Article Figures & Tables
Figures
- Figure 1
Search strategy and trial flow.
- Figure 2
Forest plot for heart failure, large versus small trials. Individual and pooled risk ratios (RRs) with 95% confidence intervals (CIs) for randomized controlled trials (RCTs) with a primary outcome that included cardiovascular outcomes and reported the number of patients in each treatment group that were admitted to hospital for heart failure as an adjudicated primary or secondary outcome, as well as smaller RCTs reporting at least 1 patient with heart failure for which outcomes were not necessarily adjudicated and patients not necessarily admitted to hospital. The pooled RRs with 95% CIs were calculated using random-effects models. Weight refers to the contribution of each study to the overall pooled estimate of treatment effect. Each square and horizontal line denotes the point estimate and 95% CI for each trial's RR. The diamonds signify the pooled RR; the diamond's centre denotes the point estimate and width denotes the 95% CI.
- Figure 3
Forest plot for heart failure by DPP-4 inhibitor. Individual and pooled risk ratios (RRs) with 95% confidence intervals (CIs) for larger and smaller randomized controlled trials (RCTs) by DPP-4 inhibitor. Interaction p values comparing RRs between pairs of subgroups of RCTs using different DPP-4 inhibitors were all nonsignificant. For the most extreme difference between saxagliptin RCTs and sitagliptin RCTs, interaction p = 0.13 (interaction p = 0.07 comparing RR for only SAVOR-TIMI 53 v. TECOS). The pooled RRs with 95% CI were calculated using random-effects models. Interaction p values were calculated using Z tests. Weight refers to the contribution of each study to the overall pooled estimate of treatment effect. Each square and horizontal line denotes the point estimate and 95% CI for each trial's RR. The diamonds signify the pooled RR; the diamond's centre denotes the point estimate and width denotes the 95% CI.
- Figure 4
Forest plot for heart failure requiring hospital admission by previous history of heart failure. Individual and pooled risk ratios (RRs) with 95% confidence intervals (CIs) for the outcome of heart failure requiring hospital admission in subgroups of patients with versus without previous heart failure in SAVOR-TIMI 53 and EXAMINE, the only RCTs that provided these data. Average rates of heart failure requiring hospital admission were 9.9% ([124+ 107 + 63 + 65]/[1056 + 1049 + 771 + 762] = 359/3638) in patients with versus 1.9% ([165 + 121 + 43 + 24]/[7224 + 7163 + 1930 + 1917] = 353/18 234) in patients without a prior history of heart failure. In this analysis, the heterogeneity in the overall analysis (I2 = 43%) is reduced (I2 = 0%) within each subgroup. The pooled RRs with 95% CIs were calculated using random-effects models. Interaction p values were calculated using Z tests. Weight refers to the contribution of each study to the overall pooled estimate of treatment effect. Each square and horizontal line denotes the point estimate and 95% CI for each trial's RR. The diamonds signify the pooled RR; the diamond's centre denotes the point estimate and width denotes the 95% CI.
Tables
First author,
year of
publicationInclusion criteria Trial duration (w) Baseline patient characteristics Patients with heart failure (n/N) Other
anti-
hyperglycemic
agent(s)A1C range
(%)Prim.
outcomeMean age
(yr)Male
(%)Caucasian
(%)BMI
(kg/m2)DM
(yr)Mean A1C
(%)On insulin
(%)DPP4 inhibitor Placebo Trials with primary cardiovascular outcomes Scirica, 2013, 2014
(SAVOR-TIMI 53)4,5 (saxagliptin)- 6.5-12 110 (med) MACE 65 67 75 31 10.3
(med)8.0 41 289/8280 228/8212 White, 2013
(EXAMINE)6,7 (alogliptin)- 6.5-11 (7-11) 76 (med) MACE 61
(med)68 73 29
(med)7.2
(med)8.0 30 106/2701 89/2679 Green, 2015
(TECOS)8,25 (sitagliptin)- 6.5-8 156 (med) MACE 66 71 68 30 9.4
(med)7.2 23 228/7332 229/7339 Trials with primary metabolic outcomes Alogliptin Nauck, 200926 Met 7-10 26 A1C 55 50 77 32 6 7.9 0 1/423 0/104 DeFronzo, 200827 - 7-10 26 A1C 53 53 67 n/r n/r 7.9 0 0/264 0/64 Pratley, 200928 Sulf 7-10 26 A1C 57 52 71 30 7.7 8.1 0 1/401 0/99 Pratley, 200929 Pio 7-10 26 A1C 55 58 74 33 7.6 8.0 0 2/397 0/97 Rosenstock, 200930 Ins ≥ 8 26 A1C 55 41 65 32 12.6 9.3 100 0/261 0/129 Rosenstock, 201031 Pio 30 7.5-11 26 A1C 53 49 80 31 3.2 8.8 0 0/327 0/163 Pratley, 201432 ± Met 7.5-10 26 A1C 54 54 68 31 4.0 8-8.5 0 0/442 0/326 DeFronzo, 201233 Met ± Pio 7.5-10 26 A1C 54 45 71 3 6.2 8.5 0 0/1037 2/516 Mita, 201634 +Var 7-9.3 104 Intimal thickness 65 58 0 25 8.6 7.3 0 0/161 0/161 Linagliptin Gomis, 201135 ±Met 7.5-10 (11) 24 A1C 55 54 67 29 n/r 8.6 0 0/259 0/130 Haak, 2012,36 201337 ±Met 7.5-10 (11) 24 A1C 55 54 67 29 2.5 8.7 0 2/428 0/363 Owens, 201138 +Met
and Sulf7-10 24 A1C 58 47 47 28 > 5 8.1 0 0/792 0/263 Taskinen, 201139 +Met 7-10 24 A1C 57 54 76 30 5 8.1 0 0/523 0/177 Thrasher 201440,41 - 7.5-11 24 A1C 54 54 0 (Black) 33 5-6 8.7 0 0/106 0/120 Barnett, 201342 - ≥ 7
(age ≥ 70)24 A1C 75 68 97 30 10-12 7.8 21 0/162 0/79 Bajaj, 201443 +Met
and Pio7.5-10 24 A1C 54 49 27 28 1-2 8.4 0 0/183 0/89 Del Prato, 201144 - 7-10 24 A1C 56 48 54 29 n/r 8 0 0/336 0/167 McGill, 201345 - 7-10 (CRI) 52 A1C 64 60 74 32 > 5 8.2 82 2/68 1/65 Yki-Jarvinen, 201346 - 7-10 52 A1C 60 52 80 31 > 5 8.3 100 3/631 2/630 ClinicalTrials.gov47 +Pio 7-10.5 84 A1C 57 53 n/r n/r n/r 8.1 0 0/392 1/409 Chen, 201548 - 7-10 24 A1C 54 59 0 (Asian) 25 n/r 8.0 0 0/200 0/99 ClinicalTrials.gov49 +Met 7-10 24 A1C 56 50 0
(Asian)26 n/r 8.0 0 0/205 0/100 DeFronzo, 201550/
Lewin, 201551±Met 7-10.5 24 A1C 55 54 74 31 ~5 8.0 0 0/545 0/551 ClinicalTrials.gov52 +Met 7.5-11 24 A1C 51 62 n/r n/r n/r n/r n/r 0/294 0/289 Wu, 201553 - 7-10 24 A1C 52 59 0
(Chinese)24 0 8.0 0 0/33 0/22 ClinicalTrials.gov54 +Empa
and Met7-10.5 24 A1C 57 57 n/r n/r n/r n/r n/r 0/238 0/240 Saxagliptin Pfützner, 201155/Jadzinsky, 200956 +Met 8-12 76 A1C 52 49 76 30 1.7 9.4 0 1/643 2/328 Barnett, 201357 +Ins ±Met 7.5-11 52 A1C 57 41 78 32 12 8.7 100 2/304 0/151 Chacra, 201158 +Sulf 7.5-10 76 A1C 55 45 59 29 7 8.4 0 1/501 0/267 DeFronzo, 200959/Rosentock, 201360 +Met 7-10 24 (208 for mort) A1C 55 51 82 31 6.5 8.0 0 3/564 2/179 Frederich, 201261 - 7-10 24 A1C 55 46 70 31 1.7 8.0 0 0/291 0/74 Hollander, 201162 +TZD 7-10.5 76 A1C 54 50 55 30 5.2 8.3 0 0/381 1/184 Nowicki, 201163 +var 7-11 (CRI) 52 A1C 67 43 100 31 16.6 8.3 75 1/85 2/85 Pan, 201264 - 7-10 24 A1C 51 55 0
