To corset or not to corset after lumbar spine fixation surgery?: A prospective randomized clinical trial and literature review

Authors: Yin Ju Chen, Jen Tsung Yang, Martin Hsiu Chu Lin, Wei Chao Huang

Abstract

Purpose:

Orthosis after lumbar fusion surgery is common. However, the evidence for benefit remains to be determined, especially in tropical areas with heavy workers. To investigate postoperative orthosis and whether it affects pain improvement, quality of life, and fusion rate.

Method:

From May 2021 to May 2022, this single-center prospective randomized clinical trial enrolled 110 patients. We excluded 9 patients, and 101 people were analyzed finally. Corset group, in which participants used a corset for 3 months postoperatively; Non-corset group, in which participants didn’t wear any orthosis. ODI and VAS scale were recorded before the 2 weeks, 1 month, 3 months, half a year, and 1 year postoperatively. The lumbar X-ray was done before the surgery, 6 months postoperatively. All complications in 1 year were recorded.

Results:

Significant decrease in VAS score in the non-corset group since post-operation day 5 (corset group 3.44 ± 1.77, non-corset group 3.36 ± 1.75, P = .0093) during admission, and also a decrease in admission duration (corset group 11.08 ± 2.39, non-corset group 9.55 ± 1.75, P = .0004) were found. There was a significantly better ODI score in the non-corset group since post-operation 1 month, while in the corset group until post-operation 3 months. Both groups had no significant difference in satisfaction, complication rates, and X-ray results, such as fusion, angular rotation, sagittal transition, and slip in the neutral position.

Conclusion:

After the transpedicular screw fixation with posterolateral fusion surgery for degenerative spondylolisthesis, non-orthosis is a safe strategy. It can reduce the admission duration and has the trend for better functional outcomes.

1. Introduction

Orthoses after spinal fusion surgery are recommended by some institutes, which can decrease motion, increase truncal support, and obtain psychological reminder effects.^[1,2]^ However, other spinal surgeons consider that intervertebral modern spinal instrumentation is much stronger than external compression orthosis.^[3–6]^ According to the American Academy of Orthopedic Surgeons guidelines and the Journal of Neurosurgery guidelines, the use of bracing following instrumented posterolateral fusion was lack of evidence,^[7]^ and was not recommended due to equivalent outcome.^[8]^ Besides, a previous meta-analysis of 4 randomized clinical trials reported there are no advantages for orthosis on disability, VAS pain score, fusion rate, or post-operation complication.^[9]^ Therefore, the practice of postoperative orthosis is inconsistent between clinical scenarios and the study.

Postoperative orthosis can cause immobility and a psychological reminder effect. However, poor compliance with an orthosis is common, such as inconvenience during labor work, especially in tropical areas, where heavy workers account for a large proportion. Besides, the cost of orthosis is a burden for the patient. The above reasons make patients prone to not wearing orthosis. Moreover, some shortcomings remain.^[3]^ Trunk muscle like multifidus, internal oblique muscle was reported to have decreased thickness. A decrease in Electromyography in the external oblique muscle and erector spinal muscle and an increase in rectus abdominis cause the patient to decondition. Increased intra-abdominal pressure can cause difficult breathing, venous engorgement of the hemorrhoid plexus, or abdominal pain. Compression sensation and muscle aches may lead to higher VAS scores.^[4,10]^

The short-term outcomes within 2 weeks were lacking. We integrated the advantages of each of the previous 4 RCTs. We got more detailed data on postoperative VAS change while still admitted, short-term degree of disability, and the relationship between heavy workers and orthosis. Our study aimed to prove the equivalence of “non-corset” compared with “corset.”^[11–14]^

2. Methods

2.1. Patient population

Patients who underwent surgery at the Chang Gung Memorial Hospital, Chiayi Branch were evaluated from May 2021 to May 2022. We collected 110 patients first and randomized them to corset or non-corset groups under computer-generated random numbers. After 9 patients were excluded, 101 patients were enrolled in our study. All patients were available for follow-up at least for 3 months. We supposed to follow-up them for 1 year. All surgery was performed by a single surgeon and post-operation order was the same protocol (Appendix 1, Supplemental Digital Content, http://links.lww.com/MD/M20), which included similar analgesics, postoperative care, and discharge standard. The inclusion criteria were patients older than 20 years, with an magnetic resonance imaging study proved symptoms compatible spondylolisthesis, with surgery level between the first lumbar to the first sacrum, with surgery level lower or equal to 4, and able to sign the permit. Exclusion criteria included previous spinal trauma history, previous spinal surgery history, spinal fracture, spine infection, spine vascular diseases, or unwillingness to be followed.

