Laparoscopic Anterior Lumbar Interbody Fusion: Technical Note
The advent of minimally invasive surgery has challenged spine surgeons to refine and perfect less invasive procedures that have comparable or superior outcomes to standard open techniques. The goal of these procedures is decreased hospitalization time and quicker return to functional status. These goals are attained by achieving the same operative result with smaller incisions and reduced tissue dissection. The development and refinement of laparoscopic intraperitoneal surgery has offered spine surgeons new opportunities for anterior approaches to the lumbar spine (26, 30, 32).
Previous reports have documented that anterior lumbar interbody fusions (ALIF) decrease perioperative blood loss and eliminate nerve root retraction compared to posterior procedures (31). It has also been shown that ALIF procedures have shorter operating times than posterior lumbar interbody fusions (PLIF) with or without pedicle screw instrumentation (24, 25). These advances have led to shorter hospitalization times and comparable fusion rates (2325).
Presently, the standard approaches for ALIF include transperitoneal or retroperitoneal exposure for interbody fusion at L45 and L5S1. More cephalad lumbar levels may also be accessible. The development of laparoscopic ALIF has allowed the spine surgeon to achieve the desired result of placement of interbody fusion devices with smaller incisions and less bowel irritation with postoperative ileus. Laparoscopic ALIF represents a viable alternative to standard open transperitoneal approaches. We review our preliminary experience with laparoscopic ALIF in a series of 32 consecutive patients and describe the lessons learned as this procedure has developed at our institution.
Materials and Methods
Over a period of two years a total of 32 patients underwent single level laparoscopic anterior lumbar interbody fusions at the L4L5 or L5S1 interspace. Operative indications included patients with degenerative disc disease characterized by segmental disc degeneration with Modic changes with or without foraminal stenosis demonstrated on MRI with a clinical history of mechanical low back pain with or without leg pain. Other candidates included patients with degenerative lumbar instability, less than grade II spondylolisthesis, postsurgical spinal instability, and pseudoarthrosis after attempted posterior fusions. (Table 1)
Table 1: Indications for Laparoscopic ALIF
Degenerative Lumbar Instability
Grade 1 or 2 Spondylolisthesis
PostSurgical Spinal Instability
Failed Posterior Fusion
All patients had failed conservative therapy, which included any combination of nonsteroidal or narcotic medications, physical therapy, or other local pain procedures. Provocative discography was utilized only in patients when the symptomatic level was in question. In those patients, needles were placed into the disc of interest as well as adjacent disc spaces, which served as a control. Sterile saline solution or contrast material was injected. Discograms were considered positive only if the injection recreated the patient's pain at the appropriate level of interest. Additionally, injection at adjacent levels had to produce either no pain or "discordant" pain. Some patients also underwent diagnostic facet blocks to help confirm the pathologic level. Patients with discordant pain dissimilar to the patient's chief complaint in terms of location and intensity were not considered appropriate candidates, did not undergo surgery, and were not included in this analysis. Patients that had evidence of posterior pathology such as a disc fragment in the canal, facet hypertrophy, lateral recess or central spinal stenosis, a history of multiple abdominal procedures or severe large vessel atherosclerosis were not considered good candidates for surgery and also were not included in this study. (Table 2)
Table 2: ContraIndications for Laparoscopic ALIF
Multiple Abdominal Procedures (relative)
Severe Atherosclerosis of the aorta and iliac vessels
The patient is placed supine on a radiolucent table. Equipment in the room is positioned to allow the surgeon an optimal view of both the carm camera and the video monitor. Pillows are placed under the hips to exentuate lumbar lordosis at the lumbosacral junction and beneath the knees to prevent hyperextension. The arms are placed at the patient's side low enough to prevent interference with the fluoroscopic view. Ankle straps are placed on the patient to prevent sliding because a steep Trendelenburg position is required during the procedure to allow the abdominal viscera to move rostrally out of the pelvis. A nasogastric tube and foley catheter are placed to decompress the stomach and bladder respectively. Both catheters are usually removed at the completion of the procedure.
Preoperatively, nonsteroidal agents are discontinued at least one week prior to surgery to minimize blood loss and promote fusion. The bowel preparation consists of one bottle of magnesium citrate initiated the night before surgery. Consent routinely includes the possibility of an open laparotomy in the event of uncontrolled bleeding or poor visualization in addition to other potential complications from laparoscopic fusion. (Table 3)
Table 3 Potential
Complications from Laparoscopic ALIF
Deep Venous Thrombosis Ileus
Implant migration Atelectasis
Nerve Root Contusion Retrograde Ejaculation
Templating the proper sized interbody implant based on an adjacent normal disc space demonstrated on a preoperative lateral lumbar radiograph is important prior to exposure. Selecting the proper size interbody implant is essential. We template over an adjacent normal disc space seen on preoperative lateral radiograph. Axial CT or MRI templates can also be utilized to confirm the selected implants fit properly within the confines of the diseased disc space. The implant size should approximate the disc height of the adjacent normal interspaces and also account for the degree of reaming and tapping into the cortical endplates. This allows for the appropriate annular tension to maintain the interbody implant under compression.
