Physical Examination of the Lumbar Spine and Sacroiliac Joint


Because of the multifactorial nature of the complaint of low back pain (LBP), patients presenting with LBP necessitate a detailed history and physical examination. Nonspecific or incorrect diagnoses can lead to ineffective treatment plans that prolong morbidity and result in suboptimal utilization of community health resources. History collection should include traumatic events, personal and family histories of cancer diagnoses, rheumatologic disorders, bowel inflammatory disorders, and skin disorders, and the review of systems should be sensitive enough to identify potential systemic symptoms. Pain-relieving and pain-exacerbating activities should be identified through history because they may provide clues toward potential mechanical relief of the back pain with physical therapy and/or increased general physical activity. The examiner must also keep in mind the importance of the sacroiliac joint (SIJ) and the hip joint within the differential diagnosis of LBP; the description of hip joint pain and examination is addressed in Chapter 8 .

Anatomy and Biomechanics

The joints of the pelvis are intrinsically stable, and proper functioning is essential for normal biomechanics about the pelvis. High forces and repetitive loads can lead to ligamentous, muscle, and bone injuries or overuse syndromes. The pelvis also adapts to injuries of the spine and the lower extremities, which may lead to maladaptive syndromes. SIJ dysfunction occurs when there is an alteration of the structural or positional relationship of the sacrum on the ilium. This dysfunction commonly originates through asymmetric force transmission, as well as degenerative changes. Despite these well-known biomechanical changes, it is still unclear whether altered motion at the SIJ is the source of pain.

The pelvis serves as the central base through which forces are transmitted both directly and indirectly ( Fig. 7.1 ). In contrast to the lumbar spine, the joints of the pelvis are relatively stable; however, repetitive load, trauma, or maladaptive compensatory gait patterns from spine or lower extremity injury can result in SIJ dysfunction. The natural degenerative changes that take place must be recognized as well. The SIJ can be affected in rheumatologic disorders such as seronegative spondyloarthropathies including reactive arthritis, psoriasis, juvenile chronic arthritis, ulcerative colitis, and Crohn disease. It can also be affected by autoimmune and infectious diseases and malignancy with and without metastasis. Radiographic changes about the SIJ have been well documented in ankylosing spondylitis (AS), including joint space narrowing, subchondral sclerosis, and osteophytosis.

Figure 7.1

Biomechanics of the sacroiliac joints during walking: a, rotation of the iliac bone at the non–weight-bearing side around a frontal horizontal axis; b, rotation of the iliac bone at the weight-bearing side around a vertical axis.

(Adapted with permission from Ombregt L, Bisschop P, Ter Veer HJ, et al. Applied anatomy of the sacroiliac joint. In: Ombregt L, Bisschop P, Ter Veer HJ, et al, eds. A System of Orthopedic Medicine. London: W.B. Saunders; 1995:694.)

Bodyweight and postural changes may create or inhibit motion at the SIJ, while the majority of indirect motion occurs secondary to the muscle groups surrounding the joint. These include the gluteals, hamstrings, hip external rotators, iliopsoas, abdominals, latissimus dorsi, quadratus lumborum, and erector spinae ( Fig. 7.2 ). Restriction of any of these muscle groups may subsequently alter mechanics about the SIJ. Acquired hyper- or hypomobility about the SIJ results in altered load transmission, which may have an impact on the other components of the lower extremity kinetic chain. Studies have shown various ranges of motion about the SIJ, but most agree that approximately 4 degrees of rotation and 1.6 mm of translation occur. The amount of joint motion decreases with age. Women develop degenerative changes that restrict motion at age 50 years, and men develop them around age 40 years. At the present time, the clinical impact of age-related degeneration about the SIJ is unclear.

Figure 7.2

Musculature about the sacroiliac joints: 1, psoas; 2, iliacus; 3, gluteus; 4, erector spinae; 5, sacroiliac joint.

(Adapted with permission from Ombregt L, Bisschop P, Ter Veer HJ, et al. Applied anatomy of the sacroiliac joint. In: Ombregt L, Bisschop P, Ter Veer HJ, et al, eds. A System of Orthopedic Medicine. London: W.B. Saunders; 1995:690.)

The presentation of SIJ dysfunction can be similar to other forms of LBP, which may make clinical diagnosis more challenging. SIJ-mediated pain can also be confounding in the appearance of lumbosacral radicular pain, with a prevalence between 15% and 30% and an unknown incidence. Ebraheim et al. via a cadaveric dissection study, determined that the fifth lumbar nerve root and lumbosacral trunk coursed across the sacroiliac at a level 2.0 ± 0.2 cm below the pelvic brim and were relatively fixed to the sacral ala with fibrous connective tissue. Bernard and Kirkaldy-Willis reported that of 1293 patients with LBP, SIJ dysfunction was thought to be the pain generator in 22.5% based on history and physical exam, with a referral pattern mainly involving the buttock and posterolateral calf, although SIJ syndrome occurred in isolation in only 61% of this subgroup. SIJ pain referral patterns have been documented after provocative SIJ injection, and show extension to the medial/lateral buttocks, greater trochanter, and the superior-lateral thigh. Pain or tenderness over the region of the posterior superior iliac spine is the most common symptom in patients with SIJ pain. The area of maximal pain has been defined from the medial buttock extending approximately 10 cm caudally and 3 cm laterally from the posterior superior iliac spine (PSIS) ( Fig. 7.3 ).

Figure 7.3

Typical pain location for sacroiliac joint dysfunction.

(Adapted with permission from George SZ, Delitto A. Management of the athlete with low back pain. Clin Sports Med. 2002;21:112.)

With these diagnostic challenges in mind, frequently used factors on history and physical examination include the determination of exacerbating activities, location pattern of the pain (midline vs paramidline), and presence or absence of the centralization/peripheralization phenomenon. On history, SIJ pain is frequently aggravated after prolonged sitting or standing and with loading of the leg of the affected side while the hip is in flexion. Transitional pain may also be a feature, such as during sit-to-stand motions.

