Eur Spine J (2015) 24:800–809 DOI 10.1007/s00586-015-3796-2
ORIGINAL ARTICLE
Incidence of vascular complications during lateral lumbar interbody fusion: an examination of the mini-open access technique Janina Kueper • Gary A. Fantini • Brendon R. Walker Alexander Aichmair • Alexander P. Hughes
•
Received: 13 August 2014 / Revised: 2 February 2015 / Accepted: 2 February 2015 / Published online: 10 February 2015 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose This article examines the incidence and management of vascular injury during Lateral Lumbar Interbody Fusion (LLIF). The details of the mini-open access technique are presented. Methods A total of 900 patients who underwent a LLIF at an average 1.94 levels (range: 1–5 levels) by one of six fellowship trained surgeons on 1,754 levels from 2006 to 2013 were identified. The incidence of intraoperative vascular injury was retrospectively determined from the Operative Records. The management of vascular injury was evaluated. The mini-open access adapted by our institution for LLIF is described. Results The incidence of major vascular complication in our series was 1/900. The incidence of minor vascular injury was 4/900. The overall incidence of vascular injury was calculated to be 0.056 % per case and 0.029 % per level. All minor vascular injuries were identified to be segmental vessel lacerations, which were readily ligated under direct visualization without further extension of the incision with no clinical sequelae. The laceration of the abdominal aorta, the major vascular complication of this series, was emergently repaired through an exploratory laparotomy. None of the patients suffered long-term sequelae from their intraoperative vascular injuries. Conclusions The mini-open lateral access technique for LLIF provides for minimal risk of vascular injury to the lumbar spine. In the rare event of minor vascular injury, the mini-open access approach allows for immediate
J. Kueper (&) G. A. Fantini B. R. Walker A. Aichmair A. P. Hughes Spine and Scoliosis Service, Hospital for Special Surgery, 535 East 70th Street, New York, NY 10021, USA e-mail: [email protected]
123
visualization, confirmation and repair of the vessel with no long-term sequelae. Keywords Lateral Lumbar Interbody Fusion Complications Vascular injury Mini-Open Approach
Introduction Lateral Lumbar Interbody Fusion (LLIF) was introduced in 2006—a minimally invasive technique of effective decompression and fusion of the lumbar spine [1]. LLIF was in part designed to circumvent the challenges experienced during direct transabdominal anterior or anterolateral retroperitoneal approaches to the lumbar spine. Vascular injury is generally deemed a secondary concern of surgeons performing a LLIF due to the lack of mobilization or manipulation of the great vessels required for direct lateral approaches to the lumbar spine [2]. Nonetheless, the approach, particularly at L4–5, is in close proximity to the common iliac vein, ascending iliolumbar vein, and segmental vessels which are potentially vulnerable to injury with this approach. The lateral surgical technique approaches the spine via a direct lateral retroperitoneal and transpsoas dissection. Various methods of lateral access have been described. The original LLIF manuscript detailed a two-incision technique [1]. The surgeon makes the first ‘‘accessory’’ incision at the junction of the erector spinae and abdominal oblique muscles. This incision allows palpation and confirmation of the retroperitoneal space. The surgeon then makes the working incision over the mid-point of the operative disk level. The tissue dilators and an expandable retractor are introduced through this working incision. The disk space is then approached through the psoas with the help of electrophysiological monitoring to avoid
Eur Spine J (2015) 24:800–809
injury to the lumbar plexus. New additional access techniques have since been described. Aghayev and Vrionis described a single incision access technique where the psoas muscle was retracted, rather than being split, to minimize postoperative sensory and lumbar plexus motor deficits [3]. Other surgeons have focused on direct visualization through a somewhat larger single incision and to avoid neural compromise due to their partial concealment [3–5]. Our institution has adopted an adaptation of these approaches— the mini-open lateral approach. While remaining minimally invasive, this approach allows for visualization, palpation, and neurophysiologic confirmation at the operative segment. It is theorized that this approach yields predictable and safe LLIF results while minimizing the risk to the neurovascular structures. Previously, our institution has published on sensorimotor complications [6–8]. This manuscript presents the incidence of vascular injury in the LLIF population and examines the mode of injury and management. This is the first technical report detailing the procedure of the mini-open access technique as it pertains to the management of vascular injury during LLIF.
Materials and methods The approval of our institution’s Institutional Review Board for this study was obtained. 900 Patients who underwent a LLIF at our institution from 2006 to 2013 utilizing a mini-open minimally invasive technique were identified. Operative reports and clinical records of all patients were reviewed retrospectively. Their demographic parameters and co-morbidities were collected. The presence of both major and minor vascular injury was explored. The operative levels and side of access were determined for all cases. Intraoperative data and postoperative data such as estimated blood loss as captured by anesthesiologic documentation, length of surgery, and length of hospital stay were recorded. The specific mode of injury and the subsequent treatment were analyzed. Mini-open technique for LLIF To avoid complications from the two-incision approach stemming from the manually guided insertion of tubular retractors with as well as the complications resulting from limited visibility, we have adopted a single incision miniopen approach for LLIF. Though inadequate for the treatment of major vascular injury as described in Case V, the approach has proven to provide a reliable surgical window for minor vascular injury management. The patient is positioned in the traditional lateral decubitus position. With regard to incision site, the principal determinant of laterality is ease of access to the L4–L5 disk
801
space. This, in turn, is dictated by coronal angulation of the L4–L5 disk space, which typically corresponds with the concavity of any lumbar curve that may exist. In the absence of same or when L4–L5 is not a target disk, operative site may be ipsilateral to any pre-existing lower extremity symptoms. The incision is centered diagonally from the upper left angle of the vertebral body distal to the index disk space to the lower right angle of the vertebral body proximal to the index disk space. It spans approximately 3 cm for a single-level LLIF, and is lengthened in small increments if additional adjacent levels are operated on. A muscle splitting approach is used, whereby external oblique, internal oblique and transversus abdominis muscles are individually split in minimal fashion, along the direction of their respective fibers. The surgical technique is one of blunt dissection, involving the use of Kittner dissectors and Wylie renal vein retractors of varying length. Every effort is made to minimize or avoid use of electrocautery, thus reducing the likelihood of denervation of the abdominal wall [8, 9]. Retroperitoneal space is then entered under direct vision and the visceral sac rotated medially. The rationale underlying the mini-open approach is tripartite and based upon 1. 2. 3.
Visualization, Palpation, and Electrophysiologic neurologic confirmation
With the psoas muscle under direct vision, palpation of the target disk space can be safely performed, as it is generally elevated relative to the mid-portion of the adjacent vertebral bodies (Fig. 1). In the setting of disk space collapse, a trough will be palpated (Fig. 2). The contour of the spine will provide tactile information as to the approximate center of the disk. The safety of the anatomic position relative to the neural elements is then confirmed with the handheld nerve detection apparatus (NeuroVision, NuVasive, Inc., San Diego, USA) (Fig. 3). Psoas fibers are then bluntly split with Kittner (peanut) dissectors and the disk space exposed through the use of Wylie renal vein retractors. Following fluoroscopic confirmation of level, handheld retraction is replaced by a self-retaining table mounted, or vertebral body anchored, retractor system. The Segmental Lumbar Artery and Vein may subsequently be visualized horizontally coursing across the approximate midline of the vertebral body. This final important component of the visualization of the index disk space allows for a safe and timely removal of the disk, without injury to the surrounding vasculature. For the remainder of the LLIF, the disk material is removed and the surgeon is able to position one cage of a size which suits the vertebral body diameter of the patients through the mini-open incision (CoRoent XL, Nuvasive, Inc. San Diego, USA) (Figs. 4, 5).
123
802
Eur Spine J (2015) 24:800–809
Fig. 2 Palpation of the disk space
Fig. 1 Visualization of the disk space
Technique of minor vessel ligation While challenging to control through any minimal access and deep incision, the mini-open approach is much more versatile in this regard. Principles of vascular control are similar to those described for the repair of iliac vein injury occurring during the course of traditional anterior spinal surgery [25]. Initial control of bleeding is obtained through direct compression of the injured vessel with a Kittner (peanut) dissector, against the aortic/caval aspect of the vertebral body. The vessels can then be looped with a right angle clamp, elevated from the vertebral body with a silk tie, doubly clipped on either side and divided. Segmental
123
Fig. 3 Electrophysiologic confirmation of a secure position
Eur Spine J (2015) 24:800–809
803
Fig. 4 a Preoperative lateral X-ray of a patient receiving a one-level stand alone LLIF at L4–5. b Preoperative anteroposterior X-ray of a patient receiving a one-level stand alone LLIF at L4–5. c Postoperative lateral X-ray of a patient receiving a one-level stand alone LLIF at L4–5. d Postoperative anteroposterior X-ray of a patient receiving a one-level stand alone LLIF at L4–5
vessels must at times be ligated without prior injury. This is generally the case in scenarios where they are deemed to be at imminent risk, for example, in the case with corpectomies, wedged vertebraes or bridging osteophytes on top of the vessels. Though segmental vessels may at those times be ligated en passant, great care is usually taken to preserve the vessels whenever possible.
Results Patient cohort Six fellowship trained spinal surgeons performed LLIF on a total of 900 patients at 1,754 levels between June, 2007
and September, 2013. The 372 men and 528 women were 62 years old on average (range: 27–91 years), with an average BMI of 28.47 (range: 17.7–48.2). A total of 476 patients were operated on through a right-sided approach, 422 patients through a left-sided approach, and two patients through a combined left and right approach for separate levels. One patient received a fusion at the level of T9–10, 16 patients at the level of T12–L1, 163 patients at the level of L1–2, 436 patients at the level of L2–3, 584 at the level of L3–4, 553 patients at the level of L4–5, and one patient at the level of L5–S1. On average, patients were operated on at 1.94 levels (range: 1–5 levels). Of the 900 patients, none of the LLIF’s was aborted due to the position of the segmental vessels or the ascending iliolumbar vein.
123
804
Eur Spine J (2015) 24:800–809
Fig. 5 a Preoperative lateral X-ray of a patient receiving a one-level instrumented LLIF at L3–4. b Preoperative anteroposterior X-ray of a patient receiving a one-level instrumented LLIF at L3–4. c Postoperative lateral X-ray of a patient receiving a one-level instrumented LLIF at L3–4. d Postoperative anteroposterior X-ray of a patient receiving a one-level instrumented LLIF at L3–4
Incidence of vascular injury Five patients sustained a vascular injury during the procedure. The incidence of vascular injury during LLIF performed at our institution was, therefore, calculated to be approximately 0.056 % per case and 0.029 % per level.
123
Four patients sustained a Segmental Artery Injury. One patient sustained a laceration of the abdominal aorta [19]. One case of segmental artery injury was diagnosed postoperatively. The patient had to return to the operating room for acute management of the bleeding vessel. All minor vessel injuries, once identified, were readily ligated with no
Eur Spine J (2015) 24:800–809
long-term sequelae. Due to the nature of the mini-open technique, the incision was not required to be lengthened for the segmental artery repair in any case. The aortic injury was repaired via an extended incision with no longterm sequelae. Vascular injury case reports Patient details are presented in Table 1. Amongst the five patients, the average number of operative levels was 2.2 (1–4). The BMI (range: 26.2–39.2) and the age of the patient at the time of surgery (range: 50–69) varied considerably. The average intraoperative estimated blood lost of the four cases of minor vascular injury was 700 ml (range: 100–1,500 ml). The intraoperative estimated that blood lost during the major vascular injury was 1,200 ml. The postoperative course of all patients was uneventful. No long-term sequelae from the intraoperative vascular injuries were recorded. Case I A 69-year-old male presented to the outpatient clinic with severe axial lower back and limited ambulation. A three-level LLIF from L2 to L5 was recommended. The L4–5 disk space was approached first utilizing the mini-open lateral approach to the spine. After the annulotomy, the trialing and finally the insertion of the PEEK implant the retractor was removed. Bleeding was noted over the L4 vertebral body. Pressure was immediately applied using a peanut cottonoid. The diagnosis of segmental artery injury was made. The segmental vertebral artery over L4 was subsequently isolated, tied off with silk ties and secured with vascular clips. The remaining two levels were operated on without further incidents. Case II A 58-year-old female presented to the outpatient clinic with a history of thoracolumbar scoliosis and lumbar spinal stenosis. A three-level LLIF from L1 to L4 was recommended. The L3–4 space, the L4–5 space and the L1–2 space were operated on without incident. To conclude, the L2–3 disk space was visualized and found to have a large lateral osteophyte bridging over the segmental artery. To allow release of the space, an osteotomy was performed during which some bleeding of the segmental artery of L3 was noted. The bleeding vessel was skeletonized and ligated with vascular clips. The remaining surgery proceeded without incident. Case III A 72-year-old male presented to the outpatient clinic with a history of a facet cyst for which he underwent a
805
foraminotomy with a cyst excision. Despite providing a temporary relief, pain in his lower extremity symptoms progressively worsened. A single-level LLIF from L3 to L4 was recommended. Despite rigorous fixation, the patient’s morbid obesity caused a shift in the patient’s position which was noted upon the attempts of placing the retractors in the appropriate position. The retractors were subsequently removed so the patient could be repositioned into a true lateral position. At this point, significant bleeding from the retroperitoneal space through the site of incision was noted. The incision was packed with gauze and the patient was then repositioned. Diagnosis of an injury to the L3 segmental artery was made. The blood vessel was crossing the vertebral body with a low trajectory and had been injured by the placement of the shim device. To control the bleeding, the vessel was secured using vascular clips. The remaining surgery proceeded without incident. Case IV A 56-year-old male presented to the outpatient clinic with a history of posterior lumbar decompression and fusion with segmental spinal instrumentation from L3 to L5 through a Posterior Lumbar Interbody Fusion (PLIF). The radiographic examination revealed junctional degeneration of the L2–3 level above the previous fusion. A single-level LLIF from L2 to L3 was recommended. The course of surgery was uneventful. On the second postoperative day, the patient complained of abdominal pain. CT-Imaging revealed a left retroperitoneal hematoma occupying the left anterior and posterior pararenal space extending inferiorly into the pericolic gutter. The patient was transferred to a general hospital where the hematoma was evacuated by the attending vascular surgeon who made the diagnosis of an injury to the L3 Segmental Artery. Henceforth, the postoperative course was uneventful. Case V A 50-year-old female presented to the outpatient clinic with severe lower back- and left lower extremity pain [10]. A three-level LLIF from L2–5, a PLIF from L5–S1 and Posterior Instrumented Fusion from T12–S1 were recommended. After the successful operation on the L4–5 level, attention was turned to the L3–4 disk space. During the placement of the implant, the proximal aspect of the PEEK cage broke. The attempt to reposition the implant resulted in the distal violation of the end-plate and the anterior cortex of the L3 vertebral body. An immediate drop of blood pressure was noted. The flank wound was sealed and cardiopulmonary resuscitation was immediately initiated. The emergency laparotomy revealed a laceration of the aortic terminus. The hematoma which had accumulated
123
123
Male
69
28.5
Parkinson’s disease; Degenerative scoliosis (25 degrees); Grade 1 Spondylolisthesis of L4–5; Spondylosis
L2–5
Left
400
10
L4
Segmental artery
Uneventful
Sex
Age at Surgery (years)
BMI
Diagnosis
Levels
Side of approach
Est. blood loss (ml)
Length of stay (days)
Level of injury
Injured vessel
Postoperative course
Patient 1
Table 1 Demographic and operative patient details
Uneventful
Segmental artery
L4
2
100
Left
L2–3
Thoracic scoliosis (40 degrees); Lumbar scoliosis (45 degrees); degenerative disk disease; spinal stenosis; spondylosis
26.5
57
Male
Patient 2
Uneventful
Segmental artery
L3
19
1,500
Right
L1–5
Grade 1 Spondylolisthesis of L3–4; spondylosis; degenerative disk disease
29.5
58
Female
Patient 3
Uneventful
Segmental artery
L3
9
800
Right
L3–4
Junctional level degeneration above a prior posterior lumbar interbody fusion from L3–L5
39.2
73
Male
Patient 4
Uneventful
Abdominal aorta
L3
38
12,000
Right
L3–5
Degenerative scoliosis (35 degrees) with coronal compensation
26.2
50
Female
Patient 5
806 Eur Spine J (2015) 24:800–809
Eur Spine J (2015) 24:800–809
within the abdomen was evacuated and the aorta was repaired with 4–0 prolene sutures. Following the closure of the abdominal wall, the patient was repositioned and the L3–4 disk space was explored. The broken implant was removed in pieces.
Discussion The mini-open access technique was developed as a potentially safer approach to the lateral lumbar spine over blind or dilator-based systems. What was originally born out of necessity when dealing with complex lumbar deformity has proven to be helpful when dealing with minor vascular complications. This technique has proven helpful when injuries to the segmental lumbar vessels were encountered, allowing for a timely control of the bleeding without the necessity of lengthening the incision. Vascular injuries in general are a rare complication of LLIF, with early studies describing their experience with LLIF reporting no vascular injuries [2, 11]. However, in more complex deformity cases, the segmental artery and vein at times approach the disk space making blind dilator approaches in these cases higher risk. Advanced techniques requiring aggressive disk space distraction as in deformities and/or use of lateral osteotomies make an awareness of segmental artery and vein location at each segment mandatory. All of the minor vascular injuries reported were diagnosed to be segmental artery lacerations. Although all of our patients diagnosed with a segmental artery laceration received immediate vessel repair and suffered no long-term sequelae from the intraoperative complication, the consequences of this injury could have been catastrophic. The segmental artery supplies blood to the vertebral bony structure, the ribs and their surrounding tissue, posterior local and epidural elements as well as the nerve roots bilaterally [12]. Usually the segmental vascular course is simple. Two segmental arteries, which arise from the dorsal surface of the aorta, encircle the vertebral body and merge on the posterior surface. At the transverse process, both arteries bifurcate into an intercostal segment which courses to the ribs and a dorsal segment which branches into several smaller vessels travels to the posterior elements and through the neural foramen into the vertebral canal [13]. The vascular arrangement and origin of the segmental artery vary widely amongst individuals [14]. Exploring the mechanisms of injury presented in this manuscript may assist in future prevention or facilitate recognition and ligation. As discussed, Case I emphasizes the importance of an awareness of the segmental location in degenerative deformities. Due to osteophyte formation and vertebral body remodeling over time, the segmental
807
vessels are often within the operative field making them vulnerable to injury either by the retractor or instruments. The removal of retractors which caused the injury in Case I has been characterized as a possible culprit of hemorrhage in spinal surgery before and should be approached with equal care in LLIF as in other procedures [25]. Furthermore, in these cases, surgeons should be cognizant that the segmental may be incorporated in chronically formed osteophytes as described in Case II. When performing osteotomies, one must be prepared to encounter segmental bleeding and perform ligation. Rarely, aberrant vessel course can occur in the absence of deformity as illustrated in Case III. The greatest risk overall, however, lies in the lack of identification of hemorrhage, as described in Case IV. In addition to potentially fatal systemic consequences, a hematoma may result in devastating neurological deterioration [15]. Though the adaption of a surgical technique such as the mini-open strives to allow for better visualization, hidden hemorrhage may occur. Since the initial symptoms of postoperative spinal epidural hematoma have been found to be presented after a median time of 2.7 h, close monitoring of every patients neurological status in the immediate postoperative period is essential [16]. Beyond the minor segmental vessels, it is theoretically possible to injure the ascending iliolumbar vein. The ascending iliolumbar vein may be encountered during a leftsided approach to the L4–L5 disk space, and may traverse the disk space and/or the cephalad aspect of L5. The ascending iliolumbar vein may be at risk for injury during performance of LLIF in the traditional closed fashion, either during retractor placement or during diskectomy. The mini-open technique allows the visualization of a vessel should it fall within the operative field. The authors recall several lateral cases that were aborted to avoid injury to the vein. Although a minor vein (diameter 2–4 mm) it can be hard to control and ligate due to a position low in the pelvis. The incidence of a problematic position of the ascending iliolumbar vein is unknown. Another unnamed vein similarly at risk can sometimes be recognized at the L4–5 level, 2–4 mm in diameter, coursing cephalocaudad immediately anterior to the lumbar plexus. Injury to either of these venous structures during performance of LLIF in the traditional closed fashion can result in significant hemorrhage which would be difficult, if not impossible, to control in the absence of conversion to an open procedure or utilization of endovascular techniques. The mini-open approach allows for prompt identification of these venous structures and facilitates avoidance or ligation and division as necessary. Given the complexity of the vasculature surrounding the spine, we must acknowledge that injuries to vascular structures smaller than the segmental vessels may not have been recognized and therefore reported, though none of the
123
808
patients in our case series other than those discussed in the case presentations were noted to have vascular complications stemming from LLIF. Minimally invasive surgery has been theorized to have several distinct advantages over traditional approaches including a lower degree of postoperative pain, decreased estimated blood loss and length of inpatient hospital stay, a faster return to the preoperative degree of daily activity and an overall lower total hospital cost over a 60-day perioperative period when compared to open surgery [17–20]. Developments in minimally invasive spine surgery, such as the laparoscopic anterior approach, have been deemed relatively safe for patients and are frequently performed [21]. Despite this, the open transabdominal anterior approach for spinal fusion of the lumbar spine continues to have several disadvantages. Complications including visceral injury, large vessel bleeding and sexual dysfunction have been reported [22–24]. In a study performed at our institution, the incidence of vascular injury which most commonly affected major venous vessels during elective anterior spinal surgery was found to be 2.9 % [25]. The lateral approach to the lumbar spine offers an improved vascular safety profile over traditional transabdominal and retroperitoneal anterior lumbar approaches. The mini-open technique presented in this manuscript presents a safe and reproducible alternative to dilator-based systems. In the setting of complex lumbar deformity cases, there may be distinct advantages of the mini-open technique in terms of vascular complication mitigation. Further, there may be cost advantages with streamlining of specialized equipment. Additional prospective comparative studies examining different surgical approaches for LLIF are warranted in the future. Conflict of interest There were no sources of financial support or perceived conflicts of interest for this study.
