{"id":6246,"date":"2012-06-13T16:01:51","date_gmt":"2012-06-13T21:01:51","guid":{"rendered":"http:\/\/www.ajnrblog.org\/?p=6246"},"modified":"2012-06-28T13:20:35","modified_gmt":"2012-06-28T18:20:35","slug":"comments-on-an-article-by-kamalian-et-al","status":"publish","type":"post","link":"https:\/\/www.ajnrblog.org\/2012\/06\/13\/comments-on-an-article-by-kamalian-et-al\/","title":{"rendered":"Comments on an Article by Kamalian et al"},"content":{"rendered":"

Published online before print April 19, 2012, doi: 10.3174\/ajnr.A3153
\nAJNR 2012 33: E94<\/span><\/p>\n

P. Schramma<\/sup><\/em><\/strong>
\n a<\/sup>Department of Neuroradiology<\/em>
\n University Medicine Goettingen<\/em>
\n Goettingen, Germany<\/em><\/p>\n

E. Klotzb<\/sup><\/em><\/strong>
\n b<\/sup>Siemens Healthcare Sector<\/em>
\n Forchheim, Germany<\/em><\/p>\n

We read with great interest the article by Kamalian et al.1<\/a><\/sup>\u00a0The authors found, in a carefully selected patient group, that CT perfusion mean transit time maps optimally distinguished benign oligemia from true \u201cat-risk\u201d ischemic penumbra. They postprocessed the same dynamic CT data with 2 commercial software packages from the same vendor, containing a standard and a delay-corrected (DC) deconvolution algorithm, respectively.<\/p>\n

For each algorithm considered separately, receiver operating characteristic (ROC) analysis yielded a significantly higher area under the curve for absolute and relative MTT than for the other CTP parameters for identification of brain tissue destined to infarct (all\u00a0P<\/em>\u00a0values < .01). The authors also found that thresholds needed to be adapted. They reported that \u201cabsolute and relative MTT thresholds for defining penumbra were 12 seconds and 249% for the standard and 13.5 seconds and 150% for the delay-corrected algorithms, respectively.\u201d That thresholds may vary considerably among different approaches and implementations is well known, but their results appear to have an internal discrepancy that requires clarification. Relative MTTs were calculated by normalizing the ischemic MTT to the one for the corresponding anatomy on the contralateral side. Therefore, one can reversely deduce that the average MTT of the normal brain by using the standard algorithm in their data was approximately 4.8 seconds (12\/2.49 seconds); the one in the delay-corrected version, however, would have to be approximately 9 seconds (13.5\/1.5 seconds). Nine seconds would be in total disagreement with basically all normal MTT values that can be found in the literature (eg, Wintermark et al2<\/a><\/sup>), which are more in the range of 4\u20136 seconds (in agreement with the results of the standard algorithm).<\/p>\n

In addition, a DC algorithm would tend to have an even shorter MTT.3<\/a><\/sup>\u00a0The gross whole-brain transit time can be estimated from the peak time difference of the arterial input function and the venous outflow function. In our experience, this difference is typically between 6 and 8 seconds; tissue MTT must be shorter. Figures 3 and 4 in a recent review article coauthored by 2 of the authors of the present study clearly confirm this.4<\/a><\/sup>If there was no accidental misreporting of numbers, this discrepancy definitely requires an explanation and discussion.<\/p>\n

Furthermore, the authors argue at length that DC algorithms are superior to standard ones because they are better adapted to the variable arrival times. We fully agree. Their data, however, appear to demonstrate the opposite. All areas under the curve and specificities for the standard algorithm are consistently higher than those for DC. If the next accurate parameter was CBF, as they report, and the difference between 0.76\/0.78 (MTT standard) and 0.73\/0.74 (CBF standard) was significant at\u00a0P<\/em>\u00a0< .01, then clearly the difference between 0.76\/0.78 (MTT standard) and 0.72\/0.71 (MTT DC) will have similar or higher significance. If one were to use the authors’ reasoning about optimal performance, one could also draw the conclusion that the CBF of standard deconvolution software (CTP3 \u201cStd,\u201d GE Healthcare) performs better than the MTT of delay-corrected software (CTP5 \u201cDC\u201d; GE Healthcare). We believe this issue requires further explanation and discussion.<\/p>\n