The informed consent and primary data were obtained before the surgery. We used a randomization sequence, and the attending surgeons were blinded before the surgery.

2.2. Procedure and intervention

We performed traditional laminectomy through a classic posterior midline incision, and inserted transpedicular screw fixation under intraoperative computed tomography navigation, and did the posterolateral fusion. The surgery was performed with pedicle screw instrumentation (Smartloc spinal fixation system, A-SPINE, United Orthopedic Corporation, Taiwan), and autologous local bone graft with some artificial bone (Sinbone bone replacement, Purzer, Purzer Pharmaceutical corporation, Taiwan) in all patients.

Patients in the corset group wore a corset with metals full-time for 12 weeks, except when taking a shower or lying. The corset type was documented in Appendix 2, Supplemental Digital Content, http://links.lww.com/MD/M21. Patients started to wear a corset on the first ambulation after the removal of hemovac. The patients in the non-corset group didn’t wear any orthosis at all.

2.3. Data collection

We followed up on VAS, ODI questionnaire, and satisfaction at each follow-up on 2 weeks, 1 month, 3 months, half a year, and 1 year after the surgery. They could answer it at the outpatient department or on the phone. Patients who were lost to follow-up in 3 months, broke the rules, or found other spine diseases were excluded. A flow diagram in Figure 1 shows the randomization, assigned interventions, exclusion reasons, and follow-up of the study participants. We finally analyzed data of 101 patients, 50 were in the corset group, 51 were in the non-corset group.

Clinical assessments were performed by 3 well-trained independent assistants, 2 in the first year and 1 in the second year, and radiographic evaluation were performed by a blinded resident doctor and a blinded attending doctor; all were not involved in patient care.

2.4. Radiographic collection

The successful fusion of posterolateral fusion is evaluated by postoperative 6 months KUB and evaluated independently by 2 doctors. Lenke classification grade A was defined as definitely solid with bilateral trabeculated stout fusion masses. Lenke classification grade B was described as possibly solid with a unilateral large fusion mass and a contralateral small fusion mass. Lenke classification grade C was probably not solid with a small fusion mass bilaterally. Lenke classification grade D was defined as definitely not solid with bone graft resorption or obvious pseudoarthrosis bilaterally.^[15,16]^

We used angular rotation, sagittal translation, and slip in neutral position change on lumbar plain film flexion and extension view to evaluate the correction loss between pre- and post-operation.^[17]^ Angular rotation was defined as the angle change during flexion and extension between the line parallel to the lower endplate of the upper vertebra and the line parallel to the upper endplate of the lower vertebra. The sagittal translation was defined as the length change of translation of the posterior border of the upper vertebrae between flexion and extension. Slip in neutral position was defined as the percentage of transition of the posterior border of the upper vertebrae in extension view.

2.5. Statistical analysis

We use G-Power (Version 3.1.6) to determine the ideal sample size. To reach an effect size of 0.5 and alpha level of 0.05 with 80% power, each group needs 51 patients. Supposing around 10% lost follow-up, an ideal total sample size was 110 patients.

We used the Fisher exact test to analyze category data, including fusion rate, complication rate, sex, occupation, and the patient numbers in Diabetes Mellitus, rheumatoid arthritis, hemodialysis, and smoking. We used the unpaired t test to evaluate VAS, ODI, change in image data (including angular rotation, sagittal translation, slip in neutral position), body mass index, bone marrow density, operative time, and admission duration. Each group’s progress of clinical outcome, including VAS and ODI, of each group was analyzed by paired t test. In order to adjust the effect of osteopenia and heavy workers, and we had 6 times of observation, we conducted repeated measurement Analysis of Variance. P < .05 was seen to be statistically significant. All statistical analyses were performed using SPSS version 22.