References 1. Ozgur BM, Aryan HE, Pimenta L, Taylor WR (2006) Extreme Lateral Interbody Fusion (XLIF): a novel surgical technique for anterior lumbar interbody fusion. Spine J 6(4):435–443 2. Youssef JA, McAfee PC, Patty CA et al (2010) Minimally invasive surgery: lateral approach interbody fusion: results and review. Spine (Phila Pa 1976) 35(26 Suppl):S302–S311. doi:10. 1097/BRS.0b013e3182023438 3. Aghayev K, Vrionis FD (2013) Mini-open lateral retroperitoneal lumbar spine approach using psoas muscle retraction technique. Technical report and initial results on six patients. Eur Spine J 22(9):2113–2119. doi:10.1007/s00586-013-2931-1 4. Yuan PS, Rowshan K, Verma RB, Miller LE, Block JE (2014) Minimally invasive lateral lumbar interbody fusion with direct psoas visualization. J Orthop Surg Res 9(1):20. doi:10.1186/ 1749-799X-9-20 5. Hardenbrook MA, Miller LE, Block JE (2013) TranS1 VEO system: a novel psoas-sparing device for transpsoas lumbar
123
Eur Spine J (2015) 24:800–809
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
interbody fusion. Med Devices (Auckl) 6:91–95. doi:10.2147/ MDER.S43746 Pumberger M, Hughes AP, Huang RR et al (2012) Neurologic deficit following lateral lumbar interbody fusion. Eur Spine J 21(6):1192–1199. doi:10.1007/s00586-011-2087-9 Taher F, Hughes AP, Lebl DR et al (2013) Contralateral motor deficits after lateral lumbar interbody fusion. Spine (Phila Pa 1976) 38(22):1959–1963. doi:10.1097/BRS.0b013 e3182a463a9 Lykissas MG, Aichmair A, Hughes AP et al (2014) Nerve injury after lateral lumbar interbody fusion: a review of 919 treated levels with identification of risk factors. Spine J 14(5):749–758. doi:10.1016/j.spinee.2013.06.066 Fantini GA, Pawar AY (2013) Access related complications during anterior exposure of the lumbar spine. World J Orthop 4(1):19–23. doi:10.5312/wjo.v4.i1.19 Aichmair A, Fantini GA, Garvin S, Beckman J, Girardi FP (2013) Aortic perforation during lateral lumbar interbody fusion. J Spinal Disord Tech [Epub ahead of print] Rodgers WB, Gerber EJ, Patterson J (2011) Intraoperative and early postoperative complications in extreme lateral interbody fusion: an analysis of 600 cases. Spine (Phila Pa 1976) 36(1):26–32. doi:10.1097/BRS.0b013e3181e1040a Prince EA, Ahn SH (2013) Basic vascular neuroanatomy of the brain and spine: what the general interventional radiologist needs to know. Semin Intervent Radiol 30(3):234–239. doi:10.1055/s0033-1353475 Shimizu S, Tanaka R, Kan S et al (2005) Origins of the segmental arteries in the aorta: an anatomic study for selective catheterization with spinal arteriography. AJNR Am J Neuroradiol 26(4): 922–928 Berenstein A, Lasjaunias P, Ter Brugge KG (2001) Spinal and spinal cord arteries and veins. In: Berenstein A, Lasjaunias P, Ter Brugge KG (eds) Surgical neuroangiography. Clinical vascular anatomy and variations, vol 2nd. Springer, New York, pp 73–164 Kao FC, Tsai TT, Chen LH, et al (2014) Symptomatic epidural hematoma after lumbar decompression surgery. Eur Spine J [Epub ahead of print] Amiri AR, Fouyas IP, Cro S, Casey AT (2013) Postoperative spinal epidural hematoma (SEH): incidence, risk factors, onset, and management. Spine J 13(2):134–140. doi:10.1016/j.spinee. 2012.10.028 Regan JJ, Aronoff RJ, Ohnmeiss DD, Sengupta DK (1999) Laparoscopic approach to L4-L5 for interbody fusion using BAK cages: experience in the first 58 cases. Spine (Phila Pa 1976) 24(20):2171–2174 Khajavi K, Shen AY (2014) Two-year radiographic and clinical outcomes of a minimally invasive, lateral, transpsoas approach for anterior lumbar interbody fusion in the treatment of adult degenerative scoliosis. Eur Spine J 23(6):1215–1223. doi:10. 1007/s00586-014-3246-6 Smith AG, Capobianco R, Cher D et al (2013) Open versus minimally invasive sacroiliac joint fusion: a multi-center comparison of perioperative measures and clinical outcomes. Ann Surg Innov Res 7(1):14. doi:10.1186/1750-1164-7-14 Singh K, Nandyala SV, Marquez-Lara A et al (2013) A perioperative cost analysis comparing single-level minimally invasive and open transforaminal lumbar interbody fusion. Spine J 13:01723–01733. doi:10.1016/j.spinee.2013.10.053 Kaiser MG, Haid RW Jr, Subach BR et al (2002) Comparison of the mini-open versus laparoscopic approach for anterior lumbar interbody fusion: a retrospective review. Neurosurgery 51(1):97–103 (discussion 103-5) Rajaraman V, Vingan R, Roth P et al (1999) Visceral and vascular complications resulting from anterior lumbar interbody fusion. J Neurosurg 91(1 Suppl):60–64
Eur Spine J (2015) 24:800–809 23. Baker JK, Reardon PR, Reardon MJ, Heggeness MH (1993) Vascular injury in anterior lumbar surgery. Spine (Phila Pa 1976) 18(15):2227–2230 24. Flynn JC, Price CT (1984) Sexual complications of anterior fusion of the lumbar spine. Spine (Phila Pa 1976) 9(5):489–492
809 25. Fantini GA, Pappou IP, Girardi FP, Sandhu HS, Cammisa FP Jr (2007) Major vascular injury during anterior lumbar spinal surgery: incidence, risk factors, and management. Spine (Phila Pa 1976) 32(24):2751–2758
123
ORIGINAL ARTICLE
Incidence of vascular complications during lateral lumbar interbody fusion: an examination of the mini-open access technique Janina Kueper • Gary A. Fantini • Brendon R. Walker Alexander Aichmair • Alexander P. Hughes
•
Received: 13 August 2014 / Revised: 2 February 2015 / Accepted: 2 February 2015 / Published online: 10 February 2015 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose This article examines the incidence and management of vascular injury during Lateral Lumbar Interbody Fusion (LLIF). The details of the mini-open access technique are presented. Methods A total of 900 patients who underwent a LLIF at an average 1.94 levels (range: 1–5 levels) by one of six fellowship trained surgeons on 1,754 levels from 2006 to 2013 were identified. The incidence of intraoperative vascular injury was retrospectively determined from the Operative Records. The management of vascular injury was evaluated. The mini-open access adapted by our institution for LLIF is described. Results The incidence of major vascular complication in our series was 1/900. The incidence of minor vascular injury was 4/900. The overall incidence of vascular injury was calculated to be 0.056 % per case and 0.029 % per level. All minor vascular injuries were identified to be segmental vessel lacerations, which were readily ligated under direct visualization without further extension of the incision with no clinical sequelae. The laceration of the abdominal aorta, the major vascular complication of this series, was emergently repaired through an exploratory laparotomy. None of the patients suffered long-term sequelae from their intraoperative vascular injuries. Conclusions The mini-open lateral access technique for LLIF provides for minimal risk of vascular injury to the lumbar spine. In the rare event of minor vascular injury, the mini-open access approach allows for immediate
J. Kueper (&) G. A. Fantini B. R. Walker A. Aichmair A. P. Hughes Spine and Scoliosis Service, Hospital for Special Surgery, 535 East 70th Street, New York, NY 10021, USA e-mail: [email protected]
123
