Introduction

Stroke is one of the diseases with the greatest impact on public health, being the third cause of mortality and the leading cause of disability in the Western world since most patients suffer sequels, which is 30% of cases disable them to perform daily activities¹,².

The proportion and frequency between the ischemic types and intracerebral hemorrhage (ICH) are variables according to a determined region; while in Spain, the rates of ICH are around to 12%³, in regions of South America such as Ecuador and Chile the ICH reach rates between 37 and 46%⁴. In Paraguay, cerebrovascular disease is the fourth cause of death⁵. However, another data about the baseline characteristics in our population are not widely known.

Among the cardiovascular risk factors related to stroke, hypertension is the most frequent and important risk factor. The prevalence is around 70% in both ischemic and hemorrhagic patients, and its optimal control is possibly one of the most important interventions in secondary prevention for both types of stroke³,⁶. In Paraguay, the prevalence of hypertension is high, with rates of 53% in men and 40% in women between 35 and 45 years, reaching up to 81% in the general population between 65 and 74 years⁷.

Another known and controversial stroke risk factor is gender, and their disparities in baseline and clinical outcomes. Several descriptive studies have shown an association of older age, frequency of hypertension, obesity, and atrial fibrillation in women than in men⁸,⁹.

Methodology

This is a descriptive, observational study in a single center. Data were collected from a prospective of stroke patients admitted in the Stroke Unit of our center from April 2015 to March 2019.

The known modifiable and non-modifiable risk factors have been recorded. In patients with hemorrhagic stroke, the etiology of hypertensive ICH has been determined according to the history of hypertension, the topography of the parenchymal lesion, and the vascular study¹⁰. The diagnosis of ischemic stroke was made according to the criteria of the Oxfordshire Community Stroke Project and the Trial of ORG 10172 in Acute Stroke Treatment¹¹.

The descriptive statistical analysis was performed for all the parameters, which included measurements of the central tendency and dispersion for the quantitative variables, as well as absolute and relative frequencies for the qualitative variables, with their 95% CI in both cases. If the data did not meet the assumptions of normality for the analysis, non-parametric statistical methods were used (Student's t-test or Mann–Whitney test, respectively). The Chi-square test was applied for the comparison between the proportions. To determine the weight of each of the risk factors between ischemic and hemorrhagic stroke, multiple logistic regression models were established, with a 95% CI. The statistical tests were performed with a significance level of 5%. The statistical program SPSS v. 23.0 was used to perform the analysis.

Results

From 996 patients with stroke, 744 (74.7%) were ischemic, and 252 (25.3%) were ICH, and 390 (39.1%) were women. The mean age was 62.8 years (SD ± 15.3). Overall, 75.4% had known antecedent of hypertension, 27% DM, and 6.4% dyslipidemia. The median baseline NIHSS was 8 (IQR: 4-12). The baseline characteristics in general and of the ischemic and hemorrhagic groups are shown in table 1.

In ischemic stroke patients, the mean age was 64.3 (SD ± 15.2) years, the median of NIHSS was 8 (IQR 4-12), and 39.8% were women. Patients with ischemic stroke were significantly older (mean years 64.3 SD ± 15.2 vs. 58.1 SD ± 14.8), had higher glycemia at admission (mean mgr./dl 168.6 SD ± 97.7 vs. 155.9 SD ± 120.3 p = 0.513), frequency of diabetes mellitus (23.5% vs. 3.4% p = 0.001), atrial fibrillation (13.6% vs. 1.3% p = 0.001), prior history of stroke (15.4% vs. 2.5% p = 0.001), prior antiplatelet treatment (8.1% vs. 1.0% p = 0.001), and lower systolic blood pressure on arrival than patients with ICH (mean mmHg, 169.8 SD ± 38.3 vs. 189.9 SD ± 43.2).

The most frequent etiology was (422 cases work-up performed) cardioembolic 41.2%, indeterminate in 24.1%, lacunar infarcts in 11.6%, and atherothrombotic in 21.5%. Hypertension was present in 91% of atherothrombotic strokes, 89.6% of lacunary strokes, 80.8% of cardioembolic, and 77.6% of indeterminate strokes (Fig. 1).

Figure 1 Graphic bar showing the number of cases of ischemic stroke etiology according to prior hypertension status. 

ICH patients had a mean age of 58.1 (SD ± 14.8) years, the median of NIHSS was 9 (IQR 5-13), and 36.9% were women. From 142 cases with etiological workup, 79.5% were due to hypertensive hemorrhage. The factors significantly related to hypertensive hemorrhage (113 cases evaluated) were hypertension (p ≤ 0.001) and the lower age stratified by groups (p = 0.006) (Fig. 2).

Figure 2 Graphic bar showing the number of cases of intracerebral hemorrhage according to stratified age group. 

