Clinical, lifestyle, socioeconomic determinants and rate of asymptomatic intracranial atherosclerosis in stroke free Pakistanis

Stroke is a leading cause of adult disability worldwide. It is also thought to be the 2nd most common cause of global mortality [1]. Among ischemic stroke sub-types, intracranial atherosclerotic disease (ICAD) is regarded as the most frequent etiology of stroke with high prevalence among Asian, African and Hispanic individuals [2]–[8]. Early data from Pakistan has also indicated a high prevalence of ICAD [9].

Symptomatic atherosclerotic disease is preceded by asymptomatic disease [10]. This provides us with an opportunity for prevention [11]. For ICAD, prevention is key since few relevant interventions exist once the disease is symptomatic, none of which are easily implementable in resource strapped regions [12]. However, in order to formulate effective mechanisms of preventing stroke, it is imperative to first identify the early determinants of ICAD. Although these determinants have been extensively studied in the Western population, data is not always applicable to a population such as the one in Pakistan due to different demographic and lifestyle characteristics. Despite the high prevalence of ICAD, data from Asian regions that bear large burdens of ICAD is relatively scarce.

The objective of this study was to determine the clinical, lifestyle, dietary and psycho-social determinants of asymptomatic ICAD in Pakistanis who do not report stroke symptoms on systematic review. These determinants would help identify high risk behavior for stroke in the Pakistani population.

Out of 283 subjects approached at the two study sites (197 at DUHS, 86 at AKUH), 200 were recruited for this study (156 at DUHS and 44 at AKUH). Of those excluded from the study most patients had stroke as an indication for MRI (37.3%) or did not consent for the study (33.7%) (Figure 1). The most common indication for the MRI was headache (52%). Out of the 200 who participated in the study, 69 (34.5%) were found to have clinically significant ICAD.

To the best of our knowledge this is the first study conducted in South Asia that identifies early determinants of ICAD by looking at asymptomatic subjects. The major finding of this study is that modifiable risk factors such as dyslipidemia, tobacco chewing and overconsumption of a Western diet, which are all increasing in prevalence in our population are associated with asymptomatic ICAD. The most commonly affected arteries were the Posterior Cerebral arteries followed by Vertebral and Middle Cerebral arteries. Of all affected arteries, about 20% showed complete occlusion in clinically silent areas. Furthermore, it also showed a high prevalence of depression and sedentary lifestyle in the study population, both of which are stroke risk factors. The findings provide us with direction for implementation of initiatives that target primary prevention.

The proportion of patients with ICAD (34.5%) was significantly higher than that reported in previous studies from other parts of the world [37]–[39] that have reported figures between 4% and 29%. Although a figure as high as this cannot be ignored, the recruitment bias introduced choosing participants with a preexisting indication for an MRI would suggest that this figure might be falsely elevated.

The study had a young (mean age 37.1 yrs) population in general but importantly there was no significant difference in the ages of the subjects with and without ICAD. The finding is not consistent with previous similar studies [10],[35],[40],[41] from other parts of the world which report an association between increasing age and ICAD. However, this observation can still be explained if the findings of a previous study on the Pakistani population [16] is taken into consideration. Atherosclerotic disease takes years to develop and the study mentioned above showed the mean age of patients of symptomatic ICAD to be 50.1 years. This essentially suggests that a significant number of young apparently healthy Pakistanis have asymptomatic ICAD which can eventually progress to symptomatic ICAD. Further, ICAD is not entirely explained by traditional risks like hypertension, diabetes, and dyslipidemia, and emerging factors like plasma homocysteine, angiotensin-converting enzyme polymorphism, and inflammatory cytokines may affect its development. These biochemical investigations were beyond the capacity of this study [42]–[44].

Stroke is common in males and this observation is more pronounced in ischemic strokes [45]–[47]. It also occurs at a younger age in males compared to females [48],[49]. With a young study population, as in this study, the same finding would be expected. This study however does not report a gender predilection. This finding, although not common, has been shown in previous studies [35],[40] from other Asian countries. Interestingly a study [50] by Jafar et al. on 3143 Pakistani adults did report that women in Pakistan were at higher risk than men for coronary artery disease due to higher number of risk factors (diabetes, hypertension, obesity, dyslipidemia). Since the same risk factors as applicable to ICAD, we theorize that the equal risk of vascular disease observed in the cardiac population also applies to the cerebrovascular risk patients.

