Introduction

The age distribution of traumatic spinal cord injury (TSCI) exhibits a bimodal pattern with two distinct peaks [1]. The first peak is observed among young adults, typically between the ages of 15 and 29. This group is particularly prone to traumatic injuries caused by motor vehicle accidents, sports-related incidents, and falls. The second peak occurs among older adults, generally over the age of 65, primarily due to the increased risk of falls associated with aging [1, 2]

Pediatric TSCI can devastate the health and social well-being of affected children, their families, and society.Although neurological recovery in children with TSCI is thought to be better than in adults [3], the prognosis for most children is bleak, with complete SCI leaving patients functions requiring full assistance to perform their physiologic functions [4]. Furthermore, diagnosing TSCI in children is more challenging because a significant percentage of these patients present with SCI without obvious radiological abnormalities (SCIWORA) on X-ray or CT scans [5]. Therefore the use of MRI is necessary to accurately diagnose these cases.

In general, the causes of pediatric TSCI worldwide are transport-related injuries, sports/leisure activities, assault, and falls [4]. These causes are frequently avoidable, albeit occasionally inevitable. To ensure adequate preventive measures, physicians and potential children and families at risk must have a thorough understanding of the epidemiology of pediatric TSCI. However, data on pediatric TSCI is scant and fragmented. Furthermore, when compared to adults the heterogeneity of data and the lack of a clear global definition of pediatric TSCI incidence necessitate extensive epidemiologic research [6, 7]. Evidence-based knowledge about pediatric TSCI epidemiology is useful for preventing traumatic events and assisting physicians in tailoring treatment and rehabilitative programs to the needs of individuals based on their prognostic status. Furthermore, epidemiologic knowledge is important to determining the global distribution of pediatric TSCI. This can assist various health administrators and organizations in providing equitable infrastructure and financial aid.

Given the limited attention to pediatrics compared to adults in existing studies, the objective of this study was to offer comprehensive insights into pediatric TSCI epidemiology, uncovering differences in incidence, etiology, and injury severity between developed and developing countries. Additionally, the study acknowledged and tackled the inconsistency in defining "pediatrics," thereby enhancing the clarity surrounding this crucial facet of pediatric healthcare.

Material and methods

This systematic review and meta-analysis were carried out following the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [8]. Furthermore, the study was pre-registered with a predefined review protocol in PROSPERO (registration number CRD42020189757).

Eligibility criteria

The inclusion criteria are presented in Table 1. The studies were included if they provided data for the pediatric population (≤ 21 years) with the diagnosis of TSCI.

Table 1 Inclusion Criteria
Full size table

Search strategy

A librarian (MG) performed the search in PubMed, Embase, Web of Science, and Scopus all from 1900 to April 2023. The search syntax was performed by combining the MeSH terms and keywords of related articles and expert opinions. The detailed search syntax is presented in Table 2. The strategy for other databases is given in Supplementary files. The last search was carried out on 15th April 2023.

Table 2 Search Strategy used for PubMed database
Full size table

Selection process

After elimination of duplicate records, two reviewers (SFM and MADO) screened the titles and abstracts independently. In case of disagreement, a third reviewer (SBJ) was consulted to reach a consensus. The full texts of all potentially eligible abstracts were retrieved and assessed based on inclusion criteria by the same two reviewers independently. The reference lists of all included studies and reviews were screened to identify any additional relevant papers.

Data collection process

Three reviewers (SBJ, SFM, and MRF) independently extracted data from included full texts in accordance with the International Spinal Cord Injury Core and Basic Data Sets for children and youth with SCI [9, 10]. Any discrepancy was resolved by consulting a fourth reviewer (AG). A specialized Excel spreadsheet tailored to the study's objectives was used as the data collection tool. Discrepancies arising from data extraction were identified through a comparison of results, and agreement was reached through consensus.

Information on participant characteristics (sample size, age, sex, etiology, level of injury, completeness of injury, Frankel grade or American Spinal Injury Association (ASIA) impairment Scale (AIS), descriptive characteristics (publication time, country, period of observation), methodological details (study design, the extent of study (national, subnational) were extracted. Additional information for data extraction can be found in Supplementary files.

Critical appraisal

The assessment of the studies' quality was conducted using the Joanna Briggs Institute critical appraisal tool designed for evaluating observational studies [11]. Further details regarding critical appraisal can be found in Supplementary files.

