Crit Rev Neurosurg (1998) 8: 333–337© Springer-Verlag 1998

Mario BrockJosef Ramsbacher

Lumbar spinal stenosis

M. Brock (½) · J. RamsbacherNeurochirurgische Klinik,Universitätsklinik Benjamin Franklin,Hindenburgdamm 30,D-12200 Berlin, Germany

Abstract Surgical treatment of spi-nal canal stenosis, a typical clinicalpicture of the elderly, has been thesubject of frequent publicationssince its description in 1954. Themostly retrospective studies havedealt with the postoperative outcomeand surgical technique on the onehand, and with the radiological find-ings on the other. Only few studiesexamine the development of spinalcanal stenosis and the resulting neu-rophysiological changes. The arti-cles selected for this review are con-cerned with the development of spi-nal canal stenosis in an animal modelas well as with improved ways ofassessing the pre- and postoperativeconditions of the patients. Iwamotoet al. [1] study the neurophysiologi-cal changes caused by spinal canalstenosis in an animal model. In thefirst part of a prospective study,Jönsson et al. [2] analyze the rele-

vance of various parameters for thetiming of surgery. The second part[3] of this study deals with parame-ters such as age and additional dis-eases, which exert a major influenceon patient outcome. Deen et al. [4]present a new and improved way ofassessing both the pre- and postop-erative clinical picture of this entity.For this purpose, they adjusted theroutine cardiological treadmill testto suit the requirements of patientswith spinal canal stenosis. The studyby Thomas et al. [5] deals with thedifferences in outcome of laminec-tomy and laminotomy, and withpostoperative instabilities. Finally,the study by Airaksinen et al. [6]investigates postoperative outcomein a large group of patients.

Key words Lumbar spinal stenosis · Treatment · Pathophysiology

Introduction

Spinal stenosis is increasingly rec-ognized as a cause of low back painin older patients, and it has becomethe most common indication forlumbar spine surgery in the elderly.Since its description in 1954 themostly retrospective studies havedealt with the postoperative out-

come, surgical technique and radio-logical findings. Only a few studieshave examined the development ofspinal canal stenosis and the result-ing neurophysiological changes.The articles selected for this reviewconcern the neurophysiologicalchanges caused by spinal canal sten-

osis in an animal model. The rele-vance of various parameters for thetiming of surgery is analysed,parameters such as age and addition-al diseases are dealt with, and a newand improved way of assessing boththe pre- and postoperative picture ofthis entity is presented.

334

[1] Lumbar spinal canal stenosis examined electrophysiologically in a rat model of chronic cauda equina compressionSpine (1997) 22:2636–2640

Information. In this experimental study, a metal coil wasimplanted at L5 in eight 3-week-old rats. Both theincreasing and decreasing action potentials were meas-ured 1 year later. The electrophysiological changes wereobserved after high-frequency stimulation. The waveshape of the increasing action potentials had three peaksat the cauda equina and two peaks at the medullary cone,followed by a broad wave. The wave shape of the decreas-ing nerve action potentials had two peaks. The mean val-ues of the decreasing and increasing conduction veloc-ities (CV) of the treated rats were slower than those inthe control group. After high-frequency stimulation at thetail or L4–5, amplitude reached a normal value after 30 s,the wave shape remaining unchanged. In the rats withspinal canal stenosis, the CV likewise normalized butwith a 10-min delay. The amplitude had not yet reacheda normal value, even after 10 min. From the differencein amplitude and CV before and after high-frequencystimulation, the authors concluded that the changescaused by impaired perfusion in the area of the nervesand cauda equina are accompanied by histological chang-es. These changes could be responsible for spinal claud-ication in patients with spinal canal stenosis.

Analysis. This study successfully simulates spinal canalstenosis in an animal model and describes a comprehen-sive electrophysiological examination that allows com-parison with a control group. The results are of highinformation value and easy to interpret. Unfortunately,the authors did not perform surgical repair of the experi-mental spinal canal stenosis and record the correspond-ing electrophysiological changes.

[2] A prospective and consecutive study of surgically treated lumbar spinal stenosis. I. Clinical features related to radiographic findingsSpine (1997) 22:2932–2937

Information. This prospective study comprising 105patients seeks a correlation between age, complaints,and radiological findings in lumbar spinal stenosis. Allpatients were interviewed and fully examined on the daybefore surgery and their X-rays assessed.

Of the 105 patients, 48 had pain at rest, which wasmilder in younger patients. A reduction in walkingrange to less than 500 m was reported by 70 patients.Babinski’s sign was absent in 70 cases. There was nocorrelation between the duration of complaints or theintensity of pain and the neurological deficits. Neuro-radiological studies revealed a complete contrast stopin 13 patients whose spinal canal had a mean diameterof 6.8 mm. The level most frequently affected wasL4/L5 in patients under 70 years of age and L3/L4 inthose over 70. Additional degenerative spondylolisthe-sis was found in 32 patients.

