INTRODUCTION
Minimally invasive spine surgery (MISS) involves using small incisions, minimizing tissue destruction, and respecting tissue planes to achieve the same surgical goals as open spine surgery [1]. MISS came about in the 20th century as surgical instrumentation and imaging techniques for spine surgery continued to develop. The Williams microdiscectomy, described in 1978, modified the traditional 6-inch incision open approach for lumbar discectomy to a much less invasive procedure [2]. In recent years, the development of MISS has significantly expanded. Tubular retractors are commonly used in MISS and allow surgeons to operate on the spine through small ports. Endoscopic spine surgery is often dubbed the future of MISS and involves using an endoscope and associated instruments through one or 2 subcentimeter ports. Advanced imaging techniques such as magnetic resonance imaging and computed tomography (CT) scans, computer-assisted navigation, and robotics have also made minimally invasive approaches to spine surgery easier and safer for patients [3-5].
With small incisions and respecting tissue planes, MISS aims to minimize damage to muscle and surrounding structures. Improved patient-reported outcomes and faster recovery periods have been associated with the MISS approach. Pokorny et al. [6] concluded that MISS reduced blood loss, hospitalization time, complications, and surgical costs compared to open spine surgery. Droeghaag et al. [7] also confirmed the cost-effectiveness of MISS by concluding that minimally invasive transforaminal lumbar interbody fusion (MIS-TLIF) is more cost-effective compared to open transforaminal lumbar interbody fusion (OTLIF). The reduction in postoperative care that follows MISS contributes to its cost-effectiveness [8]. As technology and surgeon experience evolved, MISS techniques have been successfully used to treat complex conditions such as pediatric and adult scoliosis cases with significant curves [9,10]. Intradural extramedullary and metastatic tumors have also been treated safely and effectively with MISS [11-13].
Despite the many benefits of MISS, it is important to understand its potential complications in order to improve clinical outcomes. Research describing MIS-TLIF for degenerative disc disease, spondylolisthesis, and vertebral canal stenosis describes over 89 specific complications, with the most common being radiculopathy (range, 2.8%–57.1%), screw malposition (0.3%– 12.7%), and incidental durotomy (0.3%–8.6%) [14]. Recent literature suggests complications associated with endoscopic spine surgery include dural tears, iatrogenic cord injury, incomplete decompression or recurrence, postoperative hematoma, and postoperative mechanical implant failures [15-17]. A literature review and meta-analysis assessing outcomes of open versus MISTLIF concluded that MIS-TLIF led to decreased tissue injury, blood loss, and length of hospital stay [18]. The authors also state that MIS-TLIF is a safe substitute for open spine surgery in obese patients. Ultimately, patient selection, surgeon experience, and the evolution of MISS technologies and techniques, such as augmented reality (AR) and intraoperative navigation, will continue to impact the incidence of patient complications.
METHODOLOGY
1. Literature Search
A comprehensive systematic literature review followed the Preferred Reporting Items for Systematic Reviews and Metaanalyses (PRISMA) guidelines (ref), utilizing the databases PubMed, MEDLINE, Embase via OVID, and Cochrane. The review spanned publications from January 2013 through March 2024. We employed a strategic combination of MeSH (medical subject headings) and keywords to encompass a wide array of studies addressing MISS techniques and associated complications. Keywords included “Minimally invasive,” “MISS,” “Minimally Invasive Surgical Procedures,” “tubular,” “biportal,” “uniportal,” “spine,” “surgery,” “complications,” “lumbosacral region,” “cervical,” “thoracic,” “lumbar,” “postoperative complications,” and “intraoperative complications.” Targeted searches were also conducted to pinpoint studies on minimally invasive tubular, endoscopic tubular-assisted, uniportal endoscopic, or biportal endoscopic surgeries involving the cervical, thoracic, or lumbar regions. To ensure the inclusion of unique studies, duplicate records were systematically removed. The exact content and order of the search string queries can be found in Supplementary Material 1. Additionally, a Google Scholar hand-search was conducted to identify further articles that may have yet to be captured through the initial database searches.
2. Inclusion and Exclusion Criteria
Studies were selected if they involved adult patients undergoing MISS using tubular, uniportal endoscopy, or biportal endoscopy techniques and reported on intraoperative or postoperative complications. The review focused exclusively on studies quantifying the frequency of these complications. Excluded were non-English publications, abstracts, conference presentations, editorials, expert opinions, case reports, case series with less than 10 patients, and review articles. We also excluded papers on lateral techniques. For studies reporting overlapping patient cohorts, preference was given to the report with the larger cohort.