Meta-analysis

For meta-analysis we separately analyzed developing and developed countries because it is known that there are significant differences in epidemiology of TSCI between developing and developed countries [12, 13]. We used the definition of the developing countries from the International Monetary Fund (IMF) 2021 update [14]. Only studies that were published after 1990 and had extractable data regarding age category of 0–15 years were included in meta-analysis. We considered pediatrics cut-off age of 0–15 years to be consistent with recent guidelines [7]. Based on data availability of studies we decided that the studies for 0–14 or 0–15 age groups were not sufficient for a distinct meta-analysis; however, we were able to perform meta-analysis after combining these two age groups. The details of data synthesis have been described in Supplementary files.

Results

Study selection

The PRISMA flow diagram displays the recognized studies at each stage (Fig. 1). The search resulted in the identification of 7828 records following the removal of duplications. After title screening, 681 abstracts were investigated, of which 87 full texts were included.

Fig. 1
figure 1

PRISMA 2020 flow diagram

Full size image

Methodological quality

The quality of 87 studies was evaluated. The quality score of studies ranged between 3 to 9 and only nine studies got the maximum score. Overall, 52 studies (60%) had medium quality, 27 studies (31%) had high quality and 8 studies (9%) had low quality. The details of individual study scores can be found in Supplementary files.

Review findings

Prevalence or incidence rate of TSCI in the pediatric population was reported in 42 studies [15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56] (Supplementary Table 1). The descriptive information of pediatric TSCI patients, level of injury and severity of TSCI were reported in 50 studies [15, 16, 20, 22, 23, 26, 31, 34, 36, 38,39,40,41,42, 48, 49, 52, 55, 57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88] (Supplementary Table 2). The etiology of pediatric SCI was reported in 41 studies [15, 16, 20, 22, 23, 26, 31, 34, 36, 39,40,41,42, 52, 57,58,59,60,61, 63,64,65,66,67, 69, 71,72,73,74,75,76,77,78,79,80, 82, 84,85,86,87,88] (Supplementary Table 3). Thirty-seven studies contained data for all-age TSCI (0–99 years) [17, 20,21,22,23,24,25,26,27, 30, 32, 37, 53, 54, 60, 71, 79, 81, 86, 87, 89, 59, 90,91,92,93,94,95,96,97,98,99,100,101,102,103,104]. These studies were used to estimate the proportion of pediatric TSCI (0–15 years) to all-age TSCI.

The incidence, gender, and severity of pediatric TSCI (0–15 years)

Twenty-one studies contained sufficient TSCI incidence data (number of cases and population at risk) for age category of 0–15 and were eligible for meta-analysis. The estimation of incidence in developed countries with meta-analysis among 18 studies was 4.3 cases per millions of children aged 0–15 (95% CI: 3.1; 6.0) Fig. 2). Only three studies reported TSCI incidence in developing countries including Tanzania and China where the incidence varied between 1.5 to 11.7 cases per millions of children aged 0–15 [37, 53, 54]

Fig. 2
figure 2

The pooled incidence of pediatric TSCIs in developed countries (Rate: case per million children aged 0–15)

Full size image

Taking into account gender distribution, a total of thirteen studies focusing on developed countries were incorporated into the meta-analysis. The analysis yielded an estimated male cases proportion of 67% (95% CI: 63%; 70%) within the pooled sample of 6,300 individuals from these developed countries (as presented in Appendix Fig. 1A). Only four studies reported gender proportion of 0–15 age group in developing countries. In these studies, male proportion ranged between 59% and 79% [64, 79, 82, 86].

Totally, eleven studies provided data about the completeness/incompleteness of TSCIs in age group 0–15, of which ten were related to developed countries and were included in the meta-analysis. The frequency of complete injury in the pooled sample of 6,396 individuals was 44% (95% CI: 28%; 59%, Fig. 3A) while the frequency of incomplete injury was 56% (95% CI: 40%; 72%, Fig. 3B). The frequency of complete injury among children aged 0–15 in developing countries was reported in Pakistan [64], India [82] and China [88] as 80%, 50% and 31%, respectively.