Analysis. In this study, the authors attempted to findcorrelations between radiological findings, neurologi-cal deficits, age, and duration of symptoms. It wasshown that both pain and neurological deficits dependon age, but not on radiological findings. The main rea-son for patients to undergo surgery is the marked reduc-tion in walking range and the resulting handicap.

This is the first part of an extensive study, the sec-ond part of which deals with the postoperative results.

[3] A prospective and consecutive study of surgically treated lumbar spinal stenosis. II. Five-year follow-up by an independent observerSpine (1997) 22:2938–2944

Information. This is the second part of a prospectivestudy in 105 patients who underwent surgery for spinalcanal stenosis. Laminectomy was performed in allcases, fusion in none. The follow-up examinations werecarried out 4 months, 1, 2, and 5 years later. During thefollow-up period, 19 patients had to have reoperations,which involved fusion in 4 cases, renewed decompres-sion in 13, and repairs in response to surgical compli-cations in 2. The result was rated as excellent by 43%of the patients after 4 months and by 67% after 2 years.However, after the 5-year follow-up, only 52% stillrated the result as excellent. Outcome depended on thedegree of preoperative stenosis as well as on the pre-operative duration of symptoms.

Analysis. This second part of an extensive study dem-onstrates that outcome can be predicted by a good pre-operative clinical examination and anamnesis. Thisallows adequate patient selection. The “ideal” patienthas only mild stenosis and no continuous back pain or

335

associated diseases that shorten the walking range. Oneof the most important prognostic factors, however,appears to be the duration of preoperative symptoms.Thus, pain and neurological deficits of less than 4-yearduration appear to be associated with favorable out-come. This point seems particularly important, since itallows a better timing of surgery.

[4] Use of the exercise treadmill to measure baseline functional status and surgical outcome in patients with severe lumbar spinal stenosisSpine (1998) 23:244–248

Information. This prospective study employed the rou-tine cardiological “exercise program” to evaluate com-plaints before and after surgical therapy of spinal canalstenosis. Fifty patients (34 men and 16 women) with amean age of 72 years were included. All had imagingof spinal canal stenosis and clinical evidence of spinalclaudication. Surgical treatment consisted in laminec-tomy with removal of an average of 2.52 laminae. Thepatients were tested with an exercise treadmill beforeand 3 months after surgery. Two tests were performedin each case, the first at a speed of 1.2 miles/h and thesecond at the patient’s normal walking speed. If therewere no complaints, the tests were discontinued after15 min. The results were evaluated by Wilcoxon’ssigned rank test. Five patients were excluded becauseof their poor general condition. In the 1.2 mile/h test,the first symptoms were observed after an average of1.82 min prior to surgery and after an average of11.93 min following surgery. On average, the test hadto be discontinued after 6.91 min prior to and after13.26 min following surgery. At the patient’s preferredspeed, the average preoperative times were 1.7 minuntil first symptoms and 5.84 min until discontinuation.The average postoperative times were 11.9 min and12.66 min, respectively. All patients improved signifi-cantly following surgery, both in the 1.2-mile test andin the test of individual walking speed.

A 4-grade system was established based upon thedata obtained: grade 1 comprises patients who couldwalk for 15 min without symptoms; patients in grade 2could walk for 15 min without neurological deficits;patients in grade 3 discontinued the test after 5–15 min;and patients in grade 4 discontinued the test before5 min.

Following the report of these results, the describedtests have become routine at the Mayo Clinic in Arizo-na.

Analysis. The described tests, which use the routine cardiological exercise treadmill, were modified to allow assessment of postoperative complaints inpatients with spinal canal stenosis. The analyzed param-eters are well adapted to the patient with lumbar spinalstenosis and can be assessed easily. The tests provideadequate information on the pre- and postoperativecomplaints. The data obtained are objective and quan-tifiable.

[5] Quantitative outcome and radiographic comparisons between laminectomy and laminotomyin the treatment of acquired lumbar stenosisNeurosurgery (1997) 41:567–575

Information. This retrospective study compares theoutcome after laminotomy and laminectomy in 26patients with lumbar spinal stenosis. A detailed follow-up study was performed for this purpose. Each patientwas submitted to functional x-rays of the lumbar spine.There was no significant difference between the resultsof laminotomy and laminectomy. There was no increasein postoperative spondylolisthesis in either group. How-ever, outcome depended essentially on the patient’s general condition and on the presence of associated dis-eases.

Analysis. Despite the small group of patients, this studyattempts to determine whether the surgical techniquealone is responsible for the outcome in cases of spinalcanal stenosis. It was demonstrated that, even when asufficient decompression of the spinal canal is achieved,there is no significant difference in outcome betweenlaminotomy and laminectomy. It also appears notewor-thy that the results also do not differ with regard to thepresence of postoperative spondylolisthesis. The argu-ments in favor of performing a laminotomy would thusno longer be valid.

This study must be viewed with a certain degree ofcaution, however, since all data were obtained retro-spectively and the patient population is very small. Fur-ther studies should be performed with a longer follow-up and a larger number of patients.