To evaluate the utility of the location pattern of LBP as a predictor of the pain’s source, Delpalma and coworkers performed a retrospective chart review of 170 cases with definitive diagnoses of the LBP origin. A significantly greater percentage of patients with internal disc disruption reported midline LBP (95.8%) compared with facet-mediated (15.4%) or SIJ-mediated pain (12.9%). A significantly lower percentage of patients (67.3%) with internal disc disruption reported paramidline pain compared with facet-mediated (95%) or SIJ-mediated pain (96%). The specificity of midline LBP for internal disc disruption was 74.8% compared with 28.0% for facet-mediated pain and 36.0% for SIJ-mediated pain. They concluded that the presence of midline LBP increased the probability of internal disc disruption and decreased the probability of symptomatic facet or sacroiliac dysfunction, while isolated paramidline LBP increased the probability of symptomatic facet or SIJ pain and mildly reduced the likelihood of internal disc disruption.

The centralization/peripheralization phenomenon is noted clinically as the movement of pain centrally toward the lumbar spine or peripherally toward the extremities in response to repeated lumbar movements. This phenomenon has been associated with pain from discogenic rather than sacroiliac sources and has shown high sensitivity but low specificity in patients with chronic LBP.

A comprehensive physical examination of LBP patients includes evaluation of the neurologic, vascular and musculoskeletal system. Physical examination of this region should flow through a specific sequence repeatedly by an examiner in order to avoid missing key elements. A commonly performed sequence includes: inspection, palpation, range of motion (ROM), flexibility, functional and neurologic assessments, and provocative maneuvers. Motion testing and provocative maneuvers should be performed for both the low back and for the sacroiliac and hip joints. The following sections separate physical exam of the lumbar spine from that of the SIJ.

Physical Exam of the Lumbar Spine

Inspection ( )

Examination of the low back begins with the initial observation of the patient, noting the patient’s preferred position while awaiting the physician. Patients with uncomfortable discogenic and radicular pain may be noted to be standing or moving as opposed to being seated while they wait. At other times, patients with LBP will maintain rigid postures, and motions will be noted to be hesitant to avoid bending or twisting, which may exacerbate their pain. An antalgic or listing posture should be noted if present. Patients should disrobe to either an examination gown or disposable examination shorts with an easily removed top to allow complete visualization.

Bony landmarks are useful in determining site of pain in the lumbar spine. Important landmarks include the anterior superior iliac spine (ASIS) at the level of the sacral promontory and the PSIS at the level of the spinous process of the second sacral vertebra. A commonly used reference line is created by passing a horizontal line connecting the highest points of both iliac crests, which crosses the vertebral column at the level of the L4 to L5 intervertebral space or the L4 vertebra. Notably, Duniec and coworkers performed a recent study comparing palpation guided identification of this line by an anesthetist with ultrasound evaluation before lumbar puncture and found that in 36% of cases, the palpatory exam was off by a mean of one level. This study corroborated similar results found by Locks et al. in 2010, who found that the L3 to L4 space determined anatomically in the seated position in obese and nonobese women undergoing elective cesarean section was only 49% and 53% accurate, respectively, when confirmed with ultrasound evaluation.

Forming a line through the tubercles of the iliac crests creates the intertubercular plane, which cuts the body of the fifth lumbar vertebra. The upper margin of the greater sciatic notch is opposite the spinous process of the third sacral vertebra, and slightly below this level is the posterior inferior iliac spine (PIIS). The surface markings of the posterior inferior iliac spine and the ischial spine are both situated in a line that joins the posterior superior iliac spine to the outer part of the ischial tuberosity; the posterior inferior spine is 5 cm and the ischial spine 10 cm below the posterior superior spine; the ischial spine is opposite the first portion of the coccyx. With the body erect, the line joining the pubic tubercle to the top of the greater trochanter is practically horizontal; the middle of this line overlies the acetabulum and the head of the femur.

Patients should be observed from the front, side, and rear. Anterior observation should assess head position, height symmetries of the shoulders and iliac crests, patellar directionality, and lower limb positioning. The head should sit straight on the shoulders, which themselves should be of equal height, although the hand-dominant side may be slightly lower in athletic or sporting individuals. ( Fig. 7.4A ). The iliac crest height should be equal ( Fig. 7.4B ). The patellae should be pointing anteriorly and the lower limbs straight, with any deformity such as valgus–varus misalignment noted.

Figure 7.4

Inspection. A, Assessment of shoulder height. B, Assessment of iliac crest height.

Posterior observation involves inspection of spinal, bone, and soft tissue alignment. Shifting of the pelvis or shoulders may be noted in the face of nerve root injury or gluteus medius weakness, similar to a Trendelenburg sign. With a disc herniation lateral to the nerve roots, the patient may list away from the side of the irritated nerve root in an attempt to draw the nerve root away from the disc; this is termed a lateral shift. When the herniation is medial to the nerve root, the patient may list toward the side of the lesion. The reliability of lateral shift judgments was examined by Clare and others in different levels of physical therapists with training in the McKenzie method using stable photographic slides of 45 patients with LBP, and it was found that lateral shift judgments had only moderate reliability (intraclass coefficient [ICC] range of 0.48 to 0.64).

Scoliotic curves may be evaluated by inspection of the height of the shoulders, scapula, and iliac crests. The spines of the scapulae begin at the level of the third thoracic vertebra (T3) and should be at the same angle. The inferior angles of the scapula (T8) should be equidistant from the spine. Any curve in the spine should be noted, as well as muscular asymmetry. A rib hump deformity can be noted with trunk forward flexion or may be seen through apparent scapular winging. The waist angles should be equal. Comparing the heights of the iliac crests or posterior superior iliac spines for any differences may identify pelvic obliquity, which could indicate a functional leg length discrepancy, or may stem from a spinal deformity such as scoliosis or an anomalous vertebrae. The relationship between the PSIS and the ASIS should be noted to assess for level of pelvic tilt. In addition, the gluteal folds and popliteal creases should be level.

Upon examining a patient from the side, the ear should drop a plumb line even with the tip of the shoulder and the peak of the iliac crest. A gentle lumbar lordotic curve is normal. Any exaggerated or increased curve should be noted. Exaggerated lordosis may be associated with a hip flexor contracture, weak hip extensors, or spondylolisthesis. The alignment of the lower extremities should also be observed from the side in neutral stance.

