Preoperative predictors and surgical outcome in lumbar spinal stenosis : A prospective 10-year follow-up study

(1)

IINA TUOMAINEN

Preoperative predictors and surgical outcome in lumbar spinal stenosis

Dissertations in Health Sciences

PUBLICATIONS OF

THE UNIVERSITY OF EASTERN FINLAND

(2)

(3)

Preoperative predictors and surgical outcome in lumbar spinal stenosis

A prospective 10-year follow-up study

(4)

(5)

Iina Tuomainen

Preoperative predictors and surgical outcome in lumbar spinal stenosis

A prospective 10-year follow-up study

To be presented by permission of the Faculty of Health Sciences, University of Eastern Finland for public examination in MD100 Auditorium,

Kuopio on September 24th, 2021, at 12 o’clock noon

Publications of the University of Eastern Finland Dissertations in Health Sciences

No 637

Department of Physical and Rehabilitation Medicine University of Eastern Finland, Kuopio

2021

(6)

Series Editors

Professor Tomi Laitinen, M.D., Ph.D.

Institute of Clinical Medicine, Clinical Physiology and Nuclear Medicine Faculty of Health Sciences

Professor Tarja Kvist, Ph.D.

Department of Nursing Science Faculty of Health Sciences

Professor Ville Leinonen, M.D., Ph.D.

Institute of Clinical Medicine, Neurosurgery Faculty of Health Sciences

Professor Tarja Malm, Ph.D.

A.I. Virtanen Institute for Molecular Sciences Faculty of Health Sciences

Lecturer Veli-Pekka Ranta, Ph.D.

School of Pharmacy Faculty of Health Sciences

Distributor:

University of Eastern Finland Kuopio Campus Library

P.O.Box 1627 FI-70211 Kuopio, Finland

ZZZXHIȴNLUMDVWR

Punamusta Oy Joensuu 2021

ISBN: 978-952-61-4276-0 (print) ISBN: 978-952-61-4277-7 (PDF)

ISSNL: 1798-5706 ISSN: 1798-5706 ISSN: 1798-5714 (PDF)

(7)

Author’s address: Department of Physical and Rehabilitation Medicine Kuopio University Hospital

KUOPIO FINLAND

Doctoral programme: Doctoral Programme of Clinical Research

6XSHUYLVRUV2KMDDMDW 3URIHVVRU2ODYL$LUDNVLQHQ0'3K'

Department of Physical and Rehabilitation Medicine Kuopio University Hospital

KUOPIO FINLAND

Professor Heikki Kröger, M.D., Ph.D.

Department of Surgery, Institute of Clinical Medicine, University of Eastern Finland

KUOPIO FINLAND

Docent Timo Aalto, M.D., Ph.D.

Medical Center Ikioma, MIKKELI

FINLAND

Reviewers: Docent, Eero Kyllönen, M.D., Ph.D.

Department of Physical and Rehabilitation Medicine, Medical Research Center

Oulu University

OULU FINLAND

Docent, Pekka Rantanen, M.D., Ph.D., eMBA

Department of Physical and Rehabilitation Medicine Helsinki University Hospital and University of Helsinki ESPOO

(8)

Opponent: Professor Marja Mikkelsson, M.D., Ph.D.

Faculty of Medicine University of Helsinki HELSINKI

FINLAND

(9)

Tuomainen, Iina

Preoperative predictors and surgical outcome in lumbar spinal stenosis– a prospective 10-year follow-up study

Kuopio: University of Eastern Finland

Publications of the University of Eastern Finland Dissertations in Health Sciences 637. 2021 125 p.

ISBN: 978-952-61-4276-0 (print) ISSNL: 1798-5706

ISSN: 1798-5706

ISBN: 978-952-61-4277-7 (PDF) ISSN: 1798-5714 (PDF)

ABSTRACT

Lumbar spinal stenosis (LSS) is one of the most frequent reasons for low back surgery in patients older than 65 years. Due to the aging of the population globally, incidence of LSS is likely to increase. Surgical treatment is the second- line option for symptom relief if conservative treatment has failed after three to six months. There is a gap in the literature on preoperative predictors of surgical outcomes in a long-term follow-up. Therefore, the aim of this study was to evaluate preoperative predictors and surgical outcomes of LSS in a 10-year follow-up.

A total of 102 patients who had indications for surgical treatment for LSS were operated on at Kuopio University Hospital between 2001 and 2004.

Surgical outcomes and predictive factors were assessed by questionnaires that the patients completed before the surgery and at follow-up points of WKUHHPRQWKVVL[PRQWKVRQH\HDUWZR\HDUVȴYH\HDUVDQG\HDUVDIWHU the operation.

The patients with LSS showed an improvement in symptoms up to the 10- year follow-up. At the 10-year follow-up, 68% of the patients estimated their satisfaction with the surgical outcome as “condition has been totally cured”

or “considerably improved.” During the study period, one-fourth needed a reoperation. Preoperative smokers and patients who had had a previous

(10)

OXPEDUVXUJHU\VKRZHGVLJQLȴFDQWLPSURYHPHQWLQVXEMHFWLYHGLVDELOLW\DQG SDLQXSWRWKHȴYH\HDUIROORZXSEXWQRORQJHUVKRZHGLPSURYHPHQWDWWKH

\HDUIROORZXSΖQDGGLWLRQVWDWLVWLFDOO\VLJQLȴFDQWSUHRSHUDWLYHSUHGLFWRUV for lower Oswestry Disability Index (ODI) and visual analogue scale (VAS) scores at 10 years were non-smoking status, absence of previous lumbar surgery, better self-rated health, and regular use of painkillers for less than 12 months. Furthermore, greater preoperative depressive symptoms predicted higher ODI scores in the 10-year follow-up, and preoperative dissatisfaction with life predicted higher ODI and VAS scores in the 10-year follow-up.

ΖQFRQFOXVLRQSDWLHQWVZLWK/66FDQH[SHFWVLJQLȴFDQWLPSURYHPHQWLQ symptoms up to a 10-year follow-up. Based on this study’s results, we suggest preoperative smoking cessation and screening psychological distress among patients with LSS undergoing operative treatment. These results can help medical professionals to identify which patients may be at higher risk of poor recovery from surgery even during the 10-year follow-up.

Keywords: Decompression, Surgical; Depression; Follow-Up Studies;

Lumbosacral Region; Pain; Personal Satisfaction; Preoperative Period;

Psychological Distress; Quality of Life; Surveys and Questionnaires; Spinal Stenosis; Treatment Outcome

(11)

Tuomainen, Iina

Preoperatiiviset ennustekijät ja leikkaushoidon tulokset lannerangan ahtau- mataudissa – prospektiivinen 10-vuoden seurantatutkimus

Kuopio: Itä-Suomen yliopisto

Publications of the University of Eastern Finland Dissertations in Health Sciences 637. 2021 125 s.

ISBN: 978-952-61-4276-0 (nid.) ISSNL: 1798-5706

ISSN: 1798-5706

ISBN: 978-952-61-4277-7 (PDF) ISSN: 1798-5714 (PDF)

TIIVISTELMÄ

Lannerangan ahtaumatauti (lumbar spinal stenosis, LSS) on yleisin alasel- käleikkauksen syy yli 65-vuotiailla. Leikkaushoitoa suositellaan potilaille, jos konservatiivinen hoito ei ole helpottanut oireita 3–6 kuukauden seurannassa.

Preoperatiivisten tekijöiden vaikutuksesta pitkäaikaiseen leikkaushoidon tu- lokseen on vain vähän tietoa. Tässä prospektiivisessa seurantatutkimuksessa selvitettiin lannerangan ahtaumataudin leikkaushoidon tuloksiin vaikuttavia preoperatiivisia ennustekijöitä sekä leikkaushoidon tulosta kymmenen vuoden seurannan aikana.

Potilaat (n=102), joille suositeltiin leikkausta LSS-taudin hoitoon, leikattiin Kuopion yliopistollisessa sairaalassa vuosien 2001–2004 välisenä aikana.

Preoperatiivisiä ennustekijöitä ja leikkaushoidon tulosta arvioitiin kyselylo- makkeilla, joita potilaat täyttivät ennen leikkausta, ja leikkauksen jälkeen kol- men ja kuuden kuukauden sekä, yhden, kahden, viiden ja kymmenen vuoden kohdalla.

Valtaosalla LSS-potilaista oireiston helpottuminen jatkui vielä kymmenen vuotta leikkauksen jälkeen. Potilaista 68 % arvioi tilanteensa olevan “Täy- sin parantunut” tai “Huomattavasti parempi tilanne kuin ennen leikkausta”

kymmenen vuotta leikkauksen jälkeen. LSS-potilaiden toimintakyky, kipu ja kävelymatka paranivat kymmenen vuoden seurannan loppuun. Seurannan

(12)

aikana yksi neljäsosa potilaista tarvitsi uusintaleikkauksen. Preoperatiivises- ti tupakoineiden ja aiemman alaselkäleikkauksen läpikäyneiden kipu ja toimintakyky paranivat merkitsevästi viiden vuoden seurantaan saakka, mutta ei enää kymmenen vuoden kohdalla. Preoperatiivinen tupakoimattomuus, ensimmäinen lannerangan alueen leikkaus, parempi itsearvioitu terveyden- tila, sekä kipulääkkeiden säännöllinen käyttö alle 12 kuukautta ennustivat parempaa toimintakykyä ja vähäisempää kipua kymmenen vuoden seurannan aikana. Myös preoperatiiviset korkeammat masennusoire- sekä elämän tyytymättömyyspisteet ennustivat huonompaa leikkaustulosta kymmenen vuotta leikkauksen jälkeen.

