Left ventricular mechanics assessment in amyloidosis patients: a systematic review and meta-analysis.

BACKGROUND

Over the last decade, a small number of studies have used speckle tracking echocardiography (STE) or cardiac magnetic resonance (CMR) for measuring left ventricular (LV) mechanics in patients with amyloidosis. This systematic review and meta-analysis aimed at assessing the overall influence of amyloidosis on LV global longitudinal strain (GLS) and regional longitudinal strain at basal (BLS), mid (MLS) and apical (ALS) level, respectively.

METHODS

All imaging studies assessing LV-GLS, LV-BLS, LV-MLS and LV-ALS in amyloidosis patients versus healthy controls, selected from PubMed and EMBASE databases, were included. The risk of bias was assessed by using the National Institutes of Health (NIH) Quality Assessment of Case-Control Studies. Continuous data (LV-GLS, LV-BLS, LV-MLS and LV-ALS) were pooled as a standardized mean differences (SMDs) comparing amyloidosis group with healthy controls. The overall SMDs of LV-GLS, LV-BLS, LV-MLS and LV-ALS were calculated using the random-effect model.

RESULTS

The full-texts of 13 studies with 553 amyloidosis patients and 575 healthy controls were analyzed. STE (53.8%) and CMR (46.2%) studies were separately analyzed. Average LV-GLS magnitude was significantly impaired in amyloidosis patients vs. controls in both STE (13.8±3.9 vs. 19.8±2.7%) and CMR (12.3±4 vs. 17.9±3.5%) studies. The impairment of segmental strain detected in amyloidosis patients was prevalent at basal and mid level, with relative "apical sparing." SMDs obtained for LV-GLS (SMD -1.80, 95% CI: -2.35, -1.24, P <0.001), LV-BLS (-1.98; 95% CI: -2.51, -1.45, P <0.001) and LV-MLS (-1.84; 95% CI: -2.46, -1.23, P <0.001) assessment were significantly larger than that obtained for LV-ALS (-0.72; 95% CI: -1.31, -0.13, P=0.02) measurement. Substantial heterogeneity was found among the studies assessing LV-GLS (I2=92.5%), LV-BLS (I2=91.4%), LV-MLS (I2=94.3%) and LV-ALS (I2=94.6%). Egger's test yielded a P value of 0.10, 0.20, 0.09 and 0.55 for LV-GLS, LV-BLS, LV-MLS and LV-ALS assessment respectively, indicating no publication bias. On meta-regression analysis, none of the moderators was significantly associated with effect modification for LV-GLS, LV-BLS, LV-MLS and LV-ALS (all P<0.05).

CONCLUSIONS

Amyloidosis has a large negative effect on LV-GLS, primarily related to the deterioration of segmental longitudinal strain at the basal and mid level, with relative apical sparing.