Guidance for national tuberculosis programmes on the management of tuberculosis in children: web annex: evidence summary tables

Guidance for national tuberculosis programmes on the management of tuberculosis in children, 2nd ed. Web Annex: Evidence summary tables 2 Summary of evidence reviewed by Panel in July 2012 Based on literature review from 2010 onwards N.B. Recommendations in boxes are numbered as in the final edited version of the Guidance for National TB Programmes on the Management of Tuberculosis in Children: second edition (that is submitted to GRC via e-PUB for its meeting in November 2013). Chapter 3: Diagnosis of TB in children Box 1: Recommendations regarding diagnosis of TB in children Xpert MTB/RIF for the diagnosis of pulmonary TB and rifampicin resistance in children 1.Xpert MTB/RIF should be used rather than conventional microscopy and culture as the initial diagnostic test in children presumed to have MDR TB or HIV-associated TB. 2. Xpert MTB/RIF may be used rather than conventional microscopy and culture as the initial test in all children presumed to have TB. Xpert MTB/RIF for the diagnosis of extrapulmonary TB in children 3. Xpert MTB/RIF may be used as a replacement test for usual practice (including conventional microscopy, culture, and/or histopathology) for testing of specific non-respiratory specimens (lymph nodes and other tissues) from children presumed to have extrapulmonary TB. 4. Xpert MTB/RIF should be used in preference to conventional microscopy and culture as the initial diagnostic test in testing cerebrospinal fluid specimens from children presumed to have TB meningitis. IGRAs 5. Interferon-gamma release assays (IGRAs) should not replace the tuberculin skin test (TST) in low- and middle-income countries for the diagnosis of latent TB infection in children or for the diagnostic work-up of children (irrespective of HIV status) suspected of TB disease in these settings. Commercial serodiagnostics 6. Commercial serodiagnostics should not be used in children suspected of active pulmonary or extrapulmonary TB, irrespective of their HIV status. 3 Routine HIV testing 7. Routine HIV testing should be offered to all patients, including children, with presumptive and diagnosed TB. Panel considerations on the use of Xpert MTB/RIF in children 2006 Guidance: “.. nucleic acid amplification (e.g. polymerase chain reaction) tests are not currently recommended for routine diagnosis of childhood TB, as they have been inadequately studied in children and have performed poorly in the few studies which have been done. However, this is an area that requires further research, as such tests may prove to be useful in the future” (page 7). 2011 recommendations: • Xpert MTB/RIF should be used as the initial diagnostic test in individuals suspected of having MDR-TB or HIV- associated TB. (Strong recommendation) • Xpert MTB/RIF may be considered as a follow-on test to microscopy in settings where MDR-TB or HIV is of lesser concern, especially in further testing of smear-negative specimens. (Conditional recommendation acknowledging major resource implications) This recommendation also applies to paediatric TB based on the generalization of data from adults and acknowledging the limitations of microbiological diagnosis of TB (including MDR TB) in children. Source: Automated Real-time Nucleic Acid Amplification Technology for Rapid and Simultaneous Detection of Tuberculosis and Rifampicin Resistance: Xpert MTB/RIF System: Policy Statement. World Health Organization, Geneva 2011. Comments: • Studies of evaluating Xpert in children beginning to emerge and three systematic reviews will be commissioned by WHO at the end of 2012. • Still few data from non-respiratory samples. • Lots of research in pipeline. • Challenges of obtaining respiratory samples remain. 4 Table 1: Paediatric studies published since 2010 on performance of NAAT especially Xpert MTB/RIF Reference Study Design, Location N Synopsis Major outcomes Quality assessment/ comments STARD and QUADAS Rachow 2012 Prospective cohort study TANZANIA N = 164 Suspected TB at least one of: cough, fever, weight loss/ FTT, signs of extrapulmonary TB Age 6 weeks – 14 years 51% HIV prevalence Smear microscopy and culture compared to Xpert Followed for min 12 months 41.5%: induced sputum 28 (17.1%): confirmed TB Xpert: Detected 21/28 of culture confirmed TB 100% of smear positive children 66.6% culture positive, smear negative tests Detected 4 (8.5%) cases with clinical TB, negative culture Specificity: 100% for cases where TB reliably excluded Cf LJ/ MGIT alone - 82.76% LG positive - 83.3% of MGIT positive STARD high quality Comments: Xpert → quicker testing and higher sensitivity Increased sensitivity with second and third test Not influenced by HIV Nicol 2011 Prospective cohort South Africa N = 452 Children < 15 years Suspected pulmonary TB Xpert v smear v liquid culture (2 sequential specimens) Xpert MTB/RIF tests performed twice 70 (16%): positive culture result 216 (48%): possible 166 (37%): not tuberculosis 27 (6%) +ve smear 58 (13%) +ve MTB/RIF test STARD high quality Comments: Lower sensitivity for smear negative disease (cf adult 5 HIV infected 24% 27/27 smear positive cases 25/43 smear negative culture positive cases Specificity: 98.8%; (false positive) 6 had Xpert pos/culture neg Including only 385 children with 2 results: Twice many cases as smear microscopy: 75.9% v 37.9% Detected all 22 smear positive 22/36 (61.1%) smear negative cases Spec: 98.8% Available in median 1 day (culture 12d) MTB/RIF sensitivity greater in children with HIV studies) Increased sensitivity with second test (28%) Still high proportion of false negatives high quality Problem is accuracy of those with clinical TB but not confirmed Rifampicin resistance: Identified all 70 susceptible Zar et al 2012 Prospective cohort study South Africa N = 674: 535 one IS and NPA, 396 two IS and NPAs All children < 15y 117 HIV infected 2009-11 Xpert results from IS/NPA compared with diagnosis from positive liquid culture (primary reference standard) Xpert sensitivity: 2 NPA – 65% 2 IS – 71% no sig diff incremental yield was 9 cases for IS, 11 cases for nPA) Xpert specificity 2 NPAs: 98.2 2 IS: 99.1% Smears Sensitivity: 34.5% CI 24.3-44.7 Limitations noted: Small numbr of HIV infected children Few rifampicin resistant cases Samples split between culture and Xpert 6 NPA v IS Culture yield higher in IS (84/87) v NPA (61/87) Papaventsis 2012 Retrospective cohort GREECE October 2007-Feb 2011 N = 121 Age 1-16 50: active TB 71: no infection Mixture of PTB and EPTB Pulmonary samples (gastric/ sputum/ bronchial aspirates), extrapulmonary samples tested with – microscopy, culture (LJ/ BACTEC) & Gen- probe amplified MTD® test, a nucleic acid amplification assay that detects presence of mycobacterial RNA Neg: < 30000 RLUs Pos: > 500000 RLUs Equivocal – in between and requires repeat testing Active TB in 50/121 : AFB microscopy pos 6% Culture: 16/50 (32%) 34 on clinical and radiological criteria AMTD 29 had positive results (ie 13 negative culture) All equivocal (eventually positive tests) were AFB negative Positive in 3 non-TB cases AMTD: (cf culture) Sensitivity 100% Spec 85% PPV: 50% NPV: 100% Cf clinical diagnosis 58% sensitivity 96% spec 91% PPV 76% NPV Moderate-low quality of methodology Main comment is that difficult to interpret positive results in culture confirmed cases Similar to findings of Oberhelman in Peru with PCR 7 References 1. Rachow A et al. Increased and Expedited Case Detection by Xpert MTB/RIF Assay in Childhood Tuberculosis: A Prospective Cohort Study. Clin Infect Dis 2012: 54: 1388-1397. 2. Nicol M et al. Accuracy of the Xpert MTB/ RIF test for the diagnosis of pulmonary tuberculosis in children admitted to hospital in Cape Town, South Africa: a descriptive study. Lancet Infect Dis. 2011 Nov; 11 (11): 819-24. 3. Zar HJ, et al. Rapid molecular diagnosis of pulmonary tuberculosis in children using nasopharyngeal specimens. Clinical Infectious Diseases 2012, Epub July. 4. Papaventsis et al. Impact of the Gen-Probe Amplified MTD Test on tuberculosis diagnosis in children. Int J Tuberc Lung Dis 16 (3) 384-390. Panel considerations: Publication bias was considered highly unlikely as the Xpert MTB/RIF is a new assay, produced by a single manufacturer and only available to a small number of investigators. Four studies were considered, all different comparisons, report results differently. Not appropriate or possible to pool on the basis of the published data. Not clear to the panel if pooled estimates would add to recommendation and definitely not appropriate to combine with adult data given difficulties of interpreting sensitivity and specificity. Two studies’ primary question is to report test accuracy versus ‘gold standard’ of culture; debatable how comparable the gold standard is. Use different approaches for comparing types of sputum samples. Two studies compare different sampling techniques in children – gastric lavage/nasal aspirate versus sputum/ induced sputum. Population of Nichol 2011 overlaps with the population of Zar 2012. Not clear what reference standard should be in this context – especially given definitions of clinical TB in children. Culture negative disease accepted to exist. Note that Xpert is compared to culture (which is appropriate) as the only gold standard available and sensitivity in that comparison was consistently lower than reported in adults with smear negative, culture positive disease. And then if compared to all children with clinical TB that are treated in these studies, then sensitivity is much lower. Perhaps the true result is somewhere in between? 8 All studies were undertaken in high burden hospital settings. No studies report cost and resource use. The panel would ‘grade’ the overall quality of evidence as ‘low’ without considering pooled estimates of test accuracy parameters due to the following: • Studies from high burden settings and hospitals only – very difficult to extrapolate to low burden settings, less well trained environments; • Imprecise estimates of test accuracy – small numbers for some estimates of performance parameters, variable comparator; • Test accuracy does seem to vary with method of sample collection; difficult to get summary estimate at present based on these data; and, • We only have test performance data and no clinical outcomes – the panel noted that this was accepted as a reasonable surrogate in the adult GRADE table assessment, but the panel does not agree that it is necessarily a valid surrogate in children, because of the existence of culture negative disease. The panel does not think rating for ‘inconsistency’ adds any information as the estimates are on the basis of individual studies. As stated for the adult data, the risk of publication bias is likely to be minimal – we were provided with the unpublished manuscripts. Note: At the time of its meeting (July 2012), the Panel was aware that WHO commissioned three systematic reviews to update and revise the 2011 Policy guidance on the utility of Xpert MTB/Rif for the diagnosis of tuberculosis and Rifampicin resistance in pulmonary, extrapulmonary and childhood TB. Therefore, the Panel recommended that WHO use the newest recommendations as approved by the WHO Guideline Review Committee in 2013. Panel considerations on the role of IGRAs in the diagnosis of TB in children 2006 Guidance: • “Serological tests are not currently recommended for routine diagnosis in childhood TB, as they have been inadequately studied in children and have performed poorly in the few studies which have been done.” (page 7) • No specific comment on IGRAs. 