Tissue Plasminogen Activator −7351C/T Polymorphism and Lacunar Stroke
To the Editor:
Jannes et al1 reported that a polymorphism in the tissue plasminogen activator (tPA) gene (−7351C/T) was associated with ischemic stroke in an Australian population. Stratification for stroke subtype demonstrated an association with lacunar infarction (OR: 2.7; 95% CI, 1.1 to 6.7), but not with other stroke subtypes. The authors interpreted this result as providing evidence that fibrinolytic factors play an important role in maintaining small vessel patency. The polymorphism is located within the binding site for the transcription factor Sp1, in the enhancer region of the tPA gene,1 and the TT genotype has been associated with significantly reduced vascular tPA release rates.2
This association could give clues to the pathogenesis of lacunar stroke, but first replication in independent populations is important, particularly because the original association was found in a small sample size, with only 43 patients in the lacunar subgroup. Therefore, we attempted to replicate it in a prospectively collected group of patients with well-phenotyped lacunar stroke. In addition, we determined whether the polymorphism predisposed to 1 particular type of lacunar stroke. It has been suggested that patients with larger lacunar infarcts (isolated lacunar infarction [ILI]) without leukoaraiosis may have microatheroma at the origins and proximal portions of the perforating arteries. In contrast, patients with lacunar infarction and confluent leukoaraiosis (ischemic leukoaraiosis [ILA]) may have a diffuse arteriopathy affecting the smaller perforating vessels.3 Previous studies have suggested different genetic associations in the 2 groups.4
Three hundred and twelve consecutive white patients presenting with lacunar stroke attending participating stroke services were recruited. Lacunar stroke was defined as clinical lacunar syndrome with accompanying lesion on MRI or CT. All patients had brain imaging and imaging of the carotid arteries with duplex or MR angiography. Patients with subcortical lesion ≥1.5 cm in diameter, cortical infarction of any size, a potential cardioembolic source and large-vessel disease defined as carotid, vertebral or basilar intracranial artery stenosis ≥50% were excluded. Two hundred and twenty-six (72%) had MRI and 86 (28%) had CT. In the principal center consecutive patients with lacunar stroke were recruited regardless of the presence of leukoaraiosis. To increase the number of cases with ischemic leukoaraiosis, in 4 additional centers, consecutive patients with ischemic leukoaraiosis were recruited. Patients were divided into 2 groups: ILI (absent/mild leukoaraiosis) or ILA (moderate/severe leukoaraiosis) according to a previously validated method.4 Six hundred and twenty-six age- and sex-matched controls free of symptomatic cerebrovascular disease were recruited by sampling family doctor lists from the same geographic locations as the patients. The study protocol was approved by local research ethics committees and informed consent was obtained from all participants. Genotyping was performed blinded to subtype diagnosis and case/control status.
DNA was isolated from blood samples using a commercial kit (Nucleon), and a 366bp region of DNA surrounding the tPA−7351C/T polymorphism was amplified using PCR. Digestion with Ban II restriction endonuclease produced 1 of 2 characteristic sets of fragments, depending on the presence of a C or a T allele at the SNP. The restriction fragments were separated on a 2% Micro ABgarose gel.
Results were obtained for 611(98%) controls and 301(96%) cases. The genotypes distribution was in Hardy-Weinberg equilibrium (P=0.051). There was no association between the TT genotype and lacunar stroke before or after adjustment for age, gender, hypertension, diabetes, and smoking (before adjustment OR: 0.761; P=0.206; 95% CI, 0.499 to 1.162; after adjustment OR: 1.343; P=0.187; 95% CI, 0.866 to 2.084). This was also true for both subtypes (ILI before adjustment OR: 0.796; P=0.433; 95% CI, 0.451 to 1.406; after adjustment OR: 1.141; P=0.666; 95% CI, 0.626 to 2.081; ILA OR: 1.347; P=0.280; 95% CI, 0.785 to 2.310; after adjustment OR: 1.413; P=0.234; 95% CI, 0.800 to 2.497).
