Adjacent Segment Disease After Anterior Cervical Decompression Surgery

Overview

Radiculopathy and/or myelopathy due to cervical degenerative disc disease are increasingly common pathologies in our ageing population. Both can be treated non-surgically or surgically. The most commonly used neurosurgical treatment is anterior cervical discectomy with or without fusion. The goal is to achieve neural decompression of the operated segment in both procedures. However, due to this fusion and reduced mobility of the cervical spine at the level of the intervention, adjacent segment disease may occur. This can lead to new symptoms like radiculopathy and/or myelopathy at an adjacent level which requires reoperation in about 2/3 of patients. Reoperations are burdensome for patients and have a socio-economic impact due to the costs of hospital admissions, operations, and secondary costs such as work-absenteeism. The primary objective of this retrospective study is to determine the occurrence of adjacent segment disease after a single- or multi-level anterior cervical discectomy with fusion procedure for radiculopathy and/or myelopathy in the investigators' centre and to compare this to the incidence in literature. The investigators also look at the risk of adjacent segment disease after different anterior surgical techniques, such as anterior cervical discectomy, anterior cervical discectomy with fusion and plating, and corpectomy. As a secondary outcome they aim to determine risk factors predicting the occurrence of adjacent segment disease.

1. SUMMARY Radiculopathy and/or myelopathy due to cervical degenerative disc disease are increasingly common pathologies in our ageing population. Both can be treated non-surgically or surgically. The most commonly used neurosurgical treatment is anterior cervical discectomy with or without fusion. The goal is to achieve neural decompression of the operated segment in both procedures. However, due to this fusion and reduced mobility of the cervical spine at the level of the intervention, adjacent segment disease may occur. This can lead to new symptoms like radiculopathy and/or myelopathy at an adjacent level which requires reoperation in about 2/3 of patients. Reoperations are burdensome for patients and have a socio-economic impact due to the costs of hospital admissions, operations, and secondary costs such as work-absenteeism. The primary objective of this retrospective study is to determine the occurrence of adjacent segment disease after a single- or multi-level anterior cervical discectomy with fusion procedure for radiculopathy and/or myelopathy in the investigators' centre and to compare this to the incidence in literature. The investigators also look at the risk of adjacent segment disease after different anterior surgical techniques, such as anterior cervical discectomy, anterior cervical discectomy with fusion and plating, and corpectomy. As a secondary outcome they aim to determine risk factors predicting the occurrence of adjacent segment disease. 2. INTRODUCTION AND RATIONALE Radiculopathy and myelopathy are pathologies which, amongst others, occur due to cervical disc herniation or spondylosis, also known as cervical degenerative disc disease (CDDD). Radiculopathy has been estimated to affect around 100 per 100.000 males and 60 per 100.000 females in the general population. The incidence of myelopathy is estimated to be 4 per 100.000 in the general population. The consequences of CDDD can be severe and can disable patients to an extent that it impairs working ability. Therefore, it leads to a significant socio-economic burden of disease. Cervical radiculopathy is caused by compression of the cervical nerve roots as they exit the vertebral column through the neuroforamina. Patients present with radiating pain in the arm in the corresponding dermatome, weakness and/or sensory loss, with or without associated neck pain. Cervical myelopathy is caused by a central canal stenosis, which causes compression of the spinal cord itself. Damage to the central nervous system at this level can cause loss of coordination, motor, and sensory function of both arms and legs. Treatment for radiculopathy and/or myelopathy due to CDDD can be non-surgical or surgical. The majority of patients with radiculopathy respond well to non-surgical treatment options, such as physical therapy, a neck collar, analgesic or anti-inflammatory medication. Surgery can be considered in cases of insufficient relief with these non-surgical treatment options. The reported surgery rates range from 8% to 35%. When patients have mild myelopathy, which does not progress clinically, non-surgical treatment with frequent follow-up is an option, however, controversy exists. Others suggest surgical treatment with mild myelopathy, to prevent deterioration of symptoms. Surgery is recommended in patients with moderate and severe degenerative cervical myelopathy (DCM) or with rapid clinical deterioration. Both anterior and posterior surgical techniques are used to treat radiculopathy and/or myelopathy. No clear consensus exists on which anterior technique is superior. Depending on country, centre and surgeon, different approaches are used, mostly based on experience. Anterior cervical discectomy with (ACDF) or without fusion (ACD) are the most commonly used interventions for CDDD since the compression is often located on the anterior side. Alternatives are corpectomy or ACDF with plating. In the