DOI:
Review
DOI: 10.11152/mu-1245Diagnostic value of musculoskeletal ultrasound in patients with low back pain – a review of the literature
Plamen Todorov
1, Rodina Nestorova
2, Anastas Batalov
11Medical University of Plovdiv, Rheumatology Clinic, Kaspela University Hospital, Plovdiv, 2St Irina Rheumatology Centre, Sofia, Bulgaria
Received 28.07.2017 Accepted 27.09.2017 Med Ultrason
2018, Vol. 20, No 1, 80-87
Corresponding author: Plamen Todorov
Rheumatology Clinic, Kaspela University Hospital, block 2,
64 Sofia street, Plovdiv 4002, Bulgaria Phone: +359 888566478
E-mail: [email protected]
Low back pain (LBP) occurs in various groups of the population, affects men and women equally, and is among the major reasons for rheumatological or ortho- pedic consultations. Its prevalence increases steadily with age and the rate of recurrence within one year could reach 44% [1]. Many imaging modalities are available to clinicians for evaluating LBP. The application of these modalities depends mainly on the working diagnosis, the urgency of the clinical problem, the availability, and the comorbidities of the patient [2]. Conventional radiogra- phy (CR) and computerized tomography (CT) are associ- ated with radiation exposure and show primarily the bony elements of the lower back. Their widespread use in the 20th century might be among the reasons why paraspinal
soft tissues have somehow been neglected as a cause for LBP [1]. Magnetic resonance imaging (MRI) shows both bony and soft tissue structures in the axial skeleton, but its use is hampered by the duration of the examination, its relatively high costs, its limited availability (mainly in the tertiary centers) and the contraindications for this diagnostic modality [2].
Musculoskeletal ultrasound (US) is a safe, fast, inex- pensive, and widely available imaging modality that is very well tolerated by patients [3]. It allows multiplanar and dynamic examinations of the musculoskeletal system and can show the soft tissues in great anatomical detail.
US is used by a growing number of physicians and the list of its applications in rheumatology and orthopedics is growing [4].
Therefore, the question of the possible diagnostic ap- plication of US in such a common condition as LBP is very relevant to the clinical practice. On one hand, this could decrease the radiation exposure associated with CR and CT, and reduce the costs paid for MRI. The ef- fect could be even bigger in time, as LBP is frequently a chronic or recurrent disease [1]. On the other hand, by Abstract
Patients with low back pain (LBP) frequently undergo various imaging studies in the pursuit of a more precise diagnosis.
Ultrasound (US) has the advantage of being a widely available, multiplanar, fast and radiation-free diagnostic tool. Moreover, compared to most of the other imaging modalities, it is particularly efficient in the visualization and assessment of soft tis- sues. Consequently, the question about the possible diagnostic application of US in such a common pathology as LBP is very relevant to the clinical practice. For this reason, we performed a review of the literature on the diagnostic value of US in dif- ferent conditions that could cause LBP. We hereby discuss available studies on the diagnostic application of US in spinal canal stenosis and disc herniation (probably of historical significance only), as well as in the pathology of soft tissue structures like the lumbar and pelvic ligaments, muscles and entheses, the thoracolumbar fascia and the sacroiliac joints (maybe of greater importance nowadays). The evidence for the diagnostic value of US is not equivocal, though promising for some of the causa- tive conditions, and clearly this area remains open to further research.
Keywords: low back pain; etiology; ultrasound
means of US, the perispinal/perisacral soft tissues, not well visible with CR and CT, could be evaluated in great- er detail. For this reason, we made an appraisal of the diagnostic value of US in different conditions that could cause LBP, based on the available literature. We found that contrary to the well-established role of US in guid- ing the interventions in the spine, its application as a di- agnostic method in LBP is still rather controversial. The difficulties in establishing US diagnostic value in LBP are further augmented by the fact that various structures could be a source or a cause of LBP [5]. For some of them there are studies in the literature, while for others such data is lacking (table I). Consequently, we hereby review the use of US as an imaging modality in the di- agnosis of discrete conditions and anatomical structures that could cause LBP, rather than in LBP in general.
US and the bony structures of the lumbar spine Spinal canal stenosis
In the 1970s, Porter et al were the first to use US as a diagnostic tool in LBP, measuring the diameter of the lumbar spinal canal for bony stenosis. They placed the transducer in an oblique midsagittal plane, 1 cm lateral to
the midline at the L5 to visualize the spinal canal at this level [6]. The average diameter of the canal in patients with LBP was 1.44 cm, while that of healthy controls was 1.61 cm. This difference was found to be statistically significant. In addition, a cutoff value of 1.4 cm was pro- posed, with a lower diameter of the spinal canal putting the subject in the risk category for nerve roots compres- sion and pain due to spinal stenosis. In a later study, Por- ter et al also published results on the spinal canal diam- eter measurements by the same technique in more than 700 subjects from infancy to the age of 65 years, showing that the canal was relatively wide in children, reached its maximum diameter in the late teens, and reduced slightly by late adult life [7]. Later, in a study of hospital employ- ees an odd ratio of 10.7 was calculated for taking a sick leave due to LBP, and the spinal canal size being less than the tenth percentile of the measurements [8]. At the same time, other authors had similar results in a population of mining employees [9]. This technique was even con- sidered as a pre-employment screening test for manual workers [9]. Other studies however concluded that the sonographic measurement of the spinal canal diameter by dorsal approach was operator dependent [10] and lacked reliability in elderly people [11].
Table I. Proposed sources and causes of LBP (based on [5]) and the presence of corresponding US studies.
Structure or cause US study [references] Comments Bony structures:
Vertebral bodies Kissing spines Lamina impaction Spondylolisis Spinal canal stenosis
NoNo NoNo
Yes [6-13] Conflicting data, only historical significance.
Intervertebral discs:
Internal disc disruption and disc herniation Yes [14-17] US inferior to other modalities.
Joints:
Zygapophysial (facet) joints
SIJs No
Yes [18-33] Some promising results in sacroiliitis and SIJD, no data on other possible causes for SIJ pain.
Soft tissues and others:
Thoracolumbar fascia Fat herniation
Compartment syndrome Muscles:
Muscle spasm Muscle imbalance Muscle sprain Muscle trigger points Ligaments:
PSL ILL PLC
Iliac crest pain syndrome Dural pain
Epidural plexus
Yes [35,36]
NoNo
Yes [37,39,40,42]
Yes [37,39,40]
NoNo
Yes [44,45]
Yes [46]
Yes [47]
Yes [49]
NoNo
US can show structural changes in the fascia layers.
US can show muscle contractures and morphologic features.
US shows spinal and posterior pelvic ligaments in sufficient for their assessment details.
US can detect enthesopathies at the iliac crest as a probable cause.
SIJ: sacroiliac joints; SIJD: sacroiliac joints dysfunction; PSL: posterior sacroiliac ligament; ILL: iliolumbar ligament; PLC: posterior liga- ment complex.
For this reason, Tervonen et al measured the spinal canal diameter through a transabdominal acoustic win- dow [12]. Sensitivity of 84% and specificity of 60% were reported, when again a cutoff point of 1.4 cm was used.
The measurement of the spinal canal diameter by US, however, never gained widespread acceptance and in a later 10-year prospective study, Porter et al concluded that the spinal canal size on its own, had no reliable role as a predictor or an estimator of the prognosis in LBP [13].
US and the intervertebral discs
Internal disc disruption (degeneration) and disc herniation
In 1989, Merx et al compared the results of transab- dominal US examination of the lumbar discs in 50 pa- tients with discogenic LBP to the results of lumbar mye- lography and CT [14]. The sonographic examination was inconclusive in 18% of the patients. The study revealed a sensitivity varying from 63% to 77% depending on the disc level. Comparing these results to lumbar myelog- raphy and CT scanning, the authors concluded that US should not be used as a diagnostic modality in disc de- generation disease.
At the same time Kamei et al evaluated the effective- ness of transabdominal US to diagnose herniated nucleus pulposus in 80 patients [15]. Herniation was surgically confirmed in 41 discs. Of these 78% (32 discs) were with true-positive ultrasound diagnoses, 90% (37 discs) with true-positive myelographic diagnoses, and 50% (20 discs) with true-positive neurologic diagnoses. Thus, US proved to be of added value to the neurological examina- tion, but inferior to myelography.
In 1991 Tervonen et al performed a study in which transabdominal US was compared to CT/discography [16]. They examined a total of 56 discs in 29 patients.
The sensitivity of US to recognize a discographically painful disc was 95%, but its specificity was only 38%.
US examination of lumbar discs was considered suitable only as a screening tool prior to CT/discography.
The largest study on detecting disc herniation by means of transabdominal US scanning was performed by Berth et al in 2003 on 119 patients with discogenic LBP [17]. The diagnostic accuracy of US was compared to MRI as a gold standard. It depended on the level of the disc, with sensitivity and specificity being as follows:
46% and 91% at the L3/4 level; 83% and 65% at the L4/5 level; 78% and 67% at the L5/S1 level. These results con- firmed the previous observations that the diagnostic per- formance of US in intervertebral disc herniation depend on the lumbar level.
Generally, US was found to be inferior to MRI, CT, discography, and myelography in detecting lumbar disc degeneration or herniation. Nonetheless, most of the au- thors concluded that transabdominal US could be used as a screening examination to reduce the radiation exposure or painful interventions to a certain lumbar level.
US and the sacroiliac joint
The sacroiliac joints (SIJ) could be affected by vari- ous conditions and are thought to account for up to 20%
of all cases of chronic LBP [5].
SIJ dysfunction
In SIJ dysfunction the pain is considered to be caused by too lax or too stiff joints. In 1995 Buyruk et al devel- oped a noninvasive technique, called Doppler imaging of vibration (DIV) to measure the laxity of the SIJ in an objective and repeatable way [18]. They used a special device to apply vibrations to the anterior iliac spines of subjects in supine position. The vibrations were regis- tered with an US transducer placed over the ipsilateral SIJ and processed in color Doppler US (CDUS) mode.
The intensity of the vibrations was measured indirectly in threshold units (TU). A large difference in the TU’s values between the sacral and the iliac bones (across the SIJ) indicated a lax joint, while conversely the absence of a difference pointed to a stiff joint. The results in healthy volunteers showed satisfactory interindividual reproduc- ibility and no significant differences between the right and the left SIJ [19]. Damen et al applied this technique in women with and without pregnancy related posterior pelvic pain [20]. The difference in the laxity between the left and the right SIJ was significantly higher (2.2 TU) in women with pain, compared to women without pain (0.9 TU). This asymmetric laxity correlated with higher values of pain measured on the Visual Analogue Scale (VAS) and a higher number of positive pain provocation tests. Later, the same group found that a difference of more than 3.0 TU in the laxity of the left and the right SIJ had a sensitivity of 65%, a specificity of 83%, and a posi- tive predictive value of 77% for the pregnancy related posterior pelvic pain to persist after the delivery [21].
Regardless of these promising results, a later review concluded that a thorough study of the validity of the technique was still missing and so the conclusions from DIV measurement should be drown with caution [22].
Inflammation of the SIJ
Spadaro et al used grey scale US to search for an effu- sion in the SIJ in 45 spondyloarthritis (SpA) patients and 30 controls, and compared the findings to the results of a set of clinical tests designed to assess the SIJ [23]. The transducer was placed transversely over the SIJ. They
found an effusion in 38.9% of the patients and only in 1.7% of the controls. The positive likelihood ratio (LR) for the presence of inflammatory LBP was 2.67 for the patients with an effusion. This result was higher than the positive LR of any single clinical test and equal to the positive LR of a combination of three tests being posi- tive (LR=2.88). The authors used standard US equipment with a 7.5 MHz linear transducer thus pointing to the ap- plicability of this diagnostic modality in the general rheu- matological practice.
Other researchers however used more sophisticated methods to detect SIJ inflammation. In their first study on contrast-enhanced US (CEUS), Klauser et al showed that the administration of intravenous contrast media im- proved the sensitivity from 18% to 94% and the nega- tive predictive value from 72% to 94% for sacroiliitis [24]. However, the first-generation contrast media which was used had shorter contrast duration and could lead to overestimation of contrast enhancement caused by early bubbles distraction. In a later study, the same group used second-generation microbubbles contrast agents for the detection of active sacroiliitis. They measured the depth of the contrast US enhancement in the dorsocaudal part of the SIJ in 42 patients and 21 controls [25]. CEUS showed enhancement depth into the joint cleft of 18.5 mm (range 16-22.1 mm) in all clinically active SIJs, significantly deeper compared to both inactive joints of patients (3.6 mm, range 0-12 mm), and the healthy controls (3.1 mm, range 0-7.8 mm). The agreement between CEUS and the clinical rating of SIJ was excellent (100%), both positive and negative predictive values being 100% in the detec- tion of clinically active sacroiliitis. Moreover, all inactive joints were correctly classified based on the lack of deep enhancement. The authors pointed out that while vascu- larization around the dorsal superficial SIJ could be seen in healthy subjects, the extension of this vascularity into the deeper parts of the joints was found only in sympto- matic patients. As the mean examination time was only 11 minutes and the cost was less than one third of that for contrast-enhanced MRI, the authors believed that CEUS of the SIJs might be a cost-effective first imaging study in patients with inflammatory LBP. However, this modal- ity could encounter difficulties when the joints are too narrow or with many osteophytes, as well as in patients with increased Body Mass Index (BMI). In addition, the application of the contrast itself has a list of contraindica- tions [25].
Chronologically, however, the first study that as- sessed the possible application of US in the diagnosis of SIJ inflammation, used color Doppler US [26]. This study found surprisingly high sensitivity of CDUS (100%) in patients with active sacroiliitis. The finding
lacked specificity as vascularization was also found in healthy controls and in patients with osteoarthritis. The resistive index (RI) of the vessels detected by CDUS was significantly lower in patients with sacroiliitis and it in- creased to the level of the one of controls after treatment.