3. Study Selection Process
Fig. 1 is a PRISMA diagram that illustrates the study selection process. All articles resulting from the search were exported into Rayyan (Cambridge, MA, USA), where duplicates were identified and deleted. Rayyan is professional research software widely used by collaborators for ease of study selection decisions. The study selection involved a systematic process where 2 reviewers independently screened titles and abstracts against the inclusion and exclusion criteria. The full texts of potentially relevant articles were then retrieved and assessed similarly. This screening was preceded by a pilot phase and a consensus meeting to ensure methodological rigor at each stage of the review process.
4. Data Extraction
Data regarding the type of MISS approach, patient demographics, complication rates, and detailed descriptions of complications were extracted using a structured form. Two researchers independently reviewed the data to ensure accuracy and consistency. Discrepancies were resolved through discussion, with unresolved issues referred to senior authors (GE).
RESULTS
After an electronic database search through PubMed, Embase, MEDLINE, and Cochrane, 880 references were identified. After removing 81 duplicate references, the titles and abstracts of 799 potentially relevant articles were reviewed, and 529 were excluded based on relevance. Subsequently, 270 reports were sought for retrieval, and 62 could not be retrieved. Out of the 208 reports assessed for eligibility, 71 were excluded based on the following criteria: 32 had the wrong surgery, 34 had no outcome of interest, and 5 had the wrong study design. Ultimately, 137 new studies were included in the review (Table 1). This process is illustrated in the PRISMA diagram (Fig. 1).
1. Cervical
1) Tubular
A comprehensive literature review identified 5 key studies reporting on tubular cervical spine surgery complications. The most frequently reported complications included hematomas (n=3), transient nerve root palsy (n=3), dural tears (n=2), and transient hypesthesia (n=2). Ross [19] conducted a retrospective analysis with the largest cohort (n=262) and covered a 12-year period (2001–2013), noting complications such as C5 nerve root palsy (n=3, 1.1%) and transient hypesthesia (n=2, 0.7%).
2) Uniportal
Two studies examined complications associated with uniportal cervical spine surgery, involving a total of 63 patients. The most common complications reported were transient nerve root palsy/neurapraxia (n=4) and recurrence of symptoms (n=2). Kim et al. [20] included the largest cohort (n=38) and found cases of transient nerve root palsy (n=2, 5.3%) and symptom recurrence (n=1, 2.6%).
3) Biportal
Eight studies investigated the complications associated with biportal cervical spine surgery, involving 504 patients. The most common complications reported were dural tears (n=4), epidural hematomas (n=4), motor weakness (n=3), and incomplete decompression (n=3). Jung et al. [21] included the largest cohort (n=162) and reported 3 complications: motor weakness (n=2, 1.2%) and muscle soreness (n=1, 0.6%).
2. Thoracic
1) Tubular
Three studies examined complications associated with tubular thoracic spine surgery. Across all of the 69 patients included, there were observed instances of cerebrospinal fluid (CSF) leak (n=3), pseudomeningocele formation (n=1), and durotomy (n=1). The largest study by Ross [19] included 40 patients and reported a single durotomy (n=1, 2.5%).
2) Uniportal
Four studies examined complications associated with uniportal thoracic spine surgery involving 140 patients. The most common complications were dural tears (n=4), epidural hematomas (n=3), incomplete decompression (n=2), and transient intercostal neuralgias (n=2). Ruetten et al. [22] included the largest cohort (n=55) and found that 10 patients experienced at least one adverse event. The reported complications included dural tear (n=2), epidural hematoma (n=2), transient arm dysesthesia (n=1), transient intercostal neuralgias (n=2), deterioration of myelopathy (n=1), transient deterioration of myelopathy (n=1), transient leg dysesthesia (n=1).
3) Biportal
One study examined complications associated with biportal thoracic spine surgery. In their cohort of 14 patients, Deng et al. [23] found complications of hyperalgesia (n=2, 14.3%), head and neck pain (n=2, 14.3%), and CSF leak (n=1, 7.1%).