The patient is prepped and draped in sterile fashion. Autologous bone is harvested from the anterior iliac crest usually on the right side at the beginning of the procedure. Prior to insufflation, an incision line centered on the midline approximately 1.0 to 3.0 cm wide above the pubis is marked. The fluoroscopic equipment is then brought into place before the incisions are made to verify the midline. It is important to obtain adequate fluoroscopic views for proper intraoperative visualization of the vertebral bodies and to estimate trajectory.
Four incisions are utilized. The two lower paramedian incisions allow placement of portals for the working forceps. The incision for the interbody channel and devices is centered over the midline suprapubic region and measures 2.0 4.0 cm in length. The viewing camera is placed through the curvilinear umbilical incision and is held by a robotic arm. (AESOP, Computer Motion, Inc., Goleta, California) This robotic arm responds to verbal commands from the laparoscopic surgeon and moves the camera to the surgeon's desired position. This allows the surgeon two free hands to place the instrumentation.
As the bone graft is harvested from the iliac crest, the laparoscopic surgeon gains entrance and the abdomen is insufflated. The patient is placed in steep Trendelenburg to mobilize the abdominal contents out of the pelvic inlet. If necessary, the sigmoid colon is tacked to the abdominal wall to gain better exposure.
Adequate disc space exposure is critical. The sacral promontory is identified by palpation and confirmed by fluoroscopy. The peritoneum is then opened sharply. In males, the unipolar cautery should be avoided and a blunt Kittner dissector is used with a gentle sweeping motion to mobilize the presacral sympathetic plexus. This maneuver may decrease the incidence of retrograde ejaculation, a known complication from anterior approaches. In females, monopolar electrocautery can be used to expose the anterior face of the vertebral bodies and disc space.
The first anatomical structure seen is the middle sacral artery and vein. Although these vessels do not reliably predict the midline of the vertebral body, the preoperative MRI may demonstrate their relationship to the midline (10, 16, 27). The preoperative axial images should be reviewed for the presence of eccentric anterior osteophytes, which may give the surgeon a false localization of the lateral convexities. The middle sacral artery and vein are ligated and divided. The anterior curvature of the lumbosacral junction is palpated as well as the left and right lateral convexities in conjunction with intermittent fluoroscopy to further identify the midline. The left iliac vein protrudes more anteriorly and may require more retraction. If the midline cannot be accurately identified the surgeon should consider an open conversion as higher rates of complication are more likely.
After confirmation of the midline the instrumentation phase can begin. Depending on the instrumentation system, the entry sites for the trephines are prepared. Laparoscopic approaches for interbody fusion may utilize titanium BAK cages (SpineTech, Minneapolis, MN), allograft bone dowels (Regeneration Technologies, Alachua, FL), the titanium Interfix cages (MedtronicSofamor Danek, Memphis, TN). Lordotic threaded titanium cages (Novus, MedtronicSofamor Danek, Memphis, TN) are currently under IRB evaluation. The trephine is utilized to core out soft disc material in each paramedian entry site. The trephine should enter the disc space in a trajectory parallel to the end plates to avoid eccentric placement of cages into the vertebral bone. Using the largest trephine possible optimizes maximal disc material removal. Additional disc material is removed with pituitary rongeurs.
Progressively larger distractors are then tamped into the disc space to restore the disc height to the appropriate level and to tension the annulus fibrosis. Once a distractor is in place on one side, the remaining disc material and subjacent cortical bone on the opposite side is removed with a reamer. It is better to ream to a depth longer than the selected interbody implant length to allow placement of additional bone graft anterior to the interbody device and to allow countersinking of the implant (if desired). Underreaming may promote stripping of the threaded pattern during attempted tapping. This occurs because the tap is prevented from advancing by residual disc material deep in the posterior disc space. The holes are then tapped to the premeasured depth. The implant is then inserted in proper alignment. Additional bone graft is packed anteriorly to the device. This promotes additional bone fusion, and allows the surgeon to better assess the presence of a solid arthrodeses with anterior bony bridging on postoperative followup studies. The same steps are then repeated on the other side after removing the distractor plug. Ideally the collapsed disc space should be distracted by the implant to reapproximate its original size.
