Characteristics of relief and residual low back pain after discectomy in patients with lumbar disc herniation: analysis using a detailed visual analog scale

Lumbar disc herniation (LDH) is one of the most common causes of low back pain (LBP) and sciatica. Surgical treatment is well known to be beneficial for patients with LDH who fail to respond to conservative care. Favorable results for LBP, lower extremity pain (LEP), and lower extremity numbness (LEN) in patients with LDH treated with discectomy have been demonstrated [1‐4]. In most of those studies, only a conventional visual analog scale (VAS) and the Oswestry disability index (ODI) were used to determine LBP. Those studies were unable to determine detailed changes in LBP and the characteristics of relief and residual LBP. By contrast, recent reports have clarified the characteristics of LBP using a detailed VAS for some lumbar degenerative disorders, including spondylolysis and lumbar spinal stenosis (LSS) [5‐7]. Nevertheless, there are no reports of the detailed changes and location of LBP before and after discectomy for LDH. Thus, we conducted an observational study to clarify the detailed characteristics and location of LBP, LEP, and LEN before and after discectomy for LDH using detailed VAS bilaterally.

To our knowledge, this study is the first to determine the characteristics and location of LBP using a detailed VAS we developed in patients with LDH treated with discectomy who did not experience a recurrence. According to a previous study that analyzed detailed and bilateral VAS scores for LSS patients, LBP in patients with LSS before surgery were significantly greater while the patient was standing, but pain was reduced by decompression surgery, with LBP improving equally on the affected and opposite sides [6]. The first noteworthy point of the present study is that LBP while the patients were in motion was significantly greater in those with LDH before surgery, and the LBP while they were in motion on the affected side was reduced by discectomy. This pattern of LBP relief suggests that radicular LBP is improved by nerve root decompression surgery, as indicated in previous reports [1, 6]. However, despite this similarity regarding nerve root decompression, the greater LBP that occurred in patients with LDH while they were in motion was distinct from the increased LBP found while patients with LSS were standing. For this reason, we speculate that nerve root compression in patients with LDH usually occurs with a more acute onset than that in patients with LSS. In addition, this difference in LBP characteristics may be influenced by the degree of disc and endplate degeneration in patients with LDH compared with those with LSS because patients with LDH tend to be younger than those with LSS.

Results from the present study also suggest that residual LBP after discectomy in patients with LDH who did not experience a recurrence was most pronounced while the patient was sitting. A recent report indicated that higher intradiscal pressure while sitting may result in LBP in the presence of lumbar degenerative disc diseases [11]. Pathological mechanisms of discogenic low back pain included sensory nerve ingrowth into the disc, upregulation of neurotrophic factors like nerve growth factor and inflammatory cytokines, and mechanical stress [12, 13]. Our findings of residual LBP while the patients were sitting and changes in Pfirrmann grade, when taken in combination, may suggest that the residual LBP was associated with increased load and pressure on the disc in the sitting position. Alternatively, it is also well known that Modic changes influence LBP [14]. Ohtori et al. reported favorable surgical outcomes for LDH complicated with Modic type I [15]. Although LEP improvement was obtained in patients with Modic change in our study, the residual LBP in the MC group leads us to believe that changes in load and inflammation at the endplate may also cause residual LBP while the patient is sitting.

Recent reports indicated that performing a minimally invasive discectomy using a tubular retractor under a microscope or endoscope is feasible for the treatment of LDH [16, 17]. In the present study, we compared these 3 surgical procedures, including conventional discectomy. Residual LBP at 3 months after surgery was greater in patients in group T because the baseline of LBP before surgery was significantly greater in those in this group. However, the residual LBP at 1 year follow up was equal following all 3 surgical procedures. This, along with previous reports, suggests that surgical invasion of the paraspinal muscles does not influence residual LBP [6, 16]. In addition, in the analysis of the residual LBP at 1 year follow up except for cases of bilateral laminectomy (group B), we found similarly that LBP while the patient is sitting was significantly higher than the LBP found when the patient was in motion or standing. Furthermore, recent reports indicated that the surgical procedure (open discectomy versus micro discectomy) did not influence the surgical outcome for the residual pain [18, 19]. Although including the 3 different surgical methods may present the bias in the present study, we speculate that the type of surgical procedures has less of an impact on the result for the residual LBP.

While no reports describing the relationship between surgical levels and residual LBP were found, in the present study, residual LBP was significantly greater in patients with herniations at L3–4. It is difficult to explain this phenomenon. However, we speculate that this may have been because patients with L3–4 herniations were highly complicated and also had L4–5 or L5-S disc degenerations. Further investigation with a larger sample size including multivariate modeling or mediation analysis is needed to understand this residual LBP.

The present study has several limitations. First, the present study is observational, and we did not evaluate detailed and bilateral LBP VAS scores of patients who underwent conservative treatment alone. In our study some of the patients underwent conservative treatment at another hospital, and they wished to undergo surgical treatment as soon as possible, leaving no time to evaluate further conservative treatment. Further prospective investigation will be needed to clarify this point. Second, in the present study, we investigated only ODI as a patient-based outcome and did not investigate other patient-based outcomes such as Short Form 36 (SF-36), EuroQol 5 dimension (EQ-5D), and Japanese Orthopaedic Association back pain evaluation questionnaire (JOA BPEQ). Although such patient-based outcomes as SF-36, EQ-5D, and JOA BPEQ are also important, the main purpose of this study was to determine the degree of detailed LBP under various conditions (while the patient was in motion, sitting, and standing). However, patient-based outcomes could not determine the LBP in various conditions and may include postsurgical psychogenic factors. Furthermore, excessive quantities of questionnaire may place the extra stress on the patients. To simplify the result and avoid the excessive stress on the patients, we only investigated VAS scores and ODI. Third, the present study excluded patients complicated with dynamic instability or patients with lateral herniations who underwent fusion surgery because we wanted to avoid LBP caused by instability of discs and facet joints [20]. Furthermore, the present study also excluded patients who, unfortunately, had a recurrence of herniation in the short term (less than a year postoperatively) because we wanted to determine residual LBP in the absence of herniation recurrence. If those cases had been included, the results would have been confounded because they had high VAS scores under all 3 conditions. Fourth, the present study did not evaluate sagittal alignment. Sagittal imbalance such as pelvic incidence and lumbar lordosis mismatch may contribute to postoperative LBP [21]. Using detailed VAS scores, Aoki et al. indicated that sagittal imbalance after a short segment fusion surgery resulted in residual LBP while the patient was standing [22]. Considering our finding that residual LBP while the patient was sitting was present at 1 year after discectomy, we speculate that residual LBP is less affected by sagittal alignment. Finally, the follow up after discectomy was incomplete. When patients undergo discectomy and have significant pain relief, they sometimes drop out of care at the outpatient clinic. In the present study, 18 of 114 (15.7%) patient participants dropped out. Generally, 2 years of follow-up is recommended for this type of study. However, in the present study, we were compelled to set the follow-up period to 1 year because of a decreasing follow-up rate. Further investigation, such as a prospective cohort study that follows all the cases fully will be needed to resolve this follow-up issue.

In patients with LDH without recurrence, LBP on the affected side while they are in motion was significantly greater before surgery and was reduced following discectomy. This LBP relief suggests a radicular nature to the LBP that is improved by nerve root decompression surgery. The residual LBP at 1 year after discectomy was most dominant while they were sitting. This, in combination with our findings regarding the relationship between residual LBP and changes in Pfirrmann grade and Modic type, may suggest that the residual LBP was associated with increased load and pressure on the disc in the sitting position.