There is considerable variation in the clinical tests to diagnose LSS, and “a high sensitivity is often accompanied by a low specificity and vice versa” for history and clinical exam features, which results in modest or poor predictive values (Bierma-Zeinstra et al. 2006). The overall quality of available research is poor. No golden standard exists for the diagnosis of LSS, so that determining the true sensitivity and specificity is difficult (Bierma-Zeinstra et al. 2006). For an accurate diagnosis of LSS, history findings, physical examination and radiologic findings should be considered together (Bierma-Zeinstra et al. 2006; Campos et al. 2011) and further diagnostic research is necessary (Bierma-Zeinstra et al. 2006).

The NASS (2011) published a guideline based on the best available evidence, and all recommendations concerning historical and physical findings are either of too poor or insufficient evidence to make recommandations.


  • Standing discomfort may be found in 94% of patients (Thomas 2003). To maintain comfort, patients may stand in a stooped posture (Iversen et Katz 2001).
  • There may be a straightening in the normal lumbar curve, with the patients gradually assume a kyphotic posture to minimize symptoms (Thomas 2003).


The examination should include hip, knee, ankle and foot because the entire functional chain could be (at least partly) responsible for the symptoms (Thomas 2003).

Range of motion

ROM of the spine, particularly extension followed by flexion is limited (Delitto et al. 1998; Iversen et Katz 2001), and extension might be painful (Iversen et Katz 2001). Symptoms tend to worsen with lumbar extension (Delitto et al. 1998; Jackson et Simpsons 2006) or weight-bearing , and diminishing with flexion or non-weight-bearing postures (Delitto et al. 1998).

Pain following 30 seconds of lumbar extension is reported to be associated with the diagnosis of LSS (Bayley et al. 1995; Delitto et al. 1998).

Special Tests

There is no special test to diagnose LSS. According to Bierma-Zeinstra et al. (2006) “it is not possible to draw definite conclusions about which tests are best for diagnosing lumbar spinal stenosis.”

The following tests might be used for the differential diagnosis of the claudication types:

  • Bicycle test of van Gelderen
  • Stoop Test
  • Treadmill test

For the evaluation of LSS, the Swiss Spinal Stenosis Questionnaire (SSS) seems to be the most specific current measurement tool. It measures symptom severity, physical function, and satisfaction of the patients (Albert et White, 2009). The SSS is reliable, reproducible and internally consistent when compared to other evaluation tools (Albert et White, 2009) and a validated and appropriate measure for treatment of lumbar spinal stenosis (Baisden et al. 2008).


When the anamnesis gives any functional restrictions, these should be assessed. A functional assessment for LSS could include:

  • Sitting reliving/decreasing the symptoms (Sensitivity 81-46% and specificity 16-93% (Bierma-Zeinstra et al. 2006)
  • Balance: Poor balance may also be associated with LSS (Iversen et Katz 2001).
  • Walking: Walking tolerance improves while walking with a flexed lumbar spine (Delitto et al. 1998). The gait is often wide-based in LSS patients and symptoms worsen while walking (sensitivity 71%, specificity 30%) (Bayley et al. 1995). Symptoms change when walking downhill, with a positive likelihood ratio of 3.1 (Casey et al. 2007).

No pain increase while walking is strongly associated with a low likelihood of LSS (NASS 2011).


Sensory examination should assess light touch, pinprick, and vibration, and be evaluated in the lumbar dermatome distribution (Thomas 2003). Neurological deficits are frequently reported (in 50% of cases according to Delitto et al.1998, in the majority of the subjects in the study of Iversen et Katz (2001).

Bayley et al. (1995) also strongly associates neuromuscular deficits (pinprick, vibration, reflexes, and strength) with the diagnosis of LSS. Motor weakness is present in about 1/3 of patients, with L5 myotome mostly affected (Thomas 2003).

Bierma-Zeinstra et al. (2006) state that pinprick and/or vibration deficit, weakness, and absent Achilles reflex are associated with LSS in low quality studies (with specificity above 75%).

Diagnostic Imaging

There is no golden standard to recognize LSS; it is essentially a “clinical” diagnosis (Bayley et al. 1995). A CT or MRI showing compression of nerve roots is necessary, but not enough for the diagnosis of LSS (Delitto et al. 1998, Bayley et al. 1995). “High quality” studies showed that three-sequence MRI is more sensitive than single-sequence MRI and ultrasound seems to be as accurate as CT or myelography (Bierma-Zeinstra et al. 2006).

False-positive imaging findings are well documented, so that findings from diagnostic imaging should be combined to the clinical examination before a diagnosis of LSS can be made (Delitto et al. 1998, Bayley et al. 1995). When suspecting LSS after historical and physical examination, the best available evidence (fair evidence) reports MRI to be the most appropriate imaging tool to confirm the diagnosis (NASS 2011). In case MRI cannot be performed, computed tomography myelography (CTM) is recommended  (NASS 2011). EMG and NCV lack the specificity to be used for routine diagnosis, but may be helpful in differential diagnosis (Delitto et al. 1998).

Differential diagnosis

The main symptom of LSS; lower extremity pain exacerbated by walking and relieved by sitting, can result from other conditions often seen in the elderly, like osteoarthritis of the hip (Delitto et al. 1998). This can cause making the differential diagnosis difficult. According to Campos et al. (2011), the most important differential diagnoses to be rejected are neurogenic claudication and pain due to hip conditions.

Turner and colleagues (1992) reported an average of 62% of patients in studies of surgery for LSS had complaints of neurogenic claudication. Which means that a differentiating between neurogenic and vascular claudication must be made (Jackson et Simpsons 2006), which can be difficult (Delitto et al. 1998).

Dyck and Doyles (1977) reported that a patient who is diagnosed with neurogenic claudication can pedal further with flexed spine than with spine extended. They hypothesized that if vascular claudication had been the cause, the patient would have pedalled equal distances regardless of spinal posture. They created the Bicycle test of van Gelderen. A positive test determines that the patient has neurogenic intermittent claudication (Magee 2006). Dyck (1979) described the “stoop test“ using walking instead of biking to observe the difference in walking with flexed or extended spine. Magee (2006) considers the “stoop test” as well as the treadmill test as assessment tools for neurogenic intermittent claudication.

Delitto et al. (1998) state in their review that although some patients are able to increase exercise tolerance with a flexed posture, the tests are insufficiently sensitive to be of diagnostic value but that they are of value in the differential diagnosis of LSS.

Differential diagnosis for spinal stenosis include: disc herniation, vascular claudication, peripheral neuropathy, tumor, osteoarthritis of hip/knee, osteoporotic compression fracture (Thomas 2003). To differentiate from hip arthrosis, hip ROM should be assessed, as well as provocative maneuvers for the hip (e.g. the cross over/excursion test). Pain with hip internal, external rotation may suggest hip mediated pain (Thomas 2003).


Patient History