The conventional posterior median approach of TLIF has a wide range of invasion to paraspinal muscles, which is prone to damage the posterior ramus of the spinal nerve located in the muscle and increase the intramuscular pressure, leading to the phenomenon of ischemia and denervation of paraspinal muscles, which may eventually lead to muscle atrophy, fibrosis, and fatty infiltration changes[1]. In order to reduce the injury of soft tissue and the incidence of postoperative complications such as low back pain, various minimally invasive and mini-open of TLIF were introduced and applied clinically. There is no doubt that these MIS-TLIF and MO-TLIF procedures have reported more advantages than CO-TLIF procedures in terms of the average blood loss, mean hospital stay, quality of life, and so on[2–5]. However, the current definition of minimally invasive surgery is still unclear. Even in the past two decades, numerous articles have confused MIS-TLIF with MO-TLIF. Recently, some literatures have proposed the definition of MIS-TLIF[13, 14]. Different from other types of spinal surgery, MIS-TLIF has several elements, (1)The use of some special retractors (including non-expandable and expandable), (2)Incision and approach are more conducive to screw placement and laminectomy decompression, (3)Proper use of various visualization aids such as surgical loupes, microscope or endoscope. In our MIS-TLIF procedure, we used a bilateral paramedian incision through the Wiltse approach with the aid of a Quadrant tubular retractor, which met the various elements of MIS-TLIF mentioned above.
With the wide development of MIS-TLIF, many deficiencies are emerging. Most MIS-TLIF operations are performed through Wiltse intermuscular approach. However, the clinical practice found that the paraspinal muscles in the lower lumbar spine are very hypertrophic, increasing the distance between the posterior midline and the Wiltse plane and making it relatively difficult to perform screw placement and decompression. In addition, the reliance on special retraction devices restricts the wide use of MIS-TLIF, especially in undeveloped countries and regions, and long-term use of retractors may also lead to muscle injury, ischemia, and denervation, and even lead to iatrogenic compartment syndrome[15–18]. Because of this, some surgeons have proposed to turn back to the midline incision and modify the part of surgical procedures of CO-TLIF for mini-open spinal surgery[10–12]. Their research and our study have proved that these mMO-TLIF operations also achieve similar clinical efficacy as those MIS-TLIF operations.
In our mMO-TLIF via posterior midline approach, there are three main improvement elements, (1) The surgical anatomy is modified from extensive subperiosteal dissection to "targeted limited dissection". In addition to the symptomatic side of the laminectomy decompression area, the other areas only used Cobb periosteal stripper to perform blunt dissection of the fatty plane, reduce the use of electrotome and avoid the injury of the posterior ramus of the spinal nerve. (2) Adhere to the principle of micro invasion for "point exposure" during operation without special retractor equipment. (3) The operation procedure was described as a "step-by-step" process in four screw placement areas and one laminectomy decompression area. First, the contralateral screws were placed, and then the symptomatic side screws and decompression were performed.
There are many researches about MIS-TLIF compared with CO-TLIF in the treatment of lumbar degenerative diseases. Most of these studies pointed to the obvious advantages of MIS-TLIF operation in terms of intraoperative bleeding and hospital stay. However, there was no significant difference in operation time and blood loss in most works of literature, even worse than that of CO-TLIF[19–22]. Coincidentally, the mMO-TLIF operation based on the CO-TLIF performed better than MIS-TLIF in terms of operation time and blood loss in our study. Moreover, mMO-TLIF surgery also overcomes the disadvantages of conventional surgery in terms of intraoperative bleeding and hospital stay. Although we did not set the CO-TLIF group as the control group, the perioperative parameters of the mMO-TLIF still showed a significant improvement advantage. The reason may be related to the improvement of the CO-TLIF technique and procedure. In the previous literature, the degree of incision swelling and incision length was rarely reported. Our study shows that the first 3 days after the operation is a serious period of swelling and dehydration, and detumescence drugs should be used properly. In addition, single incision and shorter incision length are also the advantages of mMO-TLIF.
ODI and VAS scores as the most commonly used evaluation indicators appear in most literature. The existing studies always showed that the scores of MIS-TLIF and CO-TLIF were significantly improved after the operation. However, when it comes to the comparison between groups at different time points, there is great divergence[21]. This is partly consistent with our current research results. Both mMO-TLIF and MIS-TLIF could significantly improve the clinical efficacy of the patients. However, there were no significant statistical differences in ODI and VAS scores between groups at the same time point. From a statistical point of view, we believe that mMO-TLIF has the same advantages as MIS-TLIF compared with CO-TLIF.
The previous literature evaluated the fusion rate between vertebrae by CT or X-ray[22]. Here, we choose CT as the evaluation standard. The fusion rate of the two groups at a 2-year follow-up was more than 90%, which coincided with other studies[23, 24]. In our research, the autogenous bone fragments made from decompressed lamina bone tissue are enough to provide ideal bone fusion without the need for additional BMP, allogeneic, or artificial bone. In addition, MRI is often used to estimate paraspinal muscle atrophy and fat infiltration, but there are few reports about MRI evaluation before and after TLIF. Luis Alberto Ortega-Porcayo et al. [25] retrospectively analyzed the MRI data of 11 patients who underwent unilateral MIS-TLIF and unilateral pedicle screw placement. By comparing the functional cross-sectional area of multifidus and erector spinalis on the operative and non-operative sides, it was found that MIS-TLIF through a mini-open tubular approach produced minimal paraspinal muscle damage. Wu et al. [26] evaluated the edema and atrophy of multifidus muscle with T2 weighted magnetic resonance imaging (MRI) at 3 different time points (preoperative, postoperative, and 1-year follow-up). They reported the safety and efficiency of a novel inextensible endoscopic tube for TLIF. Tian et al.[27] compared a modified MIS-TLIF and TLIF through the Wiltse approach with MRI score and atrophy rate of CSA and verified the advantages of the new modified operation. Our MRI results showed no significant difference in the paraspinal muscle atrophy and degree of fat infiltration between the preoperative and postoperative groups. This indicates that the mMO-TLIF has the advantages of MIS-TLIF, which can invade paraspinal muscle less.
This study has some limitations. First of all, there was no CO-TLIF operation as a control group. Although there is a lot of existing literature on the comparative study of MIS-TLIF and CO-TLIF, mMO-TLIF is a new operation method, and there is still a lack of randomized controlled parameters with CO-TLIF. Secondly, the sample size and follow-up time are relatively short. Undoubtedly, our results need to be further verified by long-term large sample size randomized controlled trials. In addition, all operations were performed by senior surgeons in the same therapy institution. Therefore, there is a lack of universality of technical capability, which may lead to partial data bias.