14.1 Epidemiology

A diagnosis of symptomatic lumbar spinal stenosis (LSS) is made if the patient has radiographic evidence of lumbar stenosis and symptoms of neurogenic claudication or lumbar radiculopathy. Neurogenic claudication presents with pain and dysfunction that starts proximally in the back and buttocks and then progresses distally into the limbs. In some cases, in a dermatomal distribution. Symptoms of neurogenic claudication are of moderate sensitivity but are highly specific to LSS. Patients usually report numbness, weakness, or discomfort radiating from the spine to the buttocks and legs while standing or walking for prolonged distances. Although radiating back and leg pain is a major characteristic of neurogenic claudication, some patients might present with neurologic deficit such as paresthesia or lower extremity weakness. Neurogenic claudication is characterized by an insidious onset and gradual progressive course over months to years. Lumbar stenosis may also present with radiculopathy that may be unilateral or bilateral, resulting in mechanical, motor, reflex, and sensory signs related to the involved levels. The symptoms of neurogenic claudication are typically exaggerated, with extension of the lumbar spine while walking or standing because spine lordosis further narrows the spinal canal. Conversely, symptoms are typically relieved by sitting or flexing the lumbar spine. Noticeable relief at rest may be reported with activities that involve a bending position (e.g., leaning on a shopping cart, lumbar flexion while riding a bicycle). This relief tends to decrease as the condition progresses. Bladder dysfunction in the form of urinary incontinence is commonly reported in the geriatric population, with noticeable improvement after decompression surgeries. ²

Although LSS is a common diagnosis in the elderly, there is a paucity of literature discussing its incidence and prevalence in different populations. In one study, LSS reached five cases per 100,000 individuals, four times more than cervical spine stenosis. ³ In another study, it was found that 14% of patients complaining of low back pain and seeking specialist care have spinal stenosis. ¹ In one study, when evaluated by imaging techniques, there was a degree of spinal stenosis in 80% of the patients above 70 years old. ⁴ In another study, the prevalence of symptomatic LSS and its association with physical performance using MRI in a population-based cohort was 9.3% overall (10.1% in men and 8.9% in women) In addition, a 6-minute walking time at a maximal pace compared to normal pace walking was a more sensitive test to recognize symptomatic LSS. ⁵

The age of onset of symptoms of lumbar stenosis crudely indicates the etiology of the stenosis, as patients with congenital stenosis tend to become symptomatic at younger ages, in the third to fifth decades of life. More commonly, LSS often becomes symptomatic in the mid-sixties and seventies. ⁶ In one study, absolute lumbar stenosis, defined as a spinal canal diameter of less than 10 mm, has a prevalence as high as 19.4% in the 60–69 year age group. ⁷

The main contraindications to surgical decompression of LSS are misdiagnosis and the inability of the patient to benefit from surgical intervention. Symptomatic LSS rarely occurs in isolation. The same degenerative and pathologic processes that cause spinal stenosis may also cause other masquerading pathology. Therefore, great care must be taken to identify other diagnoses with history, clinical examination and ancillary testing that would limit or negate the benefit of surgical intervention. The most common are: vascular claudication, hip and knee osteoarthritis, and medical comorbidities.

The most common diagnostic mimic of neurogenic claudication is vascular claudication. Vascular claudication typically presents in the distal lower extremities and progresses proximally with activity. Typically, the patient will achieve relief at rest standing still. The patient typically will not describe relief by bending forward or leaning on a shopping cart. Other clinical features that distinguish vascular from neurogenic claudication are skin pallor, decreased skin temperature, loss of hair on the legs, and diminished or absent peripheral pulses.

14.1.2 Magnetic Resonance Imaging

MRI is considered the best tool to assess for lumbar stenosis. With MRI one is able to visualize soft tissue abnormalities such as ligamentum flavum hypertrophy, disc herniations or protrusions and facet joint compromise. It also can help identify nerve roots and intrinsic cord abnormalities, edema, and demyelination. LSS is most easily seen on T2-weighted images and presents as effacement of CSF signal on T2-weighted sequences. MRI also enables surgeons to accurately localize lesions via simultaneous sagittal, para-sagittal and axial views. (Fig. 14‑1) The use of MRI is contraindicated in patients with many types of implants, including cardiac pacemakers, deep brain stimulators, and spinal cord and peripheral nerve stimulators. The presence of these implants is relatively more common in the geriatric population. The presence of these devices should be ascertained and MR-compatibility confirmed prior to ordering the imaging study. It is important to associate the patients’ presenting symptoms with the MRI findings. In one study, only one-third of patients older than 60 years with disc herniation (36%) and spinal stenosis (21%) were symptomatic. ⁸

