Investigating the Upper Limb During the First 6 Months After Sustaining a Cervical Spinal Cord Injury,

INTRODUCTION Sustaining a cervical spinal cord injury (SCI) leads to tetraplegia with paralysis in all four extremities. It is a traumatic event that radically changes an individual's ability to perform activities of daily life. Our arms and hands are critical functions since most of our daily activities depend on our hands. A person with tetraplegia also needs the upper limb (UL) to compensate for the paralysed legs in activities such as wheelchair mobility and transfers. Recovery of hand function has in several studies been depicted as the most important function to recover by people with tetraplegia.

Despite the importance of hand function, the evidence for treatment and optimization of the UL after tetraplegia is underdeveloped. Intervention studies targeting the tetraplegic UL typically focus on a single intervention and individuals in a chronic stable phase. The international clinical experience is that there are known factors that has the potential to impact the UL after c-SCI and studies investigating the incidence, timeframes and potential consequences for a functional UL in the acute and early phase are lacking.

Aim/Study rationale The overarching objective is to develop and assess a structured treatment plan for upper limb rehabilitation following cervical spinal cord injury (SCI). The ultimate aim is to empower patients with functional hands that enable daily living activities, emphasizing optimized grip strength and minimal range of motion limitations.

To achieve this, it is crucial to thoroughly investigate and describe the injury patterns, secondary complications, and biomechanical properties of the upper limb during the initial 6 months post-SCI.

METHODS Data collection and study sites This is a prospective cohort study. The study is designed to cover the absolute majority of people sustaining a cervical SCI during one year in Sweden and New Zealand. Inclusion criteria is to be found in the study design section. Following study sites will be included; in Sweden; Region Skåne, Region Västerbotten, Region Stockholm, Västra Götalandsregionen. In New Zealand, the two national spinal units in Auckland and Christchurch. Additional data will be collected from Lyndhurst Toronto rehab, KITE, Toronto, Canada. The Swiss Paraplegic Center, Notwill Switzerland will also be involved. Data will be collected at 1,2,3 and 6 months after sustained c-SCI. When available, additional assessment 1-2 weeks after injury will be included. The study aims to recruit patients for one year.

Sample size The sample size is not based on a formal sample size calculation, since this is an exploratory study investigating the injury pattern and its consequences related to the upper limb. With several assumptions due to lack of comprehensive data the estimated total number is 210 participants (420 arms) to be included and 140 participants (280 arms) to be included in the 6 months follow up. Approximately 420 arms is considered to be a feasible number to secure the primary research question; describe the injury pattern. For the secondary question, were the 6 months follow up is crucial a total of 280 arms is considered sufficient to adequately test for interactions.

Datacollection Assessment will be done by training Occupational therapists at 1 ,2, 3 (+-1w) months and 6 months (-1+2months). A detailed description of the data collection in an examiners manual will be available and all site investigators will receive specific training prior study start to secure required competence and measuring alinements between the sites. The Coordinating investigator (PI) will commence regular meetings (approximately once a month) and also chat function together with the site investigators to secure forums to clarify any questions during the data collecting phase. Outcome measures and timepoints for each measure in the study is to be found in the outcome measure section.

Site investigator verifies that each patient has consented to direct access to the original source data/hospital records, by the use of written patient information and signed Informed consent. Thereafter the site investigator will register coded original source data into the electronic system RedCap that is provided by the Coordinating investigator through Västra Götalands regionen, who is also responsible for the system. All information processed by the PI and steering group will be pseudonymized and identified with a study code. Subjects will not be identified by their names, but by an identification code in RedCap, and in other relevant study documents.

Data analysis Sahlgrenska University Hospital (SUH), as part of Region Västra Götaland, is the entity responsible for research and Personal Data Controller and will store the RedCap system (the eCRF-system used in the study). SUH will also be in charge of the statistical analyses and will be a Personal Data Processor in the study.

Demographics and timepoint specific characteristics will be summarized descriptively and changes over time will be described. Descriptive statistics that will be presented are number of observations and proportions for categorical variables, and number of observations, means, standard deviations, medians, quartiles and ranges for continuous variables.

Linear regression will be used to investigate whether the functional outcome (GRASSP, SCIM and TUAQ) is dependent on baseline characteristics. The Jonckheere-Terpstra test will be used to determine if there is a statistically significant trend between ordinal independent variables and a continuous or ordinal dependent variables. All tests will be two-tailed and conducted at 0.05 significance level.

The exact co-variates will be defined in the statistical analysis plan (SAP) and will include e.g.:

• Age
• Sex
• Severity of injury (ISNCSCI; C1,C2, C3, C4, C5, C6, C7) (AIS; A,B,C,D)
• Secondary complications (stiffness; prevalence of restrictions over 20 degrees,) (spasticity; prevalence of MAS 0, 1-2 and 3-4) (swelling; prevalence of increased circumference >1cm )
• Remaining motor function (total number of innervated muscles (1-5), total number of muscles >MRC 3, grouping data into flexors and extensors)
• Motor excitability (number of innervated muscles (=2), partly denervated muscles (=1) and denervated muscles (=0), grouping data into flexors and extensors.
• Biomechanical properties (resting position, hand opening/closing and pinch opening/closing, respectively; 0= no/minimal opening/closing, 1= some but not functional, 2= over half opening/closing, 3=3/4 open/closing, 4=normal) Multivariate analyze will be used to investigate explanatory factors to grip ability (GRASSP) and patients' satisfaction with grip ability (TUAQ) at 6 months. The aim is to assess differences between sub-groups with respect to the secondary endpoints using Cox regression models including Grip ability variable and subgroup factor and also the interaction between these variables as independent variables will be tested. The subgroups will be based on for example sex, age, injury severity, secondary complications, biomechanical properties and will be further pre-specified in the SAP.

Categorical endpoints will be analyzed using unadjusted and adjusted logistic regression models. Odds ratios (OR) and 95% CIs will be presented. Linear models will be used to analyze continuous variables, with logarithmic transformations applied if needed. Mean values and 95% CIs and OR will be presented.

Prior to start of data analysis, a separate statistical analysis plan (SAP) describing the intended statistical analyses and derivation of variables in greater detail, will be finalized.