visualization, confirmation and repair of the vessel with no long-term sequelae. Keywords Lateral Lumbar Interbody Fusion Complications Vascular injury Mini-Open Approach
Introduction Lateral Lumbar Interbody Fusion (LLIF) was introduced in 2006—a minimally invasive technique of effective decompression and fusion of the lumbar spine [1]. LLIF was in part designed to circumvent the challenges experienced during direct transabdominal anterior or anterolateral retroperitoneal approaches to the lumbar spine. Vascular injury is generally deemed a secondary concern of surgeons performing a LLIF due to the lack of mobilization or manipulation of the great vessels required for direct lateral approaches to the lumbar spine [2]. Nonetheless, the approach, particularly at L4–5, is in close proximity to the common iliac vein, ascending iliolumbar vein, and segmental vessels which are potentially vulnerable to injury with this approach. The lateral surgical technique approaches the spine via a direct lateral retroperitoneal and transpsoas dissection. Various methods of lateral access have been described. The original LLIF manuscript detailed a two-incision technique [1]. The surgeon makes the first ‘‘accessory’’ incision at the junction of the erector spinae and abdominal oblique muscles. This incision allows palpation and confirmation of the retroperitoneal space. The surgeon then makes the working incision over the mid-point of the operative disk level. The tissue dilators and an expandable retractor are introduced through this working incision. The disk space is then approached through the psoas with the help of electrophysiological monitoring to avoid
Eur Spine J (2015) 24:800–809
injury to the lumbar plexus. New additional access techniques have since been described. Aghayev and Vrionis described a single incision access technique where the psoas muscle was retracted, rather than being split, to minimize postoperative sensory and lumbar plexus motor deficits [3]. Other surgeons have focused on direct visualization through a somewhat larger single incision and to avoid neural compromise due to their partial concealment [3–5]. Our institution has adopted an adaptation of these approaches— the mini-open lateral approach. While remaining minimally invasive, this approach allows for visualization, palpation, and neurophysiologic confirmation at the operative segment. It is theorized that this approach yields predictable and safe LLIF results while minimizing the risk to the neurovascular structures. Previously, our institution has published on sensorimotor complications [6–8]. This manuscript presents the incidence of vascular injury in the LLIF population and examines the mode of injury and management. This is the first technical report detailing the procedure of the mini-open access technique as it pertains to the management of vascular injury during LLIF.
Materials and methods The approval of our institution’s Institutional Review Board for this study was obtained. 900 Patients who underwent a LLIF at our institution from 2006 to 2013 utilizing a mini-open minimally invasive technique were identified. Operative reports and clinical records of all patients were reviewed retrospectively. Their demographic parameters and co-morbidities were collected. The presence of both major and minor vascular injury was explored. The operative levels and side of access were determined for all cases. Intraoperative data and postoperative data such as estimated blood loss as captured by anesthesiologic documentation, length of surgery, and length of hospital stay were recorded. The specific mode of injury and the subsequent treatment were analyzed. Mini-open technique for LLIF To avoid complications from the two-incision approach stemming from the manually guided insertion of tubular retractors with as well as the complications resulting from limited visibility, we have adopted a single incision miniopen approach for LLIF. Though inadequate for the treatment of major vascular injury as described in Case V, the approach has proven to provide a reliable surgical window for minor vascular injury management. The patient is positioned in the traditional lateral decubitus position. With regard to incision site, the principal determinant of laterality is ease of access to the L4–L5 disk
801
space. This, in turn, is dictated by coronal angulation of the L4–L5 disk space, which typically corresponds with the concavity of any lumbar curve that may exist. In the absence of same or when L4–L5 is not a target disk, operative site may be ipsilateral to any pre-existing lower extremity symptoms. The incision is centered diagonally from the upper left angle of the vertebral body distal to the index disk space to the lower right angle of the vertebral body proximal to the index disk space. It spans approximately 3 cm for a single-level LLIF, and is lengthened in small increments if additional adjacent levels are operated on. A muscle splitting approach is used, whereby external oblique, internal oblique and transversus abdominis muscles are individually split in minimal fashion, along the direction of their respective fibers. The surgical technique is one of blunt dissection, involving the use of Kittner dissectors and Wylie renal vein retractors of varying length. Every effort is made to minimize or avoid use of electrocautery, thus reducing the likelihood of denervation of the abdominal wall [8, 9]. Retroperitoneal space is then entered under direct vision and the visceral sac rotated medially. The rationale underlying the mini-open approach is tripartite and based upon 1. 2. 3.
Visualization, Palpation, and Electrophysiologic neurologic confirmation
With the psoas muscle under direct vision, palpation of the target disk space can be safely performed, as it is generally elevated relative to the mid-portion of the adjacent vertebral bodies (Fig. 1). In the setting of disk space collapse, a trough will be palpated (Fig. 2). The contour of the spine will provide tactile information as to the approximate center of the disk. The safety of the anatomic position relative to the neural elements is then confirmed with the handheld nerve detection apparatus (NeuroVision, NuVasive, Inc., San Diego, USA) (Fig. 3). Psoas fibers are then bluntly split with Kittner (peanut) dissectors and the disk space exposed through the use of Wylie renal vein retractors. Following fluoroscopic confirmation of level, handheld retraction is replaced by a self-retaining table mounted, or vertebral body anchored, retractor system. The Segmental Lumbar Artery and Vein may subsequently be visualized horizontally coursing across the approximate midline of the vertebral body. This final important component of the visualization of the index disk space allows for a safe and timely removal of the disk, without injury to the surrounding vasculature. For the remainder of the LLIF, the disk material is removed and the surgeon is able to position one cage of a size which suits the vertebral body diameter of the patients through the mini-open incision (CoRoent XL, Nuvasive, Inc. San Diego, USA) (Figs. 4, 5).