(Asian)26 1 8.1 0 0/284 0/284 Rosenstock, 200965/Rosenstock, 201360 - 7-10 24 (208 for mort) A1C 53 51 85 32 2.6 7.9 0 1/306 0/95 Moses, 201466 +Met and Sulf 7-10 24 A1C 57 60 45 29 n/r 8.3 0 0/129 0/128 Rosenstock, 201567 +Met and dap 8-12 24 A1C 54 49 70 32 7.3 8.9 0 0/179 0/179 Matthaei, 201568 +Met and dap 8-11.5 24 A1C 55 47 88 31 7.7 7.9 0 1/ 53 2/ 62 Sitagliptin Charbonnel, 200669 +Met 7-10 24 A1C 55 57 64 31 4.6 8.0 0 1/464 0/237 Hermansen, 200770 +Sulf ±Met 7.5-10.5 24 A1C 56 53 63 31 8.8 8.3 0 1/222 0/219 Henry, 201471 +Pio 7.5-11 54 A1C 52 56 67 31 3.9 8.9 0 2/691 0/693 Raz, 200872 +Met 8-11 30 A1C 55 46 44 30 7.9 9.2 0 0/96 0/94 Vilsbøll, 201173 +Ins ±Met 7.5-11 24 A1C 58 51 70 31 12 8.7 100 0/322 2/319 Williams-Herman, 201074/ Goldstein, 200775 ±Met 7.5-11 104 A1C 53 49 52 32 4.5 8.8 0 1/372 0/364 Yoon, 2012,76 201177 +Pio 8-12 54 A1C 51 54 52 30 2.3 9.5 0 0/261 0/259 ClinicalTrials.gov78 - 7-10 24 A1C 49 56 0 (India) n/r n/r n/r 0 0/107 0/106 Yang, 201179 +Met 7-10 24 A1C 54 48 0 (Asian) 26 5.1 7.8 0 0/283 0/287 ClinicalTrials.gov80 +Met 6.5-11 24 A1C 55 26 n/r 29 8.4 8.5 0 0/36 0/9 Barzilai, 201181 - 7-10 (age ≥ 65) 24 A1C 72 47 79 31 7.1 7.8 0 0/102 0/104 Yang, 201282 +Met 7.5-11 24 A1C 55 51 0 (Chi-nese) 25 6.9 8.5 0 0/197 0/198 Rosenstock, 200683 +Pio 7-10 24 A1C 56 56 73 31 6.1 8.0 0 0/175 0/178 Aschner, 200684 - 7-10 24 A1C 54 52 51 30 4.4 8.0 0 0/488 0/253 Fonseca, 201385 +Met +Pio 7.5-11 26 A1C 56 62 50 30 9.8 8.7 0 0/157 0/156 Olansky, 201186 +Met ≥ 7.5 44 A1C 50 57 80 33 3.4 9.9 0 1/625 0/621 ClinicalTrials.gov87 +Met 6/6.5-9 44 β-cell function 61 67 62 33 3.0 6.1 0 0/10 0/11 Dobs, 201388 +Met and Rosi 7.5-11 54 A1C 55 58 51 30 9.3 8.8 0 0/170 0/92 Moses, 201689 +Met and Sulf 7.5-10.5 24 A1C 55 46 n/r n/r n/r 8.4 0 0/210 0/212 Lavalle-González, 201390 - 7-10.5 26 A1C 55 48 72 32 6.8 7.9 0 1/366 0/183 Mathieu, 201591 +Glar ±Met 7.5-11 24 A1C 59 48 70 32 13.5 8.7 100 0/329 0/329 Roden, 201392 - 7-10 76 A1C 55 59 34 28 < 5 7.9 0 1/223 0/223 ClinicalTrials.gov93 +Acar n/r 24 A1C 57 51 n/r n/r n/r 8.1 0 0/191 0/189 Skrivanek, 201494 +Met 7-9.5 26 A1C 54 49 51 31 7 8.1 0 0/355 0/177 ClinicalTrials.gov95 +Met n/r 156 A1C 55 47 72 n/r n/r n/r n/r 1/302 0/101 Ji, 201696 ±Met 7(0.5)-10(11) 24 A1C 53 61 0 (Chi-nese) n/r n/r 8.7 0 0/367 0/376 ClinicalTrials.gov97 +Met and Ins 7-10 24 A1C 56 52 49 29 9.1 8.4 100 0/383 0/192 ClinicalTrials.gov98 +Sulf ±Met 7.5-11 24 A1C 57 50 n/r n/r n/r 8.6 0 0/248 0/249 ClinicalTrials.gov99 +Ins ±Met 7.5-11 24 A1C 58 53 0 (Chi-nese) n/r n/r n/r 100 0/234 0/233 ClinicalTrials.gov100 +Met 7.1-11 28 A1C 54 56 81 32 8.2 8.5 0 0/122 0/61 Ishikawa, 2014101 - < 6.5 52 Carotid artery intima thickness 71 85 0 (Jap-anese) 25 n/r 5.6 0 0/37 0/39 Derosa, 2012102-104 +Met > 8.0 52 A1C 55 48 100 29 0.5 8.1 0 0/91 0/87 Derosa, 2014105 +Var > 7.0 104 A1C n/r 49 100 29 n/r 8.1 0 0/102 0/103 Chien, 2011106 +Var ≥ 7.0 24 A1C 73 58 0 (Chi-nese) 26 13.6 9.7 0 0/49 0/48 Weinstock, 2015107 +Met 7/8-9.5 26 A1C 54 49 51 31 7.0 8.1 0 0/315 0/177 Mita, 2016108 +Var ≥ 6.6 104 Intimal thickness 64 60 0 (Jap-anese) 25 17.3 8.1 100 0/137 0/137 ClinicalTrials.gov109 - n/r 24 A1C 60 69 0 (Jap-anese) n/r n/r n/r 0 0/164* 0/82 Vildagliptin Bosi, 2007110 +Met 7.5-11 24 A1C 54 57 74 33 6.3 8.4 0 0/360 0/181 Bosi, 2009111 +Met 7.5-11 24 A1C 53 58 73 31 4.1 8.7 0 0/292 0/292 Fonseca, 2007112 +Ins 7.5-11 24 A1C 59 51 71 33 14.7 8.4 100 0/144 0/152 VIVIDD, 2014113 - 6.5-10 (LVEF < 40) 52 LVEF 63 77 n/r 29 n/r 7.8 34 13/128 10/125 Scherbaum, 2008114,115 - 6.2-7.5 104 A1C 63 59 99 30 2.6 6.7 0 0/156 0/150 Strain, 2013116 - 7-10 (age > 70) 24 A1C 75 45 97 30 11.4 7.9 0 1/139 1/139 Vollmer, 2009117 +Met 6.5-8 24 A1C 61 46 96 n/r n/r 7.2 0 0/274 0/131 Garber, 2007118 +Pio 7.5-11 24 A1C 54 43 69 32 4.7 8.7 0 1/305 1/158 Garber, 2008119 +Sulf 7.5-11 24 A1C 58 59 69 31 7.1 8.5 0 0/339 0/176 Yang, 2015120 +Sulf 7.5-11 24 A1C 59 57 0 (Chi-nese) 25 6.9 8.7 0 0/143 0/136 Foley, 2011121/Bunck, 2012122 - ≤ 7.5 52 β-cell function 57 59 93 30 1.0 6.0 0 0/29 0/30 Pan, 2012123 +Met 7-10 24 A1C 54 47 0 (Chi-nese) 25 5 8.1 0 0/294 0/144 Lukashevich, 2014124 +Met ±Sulf 7.5(8.5)-11 24 A1C 55 48 23 28 7.3 8.8 0 0/157 0/160 Macauley, 2015125 +Met ≤ 7.6 26 Liver TG + Ins Sens 61 64 n/r 30 1.0 6.0 0 0/20 0/19 Ahren, 2004,126 2005127 +Met 7-9.5 52 A1C 57 37 99 30 5.6 7.7 0 0/56 0/51 Goodman, 2009128 +Met 7.5-11 24 A1C 55 58 66 32 n/r 8.6 0 0/248 0/122 Ito, 2011129 - > 7.0 (ESRD) 24 A1C 67 69 0 (Jap-anese) 23 n/r 6.7 0 0/30 0/21 Derosa, 2012130-132 +Met 8.1-10.9 52 A1C+ β-cell func-tion 53 51 100 28 0.5 8.2 0 0/84 0/83 Zografou, 2015133 +Met 7-9 26 Arter. stiffness 54 59 n/r 32 n/r 8.1 0 0/32 0/32 Gemigliptin Yang, 2013134 - 7-11 24 A1C 53 58 0 (Ind/Korean) 26 3 8.3 0 0/87 0/87 Teneligliptin ClinicalTrials.gov135 +Met 7-10 24 A1C 58 56 0 0/359 0/88 Anagliptin Yang, 2015136 - 6.5-10 24 A1C 56 54 0 (Korean) 25 3.6 7.1 0 0/68 1/40 Note: acar = acarbose, BMI = body mass index, CRI = chronic renal insufficiency, dap = dapagliflozin, DPP-4 = dipeptidyl peptidase-4, empa = empagliflozin, ESRD = end-stage renal disease, glar = glargine, Ind = Indian, ins = insulin, LVEF = left ventricular ejection fraction, MACE = major adverse cardiovascular events, med = median, met = metformin, n/N = number of patients with heart failure/total number of patients, n/r = not reported, pio = pioglitazone, rosi = rosiglitazaone, sulf = sulfonylurea, TZD = thiazolidinedione, var = various hypoglycemic agents.
