WHO/HIV/2013.30
© World Health Organization 2013
GRADE tables: What ARV regimen to start (once-daily NNRTI regimens) for adults living with HIV? Author(s): George W. Rutherford, Alicen B. Spaulding Date: 2012-09-28 Question: TDF/FTC/EFV versus ZDV/3TC/EFV for initial treatment of HIV infection Settings: Multiple high-income and low- and middle-income country settings Bibliography: Arribas 2008, Avihingsanon 2010, Campbell 2012, Pozniak 2006.
Quality assessment
No. of patients
Effect Quality Importance
No. of studies
Design
Risk of bias
Inconsistency
Indirectness
Imprecision
Other Relative TDF/FTC/EFV ZDV/3TC/EFV considerations (95% CI)
Absolute
Mortality (96 weeks) 1 randomized no serious no serious trials risk of inconsistency bias1,2 no serious indirectness very serious3 none 2/258 (0.78%) 2/259 (0.77%) RR 1 (0.14 0 fewer per 1000 (from to 7.07) 7 fewer to 47 more) ⊕⊕OO LOW CRITICAL
Mortality (median 184 weeks) 1 randomized no serious no serious trials risk of bias inconsistency no serious indirectness very serious3 none 18/525 (3.4%) 20/517 (3.9%) RR 0.89 (0.47 to 1.66) 4 fewer per 1000 (from 21 fewer to 26 more) ⊕⊕OO LOW CRITICAL
Disease progression (median 184 weeks) 1 randomized no serious no serious trials risk of bias inconsistency no serious indirectness very serious3 none 10/525 (1.9%) 15/517 (2.9%) RR 0.66 10 fewer per 1000 (0.3 to 1.45) (from 20 fewer to 13 more) ⊕⊕OO LOW CRITICAL
Serious adverse events (median 184 weeks) 1 randomized no serious no serious trials risk of bias inconsistency no serious indirectness serious imprecision6 none 116/525 (22.1%) 115/517 (22.2%) RR 0.99 (0.79 to 1.25) 2 fewer per 1000 (from CRITICAL ⊕⊕⊕O 47 fewer to 56 more) MODERATE
Viral response (<400 copies/ml) (144 and 183 weeks)
1 This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS
WHO/HIV/2013.30
© World Health Organization 2013
2
randomized no serious no serious trials risk of inconsistency bias1,2
no serious indirectness
no serious imprecision
none
581/783 (74.2%)
624/776 (80.4%)
RR 1.01 (0.67 to 1.52)
8 more per 1000 (from 265 fewer to 418 more)
⊕⊕⊕⊕ HIGH
CRITICAL
Immune response (48 weeks) (better indicated by higher values) 1 randomized no serious no serious trials risk of inconsistency bias1,4 no serious indirectness very serious5 none 10 6 – Mean difference 72 higher (79.82 lower to 223.82 higher) ⊕⊕OO LOW CRITICAL
Immune response (144 and 183 weeks) (better indicated by higher values) 2 randomized no serious no serious trials risk of inconsistency bias1,2 no serious indirectness no serious imprecision none 783 776 – Mean difference 17.05 higher (27.99 lower to 62.09 higher) ⊕⊕⊕⊕ HIGH CRITICAL
Retention on treatment (96 weeks) 1 randomized no serious no serious trials risk of inconsistency bias1,2 no serious indirectness no serious imprecision none 232/258 (89.9%) 231/259 (89.2%) RR 1.01 (0.95 to 1.07) 9 more per 1000 (from 45 fewer to 62 more) ⊕⊕⊕⊕ HIGH CRITICAL