In the multivariate analysis adjusted for sex, hypertension, diabetes mellitus, previous stroke, and atrial fibrillation, age < 60 years old (OR: 4.893, 95% CI: 1.772-13.509, p = 0.029) and a higher systolic blood pressure at admission (OR: 1.098, 95% CI 1.044-1.155, p = 0.009) were independent factors associated with ICH.

In relation to gender, hypertension was present in 79.3% in men and 83.5% in women. In men, smoking habit was significantly more frequent (23.9% vs. 3.4% p < 0.001) and they presented more ischemic heart disease (7.9% vs. 2.3% p = 0.001) than women. Men were younger (61.75 SD ± 13.6 vs. 64.16 SD ± 17.4 p = 0.003), being in the ICH group, even more remarkable the difference (mean years 55.41 [SD ± 14.8 vs. 62.48 [SD ± 15.2 p ≤ 0.001) (Fig. 3).

Figure 3 Bars showing the mean age according to gender in overall A: ischemic; B: intracerebral; and C: stroke patients. 

Discussion

This is a descriptive study, with data from a prospective cohort over a period of 41 months. In general, there is a higher proportion of men than women with stroke, and the mean age is significantly lower than in developed countries¹,³, but similar to previous studies in Buenos Aires and Brasilia¹²,¹³. Hypertension is the most frequent risk factor, slightly higher than other series³,¹²-¹⁴. Another risk factors rate such as DM, previous stroke, and atrial fibrillation were similar to prior studies; however, the frequency of dyslipidemia, known ischemic cardiopathy, and smoking was strikingly lower³,¹²-¹⁴. Possible explanations (in addition to the inherent biases and limitations from the nature of this study) are that our center, although it is located in an urban center, is a reference center that receives patients from the rural area also, and this could affect our results. Another reason would be the lack of knowledge of the risk factors of our population (less than half of men know about their blood cholesterol levels)⁷.

The rate of HIC is significantly higher than that described in developed countries³,¹² and is consistent with results from our region in studies conducted in Chile, Ecuador, and Argentina¹⁴-¹⁶. These results are explained by the high prevalence rate of hypertension in our population, especially manifests at early age in adult⁷.

In relation to ischemic stroke, the etiology by TOAST (in 422 cases) showed a high rate of cardioembolic (41.2%), however a lower frequency of atheromatous disease (21.5%) and lacunar (11.6%). These results are more similar to those found in Santiago¹⁴ than in those shown in Brasilia, Guayaquil, and Buenos Aires¹³,¹⁵,¹⁶. It should be noted that in South America the frequency in relation to the etiology of stroke is very heterogeneous and diverse, mainly due to ethnic, geographical, and sociocultural differences⁴. Finally, the relatively high rate of indeterminate etiology would be related in many cases to an incomplete study and is a clear limitation in the present study.

Interestingly, while hypertension rates are high (79.3%) in ischemic patients, the percentage of lacunar strokes for presumed hypertensive microangiopathy as a cause is lower than expected (11.6%). In contrast, hypertensive hemorrhage represents the cause of up to 79.5% of ICH. Previous studies in relation to this possible common substrate have shown that the cerebral vascular phenotype (ICH vs. lacunar infarcts) due to hypertensive microangiopathy would be related in part to age and cholesterol levels¹⁷,¹⁸. Therefore, a possible explanation to our findings could be related to age (remarkably younger patients) and the lower frequency of dyslipidemia (5.6%) known in our population. Another possible explanation would be in relation to the less severe clinical deficit associated with lacunar stroke¹⁹, which may affect the arrival or referral of these patients. These results are consistent in the region with previous studies in Santiago de Chile and Bogotá¹³,²⁰. Unfortunately, the body mass index, which has been shown to be a prognostic factor for the development of ICH or lacunar infarctions in hypertensive microangiopathy¹⁸, has not been evaluated in the present study. Future prospective and comparative studies should be taken into account to confirm these findings.

Regarding gender, women presented a stroke less frequently, with greater age (according to the literature)⁸,²¹,²², and significantly less known cardiovascular risk factors than men, which is also consistent with previous studies. Factors such as hypertension and atrial fibrillation, which in other series showed a significantly higher frequency in women, have not been seen in our study⁸,⁹. The age difference in HIC is remarkable among genders, being the higher hypertension prevalence in early age⁷ and the hypothetical hormonal preventive effect in women²⁰,²¹ probable reasons for this finding.

As a relevant limitation at the present study; it is a monocentric study, involving patients from urban and rural areas that may not represent a specific geographic area.

Conclusion

In this prospective study in our country, hypertension is a risk factor highly related to stroke with a higher frequency of hypertensive hemorrhages in younger patients than other series, especially in men. The frequency of different stroke subtypes would be related to environmental, sociocultural, and biological factors. Multicenter and comparative studies are warranted to confirm these findings. The intensification of public health programs focused on the control of risk factors, especially hypertension, is necessary to prevent cerebrovascular disease in our environment.