*Randomized patients to 3 groups comparing both sitagliptin and omarigliptin to placebo; 0/166 heart failure events in the omarigliptin group.
- Table 2: Trial and baseline patient comorbidities and medications for RCTs with primary cardiovascular outcomes*
Trial name SAVOR-TIMI 53 EXAMINE TECOS Trial characteristics DPP-4 Inhibitor Saxagliptin Alogliptin Sitagliptin Number of patients 16 492 5380 14 724 Enrolment period October 2009-March 2013 May 2010-December 2011 December 2008-July 2012 Median follow-up (yr) 2.1 1.5 3.0 Main inclusion criteria A1C 6.5%-12.0% 6.5%-11.0%
(7%-11% if on insulin)6.5%-8.0% Clinical Established CV disease or age > 55/60 (male/female)
and one other CV risk factorACS in previous 15-90 d Established CV disease and ≥ 50 years old Patient comorbidities Hypertension, % 81 83 86 Dyslipidemia, % 71 n/r 77 Current smoker, % 13 14 11 Prior MI, % 38 88† 43 Prior PCI, % 27 63† 39 Prior CABG, % 24 13† 25 Heart failure, % 13 28 18 Atrial fibrillation, % 7 n/r 8 (incl. AFlutter) Stroke, % 13 7 17 (+4% TIA) Peripheral arterial disease, % 12 10 17 eGFR (mL/min/1.73 m2) 73
(excluded dialysis)71
(excluded dialysis)75 (excluded < 30) Medications ASA, % 75 91 79 Any anti-platelet, % 81 n/r (80% thienopyridine) n/r (22% clopidogrel/ticlopidine, 7% vit K antagonist) Statin, % 78 90 80 ACE Inhibitor/ARB, % 79 (total) 82 (total) 79 Beta-blocker, % 61 82 64 Other antihypertensive agents, % 41 22 CCB
37 diuretic34 CCB
41 diureticNote: ACE = angiotensin converting enzyme, ACS = acute coronary syndrome, ARB = angiotensin receptor blocker, Aflutter = atrial flutter, ASA = acetylsalicylic acid, CABG = coronary artery bypass grafting, CAD = coronary artery disease, CCB = calcium channel blocker, CV = cardiovascular; eGFR = estimated glomerular filtration rate, MI = myocardial infarction, n = number of patients, n/r = not reported, PCI = percutaneous coronary intervention, RCT = randomized controlled trial, TIA = transient ischemic attack, vit = vitamin.
*The 3 trials used virtually identical heart failure definitions: patients were required to be admitted to hospital or have an emergency department visit of more than 12 hours with clinical manifestations of heart failure, defined as at least 1 of new or worsening dyspnea, orthopnea, paroxysmal nocturnal dyspnea, peripheral edema, bibasilar rales on pulmonary examination, jugular venous distention, new third heart sound, or radiographic evidence of heart failure; and receive at least 1 of intravenous treatment with a diuretic, inotrope, or vasodilator therapy, ultrafiltration or dialysis, or mechanical or surgical intervention (including heart transplant) specifically directed as treatment for their heart failure. The other smaller RCTs did not provide definitions, or specify whether the patients with heart failure required hospital admission or whether this outcome was subject to blinded adjudication.
†Includes index event before enrolment.
