Clinical Infectious Diseases Advance Access published August 25, 2014
1
ACQUIRED RIFAMPICIN RESISTANCE IN THRICE-WEEKLY TB THERAPY: IMPACT OF HIV AND ANTIRETROVIRAL THERAPY
cr ipt
us
Centre), Chetpet, Chennai, India
Govt. Hospital of Thoracic Medicine, Tambaram Sanatorium, Chennai, India
3
Govt. Stanley Hospital, Tondiarpet, Chennai, India
an
2
CORRESPONDENCE AND REQUESTS FOR REPRINTS SHOULD BE ADDRESSED TO: Senior and corresponding Author, Dr. Soumya Swaminathan Scientist ‘G’ &
M
Director, National Institute for Research in Tuberculosis, No.1, Mayor Sathiyamoorthy Road, Chetput, Chennai, India 600031. E-mail:
pt ed
[email protected], Telephone Number: +914428369600, Fax no: +914428362528
Alternate corresponding author: Dr. Narendran Gopalan, Email: [email protected], Telephone number: +914428369545
Summary
ce
Risk factors for Acquired Rifampicin Resistance (ARR) among TB patients on thriceweekly anti-TB therapy were baseline isoniazid resistance and HIV infection. Among
Ac
HIV+ patients, higher mycobacterial burden and lower CD4 cell count, but not HAART, were significantly associated with ARR.
© The Author 2014. Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please e-mail: [email protected].
Downloaded from http://cid.oxfordjournals.org/ at Univ. of Massachusetts/Amherst Library on September 27, 2014
Gopalan Narendran1, Pradeep Aravindan Menon1, Perumal Venkatesan1, Krishnamoorthy Vijay1, Chandrasekaran Padmapriyadarsini1, Santhanakrishnan RameshKumar1, Kannabiran Perumal Bhavani1, Lakshmanan Sekar1, Sivaramakrishnan Narayan Gomathi1, Chockalingam Chandrasekhar2, Satagopan Kumar2, Rathinam Sridhar3, and Soumya Swaminathan1 1 National Institute for Research in Tuberculosis (formerly Tuberculosis Research
2
Abstract Introduction: Risk factors for Acquired Rifampicin Resistance (ARR) in HIV-TB coinfection, in the Highly Active Antiretroviral Therapy (HAART) era, needs evaluation.
us
Methods: This cross protocol analysis included rifampicin susceptible, newly diagnosed pulmonary TB patients with and without HIV, enrolled in clinical trials (who took > 80% of medication), at the National Institute for Research in Tuberculosis between 1999 and 2013. All patients received rifampicin and isoniazid for six months reinforced with pyrazinamide and ethambutol in the first two months, given thrice-weekly throughout along with HAART in one group. Outcomes were categorized and multivariate logistic regression analysis performed to identify risk factors for ARR.
M
an
Results: The per-protocol results included TB patients; 246 HIV-negative (HIV-TB+), 212 HIV patients without HAART (non-HAART) and 116 HIV patients on HAART. Median CD4 counts of the latter two groups were 150 and 93 (cells/mm3) while the median viral loads were 147,000 and 266,000 copies/ml respectively. Compared to HIV-TB+, the relative risk (RR) for an unfavourable response in the co-infected, nonHAART and HAART groups were 5.1 (95%CI 1.7-14.8), p 18 years in the two HIV positive cohorts were enrolled. All study subjects had to satisfy the clinical, sociological and
Ac
laboratory eligibility criteria for enrolment. DNA fingerprinting was performed on paired isolates from stored cultures (at baseline and at the time of failure with bacteriological confirmation). Three molecular methods namely IS6110 analysis, mycobacterial interspersed repetitive unit-variable number tandem repeat (MIRUVNTR) typing, and spacer oligonucleotide typing (spoligotyping) were used for
Downloaded from http://cid.oxfordjournals.org/ at Univ. of Massachusetts/Amherst Library on September 27, 2014
This included HIV testing, Hepatitis B and C serology, complete blood count
6 identification of the mycobacterial strain on these paired isolates. The trials were individually approved by the Institutional Ethics Committee. Written informed
cr ipt
consent was obtained from all study subjects prior to enrolment.