9 Recent recommendations: • “IGRAs should not replace the TST in low- and middle-income countries for the diagnosis of latent TB infection in children, nor for the diagnostic work-up of children (irrespective of HIV status) suspected of active TB in these settings (strong recommendation, very low quality of evidence). It should also be noted that there may be additional harms associated with blood collection in children and that issues such as acceptability and cost had not been adequately addressed in any studies” • “The quality of evidence for commercial serodiagnostic tests was very low, with harms/risks far outweighing any potential benefits (strong recommendation). It is therefore recommended that these tests should not be used in individuals suspected of active pulmonary or extra-pulmonary TB, irrespective of their HIV status. This recommendation also applies to paediatric TB based on the generalisation of data from adults.” Source: Use of tuberculosis interferon-gamma release assays (IGRAs) in low and middle income countries: Policy Statement, World Health Organization, Geneva 2011. Commercial serodiagnostic tests for diagnosis of tuberculosis: Policy Statement. Geneva, World Health Organization 2011. Comments: • Studies of IGRAs (ELISPOT and QFN) in children published over last 10 years and systematic reviews and meta-analyses of these studies listed below. • Note that IGRA is NOT a diagnostic test for disease – though has been evaluated in that context being used as TST is used – to indicate infection. • Consistent findings from recent and well conducted systematic reviews: heterogeneity of studies and small numbers of confirmed with inherent limitations of reference standards for infection and disease noted. Notwithstanding, no clear evidence that IGRAs should replace the role of TST in assessment of children for infection or as part of diagnostic assessment, especially in TB-endemic countries. • IGRAs may have an additional role in certain scenarios but are expensive and technically difficult. • Additional studies since systematic reviews also have marked heterogeneity of methodology, limitations and variability in quality – and do not change overall conclusion. • Small number of studies suggest IGRAs have increased sensitivity in HIV infected children (over TST). 10 Table 2: Meta-analyses of paediatric studies reviewing the use of IGRAs Reference N, patient characteristics Synopsis Major outcomes Quality assessment/ comments Mandalakas 2011 31 articles → 4122 children with data analysed 1998-June 2010 Children < 18 years included 18 countries; incidence >25/100,000 in 10 countries Definite TB: +ve culture/ smear microscopy/ histological or nucleic acid amplification Probable TB: three of: radiology, sx, other radiology, exposure, response to therapy Heterogeneous methodology TSTs and IGRAs – similar accuracy Test failure - QFT: 0-7% - ELISPOT: 0-21% - TST: 0-11% Rate of indeterminate results • QFT: 6.5% • ELISPOT: 3.5% • Association with young age, helminth coinfection, immune suppression Latent TB infection, pooled odds ratios for association with dichotomous exposure: • TST >5mm: 1.3, (95% CI: 0.7-2.7) • TST >10mm: 1.9 ( 0.98 – 3.8) • TST >15mm: 1.8 (0.7-5.0) • QFT: 3.5 (1.9-6.7) • ELISPOT: 1.3 (0.8-2.3) For low-middle countries: • IGRAS positively correlated with exposure • TST was not Clear exclusion criteria and independent selection process with defined reference standards QUADAS assessment tool used Estimated inter-study heterogeneity Screened for industry involvement Main comments: Most used active TB (confirmed and probable) as a reference standard in the absence of gold standard (confirmed) diagnosis Small numbers in subgroup analyses No studies listed in Table 1.2 included Association of test response with exposure—categorized dichotomously or as a gradient—was similar for all tests. The sensitivity and specificity of all tests were similar in diagnosing the disease. 11 For active TB: • Sensitivity and specificity slightly higher than IGRAS • CI overlapping ↓ sensitivity with age < 5 years, HIV infection, > 50% BCG vaccination Stratified analysis suggested lower sensitivity for all tests in young or HIV infected children. Heterogeneous methodology limited the comparability of studies and the interpretation of results. Sun 2011 16 articles N: 598 patients, 432 controls with no exposure Jan 2000- Jan 2011 Studies use of IGRAs for diagnosis of active disease Calculation of sensitivity , specificity for each individual study, pooled on forest plot Total estimates of sensitivity and specificity were weighted by sample size No HIV infection ELISA studies (9) – all in low incidence countries - 70% pooled sensitivity - 100% pooled specificity ELISPOT studies (10) – similar sensitivities between low/ high incidence studies - 62% pooled sensitivity - 90% pooled specificity TST (12) - similar sensitivities between low/ high incidence studies - 71% pooled sensitivity - 56% overall specificity, 49% if BCG vaccination Subanalysis of culture confirmed v clinically diagnosed TB ↓ sensitivity of all tests: ELISA (85%), ELISPOT (76%) and TST (85%) versus for ELISA (64%), ELISPOT (58%) and TST (66%) but no mention of numbers of confirmed Agreement between tests - Higher in non-BCG vaccinated Clear exclusion criteria Two independent reviewers Evaluated heterogeneity No independent assessment tool Main comments: High levels of heterogeneity amongst studies No mention of numbers of confirmed No studies in Table 1.2 included Agreement between the TST and IGRAs in non-BCG-vaccinated children is higher than in BCG-vaccinated. TST and IGRAs have similar sensitivity in the diagnosis of active TB,. Specificity of IGRAs is greater than TST particularly if previous BCG vaccination 12 Machingaidze 2011 Included 20 articles Pre January 2010 Only pediatric studies Studies use of IGRAS for diagnosis of latent infection and disease Excluded studies with children that were HIV positive For diagnosis of LTBI: QFT Sensitivity, pooled: 66% (95%CI 27- 72%) with 74.8% heterogeneity. Sensitivity was reduced in high incidence settings QFT specificity: statistics not presented For active disease (6 studies: 2 high incidence, 4 low incidence): Sensitivity: ranged from 53-94%, no diff noted Combination of tests improved sensitivity TST and QFT concordance: Ranged from 0.17-0.86 Indeterminate results: ↑ in younger children Clear exclusion criteria Two independent reviewers Evaluated heterogeneity Quality assessment done but approach not defined Main comments: Low number of culture confirmed cases in studies (6-49) Heterogeneity amongst studies No clear evidence that IGRAs should replace TST for detecting LTBI Sensitivity of the IGRA for TB disease was no different from TST, and significantly reduced in high-burden TB compared with low-burden TB settings. Sester 2011 4/27 studies included children ; 2 UK, 1 S/Africa, 1 Italy Jan 2001- November 2009 Indeterminate results was 3.6% (v 12.6% in adults) QFT Sensitivity in children 79.9 ± 20.9% Specificity (only reported in one study) 85.8% TSPOT Sensitivity in children 42.2 ± 11% Specificity (only reported in one study) 74% Studies included (Bamford 2010, Bianchi 2009, Kampmann 2009, Nicol 2009) – reviewed in above analyses 13 References 1. Mandalakas AM, Detjen AK, Hesseling AC, Benedetti A, Menzies D. Interferon-gamma release assays and childhood tuberculosis: systematic review and meta-analysis. INt J Tuberc Lung Dis. 2011 Aug; 15 (8) 1018-32. 2. Sun L. Xiao J. Miao Q. Feng WX. Wu XR. Yin QQ. Jiao WW. Shen C. Liu F. Shen D. Shen AD. Interferon gamma release assay in diagnosis of pediatric tuberculosis: a meta-analysis. FEMS Immunology & Medical Microbiology. 63(2):165-73, 2011 Nov. 3. Machingaidze, Shingai Wiysonge, Charles Shey, Gonzalez-Angulo, Yulieth Hatherill, Mark, Moyo, Sizulu MB; Hanekom, Willem Mahomed, Hassan Mmed. The Utility of an Interferon Gamma Release Assay for Diagnosis of Latent Tuberculosis Infection and Disease in Children: A Systematic Review and Meta-analysis. Pediatric Infectious Disease Journal. 30(11):pg. 694-700. 4. Sester M. Sotgiu G. Lange C. Giehl C. Girardi E. Migliori GB. Bossink A. Dheda K. Diel R. Dominguez J. Lipman M. Nemeth J. Ravn P. Winkler S. Huitric E. Sandgren A. Manissero D. Interferon-gamma release assays for the diagnosis of active tuberculosis: a systematic review and meta-analysis. [Review] European Respiratory Journal. 37(1):100-11, 2011 Jan. Table 3: Studies since 2010 comparing IGRAs/ TST , excluded those reviewed in above meta-analyses Reference Study design, Location N, patient characteristics Synopsis Major outcomes Quality assessment Dayal 2012 Case control, Diagnostic evaluation India Evaluation of QFT-G N = 150 82 cases 48 controls 20 PPD +ve controls < 18 years old Recruited from casualty/ V small number of culture confirmed cases : n=13 Of 13 culture confirmed by LJ or BacT or both, QFT positive in 8, negative in 3 and indeterminate in 2 Of 20 controls with TST positive but no disease, QFT positive in 8, negative in 9 and STARD – low quality Numbers small of confirmed cases Material and methods poorly explained Patient selection and time frame unclear Choice of controls unclear Almost no lab methodology Difficult to know what QFN was being compared to in various analyses Discussion does not address main 14 Wards of Pediatric Hospital indeterminate in 3 Of 48 TST neg controls, QFT positive in 14, negative in 23 and indeterminate in 11 QFT less affected than TST by BCG vaccination limitations well and makes assumptions e.g. very unusual to have many smear-positive culture-negative cases Shah 2011 Prospective cohort study South Africa 270 pediatric patients → 196 at 6 months 6 months – 16 years No prior diagnosis of active TB or LTBI 6 month follow up of children positive TST (≥5 mm) / QFT (household contacts of adults with newly dx PTB) Did not make comparison with culture results – therefore no assessment of test sensitivity/ specificity Does not really provide analysis No difference between TST and QFN at baseline in exposed children Changes in test results: • QFT: 29% → 38% (32% → 44% if loss to follow up excluded) • TST≥ 5mm: 28% → 33% +ve (no significant difference; p = 0.08) • TST ≥10 mm: 22% → 27% BUT only TST neg at Newcastle-Ottawa QAS – 6/9 stars Selection: representative (contacts of active PTB) no non-exposed cohort exposure assessed by: interview outcome of interest not present at start of study Comparability Comparison was between results at baseline versus results at 6 months All TB-exposed – no unexposed controls Outcome Assessment of outcome: prevalence of infection = record linkage/independent assessment Follow up = 6 months 74 (27%) loss to follow up and reported separately 15 of efficacy of IGRA baseline were retested (conversion 10% for TST ≥ 10 mm - Serial testing for contact investigation suggested Comments: Moderate quality but not an analysis of IGRA in context of diagnosis Yassin 2011 Cross sectional, case-control comparison ETHIOPIA N = 813: 322 children with suspected TB, 335 children in contact, 156 controls Stratified children into confirmed (n=28), probable (n=136), unlikely TB (n=131) contacts (n=335), controls (n=156) Compared test results (TST, IP10 , QFN-in tube) TST – negative, intermediate (5- 10) and positive (≥ 10 mm) TST pos: confirmed 78.6%, probable 59.3%, unlikely 28.7%, contacts 54.1%, controls 12.8% QFN pos : confirmed 59.3%, probable 37.5%, unlikely 28.1%, contacts 44.7%, controls 13.1% Concordance between test varied across group. TST+/QFN- in probable and contacts; TST-/QFN+ in unlikely and controls Agreement between TST and IGRA varied STARD – high quality Representation: enrolled consecutively from major health service providers Selection of non-exposed: from community , without known contact Exposure determined by medical investigations/ interview Large numbers Tests not blinded High numbers of indeterminate Comments: High quality but usual limitation of relatively small numbers of confirmed Low sensitivity of QFN even in confirmed High numbers of indeterminates (16- 18% in confirmed and contacts; and 36% in unlikely and probable) “given the high cost of the IGRAS and the high proportion of indeterminate results obtained, control programmes should refrain from incorporating these 16 from 81% (k = 0.49) in confirmed TB, 73% (k= 0.40) in probable TB and 66% (k =0.23) in unlikely TB, 70% (k = 0.39) in contacts and 81% (k= 0.24) in controls. No difference in IP10 concentrations between HIV negative and HIV positive , P >0.1 tests until clear advantages over the TST are demonstrated” Thomas 2010 Cross sectional, comparison BANGLADESH N = 312, 11-15.3 years No known history of TB disease QFT v TST (≥10 mm) for LATENT INFECTION , effect of malnutrition (31%) and helminth infection (47%) TST positive - 33% QFT positive - 35%, negative - 40%, and indeterminate - 25% Moderate agreement (K=0.55) STARD – methodology good for study purpose BUT for this purpose data low quality because not a direct comparison of IGRA with TB infection/disease limited to adolescents