Nerve Root Anomalies: Implications for Transforaminal Lumbar Interbody Fusion Surgery and a Review of the Neidre and Macnab Classification System

Neurosurg Focus 35 (2):E9, 2013 ©AANS, 2013

Nerve root anomalies: implications for transforaminal lumbar interbody fusion surgery and a review of the Neidre and Macnab classification system

Shane M. Burke, B.S.,1,2 Mina G. Safain, M.D.,1,2 James Kryzanski, M.D.,1,2 and Ron I. Riesenburger, M.D.1,2 1Department of Neurosurgery, Tufts Medical Center; and 2Department of Neurosurgery, Tufts University School of Medicine, Boston, Massachusetts

Lumbar nerve root anomalies are uncommon phenomena that must be recognized to avoid neural injury during surgery. The authors describe 2 cases of nerve root anomalies encountered during mini-open transforaminal lumbar interbody fusion (TLIF) surgery. One anomaly was a confluent variant not previously classified; the authors suggest that this variant be reflected in an amendment to the Neidre and Macnab classification system. They also propose strategies for identifying these anomalies and avoiding injury to anomalous nerve roots during TLIF surgery. Case 1 involved a 68-year-old woman with a 2-year history of neurogenic claudication. An MR image demonstrated L4–5 stenosis and spondylolisthesis and an L-4 nerve root that appeared unusually low in the neural foramen. During a mini-open TLIF procedure, a nerve root anomaly was seen. Six months after surgery this patient was free of neurogenic claudication. Case 2 involved a 60-year-old woman with a 1-year history of left L-4 radicular pain. Both MR and CT images demonstrated severe left L-4 foraminal stenosis and focal scoliosis. Before surgery, a nerve root anomaly was not detected, but during a unilateral mini-open TLIF procedure, a confluent nerve root was identified. Two years after surgery, this patient was free of radicular pain. (http://thejns.org/doi/abs/10.3171/2013.2.FOCUS1349)

Key Words • anomaly • confluent nerve root • fusion • transforaminal • transforaminal lumbar interbody fusion

pinal nerve root anomalies are well-described but uting branches, patients with these anomalies might be uncommon phenomena that must be recognized to more susceptible to radicular symptoms and often re- avoid inadvertent nerve injury during surgery. Pre- quire aggressive decompressive surgery to maximize the Svious studies have determined that the incidence of these chances of a good clinical outcome.2,4–7,10–12,15,16,21,22,26,30,36 anomalies varies: 1.3% when found during surgery,36 The unique challenges presented by nerve root anoma- 2.0%–6.7% when found by imaging before surgery,1,26–28 lies can make standard procedures more complicated and 14 and 14.0% when found during cadaver dissection. can increase the risk for iatrogenic injury.1,14,15,20,24,29,33,34,36 Several classification systems describe the different Therefore, preoperative diagnosis and identification of morphological appearances of these anomalies, includ- nerve root anomalies is important and helps the surgeon ing the Postacchini system, the Kadish and Simmons plan a safe surgical approach. However, even MRI, which system, and the Neidre and Macnab system.14,18,25,28 The 25 has been shown to provide the resolution necessary to system developed by Neidre and Macnab is the most successfully classify anomalies,1,33 can fail to detect small frequently cited for lumbosacral nerve root classification. anastomotic branches between nerve roots.1 Thus, when It divides anomalies into 3 types: conjoined, redundant, preoperative imaging fails to detect a nerve root anomaly, and anastomotic. Of these, conjoined nerve roots are by far the most common, followed by anastomotic anoma- intraoperative detection is imperative if nerve root injury lies.3,5,14,21,28,29,32,34,35 is to be avoided. Because of the larger cross-sectional area of anoma- Some of the commonly treated degenerative con lous nerve roots and a tethering effect from the contrib- ditions that co-occur with nerve root anomalies include recurrent disc herniation,5,8,22,25,29,30,32,33 spinal/foraminal Abbreviation used in this paper: TLIF = transforaminal lumbar stenosis,3,17,20 and spondylolisthesis.5,6,30,31 One operation interbody fusion. that is used to treat these conditions is TLIF.24 Given that