In conclusion, this study does not support the tPA polymorphism being a risk factor for lacunar stroke. It is most likely that the original association was a false-positive attributable to small sample size.
This work was funded by the British Heart Foundation (PG03/070) and Stroke Association. Thanks also to Robyn Labrum for helpful discussions.
Jannes J, Hamilton-Bruce MA, Pilotto L, Smith BJ, Mullighan CG, Bardy PG, Koblar SA. Tissue Plasminogen Activator −7351C/T Enhancer Polymorphism Is a Risk Factor for lacunar Stroke. Stroke. 2004; 35: 1090–1094.
Boiten J, Lodder J, Kessels F. Two clinically distinct lacunar infarct entities? A hypothesis. Stroke. 1993; 24: 652–656.
Hassan A, Hunt BJ, O’Sullivan M, Bell R, D’Souza R, Jeffery S, Bamford JM, Markus HS. Homocysteine is a risk factor for cerebral small vessel disease, acting via endothelial dysfunction. Brain. 2004; 127: 212–219.
We thank Armstrong and colleagues for acknowledging the importance of the original study, which found a positive association between the tPA –7351 TT genotype and lacunar stroke in an Australian population. As we concluded in the article, this association was made after subgroup analysis and was thus interpreted as hypothesis generating. We agree that confirmation in a larger, well-designed study is critical. It is important to reiterate that the tPA −7351 C/T polymorphism possesses a plausible biological reason to cause cerebral lacunar infarction, and it has been implicated in other arterial disease phenotypes, namely coronary artery disease.1
The current 2 studies of the tPA –7351 C/T polymorphism in cerebrovascular disease illustrate the conundrum of genetic association studies, namely conflicting results. Many times the conflicting results may arise from differences in study power or methodology. It should be remembered that the results from all such studies may only be interpreted as supporting or negating a null hypothesis and do not indicate causation in linking a genotype with a phenotype. We acknowledge that the present study has a larger sample size than the first study. However, there still exist a number of methodological concerns within the present study, which may call to question the authors’ conclusion.
First, the selection of controls should be representative of the general population from which the cases originate. It may be argued that the recruitment of controls using family doctor lists is not representative of the general population and could result in bias toward an unhealthy cohort with cardiovascular disease. In the past, many genetic association studies had controls obtained from Red Cross blood banks, but it is now recognized that this may significantly confound a study’s finding. We took this into consideration and designed the first study such that controls were recruited randomly from the electoral role of the Adelaide metropolitan area from where the stroke patients resided. We believe this to be a major strength of the original study.
Second, the ethnicity of controls in the present study was not reported, which raises the question of population stratification bias. This may be of particular relevance to the present study as a large number of residents in the greater London region are likely of noncaucasoid origin.
Third, in the current study the classification of lacunar stroke into subtypes dependent on the presence or absence of leukoaraiosis is of questionable relevance. Histopathological studies have shown that lacunar stroke is caused by either proximal microatheroma or lipohyalinosis, 2 entities which are distinctly different to the hyaline arteriosclerosis that underpins leukoaraiosis.2 These pathological subtypes are also likely to have different genetic determinants. The tPA –7351 C/T polymorphism is hypothesized to influence the risk of stroke by its effect on lysis of an occluding thrombus irrespective of the underlying vessel pathology.
Fourth, the current study also differed in methodology from the original study in that an adjustment for other important confounding influences was not considered. These possible confounders included medication usage, hyperlipidemia, and family history of cerebrovascular disease.
In sum, the conflicting results from these 2 studies investigating the association of tPA –7351 C/T polymorphism in lacunar infarction emphasizes that further large-scale prospective studies are required to clarify this current paradox. We contend that both studies have methodological deficiencies and thus the hypothesis remains valid.