Retention on treatment (median 184 weeks) 1 randomized no serious no serious trials risk of bias inconsistency no serious indirectness no serious imprecision none 444/526 (84.4%) 418/517 (80.9%) RR 1.04 32 more per 1000 (from (0.99 to 1.1) 8 fewer to 81 more) ⊕⊕⊕⊕ HIGH CRITICAL
Switching (96 weeks) 1 randomized no serious no serious trials risk of inconsistency bias1,2 no serious indirectness very serious4 none 2/232 (0.86%) 2/231 (0.87%) RR 1 (0.14 0 fewer per 1000 (from to 7.01) 7 fewer to 52 more) ⊕⊕OO LOW CRITICAL
1 2
Two studies (three publications) funded by industry. Three out of four studies open-label. 3 Very few events 4 One non-blinded study. 5 Very small sample size. 6 Few events. Author(s): George W. Rutherford, Alicen B. Spaulding Date: 2012-09-28
2 This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS
WHO/HIV/2013.30
© World Health Organization 2013
Question: TDF/FTC/EFV versus d4T/3TC/EFV for initial treatment of HIV infection Settings: Argentina, Brazil, Dominican Republic, France, Germany, Italy, Spain, Switzerland, United Kingdom, United States Bibliography: Gallant 2004
Quality assessment
No. of patients
Effect Quality Importance
No. of studies
Design
Risk of bias
Inconsistency
Indirectness
Imprecision
Other Relative TDF/FTC/EFV d4T/3TC/EFV considerations (95% CI)
Absolute
Mortality (144 weeks) 1 randomized no serious no serious trials risk of inconsistency bias1 no serious indirectness very serious2 none 1/299 (0.33%) 3/303 (0.99%) RR 0.34 7 fewer per 1000 (from (0.04 to 3.23) 10 fewer to 22 more) ⊕⊕OO LOW CRITICAL
Serious adverse events (144 weeks) 1 randomized no serious no serious trials risk of inconsistency bias1 no serious indirectness serious3 none 81/299 (27.1%) 76/301 (25.2%) RR 1.07 18 more per 1000 CRITICAL ⊕⊕⊕O (0.82 to 1.4) (from 45 fewer to 101 MODERATE more)
Viral response (<50 copies/ml) (144 weeks) 1 randomized no serious no serious inconsistency trials risk of bias1 no serious indirectness no serious imprecision none 203/299 (67.9%) 188/303 (62%) RR 1.09 56 more per 1000 (0.97 to 1.23) (from 19 fewer to 143 more) ⊕⊕⊕⊕ HIGH CRITICAL
Immune response (144 weeks) (better indicated by higher values) 1 randomized no serious no serious inconsistency trials risk of bias1 no serious indirectness no serious imprecision none 299 303 – Mean difference 20 lower (43.97 lower to 3.97 higher) ⊕⊕⊕⊕ HIGH CRITICAL
Retention on treatment (144 weeks) 1 randomized no serious no serious trials risk of inconsistency bias1 no serious indirectness no serious imprecision none 217/299 (72.6%) 203/303 (67%) 54 more per 1000 RR 1.08 (0.97 to 1.2) (from 20 fewer to 134 more) ⊕⊕⊕⊕ HIGH CRITICAL
3 This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS
WHO/HIV/2013.30
© World Health Organization 2013
Switching (144 weeks) 1 randomized no serious no serious trials risk of inconsistency bias1 no serious indirectness very serious2 none 3/299 (1%) 0/303 (0%) RR 7.09 (0.37 to 136.74) – ⊕⊕OO LOW CRITICAL
1 2
Study funded by industry. Very few events. 3 Few events.
