High Plasma Brain Natriuretic Peptide Level in Thromboembolism Patients Associated With Nonvalvular Atrial Fibrillation: Cause or Effect?
To the Editor:
We read with great interest the recent article by Shimizu et al1 entitled “High Plasma Brain Natriuretic Polypeptide Level as a Marker of Risk for Thromboembolism in Patients With Nonvalvular Atrial Fibrillation.” Although in this article the authors concluded that plasma brain natriuretic peptide (BNP) may be a useful marker to predict vulnerability to thromboembolism in patients with nonvalvular atrial fibrillation (AF), we have some reservations about the word predict or predictor used in this report.
The authors mentioned that there was a significant difference in plasma BNP levels between the group with an embolic event and the group with no embolic event (126±53 versus 84±45 ng/L) and that overall analysis of the continuous variables with multiple logistic regression analysis revealed that the plasma BNP level was an independent predictor of thromboembolic complication. However, it seems doubtful whether plasma BNP level may be a useful predictor of thromboembolic complication for reasons mentioned below; we believe there is some statistical relationship between the plasma BNP levels and the occurrence of the thromboembolic events in this study. If the blood sampling for BNP measurements had been performed before thromboembolic events in all the patients of the embolic event group, we would agree with the authors that plasma BNP level might be a useful predictor of thromboembolic complication. In this study, unfortunately, the blood sampling was performed before the events in only 3 of 11 patients who had a history of thromboembolic events and after the events in the other 8 patients.
BNP has been considered to be secreted mainly from the heart, especially the left ventricle2 or the left atrium.3 Recent evidence,4 however, showed that there was some immunoreactivity of BNP also through the brain, including the cerebral cortex, thalamus, cerebellum, pons, and hypothalamus, thus indicating that BNP secretion may be induced by pathological processes involving these regions. Actually, there has recently been increasing evidence that plasma BNP level increases after subarachnoid hemorrhage,5 especially in patients with symptomatic cerebral vasospasm.6 In addition, Saper et al7 demonstrated that the internal carotid artery and the proximal portions of the middle and anterior cerebral and posterior communicating arteries were the most intensely innervated by BNP-immunoreactive fibers in the rat. These findings suggest that an ischemic insult to brains may also induce BNP secretion. In this study 7 of 11 patients who had a history of systemic thromboembolic events had a history of cerebral infarction, thus possibly leading to an increase of plasma BNP level. We therefore think that further prospective studies are required to assess the usefulness of measuring plasma BNP levels in detecting patients at high risk for thromboembolic complications in nonvalvular AF.
- ↵Shimizu H, Murakami Y, Inoue S, Ohta Y, Nakamura K, Katoh H, Sakne T, Takahashi N, Ohata S, Sugamori T, Ishibashi Y, Shimada T. High plasma brain natriuretic polypeptide level as a marker of risk for thromboembolism in patients with nonvalvular atrial fibrillation. Stroke. 2002; 33: 1005–1010.
- ↵Yasue H, Yoshimura M, Sumida H, Kikuta K, Kugiyama K, Jougasaki M, Ogawa H, Okumura K, Mukoyama M, Nakao K. Localization and mechanism of secretion of B-type natriuretic peptide in comparison with those of A-type natriuretic peptide in normal subjects and patients with heart failure. Circulation. 1994; 90: 195–203.
- ↵Sviri GE, Feinsod M, Soustiel JF. Brain natriuretic peptide and cerebral vasospasm in subarachnoid hemorrhage: clinical and TCD correlations. Stroke. 2000; 31: 118–122.
We appreciate the interesting comments of Dr Fukui and his colleagues regarding our recent article.1 Previous investigations have shown that patients demonstrate an elevation of brain natriuretic polypeptide (BNP) several days after subarachnoid hemorrhage.2,3⇓ We performed sampling using catheters in patients with nonvalvular atrial fibrillation (AF) and reported that the left atrium mainly secretes BNP in AF.4 On the other hand, the mechanism and source of BNP are not yet clarified in patients with subarachnoid hemorrhage. Patients with subarachnoid hemorrhage sometimes demonstrate cardiac damage, including serial changes in ECGs and wall motion abnormality.5,6⇓ Some of these conditions could stimulate cardiac BNP secretion. In our study blood sampling for BNP assay was performed in an outpatient clinic at least 6 months after the onset of cerebral embolism (range, 6 months to 5 years). Moreover, plasma BNP levels in the patient group with cerebral infarction (n=7) were 121±53 ng/L. This is not statistically different, however, from that in the group with other arterial embolism (120±45 ng/L; n=6). Taken together, we think it is unlikely that BNP is secreted from the brain in these patients with chronic healed cerebral infarction.
We used a logistic regression model for data analysis. This method is usable in a clinical investigation designed to retrospectively detect factor(s) predicting a certain dichotomous variable.7 We therefore can state that “BNP predicts thromboembolism.” We hope that the clinical usefulness of BNP in predicting the risk of arterial embolism is verified by a prospectively designed study in a larger patient population with nonvalvular AF.
- ↵Simizu H, Murakami Y, Inoue S, Ohta Y, Nakamura K, Katoh H, Sakane T, Takahashi N, Ohata S, Sugamori T, Ishibashi Y, Shimada T. High plasma brain natriuretic polypeptide level as a marker of risk for thromboembolism in patients with nonvalvular atrial fibrillation. Stroke. 2002; 33: 1005–1010.
- ↵Sviri GE, Feinsod M, Soustiel JF. Brain natriuretic peptide and cerebral vasospasm subarachnoid hemorrhage: clinical and TCD correlations. Stroke. 2000; 31: 118–122.
- ↵Rudehill A, Olsson GL, Sundqvist K. ECG abnormalities in patients with subarachnoid haemorrhage and intracranial tumors. J Neurol Neurosurg Psychiatry. 1987; 50: 1375–1381.
- ↵Kleinbaum DG. Logistic Regression: A Self-Learning Text. New York, NY: Springer-Verlag; 1996.