high proportion of indeterminates – associated with malnutrition and helminth infection Diel 2011 Prospective observational cohort study Comparison to develop TB N =1417 Close contacts of smear positive cases 141 = children Study population tested with QFT, monitored for progression to TB QFT pos 20.8% , TST > 10 mm: 25.4% 19 developed active TB: 100% QFT positive Newcastle-Ottawa QAS – 6/9 stars Selection Cases – close contacts of AFB positive cases Comparison with QFT or TST negatives at baseline Ascertainment of exposure: QFT 17 HAMBURG < 16 years 106 children had both TST and QFT 90% TST +ve > 5mm, 53% TST +ve > 10 mm Results in children: QFT pos 22% TST >5mm 38% TST > 10 mm 19% 6 developed active TB 100% QFT positive 100% TFT > 5 mm 4/6 TFT > 10 mm no significant difference in progression rates Children v adult progression rate after QFT positive: 28.6% v 10.3% ( p = 0.03) Age and IFN-gamma level significantly associated with disease positive , structured interview, radiography, culture + clinical suspicion Confirmed cases with PCR testing Comparability No control cohort of non-exposed Comparison of those developed TB between TST and QFN at baseline in untreated contacts Outcome Independent blind assessment Follow up: four years 15 of 141 children were LTFU Comment: High quality methodology BUT answers a different question i.e. not a direct comparison of IGRA with TB infection/disease Small numbers of children Markova 2011 Prospective cohort study Bulgaria N = 68 Active TB < 16 years Analysis of effect of age on IGRA response TST negative < 5mm, Positive > 15 mm 68 children; 17 Culture positive QFN positive 13 (77%) TST ≥ 10 mm 14 (82%) 33 culture negative Low quality data Aims to answer a different question Small study – aim to determine effect of age on IFN responses not compare diagnostic tests 18 children with clinical TB QFN positive 20 (61%) TST ≥ 10 mm 13 (39%) Altet- Gomez 2011 Prospective cohort study SPAIN N = 166 < 15 years old 98 children from contact tracing, 68 with TST (≥ 5 mm) identified on public health screenings TSPOT v QFT v TST to diagnose active disease Active disease: culture / presence of sx and radiology / +ve TST responding to antituberculous chemo Overall: TST 149 (89.8%) positive Tspot: 64 (38.6%) QFT 61 (36.7%) IGRAs concorded in 146/166 No factors associated with discordance 14 cases of active TB: TSPOT 11 (79%) QFT 9 (64%) Positive concordance in 8 (57%) Together sens/ spec: 79%, 100% TST and IGRAs Agreement higher in non-BCG vaccinated children Low quality data for the question of TB diagnosis in children in TB endemic setting Low TB endemic setting Usual common discordance in that setting TST+/IGRA – Small numbers of children with confirmed disease Adds little to data from low TB endemic settings Setiawati 2011 Cross sectional study, INDONESIA 37 children 1-15 years Household TST ≥10mm, IGRA, gastric aspirate culture Comparison of tests Overall agreement: moderate (k = 0.526) STARD low quality data Small numbers 19 contact with AFB +ve case Analysis of test concordance with BCG use of BCG scar does not present analysis of IGRA v culture results Without BCG: high level of agreement (k = 0.857) • BCG scars: no significant agreement (k = 0.251, p = 0.226) Exposure ascertained by medical records Use of BCG scar Cross-sectional observational study conducted in health facility No control References 1. Dayal R. Verma V. Sharma B. Kumar G. Kumar N. Gupta R. Katoch VM. Joshi B. Chauhan DS. Diagnostic value of interferon- gamma release assays (QuantiFERON-TB Gold[REGISTERED] In Tube) in childhood tuberculosis. Indian Journal of Pediatrics. 79(2):183-7, 2012 Feb. 2. Shah M. Kasambira TS. Adrian PV. Madhi SA. Martinson NA. Dorman SE. Longitudinal analysis of QuantiFERON-TB Gold In-Tube in children with adult household tuberculosis contact in South Africa: a prospective cohort study. PLoS ONE [Electronic Resource]. 6(10):e26787, 2011. 3. Yassin MA, Petrucci R, Garie KT, Harper G, Arbide I, Aschalew M, Merid Y, Kebede Z, Bawazir AA, Abuamer NM, Cuevas LE. Can interferon-gamma or interferon-gamma-induced-protein-10 differentiate tuberculosis infection and disease in children of high endemic areas? PLoS One. 2011;6(9):e23733. Epub 2011 Sep 23. 4. Thomas TA et al. Malnutrition and Helminth Infection affect performance of an interferon gamma release assay. Pediatrics 2010. 126 5. Diel R. Loddenkemper R. Niemann S. Meywald-Walter K. Nienhaus A. Negative and Positive Predictive Value of a Whole- Blood Interferon-{gamma} Release Assay for Developing Active Tuberculosis: An Update. American Journal of Respiratory & Critical Care Medicine. 183(1):88-95, 2011 Jan 1. 6. Markova R. Drenska R. Minchev P. Todorova Y. Ciccozzi M. Amicosante M. Association of age with the level of response in the Quantiferon-TB Gold In-Tube assay for children with active tuberculosis. New Microbiologica. 34(1):81-5, 2011 Jan. 7. Altet-Gomez N et al. Diagnosing TB infection in children: analysis of discordances using in vitro tests and the tuberculin skin test. Eur REspir J 2011; 27: 1166-1174. 8. Setiawati L. Endaryanto A. Kusumadewi A. Lestari P. Effect of BCG vaccination and non-tuberculous Mycobacterium infection on interferon gamma specific assay and a tuberculin skin test among children with a tuberculosis contact in Surabaya, Indonesia. Southeast Asian Journal of Tropical Medicine and Public Health. 42.6 Nov 2011. 20 Table 4: Studies noting improved performance of IGRA in HIV infected children with TB disease Reference Study design, Location N, patient characteristics Synopsis Major outcomes Quality assessment STARD and QUADAS Liebeschetz 2004 Prospective blinded study South Africa N = 293 Suspected TB inpatient and outpatient 57 confirmed 76 highly probable 13 not TB HIV test 164 – 46% positive Patients assessed clinically, by ELISPOT and TST Clinicians blinded to ELISPOT result ELISPOT v TST: 73% (n = 133) v 36% (n = 116) for confirmed and ‘highly probable TB ELISPOT more likely to be positive for young age, HIV and malnourished Only 60% of eligible children were enrolled Not all confirmed were culture positive 31 ELIPOT results not available – technical difficulties Not all had TST Not all HIV tested Davies 2009 Prospective South Africa N = 188, 22 definite TB All HIV-infected Hospital patients with suspected TB Evaluates ELISPOT for diagnosis of TB in HIV infected children Clinicians not blinded ELISPOT v TST: 66% vs 33% for definite 64 %v 29% for definite or probable TB ELISPOT particularly better for those with marked immunosuppression Small numbers of definite and not all definite were culture positive High LTFU of TST e.g. 15 of 22 has TST done and read Stavri 2009 Cohort study Romania N = 36 Confirmed TB (micro/histo) Comparison of ELISPOT, TST AND QFT for diagnosis in HIV infected children Positive: 39% TST; 47% QFN; 11% ELISPOT – tendency for better sensitivity if low CD4 Only abstract available Only 36 patients but all confirmed 21 References 1. Liebeschetz S, Bamber S, Ewer K, Deeks J, Pathan AA, Lalvani A. Diagnosis of tuberculosis in South African children with a T-cell based assay: a prospective cohort study. Lacet 2004 Dec; 364 (9452): 2196-203 2. Davies MA, et al. Detection of tuberculosis in HIV infected children using an enzyme-linked immunospot assay. AIDS 2009 23: 961-969. 3. Stavri H, et al. Comparison of tuberculin skin test with a whole-blood interferon gamma assay and ELISA, in HIV positive children and adolescents with TB. Roum Arch Microbiol Immunol. 2009: 68:14-9. (QFN and ELISPOT). Panel considerations on “Routine HIV testing should be offered to all patients with presumptive and diagnosed TB” 2006 guidance: “HIV testing is indicated for all TB patients as part of their routine management in areas with a high prevalence of HIV infection in the general population (>1%) , where TB and HIV are likely to coexist. Source of recent recommendation: Guidance on provider-initiated HIV testing and counseling in health facilities. World Health Organization, Geneva 2007. WHO policy on collaborative TB/HIV activities. Guidelines for national programmes and other stakeholders, 2012. World Health Organization, Geneva, 2012. Comments: • High prevalence of HIV among children with TB. • HIV infected children are at increased risk of TB. • HIV infection impacts upon diagnosis and outcome in children with TB. • HIV infected children require additional interventions in addition to TB diagnosis/treatment. • Current data that includes children with confirmed disease is mainly from TB-HIV endemic setting. • Most data from HIV endemic setting. • Consider implications in low HIV setting. Main question for the panel: should this be a blanket recommendation/should it remain for HIV endemic setting? 22 Table 5: Key references regarding routine HIV testing in the setting of presumptive/ diagnosed TB Reference Synopsis Major outcomes Quality assessment / additional comments 2012 WHO Policy on Collaborative TB Ref p. 26 Recommendation C.1: Routine HIV testing should be offered to all patients with presumptive and diagnosed TB (strong recommendation, low quality of evidence) Based on evidence from HIV settings 2010 Guidance on TB /HIV Joint WHO IULTD Ref p.13 Recommendation: HIV testing is recommended for all children who are TB suspects or TB patients Collaborative document World Health Organization/ UNAIDS 2007. Guidance on provider initiated HIV testing Ref p.8-9 Recommendations for all epidemic types: “Health care providers should recommend HIV testing and counselling as part of the standard of care to: adolescents/children who present to health facilities with signs, symptoms or medical conditions that could indicate HIV infection... This includes tuberculosis infection” Pre-GRADE Hesseling et al 2009 Prospective cohort (hospital) surveillance 2004-2006 South Africa Includes all culture confirmed TB cases (n = 245) Incidence of TB among HIV infected infants: 1596/100,000 compared to: 66/100,000 for HIV uninfected infants RR 24 (95% CI 17-34) Incidence calculated on basis of population estimates Schaaf 2007 Retrospective record review 2003-2005 (pre-ART era) N = 596, all culture confirmed cases: Of 414 HIV tested, 133 were HIV infected Cape Town, South Africa Retrospective Very large series 23 Wiseman CA et al 2011 Retrospective 52 HIV-infected infants HIV-infected infants at risk of TB disease and TB-related death 17 32.7%) died 10 (19.2%) died before ART initiated HIV disease stage associated with death (p = 0.004) Retrospective Viani 2011 Cohort study prospective Mexico 73 HIV-infected children incl as ART coverage increased TB second commonest cause of death after pneumonia Mexico Proportion of patients on HAART 1998-2001: 60% 2002-2003:75% 2004-2005:83% 2006-2007:94% (sig diff: p <0.001) TB – second most common cause of death (22%) after pneumonia Abstract only Evidence from outside of Africa and lower TB endemic setting Autopsy studies TB a common cause of death in HIV-infected children (pre- ART) Zambia, Botswana, Zimbabwe Selection bias Hospital-based studies of children with TB in high HIV setting 30-60% of children with TB are HIV-infected (pre-ART) Selection bias Hospital based studies References: 1. World Health Organization. WHO policy on collaborative TB/HIV activities. Guidelines for national programmes and other stakeholders. Geneva, 2012. 24 2. World Health Organization & UNAIDs. Guidance on provider-initiated HIV testing and counselling in health facilities. World Health Organization, Geneva 2007. 3. Hesseling AC et al. High incidence of TB among HIV-Infected infants: evidence from a South African Population-Based study highlights the need for improved tuberculosis control strategies. Clinical Infectious Diseases 2009; 48: 108-114. 4. Schaaf HS et al. Culture-confirmed childhood tuberculosis in Cape Town, South Africa: a review of 596 cases. BMC Infectious Diseases, 2007. 5. Viani RM, Araneta MR, Lopez G, Chacón-Cruz E, Spector SA. Clinical Outcomes and Hospitalizations among Children Perinatally Infected with HIV-1 in Baja California, Mexico. J Int Assoc Physicians AIDS Care (Chic). 