123
802
Eur Spine J (2015) 24:800–809
Fig. 2 Palpation of the disk space
Fig. 1 Visualization of the disk space
Technique of minor vessel ligation While challenging to control through any minimal access and deep incision, the mini-open approach is much more versatile in this regard. Principles of vascular control are similar to those described for the repair of iliac vein injury occurring during the course of traditional anterior spinal surgery [25]. Initial control of bleeding is obtained through direct compression of the injured vessel with a Kittner (peanut) dissector, against the aortic/caval aspect of the vertebral body. The vessels can then be looped with a right angle clamp, elevated from the vertebral body with a silk tie, doubly clipped on either side and divided. Segmental
123
Fig. 3 Electrophysiologic confirmation of a secure position
Eur Spine J (2015) 24:800–809
803
Fig. 4 a Preoperative lateral X-ray of a patient receiving a one-level stand alone LLIF at L4–5. b Preoperative anteroposterior X-ray of a patient receiving a one-level stand alone LLIF at L4–5. c Postoperative lateral X-ray of a patient receiving a one-level stand alone LLIF at L4–5. d Postoperative anteroposterior X-ray of a patient receiving a one-level stand alone LLIF at L4–5
vessels must at times be ligated without prior injury. This is generally the case in scenarios where they are deemed to be at imminent risk, for example, in the case with corpectomies, wedged vertebraes or bridging osteophytes on top of the vessels. Though segmental vessels may at those times be ligated en passant, great care is usually taken to preserve the vessels whenever possible.
Results Patient cohort Six fellowship trained spinal surgeons performed LLIF on a total of 900 patients at 1,754 levels between June, 2007
and September, 2013. The 372 men and 528 women were 62 years old on average (range: 27–91 years), with an average BMI of 28.47 (range: 17.7–48.2). A total of 476 patients were operated on through a right-sided approach, 422 patients through a left-sided approach, and two patients through a combined left and right approach for separate levels. One patient received a fusion at the level of T9–10, 16 patients at the level of T12–L1, 163 patients at the level of L1–2, 436 patients at the level of L2–3, 584 at the level of L3–4, 553 patients at the level of L4–5, and one patient at the level of L5–S1. On average, patients were operated on at 1.94 levels (range: 1–5 levels). Of the 900 patients, none of the LLIF’s was aborted due to the position of the segmental vessels or the ascending iliolumbar vein.
123
804
Eur Spine J (2015) 24:800–809
Fig. 5 a Preoperative lateral X-ray of a patient receiving a one-level instrumented LLIF at L3–4. b Preoperative anteroposterior X-ray of a patient receiving a one-level instrumented LLIF at L3–4. c Postoperative lateral X-ray of a patient receiving a one-level instrumented LLIF at L3–4. d Postoperative anteroposterior X-ray of a patient receiving a one-level instrumented LLIF at L3–4
Incidence of vascular injury Five patients sustained a vascular injury during the procedure. The incidence of vascular injury during LLIF performed at our institution was, therefore, calculated to be approximately 0.056 % per case and 0.029 % per level.
123
Four patients sustained a Segmental Artery Injury. One patient sustained a laceration of the abdominal aorta [19]. One case of segmental artery injury was diagnosed postoperatively. The patient had to return to the operating room for acute management of the bleeding vessel. All minor vessel injuries, once identified, were readily ligated with no
Eur Spine J (2015) 24:800–809
long-term sequelae. Due to the nature of the mini-open technique, the incision was not required to be lengthened for the segmental artery repair in any case. The aortic injury was repaired via an extended incision with no longterm sequelae. Vascular injury case reports Patient details are presented in Table 1. Amongst the five patients, the average number of operative levels was 2.2 (1–4). The BMI (range: 26.2–39.2) and the age of the patient at the time of surgery (range: 50–69) varied considerably. The average intraoperative estimated blood lost of the four cases of minor vascular injury was 700 ml (range: 100–1,500 ml). The intraoperative estimated that blood lost during the major vascular injury was 1,200 ml. The postoperative course of all patients was uneventful. No long-term sequelae from the intraoperative vascular injuries were recorded. Case I A 69-year-old male presented to the outpatient clinic with severe axial lower back and limited ambulation. A three-level LLIF from L2 to L5 was recommended. The L4–5 disk space was approached first utilizing the mini-open lateral approach to the spine. After the annulotomy, the trialing and finally the insertion of the PEEK implant the retractor was removed. Bleeding was noted over the L4 vertebral body. Pressure was immediately applied using a peanut cottonoid. The diagnosis of segmental artery injury was made. The segmental vertebral artery over L4 was subsequently isolated, tied off with silk ties and secured with vascular clips. The remaining two levels were operated on without further incidents. Case II A 58-year-old female presented to the outpatient clinic with a history of thoracolumbar scoliosis and lumbar spinal stenosis. A three-level LLIF from L1 to L4 was recommended. The L3–4 space, the L4–5 space and the L1–2 space were operated on without incident. To conclude, the L2–3 disk space was visualized and found to have a large lateral osteophyte bridging over the segmental artery. To allow release of the space, an osteotomy was performed during which some bleeding of the segmental artery of L3 was noted. The bleeding vessel was skeletonized and ligated with vascular clips. The remaining surgery proceeded without incident. Case III A 72-year-old male presented to the outpatient clinic with a history of a facet cyst for which he underwent a
805
foraminotomy with a cyst excision. Despite providing a temporary relief, pain in his lower extremity symptoms progressively worsened. A single-level LLIF from L3 to L4 was recommended. Despite rigorous fixation, the patient’s morbid obesity caused a shift in the patient’s position which was noted upon the attempts of placing the retractors in the appropriate position. The retractors were subsequently removed so the patient could be repositioned into a true lateral position. At this point, significant bleeding from the retroperitoneal space through the site of incision was noted. The incision was packed with gauze and the patient was then repositioned. Diagnosis of an injury to the L3 segmental artery was made. The blood vessel was crossing the vertebral body with a low trajectory and had been injured by the placement of the shim device. To control the bleeding, the vessel was secured using vascular clips. The remaining surgery proceeded without incident. Case IV A 56-year-old male presented to the outpatient clinic with a history of posterior lumbar decompression and fusion with segmental spinal instrumentation from L3 to L5 through a Posterior Lumbar Interbody Fusion (PLIF). The radiographic examination revealed junctional degeneration of the L2–3 level above the previous fusion. A single-level LLIF from L2 to L3 was recommended. The course of surgery was uneventful. On the second postoperative day, the patient complained of abdominal pain. CT-Imaging revealed a left retroperitoneal hematoma occupying the left anterior and posterior pararenal space extending inferiorly into the pericolic gutter. The patient was transferred to a general hospital where the hematoma was evacuated by the attending vascular surgeon who made the diagnosis of an injury to the L3 Segmental Artery. Henceforth, the postoperative course was uneventful. Case V A 50-year-old female presented to the outpatient clinic with severe lower back- and left lower extremity pain [10]. A three-level LLIF from L2–5, a PLIF from L5–S1 and Posterior Instrumented Fusion from T12–S1 were recommended. After the successful operation on the L4–5 level, attention was turned to the L3–4 disk space. During the placement of the implant, the proximal aspect of the PEEK cage broke. The attempt to reposition the implant resulted in the distal violation of the end-plate and the anterior cortex of the L3 vertebral body. An immediate drop of blood pressure was noted. The flank wound was sealed and cardiopulmonary resuscitation was immediately initiated. The emergency laparotomy revealed a laceration of the aortic terminus. The hematoma which had accumulated
123
123
Male
69
28.5
