Correlation between the degree of pain relief following discoblock and short-term surgical disability outcome among patients with suspected discogenic low back pain

Introduction

Low back pain (LBP) is one of the most prevalent health problems in the world [1], and commonly arises from intervertebral disc degeneration (IVD). The pain associated with IVD is referred to as discogenic if no clear pathology or anatomical deformity is present. The diagnostic criteria of discogenic LBP are unclear [2]. Non-invasive methods such as physical examination and radiological imaging are either unsensitive or unspecific [3]. Lumbar discography is a widely used diagnostic procedure, and involves the injection of a contrast agent into the disc of interest. The concordance of the patient’s pain response and internal anatomy of the disc are subsequently evaluated [4]. Despite being more accurate than non-invasive methods, discography may be associated with high false-positive rates [4], [5], [6], complications [7], and poor predictive value for successful operative treatment [8, 9]. It also exposes patients to considerably strong pain sensation [4].

Discoblock is an alternative diagnostic method whereby an anesthetic agent is injected into the disc of interest and the result is considered positive when LBP decreases [10]. Liu et al. recently described notable clinical improvement among patients with discography- and discoblock-confirmed LBP 12 months after oblique lumbar interbody fusion, based on Visual Analog Scale (VAS) pain scores and Oswestry Disability Index (ODI) values. The patients underwent the surgery following positive discography and positive discoblock [11]. However, patients were excluded if their discography was negative, which makes it difficult to interpret whether discoblock is a suitable stand-alone selection criterion for surgery. In a Japanese study, patients underwent anterior interbody fusion surgery after their discogenic LBP was confirmed by either discography or discoblock in a randomized fashion. Three years after surgery, the discoblock group had superior surgical results to those of the discography group, as evaluated using VAS, ODI, and the Japanese Orthopedic Association Score (JOAS) [10]. Notably, both studies categorized discoblock results in a binary manner without measuring the degree of post-procedural pain relief.

As scientific evidence of the usefulness of discoblock as a selection criterion for operative treatment of discogenic LBP is scarce, the present study evaluated whether the degree of pain relief following discoblock (ΔNRS) was a good predictor of the disability outcome of subsequent fusion or total disc replacement (TDR) surgery, based on short-term ODI scores.

Methods

Results

General characteristics

Fifteen patients underwent operations. Figure 1 illustrates the exclusion rates and reasons. Three (20%) patients underwent posterolateral fusion, two (13.3%) underwent TLIF, four (26.7%) underwent ALIF, and six (40.0%) underwent TDR. The spinal levels operated on were L5/S1 (n=9; 60%), L4/L5 (n=5; 33.3%) and L3/L4 (n=1; 6.7%). Table 1 presents the patient demographics.

Figure 1:

Flow chart of the study setting. TDR, total disc replacement.

Table 1:

Demographic characteristics.

	Total	Fusion	TDR
No. of patients (%)	15 (100%)	9 (60%)	6 (40%)
No. of male sex (%)	8 (53.3%)	6 (66.7%)	2 (33.3%)
Age range (mean ± SD), yr	21.2–59.0 (44.9 ± 11.3)	21.2–59.0 (47.3 ± 13.0)	31.2–53.1 (41.5 ± 7.7)
No. of non-smokers (%)	9 (60%)	5 (55.6%)	4 (66.7%)
Follow-up time range (mean ± SD), mo	3.0–18.5 (10.8 ± 5.3)	3.0–18.5 (9.2 ± 6.2)	12.0–16.0 (13.3 ± 2.1)

1. TDR, total disc replacement; SD, standard deviation; yr, years; mo, months.

Correlation analysis

We found a Spearman’s rho of 0.57 (p=0.026) between ΔNRS and ΔODI% for the combined group. The individual Spearman’s rho values were 0.85 (p=0.004) for the fusion group and 0.62 (p=0.191) for the TDR group. Figure 2 provides a graphical representation.

Figure 2:

Scatter box showing the positive correlations between the ΔNRS and the percentual change in ODI score for the combined fusion and TDR group, and the fusion and TDR groups separately. TDR, total disc replacement; ΔNRS, degree of pain relief following discoblock; ΔODI%, percentual change in ODI score following surgery. *Indicates a line of percentual decrease of 30% in ODI score. †p=0.026, ‡p=0.004.

The postoperative ODI score did not significantly correlate with the pre-procedural NRS pain score or the preoperative ODI score (rho −0.30; p=0.276 and rho 0.31; p=0.257 for the combined group; rho −0.52; p=0.156 and rho 0.19; p=0.620 for the fusion group; rho 0.058; p=0.913 and rho 0.20; p=0.700 for the TDR group).

Discussion

To our knowledge, this is the first study to compare the ΔNRS with the disability outcome of operative treatment of discogenic LBP. We found a moderate positive correlation between the ΔNRS and short-term surgical disability outcome measured as a decrease in the postoperative ODI score for the combined group. Thus, our results suggest that the ΔNRS can be utilized to predict surgical disability outcome following lumbar fusion or TDR surgery of discogenic LBP.

Further, the correlation between the ΔNRS and postoperative ODI score appeared stronger in the fusion group than in the TDR group (rho 0.85 vs. 0.62, respectively). To the authors’ knowledge, the correlation between discoblock result and TDR surgery outcome has not previously been assessed. Although we used mixed fusion techniques, our results are consistent with previous studies that have reported positive discoblock results to have a predictive value for superior anterior and oblique lumbar fusion outcomes [10, 11]. Thus, we believe that discoblock, among other standard methods (i.e., physical examination and MRI), is a viable selection criterion for the operative treatment of discogenic LBP, especially when fusion surgery is under consideration.

We recorded pain relief following discoblock using a numerical continuous NRS pain score, as opposed to previous studies, which considered discoblock to be positive when a patient has obtained any relief at all for their LBP [10, 11]. In the present study, only one of six (16.7%) patients whose pain relief following discoblock was moderate (i.e., ΔNRS 0–4) exceeded the clinically significant decrease of 30% in the ODI score [17] following surgery (Figure 2). In addition to using the NRS pain score to report the discoblock results, we suggest a discoblock result of ΔNRS >−4 as a disqualifying factor for the operative treatment of discogenic LBP. In addition, using the NRS pain score may reduce the subjectivity of discoblock.

The complication group and the increasing ΔODI% group both included three patients. These groups were overlapped by one patient who underwent TDR with a ΔNRS of −6 and had an inferior ΔODI% of +60.0%. On the two other patients whose ΔODI% increased, discoblock seemed to work as hypothesized (ΔNRS −2 and −2.5; ΔODI% +17.4% and +5.0%, respectively). Conversely, discoblock, as a preoperative test, predicted postoperative complications poorly; patients suffering from complications had ΔNRSes of −2; −5; and −6. As operative complications have a broad variety of possible sources including surgical approach- and device-related [18], we keep discoblock’s inability to predict operative complications reasonable. Furthermore, the complications might also account for the insignificance of the ΔODI among TDR patients. Of the three patients who suffered postoperative complications, two had undergone TDR surgery. They had minimal to no improvement in disability (ΔODI −8 and +30) after a somewhat notable pain relief following discoblock (ΔNRS −5 and −6, respectively). Constituting a third of the TDR group, these patients might have caused the insignificance.

We found no statistically significant correlations between postoperative ODI score and pre-procedural NRS pain score or preoperative ODI score, both of which are identified as patient-related baseline factors that affect the surgical disability outcome of spine surgery [19]. However, as many other confounding factors are present in spine surgery [19, 20] and surgical treatment is generally considered a controversial treatment for discogenic LBP [21], determining discoblock’s predictive value solely on the basis of surgical outcomes may be problematic. This issue is also known as the “gold standard” dilemma. Since no objective standardized method is available for diagnosing discogenic LBP or for comparing to novel diagnostic criterion, validating discoblock or any other diagnostic method is challenging [5]. Therefore, further research on the diagnostics of discogenic LBP is needed.

The unclear diagnostic criteria of discogenic LBP naturally also affect our study setting. Intervertebral disc, facet joint, and sacroiliac joint (SI joint), in order of estimated prevalence, are potential pain generators of LBP. In addition to intra-structural injections and SI joint scintigraphy, the best readily available diagnostic methods to diagnose said structures as the LBP source are physical tests and MRI, both of which our study sample underwent, interpreted by a spine surgeon [3]. As we injected only suspected painful intervertebral discs based on MRI and clinical evaluation, and SI joint originating pain occurs more caudad to lumbar spine [22], we think we defined our study sample as accurately as possible to discogenic LBP patients. Nevertheless, we are unable to completely rule out a clinical bias regarding false assumptions of discogenic LBP as aforementioned diagnostic methods are informationally defective and the complexity of LBP is known [3, 23, 24].

Modic changes are lesions in vertebral endplates seen in MRI as changes in signal intensities. Briefly, Modic change type I represents active inflammation and fibrovascular growth; type II, fatty replacement; and type III, sclerotic conversion of the adjacent bone marrow [25], [26], [27]. These changes are connected to endplate defects and disc degeneration [28, 29]. A positive correlation has been reported between Modic changes and both positive discoblock [30, 31] and superior fusion outcomes [32], [33], [34]. In the current study, the greater response to discoblock might be attributable to the inflammatory phase of the adjacent bone marrow, that is, to type I Modic changes. Therefore, it is possible that our results are related to Modic changes. However, further evaluation is needed in the future.

Our study has some limitations. The design of the study was retrospective and had a limited sample size. We might have introduced a selection bias to our study, as the patients who underwent discoblock might have expressed more severe symptoms or have had discrepancy in their diagnostics. Regarding discoblock, some patients’ pre-procedural NRS pain score could have been low and therefore influenced the magnitude of the reduction in the NRS pain score. In addition, our study lacked a control group, and we could only analyze short-term surgical results in the absence of other common outcome measures besides the ODI score. Although common, the mixed fusion techniques might bias our results. We could not analyze ΔODI% in the fusion subgroups due to our limited sample size. It should also be noted that albeit patients were given a strong recommendation to quit nicotine preoperatively, the operating spine surgeon made the decision patient-wise if continued smoking was a disqualification criterion for surgery. Additionally, the time interval between the assessment of preoperative ODI score and surgery was relatively long for some patients (maximum 228 days). However, patients were re-examined by the operating spine surgeon in one to two weeks’ range before the operation to confirm LBP symptoms unchanged.

In conclusion, we found a moderate, positive correlation between the ΔNRS and short-term surgical disability outcome among patients who underwent fusion or TDR surgery for discogenic LBP. The correlation was very strong in the fusion group. Thus, our findings suggest that discoblock is a useful predictive criterion for disability outcome prior to surgery for discogenic LBP, especially when stabilizing spine surgery is under consideration.