2011;10(4): 223-8. Panel considerations regarding scoring systems 2006 Guidance: The first edition of the guidance did contain a box on key features suggestive of TB (see below) Box 4 Key features suggestive of TB The presence of three or more of the following should strongly suggest a diagnosis of TB: • chronic symptoms suggestive of TB • physical signs highly of suggestive of TB • a positive tuberculin skin test • chest X-ray suggestive of TB Comments: • This was not on evidence base and was not intended to be another “scoring system” but has been used as such “three or more”. • “WHO scoring system” is often quoted but WHO has never developed or validated a scoring system. Often refer to Harries AD, et al. TB/HIV Manual – which includes a scoring system based on Keith Edwards system developed in PNG. • Updated guidance should state in text: “In some countries, score charts are used for the diagnosis of TB in children, although they have rarely been evaluated or validated against a "gold standard". Therefore, they should be used as screening tools and not as the means of making a firm diagnosis. Score charts perform particularly poorly in children suspected of pulmonary TB (the most common form) and in children who are also HIV-infected.” 25 Question to panel: should Box 4 be omitted from the second edition of the guidance? Should there be a statement about scoring systems INCLUDING a negative statement? Table 6: Studies on scoring systems for diagnosis of TB in children Reference Study type Location Patient characteristics Synopsis Major outcomes Quality assessment/comments Hatherill 2010 South Africa N=1445 < 2 years old; Children suspected of pulmonary TB in community underwent further investigation Evaluated case frequencies yielded by 9 different scoring systems 2001-2006 Significant differences noted between most scoring systems Frequency of case detection did not correlate to high agreement between systems Even high frequency detection had fair to moderate agreement (K = < 0.4, 0.4-0.6) Variation of TB cases was from 7 to 89% Only major study that tries to compare and validate study population is community based ; does not account for setting in which some systems were designed Graham 2011 Review NA Summarises scoring systems Limited validation (in absence of gold standard for diagnosis) May lack practicality in application/ use An update – similar to Hesseling et al 2002 Includes recent system from Brazil which performs poorly in recent study 2012 Hesseling 2002 Review NA Evaluated 16 diagnostic approaches including point scoring systems, diagnostic classifications, diagnostic algorithms Noted variability in terms used Gold standard – typically bacteriological, microscopy or culture Mostly in hospital settings Comprehensive review Lack of comparison to gold standard and adaptations for HIV/ malnutrition 26 References 1. Graham, S. The Use of Diagnostic Systems for Tuberculosis in Children. Indian Journal of Pediatrics (March 2011) 78:334- 339. 2. Hesseling AC, Schaaf HS, Gie RP, Starke JR, Beyers N. A critical review of diagnostic approaches used in the diagnosis of childhood tuberculosis. International Journal of Tuberculosis Lung Disease 6(12):1038-1045. 3. Hatherill M, Hanslo M, Hawkridge T, Little F, Workman L, Mahomed H, Tameris M, Moyo S, Geldenhuys H, Hanekom w, Geiter L, Hussey G. Structured approaches for the screening and diagnosis of childhood tuberculosis in a high prevalence region of South Africa. Bulletin of the World Health Organization 2010; 88: 312-320. Additional references regarding diagnosis - not specifically related to key recommendations Table 7: Additional studies regarding diagnosis of TB in children Reference Study Design, Location N Synopsis Major outcomes Perez- Porcuna 2012 Prospective Brazil N = 32 children and 61 respiratory samples Children 0-5 years with clinically suspected TB and in contact with a new case of pulmonary TB within last 6 months Comparison of MODS with LJ Two samples by induced sputum by NPA → sputum smear, LJ culture, liquid culture in MGIT and PCR Sputum smear pos: 0 Solid cultures (LJ): 3/31 (9.7%) – at least one positive Liquid cultures (MGIT): More rapid than solid (p< 0.001) 7/31 (22.6%) at least one positive 4 of these “unlikely” with clinical score Concordance b/w LJ, MGIT moderate (0.540) PCR: more positives than microbiologic cultures no concordance with microbiological culture, clinical data Poor concordance between culture and clinical 27 scores. Oberhelman 2010 Case control PERU N = 456; 218 cases, 238 controls < 12 years, Stegen Toledo score > 5 points, no anti-TB therapy April 2002- January 2007 Comparison of NPA, stool and gastric aspirates Comparison of MODS v LJ culture Sensitivity, specificity and predictive values of a PCR assay Confirmed by LJ or MODS N=22 Source of specimen: - gastric aspirate – all 22 - NPA: 12 - Stool – 4 58 total specimens 37% from a second gastric aspirate MODs v LJ Culture times: MODs median 10 days v 25, p = 0.0001 PCR results: Healthy controls had positive Positive in 13/21 (62%) culture positive Positive in 26/128 (20%) culture negative PPV 50% from gastric aspirate Hepple 2012 Systematic review 23 studies, total 3127 participants 15 adults, 6 children, 1 both, 1 unspecified Pulmonary TB Microscopy v culture (reference standard) Obtaining samples: - Induction: 76% - 100% Microscopy: - 0-100% positive Culture: - 2.4% - 100% positive References 1. Perez- Porcuna Tm, Ascaso C, Ogusku MM, Abellana R, Malheiro A, Quinco P, Antunes I, Monte R, Tavares M, Garrido M, Buhrer-Sekula S, Martinez-Espinosa FE. Evaluation of New Strategies for the Diagnosis of Tuberculosis among Pediatric Contacts of Tuberculosis Patients. The Pediatric Infectious Disease Journal Publish ahead of print. 2. Oberhelman RA, Soto-Castellares G, Gilman RH, Cavides L, Castillo ME, Kolevic L, Del PIno T, Saito M, Salazar-Lindo E, Negron E, Monteegro S, Laguna-Torres VA, Moore DA, Evans CA. Diagnostic approaches for paediatric tuberculosis by use 28 of different specimen types, culture methods, and PCR: a prospective case-control study. Lancet Infect Dis. 2010 Sep; 10 (9) 612-20. 3. Hepple, P, Ford N, McNerney R. Microscopy compared to culture for the diagnosis of tuberculosis in induced sputum samples: a systematic review. Int J Tuberc Lung Dis 16 (5): 579-588. 29 Chapter 4: Treatment of children with TB Box 2: Recommendations regarding treatment of children with TB 8. The following dosages of anti-TB medicines should be used daily for the treatment of TB in children: isoniazid (H) 10 mg/kg (range 7–15 mg/kg); maximum dose 300 mg/day rifampicin (R) 15 mg/kg (range 10–20 mg/kg); maximum dose 600 mg/day pyrazinamide (Z) 35 mg/kg (range 30–40 mg/kg) ethambutol (E) 20 mg/kg (range 15–25 mg/kg) 9. 9. Children with suspected or confirmed pulmonary TB or tuberculous peripheral lymphadenitis who live in settings with low HIV prevalence and/or low prevalence of isoniazid resistance and children who are HIV-negative, can be treated with a three-drug regimen (HRZ) for 2 months followed by a two-drug (HR) regimen for 4 months at the dosages specified in Recommendation 8. 10. Children with suspected or confirmed pulmonary TB or tuberculosis peripheral lymphadenitis and/or children with extensive pulmonary disease, living in settings where the prevalence of HIV is high and/or the prevalence of isoniazid resistance is high should be treated with a four-drug regimen (HRZE) for 2 months followed by a two-drug regimen (HR) for 4 months at the dosages specified in Recommendation 8. 11. Infants aged 0–3 months with suspected or confirmed pulmonary TB or tuberculous peripheral lymphadenitis should be promptly treated with the standard treatment regimens, as described in recommendation 9 or 10. Treatment may require dose adjustment to reconcile the effect of age and possible toxicity in young infants. The decision to adjust doses should be taken by a clinician experienced in managing paediatric TB. 12. During the continuation phase of treatment, thrice-weekly regimens can be considered for children known not to be HIV-infected and living in settings with well-established directly-observed therapy (DOT). 13. Streptomycin should not be used as part of first-line treatment regimens for children with pulmonary TB or tuberculous peripheral lymphadenitis. 14. Children with suspected or confirmed tuberculous meningitis and children with suspected or confirmed osteoarticular TB should be treated with a four-drug regimen (HRZE) for 2 months, followed by a two-drug regimen (HR) for 10 months, the total duration of treatment being 12 months. The doses recommended for the treatment of tuberculous meningitis are the same as those described for pulmonary TB. 30 Doses of first line anti-TB drugs are revised to achieve optimal concentration levels 2006 guidance: Recommended doses of first-line anti-TB drugs for adults and children were as follows: Drug Recommended dose Daily Three times weekly Dose and range (mg/kg body weight) Maximum (mg) Dose and range (mg/kg body weight) Daily maximum (mg) Isoniazid 5 (4–6) 300 10 (8–12) – Rifampicin 10 (8–12) 600 10 (8–12) 600 Pyrazinamide 25 (20–30) – 35 (30–40) – Ethambutol children 20 (15–25)* adults 15 (15–20) – 30 (25–35) – Streptomycin** 15 (12–18) – 15 (12–18) – * The recommended daily dose of ethambutol is higher in children (20 mg/kg) than in adults (15 mg/kg), because the pharmacokinetics are different (peak serum ethambutol concentrations are lower in children than in adults receiving the same mg/kg dose). Although ethambutol was frequently omitted from treatment regimens for children in the past, due in part to concerns about the difficulty of monitoring for toxicity (particularly for optic neuritis) in young children, a literature review indicates that it is safe in children at a dose of 20 mg/kg (range 15–25 mg/kg) daily (3). ** Streptomycin should be avoided when possible in children because the injections are painful and irreversible auditory nerve damage may occur. The use of streptomycin in children is mainly reserved for the first 2 months of treatment of TB meningitis. Recent guidance: Rapid Advice: Treatment of tuberculosis in children. World Health Organization, Geneva 2010. WHO recommends the following dosages of antituberculosis medicines for the treatment of tuberculosis in children: isoniazid (H) – 10 mg/kg (range 10–15 mg/kg); maximum dose 300 mg/day 31 rifampicin (R) – 15 mg/kg (range 10–20 mg/kg); maximum dose 600 mg/day pyrazinamide (Z) – 35 mg/kg (30–40 mg/kg) ethambutol (E) – 20 mg/kg (15–25 mg/kg) Comments: • Recognition from PK studies that 2006 recommended dosages resulted in low levels of drug in young children. • Recognition that in groups at risk of poor outcome ( HIV infected children, young infants) adequate levels may result in better outcomes. • Data used to inform revised recommendations was from PK studies, not clinical outcome studies. • Careful consideration was given to whether increased dosages would result in increased toxicity and it was concluded that the new recommendations were within ranges with excellent safety profile. • Therefore newly recommended dosages and ranges were published in 2010. • This has caused challenges for implementation using the available fixed dose combinations and a preferred fixed dose combination has been submitted to WHO and is awaiting interest from manufacturers. This is HRZ 50:75:150 – note that H:R is 2:3. • A problem has occurred because although the Isoniazid dosage is 10 mg/kg the range is currently recommended as 10-15 mg/kg, which means that the recommended doses is at the lower end of the range. This has caused problems for dosing schedules. It must be considered whether a wider range (e.g. 5-15 mg/kg) should be recommended with 10 mg/kg as the midpoint. Question to the panel: Should the range of INH be revised to have10 mg/kg as midpoint with the lower end of range at 7 mg/kg? Table 8: Studies published since 2010 on performance/ pharmacokinetics on first line anti-tuberculosis drugs Reference Study design, location N, pt demographics Synopsis Treatment regimen Key outcomes Quality assessment/comments ISONIAZID, PYRAZINAMIDE AND RIFAMPICIN Thee 2011 Prospective case-control study, N = 20 Assessed 2 weeks after treatment RHZ at previous recommended INH (n = 20) Higher Cmax 10/20 receiving 5 mg/kg High quality data Conducted in 32 South Africa All Children < 2 years old 5 HIV +ve All received H and Z – only 11 received R in regimen commenced, sampled at 0.5, 1.5, 3 and 5 hours after dosing with previous recommendations Repeated after one week with new recommendations dosages and later RHZ at