Tämän tutkimuksen perusteella lannerangan ahtaumataudin leikkaushoi- to hyödyttää valtaosaa potilaista kymmenen vuotta leikkauksen jälkeen. Tut- kimustulosten perusteella tupakoinnin lopettaminen ennen leikkausta olisi tärkeää, sekä psykiatristen tekijöiden tunnistamista suositellaan LSS-potilail- la, jotka tarvitsevat leikkaushoitoa. Tutkimustulosten avulla voidaan tunnis- taa paremmin ne potilaat, joiden riski huonompaan leikkaustulokseen on kohonnut jopa kymmenen vuotta leikkauksen jälkeen.

Avainsanat: elämänlaatu; kipu; kirurgia; kyselytutkimus; lanneranka;

leikkaushoito; masennus; oireet; preoperatiivinen hoito; selkä;

seurantatutkimus; tyytyväisyys

(13)

ACKNOWLEDGEMENTS

This study was carried out in the Department of Physical and Rehabilitation Medicine, Kuopio University Hospital. The study originally started in 2001 as the “Ennustenoosi” project, which I had the privilege to work on. I would like to express my gratitude to all the people who have taken part in this project.

My sincere thanks go to all my supervisors. Professor Olavi Airaksinen, WKDQN\RXIRU\RXUVLJQLȴFDQWJXLGDQFHDQGKHOSΖZLOODOZD\VEHJUDWHIXOIRU

\RXULQWURGXFLQJPHWRWKHVFLHQWLȴFZRUOGΖZRXOGDOVROLNHWRDFNQRZOHGJH Professor Heikki Kröger, who always helped me and thoroughly answered my questions. I am grateful to docent Timo Aalto, who guided me through every challenge I faced in this project. Thank you, all my supervisors, for HQFRXUDJLQJPHWRȴQLVKWKLVSURMHFW

I would like to express my gratitude to all my co-authors for their contributions to the research project: Professor Heimo Viinamäki, Professor Ville Leinonen, Docent Marinko Rade, Docent Sanna Sinikallio, Maarit Pakarinen, PhD, and Janne Pesonen, MD. Special thanks go to Heimo, Ville, and Sanna, who thoroughly answered my questions when needed. I would like to thank you, Maarit, for your help and guidance, especially in the beginning. I want to express my gratitude also to Marinko and Janne for your constructive comments and help.

Sincere acknowledgements belong to my reviewers, Professor Eero Kyllönen and Docent Pekka Rantanen. With your critical comments and valuable guidance, I was able to upgrade this thesis to the next level. I am grateful to have Professor Marja Mikkelsson from the University of Helsinki as my opponent.

I want to thank my friends who read my papers and, therefore, special thanks belong to Aino Rönkkö, Henna Ketola, Jonna Weisell, Heidi Virranta, and Heidi Hyttinen Also, I thank all of my friends who supported me throughout this project: Sonja Leskinen, Laura Honkalampi, Emmi Korpi, Verna Karen, Anita Ylimäki, Noora Miettinen, Anna Partanen, Elisa Pohjola, Eveliina Miettinen, and Milja Sarapää.

(14)

I would like to express my gratitude to my grandparents: Irma and Osmo Tuomainen and Taimi and Heikki Kastelli. I will always be grateful to my family: my mom Eija Kastelli, my dad Rauno Tuomainen and my brother Erno Tuomainen—thank you for your endless support and encouragement.

Finally, I will always be grateful to Niilo Ågren, without whose support and help I could not have gone beyond my limits.

I am grateful for the funding I received from The Research Committee of the Kuopio University Hospital Catchment Area for the State Research Funding, the Finnish Cultural Foundation Grant, the North Savo Regional Fund, and The Finnish Medical Foundation. I am also thankful for my position in the UEF PhD Doctoral Program of Clinical Research.

Kuopio, July 2021 Iina Tuomainen

(15)

LIST OF ORIGINAL PUBLICATIONS

This dissertation is based on the following original publications:

I Tuomainen I, Aalto T, Pesonen J, Rade M, Pakarinen M, Leinonen V, Kröger H & Airaksinen O. Unfolding the outcomes of surgical treatment of lumbar spinal stenosis—A prospective 5- and 10- year follow-up study. European Spine Journal. 2020; 29(9):2231–42.

II Tuomainen I, Pesonen J, Rade M, Pakarinen M, Leinonen V, Kröger H, Airaksinen O & Aalto T. Preoperative predictors of better long-term functional ability and decreased pain following LSS surgery: A prospective observational study with a 10-year follow-up period. Spine. 2020;45(11):776–83.

III Tuomainen I*, Pakarinen M*, Aalto T, Sinikallio S, Kröger H, Viinamäki H & Airaksinen O. Depression is associated with the long-term outcome of lumbar spinal stenosis surgery: A 10-year follow-up study. The Spine Journal. 2018;18(3):458–63.

IV Pakarinen M, Tuomainen I, Koivumaa-Honkanen H, Sinikallio S, Lehto SM, Airaksinen O, Viinamäki H & Aalto T. Life

dissatisfaction is associated with depression and poorer surgical outcomes among lumbar spinal stenosis patients. International Journal of Rehabilitation Research. 2016;39(4):291–5.

*Equal contribution

The publications were adapted and reprinted with the permission of the copyright owners.

(16)

(17)

points according to the depressive burden status at the 10-year follow-up ...68 Table 6. Mixed model for longitudinal ODI score and VAS score ...69 Table 7. Clinical characteristics at the 10-year follow-up in relation to the life dissatisfaction burden ...70 7DEOH 7KHULVN25&ΖRIDSRRUVXUJLFDORXWFRPHGHȴQHGE\

disability and pain intensity (VAS) at the 10-year follow-up according to the logistic regression models ...71

(21)

LIST OF FIGURES

Figure 1. Spinal canal. Adapted from Lurie et al. (7). ...26 Figure 2. A pathological pathway of disc degeneration (91). ...30 )LJXUH $SDWKRORJLFDOSDWKZD\RIOLJDPHQWXPȵDYXP/)K\SHUWURSK\

(87). ...31 )LJXUH $ȵRZFKDUWLOOXVWUDWLQJWKHVWXG\SRSXODWLRQȇVSURJUHVVGXULQJ

the follow-up periods. ...52 )LJXUH 'LVWULEXWLRQRIUHVSRQVHVWRWKHTXHVWLRQȊ+RZVDWLVȴHGDUH\RX

with the surgical outcomes?” ...60 Figure 6. The line charts of disability (ODI); overall pain (VAS); low back

pain at rest (NRS-11); leg pain while walking (NRS-11); and walking ability (meters). ...62 Figure 7. Mean of ODI and VAS scores in patients who preoperatively

VPRNHGDWGL΍HUHQWWLPHSRLQWV ...64 Figure 8. Mean of ODI and VAS scores in patients who had a previous

OXPEDURSHUDWLRQDWGL΍HUHQWWLPHSRLQWV ...65

(22)

ABBREVATIONS

LSS lumbar spinal stenosis IVD intervertebral disc

$) DQQXOXVȴEURVXV NP nucleus pulposus /) OLJDPHQWXPȵDYXP

LP leg pain

LBP low back pain

CT computed tomography MRI magnetic resonance imaging EMG electromyography

PAD peripheral arterial disease ABI ankle-brachial index

PROMS patient-reported outcome measures ODI Oswestry Disability Index

VAS visual analogue scale NRS-11 numeric rating scale – 11 IPD interspinous process device SPWT self-paced walking test

ASA American Society of Anesthesiologist BDI Beck Depressive Inventory

LS life satisfaction WAI Work Ability Index BMI body mass index QQ quantile-quantile ANOVA analysis of variance SD standard deviation

&Ζ FRQȴGHQFHLQWHUYDO 16 QRWVLJQLȴFDQW

IQR interquartile range NA not applicable

MIC minimal important change

(23)

1 INTRODUCTION

Neurogenic claudication due to lumbar spinal stenosis (LSS) (i.e., narrowing RIWKHVSLQDOFDQDOZDVȴUVWGHVFULEHGE\9HUELHVWLQ+HGHVFULEHG seven patients whose symptoms appeared only while standing and walking, but the patients were asymptomatic while resting (1). These symptoms arose from the narrowing of the spinal canal, nerve root canals, or intervertebral foramina, which can produce pressure around the nerve structures caused by bone or soft tissue (2).

It has been estimated that 102 million individuals worldwide are diagnosed with spinal stenosis annually, and it is one of the most common lumbar degenerative diseases (3). Furthermore, LSS is the most frequent reason for low back surgery in patients older than 65 years (4).

/66LVGHȴQHGDVDQDUURZHGGLDPHWHURIWKHVSLQDOFDQDOODWHUDOQHUYH canals, or neural foramina (5). Stenosis may place in the bony canal alone or the dural sac or both (2). The most common reason for LSS is degenerative changes in the spinal canal (6). LSS can cause symptoms that can lead to debilitating conditions (7). Older patients with LSS are at higher risk for functional-ability limitations and greater pain experience compared to controls (8). One study found that diagnosis of spinal stenosis and smoking were additional risk factors for persistent disability among elderly adults who schedule primary care visits for back pain. Among these patients, depressive symptoms and longer duration of symptoms were also related to persistent disability (9).

