Permission to conduct this study was granted by our Institutional Review Board (approval number: 4-2016-0462). This prospective observational cohort study was conducted from August 16, 2016, to October 30, 2018, as part of the K-SPORT by the National Research and Development Program of the Ministry of Health and Welfare of Korea. A total of 142 consecutive patients with visual analog scale (VAS) spinal pain score of >5 despite conservative treatments such as medications and physical therapies over 3 months at a single institution were included. For these patients, either spinal injection or conservative treatment was performed according to the patients’ preference, and the clinical outcomes of both groups were investigated and analyzed at 1, 3, and 6 months. Diagnosis of lumbar spinal stenosis or herniated lumbar disc was made based on patients’ symptoms, physical examination, and radiological findings on magnetic resonance imaging. Patients with neurological deficits and intractable pain at the time of diagnosis; those who were younger than 18 years; those who had previous lumbar spine surgery, vertebral fractures, spinal infections, tumors, inflammatory spinal diseases, pregnancy, or serious comorbidities at diagnosis; and those with gradually aggravating spinal pain or neurological symptoms that needed pain-alleviating procedure or surgery during the 3 months were excluded.

The baseline patients’ characteristics including age, sex, weight, height, and comorbidities were prospectively collected and analyzed (Table 1). Spinal pain was assessed using VAS (scores range from 0 to 10, with higher scores indicating more severe pain), and quality of life was assessed using the Euro Quality of Life-5 Dimensions (EQ-5D; scores range from 1 to 5, with higher scores reflecting better quality of life) and the 36-item Short-Form Health Survey (SF-36, which address physical and mental subscales; scores range from 0 to 100, with higher scores reflecting lower disability). The Oswestry Disability Index (ODI; scores range from 0 to 100, with higher scores reflecting more disability) was used to assess physical function. The primary outcomes were changes in VAS (back: VAS_back; leg: VAS_leg), SF-36, EQ-5D, and ODI scores from baseline to months 1, 3, and 6, as assessed after treatment of back and leg pain. The secondary outcome was the number of successful responders to treatment at each follow-up period. Successful response was defined depending on previous studies with some modifications [5,7,16]. Successful response was defined as 50%(or 4-point) reduction from baseline VAS_back or VAS_ leg and no increase from baseline SF-36 and ODI or 30% (or 2-point) reduction from baseline VAS_back or VAS_leg with simultaneous 30%(or 3-point) reduction in ODI from baseline or 30%(or 20-point) increase in SF-36 from baseline.

The conservative treatment used in this study included physical therapy, therapeutic exercise of the lower back, and medical treatment. Physical therapy included hot pack, ultrasound, and transcutaneous electrical nerve stimulation. Back exercise was systematic exercise that was conducted at an exercise clinic. Medical treatment was prescription of nonsteroidal anti-inflammatory drugs, anticonvulsants, antidepressants, muscle relaxants, and opioids, as necessary.

The spinal injections used in this study include transforaminal selective epidural steroid injections, medial branch block, caudal block, percutaneous epidural neuroplasty, balloon neuroplasty, and transforaminal epiduroscopic laser annuloplasty [4,12,14,18,21,23]. The number of patients who underwent percutaneous epidural neuroplasty was the highest among the spinal injection group (1 month: 54, 70%; 3 months: 46, 74%; 6 months: 32, 68%). Balloon neuroplasty, which induces epidural decompression and adhesiolysis, was performed using gentle side-to-side movement of the catheter with intermittent ballooning. Transforaminal epiduroscopic laser annuloplasty, a recent development, enables simultaneous decompression and thermal annuloplasty. Percutaneous epidural neuroplasty is a catheter procedure in which the catheter is placed directly into the lesion site compromising the nerve root. All conventional procedures were performed by one physician in compliance with the Spine Intervention Society protocol [22].

Statistical analyses were carried out using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA). A p-value <0.05 was considered significant at two-tailed test. Continuous variables were presented as mean±standard deviation or median (Q1-Q3) depending on whether the normality assumption was met or not. If the normality assumption was not violated, then an independent two-sample t-test was adapted to compare the outcomes between the two groups. In case of violation of the normality assumption, the comparison was conducted using Mann-Whitney U-test. Categorical variables were compared using chi-squared test or Fisher’s exact test, as appropriate, and the results were reported as frequency along with the percentage in parenthesis. Significant p-values at significance level of α=0.05 are presented in boldface. For patients who finished up to 6 months of follow-up, we utilized a linear mixed model to investigate whether there are significant changes on the primary and secondary outcomes from baseline over time. Multiple comparisons were adjusted with Bonferroni method.

There were statistically significant improvements after treatment in all categories of both groups at 6-month follow-up. VAS_leg was significantly more decreased in the spinal injection group than in the conservative treatment group at 1 month after the start of treatment (spinal injection group: 2.94±0.23; conservative treatment group: 3.8±0.19; p=0.0096). However, there was no statistical difference between the spinal injection and conservative treatment groups after 3 months. All EQ-5D scores gradually decreased until 3 months of follow-up and then increased again in the conservative treatment group. Similarly, the EQ-5D scores in the spinal injection group decreased until 1 month of follow-up and then increased again. The best therapeutic effect was obtained at 1 month in the spinal injection group and at 3 months in conservative treatment group. In both groups, the SF-36 and ODI scores decreased until 3 months of follow-up, showing improvement in clinical outcome and then increasing again (Fig. 2). In particular, the level of anxiety in the EQ-5D score was significantly lower in the spinal injection group at 1-month follow-up, but there was no significant difference between the two groups after 3-month follow-up.

The treatment results of the conservative and spinal injection groups were compared by two success criteria. According to criteria 1, 73 (38%), 68 (48%), and 57 patients (40%) were successful responders at the 1-month, 3-month, and 6-month follow-up periods, respectively. According to criteria 2, 109 (57%), 106 (63%), and 85 patients (59%) were successful responders at the 1-month, 3-month, and 6-month follow-up periods, respectively (Table 2). There was no statistically significant difference in the rate of successful responders between the spinal injection and conservative treatment groups for each follow-up period; however, the percentage of successful responders tended to be higher in the spinal injection group at 1 month (Table 3).

The starting points of the clinical outcomes, especially VAS, differed. Patient who were in more pain and had worse functional performance preferred spinal injection. Therefore, the selection of treatment depends on patients’ preference, and not on objective criteria. After enough provision of information of spinal injection and conservative treatment, including treatment costs, effectiveness and plan, the method of treatment was selected based on patients’ preference.

The treatment method of both groups was heterogeneous. A large number of patients underwent percutaneous epidural neuroplasty. Furthermore, the medication prescribed for the conservative treatment group was not homogeneous, as the drugs were prescribed depending on the degree and pattern of patients’ symptoms.

The primary and secondary outcomes of the spinal injection group were underestimated because crossover patients were excluded from this study. Crossover patients who switched their treatment method, from conservative treatment to spinal injection, imply failure of conservative treatment. In other words, the rate of successful conservative treatment may be overestimated.