4 This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS
WHO/HIV/2013.30
© World Health Organization 2013
Author(s): George W. Rutherford, Alicen B. Spaulding Date: 2012-09-28 Question: TDF/FTC/EFV versus TDF/FTC/ATV/r for initial treatment of HIV infection Settings: Hong Kong SAR, Thailand, Australia, Canada, Puerto Rico, United States Bibliography: Daar 2011, Puls 2010
Quality assessment
No. of patients
Effect Quality Importance
No. of studies
Design
Risk of bias
Inconsistency
Indirectness
Imprecision
Other Relative TDF/FTC/EFV TDF/FTC/ATV/r considerations (95% CI)
Absolute
Mortality (48 weeks) 1 randomized no serious no serious trials risk of bias inconsistency no serious indirectness very serious1 none 0/107 (0%) 2/115 (1.7%) RR 0.21 (0.01 to 4.42) 14 fewer per 1000 CRITICAL ⊕⊕⊕O (from 17 fewer to 59 MODERATE more)
Mortality (median 138 weeks) 1 randomized no serious no serious trials risk of inconsistency bias2 no serious indirectness3 very serious1 none 6/465 (1.3%) 6/464 (1.3%) RR 1 (0.32 to 3.07) 0 fewer per 1000 CRITICAL ⊕⊕⊕O (from 9 fewer to 27 MODERATE more)
Serious adverse events (48 weeks) 1 randomized no serious no serious trials risk of bias inconsistency no serious indirectness very serious1 none 15/107 (14%) 15/115 (13%) RR 1.07 (0.55 to 2.09) 9 more per 1000 (from 59 fewer to 142 more) ⊕⊕OO LOW CRITICAL
Serious adverse events (median 138 weeks) 1 randomized no serious no serious inconsistency trials risk of bias2 no serious indirectness3 no serious imprecision none 141/465 (30.3%) 147/464 (31.7%) RR 0.96 (0.79 to 1.16) 13 fewer per 1000 (from 67 fewer to 51 more) ⊕⊕⊕⊕ HIGH CRITICAL
Viral response (<50 copies/ml) (48 weeks)
5 This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS
WHO/HIV/2013.30
© World Health Organization 2013
1
randomized no serious no serious trials risk of bias inconsistency
no serious indirectness
serious4
none
97/115 (84.3%)
93/107 (86.9%)
RR 0.97 (0.87 to 1.08)
26 fewer per 1000 CRITICAL ⊕⊕⊕O (from 113 fewer to 70 MODERATE more)
Immune response (48 weeks) (better indicated by higher values) 1 randomized no serious no serious trials risk of bias inconsistency no serious indirectness serious4 none 115 107 – Mean difference 5 CRITICAL ⊕⊕⊕O lower (44.49 lower to MODERATE 34.49 higher)
Immune response (96 weeks) (better indicated by higher values) 1 randomized no serious no serious trials risk of inconsistency bias2 no serious indirectness3 no serious imprecision none 464 465 – Mean difference 31 lower (50.29 to 11.71 lower) ⊕⊕⊕⊕ HIGH CRITICAL
Adherence (100% at 48 weeks) 1 randomized no serious no serious trials risk of bias inconsistency no serious indirectness serious4 none 110/115 (95.7%) 100/107 (93.5%) RR 1.02 (0.96 to 1.09) 19 more per 1000 CRITICAL ⊕⊕⊕O (from 37 fewer to 84 MODERATE more)
Retention on treatment (48 weeks) 1 randomized no serious no serious trials risk of bias inconsistency no serious indirectness serious4 none 111/115 (96.5%) 104/107 (97.2%) RR 0.99 (0.95 to 1.04) 10 fewer per 1000 CRITICAL ⊕⊕⊕O (from 49 fewer to 39 MODERATE more)
Retention on treatment (median 138 weeks) 1 randomized no serious no serious trials risk of inconsistency bias2 no serious indirectness3 no serious imprecision none 343/464 (73.9%) 342/465 (73.5%) RR 1.01 (0.93 to 1.09) 7 more per 1000 (from 51 fewer to 66 more) ⊕⊕⊕⊕ HIGH CRITICAL
Switching (48 weeks) 1 randomized no serious no serious trials risk of bias inconsistency no serious indirectness very serious1 none 1/115 (0.87%) 0/107 (0%) RR 2.79 (0.12 to 67.83) – ⊕⊕OO LOW CRITICAL
6 This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS
WHO/HIV/2013.30
© World Health Organization 2013
Switching (median 138 weeks) 1 randomized no serious no serious trials risk of inconsistency bias2 no serious indirectness3 very serious1 none 42/464 (9.1%) 33/465 (7.1%) RR 1.28 (0.82 to 1.98) 20 more per 1000 (from 13 fewer to 70 more) ⊕⊕OO LOW CRITICAL
1 2
Very few events One study open-label. 3 One study conducted primarily in high-income country settings. 4 Few events.