Patients are mobilized early and are allowed to ambulate on postoperative day one. Diet is advanced as tolerated provided there are adequate bowel sounds present. A hard or soft lumbar orthosis is utilized depending on the degree of preoperative instability, the presence of osteoporosis, and size of the patient. Patients usually are discharged on postoperative day two. Followup radiographs are obtained prior to discharge, at six weeks, 3 and 9 months, and at one and two years. Determination of solid arthrodeses can be problematic. A distinction must be made between "stability" versus "arthrodeses". Previously published studies of titanium cages determine "fusion" by absence of lucency around the implant and absence of movement on flexionextension films (21, 22). Presence of bone within the cage does not necessarily imply sold arthrodeses. Solid bridging bone (between the vertebral endplates) anterior to the cage may signify a solid arthrodeses. We believe evidence of fusion is best demonstrated on reformatted thin cut saggital CT scans which demonstrate: (1) incorporation of cancellous bone within the cage/dowel (2) subchondral increase in endplate sclerosis and (3) a solid arthrodeses in the bone graft placed anterior or posterior to the cage/dowel. With allograft cortical bone dowels incorporation of the threads can be seen, combined with a progressive diminution in the "halo" seen on the AP radiographs. An advantage of utilizing bone dowels is a more direct assessment of arthrodeses. Evidence that suggests pseudoarthrosis on plain AP and lateral radiographs include lucency around the device, persistent movement at the instrumented level, or lack of a solid bone mass. Movement can be defined by translational change in position or change in angulation of the disc space on flexionextension lateral radiographs.
In our series of 32 patients there was one disc space infection with a bone dowel successfully treated with intravenous antibiotics. Early in the experience we have one nerve root contusion following lateral cut out of the disc space. This occurred because we had not properly selected the exact midline from the start of the case. Two iliac vein thromboses occurred requiring anticoagulation. These thromboses were felt to be due to retraction of the vessels. Three of fifteen men reported retrograde ejaculation at longest followup. Two other procedures were converted to an open transperitoneal approach from adhesions, and one for better visualization of a spinal deformity. There were no conversions for bleeding.
Other major complications not encountered in our series but are potential problems in performing this procedure include vascular injury requiring conversion to an open procedure to control bleeding. Minor complications reported in the literature include ileus and retrograde ejaculation. There has been no series reporting problems with impotence or lack of orgasm. We feel it is important to fully inform the male patient preoperatively regarding the risk of retrograde ejaculation.
Zucherman and Zdeblick were the first to perform the technique of laparoscopic anterior interbody fusion (34). In their first reported series of 17 patients they had 3 technical complications, 2 of which required open conversion to laparotomy secondary to venous bleeding from the middle sacral vein. Subsequently, numerous reports of this technique have been described (19, 1115, 1720, 28, 29). In a prospective randomized series comparing posterolateral fusion using rigid pedicle screw instrumentation with and without interbody grafting to laparoscopic BAK fusion cages Zdeblick demonstrated higher fusion rates and shorter hospitalization and return to work (33). Their study confirmed significant cost advantages and outcomes compared to posterior open techniques in patients with L5S1 disease.
When comparing results relative to interbody fusion it is important to understand the difference between fusion and stability. Fusion is based on the radiographic findings discussed previously such as: increase in subchondral endplate sclerosis, bridging bone incorporating the anterior bone graft, and diminution in the "halo" seen on AP radiographs. Stability is based on radiographic evidence of nonmovement on flexionextension roentgenograms. It may be possible to have a stable construct without fusion. Stability rates for threaded interbody cages appear excellent and approach 98% at three years in the BAK series. Several large clinical trials have demonstrated marked improvement in pain (10, 21, 22).
Laparoscopic ALIF is a safe and effective method of placing threaded cylindrical lumbar interbody implants (12, 22, 33, 34). The benefits of this procedure over posterior fusion procedure include the elimination of nerve root retraction and possible dural injury. There is less muscular dissection and thus decreased blood loss and postoperative pain. The drawbacks of the procedure include a steep learning curve and the low, but real, incidence of retrograde ejaculation. Larger series with longterm followup will bear out whether one procedure is superior to another or whether they are complimentary techniques. Keys to avoiding complications include proper identification of the midline, applying an initial "toe in" force to discourage lateral cut out, and flexibility to convert to an open approach in the face of inadequate exposure. Laparoscopic fusion with standalone interbody devices may have limited success in treating patients with mobile instability and/or highgrade spondylolisthesis. In these cases, supplemental posterior stabilization should be considered.