Treatment details
All TB patients were treated with a thrice-weekly intermittent regimen consisting of
us
> 60 kg) for 6 months, reinforced with ethambutol (E) 1200 mg and pyrazinamide (Z) 1500 mg in the first two months of therapy (2EHRZ3/4HR3). All TB drugs were
an
assayed for content as part of quality assurance. Treatment was fully supervised in all except the NCT0376012 trial, where the continuation phase had only one out of three
M
doses supervised. Patients selected from the NCT0033206 study were initiated on HAART at two months of ATT. The HAART regimen consisted of didanosine (250/400 mg for body weight < or > 60kg), lamivudine (300 mg) along with efavirenz
pt ed
(600 mg) given once-daily [13]. Those patients included from the NCT0933790 trial [14] had either tenofovir (300 mg OD) or zidovudine (300mg BD), along with lamivudine (150 mg BD) and efavirenz (600 mg OD), initiated within the first 2-8 weeks (based on prevailing ART guidelines) [17, 18]. All HIV-infected patients were
ce
given co-trimoxazole DS 1 tablet daily if their CD4 count was < 350 cells/mm3, in addition to pyridoxine 10 mg daily.
Ac
Outcome definition
A cure or favourable response to treatment was defined as all available cultures (6 in number) in the last two months of treatment being negative. Unfavourable responses included failures and deaths due to TB. Failures were further classified as “bacteriological” when patients had at least 2 positive cultures in the last 2 months of
Downloaded from http://cid.oxfordjournals.org/ at Univ. of Massachusetts/Amherst Library on September 27, 2014
isoniazid (H) 600 mg and rifampicin (R) 450 mg or 600 mg (based on weight < 60 or
7 treatment or persistent culture positivity after 4 months of treatment, and “clinical” when clinical and /or radiolographic deterioration mandated change or extension of
cr ipt
ATT, without accompanying positivity in sputum culture.
Analysis
Per-protocol results from the three groups were analysed using SPSS Version 20. Chi
us
relative risks for ARR and unfavourable responses were calculated taking HIVnegative patients with TB as the standard, and comparing it with the two HIV positive
an
groups with and without HAART. A multiple logistic regression model was used to identify factors contributing to ARR, after adjusting for covariates. Baseline variables
M
that were considered important (age, sex, weight, CD4, HAART, culture grade and INH resistance) were used to fit the regression model, and odds ratio with 95% confidence intervals (95%CI) computed. Further, the backward elimination method
pt ed
was used to identify the significant covariates associated with ARR. Regression analysis was done for the whole group as well as within the HIV positive group separately.
ce
RESULTS
This retrospective cross protocol analysis included data from 246 patients with HIV-
Ac
negative TB (HIV-TB+), 212 HIV-infected patients without HAART (HIV+TB+, non-HAART) and 116 HIV-infected patients on HAART (HIV+TB+HAART). Baseline/pre-treatment characteristics of the three groups, including their drug susceptibility patterns, are shown in Table 1. The bacteriologic failures among HIVTB+, HIV+TB+, non-HAART and HIV+TB+HAART were 4% (9/246), 9% (19/212)
Downloaded from http://cid.oxfordjournals.org/ at Univ. of Massachusetts/Amherst Library on September 27, 2014
square /Fischer’s exact test was used for testing differences between proportions. The
8 and 5% (6/116) respectively. Details of TB treatment outcomes are provided in Table 2.
cr ipt
In addition, four patients (three with lesions in the brain and one with clinical and radiological deterioration in the chest X-ray) in the HIV+TB+, non-HAART group as
well as one patient (who developed an open pneumothorax) in the HIV+TB+HAART group, were classified as clinical failures, as their clinical condition necessitated
us
group required ATT extension beyond 6 months.
Compared to the HIV-negative group, the relative risk of an unfavourable response
an
was 5.1 (95% CI 1.7-14.8), p=0.00009 in the HIV-positive group without HAART and 2.12 (95% CI 0.86-5.2), p=0.13 with HAART. (Table2). Within the HIV group,
M
the relative risk for unfavorable response was 2.4 times more (95% CI 1.2-4.7) p=0.0045 in the absence of HAART. The month-wise sputum culture negativity in the 3 groups is depicted in Figure 1.