The skin may reveal ecchymosis after blunt trauma, erythema with infection or inflammation, or rashes with shingles or infection. Atrophy of the tissues may be noted with the presence of nerve root or peripheral nerve injury.


Palpatory examination in ambulatory patients begins in the standing position, usually with the examiner placing their fingers along the tops of the iliac crests with thumbs directed toward the midline L4 to L5 interspinous level. Palpation as such will allow further assessment of iliac crest height and symmetry in patients with obese body habitus that obscures visual examination. While the patient is standing, the spinous processes are palpated sequentially, not only to evaluate for pain but also for the presence of a possible step-off deformity from one level to the next, which may be indicative of spondylolisthesis. Palpation of the paraspinal musculature is performed to identify tender or trigger points as well as regions of muscle spasm; tenderness may be found in regions of referred pain such as the gluteal musculature. Midline palpation may elicit pain with symptomatic intervertebral disc disorders. Tenderness to palpation or percussion of the vertebrae should be noted because it may be suggestive of metastasis, compression fractures, or osteomyelitis and should be correlated with the patient’s history. Deyo and others described palpation of soft tissue and bony tenderness as having both poor reproducibility (kappa coefficient [κ] = 0.40) and specificity; however, it is essential to perform. To complete the examination, palpation of the posterior superior iliac spines, iliac crests, the greater trochanters, and ischial tuberosities are all important in localizing the etiology of the patient’s symptoms.

Range of Motion

Lumbar ROM is used to measure impairment and identify restrictions in patients with LBP. The literature supports greater accuracy of passive range assessment over active range assessment. Motion must be assessed in all planes, document any side-to-side differences. ROM may be affected by age and sex. Total sagittal ROM, flexion angle, and extension angle decline as age increases. A comprehensive normative database of lumbar ROM indices was determined by Troke and associates, who examined 405 asymptomatic subjects ( Table 7.1 ).

Table 7.1

Maximum and Minimum Median Ranges of Lumbar Spinal Motion Across All Subjects (Overall Age Range of Subjects, 16 to 90 Years)

Reproduced with permission from Troke M, Moore AP, Maillardet FJ, et al. A normative database of lumbar spine ranges of motion. Man Ther 2005;10:198-206.

Movement Male Female
Max Min Max Min
(Median of Values) (Deg) (Median of Values) (Deg)
Flexion 73 40 68 40
Extension 29 7 28 6
Right lateral flexion 28 15 27 14
Left lateral flexion 28 16 28 18
Right axial rotation 7 7 8 8
Left axial rotation 7 7 6 6

Mellor and colleagues found greater variations in proportional motion between lumbar vertebrae in subjects with LBP versus controls using quantitative fluoroscopy. In a systematic review and meta-analysis, Laird and associates found that on average, persons with LBP have reduced lumbar ROM and proprioception and move more slowly than healthy counterparts.

Studies have measured lateral flexion from different reference points, which has led to differing normative data. Using T12–L1 to L5–S1 as the reference points for measuring lateral flexion, the range on both sides was 49 to 77 degrees, and Pearcy and associates measured the segmental ROM of the spine ( Table 7.2 ).

Table 7.2

Segmental Range of Motion (in Degrees)

Reproduced with permission from McGill SM. Low Back Disorders: Evidence-based Prevention and Rehabilitation. Champaign, IL: Human Kinetics; 2002.

Level Flexion Extension Lateral Bending Axial Twist
L1/L2 8 5 6 2
L2/L3 10 3 6 2
L3/L4 12 1 8 2
L4/L5 13 2 6 2
L5/S1 9 5 3 5

A common starting measurement is forward flexion, which is composed of both lumbar and pelvic movement. It is generally considered that the first 60 degrees takes place in the lumbar spine, while any further motion takes place in the hips; however, strict measurements validating this have not been performed. Torso flexion is accomplished with a combination of hip and lumbar spine motion. Caillet has demonstrated that the initial 45 degrees of trunk flexion is essentially the reversal of lumbar lordosis and that the remainder of the motion is a result of pelvic rotation.

Radiographic analysis is considered the gold standard for determining ROM; there is no physical examination gold standard to measure ROM of the lumbar spine in the peer-reviewed literature. Studies have used external measurements with comparison with plain radiographs with inconsistent findings. Mayer and coworkers and Saur and associates have found good correlation between these measurements. Mayer and coworkers found no significant differences between radiographic ROM measurements and noninvasive inclinometer techniques. Saur and associates found a very close correlation ( R = 0.93) of ROM taken with and without radiologic evaluation using an inclinometer. A systematic review by Littlewood and May indicated limited positive evidence that the double inclinometer method is valid for measuring total lumbar ROM, conflicting evidence for double inclinometer measurement of flexion range, and limited evidence that the modified–modified Schober test is not valid for measurement of lumbar flexion range.


The finger-to-floor distance integrates lumbar spine and hip ROM. Patients stand with their knees extended and bend forward towards the floor. The distance between the fully extended middle finger and floor can be measured or visually estimated. When used quantitatively, the intratester reliability has been shown to be 76% and the intertester reliability 83% for using this method. The specificity of this method has been shown to be 88.8% with a sensitivity of 45.3%. Robinson and Mengshoel obtained a smallest detectable change of 9.8 cm at which an examiner can be certain a change has occurred. In a study evaluating 111 adolescents aged 12 to 14 years, finger-to-floor distance achieved an acceptable intrarater and interrater agreement (ICC ≥ 0.75), although the study was limited to comparison between two experienced chiropractors. Observation of the maneuver can expose limitations in pelvic mobility if the motion is noted to initiate and maintain primarily in the lumbar spine, potentially secondary to tight hamstring musculature.

Right and left lateral flexion of the lumbar spine is assessed similarly. Thomas et al. evaluated 344 patients with new-onset LBP and 118 individuals without LBP. Lateral flexion was measured as the distance covered by the fingertips on the lateral thigh. Right lateral flexion had a sensitivity of 23% and a specificity of 94%, while left lateral flexion had a sensitivity of 26% and a specificity of 92%.

