Minimally Invasive Transforaminal Lumbar Interbody Fusion Ameliorates Persistent Pain After Lumbar Fusion Surgery

Background: Approximately 4-20% patients with degenerative lumbar diseases showed persistent pain after lumbar fusion surgery that may develop into failed back surgery syndrome (FBSS), and this persistent pain may be related to the postoperative increased release of inammatory mediators. Minimally invasive transforaminal lumbar interbody fusion (MIS-TLIF) can obviously reduce the intraoperative soft tissue trauma. The aim of this study is to investigate the persistent pain in the patients with degenerative lumbar diseases undergoing MIS-TLIF compared with conventional ‐ invasive TLIF. and methods: This study retrospectively included 146 patients (MIS-TLIF vs. conventional ‐ invasive TLIF: 56 vs. 90), and the incidence of persistent pain were evaluated. Furthermore, inammation related markers in both blood and drainage uid samples, including white blood cell (WBC) count, C-reactive protein (CRP), creatine kinase (CK), interleukin-6 (IL-6) and IL-1β, were tested before and after operation.

Local in ammatory substance accumulation may be potential cause for postoperative persistent pain, and MIS-TLIF may reduce this in ammatory accumulation at the surgical site and subsequently reduce the risk of persistent pain.

Background:
Transforaminal lumbar interbody fusion (TLIF) is a routine and effective operation for degenerative spinal diseases [1]. While TLIF usually has excellent clinical outcomes, serious complications can occur [1][2][3]. Among these, postoperative persistent pain of low back/lower limbs after TLIF is a devastating complication that may develop into failed back surgery syndrome (FBSS) [4][5][6], which lead to higher morbidity/mortality rates and greater healthcare costs. Given the clinical outcomes and economic impact of both postoperative persistent pain and subsequent FBSS, efforts to minimize persistent pain in the early postoperative period have a high priority. Therefore, many previous studies have investigated potential mechanism of this postoperative persistent pain of low back/lower limbs [7][8][9][10], and recent studies demonstrated that increased release of in ammatory mediators caused by surgical trauma may be one of the possible reasons for this persistent pain [9,10], which was further supported by intraoperative application of the epidural steroid can effectively reduce postoperative persistent pain since steroid can reduce the surgical local in ammatory substances accumulation.
In the last few decades, an increasing interest in that minimally invasive spine surgery has been observed [11][12][13][14]. The advantages of the minimally invasive TLIF (MIS-TLIF) include smaller incisions, less blood loss, less damage to the dorsal musculature, shorter hospitalization times, reduced perioperative morbidity and better rehabilitation, thus providing a bene cial alternative to conventional surgical approaches to treat degenerative lumbar diseases [11,12]. However, fewer studies involving the impact of MIS-TLIF on the postoperative persistent pain in patients with degenerative lumbar diseases have been conducted, although this type of study may guide clinicians to explore better treatment that allow better prognoses in patients with degenerative lumbar diseases.
The aim of this study was to investigate the incidence of the postoperative persistent pain in patients undergoing MIS-TLIF compared with those undergoing conventionalinvasive TLIF, and the potential possibilities of persistent pain after lumbar fusion surgeries were also analyzed in this study.

Methods:
Subjects: This retrospective cohort analysis included a total of 146 patients with lumbar spinal stenosis and/or lumbosacral disc herniation. In the present study, fty-six patients underwent MIS-TLIF [15,16], and the other 90 patients underwent conventionalinvasive TLIF [1] (Table 1). All patients were recruited in SongJiang district central Hospital from October 2016 to September 2019. The study protocol was approved by Human Ethics Committees (Songjiang district central hospital, Shanghai, China; SJ2020-KY014). All subjects gave informed consent. The inclusion criteria for patients in this study included (1) low back discomfort with referral of pain/paresthesias into lower limb and/or neuronal intermittent claudication (2)  Surgical procedures: Procedure for minimally-invasive TLIF In the present study, the MIS-TLIF was performed by the same spine surgeon experienced in this technique. After the general anesthesia, the patient is placed in a knee-chest position, and intervertebral segment to treat is located by a positioning needle under the C-arm uoroscopy. After performing a 2-to 3-cm skin incision approximately 3 cm away from spinous process, the guidewire was inserted along the wiltse approach into the facet joint, followed by placement of the expansion sleeve and assembly of the minimally invasive access system. The inferior and superior facets were resected under direct visualization, and then a discectomy was performed. A polyetheretherketone (PEEK) cage lled with harvested local was placed after endplate preparation. Bilateral pedicle screws were inserted through wiltse approach by minimally invasive access system, and both compression and xation over the intervertebral space was completed afterward.
Procedure for conventionalinvasive TLIF During the conventionalinvasive TLIF procedure [1], a midline incision was made. The soft tissues were cut in layers, and the paraspinal muscles were stripped from both sides of the spinous process to expose the lamina and the articular process. After the pedicle screw was implanted, the inferior and superior facets of the intended levels were resected under direct visualization, and then a discectomy was performed. A polyetheretherketone (PEEK) cage lled with harvested local was placed after endplate preparation, and compression over the intervertebral space was completed afterward.
Postoperative Management: Antibiotics were given at postoperative 48 hours to prevent infection, and both hormone and non-steroidal anti-in ammatory drugs were not used in all patients in this study after operation. After one day of bed rest, the patients were allowed to walk with the protection of a waist brace. When the patients presented with postoperative persistent pain, the non-steroidal anti-in ammatory drugs were used until the Numerical Rating Scale(NRS)scores are lower than 3. A drainage tube was placed for 1-2 days. Patients were usually discharged from hospital 5-10 days after operation Perioperative assessment: Assessment of pain: Both low back and lower limb pain in all patients were measured by NRS scores before, 3 days, 2 weeks and 6 months after operation. In this study, the persistent pain was de ned as similar or more serious pain at the primary site or other sites in low back/lower limbs after operation. NRS scores is a continuous scale composed anchored by a score of zero, indicated no pain, and a score of 10, represented the worst pain.
Assessment of venous blood and drainage uid samples: Venous blood samples were obtained before, 1 day, 3 days and 6 days after operation, and the wound drainage uid was collected from 1-3 days after operation according to the drainage time. For the blood samples, white blood cell (WBC) count, interleukin-6 (IL-6), C-reactive protein (CRP) and creatine kinase (CK) were measured. In the drainage uid samples, both interleukin (IL)-1β and IL-6 were measured.
Statistical methods: The measurements were analyzed using SPSS version 18.0 (IBM, USA). Measurements between the cases in two patient groups were compared by the independent t-tests, and the same statistical method was also used to analyze the measurements between the cases with or without persistent pain in both patient groups. The frequencies of postoperative persistent pain between two patient groups were compared by chi-square tests. The correlations between the extent of persistent pain and in ammation related markers in both blood and drainage uid samples were analyzed by Pearson correlation coe cient analysis. In all instances, a P-value < 0.05 was considered signi cant.

Results:
There was no statistical difference in patient characteristics between patients who accepted MIS-TLIF or conventionalinvasive TLIF (Table 1, P > 0.05), and both NRS scores and all measurements of blood samples were similar between these two patient groups before operation ( Fig. 1, P > 0.05).
Compared with the patients accepting conventionalinvasive TLIF, those undergoing MIS-TLIF showed obviously less intraoperative bleeding and total amount of postoperative drainage (Table 1, P < 0.05). In contrast, there was no difference of operative time and drainage time between these two groups ( Table 1, P > 0.05). Obviously lower CK and IL-6 in the blood samples were observed in the patients undergoing MIS-TLIF than those in the patients accepting conventionalinvasive TLIF in all postoperative assessments ( Fig. 1, P < 0.05), and the patients undergoing MIS-TLIF also presented with signi cantly lower IL-6 and IL-1β in the drainage uid samples compared to those undergoing open discectomy (Fig. 2, P < 0.05). Furthermore, similar measurements of both WBC and CRP were observed in both patient groups in all postoperative assessments (Fig. 1, P > 0.05).
Importantly, signi cantly larger number of the patients who underwent conventionalinvasive TLIF presented with postoperative persistent pain compared to the patients undergoing MIS-TLIF (4/56, 7.1% vs. 20/90, 22.2%; P < 0.05), and both 16 patients accepting conventionalinvasive TLIF (16/20, 80.0%) and 3 patients undergoing MIS-TLIF (3/4, 75.0%) presented with more severe and extensive lumbar/lower limb pain compared to the pre-operative pain. In both treatment groups, the patients with persistent pain presented with higher IL-6 and IL-1β in the drainage uid samples, as well as the greater IL-6 in the venous blood samples, compared with those without persistent pain ( Table 2,    Discussion: The results of this study demonstrated a signi cant difference in the incidence of postoperative persistent pain between the patients who accepted MIS-TLIF or conventionalinvasive TLIF, and an obvious correlation between this persistent pain and surgical local in ammatory substances accumulation was also identi ed in this study.
Consistent with previous studies [17][18][19], both less intraoperative bleeding and postoperative drainage, as well as lower CK, were observed in patients undergoing MIS-TLIF compared to those undergoing conventionalinvasive TLIF, suggesting less surgical trauma in the MIS-TLIF group. Previous studies demonstrated that nerve root edema caused by intraoperative traction injury may be possible reason for postoperative lower limb pain [20,21]. However, most patients in this study showed obviously more extensive coverage of postoperative pain than those supplied by intraoperative decompressed nerve root.
Furthermore, recently published studies demonstrated that postoperative persistent pain and the resulting FBSS even may occur at the other sites, not the preoperative involved site, in both low back and lower limbs [4][5][6]. These results collectively argued against the intraoperative over traction of nerve root is the main cause for the postoperative persistent pain and FBSS.
These are increasing evidences that in ammatory stimulation presented with greater correlation with the radicular and low back pain compared to the mechanical stimulation [22][23][24][25][26]. Although in ammatory substances in the blood sample are similar between two patient groups in this study, the patients undergoing MIS-TLIF presented with obviously fewer in ammatory substances in the drainage uid than those in the conventionalinvasive TLIF patient group. These ndings suggested that postoperative persistent pain may be mainly ascribed to the stimulation of local in ammatory substance accumulation rather than systemic in ammatory response caused by surgical trauma, which was further supported by signi cant correlation between the in ammatory markers in drainage uid and the extent of postoperative persistent pain in postoperative third day in this study. Therefore, di culty in aggregating of in ammatory substances around the nerves after operation may be the main reason for the low incidence of persistent pain after MIS-TLIF.
According to the previous studies, in ammatory substances may be released by the locally damaged tissue around the surgical site [27,28], that may stimulate the nerve roots and cause persistent pain. Different from the conventionalinvasive TLIF, MIS-TLIF is performed under the minimally invasive access system during the operation [15,16], which may effectively prevent the aggregation of local in ammatory substances thorough protecting surrounding tissues. Therefore, less in ammatory exudation at the surgical site caused by smaller surgical wounds in MIS-TLIF may be another reason for reduced aggregation of local in ammatory substances around nerve roots.
The ndings of this study should be interpreted with caution. The half-life of IL-6 or IL-1β is quite short that may affect the analysis results, and in ammatory substances in the drainage uid can only indirectly re ect the local in ammation. Another clinical limitation of this study is low sample size.
Therefore, more signi cant results might be achieved in future study with establishment of both more suitable marker and an increased number of cases.

Conclusion:
The results of the current study support the view that local in ammatory substance accumulation is a potential cause for postoperative persistent pain in patients with degenerative lumbar diseases. Therefore, perioperative management in patients with degenerative lumbar diseases should account for local in ammatory response. Importantly, differences in postoperative results between the MIS-TLIF and conventionalinvasive TLIF groups suggested that MIS-TLIF may effectively reduce the local in ammatory substances at the surgical site, reduce the risk of postoperative persistent pain and resulting FBSS. Ethics approval and consent to participate The study protocol was approved by Human Ethics Committees (Shanghai Songjiang District Central Hospital). All subjects gave informed consent.

Consent for publication
Not applicable.

Availability of data and materials
All data generated or analysed during this study are included in this published article.