TY - JOUR
T1 - Corrigendum to “Diffusion tensor image segmentation of the cerebrum provides a single measure of cerebral small vessel disease severity related to cognitive change” (NeuroImage
T2 - Clinical (2017) 16 (330–342), (S221315821730205X), (10.1016/j.nicl.2017.08.016))
AU - Williams, Owen A.
AU - Zeestraten, Eva A.
AU - Benjamin, Philip
AU - Lambert, Christian
AU - Lawrence, Andrew J.
AU - Mackinnon, Andrew D.
AU - Morris, Robin G.
AU - Markus, Hugh S.
AU - Charlton, Rebecca A.
AU - Barrick, Thomas R.
PY - 2019
Y1 - 2019
N2 - The authors regret that due to a data coding error, cross-sectional effects from linear mixed effects models were stored, presented and interpreted as longitudinal effects (i.e. interactions with the time variable). As a result, the results presented in Tables 5 and 6 [Table presented] do not represent the associations between change in MRI markers and change in executive function (EF) and information processing speed (IPS) but show the baseline associations. The true longitudinal associations are provided in the corrected Tables 5 and 6. The primary difference in the results is that while change in DSEG θ is significantly associated with change in EF and IPS in univariable analyses, it does not remain significant in the multivariable analyses as previously stated. Furthermore, changes in histogram metrics of FA and MD were not associated to EF and only change in FA NPH was significantly associated with IPS, whereas all four metrics were originally reported to be associated with change in EF and IPS. The changes in the results do effect some of the conclusions in the original article. While DSEG θ is a valid measure of cerebral small vessel disease (SVD) and is associated with change in EF and IPS, we must retract the statement that DSEG θ “provides the strongest predictor of cognitive change” as, in the multivariable models, lacunar infarcts and FA NPH were associated with change in EF and IPS respectively. Nevertheless, DSEG θ provides a marker of SVD severity that is related to changes in cognitive performance. The authors would like to apologize for any inconvenience caused by this error.
AB - The authors regret that due to a data coding error, cross-sectional effects from linear mixed effects models were stored, presented and interpreted as longitudinal effects (i.e. interactions with the time variable). As a result, the results presented in Tables 5 and 6 [Table presented] do not represent the associations between change in MRI markers and change in executive function (EF) and information processing speed (IPS) but show the baseline associations. The true longitudinal associations are provided in the corrected Tables 5 and 6. The primary difference in the results is that while change in DSEG θ is significantly associated with change in EF and IPS in univariable analyses, it does not remain significant in the multivariable analyses as previously stated. Furthermore, changes in histogram metrics of FA and MD were not associated to EF and only change in FA NPH was significantly associated with IPS, whereas all four metrics were originally reported to be associated with change in EF and IPS. The changes in the results do effect some of the conclusions in the original article. While DSEG θ is a valid measure of cerebral small vessel disease (SVD) and is associated with change in EF and IPS, we must retract the statement that DSEG θ “provides the strongest predictor of cognitive change” as, in the multivariable models, lacunar infarcts and FA NPH were associated with change in EF and IPS respectively. Nevertheless, DSEG θ provides a marker of SVD severity that is related to changes in cognitive performance. The authors would like to apologize for any inconvenience caused by this error.
UR - http://www.scopus.com/inward/record.url?scp=85067437503&partnerID=8YFLogxK
U2 - 10.1016/j.nicl.2019.101742
DO - 10.1016/j.nicl.2019.101742
M3 - Comment/debate
C2 - 31235449
AN - SCOPUS:85067437503
SN - 2213-1582
VL - 23
JO - NeuroImage: Clinical
JF - NeuroImage: Clinical
M1 - 101742
ER -