Schober Test and Modifications Thereof

In 1937, Schober first described a test to measure segmental motion of the lumbar spine. It is described as follows:

The first sacral spinous process is marked, and a mark is made about 10 cm above this mark. The patient then flexes forward, and the increased distance is measured. If there is normal motion of the lumbar spine with absence of disease, there should be an increase of 4–5 cm.

This test is used only to measure flexion, with intratester variation reported to be just 4.8%. The Schober test has been criticized because of the difficulty in isolating surface landmarks through different depths of subcutaneous tissue. A smallest detectable change of 1.8 cm was determined by Robinson and Mengshoel. In a study comparing subjects with ankylosing spondylitis with healthy control participants, Rezvani and colleagues found excellent intrarater reliability but only a weak correlation between the Schober test and radiographically analyzed spinal motion.

Modified Schober Test

In 1969, Moll and Wright modified the Schober technique for the assessment of patients with ankylosing spondylitis and others, with the addition of a mark 5 cm below S1. It is described as follows:

With the subject standing erect but relaxed, a point is drawn with a skin marker at the spinal intersection of a line joining the dimples of Venus (S1). Additional marks are made 10 cm above and 5 cm below S1. Subjects are asked to bend forward. The distance between the marks 10 cm above and 5 cm below S1 is measured.

The rationale for this modification was an observation that, on forward flexion, both the lumbosacral junction and superiorly placed 10-cm skin marks tended to move less relative to the spinous processes and the skin than the previously used mark 5 cm inferior to the sacrum ( Fig. 7.5 ).

Figure 7.5

A, Modified Schober technique (neutral standing). B, Modified Schober technique (full flexion).

Reynolds found this measurement of motion to have good reliability in flexion and extension, with reasonable variation in flexion. Fitzgerald and others reported a Pearson correlation coefficient of 1.0 for lumbar flexion and 0.88 for lumbar extension in a study of young healthy subjects. Rezvani and associates found that the modified Schober test reflected spinal mobility better than the original Schober test.

Gill and colleagues concluded that the modified Schober method was the most reproducible method of measurement. The coefficient of variation (CV) was excellent, ranging from 0.9% for flexion and 2.8% for extension.

Miller and associates noted overall good interrater reliability ( R = 0.71) of the modified Schober method. Potential errors that affect the reliability of this clinical test were (1) the presence or absence of dimples of Venus, (2) anatomic location of the dimples of Venus, (3) anatomic variability of the 10 cm line, (4) problems introduced by skin distraction, and (5) problems in developing a normative database. Stankovic and coworkers noted an intrarater correlation coefficient of 0.95 and an interrater correlation coefficient of 0.94 utilizing the modified Schober test. Thomas and others determined the specificity to be 95% and sensitivity to be 25% when comparing patients with and without LBP.

Viitanen and associates compared the modified Schober test with the thoracolumbar flexion measure using a simple tape method and correlated the results with radiologic changes in patients with ankylosing spondylitis (AS). The Schober test and tape methods correlated fairly highly (modified Schober R = 0.71, 0.62; tape method R = 0.49, 0.42) with radiologic changes. Macrae and Wright found the validity of modified Schober against radiographs was strong ( r = 0.97) while Rahali-Khachlouf found it to be moderate ( r = 0.59).

In conclusion, the modified Schober test has been found to be moderately reproducible and specific but not sensitive when examining patients with and without LBP, and at least moderately reflects lumbar spine motion compared to radiographs.

Modified–Modified Schober Test

The modified–modified Schober technique was first described by Van Adrichem and Van der Korst in 1973. Williams and associates reported interrater reliability, with Pearson correlation coefficients from 0.72 for flexion and 0.76 for extension, using the further modified Schober technique.

Tousignant and colleagues examined 31 subjects with LBP and found moderate validity with the gold standard of radiography, with a Pearson correlation coefficient of 0.67. This is in contrast to Rezvani and coworkers who found it reflected spinal motion poorly in subjects with ankylosing spondylitis. Therefore, the modified–modified Schober (MMS) test may be more a valid ROM measurement in patients with LBP of nonspondylarthritic origin.


An inclinometer is a handheld, circular, fluid-filled disc with a weighted gravity pendulum that remains vertically oriented. The inclinometer technique has been described using single and dual techniques, with dual inclinometry demonstrating superior results. Dual inclinometry requires one inclinometer to be placed on the sacrum to measure hip motion and the other placed on the first lumbar vertebra to measure hip and lumbar ROM. Loebl described a method of measuring four spinal segments with one inclinometer placed on the T12 spinous process and the other at a point 15 cm above the S1 spinal level. This test was based on the assumption that the curvature of the spine can be determined by the angle formed by the tangent of one point on the curve with the tangent of another point on the curve. ROM is then measured by calculating the differences between angles measured while the back is in neutral, flexed, and extended positions.

The AMA Guides to the Evaluation of Permanent Impairment, 5th edition , recommends the use of dual inclinometry for measuring lumbar ROM with techniques as described by Loebl. Nattrass evaluated the validity of spinal ROM methods as outlined in the second and fourth editions of the AMA Guidelines for Impairment and Disability. Comparing goniometry and dual inclinometry, interrater and intrarater reliability were found to be poor (Pearson correlation coefficient ranges from −0.38 to 0.54) with measurement error for thoracolumbar and lumbar movements as large as ± 30 degrees, with the smallest error being 9 degrees. Reliability measures have varied greatly in the literature ( Table 7.3 ).

Table 7.3

Tests of Lumbar Spine Motion

Test Description Reliability/Validity Tests Comments
Fingertips to floor The subject is asked to stand erect with knees extended and bend forward as far as possible. The distance between the middle finger and the floor is measured with a measuring tape.

  • Merrit et al. 1986

  • Intratester reliability: 76%

  • Intertester reliability: 83%

Not specific for lumbar spine because it assesses both lumbar and hip motion
The specificity of this method has been shown to be 88.8% with a sensitivity of 45.3%. Smallest detectable change of 9.8 cm at which an examiner can be certain a change has occurred
Lateral flexion of lumbar spine can be measured in a similar manner.