Fig. 3
figure 3

Severity (completeness vs incompleteness) of pediatric TSCIs meta-analysis in developed countries A Complete injuries B Incomplete injuries

Full size image

Regarding tetraplegia versus paraplegia we considered six studies related to developed countries that reported the severity of injury. The proportion of complete tetraplegia and incomplete tetraplegia in the pooled sample of 222 individuals in six included studies was 15% (95% CI: 8.1%; 23%, Appendix Fig. 1B) and 27% (95% CI: 14%; 43%, Appendix Fig. 1C), respectively. The complete paraplegia and incomplete paraplegia proportion in the pooled sample of 222 individuals in 6 included studies was 28% (95% CI: 11%; 55%, Appendix Fig. 1D) and 24% (95% CI: 16%; 32%, Appendix Fig. 1E), respectively. The frequency of paraplegia and tetraplegia among the 0–15 age group in developing countries were reported in Pakistan, and were 86% and 14%, respectively [64].

Etiologies of TSCIs

In the meta-analysis encompassing 16 studies focused on developed countries, the leading etiology of TSCIs was transport injuries, constituting the highest relative frequency. Specifically, the relative frequency of transport injuries within the pooled sample of 7,368 individuals was 50% (95% CI: 42%; 57%, as demonstrated in Fig. 4A). Notably, among five studies from developing countries, the relative frequency of transport injuries in the pooled sample of 401 individuals was 24% (95% CI: 13%; 37%, as indicated in Fig. 4A*).

Fig. 4
figure 4

Meta-analysis of the etiology of pediatric TSCIs. A Transport injuries in developed countries A* Transport injuries in developing countries. B Falls in developed countries B* Falls in developing countries. C Sports-leisure activity in developed countries C*Sports-leisure activity in developing countries. D Assault in developed countries D* Assault in developing countries

Full size image

Regarding falls, a meta-analysis of 16 studies pertaining to developed countries revealed a frequency of 10% (95% CI: 7.1%; 15%) within the pooled sample of 7,368 individuals (as shown in Fig. 4B). In contrast, within the context of developing countries, the relative frequency of falls among five studies comprising 401 individuals was notably higher at 31% (95% CI: 20%; 42%, as illustrated in Fig. 4B*).

The analysis of TSCI etiologies related to sports, involving twelve studies associated with developed countries, unveiled a frequency of 16% (95% CI: 12%; 21%) within the pooled sample of 7,293 individuals (depicted in Fig. 4C). Among studies from developing countries, involving 401 individuals across five studies, the relative frequency of sports-related injuries was 14% (95% CI: 1.7%; 34%, as presented in Fig. 4C*).

Regarding assaults as an etiology, among 10 studies covering developed countries and a pooled sample of 7,240 individuals, the frequency of assault-related TSCIs was 11% (95% CI: 7.1%; 17%, shown in Fig. 4D). Notably, within two studies from Brazil and Pakistan, assault emerged as the primary cause of TSCIs. Among four studies from developing countries encompassing 359 individuals, the relative frequency of assault-related injuries was 18% (95% CI: 5.1%; 37%, as seen in Fig. 4D*).

Furthermore, the frequency of other/unrecognized etiologies within the pooled sample of 7,293 individuals from 13 studies in developed countries was 14% (95% CI: 8.5%; 21%, as presented in Appendix Fig. 1H). In the context of developing countries, this relative frequency was notably lower at 5.6% (95% CI: 0.7%; 13%, as depicted in Appendix Fig. 1H*).

Level of TSCIs

In the meta-analysis focused on injury level, our consideration was limited to studies that categorized injuries into three distinct groups: cervical, thoracic, or lumbosacral. Within the pooled dataset of 2,897 individuals from six studies centered on developed countries, the frequency of cervical-level injuries was found to be 50% (95% CI: 41%; 58%, as shown in Appendix Fig. 1I). Moreover, thoracic-level injuries constituted 35% (95% CI: 23%; 47%, as presented in Appendix Fig. 1J), while lumbosacral-level injuries accounted for 13% (95% CI: 6.9%; 20%, as illustrated in Appendix Fig. 1K). Regarding developing countries only three studies had level specific data. In India [82] cervical level was injured in 50% of children 0–15 years. However thoracic level was more prevalent than cervical level in studies from Pakistan [62] and China [88].

TSCI proportion in pediatrics

For meta-analysis, 21 studies provided data about the proportion of pediatric TSCI (0–15 years) to all-age (0–99 years) TSCI in developed countries, while there were 16 studies related to developing countries. The frequency of TSCI related to pediatrics in the pooled sample of 130,945 individuals in developed countries was 3% (95% CI: 2.2%; 3.9%, Fig. 5A) while in developing countries in the pooled sample of 86,669 was 4.5% (95% CI: 2.8%; 6.4%, Fig. 5B).