investigators' centre ACDF is the procedure of choice in most patients with single- or multilevel CDDD and radiculopathy and/or myelopathy. ACD and ACDF both have good short-term clinical results. However, in the long-term patient satisfaction drops, which is probably for a large part due to adjacent segment disease (ASD). This occurs in approximately 25% of patients during 10 years follow-up and over 2/3 of these patients need additional surgery. ASD is thought to be a consequence of fusion of two or more vertebra which eliminates mobility at the level of the intervention. Adjacent levels compensate for this reduced mobility by increasing their mobility and thus increasing their risk at disc degeneration. This risk is not only determined by the biomechanical stress on the adjacent level caused by fusion. Changes in the anatomy at the adjacent level with the initial surgery and the natural course of the adjacent disc also plays a role. Anterior cervical discectomy with arthroplasty (ACDA) was developed in an effort to reduce the incidence of ASD by preserving physiological motion in the operated segment. ACDA has been confirmed to ACDF for treatment of two-level CDDD, a significantly lower re-operation rate in the ACDA group compared to the ACDF group was found after two years (3.1% versus 11.4%, respectively). Re-operation rates for ACDA and ACDF after 7 years were compared. They reported reoperation rates of 3.7% for ACDA versus 13.6% for ACDF in single-level CDDD patients. The rate of reoperation was also significantly lower in the two-level ACDA group compared to ACDF, with rates of subsequent surgery at the index level of 4.4% versus 16.2% and rates of adjacent level surgery of 4.4% versus 11.3%, respectively. In a systematic review from 2017 multilevel ACDA was proven at least as safe and effective as ACDF, with preservation of cervical motion and potentially with fewer reoperations expected. The goal of this study is to determine the risk of ASD after ACD(F) for radiculopathy and/or myelopathy in the study population, and to compare this to the rates reported in literature. The expectation is that the occurrence in the study population is similar to the occurrence in the general population. The investigators will also look at the risk of ASD after different anterior surgical techniques, e.g. ACDF with plating, ACD, corpectomy. These are less commonly used procedures in the study centre. Due to the biomechanical stress induced by fusion, higher rates of ASD are expected after ACDF with plating and corpectomy than after ACDF or ACD. As a secondary outcome the investigators aim to determine risk factors predicting the occurrence of adjacent segment disease. Age and degeneration of the cervical spine are expected to play a role. Therefore, baseline characteristics will be assessed and the Kellgren's score on pre-operative X-rays will be determined. 3. METHODS 3.1 Study design A retrospective cohort study will be conducted in the Zuyderland Medical Center, Heerlen and Maastricht UMC+, Maastricht. Patients undergoing anterior cervical decompression surgery in the past 10 years will be analysed. The occurrence of ASD will be the primary outcome. Baseline characteristics will be assessed to determine predictive factors for ASD. Data will be assessed in the electronic patient files and a database will be constructed. This study is expected to take about a year from the METC approval to submitting the article. 3.2 Study selection 3.2.1 Study population and selection The study population consists of adult patients who underwent anterior cervical decompression for radiculopathy and/or myelopathy due to CDDD in the Zuyderland MC Heerlen or MUMC+ Maastricht in the past 10 years. The following anterior surgical decompression techniques will be included; ACD, ACDF, ACDF with plating, and corpectomy. The data manager for Zuyderland Medical Center and Maastricht UMC+ will conduct a search of the electronic patients records in this period using DBC/DOT codes for above mentioned neurosurgical interventions. 3.2.2 Sample size There will be around 100 patients per year that can be included in the study which gives an estimated total sample size of approximately 1000 patients. As this is a retrospective cohort study, the number of patients that can be included is set. With the given sample size and power, the smallest possible difference this study can show could be calculated. However, in this study this is not relevant. Based on previous studies, as shown in the hypothesis, an incidence between 10% and 20% is expected. With a sample size of 1000 and an incidence of 10%, the 95% confidence interval (including continuity) is 8,25%-12,07% (width 3,82%). With a sample size of 1000 and an incidence of 20%, the 95% confidence interval (including continuity) is 17,59-22,64% (width 5,05%). Concluding: with an incidence of 10-20% and a sample size of 1000, the width of the confidence intervals will be 3,82-5,05%. This can be considered small, which indicates a sufficient sample size. To examine the effect of parameters that may have an influence on the occurrence of ASD, 10 events and 10 non-events are needed per parameter. If the incidence were to be 10% with a sample size of 1000, there are 100 events. This means 10 parameters can be included in the analysis. 