Limits...
Feature tracking measurement of dyssynchrony from cardiovascular magnetic resonance cine acquisitions: comparison with echocardiographic speckle tracking.

Onishi T, Saha SK, Ludwig DR, Onishi T, Marek JJ, Cavalcante JL, Schelbert EB, Schwartzman D, Gorcsan J - J Cardiovasc Magn Reson (2013)

Bottom Line: Analysis of left ventricular (LV) mechanical dyssynchrony may provide incremental prognostic information regarding cardiac resynchronization therapy (CRT) response in addition to QRS width alone.These were large (up to 100% or more) relative to the small mean delays measured in more synchronous patients, but acceptable (mainly <25%) in those with mean delays of >200 ms.The clinical usefulness of the method, for example in predicting prognosis in CRT patients, remains to be investigated.

View Article: PubMed Central - HTML - PubMed

Affiliation: The University of Pittsburgh, 200 Lothrop Street, Pittsburgh, PA, USA. gorcsanj@upmc.edu.

ABSTRACT

Background: Analysis of left ventricular (LV) mechanical dyssynchrony may provide incremental prognostic information regarding cardiac resynchronization therapy (CRT) response in addition to QRS width alone. Our objective was to quantify LV dyssynchrony using feature tracking post processing of routine cardiovascular magnetic resonance (CMR) cine acquisitions (FT-CMR) in comparison to speckle tracking echocardiography.

Methods: We studied 72 consecutive patients who had both steady-state free precession CMR and echocardiography. Mid-LV short axis CMR cines were analyzed using FT-CMR software and compared with echocardiographic speckle tracking radial dyssynchrony (time difference between the anteroseptal and posterior wall peak strain).

Results: Radial dyssynchrony analysis was possible by FT-CMR in all patients, and in 67 (93%) by echocardiography. Dyssynchrony by FT-CMR and speckle tracking showed limits of agreement of strain delays of ± 84 ms. These were large (up to 100% or more) relative to the small mean delays measured in more synchronous patients, but acceptable (mainly <25%) in those with mean delays of >200 ms. Radial dyssynchrony was significantly greater in wide QRS patients than narrow QRS patients by both FT-CMR (radial strain delay 230 ± 94 vs. 77 ± 92* ms) and speckle tracking (radial strain delay 242 ± 101 vs. 75 ± 88* ms, all *p < 0.001).

Conclusions: FT-CMR delivered measurements of radial dyssynchrony from CMR cine acquisitions which, at least for the patients with more marked dyssynchrony, showed reasonable agreement with those from speckle tracking echocardiography. The clinical usefulness of the method, for example in predicting prognosis in CRT patients, remains to be investigated.

Show MeSH

Related in: MedlinePlus

Regression lines and Bland Altman plots. Radial dyssynchrony by feature tracking CMR was significantly correlated with that by speckle tracking echocardiography indexes: anteroseptal (ant-sep) to posterior delay by feature tracking CMR versus speckle tracking echocardiography. Bland Altman plot showed limits of agreement between different software for ant-sep to posterior wall delay as well. ST echo = speckle tracking echocardiography, FT CMR = feature tracking CMR.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4016574&req=5

Figure 3: Regression lines and Bland Altman plots. Radial dyssynchrony by feature tracking CMR was significantly correlated with that by speckle tracking echocardiography indexes: anteroseptal (ant-sep) to posterior delay by feature tracking CMR versus speckle tracking echocardiography. Bland Altman plot showed limits of agreement between different software for ant-sep to posterior wall delay as well. ST echo = speckle tracking echocardiography, FT CMR = feature tracking CMR.

Mentions: Imaging data were suitable in 100% (72/72) and 93% (67/72) for quantitative feature tracking CMR and speckle tracking echocardiography radial dyssynchrony analysis, respectively (p = 0.05). Over the relatively wide range of strain delays studied, radial dyssynchrony by FT-CMR correlated with that by speckle tracking echocardiography as follows: anteroseptal to posterior wall delay (r = 0.93, p < 0.0001; Figure 3). The Bland Altman plot showed moderate agreement of feature tracking CMR with speckle tracking echocardiography for anteroseptal to posterior wall delay, but lack of consistent agreement for values less than 100 ms and variability above 200 ms mainly within 25% difference levels (Figure 3). However, anteroseptal to posterior wall delays < 100 ms are seen in normal controls and not of clinical significance with a delay > 130 ms considered as dyssynchronous [9,19]. As expected, radial dyssynchrony was significantly greater in patients with wide QRS than in patients with narrow QRS duration both by feature tracking CMR (anteroseptal to posterior wall delay 230 ± 94 vs 77 ± 92* ms) and by speckle tracking echocardiography (anteroseptal to posterior wall delay 242 ± 101 vs 75 ± 88* ms, all *p < 0.001). Intra- / inter-observer variabilities of feature tracking CMR radial dyssynchrony were respectively 4 ± 4% / 5 ± 6% for anteroseptal to posterior wall delay. Intra- / inter-observer variabilities of speckle tracking echocardiography radial dyssynchrony were respectively 8 ± 9% / 9 ± 10% for anteroseptal to posterior wall delay.


