Secondary Logo

Journal Logo


Ultrasound versus fluoroscopic-guided epidural steroid injections in patients with degenerative spinal diseases

A randomised study

Evansa, Irina; Logina, Inara; Vanags, Indulis; Borgeat, Alain

Author Information
European Journal of Anaesthesiology (EJA): April 2015 - Volume 32 - Issue 4 - p 262-268
doi: 10.1097/EJA.0000000000000103
  • Free



Epidural steroid injections have been used for treatment of lower back pain and radiated pain in the legs for many years. The current Practice Guidelines for Chronic Pain Management suggest that transforaminal epidural injections should be performed under appropriate image guidance to confirm correct needle position and to avoid dramatic complications; image guidance is also considered for interlaminar epidural injections.1

Fluoroscopic-guided interlaminar epidural steroid injections with contrast epidurography are fast to perform and increase accuracy of needle placement and target delivery of the injected medications to the site of disease.2 Fluoroscopy is also a helpful image guidance tool for patients whose anatomical features pose particular challenges (such as degenerative changes, obesity, malformation).3 The most reasonable exception would be initial ‘blinded’ interlaminar lumbar epidural steroid injections in younger, nonobese, nonoperated patients. However, in a survey of the technical aspects of epidural steroid injections, Cluff et al.4 found that only 30% of lumbar epidurals were performed under fluoroscopy in academic institutions, whereas in private practice the rate was 77%. Possible explanations for this difference include lack of fluoroscopic-guidance resources and the existence of patients for whom this method might be contraindicated due to radiation exposure or allergy to contrast agent.

Ultrasound has recently been used to facilitate lumbar spinal and epidural anaesthesia. Prepuncture ultrasound spine examinations are fast to perform and allow identification of the level of the spine at which the puncture should be performed, and the optimal puncture site.5 The main limitation of this technique is that the operator remains ‘blind’ during the actual needle insertion and may encounter difficulty in directing the needle through the interlaminar space. Nonetheless, the ultrasound technique could be an attractive alternative to fluoroscopy for performing interlaminar lumbar epidural steroid injections if the patient is allergic to iodine or fluoroscopy is unavailable.

The current study was designed to evaluate procedure accuracy, technical difficulty, analgesic and functional improvements of ultrasound-assisted interlaminar epidural steroid injections compared with fluoroscopically guided injection for patients with lower back pain with radiculopathy secondary to degenerative spine diseases through a prospective randomised study.

Materials and methods

Ethical approval for this study [Protocol E-9(2)] was provided by the Ethical Committee of Stradins University, Riga, Latvia (Chairperson Prof O. Bruvers) on 14 January 2010. The study took place at the Department of Anaesthesiology of Riga Stradins University Hospital, Latvia, from January 2010 to September 2012. Written informed consent was obtained from all participants’ during consultation.

We prospectively enrolled adult patients with a history of chronic lower back pain and unilateral or bilateral lumbosacral radiculopathy due to degenerative spine diseases with previous failed conservative treatment. All patients were booked for three injections at the first consultation. Inclusion criteria were patients with a history of chronic function limiting lower back and extremity pain assessed as at least six on a scale of 0 to 10, and pain duration of more than 3 months. Patients with intervertebral disc degeneration, degenerative spondylolisthesis, disc herniation or spinal canal stenosis, who were competent to understand the study protocol and provided voluntary, written informed consent, and able to participate in outcome measurements were enrolled. Exclusion criteria included patients with history of previous lumbar spinal surgery, lumbar epidural steroid injections in the past year, allergy to any anaesthetic or corticosteroid, allergy to contrast dye, concurrent use of systemic steroid medications, opioid addiction and coagulopathy.

Patients were assigned according to a computerised randomisation list to either a fluoroscopy-guided group or an ultrasound-guided epidural steroid injection group. Codes were kept in sealed envelopes and the group assignment was revealed to the investigator on the day of the booked procedure.

All patients were placed in the prone position on a radiology table with a pillow under the hips, under appropriate monitoring with intravenous access and with sedation as required. All patients received interlaminar median or paramedian epidural steroid injection either between L5 and S1 or at a higher level based on clinical and radiological manifestations. Right versus left paramedian or median puncture sides were determined according to the clinical examination and the results of diagnostic imaging studies. If the initial injection resulted in significant pain relief, the second and third injections were performed using the same technique as the first. A single investigator (I.E.), experienced in both ultrasound and fluoroscopy techniques, performed all injections. Although patients and investigators were aware of the group allocation, the outcome assessors and data analysts were blinded.