14.2 Biomechanical Considerations

LSS can be attributed to several anatomical findings. As a consequence of the degeneration of disks, ventral compression can be caused by centrally located herniated discs or disc bulges. The degenerative process leads to loss of disk height that forces infoldings of the ligamentum flavum and formation of spurs, osteophytes or, in rare occasions, a consequent hypertrophic posterior longitudinal ligament. The subsequent mechanical compression and vascular insufficiency in the lumbar nerve roots and cauda equina contribute to the set of symptoms known as neurogenic claudication. The symptoms of neurogenic claudication are typically exaggerated with extension of the lumbar spine such as with walking or standing due to increased spine lordosis resulting in further narrowing of the spinal canal. Conversely, symptoms are typically relieved by sitting or flexing the lumbar spine.

14.3 Treatment Options

Treatment of symptomatic LSS employs an array of both surgical and conservative options, including physical therapy, membrane stabilizers, analgesics and pain interventions. Commonly, many patients with symptoms of variable severities that are not associated with neurologic deficits are initially treated conservatively. Special consideration should be given to medical treatment in the elderly population, not only to primarily treat the symptoms of LSS, but also to optimize the medical perioperative care for associated comorbidities—cardiac, pulmonary, renal, and endocrine—to achieve the best outcomes. ⁹ The spine surgeon should give special attention to patient’s bone health, given the high prevalence of bone disease in the elderly population. ⁹

Understanding the natural history of LSS is a key to managing patients due to the variable severities of presenting symptoms. Although LSS in the elderly is primarily due to degenerative changes in the spine, this fact does not anticipate an inevitable worsening of the symptoms. In fact, according to the North American Spine Society (NASS) guidelines, the natural history of one-third to one-half of mild-to-moderate LSS patients was favorable. ¹⁰ In addition, other reviews reported improvement in one-third of the cases as well as no change of severity during 8 years of follow-up, making the rapid progression of pain or disabilities in such patients seem to be rare. ¹¹ Similarly, another study reported no dramatic progression in LSS symptoms in 60% of patients, despite MRI-confirmed progressive stenosis of the spinal canal. ¹²

14.3.1 Surgical Management

The natural history of LSS shows that the course of symptom progression is gradual and occurs over years. It is unusual to encounter a patient who has a rapid deterioration in symptoms, and it may indicate an association with other pathology, such as tumors, compression fractures, or disc herniaitons. Therefore, surgery is almost always elective and planned after discussing the severity of symptoms and the expected outcomes of surgery with the patient. ¹³ Although there is no high-quality evidence of its effects, a trial of nonoperative treatment options is usually performed prior to considering surgical intervention. ^{1 , 14} Several studies have compared surgery to other nonoperative management. The Spine Patient Outcomes Research Trial (SPORT) demonstrated an early significant benefit of surgery over nonoperative treatment. However, this advantage decreased over time to provide no difference in 6-year and 8-year follow-up. ¹⁵ The conclusions drawn were considered low quality because of the high rate of crossover from the conservative treatment group to surgery.

The goal of surgery is to decompress the nerve roots that are trapped within the stenotic spinal canal, which gives a chance for pain relief and functional recovery in cases with neural deficits. The decompression surgeries have many surgical approaches and techniques that vary depending on the level, extent, and severity of the stenosis as well as of the associated conditions. It also depends on the surgeon’s preference and the patient’s history with spine surgery.

These techniques involve unilateral and bilateral laminectomy and bilateral decompression through unilateral laminectomy, in addition to different methods of laminoplasty and minimally invasive surgery. However, current evidence does not prefer any of the surgical approaches. ¹⁶ A systematic review in 2015 compared the effects of three of the new approaches—unilateral laminotomy for bilateral decompression, bilateral laminotomy, and split-spinous process laminotomy—to conventional laminectomy. The results showed similarity in outcomes of functional disability, perceived recovery, and leg pain after all approaches, although the evidence was low to very low quality. It highlighted the poor methodology of the currently available studies of these new approaches and the lack of long-term outcomes data. ¹⁷ In another as-treated analysis of SPORT, predominant leg pain patients improved significantly more with surgery than predominant LBP patients, who still improved significantly more with surgery than with nonoperative treatment. ¹⁸