First author,
year of publicationTotal no.
of patients randomizedNo.
of centresTrial duration, wk Blinded* AC ITT No early stopping for benefit < 5% loss to follow up (%) Primary cardiovascular outcomes Scirica, 2013
(SAVOR-TIMI 53)4,5 (saxagliptin)16 492 788 110* Yes Yes Yes Yes Yes (0.9) White, 2013
(EXAMINE)6,7 (alogliptin)5380 898 76* Yes Yes Yes Yes Yes (0.9) Green, 2015
(TECOS)8,25 (sitagliptin)14 724 673 156* Yes Yes Yes Yes Yes (0.4) median* Alogliptin DeFronzo, 200827 329 67 26 Yes Yes Yes Yes Yes (0.3) Nauck, 200926 527 115 26 Yes Yes Yes Yes Yes (0) Pratley, 200928 500 124 26 Yes Yes Yes Yes Yes (0) Pratley, 200929 493 125 26 Yes Yes Yes Yes Yes (0) Rosenstock, 200930 390 110 26 Yes Yes Yes Yes Yes (0) Rosenstock, 201031 490 161 26 Yes Yes Yes Yes Yes (0) Pratley, 201432 784 198 26 Yes Yes Yes Yes Yes (2.0) DeFronzo, 201233 1554 327 26 Yes Yes Yes Yes Yes (0.1) Mita, 201634 341 11 104 No Yes Yes Yes No (5.6) Linagliptin Gomis, 201135 389 43 24 Yes Yes Yes Yes Yes (0) Haak, 201236, 201337 791/567 133/112 24/54 Yes Yes Yes Yes Yes (0/0.2) Owens, 201138 1058 100 24 Yes Yes Yes Yes Yes (0.3) Taskinen, 201139 701 82 24 Yes Yes Yes Yes Yes (0.1) Thrasher, 201440,41 226 93 24 Yes Yes Yes Yes Yes (0) Barnett, 201342 241 33 24 Yes Yes Yes Yes Yes (0) Bajaj, 201443 272 52 24 Yes Yes Yes Yes Yes (0) Del Prato, 201144 503 66 24 Yes Yes Yes Yes Yes (0) McGill, 201345 133 53 52 Yes Yes Yes Yes Yes (0) Yki-Jarvinen, 201346 1261 167 52 Yes Yes Yes Yes Yes (0) ClinicalTrials.gov47 936 132 84 Yes Yes Yes Yes Yes (0) Chen, 201548 300 19 24 Yes Yes Yes Yes Yes (0.3) ClinicalTrials.gov49 306 19 24 Yes Yes Yes Yes Yes (0.3) DeFronzo, 201550/Lewin, 201551 1404 211 24 Yes Yes Yes Yes Yes (3.0) ClinicalTrials.gov52 730 56 24 Yes Yes Yes Yes Yes (0) Wu, 201553 57 1 24 Yes Yes Yes Yes Yes (3.5) ClinicalTrials.gov54 708 114 24 Yes Yes Yes Yes n/r Saxagliptin Pfützner, 201155/Jadzinsky, 200956 1306 211 76 Yes Yes Yes Yes Yes (0.2) Barnett, 201357 457 80 52 Yes Yes Yes Yes Yes (0.4) Chacra, 201158 768 115 76 Yes Yes Yes Yes Yes (0) DeFronzo, 200959/Rosenstock, 201360 743 154 24
(208 for mortality)Yes Yes Yes Yes Yes (0) Frederich, 201261 366 72 24 Yes Yes Yes Yes Yes (0.3) Hollander, 201162 565 133 76 Yes Yes Yes Yes Yes (0) Nowicki, 201163 170 75 52 Yes Yes Yes Yes Yes (0) Pan, 201264 568 40 24 Yes Yes Yes Yes Yes (0) Rosenstock, 2009,65 201360 401 135 24 (208 for mortality) Yes Yes Yes Yes Yes (0) Moses, 201466 257 35 24 Yes Yes Yes Yes Yes (0) Rosenstock, 201567 534 139 24 Yes Yes Yes Yes Yes (0) Matthaei, 201568 315 84 24 Yes Yes Yes Yes Yes (0) Sitagliptin Charbonnel, 200669 701 100 24 Yes Yes Yes Yes Yes (0) Hermansen, 200770 441 74 24 Yes Yes Yes Yes Yes (0) Henry, 201471 1615 256 54 Yes Yes Yes Yes Yes (0) Raz, 200872 190 24 30 Yes Yes Yes Yes Yes (0) Vilsbøll, 201173 641 100 24 Yes Yes Yes Yes Yes (0) Wiiliams-Herman, 201074/