Fluoroscopy-guided technique

The area for puncture was prepared using an iodine-based antiseptic solution. The physician used intermittent fluoroscopy to localise the median or paramedian interlaminar space and then performed a local anaesthetic injection to infiltrate the injection point. Next, an 18-gauge Tuohy-type epidural needle (B. Braun, Melsungen, AG, Germany) was inserted in the anaesthetised injection point area. The needle was advanced directly perpendicular to the skin in a posterior-anterior direction and the epidural space was identified using the loss of resistance technique. Once loss of resistance was obtained, 1 ml of contrast medium (Omnipaque 300; GE Healthcare, Oslo, Norway) was injected to confirm epidural flow and to rule out intravascular, intrathecal or soft-tissue infiltration. Fluoroscopy was used for interspace visualisation, at each needle adjustment according to the protocol, and for the contrast injection phase. Images were obtained using an anterior-posterior projection to confirm that the epidural space had been entered, and in case of difficulty, a lateral projection was used. Then, corticosteroid (methylprednisolone acetate 80 mg), along with 4 ml of 1% lidocaine, was injected into the epidural space (total volume 5 ml).

Ultrasound-guided technique

Ultrasound imaging of the lumbar spine was performed prior to epidural needle insertion to visualise the interlaminar space and mark the injection site. We used a BK Medical Flex Focus ultrasonograph equipped with a 5.0 MHz curved array probe (Herlev, Denmark). The nonsterile probe was oriented longitudinally and a parasagittal oblique view was performed to mark the paramedian injection approach and interlaminar space by counting upwards from the sacrum line. The transducer was angled towards the midline to visualise the epidural space, spinal canal and the posterior aspect of the vertebral body. The probe was then rotated to obtain a median transverse view of the lumbar spine, and the same structures were identified and marked for the midline injection approach. When we found the acoustic window showing the best view of the epidural space, it was moved to the middle of the screen and the skin was marked laterally at the midpoints of each end of the probe. The intersection of four marked lines was interpreted as the puncture site. The angle of probe providing the best image of the interspace structures was visually noted. The distance from the skin to the epidural space was measured using a built-in caliper.5 The ultrasound transducer was then removed, and the gel cleaned from the skin. The area for puncture was prepared using an iodine-based antiseptic solution and a local anaesthetic injection performed to infiltrate injection area. An 18-gauge Tuohy-type epidural needle was advanced through the ultrasound-determined insertion point at an angle reproducing the direction of the ultrasound beam and depth of the epidural space during the prescan technique. After the needle reached the epidural space and loss of resistance was felt, an aspiration test was performed to rule out intravascular or intrathecal injection and 1 ml of the contrast medium was injected. The anterior-posterior images were obtained to confirm the distribution of the injected contrast medium with C-arm fluoroscopy. The same amount of corticosteroid as for the fluoroscopic group was injected after confirmation of the proper shadowed contrast of the epidural. Speed of the corticosteroid injection was approximately 1 ml s−1.

The primary outcome measures were total procedure time (the time to establish landmarks and perform the epidural steroid injection); the number of needle insertion attempts required for successful epidural puncture with or without the necessity of puncturing the neighbouring intervertebral space; and the number of needle passes (each redirection of the needle after skin puncture, preceded by complete withdrawal from the patient's skin) required for successful epidural puncture. The secondary outcome measures were visual analogue scale (VAS) for pain (0 to 10) where 0 is no pain and 10 is the worst pain imaginable, and the Oswestry Low Back Pain Disability Index (ODI), a multipurpose questionnaire of 10 items used to assess functional health status and well being of adults. Each item is scored from zero to five, with higher values representing greater disability. The total score is multiplied by two and expressed as a percentage,6 and complications recorded during the procedure and during the first 24 h afterwards. Positioning the patient, starting the fluoroscopy-guided and ultrasound-guided machine, entering patient data into the devices, local anaesthetic injections and epidural space confirmation under fluoroscopic guidance in the US group have not been considered. The duration of follow-up of all patients was 1 and 3 months after the last procedure.