3.3 Data selection Baseline data will be collected from the electronic patient file (EPD). Treatment data such as the indication and level of surgery will be documented, as well as complications related and not-related to the surgery. When interpreting the clinical outcome after the surgery, the following has to be taken into account. For radiculopathy, a good clinical outcome means improvement. A poor outcome is either stabilisation or deterioration. For myelopathy a good outcome is stabilisation (or in some cases improvement) and only deterioration is considered as a poor outcome. Baseline and treatment data will be collected to evaluate whether they attribute to the risk on developing ASD. Baseline data * Age (years) * Gender (male/female) * Body Mass Index (BMI) (kg/m2) * Smoking (yes/no) * Pre-operative duration of symptoms (weeks) * Indication (myelopathy/radiculopathy/both) * Level of CDDD (C3-C4, C4-C5, C5-C6, C6-C7, C7-Th1) * Post-operative use of pain medication (yes/no) * Kellgren's score (0-4) Treatment data * Intervention (ACDF/ACDF with plating/ACD/corpectomy) * Level of surgery (C3-C4, C4-C5, C5-C6, C6-C7, C7-Th1) * Single or multilevel (1-level/2-level/more than 2 levels) * Follow-up duration (months) Outcome parameters * Radiculopathy (good/poor) * Myelopathy (good/poor) * Re-operation for ASD (yes/no) * New symptoms (yes/no, radiculopathy/myelopathy/both, level) Complications * Related to surgery (none/failure of treatment/death) * Other (death) 3.4 Data analysis Statistical analyses will be carried out using IBM SPSS statistics 25. Summary statistics will be calculated for demographic, treatment and outcome variables. T-tests will be used for continuous variables. X2 tests will be used for categorical variables. ANOVA-analyses will be used is case of more than two groups. Analyses will be two-sided. P values of \< 0.05 will be considered significant differences. Descriptive statistics will be used to assess the incidence of ASD in the study population. A logistic regression analysis will be performed with regard to the outcome parameters. A survival analysis will be performed to evaluate the risk factors for ASD over time. 3.5 Data management A coded excel file will be created with obtained patient data. The key to the code will be stored separately and safely in the MUMC+. Information and results will be stored in SPSS where only patient numbers will be mentioned. Names and other personal information, such as date of birth will be left out of this file. Data and results will be stored on password protected computers and a hard drive secured by the hospital. The files are only available for the investigators involved in the study (Dr. A. Smeets, Drs. V. Schuermans and Drs. N. Wijsen) and will not leave the hospital. Handling of personal data will comply with the Dutch Personal Data Protection Act (Wbp). Data will be stored for 15 years in the concerning centre. 4. ETHICAL CONSIDERATIONS The study protocol will be submitted to the Medisch Ethische Toetsings Commissie van Zuyderland en Zuyd Hogeschool Zuderland The Medical Research Involving Human Subject Act does not apply to the above-mentioned study (niet WMO-plichtig onderzoek). The 'geen bezwaarregeling"; is applicable seen the impossible workload of contacting an estimated 800 patients to ask for informed consent. Moreover, a large amount of these patients might have changed contact information or may have deceased. The patients and their lives will not be negatively influenced seen all data is anonymized. When patients are contacted, they are asked whether they are willing to participate in to the study and to answer one question about their previous neurosurgical treatment. This question is not considered to be burdensome for the patients. Moreover, this study cannot be conducted without the data of these patients and the results are of public importance. Therefore, it is reasonable to proceed with the study without obtaining written informed consent. The results obtained in this study will give important insights in the risk of ASD after different surgical techniques. This might lead to other primary treatment techniques, which might prevent re-operations and hospital admissions for future patients. Before using patients' data, records will be checked for objection to participation in research. All data will be anonymized. A neurosurgeon or resident neurosurgery with a treatment relationship to the patient will contact them by phone to ask for re-operation in a different centre. Patients will not be burdened by this study, as they will only have to reply to one phone call. Patients will not receive a reimbursement for participation. 5. VALORISATION AND PUBLICATION The results of this study will lead to more insights in the reoperation rates for adjacent segment disease in the study population after anterior cervical decompression surgery CDDD. If a relation can be seen between age and Kellgren score and the risk for ASD, a different counselling or treatment choice can be made for this population. If a specific age group seems to have a lower risk on ASD, they might not experience benefit from motion-sparing surgery. Moreover, the investigators might be able to determine a difference in re-operation rates for ASD in different anterior surgical techniques, which can also influence treatment decisions. Results of this study will be submitted to peer-reviewed open access journals. In this publication, data will be anonymized and will also be submitted when they turn out to be negative. None of the involved parties have veto about whether or not data is published.