3. Lumbar
1) Tubular
Fifty-seven studies examined complications associated with tubular lumbar surgery. Across all 7,495 patients, the most common complications recorded were incidental dural injuries (n=266), postoperative neuropathic conditions (n=59), disc herniation recurrences (n=30), and urinary tract infections (n=27). Staartjes et al. [24] included the largest cohort of patients (n=1,241) within the tubular lumbar surgery group and recorded the following complications: durotomy (n=47, 3.8%), spondylodiscitis (n=4, 0.3%), iatrogenic nerve root lesion (n=3, 0.2%), wound infection (n=1, 0.1%), wrong side incision (n=1, 0.1%), conversion to open surgery (n=2, 0.2%) excessive blood loss >500 mL (n=1, 0.1%), and phlebitis (n=1, 0.1%).
2) Uniportal
Sixteen studies examined complications associated with uniportal lumbar surgery. Across the 1,146 patients included, the most common complication recorded was incidental dural injury (n=29). Lee et al. [25] included the largest cohort of patients (n=213) and reported the following complications: transient dysesthesia (n=12, 5.6%), motor weakness (n=3, 1.4%), and durotomy (n=6, 2.8%).
3) Biportal
Fifty-six studies examined complications associated with biportal endoscopic surgery. Across the 4,002 patients included, the most common complications recorded included incidental dural injuries (n=69), hematomas (n=39), and neuropathic conditions (n=34). Kim et al. [19] included the largest cohort of patients (n=797) and recorded the following complication: hematoma (n=5, 0.62%), lesion recurrence (n=16, 2%), incomplete operation (n=8, 1.0%), dural tear (n=3, 0.4%), instability (n=2, 0.3%), and infection (n=1, 0.1%).
Table 2 summarizes the common complications in MISS.
DISCUSSION
1. Intraoperative
1) Nerve injury
Nerve injuries arising from spinal surgery can be transient or permanent. They can have direct causes, such as trauma, which may include unintentional nerve contact with instruments, or indirect causes, such as hematomas or swelling. Patients often report pain, weakness, or numbness when experiencing a nerve injury. Compared to traditional open spinal surgery, a MISS approach provides the advantage of decreased anatomical disruption, which should theoretically diminish the probability of disturbing spinal nerves. On the other hand, MISS can sometimes provide a limited field of view and a smaller working field, hypothetically contributing to an increased risk of nerve damage. One systematic review and meta-analysis revealed 9 instances (0.24%) of transient nerve root injury complications in 3,673 biportal endoscopic spine cases [16]. A comprehensive review and meta-analysis evaluating MISS versus traditional open surgery for cervical and lumbar discectomy in 1,590 patients found that MISS had a higher incidence of nerve root injury compared to open surgery [26].
Given the small corridors that MIS surgeons work through, intraoperative neuromonitoring (IONM) can be critical for detecting potential nerve injury or damage in advance. IONM can track nerve function in real-time throughout a procedure. A recent retrospective study assessed the feasibility of using IONM for dysesthesia prophylaxis in endoscopic spine surgery. An experimental group underwent IONM with somatosenory evoked potentials and transcranial motor evoked potentials, while a control group did not. Postoperative dysesthesia occurred in 2 patients in the experimental group with IONM evidence of compression of the exiting nerve root’s dorsal root ganglion and in 6 patients in the control group. The authors believe this compression was due to the initial placement or manipulation of the endoscopic working cannula and instruments [27]. Li et al. [28] demonstrated that IONM is feasible within Kambin’s triangle using a stimulation probe under fluoroscopic and robotic guidance in 34 TLIFs. Vitale et al. [29] developed a checklist for items to consider in response to patient IONM changes. Overall, Laratta et al. [30] detected a 296% increase in the use of IONM in spine surgery between 2008 and 2014, relating the rise to 1 in every 200 IONM cases in which a consequent neurological deficit is avoided, and an estimated $120,000 is saved.
2) Durotomy
The literature points to a slightly higher incidence of symptomatic durotomies in open spinal surgery when compared with MISS. Lower durotomy rates in MISS could be due to the preservation of the paraspinal musculature with the smaller resultant dead space allowed for CSF to collect [31]. A systematic review assessing complications between MIS-TLIF and O-TLIF in 14 studies found 6 out of 455 and 7 out of 446 cases of dural tears, respectively [32]. Another study documented 6.4% versus 15.9% risk of durotomy (p<0.05) in 240 patients undergoing MISS and open spinal surgery, respectively [33]. One 10-year systematic review on unintended durotomies in lumbar degenerative spinal surgery reported a 7.20% risk in MISS versus 7.02% rate in open surgery; however, this difference was not statistically significant [34].