\n

References<\/h3>\n
    \n
  1. Kamalian\u00a0S, Kamalian\u00a0S, Konstas\u00a0AA, et al.\u00a0CT perfusion mean transit time maps optimally distinguish benign oligemia from true \u201cat-risk\u201d ischemic penumbra, but thresholds vary by postprocessing technique.\u00a0AJNR Am J Neuroradiol<\/abbr>\u00a02012;\u00a033:\u00a0545\u201349\u00a0\u00bb Abstract\/FREE\u00a0Full Text<\/a><\/li>\n
  2. Wintermark\u00a0M, Flanders\u00a0AE, Velthuis\u00a0B, et al.\u00a0Perfusion-CT assessment of infarct core and penumbra: receiver operating characteristic curve analysis in 130 patients suspected of acute hemispheric stroke.\u00a0Stroke<\/abbr>\u00a02006;\u00a037:\u00a0979\u201385 \u00bb Abstract\/FREE\u00a0Full Text<\/a><\/li>\n
  3. Abels\u00a0B, Klotz\u00a0E, Tomandl\u00a0BF, et al.\u00a0Perfusion CT in acute ischemic stroke: a qualitative and quantitative comparison of deconvolution and maximum slope approach.\u00a0AJNR Am J Neuroradiol<\/abbr>\u00a02010;\u00a031:\u00a01690\u201398 \u00bb Abstract\/FREE\u00a0Full Text<\/a><\/li>\n
  4. Konstas\u00a0AA, Wintermark\u00a0M, Lev\u00a0MH.\u00a0CT perfusion imaging in acute stroke. Neuroimaging Clin N Am<\/abbr>\u00a02011;\u00a021:\u00a0215\u201338,\u00a0ix\u00a0\u00bb CrossRef<\/a>Medline<\/a><\/li>\n<\/ol>\n

    Reply<\/h2>\n

    Published online before print April 19, 2012, doi: 10.3174\/ajnr.A3154
    \nAJNR 2012 33: E95<\/span><\/p>\n

    We thank Drs Schramm and Klotz for their interest in our article and this important topic.1<\/a><\/sup>\u00a0They are indeed correct that based on our reported delay-insensitive CTP5 software package raw data, \u201cone can reversely deduce that the average MTT of the normal brain \u2026 would have to be approximately 9 seconds \u2026 in total disagreement with basically all normal MTT values that can be found in the literature. \u2026\u201d<\/p>\n

    In brief, this discrepancy is explained by the fact that the CTP5 delay-insensitive postprocessing software used for our study\u2014as noted in our \u201cMaterials and Methods\u201d section\u2014was a research version that required appropriate DICOM scaling of the raw data values if these were to be used for absolute quantitation of the CTP parameters; this scaling was not automatically performed by the Analyze third-party software package (Analyze 8.1; Analyze-Direct, Mayo Clinic, Rochester, Minnesota) used for our analyses. (The CTP5 beta software was intended for research use only and has since been replaced by the now commercially available CTP 4D software [http:\/\/www.gehealthcare.com\/usen\/ct\/products\/docs\/CT_Clarity_062411_pg48-49.pdf<\/a>], for which this scaling factor is not an issue).<\/p>\n

    For absolute quantification of the CTP5 MTT maps, the required scaling factor is 2, meaning that the correct \u201creversely deduced\u201d average MTT value derived from our results is, in fact, approximately 4.5 seconds. This value is not only in agreement with the literature but is also smaller than the 4.8 seconds derived by using the delay-sensitive (CTP3) standard algorithm and is in keeping with the expected results as outlined in Drs Schramm and Klotz’s letter. (Additionally, if we apply this scaling, our absolute quantitative delay-corrected MTT threshold for \u201ctrue\u201d ischemic penumbra becomes 6.75, rather than 13.5 seconds.)<\/p>\n