/DFNRIV\PSWRPFRUUHODWLRQZLWKUDGLRORJLFDOȴQGLQJVDQGDV\PSWRPDWLF /66ȴQGLQJVLQUDGLRORJLFDOLPDJLQJKDVEHHQJHQHUDWHGKHWHURJHQHLW\LQ study groups. A shared decision-making approach is recommended when considering treatment options in patients with LSS (7,10). Surgical treatment is usually suggested when moderate to severe symptoms with positive ȴQGLQJVLQUDGLRORJLFLPDJLQJSHUVLVWGHVSLWHFRQVHUYDWLYHWUHDWPHQWDIWHU three to six months (10–12). Decompression procedure is a conventional surgical treatment, but spinal fusion and minimal invasive techniques can

(24)

be used. One study showed that decompressive surgery may be the most FRVWH΍HFWLYHWUHDWPHQWFKRLFHIRUSDWLHQWVZLWK/66

Surgical treatment has been shown to be superior to conservative WUHDWPHQWDWOHDVWIRUIRXUWRVL[\HDUVZLWKWKHEHQHȴWVGHFUHDVLQJRYHUWLPH Ȃ3UHRSHUDWLYHIDFWRUVWKDWPD\KDYHDQH΍HFWRQVXUJLFDORXWFRPHV have been found, but the follow-up periods in previous studies investigating associations other than psychological factors with the surgical outcomes have been up to four years (21–35). For example, preoperative smoking has shown WRKDYHQHJDWLYHH΍HFWRQSRVWRSHUDWLYHRXWFRPHVXSWRIRXU\HDUIROORZXS (26–28). Furthermore, depression and life dissatisfaction have been reported WRKDYHDQHJDWLYHLPSDFWRQVXUJLFDORXWFRPHVXSWRWKHȴYH\HDUIROORZXSV for patients undergoing decompressive surgery due to LSS (24,36–43).

(25)

2 REVIEW OF THE LITERATURE

2.1 Anatomy and function of the lumbar spine

2.1.1 The bony structure of the lumbar spine

The human spine consists of 33 vertebrae, comprising seven cervical YHUWHEUDHWKRUDFLFYHUWHEUDHȴYHOXPEDUYHUWHEUDHȴYHPHUJHGVDFUDO vertebrae, and four coccygeal bones (44,45). The cervical vertebrae from three to seven, thoracic, and lumbar vertebrae are similarly shaped (45).

The sacrum forms a triangular bone shape, and its terminal combines to the coccyx (44). The main functions of the vertebral column are to cover neural and vascular structures, to stabilize the body, and allows the movement of the trunk. The lumbar spine is intended to maintain the compression load of the upper body (45,46).

The lumbar vertebrae consist of a body, two pedicels, two lamina, four articular facets, two processes transversus and a spinous process. Between two bodies of the vertebrae is an intervertebral disc. In addition to the disc, two adjacent vertebrae are connected to each other by articular facet joints.

Facet joints consist of inferior articular processes of the vertebra above the joint and superior articular processes of the vertebra below the joint, and their function is to quantify the range and type of movement of the spine (45,47).

2.1.2 The connective tissue structures of the lumbar spine

$QLQWHUYHUWHEUDOGLVFΖ9'KDVWKUHHGL΍HUHQWVWUXFWXUHVΖWFRQVLVWVRIWZR OD\HUVDQQXOXVȴEURVXV$)LVWKHH[WHUQDOȴEURXVULQJDQGWKHLQWHUQDOFRUH LVQXFOHXVSXOSRVXV13$)LVDQHODVWLFULQJWKDWFRQWDLQVFROODJHQȴEHUV ZKLFKDUHDUUDQJHGLQOD\HUVΖQDGGLWLRQWRHODVWLQȴEHUVFROODJHQȴEHUVW\SHΖ DQGΖΖDUHPDMRULW\ȴEHUVLQWKHΖ9'13FRQWDLQVFROODJHQȴEHUVHODVWLQȴEHUV and proteoglycans. Proteoglycans generate hydrostatic pressure, leading to water-containing gel. Therefore, the function of the disc is to support compressive loads and enable spine movements. The third structure in the

(26)

disc is an end plate, which contains hyaline cartilage, and it is a horizontal structure located above and below the two layers. The end plate combines the IVD and vertebra body together (45,48–51).

7KHSDUDVSLQRXVPXVFOHVFDQEHGLYLGHGLQWRWKHGHHSHUDQGVXSHUȴFLDO OD\HUV7KHGHHSOD\HUFDQEHVXEGLYLGHGLQWRVXSHUȴFLDODQGGHHSJURXSV7KH VXSHUȴFLDOJURXSFRQWDLQVSDUDVSLQRXVHUHFWRUVSLQDHJURXSZKRVHIXQFWLRQ is to maintain erect posture. Deep group contains transversospinalis muscle groups, which provide lateral stability and rotation of the spine. Deeper than transversospinalis muscles are the interspinal and intertransverse muscles (45).

2.1.3 The neural structures and formation of the lumbar spinal canal

The bony ring of the spinal canal forms from the vertebral body, the pedicles, and the lamina. The spinal canal is surrounded anteriorly by the body of vertebrae and discs, laterally by the pedicles, and posterolaterally by the ODPLQDDQGOLJDPHQWXPȵDYXP/)ΖQVLGHWKHVSLQDOFDQDOLVWKHVSLQDOFRUG which is part of the central nervous system and contains grey and white matter. Grey matter includes the motor neurons in the anterior horn and sensory neurons in the posterior horn, and the intermediate grey zone includes autonomic neurons. In the middle of the central canal and around the grey matter is white matter, which consists of ascending, descending, and propriospinal neural pathways. The central canal contains cerebrospinal ȵXLGȂ

The spinal cord is covered by three membranes: dura mater, arachnoid, DQGSLDPDWHUZKLFKDUHSODFHGIURPVXSHUȴFLDOWRGHHSHVWUHVSHFWLYHO\7KH epidural space, which contains a layer of fat, is located between the dura mater and vertebral canal. The spinal cord terminates at the L1 level, and below it is the cauda equina, which is a bundle of descending nerve roots.

The spinal cord and cauda equina are covered by dural sac, the membrane that consists of dura mater and arachnoid mater (55,56).

The soft tissues in the spinal canal are the cauda equina, the three membranes, the dural sac, the epidural fat, and the epidural veins (54).

LF is attached to the lamina of the vertebrae, and it is one of the clinically

(27)

meaningful ligaments in the spinal column. In addition to LF, several ligaments exist within the vertebral column, and their function is to control and stabilize the spinal column motion (45).

The spinal cord consists of segments, in which a dorsal root and a ventral root combine to form a spinal nerve. Spinal nerves leave from the spinal canal WKURXJKLQWHUYHUWHEUDOIRUDPHQ7KHUHDUHVSLQDOQHUYHVȴYHRIZKLFKDUH named after the lumbar area (56). Sacral roots leave the spinal canal through sacral foramen (44).

2.1.4 Vascular structure

The spinal cord receives arterial supply from the segmental arteries, the anterior spinal artery, and two posterior spinal arteries. The spinal trunks of segmental arteries divide to the anterior and posterior radicular arteries after entering the spinal canal through intervertebral foramen. The radicular arteries supply the dura and nerve root at each level. The radicular arteries branch several levels as the radiculomedullary arteries supplying the spinal cord through the anterior and posterior spinal arteries. Arterial branches form various anastomoses between each other, and the arterial plexuses form the arterial network on the entire surface of the spinal cord between WKHSRVWHULRUVSLQDODQGDQWHULRUVSLQDODUWHULHVb

The anterior median vein follows the anterior spinal artery. The posterior spinal veins receive drainage from the radial veins of the dorsal spinal cord. The anterior and posterior median spinal veins combine to form radiculomedullary veins, which follow the anterior and posterior spinal nerve root. The anterior and posterior spinal veins are united in the plexus in epidural space via the anterior and posterior radiculomedullary veins (57).

(28)

Figure 1. Spinal canal. Adapted from Lurie et al. (7).

2.2 LSS

2.2.1 Epidemiology

Approximately 102 million individuals are diagnosed with spinal stenosis yearly, and the highest incidence was seen in Europe (3). Prevalence of LSS increases with age, and the frequency of diagnosis is higher especially among patients older than 60 years (58). Furthermore, only 9.8% of patients with diagnoses of LSS are younger than 51 years (59).

In the United States, prevalence of LSS with coexisting spondylolisthesis and scoliosis increased nationwide from 2004 to 2009. In the same time period, 63.3% of patients who were discharged from hospitals for LSS were over 65 years old (60). Likewise, during the 2000s, the surgical hospitalization rate for /66LQFUHDVHGE\LQWKH8QLWHG6WDWHV$FRKRUWVWXG\UHȵHFWLQJWKH Japanese general population revealed that 76% of the study population had

(29)

radiographically evaluated stenosis in the lumbar spine. However, only 10%

of them had symptoms due to LSS (62).

