Minimally invasive versus open transforaminal lumbar interbody fusion: a prospective, controlled observational study of short-term outcome

The treatment of lumbar degenerative spinal disk disease with or without segmental instability is still a matter of debate [32]. Nevertheless, the rate of lumbar fusion surgery is rising dramatically compared to other musculoskeletal surgical procedures [6]. From 2004 to 2015, there has been a 62% increase in the USA regarding lumbar fusion procedures for degenerative conditions, with approximately a 32% increase in the population rate [30]. Although surgical treatment has shown advantages compared to conservative treatment in degenerative spondylolisthesis [13, 33], patient self-reported outcome data after spondylodesis for degenerative conditions shows unsatisfactory results in 30–40% of patients in randomized controlled trials [4, 13]. Moreover, complication rates of up to 20% in the often elderly population were reported in multisegmental procedures [5]. Complication rates appear to increase with age, blood loss, duration of surgery, and the number of levels treated.

With the introduction of minimally invasive spine surgery (MISS), better outcomes were expected by substantially reducing tissue damage as well as achieving the same treatment goals compared to a traditional open approach [23, 56]. Over time, many different MIS-TLIF techniques were described, which differ in the type of access or in the type of retractor system used [28, 53]. In general, MIS-TLIF is defined by some key features including the use of a nonexpandable or expandable tubular retractor, performing a paramedian or lateral incision, and the use of a microscope or endoscope for visualization [28]. With these special techniques, it is possible to approach the same anatomic landmarks as in open lumbar surgery, without using big skin incisions with massive muscle trauma [23, 53, 56], leading to more favorable results regarding intraoperative blood loss, need for blood transfusions, shorter hospitalization times, and less postoperative pain [21, 25, 43].

Nowadays, MISS is constantly increasing. Analyses of the few comparative studies suggest that MISS, in comparison to open surgery, does not show any advantages with regard to the most common clinical outcome instruments like ODI and VAS at follow-ups of 6 month and longer [21]. These results indicate that potential advantages of MISS are mainly found in the early postoperative phase. Novel techniques, however, are often associated with significant publication bias and it has been argued that MIS-TLIF requires more nerve root manipulation [7] and/or they lead to insufficient decompression. These parameters have hardly been studied prospectively so far, although they may be most relevant for elderly and comorbid patients. These patients would benefit significantly from decreased blood loss and earlier mobilization postoperatively [42], but would suffer from increased radicular problems. These differences might also have a considerable socio-economic impact, as recent investigations report the majority of cost savings due to a more rapid mobilization and discharge after MISS [1, 15, 30, 41].

The COMOSA (comparing a minimally invasive to open instrumented spondylodesis approach) study was designed as a single-center prospective cohort study and was intended as a preliminary study on evaluation tools for a randomized, controlled study. This study was approved by the local ethics committee (ID: UN4624; session number 310/4.6) to ensure the standards of good clinical practice (GCP) and was performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki. All participants signed a written consent form. Afterwards, study participants were subdivided into a minimally invasive (MIS) and an open (O) non-randomized cohort, so that the group assignment, whether MIS or O, was done by the surgeon. Only surgeons, who are experienced in both techniques, performed the operations. The O-TLIF (open transforaminal lumbar interbody fusion) group was treated using a traditional open procedure [16, 17]. The MIS-TLIF (minimal invasive transforaminal lumbar interbody fusion) group was treated by performing the essential steps of MIS-TLIF including decompression, with central or bilateral decompression (using the over-the-top decompression technique), discectomy, and interbody graft insertion using an expandable retractor system (Pipeline; DePuy Synthes Spine, PA, USA), as well as pedicle screw and rod placement (VIPER® 2 MIS Spine System; DePuy Synthes Spine, PA, USA) [28]. The interbody cage and the corresponding screws were placed from one side through a mini-open approach using an oblique cage system (CONCORDE® interbody system; DePuy Synthes Spine, PA, USA). The contralateral screws were placed percutaneously. The postoperative management was not affected by the treatment group and followed institutional standards. The purpose of the study was the observation of the surgical and the early postoperative period up to 12 weeks. A set of outcome instruments that are well established and validated in geriatric patients such as the Timed Up and Go test (TUG) and the Barthel Index (BI) were chosen to investigate the postoperative short-term course [40, 48].

Advantages and disadvantages of MIS-TLIF have been controversially discussed, as the results vary widely in the literature and publication bias may be an potential issue [14]. Before minimally invasive techniques in spinal surgery gained popularity, the traditional open approach represented the “gold standard” for lumbar instrumented fusion. This open approach involves large incisions and deattachment of healthy muscle tissue from bone resulting in increased muscle trauma, blood loss, and associated postoperative pain [39]. This muscle trauma leads to muscle edema, decreased muscular performance, and potentially denervation [24, 31]. These findings might be associated with prolonged hospital stays and increased postoperative complications [36, 46]. To overcome these problems, the minimally invasive approach for lumbar instrumented fusion has become increasingly attractive, as it provides potential benefits in terms of reduced paraspinal muscle trauma, resulting in decreased loss of blood, faster recovery rates, and reduced surgical site infection rates [39, 45, 55]. In contrast, the long-term observations comparing minimally invasive to open TLIF procedures failed to reveal significant differences, so that the beneficial effect of MIS procedures seems to be in the first weeks after surgery. Some studies examined functionality and pain after 3 months [35, 43], but the short-term period with analysis of multiple time points during the first 12 weeks postoperatively is still underreported. One study reported the average time until walking or standing-up postoperatively with MISS was 3.2 days compared to 5.4 days with open surgery [47]. Another study reported a significant reduction of muscle injury and systemic inflammatory markers during the acute postoperative period with MISS [25]. The authors suggested that MISS may play an important role in preventing medical morbidity after spinal surgery [25]. Others reported decreased pain, stress, fatigue, and mood disturbance 6 weeks postoperatively [50] and reduced surgical site infections [5, 37, 49] compared to patients who underwent open surgery. Postoperative narcotic use and return to work were found to be the most clinically relevant factors as both reduced twofold by MISS [1]. Despite these few reports about the postoperative short follow-up, there is no comparative study that focusses on the first 12 weeks postoperatively in detail.