336

[6] Surgical outcome of 438 patients treated surgically for lumbar spinal stenosisSpine (1997) 22:2278–2282

Information. This retrospective study is based on thefollow-up of 438 patients (183 women, mean age 54years; 255 men, mean age 53 years) who underwent sur-gery for spinal canal stenosis between 1974 and 1987.Of the total number of patients, 33 had already died and26 could not be traced. The Oswestry low-back disabil-ity questionnaire was used for evaluation. This ques-tionnaire includes ten categories such as pain intensity,social life, sexual life, travelling, standing, sitting,walking, rising, and personal hygiene. The results weregraded as excellent or good for patients with 0–40 pointsand poor or very poor for those with 41–100 points.

The main preoperative symptoms were pain in thelower extremities (58%), spinal claudication (23%),mixed symptoms (18%), and back pain (1%). The meanpreoperative duration of pain was 7 years. Thirty-twopatients had not undergone conservative treatment priorto surgery; 112 had received nonsteroidal anti-inflam-matory and analgesic drugs, and 222 had been submit-ted to physical therapy. Another 64 patients had alreadyhad 3 weeks of conservative treatment. Diagnosis was

confirmed by computed tomography or myelography.Surgery consisted in a laminectomy, widened laterallyfor nerve root decompression. On average 1.5 laminaewere removed. In 9% of the patients, only an extendedinterlaminar fenestration was performed. Peri- or post-operative complications occurred in 11% of the cases.Outcome was excellent or good in 42% and poor or verypoor in 58%. At the time of follow-up, 25% required noanalgesics, and 39% could walk for more than 1 mile,while 5% needed a walking stick, and 1% were bedrid-den.

Patients with additional diseases such as diabetesmellitus and hip-joint arthrosis had a significantly high-er disability score. The average score was 28 in patientswith a complete and 30 in patients with subtotal con-trast stop; it was 35 in the presence of a spinal canalstenosis narrower than 10 mm, 36 for one 10–12 mm,and 37 for a canal wider than 12 mm.

Analysis. This retrospective study presents the follow-up of a large patient population and a detailed reviewof the literature. It demonstrates the suitability of theOswestry disability score for evaluating postoperativecomplaints. However, the study does not provide muchnew information.

Synthesis

The development of spinal canalstenosis has not yet been fully eluci-dated. Although predisposing fac-tors are known, the process ultimate-ly leading to stenosis still remains tobe clarified. Moreover, there is as yetno precise definition of the term spi-nal canal stenosis. The literaturecontinues to define spinal canal sten-osis on the basis of diameter of thecanal. In addition, the ideal time forsurgery is still uncertain, and theform of treatment remains as contro-versial as the development and def-inition of this disease. Favored pro-cedures include laminotomy, lami-

nectomy, and extended interlaminarfenestration. There are numerousstudies on the outcome of lumbarspinal stenosis, but no valid compar-ison between pre- and postoperativecomplaints. The reviewed studiestouched upon some of the unsolvedproblems. The experimental study[1] reviewed, together with evidencefrom the literature, substantiates thehypothesis that spinal canal stenosiscauses complaints through reducedperfusion. Major progress withregard to surgical indication seemsto have been made in the study byJönsson et al. [2], which makes it

possible to pinpoint an optimal timefor surgery. The study comparinglaminotomy and laminectomy [5]suggests that sufficient spinal canaldecompression is the essential stepof the surgical treatment. It alsoshows that postoperative instability,so frequently discussed, is not influ-enced by either procedure. All clin-ical studies show, however, thatgood long-term results are obtainedby surgery only in 50–60% of cases.There is a need for prospective stud-ies in order to eventually optimizethe treatment of this entity.

337

Papers reviewed1. Iwamoto H, Matasuda H, Noriage A,

Yamano Y (1997) Lumbar spinal canalstenosis examined electrophysiologicallyin a rat model of chronic cauda equinacompression. Spine 22:2636–2640

2. Jönsson B, Annertz M, Sjöberg C,Strömqvist B (1997) A prospective andconsecutive study of surgically treatedlumbar spinal stenosis. I. Clinical featuresrelated to radiographic findings. Spine22: 2932–2937

3. Jönsson B, Annertz M, Sjöberg C,Strömqvist B (1997) A prospective andconsecutive study of surgically treatedlumbar spinal stenosis. II. Five-year fol-low-up by an independent observer. Spine22: 2938–2944

4. Deen HG, Zimmerman RS, Lyons MK,McPhee MC, Verheijde JL, Lemeens SM(1998) Use of the exercise treadmill tomeasure baseline functional status andsurgical outcome in patients with severelumbar spinal stenosis. Spine 23:244–248

5. Thomas NW, Rea GL, Pikul BK, MervisLJ, Irsik R, McGregor JM (1997) Quan-titative outcome and radiographic compa-risons between laminectomy and lamin-otomy in the treatment of acquired lum-bar stenosis. Neurosurgery 41: 567–575

6. Airaksinen O, Herno A, Turunen V, SaariT, Suomlainen O (1997) Surgical out-come of 438 patients treated surgically for lumbar spinal stenosis. Spine 22:2278–2282