Goldstein, 2007751091 140/117 104 Yes Yes Yes Yes Yes (0) Yoon, 2012,76 201177 520 60/28 54 Yes Yes Yes Yes Yes (0) ClinicalTrials.gov78 213 9 24 Yes Yes Yes Yes Yes (0) Yang, 201179 570 40 24 Yes Yes Yes Yes Yes (0) ClinicalTrials.gov80 68 1 24 No n/r n/r n/r n/r Barzilai, 201181 206 52 24 Yes Yes Yes Yes Yes (0) Yang, 201282 395 17 24 Yes Yes Yes Yes Yes (0) Rosenstock, 200683 353 71 24 Yes Yes Yes Yes Yes (0) Aschner, 200684 741 111 24 Yes Yes Yes Yes Yes (0) Fonseca, 201385 313 58 26 Yes Yes Yes Yes Yes (0) Olansky, 201186 1250 209 44 Yes Yes Yes Yes Yes (0.3) ClinicalTrials.gov87 21 1 44 Yes Yes Yes Yes Yes (0) Dobs, 201388 278 41 54 Yes Yes Yes Yes No (5.8) Moses, 201689 427 Multi 24 Yes Yes Yes Yes Yes (1.2) Lavalle-González, 201390 1284 169 26 Yes Yes Yes Yes Yes (0) Mathieu 201591 660 Multi 24 Yes Yes Yes Yes Yes (0.3) Roden, 201392 899 124 76 Yes Yes Yes Yes Yes (1.1) ClinicalTrials.gov93 381 Multi 24 Yes Yes Yes Yes Yes (0) Skrivanek, 201494 1202 99 26 Yes Yes Yes Yes Yes (0) ClinicalTrials.gov95 1049 386 156 Yes Yes Yes Yes Yes (3.5) Ji, 201696 744 Multi 24 Yes Yes Yes Yes Yes (0.1) ClinicalTrials.gov97 582 60 24 Yes Yes Yes Yes Yes (1.2) ClinicalTrials.gov98 498 Multi 24 Yes Yes Yes Yes Yes (0.2) ClinicalTrials.gov99 467 28 24 Yes Yes Yes Yes Yes (0) ClinicalTrials.gov100 366 60 28 Yes Yes Yes Yes Yes (0.3) Ishikawa, 2014101 80 1 52 No Uncl. Yes Yes Yes (3.8) Derosa, 2012102-104 178 1 52 Yes Yes Yes Yes n/r Derosa, 2014105 205 Multi 104 Yes Yes Yes Yes n/r Chien, 2011106 97 1 24 No Uncl. Yes Uncl. Yes (0) Weinstock, 2015107 1098 111 26 Yes Yes Yes Yes Yes (0) Mita, 2016108 282 12 104 No Yes Yes Yes Yes (2.8) ClinicalTrials.gov109 414 Multi 24 Yes Yes Yes Yes Yes (0.5) Vildagliptin Bosi, 2007110 544 106 24 Yes Yes Yes Yes Yes (0.6) Bosi, 2009111 1179 > 250 24 Yes Yes Yes Yes Yes (0.8) Fonseca, 2007112 296 68 24 Yes Yes Yes Yes Yes (0) VIVIDD, 2014113 254 94 52 Yes Yes Yes Yes Yes (0.4) Scherbaum, 2008114,115 306 69 52/104 Yes Yes Yes Yes Yes/No (0/57) Strain, 2013116 278 45 24 Yes Yes Yes Yes Yes (0) Vollmer, 2009117 405 94 24 Yes Yes Yes Yes Yes (0) Garber, 2007118 463 123 24 Yes Yes Yes Yes Yes (0) Garber, 2008119 515 114 24 Yes Yes Yes Yes Yes (0) Yang, 2015120 279 18 24 Yes Yes Yes Yes Yes (0) Foley, 2011121/Bunck, 2012122 59 1 52 Yes Yes Yes Yes Yes (0) Pan, 2012123 438 17 24 Yes Yes Yes Yes Yes (0) Lukashevich, 2014124 318 55 24 Yes Yes Yes Yes Yes (0.3) Macauley, 2015125 44 1 26 Yes Yes Yes Yes No (11.4) Ahren, 2004,126 2005127 107 4 52 Yes Yes Yes Yes Yes (0) Goodman, 2009128 370 67 24 Yes Yes Yes Yes Yes (0) Ito, 2011129 60 1 24 No Yes Yes Yes No (11.7) Derosa, 2012130-132 167 4 52 Yes Yes Yes Yes Yes (0) Zografou, 2015133 64 1 26 No Uncl. Yes Uncl. n/r Gemigliptin Yang, 2013134 182 14 24 Yes Yes Yes Yes Yes (4.2) Teneligliptin ClinicalTrials.gov135 448 45 24 Yes Yes Yes Yes Yes (0.2) Anagliptin Yang, 2015136 109 25 24 Yes Yes Yes Yes Yes (0.9) Note: AC = allocation concealment, ITT = intention-to-treat analysis, multi = multicentre, n/r = not reported, uncl = unclear.
