Not IRB approved

Tuberculosis (TB) is one opportunistic infection often seen in HIV individuals. In 2013, there were an estimated 31,800 HIV-TB co-infection cases and 6,100 HIV-related deaths due to TB in the Americas. Due to the non-specific nature of its clinical symptoms, TB can be confused with various diseases such as histoplasmosis, sarcoidosis, lymphoma, and pneumonia. In Panama, where Histoplasma capsulatum is endemic, diagnosing TB versus histoplasmosis based on clinical symptoms can be difficult. In Panama, approximately 7.65% of HIV patients are co-infected with histoplasmosis, and there is a 30% mortality rate in HIV-histoplasmosis patients in Latin America. Due to similar clinical features, misdiagnosis of active TB and disseminated histoplasmosis in endemic regions may lead to incorrect antibiotic management, which in turn results in unnecessary toxicity, antibiotic resistance, and monetary expenditures. The investigators interests lie in increasing TB diagnostic accuracy using a simple urine dipstick test and evaluating physician response to new diagnostic testing, in order to reduce misdiagnosis and improve health outcomes in the HIV population.

The gold standard for TB diagnosis, mycobacterial culture, is limited by a slow turnaround time, the need for skilled technicians and biosafety level 3 facilities. Another diagnostic test, Xpert MTB/RIF, produces same-day results, boasts a specificity of 98%, but is limited in the HIV population by its sensitivity of 67% (with a single sputum sample). High costs associated with this test have presented a major obstacle to its routine use (US$65,500 for a 16-module instrument, and US$9.98 to US$18.00 per cartridge). The urine lipoarabinomannan (LAM) Ag test has a specificity of 98.6% and a sensitivity of 66.7% in patients with a CD4 count below 50. LAM is a 17.5kD glycolipid and virulence factor of mycobacteria. The test's sensitivity improves in advanced HIV cases because of the higher bacillary burden, more frequent disseminated TB, and greater concentration of urine antigen due to less antigen-antibody formation, and HIV nephropathy. The test strip contains its own positive and negative controls for quality assurance, and can utilize fresh urine samples stored at room temperature for up to 8 hours after collection. In the right clinical setting, the urine LAM Ag test could have an important additive benefit in TB diagnosis in HIV individuals when combined with the current standards of care.

This study is investigating how physician management and patient outcomes change when the urine LAM test is added to the current TB diagnostic tests (mycobacterial cultures and occasional use of Xpert MTB/RIF). The national referral hospital in Panama, Hospital Santo Tomás, is an ideal study site given its dedicated HIV inpatient service and high TB and histoplasmosis burden. Histoplasmosis diagnosis is often based on clinical symptoms alone, and less frequently confirmed by skin or bone marrow biopsy. Due to diagnostic uncertainty, 40% of HIV patients with fever and cough receive dual TB and histoplasmosis therapy. Their current standard of TB care includes sputum AFB microscopy and culture. However, rapid results are only available for the sputum AFB, which has a sensitivity of approximately 30%. The urine LAM Ag test is not presently used there. When combined with the urine LAM Ag test, sputum AFB has a sensitivity of 25% (if CD4>200) to 72% (if CD4<50). Over the next year, investigators plan to study changes in physician diagnostic classification and management decisions before and after the introduction of the urine LAM Ag test. This information would not only aid in TB diagnosis, but also in the investigators understanding of how physicians integrate new information.

The main objective is to determine the effect of urine LAM Ag test results in reducing dual TB/histoplasmosis therapy, by comparing the rate and duration of dual therapy in the urine LAM group to a retrospective control group. The investigators second objective is to investigate barriers to incorporating the urine LAM Ag test and how the results impact physicians' approaches to treatment using a physician questionnaire.

A clinical trial study will be conducted using HIV positive patients from June 2016 to June 2017 who present with pertinent clinical criteria at Hospital Santo Tomás. Following urine LAM Ag test administration, results will immediately be provided to the treating physicians. The control group will consist of retrospectively selected patients from the Hospital Santo Tomás database who fit the same inclusion and exclusion, matched for age, sex, and clinical severity. Patient outcome data will be collected from the day of hospital admittance to their time of discharge. Following the conclusion of data collection, consenting physicians will be interviewed on benefits and barriers to incorporating the urine LAM Ag test within their practice.

Inclusion Criteria:

• HIV positive adult, and

• Admitted to Hospital Santo Tomas, and

• history of fevers, and

• Two or more of the following symptoms:

  • cough
  • shortness of breath
  • night sweats
  • weight loss
  • fatigue
  • loss of appetite

Exclusion Criteria:

• Under 18 yrs of age, or
• Already on TB therapy, or
• Anuric