Ruban and O’Toole [35] reported on their management of incidental durotomies in MISS. They devised and followed a repair and management system for unintended durotomies. A primary repair was attempted using 4-0 Nurolon Ethicon sutures for full-thickness durotomies. Fibrin glue was used if a watertight closure could be achieved; otherwise, a muscle graft or collagen matrix was employed with fibrin glue. If primary repair was not feasible, blood-soaked gel foam and fibrin glue were used. For partial-thickness tears, only fibrin glue was applied. Two modified needle drivers and a bayoneted Chitwood Knot Pusher were used for all primary repairs, with bed rest emphasized for all cases.
Additionally, Boukebir et al. [36] reported that 9.9% of patients experienced incidental durotomies during tubular procedures. These tears were repaired intraoperatively with fibrin glue, DuraSeal, or direct closure, resulting in no postoperative CSF leaks or infections. The study highlighted a significant learning curve, with earlier procedures having higher dural injury rates (27.3%) than later ones (1.8%). Effective techniques included using a ball-tip instrument to separate the dura from the ligamentum flavum and the Scanlan endoscopic dural repair set with a 4-0 Nurolon TF-5 suture for primary repairs.
A multicenter study on managing dural tears in lumbar endoscopic surgery documented 698 durotomies in 64,470 cases (~1%) [37]. Most durotomies were repaired with sealants, with DuraSeal being the most commonly used sealant. Two-thirds of endoscopic spine surgeons who used sealants reported an uneventful postoperative course in their patients after an incidental dural tear. The remaining one-third observed issues related to durotomy in their patients but at a low incidence.
Telfeian et al. [38] reported 4 incidental durotomies in 907 patients who underwent transforaminal endoscopic spine surgery over 4 years. Of the 4, one was treated with the Durepair Regeneration Matrix (Medtronic, Minneapolis, MN, USA), and another with a Duragen patch (Integra, Princeton, NJ, USA). Of the remaining 2 durotomies, one was treated with 24-hour bed rest, and the other was left untreated because it was considered too small. No subsequent CSF leaks were reported.
Nam et al. [31] reviewed 3 incidental durotomies from endoscopic lumbar decompressions. All tears were repaired with TachoSil (Corza Medical, Westwood, MA, USA). They reported that all 3 patients improved after durotomy repair with Tacho-Sil, and no complications occurred after discharge during follow-up. Derman et al. [39] discussed 3 cases of incidental durotomies during uniportal endoscopic spine surgery successfully repaired with collagen matrix inlay graft. In a retrospective review of 27 patients with incidental durotomy following endoscopic stenosis lumbar decompression, Kim et al. [40] recommended endoscopic patch blocking for patients with less severe dural tears and good prognosis, and consideration of conversion to open repair in patients with more serious tears and with fair to poor prognosis. Heo et al. [41] reported successful dural tear repairs with nonpenetrating titanium vascular anastomosis clips in biportal endoscopic lumbar surgery.
With innovative materials and technologies, MIS surgeons have been able to treat incidental durotomies with much reported success. However, the difficulty of such repairs must be considered in that MIS is performed through a small tube, which limits the use of standard dural repair instruments [42]. Furthermore, minor tears may not always be identified by the surgeon and not require incidental durotomy repair kits. This leads to the potential underreporting of durotomy within the minimally invasive setting [43].
3) Hardware malplacement
Not much has been reported on the incidence of suboptimally placed instrumentation when comparing MISS and open spinal surgery. Schmidt et al. [44] reported a 90% screw placement accuracy with the use of a navigated MIS Single Step Pedicle Screw System. Other MIS studies have reported high screw placement accuracies when comparing preoperatively planned to implanted screws using imaging [45]. A recent review on longterm reoperation rates after open versus MIS surgery documented a 28% versus 14% rate [46].
Navigation is also a valuable tool that can prevent reoperation. Navarro-Ramirez et al. [47] reported using a portable intraoperative CT 3-dimensional (3D) navigation system for 117 cases of various indications in all spine regions. The authors achieved a pedicle screw placement accuracy of 99%.
Navigation also requires collaboration among surgeons, nurses, and technicians. The burden of set-up and calibration of equipment and verification of imaging data accuracy and surgical tool functionality is shared, making the technology quite easy to use over time. Virk and Qureshi demonstrate this increased accuracy in addition to decreasing intraoperative radiation exposure [48].