*Patient, caregiver and outcome assessor blinding for heart failure outcome.
Meta-analyses, first author, year Analysis included only RCTs in which randomized groups differ by DPP-4 inhibitor treatment to avoid the confounding effect of other medications Statistically compared HF outcomes between different DPP-4 inhibitors Additional ≥ 24-week follow-up DPP-4 inhibitor v. placebo RCTs/enrolled patients with HF events included in the current meta-analysis Avoidance of inadvertent double counting of some included RCTs Inclusion of all* HF outcomes for EXAMINE Trial7 Inclusion of most recently published HF results for VIVIDD Trial113 Main conclusions Including TECOS Current meta-analysis Yes Yes (reference) Yes Yes Yes •13% increase in HF risk only statistically significant (p = 0.03) if results of smaller RCTs added to large cardiovascular safety RCTs
•differences between agents not statistically significant (interaction p = 0.07-0.12)Li, 2016151 No No 10 RCTs/
5541 patientsYes No Yes •12% increase in HF risks¶ (p = 0.05) pooling HF hospitalization outcomes from 5 RCTs only; no significant increase in HF for the remaining RCTs v. all comparators Abbas, 2016149 Yes† No 29 RCTs/
18 097
patients†Yes No -‡ •non-significant 11% increase in HF risk (p = 0.19) pooling only the 3 large cardiovascular safety RCTs but not including all HF outcomes for EXAMINE Kongwat-charapong, 2016150 No No 4 RCTs/
1639 patientsYes Yes No •non-significant 11% increase in HF risk (p = 0.06) v. all comparators
•highlighted increase in HF for saxagliptin but differences not statistically compared with other agentsPre-TECOS Monami, 2014152 No No 8 RCTs/
17 463 patientsNo
(double counted
NCT0102839176
which was an extension
of NCT0039763177)No -‡ •19% increase in HF odds (p = 0.015) v. all comparators
•highlighted increase in HF for saxagliptin but differences not statistically compared with other agentsWu, 2014154 No§ No 15 RCTs/
19 339
patientsNo
(double counted
2 publications
for NCT0032701555,56)No No •16% increase in HF risk (p = 0.04) v. all comparators and 17% increase (p = 0.03) v. only placebo comparators Savarse, 2015153 No No 9 RCTs/
18 055 patientsNo
(double counted NCT0091577237
which was an extension
of NCT0079816136)No No •16% increase in HF risk (p = 0.03) v. all comparators pooling long-term follow up RCTs but no increase pooling short-term follow up RCTs Note: DPP-4 = dipeptidyl peptidase-4, HF = heart failure, RCT = randomized controlled trial.
*Some of the previously published meta-analyses149,151-154 included only heart failure hospital admissions that were counted in the analysis of the composite end point reported in the abstract of the follow-up publication for EXAMINE focusing on heart failure outcomes,7 instead of all admissions for heart failure reported in the main body of this publication. Including all hospital events results in a higher risk for alogliptin for this RCT (RR 1.18 v. RR 1.07).
†Abbas 2016149 only included the 3 large cardiovascular RCTs (SAVOR-TIMI 53,4,5 EXAMINE6,7 and TECOS8).
‡VIVIDD113 was not included in Monami 2014152and Abbas 2016.149
§For Wu 2014,154 comparison to placebo trials was included as a secondary analysis.
¶Li 2016151 actually reported 13% increase in odds using Peto odds ratios (p = 0.05), which corresponds to a 12% increase in risk using risk ratios (p = 0.05).
In this issue
Article tools
