Caudal epidural steroid injection for chronic low back pain: A prospective analysis of 107 patients

BACKGROUND: Low back pain affects 80% of people worldwide at least once in a lifetime and reduces the quality of life and causes absence from work. OBJECTIVE: To evaluate the pain and functional status of patients with lumbar disc disease who received blind caudal epidural injections (CEI) for pain relief. METHODS: The records of 107 patients who had been given CEI between September 2017 and January 2018 were retrospectively analyzed. The inclusion criteria were age > 18 years, > 3-month history of low back pain, and diagnosis of lumbar disc disease by magnetic resonance imaging. The epidural injection solution consisted of 2 mL of betamethasone sodium and 8 mL saline. Follow-up examinations were conducted 3 and 6 months post-injection and the patients were evaluated using a visual analog scale (VAS) and the Oswestry Disability Index (ODI). RESULTS: The most common disc pathology was at the L4–L5 level. The VAS and ODI scores indicated significantly reduced pain at 3 and 6 months compared with the pre-injection baseline. Two patients experienced total anesthesia and paresis of the lower limbs, but recovered fully after 2 weeks. Blood was aspirated during the injection in two patients, but second-attempt injections were successful in both cases. No other complications were observed. CONCLUSION: Our results suggest that the blind method is safe for administering CEI to patients with chronic low back pain in the absence of radiological screening and results in significant pain relief with improved functional capacity.

Keywords: Low back pain; caudal injection; steroid; conservative treatment; lumbar injection; chronic back pain

Low back pain affects 80% of people worldwide at least once in a lifetime and reduces quality of life and causes absence from work [[1]]. Low back pain is defined as chronic when it persists for 12 weeks or more [[3]]. Most patients are treated conservatively, with oral anti-inflammatory agents, physical therapy, and exercise. When these methods fail, epidural injection (transforaminal, interlaminar, or caudal) is an option [[2], [4]].

Caudal epidural injections are administered to patients with lumbar spinal stenosis, lumbar spinal radiculopathy, postlaminectomy pain, and nonspecific chronic low back pain [[1]]. Most of such injections are administered under fluoroscopic or ultrasonographic guidance, but blind administration guided by anatomical landmarks is also performed [[1]]. The rates of accurate blind injection are $>$ 96% in children (mostly in the setting of surgical anesthesia) and 68–75% in adults [[1], [3], [5]]. The most common complications of epidural injection are repeated attempts, hypotension, and bleeding at the injection site [[2], [4], [6]].

We investigated the efficiency of caudal epidural injections without radiological guidance. This study retrospectively evaluated complications and evaluated pain and functional status in patients with lumbar disc disease who received blind caudal epidural injections for pain relief.

After approval was obtained by the hospital ethics review board, we screened the records of patients who had been given caudal epidural injections by an experienced physiatrist in a teaching hospital between September 2017 and January 2018. The inclusion criteria were age $>$ 18 years, $>$ 3-month history of low back pain despite conservative treatment, inflammatory markers in the normal ranges, surgical treatment not indicated, and degenerative lumbar disc disease diagnosis by magnetic resonance imaging. Pregnant women, patients with a previous lumbar injection, lumbar disc disease due to secondary reasons (neoplasia, infection, inflammation etc.), infection at the injection site, severe motor deficit, or progressive neurological deficit were excluded.

The patients were followed in the outpatient clinic or by telephone interview 3 and 6 months post-injection. Patients who did not complete both follow-up examinations and those who had undergone surgery after the injection were excluded. All patients were informed about the study and provided informed consent.

Each patient was evaluated before and immediately after the injection, as well as during follow-up, using a visual analog scale (VAS) and the Oswestry Disability Index (ODI) [[6]]. The VAS was a 10 cm line with anchor statements from left (no pain) to right (strongest pain). The patients were asked to show their pain intensity on the line [[8]]. The ODI is a self-completed questionnaire concerning the intensity of pain and various daily life activities, each followed by six statements describing different potential scenarios in the patient's life relating to the topic. The scores for all questions answered are summed to obtain the index (range 0 to 100). Zero equals no disability and 100 is the maximum disability possible.

Graph: Figure 1.Anatomical landmarks.

Each injection was administered as follows. Before the epidural injection, 1 g cefazolin was injected intramuscularly. The patient then rested for 30 min in the prone position. To avoid discomfort at the injection site, a pillow was placed under the patient's abdomen. The injection site was sterilized with povidone-iodine solution, and an equilateral triangle was drawn between the bilateral posterior superior iliac spines and sacral hiatus (Figs 1 and 2). Local anesthesia was provided at the injection site with 4 mL prilocaine diluted in 10 mL saline solution. The epidural injection solution consisted of 2 mL betamethasone sodium and 8 mL saline. A 20-gauge, 90-mm spinal needle was used to administer the epidural injection; it was inserted at a 45 ${}^{\circ }$ angle through the sacral horn. After confirming passage through the sacrococcygeal ligament by a "popping" sound, the spinal needle was gently retracted until half of its length had entered the epidural space at a 30 ${}^{\circ }$ angle [[9]].

Graph: Figure 2.Caudal epidural injection.

The absence of a vascular breach was determined by aspiration. If no blood was observed, then half of the prepared solution was infused. After the absence of vascular invasion had been confirmed by repeat aspiration, the rest of the solution was infused. The injection site was closed after sterilization with povidone-iodine. Patients were monitored for complications, such as paresis and dizziness, for 1 h after the injection.

The mean, standard deviation, median, range, frequency, and ratio values were used as descriptive statistics of the data. The distribution of the variables was assessed with the Kolmogorov-Simonov test. The Wilcoxon test was used in the analysis of dependent quantitative data. The McNemar test was used to analyze qualitative dependent data. $P$ -value $<$ 0.05 was deemed statistically significant. SPSS 22.0 was used in the analyses.

Of the 162 patients who were given epidural injections, 27 could not subsequently be reached by telephone and 12 underwent surgery after the injection. For the remaining 123 patients, follow-up examinations were performed in the outpatient clinic (Fig. 3). Of these, 16 patients did not complete the clinical follow-up. Thus, the analysis included data from 107 patients (86 females, 21 males), most of whom worked as laborers. The most commonly detected disc pathology was at the L4–L5 level (Table 1).

Table 1 Patient demographics

Graph: Figure 3.Flow chart of patient follow-up.

The VAS and ODI scores indicated significantly reduced pain at 3 and 6 months compared with the pre-injection baseline ( $p<$ 0.05; Table 2).

Table 2 VAS and ODI scores

w: Wilcoxon test.

Two patients experienced total anesthesia and paresis of the lower limbs, but recovered fully after 2 weeks. Blood was aspirated during the injection in two patients, but second-attempt injections were successful in both cases. No other complications were observed.

Epidural injections are often administered blindly because of the lack of radiological screening equipment and to reduce treatment costs [[10]]. The most commonly used blind technique is palpation of the sacral horn, followed by local anesthesia administration and confirmation of sacrococcygeal ligament penetration by the spinal needle [[10]]. The reported misinjection rate with the blind method is 25–30% [[10], [12]], with a higher risk associated with the presence of anatomical variation [[10]].

In our study, an equilateral triangle was drawn between the apex of the sacral hiatus and the bilateral posterior superior iliac spines while the patient was in the prone position. Senoglu et al. used the same landmarks, but did not report clinical outcomes, precluding comparison with our results [[10]].

The most common complications of blind injections are needle misplacement, subarachnoid puncture, and intrathecal or intravascular injection [[10], [12]]. The reported rate of incorrect injection site localization with repeated injection attempts is 11–42% [[10], [12]]. In our study, needle misplacement occurred in two patients, but the injection was not administered in either of these cases because aspirated blood indicating incorrect localization was detected. Successful placement was achieved in both cases with a second attempt.

Doo et al. compared guided caudal injection after penetration of the needle 1 cm into the sacral canal (conventional method) with injection just after penetration of the sacrococcygeal ligament, as performed in our study, but under ultrasonographic guidance [[5]]. The rates of intravascular injection were 24% with the conventional method and 0% with the sacrococcygeal method [[5]]. Manchanti et al. reported that intravascular injection occurred in 14% of the patients who received epidural injections under fluoroscopic guidance and numbness in 1% [[13]]. In our study, two patients experienced lower-extremity numbness, but recovered fully within 2 weeks.

Sing et al. compared selective nerve blockade with caudal epidural injection in patients with chronic lumbar radiculopathy, and reported similar clinical outcomes, although they found that caudal epidural injection was safer [[14]]. The clinical outcomes in that study were similar to ours, but the authors used C-arm fluoroscopy and each patient was given three injections at 3-week intervals.

A previous prospective study compared the accuracy of blind placement of a caudal epidural needle with that of radiocontrast epidurogram-guided placement. The surgical miss rate in the blind placement group was 26%. Therefore, the authors recommended radiological guidance and the use of an epidurogram for the administration of caudalepidural injections [[12]].

In a systematic review, epidural injections were found to alleviate pain and improve functional capacity [[15]]. In our patients, the post-injection clinical outcomes were favorable. Our study showed that the blind method can be useful for administering caudal epidural injections to patients with chronic low back pain in the absence of radiological screening.

The potential limitations of our study include our inability to identify potential anatomical variation in the patients due to the lack of radiological visualization. In addition, the study lacked a prospective design and a control group. Nevertheless, our results demonstrate that favorable clinical outcomes and low complication rates can be achieved with caudal epidural injection using the blind method guided by anatomical landmarks.

The most common complications of blind injections are needle misplacement, subarachnoid puncture, and intrathecal or intravascular injection. In our study, needle misplacement occurred in two patients, but the injection was not administered in either case because aspirated blood indicating incorrect localization was detected. Two patients experienced lower-extremity numbness 30 minutes after the injection, but recovered fully within 2 weeks.

In our patients, the post-injection clinical outcomes were favorable. Our study showed that the blind method is safe and useful for administering caudal epidural injections to patients with chronic low back pain in the absence of radiological screening and may be suggested to patients in whom surgery is not indicated and those who have had no benefit from conservative treatment.

None to report.