Fig. 5
figure 5

The pooled Frequency of TSCIs related to pediatrics meta-analysis in A developed countries and B developing countries

Full size image

SCIWORA (0–17 years)

Based on data availability we calculated the frequency of SCIWORA in 0–17 years. In a pooled sample of 11,299 individuals from 7 included studies related to developed countries the frequency of SCIWORA was 35% (95% CI: 19%; 54%, Appendix Fig. 1L). There were no studies of SCIWORA in this age group in developing countries.

Sensitivity analysis

The results of sensitivity analysis are provided in supplementary files.

Discussion

In this comprehensive epidemiological study, we sought to bridge the difference among reports on the incidence of pediatric TSCI using a meta-analytical approach. Importantly, this epidemiological data can support and lead clinical and policy efforts to improve TSCI in children among marginalized groups. New et al. reviewed pediatric TSCI in 25 countries and found that pediatric TSCI incidence varied from 3.3–13.2 case per million population in different regions [7]. According to our meta-analysis, the estimated incidence of pediatric TSCI among developed countries was 4.3 cases per million children aged 0–15 (95% CI: 3.1–6.0 cases per million children aged 0–15).

Regarding the trend of TSCI over time we found an upward trend in two studies in Norway, and a constant or decreasing trend in seven studies form Australia, Netherlands, Spain and the U.S (Supplementary Table 1). The increasing trend in incidence of pediatric TSCI observed in Norway can be explained by a decrease in the rate of prehospital mortality, and higher number of transport-related injuries and more sports/leisure activities in pediatric population [20, 21]. It is also important to analyze the reasons behind the decreasing pattern of TSCI incidence. In the United States, Saunders et al. observed a significant decrease in incidence of pediatrics TSCI from 1998–2012 only in Whites. They also showed higher incidence of assault-related injuries in non-Whites in comparison to Whites (29% vs 5%) during the study period. Therefore, they suggested that future preventive measures should be focused on assault-related injuries in non-White pediatric population [39]. Two studies also observed a decreasing trend in incidence of pediatric TSCI in Spain, but they related this decrease to the decline in the frequency of transport-related injuries as a result of improvement in road-safety, and implementation of awareness raising campaigns [16, 22]. Lystad et al. observed that over a period of 10 years, the hospitalization rates for spinal cord and nerve injuries declined by 2.8% per year in Australia. However, contrasting trends were observed for spinal fractures and spinal sprains/strains, with rates increasing by 1.1% and 4.7% respectively. The authors also observed an increase in female hospitalization especially as a result of equestrian injuries [55].

The reported proportion of pediatric TSCI to all-age TSCI varies among individual studies and literature reviews. However, the rate of pediatric TSCI in most data are heterogeneous and mixed [6, 71, 105, 106]. The pooled proportion of pediatric TSCI in this study was 3% and 4.5% in developed and developing countries, respectively. These results suggest that the proportion of pediatric TSCI injuries is higher in developing countries compared to developed countries. This finding aligns with the findings of Ding et al. study which similarly highlighted that children from families with lower socioeconomic status face an increased risk of encountering SCI [107].

We observed a male predominance for pediatric TSCI. However, seemingly there is an increase in the trend of female spine injuries due to the recent increase in female sport activities since the latter part of the twentieth century [55].

De Vivo and Vogel found complete injury to be highest in children aged from 0 to 5, while incomplete injury was more common in ages between 13 and 15 years [66]. However, we found that most children are more susceptible to incomplete injury (56%) than complete injury (44%). The higher incidence of incomplete injury compared with complete injury can be attributed to the lower number of traffic accidents and improved initial care and treatment modalities in developed countries [22]. In our meta-analysis the frequency of complete paraplegia > incomplete tetraplegia > incomplete paraplegia > complete tetraplegia was observed in developed countries. Some publications discovered significant correlation between tetraplegia and incomplete injury, and paraplegia and complete injury [34, 60, 66, 78]. However, the exact explanation for this correlation is yet to be elucidated.