7 This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS
WHO/HIV/2013.30
© World Health Organization 2013
Author(s): George W. Rutherford, Alicen B. Spaulding Date: 2012-09-28 Question: Should TDF/FTC/EFV versus ABC/3TC/EFV be used for initial treatment of HIV infection? Settings: United States, Puerto Rico, 13 European countries Bibliography: Daar 2011, Post 2010, Sax 2009
Quality assessment
No. of patients
Effect Quality Importance
No. of studies
Design
Risk of bias
Inconsistency
Indirectness
Imprecision
Other Relative TDF/FTC/EFV ABC/3TC/EFV considerations (95% CI)
Absolute
Mortality (median 138 weeks) 1 randomized no serious no serious trials risk of bias1 inconsistency no serious indirectness2 very serious5 none 6/464 (1.3%) 11/465 (2.4%) RR 0.55 11 fewer per 1000 (0.2 to 1.47) (from 19 fewer to 11 more) ⊕⊕OO LOW CRITICAL
Disease progression (48 weeks) 2 randomized no serious no serious trials risk of inconsistency bias1,4 no serious indirectness2 very serious5 none 22/596 (3.7%) 29/593 (4.9%) RR 0.82 (0.37 to 1.81) 9 fewer per 1000 (from 31 fewer to 40 more) ⊕⊕OO LOW CRITICAL
Serious adverse events (48 weeks) 1 randomized no serious no serious trials risk of inconsistency bias1,4 no serious indirectness2 very serious5 none 38/197 (19.3%) 56/195 (28.7%) RR 0.67 (0.47 to 0.96) 95 fewer per 1000 (from 11 fewer to 152 fewer) ⊕⊕OO LOW CRITICAL
Serious adverse events (median 138 weeks) 1 randomized no serious no serious trials risk of bias1 inconsistency no serious indirectness2 no serious imprecision none 147/464 (31.7%) 187/465 (40.2%) RR 0.79 (0.66 to 0.94) 84 fewer per 1000 (from 24 fewer to 137 fewer) ⊕⊕⊕⊕ HIGH CRITICAL
Viral response (<50 copies/ml) (48 weeks)
8 This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS
WHO/HIV/2013.30
© World Health Organization 2013
1
randomized no serious no serious trials risk of inconsistency bias1,4
no serious indirectness2
serious imprecision2
none
137/197 (69.5%)
114/195 (58.5%)
RR 1.19 (1.02 to 1.38)
111 more per 1000 (from 12 more to 222 more)
⊕⊕⊕⊕ HIGH
CRITICAL
Immune response (48 weeks) (better indicated by higher values) 1 randomized no serious no serious trials risk of inconsistency bias1,4 no serious indirectness2 no serious imprecision none 197 195 – Mean difference 20 lower (49.7 lower to 9.7 higher) ⊕⊕⊕⊕ HIGH CRITICAL
Immune response (median 138 weeks) (better indicated by higher values) 1 randomized no serious no serious trials risk of bias1 inconsistency no serious indirectness2 no serious imprecision none 464 465 – Mean difference 30 lower (49.29 to 10.71 lower) ⊕⊕⊕⊕ HIGH CRITICAL
Retention on treatment (48 weeks) 1 randomized no serious no serious trials risk of inconsistency bias1,4 no serious indirectness2 no serious imprecision none 149/197 (75.6%) 129/195 (66.2%) 93 more per 1000 RR 1.14 (1.01 to 1.3) (from 7 more to 198 more) ⊕⊕⊕⊕ HIGH CRITICAL
Retention on treatment (median 138 weeks) 1 randomized no serious no serious trials risk of bias1 inconsistency no serious indirectness2 no serious imprecision none 343/464 (73.9%) 324/465 (69.7%) RR 1.06 (0.98 to 1.15) 42 more per 1000 (from 14 fewer to 105 more) ⊕⊕⊕⊕ HIGH CRITICAL
Switching (median 138 weeks) 1 randomized no serious no serious trials risk of bias1 inconsistency no serious indirectness2 serious3 none 42/464 (9.1%) 60/465 (12.9%) RR 0.7 (0.48 to 1.02) 39 fewer per 1000 CRITICAL ⊕⊕⊕O (from 67 fewer to 3 MODERATE more)
1 2
All studies open-label. All studies conducted only in high-income countries. 3 Few events and/or wide confidence interval. 4 One study funded by industry. 5 Very few events. Author(s): George W. Rutherford, Alicen B. Spaulding Date: 2012-09-28
9 This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS
WHO/HIV/2013.30
© World Health Organization 2013
Question: TDF/FTC/EFV versus TDF/FTC/rilpivirine for initial treatment of HIV infection Settings: Multiple international clinical sites Bibliography: Molina 2010
Quality assessment
No. of patients
Effect Quality Importance
No. of studies
Design
Risk of bias
Inconsistency
Indirectness
Imprecision
Other Relative TDF/FTC/EFV TDF/FTC/rilpivirine considerations (95% CI)
Absolute
Mortality (48 weeks) 1 randomized no serious no serious trials risk of inconsistency bias1 no serious indirectness very serious2 none 1/348 (0.29%) 0/346 (0%) RR 2.98 (0.12 to 72.97) – ⊕⊕OO CRITICAL LOW
Serious adverse events (48 weeks) 1 randomized no serious no serious trials risk of inconsistency bias1 no serious indirectness very serious2 none 31/348 (8.9%) 23/346 (6.6%) RR 1.34 (0.8 23 more per 1000 ⊕⊕OO CRITICAL to 2.25) (from 13 fewer to 83 LOW more)
Viral response (<50 copies/ml at 48 weeks) 1 randomized no serious no serious trials risk of inconsistency bias1 no serious indirectness no serious imprecision none 285/348 (81.9%) 287/346 (82.9%) RR 0.99 (0.92 to 1.06) 8 fewer per 1000 ⊕⊕⊕⊕ CRITICAL (from 66 fewer to 50 HIGH more)
Immune response (48 weeks) (better indicated by higher values) 1 randomized no serious no serious trials risk of inconsistency bias1 no serious indirectness no serious imprecision none 348 346 – Mean difference 14 ⊕⊕⊕⊕ CRITICAL lower (36.32 lower to HIGH 8.32 higher)
Adherence (95% at 48 weeks) 1 randomized no serious no serious trials risk of inconsistency bias1 no serious indirectness no serious imprecision none 229/348 (65.8%) 236/346 (68.2%) RR 0.96 (0.87 to 1.07) 27 fewer per 1000 ⊕⊕⊕⊕ CRITICAL (from 89 fewer to 48 HIGH more)
10 This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS
WHO/HIV/2013.30
© World Health Organization 2013
Retention on treatment (48 weeks) 1 randomized no serious no serious trials risk of inconsistency bias1 no serious indirectness no serious imprecision none 288/348 (82.8%) 296/346 (85.5%) RR 0.97 (0.91 to 1.03) 26 fewer per 1000 ⊕⊕⊕⊕ CRITICAL (from 77 fewer to 26 HIGH more)
Switching (48 weeks) 1 randomized no serious no serious inconsistency trials risk of bias1 Study funded by industry. Very few events. no serious indirectness very serious2 none 1/348 (0.29%) 0/346 (0%) RR 2.98 (0.12 to 72.97) – ⊕⊕OO CRITICAL LOW
1 2
11 This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS
WHO/HIV/2013.30
© World Health Organization 2013
Author(s): George W. Rutherford, Alicen B. Spaulding Date: 2012-09-28 Question: TDF/3TC/EFV versus TDF/ddI/EFV for initial treatment of HIV infection Settings: Italy Bibliography: Torti 2005
Quality assessment
No. of patients
Effect Quality Importance
No. of studies
Design
Risk of bias Inconsistency
Indirectness
Imprecision
Other Relative TDF/3TC/EFV TDF/ddI/EFV considerations (95% CI)
Absolute
Viral response (<50 copies/ml) (28 weeks) 1 randomized trials no serious no serious risk of bias1 inconsistency no serious indirectness2 very serious3,4 none 10/10 (100%) 6/11 (54.5%) RR 1.76 415 more per 1000 (from ⊕⊕OO CRITICAL (1.03 to 3.01) 16 more to 1000 more) LOW