revised dosages had low Cmax (< 3 ug/ml) but nil on 10 mg/kg PZA (n = 20) Higher C max 15/20 on 25 mg/kg had low Cmax (< 35 ug/ml), 1 on 35 mg/kg RMP (n=11) Higher C max 6/11 on 10 mg/kg had low Cmax (< 8 ug/ml) v 3 on 15 mg/kg Other key findings: RMP not shown to interfere with pharmacokinetics of others No diff in pharmacokinetics between HIV and non- HIV representative population and all very young Identical controls Exact dosage uncertain ISONIAZID Roy 2010 Prospective cohort study India N = 20 5-12 years Comparison according to nutrition status, INH levels measured pre- dose and 1, 2, 4, 6, 8 and 24 hours INH 5 mg/kg No significant differences in levels between underweight and normal – tendency to higher in undernourished peak INH concentration Short communication Used weight-for-age Z score to grade malnutrition – not ideal Numbers in each group not reported 33 after INH (Cmax) was >3 μg/ml in all the patients, “about 150–200 times the MIC of INH of 0.025–0.05 μg/ml” thought to be all slow acetylator population – not known Thee, 2010 Retrospective case series Germany Study 1: 20 children 1- 13 years Study 2: 25 children 1- 13 years - Comparison to adults in study n = 24, 22 Studies of INH levels in children v adults Study 1: INH: 5mg/kg BW orally and 0,1,2,4,6 hours sampling Study 2: Subcutaneous INH 5 mg/kg BW then 200 mg/m2 BSA In same children sampled once at 4 hours INH levels < than adults, esp in children < 8 years old Dosing according to BSA: similar serum levels to adults ; recommends 200 mg/m2 BSA Reporting data from studies in 1960 and 1961 Studies different sampling times – sampled once only in second study Children also receiving PAS and streptomycin Use of subcut INH in study 2 – results may be less generalizable PYRAZINAMIDE McIlleron 2011 Prospective case series, South Africa N = 34 3months – 13 years Plasma concentrations measured 1 month post treatment (0.75, 1.5, 3.0, 4.0 and 6.0 hours after observed treatment dose) PZA, varied dose Aimed to deliver 15-30 mg/kg Linear relationship between dose and levels with not reaching adequate C max until up and above 25 mg/kg dose Peak concentrations were insufficient (< 20 High quality PK data Observed treatment Brief report 34 Median dose: 23 mg/kg mg/L) in 5 (15%) of children; most likely if dose < 20 mg/kg Most Cmax < 35 mg/L Concentration levels not associated with age, kwashiorkor, HIV infection No evidence of hepatitis; 2 children with transiently elevated ALTs OTHERS Donald 2012 Review Reviews 8 PK studies of PZA in children, and compares to adult PK data All predate 2010 except McIlleron et al 2011 Cmax in adults and children are equivalent when same dose given Any less than 30 mg/kg will result in younger children having low concentrations (< 20 ug /mL) Varied results in patients with HIV infection Inclusive and comprehensive as part of review to inform recent update 2010 Donald 2011 Review Reviews 9 paediatric PK studies of children not established on RIF and 6 studies of children established on RIF All pre-1997 except Schaaf 2009 Compares to healthy and “adults, both volunteers and patients established on RMP reach higher Cmax values than children; children established on RMP require approximately twice the mg/kg body weight dosage of RMP to reach serum concentrations equivalent to those of adults.” Comprehensive as part of review to inform recent update 2010 35 diseased adults Review suggests that dosage for children should be in range 10-20 mg/kg Donald 2011 Review Extensive review of literature and selected studies from adults for comparison of hepatotoxicity Amongst 12,708 children receiving chemoprophylaxis, mainly with INH alone, but also other combinations of INH, RMP and PZA only 1 case (0.06%) of jaundice was recorded and abnormal liver functions documented in 110 (8%) of the 1225 children studied. Excluding tuberculous meningitis (TBM) 8984 were children treated for tuberculosis disease and jaundice documented in 75 (0.83%) and abnormal liver function tests in 380 (9.9%) of the 3855 children evaluated. Amongst 717 children treated for TBM, however, jaundice occurred in 72 (10.8%) and abnormal LFT were recorded in 174 (52.9%) of those studied. Case reports document the occurrence of antiTB drug induced hepatotoxicity (ADIH) in at least 63 children. Signs and symptoms of ADIH were frequently ignored in the recorded cases. ADIH can occur in children at any age or at any dosage of INH, RMP or PZA, but the incidence of. ADIH is considerably Comprehensive as part of review to inform recent update 2010 209 references Range of mg/kg used ep fro INH Largest body of evidence is for INH 36 lower in children than in adults. Children with disseminated forms of disease are at greater risk of ADIH. The use of the higher dosages of INH, RMP and PZA recently recommended by WHO is unlikely to result in a greater risk of ADIH in children. Jeena 2011 Clinical trial simulations for 10,000 children 10,000 ‘simulated children’ < 10 years old Algorithim used to intimate clinical variability in 10,000 children Variable doses of INH: 2.5, 5, 7.5, 10. 15, 20 mg/kg day Suggests variation is so great that standardized therapies are impractical → need to consider, age, disease process and acetylation speed Theoretical Table 9: Paediatric PK studies of second line anti-TB drugs Reference Study design, location Number, Description Treatment regimen Key outcomes Thee 2011 Prospective case series, South Africa 31: 10 < 2 years 11 2-6 years 10 6-12 years ETH 15-20 mg/kg (16) ETH 15-20 mg/kg AND RIF (15) Levels at 0,1,2,3,4 and 6 hours following 1 month and 4 months of therapy Lower ETH concentrations seen in children < 2 ys compared to children 6-12 years HIV associated with reduced exposure (AUC) at both time points No difference in PK parameters for ETH when receiving RMP High quality PK methodology and data 37 No significant difference of ETH concentrations at 1 and 4 months Oral dose of 15-20 mg/kg gives adequate levels References 1. Thee S. Seddon JA. Donald PR. Seifart HI. Werely CJ. Hesseling AC. Rosenkranz B. Roll S. Magdorf K. Schaaf HS. Pharmacokinetics of isoniazid, rifampin, and pyrazinamide in children younger than two years of age with tuberculosis: evidence for implementation of revised World Health Organization recommendations. Antimicrobial Agents & Chemotherapy. 55(12):5560-7, 2011 Dec. 2. Roy V. Gupta D. Gupta P. Sethi GR. Mishra TK. Pharmacokinetics of isoniazid in moderately malnourished children with tuberculosis. International Journal of Tuberculosis & Lung Disease. 14(3):374-6, 2010 Mar. 3. Thee S. Detjen AA. Wahn U. Magdorf K Isoniazid pharmacokinetic studies of the 1960s: considering a higher isoniazid dose in childhood tuberculosis. Scandinavian Journal of Infectious Diseases. 42(4):294-8, 2010 Apr 1. McIlleron H, Willemse M, Schaaf HS, Smith PJ, Donald PR. Plasma concentrations of pyrazinamide in young children with tuberculosis. Pediatr Infect Dis 2011;30:262e5. 2. Donald, PR. Maritz, JS. Diacon, A.H. Pyrazinamide pharmacokinetics and efficacy in adults and children. Tuberculosis 2012; 92: 1-8 3. Donald PR. Maritz JS. Diacon, A.H. The pharmacokinetics and pharmacodynamics of rifampicin in adults and children in relation to the dosage recommended for children. Tuberculosis 2011; 91: 196-207 4. Donald PR. Antituberculosis drug-induced hepatotoxicity in children. Pediatr Rep 2011; 3: e16 5. Jeena PM. Bishai WR. Pasipanodya JG. Gumbo T. In silico children and the glass mouse model: clinical trial simulations to identify and individualize optimal isoniazid doses in children with tuberculosis. Antimicrobial Agents & Chemotherapy. 55(2):539-45, 2011 Feb. 6. Thee S. Seifart HI. Rosenkranz B. Hesseling AC. Magdorf K. Donald PR. Schaaf HS. Pharmacokinetics of ethionamide in children. Antimicrobial Agents & Chemotherapy. 55(10):4594-600, 2011 Oct. 38 Panel considerations on “Intermittent regimens should not be used to treat children with suspected/ confirmed pulmonary TB or TB peripheral lymphadenitis living in settings with HIV prevalence” Table 10: Intermittent regimens Reference Study design, location Number, Patient demographics Treatment regimen Key outcomes Menon 2010 Systematic review 4 RCTs; 466 children Children < 16 years old Intermittent therapy v Daily therapy Lower cure rates in children on twice weekly regimen v daily regimen OR: 0.27, 95% 0.15 – 0.51 Chang 2011 Systematic review 32 studies, pre-2010 3 studies on HIV Adults HIV: 3 studies – one systematic review, 2 retrospective cohort analyses: • All suggest intermittent treatment reduces efficacy • Level of evidence 1+ • Grade of recommendation A Child studies: one study, not HIV specific Comment: Studies highlight benefit of daily therapy in relation to treatment outcome, not to HIV prevalence. References 1. Menon PR. Lodha R. Sivanandan S. Kabra SK. Intermittent or daily short course chemotherapy for tuberculosis in children: meta-analysis of randomized controlled trials. Indian Pediatrics. 47(1):67-73, 2010 Jan. 2. Chang KC. Leung CC. Grosset J. Yew WW. Treatment of tuberculosis and optimal dosing schedules. [Review] Thorax. 66(11):997-1007, 2011 Nov. 39 Other studies related to treatment of TB – not directly related to updated recommendations Including vitamin supplementation in treatment with active tuberculosis Table 11: Other studies related to treatment of TB Including vitamin supplementation in treatment with active tuberculosis Reference Type of study Synopsis Results Quality/ comments Mehta 2011 Tanzania RCT Multivit (excl Fe and Zn) versus placebo 2005-7 255 children randomized to vitamins/ placebo → primary endpoint = weight gain Median age 1.5 years 87 (34%) HIV-infected No significant difference in weight overall Improved levels of haemoglobin associated with supplement Nb – subgroups: 6 week- 6 month: 1.08 v 0.46 kg, p 0.01 Increased height in HIV-infected children with multivitamin Low quality Very short trial period ( first 8 weeks of anti-TB therapy), small number Trial terminated early due to funding constraints Any effect in subgroup not necessarily TB related No mention of ART use in HIV-infected Sinclair 2011 Cochrane Mainly adult References: 1. Mehta S. Mugusi FM. Bosch RJ. Aboud S. Chatterjee A. Finkelstein JL. Fataki M. Kisenge R. Fawzi WW. A randomized trial of multivitamin supplementation in children with tuberculosis in Tanzania. Nutrition Journal. 10:120, 2011. 2. Sinclair D. Abba K. Grobler L. Sudarsanam TD. Nutritional supplements for people being treated for active tuberculosis. [Review] [Update of Cochrane Database Syst Rev. 2008;(4):CD006086; PMID: 18843702] Cochrane Database of Systematic Reviews. (11):CD006086, 2011. 40 Chapter 5: Prevention of TB in Children Box 3: Recommendations regarding prevention of TB in children: BCG vaccination BCG vaccination 15. In settings where TB is highly endemic or where there is high risk of exposure to TB, a single dose of BCG vaccine should be given to all infants. 16. In children who are known to be HIV-infected, BCG vaccine should not be given. 