Statistical analysis

In a pilot study of 30 patients having fluoroscopic-controlled epidural steroid injections for lower back and extremity pain, the mean ± SD procedure time was 345 ± 150 s. The sample size was based on a two-sided test with an alpha of 0.05 and a power of 80%. To obtain a 100 s reduction, a minimum of 35 patients per group was required.7 To increase the power, we decided to include 60 patients per group. Statistical analysis was performed using SPSS ver. 19.0 (SPSS Inc., Chicago, Illinois, USA). Discrete variables side-effects were analysed with the Chi-square test. Nonnormally distributed data were compared using the Mann–Whitney test. Results are expressed as mean ± SD, if not otherwise specified. For all determinations, a P value of less than 0.05 was considered to be significant. At each time point, VAS and ODI were compared by repeated-measure analysis of variance (ANOVA).


One-hundred and thirty-seven patients were assessed for eligibility and 120 patients were randomised (aged 47 to 89 years; weight 50 to 120 kg; height 152 to 190 cm). We were unable to complete the study protocol in eight patients because of acute disc herniation (n = 3), osteoporotic vertebral fracture (n = 2) or other injection leading to protocol violation (n = 3). Fifty-six patients in each group completed the study and their data were analysed (Fig. 1). Lumbar interlaminar procedures were performed in 90% of cases at the L5/S1 level and 10% of cases at L4/5 level and the median [interquartile range, IQR] interval between the injections was 3 [2 to 5] weeks. All patients received three consecutive therapeutic injections.

Fig. 1
Fig. 1:
Patient flow diagram.

Basic demographic characteristics, persistence of symptoms and back pain distribution and the types of morphological changes of the lumbar spine are summarised in Table 1.

Table 1
Table 1:
Patient baseline characteristics

The time required to perform the procedure was the same in the two groups, as were the number of needle insertion attempts or the number of needle passes required to identify the epidural space during the procedure (Table 2). Successful epidural puncture was achieved during the first needle insertion attempt in 73% of patients in the fluoroscopy-guided group and 80% in the ultrasound-guided group (P > 0.05); the corresponding values for successful epidural puncture at the first needle pass were 18 and 27% (P > 0.05).

Table 2
Table 2:
Procedural data

Table 3 illustrates that, overall, the fluoroscopy-guided and ultrasound-guided groups produced similar clinically significant improvements in pain and functional status (ANOVA, P < 0.01). The incidences of side-effects such as dizziness, pain at injection site and face flushing were similar in the two groups (eight in the fluoroscopy and seven in the ultrasound groups). There was a single case of intrathecal contrast injection in the fluoroscopy group, in which the patient remained asymptomatic; in all other cases, epidural spread of contrast was confirmed.

Table 3
Table 3:
Changes in pain (VAS scale) and Oswestry disability index after three epidural steroid injections


This prospective study demonstrated that both ultrasound-assisted and fluoroscopy-guided interlaminar epidural steroid injections are suitable and provide similar analgesic and functional improvements in patients with degenerative spine diseases.

Degenerative spine diseases, such as intervertebral disc degeneration, spinal stenosis and degenerative spondylolisthesis, are common conditions that cause chronic lower back pain and pain referred to the legs.8 Lumbar injections are commonly used in the management of lower back pain with or without radiculopathy as well as for degenerative spine diseases including lumbar spinal stenosis. Currently, guidelines for lumbar steroid injection recommend introduction of the needle under fluoroscopic guidance; however, visualisation is strongly recommended for transforaminal epidural steroid injections to minimise the risk of injury of the artery of Adamkiewicz or to nerve roots.9 Direct fluoroscopic needle visualisation reduces procedure time, helps the practitioner choose the correct level for injection and reduces the number of needle redirection attempts.10 However, if fluoroscopy is not available, ultrasound is considered to be an alternative method that is superior to ‘blind’ injection.

In a prospective study of 50 patients with radiographic control, after ultrasound prescan of the L2/3 to L4/5 intervertebral spaces, fully trained anaesthesiologists were able to determine the location of specific lumbar interspaces accurately in 70% of patients when ultrasound examination was used compared with 30% of patients based on palpation examination.11 In another study of 17 patients, ultrasonographic location of L3/4 intervertebral spaces was correct in 76% of cases compared with control MRI, with a difference of one level in the remaining 24%.12 Previous studies have shown incorrect interspaces, mostly in L2/3 and L3/4 levels, with a difference of one interspace because of difficulty counting from the sacrum line.11,12 Most epidural injections for treatment of chronic pain due to degenerative spinal changes are performed at the L5/S1 level and this is the first interspace to be identified after the sacrum line.13 Using this level as a starting point in the ultrasound group, we subsequently identified the L4/5 level, so frequency of mistakes should be lower than counting upward to the L2/3 and L3/4 levels.

Unilateral or bilateral symptoms depend on the segment and side of spine involved in the pathological process. In the fluoroscopy-guided injection group for bilateral radicular pain, we used the median approach, and for unilateral pain, we used the paramedian approach, as described in previous studies.2 The approach in the ultrasound-assisted group was transverse midline and paramedian oblique for bilateral and unilateral pain, respectively.14,15

Some previous studies have compared the time required to perform landmark, ultrasound, fluoroscopically or computed tomography (CT)-guided spinal injections with different rates of success.16,17 However, in our study population that included patients with BMI values up to 32 kg m-2 and different spinal conditions, the time to perform the procedure did not differ between the groups. Grau et al.18 investigated the total number of needle insertion attempts required for successful neuraxial puncture after ultrasound examination of the spine in young patients with abnormal spinal anatomy or previous spinal surgery. They found that the mean ± SD number of needle redirections was 1.5 ± 0.9, but older, obese adults required a greater number of needle redirections.17 According to our data, older, obese patients with presumed difficulty due to degenerative spinal changes did not need more redirections of the needle in partly blind, ultrasound-assisted epidural injections than the real-time fluoroscopic group injections.

Chin et al.17 reported successful first needle insertion attempts using an ultrasound-assisted technique in 65% and first needle pass in 27% of cases in a study on patients with difficult surface anatomy landmarks and BMI more than 35 kg m-2 scheduled for spinal anaesthesia. In our study, the first needle insertion attempt success rate was 73 and 80%, and the first needle pass rate was 18 and 27%, in the fluoroscopy and ultrasound-guided groups, respectively, with no significant difference between the groups.

We followed up our patients for 3 months on the basis of previous review articles, which concluded that epidural steroid injections were efficacious in chronic lower back and lower extremity pain syndromes at intermediate-term follow-up (2 weeks to 3 months); however, there is limited evidence for long-term relief.19–21 Our study demonstrates similar analgesic effectiveness and improvement of functional status in both the ultrasound and fluoroscopy epidural steroid injection groups, lasting for 1 month. We observed that due to degenerative lumbar events resulting in spread of inflammation and multiple pain generators in the present study, only half of the patients benefited from injections at 3 months. This finding was also reported by Briggs et al.3 who observed the effect of injection treatment on pain relief among adults (60 years and over) diagnosed with degenerative lumbar spinal stenosis. In general, pain scores in their study improved substantially 1 month after treatment. Three months after treatment, the improvement was also seen, but was not as significant as after 1 month. The authors concluded that injected pain medications did not provide a lasting effect, but were still providing some pain relief after 3 months. However, according to our data, the decrease in pain and Oswestry disability scores was not different between groups at any time.

All types of epidural steroid injections can be associated with complications and adverse events. In our study, we had no serious complications, but we did observe mild side-effects that are in accordance with previous results.22,23

Our study has some limitations. It was not a double-blinded control study. It is difficult to conduct such a study with modalities such as ultrasound or fluoroscopy. The same physician conducted the treatment procedures and our results may reflect the experience of that practitioner, thus limiting the generalisability of our findings. The presence of degenerative changes of the spine in our population reduces steroid efficacy due to multiple pain generators. Additional studies may be required with reduced limitations in the future.

In conclusion, the results of this study suggest that ultrasound-assisted epidural steroid injections are feasible with similar analgesic effect and take about the same time to perform as fluoroscopy-guided injections. Ultrasound may not replace fluoroscopy-controlled epidural steroid injections, but could be an acceptable method when fluoroscopy is not available.

Acknowledgements relating to this article

Assistance with the study: the authors would like to thank the nurses of the Department of Anaesthesiology for their support.

Financial support and sponsorship: none.

Conflicts of interest: none.

Presentation: preliminary data presented at the ESRA Congress, Glasgow, 2013.


1. American Society of Anesthesiologists Task Force on Chronic Pain Management and the American Society of Regional Anesthesia and Pain Medicine. Practice guidelines for chronic pain management. Anesthesiology 2010; 112:1–24
2. Candido KD, Raghavendra MS, Chinthagada M, et al. A prospective evaluation of iodinated contrast flow patterns with fluoroscopically guided lumbar epidural steroid injections: the lateral parasagittal interlaminar epidural approach versus the transforaminal epidural approach. Anesth Analg 2008; 106:638–644.
3. Briggs VG, Li W, Kaplan MS, et al. Injection treatment and back pain associated with degenerative lumbar spinal stenosis in older adults. Pain Phys 2010; 13:347–355.
4. Cluff R, Mehio AK, Cohen SP, et al. The technical aspects of epidural steroid injections: a national survey. Anesth Analg 2002; 95:403–408.
5. Chin KJ, Karmakar MK, Peng P. Ultrasonography of the adult thoracic and lumbar spine for central neuraxial blockade. Anesthesiology 2011; 114:1459–1485.
6. Fairbank JC, Couper J, Davies JB, O’Brien JP. The Oswestry low back pain disability questionnaire. Physiotherapy 1980; 66:271–273.
7. Browner WS, Newman TB, Cummings SR, Hulley SB. Hulley SB, Cummings SR, Browner WS, Grady D, Hearst N, Newman TB. Estimating sample size and power. Designing clinical research: an epidemiologic approach. Philadelphia: Lippincott, Williams & Wilkins; 2001. 65–84.
8. Porter RW. Spinal stenosis and neurogenic claudication. Spine 1996; 21:2046–2052.
9. Murthy NS, Maus TP, Behrns CL. Intraforaminal location of the great anterior radiculomedullary artery (artery of Adamkiewicz): a retrospective review. Pain Med 2010; 11:1756–1764.
10. Silbergleit R, Mehta BA, Sanders WP, Talati SJ. Imaging-guided injection techniques with fluoroscopy and CT for spinal pain management. Radiographics 2001; 21:927–939.
11. Furness G, Reilly MP, Kuchi S. An evaluation of ultrasound imaging for identification of lumbar intervertebral level. Anesthesia 2002; 57:266–283.
12. Watson MJ, Evans S, Thorp JM. Could ultrasonography be used by an anaesthetist to identy a specified lumbar interspace before spinal anaesthesia? Br J Anaesth 2003; 90:509–511.
13. Smith CC, Booker T, Schaufele MK, Weiss P. Interlaminar versus transforaminal epidural steroid injections for the treatment of symptomatic lumbar spinal stenosis. Pain Med 2010; 11:1511–1515.
14. Karmakar MK, Li X, Kwok WH, et al. Sonoanatomy relevant for ultrasound-guided central neuraxial blocks via the paramedian approach in the lumbar region. Br J Radiol 2012; 85:262–269.
15. Chin KJ, Ramlogan R, Arzola C, et al. The utility of ultrasound imaging in predicting ease of performance of spinal anesthesia in an orthopedic patient population. Reg Anesth Pain Med 2013; 38:34–38.
16. Obernauer J, Galiano K, Gruber H, et al. Ultrasound-guided versus computed tomography-controlled facet joint injections in the middle and lower cervical spine: a prospective randomized clinical trial. Med Ultrason 2013; 15:10–15.
17. Chin KJ, Perlas A, Chan V, et al. Ultrasound imaging facilitates spinal anesthesia in adults with difficult surface anatomic landmarks. Anesthesiology 2011; 115:94–101.
18. Grau T, Leipold RW, Conradi R, Martin E. Ultrasound control for presumed difficult epidural puncture. Acta Anaesthesiol Scand 2001; 45:766–771.
19. Parr AT, Diwan S, Abdi S. Lumbar interlaminar epidural injections in managing chronic low back and lower extremity pain: a systematic review. Pain Phys 2009; 12:163–188.
20. Abdi S, Datta S, Trescot AM, et al. Epidural steroids in the management of chronic spinal pain: a systematic review. Pain Phys 2007; 10:185–212.
21. Benyamin RM, Manchikanti L, Parr AT, et al. The effectiveness of lumbar interlaminar epidural injections in managing chronic low back and lower extremity pain. Pain Phys 2012; 12:363–404.
22. Goodman BS, Posecion LWF, Mallempati S, Bayazitoglu M. Complications and pitfalls of lumbar interlaminar and transforaminal epidural injections. Curr Rev Musculoskelet Med 2008; 1:212–222.
23. Manchikanti L, Malla Y, Wargo BW, et al. A prospective evaluation of complications of 10 000 fluoroscopically directed epidural injections. Pain Phys 2012; 15:131–140.
© 2015 European Society of Anaesthesiology