SIZE study: study protocol of a multicentre, randomised controlled trial to compare the effectiveness of an interarcuair decompression versus extended decompression in patients with intermittent neurogenic claudication caused by lumbar spinal stenosis

Introduction Intermittent neurogenic claudication (INC) is often caused by lumbar spinal stenosis (LSS). Laminectomy is considered a frequently used surgical technique for LSS. Previous studies have shown that laminectomy can potentially cause lumbar instability. Less invasive techniques, preserving midline structures including the bilateral small size interarcuair decompression, are currently applied. Due to lack of evidence and consensus, surgeons have to rely on their training and own experiences to choose the best surgical techniques for their patients. Hence, an observer and patient blinded multicentre, randomised controlled trial was designed to determine the effectiveness and cost-effectiveness of bilateral interarcuair decompression versus laminectomy for LSS. Methods and analysis 174 patients above 40 years with at least 12 weeks of INC will be recruited. Patients are eligible for inclusion if they have a clinical indication for surgery for INC with an MRI showing signs of LSS. Patients will be randomised to laminectomy or bilateral interarcuair decompression. The primary outcome is functional status measured with the Roland-Morris Disability Questionnaire at 12 months. Secondary outcomes consist of pain intensity, self-perceived recovery, functional status measured with the Oswestry Disability Index and a physical examination. Outcome measurement moments will be scheduled at 3 and 6 weeks, and at 3, 6, 12, 18, 24, 36 and 48 months after surgery. Physical examination will be performed at 6 weeks, and 12, 24 and 48 months. An economic evaluation will be performed and questionnaires will be used to collect cost data. Ethics and dissemination The Medical Ethical Committee of the Erasmus Medical Centre Rotterdam approved this study (NL.65826.078.18). The results will be published in an international peer-reviewed journal. Trial registration number ClinicalTrials.gov (NCT03480893). IRB approval status MEC-2018-093.

risk of an insufficient decompression may be higher, possibly resulting in reoperations due to residual 2 stenosis [15][16][17] . 3 The presumption is often made that, after a wide decompression, recurrence of the complaints are 4 scarce. The opinions on this matter are diverse and a direct comparison has not yet been performed.
5 Therefore, a randomised controlled trial is needed to clarify whether bilateral interarcuair 6 decompression is effective and cost-effective compared to a classical laminectomy.

Study objectives
9 The primary objective of this study is to determine whether bilateral interarcuair decompression is more 10 effective than laminectomy in patients with INC caused by LSS. The secondary objective is to evaluate if 11 bilateral interarcuair decompression is more cost-effective than laminectomy.

Study design
14 The study is designed as a multi-centre, patient and observer blinded randomised controlled trial with 15 an economic evaluation alongside. Participants will be allocated to one of the two groups: laminectomy Patients who meet any of the following criteria will be excluded from participating in this study. 11 Eligible patients will be randomised in a 1:1 ratio to the two groups.

13
Intervention: bilateral small size interarcuair decompression 14 The intervention consists of the interarcuair decompression, otherwise known as a bilateral 15 laminotomy, limited bony decompression or a interlaminar decompression. General anaesthesia is to be 16 administered. A midline skin incision is made, after which the paravertebral muscles are dissected 17 subperiosteally and retracted bilaterally. Decompression will be applied through decompression of the 18 ligamentum flavum and partial laminotomy if necessary. The lateral recess will be opened bilaterally and 19 a partial medial facetectomy will be performed in order to maintain stability of the segments. The  Usual care consists of conventional laminectomy, otherwise known as wide bony decompression.

5
General anesthesia is to be administered. A midline skin incision is made over the spinous processes.
6 The laminae of the affected level(s) are exposed subperiosteally, and the supraspinous ligament is 7 incised. The spinous process is removed. The supra and interspinous ligament of the affected level is 8 removed by drill or Kerrison punched. Both laminae are removed at the affected level, leaving the facet 9 joint intact. The lateral recess is opened bilaterally and a medial facetectomy will be performed in order 10 to maintain stability of the segments 15 . To clarify, when a single level stenosis is present (e.g. L4-L5), 11 both laminae L4 and L5 will be removed. When a double level stenosis (e.g. L3-L4 and L4-L5) is present, 12 three laminae (L3, L4 and L5) will be removed. The wound is closed in layers with our without a suction 13 drain. Patients will be operated with loupe magnification or microscope depending on the surgeon's 14 preference. 15