Several later studies also used Doppler US to detect vascularization in the region of the dorsal SIJ as a sign of inflammation. Although universally they showed an increased vascularization in sacroiliitis, one had to bear in mind the fact that vessels here were also commonly found in healthy subjects. For example, Pakkafahli et al reported vascularization near the dorsal part of the sac- roiliac joint, caused by branches of the sacral arteries, in 13 out of 23 healthy controls [27] and Klauser at al found vascularization in 12 out of 21 healthy controls [25].
Furthermore, in a Doppler US study, McGrath et al used vascular signature of the dorsal sacral arteries to identify dorsal sacral nerve rami beneath the posterior long sac- roiliac ligament in healthy individuals [28]. The vessels accompanying nerves could be identified in 62% of the subjects, again in correlation with the above figures.
Unlu et al used CDUS to assess not only the SIJs, but also the thoracic and lumbar paravertebral regions in 39 active Ankylosing spondylitis (AS) patients and 14 con- trols [29], placing the transducer in the transverse plane over the SIJ and then successively at each spinal level. In patients with active AS, the RI values were significantly lower for all three regions compared to controls. In the subset of patients who underwent anti-TNF therapy, the values of the RI significantly increased for the lumbar region and the SIJs. This finding was confirmed later by Jiang et al, who used power Doppler (PDUS) to assess the SIJs of 55 AS patients, before and after a three-month treatment with Infliximab, and measured the RI of the vessels [30]. They found that compared to baseline, few- er SIJs exhibited blood flow signals after the treatment.
Hu et al performed a larger study by Doppler US on the SIJs and the major entheses in the lower limbs of 161 patients with AS, dividing them in an active and inactive group [31]. In the active group (Bath Ankylosing Spon- dylitis Disease Activity Index (BASDAI) value more than 4.0) 90.7% of patients had vascularization around the dorsal SIJs, compared to 38.5% in the inactive group (with BASDAI less than 4.0), RI in the active group be- ing significantly lower. There was a good agreement be- tween Doppler US of the SIJs and BASDAI values, while the agreement between Doppler US of the entheses and BASDAI was poor.
The vascularity pattern of the dorsal SIJs was inves- tigated further by Zhu et al using CDUS in 68 patients with AS and 35 controls [32]. They described the oc- currence of three different spectral Doppler patterns, the
one indicative of a venous flow being significantly more common in AS patients with active sacroiliitis. However, in most cases two or more flow patterns were simulta- neously observed making CDUS findings’ interpretation more difficult.
Finally, Ghosh et al conducted a study that combined CD and grey scale US of the sacroiliac joints in 29 pa- tients with inflammatory LBP and normal X-ray and 32 controls, comparing the findings with MRI [33]. They found that the observation of three or more flow signals on CDUS and a RI below 0.605 correlated well with the MRI proven cases of sacroiliitis. As far as sensitiv- ity and specificity were concerned, hyperechogenicity of the joint space was the most specific parameter (95%), while the most sensitive one was low RI value (94.7%).
Although the study compared nonradiographic sacroili- itis on US to MRI and proposed cutoff values for the first time, it had the limitation of the small sample size, the subjectivity of counting flow signals and the quality of the equipment.
In all the mentioned studies the RI of the vessels was found to be lower in active sacroiliitis and to in- crease with treatment. However, it is worth noticing that McGrath et al in their study of healthy volunteers found that the RI increased significantly with each caudal sa- cral level (i.e. from S1 to S3) [28]. Thus, the exact sacral level at which the measurement of RI was taken might be of importance, while none of the studies above provided such data.
US and the soft tissues in the lumbar and sacral regions
The thoracolumbar fascia
The thoracolumbar fascia is especially strong in the lower lumbar region, where it envelopes the muscles of the back and is continuous with the gluteal fascia [34]. Its posterior layer has a composite structure and its altera- tion in patients with chronic LBP was studied sonograph- ically by Langevin et al [35]. They evaluated 60 patients with LBP and 47 controls in a standardized manner, plac- ing a 10 MHz linear transducer in the longitudinal plane on a point 2 cm lateral to the midpoint at L2-L3 level.
They measured the thickness and echogenicity of the
“perimuscular zone”. This zone was defined as an echo- genic layered structure located closest to the muscles’
dorsal surface and separated from the nearest, more su- perficial echogenic layer, by more than 2 mm. In subjects with chronic LBP this zone was found to be 25% thicker and with greater echogenicity, compared to controls af- ter adjusting for BMI. However, they could not explain if that was the cause or the result of the chronic LBP.
Moreover, the structure under investigation in this study was not well anatomically defined, as the “perimuscular zone” at L2-L3 level would have included tendinous or aponeurotic parts from different muscle groups, as well as perifascial fat tissue and the fascia itself.