pt ed
Comparison of the drug susceptibility pattern among bacteriologically confirmed cases, at pretreatment and at failure, in each of the groups is provided in Table 3. All 19 patients who failed therapy in the HIV+TB+, non-HAART group had emergence of ARR. Fifteen out of the 19 patients developed MDR-TB (resistance to both H and
ce
R) with four developing R mono-resistance. Three out of the six bacteriological failures in the HIV+TB+HAART group and one out of the nine failures in the HIV-
Ac
TB+ group, had emergence of ARR. The relative risk of ARR was 21.1 (95%CI 2.6184), p
1
ACQUIRED RIFAMPICIN RESISTANCE IN THRICE-WEEKLY TB THERAPY: IMPACT OF HIV AND ANTIRETROVIRAL THERAPY
cr ipt
us
Centre), Chetpet, Chennai, India
Govt. Hospital of Thoracic Medicine, Tambaram Sanatorium, Chennai, India
3
Govt. Stanley Hospital, Tondiarpet, Chennai, India
an
2
CORRESPONDENCE AND REQUESTS FOR REPRINTS SHOULD BE ADDRESSED TO: Senior and corresponding Author, Dr. Soumya Swaminathan Scientist ‘G’ &
M
Director, National Institute for Research in Tuberculosis, No.1, Mayor Sathiyamoorthy Road, Chetput, Chennai, India 600031. E-mail:
pt ed
[email protected], Telephone Number: +914428369600, Fax no: +914428362528
Alternate corresponding author: Dr. Narendran Gopalan, Email: [email protected], Telephone number: +914428369545
Summary
ce
Risk factors for Acquired Rifampicin Resistance (ARR) among TB patients on thriceweekly anti-TB therapy were baseline isoniazid resistance and HIV infection. Among
Ac
HIV+ patients, higher mycobacterial burden and lower CD4 cell count, but not HAART, were significantly associated with ARR.
© The Author 2014. Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please e-mail: [email protected].
Downloaded from http://cid.oxfordjournals.org/ at Univ. of Massachusetts/Amherst Library on September 27, 2014
Gopalan Narendran1, Pradeep Aravindan Menon1, Perumal Venkatesan1, Krishnamoorthy Vijay1, Chandrasekaran Padmapriyadarsini1, Santhanakrishnan RameshKumar1, Kannabiran Perumal Bhavani1, Lakshmanan Sekar1, Sivaramakrishnan Narayan Gomathi1, Chockalingam Chandrasekhar2, Satagopan Kumar2, Rathinam Sridhar3, and Soumya Swaminathan1 1 National Institute for Research in Tuberculosis (formerly Tuberculosis Research
2
Abstract Introduction: Risk factors for Acquired Rifampicin Resistance (ARR) in HIV-TB coinfection, in the Highly Active Antiretroviral Therapy (HAART) era, needs evaluation.
us
Methods: This cross protocol analysis included rifampicin susceptible, newly diagnosed pulmonary TB patients with and without HIV, enrolled in clinical trials (who took > 80% of medication), at the National Institute for Research in Tuberculosis between 1999 and 2013. All patients received rifampicin and isoniazid for six months reinforced with pyrazinamide and ethambutol in the first two months, given thrice-weekly throughout along with HAART in one group. Outcomes were categorized and multivariate logistic regression analysis performed to identify risk factors for ARR.
M
an
Results: The per-protocol results included TB patients; 246 HIV-negative (HIV-TB+), 212 HIV patients without HAART (non-HAART) and 116 HIV patients on HAART. Median CD4 counts of the latter two groups were 150 and 93 (cells/mm3) while the median viral loads were 147,000 and 266,000 copies/ml respectively. Compared to HIV-TB+, the relative risk (RR) for an unfavourable response in the co-infected, nonHAART and HAART groups were 5.1 (95%CI 1.7-14.8), p 18 years in the two HIV positive cohorts were enrolled. All study subjects had to satisfy the clinical, sociological and
Ac
laboratory eligibility criteria for enrolment. DNA fingerprinting was performed on paired isolates from stored cultures (at baseline and at the time of failure with bacteriological confirmation). Three molecular methods namely IS6110 analysis, mycobacterial interspersed repetitive unit-variable number tandem repeat (MIRUVNTR) typing, and spacer oligonucleotide typing (spoligotyping) were used for
Downloaded from http://cid.oxfordjournals.org/ at Univ. of Massachusetts/Amherst Library on September 27, 2014
This included HIV testing, Hepatitis B and C serology, complete blood count
6 identification of the mycobacterial strain on these paired isolates. The trials were individually approved by the Institutional Ethics Committee. Written informed
cr ipt
consent was obtained from all study subjects prior to enrolment.