3. Result

Pre-operation data analysis was shown in table 1. There were no significant differences in sex, age, body mass index, bone marrow density, patient numbers in Diabetes Mellitus, rheumatoid arthritis, hemodialysis, smoking, and pre-operation VAS, ODI. The level of the surgery, and the distribution of back pain dominant or limbs pain dominant also showed no significant differences.

The perioperative outcome was shown in Table 1 and Figure 2A. There were no significant differences in operation time, blood loss, or during admission VAS for the first post-operation 4 days. The VAS score in both groups showed significant decrease compared to pre-operation 1 on post-operation day 2. However, a significant more decrease in VAS score in the non-corset group was found on post-operation day 5 (corset group 3.44 ± 1.77, non-corset group 3.36 ± 1.75, P = .0093), post-operation day 6 (corset group 3.52 ± 1.81, non-corset group 2.71 ± 1.6, P = .0184), post-operation day 7 (corset group 3.14 ± 1.77, non-corset group 2.24 ± 1.31, P = .0043), and also decrease of admission duration was found (corset group 11.08 ± 2.39, non-corset group 9.55 ± 1.75, P = .0004).

The VAS score in both groups improved significantly in 2 weeks, 1 month, 3 months, 6 months, and 1 year postoperatively (Fig. 2B). Though the VAS score in the non-corset group was slightly higher than the corset group preoperatively, it was somewhat lower postoperatively in non-corset group, and didn’t reach significant yet.

The ODI scores in both groups increased in 2 weeks postoperatively, and started to have a significant decrease in the non-corset group since 1 month postoperatively (corset group 40.3 ± 11.53–38.71 ± 11.45, P = 048; non-corset group 44.71 ± 11.79–38.81 ± 12.55, P = .006), and in corset group since 3 months postoperatively (Fig. 2C).

Regarding radiographic outcome, the fusion reached Lenke classification grade A in 6 months postoperatively was 83.7% in the corset group, and 90.1% in the non-corset group, which had no significant difference. Angular rotation, Sagittal transition, and Slip in neutral position also showed no difference between 2 groups (Table 2).

The Satisfaction of both groups was almost >90, and there was no significant difference between them when just discharge, 2 weeks, 1 month, 3 months, 6 months, and 1 year postoperatively. (Fig. 2D).

In the subgroup analysis of heavy workers or light workers, both the corset group and the non-corset group had no difference in VAS (P = .333), or ODI (P = .093) preoperatively, 2 weeks, 1 month, 3 months, 6 months, 1 year postoperatively. Though the ODI of the light workers was more in the non-corset group, the follow-up ODI 2 weeks, 1 month, 3 months, 6 months, and 1 year postoperatively showed no significant differences. So does the VAS (Appendix 3, Supplemental Digital Content, http://links.lww.com/MD/M22).

In the subgroup analysis of osteopenia and osteoporosis (T score ≤ −1) or normal bone marrow density, both the corset group and the non-corset group had no difference in VAS (P = .414), or ODI (P = .105) preoperatively, 2 weeks, 1 month, 3 months, 6 months, 1 year postoperatively (Appendix 4, Supplemental Digital Content, http://links.lww.com/MD/M23).

There were 22 patients in the corset group mentioned that there were discomforts when wearing a corset; 7 complained about itchiness, 3 complained about rash, 6 complained about cosmetic or social problems, 6 complained about muscle aches, 7 complained about compression sensation (Appendix 5, Supplemental Digital Content, http://links.lww.com/MD/M24).

Complication data was collected, including perioperative and postoperative complications within 1 year of the surgery. There was 1 patient in the corset group, and 2 patients in the non-corset group observed incidental durotomy. Two patients with superficial wound infection under prolonged antibiotic use were observed in the corset group. There were 3 patients in the corset group and 1 patient in the non-corset group undergoing revision surgery due to adjacent spondylolisthesis (Table 3).

4. Discussion

The practice in postoperative orthosis after lumbar spine fusion is inconsistent between providers. In our randomized clinical trial study, there is no advantage of wearing postoperative orthosis (Appendix 6, Supplemental Digital Content, http://links.lww.com/MD/M25), similar to the previous 4 RCTs in the meta-analysis and another RCT for MIS TLIF.^[18]^ It cannot lead to better pain improvement, functional outcomes, low complication, or high fusion rates. In comparison, the patients in the non-corset group had significant early pain improvement on postoperative day 5 to day 7, which implies a shorter length of stay in the hospital. The ODI also improved dramatically in the postoperative 1 month, earlier than the corset group, which was 3 months postoperatively. This implication of wearing orthosis post lumbar fusion procedure has no role in early recovery. Early core muscle support by itself instead of orthosis may benefit more during the recovery periods.

This study is the only RCT aimed at tropical areas populations who had poor compliance with orthosis, which worked in agriculture, forestry, fishery, animal husbandry, mining, industry, and processing industries, took 65.3%, much more than the Taiwanese average of 33.6%. We also dichotomized them as heavy workers vs light workers. Neither heavy workers nor light workers changed the pain or function outcome with the corset use. Moreover, we recorded the discomfort of wearing the orthosis, including itchiness, rash, cosmetic or social problems, muscle aches, and compression sensation. Our study revealed that almost half of all patients complained of the above discomfort without changing any functional outcome or pain score. Postoperative orthosis is unnecessary because of its disadvantages and because it has no pain or functional improvement in the short or long term.

X-ray of Lumbar flexion and extension lateral view was used for follow-up due to the same accuracy of evaluating fusion compared to CT.^[19]^ Plain X-ray films and CT scans had a sensitivity of 100% for pseudoarthrosis and a negative predictive value of 100% for healed fusion. Otherwise, they can provide more information, including angular rotation, sagittal translation, and slip in a neutral position. However, an earlier study advocated that postoperative lumbar immobilization with a rigid lumbar orthosis worked on the consolidation of posterolateral lumbosacral fusions.^[20]^ There is no significant difference in the fusion reached Lenke classification grade A, which means it is definitely solid with bilateral stout fusion mass present between 2 groups in our study. No matter the fusion rate or correction of spondylolisthesis, the corset group and the non-corset group had no significant difference in the radiographic outcomes.

The complication rate was low in both groups. ^[15]^ We also recorded 3 patients in the corset group and 1 patient in the non-corset group undergoing revision surgery due to adjacent spondylolisthesis in 1-year follow-up postoperatively. No implant dislodgement was found, such as screw loosening or broken screw. Non-orthosis did not lead to a higher complication rate.

The compliance of orthosis is the first limitation of our study. Even if we tried to monitor patients and teach them the adequate usage of the corset daily during admission or each follow-up period. but we still needed help to obtain thorough compliance when they were out of sight. Second, we did not analyze the imaging study of the trunk muscle thickness, which is also our limitation. The core muscle is closely related to our spinal stability and back pain(VAS). Third, the performance bias is still unavoidable; although the surgeon was blinded preoperatively, the surgeon, the patients, and even the evaluator could not be blinded postoperatively during the study. Fourth, the power of our research could be escalated for the number of patients in our study needed to be higher, and only 1 year follow-up period. Further long-term high-quality multicenter or register studies was required.

5. Conclusion

A non-orthosis is a safe strategy after the transpedicular screw fixation with posterolateral fusion surgery for degenerative spondylolisthesis. The length of stay in the hospital is shorter, with early pain improvement and better functional outcomes.

Author contributions

Conceptualization: Yin Ju Chen, Jen Tsung Yang.

**Data ** Yin Ju Chen, Martin Hsiu Chu Lin.

**Formal ** Yin Ju Chen, Wei Chao Huang.

Investigation: Yin Ju Chen.

Methodology: Yin Ju Chen, Wei Chao Huang.

**Project ** Yin Ju Chen, Jen Tsung Yang.

Supervision: Martin Hsiu Chu Lin, Wei Chao Huang.

**Writing – original ** Yin Ju Chen.

**Writing – review & ** Wei Chao Huang.

Supplementary Material