The authors partly addressed this issue in a later study [36] where they divided the “perimuscular zone” in two hyperechoic layers made of fibers originating from dif- ferent muscles and separated by an echolucent layer.
They used ultrasound elastography to evaluate tissue displacement during passive trunk motions on an articu- lated couch and calculated the shear strain stress. They found that the shear strain of the thoracolumbar fascia was about 20% less in subjects with chronic (more than 12 months) LBP than in controls. Also, overall males had significantly lower shear strain than females, and only in males there was a correlation between the thoracolumbar fascia shear strain and the thickness and echogenicity of the “perimuscular zone”. Mean thicknesses of the “peri- muscular zone” were 0.49 mm and 0.41 mm respectively for the male and female patients with LBP and 0.37 mm and 0.41 mm for the male and female control subjects.
The muscles of the lower back and pelvis
There is a considerable number of papers concerning the so-called rehabilitative ultrasound imaging (RUSI), in which US is used to measure the size (and sometimes to evaluate the structure and contractions) of the paraverte- bral lumbar muscles [37]. As this technique is not exactly diagnostic and has been extensively reviewed elsewhere [38], it will be mentioned here only briefly. In RUSI, pa- tients are examined in prone position with the transducer placed in a transverse plane over the spinous process and moved laterally to each paravertebral side [39]. Both low frequency linear and curvilinear transducers could be used [40]. The muscle dimensions, shape, and size were taken into consideration and the cross-sectional area of the paraspinal muscles was calculated. The shape and the size of the muscles were similar in healthy individuals of both sexes, and the between-side differences were found to be below 10% [37]. The extent of the lumbar muscles’
atrophy associated with LBP, side to side differences, as well as the muscle group predominantly affected could be well studied by RUSI [37].
The Piriformis syndrome is among the etiologies that are less frequent in LBP, thought to be caused by the thickening and contracture of the piriformis muscle [41].
Our group performed a pilot study on the sonographic features of the Piriformis syndrome using a 7.5-10 MHz linear transducer placed longitudinally to the axis of the muscle [42]. We found out a significant between-side dif- ference in the piriformis muscle’s thickness in patients with pain, compared to the control subjects where there
was a marked symmetry between the left and the right muscles.
The ligaments of the lower back and pelvis
The posterior sacroiliac ligament (PSL) might be among the LBP generators particularly in pregnancy re- lated posterior pelvic pain [43]. First, Moore et al did an anatomical US study of PSL’s long part in cadavers and in healthy young women [44]. Subjects were examined in lateral decubital position using linear 7.5-10 MHz transducer. The ligament was continuously visualized as a hyperechoic fibrillar structure between the poste- rior iliac spine and the lateral aspect of the third sacral transverse process. Its mean length was 37.9±2.4 mm, its mean thickness was 1.57±0.38 mm, and the median an- gle to the posterior superior iliac spine: 18.5°. There was also a good overall agreement between the PSL length and the thickness in cadavers measured by US and at the dissection.
In 2011, Le Goff et al performed another ultrasound study of the PSL in 20 middle-aged volunteers of both sexes using 5-12 MHz transducer [45]. They separately assessed the short and the long portion of the ligament.
The median length of the long part was 34.2 mm on the left and 35.6 mm on the right. The ligament was assessed in prone position and was visualized as a fibrillar struc- ture between the inferior part of the posterior superior iliac spine (PSIS) and the lateral side of the third sacral vertebra. Assessment of the short portion of the PSL re- quired more transverse positioning of the probe directly over the sacroiliac joint. It was seen as a fibrillar structure attached to the posterior tuberosity of the ilium and to the sacrum covered by the multifidus muscles. Its me- dian length was 23.1 mm and 21.7 mm on the left and right respectively. Dorsal sacral nerve rami normally pass through the space beneath the PSL [29]. As their en- trapment in this space was hypothesized to be among the causes for LBP [45], the authors also measured the aver- age area between the ligament and the underlying sacrum (1.19 cm2 on the right, 1.33 cm2 on the left).
Another important ligament of the lower back, namely the iliolumbar ligament (ILL), was studied by Wolenski at al [46]. The authors measured the ligament’s thickness in 41 patients and 43 control subjects by US at a point 0.5 cm above the iliac crest, with the transducer placed in the oblique plane. The median thickness was 3.4 mm in patients’ group and 2.2 mm in controls’ group and the difference was found to be significant. However, as in the above-mentioned studies on the thoracolumbar fascia, it remained controversial whether the increased thickness of the ligament was a cause or an effect of the LBP.
Thus, the main clinical impact of these studies is that they provide good normative data for possible future US
assessment of pathologies in the PSL and ILL in patients with LBP.
The posterior ligament complex (PLC) includes the supraspinous and interspinous ligaments, the ligamen- tum flavum, and the facet joint capsules. In a recent re- view Alcalá-Cerra et al evaluated the results of four stud- ies on the diagnostic capability of US to detect injuries of these structures in concomitant vertebral fractures [47].
The following US features were considered as pathologi- cal: the presence of any disruption of the first echogenic layer; the disruption of the continuous hyperechoic line between spinous processes; the detection of hypoechoic cysts (hematoma); the inhomogeneous arrangement of the ligaments’ and muscles’ fibers; the identification of avulsed bony fragments. The estimation of the diagnostic accuracy of US using MRI as a gold standard yielded the following results: sensitivity – 90%; specificity – 79%;
positive likelihood ratio – 3.63; negative likelihood ratio – 0.18; and diagnostic odds ratio – 24.24. The authors concluded that US is a highly accurate imaging modality in PLC injuries. However, in most of the studies only the interspinous and supraspinous ligaments were assessed, so the good diagnostic performance of US should be considered only for injuries of these structures and not for lesions of the ligamentum flavum and the facet joint capsules [47].
Iliac crest pain syndrome
Enthesopathies/tendinopathies cause a considerable number of regional pain syndromes such as Great Tro- chanter Pain Syndrome or Tennis Elbow [48], but their role as pain generators in LBP has not been established.
Our group evaluated by US the iliac entheses of the erec- tor spine (ES) muscle in 60 LBP patients and 50 con- trols [49]. All patients had clinical signs indicative of LBP caused by a unilateral Iliac Crest Pain Syndrome (ICPS) – a regional pain syndrome described by Collee et al [50]. ICPS is a frequent cause for LBP, its etiology remains obscure and its location coincides with the ES attachment site [5]. In our study, we used linear 10-12 MHz transducer placed in the oblique plane parallel and perpendicular to the terminal ES tendon. The entheses at the painful side were significantly thicker when com- pared to both the nonpainful entheses in the patients, and to the same structures in the control subjects. Moreover, painful entheses had significantly more elementary so- nopathological lesions (as defined by Terslev et al [51]):
hypoechogenicity, altered fibrillar structure, cortical ir- regularities, and hyperechoic foci than the entheses in both control groups. Our team concluded from this pilot study that enthesopathy of ES could be a discrete cause for LBP (and the underlying lesion of ICPS), identifiable by US.
Conclusions
LBP is clearly a complex, multifactorial condition and identifying its most probable cause in any patient requires a vast armamentarium of imaging and clini- cal methods. At present the number of studies on the diagnostic capability of US in LBP is relatively low, compared to the conventional modalities, and does not allow the drawing of definite conclusions. Nevertheless, the numerous advantages of US such as the low cost, the lack of radiation exposure, the excellent visualiza- tion of the soft tissues, makes it more likely that US will be increasingly used by physicians who will exam- ine or treat LBP patients. For this reason, more studies are required, and the American Institute of Ultrasound in Medicine (AIUM) “urges investigators to perform properly designed research projects to evaluate the ef- ficacy of diagnostic spinal ultrasound examinations”
[52]. At the same time, AIUM states that currently the use of spinal/paraspinal US in adults for diagnostic or screening evaluations should be considered only inves- tigational.
Conflict of interest: none
References
1. Borg-Stein J, Wilkins AN. Soft tissues determinants of low back pain. Curr Pain Headache Rep 2006;10:339-344.
2. Patel ND, Broderick DF, Burns J et al. ACR Appropriateness Criteria Low Back Pain. J Am Coll Radiol 2016;13:1069- 1078.
3. Naredo E. Ultrasound in rheumatology: two decades of rapid development and evolving implementation. Med Ul- trason 2015;17:3-4.
4. Ruta S, Reginato AM, Pineda C, Gutierrez M. General ap- plications of ultrasound in rheumatology: why we need it in our daily practice. J Clin Rheumatol 2015;21:133-143.
5. Bogduk N. Clinical anatomy of the lumbar spine and sa- crum. Elsevier, 2005.
6. Porter RW, Wicks M, Ottewell D. Measurement of the spi- nal canal by diagnostic ultrasound. J Bone Joint Surg Br 1978;60-B:481-484.
7. Porter RW, Hibbert C, Wellman P. Backache and the lumber spinal canal. Spine (Phila Pa 1976) 1980;5:99-105.
8. Anderson DJ, Adcock DF, Chovil AC, Farrell JJ. Ultra- sound lumbar canal measurement in hospital employees with back pain. Br J Ind Med 1988;45:552-555.
9. Macdonald EB, Porter R, Hibbert C, Hart J. The relation- ship between spinal canal diameter and back pain in coal miners. Ultrasonic measurement as a screening test? J Oc- cup Med 1984;26:23-28.
10. Macdonald EB. Ultrasonic measurement of the lumbar spi- nal canal. J Occup Med 1984;26:412.
11. Howie DW, Chatterton BE, Hone MR. Failure of ultra- sound in the investigation of sciatica. J Bone Joint Surg Br 1983;65:144-147.
12. Tervonen O, Koivukangas J. Transabdominal ultrasound measurement of the lumbar spinal canal. Its value for evaluation of lumbar spinal stenosis. Spine (Phila Pa 1976) 1989;14:232-235.
13. Porter RW, Bewley B. A ten-year prospective study of ver- tebral canal size as a predictor of back pain. Spine (Phila Pa 1976) 1994;19:173-175.
14. Merx JL, Thijssen HO, Meyer E, Chung RW. Accuracy of ultrasonic evaluation of lumbar intervertebral discs by an anterior approach. Neuroradiology 1989;31:386-390.
15. Kamei K, Hanai K, Matsui N. Ultrasonic level diag- nosis of lumbar disc herniation. Spine (Phila Pa 1976) 1990;15):1170-1174.
16. Tervonen O, Lähde S, Vanharanta H. Ultrasound diagno- sis of lumbar disc degeneration. Comparison with com- puted tomography/discography. Spine (Phila Pa 1976) 1991;16:951-954.
17. Berth A, Mahlfeld K, Merk HR. Transabdominal ultra- sonography in the diagnosis of lumbar disc herniation. Ul- traschall Med 2003;24:383-387.
18. Buyruk HM, Snijders CJ, Vleeming A, Laméris JS, Hol- land WP, Stam HJ. The measurements of sacroiliac joint stiffness with colour Doppler imaging: a study on healthy subjects. Eur J Radiol 1995;21:117-1121.
19. Buyruk HM, Stam HJ, Snijders CJ, Laméris JS, Holland WP, Stijnen TH. Measurement of sacroiliac joint stiffness in peripartum pelvic pain patients with Doppler imaging of vibrations (DIV). Eur J Obstet Gynecol Reprod Biol 1999;83:159-163.
20. Damen L, Buyruk HM, Guler-Uysal F, Lotgering FK, Sni- jders CJ, Stam HJ. Pelvic pain during pregnancy is associ- ated with asymmetric laxity of the sacroiliac joints. Acta Obstet Gynecol Scand 2001;80:1019-1024.
21. Damen L, Buyruk HM, Guler-Uysal F, et al. The prog- nostic value of asymmetric laxity of the sacroiliac joints in pregnancy-related pelvic pain. Spine (Phila Pa 1976) 2002;27:2820-2824.
22. de Groot M, Spoor CW, Snijders CJ. Critical notes on the technique of Doppler imaging of vibrations (DIV). Ultra- sound Med Biol 2004;30:363-367.
23. Spadaro A, Iagnocco A, Baccano G, Ceccarelli F, Sa- batini E, Valesini G. Sonographic-detected joint effusion compared with physical examination in the assessment of sacroiliac joints in spondyloarthritis. Ann Rheum Dis 2009;68:1559-1563.
24. Klauser A, Halpern EJ, Frauscher F, et al. Inflammatory low back pain: high negative predictive value of contrast- enhanced color Doppler ultrasound in the detection of inflamed sacroiliac joints. Arthritis Rheum 2005;53:440- 25. 444.Klauser AS, De Zordo T, Bellmann-Weiler R, et al. Fea- sibility of second-generation ultrasound contrast media in the detection of active sacroiliitis. Arthritis Rheum 2009;61:909-916.
26. Arslan H, Sakarya ME, Adak B, Unal O, Sayarlioglu M.
Duplex and color Doppler sonographic findings in active sacroiliitis. AJR Am J Roentgenol 1999;173:677-680.
27. Pekkafahli MZ, Kiralp MZ, Basekim CC, et al. Sacroiliac joint injections performed with sonographic guidance. J Ul- trasound Med 2003;22:553-559.
28. McGrath CM, Jeffery R, Stringer MD. The dorsal sacral rami and branches: Sonographic visualization of their vas- cular signature. Int J Osteopath Med 2011;15:3-12.
29. Unlu E, Pamuk ON, Cakir N. Color and duplex Doppler sonography to detect sacroiliitis and spinal inflamma- tion in ankylosing spondylitis: can this method reveal re- sponse to anti-tumor necrosis factor therapy? J Rheumatol 2007;34:110-116.
30. Jiang Y, Chen L, Zhu J, et al. Power Doppler ultrasonogra- phy in the evaluation of infliximab treatment for sacroili- itis in patients with ankylosing spondylitis. Rheumatol Int 2013;33:2025-2029.
31. Hu Y, Zhu J, Xue Q, Wang N, Hu B. Scanning of the sac- roiliac joint and entheses by color Doppler ultrasonogra- phy in patients with ankylosing spondylitis. J Rheumatol 2011;38:1651-1655.
32. Zhu J, Xing C, Jiang Y, Hu Y, Hu B, Wang N. Evaluation of complex appearance in vascularity of sacroiliac joint in ankylosing spondylitis by color Doppler ultrasonography.
Rheumatol Int 2012;32:69-72.
33. Ghosh A, Mondal S, Sinha D, Nag A, Chakraborty S. Ultra- sonography as a useful modality for documenting sacroili- itis in radiographically negative inflammatory back pain: a comparative evaluation with MRI. Rheumatology (Oxford) 2014;53:2030-2034.
34. Benjamin M. The fascia of the limbs and back – a review. J Anat 2009;214:1-18.
35. Langevin HM, Stevens-Tuttle D, Fox JR, et al. Ultrasound evidence of altered lumbar connective tissue structure in human subjects with chronic low back pain. BMC Muscu- loskelet Disord 2009;10:151.
36. Langevin HM, Fox JR, Koptiuch C, et al. Reduced thora- columbar fascia shear strain in human chronic low back pain. BMC Musculoskelet Disord 2011;12:203.
37. Stokes M, Hides J, Elliott J, Kiesel K, Hodges P. Rehabili- tative ultrasound imaging of the posterior paraspinal mus- cles. J Orthop Sports Phys Ther 2007;37:581-595.
38. Heidari P, Farahbakhsh F, Rostami M, Noormohammad- pour P, Kordi R. The role of ultrasound in diagnosis of the causes of low back pain: a review of the literature. Asian J Sports Med 2015;6:e23803.
39. Hides JA, Cooper DH, Stokes MJ. Diagnostic ultrasound imaging for measurement of the lumbar multifidus mus-
cle in normal young adults. Physiother Theory and Pract 1992;8:19-26.
40. Whittaker JL, Teyhen DS, Elliott JM, et al. Rehabilitative ultrasound imaging: understanding technology and its ap- plications. J Orthop Sports Phys Ther 2007;37:434-449.
41. Hopayian K, Song F, Riera R, Sambandan S. The clinical features of the piriformis syndrome: a systematic review.
Eur Spine J 2010;19:2095-2109.
42. Todorov P, Nestorva R, Batalov A. Sonographic features of the piriformis syndrome. Ann Rheum Dis 2014;73(Suppl 2):296-297.
43. Vleeming A, Mooney V, Stoeckart R. Movement, stability and lumbopelvic pain. Integration of research and therapy.
Churchill Livingstone Elsevier, 2007.
44. Moore AE, Jeffery R, Gray A, Stringer MD. An anatomical ultrasound study of the long posterior sacroiliac ligament.
Clin Anat 2010;23:971-977.
45. Le Goff B, Berthelot J-M, Maugars Y. Ultrasound assess- ment of the posterior sacroiliac ligaments. Clin Exp Rheu- matol 2011;29:1014-1017.
46. Wolenski L, Bizzi E, Galletti S, Galletti R, Marchi A. Di- agnostic classification of ileum-lumbar ligamentite. Ann Rheum Dis 2015;74(Suppl 2):1255.
47. Alcalá-Cerra G, Paternina-Caicedo Á, Gutiérrez-Paternina J, Moscote-Salazar L, Alvis-Miranda H, Sabogal-Barrios R. Diagnostic accuracy of ultrasound for detecting posteri- or ligamentous complex injuries of the thoracic and lumbar spine: A systematic review and meta-analysis. J Craniveret- bral Junction Spine 2013;4:25-31.
48. Benjamin M, Toumi H, Ralphs JR, Bydder G, Best TM, Milz S. Where tendons and ligaments meet bone: attach- ment sites (‘entheses’) in relation to exercise and/or me- chanical load. J Anat 2006;208:471-490.
49. Todorov P, Nestorova R, Batalov A. Use of musculoskel- etal ultrasound to detect painful enthesopathies in low back pain. Proceedings of the 9th Intradisciplinary World Congress on Low Back and Pelvic Girdle Pain, Singapore, 2016:220-223.
50. Collee G, Dijkmans BA, Vandenbrucke JP, Cats A. Iliac crest pain syndrome in low back pain: frequency and fea- tures. J Rheumatol 1991;18:1064-1067.
51. Terslev L, Naredo E, Iagnocco A, et al. Defining enthesitis in spondyloarthritis by ultrasound: results of a Delphi pro- cess and of a reliability reading exercise. Arthritis Care Res (Hoboken) 2014;66:741-748.
52. The American Institute for Ultrasound in Medicine. Offi- cial Statement: Nonoperative spinal/paraspinal ultrasound in adults. April 2, 2014, http://www.aium.org/officialState- ments/18, accessed April 24, 2017.