Neurosurg Focus / Volume 35 / August 2013 1

Unauthenticated | Downloaded 10/07/21 05:33 PM UTC S. M. Burke et al. anomalies are frequently not identified on preoperative images, injury to an unrecognized anomalous nerve root can easily occur during lumbosacral decompression and fusion operations.1,9,14,15,20,23,24,29,33,34,36 Therefore, spine surgeons should be adequately prepared to identify these congenital variations during surgery. To our knowledge, only 2 articles in the English- language literature describe nerve root anomalies of any form encountered during TLIF procedures.6,30 However, neither article clearly addresses strategies to both identify these anomalies and avoid injuring the nerves during TLIF. Among 200 TLIFs performed by our senior author (R.I.R.), 2 cases of nerve root anomalies were encountered; this incidence is consistent with the reported incidence of these anomalies.36 Our objectives are to describe these 2 cases of nerve root anomalies encountered during TLIF, to add a new category (describing an unclassified variant) to the Neidre and Macnab classification system, and to propose an appropriate course of action for identifying and preserving anomalous nerve roots encountered during TLIF procedures.

Case Reports Fig. 1. Case 1. MR images showing L4–5 central stenosis and spondylolisthesis. A: Midline sagittal T2-weighted image demonstrating Case 1 Grade 1 L4–5 spondylolisthesis and nonsymptomatic L5–S1 spondylolisthesis. B: Axial T2-weighted image showing right L4–5 foraminal stenosis. C: Right parasagittal T2-weighted image demonstrating a History and Examination. The patient was a 68-year- low-lying right L-4 nerve root (arrow). D. Right parasagittal T2-weight- old woman with a 2-year history of neurogenic claudica- ed image demonstrating a common trunk (arrows) from which both the tion involving both lower extremities. Medications and 2 L-4 and L-5 nerve roots emanate. Dynamic movement of the image (not epidural steroid injections provided no benefit. Examina- shown) enabled tracking of the L-4 and L-5 nerve roots into 1 common tion revealed 4+/5 strength in her right quadriceps and an- nerve root sleeve. terior tibialis muscles. An MR image (Fig. 1) demonstrated L4–5 stenosis and spondylolisthesis. While reviewing not possible. Therefore, right L4–5 pedicle screws and the MR image, we noted that the right L-4 nerve root rod were placed. An additional small Wiltse incision was seemed to be unusually low in the neural foramen. On a made on the left side, allowing placement of the L4–5 parasagittal view, the right L-4 nerve root also seemed to pedicle screw and rod. show a common trunk with the L-5 nerve root. Therefore, before surgery, the patient was thought to most likely har- Postoperative Course. Six months after surgery, the bor a Type 1B Neidre and Macnab anomaly. patient reported resolution of her neurogenic claudication. Postoperative radiographs are shown in Fig. 3. Operation. The patient had right L-4 nerve root com- pression as well as spinal stenosis. Therefore, we recom- mended decompression and TLIF with the caveat that an interbody graft might not be possible, given the abnor- mally low location of the L-4 nerve root in the right L4–5 foramen. The surgical approach was a right mini-open Wiltse approach. Intraoperatively, while performing the L4–5 laminectomy and right facetectomy, the surgeon (R.I.R.) noted almost complete absence of the ligamentum flavum (Video 1). Video 1. Clip showing surgery for Case 1. Click here to view with Media Player. Click here to view with Quicktime. After the facetectomy was performed, the L-4 nerve root appeared to be unusually low in the foramen and was found to emerge from a common trunk with the L-5 nerve root (Fig. 2, Video 1), as had been observed on preopera- Fig. 2. Case 1. Intraoperative image showing the conjoined nerve tive images. The L-4 nerve root traveled just rostral to root. Left is caudal and right is rostral. The conjoined nerve root is la- the L-5 pedicle, and the L-5 nerve root traveled just cau- beled L4-L5*. The L-4 nerve root travels just rostral to the L-5 pedicle, dal to the L-5 pedicle. Given that the L-4 nerve root was and the L-5 nerve root travels just caudal to the L-5 pedicle. The dura directly over the L4–5 disc space, interbody fusion was and L-5 pedicle are noted for orientation.

