Active tuberculosis and venous thromboembolism: association according to international classification of diseases, ninth revision hospital discharge diagnosis codes.

BACKGROUND: Infections are risk factors for venous thromboembolism (VTE), especially if severe and acute. The role of chronic infections such as active tuberculosis is ill defined, although several case reports and small series have suggested an association between tuberculosis and VTE. METHODS: Using data from the Premier Perspective database (27 659 947 admissions), we performed a multivariate analysis to assess the specific VTE risk associated with tuberculosis. The analysis was adjusted on classic risk factors for VTE. RESULTS: The prevalence of VTE among patients with active tuberculosis was 2.07% (95% confidence interval [CI], 1.62%-2.59%). In a multivariate analysis model, adults with active tuberculosis had a greater risk of VTE than those without (odds ratio, 1.55 [95% CI, 1.23-1.97], P < .001), close to the previously reported risk associated with neoplasia. No particular link was found between pulmonary tuberculosis and pulmonary embolism, or between extrapulmonary tuberculosis and deep vein thrombosis. This may suggest the preponderant role of a systemic hypercoagulable state over an intrathoracic venous compression mechanism. In-hospital mortality of patients with both active tuberculosis and VTE (11/72 [15%]) was higher than mortality of patients with only active tuberculosis (92/3413 [2.7%]) or only VTE (5062/199 480 [2.5%]) (P < .001). Pulmonary embolism was more frequent in black patients, suggesting that this population, which is also more likely to suffer from tuberculosis, should be followed carefully. CONCLUSIONS: Tuberculosis must be considered as a pertinent risk factor for VTE and should be included in thromboembolism risk evaluation similar to any acute and severe infection.

Outbreak of Human Metapneumovirus in a Nursing Home: A Clinical Perspective.

OBJECTIVES: To describe a human metapneumovirus (hMPV) outbreak occurring in a nursing home for older adults and to identify the risk factors associated with the clinical infection. DESIGN: A retrospective, case-controlled study. SETTING AND PARTICIPANTS: A French nursing home for older adults between December 27, 2014 and January 20, 2015. Probable cases were residents presenting at least 1 respiratory symptom or 1 constitutional symptom. Confirmed cases identified in the same way as probable cases but with a positive RT-PCR test for hMPV. Controls were residents with no symptoms of respiratory infection. MEASURES: Identification of hMPV was realized on nasal swab samples by RT-PCR. RESULTS: Seventy-eight older people were resident at the time of the outbreak. Three of the 4 tested were positive for hMPV by RT-PCR and negative for 13 other viruses or bacteria. All probable infected residents presented cough; other symptoms were scarcer. An inflammatory response was present, with median C-reactive protein at 50 mg/L. The median duration of the illness was 7 days. The rate of infection among residents was high (51%), with 5 hospitalizations (12.5%) and 1 death (2.5%). In multivariate analysis, vaccination against influenza virus appeared to emerge as associated with a probable hMPV infection, but this might be an artifact, as the proportion of unvaccinated residents was low (15%). A clear infected population profile was hard to define, although limited autonomy and low ADL score may play a role. Basic hygiene precautions were reinforced, but droplet precautions seemed difficult to apply rigorously to this population. CONCLUSIONS/IMPLICATIONS: Clinical and biological presentations were nonspecific. The rate of infection was high, highlighting the need for the rapid introduction of strict precautions to contain the infection.