2.2.2 (WLRORJ\DQGFODVVLȴFDWLRQRI/66

Etiology in LSS can be divided into subgroups of congenital or acquired stenosis or a combination of both (2). In congenital lumbar stenosis, patients have a shorter pedicle length and, as a consequence, a smaller cross-sectional spinal canal area in radiologic imaging studies. Furthermore, patients with congenital stenosis are usually younger, have fewer degenerative changes, and deal with stenosis at multiple levels (63). Acquired stenosis is caused most commonly by degenerative changes on the lumbar spine but disc herniation, vertebral displacement, spondylolisthesis or spondylolysis with FRH[LVWLQJROLVWKHVLVWUDXPDV3DJHWȇVGLVHDVHWXPRUVVRIWWLVVXHFDOFLȴFDWLRQ RVVLȴFDWLRQDQGȵXRURVLVGL΍XVHLGLRSDWKLFK\SHURVWRVLVDQGVFROLRVLVFDQ also cause a narrowed lumbar canal (2,64,65).

Only few studies have explored genetic predispositions of LSS. Degenerative LSS may have a multi-factorial origin due to the presence of multiple single QXFOHRWLGH SRO\PRUSKLVPV DQG GL΍HUHQW JHQHWLF SUHGLVSRVLWLRQV LQ WKH upper and lower lumbar areas (66). It has been estimated that there is 66.9%

heritability for LSS and 81.2% heritability for the smaller dural sac cross- sectional area (67). A small-sample-size study in Finland demonstrated a relationship between genetic factors and pathology of degenerative stenosis (68).

Low birth weight, maternal smoking, low placental weight, and lower VRFLRHFRQRPLFVWDWXVPLJKWKDYHDQLQȵXHQFHRQGLPLQLVKHGJURZWKRIWKH lumbar vertebral canal in childhood, and therefore they may be risk factors for developing spinal stenosis in adult life (69). Incidence of LSS is high among patients with skeletal dysplasia achondroplasia (70). In later life, heavy repeated physical movements—for example, heavy lifting at work—and other SK\VLFDOOLIHVW\OHVD΍HFWLQJWKHVSLQDOFDQDOPD\KDYHDQDVVRFLDWLRQZLWKWKH development of LSS. In females, a higher amount of labor may also have a correlation to acquired LSS (71). Furthermore, obese and overweight people are at higher risk for developing LSS (71,72).

(30)

LSS can be divided anatomically into central or lateral stenosis. Central stenosis occurs at the intervertebral level, and lateral stenosis can occur at either the nerve root canal (or lateral recess) or the intervertebral foramen (or neural foramen). Nerve root exits from the spinal canal through canal that consists of the lateral recess and neural foramen. The lateral lumbar spinal canal can also be divided into three anatomic zones: entrance zone, midzone, and exit zone (73).

The entrance zone of the lateral lumbar canal is placed medial to the pedicle and is below the superior articular process of the zygapophyseal joint.

The entrance zone area is also known as the lateral recess area. The midzone of the lateral canal is below the pars interarticularis of the lamina and under the pedicle. The exit area of the lumbar spinal canal is intervertebral or neural foramen (73). Lateral and central stenosis can appear together, characterized as a mixed stenosis (74).

2.2.3 Degeneration process

The degeneration process commonly takes place in IVDs, facet joints, and OLJDPHQWXP ȵDYXP (SLGHPLRORJ\ LQ GLVF GHJHQHUDWLRQ YDU\ EHFDXVH RI WKHODFNRIDVWDQGDUGL]HGGHȴQLWLRQRIGLVFGHJHQHUDWLRQLQUHVHDUFK Nevertheless, the prevalence of degeneration processes in the spine is likely to increase because the degree of disc degeneration and facet joint arthrosis are age-related and rise during aging. Degeneration processes regularly take place at L4–S1 vertebral levels (75–77). Thickening of the LF is also age-related, DQG DIWHU WKH DJH RI WKLFNHQLQJ RI OLJDPHQWXP VLJQLȴFDQWO\LQFUHDVHV at levels L3–L4. The thickness of the LF is greater at levels L3–4 and L4–5 compared to other lumbar levels (78,79).

/LIHVW\OHȃIRUH[DPSOHVPRNLQJDQGXSULJKWSRVWXUHȃKDYHDQH΍HFWRQGLVF GHJHQHUDWLRQΖQWZLQVWXGLHVKHULWDELOLW\KDVSOD\HGDVLJQLȴFDQWUROH in etiology of disc degeneration (75). Furthermore, some genotype and allele frequencies have been found to be associated with lumbar disc degeneration (81,82). Some genetic expressions, multiple genetic polymorphisms, and signal pathways have been found to have an association with presence of LQFUHDVHGLQȵDPPDWRU\F\WRNLQHVDQGJURZWKIDFWRUVȂ+RZHYHU

(31)

evidence of the genetic association with disc degeneration is mostly fragile (88,90).

2.2.3.1 Degeneration of IVDs

The main pathway in the degeneration process is reduced expression of proteoglycans and collagen type II, which leads to reduced capacity to absorb water (91). Therefore, hydrostatic pressure in nucleus pulposus declines, and dehydration occurs, which leads to disc height loss (48). The degeneration SURFHVVDQGPHFKDQLFDOSUHVVXUHFDXVHVWKHLQȵDPPDWLRQSURFHVVUHVXOWLQJ LQHOHYDWHGOHYHOVRILQȵDPPDWRU\F\WRNLQHVPRVWO\Ζ/E and TNF-alfa. The LQȵDPPDWLRQ SURFHVV DFWLYDWHV D F\FOH WKDW OHDGV WR WLVVXH GDPDJH DQG LQȵDPPDWLRQWKDWFDXVHVFHOOGHDWKRI13FHOOVDQGUHGXFWLRQRISURWHRJO\FDQV DQGFROODJHQW\SHΖΖΖQDGGLWLRQWKHVHFKDQJHVLQȵXHQFHWKHGLVFȇVDELOLW\

WRFRSHZLWKPHFKDQLFDOSUHVVXUHDQGȵH[LELOLW\ZKLFKFDXVHVGLVWXUEDQFHV in stress distributions between NP, AF, and end plates. These changes have DQH΍HFWRQGLVFVWUXFWXUHOHDGLQJWRDPRUHȴEURWLFVWUXFWXUHȃIRUH[DPSOH tears and higher levels of collagen type I and type III in AF and sclerosis or fractures in end plates (50,91–93). Immobilization and extensive exercise have been shown to be harmful to disc metabolism (94). Kauppila et al. found that atherosclerosis in the abdominal aorta and arteries supplying the lumbar spine were related to disc degeneration (95).

(32)

Figure 2. A pathological pathway of disc degeneration (91).

2.2.3.2 'HJHQHUDWLRQRIIDFHWMRLQWVDQGOLJDPHQWXPȵDYXP

Decreased disc height due to the degeneration process contributes to restricted motion of the spine and unphysiological loading to joints in the vertebral column. Degenerative discs also induce a caudal dislocation of the inferior articular process. These changes can lead to degenerative changes in facet joints (49). Facet joint arthrosis has been shown to be higher in patients with stenosis compared to a control group (77).

/)K\SHUWURSK\FRQVLVWVRIDKHWHURJHQHRXVSURJUHVVLRQVWDWHRIȴEURVLV (83), which is the main reason for LF hypertrophy (78). LF mostly consists of HODVWLFȴEHUVDQGLQDGGLWLRQWRȴEURVLVORVVRIHODVWLFȴEHUVRFFXUVLQ hypertrophied LF (78). Hypertrophy of the ligamentum can occur alone or in association with facet joint degeneration (76). Thickening of the LF initially results from mechanical stress that induces tissue damage. In damaged WLVVXHH[SUHVVLRQRIJURZWKIDFWRUVDQGLQȵDPPDWRU\F\WRNLQHVLQFUHDVH OHDGLQJWRVFDUIRUPDWLRQGXULQJWKHUHSDLULQJSURFHVV7KLVOHDGVWRȴEURVLV

(33)

of thicked LF (87). Moreover, angiogenesis has been shown to have a role in the pathogenic process in hypertrophy after mechanical stress (96).

'LDEHWHV PD\ LQFUHDVH WKH ORVV RI HODVWLF ȴEHUV ΖW KDV DOVR EHHQ suggested that cytokines, such as IL-6 concentration, are higher in FHUHEURVSLQDOȵXLGLQSDWLHQWVZLWK/66DQGWKHFRQFHQWUDWLRQLVDVVRFLDWHG with the radiological severity of stenosis (98). One study reported higher IL-6 serum levels in patients with diagnoses of spinal stenosis or degenerative disc disease compared to controls and patients with disc herniation (99).

Figure 3. $SDWKRORJLFDOSDWKZD\RIOLJDPHQWXPȵDYXP/)K\SHUWURSK\

(87).

2.2.4 Pathophysiology of LSS

The most common etiology in LSS is degenerative changes in the lumbar VSLQDOFDQDO7KHGHJHQHUDWLYHSURFHVVSUHGRPLQDQWO\D΍HFWVΖ9'VDQG/) UHVXOWLQJLQGLVFKHLJKWORVVDQGWKLFNHQLQJRIȵDYXPUHVSHFWLYHO\

Decreased disc height during the degeneration process can cause a bulking of LF toward the spinal canal (100). Degenerative changes in facet joints cause facet joint arthrosis (49). These changes lead to the narrowing of the central canal and intervertebral foramina. Reduced space available for the neural and vascular elements results from narrowing of the bone canals of the lumbar spine, resulting in compression of neural structures (2,10).