17. In infants whose HIV status is unknown and who are born to HIV-positive mothers and who lack symptoms suggestive of HIV, BCG vaccine should be given after considering local factors. Panel considerations on BCG vaccination 2006 Guidance: • “The WHO Expanded Programme on Immunization recommends BCG vaccination as soon as possible after birth in countries with a high TB prevalence. High TB-prevalence countries are those not meeting the criteria for low TB prevalence”. • “A child who has not had routine neonatal BCG immunization and has symptoms of HIV disease/acquired immunodeficiency syndrome should not be given BCG because of the risk of disseminated BCG disease.” Recent evidence: Global Advisory Vaccine Use of BCG vaccine in HIV-infected infants –Weekly Epidemiological Report, January 2010, no.4, 32-33. “The new data do not provide arguments for modifying the current policy recommended by WHO (2007 ) which has also been supported by recent statements from the International Union Against Tuberculosis and Lung Disease and the Stop TB Partnership Childhood TB Subgroup. The operational difficulties in implementing the WHO recommendations were noted, in particular the delayed vaccination approach, which might be possible to implement only in situations where: good TB and HIV surveillance systems for pregnant women and infants existed; where strategies for the prevention of mother-to-child transmission of HIV were 41 operating optimally and were closely linked to well-functioning EPI programmes with good follow-up of all infants; and in situations where HAART coverage for mothers and children is high.” Comments: • Strong evidence for BCG as protective especially for disseminated TB and leprosy. • Recent additional evidence that also protects against infection with TB (?) • A study from Cape Town reported increased risk of BCG disease in HIV infected children, which was often fatal. • A statement was then made by Strategic Advisory Group of Experts (SAGE) that BCG should not be used in HIV infected children. This was not a formal WHO recommendation and predated the guidelines review committee. • Risk of BCG disease low (1% of 1%) and now falling further because of PMTCT and early ART. Question for the panel: Should a formal recommendation be made regarding BCG vaccination in HIV infected children/ should this remain a ‘guidance’ suggestion for case by case decision making? Recommendation: BCG vaccine should not be used in children who are known to be HIV infected Table 12: Studies regarding risk of BCG disease in HIV infected children Reference Study type, location Pt characteristics Synopsis Major outcomes Quality assessment/ Comments BCG disease pre-ART Azzopardi 2009 Systematic review Variation in case-finding and reporting of cases between studies 16 observational studies, 7 case reports and 4 case series → 69 cases of disseminated BCG disease in HIV-infected children 29 of 36 with outcome reported died (81%) Antimycobacterial treatment significantly improved survival (5.9 v 0.18 months, p = 0.05) 49 disseminated BCG disease in observational studies : 47 High quality systematic review Range of rates identified between studies and various limitations and interpretations discussed 42 235 loco-regional disease – 103 HIV- infected (61 BCG IRIS) Most cases with BCG Danish strain HIV infected, 2 had immunodeficiencies (dx required culture/ PCR/ biochemical speciation) - Table 5 below Highest rate in South Africa studies Suggestion rate reduced if on ART Fernandes 2009 Prospective cohort study Brazil n = 207 Gp 1 (n = 141) exposed to vertical transmission; < 15 days Gp 2: HIV infected (n =66) 15 days – 10 years Analysis of disseminated BCG disease Observed from August 2000-Feb 2008 Brazil BCG strain used No cases of regional BCG disease in exposed/ uninfected children No cases of disseminated BCG disease in HIV-infected Local BCG disease in HIV infected children 4.5% (3/66) All occurred in children < 1 year 1 local case assoc w severe immunodeficiency 2 cases part of IRIS Sound methodology but low study numbers – only 66 HIV-infected Hesseling 2009 Prospective surveillance South Africa expanded surveillance for disseminated BCG to three hospitals servicing the Western Included measured (rather than estimated) vertical transmission rates to estimate the denominator 32 cases of disseminated BCG were reported, with a pooled risk of 992 per 100 000 (95% CI 567–1495). Confirmed cases High quality Prospective surveillance BCG confirmed 43 Cape Province BCG vaccine coverage was based on census and health survey data and taken as 98–99%. Highest rate recorded Mansoor 2009 Immunogenicity study South Africa 20 HIV infected 25 HIV exposed, uninfected 23 HIV unexposed Recruited between 2003- 2006 Assessed Cd4 and Cd8 responses in HIV infected, exposed and control infants ART not available for participants Lower T cell response noted in HIV infected children: In all groups, T cell response peaked at 3 months, then waned over 9 month follow up (suggesting central memory Total CD4T cell response significantly lower in HIV + infants Difficult to interpret data for this purpose In vitro immunogenicity does not necessarily correlate with protective efficacy BCG-IRIS Fernandes 2010 Comment based on series below Summary of previous study with comment Low CD4 count appeared to be an independent risk factor for adverse reaction Authors argue vaccination should not necessarily be contraindicated Nuttall 2008 Retrospective case series Children started on ART Median age of start was 5 months 21 of 352 (6%) developed complications – all but one was ipsilateral axillary adenopathy Median of 34 days Risk factors: young and high viral load BCG confirmed by culture in majority Retrospective Majority confirmed First major series of BCG-IRIS 44 after HAART commenced References 1. Global Advisory Vaccine Use of BCG vaccine in HIV-infected infants –Weekly Epidemiological Report, January 2010, no.4, 32-33. 2. Azzopardi P, Bennett CM, Graham SM, Duke T. Bacille Calmette-Guerin vaccine-related disease in HIV infected children: a systematic review. International Journal of Tuberculous Lung disease 13 (11):1331-1344. 3. Fernandes RC, de Araujo LC, Medina-Acosta E. Reduced rate of adverse reactions to the BCG vaccine in children exposed to the vertical transmission of HIV infection and in HIV-infected children from an endemic setting in Brazil. Eur J Pediatr 2009;168:691–6. 4. Hesseling A C, Johnson L F, Jaspan H, et al. Disseminated bacilli Calmette–Guérin disease in HIV-infected South African infants. Bull World Health Organ 2009; 87: 505–511. 5. Mansoor N, Scriba T, de Kock M, Tameris M, Abel B, Keyser A, Little F, Soares A, Gelderbloem S, Mlenjeni S, DenationL, Hawkridge A, Boom WH, Kaplan G, Hussey G, Hanekom W. HIV-1 infection in infants severely impairs the immune response induced by Bacille Calmette-Guerin Vaccine. Journal of Infectious Diseases 2009: 199-298. 6. Fernandes RC, Medina-Acosta E. Complications of BCG in children treated with HAART. Int J Infect Dis 2010. 7. Nuttall JJ et al. Bacillus Calmette–Guérin vaccine-induced complications in children treated with highly-active antiretroviral therapy. International Journal of Infectious Diseases 2008; 12: e99-105. Additional references: benefit of BCG vaccination Table 13: Studies regarding the benefit of BCG vaccination Reference Study type Location Patient Synopsis Major outcomes Trunz 2006 Numerical evaluation of NA Numerical evaluation of efficacy of BCG vaccination in preventing tuberculous meningitis, milary tuberculosis in 2002 100.5 million BCG vaccinations 29729 cases (1/ 3435) of tuberculous meningitis prevented 45 Does not include protective affect against pulmonary/ extrapulmonary/ leprosy (pre HIV) 11486 cases (1/9314) of milary tuberculosis Worldwide cost is < US$200 per year of healthy life gained ( less than average annual income) Most cost effective in SE Asia, Africa, Western pacific region Rodrigues 2007 Case control 226 cases, 857 controls BCG vaccination compared in cases v control at different ages Pre-HIV Protection strong up to 30, persisting protection up to 40 for leprosy < 30: 86% 30-40: 54% >40: 32% Colditz GA 1995 Meta-analysis Systematic review → relative risk/odds ratio for TB in vaccinated v unvaccinated infants Random effects model → indicated that better studies = greater efficacy, Pre-HIV RCT: Overall protective effect: 0.74, .62- 0.83 Case controls: 0.52, 0.38-0.64 Meningitis (5 studies): .64, .30-0.82 Disseminated TB(3 studies) 0.78, 0.58-0.88 References 1. Trunz B, Fine P, Dye C. Effect of BCG vaccination on childhood tuberculous meningitis and military tuberculosis worldwide: a meta-analysis and assessment of cost-effectiveness. Lancet 2006: 367: 1173-1180. 2. Rodrigues LC, Kerr-Pontes LRS, Frietas MVC Barreto ML. Long lasting BCG protection against leprosy. Vaccine 25 (2007): 6842-6844. 3. Colditz GA, et al. The efficacy of bacillus Calmette-Guérin vaccination of newborns and infants in the prevention of tuberculosis: meta-analyses of the published literature. Pediatrics. 1995 Jul;96(1 Pt 1):29-35. 46 Other key documents: • World Health Organization. Safety of BCG vaccine in HIV-infected children. Weekly epidemiological record 19 January 2007, 3 82 17-24. • International Union against Tuberculosis and Lung Diseases BCG Working Group: Consenus IUALTD Statement on the Revised World Health Organization Recommendations Regarding BCG Vaccination in HIV-infected infants. 4 July 2008. From Azzopardi P et al (2009), the Panel looked at the following table on the Rates of disseminated BCG in studies with well-defined denominator populations and microbiologically confirmed BCG disease. 47 48 Panel considerations on contact Screening and Management Box 4: Recommendations regarding prevention of TB in children: contact screening and management Contact screening and management 18. Clinical evaluation of household and close contacts for active TB should be done on the basis of their risk for having or developing active TB or for the potential consequences of the disease if it develops. Priority should be given to contacts who are: − children with symptoms suggestive of TB; − children <5 years of age; − children with known or suspected immunocompromising conditions (especially those living with HIV); and − child contacts of index cases with multidrug-resistant or extensively drug-resistant TB (proven or suspected) 19. It is recommended that contact investigation be conducted for household and close contacts when the index case has any of the following characteristics: − has sputum smear-positive pulmonary TB; − has multidrug-resistant or extensively drug-resistant TB (proven or suspected); − is a person living with HIV; or − is a child <5 years of age 20. Contact investigation may be conducted for household and close contacts of all other index cases with pulmonary TB, in addition to the index cases covered in Recommendation 19. Isoniazid preventive therapy (IPT) 21. Children <5 years of age who are household or close contacts of people with TB and who, after an appropriate clinical evaluation, are found not to have active TB should be given 6 months of IPT (10 mg/kg per day, range 7–15 mg/kg, maximum dose 300 mg/day). 22. In settings of high HIV prevalence, all household and close contacts of people with TB should be counselled and tested for HIV. 23. In settings of low HIV prevalence, all household members and close contacts of people with TB who have symptoms compatible with TB disease may be offered counselling and testing for HIV as part of their clinical evaluation. 49 24. All household contacts of an index case who is a person living with HIV should be counselled and tested for HIV. 2006 Guidance: It is recommended that all NTPs screen household contacts for symptoms of disease and offer isoniazid preventive therapy (i.e. daily isoniazid for at least 6 months) to children aged less than 5 years and all HIV-infected children who are household contacts. Some programmes screen and provide isoniazid preventive therapy to all children and adults who are household contacts. Recent recommendations: Recommendations for the investigation of contacts of persons with infectious tuberculosis in low and middle income countries, Geneva, World Health Organization, 2012. Comments: • Guidelines to be published on contact screening in 2012 essentially do not differ from these previous 2006 recommendations. • The main development of the 2006 guidance was the introduction of symptom based screening approach (Figure 1) which replaced previous recommendations that included the use of TST and CXR as part of screening. Table 14: Studies evaluating symptom based screening Reference Study type, location Pt characteristics Synopsis Major outcomes Quality assessment/ comments Gupta 2011 Retrospective analysis + prospective study ( 1 year post partum) India 799 HIV infected women Studied between 2002-2008 Evaluated symptom screening (cough, fever, night sweats, weight loss) at 1 week before, 2 weeks after delivery TST (> 5 mm +ve), chest radiograph (if TST positive) 94% screening rate 778 had TST, 21.1% positive 11 cases (1.4%) active TB identified (5 smear negative, culture positive, 4 smear positive, 2 clinical/radiologic TB) Original and high quality data but not children Pregnant women 50 Sputum sample taken if active TB suspected Sx based screening Negative predictive value : 99.3% Specificity: 90.9% Kruk 2008 Prospective observational South Africa n = 252 < 5 years old, in household contact with adult contacts with pulmonary TB on sputum smear/ culture Jan 2004-December 2004 Children screened with symptoms, TST, CXR 176 asymptomatic 33 (13.1%) received TB treatment 8/33 (24.2%) treated for TB with uncomplicated hilar adenopathy on CXR Negative predictive value: 95.6% (all children treated for tb) 97.2% (radiologically certain TB) 100% if excluded aymptomatic hilar adenopathy Small numbers Not microbiologically confirmed Difficulty with interpretation of CXR Suggests a safe approach Marais 2006 Prospective study South Africa N – 428 Children < 13 years Feb 2003- Jan 2005 (overlaps with below) Documentation of symptoms Uses a cluster of symptoms for “strict” symptom criteria 197 PTB (96 bacteriological, 75 radiological, 26 probable) 3 symptoms (cough > 2/52 +FTT + fatigue) in HIV UNINFECTED - Sens 62.6% - Spec 90.2% - PPV: 82.3% Not evaluating symptom based screening specifically 51 Poor diagnostic value in HIV infected children References 1. Gupta A, Chandrasekhar A, Gupte N, et al. Symptom screening among HIV-infected pregnant women is acceptable and has high negative predictive value for active tuberculosis. Clin Infect Dis 2011: 53; 1015-1018. 