4. De Peretti F, Hovorka I, Fabiani P, Argenson C: New possibilities in L2L5 lumbar arthrodeses using a lateral retroperitoneal approach assisted by laparoscopy: preliminary results. European Spine Journal 5(3): 210216, 1996.
9. Husson JL, Le Huec JC, Trebuchet G, Lesprit E, Bossis JM: Interbody arthrodeses of the lumbar vertebrae using retroperitoneal videoendoscopy. A preliminary study of 38 cases. Chirurgie 123(5): 4919, 1998.
11. Lazennec JY, Pouzet B, Ramare S, Mora N, Hansen S, Saillant G, Benazet JP: Possibilities of anterior approach to the lumbar spine by minimal retroperitoneal access. Anatomical bases. Technical principles and initial results. Chirurgie 122 (89): 468477, 1997.
15. McAfee PC, Regan JR, Zdeblick T, Zuckerman J, Picetti GD, Heim S. Geis WP, Fedder IL: The incidence of complications in endoscopic anterior thoracolumbar spinal reconstructive surgery. A prospective multicenter study comprising the first 100 consecutive cases. Spine 20(14): 16241632, 1995.
18. Ogon M, Maurer H, Wimmer C, Landauer F, Sterzinger W, Krismer M: Minimally invasive approach and surgical procedures in the lumbar spine. Orthopade 26(6): 553561, 1997.
19. Olinger A, Hildebrandt U, Pistorius G, Lindemann W, Menger MD: Laparoscopic 2level fusion of the lumbar spine with Bagby and Kuslich implants. Chirurg 67(4): 348350, 1996.
20. Olsen D, McCord D, Law M: Laparoscopic discectomy with anterior interbody fusion of L5S1. Surgical Endoscopy 10(12): 11581163, 1996.
22. Ray CD: Threaded titanium cages for lumbar interbody fusions. Spine 22(6): 667680, 1997.
23. Reagan JJ, McAfee PC, Guyer RD, Aronoff RJ: Laparoscopic fusion of the lumbar spine in a multicenter series of the first 34 consecutive patients. Surgical Laparoscopy, Endoscopy & Percutaneous Techniques 6(6): 459468, 1996.
24. Reagan JJ, Yuan H, McAfee PC: Laparoscopic fusion of the lumbar spine: minimally invasive spine surgery. A prospective multicenter study evaluating open and laparoscopic lumbar fusion. Spine 24(4): 402411, 1999.
25. Regan JJ, Zdeblick T, Zucherman J, et. al: Comparison of open versus laparoscopic fusion of the lumbar spine using the BAK( threaded fusion cage. 10th Annual Meeting, North American Spine Society, Washington, DC, October 1821, 1995, p. 131.
26. Riley LH, Eck JC, Yoshida H, Toth JM, Nguygen C, Lim TH, McGrady LM: Laparoscopic assisted fusion of the lumbosacral spine. A biomechanical and histologic analysis of the open versus laparoscopic technique in an animal model. Spine 22(12): 14071412, 1997.
27. Sachs BL, Schwaitzberg SD: Lumbosacral (L5S1) discectomy and interbody fusion technique. In Atlas of Endoscopic Spine Surgery: Regan JF, McAfee PC, Mack MF (eds). Quality Medical Publishing, Inc., St. Louis, 275291, 1995.
28. ScholzJager A, Kemen M, Willburger RE, Wittenberg RH, Steffen R, Zumtobel V: Video laparoscopic transperitoneal exposure of the lumbar spine for ventral fusion. Langenbecks Archiv fur Chirurgie Supplement Kongressband 113: 956959, 1996.
29. Silcox DH: Laparoscopic bone dowel fusions of the lumbar spine. Orthopedic Clinics of North America 29(4): 655663, 1998.
30. Weiner BK, Fraser RD: Spine updatelumbar interbody cages. Spine 23(5): 634640, 1998.
31. Zdeblick TA, Ulschmid S, Dick JC: The surgical treatment of L5S1 degenerative disc disease. A prospective randomized study. 10th Annual Meeting, North American Spine Society, Washington, DC, October 1821, 1995, p. 168.
32. Zdeblick TA, Warden KE, Zon D, McAfee PC, Abitbol JJ: Anterior spinal fixators: A biomechanical in vitro study. Spine 18(5): 513517, 1993.
34. Zucherman JF, Zdeblick TA, Bailey SA, Mahvi D, Hsu KY, Kohrs D: Instrumented laparoscopic spinal fusion: preliminary results. Spine 20(18): 20292035, 1995.