Spinal fusion in LSS patient with decompression surgery has been a major controversy in the surgical management of LSS. It has been recommended for patients with associated degenerative spondylolisthesis, recurrent stenosis despite decompression, instability or deformity. ¹³ (Fig. 14‑2) According to one retrospective cohort, the rate of complex lumbar fusion surgeries in the elderly has increased 15-fold (from 1.3 to 19.9 per 100,000 beneficiaries) compared to decompression alone or simple fusion surgeries. This increase was associated with an increase in the rate of major complications, 30-day mortality, costs, and resource use compared to decompression alone. ¹⁹ In another retrospective cohort, posterior lumbar interbody fusion (PLIF) was compared to decompressive laminectomy and flavectomy without fusion (DLF) in elderly patients (>75 years old). It showed a statistically significant decrease in LBP and lower recurrence after PLIF versus DLF. Moreover, the study concluded that lumbar surgery is safe and justifiable even in elderly patients > 75 years old and recommended lumbar fusion surgery in patients with predominant back pain. ²⁰ A clinical practice guideline update on 2013 recommended that “in the absence of deformity or instability, lumbar fusion has not been shown to improve outcomes in patients with isolated stenosis, and is not recommended” ¹⁰ . According to NASS guidelines, in the absence of instability or deformity, decompression alone is recommended for patients with prominent leg symptoms. ²¹

Use of intraspinous spacers as an alternative to lumbar decompression has been advocated in the literature. Patients whose symptoms improve with forward flexion and have single-level stenosis are considered candidates for this procedure. Placement of the spacer between the spinous processes of the stenotic level forces the spine into focal kyphosis, thereby enlarging the diameter of the canal. While initial studies showed improvement in claudicant symptoms, symptom relief may not be as durable as conventional decompressive surgery, with high failure rates in medium- to long-term follow up. ²²

14.4 Benefit and Risks

When considering surgical intervention for symptomatic lumbar stenosis, a critical appraisal of the risks and benefits of surgery is mandatory. The benefits of surgical intervention in the appropriately selected patient, as described above, include improved posture, improved ambulatory distance, and reduced pain and discomfort. The risks of surgery can be divided into two general catagories: risks related to having surgery and risks related to the specific operation. Surgical risks include perioperative cardiovascular events, pulmonary events, venous thromboembolism, and, in older adults, cognitive impairment. Risks specific to lumbar decompression include injury to the neural elements, durotomy, postoperative pseudomeningocele, postoperative epidural hematoma, iatrogenic instability, and postoperative wound infection. If a fusion is performed, additional risks include pseudoarthrosis, hardware failure/fracture, and iatrogenic deformity.

14.5 Pitfalls, Complications and Avoidance

While the presentation and basic treatment paradigms for the management of symptomatic lumbar stenosis are not age specific, the increased incidence of medical comorbidities, challenges with mobility, presence of frailty, cognitive decline, and complex social situations must be assessed prior to embarking on a treatment plan. In general, medical management has a lower risk profile than surgical intervention and should be considered first. However, in patients whom symptoms do not improve, surgery should be considered.

Frailty and advanced patient age correlate with increased risk of postoperative morbidity and poor outcome. They have also been associated with increased cost of care in the immediate postoperative period. ^{23 , 24} Despite the risk of comorbidity, in at least one study the most frail patients achieved the biggest improvement in self-reported outcome measures (ODI). ²⁴ Therefore, if the risks of surgery can be effectively mitigated, surgery in even the frail elderly could be considered. ²⁵ Strategies for risk reduction include rehabilitation, nutritional optimization, and comanagement perioperatively with a geriatrician. ²⁶

Prior to surgical intervention, social factors must be assessed to optimize recovery and reduce the risk of readmission. Patients with poor mobility, those who live alone or have declining cognitive function, may require placement in a skilled nursing facility postoperatively. This possibility should be discussed with the patient and his/her family prior to surgical intervention. End-of-life issues, such as durable power of attorney for health care and do-not-resuscitate orders should be addressed.

When fusion is being considered, the patient’s bone quality needs to be assessed. Relative osteopenia/osteoporosis rates are high in the elderly. Achieving good fixation can be challenging in these cases, as can be seen by the higher rates of hardware failure and proximal junctional kyphosis in the frail and elderly. ²⁴ Bone health should be assessed and optimized prior to surgical intervention to minimize this risk whenever possible. Strategies to improve fixation and outcomes of fusion surgery include cement augmentation, partial correction of deformity, and focal fusion without deformity correction in select cases. (Fig. 14‑3)

14.6 Outcomes/Evidence

There is a lack of consensus on either conservative or surgical treatments. This can be explained by the paucity of high-quality randomized trials investigating the efficacy of the various conservative treatment options that are being utilized. ¹ A systematic review of LSS confirmed with MRI and presenting with neurogenic claudication concluded that there is no significant evidence supporting any type of conservative treatment. ²⁷ Moreover, there was a lack of clear description of nonsurgical treatment protocols, which makes treatment outcomes analysis difficult. ¹

Most patients with LSS and neurogenic claudication are treated with a course of conservative treatment before surgical intervention. ²⁷ In the elderly population, patients commonly present with other severe comorbidities which may make surgery a risky choice. ^{28 , 29} Such cases are usually treated by multimodality pain centers. ³⁰ If conservative measures fail and surgery is considered, patients must be selected carefully based on the imaging findings and the expected outcomes based on their overall health.

Few randomized controlled trials have studied the conservative (nonoperative) strategies of LSS treatment. A systematic review of nonoperative interventions in imaging-confirmed symptomatic LSS patients failed to conclude a sufficient evidence to recommend any nonoperative strategy. Those strategies included calcitonin, prostaglandins, gabapentin, methylcobalamin, epidural steroid injections, exercise, and multimodal nonoperative treatment. It also noted that the lack of clearly defined descriptions of these protocols is preventing data pooling and analysis for better testing of the outcomes. ²⁷

Pharmacologic treatments of LSS include drugs such as NSAIDs, acetaminophen, gabapentin, prostaglandins, and vitamin B1. However, the only available evidence is of low to very low quality provided by single-center small trials. ^{27 , 31}

Several studies have suggested that NSAIDs do not offer more relief than acetaminophen. ³² Prostaglandins are theorized to enhance blood supply to the nerve roots, but a single-center small trial did not offer more than a low- to very low-quality evidence of symptom relief. ³³

In a retrospective cohort study, quality of life (QOL) of LSS patients with concordant neurogenic claudication was evaluated after treatment with membrane-stabilizing agents (MSAs) (701 patients) vs. conservative treatment without MSAs (2,104 patients). The results suggested better QOL for the patients treated with MSAs compared to the other group. ³⁴ A recent retrospective cohort, by the same institution, sought to create a prediction model for MSAs posttreatment outcomes. It concluded that greatest effects of MSAs will be in those with worse QOL, those with less depression, those who are married, and in patients with high socioeconomic state. ³⁵

Another randomized controlled trial compared a group of patients receiving the standard medical care (therapeutic exercise, lumbosacral closet with steel bracing, and NSAIDS) against a group that received gabapentin in addition to the standard medical treatment. Gabapentin resulted in improvements in walking distance, pain scores, and sensory deficit recovery. ³⁶

In the past, calcitonin was suggested to cause improvement of symptoms, but a recent meta-analysis showed no better effect than acetaminophen or placebo. ³⁷

Opioids have been used for pain control, but their role in the long-term control of symptoms is unclear. ¹ A randomized, double-blinded crossover trial tested the efficacy of oxymorphone and propoxyphene/acetaminophen in LSS with neurogenic claudication. The study failed to demonstrate a benefit from the tested opioids. ³⁸ Corticosteroids are often used in LSS conservative management schemes although there is still no high-quality evidence of their efficacy in LSS. ¹

Physiotherapy is commonly utilized in the nonoperative treatment of LSS. It includes exercise that can be for increasing flexibility, core strengthening, or aerobic conditioning to increase exercise tolerance. Bracing, corsets and lumbar semi-rigid orthosis have been utilized in some cases. Other pain-relieving methods such as applying heat or ice, electric stimulation, massage or ultrasonic waves, as well as spine manipulation and postural instructions, are often included in medical management programs. ¹ In a systematic review of nonoperative treatment options, there was low-quality evidence from a single trial that exercise is of short-term benefit for leg pain and function compared to no treatment. ²⁷ In another systematic review of physical therapy trials in LSS treatment, it could not be concluded which physical therapy treatment was superior for LSS. It also found low-quality evidence suggesting that physical therapy modalities such as ultrasound, Transcutaneous Electrical Nerve Stimulation (TENS), and heat packs have no additional effect to exercise. Moreover, surgery leads to better long-term (2-year) outcomes for pain and disability but not for walking distance. ³⁹ In a secondary analysis of the SPORT study, the relationship between physical therapy and the long-term prognosis for LSS patients was evaluated. It concluded that physical therapy was associated with a reduced likelihood of patients’ receiving surgery within 1 year. ⁴⁰

Epidural steroid injections are commonly used to treat symptoms of LSS, in spite of the variability of the results of the trials. In one meta-analysis that included eight trials involving patients with symptomatic LSS, epidural corticosteroid injections did not have a clear effect on spinal stenosis. ⁴¹ Another meta-analysis in 2015 that included 10 randomized controlled trials demonstrated a minimal evidence that epidural steroid injections have better effects than Lidocaine alone. It also demonstrated a fair amount of short-term and long-term benefits in LSS patients. ⁴² The steroid injections are performed utilizing three main approaches in the lumbar spine—caudal, interlaminar and transforaminal—but the literature discussing the superiority of one approach over another is sporadic. ⁴³

In an analysis of two randomized controlled trials for both caudal and interlaminar injections in the management of pain and disability of central LSS, results at 2 year follow up showed significant improvement with caudal and interlaminar epidural approaches, either with local anesthetic only, or with steroids at 2 years. In addition, the interlaminar approach provided significantly better results than the caudal injection in that management setting. ⁴³ It is commonly believed that epidural steroid injections are primarily helpful only for short-term pain relief and disability improvement. Notably, other trials also concluded that pain sensitivity in patients with LSS has no effect on the degree of pain relief after epidural steroid injections. ^{44 , 45} It is important to mention that epidural injections in general bear a risk of introducing infections to the CNS. This should be kept in mind when making the decision to proceed with injections in the aging population, where there may be a higher incidence of depressed immunity. ³⁵

Among the other treatments that have some popularityis spinal manipulation. It is commonly used to treat chronic low back pain, although until recently it was considered contraindicated for confirmed LSS. ⁴⁶ Recent studies showed very low-quality evidence that flexion distraction manipulation could be beneficial for LSS. ^{22 , 31 , 47} Acupuncture is another common treatment, although there is limited evidence that it is beneficial for LSS. ²³

A systematic review in 2014 evaluated the effectiveness of spine surgery for symptomatic LSS in the elderly and the safety of surgery in the same population, with emphasis on the incidence of perioperative complications. The results showed significant overall improvement of pain and disability. ²⁴ Moreover, the rate of perioperative complication such as durotomy, wound infection, and general mortality was infrequent. It also showed less favorable complication rates and outcomes for obese and diabetic patients. ²⁴ Another systematic review in 2011 that included five studies on a total of 918 patients compared the effectiveness of surgery vs. conservative treatment. In all five trials, surgery showed superiority in pain relief, disability, and quality of life, although not for walking ability. The advantage of surgery was noticeable at 3 to 6 months and remained for up to 2 to 4 years, but the differences tended to be smaller by the end of that period. ³⁰ Another randomized controlled trial involved 94 patients and compared outcomes of nonoperative management of 44 patients vs. decompression surgery for 50 patients, of which 10 had additional fusions. It concluded that surgery achieved better pain relief and less overall disability than nonoperative management over 2 years of follow-up. ²⁵ Moreover, using a design of a randomized trial concurrently with an observational cohort study, SPORT provided 4-year results that compared surgery to nonoperative care in LSS. It showed clinically significant advantages of surgery that were maintained for 4 years. These advantages included significantly less bodily pain and Oswestry Disability Index (ODI) scores. ²⁶

On the other hand, a case series compared outcomes using ODI and the surgeon’s clinical assessment of nonoperative treatment of 54 matched pairs of patients with LSS vs. laminectomy. It concluded no statistically significant difference in outcomes between them. ⁴⁸ In addition, in another nonrandomized cohort of patients with LSS, decompression (54 patients) and decompression with fusion for degenerative spondylolisthesis (42 patients) were compared to nonoperative treatment (29 patients). Both surgical interventions groups showed better outcomes and scored better on Roland Morris Disability Questionnaires for outcomes assessment. ⁴⁹