SCIWORA may be explained by the difference in elasticity between the spinal column and the spinal cord [100]. Notably, children below 8 years are at risk of SCIWORA and consequently complete or severe partial cord lesions [68, 70, 76, 105, 108, 109]. We estimated the frequency of SCIWORA to be 35% in developed countries. It indicates that a significant proportion of individuals in the studied population experienced spinal cord injuries without accompanying radiographic abnormalities. The best diagnostic technique for SCIWORA is MRI, which allows for the evaluation of extradural spaces as well as the integrity of the spinal ligaments [110].

Craniocervical junction and atlanto-axis (C1-C2) injuries are more common in children below 8 or pre-teens[3, 66, 111]. However, C4-C5 injuries are observed most in teens and adults[3]. The large head-to-torso ratio and the horizontally aligned vertebra in younger children account for the higher proportion of cervical injuries [4, 111, 112]. Interestingly, the craniocervical junction and atlanto-axis joint become more secure and less prone to injuries, as the child grows. [111]. By 8 to 10 years of age, the spine is thought to acquire a more adult position, with cervical fractures in children in this age group having a comparable spectrum to adult fractures [103]. While certain studies have suggested that the percentage of cervical injuries could range between 60% and 80% [105, 113], we found a relatively lower frequency of 50% for cervical injuries, with subsequent rates of 35% for thoracic injuries and 13% for lumbosacral injuries. Variations in the definitions used for categorizing and classifying injuries, as well as regional disparities in injury patterns, can contribute to differences in reported percentages.

The etiology of TSCI in pediatrics tends to differ from developing and developed countries. In our meta-analysis, we found that the most prevalent cause of pediatric TSCI in developed countries was transport injuries and falls were the least common while in developing countries falls were the leading etiology. This could be due to factors such as inadequate infrastructure, unsafe living conditions, or lack of safety measures in homes and public spaces. Falls from heights, such as from trees or buildings, may be more common in these settings. In addition assault related injuries were prevalent in developing countries such as Brazil [63] and Pakistan [64]. This suggests that intentional acts of violence, including physical abuse or assaults, contribute significantly to the occurrence of TSCI in pediatric populations in these regions. Sports-related injuries were found to be common causes of pediatric TSCI in both developing and developed countries. These injuries often occur in older children who engage in sports activities. Participation in contact sports or high-risk activities without proper safety precautions can lead to spinal cord injuries. Giving careful consideration to sports-related injuries is of utmost importance in the context of children. This significance stems from the fact that children have a relatively higher head-to-body ratio and their paraspinal muscle support hasn't fully matured [114]. As a result, they are more vulnerable to cervical injuries, which could potentially lead to paralysis and respiratory complications, ultimately carrying the risk of fatal outcomes [115].

Understanding the variation in etiology of pediatric TSCI in different countries is crucial for developing targeted prevention strategies, improving infrastructure and safety measures, and providing appropriate medical care. Our findings highlights the need for context-specific approaches to prevent and manage pediatric TSCI based on the prevalent causes in each country or region.

Limitations

The most important limitation of our study was the fact that the outcome reporting of studies was highly heterogeneous. Various age cut-offs were used as definition of pediatric population, furthermore different age sub-groups were used which precluded us from performing a sub-group meta-analysis of age-specific level, severity and etiology of injury. Additionally, there were only few studies from developing countries, which hindered us from performing a meta-analysis of incidence rate, gender proportion, injury completeness, severity, and level of TSCI in these countries. Overall, we could find data only for 29 countries, most of which were developed countries, which can reduce the generalizability of data. Most of the included studies were medium or low quality which can increase the risk of bias in the published articles.

Conclusion

This systematic review, to our knowledge, is the first to use a meta-analytic approach to describe the epidemiology of pediatric TSCI. According to our meta-analysis, the proportion of pediatric TSCI to all age TSCI was higher in developing countries than in developed countries. Developed countries have an estimated incidence of 4.3 cases per million children aged 0–15. Regarding the level of TSCI, cervical injuries were more than thoracic and lumbosacral injuries in developed countries. The most common etiology of TSCI were transport-related injuries in developed countries and falls in developing countries. The distinct epidemiological features identified in TSCIs have the potential to guide the development of targeted and cost-effective preventive strategies, aimed at reducing the incidence and overall impact of TSCIs in the pediatric population.

Implications for future studies

In order to improve the quality of studies and to ensure the comparability of data, we encourage global spine fracture and SCI registry and future registry-based studies to use a standardized form of data reporting. The International Spinal Cord Injury Core data set provides useful recommendations in this regard [9, 10].