When looking at the lifestyle factors, our study findings suggest that a high socioeconomic status is significantly associated with stroke. Review studies [51],[52] that have specifically addressed the association between socioeconomic status and stroke have documented that it is in fact low socioeconomic status that is associated with higher incidence and severity of stroke. We suggest that lifestyle choices associated with the higher socioeconomic status such as adoption of Western diet have contributed to a higher incidence of stroke in our transitional population [53]. What adds weight to this argument is that the western diet has also been shown to be significantly associated with ICAD in our study (Adjusted PR 2.86 (1.11-7.41) for quintile 4 and Adjusted PR 3.04 (1.15-7.99) for quintile 5). There is documented evidence of the proinflammatory effects of Western diet [54]–[57] especially when taken in large quantities. The finding emphasizes the need to address the adaptation of Western diet before it adds to the burden of non communicable diseases in Pakistan which is now becoming endemic.

The use of smokeless tobacco is Pakistan is rampant with an estimated prevalence between 16% and 26% [58]–[61]. Overall prevalence of smokeless tobacco use was 12% in our study which was higher in participants with ICAD compared to those without. The association was of ICAD was found to be significant after multivariate analysis (Adjusted PR 3.27) reiterating the increased risk of stroke with use of smokeless tobacco which has documented before [62].

The causal relationship of diseases such as dyslipidemia, hypertension and diabetes with ICAD is well studied and hence we did not look to screen patients for these diseases. The presence or absence of these diseases was mostly based on history and a single blood pressure reading and random blood sugar test. The ascertainment bias resulting from this methodology could have contributed to the finding that apart from dyslipidemia, no other diseases such as hypertension and diabetes showed an association with ICAD. At the same time what also needs to be considered is the fact that a lot of these subjects were young and they might develop these diseases in the future and ICAD has simply manifested earlier due to lifestyle choices.

The association of depression with intracranial disease has been studied before and it has been suggested that small brain infarctions can cause depression [63],[64]. Through our study we attempted to establish an association, if any, of ICAD and depression. Recent data [65]–[68] reports the prevalence of depression among the residents of Karachi between 16% and 70%. Taking this into consideration we used a high cut off score for the SRQ (>8) to increase the specificity (Positive predictive value is 98% at this SRQ score). The study however still showed depression in more than half the subjects. With such a high figure, distributed equally across the study population, it is very difficult to establish any association. Similarly, more than 80% of the participants had a low physical activity level. With a prevalence as high as this, it is not possible to establish an association.

We feel that our preliminary observations have impact and relevance in a global context.

In a resource strapped country like Pakistan, we feel the key to tackling a health issue like ICAD lies in primary prevention. Evidence currently available for the therapies of ICAD [69]–[73] from developed countries shows that these interventions are not only expensive but often ineffective, and difficult to execute at best in low middle income settings.

Our suggestion to focus on controlling non communicable risk factors to reduce the burden of disease is not novel to low middle income regions. Our review of these initiatives reveals interventional projects in similar settings that addressed non communicable diseases and showed promising results [74]–[79]. What is lacking are holistic, stroke based or multiple risk initiatives.

Our study has several strengths. Our participants were chosen from the general population rather than from a high risk population. A detailed review of asymptomatic status was done by using a validated questionnaire (QVSFS) which to rule out clinical stroke with 95% sensitivity and specificity. This tool has also been used in Pakistan previously [16]. Rigorous training was given to all data collectors so that a standard protocol was implemented at both centers. A round the clock back up was available by stroke neurologists further adding clarity to the participant stroke free status. Both study centers were also equipped with a similar MRI scanner (1.5 Tesla) using standardized protocol thus reducing detection bias. Centralized coding system for MRA and specialized software was used for storing and image viewing. This study also assessed a number of novel risk factors for ICAD like socio-demographic factors, physical activity, diet and depression that have not always been explored using previously validated tools.

Our study has certain limitations. Since this is a cross sectional study, temporality cannot be assessed and no conclusions can be deduced regarding causality.There is a recruitment bias when we selected our participants from those who had a preexisting indication for an MRI. Secondly, the medical risks reported here (Hypertension and Diabetes) are self-reported, so that there may have been under-reporting and recall bias. This study also has a diagnostic limitation since we did not use a 4 vessel angiogram for diagnosing ICAD. This is because 4 Vessel angiography has a 1.5 - 3% risk of stroke [80]. Using this on otherwise healthy stroke free patients would be unethical. We used the MRI instead, which is a noninvasive tool with a high negative predictive value [81]. Using biochemical investigations such as homocysteine levels would have been relevant to identifying determinants of ICAD; however this was beyond the scope of this study, as was the opportunity to do risk loading stratified analysis based on degree of stenosis or the presence or absence of stroke and stroke brain volumes, both of which are avenues for further investigations. Furthermore, our data essentially applies to urban transitional Pakistan since rural based populations are not well represented in this study.

In conclusion, we have demonstrated that the distribution of ICAD in young Pakistanis is frequent and equally distributed across gender and is associated with modifiable risks. Future studies should focus on longitudinal assessment of ICAD, and holistic risk reduction based interventions on younger populations.

Ayeesha Kamran Kamal and Farzin Majeed is Joint First Authors.