Immune response (28 weeks) (better indicated by higher values) 1 randomized trials no serious no serious risk of bias1 inconsistency no serious indirectness2 very serious3 none 10 11 – Mean difference 70 higher ⊕⊕OO CRITICAL (58.46 lower to 198.46 LOW higher)
Adherence (100% at 28 weeks) 1 randomized trials no serious no serious risk of bias1 inconsistency no serious indirectness2 very serious3,4 none 10/10 (100%) 11/11 (100%) RR 1 (0.84 to 0 fewer per 1000 (from 160 ⊕⊕OO CRITICAL 1.19) fewer to 190 more) LOW
Retention on treatment (28 weeks) 1 randomized trials no serious no serious risk of bias1 inconsistency no serious indirectness2 very serious3,4 none 10/10 (100%) 11/11 (100%) RR 1 (0.84 to 0 fewer per 1000 (from 160 ⊕⊕OO CRITICAL 1.19) fewer to 190 more) LOW
Switching (28 weeks) 1 randomized trials no serious no serious risk of bias1 inconsistency no serious indirectness2 very serious3,4 none 0/10 (0%) 1/11 (9.1%) RR 0.36 58 fewer per 1000 (from 89 ⊕⊕OO CRITICAL (0.02 to 8.03) fewer to 639 more) LOW
12 This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS
WHO/HIV/2013.30
© World Health Organization 2013
1 2
Study an open-label study. Study conducted in a high-income country. 3 Very small sample size. 4 Very few events Author(s): George W. Rutherford, Alicen B. Spaulding Date: 2012-09-28 Question: TDF/XTC/EFV versus two NNRTIs/EFV for initial treatment of HIV infection Settings: Multiple international clinical sites Bibliography: Arribas 2008, Avihingsanon 2010, Campbell 2012, Daar 2011, Gallant 2004, Gallant 2006, Post 2010, Pozniak 2006, Sax 2009, Torti 2005
Quality assessment
No. of patients
Effect Quality Importance
No. of studies
Design
Risk of bias Inconsistency
Indirectness
Imprecision
Other Two TDF/XTC/EFV considerations NRTIs/EFV
Relative (95% CI)
Absolute
Mortality (96 weeks) 1 randomized no serious trials risk of bias1,2 no serious inconsistency no serious indirectness3 very serious4 none 2/258 (0.78%) 2/259 (0.77%) RR 1 (0.14 0 fewer per 1000 (from 7 ⊕⊕OO CRITICAL to 7.07) fewer to 47 more) LOW
Mortality (median 138–184 weeks) 3 randomized no serious trials risk of bias1,2 no serious inconsistency no serious indirectness3 very serious4 none 25/1288 (1.9%) 34/1285 (2.6%) RR 0.74 (0.44 to 1.24) 7 fewer per 1000 (from 15 ⊕⊕OO CRITICAL fewer to 6 more) LOW
Disease progression (48 weeks) 2 randomized no serious trials risk of bias1,2 no serious inconsistency no serious indirectness3 very serious4 none 22/596 (3.7%) 29/593 (4.9%) RR 0.82 (0.37 to 1.81) 9 fewer per 1000 (from 31 ⊕⊕OO CRITICAL fewer to 40 more) LOW
Disease progression (median 184 weeks) 1 randomized no serious no serious trials risk of bias inconsistency no serious indirectness very serious4 none 10/525 (1.9%) 15/517 (2.9%) RR 0.66 (0.3 10 fewer per 1000 (from ⊕⊕OO CRITICAL to 1.45) 20 fewer to 13 more) LOW
13 This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS
WHO/HIV/2013.30
© World Health Organization 2013
Serious adverse events (48 weeks) 1 randomized no serious trials risk of bias1,2 no serious inconsistency no serious indirectness3 very serious4 none 38/197 (19.3%) 56/195 (28.7%) RR 0.67 (0.47 to 0.96) 95 fewer per 1000 (from ⊕⊕OO CRITICAL 11 fewer to 152 fewer) LOW
Serious adverse events (median 138–184 weeks) 3 randomized no serious trials risk of bias1,2 no serious inconsistency no serious indirectness3 no serious imprecision none 344/1288 (26.7%) 378/1283 (29.5%) RR 0.92 (0.76 to 1.12) 24 fewer per 1000 (from ⊕⊕⊕⊕ CRITICAL 71 fewer to 35 more) HIGH