  • Thomas et al. 1998

  • Right lateral flexion:

    • Sensitivity: 23%

    • Specificity: 94%

  • Left lateral flexion:

    • Sensitivity: 26%

    • Specificity: 92%

Compared spinal ROM in 344 patients with new-onset LBP with 118 without LBP. The subject stood with head and buttocks pressed against the wall with no knee flexion, and was asked to bend sideways.
Schober test The first sacral spinous process is marked, and a mark is made about 10 cm above this mark. The patient then flexes forward, and the increased distance between marks is measured.

  • Biering-Sorensen 1984

  • Intratester variation reported to be only 4.8%.

  • Schober test is only used to measure flexion

  • Criticized for difficulty in isolating surface landmarks with different tissue depths

  • Rahali-Khachlouf et al. 2001

  • Reliability:

    • Strong intrarater: ( r = 0.96)

    • Strong interrater: ( r = 0.90)

  • Robinson and Mengshoel 2014

  • Smallest detectable change of 1.8 cm

  • Validity correlation with radiographic data:

    • Macrae and Wright 1969

    • Strong ( r = 0.90)

    • Rahali-Khachlouf et al. 2001

    • Moderate ( r = 0.68)

Modified Schober test With the subject standing erect but relaxed, a point is drawn with a skin marker at the spinal intersection of a line joining the dimples of Venus (S1). Additional marks are made 10 cm above and 5 cm below S1. Subjects are asked to bend forward. The distance between the marks 10 cm above and 5 cm below S1 is measured.

  • Reynolds 1975

  • Pearson correlation coefficients:

    • 0.59 for lumbar flexion

    • 0.75 for extension

  • COV:

    • 11.65% for flexion

    • 21.57% for extension

  • Fitzgerald et al. 1983

  • Pearson correlation coefficients:

    • 1.00 for lumbar flexion

    • 0.88 for lumbar extension

Study performed on young healthy subjects, may not translate to older individuals

  • Gill et al. 1988

  • COV:

    • 0.9% for flexion

    • 2.8% for extension

Concluded that modified Schober method was most reproducible method of measurement as compared to fingertip-to-floor, modified–modified Schober, inclinometer, and photometric techniques

  • Miller et al. 1992

  • Interrater reliability: 0.71

  • Potential errors:

    • (1)

      Presence or absence of dimples of Venus

    • (2)

      Anatomic location of the dimples of Venus

    • (3)

      Anatomic variability of the 10-cm line

    • (4)

      Problems introduced by skin distraction

    • (5)

      Problems in developing a normative database

  • Rahali-Khachlouf et al. 2001

  • Interrater reliability: 0.92

  • Intra-rater reliabilitiy: 0.96

  • Stankovic et al. 1999

  • Intrarater CC: 0.95

  • Interrater CC: 0.94

  • Thomas et al. 1998

  • Specificity: 95%

  • Sensitivity: 25%

Comparing patients with and without LBP

  • Validity correlation with radiographic data:

  • Viitanen et al. 1999

    • Moderate to strong

    • ( r = 0.71, 0.62)

    • Macrae and Wright 1969

    • Strong ( r = 0.97)

    • Rahali-Khachlouf et al. 2001

    • Moderate ( r = 0.59)

In patients with AS
Modified–modified Schober test The PSISs are identified and a mark is made on the midline of the lumbar spines horizontal to the PSIS. Another mark is placed on the spinous processes 15 cm superior to the PSIS line. A tape measure is aligned between the two marks, and the patient is asked to bend forward or backward depending on the motion being measured. The new distance between the markings is measured. The difference between the two measurements is recorded.

  • Van Adrichmem and Van der Korst

  • Pearson correlation coefficients:

    • Lumbar flexion: 0.78 and 0.89

    • Lumbar extension: 0.69 and 0.91

  • Interrater reliability:

    • Flexion: 0.72

    • Extension 0.76

  • Williams et al. 1993

  • Interrater reliability:

    • Flexion: 0.72

    • Extension: 0.76

  • Rezvani et al. 2012

  • Intrarater reliability:

    • Flexion: 0.97

  • Validity correlation with radiographic data:

    • Tousignant et al. 2005

    • Moderately valid

    • Pearson correlation coefficient 0.67

    • Rezvani et al. 2012

    • No validity established

Inclinometer Dual inclinometry requires one inclinometer to be placed on the sacrum to measure hip motion and the other placed on the first lumbar vertebra to measure hip and lumbar range of motion. ROM is then measured by calculating the differences between angles measured while the back is in neutral, flexed, and extended positions.

  • Nattrass et al. 1999

  • Pearson correlation coefficient ranged from −0.38 to 0.54

  • Measurement error for thoracolumbar and lumbar movements as large as ±30 degrees, with the smallest error being 9 degrees.

Evaluated the validity of spinal ROM methods as outlined in the second and fourth editions of the AMA Guidelines for Impairment and Disability ; compared goniometry and dual inclinometer

  • Reynolds 1975

  • Intertester reliability:

    • Flexion, R = 0.76

    • Extension, R = 0.87

Standard inclinometer

  • Burdett et al. 1986

  • Intertester reliability:

    • Flexion, R = 0.73

    • Extension R = 0.15

Gravity inclinometer

  • Merrit et al. 1986

  • Intertester reliability:

    • CV flexion: 9.6

    • CV extension: 65.4

  • Intratester reliability:

    • CV flexion: 13.4

    • CV extension: 50.7

Standard inclinometer

  • Dillard et al. 1991

  • Intertester reliability:

    • Flexion, R = 0.78

    • Extension, R = 0.27

  • Intratester reliability:

    • Lateral flexion, R = 0.66

Dual inclinometer

  • Saur et al. 1996

  • Intertester reliability:

    • Flexion, R = 0.88

    • Extension, R = 0.94

Standard inclinometer

  • Ng et al. 2001

  • Intratester reliability:

    • Flexion, R = 0.87

    • Extension, R = 0.92

    • Right lateral flexion, R = 0.96

    • Left lateral flexion, R = 0.94

    • Right axial rotation, R = 0.96

    • Left axial rotation, R = 0.94

Modified inclinometer with pelvic restraint

  • Dopf et al. 1994

  • Intratester reliability:

    • R = 0.93

  • Nattrass et al. 1999

  • Intratester reliability

    • Flexion, R = 0.90

    • Extension, R = 0.70

    • Lateral flexion, R = 0.89–0.90

Dual inclinometer

  • Williams et al. 1993

  • Intratester reliability:

    • Flexion, R = 0.13–0.87

    • Extension, R = 0.28–0.66

  • Keeley et al. 1986

  • Intratester reliability:

    • Extension, R = 0.90–0.96

Dual inclinometer

  • Portek et al. 1983

  • Intratester reliability:

    • Flexion, R = 0.86

Standard inclinometer

  • Mellin 1986,1987

  • Intratester reliability:

    • Flexion, R = 0.86

    • Extension, R = 0.93

    • Lateral flexion, R = 0.6–0.85

Dual inclinometer

  • Gill et al. 1994

  • Intratester reliability:

    • CV flexion: 9.3–33.9

    • CV extension: 3.6–4.7

Dual inclinometer

AS, Ankylosing spondylitis; CC, correlation coefficient; CV, coefficient of variation; LBP, low back pain; PSIS, posterior superior iliac spine; ROM, range of motion.

With the development and use of computerized digital inclinometers (CDIs), their reliability and validity have been studied. MacDermid and coworkers measured sagittal plane ROM in 20 subjects with LBP and 20 without, comparing CDIs with the MMS test. They found high to very high intratrial reliability for both methods (ICC 0.85 to 0.96 for CDI and 0.84 to 0.98 for MMS); however, intertrial reliability of the CDI was poor to moderate, and poor correlations were found between the CDI and MMS for flexion measurements. Bedekar and associates found moderate to high reliability of an iPod with goniometer software, but they did not compare it with standard measurement devices. Kolber and colleagues compared an iPhone inclinometer with dual bubble inclinometry in 30 asymptomatic subjects and found good intrarater and interrater reliability with nearly equivalent ICCs for bubble inclinometry (≥0.81) and the iPhone application (≥0.80). Validity between the two devices was good with an ICC ≥ 0.86, but it was noted that individual differences of up to 18 degrees may exist when devices were used interchangeably.

The inclinometer can be used to measure lateral flexion, which is determined by subtracting measurements between the T12 and sacral inclinometers, with good to excellent intrarater reliability (0.60 to 0.96). Ohlen and others measured 82 degrees of lateral flexion with a COV of 19%.

In conclusion, the inclinometer is moderately reliable but is not consistent in patients with LBP.

Neurologic Examination

Manual Motor Testing ( )

Strength assessment is typically performed in a sequential manner, evaluating muscle groups innervated by different peripheral nerves and nerve roots. Lumbar radiculopathy is usually characterized by weaknesses affecting two or more muscles from the same spinal segment but different peripheral nerves. For example, an L5 radiculopathy may affect both the dorsiflexors of the foot and toes (peroneal nerve) and abduction of the hip (superior gluteal nerve). The strength examination should include the assessment of the hip flexors (L1–L3), quadriceps (L2–L4), tibialis anterior (L4–L5), extensor hallucis longus (L5), and gastrocnemius-soleus (S1). The latter muscle may be assessed via the performance of 10 toe raises (unilaterally) or the ability to ambulate on toes. Functional testing of the hip abductors should be performed with the patient standing on one leg to evaluate for the presence of a Trendelenburg sign, noted as a sagging of the iliac crest on the unloaded side.

Several studies have looked at the sensitivity and specificity of muscle strength testing in patients with lumbar radiculopathy ( Table 7.4 ). Kerr and coworkers demonstrated reduced ankle dorsiflexion in 54% and plantar flexion in 13% of those with lumbar disc protrusions from L4 to S1 with an overall specificity of 89%. Weakness of the extensor hallucis longus had a sensitivity of 12% to 51% with a specificity of 72% to 91% for detecting L5 radiculopathy, whereas weak ankle plantar flexors had an overall specificity between 26% and 99% in detecting S1 radiculopathy. In a cross-sectional study, Ortiz-Corredor correlated physical exam findings to abnormal electromyogram (EMG) findings in LBP patients and found weak plantar flexors were highly specific (97.5%), but weak quadriceps were poorly sensitive (35.67%). In a small retrospective chart review of 28 patients, Iizuka and colleagues found that the most common condition causing a foot drop (manual muscle testing grade of 0 to 3) was compression of two nerve roots. A systematic review and meta-analysis by Al Nezari and others in 2013 determined that motor testing for paresis showed low pooled sensitivities (22% to 40%) and moderate specificities (62% to 79%) for surgically and radiologically determined disc herniations, while motor testing in muscle atrophy exhibited a pooled sensitivity of 32% and a specificity of 76% for surgically determined disc herniations.

Table 7.4

Neurologic Exam as a Test for Lumbar Disc Herniation

Test Description Reliability/Validity Tests Comments
Muscle strength testing Assessment of strength must be performed in a sequential manner, evaluating muscle groups innervated by different peripheral nerves and nerve roots. Radiculopathy is characterized by weaknesses affecting two or more muscles from the same spinal segment but different peripheral nerves.

  • Spangfort 1971

  • Ankle dorsiflexor weakness:

    • Sensitivity: 49%

    • Specificity: 54%

Tested ankle dorsiflexor in 2504 patients; 70–90% of patients with weakness had HNP at L4/L5 level

  • Hakelius and Hindmarsh 1970 and Hakelius 1972

  • Ankle dorsiflexor weakness:

    • Sensitivity: 20%

    • Specificity: 82%

  • Ankle plantar flexor weakness:

    • Sensitivity: 6%

    • Specificity: 95%

The examination should include hip flexors (L1–L3), quadriceps (L2–L4), tibialis anterior (L4–L5), extensor hallucis longus (L5), and the gastrocnemius/soleus complex (S1).

  • Great-toe extensor weakness:

    • Sensitivity: 37%

    • Specificity: 71%

  • Quadriceps weakness:

    • Sensitivity: <1%

    • Specificity: 99%

  • Kerr et al. 1988

  • Ankle dorsiflexor weakness:

    • Sensitivity: 54%

    • Specificity: 89%

  • Ankle plantar flexor weakness:

    • Sensitivity: 13%

    • Specificity: 100%

  • Kortelainen et al. 1985

  • Ankle dorsiflexor weakness:

    • Sensitivity: 57%

  • 32% HNP at L4/L5

  • 57% HNP at L5/S1

  • Knutsson 1961

  • Ankle dorsiflexor weakness:

    • Sensitivity: 63%

    • Specificity: 52%

For L5 root specifically, the sensitivity was 76%, and the specificity was 52%.

