Limitations
Long-term follow-up in our study was incomplete (87% of the patients
included in 2DSPER had long-term follow-up). Long-term follow-up,
however, was available in 99% of the patients included in the 6 centers
participating in the currents study, thus minimizing selection bias.
Given the unequivocal nature of our results (Figure 2), it is highly
unlikely that complete follow-up would significantly alter our findings.
GLS is dependent on 2D image quality.22,27 In 2DPSER
mid-range systems, frequently utilized in the ED, were used. We do not
know whether using high-end machines would have significantly improved
GLS accuracy and its predictive power. However, only patients with
adequate 2D image quality were included in 2DSPER, hence making our
findings relevant to everyday practice. Furthermore, GLS did not predict
outcome even in the subgroup of patients with optimal 2D echo image
quality.