In addition to a narrowed spinal canal, three pathophysiology mechanisms have been suggested as reasons for symptoms due to LSS (101). According WRDSRVWXUDOWKHRU\GXULQJȵH[LRQWKHUHLVPRUHVSDFHDYDLODEOHDURXQGWKH neural structures in the spina canal, lateral recess, and neural foramina. On the FRQWUDU\WKHUHDUHVLJQLȴFDQWUHGXFWLRQVLQWKHVHDUHDVLQOXPEDUH[WHQVLRQ resulting in the narrowing of the canals around the neural structures (102).

As the spinal canal is already narrowed by surrounding tissues, movements of cauda equina nerve roots are constricted, especially in lumbar extension.

(34)

During walking, the mechanical compression increases around the cauda equina, and this compression occurs repeatedly. Acute repeated compression PD\FDXVHFKURQLFFRPSUHVVLRQOHDGLQJWRGHJHQHUDWHGQHUYHȴEHUV

$FFRUGLQJ WR D YHQRXV WKHRU\ D PLOG SUHVVXUH EORFN DW WZR GL΍HUHQW levels can cause vena congestion between these occlusions, which thereby LQGXFHVDUWHULDOEORRGȵRZGXHWRKLJKHUSUHVVXUH'HVSLWHWKHKLJKHUDUWHULDO SUHVVXUH WKH EORRG ȵRZ WKH R[\JHQ VXSSO\ DQG WKH QXWULWLRQ GHFOLQH and metabolites increase due to vena congestion (64). Higher demand of oxygen during exercise also causes arterial vasodilation, which can lead to an increased block to an already narrowed spinal canal. Increased capillary pressure caused by vena congestion can cause intraradical oedema, along ZLWK GDPDJH WR WKH EORRGQHUYH EDUULHU DQG DQ LQȵDPPDWRU\ UHVSRQVH

$FFRUGLQJWRDQLVFKHPLFWKHRU\LQDGHTXDWHEORRGȵRZFDXVHV symptoms due to LSS when the higher demand for oxygen cannot be met during exercise (101).

2.2.5 Diagnosis of LSS

$IUDJLOHDVVRFLDWLRQEHWZHHQUDGLRORJLFDOȴQGLQJVDQGV\PSWRPVGXHWR/66 as well as the lack of standard criteria for diagnosis and inclusion criteria for study populations, has resulted in heterogenous study groups of patients ZLWK /66 7KHUHIRUH FODVVLȴFDWLRQ FULWHULD IRU QHXURJHQLF FODXGLFDWLRQ resulting from LSS have been developed to improve homogeneity of studies.

The N-CLASS criteria includes patient-reported symptoms such as pain in both legs, leg pain revealed by sitting, leg pain decreased by leaning forward RU ȵH[LQJ WKH VSLQH DQG FOLQLFDO ȴQGLQJV QHJDWLYH VWUDLJKW OHJ WHVW DQG positive 30s extension tests). Age over 60 years is associated with neurogenic FODXGLFDWLRQGXHWR/66ΖQDGGLWLRQWRDJHKLVWRU\ȴQGLQJVVXFKDV bilateral symptoms, leg pain more than back pain, pain during walking or standing, and pain relief upon sitting have been found to be the best evidence in clinical diagnostic accuracy (106,107). Furthermore, patients may experience motor or sensory disturbances while walking and low back pain (108).

7KHUHLVDOVRFRQVLGHUDEOHGLDJQRVWLFYDULDWLRQRITXDOLȴFDWLRQFULWHULDIRU research study groups. The most common criteria that has been used in

(35)

study groups are (1) age older than 50; (2) presence of accompanying back, EXWWRFNRUOXPERSHOYLFSDLQFKDUDFWHULVWLFRIUDGLDWLQJSDLQGHȴQLWLRQ RIQHXURJHQLFFODXGLFDWLRQUDGLRORJLFDOFRQȴUPDWLRQRIVWHQRVLVH[SOLFLW description of radiological stenosis; (7) correlation between symptoms and OHYHORIUDGLRORJLFDOVWHQRVLVVSHFLȴFLQFOXVLRQRUH[FOXVLRQRIVRPHRWKHU UDGLRORJLFDOȴQGLQJVDQGH[FOXVLRQFULWHULDRIRWKHUVSLQDORUQRQVSLQDO problems. However, no inclusion or exclusion criteria regarding clinical examination were found (109).

2.2.6 &OLQLFDOV\PSWRPVDQGȴQGLQJV

The most common symptom in LSS is a neurogenic claudication (12).

1HXURJHQLFFODXGLFDWLRQD΍HFWVZDONLQJDELOLW\ȃZKHQSDWLHQWVIHHOGLVFRPIRUW pain, or numbness in lower extremities and when patients have to stop walking because of the symptoms. Symptoms arise due to compression of WKHQHXUDOVWUXFWXUHVLQWKHOXPEDUVSLQDOFDQDODQGRULQIRUDPLQD3DWLHQWV XVXDOO\IHHOV\PSWRPUHOLHIZKHQEHQGLQJIRUZDUGȃOXPEDUVSLQHLQȵH[LRQ $ ȵH[HG OXPEDU VSLQH SURGXFHV PRUH VSDFH IRU QHXUDO VWUXFWXUHV and therefore releases the compression of neural structures. On the other hand, lumbar extension—for example, walking downhill—usually worsens the symptoms due to LSS (102).

In addition to neurogenic claudication, the most prevalent symptoms are low back pain and leg pain. Pain relief when seated and pain exacerbating in OXPEDUH[WHQVLRQDUHIDPLOLDUFOLQLFDOȴQGLQJVLQSDWLHQWVZLWK/66ΖW has been shown that approximately 40% of patients with LSS have equal pain in their low back and lower extremes, whereas one-third reported predominant leg pain, and one-fourth reported dominating pain in their low back (30).

Sigmundsson et al. found that patients with LSS and spondylolisthesis had higher back-pain levels compared to leg-pain levels, when the comparison was made between them and patients with central spinal stenosis or lateral recess stenosis who had indications for surgery. In addition, patients with lateral recess stenosis had the lowest back-pain intensity, and their walking distance was better. Quality of life and disability scores were similar between WKHVHWKUHHGL΍HUHQWGLDJQRVLVJURXSV1HXURJHQLFFODXGLFDWLRQLVD

(36)

common symptom in patients with central stenosis, and radicular symptoms often result from lateral spinal stenosis (73).

Physical examinations can be normal in patients with LSS—thus, a patient’s history and symptoms are more useful when diagnosing LSS (107). However, ȴQGLQJVGXULQJSK\VLFDOH[DPLQDWLRQVVXFKDVZLGHEDVHGJDLWDEQRUPDO Romberg test results, leg pain after 30 seconds in lumbar extension, and QHXURPXVFXODUGHȴFLWVKDYHEHHQVKRZQWREHPRVWVWURQJO\UHODWHGWRD diagnosis of LSS (105,110). Symmetric arterial pulses in lower extremities are usually present, and lower-extremity weakness may be found in patients with LSS (108).

0RVWFRPPRQO\SDWLHQWVKDYHVX΍HUHGV\PSWRPVIRUWKUHHWRVL[PRQWKV before seeing a specialist, but one-third have had symptoms for more than one year (112). Co-presence of degenerative scoliosis of the lumbar spine can correlate to symptom severity (113). The natural course of symptoms seldomly aggravates, especially in patients with mild to moderate symptoms due to LSS. However, a dural sac area less than 50 mm²LQUDGLRORJLFDOȴQGLQJV may imply the lack of spontaneous symptom improvement over time (113).

2.2.7 5DGLRORJLFDOȴQGLQJVLQ/66

Plain radiographs, myelography, computed tomography scan (CT), and magnetic resonance imaging (MRI) are radiological modalities that can be used when evaluating spinal stenosis. All of the radiological imaging can be completed in statistic and dynamic positions with the aim of improving GL΍HUHQWD[LDOORDGLQJWRHYDOXDWHGL΍HUHQWW\SHVRIVWHQRVLV

0\HORJUDSK\ZDVWKHȴUVWWHFKQLTXHXVHGLQGLDJQRVLQJVSLQDOVWHQRVLVDV Verbiest declared the measurement of the anteroposterior diameter of the dural sac in absolute stenosis 10 mm or less and 12 mm for relative stenosis (54). However, due to its invasive methodology, myelography is rarely used nowadays (54,107). MRI is the most useful imaging test for LSS (107). MRI provides the most accurate sequences of the spinal cord and soft tissues of the spine. However, CT provides a better evaluation for bone tissues compared to MRI (114).

ΖQ 05Ζ DQG &7 LPDJLQJ VWXGLHV FHQWUDO VWHQRVLV FDQ EH GHȴQHG E\ D measurement of the anteroposterior diameter of the spinal canal from the

(37)

sagittal segments. From the transverse segments, radiological stenosis can be evaluated by a measurement of the transverse diameter of the spinal canal and the cross-sectional area of the dural sac (115–117). The dural sac DUHDHQFRPSDVVHVWKHDUHDRIFHUHEURVSLQDOȵXLGLQVLGHWKHGXUDPDWHU$

cross-sectional area of less than 100 mm²or more at two or more levels is considered related to symptoms of stenosis (118,119). The critical measure that interferes with the normal function of neural structures is considered to be less than 75 mm² (120,121). Furthermore, indentifying a relation between FDXGDHTXLQDDQGFHUHEURVSLQDOȵXLGPD\LPSURYHGHȴQLQJWKHDFFXUDQF\

the degree of radiological stenosis (117,122).