  • Lauder 2002

  • Great-toe extensor weakness:

    • Sensitivity: 61%

    • Specificity: 55%

  • Quadriceps weakness:

    • Sensitivity: 40%

    • Specificity: 89%

  • Ankle plantar flexor weakness:

    • Sensitivity: 47%

    • Specificity: 76%

  • Ortiz-Corredor 2003

  • Ankle plantar flexor weakness:

    • Specificity: 97%

  • Quadriceps weakness:

    • Sensitivity 35.67%

Compared EMG with physical exam
Muscle stretch reflex Tapping the test tendon stretches the spindle and activates its fibers. Afferent projections from these fibers synapse with the alpha motor neurons, which in turn send impulses to the skeletal muscles, resulting in a brief contraction. This contraction is graded on a standard scale.

  • Spangfort 1971

  • Ankle:

    • Sensitivity: 50%

    • Specificity: 62%

  • Patella:

    • Sensitivity: 4%

    • Specificity: 97%

  • Ankle: HNP at L5/S1 level in 80%–90% for ages 20–45 years and 60% older than 50 years

  • Patella: Sensitivity of 50% in L3/L4 HNP. In 67% of cases of impairment, HNP is at L4/L5 and L5/S1 levels.

  • Hakelius and Hindmarsh 1970 and Hakelius 1972

  • Ankle:

    • Sensitivity: 52%

    • Specificity: 63%

  • Patella:

    • Sensitivity: 7%

    • Specificity: 93%

  • Knutsson 1961

  • Ankle:

    • Sensitivity: 56%

    • Specificity: 57%

  • Patella:

    • Sensitivity: 15%

    • Specificity: 67%

For S1, the sensitivity was 79%, and the specificity was 62%.

  • Kerr et al. 1988

    • Sensitivity: 48%

    • Specificity: 89%

For L3–L4, the sensitivity was 10%, and the specificity was 85%.

  • Lauder 2002

  • Ankle:

    • Sensitivity: 47%

    • Specificity: 90%

  • Patella:

    • Sensitivity: 50%

    • Specificity: 93%

For L5–S1, the sensitivity was 78%, and the specificity was 88%.

  • Kortelainen et al. 1985

    • Sensitivity: 7%

  • Suri et al. 2011

  • Ankle:

    • Sensitivity: 33%

    • Specificity: 91%

    • Likelihood ratio: 3.9

In 85% of cases of impairment, HNP is at L4–L5 and L5/S1.

  • Patella:

    • Sensitivity: 39%

    • Specificity: 95%

    • Likelihood ratio: 7.7

For L5 impingement

  • Iversen et al. 2013

  • Ankle:

    • Sensitivity: 44%

    • Specificity: 61%

For L4 impingement
For L4 impingement

  • Patella:

    • Sensitivity: 67%

    • Specificity: 83%

For S1 impingement
Sensory Exam The sensory examination should cover the bilateral lower extremities to evaluate for dermatomal or diffuse sensory loss. Sensation is evaluated using different modalities, including vibration, proprioception, temperature, light touch, and pinprick.

  • Kerr et al. 1988

    • Sensitivity: 16%

    • Specificity: 86%

  • Knutsson 1961

    • Sensitivity: 29%

    • Specificity: 67%

  • Kosteljanetz et al. 1984

    • Sensitivity: 66%

    • Specificity: 51%

  • Kortelainen et al. 1985

    • Sensitivity: 38%

  • Lauder 2002

    • Sensitivity: 50%

    • Specificity: 62%

  • Suri et al. 2011

  • Anterior thigh pinprick:

    • Sensitivity: 50%

    • Specificity: 96%

    • Positive likelihood ratio: 13

For L2 impingement

  • Medial ankle pinprick:

    • Sensitivity: 31%

    • Specificity: 100%

    • Positive likelihood ratio: infinity

For L4 impingement

EMG, Electromyogram; HNP, herniated nucleus pulposus.

Sensory Examination ( )

The sensory exam should cover the bilateral lower extremities to evaluate for true dermatomal or more diffuse sensory loss as seen in peripheral neuropathies with a “stocking” distribution of loss. Dermatomes define the area of skin innervated by a single nerve root or peripheral nerve ( Fig. 7.6 ). Sensation can be evaluated using many different modalities, including vibration, proprioception, temperature, light touch, and pinprick. The latter two are more commonly used in the neurologic evaluation, although vibration may be more sensitive for injury to large-diameter sensory nerves and proprioception loss may be indicative of posterior column dysfunction such as vitamin B 12 deficiency and syphilis. The sensitivity and specificity of the sensory examination in the diagnosis of lumbar disc herniation has been described in the literature (see Table 7.4 ). Suri and coworkers examined pinprick testing at the anterior thigh, medial knee, medial ankle, great toe, and lateral foot for sensitivity and specificity for mid lumbar and low lumbar nerve root impingement as well as level-specific nerve root impingement. For mid lumbar and low lumbar impingement, sensitivities ranged from 0% to 21%, though specificities ranged from 79% to 100%, with the highest specificities and significant likelihood ratios observed for the medial knee and medial ankle in mid lumbar nerve root impingement. For level-specific impingement, anterior thigh pinprick sensation was 50% sensitive and 96% specific and had a positive likelihood ratio (LR) of 13 for L2 impingement, while medial ankle pinprick was 31% sensitive and 100% specific and had an LR approaching infinity for L4 nerve root impingement.

Figure 7.6

Dermatomes and peripheral nerve distribution of the lower extremities.

(Adapted with permission from Borenstein, Wiesel, and Boden; Low Back and Neck Pain: comprehensive diagnosis and management. 3rd edition (2004). Chapter 5, Figure 5-14, page 123.)