With respect to navigation and robotics, Guillotte et al. [49] reported 100% of pedicle screws and 100% of interbodies were placed satisfactorily using the Globus Excelsius GPS. No instrumentation required replacement or revision intraoperatively. Saway et al. [50] concluded that single and multilevel robotic endoscopic TLIF is a safe and efficacious approach with comparable outcomes to open and other minimally invasive approaches. AR is the newest technology on the market that can assist with accurate hardware placement in addition to multiple other benefits.
2. Postoperative
1) Infection
Spinal surgical site infections (SSIs) can be difficult to manage and may require multiple hospitalizations, prolonged antibiotic therapy, repeated surgeries for wound debridement, or implant removal [51]. MISS studies have documented favorable outcomes regarding SSIs. Kulkarni et al. [51] retrospectively reviewed 1,043 patients who underwent 763 noninstrumented MISSs and 280 MIS-TLIF procedures for an overall SSI rate of 0.29%. Mueller et al. [52] retrospectively reviewed 961 MIS and 481 open cases and reported statistically different SSI rates of 0.5% and 3.3% (p=0.0003).
Intraoperatively, intravenous antibiotic prophylaxis is effective for SSI reduction [53]. The administration of a broad-spectrum antibiotic such as cefazolin half an hour before the skin incision with readministration every 4 hours during long surgeries is standard practice [54]. Skin antisepsis also reduces the probability of patients developing SSIs [55]. Proven closure and dressing protocol and the reduction of intraoperative blood loss can also reduce the risk of SSIs [56]. O’Toole et al. [57] propose 4 possible intraoperative causes of reduced infection rates in MISS compared to open spinal surgery: reduced contamination surface area, prevention of skin flora contamination via tubular retractors, smaller incisions decreasing skin dehiscence risk, and reduced operative site dead space limiting postoperative wound seromas or hematomas, thereby potentially lowering the chance of SSIs.
2) Pain
An important objective of MISS is to decrease postoperative pain with smaller incisions and reduced tissue disruption. Hockley et al. [58] retrospectively reviewed 172 TLIF cases (109 open, 63 MIS) and reported a shorter operative time, decreased blood loss, and less inpatient opioid usage for MIS-TLIF. A systematic literature review that sought to assess return to work and narcotic use following MIS and open lumbar spinal fusions concluded that patients who underwent MISS generally returned to work quicker than patients who had open spinal surgery. MISS patients also required less postoperative narcotic use for pain control [59]. A subsequent narrative review reported improved outcomes with MIS-TLIF compared to open-TLIF regarding intraoperative bleeding, hospital stay, time to ambulation, postoperative narcotic use, and time to resume work [19]. Three Quality Outcome Database (QOD) studies have assessed MIS versus open approaches for fusions and decompressions. Mooney et al. reported that an MIS approach to lumbar spinal fusions was associated with a greater decrease in leg and back pain at threeand twelve-month follow-up time points [60]. A statistically significant decrease in leg pain was reported in MIS compared to open fusion for grade 1 degenerative lumbar spondylolisthesis [61]. The final QOD study assessing MIS versus open decompression for low-grade spondylolisthesis reported no significant differences in back or leg pain outcomes. However, both approaches led to favorable patient outcomes [62].
3) Long-term
With decreased muscle, bony and ligamentous disruption, MIS surgeons believe that a minimally invasive approach should decrease the risk of adjacent segment disease (ASD). Despite this, a systematic review assessing the long-term outcomes of MIS-TLIF and open-TLIF in 16 studies reported no significant difference in ASD at a minimum follow-up of 2 years (12.6% vs. 12.40%, p=0.50) [63]. However, a separate study by Li et al. [28] reviewed 9 trials comprising 770 patients and reported a significantly lower adjacent segment pathology (ASP) incidence rate in patients who underwent an MIS procedure compared with an open procedure (p=0.0001). Single-level lumbar interbody fusion was performed in 6 trials of 408 patients, and Li et al. [28] reported a lower ASP incidence rate in the MIS group (p=0.002) for these studies. Their pooled data analysis favored the MIS approach for ASD and adjacent segment degeneration avoidance.
A systematic review of 24 studies, including 2,496 patients evaluating open laminectomies and minimally invasive bilateral canal enlargement, reported that instability was observed most frequently in those with preexisting spondylolisthesis (12.6%) and those treated with an open laminectomy (12%) [64]. The review also concluded that instability following lumbar decompressions would likely occur less frequently using minimally invasive techniques.
