International surveillance study in acute spinal cord injury confirms viability of multinational clinical trials

Abstract

Background: The epidemiological international landscape of traumatic spinal cord injury (SCI) has evolved over the last decades along with given inherent differences in acute care and rehabilitation across countries and jurisdictions. However, to what extent these differences may influence neurological and functional recovery as well as the integrity of international trials is unclear. The latter also relates to historical clinical data that are exploited to inform clinical trial design and as potential comparative data.

Methods: Epidemiological and clinical data of individuals with traumatic and ischemic SCI enrolled in the European Multi-Center Study about Spinal Cord Injury (EMSCI) were analyzed. Mixed-effect models were employed to account for the longitudinal nature of the data, efficiently handle missing data, and adjust for covariates. The primary outcomes comprised demographics/injury characteristics and standard scores to quantify neurological (i.e., motor and sensory scores examined according to the International Standards for the Neurological Classification of Spinal Cord Injury) and functional recovery (walking function). We externally validated our findings leveraging data from a completed North American landmark clinical trial.

Results: A total of 4601 patients with acute SCI were included. Over the course of 20 years, the ratio of male to female patients remained stable at 3:1, while the distribution of age at injury significantly shifted from unimodal (2001/02) to bimodal distribution (2019). The proportional distribution of injury severities and levels remained stable with the largest percentages of motor complete injuries. Both, the rate and pattern of neurological and functional recovery, remained unchanged throughout the surveillance period despite the increasing age at injury. The findings related to recovery profiles were confirmed by an external validation cohort (n=791). Lastly, we built an open-access and online surveillance platform ("Neurosurveillance") to interactively exploit the study results and beyond.

Conclusions: Despite some epidemiological changes and considerable advances in clinical management and rehabilitation, the neurological and functional recovery following SCI has remained stable over the last two decades. Our study, including a newly created open-access and online surveillance tool, constitutes an unparalleled resource to inform clinical practice and implementation of forthcoming clinical trials targeting neural repair and plasticity in acute spinal cord injury.

Keywords: Aging; Benchmark; Epidemiological shift; Functional recovery; Neurological recovery; Spinal cord injury; Surveillance study.

Fig. 1

Study overview and result from the main cohort. A Flowchart of the included and excluded patients that were originally enrolled in the European Multi-Center Study about Spinal Cord Injury (EMSCI) study. Almost 90% of the EMSCI patients met our inclusion criteria. B Number of patients recruited between 2001 and 2019 per country. The majority of patients were admitted to centers in Germany, Switzerland, and Czech Republic. Note: The Indian center joined the EMSCI network only in 2011. C Annual ratio between female and male individuals with spinal cord injury enrolled in the EMSCI. Between 2001 and 2019, the ratio between men and women sustaining a traumatic or ischemic spinal cord injury remained comparable at 3:1. D Change in distribution of age at injury. Over the last two decades, a shift in age at injury was observed for individuals with spinal cord injury. In comparison to early 2000s, which were characterized by a unimodal distribution, the proportion of elderly people sustaining a traumatic spinal cord injury increased significantly. E Baseline injury severity. While there are some fluctuations, the proportions of injury severities, as measured by AIS scores, remained constant across the study period. F Baseline level of injury. The proportion of cervical, thoracic, and lumbar injuries did not significantly change as a function of time

Fig. 2

Neurological and functional recovery throughout the surveillance period. The recovery trajectory profiles of A the motor function, B functional independence, and C walking function remained comparable across the surveillance period. In other words, the degree a person with spinal cord injury spontaneously recovers motor and walking function as well as functional independence within 1-year post-injury is the same now as it was two decades ago. The solid lines represent the fitted models and the shaded areas the standard error. The inserted boxes illustrate the robustness of the recovery profiles across all years for patients with AIS-C injuries. For all other injury severities, please refer to the supplementary material section

Fig. 3

Comparison of sensorimotor recovery between data sources. A The pattern and degree recovery of motor and B sensory function of patients enrolled in the EMSCI were comparable to those of patients from the Sygen study (C and D). The heat plots and the number in the tiles represent the mean of motor and sensory scores, respectively. The progression of upper extremity motor scores is only shown for individuals with a tetraplegic spinal cord injury. Note: Individuals with paraplegic spinal cord injury have, by definition, full function in the upper extremities (i.e., UEMS of 50)

Fig. 4

Overview of the validation study. A The ratio between male and female individuals with a spinal cord injury. Depending on the year, the ratio of male and female spinal cord injury individuals changed between 3:1 and 4:1. B Distribution of age at injury. Throughout the clinical trial period, there was no change in distribution of age at injury. Important to note, the average age at injury of the Sygen clinical trial cohort, independent of sex, was significantly lower compared to the EMSCI cohort. C Baseline injury severity and D injury level: The proportions of injury characteristics remained constant between 1992 and 1997. E Motor and F sensory recovery stratified by AIS grade and plegia (i.e., paraplegia or tetraplegia). The solid lines represent the fitted models and the shaded areas the standard error