Reflex Examination ( )

A reflex is the involuntary contraction of muscles induced by a specific stimulus. Tendon reflex activity depends on the status of the alpha motor neurons, the muscle spindles and their afferent fibers, and the gamma neurons whose axons terminate on intrafusal muscle fibers within the spindles. Rapid distension of the tendon stretches the spindle and activates its fibers. Afferent projections form these fiber synapses with the alpha motor neurons, which in turn send impulses to the skeletal muscles resulting in the familiar brief muscle contraction or monophasic stretch reflex.

The reflex is usually named after the muscle being tested. Common stretch reflexes include the quadriceps or patellar reflex involving the L2 to L4 spinal levels, medial hamstring reflex at the L5 level, and Achilles reflex involving the S1 level ( Fig. 7.7 ). A 5-point grading system recommended by the National Institute of Neurological Disorders and Stroke is the most commonly used scale ( Table 7.5 ). Eliciting reflexes may be difficult in the patient who is unable to relax. In 1885, Jendrassik described a technique to elicit reflexes by having the patient “hook together the flexed fingers of his right and left hands and pull them apart as strongly as possible” while the clinician taps on the tendon; this enhances the reflexes of patients.

Figure 7.7

A, Medial hamstring reflex. B, Prone technique for eliciting Achilles reflex.

Table 7.5

Classification of Muscle Stretch Reflexes

Reproduced with permission from Nicholas J. Talley and Simon O’Connor. Clinical examination: A systematic guide to physical diagnosis , 7e. Churchill Livingstone, 2014.

0 Absent
+ Present but reduced
++ Normal
+++ Increased, possibly normal
++++ Greatly increased, often associated with clonus

Absent or exaggerated reflexes by themselves do not signify neurologic disease. In older adults, up to 50% without neurologic disease lack an Achilles reflex bilaterally. Small percentages (3% to 5%) of normal individuals have generalized hyperreflexia.

Absent or exaggerated reflex is significant only when it is associated with one of the following clinical settings:

  • 1.

    The absent reflex is associated with other findings of lower motor neuron disease.

  • 2.

    The exaggerated reflex is associated with other findings of upper motor neuron disease.

  • 3.

    The reflex amplitude is asymmetric.

  • 4.

    The reflex is unusually brisk compared with reflexes from a higher spinal level.

Deyo and others reviewed the relevance of the physical examination in patients with LBP (see Table 7.4 ). Andersson and Deyo noted a specificity of 0.60 and a sensitivity of 0.50 for the Achilles reflex in diagnosing lumbar disc injury. In patients with a high probability (> 60%) of a herniated disc (which is based on the clinicians’ patient population and the prevalence of LBP), the positive predictive value for an impaired Achilles reflex was 0.65 with a negative predictive value of 0.44. However, in patients with a low probability of a herniated disc, the positive predictive value was 0.04 with a negative predictive value of 0.98. Suri and associates determined a 39% sensitivity, 95% specificity, and significant LR of 7.7 for L4 impingement if the patellar reflex was affected, and a 33% sensitivity, 91% specificity, and LR of 3.9 for L5 impingement if the ankle reflex was affected. Iversen and coworkers found the knee reflex to be 67% sensitive and 83% specific for L4 impingement and the ankle reflex to be 44% sensitive and 61% specific for S1 impingement.

Provocative Lumbar Spine Maneuvers

Straight-Leg Raise Test ( )

Lasègue’s sign, or straight-leg raise (SLR) test, was initially described by J. J. Frost, a student of Charles Lasègue in 1881, who memorialized his mentor through the naming of this test. In the described test, pain was induced in the distribution of the sciatic nerve upon lifting the leg while maintaining it extended via pressure on the knee. Frost suspected that the activation of pain during this maneuver was the result of pressure from the hamstring on the nerve. Lazarevic described this test in 1880, 1 year earlier than Frost, after he observed six patients to have increased pain during stretching of the sciatic nerve.

Lazarevic described a three-step approach to the test :

In the first step, the patient was asked to flex forward, maintaining his knees straight. In step 2, the patient was asked to lie supine and his trunk was slowly brought into flexion, maintaining his knees extended. In step 3, the supine patient had his leg raised with the knee extended. Elevation of the leg was stopped when the patient began to feel pain, and the angle of elevation and amount of pelvic movement was recorded. The patient was then asked to indicate the distribution of pain. All three maneuvers were noted to reproduce discomfort in the sciatic nerve distribution.

In 1884, Lucien de Beurmann concluded that during the lifting of a stretched leg, pain is evoked by stretching of the nerve rather than from the compression of the muscle. Inman and Saunders noted a 2- to 7-mm distal migration of the spinal nerve roots during performance of straight-leg raising. Falconer and coworkers described a 2- to 6-mm downward migration of the nerve roots through their respective foramen during the test. Goddard and Reid noted the L5 nerve root to move 3 mm and the S1 nerve root to move 4 to 5 mm during performance of SLR. In a recent controlled in vivo radiologic study, Rade and colleagues scanned 16 asymptomatic subjects with a 1.5 T magnetic resonance imaging (MRI) scanner; the displacement of the medullar cone during the SLR was quantified and noted to displace caudally in the spinal canal by 2.31 ± 1.2 mm with right SLR and 2.35 ± 1.2 mm with left SLR, which was speculated to be directly proportional to the sliding of the L5 and S1 nerve roots.

The classic SLR test is considered positive when the supine leg is elevated to between 30 and 70 degrees and pain is reproduced down to the posterior thigh below the knee. Pain below 30 degrees is not considered to be related to nerve root irritation. Kosteljanetz and others noted prolapsed discs in 45 of 52 individuals diagnosed during surgery with a positive SLR on physical examination. In addition, no “typical” Lasègue’s sign with pain into the leg was noted beyond 70 degrees of leg elevation. SLRs with pain induced beyond 70 degrees of leg elevation is not believed to be due to nerve root tension. Rather, it is thought likely to be related to tightness within the hamstrings or gluteal muscles ( Fig. 7.8 ).

Jul 23, 2019 | Posted by in MUSCULOSKELETAL MEDICINE | Comments Off on Physical Examination of the Lumbar Spine and Sacroiliac Joint

Full access? Get Clinical Tree

Get Clinical Tree app for offline access