Our study reports advantages of the MIS-TLIF procedure regarding operative time, intraoperative as well as postoperative blood loss, and some functional outcome scores. In terms of VAS and NRS, the MIS group revealed significantly more leg pain but a trend towards faster improvement of back pain compared to the open procedure. The patient-reported outcome measures (PROMs) demonstrated no statistically significant intergroup difference after 12 weeks.

Significantly more leg pain (VAS and NRS leg) of patients treated with MIS-TLIF was apparent at nearly all postoperative time points within the evaluated period of 12 weeks. The cause of increased leg pain in the MIS cohort may be explained by the greater nerve root retraction due to the minimally invasive approach. In the case of an open TLIF procedure, the facet joint is commonly widely resected and Kambin`s triangle may be approached with less retraction of the traversing and exiting nerve root as well as the dural sac without a predefined lateromedial angulation of a minimally invasive installed retractor [53]. By choosing a too medial MIS-TLIF approach, the dural sac and the traversing nerve root might be stressed. In case of a far lateral approach, Kambin’s triangle is approached in a flat angle leading to a small working corridor for cage implantation as well as potential irritation of the exiting nerve root. Two previous studies examined VAS leg after 6 and 24 months without significant differences [1, 38]. In contrast, immediate postoperative low back pain showed no significant differences between the two cohorts, even though back pain tended to improve faster in the minimally invasive treatment group potentially related to less muscle trauma. Our results are in line with the existing literature observing the long-term follow-up, as most studies have observed less postoperative pain in the MIS-TLIF cohort [10, 39] than in the open group mostly, without statistically differences.

In addition to that, the learning curve of MIS procedure cannot be neglected. MIS-TLIF requires mastering new techniques, which is thought to increase the length of operation due to the learning curve and also effects other perioperative factors such as complication rate [26, 34].

Functionality and disability are important factors after spinal surgery, so that most of the published studies evaluated the Oswestry Disability Index between MIS and open procedures between 3 and 24 months postoperatively [9]. To summarize their long-term findings (3 and 24 months postoperatively), MIS patients showed lower postoperative ODI scores compared to open-treated patients at nearly all study time points [11, 35, 47]. We can mostly agree with these findings, but a 1:1 comparison could not be drawn due to the lack of studies evaluating the short postoperative course. Focusing on the fact that the ODI depends on patients’ responsiveness, we additionally used the TUG test. There are recent studies demonstrating the reliability and the ease use of the TUG test [18, 51] for the measurement of functionality in case of spinal disorders. Considering the TUG, the MIS cohort also tended to outperform the open cohort. These findings might be also explained by the minimized surgical muscle trauma leading to reduced postoperative low back pain and consequently to a higher functionality state [39].

Summarizing the majority of studies in terms of surgical time, there were no significant differences between MIS and open TLIF [15, 26, 54], which coincide with our observation. Nevertheless, duration of surgery might only have a limited significance on the superiority of one technique, as it strongly depends on the surgeon’s routine [34].

Regarding peri- and postoperative blood loss as well as complication rate, we can support the results of prior studies [10, 22, 26], as in our observation the MIS cohort had a significantly reduced blood loss and no statistically significant difference in terms of complication rate. However, it should be mentioned that there are also many cofactors (multi-level fusion, preoperative hemoglobin, male gender, and body mass index) related to a higher loss of blood [20].

In terms of PROMs, they are less frequently recorded in the existing literature regarding the comparison of MIS and open TLIF procedures. In general, the most common PROMs consist of ODI, COMI, GDS, and health status questionnaire as well as the EQ-5D. Recent studies only reported on the Short-Form 36 (SF-36) [27, 44, 46], the Short-Form12 (SF-12) [27], and the EQ-5D [1, 37]. These findings showed no statistical significance in any of the reviewed studies in the long-term follow-up [15]. This observation also applied for the short-term follow-up within 12 weeks, without significant differences between both groups.

This study showed similar postoperative outcomes for the open versus the MIS-TLIF approach, yet some relevant differences were demonstrated. Especially regarding functionality, the MIS cohort had better outcomes in ODI and TUG scores but seems to carry a risk of increased nerve root manipulation, as shown by increased postoperative leg pain. In experts, this risk may be negligible, but for beginners, it might be reduced by the use of a distractable cage and by the increased use of training simulators to get more experienced in MIS techniques. Nevertheless, further randomized, controlled trials and a long-term follow-up are necessary to provide advantages and disadvantages of the long-term effects of these two techniques.