2. Kruk, A, Gie, R, Schaaf HS, Marais, BJ. Symptom-based screening of child tuberculosis contacts: improved feasibility in resource-limited settings. Pediatrics 2008; 121 e1646-1652. 3. Marais BJ, Gie RP, Hesseling AC, Schaaf HS, Lombard C, Enarson DA, Beyers N. A refined symptom-based approach to diagnose pulmonary tuberculosis in children. Pediatrics. 2006 Nov;118(5):e1350-9. 52 Chapter 6: Management of TB in children living with HIV Box 5: Recommendations on the management of TB in children living with HIV 25. Children living with HIV who are more than 12 months of age and who are unlikely to have TB disease on symptom-based screening and who have no contact with a TB case: • should be offered 6 months of IPT (10 mg/kg per day, range 7–15 mg/kg, maximum dose 300 mg/day) as part of a comprehensive package of HIV prevention and care services if living in settings with a high TB prevalence • might be offered 6 months of IPT (10 mg/kg per day, range 7–15 mg/kg, maximum dose 300 mg/day) as part of a comprehensive package of HIV prevention and care services if living in settings with a medium or low TB prevalence 26. Children with suspected or confirmed pulmonary TB or tuberculous peripheral lymphadenitis living in settings with a high HIV prevalence (or with confirmed HIV infection) should not be treated with intermittent regimens (that is, twice-weekly or thrice-weekly doses) 2006 Guidance: “Most current international guidelines recommend that TB in HIV-infected children should be treated with a 6-month regimen as in HIV-uninfected children.” However, no specific treatment recommendation for HIV infected children/ HIV endemic areas was made. Six month treatment regimens containing Rifampicin in the continuation phase were recommended for all new cases of TB (including TB meningitis and osteoarticular disease) or duration of 8 month treatment including ethambutol in the continuation phase. Anti-retroviral treatment: Because recommendations on combinations of anti-TB drugs and antiretroviral drugs are frequently revised, it is advisable to obtain the most recent information from the WHO web site (http://www.who.int/hiv/mediacentre). The clinical and immunological condition of the HIV-infected child should guide the decision whether to: o start ART treatment soon (2–8 weeks) after the start of anti-TB treatment; o delay ART until after completion of the initial phase of anti-TB treatment; o delay start of ART until anti-TB treatment is completed. Daily cotrimoxazole prophylaxis (20 mg trimethoprim (TMP) + 100 mg sulfamethoxazole (SMX) if under 6 months of age; 40 mg TMP + 200 mg SMX if aged under 5 years; 80 mg TMP + 400mg SMX if 5 years or older) prolongs survival in HIV-infected children and reduces the incidence of respiratory infections and hospitalization. 53 Recent recommendations: WHO policy on collaborative TB/HIV activities. Guidelines for national programmes and other stakeholders, 2012. World Health Organization, Geneva, 2012. Guidelines on co-trimoxazole prophylaxis for HIV-related infections among children, adolescents and adults. Recommendations for a public health approach. World Health Organization, Geneva, 2006. Guidelines for intensified case-finding for tuberculosis and isoniazid preventive therapy for people living with HIV in resource- constrained settings. World Health Organization, Geneva, 2011. Comments: A number of new recommendations have been published relating to TB/HIV where the recommendations have been informed largely by data from studies in adults and clinical / epidemiological challenges for TB/HIV in adults This would include: • the addition of ethambutol for all children in an HIV-endemic setting. Given that most children have paucibacilary disease the routine use of a fourth drug would seem less necessary and yet it is important that there is consistency in treatment regimens from a public health perspective. • the use of IPT for children > 12 months who are NOT contacts of a TB case. Data listed below are contradictory and there are different epidemiological issues that need consideration. • There is experience with early commencement of ARTs improving outcome for TB in those living with HIV but no data that critically compares different starting approaches in children. Panel considerations on 6 months of IPT for children OVER 12 months Source: WHO guidelines for intensified tuberculosis case-finding and isoniazid preventive therapy for people living with HIV in resource constrained settings, 2011. Summary of findings and quality of evidence: preventive therapy for children living with HIV Should INH prophylaxis (six months) versus placebo be used in HIV-infected children (PPD-positive or TB-exposed)? 54 Table 15: Studies regarding the use of IPT in children with HIV infection Reference Study type, location Pt characteristics Synopsis Major outcomes Quality assessment/comments Frigati 2011 Retrospective Cohort analysis of RCT below (Zar et al) SOUTH AFRICA 298 117 placebo, 181 INH HIV infected Older than 8 weeks Jan 2003-May 2004 RCT and then all received INH with increasing numbers started on ART until 2007 INH 10 mg/kg/ day and cotrimoxazole / placebo and cotrimoxazole daily or three times a week ART: NRTI x2 + protease inhibitor ; 12 months after trial commenced all placed on ART with national program Trial from Jan 2003- Dec 2007 Median follow-up: 20 months Follow-up 208 pyo INH alone 233 pyo INH plus ART 13 placebo and ART 42 placebo, no ART INH alone: reduced risk of TB by 0.22, 0.09-0.53 ART alone: reduced risk of TB by 0.35 (0.07-1.15) INH+ART reduced risk of TB by 0.11 (0.04-0.32) cf children receiving neither High proportion with severe disease clinical and by CD4 count Different epidemiological scenario TB incidence ART use So are data generalizable? Low pyo placebo Not randomised Not known if groups had similar risks at baseline Madhi 2012 RCT South Africa N =1352; 548 HIV infected, 805 uninfected recruited at infancy All had BCG at birth INH 10-20 mg/kg per day or placebo for 96 weeks Dec 2004-June 2008 Johannesburg, Cape Town and Durban Development of TB/death in HIV infected children: INH v no INH : 19.0% v 19.3%, p = 0.93 HIV uninfected children: High quality RCT Large numbers ART initiated in 99% 55 development of TB disease/ death: 10% v 11%, p = 0.44 No significant difference in TB related outcomes with IPT TB rates higher in HIV infected than uninfected: 121 v 41 per 100,000 Zar 2006 Prospective RCT South Africa N = 263 Children > 8 weeks HIV infected, > 2.5 kg Plan was for 2 years duration of intervention with regular follow-up INH 10 mg/kg/ day and cotrimoxazole / placebo and cotrimoxazole daily or three times a week Jan 2003-May 2004 No diff in numbers receiving HAART Study stopped early DSMB – so small numbers132 in each group 29% < 12 months (not in range for current recommendations) 9% receiving ART Mortality lower in INH group: 11 (8%) v 21 (16%) Hazard ratio (0.46, 0.22-0.95) Similar reduction regardless of daily/ three times a week therapy 5/132 v 13/131 (HR 0.28, 0.10-0.78, p = 0.005) Causes of death: sepsis, pneumonia, gastroenteritis, unknown Total child months of High quality Placebo-controlled Groups similar Exclusion criteria based on safety BUT Different epidemiological scenario TB incidence ART use So are data generalizable? IPT protective against TB (small numbers) BUT this does not clearly 56 9% TST positive at enrolment 16% prior TB treatment 88% CDC B or C – symptomatic – these are sick children intent to treat follow- up in the two groups for the incidence of TB 667 months placebo 839 months in INH group. OR 7.2 cases of TB annually per 100 children in the isoniazid group compared with 23.4 cases in the placebo group IRR 0.31 (0.09 to 0.91). The protective effect of occurred in all categories of severity of clinical disease in children aged > 1 year and in both dose regimens. All five cases of culture confirmed TB in the placebo group. explain the reduction in mortality for those on INH “results support IPT for children that cannot access HAART” No cost effectiveness analysis No info on long term protection References 1. Frigati LJ. Kranzer K. Cotton MF. Schaaf HS. Lombard CJ. Zar HJ. The impact of isoniazid preventive therapy and antiretroviral therapy on tuberculosis in children infected with HIV in a high tuberculosis incidence setting. Thorax. 66(6):496- 501, 2011 Jun. 57 2. Madhi SA. Nachman S. Violari A. Kim S. Cotton MF. Bobat R. Jean-Philippe P. McSherry G. Mitchell C. P1041 Study Team. Primary isoniazid prophylaxis against tuberculosis in HIV-exposed children. New England Journal of Medicine. 365(1):21-31, 2011 Jul 7. 3. Zar HJ, Cotton, MF, Strauss, S, Karpakis J, Hussey, G, Schaaf, S. Rabie, H., Lombard,C. Effect of isoniazid prophylaxis on mortality and incidence of tuberculosis in children with HIV: randomised controlled trial. British Journal of Medicine 2006 Nov 3. Panel considerations: The Panel considered recent evidence that was not published at the time of the development of this recommendation (although it was taken into account in the unpublished format) and noted that a large, prospective, randomized controlled trial (Madhi et al., 2012) did not show any benefit of primary prophylaxis. Review of the original trial (Zar et al., 2006), which did show benefit, found that it represented an epidemiological context that may not be widely representative (coming from one of the highest TB burden settings). In addition, the Panel noted that the epidemiological context of that trial predated the introduction of ART in children living with HIV, and is therefore not representative of the situation in 2013. Finally, that study was stopped early, following interim analysis. There was also an acknowledged difficulty in interpreting that study: the reduced mortality reported was not a reduction in deaths attributed primarily to TB. A recent publication (Frigati et al., 2011) that was reviewed was an observational cohort study of the survivors of the original trial, all started on IPT and with ART introduced over time. IPT was reported as providing benefit additional to that of ART in protecting against TB, but the quality of evidence was considered low because of study design and other confounders changing over time. Note: The above findings were discussed in a subsequent conference call between the Panel of this guidance and the Panel of the 2011 Guidelines for intensified case-finding for tuberculosis and isoniazid preventive therapy for people living with HIV in resource- constrained settings. During the conference call with the two panels, it was agreed to split the recommendation for two different settings (settings with a high TB prevalence and settings with a medium or low TB prevalence). In high TB prevalence settings, 6 months of IPT may have additional benefits to that of ART in protecting against TB. However, in settings with a medium to low prevalence of TB, IPT might be offered considering resource implications. The Panel recommended further research in this area. In addition, it was agreed to change the quality of evidence was changed to low because the evidence is conflicting between the two prospective randomized trials. 