Another recent review reported the following percentages for recurrent disc herniations: open discectomy: 4.8%, microdiscectomy: 5.1%, microendoscopic discectomy: 3.9%, and full endoscopic discectomy: 3.5% [65]. These numbers do not vary significantly. Therefore, no clear indication of which technique reduces recurrent disc herniation incidence exists.
4) AR in MISS
In MISS, AR is used to overlay relevant anatomical landmarks and preplanned screw trajectories on actual patient anatomy. Surgeons can experience real-time feedback of instruments in space and in relation to real anatomy [66]. AR projections can be displayed on wearable headsets or directly into the surgeon’s microscope with AR application software [11]. The advent of AR in spine surgery began with the development of the head-up display (HUD) system, which overlayed CT imaging on the eyepiece of the operating microscope [67]. The visualization of spinal tumor resection planes, in osteotomies, and during MIS transvertebral cervical foraminotomies with the HUD system has been well documented [68,69]. To address some of the HUD system’s constraints, AR-head-mounted displays (HMDs) were subsequently developed. The Augmedics Xvision Spine system is the first and only AR-HMD to have been approved by the U.S. Food and Drug Administration [70,71]. Companies like Brainlab have taken a similar approach to HUD systems utilizing special software to display AR projections into the operative microscope’s field of view.
AR can be combined with intraoperative navigation to facilitate surgical precision and drastically reduce the need for fluoroscopy. With Brainlab software, for example, AR anatomical and screw trajectory planning can be completed on a preoperative CT scan. Preoperative data can be fused with an intraoperative CT scan using elastic fusion software with digital correction. Following fusion, the planned imaging is projected into the operative microscope and adjusted according to the surgeon’s preferences [11]. Brainlab’s software provides 3D intraoperative navigation, which enhances visibility and effectively orients the surgeon. Using a navigated pointer, surgeons can verify exactly where they are located within a patient’s spine at any given moment. This feature is especially crucial for complex MIS cases.
Sommer et al. [11] combined the use of AR with “total 3D navigation” using Brainlab software to perform 10 MIS-TLIFs. Total 3D navigation, coined by Lian et al. [72], refers to the use of navigation for all steps of an MIS procedure, from pathology localization, incision planning to screw placement, tubular decompression, cage placement, and rod measurement without the need for any intraoperative fluoroscopy. The operating surgeon could observe all preoperatively planned imaging in the microscopic field of view. AR implementation added only 1.3±0.37 minutes to the overall procedure time, and there were no reported intraoperative or postoperative complications. All 10 patients reported favorable outcomes at an average of 8.4±2.4 months. In a separate study, Sommer et al. [11] also reported on using AR with navigation for MIS and open resection of benign intradural extramedullary tumors—preoperative AR planning involved marking tumor margins with the Brainlab “smart brush” function. All surgeries were successful as with the TLIF cases performed with AR and navigation.
5) Avoiding complications with AR in MISS
AR’s surgical mapping feature is perhaps the strongest tool that can be used in the surgeon’s attempt to avoid intraoperative and postoperative complications. As mentioned, MIS surgeons can use software such as Brainlab Elements to study patient images in 3D and accurately mark or outline tumors and structures at risk, and plan trajectories. Anatomy varies between patients; therefore, experience with approaches to spinal pathology can fall short. With the added benefit of in-depth interaction with and reviewing patient anatomy and disease preoperatively, surgeons can foresee potential intraoperative difficulties and be prepared for challenges.
With the projection of these preoperatively planned models into a microscope or headset, MIS surgeons can be precise with their techniques—many complications in MISS stem from unnecessary contact with unaffected anatomy. For example, screws placed not completely within the pedicle can cause serious neurologic or vascular injury or a CSF leak [73,74]. With enhanced preoperative study and intraoperative visualization of vertebral body anatomy, surgeons are more likely to be more accurate with screw implantation and avoid lateral, medial, cranial, or caudal breaches [70,75,76]. Butler et al. [77] discuss the potential of minimizing damage to key structures surrounding the level(s) operated on in the prevention of ASD. AR can easily facilitate that objective. In complex cases such as tumor resections where the risk of complication is higher, AR has been proven to be a valuable tool [11,78].