Parkinson’s disease; Degenerative scoliosis (25 degrees); Grade 1 Spondylolisthesis of L4–5; Spondylosis
L2–5
Left
400
10
L4
Segmental artery
Uneventful
Sex
Age at Surgery (years)
BMI
Diagnosis
Levels
Side of approach
Est. blood loss (ml)
Length of stay (days)
Level of injury
Injured vessel
Postoperative course
Patient 1
Table 1 Demographic and operative patient details
Uneventful
Segmental artery
L4
2
100
Left
L2–3
Thoracic scoliosis (40 degrees); Lumbar scoliosis (45 degrees); degenerative disk disease; spinal stenosis; spondylosis
26.5
57
Male
Patient 2
Uneventful
Segmental artery
L3
19
1,500
Right
L1–5
Grade 1 Spondylolisthesis of L3–4; spondylosis; degenerative disk disease
29.5
58
Female
Patient 3
Uneventful
Segmental artery
L3
9
800
Right
L3–4
Junctional level degeneration above a prior posterior lumbar interbody fusion from L3–L5
39.2
73
Male
Patient 4
Uneventful
Abdominal aorta
L3
38
12,000
Right
L3–5
Degenerative scoliosis (35 degrees) with coronal compensation
26.2
50
Female
Patient 5
806 Eur Spine J (2015) 24:800–809
Eur Spine J (2015) 24:800–809
within the abdomen was evacuated and the aorta was repaired with 4–0 prolene sutures. Following the closure of the abdominal wall, the patient was repositioned and the L3–4 disk space was explored. The broken implant was removed in pieces.
Discussion The mini-open access technique was developed as a potentially safer approach to the lateral lumbar spine over blind or dilator-based systems. What was originally born out of necessity when dealing with complex lumbar deformity has proven to be helpful when dealing with minor vascular complications. This technique has proven helpful when injuries to the segmental lumbar vessels were encountered, allowing for a timely control of the bleeding without the necessity of lengthening the incision. Vascular injuries in general are a rare complication of LLIF, with early studies describing their experience with LLIF reporting no vascular injuries [2, 11]. However, in more complex deformity cases, the segmental artery and vein at times approach the disk space making blind dilator approaches in these cases higher risk. Advanced techniques requiring aggressive disk space distraction as in deformities and/or use of lateral osteotomies make an awareness of segmental artery and vein location at each segment mandatory. All of the minor vascular injuries reported were diagnosed to be segmental artery lacerations. Although all of our patients diagnosed with a segmental artery laceration received immediate vessel repair and suffered no long-term sequelae from the intraoperative complication, the consequences of this injury could have been catastrophic. The segmental artery supplies blood to the vertebral bony structure, the ribs and their surrounding tissue, posterior local and epidural elements as well as the nerve roots bilaterally [12]. Usually the segmental vascular course is simple. Two segmental arteries, which arise from the dorsal surface of the aorta, encircle the vertebral body and merge on the posterior surface. At the transverse process, both arteries bifurcate into an intercostal segment which courses to the ribs and a dorsal segment which branches into several smaller vessels travels to the posterior elements and through the neural foramen into the vertebral canal [13]. The vascular arrangement and origin of the segmental artery vary widely amongst individuals [14]. Exploring the mechanisms of injury presented in this manuscript may assist in future prevention or facilitate recognition and ligation. As discussed, Case I emphasizes the importance of an awareness of the segmental location in degenerative deformities. Due to osteophyte formation and vertebral body remodeling over time, the segmental
807
vessels are often within the operative field making them vulnerable to injury either by the retractor or instruments. The removal of retractors which caused the injury in Case I has been characterized as a possible culprit of hemorrhage in spinal surgery before and should be approached with equal care in LLIF as in other procedures [25]. Furthermore, in these cases, surgeons should be cognizant that the segmental may be incorporated in chronically formed osteophytes as described in Case II. When performing osteotomies, one must be prepared to encounter segmental bleeding and perform ligation. Rarely, aberrant vessel course can occur in the absence of deformity as illustrated in Case III. The greatest risk overall, however, lies in the lack of identification of hemorrhage, as described in Case IV. In addition to potentially fatal systemic consequences, a hematoma may result in devastating neurological deterioration [15]. Though the adaption of a surgical technique such as the mini-open strives to allow for better visualization, hidden hemorrhage may occur. Since the initial symptoms of postoperative spinal epidural hematoma have been found to be presented after a median time of 2.7 h, close monitoring of every patients neurological status in the immediate postoperative period is essential [16]. Beyond the minor segmental vessels, it is theoretically possible to injure the ascending iliolumbar vein. The ascending iliolumbar vein may be encountered during a leftsided approach to the L4–L5 disk space, and may traverse the disk space and/or the cephalad aspect of L5. The ascending iliolumbar vein may be at risk for injury during performance of LLIF in the traditional closed fashion, either during retractor placement or during diskectomy. The mini-open technique allows the visualization of a vessel should it fall within the operative field. The authors recall several lateral cases that were aborted to avoid injury to the vein. Although a minor vein (diameter 2–4 mm) it can be hard to control and ligate due to a position low in the pelvis. The incidence of a problematic position of the ascending iliolumbar vein is unknown. Another unnamed vein similarly at risk can sometimes be recognized at the L4–5 level, 2–4 mm in diameter, coursing cephalocaudad immediately anterior to the lumbar plexus. Injury to either of these venous structures during performance of LLIF in the traditional closed fashion can result in significant hemorrhage which would be difficult, if not impossible, to control in the absence of conversion to an open procedure or utilization of endovascular techniques. The mini-open approach allows for prompt identification of these venous structures and facilitates avoidance or ligation and division as necessary. Given the complexity of the vasculature surrounding the spine, we must acknowledge that injuries to vascular structures smaller than the segmental vessels may not have been recognized and therefore reported, though none of the
123
808
patients in our case series other than those discussed in the case presentations were noted to have vascular complications stemming from LLIF. Minimally invasive surgery has been theorized to have several distinct advantages over traditional approaches including a lower degree of postoperative pain, decreased estimated blood loss and length of inpatient hospital stay, a faster return to the preoperative degree of daily activity and an overall lower total hospital cost over a 60-day perioperative period when compared to open surgery [17–20]. Developments in minimally invasive spine surgery, such as the laparoscopic anterior approach, have been deemed relatively safe for patients and are frequently performed [21]. Despite this, the open transabdominal anterior approach for spinal fusion of the lumbar spine continues to have several disadvantages. Complications including visceral injury, large vessel bleeding and sexual dysfunction have been reported [22–24]. In a study performed at our institution, the incidence of vascular injury which most commonly affected major venous vessels during elective anterior spinal surgery was found to be 2.9 % [25]. The lateral approach to the lumbar spine offers an improved vascular safety profile over traditional transabdominal and retroperitoneal anterior lumbar approaches. The mini-open technique presented in this manuscript presents a safe and reproducible alternative to dilator-based systems. In the setting of complex lumbar deformity cases, there may be distinct advantages of the mini-open technique in terms of vascular complication mitigation. Further, there may be cost advantages with streamlining of specialized equipment. Additional prospective comparative studies examining different surgical approaches for LLIF are warranted in the future. Conflict of interest There were no sources of financial support or perceived conflicts of interest for this study.