Use of co-intervention
16 Pain medication will be provided to patients after surgery, should this be necessary. Further, the use of 17 co-interventions will be tracked by cost questionnaires, in which medication usage and any health care 18 utilization is monitored throughout the follow-up period. The primary outcome is a functional status measured using the Roland-Morris Disability Questionnaire 2 (RMDQ). This is a 24-point questionnaire that is designed to assess the functional status in patients with 3 low back pain. This outcome measure has been identified as one of the most commonly used outcomes 4 in a population with chronic low back pain 21-25 .
5 Secondary outcomes 6 Oswestry Disability Index (ODI) 7 The ODI is one of the principal condition-specific outcome measures used in de management of spinal 8 disorders 26 . The current version 2.1a will be used 27 . It consists of 10 questions, each with six possible 9 answers and each answer option receives a score of 0 to 5 points, yielding a score range between 0 and 10 50, which is scaled to a 100% range. The questions focus on a range of daily physical functions and how 11 the back or leg pain is affecting the patient's ability to manage in everyday life.  Patients' perceived recovery 5 Recovery will be measured using a seven-point Likert scale. The score on this scale varies from 6 'complete recovery' to 'worse than ever'. The outcomes 'complete recovery', 'almost complete 7 recovery' and 'moderate recovery' will be considered as recovered 30 .
Patients' satisfaction of change and satisfaction of treatment will be assessed using a seven-point Likert 10 scale, ranging from 'completely satisfied' to 'completely dissatisfied'. 'Completely satisfied', 'almost 11 completely satisfied' and 'moderately satisfied' will be defined as good outcome.

12
Physical examination 13 A physical examination will be performed before surgery, at 6 weeks, 12, 24 and 48 months after 14 surgery. This examination will include a neurologic examination, the Timed Up and Go Test (TUG), the 15 Timed Chair-Stand Test and patient's weight. We will assess if there is a difference in muscle strength 16 between the affected leg and the other leg. If both legs are affected, we will note the muscle strength 17 and monitor the progress. The neurologic examination will include the patellar and the Achilles tendon reflexes; the muscle 20 strength of the quadriceps, the iliopsoas, the hamstrings, the gastrocnemicus and the tibialis 21 anterior muscle, which will be noted as a difference in strength between both legs. The tendon 22 reflexes have to be measured in a sitting upright position with both feet hanging above the floor. We will evaluate if there is a difference in muscle strength between the affected leg and 2 the other leg.

b. Timed Up and Go Test (TUG)
4 Patients are asked to sit in a chair with their arms resting on the armrests. On direction of the 5 research nurse, patients have to walk as fast as possible (without running) to a wall at 3-m 6 distance of the chair. When they have reached that wall, a turn of 180 degrees is made and they 7 walk directly back to the chair and sit down. The research nurse will measure the time 8 (in seconds) it takes for a patient, to perform this action. The time is stopped when the patients 9 are seated again 31 . Patients need to perform this test, three times. Patients are allowed to wear 10 their regular shoes and use a walking aid, if necessary. The TUG test has been described as a 11 valid, reliable tool in the assessment of objective functional impairment in patients with lumbar 12 degenerative disc disease 32 33 . However, this has not been done for LSS. The validation for LSS 13 will be conducted in this study.
14 c. Timed Chair-Stand Test

15
Patients are asked to sit down on an armless chair, firmly placed against a wall, with their arms 16 folded across their chest. They are instructed to place their feet flat on the ground. To become 17 familiarized with the exercise, the patients are asked to stand up and sit down again without 18 using their arms. If assistance is required during this action, or this movement cannot be 19 completed by the patient, the test is abandoned. If the manoeuvre is possible, the patient will 20 be asked to, on the word "go", stand up entirely and sit down again, 5 times as quick as possible.

21
The research nurse will time the 5 movements in seconds from the command "go" to the 22 completed fifth stand. The measured time will be noted as the score. If the patient is unable to 23 perform the test 5 times, a maximum score will be noted of 30 seconds 34 35 .

24
Functional lumbar X-rays Functional X-rays will be obtained from all patients. The X-ray will be made from AP and lateral position 2 to assess spondylolisthesis. Degenerative spondylolisthesis is defined as a vertebral slip of at least 3 3 mm 36 . After the measurement of the vertebral slip, this will be classified according to the Meyerding 4 classification. This will be obtained at 6 weeks postoperatively, for level verification and for verification 5 of correct procedure and after two years postoperatively to look for possible instability. 6 EuroQol-5-dimension-5-levels 7 The EuroQol (EQ-5D-5L) is a validated instrument to assess health-related quality of life (QOL problems, moderate problems, major problems, severe problems, unable to). The Dutch tariff of the EQ-16 5D-5L will be used to calculate the quality-adjusted life years (QALYs) 37 38 .