Feature tracking measurement of dyssynchrony from cardiovascular magnetic resonance cine acquisitions: comparison with echocardiographic speckle tracking.

Onishi T, Saha SK, Ludwig DR, Onishi T, Marek JJ, Cavalcante JL, Schelbert EB, Schwartzman D, Gorcsan J - J Cardiovasc Magn Reson (2013)

Regression lines and Bland Altman plots. Radial dyssynchrony by feature tracking CMR was significantly correlated with that by speckle tracking echocardiography indexes: anteroseptal (ant-sep) to posterior delay by feature tracking CMR versus speckle tracking echocardiography. Bland Altman plot showed limits of agreement between different software for ant-sep to posterior wall delay as well. ST echo = speckle tracking echocardiography, FT CMR = feature tracking CMR.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4016574&req=5

Figure 3: Regression lines and Bland Altman plots. Radial dyssynchrony by feature tracking CMR was significantly correlated with that by speckle tracking echocardiography indexes: anteroseptal (ant-sep) to posterior delay by feature tracking CMR versus speckle tracking echocardiography. Bland Altman plot showed limits of agreement between different software for ant-sep to posterior wall delay as well. ST echo = speckle tracking echocardiography, FT CMR = feature tracking CMR.
Mentions: Imaging data were suitable in 100% (72/72) and 93% (67/72) for quantitative feature tracking CMR and speckle tracking echocardiography radial dyssynchrony analysis, respectively (p = 0.05). Over the relatively wide range of strain delays studied, radial dyssynchrony by FT-CMR correlated with that by speckle tracking echocardiography as follows: anteroseptal to posterior wall delay (r = 0.93, p < 0.0001; Figure 3). The Bland Altman plot showed moderate agreement of feature tracking CMR with speckle tracking echocardiography for anteroseptal to posterior wall delay, but lack of consistent agreement for values less than 100 ms and variability above 200 ms mainly within 25% difference levels (Figure 3). However, anteroseptal to posterior wall delays < 100 ms are seen in normal controls and not of clinical significance with a delay > 130 ms considered as dyssynchronous [9,19]. As expected, radial dyssynchrony was significantly greater in patients with wide QRS than in patients with narrow QRS duration both by feature tracking CMR (anteroseptal to posterior wall delay 230 ± 94 vs 77 ± 92* ms) and by speckle tracking echocardiography (anteroseptal to posterior wall delay 242 ± 101 vs 75 ± 88* ms, all *p < 0.001). Intra- / inter-observer variabilities of feature tracking CMR radial dyssynchrony were respectively 4 ± 4% / 5 ± 6% for anteroseptal to posterior wall delay. Intra- / inter-observer variabilities of speckle tracking echocardiography radial dyssynchrony were respectively 8 ± 9% / 9 ± 10% for anteroseptal to posterior wall delay.

Bottom Line: Analysis of left ventricular (LV) mechanical dyssynchrony may provide incremental prognostic information regarding cardiac resynchronization therapy (CRT) response in addition to QRS width alone.These were large (up to 100% or more) relative to the small mean delays measured in more synchronous patients, but acceptable (mainly <25%) in those with mean delays of >200 ms.The clinical usefulness of the method, for example in predicting prognosis in CRT patients, remains to be investigated.

View Article: PubMed Central - HTML - PubMed

Affiliation: The University of Pittsburgh, 200 Lothrop Street, Pittsburgh, PA, USA. gorcsanj@upmc.edu.

ABSTRACT

Background: Analysis of left ventricular (LV) mechanical dyssynchrony may provide incremental prognostic information regarding cardiac resynchronization therapy (CRT) response in addition to QRS width alone. Our objective was to quantify LV dyssynchrony using feature tracking post processing of routine cardiovascular magnetic resonance (CMR) cine acquisitions (FT-CMR) in comparison to speckle tracking echocardiography.

Methods: We studied 72 consecutive patients who had both steady-state free precession CMR and echocardiography. Mid-LV short axis CMR cines were analyzed using FT-CMR software and compared with echocardiographic speckle tracking radial dyssynchrony (time difference between the anteroseptal and posterior wall peak strain).

Results: Radial dyssynchrony analysis was possible by FT-CMR in all patients, and in 67 (93%) by echocardiography. Dyssynchrony by FT-CMR and speckle tracking showed limits of agreement of strain delays of ± 84 ms. These were large (up to 100% or more) relative to the small mean delays measured in more synchronous patients, but acceptable (mainly <25%) in those with mean delays of >200 ms. Radial dyssynchrony was significantly greater in wide QRS patients than narrow QRS patients by both FT-CMR (radial strain delay 230 ± 94 vs. 77 ± 92* ms) and speckle tracking (radial strain delay 242 ± 101 vs. 75 ± 88* ms, all *p < 0.001).

Conclusions: FT-CMR delivered measurements of radial dyssynchrony from CMR cine acquisitions which, at least for the patients with more marked dyssynchrony, showed reasonable agreement with those from speckle tracking echocardiography. The clinical usefulness of the method, for example in predicting prognosis in CRT patients, remains to be investigated.

Show MeSH
Related in: MedlinePlus