Clinical Drug Experience Knowledgebase

Randomized Clinical Trial to Assess the Efficacy of Short Course Intermittent Regimens for the Treatment of HIV-Associated Tuberculosis

All HIV positive patients seeking care at one of the study centers, above the age of 15 years, not suffering from a serious illness, non-pregnant and diagnosed with TB will be briefly explained the treatment trial.Patients willing to participate in the trial will be asked to provide written consent. Patients who refuse participation in the study will be managed according to the RNTCP guidelines.

All patients who provide written consent to participate in the treatment trial will be interviewed by a social worker using a standard questionnaire. Patients on anti-retroviral drugs will be excluded from the study. Blood samples will be collected from patients meeting initial eligibility criteria to test for laboratory eligibility criteria. Laboratory investigations will include complete heamogram,Renal function tests, Liver function tests, random blood sugar, urine albumin and urine sugar. Patients who fulfill laboratory criteria (hemoglobin >70 g/L, granulocyte count >1.1 X 109/L, platelet count > 100X 109/L, serum alanine amino transferase concentration <2.5 times the upper limit of normal, serum creatinine concentration <1.1mg%, random blood sugar < 140 mg/dl) will be enrolled in to the study.

Randomization and Dosing:

All patients enrolled into the treatment trial will receive supervised directly observed treatment during the intensive phase. At treatment initiation, each patient will be counseled about the importance of treatment regularity. While patients are undergoing intensive phase treatment, they will be randomized either to the standard regimen or to the extended regimen as soon as CD4 results are available. Randomization will be done according to a permuted block scheme in blocks of four, stratified by CD4 counts (> 200 and ≤200) and by smear grade (0, 1+, 2+, or 3+). The treatment assignment list will be generated before the start of trial and sequentially numbered sealed envelopes containing the treatment assigned will be prepared independently.

The treatment regimens in each arm of the trial will be as follows:

Category I RNTCP regimen 2EHRZ3/4RH3 Trial regimen 2EHRZ3/7RH3 Category II RNTCP regimen 2SEHRZ3/1EHRZ3/5EHR3 Category III RNTCP regimen 2HRZ3/4RH3. All anti-TB drugs will be administered as per the RNTCP strategy of DOTS. Patients residing in Chennai and Madurai will attend the respective centers/ sub-centers three times a week for the intensive phase of treatment (first two/three months) and then once a week during the continuation phase (four to seven months). Dosages will be as per RNTCP manual and may be modified if the patient is extremely debilitated (weight < 30 kg). Patients in all treatment arms will receive 10 mg of Pyridoxine on treatment days and Co-trimoxazole DS 1 tablet daily.

During chemotherapy, patients will be called to the clinic for monthly clinical evaluation. During follow-up visits at the clinic (monthly up to 24 months, every 3 months after that), patients will be thoroughly evaluated for likely drug toxicity and the information on adverse effects will be recorded on a standardized toxicity chart.

Compliance with therapy will be measured by checking treatment cards, DOTS provider notebooks and empty drug packets. Additionally, spot urine examination will be performed to check for acetyl isoniazid and rifampicin levels at each monthly.

Statistical analysis:

The intention-to-treat approach will be used for analyzing the data for primary and secondary end points. Annual interim analyses will be done to ensure timely identification of any significant risks or benefits to patients.

Comparisons of categorical variables will be done by chi-square test and Fisher's exact test. Continuous variables will be compared by t test or by Wilcoxon's rank sum test. Survival estimates will be made by the Kaplan-Meier method. Comparison of Kaplan-Meier survival curves will be made with the log-rank test. Multivariate analyses will be performed using Cox's proportional Hazards model.