Validating a Brief Screening Tool With Biomarkers for Prediction of Chronic Post Traumatic Neck Pain

Background Post-traumatic neck pain sustained after accidents are frequent (1) and represents a huge human and societal burden that can cause significant long-term pain and disability (2-4). It has been estimated that there are 6,000 new annual cases of traffic accidents resulting in post-traumatic neck pain in Denmark (5). However, recent international research indicates that the figure may be more than double (6-8). While up to half of those injured continue to experience symptoms a year after the accident (1-3), 15-25% experience significantly more severe symptoms and functional impairment (9,10) and are at risk of leaving the labour market. The accumulated cost of post traumatic neck pain in Denmark has not been evaluated in a scientific study but it is estimated the direct and indirect economic costs of post traumatic neck pain are approximately DKK 14 billion annually (11).

Post-traumatic neck pain after accident is often accompanied by a myriad of symptoms such as headaches, fatigue, concentration problems, and sensitivity to light and sound (4,12-16). A number of studies also point to frequent psychological symptoms after injuries such as symptoms of stress, depression, and anxiety (15-17), and early post-traumatic stress has been reported to predict poorer recovery over time (18,19).

Previous studies have identified the same risk factors after whiplash injuries (20-26). Risk factors are age, high pain intensity, hypersensitivity to pain, and post-traumatic arousal symptoms (i.e. increased stress) immediately after the accident. The results suggest that the lack of efficacy of existing treatments (27-30) :is caused by not specifically targeting risk patients and the underlying related mechanisms.

Ritchie and colleagues have developed and validated a clinical prognostic screening tool to identify patients at risk of developing post traumatic neck pain (31,32). The optimal use of the screening tool is immediately after accidents. Based on simple screening questions that can be answered in less than 5 minutes regarding pain and related functional limitations, post-traumatic stress disorder (hyperarousal) and age, the screening tool can with 90% accuracy be used to identify those that do not improve after the accident and with 80% accuracy those who recovers completely. However, there is a need to test the screening tool in a Danish context in the emergency department and investigate whether biomarkers for stress can improve the predictive value of the tool.

In Denmark, the standard emergency department treatment after non traumatic accident is a thorough physical evaluation coupled with good advice, which is appropriate for a significant proportion of patients. Unfortunately, it is difficult to distinguish a normal reaction from a treatment-requiring reaction, which in turn can lead to a worsening of traumatic neck pain. In this instance the prognostic tool (31,32) will be simple and quick to use. Collection of objective biomarkers for early stress response immediately after the accident will probably strengthen the value of the tool and provide more insight into causal mechanisms in the development of chronic post-traumatic neck pain. Moreover, biomarkers such as heart rate variability (HRV) can be collected immediately after an accident and thereby predict the risk of developing PTSD. Using questionnaires this information would take weeks to collect risking delayed identification of risk patients.

A significant and recurring finding is that stress response immediately after an accident is a significant factor in the development of chronic post-traumatic neck pain. Post-traumatic arousal is associated with increased sensitivity to pain, catastrophic thinking, and misinterpretation of common bodily symptoms (33,34). An important biomarker for autonomic dysregulation (stress response) and development of post-traumatic stress symptoms is low HRV accompanied by high resting heart rate (35-41). Conversely, a high HRV is associated with cardiovascular recovery and improved self-regulation during stressful situations (42). To the best of our knowledge, only one study examined HRV as a predictor of PTSD after severe traffic trauma and found that HRV measured two days after an accident correctly identified 72% of those who developed PTSD after the accident (43).

Questionnaires Participants will be asked to answer questionnaires regarding quality of life, sleep, comorbidities, PTSD symptoms and physical activity. All questionnaires are validated in Danish. Questionnaires will be pilot tested at both sites before given to participants. Time consumption is estimated to 30 minutes and will be completed using a tablet when attending the emergency department and online during follow-up at 6 and 12 months respectively.

Clinical assessment The study uses a combination of different measurements such as self-reported variables, biomarkers (HRV), pain sensitivity tests (QST) and register data. Data will be collected at different time points.

During the initial visit in the emergency department a clinical assessment of the participants active range of motion of the cervical spine and upper and lower limbs will be conducted. Any pain, restriction in range of motion or discomfort will be noted. A complete neurological assessment of the upper and lower extremities will be performed. Isometric muscle strength and tendon reflexes will also be assessed. The clinical assessment including measurements of range of motion will be repeated during the follow up visit 1 week after the accident.

HRV HRV, resting heart rate, respiration frequency and sleep patterns form the study's biomarker for stress and are measured with the Firstbeat Bodyguard 2 tracker (Firstbeat Technologies Ltd., Finland) which records data continuously over four days. Measurements of HRV will be carried out at baseline, 3,6,12 month. HRV measurements can be used as a measurement of stress and recovery (45).

