Quantifying Race and HIV Status among Intestinal Tuberculosis Patients Hospitalized in the United States: 2005-2009

Abstract

Purpose: TB remains a major cause of morbidity and mortality across the developing world. In the United States, overall TB incidence and prevalence are decreasing with an estimated incidence of 3.6 per 100,000 in 2011 [CDC]. In the US approximately 9% of TB patients tested are HIV positive. This study quantifies intestinal TB among hospitalized patients with and without HIV infection between 2005 and 2009. Methods: Data from the Nationwide Inpatient Sample (NIS) from 2005-2009 were examined. All admissions associated with a diagnosis of intestinal TB (ICD-9-CM: 014.80-014.86) were included and individuals with comorbid HIV infection (ICD-9-CM: 042) were identified. Demographic factors, surgical procedures and in-hospital mortality were examined. Standard descriptive statistical analyses were conducted. Patient race is missing for approximately 25% of NIS admissions; therefore patient race was examined in the subset of admissions where available. Results: After weighting to approximate the national population, 197,476,074 hospitalizations were examined of which a total of 1,470 were associated with a diagnosis of intestinal TB. Of these, 408 (27.8%) were associated with HIV infection. HIV positive patients were younger (39.0 vs. 46.9 years, p<0.001) and more likely to be male (79.9% vs. 48.0%, p<0.001) compared with non-infected patients. Race data were available for 1,213 patients, 21% of whom were white, 22% black, 30% Hispanic, 16% Asian, and 11% other. HIV infection was present in 28.7% of white, 51.4% of black, 37.6% of Hispanic, 5.0% of Asian and 31.2% of other patients with intestinal TB. Hospital admission-based rates (per 100,000 admissions) of intestinal TB were: 0.24 for white, 1.30 for black, 1.86 for Hispanic, 2.06 for other and 5.09 for Asian patients. Surgical resection was not common (7.8%), however HIV negative patients were significantly more likely to undergo resection than HIV positive patients (9.8% vs. 2.5% respectively, p=0.038). In-hospital mortality was twice as common among intestinal TB patients compared with all other admissions (4.2% vs. 2.0%, p=0.007) but not associated with either HIV status or patient race. Conclusion: Intestinal TB is rare but not inconsequential among inpatients in the United States and is not commonly treated with surgical resection. HIV infection was reported in less than 1/3 of inpatient admissions with intestinal TB. On a per-admission basis, intestinal TB was 21 times more common in Asian patients than white patients despite HIV infection being 5 times more common among white patients. Intestinal tuberculosis mimics other gastrointestinal disorders and should be considered in patients with non-specific abdominal symptoms, especially prior to implementing immunosuppressive therapy.