Viral response (<50 copies/ml) (28 and 48 weeks) 3 randomized no serious trials risk of bias1,2 no serious inconsistency no serious indirectness3 no serious imprecision none 157/217 (72.4%) 125/212 (59%) RR 1.22 130 more per 1000 (from ⊕⊕⊕⊕ CRITICAL (1.07 to 1.4) 41 more to 236 more) HIGH
Viral response (<400 copies/ml) (144 and 183 weeks) 3 randomized no serious no serious trials risk of bias2 inconsistency no serious indirectness3 no serious imprecision none 784/1082 (72.5%) 812/1079 (75.3%) RR 1.03 (0.78 to 1.36) 23 more per 1000 (from ⊕⊕⊕⊕ CRITICAL 166 fewer to 271 more) HIGH
Immune response (28 and 48 weeks) (better indicated by higher values) 3 randomized no serious trials risk of bias1,2 no serious inconsistency no serious indirectness3 no serious imprecision none 217 212 – Mean difference 10.2 higher (53.71 lower to 74.1 higher) ⊕⊕⊕⊕ CRITICAL HIGH
Immune response (median 138–183 weeks) (better indicated by higher values) 4 randomized no serious trials risk of bias1,2 no serious inconsistency no serious indirectness3 no serious imprecision none 1546 1544 – Mean difference 4.28 lower (31.96 lower to 23.39 higher) ⊕⊕⊕⊕ CRITICAL HIGH
Adherence (100% at 28 weeks) 1 randomized no serious no serious trials risk of bias inconsistency no serious indirectness3 very serious4 none 10/10 (100%) 11/11 (100%) RR 1 (0.84 to 1.19) 0 fewer per 1000 (from ⊕⊕OO CRITICAL 160 fewer to 190 more) LOW
14 This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS
WHO/HIV/2013.30
© World Health Organization 2013
Retention on treatment (48 weeks) 1 randomized no serious trials risk of bias1,2 no serious inconsistency no serious indirectness3 no serious imprecision none 149/197 (75.6%) 129/195 (66.2%) RR 1.14 93 more per 1000 (from 7 ⊕⊕⊕⊕ CRITICAL (1.01 to 1.3) more to 198 more) HIGH
Retention on treatment (96 weeks) 1 randomized no serious trials risk of bias1,2 no serious inconsistency no serious indirectness3 no serious imprecision none 232/258 (89.9%) 231/259 (89.2%) RR 1.01 (0.95 to 1.07) 9 more per 1000 (from 45 ⊕⊕⊕⊕ CRITICAL fewer to 62 more) HIGH
Retention on treatment (median 138–184 weeks) 4 randomized no serious trials risk of bias1,2 no serious inconsistency no serious indirectness3 no serious imprecision none 1014/1299 (78.1%) 956/1296 (73.8%) RR 1.05 37 more per 1000 (from 7 ⊕⊕⊕⊕ CRITICAL (1.01 to 1.1) more to 74 more) HIGH
Switching (28 weeks) 1 randomized no serious no serious trials risk of bias inconsistency no serious indirectness3 very serious4,6 none 0/10 (0%) 1/11 (9.1%) RR 0.36 (0.02 to 8.03) 58 fewer per 1000 (from ⊕⊕OO CRITICAL 89 fewer to 639 more) LOW
Switching (96 weeks) 1 randomized no serious trials risk of bias1,2 no serious inconsistency no serious indirectness3 very serious4 none 2/232 (0.86%) 2/231 (0.87%) RR 1 (0.14 0 fewer per 1000 (from 7 ⊕⊕OO CRITICAL to 7.01) fewer to 52 more) LOW
Switching (median 138–144 weeks) 2 randomized no serious trials risk of bias1,2 no serious inconsistency no serious indirectness3 very serious4 none 45/763 (5.9%) 60/768 (7.8%) RR 1.38 (0.17 to 11.1) 30 more per 1000 (from ⊕⊕OO CRITICAL 65 fewer to 789 more) LOW
1 2
Some open-label studies. Some studies funded by industry. 3 Some studies conducted in high-income countries only. 4 Very few events. 5 Few events. 6 Very small sample size.