Quantitative sensory testing

The investigators l perform pressure pain thresholds and conditioned pain modulation at baseline and 3 month. Conditioned pain modulation will consist of pressure pain threshold followed by cold pressor test and pressure pain threshold. Pressure pain threshold will be performed in the lower limb (upper anterolateral aspect of the tibia). Immediately after the pressure pain threshold test participants will be asked to submerge their non-dominant hand in 0-2 degrees cold water for a maximum of 2 minutes or until pain is intolerable. Immediately here after the pressure pain threshold will be performed once again. Pressure pain threshold will be measured using an algometer (Somedic, Sweden). QST will be carried out on the non-dominant side. The protocol for QST will conform to the German Network for Neuropathic Pain protocol (46).

Follow up at 3, 6, and 12 months The participants will be contacted by phone at 3, 6 and 12 months. At 3 months another appointment for onsite testing will be arranged. At 6 and 12 months the FirstBeat device will be sent to participants by mail and participants will be asked to repeat the previous questionnaires online.

Sample size calculation and statistical analysis plan Based on previous cohort studies, it is expected that at least 25% of the participants will develop moderate to severe chronic symptoms (20,21,23,24,26). A more conservative estimate of 15% is used as the starting point for power calculation (easyROC web-tool v1.3.1). At an expected AUC of 0.80 and an allocation ratio of 7, a sample of 104 patients will be sufficient to give a statistical strength of 90% and yield a significant result (alpha 0.01). Drop-out rates and missing data are expected, hence an expected drop-out rate at 12-month follow-up of up to 25% is estimated. Thus, a total of 155 patients will be included. Distribution of normality will be examined using histogram and qq-plots. Parametric and non-parametric analysis will be applied as applicable. Participants' characteristics will be described at all time points.

The prediction tool will be validated by analysing its ability to predict those who recover from the traffic accident (NDI < 10%) and those who develop moderate to severe symptoms (NDI > 32%) at 12 months follow-up. The precision of the tool will be examined with receiver operator characteristics (ROC) analysis, and sensitivity and specificity are calculated by various cut-off criteria. In addition, logistic regressions will be performed as applicable with and without biomarkers in order to establish the best prediction model. Health economic costs are compared between groups. Labour market attachment, consumption of medicines and health services will serve as a proxy of recovery after 12 months for the groups identified by the prediction tool. All register data will be collected after 12-month follow-up has ended for all participants.

A multivariable regression analysis will be used to investigate the association between PTSD and HRV and participant data adjusted for relevant cofounders. Relevant confounders will be identified using directed acyclic graph (DAG)(47) specifically the dagitty tool for R package (48). Missing data will be managed by applying multiple imputation strategies. All data will be analysed using Stata (V.17, StataCorp, Texas, USA) and R (version 4.3.1 (Beagle Scouts)). The result of the analysis will be reported in accordance with STROBE statement (49) for cohort studies.

Ethics and dissemination The study is approved by the regional Ethics Committee (S-20230037) and the University of Southern Denmark, the study will also be registered at ClinicalTrials.gov. The study will conform with GDPR and the Data Protection Act as well as the declaration of Helsinki (50,51). Access to register data will be applied for via Statistics Denmark. Adverse events (AE) will be reported according to the regional Ethics Committee. Furthermore, AE will be recorded at all follow-up visits by asking participants about potential AEs using open questions to make sure all AEs are recorded. An AE is defined as any unwanted experience during follow-up leading to interaction with the healthcare system (General practitioner or hospital) due to symptom worsening.

Apart from possible short term pain flare up following the QST procedure there are no known significant long-term side effects of the data collection process. The design of this study does not allow for the minimum requirement of 24 hours of reflection time for participants. This procedure has been used before in the emergency department, and the procedure is approved by the regional Ethics Committee.

The results of the project will be published in peer-reviewed scientific journals as well as non-scientific dissemination in the form of articles in patient and professional journals and through press releases and social media. Practical workshops will be held to train interested professionals in the screening method - should this prove useful and effective. The result of this project will be published regardless of the magnitude or direction of the result. Co-authorships will be based on the ICMJE recommendations (52).

Perspectives and implications The results of this study should be of interest to clinicians, researchers, legal and insurance companies and policymakers and will strengthen both national and international knowledge about early identification of patients in high risk of developing chronic neck pain and the role of biomarkers. To our knowledge, this project is the first of its kind in a European context. If the findings of this study are positive, the investigators expect it will allow us to target resources more efficient and possibly cut the human cost of post traumatic neck pain.

Funding The project has received financial support from the University of Southern Denmark, and the Næstved-Slagelse-Ringsted Hospital Research Fund. None of the funders of this project will have influence on whether and how results are published, just as funders will have no influence on the choice of research design, content and the preparation of this protocol.

"These materials have received financial support from The Danish Victims Fund The execution, content, and results of the materials are the sole responsibility of the authors. The analysis and viewpoints that have been made evident form the materials belong to the authors and do not necessarily reflect the views of The Council of The Danish Victims Fund".