Lateral stenosis can be illustrated by a measurement of height and length of the recess or a nerve root compression of lateral recess. In addition, angle of lateral recess can be used to evaluate lateral stenosis. Foraminal stenosis can be characterized by measurements of the foraminal diameter or foraminal nerve root impingement (115–118).

In addition to former criteria, disc protrusion, lack of perineural intraforaminal fat, hypertrophic facet joint degeneration, and hypertrophy of the LF can be used in determining the radiological degree of stenosis.

+RZHYHUWKHUHLVDODFNRIXQLW\IRUGHȴQLQJGLDJQRVWLFFULWHULDIRU/66LQ radiological image investigations (115–117,123).

$VVRFLDWLRQV EHWZHHQ UDGLRORJLFDO ȴQGLQJV DQG WKH GHJUHH RI VHYHULW\

of clinical symptoms have seldom been found (124,125). For example, no associations were found between radiological degree of severity of central LSS and clinical symptoms such as pain, disability, or walking ability (126).

Moreover, a narrowed dural sac cross-sectional area had a weak association with the presence of low back pain (127). Lateral lumbar spinal canal stenosis seen on MRI were shown to be associated with electromyography (EMG) ȴQGLQJVEXWQRWZLWKSDWLHQWVȇV\PSWRPVLQUHIHUHQFHWRGLVDELOLW\EDFNRU leg pain, or walking distance (128). The lack of correlation between symptoms DQGUDGLRORJLFDOȴQGLQJVPD\EHGXHWRLPDJLQJSDWLHQWVLQDVXSLQHSRVLWLRQ When imaging patients are subjected to axial loading or are in an upright position, results can provide more accurate associations between the imaging and clinical symptoms (129–131). In addition, axial loading in MRI establishes DGGLWLRQDOLQIRUPDWLRQWKDWFDQD΍HFWWKHWUHDWPHQWGHFLVLRQ

(38)

In addition to imaging testing, electrodiagnostic studies can be done, which can show reduced nerve root conduction (101). A study by Haig et al.

demonstrated that electromyography results can distinguish symptomatic and asymptomatic patients with LSS, which was not reliable in MRI imaging (133). Therefore, electrodiagnostic tests in patients with atypical symptoms RI/66KDYHEHHQVXJJHVWHGWRFRQȴUPFOLQLFDOVLJQLȴFDQFHRIUDGLRORJLFDO ȴQGLQJV

2.2.8 'L΍HUHQWLDOGLDJQRVLVLQ/66

7KH PRVW FRPPRQ GL΍HUHQWLDO GLDJQRVLV LQ /66 LV YDVFXODU FODXGLFDWLRQ caused by peripheral arterial disease (PAD), where patients have to stop walking due to leg pain (135). In patients who have atypical claudication, the DQNOHEUDFKLDOLQGH[$%ΖKDVVKRZQWRKDYHJRRGVHQVLWLYLW\DQGVSHFLȴFLW\

IRUGL΍HUHQWLDOGLDJQRVLVEHWZHHQ3$'DQG/66

In patients with disc herniation as their main diagnosis, factors such as greater leg-pain intensity, greater disability, and pain in the anterior thigh and leg are more frequently present than in patients with LSS (137). Furthermore, patients with hip osteoarthritis may have similar symptoms to patients with LSS (138).

The most common comorbid diseases reported in LSS patients are FRQFRPLWDQWMRLQWGLVHDVHK\SHUWHQVLRQKHDUWGLVHDVHVWRPDFKLQWHVWLQDO problems, diabetes, and osteoporosis (112). Diabetes and low ABI measures were found to have an association with moderated symptomatic LSS (139).

2.2.9 Conservative treatment

Conservative treatment in patients with LSS can consist of oral medication, epidural injections or blocks, and physical therapy as well as a combination of these conservative managements (140). Oral medication commonly includes SURVWDJODQGLQVJDEDSHQWLQQRQVWHURLGDODQWLLQȵDPPDWRU\GUXJV16$Ζ' paracetamol, opioids, antidepressant, vitamin B₁₂, and corticosteroids (7,141–

143).

Physical therapy can include muscle stretching and exercise, balance, or muscle-coordination training, walking with or without canes, or treadmill ambulation exercise. Pain-relieving management such as transcutaneous

(39)

electrical nerve stimulation, heat or ice, ultrasound, and manual therapy are reported to be part of physical therapy. Furthermore, postural instructions, braces and corsets, or lumbar semi-rigid orthosis can be part of the conservative treatment for patients with LSS (142,144). In a US cohort study (SPORT), the most frequent non-surgical treatments used were spinal injections and physical therapy (145).

7KHUH DUH RQO\ VPDOO GL΍HUHQFHV EHWZHHQ WKH EHQHȴWV RI GL΍HUHQW conservative treatments. Receiving physical therapy did not guarantee a better outcome compared to not receiving physical therapy (145). A review E\0DFHGRHWDOIRXQGVLPLODUWUHDWPHQWRXWFRPHVEHWZHHQWKHGL΍HUHQW physical therapy strategies (144). However, a study by Fritz et al. showed a lower likelihood to undergo surgical treatment within one year in patients with LSS who received physical therapy (145). There is only very-low-to-low- TXDOLW\HYLGHQFHIRUH΍HFWLYHQHVVRIFRQVHUYDWLYHWUHDWPHQWDQGWKHUHIRUH no recommendations could have been made for clinical practice (140). In addition, a study by Parker et al. revealed a failure to achieve sustainable results in conservative management up to a two-year follow-up among patients who had an indication to operative treatment (146).

Only few studies have investigated long-term conservative treatment outcomes and predictors or risk factors of conservative treatment outcomes. In a seven-year follow-up, approximately 40% of patients with mild-to-moderate symptoms had a deteriorated clinical status after conservative treatment, DQG(0*ȴQGLQJVZHUHIRXQGWREHDSUHGLFWRURIDQXQVDWLVIDFWRU\RXWFRPH In addition, severe intermittent claudication was shown to be a risk factor of an unsuccessful conservative treatment outcome (147). One study has GHVFULEHGWKHQHJDWLYHH΍HFWRIDKLJKHUQXPEHURIFRPRUELGLWLHVRQWKH conservative treatment outcome in a 12-year follow-up (148).

2.2.10 Operative treatment

In the United States, the total amount of charges for surgery in patients with LSS increased from 2010 to 2014. Furthermore, approximately 88%

of patients with LSS who were discharged from the hospital underwent operative treatment in 2014 (149). In Sweden, the mean rate of operative treatment for LSS was 9.7 per 100,000 inhabitants per year, and it increased

(40)

threefold from 1987 to 1993 (150). In Norway, surgical procedures due to LSS more than tripled from 1999 to 2013. During the same time period, the annual prevalence of surgery due to LSS quadrupled among patients over 65 years in Norway (151). Over the past 20 years, the amount of complex fusion procedures has elevated, while the rate of simple decompression procedures has diminished in the United States (60,149,152).

Based on the study results of the latest meta-analysis, no clear guidance for treatment choice in patients with LSS could have been done (153), and decisions for the choice of treatment can be subjective and vary among SK\VLFLDQV 7KHUH FDQ EH GL΍HUHQFHV LQ VXUJLFDO DSSURDFKHV regarding concomitant fusion procedures especially between geographic regions (155,156). One study revealed that one-year postoperative disability LQSDWLHQWVZLWK/66VLJQLȴFDQWO\YDULHGEHWZHHQGL΍HUHQWFOLQLFLDQV However, similar approaches to surgery have also been found. For examples, similarities of indications for surgery between the United States and Norway were found (156), and there was a consensus on surgical approaches to degenerative spondylolisthesis with stenosis among 30 orthopedic surgeons and neurosurgeons (155).

Operative treatment is often recommended as a second line of symptom relief if conservative treatment fails to improve symptoms in three to six months (10,11,14). More importantly, surgery is considered when symptoms GXHWR/66DUHSUHVHQWZLWKSRVLWLYHȴQGLQJVIURPGLDJQRVWLFLPDJLQJ Patients with LSS can expect similar health-related quality of life after operative treatment as patients with hip or knee arthroplasty (158), and predictable FRVWH΍HFWLYHQHVV LV GHVFULEHG DV VLPLODU EHWZHHQ WKHVH WZR JURXSV XS WRWKHȴYH\HDUIROORZXS+RZHYHUWKHUHLVRQO\ORZOHYHOHYLGHQFH that surgical outcomes in reference to disability, leg pain, and back pain are better than conservative treatment in patients with LSS in long-term follow- ups (153). Therefore, a shared decision-making process with the patients is recommended when considering operative treatment (10).

2.2.11 Surgical techniques

Decompression is the most common surgical intervention for patients with LSS. The aim of the decompressive procedure is to reduce the pressure

(41)

RI WKH QHXUDO VWUXFWXUH E\ WDNLQJ R΍ ERQH DQG VRIW WLVVXHV IURP DURXQG the neural elements (160). Laminectomy is the conventional technique of the decompression procedure, especially in patients without instability or deformity (161). Less invasive decompression techniques are uni- and bilateral laminotomies and spinous process-splitting laminectomies (160,162). In addition to decompression alone, spinal fusion plus decompression and interspinous process spacers are options for surgical techniques (163).

