Abstract
BACKGROUND: Pulmonary tuberculosis infection can manifest in different states, including subclinical tuberculosis. It is commonly defined as confirmed tuberculosis without the classic symptoms (commonly, persistent cough for ≥2 weeks). This narrow definition likely poses limitations for surveillance and control measures. The aims of the current study were to characterize the clinical presentation of tuberculosis; estimate the prevalence of subclinical tuberculosis among individuals with bacteriologically confirmed tuberculosis, using various definitions; and investigate risk factors for subclinical as opposed to clinical tuberculosis in a population-based survey. METHODS: We conducted a secondary analysis of data from a nationally representative tuberculosis prevalence survey from Zambia in 2013-2014, in which participants were screened for tuberculosis based on chest radiographic findings and symptoms. Tuberculosis was defined as culture-positive or GeneXpert MTB/RIF test-positive sputum. Risk factors for subclinical tuberculosis were assessed by means of multivariable logistic regression. RESULTS: Of 257 participants with confirmed tuberculosis, 104 (40.5%) were without cough persisting ≥2 weeks. Only 23 (22.1%) of these did not present with any other common symptoms. Those without cough persisting ≥2 weeks frequently reported other symptoms, particularly chest pain (46.2%) and weight loss (38.5%); 36 (34.6%) reported experiencing other symptoms persisting ≥4 weeks. Female subjects were more likely to report no cough persisting ≥2 weeks, as were relatively wealthier individuals. CONCLUSIONS: The commonly used definition of subclinical tuberculosis includes a large proportion of individuals who have other tuberculosis-suggestive symptoms. Requiring cough ≥2 weeks for tuberculosis diagnosis likely misses many active tuberculosis infections and allows a large reservoir of likely transmissible tuberculosis to remain undetected.