With AR in MISS, surgeons can observe patient anatomy in 3D projected over the standard field of view. An enhanced perception of critical structures, including nerve roots and the dura, can give surgeons a better sense of confidence and control in complex cases where the probability of compromising such delicate structures is higher. AR can also be coupled with intraoperative navigation to improve surgical workflow further and avoid complications.
In addition to its direct operative benefit, AR is a valuable resource for surgical education. Schmidt et al. [79] assessed the utility, accuracy, efficiency, and precision of AR-guided MIS-TLIF and sought to determine the technology’s impact in spine surgery training. Twelve neurosurgical residents at 2 sites performed a 1-level TLIF on a high-fidelity lumbar spine simulation model, each with and without AR projection (which included highlighted landmarks) into a microscope. The National Aeronautics and Space Administration (NASA) task load index was administered to all residents postoperatively. Results demonstrated that AR-guided procedures consistently impacted resident anatomical orientation and workload experience. According to the NASA task load index, procedures completed without AR required a significantly higher mental demand (p=0.003) than with AR. Residents also reported that it was significantly more difficult for them to accomplish their level of performance without AR (p=0.019). Virtual reality (VR) also has important educational value. In a separate study, neurosurgery residents completed minimally invasive lateral interbody fusions using a VR system for 3 simulations over 6 weeks. Performance scores improved for the majority of participants. All participants also demonstrated improved comfort with important surgical landmarks and confidence in performing the procedure without supervision [80].
The adjustment to using AR in MISS is the toughest challenge of implementing the technology into practice. Kann et al. [81] have completed extensive research on the use of the AR-HMD. They describe the biggest challenge of its integration into practices as the limited number of experienced surgeons proficient in its usage since the technology is relatively new. To become experienced with AR technology, surgeons need to undergo extensive training. No standardized curricula for AR/VR surgical tools are endorsed by major neurosurgery or orthopedic specialty organizations. Those more advanced in their careers and have built successful practices without modern technology like AR may also be more reluctant to implement it. Adopting AR into surgical workflow also challenges operating room (OR) staff. OR anesthesiologists, nurses, and techs must also consider their role with newly implemented technology and how their routine will change.
Acquiring and maintaining AR/VR systems is expensive. Software, training, and technical support can become burdens for hospitals, especially those not well-resourced or funded. High initial costs associated with high-fidelity simulators have already led to significant variability in simulator availability across institutions in the United States [82]. While AR is costly, there are clearly significant benefits to its adoption. AR devices can also challenge patient data safety [83]. These systems can collect and store vast patient information on company servers. Given the novelty of these systems, there is also a lack of developed safeguards against attacks on sensitive patient information. More work must be done to do so.
STRENGTHS AND LIMITATIONS
The primary strength of this narrative review is its comprehensive evaluation of complications in MISS across cervical, thoracic, and lumbar regions. By incorporating data from a wide range of studies, the review provides a detailed understanding of complication profiles associated with different MISS techniques, including tubular, uniportal, and biportal endoscopic surgeries, at each spine region. The reviewed literature includes data from various countries, surgical settings, and patient populations over the past decade, providing a broad and diverse perspective.
Despite its strengths, this review has several limitations. Many included studies were retrospective, introducing selection bias and limiting causal inferences. Heterogeneity among studies regarding patient populations, surgical techniques, and outcome measures poses a challenge in drawing definitive conclusions. Furthermore, variability in surgeon experience with MISS techniques, particularly for newer approaches like biportal endoscopy, complicates the assessment of complication rates. The subjective nature of some reported complications, such as postoperative headaches, introduces potential interviewer bias. Additionally, some studies’ lack of detailed patient demographics and clinical characteristics limits understanding of their impact on complication rates. Future research should incorporate prospective, randomized controlled trials with standardized complication reporting and detailed patient characteristics.
CONCLUSION
Complications in any surgical procedure are inevitable. With MISS, surgeons recognize this fact and seek to diminish this risk with approaches that preserve anatomy and stability while targeting indicated degenerative pathology. Various studies demonstrate that MISS provides many benefits to patients that, in many cases, are superior to open surgery. With MISS, there is ample room for innovation to minimize potential complications, as demonstrated by novel, proven approaches like endoscopy and technological advancements such as AR/AR and robotics. As MISS develops rapidly, surgeons and industry partners must continue to reassess aids and technologies and keep the consequential mission of providing patients with the best spine care preeminent in all considerations.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement
Print