    Hence, the values we reported for MTT penumbral thresholds were specific to our postprocessing\/analysis platforms and were not intended for use to \u201cback-calculate\u201d absolute quantitative MTT parameter values in clinically normal brain. Indeed, the most important conclusions of our article underscore this point\u2014that the CTP threshold values reported in the literature are platform-specific, are not standardized, and hence are not necessarily generalizable to acquisition and postprocessing protocols other than those specifically under investigation (in our case, CTP3 and CTP5). Moreover, absolute quantification of CTP parameter values is highly variable and critically dependent on many factors, such as correct placement of a venous scaling region of interest and estimation of hematocrit. For these reasons, we favor the use of relative, rather than absolute, perfusion values for our clinical stroke work.<\/p>\n

    The aims of our study were the following: 1) to determine which CTP map or maps optimally distinguish benign oligemia from true \u201cat-risk\u201d penumbra, and 2) to confirm earlier reports suggesting that specific threshold values might vary according to the postprocessing platform used. Despite the considerations discussed above, we found\u2014by using both the delay-sensitive and the delay-insensitive software\u2014that both relative and absolute MTTs were the most accurate CTP maps for determining \u201ccritical\u201d penumbra. This result was irrespective of the scaling factors required for absolute quantification or other potentially confounding technical differences between the commercial and beta versions of the software, which were outside the scope of our study (eg, the degree of image noise). In this regard, it was not our goal to compare the accuracy of the delay-sensitive-versus-delay-insensitive platforms. Had this been the case, we would have studied a more homogeneous highly selected patient cohort, all with significant proximal large-vessel occlusions (ICA and\/or M1) so as to target the marked contrast arrival-time differences between regions with otherwise similar baseline cerebral blood flow.<\/p>\n

    We appreciate this opportunity to clarify our work, apologize for any confusion in the interpretation of our results, and again thank Drs Schramm and Klotz for helping to highlight these important issues.<\/p>\n

    \n

    Reference<\/h3>\n
      \n
    1. Kamalian\u00a0S, Kamalian\u00a0S, Konstas\u00a0AA, et al.\u00a0CT perfusion mean transit time maps optimally distinguish benign oligemia from true \u201cat-risk\u201d ischemic penumbra, but thresholds vary by postprocessing technique.\u00a0AJNR Am J Neuroradiol<\/abbr>\u00a02012;33:545\u201349\u00a0\u00bb Abstract\/FREE\u00a0Full Text<\/a><\/li>\n<\/ol>\n<\/div>\n<\/div>\n","protected":false},"excerpt":{"rendered":"

      Published online before print April 19, 2012, doi: 10.3174\/ajnr.A3153 AJNR 2012 33: E94 P. Schramma aDepartment of Neuroradiology University Medicine Goettingen Goettingen, Germany E. Klotzb bSiemens Healthcare Sector Forchheim, Germany We read with great interest the article by Kamalian et<\/p>\n","protected":false},"author":465,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[176],"tags":[],"_links":{"self":[{"href":"https:\/\/www.ajnrblog.org\/wp-json\/wp\/v2\/posts\/6246"}],"collection":[{"href":"https:\/\/www.ajnrblog.org\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.ajnrblog.org\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.ajnrblog.org\/wp-json\/wp\/v2\/users\/465"}],"replies":[{"embeddable":true,"href":"https:\/\/www.ajnrblog.org\/wp-json\/wp\/v2\/comments?post=6246"}],"version-history":[{"count":4,"href":"https:\/\/www.ajnrblog.org\/wp-json\/wp\/v2\/posts\/6246\/revisions"}],"predecessor-version":[{"id":6301,"href":"https:\/\/www.ajnrblog.org\/wp-json\/wp\/v2\/posts\/6246\/revisions\/6301"}],"wp:attachment":[{"href":"https:\/\/www.ajnrblog.org\/wp-json\/wp\/v2\/media?parent=6246"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.ajnrblog.org\/wp-json\/wp\/v2\/categories?post=6246"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.ajnrblog.org\/wp-json\/wp\/v2\/tags?post=6246"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}