6
Hence the sample size of this study is based on a superiority design, using a delta of 3, an assumed true 7 mean difference of 0 and a pooled standard deviation of 6. Using an alpha of 0.05, and a power of 90%, 8 we calculated a sample size of 69 patients per group. Accounting for a 20% loss in follow-up, we will 9 therefore enrol 174 patients (87 patients per group).

11
Computerized block randomisation will take place after the patient has been anaesthetised. Allocation 12 concealment will be assured by using ALEA, a web-based data system supervised by the Clinical Trial  1  All patient reported outcomes will be collected preoperatively and after 3 weeks, 6 weeks, 3 months, 6   2  months, 12 months, 18 months, 24 months, 36 months and 48 months (see Table 1). Patients will 3 receive an email with a link, reminding them of the upcoming data collection and requesting them to fill 4 in the web-based questionnaires. If the questionnaires are not completed, the patient will receive a 5 reminder after one and two weeks. After two weeks, non-responders will be contacted by phone.
6 Data management 7 Data from initial visits to the hospital, follow-up visits and questionnaires will be entered into a database 8 via an electronic data system (Gemstracker). This data will be noted and analysed by using coded 9 information (study numbers) without any personal identifiers. Data will be stored via the secure data 10 management system of the trial coordination centre of the Erasmus MC.

Statistical analysis
13 Patient characteristics will be summarized per treatment group to determine whether the 14 randomisation was balanced. Continuous variables will be presented using means and standard 15 deviations, or in the case of non-normally distributed variables using medians and interquartile ranges.

16
Normality will be graphically assessed. Categorical variables will be presented using counts and 17 percentages. The primary analysis will study differences in the primary outcome measure (RMDQ). The 18 primary endpoint of interest is 12 months after randomisation. The results of the RMDQ will be 19 assessed using generalized linear mixed models (GLMM), to account for the repeated measurements 20 within patients and the multi-centre stratification.
3 Consent or assent 4 Surgeons will introduce the trial to patients at the outpatient clinic and will hand them a patient 5 information brochure regarding the main aspects of the trial. Patients will be given five working days to 6 decide if they want to participate in the trial. Trained research nurses will call patients after one week to 7 ask if they wish to partake in the trial. After verbal consent, an appointment will be made with the 8 research nurse, where informed consent forms will be obtained.

Instructions to authors
Complete this checklist by entering the page numbers from your manuscript where readers will find each of the items listed below.
Your article may not currently address all the items on the checklist. Please modify your text to include the missing information. If you are certain that an item does not apply, please write "n/a" and provide a short explanation.
Upload your completed checklist as an extra file when you submit to a journal.
In your methods section, say that you used the SPIRITreporting guidelines, and cite them as: 195 The ODI is one of the principal condition-specific outcome measures used in de management of spinal 196 disorders 28 . The current version 2.1a will be used 29   be asked to, on the word "go", stand up entirely and sit down again, 5 times as quick as possible.

250
The research nurse will time the 5 movements in seconds from the command "go" to the 251 completed fifth stand. The measured time will be noted as the score. If the patient is unable to 252 perform the test 5 times, a maximum score will be noted of 30 seconds 36 37 .

254
This test is performed to assess the gait pattern, walking speed and physical endurance of 255 patients. The patient will be requested to walk at such a speed that at the end of 6 minutes, the 256 patient will have the feeling of maximum output 38 . Before taking the test, the patient will be 257 given the instructions to install the 6MWT application on their cell phone 39 40 . During these 6 258 minutes this application will measure the walking distance. During the walking test, the patient 259 is permitted to use a walking aid and/or orthosis which will be noted.

261
Functional lumbar X-rays Functional X-rays will be obtained from all patients. The X-ray will be made standing from AP and lateral 263 position to assess spondylolisthesis. X-rays will also be made in maximum flexion and extension position.

264
Degenerative spondylolisthesis is defined as a vertebral slip of at least 3 mm 41 . After the measurement 265 of the vertebral slip, this will be classified according to the Meyerding classification. This will be obtained 266 at 6 weeks postoperatively, for level verification and for verification of correct procedure and after two 267 years postoperatively to look for possible instability. 268 EuroQol-5-dimension-5-levels

269
The EuroQol (EQ-5D-5L) is a validated instrument to assess health-related quality of life (QOL Based on a diary with made costs, which will be completed by patients, insight will be obtained in the 281 need for care, working participation and direct-and indirect medical cost. Direct medical cost will be 282 estimated on basis of the cost centre method. In addition it is expected that the study population will 283 consist of elderly people who may be retired. For the cost calculation additional costs, such as help in 284 house, transport, help from family etc. will be taken in account. progressive neurological deficit and secondary bleeding) will be carried out by the surgeon and the 307 research nurse shortly after discharge.