7KHUHDUHRQO\ORZTXDOLW\PHWDDQDO\VHVWKDWKDYHUHSRUWHGGL΍HUHQFHV LQVXUJLFDORXWFRPHVEHWZHHQGL΍HUHQWVXUJLFDOSURFHGXUHV

%DVHG RQ VWXG\ UHVXOWV WKHUH DUH QR GL΍HUHQFHV LQ VXUJLFDO RXWFRPHV EHWZHHQGL΍HUHQWVXUJLFDOWHFKQLTXHVȂ)XVLRQSURFHGXUHVKDYH demonstrated a higher risk of major complications, postoperative mortality, and hospitalization (152).

Decompression with fusion is usually recommended when concomitant spondylolisthesis or spinal instability are present (167,168). After lumbar decompression alone, instability can develop, especially among patients with spondylolisthesis (169). Degenerative spondylolisthesis, spinal instability in radiologic investigations, assumption of postoperative instability, spinal deformity, recurrent stenosis, or foraminal stenosis with compressed exiting nerves caused by postsurgical disc collapse are indications for additional fusion (170,171).

2.3 Outcome measures in LSS

The operative aim is to relieve pressure around the neural structures and, as a consequence, reduce patients’ experienced pain that restricts functional ability and walking capacity (12). Therefore, surgical outcome can be measured by pain relief, improvement of disability, and patients’

expectations of treatment (172). During the last decade, the top six most- frequent patient-reported outcome measures (PROMs) in spine-related research were the visual analogue scale (VAS), Oswestry Disability Index (ODI), Short Form 36, Japanese Orthopaedic Association Outcome Questionnaire, Neck Disability Index, and Scoliosis Research Society-22 (173). A review by

(42)

'H9LQHHWDOFRQFOXGHGWKDWWKHWUHDWPHQWVSHFLȴFRXWFRPHPHDVXUHVRI pain and spine-related function ability are sensitive for changes in outcome VFRUHV DQG WKHUHIRUH DUH PRUH UHVSRQVLYH WR WUHDWPHQW H΍HFWV LQ VSLQDO research (174). Austevoll et al. concluded that PROMs’ follow-up point score or the percentage-change score from preoperative to postoperative follow- XSSRLQWVGHVFULEHPRVWDFFXUDWHO\WKHH΍HFWRIRSHUDWLYHWUHDWPHQW Furthermore, surgical outcomes evaluated by validated PROMs allows the FRPSDULVRQRIRXWFRPHVEHWZHHQGL΍HUHQWVWXGLHV

2.3.1 Satisfaction with the surgical outcome

0HDVXULQJVDWLVIDFWLRQZLWKWKHWUHDWPHQWPD\UHȵHFWSDWLHQWVȇVDWLVIDFWLRQ with the overall care and service or satisfaction with the treatment outcome 3DWLHQWVȇUHVSRQVHVWR3520TXHVWLRQVUHȵHFWPRUHRIWHQWKHFXUUHQW level of symptoms rather than the change of symptoms from the preoperative phase to the present time. Therefore, patients’ global assessments of pain LPSURYHPHQWDIWHUVXUJHU\PD\EHDQHɝFLHQWWRROWRPHDVXUHDVXFFHVVIXO outcome of degenerative lumbar spinal surgery (178). Furthermore, patients’

global assessments of surgical outcome can provide the evaluation of patients’

experiences of treatment in reference to expectations (179).

2.3.2 Disability

In spinal research, the most common functional ability measures used are the ODI and the Roland-Morris Disability Questionnaire (180,181). The ODI questionnaire measures subjective disability, and the scale ranges from 0% to 100%, where 0% indicates no disability and 100% indicates extreme disability. Therefore, a higher ODI score means a greater disability (182,183).

The questionnaire includes one question about pain and nine questions about activities of daily living (184). The ODI is a valuable measure to indicate patients’ subjective symptoms according to functional restrictions (185). The ODI scores are also found to be associated with walking capacity in patients with LSS (186). A 30% change from baseline or threshold points of 22 or less after treatment are proposed to associate with a satisfactory outcome (187,188).

(43)

2.3.3 Painful symptoms

Patients with LSS can experience low back pain and leg pain (30,105–

107,110,111), and leg pain and low back pain are usually evaluated separately.

Furthermore, overall pain can be assessed. Painful symptoms due to LSS are usually relieved by sitting or resting and exacerbated by standing or walking (30,105–107,110,111). Therefore, painful symptoms can be measured separately while resting or sitting and while walking or standing.

Pain intensity in back-related conditions can be assessed by the VAS and the numeric rating scale-11 (NRS-11), and these measures are recommended because of their easy administration and responsiveness to change (181).

The VAS consists of a scale from 0 mm to 100 mm, where 0 mm means no pain at all and 100 mm indicates the most intense sensation or most unpleasant pain imaginable (189). Similarly, the NRS-11 scale ranges from 0 to 10, where 0 indicates no pain and 10 the worst possible pain in the NRS- TXHVWLRQQDLUH7KH9$6LVFRQVLGHUHGEHQHȴFLDOIRUHYDOXDWLQJSDLQ intensity and changes after intervention (191). Furthermore, the VAS score is easy to use, and it is also a valid and reliable measure to assess patients’

pain intensity and unpleasantness of pain (189). In a two-year follow-up, a pain level of 3 or less was suggested as a satisfactory result after surgery for GHJHQHUDWLYHGLVRUGHUV$SSUR[LPDWHO\WZRVFRUHVGL΍HUHQFHRUD reduction from baseline in the NRS-11 scale is proposed to imply clinically important improvement (187,193). A small association has also been found between the change in pain outcome and the change in health-related quality of life after lumbar spinal surgery (174).

2.3.4 Walking ability

Walking ability can be evaluated by a self-reported questionnaire, treadmill test, or a self-paced walking test (SPWT) (194). An SPWT is thought to correlate more accurately to walking capacity (194). Self-reported walking measures in patients with LSS have also been shown to be related mostly to walking capacity in meters but not walking performance in daily life (195). Furthermore, a better postoperative walking capacity was shown to have an association with satisfaction with the surgical outcome in LSS (196). Although neurogenic claudication resulting in constrictions on walking ability is a classic symptom

(44)

RI/66WKHUHLVRQO\YHU\ORZWRORZTXDOLW\HYLGHQFHRIH΍HFWVRIERWKQRQ operative and operative treatments on walking ability. Based on a review E\$PPHQGROLDHWDOGLUHFWGHFRPSUHVVLRQKDGVLPLODUH΍HFWVRQZDONLQJ ability as conservative treatment, and there is only low-quality evidence that interspinous process device (IPD) surgery may improve walking ability better than non-operative treatment (197).

2.4 Surgical outcome in LSS

2SHUDWLYHWUHDWPHQWKDVEHHQVKRZQWREHH΍HFWLYHLQLPSURYLQJV\PSWRPV due to LSS both in short-term and long-term follow-ups. Surgery has been shown to provide better outcomes for at least four to six years, with the EHQHȴWVGHFUHDVLQJRYHUWLPHZKHQFRPSDUHGZLWKFRQVHUYDWLYHWUHDWPHQW Ȃ 2SHUDWLYH WUHDWPHQW KDV DOVR EHHQ GHVFULEHG DV DQ H΍HFWLYH treatment option for patients over 80 years (198,199). Nevertheless, a recent UHYLHZFRQFOXGHGWKDWLWLVXQFOHDULIRSHUDWLYHWUHDWPHQWLVPRUHH΍HFWLYH than non-operative treatment in the long term (153).

When outcomes were compared to conservative treatment, operative treatment provided greater improvement in physical function and painful symptoms in a four-year follow-up (200,201). Furthermore, a study by Held et al. showed a higher quality of life at the one-year follow-up in patients who underwent surgery compared to patients who received conservative treatment (201). In addition to improved leg pain and disability, decompression surgery has also been shown to improve associated low back pain (202–204).

$IWHUDIRXU\HDUIROORZXSQDUURZLQJRIWKHEHQHȴWVRIVXUJLFDORXWFRPHVLQ reference to pain and disability has been reported up to a 10-year follow-up when compared to conservative treatment of LSS (14–19).

ΖQIROORZXSVWXGLHVODVWLQJPRUHWKDQWZR\HDUVVLJQLȴFDQWLPSURYHPHQWV of symptoms after surgery up to ten years have been described (205–208).

A meta-analysis concluded that patients with LSS experienced the most VXEVWDQWLDO V\PSWRP UHOLHI LQ WKH ȴUVW WKUHH PRQWKV SRVWRSHUDWLYHO\

)XUWKHUPRUHVRPHLPSURYHPHQWLQV\PSWRPVRFFXUUHGIRUXSWRȴYH\HDUV (206). A study by Fekete et al. showed that achieved surgical outcome results

(45)

DIWHURQH\HDUZHUHVWDEOHXSWRDȴYH\HDUIROORZXS0DQQLRQHWDO found that disability, leg pain, and low back pain improved two months after VXUJHU\DQGWKHRXWFRPHUHVXOWVUHPDLQHGURXJKO\VWDEOHXSWRWKHȴYH\HDU follow-up point (208). On the other hand, a worsening trend in leg pain and low-back-pain intensity has also been reported (205,206).

