An international perspective on young stroke incidence and risk factors: a scoping review – BMC Public Health

Search of four databases identified a total of 5750 articles. After screening and removal of duplicates, 471 articles were included in this review. Details of search results per databases have been reported (Fig. 1) in line with the Preferred Reporting Items for Systematic Reviews and Meta-analyses extension for scoping review (PRISMA-ScR). The extracted data is reported in three sections: characteristics of included studies, risk factors of stroke and incidence of stroke. Furthermore, the retrieved data is reported individually according to age group categories of stroke, namely perinatal, paediatric, and young adults. Key indicators reported for each age group category of stroke include research design, categorisation of the number of studies by type of stroke, risk factors, countries of study, and income level.

Characteristics of included studies

Among the 471 articles included, the year of publication ranged from 2012 to 2022, while the reported study periods were from 1969 to 2020. There were 224 (47.6%) cohort studies, 164 (34.8%) case studies/case series, 35 (7.4%) reviews, 30 (6.4%) case-control and a combination of designs constituted 18 (3.8%). Concerning total reports on incidence and risk factors, some studies reported on more than one age group, and others reported on both incidence and risk factors therefore data for studies reporting on more than one age group had to be duplicated for each age group category. The same approach was adopted for studies on both risk factors and incidence, thereby conflating the total number of studies included. In view of that, there were 822 reports on incidence and risk factors, with 518/822 (63%) reports on risk factors and 304/822 (37%) reports on incidence. A high proportion of studies (n = 470/822, 57.1%) were in paediatric stroke, with young stroke and perinatal stroke studies contributing (n = 233/822, 28.4%) and (n = 119/822, 14.5%) respectively. Data were retrieved from 50 different countries, with most of the studies (n = 199/471, 42.3%) conducted in high-income countries, studies from upper and middle income were (n = 38/471, 8.1%), lower middle-income (n = 39/471, 8.3%), low-income (n = 3/471, 0.6%), international study (n = 7/471, 1.5%) and a further 185/471 (39.3%) articles did not state the country of research. Most of the studies were conducted in the USA (n = 69) followed by UK (n = 21) and China (n = 19). Full details on the authors, aim, sample size, gender, and data collection method can be found in the supplementary file Addendum S3. Figure 2 shows the percentage of studies on risk factors and incidence per age group category.

Percentage of studies on risk factors and incidence by age category

Risk factors of stroke

Risk factors are attributes, characteristics, or exposures that heighten an individual’s probability of developing a disease or health disorder [32]. Conversely, aetiology pertains to the actual cause or the specific factors that instigate the onset of a disease [33].

Despite the distinct meanings of these terms, they were often used interchangeably across studies reviewed. For example, congenital heart disease was classified as an aetiology of stroke by some authors, while others labelled it as a risk factor—this interchangeable use of terms introduced challenges in distinguishing between the two. Consequently, for this review, we included studies reporting the aetiology of stroke under the broader category of risk factors.

There were (n = 259/471) studies on risk factors and (n = 39/471) on both incidence and risk factors. The studies that reported on both the incidence and risk factors were included in the group of studies that focused on risk factors. In addition, studies reporting on more than one age group were also duplicated therefore total reports on risk factors for stroke was 518/822 (63%) across the age groups. There were varying ways studies reported the risk factors of stroke. For example, some research reported risk factors based on either haemorrhagic or ischaemic stroke, while others were based on general stroke, including both types of stroke. The reporting of the risk factors was categorised based on the age groups.

Risk factors for perinatal stroke

Out of the total studies on risk factors, 68 out of 518 (13.1%) were reported on perinatal stroke. The majority of these studies were cohort studies (58.8%, 40/68), while case-control studies made up the smallest proportion (10.2%, 7/68). Furthermore, a considerable proportion of the studies (48.5%, 33/68) focused on ischaemic stroke. Studies from high-income countries accounted for most of the data on risk factors for perinatal stroke and the top four (4) countries reporting were the USA (n = 14), United Kingdom (UK) (n = 4), Canada (n = 4) and Australia (n = 4). There were multiple risk factors for perinatal stroke and the top three (3) frequently reported were infections (neonatal sepsis, meningitis, necrotising enterocolitis, maternal sepsis), cardiac conditions (congenital heart disease, cardiac defects, aortic coarctation) and intrapartum/foetal related factors (Apgar score 2 provides a summary of risk factors for perinatal stroke. Further details of study characteristics can be found in supplementary file Addendum S4.

