Time-trend treatment effect of cardiac resynchronization therapy with or without defibrillator on mortality: a systematic review and meta-analysis

Figures

• 

  Figure 1. PRISMA flow chart of searching for publications. HR, hazard ratio.

• 

  Figure 2. (A) Risk of all-cause mortality based on hazard ratio in CRT-D vs. CRT-P patients. Forest plot of studies with data on all-cause mortality using hazard ratios. The analysis included 18 studies comparing 36 421 CRT-D patients with 26 473 CRT-P patients. The HR was 0.74 (95% CI: 0.66–0.82). (B) Risk of all-cause mortality based on adjusted hazard ratio in CRT-D vs. CRT-P patients. Forest plot of studies with data on all-cause mortality using adjusted hazard ratios. The analysis included 22 studies comparing 40 434 CRT-D patients with 33 054 CRT-P patients. The aHR was 0.85 (95% CI: 0.76–0.94). (C) Risk of all-cause mortality based on PSM in CRT-D vs. CRT-P patients. Forest plot of studies with data on all-cause mortality using PSM. The analysis included eight studies comparing 13 220 CRT-D patients with 13 220 CRT-P patients. The HR was 0.83 (95% CI: 0.80–0.87). aHR, adjusted hazard ratio; CI, confidence interval; CRT-D, cardiac resynchronization therapy with defibrillator; CRT-P, cardiac resynchronization therapy with pacemaker.

• 

  Figure 3. (A) Risk of mortality from progressions of heart failure. Forest plot of studies with data on heart failure mortality using hazard ratios. The analysis included three studies comparing 2618 CRT-D patients with 2105 CRT-P patients. The HR was 0.59 (95% CI: 0.41–0.85). (B) Risk of mortality from sudden cardiac death. Forest plot of studies with data on sudden cardiac death using hazard ratios. The analysis included five studies comparing 3475 CRT-D patients with 2959 CRT-P patients. The HR was 0.45 (95% CI: 0.32–0.62). (C) Risk of mortality from cardiovascular mortality. Forest plot of studies with data on cardiovascular mortality using hazard ratios. The analysis included four studies comparing 28 452 CRT-D patients with 21 382 CRT-P patients. The HR was 0.68 (95% CI: 0.49–0.94). CI, confidence interval; CRT-D, cardiac resynchronization therapy with defibrillator; CRT-P, cardiac resynchronization therapy with pacemaker; HR, hazard ratio.

• 

  Figure 4. (A) Risk of mortality by aetiology in CRT-D vs. CRT-P patients, Forest plot of studies with data on all-cause mortality by ischaemic and non-ischaemic aetiology using hazard ratios. The ischaemic analysis included five studies comparing 3171 CRT-D patients with 1643 CRT-P patients. The HR was 0.80 (95% CI:0.67–0.94). The non-ischaemic analysis included seven studies comparing 7021 CRT-D patients with 3248 CRT-P patients. The HR was 0.95 (95% CI: 0.79–1.15), but there was no mortality difference between CRT-D and CRT-P patients. (B) Risk of mortality over 75 years in CRT-D vs. CRT-P patients. Forest plot of studies with data on all-cause mortality by age. Only patients above 75 years were included. The analysis included six studies comparing 3623 CRT-D patients with 1788 CRT-P patients. The HR was 1.08 (95% CI: 0.96–1.21) There was no mortality difference between CRT-D and CRT-P patients. CI, confidence interval; CRT-D, cardiac resynchronization therapy with defibrillator; CRT-P, cardiac resynchronization therapy with pacemaker; HR, hazard ratio.

• 

  Figure 5. (A) Time-trend variation of all-cause mortality by device type. To assess the temporal effect on all-cause mortality HRs, we first took each study’s reported timespan (in years) and calculated the midpoint for each time period. These central values were used in a meta-regression. Hazard ratios were slightly decreased, and CRT-D showed a better treatment effect of CRT-D on mortality could be observed over the years. (B) Time-trend variation of ischaemic aetiology. The percentage of ischaemic patients was shown over time. A trend could be observed for a higher prevalence of ischaemic patients among CRT-P candidates. (C) Time-trend variation of reported mean age. A time trend in age was shown, and the mean age of CRT-P patients increased, which was not such pronounced in the CRT-D cohorts. CRT-D, cardiac resynchronization therapy with defibrillator; CRT-P, cardiac resynchronization therapy with pacemaker; HR, hazard ratio.