308
Reoperation is considered a bad outcome. The incidence of reoperation as well as perioperative 309 morbidity will be assessed in both groups, using surgical parameters such as blood loss, time of surgery 310 and length of hospital stay.

311
We used the SPIRIT checklist when writing our report 44 .   All patient reported outcomes will be collected preoperatively and after 3 weeks, 6 weeks, 3 months, 6 340 months, 12 months, 18 months, 24 months, 36 months and 48 months (see Table 1). Patients will 341 receive an email with a link, reminding them of the upcoming data collection and requesting them to fill 342 in the web-based questionnaires. If the questionnaires are not completed, the patient will receive a 343 reminder after one and two weeks. After two weeks, non-responders will be contacted by phone.

344
Data management

345
Data from initial visits to the hospital, follow-up visits and questionnaires will be entered into a database 346 via an electronic data system (Gemstracker). This data will be noted and analysed by using coded 347 information (study numbers) without any personal identifiers. Data will be stored via the secure data 348 management system of the trial coordination centre of the Erasmus MC.

Statistical analysis
351 Patient characteristics will be summarized per treatment group to determine whether the 352 randomisation was balanced. Continuous variables will be presented using means and standard 353 deviations, or in the case of non-normally distributed variables using medians and interquartile ranges.

354
Normality will be graphically assessed. Categorical variables will be presented using counts and 355 percentages. The primary analysis will study differences in the primary outcome measure (RMDQ). The 356 primary endpoint of interest is 12 months after randomisation. The results of the RMDQ will be 357 assessed using generalized linear mixed models (GLMM), to account for the repeated measurements 358 within patients and the multi-centre stratification.

359
The secondary outcomes (ODI, NRS leg and back pain score, TUG test, TCST and the SF-36) will be 360 similarly assessed using GLMMs. In all analyses the primary assessment of treatment effect will be the 361 estimate of the main effect within the appropriate model at 12 months, adjusted for the stratification Surgeons will introduce the trial to patients at the outpatient clinic and will hand them a patient 411 information brochure regarding the main aspects of the trial. Patients will be given five working days to 412 decide if they want to participate in the trial. Trained research nurses will call patients after one week to 413 ask if they wish to partake in the trial. After verbal consent, an appointment will be made with the 414 research nurse, where informed consent forms will be obtained (see appendix 1).

416
All study-related information will be stored securely at the study site. All participant information will be 417 stored in locked file cabinets with limited access. All reports, data collection (e.g. CRFS) and 418 administrative forms will be entered into an online data system. All these documents will be identified 419 by a coded study number only to maintain the patient confidentiality. These data will be stored for at 420 least 15 years.
Ik had genoeg tijd om te beslissen of ik meedoe.
-Ik geef toestemming voor het opvragen van informatie bij de specialist die mij behandelt over de wervelkanaalstenose.

Instructions to authors
Complete this checklist by entering the page numbers from your manuscript where readers will find each of the items listed below.
Your article may not currently address all the items on the checklist. Please modify your text to include the missing information. If you are certain that an item does not apply, please write "n/a" and provide a short explanation.
Upload your completed checklist as an extra file when you submit to a journal.

111
The presumption is often made that, after a wide decompression, recurrence of the complaints are 112 scarce. The opinions on this matter are diverse and subject to debate. 19 . Therefore, a randomised 113 controlled trial is needed to clarify whether bilateral interarcuair decompression is effective and cost-114 effective compared to a classical laminectomy.

116
Study objectives

117
The primary objective of this study is to determine whether bilateral interarcuair decompression is more 118 effective than laminectomy in patients with INC caused by LSS. The secondary objective is to evaluate if 119 bilateral interarcuair decompression is more cost-effective than laminectomy.