15 This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS
WHO/HIV/2013.30
© World Health Organization 2013
Author(s): George W. Rutherford, Alicen B. Spaulding Date: 2012-09-28 Question: TDF/XTC/EFV versus TDF/XTC/PI/r for initial treatment of HIV infection Settings: Multiple international clinical sites Bibliography: Daar 2011, Molina 2010, Puls 2010
Quality assessment
No. of patients
Effect Quality Importance
No. of studies
Design
Risk of bias Inconsistency
Indirectness
Imprecision
Other Relative TDF/XTC/EFV TDF/XTC/PI/r considerations (95% CI)
Absolute
Mortality (48 weeks) 2 randomized no serious no serious trials risk of bias1 inconsistency no serious indirectness very serious2 none 1/455 (0.22%) 2/461 (0.43%) RR 0.76 (0.06 to 10.02) 1 fewer per 1000 (from 4 ⊕⊕OO CRITICAL fewer to 39 more) LOW
Mortality (median 138 weeks) 1 randomized no serious trials risk of bias1,3 no serious inconsistency no serious indirectness4 very serious2 none 6/465 (1.3%) 6/464 (1.3%) RR 1 (0.32 to 0 fewer per 1000 (from 9 ⊕⊕OO CRITICAL 3.07) fewer to 27 more) LOW
Serious adverse events (48 weeks) 2 randomized no serious no serious trials risk of bias1 inconsistency no serious indirectness very serious2 none 46/455 (10.1%) 38/461 (8.2%) RR 1.23 19 more per 1000 (from ⊕⊕OO CRITICAL (0.82 to 1.86) 15 fewer to 71 more) LOW
Serious adverse events (median 138 weeks) 1 randomized no serious trials risk of bias1,3 no serious inconsistency no serious indirectness4 no serious imprecision none 141/465 (30.3%) 147/464 (31.7%) RR 0.96 13 fewer per 1000 (from ⊕⊕⊕⊕ CRITICAL (0.79 to 1.16) 67 fewer to 51 more) HIGH
Viral response (<50 copies at 48 weeks) 2 randomized no serious no serious trials risk of bias1 inconsistency no serious indirectness no serious imprecision none 382/463 (82.5%) 380/453 (83.9%) RR 0.98 17 fewer per 1000 (from ⊕⊕⊕⊕ CRITICAL (0.93 to 1.04) 59 fewer to 34 more) HIGH
16 This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS
WHO/HIV/2013.30
© World Health Organization 2013
Immune response (48 weeks) (better indicated by higher values) 2 randomized no serious no serious trials risk of bias1 inconsistency no serious indirectness no serious imprecision none 463 453 – Mean difference 11.82 ⊕⊕⊕⊕ CRITICAL lower (31.25 lower to HIGH 7.61 higher)
Immune response (96 weeks) (better indicated by higher values) 1 randomized no serious trials risk of bias1,3 no serious inconsistency no serious indirectness4 no serious imprecision none 464 465 – Mean difference 31 lower (50.29 to 11.71 lower) ⊕⊕⊕⊕ CRITICAL HIGH
Adherence (95–100% at 48 weeks) 2 randomized no serious no serious trials risk of bias1 inconsistency no serious indirectness no serious imprecision none 339/463 (73.2%) 336/453 (74.2%) RR 1 (0.93 to 0 fewer per 1000 (from ⊕⊕⊕⊕ CRITICAL 1.08) 52 fewer to 59 more) HIGH
Retention on treatment (48 weeks) 2 randomized no serious no serious trials risk of bias1 inconsistency no serious indirectness no serious imprecision none 399/463 (86.2%) 400/453 (88.3%) RR 0.98 18 fewer per 1000 (from ⊕⊕⊕⊕ CRITICAL (0.95 to 1.02) 44 fewer to 18 more) HIGH
Retention on treatment (median 138 weeks) 1 randomized no serious trials risk of bias1,3 no serious inconsistency no serious indirectness4 no serious imprecision none 343/464 (73.9%) 342/465 (73.5%) RR 1.01 7 more per 1000 (from ⊕⊕⊕⊕ CRITICAL (0.93 to 1.09) 51 fewer to 66 more) HIGH
Switching (48 weeks) 2 randomized no serious no serious trials risk of bias1 inconsistency no serious indirectness very serious2 none 2/463 (0.43%) 0/453 (0%) RR 2.89 (0.3 to 27.61) – OOOO CRITICAL LOW
Switching (median 138 weeks) 1 randomized no serious trials risk of bias1,3 Two studies funded by industry. Very few events. no serious inconsistency no serious indirectness4 very serious2 none 42/464 (9.1%) 33/465 (7.1%) RR 1.28 20 more per 1000 (from OOOO CRITICAL (0.82 to 1.98) 13 fewer to 70 more) LOW
1 2
17 This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS
WHO/HIV/2013.30
© World Health Organization 2013
3 4
One study open-label. One study conducted in high-income countries only.
.
18 This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS