[Skip to Navigation]
Views 182
Citations 0
Comment & Response
May 2017

A New Sign of Intracerebral Hematoma Expansion—Reply

Author Affiliations
  • 1Hemorrhagic Stroke Research Program, Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston
  • 2Department of Neuroradiology, Université Paris Descartes, INSERM S894, Centre Hospitalier Sainte-Anne, Paris, France
JAMA Neurol. 2017;74(5):609-610. doi:10.1001/jamaneurol.2017.0073

In Reply We thank Xia et al for their letter on our recently published article1 in which they discuss key aspects and potential limitations of our study. Published studies with closely tangential or similar analyses (eg, an analysis between the extent of acute bleeding and magnetic resonance imaging markers of small vessel disease burden) have shown contradictory results, as mentioned in the Discussion in our article1 and the corresponding Editorial by Smith.2 Specifically, the association between cerebral microbleeds (CMBs) burden and bleeding extent has been demonstrated to range from direct (and important)3 to inverse (and moderate).1,4 The reason for this discrepancy remains unclear and may result from different patient populations, study selection criteria, amalgamation paradoxes, or “chance only” findings.2 Another important consideration is selection bias since patients with more severe hemorrhages were less likely to be included (these patients may bear a higher CMBs burden), which is an inherent and well-known factor of all intracerebral hemorrhage studies that rely on clinical imaging. Despite those potential limitations, several recent reports offer harmonious results suggesting that CMBs (at least those related to cerebrovascular amyloid deposition) reflect thicker vessels that are shielded by amyloid and are thus resistant to ongoing bleeding through secondary ruptures in the acute phase. For instance, the findings that a higher burden of CMBs is related to a lower prevalence of spot sign (a strong marker of continued bleeding) and that microbleeds and macrobleeds represent distinct entities neuropathologically5 (with the implication that patients with higher counts of CMBs are less likely to develop macrohemorrhages) support our results. Notably, our analysis covered the spectrum of small vessel disease magnetic resonance imaging markers (eg, cortical superficial siderosis), which was often not the case in previous reports, likely affecting the associations with hematoma expansion.

Add or change institution