179
To clarify, when a single level stenosis is present (e.g. L4-L5), both laminae L4 and L5 will be removed.

180
When a double level stenosis (e.g. L3-L4 and L4-L5) is present, three laminae (L3, L4 and L5) will be 181 removed. The wound is closed in layers with or without a suction drain. Patients will be operated with 182 loupe magnification or microscope depending on the surgeon's preference. 183

Use of co-intervention
184 Pain medication will be provided to patients after surgery, should this be necessary. Further, the use of 185 co-interventions will be tracked by cost questionnaires, in which medication usage and any health care 186 utilization is monitored throughout the follow-up period.  196 The ODI is one of the principal condition-specific outcome measures used in de management of spinal 197 disorders 28 . The current version 2.1a will be used 29   be asked to, on the word "go", stand up entirely and sit down again, 5 times as quick as possible.

251
The research nurse will time the 5 movements in seconds from the command "go" to the 252 completed fifth stand. The measured time will be noted as the score. If the patient is unable to 253 perform the test 5 times, a maximum score will be noted of 30 seconds 36 37 .

255
This test is performed to assess the gait pattern, walking speed and physical endurance of 256 patients. The patient will be requested to walk at such a speed that at the end of 6 minutes, the 257 patient will have the feeling of maximum output 38 . Before taking the test, the patient will be 258 given the instructions to install the 6MWT application on their cell phone 39 40 . During these 6 259 minutes this application will measure the walking distance. During the walking test, the patient 260 is permitted to use a walking aid and/or orthosis which will be noted. Functional lumbar X-rays 264 Functional X-rays will be obtained from all patients. The X-ray will be made standing from AP and lateral 265 position to assess spondylolisthesis. X-rays will also be made in maximum flexion and extension position.

266
Degenerative spondylolisthesis is defined as a vertebral slip of at least 3 mm 41 . After the measurement 267 of the vertebral slip, this will be classified according to the Meyerding classification. This will be obtained 268 at 6 weeks postoperatively, for level verification and for verification of correct procedure and after two 269 years postoperatively to look for possible instability. 270 EuroQol-5-dimension-5-levels

271
The EuroQol (EQ-5D-5L) is a validated instrument to assess health-related quality of life (QOL Based on a diary with made costs, which will be completed by patients, insight will be obtained in the 283 need for care, working participation and direct-and indirect medical cost. Direct medical cost will be 284 estimated on basis of the cost centre method. In addition it is expected that the study population will 285 consist of elderly people who may be retired. For the cost calculation additional costs, such as help in 286 house, transport, help from family etc. will be taken in account. progressive neurological deficit and secondary bleeding) will be carried out by the surgeon and the 309 research nurse shortly after discharge.

310
Reoperation is considered a bad outcome. The incidence of reoperation as well as perioperative 311 morbidity will be assessed in both groups, using surgical parameters such as blood loss, time of surgery 312 and length of hospital stay.

313
We used the SPIRIT checklist when writing our report 44 .

342
All patient reported outcomes will be collected preoperatively and after 3 weeks, 6 weeks, 3 months, 6 343 months, 12 months, 18 months, 24 months, 36 months and 48 months (see Table 1). Patients will 344 receive an email with a link, reminding them of the upcoming data collection and requesting them to fill 345 in the web-based questionnaires. If the questionnaires are not completed, the patient will receive a 346 reminder after one and two weeks. After two weeks, non-responders will be contacted by phone.

347
Data management

348
Data from initial visits to the hospital, follow-up visits and questionnaires will be entered into a database 349 via an electronic data system (Gemstracker). This data will be noted and analysed by using coded 350 information (study numbers) without any personal identifiers. Data will be stored via the secure data 351 management system of the trial coordination centre of the Erasmus MC.

Statistical analysis
354 Patient characteristics will be summarized per treatment group to determine whether the 355 randomisation was balanced. Continuous variables will be presented using means and standard 356 deviations, or in the case of non-normally distributed variables using medians and interquartile ranges.

357
Normality will be graphically assessed. Categorical variables will be presented using counts and 358 percentages. The primary analysis will study differences in the primary outcome measure (RMDQ). The 359 primary endpoint of interest is 12 months after randomisation. The results of the RMDQ will be 360 assessed using generalized linear mixed models (GLMM), to account for the repeated measurements 361 within patients and the multi-centre stratification.
Ik had genoeg tijd om te beslissen of ik meedoe.
-Ik geef toestemming voor het opvragen van informatie bij de specialist die mij behandelt over de wervelkanaalstenose.

Instructions to authors
Complete this checklist by entering the page numbers from your manuscript where readers will find each of the items listed below.
Your article may not currently address all the items on the checklist. Please modify your text to include the missing information. If you are certain that an item does not apply, please write "n/a" and provide a short explanation.
Upload your completed checklist as an extra file when you submit to a journal.