We agree with Di Napoli that the levels of CRP obtained at the 3rd, 10th, and 16th day in our study may not all have the same clinical significance. If we had checked the serum CRP value immediately after the onset of ischemic stoke in all patients, we would have obtained clearer results. However, patients with cerebral infarction associated with POEMS syndrome are rarely encountered. It was not possible to determine the CRP level at the same time point in every patient with ischemic stroke associated with POEMS syndrome. Fibrinogen level may be elevated in the days before stroke in some patients without POEMS syndrome.1 Hyperfibrinogenemia is known as an independent risk factor for ischemic stroke,2- 4 and such patients may be at increased risk. The patients in our study who had hyperfibrinogenemia before stroke might have an increased risk of cerebral ischemia. Although the fibrinogen levels in our patients were within the range reported by Di Napoli et al,5 those in 2 patients were higher than the mean fibrinogen level in their study. This difference might not be statistically significant. There seems to be a trend toward higher fibrinogen levels in patients with stroke who have POEMS syndrome. In our patients, recurrent ischemic infarcts were found in the end artery border zone areas without corresponding large-artery stenosis, which suggests that infarcts in POEMS syndrome may be attributable to the state of hypercoagulability. In addition, other "hypercytokinemic" molecules may be involved in the occurrence of ischemic stroke in POEMS syndrome, such as cytokines or M protein. A direct comparison of serum fibrinogen or CRP levels between ischemic stroke with POEMS syndrome and the "usual" ischemic stroke may have many undiscovered pitfalls; the level of acute-phase reactant in POEMS syndrome is a result of the reaction cascade of other abnormal molecules such as circulating immune complexes.
Kang K, Chu K, Roh J. In reply. Arch Neurol. 2004;61(1):155–156. doi: