A novel mini-open transforaminal lumbar interbody fusion for lumbar degenerative diseases: technical note and preliminary results

Transforaminal lumbar interbody fusion (TLIF) has been proven as an excellent treatment option for various spinal pathological conditions, such as degenerative disk disease and deformity [1]. Surgeons have been pursuing minimally invasive technology as soon as TLIF was introduced and many techniques have been developed. Following technical advances in surgical instruments, minimally invasive surgery has attracted considerable interest as an alternative to open surgery. Endoscopic techniques, such as full endoscopic technique and unilateral biportal endoscopic technique, have increased in popularity among surgeons in the past decades. The clinical outcomes among lumbar interbody fusion procedures still remain controversial. Some studies have shown good to early outcomes after endoscopic lumbar interbody fusion. Some surgeons believe that the endoscopic lumbar interbody fusion will become more prevalent and even take the place of open procedure. We consider this to be an incomprehensive opinion. The open surgery can be modified and still has advantages over the next few years. We would like to share a novel mini-open TLIF with a smaller midline incision, limited subperiosteal dissection and percutaneous pedicle screws (PPS), which we refer to as mini-open TLIF (MO-TLIF). The objective of this study was to present the feasibility of this procedure and the preliminary results.

This study included 96 patients (47 males and 49 females) with lumbar degenerative diseases, with a mean age of 54.8 ± 17.4 years (range 22–89 years); the demographic characteristics of patients are shown in Table 1. Operation time of single level was 112.6 ± 15.2 min (range 75–160 min), operation time of multilevel was 140.1 ± 16.3 min (range 112–185 min), intraoperative blood loss of single level was 64.5 ± 30.2 ml (range 35–125 ml), intraoperative blood loss of multilevel was 116.3 ± 30.4 ml (range 60–215 ml), and calculated total blood loss in single segment was 366.9 ± 154.8 ml and was 431.2 ± 175.9 ml in multilevel. Postoperative ambulatory time of single segment was 1.7 ± 0.4 days (range 1–3 days), and multilevel ambulatory time was 2.0 ± 0.5 days (range 1–5 days). The perioperative data are shown in Table 2. Follow-up period was 13.2 ± 2.1 months (range 12–18 months). The VAS score for preoperative leg pain was 6.65 ± 0.86, the VAS for preoperative back pain was 5.28 ± 1.19, the VAS for leg pain 1 week postoperatively was 1.67 ± 0.72, the VAS for back pain 1 week postoperatively was 2.34 ± 0.93, the VAS for leg pain 1 year postoperatively was 0.45 ± 0.56, the VAS for back pain 1 year postoperatively was 0.98 ± 0.66, the preoperative ODI (%) was 65.77 ± 10.01, the ODI (%) 1 week postoperatively was 32.95 ± 8.52, and the ODI (%) 1 year postoperatively was 9.38 ± 3.33. The VAS and ODI scores before and after surgery were significantly different (P < 0.0001) and reached minimal clinically important difference (MCID), as shown in Table 3. The CSA for preoperative lumbar muscles on the symptomatic side was 2088.4 ± 226.7 mm² and that on the asymptomatic side was 2081.8 ± 238.6 mm². There was no statistically significant difference. The CSA for lumbar muscles on the symptomatic side postoperatively which was measured by MRI 1 year after operation was 2077.9 ± 225.5 mm² and that on asymptomatic side was 2076.1 ± 235.5 mm². Atrophy rate was 2.5% for symptomatic side and was 1.2% for asymptomatic side, which had no statistically significant difference (P > 0.05). The properties of fat infiltration on symptomatic side were 22.14% ± 9.21% preoperatively and 22.09% ± 9.04% postoperatively. The properties of fat infiltration on asymptomatic side were 21.78% ± 8.71% preoperatively and 22.20% ± 9.19% postoperatively. The detailed paraspinal muscle data are presented in Tables 4 and 5. At the last follow-up, 93 (93/96) cases demonstrated solid fusion through CT scanning and flexion–extension lateral radiography, with a fusion rate of 96.88%.

In the past decade, the technologies of tubular retractors and endoscopes have shown dramatic development [3]. Various procedures such as minimally invasive TLIF (MIS-TLIF), percutaneous endoscopic TLIF (PE-TLIF) and unilateral biportal endoscopic TLIF (BE-TLIF) have become popular alternations to conventional open surgery for lumbar degenerative diseases requiring interbody fusion [4‐8]. These techniques are superior to traditional open surgery in trauma, bleeding volume and postoperative rehabilitation [9]. Some surgeons believe that the endoscopic lumbar interbody fusion will become more prevalent even take the place of open procedure. However, we proposed that open surgery can be modified and still has advantages. Previous literature demonstrated mini-open TLIF which is similar to MIS-TLIF using Wiltse approach and expandable tubular retractor [10‐12]. Our study presents a novel mini-open technique with PPS, smaller midline incision and limited subperiosteal dissection to access the less damage of paraspinal muscles, which is not reported previously as far as we know. MO-TLIF procedure not only aims to acquire similar clinical outcomes and minor damage to muscles compared with other minimally invasive surgeries, such as endoscopic techniques, but also provides a customary alternative for traditional surgeons with a smooth learning curve.

MO-TLIF can be performed with or without visualization aids such as surgical loupes, or microscopes and can handle not only single-level but also multilevel diseases. In the operation of single-level, the paraspinal muscles are subperiosteally dissected from the spinous processes and laminae with a 3 cm incision. In comparison, the systematic review involving 28 studies and 1,475 patients published by Lei Zhu et al. [13] reported that PE-TLIF shows advantages in less surgical trauma and early postoperative relief of back pain and the medium- to long-term clinical outcomes were similar to MIS-TLIF. This meta-analysis shows the mean operative time was 155 min for the PE-TLIF and the mean intraoperative blood loss was 101.1 ml. The mean operative time of MIS-TLIF was 181.1 min, and the mean intraoperative blood loss was 174 ml. The MO-TLIF presents the similar clinical outcomes in terms of VAS scores, ODI scores, bleeding volume, postoperative ambulation time in single-segment treatment and shorter operation time compared with PE-TLIF, which means less surgical trauma than MIS-TLIF. We attempted Wiltse approach in a group of patients previously [14], but it was tough to dissect muscle ideally and took a long time to place tubular retractors, especially in muscular patients, which was not as comfortable as mentioned in certain studies. Meng, Fanjian et al. revealed that there was no significant difference in the deposition of adipose tissue in the paravertebral muscles and the area of paravertebral muscles at the last follow-up (P > 0.05) for limited subperiosteal dissection and Wiltse approach [14]. Moreover, MO-TLIF is different from PE-TLIF, which consumes a lot of water for establishing water channel. It would become more meaningful in future if water resources were scarce.

To the best of our knowledge, few studies have reported endoscopic lumbar interbody fusion for multilevel diseases. MO-TLIF can be advantageously applied to treat multilevel lumbar degenerative diseases. The incision can treat the two segments by lengthening 1 cm (to proximally or distally) and 2 cm for three segments. The 49 patients who underwent multilevel operation reveal significant relief of low back pain and leg pain. Preoperatively, most patients presented with intermittent claudication and radicular pain in the lower extremities, and the mean VAS score for leg pain was higher than the mean VAS score for back pain. The postoperative VAS score (back pain and leg pain) and ODI both improved significantly compared to pre-surgery scores. Parker et al. revealed that MCID scores following TLIF are 2.1 points for back pain VAS, 2.8 points for leg pain VAS and 14.9 points for ODI. The VAS and ODI of MO-TLIF reached the MCID scores [15]. The mean operation time of multilevel was 140.1 ± 16.3 min. A slight increase in incision length, blood loss and operation time can achieve the same clinical outcomes as single-segment treatment.

The CSA of symptomatic side (approach side) and asymptomatic side was measured by MRI preoperatively and postoperatively (Table 4). The CSA was measured at the level of the center of the intervertebral disk and calculated the average in multilevel. Preoperative CSA was similar on both sides with no statistical difference (P > 0.05; Table 5). The postoperative CSA showed slight atrophy in both sides but was not significant (P > 0.05). There are no statistical differences of CSA in postoperatively bilateral paraspinal muscles (Fig. 4). Atrophy rate was 2.5% for symptomatic side and 1.2% for asymptomatic side, which demonstrates that limited subperiosteal dissection did not increase atrophy of paraspinal muscles compared with PPS.

Intraoperative blood loss was estimated by calibrated collection bags, which may have bias. The Nadler’s formula was used to evaluate the blood loss, which takes into account height, weight and sex for calculation [16‐22]. However, all blood loss estimation formulas showed a significant tendency to overestimate blood loss according to reference [23]. The total blood loss is 366.9 ± 154.8 mL in single-level operation and 431.2 ± 175.9 mL in multilevel. None of the 96 patients received blood transfusions during the perioperative period.

The main complications included wound infection (n = 10), dural tear (n = 4) and wrong surgical segment (n = 2). No neurological deficit, cage subsidence or screw loosening occurred. It is necessary to reconfirm segment by fluoroscopy after dissection. Ventral and the excessively thick lateral ligamentum flavum were removed. Part of dorsal ligamentum flavum was retained to keep spinal canal intact, thus reducing the possibility of epidural scar formation and shortening the operation time. This technique can also achieve decompression of contralateral central spinal canal and lateral recess by removing the root of spinous process.

This research aims to present a new open TLIF without special surgical instruments and aqueous medium to achieve similar minimally invasive effect compared with PE-TLIF and MIS-TLIF. The most outstanding advantages of MO-TLIF are treatment for multiple segments and amicable learning curve.

This study has some limitations. First, this study did not include a multicenter trial, and there was an absence of a control group. Second, the overall follow-up period was too short (ranging from 12 to 18 months) to determine the clinical efficacy of MO-TLIF. Finally, the blood loss calculation lacks universal assessment methodologies, which may cause a bias.

Under appropriate patient selection and surgical indications, mini-open TLIF is effective and feasible for the treatment of lumbar degenerative diseases, especially multilevel diseases, with minor muscle damage and shorter operation time. It is an optional method of lumbar interbody fusion.