Degenerative cervical myelopathy (DCM) is a disease characterized by gradual cervical spinal cord compression from degenerative changes of the spine (spondylosis), leading to neurologic dysfunction and disability. Although variable, clinical findings associated with DCM include gait problems, loss of fine motor dexterity, and eventual bladder and bowel control dysfunction. DCM is one of the most common causes of spinal cord dysfunction worldwide; however, there is a lack of epidemiological data on DCM currently available. The treatment for DCM is surgical, to halt disease progression and improve symptoms. Further, early surgical treatment (vs late) of DCM is associated with improved outcomes and decreased patient suffering. Despite this, diagnosis of DCM is often tricky, resulting in longer wait times diagnosis and treatment. There is a significant lack of literature defining the delay in diagnosis of DCM and the factors associated with a longer time from symptom onset to diagnosis.
The objective of this project is to describe the time course to diagnosis and treatment for patients with DCM. It will further evaluate the factors associated with a delayed diagnosis of DCM. Overall, the goal is to help guide clinicians with the early recognition of DCM, which could shorten the delay to surgery and improve outcomes.
The objectives of this project are as follows:
1. Describe the time from symptom onset of DCM to diagnosis and treatment by a specialist that manages DCM in the general population of the UK.
2. Assess factors associated with a delay in the diagnosis and treatment of DCM by a specialist.
Methods: This study is a retrospective cohort study centred on the UK population covered by the CPRD suffering from DCM. The CRPD will be used to analyze the time to diagnosis and/or surgical treatment for patients with DCM by a specialist. Patients will be included using procedure and diagnostic codes for DCM. The past medical codes for these patients, including previous diagnoses, testing, specialist referrals, and prescriptions, will be reviewed to identify the date of initial presentation to a primary care physician. Aggregate linked data sources (including HES Emergency and diagnostic imaging datasets) will be used to analyze variables that will help delineate the factors associated with delayed diagnosis in DCM. Survival analysis using Kaplan Meier plots will be used to describe the time from an initial visit to diagnosis or surgical treatment. Risk factors for a delay in DCM diagnosis by a specialist will be assessed using a Cox Proportional Hazards Model.
Conclusions: Our goal is to define the clinical window between initial presentation for patients with DCM to surgical intervention. Further, we seek to identify risk factors within this patient population that may be associated with a delay in diagnosis and treatment. Faster time from symptom onset to surgery is associated with improved outcomes. Given the significant individual and societal burden of spinal cord injuries (and DCM in particular), we hope to use this information to help guide clinical pathways and aid in the early diagnosis (and eventually treatment) of patients with DCM.
The outcomes to be measured will be the date of DCM diagnosis by a neurosurgeon, orthopedic surgeon, physical medicine & rehabilitation (PM&R) specialist, or neurologist. We will measure the time from initial presentation at a primary care clinic to the first date of diagnosis of DCM by an appropriate physician or a surgical intervention for DCM. The initial presentation at a primary care clinic is defined as the first time a patient presents for suspected myelopathy; this will be the first general practice visit for a gait-related problem, cervical spondylosis, cervical radiculopathy, carpal tunnel syndrome, peripheral nerve entrapment in the upper extremity, neuropathy or cervicalgia. Further, any patient with a cervical spine MRI in the 5 years prior to the diagnosis will have their t=0 at that time.
We will also assess the prevalence of faulty diagnoses, diagnostic imaging tests, medication prescriptions and physician referrals that are used to manage these patients in general practice prior to their referral to a specialist to describe the potential causes for a delay in diagnosis by a specialist.
The diagnosis of DCM is made overwhelmingly by specialists. In this study, the diagnosis of DCM will be given if:
1) A formal diagnosis of DCM is made by a specialist using codes identified in the appendix, which clearly identifies cervical myelopathy (i.e. “cervical myelopathy”, “cervical disc prolapse with myelopathy”.
OR
2) A surgical intervention code that is used for myelopathy surgery (i.e. “anterior cervical decompression and fusion”) is given WITH a diagnosis suspicious of myelopathy by a specialist (i.e. “cervical spinal stenosis”).
We acknowledge that the diagnosis of DCM by a specialist is not the gold-standard for diagnosis of the pathology. However, there is no current gold-standard for DCM diagnosis; diagnosis algorithms using administrative health data has been measured to have a sensitivity of 97% and a specificity of 60%, with a high positive predictive value (PPV) of 92%6. Given this high PPV, the probability of truly having DCM if a diagnosis of DCM if given to you by specialist is high. Within the context of our research, this implies with reasonable certainty that those who receive a diagnosis of DCM by an appropriate specialist truly suffer from DCM. Ultimately, the lack of a true gold-standard may affect the external validity of the study if many patients with DCM are missed, or if the somewhat lower specificity (60%) increases false positives.
Samy Suissa - Chief Investigator - Sir Mortimer B Davis Jewish General Hospital
Samy Suissa - Corresponding Applicant - Sir Mortimer B Davis Jewish General Hospital
Hui Yin - Collaborator - Sir Mortimer B Davis Jewish General Hospital
Lior Elkaim - Collaborator - McGill University
Oliver Lasry - Collaborator - McGill University
HES Accident and Emergency;HES Admitted Patient Care;HES Diagnostic Imaging Dataset;Patient Level Index of Multiple Deprivation Domains