Treatment details
All TB patients were treated with a thrice-weekly intermittent regimen consisting of
us
> 60 kg) for 6 months, reinforced with ethambutol (E) 1200 mg and pyrazinamide (Z) 1500 mg in the first two months of therapy (2EHRZ3/4HR3). All TB drugs were
an
assayed for content as part of quality assurance. Treatment was fully supervised in all except the NCT0376012 trial, where the continuation phase had only one out of three
M
doses supervised. Patients selected from the NCT0033206 study were initiated on HAART at two months of ATT. The HAART regimen consisted of didanosine (250/400 mg for body weight < or > 60kg), lamivudine (300 mg) along with efavirenz
pt ed
(600 mg) given once-daily [13]. Those patients included from the NCT0933790 trial [14] had either tenofovir (300 mg OD) or zidovudine (300mg BD), along with lamivudine (150 mg BD) and efavirenz (600 mg OD), initiated within the first 2-8 weeks (based on prevailing ART guidelines) [17, 18]. All HIV-infected patients were
ce
given co-trimoxazole DS 1 tablet daily if their CD4 count was < 350 cells/mm3, in addition to pyridoxine 10 mg daily.
Ac
Outcome definition
A cure or favourable response to treatment was defined as all available cultures (6 in number) in the last two months of treatment being negative. Unfavourable responses included failures and deaths due to TB. Failures were further classified as “bacteriological” when patients had at least 2 positive cultures in the last 2 months of
Downloaded from http://cid.oxfordjournals.org/ at Univ. of Massachusetts/Amherst Library on September 27, 2014
isoniazid (H) 600 mg and rifampicin (R) 450 mg or 600 mg (based on weight < 60 or
7 treatment or persistent culture positivity after 4 months of treatment, and “clinical” when clinical and /or radiolographic deterioration mandated change or extension of
cr ipt
ATT, without accompanying positivity in sputum culture.
Analysis
Per-protocol results from the three groups were analysed using SPSS Version 20. Chi
us
relative risks for ARR and unfavourable responses were calculated taking HIVnegative patients with TB as the standard, and comparing it with the two HIV positive
an
groups with and without HAART. A multiple logistic regression model was used to identify factors contributing to ARR, after adjusting for covariates. Baseline variables
M
that were considered important (age, sex, weight, CD4, HAART, culture grade and INH resistance) were used to fit the regression model, and odds ratio with 95% confidence intervals (95%CI) computed. Further, the backward elimination method
pt ed
was used to identify the significant covariates associated with ARR. Regression analysis was done for the whole group as well as within the HIV positive group separately.
ce
RESULTS
This retrospective cross protocol analysis included data from 246 patients with HIV-
Ac
negative TB (HIV-TB+), 212 HIV-infected patients without HAART (HIV+TB+, non-HAART) and 116 HIV-infected patients on HAART (HIV+TB+HAART). Baseline/pre-treatment characteristics of the three groups, including their drug susceptibility patterns, are shown in Table 1. The bacteriologic failures among HIVTB+, HIV+TB+, non-HAART and HIV+TB+HAART were 4% (9/246), 9% (19/212)
Downloaded from http://cid.oxfordjournals.org/ at Univ. of Massachusetts/Amherst Library on September 27, 2014
square /Fischer’s exact test was used for testing differences between proportions. The
8 and 5% (6/116) respectively. Details of TB treatment outcomes are provided in Table 2.
cr ipt
In addition, four patients (three with lesions in the brain and one with clinical and radiological deterioration in the chest X-ray) in the HIV+TB+, non-HAART group as
well as one patient (who developed an open pneumothorax) in the HIV+TB+HAART group, were classified as clinical failures, as their clinical condition necessitated
us
group required ATT extension beyond 6 months.
Compared to the HIV-negative group, the relative risk of an unfavourable response
an
was 5.1 (95% CI 1.7-14.8), p=0.00009 in the HIV-positive group without HAART and 2.12 (95% CI 0.86-5.2), p=0.13 with HAART. (Table2). Within the HIV group,
M
the relative risk for unfavorable response was 2.4 times more (95% CI 1.2-4.7) p=0.0045 in the absence of HAART. The month-wise sputum culture negativity in the 3 groups is depicted in Figure 1.
pt ed
Comparison of the drug susceptibility pattern among bacteriologically confirmed cases, at pretreatment and at failure, in each of the groups is provided in Table 3. All 19 patients who failed therapy in the HIV+TB+, non-HAART group had emergence of ARR. Fifteen out of the 19 patients developed MDR-TB (resistance to both H and
ce
R) with four developing R mono-resistance. Three out of the six bacteriological failures in the HIV+TB+HAART group and one out of the nine failures in the HIV-
Ac
TB+ group, had emergence of ARR. The relative risk of ARR was 21.1 (95%CI 2.6184), p