Risk factors for paediatric stroke

Risk factors for paediatric stroke were reported in 305/518 (58.9%) of total studies on risk factors. Of paediatric stroke studies reviewed, cohorts and case reports accounted for 146/305 (47.9%) and 100/305 (32.8%), respectively. There were 140/305 (45.9%) studies on ischaemic strokes, 72 on haemorrhagic strokes, and 93 on both haemorrhagic and ischaemic strokes. Most of the studies (n = 128/305, 41.9%) were conducted in high-income countries, with only one study from a low-income country; 124/305 (40.7%) of the paediatric stroke studies had the country of research not stated. The top four (4) countries reporting data on risk factors for paediatric stroke were the USA (n = 43), China (n = 13), the UK (n = 10) and Canada (n = 10). There were multiple risk factors for paediatric stroke and the top three (3) frequently reported were vasculopathies (arteriovenous malformation, Moyamoya disease, cervical arterial dissection), infections (otitis media, Human immunodeficiency Virus (HIV) infection, meningoencephalitis) and cardiac conditions (aortic coarctation, patent foramen ovale, cardiomyopathy). Table 3 provides a summary of risk factors for paediatric stroke. Further details of study characteristics can be found in supplementary file Addendum S5.

Risk factors for young stroke

Risk factors for young stroke were reported in 145/518 (28.0%) of the total studies on risk factors. Out of the young stroke studies reviewed, cohorts and case reports accounted for 71/145 (48.9%) and 55/145 (37.9%), respectively. Studies conducted on ischaemic stroke constituted a high proportion of 55 (45.9%) with 42 haemorrhagic strokes and 48 (both haemorrhagic and ischaemic stroke). Studies from high-income countries accounted for majority 60/145 (41.3%) of the studies, with only one study from a low-income country. Almost a quarter (n = 52/145, 35.9%) of the young stroke studies did not state the country of research. The top four (4) countries reporting data on risk factors for young stroke were the USA (n = 14), China (n = 9), India (n = 7) and UK (n = 6). There were multiple risk factors for young stroke and the top three (3) frequently reported were chronic conditions (diabetes mellitus, hypertension, migraine, nephrotic syndrome), vasculopathies (arteriovenous malformation, cerebral artery aneurysm, Moyamoya disease) and cardiac conditions (coarctation of the aorta, congenital heart disease, atrial fibrillation). Table 4 provides a summary of risk factors for young stroke. Further details of study characteristics can be found in supplementary file Addendum S6.

Incidence of stroke

Data on the incidence of stroke among the younger population ≤ 30 years old was reported in 304/822 studies. Some studies reported incidence data only while others reported on both incidence and risk factors. Each of the studies on both incidence and risk factors were duplicated for risk factors and incidence. Hence, the high number of studies exceeding the overall total. Data on the incidence of paediatric stroke was the most reported (n = 165/304, 54.3% ), followed by young stroke (n = 88/304, 28.9%) and perinatal stroke (n = 51/304, 16.8%).

Incidence by stroke type

Most of the incidence data reported were on both haemorrhagic and ischaemic stroke across all categories of stroke. Types of stroke by category is presented in Fig. 3.

Number of studies on the incidence of stroke by stroke type and age category

Incidence of stroke by income level

Majority (n = 187/304, 61.5%) of the incidence data were from high-income countries. The top 3 reported countries for perinatal stroke were USA (n = 9), Canada (n = 6) and Australia (n = 4). For paediatric stroke studies, USA (n = 35), Canada (n = 10) and China (n = 8) were the top 3 most reported countries. USA (n = 9), Australia (n = 6) and China (n = 6) were the top 3 countries reporting the incidence of stroke among young people.

Calculation of incidence of stroke across perinatal, paediatric and young adult age groups

Stroke incidence data were heterogeneously reported across all categories of stroke in terms of the unit of measurement, population sampled and age groupings. Studies reporting on the incidence of stroke across perinatal, paediatric, and young adults had varying age ranges. Figures 4 and 5, and 6 present details about variations in age groupings for incidence data.

Heterogeneity of reporting perinatal stroke incidence data

Heterogeneity of reporting paediatric stroke incidence data

Heterogeneity of reporting young stroke incidence data

Heterogeneity of reporting based on the unit of measurement and sampled population

The reporting of incidence across all age categories varied, including a mix of hospital-based [34,35,36,37,38] and population-based data [39,40,41] with measurements such as per 100,000 person-years (py), per 1000 population, percentages and absolute numbers. Table 5 provides details of the variations in the unit of measure across the categories of stroke. A summary of heterogeneity across stroke categories is presented in Fig. 7.

Summary of the Percentage of Studies on Incidence Per Unit of Measurement

**Other measurement includes percentages, absolute numbers, per 1000 LB, per 1000 births, per million admissions.