Low Plasma Antioxidant Activity Is Associated With High Lesion Volume and Neurological Impairment in Stroke
To the Editor:
We read with great interest the article by Leinonen et al1 in the January 2000 issue of Stroke. In this study, low total peroxyl radical-trapping potential (TRAP) of plasma, but not of cerebrospinal fluid, was found to be associated with high lesion volume and high neurological impairment. An inverse correlation between plasma TRAP and infarct volume (r=−0.53, P=0.01 for TRAP and r=−0.56, P=0.007 for TRAPunid), in addition to higher lesion volume in patients with a lower-than-mean level of plasma TRAP (1190 μmol/L) than patients with a TRAP level higher than that (15.3±14.1 versus 5.6±7.1, P=0.03), was the affirmative underlying statistical findings for this result. The infarct volume–plasma TRAP level correlation analysis was performed in 21 patients with right hemisphere acute infarction. Lesion volume was <1.5 cm3 in 6 patients. Their NIH Stroke Scale (NIHSS) scores, ranging from 1 to 4 (mean 2.2±1.5), suggest involvement of only 1 or 2 neurological symptom domains in the suspicion of lacunar infarction. Moreover, 4 of these patients were hypertensive. When we examined differences in plasma TRAP levels between these patients and 12 patients with infarction >3 cm3, using the numbers from Table 1 of Leinonen et al1 and the Mann-Whitney U test, we found that the small infarct group had higher plasma TRAP levels than the larger group (1377.0±219 versus 1102±151, P=0.025). When the small infarct group was omitted, the correlation between infarct volume and plasma TRAP levels was not significant (in patients with infarction >1.5 cm3, n=15, r=−0.140, P=0.619; in patients with infarct volume >3 cm3, n=12, r=0.094, P=0.773, by the Pearson correlation test). Similarly, NIHSS score was not correlated to the plasma TRAP levels in these patient groups (in the patients with infarction >1.5 cm3, n=15, r=−0.314, P=0.253; in the patients with infarct volume >3 cm3, n=12, r=0.115, P=0.722, by the Pearson correlation test). We thought that omission of lacunar infarction was important because of the title of the article by Leinonen et al1 : “Low Plasma Antioxidant Activity Is Associated With High Lesion Volume and Neurological Impairment in Stroke”. By reanalyzing of their data, this statement did not seem to be relevant to territorial infarctions. If so, their paper may mislead casual readers.
According to aggregated data from a large number of experimental in addition to a limited number of human stroke studies, the generation of free radicals and/or antioxidant depletion leading to oxidative stress plays an important pathophysiological role in ischemic brain injury. Currently, there is evidence that oxidative damage to membrane lipids and proteins is increased during cerebral ischemia and reperfusion. Even though it remains difficult to determine which part of the damage is caused by hypoxia and which part by reperfusion, it seems possible that this type injury may occur in all types of territorial infarction. However, lacunar infarction shows very different pathogenetic properties. The presence of ischemic penumbra and occurrence of reperfusion are not, indeed, expected in lacunar infarctions. Apart from demonstration of elevation of plasma lipid peroxides in diabetics with multiple lacunar infarction2 and lowered vitamin A/cholesterol and carotenoids/cholesterol ratios in small-artery ischemic stroke3 in some studies, the oxidative injury similar to territorial infarct is never believed to exist in lacunar infarcts. Also, in some studies of oxidative injury in ischemic stroke, patients with lacunar infarction have served as the control group.4
- Copyright © 2000 by American Heart Association