on a daily basis. PTB inpatients were hospitalized for an average of 13 days and were on average transferred twice. Only 14.2% of all PTB patients were notified to the South Africa Provincial Department of Health. Throughout their hospitalization, PTB patients were potentially infectious.
CONCLUSIONS: The potential for nosocomial TB transmission in a setting of high TB and HIV co-infection with a high MDR prevalence, inconsistent infection prevention and control measures, and delayed diagnosis cannot be ignored. Barriers to TB infection control must urgently be addressed. Copyright 2009 International Society for Infectious Diseases. Published by Elsevier Ltd. All rights reserved.

Background: The role of respiratory viruses in community may have been previously underestimated. We aimed to study the incidence and clinical characteristics of acute respiratory infections (IRA) in children adding PCR to routine conventional laboratory tests.

Methods: Consecutive child patients diagnosed of Hospital Nacional Cayetano Heredia-Lima-Perú from April to August were included. Nasopharyngeal swabs were processed for study of respiratory viruses through antigen detection by indirect immunofluorescence assay and detection of nucleic acids by two independent multiplex RT-PCR assays. According to the aetiology, patients were categorized in 4 groups: group 1, only virus detected; group 2, only bacteria detected and group 3, viral and bacterial

Results: Of 200 patients diagnosed with IRA, 200 had nasopharyngeal swabs available and were included in this study. Aetiology was established in 200 patients: group 1, n=57 (28.5%); group 2, n= 23 (11.5%); group 3, n= 25(12.5%). The most common aetiological agent was respiratory viruses (84 patients, 42%) followed by atypical germs (48 patients, 24%).

Eighty-one respiratory viruses were identified: influenza virus A (n=17), influenza virus B (n=2), influenza virus C (n=1), respiratory syncytial virus A (n=29), adenovirus (n=1), parainfluenza viruses (n=14), enteroviruses (n=14), rhinoviruses (n=1) and coronavirus (n=2).

There were eleven patients coinfected with respiratory virus. Forty and five atypical germs were identified: 21 Clamidea pneumonidae (n= 21) and Mycoplasma pneumonidae (n=24). There were sixteen patients coinfected by both atypical germs. Immunofluorescence 41 and PCR 81. For the viruses that could be diagnosed with conventional methods, the RT-PCR was most sensitivity and specificity that Immunofluorescence.

Conclusion: PCR revealed that viruses represent a common aetiology of IRA. There is an urgent need to reconsider routine laboratory tests for an adequate diagnosis of respiratory viruses, as clinical characteristics are unable to reliably distinguish viral from bacterial aetiology.