• Time to Development of Tuberculosis (TB) Disease or Death Among HIV-infected Children [ Time Frame: Through to week 96 ] [ Designated as safety issue: No ]
  Criteria for diagnosis with TB disease: Definite-isolation of Mycobacterium TB (M.tb) or +ve stain on cerebrospinal fluid (CSF); Probable- +ve acid fast bacilli (AFB) stain on fluids/tissues other than CSF and sufficient clinical criteria/radiographic evidence suggestive of TB; Possible-abnormal chest radiograph suggestive of pulmonary TB (PTB) and either a +ve tuberculin skin test (TST) or minimum score on algorithm for clinical TB. Records reviewed by Endpoint Review Group. Results report percent of participants reaching TB disease/death by week 96 calculated using the Kaplan-Meier method.
  
  
• Time From Randomization to Development of TB Infection or Death Among Perinatally Exposed, HIV-uninfected Children [ Time Frame: Through to week 96 ] [ Designated as safety issue: No ]
  Criteria for diagnosis with TB infection were outlined in the protocol. TB infection included TB disease (see primary outcome measure 1 for definition) and latent TB infection. Latent TB infection was diagnosed by a positive tuberculin skin test (TST) based on a purified protein derivative (PPD) performed at week 96. Participant records were reviewed by an Endpoint Review Group to verify that participants had met the criteria for TB infection. Results report percent of participants reaching TB infection or death by week 96 calculated using the Kaplan-Meier method.
  
  

• Time From Randomization to Development of TB Infection or Death Among HIV-infected Children [ Time Frame: Through to week 96 ] [ Designated as safety issue: No ]
  Criteria for diagnosis with TB infection were outlined in the protocol. TB infection included TB disease (see primary outcome measure 1 for definition) and latent TB infection. Latent TB infection was diagnosed by a positive TST based on a PPD performed at week 96. Participant records were reviewed by an Endpoint Review Group to verify that participants had met the criteria for TB infection. Results report percent of participants reaching TB infection or death by week 96 calculated using the Kaplan-Meier method.
  
  
• Time From Randomization to HIV Disease Progression or Death Among HIV-infected Children [ Time Frame: Through to week 96 ] [ Designated as safety issue: No ]
  HIV disease progression was defined as any advancement in Centers for Disease Control (CDC) disease category from entry or death. If a participant was CDC disease category C at entry progression was defined as death. Results report percent of participants with HIV progression or death by week 96 calculated using the Kaplan-Meier method.
  
  
• Time From Randomization to Development of TB Disease or Death Among Perinatally Exposed, HIV-uninfected Children [ Time Frame: Through to week 96 ] [ Designated as safety issue: No ]
  Criteria for diagnosis with TB disease were: Definite-isolation of M.tb or positive stain on CSF; Probable-positive AFB stain on fluids/tissues other than CSF and sufficient clinical criteria/radiographic evidence suggestive of TB; Possible-abnormal chest radiograph suggestive of PTB and either a +ve TST or minimum score on algorithm to diagnose clinical TB. All records were reviewed by an Endpoint Review Group to verify that participants had met the criteria for TB disease. Results report percent of participants reaching TB disease/death by week 96 calculated using the Kaplan-Meier method.
  
  
• Time From Randomization to Development of TB Disease Among HIV Infected and Perinatally Exposed, HIV-uninfected Children [ Time Frame: Through to week 96 ] [ Designated as safety issue: No ]
  Criteria for diagnosis with TB disease were: Definite-isolation of M.tb or positive stain on CSF; Probable-positive AFB stain on fluids/tissues other than CSF and sufficient clinical criteria/radiographic evidence suggestive of TB; Possible-abnormal chest radiograph suggestive of PTB and either a +ve TST or minimum score on algorithm to diagnose clinical TB. All records were reviewed by an Endpoint Review Group to verify that participants had met the criteria for TB disease. Results report percent of participants reaching TB disease/death by week 96 calculated using the Kaplan-Meier method.
  
  
• Time From Randomization to Development of TB Infection Among HIV-infected and Perinatally Exposed, HIV-uninfected Children [ Time Frame: Through to week 96 ] [ Designated as safety issue: No ]
  Criteria for diagnosis with TB infection were outlined in the protocol. TB infection included TB disease (see primary outcome measure 1 for definition) and latent TB infection. Latent TB infection was diagnosed by a positive TST based on a PPD performed at week 96. Participant records were reviewed by an Endpoint Review Group to verify that participants had met the criteria for TB infection. Results report percent of participants reaching TB infection by week 96 calculated using the Kaplan-Meier method.
  
  
• Time From Randomization to Death Among HIV-infected and Perinatally Exposed, HIV-uninfected Children [ Time Frame: Through to week 96 ] [ Designated as safety issue: Yes ]
  Deaths from any cause were included. Results report percent of participants dying by week 96 calculated using the Kaplan-Meier method.
  
  
• Time From Randomization to First New Grade 3 or Worse Adverse Event Among HIV-infected and Perinatally Exposed, HIV-uninfected Children [ Time Frame: Through to week 96 ] [ Designated as safety issue: Yes ]
  Signs, symptoms and laboratory values were graded according to the Division of AIDS Adverse Event Grading System. Any event of grade 3 or higher not present at entry that occurred after randomization was classified as a new event. Results report percent of participants with a new event by week 96 calculated using the Kaplan-Meier method.
  
  

Tuberculosis (TB) and the Human Immunodeficiency Virus (HIV) are major public health problems in southern Africa, and the incidence of TB in South Africa is among the highest in the world. TB is caused by the highly contagious bacterium Mycobacterium tuberculosis. The use of Isoniazid (INH) prophylaxis in adults has been associated with reduced risk of TB disease in high-risk populations. Delay in initiating INH prophylaxis in children has resulted in more cases of childhood TB infection. This study evaluated the effectiveness of INH prophylaxis in preventing TB infection in infants born to HIV-infected mothers in southern Africa.

Infants were randomly assigned to receive either INH or placebo by mouth daily, beginning between the 91st and 120th day of life, and at least 90 days after Bacille Calmette-Guerin (BCG) vaccination. At sites in South Africa, HIV-infected infants received daily trimethoprim/sulfamethoxazole (TMP/SMX) as Pneumocystis jiroveci pneumonia (PCP) prophylaxis until at least 1 year of age; HIV-uninfected infants received TMP/SMX until at least 6 months of age.

The study was to follow participants for 192 weeks. Study visits occurred at study entry and every 12 weeks until week 192. A physical exam and blood collection occurred at each study visit. Infants were assessed for peripheral neuropathy every 12 weeks until week 96 and for TB at weeks 96, 144, and 192. The study also assessed medication adherence.

As of November 12, 2008, follow-up was revised. All participants were permanently discontinued from study follow-up by February 28, 2009 and no later than May 31, 2009. Only clinical evaluations were performed for all participants. For HIV-infected participants, the study drug was stopped at the next scheduled visit. For HIV-uninfected subjects, the study drug was discontinued immediately.