References 1. Ozgur BM, Aryan HE, Pimenta L, Taylor WR (2006) Extreme Lateral Interbody Fusion (XLIF): a novel surgical technique for anterior lumbar interbody fusion. Spine J 6(4):435–443 2. Youssef JA, McAfee PC, Patty CA et al (2010) Minimally invasive surgery: lateral approach interbody fusion: results and review. Spine (Phila Pa 1976) 35(26 Suppl):S302–S311. doi:10. 1097/BRS.0b013e3182023438 3. Aghayev K, Vrionis FD (2013) Mini-open lateral retroperitoneal lumbar spine approach using psoas muscle retraction technique. Technical report and initial results on six patients. Eur Spine J 22(9):2113–2119. doi:10.1007/s00586-013-2931-1 4. Yuan PS, Rowshan K, Verma RB, Miller LE, Block JE (2014) Minimally invasive lateral lumbar interbody fusion with direct psoas visualization. J Orthop Surg Res 9(1):20. doi:10.1186/ 1749-799X-9-20 5. Hardenbrook MA, Miller LE, Block JE (2013) TranS1 VEO system: a novel psoas-sparing device for transpsoas lumbar
123
Eur Spine J (2015) 24:800–809
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
interbody fusion. Med Devices (Auckl) 6:91–95. doi:10.2147/ MDER.S43746 Pumberger M, Hughes AP, Huang RR et al (2012) Neurologic deficit following lateral lumbar interbody fusion. Eur Spine J 21(6):1192–1199. doi:10.1007/s00586-011-2087-9 Taher F, Hughes AP, Lebl DR et al (2013) Contralateral motor deficits after lateral lumbar interbody fusion. Spine (Phila Pa 1976) 38(22):1959–1963. doi:10.1097/BRS.0b013 e3182a463a9 Lykissas MG, Aichmair A, Hughes AP et al (2014) Nerve injury after lateral lumbar interbody fusion: a review of 919 treated levels with identification of risk factors. Spine J 14(5):749–758. doi:10.1016/j.spinee.2013.06.066 Fantini GA, Pawar AY (2013) Access related complications during anterior exposure of the lumbar spine. World J Orthop 4(1):19–23. doi:10.5312/wjo.v4.i1.19 Aichmair A, Fantini GA, Garvin S, Beckman J, Girardi FP (2013) Aortic perforation during lateral lumbar interbody fusion. J Spinal Disord Tech [Epub ahead of print] Rodgers WB, Gerber EJ, Patterson J (2011) Intraoperative and early postoperative complications in extreme lateral interbody fusion: an analysis of 600 cases. Spine (Phila Pa 1976) 36(1):26–32. doi:10.1097/BRS.0b013e3181e1040a Prince EA, Ahn SH (2013) Basic vascular neuroanatomy of the brain and spine: what the general interventional radiologist needs to know. Semin Intervent Radiol 30(3):234–239. doi:10.1055/s0033-1353475 Shimizu S, Tanaka R, Kan S et al (2005) Origins of the segmental arteries in the aorta: an anatomic study for selective catheterization with spinal arteriography. AJNR Am J Neuroradiol 26(4): 922–928 Berenstein A, Lasjaunias P, Ter Brugge KG (2001) Spinal and spinal cord arteries and veins. In: Berenstein A, Lasjaunias P, Ter Brugge KG (eds) Surgical neuroangiography. Clinical vascular anatomy and variations, vol 2nd. Springer, New York, pp 73–164 Kao FC, Tsai TT, Chen LH, et al (2014) Symptomatic epidural hematoma after lumbar decompression surgery. Eur Spine J [Epub ahead of print] Amiri AR, Fouyas IP, Cro S, Casey AT (2013) Postoperative spinal epidural hematoma (SEH): incidence, risk factors, onset, and management. Spine J 13(2):134–140. doi:10.1016/j.spinee. 2012.10.028 Regan JJ, Aronoff RJ, Ohnmeiss DD, Sengupta DK (1999) Laparoscopic approach to L4-L5 for interbody fusion using BAK cages: experience in the first 58 cases. Spine (Phila Pa 1976) 24(20):2171–2174 Khajavi K, Shen AY (2014) Two-year radiographic and clinical outcomes of a minimally invasive, lateral, transpsoas approach for anterior lumbar interbody fusion in the treatment of adult degenerative scoliosis. Eur Spine J 23(6):1215–1223. doi:10. 1007/s00586-014-3246-6 Smith AG, Capobianco R, Cher D et al (2013) Open versus minimally invasive sacroiliac joint fusion: a multi-center comparison of perioperative measures and clinical outcomes. Ann Surg Innov Res 7(1):14. doi:10.1186/1750-1164-7-14 Singh K, Nandyala SV, Marquez-Lara A et al (2013) A perioperative cost analysis comparing single-level minimally invasive and open transforaminal lumbar interbody fusion. Spine J 13:01723–01733. doi:10.1016/j.spinee.2013.10.053 Kaiser MG, Haid RW Jr, Subach BR et al (2002) Comparison of the mini-open versus laparoscopic approach for anterior lumbar interbody fusion: a retrospective review. Neurosurgery 51(1):97–103 (discussion 103-5) Rajaraman V, Vingan R, Roth P et al (1999) Visceral and vascular complications resulting from anterior lumbar interbody fusion. J Neurosurg 91(1 Suppl):60–64
Eur Spine J (2015) 24:800–809 23. Baker JK, Reardon PR, Reardon MJ, Heggeness MH (1993) Vascular injury in anterior lumbar surgery. Spine (Phila Pa 1976) 18(15):2227–2230 24. Flynn JC, Price CT (1984) Sexual complications of anterior fusion of the lumbar spine. Spine (Phila Pa 1976) 9(5):489–492
809 25. Fantini GA, Pappou IP, Girardi FP, Sandhu HS, Cammisa FP Jr (2007) Major vascular injury during anterior lumbar spinal surgery: incidence, risk factors, and management. Spine (Phila Pa 1976) 32(24):2751–2758
123