58 Panel considerations on the use of antiretroviral therapy (ART) Table 16: Studies regarding use of ART in TB/HIV infected children Reference Study type, Location Pt Characteristics Synopsis Major outcomes Comments Jensen 2012 Retrospective study Spain 1307 HIV infected children, 5228 HIV uninfected children < 17 years of age treated at public hospitals Data based study, identified from hospital ; divided into 3 calendar periods of HAART therapy A: 97-99 B: 2000-02 C:03-08 No direct correlation in clinical outcomes Rates of TB decreased with increased use of HAART • A – B 3.53 → 0.84/1000 HIV infx child, p = 0.016 • A-C 3.36-0.32/1000 HIV infx child, p = 0.002 Reports epidemiological trends; cannot account for patients not hospitalised, individual treatment regimens NO clear explanation for segregation of years (other than early period HAART, middle, late) Wiseman CA 2011 Retrospective study South Africa N = 245 (culture confirmed TB): 52HIV infected 71 unknown HIV 122 HIV negative Surveillance data collected from infants diagnosed with culture confirmed TB Jan 2004- Dec 2006, followed up to Jan 2008 37/52 HIV infected infants commenced on ART: - 4 on ART at time of diagnosis - 4 commenced within 2 weeks of anti TB Rx - 11 b/w 2wks-3 months of treatment - 3 between 3-6 Similar survival regardless of time commencement of ART may support new recommendations. No clear time periods stated 59 months - 14 > 6 months - 1 not recorded 17 (32.7%) died - 10 (19.2%) died before ART initiated - HIV disease stage associated with death (p = 0.004) - NO difference b/w survival in ART started before TB diagnosis/ during treatment (p = 0.449) - Younger age at ART initiation assoc. with ↑ mortality – (median age 7 v 14 months, p = 0.0273) Bakeera 2011 Retrospective cohort Uganda N = 1806 HIV infected children, 311 diagnosed with TB < 18 years of age Children seen between Jan2003- Study of HIV infected children, data on children on ART and anti TB treatment reviewed TB dx – not standardized (clinical/ culture) Of those diagnosed with TB - Advanced HIV more common (CD4<15% in 90% - 171 (9.5%) diagnosed before ART - 140 (8.6%) diagnosed after Children with TB/ HIV NOT on ART would not be captured 60 July 2006, F /U til Jan 2008 Children seen from Jan 2006 were screened for TB ART commenced Once CD4 controlled for, ↓ risk of TB with age (each year 6.7% reduction in risk, 95% CI 1.9-9.3, p = 0.003) In those developing TB post ART, rate of dx highest in first 100 days After TB screening initiated in 2006, proportion of newly dx TB with ART commencement decreased by 70% (95CI75-82, p < 0.001) Marais 2011 Review ART reduces TB rates, reduce risk of TB (48-51): Summary article. Cites key references that have shown general reduction of TB rates with ART therapy (not related to onset of ART treatment) Karim 2010 Open label RCT South Africa N = 642 (adults) ➢ 18 years of age Patients with TB and HIV assigned to : A:ART within 4 weeks of TB therapy Of 642 patients, 338 still in active follow up - 52 died - 134 completed - 56 withdrawn 61 Recruited from 2005-2008 B: ART within 4 weeks post intensive C: ART after TB therapy TB – AFB confirmed - 62 loss to follow up (4 months without a visit) – 35 known to be alive A, B: 429 , C: 213 Reduction in rate of death: A +B: 12.1/100 person years →C: 5.4/100 py RR of 56%, HR 0.44 95CI 0.25-0.79, p = 0.003 Mortality lower in combined groups in all CD4 counts Walters 2008 Retrospective review N = 290 All children initiated on HAART from Jan 2003-December 2005 Data based study: recruitment from pharmacy database and HAART register and HIV clinic. Info collected from medical records TB : Bacteria/ radiological/ clinical, previous (9 months pre-HAART) Treated with standard 48% received TB treatment 137 TB episodes in 136 children (140, 4 no data) - 116 prior to HAART - 21 after HAART (10 likely due to IRIS) Per 100 patient years: - 53.3 in 9 months prior to HAART - 6.4 post HAART - OR: 16.6, 95% CI 12.5-22.4 Management often not standardised/ guideline 62 (3 drug)/ expanded regimen TB outcomes – positive i: - 60% Tb pre- HAART - 78% < 6 months post HAART - 86% > 6 months post HAART TB mortality: HAART v HAART after diagnosis - 4.8% v 12.9% mortality Akslip 2007 Prospective cohort study Thailand N = 329, 225 (10%) known HIV, 104/680 who underwent testing found to be HIV infected Adult patients Seen from Feb 2003-January 2004 Treatment according to clinicians judgement Primary outcome: benefit of antiretroviral therapy before or during TB treatment (not stratified into time of commencement) 290 had final outcome data : For those on ART: - 5/71 (7% died v 94/219 (43%) who did not receive ART, RR 0.2, 95%CI 0.1-0.4 As above, no clear segregation into how time of commencement affects outcome Cotton 2004 Review South AFrica Apart from duration of treatment, principles are the same Cites Studies showing that ART restores anti- At time of this article, HAART was delayed until TB was treated 63 tuberculosis immunological activity References 1. Jensen J. Alvaro-Meca A. Micheloud D. Diaz A. Resino S. Reduction in mycobacterial disease among HIV-infected children in the highly active antiretroviral therapy era (1997-2008). Pediatric Infectious Disease Journal. 31(3):278-83, 2012 Mar. 2. Wiseman CA. Schaaf HS. Cotton MF. Gie RP. Jennings T. Whitelaw A. Roux P. Hesseling AC. Bacteriologically confirmed tuberculosis in HIV-infected infants: disease spectrum and survival. International Journal of Tuberculosis & Lung Disease. 15(6):770-5, 2011 Jun 3. Bakeera-Kitaka S. Conesa-Botella A. Dhabangi A. Maganda A. Kekitiinwa A. Colebunders R. Boulware DR. Tuberculosis in human immunodeficiency virus infected Ugandan children starting on antiretroviral therapy. International Journal of Tuberculosis & Lung Disease. 15(8):1082-6, 2011 Aug. 4. Marais BJ. Rabie H. Cotton MF. TB and HIV in children - advances in prevention and management. [Review] Paediatric Respiratory Reviews. 12(1):39-45, 2011 Mar 5. Karim SSA et al. Timing of initiation of Antiretroviral Drugs during Tuberculosis Therapy. NEJM 2010 362;8, 697-706 6. Martinson NA et al. HAART and risk of tuberculosis in HIV infected South African children: a multi-site retrospective cohort. International Journal of Tuberculous Lung Disease 2009. 13 (7):862-867 7. Walters E et al. Clinical presentation and outcome of tuberculosis in human immunodeficiency virus infected children on anti- retroviral therapy. BMC Pediatrics. 2008:1 8. Akksilp S. Karnkawinpong O. Wattanaamornkiat W. Viriyakitja D. Monkongdee P. Sitti W. Rienthong D. Siraprapasiri T. Wells CD. Tappero JW. Varma JK. Antiretroviral therapy during tuberculosis treatment and marked reduction in death rate of HIV- infected patients, Thailand. Emerging Infectious Diseases. 13(7):1001-7, 2007 Jul. 9. Cotton MF. Schaaf HS. Hesseling AC. Madhi SA. HIV and childhood tuberculosis: the way forward. [Review] [65 refs]International Journal of Tuberculosis & Lung Disease. 8(5):675-82, 2004 May. Note: The final edited version of the guidance is coherent with the Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection: recommendations for a public health approach. Geneva, World Health Organization, 2013. 64 Chapter 7: Multi-drug resistant tuberculosis in children Box 6: Recommendation for the management of MDR-TB in children 27. Children with proven or suspected pulmonary TB or tuberculous meningitis caused by multidrug-resistant bacilli can be treated with a fluoroquinolone in the context of a well-functioning MDR-TB control programme and within an appropriate MDR-TB regimen. The decision to treat should be taken by a clinician experienced in managing paediatric TB. 2006 Guidance: Treatment is difficult – specialist referral is advised. Some basic principles of treatment are as follows: • Do not add a drug to a failing regimen. • Treat the child according to the drug susceptibility pattern (and using the treatment history) of the source case’s M. tuberculosis strain if an isolate from the child is not available. • Use at least four drugs certain to be effective. • Use daily treatment only; directly observed therapy is essential. • Counsel the child’s caregiver at every visit, to provide support, advice about adverse events and the importance of compliance and completion of treatment. • Follow-up is essential: clinical, radiological and bacteriological (mycobacterial culture for any child who had bacteriologically confirmed disease at diagnosis). • Treatment duration depends on the extent of the disease, but in most cases will be 12 months or more (or at least 12 months after the last positive culture). • With correct dosing, few long-term adverse events are seen with any of the more toxic second-line drugs in children, including ethionamide and the fluoroquinolones. 65 Comments: • Updated guidelines will include a chapter on MDR-TB (rather than an annex). • In terms of guidance, there are no changes/recommendations specific to children made; the principles remain the same as in 2006, both in regards to those with disease and in those who are contacts. • MDR-TB in children is now the focus of research and we await clinical studies to provide better evidence in children on issues such as duration of regimens and optimal drug choices. • The current evidence would suggest that children with MDR-TB that receive treatment have at least as good outcomes as adults and tolerate the drugs at least as well. Panel considerations with respect to second-line treatment regimens Table 17 Studies on second-line treatment regimens Reference Study type, location Synopsis Major outcomes/ recommendations Schaaf 2011 Review - Resistance often concordant between contacts → “special effort has to be made to obtain specimens for culture and DST in any child TB suspect who report recent close contact with someone diagnosed with drug-resistant TB” - Dx: MODS, NAAT (GenoTYpe) - Rx regimen: as notated in updated guidelines - Duration – remained Ettehad 2012 Review 8 studies, n = 315 < 16 years - Study characteristics: o Origins: Peru, Spain, USA, S. Africa, Latvia (4 high, 4 low incidence) o N: 8-111 o Age range: 25-132 months o Treatment regimens: all individualised, guided by DST. ▪ “high rates of treatment completion” ▪ Duration ranged from 6-34 months - Primary outcome was treatment success (cure/ completion/ death/ loss to follow up) 66 o Pooled success rate: 81.67% (95CI 72.54-90.80) o Higher in studies which most patients had injectable drugs: 87.2%, 95% CI 74.7-99.8 v 62.6%, 45.3-80.8%, p = 0.02 o Effect of other covariates (DST testing, treatment duration, age, HIV status) had non-significant effects - Other outcomes: o 22.9 deaths total (5.9%) o 19 defaults (6.2%) o 182 adverse events (39.1%); inc N/V, hearing loss, psychiatric effects, hypothyroidism References 1. Schaaf HS, Marais BJ. Management of multidrug-resistant tuberculosis in children: a survival guide for paediatricians. Paediatric Respiratory Reviews 12 (2011) 31-38. 2. Ettehad D, Schaaf HS, Seddon JA, Cooke GS, Ford N. Treatment outcomes for children with multidrug-resistant tuberculosis: a systematic review and meta-analysis. Lancet Infect Dis. 2012 Jun;12(6):449-56. Epub 2012 Feb 27. Table 18: Studies on management of MDR-TB Contacts Reference Study type, location Key points Seddon 2012 Review • Assessment must include: establishing presence of disease/ infection (evaluating hx of exposure, TST/ IGRA) and risk of progression (extent of lung disease in source case, sputum +ve/-ve, physical proximity of source case • Preventive treatment in MDR o Few studies have assessed MDR contacts (no RCTs) o WHO guideline suggests Isoniazid should be used in case of susceptibility o Varied recommendations currently o Use of source case susceptibility – limited data o Options: IPE, Ethionamide, Fluroquinolones o No studies to compare: ▪ INH/ RIF resistance = the other agent should be used 67 ▪ Monotherapy with fluoroquinolone ▪ Multiple drugs – standard/ individualised • General recommendation: high dose isoniazid + fluoroquinolone for 6 months References 1. Seddon JA, Godfrey-Faussett P, Hesseling AC, Gie RP, Beyers N, Schaaf HS. Management of children exposed to multidrug-resistant Mycobacterium tuberculosis. Lancet Infect Dis. 2012 Jun;12(6):469-79. Epub 2012 Feb 27. Other key references for MDR-TB • World Health Organization. Guidelines for the programmatic management of drug-resistant tuberculosis. Emergency update 2008. WHO, Geneva, Switzerland, 2008. WHO/HTM/TB/2008.402. 68 WHO/HTM/TB/2014.05 © World Health Organization, 2014 All rights reserved.