Geert Kleinnibbelink

RV Strain in Pulmonary Hypertension 6 151 Echocardiographic assessment All studies reported that strain was calculated from 2D or 3D grey scale apical 4-chamber orientation, whilst one study performed both 2D and 3D-strain imaging. 28 Strain was calculated with a variety of software packages (EchoPAC, GE Medical Systems, n=8; Syngo vector velocity imaging, Siemens, n=2; 2D cardiac performance analysis, TomTec, n=1). 10 out of 11 studies determined a multivariate HR for RVFWS, while 4 out of 11 studies determined the multivariate HR for RVGLS. Half of the studies (6 out of 11) reported the methods applied for image optimization (i.e. adjustment of image sector width, gain and greyscale), while 9 out of 11 studies reported a frame-rate of > 40 frames/s for strain analysis. Combined endpoint Seven studies adopted a combined endpoint of mortality and PH-related events and had a mean follow-up time of 26±17 months. 7, 8, 10, 11, 21, 26, 29 All but one 26 study revealed a significant HR after univariate analysis. After multivariate analysis, four studies revealed a significant HR for mortality and PH-related events 7, 8, 10, 26 , while HR did not achieve statistical significance in three studies. 11, 21, 29 Combining all multivariate HRs in our meta-analysis revealed that a relative reduction of 19% (range -5 to -31%) of RVLS significantly increased the risk (HR: 1.22, 95%CI: 1.07-1.40) for the combined endpoint of mortality and PH-related events ( Figure 2 ). Studies with a relative reduction below 10% of RVLS tended to be insignificant after multivariate analysis while studies with a relative reduction larger than 10% of RVLS did present significantly higher HR’s after multivariate analysis ( Figure 2 ). All-cause mortality Using data from univariate analysis, all five studies revealed a significant increased HR for RVLS in the prediction for future all-cause mortality after a mean follow-up time of 30±9 months. Multivariate analysis revealed that a lower RVLSwas associatedwith a significantly higher HR for all-cause mortality in all studies. 9, 10, 15, 23, 28 Combining all multivariate HRs, our meta-analysis revealed that a relative reduction of 22% (range -10 to -33%) of RVLS was associated with an increased risk (HR: 2.96, 95%CI: 2.00-4.38) for all-cause mortality ( Figure 3 ). No clear relation between a larger relative reduction in % of RVLS and HR was present ( Figure 3 ).