In a two-year follow-up, approximately 47% of patients with LSS were very RUVRPHZKDWVDWLVȴHGZLWKWKHV\PSWRPVWDWHDIWHUVXUJHU\ΖQ6ZHGLVK VWXG\RIWKHSDWLHQWVZHUHVDWLVȴHGZLWKWKHVXUJLFDOUHVXOWVLQWKH two-year follow-up (35). In a Finnish study, 64% and 61% patients assessed their satisfaction with the surgical outcome as excellent or good at the three- and 24-month follow-ups, respectively (209). In a study by Yamashita et al., RIWKHVWXG\SDWLHQWVUHSRUWHGWREHYHU\VDWLVȴHGZLWKWKHUHVXOWVRI surgery during the average three-year follow-up (196). Furthermore, in their ȴYH\HDUIROORZXSVWXG\0DQQLRQHWDOUHSRUWHGWKDWRIWKHSDWLHQWV achieved a good surgical outcome in the global outcome measure (208). Atlas et al. showed a 54% satisfaction rate at eight to 10 years after the surgery (16).

On the contrary, a study by Hebert et al. demonstrated that 30%–40% of the SDWLHQWVJDLQHGRQO\OLWWOHRUQREHQHȴWIROORZLQJGHFRPSUHVVLYHVXUJHU\LQ a two-year follow-up (210). Furthermore, a study by Nerland et al. concluded that there is a 9% risk of worsening pain and disability one year after surgery among patients with LSS who had undergone operative treatment (23).

Reoperation rates vary between 7%–10% in the two-year follow-up ȂXSWRWKHȴYH\HDUIROORZXSDQGȂ 25% in follow-ups after more than seven years (16,208,214,215). Surgical outcomes among patients with reoperation have been reported to be worse than patients with only one surgical intervention (213,214).

2.5 Preoperative predictors of surgical outcome in LSS

2.5.1 Patient-related variables

One preoperative predictor can have an association with one or more surgical outcomes (i.e., disability, painful symptoms, or satisfaction with the RXWFRPH7KHH΍HFWRIDSUHRSHUDWLYHSUHGLFWRURQWKHVXUJLFDORXWFRPHFDQ

(46)

be positive, negative, or neutral (22). Furthermore, preoperative predictors can be assessed as binary or continuous variables (209). Previous studies KDYH GHWHUPLQHG SRVVLEOH SDWLHQWUHODWHG IDFWRUV D΍HFWLQJ WKH VXUJLFDO outcome of LSS in reference to physical function and painful symptoms, as GHVFULEHGLQPRUHGHWDLOLQ7DEOH)ROORZXSSHULRGVYDU\XSWRȴYH\HDUV and there is a lack of studies determining preoperative predictors of better surgical outcomes up to the 10-year follow-ups.

A study by Sanden et al. showed that smokers had worse walking abilities, lower satisfaction rates with surgical outcomes, and lower quality of life at the WZR\HDUIROORZXS+RZHYHUVPRNHUVVWLOODFKLHYHGVLJQLȴFDQWSRVWRSHUDWLYH improvement in disability and painful symptoms after two years (27). Pearson HWDOFRQFOXGHGWKDWVPRNHUVJDLQHGVLPLODUEHQHȴWVIURPRSHUDWLYHDQG conservative treatments at the four-year follow-up (28).

Various results for associations of obesity with the surgical outcome in patients with LSS have been reported. A study by McGuire et al. described similar outcomes in highly obese (BMI ൒NJPDQGQRQREHVHSDWLHQWV after operative treatment of LSS (216). Similarly, subgroup analyses from 63257FRKRUWVWXGLHVGHPRQVWUDWHGWKDWKLJKHU%0ΖGLGQRWLQȵXHQFHVXUJLFDO outcome in a four-year follow-up period (216,217). Knutsson et al. reported inferior surgical outcome results for obese patients with LSS, and obesity was related to a higher degree of dissatisfaction. However, they still concluded WKDWREHVHSDWLHQWVJDLQHGVLJQLȴFDQWLPSURYHPHQWLQV\PSWRPVDWWKHWZR year follow-up (35). Athiviraham et al. showed a negative association of higher BMI with the surgical outcomes in patients with LLS (32).

Many studies have described a negative association of longer duration of symptoms with the surgical outcome (25,31,33,34). On the other hand, few studies have shown a neutral association between preoperative prolonged symptom duration and the surgical outcome in patients with LSS (32,218).

Katz et al. showed that better self-rated health preoperatively had a strong positive impact on improved postoperative symptom severity, walking- capacity scales, and satisfaction with the outcome (21).

(47)

Table 1. Literature review of possible preoperative patient-related predic- WRUVWKDWKDYHEHHQVKRZQWRKDYHDQH΍HFWRQSRVWRSHUDWLYHDIXQFWLRQDO status and (b) painful symptoms in patients with LSS. Psychological factors are excluded.

1a

Better functional ability Worse functional ability

• less comorbidity (22)

• more severe central stenosis (22)

• higher expectations of physical function (22)

• duration of symptoms less than 33 months (33)

• non-smoking (26–28)

• better self-rated health (26)

• painkiller use for less than 12 months

• duration of symptoms less than 12 (26) months (31)

• greater disability (24,32)

• predominant leg pain (28,30)

• baseline ODI score less than 56 (28)

• neuroforaminal stenosis (28)

• not lifting at work (28)

• EDVHOLQHQHXURORJLFDOGHȴFLW

• higher pain (24)

• higher BMI (32,35)

• duration of leg pain more than two years (34)

• duration of symptoms more than one year (25)

• worse baseline function (23,29,34)

• younger age (23)

• smoking (23,28)

• American Society of Anesthesiologist (ASA) grade of 3 or greater (23)

• prior surgery to the lumbar spine (23,24)

• higher pain sensitivity (29)

• low back pain (24)

• analgesic use (25)

1b

Less painful symptoms More painful symptoms

• shorter duration of symptoms in lateral stenosis (22)

• more severe central stenosis (22)

• duration of symptoms less than 33 months (33)

• non-smoking (27)

• predominant leg pain (30)

• GLVRUGHULQȵXHQFLQJZDONLQJDELOLW\

• scoliosis (22)(22)

• higher expectations of pain relief (22)

• intermittent or regular use of analgesic (34)

• duration of leg pain more than two years (34)

• shorter walking distance (34)

• low health-related quality of life (34)

• higher BMI (35)

• longer duration of preoperative symptoms (25)

(48)

2.6 Psychological factors in LSS surgery

Mental disorders and subjective mental health are related to common musculoskeletal disorders (219). One study estimated that psychological variables may explain 20% of the variety in disability scores in patients who XQGHUZHQWVSLQDOVXUJHU\)XUWKHUPRUHVXUJLFDORXWFRPHLVD΍HFWHG by multimodal variables, but psychological factors such as presurgical somatization, depression, anxiety, and poor coping have been shown to be associated with poorer surgical outcome in spinal surgery (221).

2.6.1 Depression

The pathophysiology of depression includes complex multidimensional pathways, some of which are considered to be immunological pathways, such as disturbances in the immune system (222,223). It has been estimated that the prevalence of depression is approximately 20% during the average lifespan (224). Prevalence of depression is 2.6% among the Finnish population over 65 years of age (225). Furthermore, depression was assessed to be in the WRSȴYHFDXVHVRIGLVDELOLW\DPRQJSDWLHQWVDJHGEHWZHHQDQG\HDUV in 2015 (226).

Approximately 20%–40% of patients with LSS undergoing operative treatment have been described to have depressive symptoms (227–229).

Furthermore, it has been reported that 17%–18% of patients with LSS have depressive symptoms postoperatively (228,230).

0DQ\ VWXGLHV KDYH GHPRQVWUDWHG D QHJDWLYH H΍HFW RI SUHRSHUDWLYH depression on surgical outcome in reference to disability among patients ZLWK/66XSWRDȴYH\HDUIROORZXSȂ+RZHYHUWKHH΍HFWRI preoperative depression on postoperative pain is less clear (43).

Patients without depression reported a greater improvement in disability after surgery than patients with continuous depression, and the same trend was VHHQLQIROORZXSSHULRGVODVWLQJIURPRQHWRȴYH\HDUV+RZHYHU SDWLHQWVZLWKSUHRSHUDWLYHGHSUHVVLRQFDQJDLQVLJQLȴFDQWLPSURYHPHQWVLQ pain and disability levels postoperatively (40), and patients’ preoperative depressive symptoms may improve after operative treatment (37,228,231).

In addition, patients who recovered from depression postoperatively can

Preoperative predictors and surgical outcome in lumbar spinal stenosis : A prospective 10-year follow-up study

IINA TUOMAINEN

Preoperative predictors and surgical outcome in lumbar spinal stenosis

Dissertations in Health Sciences

Preoperative predictors and surgical outcome in lumbar spinal stenosis

Preoperative predictors and surgical outcome in lumbar spinal stenosis

ACKNOWLEDGEMENTS

LIST OF ORIGINAL PUBLICATIONS

TABLE OF CONTENTS

1 INTRODUCTION

2 REVIEW OF THE LITERATURE