2 Neurosurg Focus / Volume 35 / August 2013

Unauthenticated | Downloaded 10/07/21 05:33 PM UTC Nerve root anomalies: implications for TLIF

Fig. 3. Postoperative anteroposterior and lateral radiographs of the patients in Case 1 (A and B) and Case 2 (C and D). Arrows indicate that the patient was standing (weight bearing) during radiography.

Case 2

Fig. 4. Case 2. MR images showing severe left L4–5 foraminal ste- History and Examination. The patient was a 60-year- nosis. A: Midline sagittal T2-weighted image demonstrating moderate old woman with a 1-year history of left lower-extremity L4–5 central stenosis. B: Axial T2-weighted image showing severe radicular pain in an L-4 distribution. Three epidural ste- left L4–5 foraminal stenosis, facet hypertrophy, and moderate central roid injections had provided no lasting benefit. Exami- canal stenosis. C: Left parasagittal T2-weighted image demonstrat- nation revealed that she had left L-4 radicular pain and ing severe L4–5 foraminal stenosis (arrow). No nerve root anomaly is Grade 4+/5 strength in her left quadriceps and anterior identified. tibialis muscles. An MR image (Fig. 4) demonstrated severe left L-4 foraminal stenosis. A CT image indicated without complication. Two years after surgery, the patient focal scoliosis contributing to the foraminal stenosis. A was free of radicular pain. Postoperative radiographs are nerve root anomaly was not detected on preoperative im- shown (Fig. 3). ages. Operation. Unilateral TLIF surgery was performed Discussion through a left mini-open Wiltse incision. After removal of the left L4–5 facet joint, the L-4 nerve root seemed The 2 cases described here highlight the unique chal- tethered caudally and was unusually difficult to mobilize. lenges that nerve root anomalies present during TLIF Therefore, the area just rostral to the L-5 pedicle, which surgery. Although the morphologic appearance of the is typically a safe zone, was explored. An additional, anomaly encountered in Case 1 is well described by the smaller L-4 nerve root that branched off the thecal sac Neidre and Macnab classification system, the anomaly was observed over the L4–5 disc space (Fig. 5, Video 2). encountered in Case 2 is not described by this system. Therefore, we first review the 3 categories currently de- Video 2. Clip showing surgery for Case 2. Click here to view with Media Player. Click here to view with Quicktime. scribed by Neidre and Macnab, and then we propose a fourth class of confluent nerve roots that adequately The larger, more rostral L-4 root and the smaller, caudal describes the anomaly encountered in Case 2. We then L-4 root branched off the thecal sac and joined to form discuss identification and surgical management of these a confluent root distally. After the anatomy was properly anomalies during TLIF surgery. identified, the two L-4 nerve roots were mobilized ros- trally, allowing placement of an interbody graft. Neidre and Macnab Classification Postoperative Course. The operation was completed Type 1. Type 1 anomalies are conjoined nerve roots

Neurosurg Focus / Volume 35 / August 2013 3

Unauthenticated | Downloaded 10/07/21 05:33 PM UTC S. M. Burke et al.

Type 3. Type 3 anomalies involve otherwise normal nerve roots that have an anastomosis or bridging connec- tion between 2 adjacent roots (Fig. 6). Proposed Type 4. Our proposed Type 4 anomaly is a confluent nerve root anomaly, as observed in Case 2. In this patient, a normal-caliber L-4 nerve root exited the thecal sac caudal to the L-4 pedicle. Another, smaller L-4 nerve root exited the thecal sac caudal to the normal- caliber L-4 root. These 2 roots joined distally to form a confluent L-4 root that exited the L4–5 neural foramen (Fig. 7). To clarify, a confluent nerve root anomaly consists of 2 nerve roots that arise separately from the thecal sac and join together distally. This anomaly differs from a conjoined nerve root anomaly, in which 2 nerve roots arise from a common dural sheath and separate distally. Fig. 5. Case 2. Intraoperative image showing the confluent nerve root. Left is rostral and right is caudal. The larger, more rostral L-4 (L-4) To our knowledge, a confluent nerve root anomaly 16 root and the smaller, caudal L-4 (L-4*) root can be seen branching off was first described by Keon-Cohen in 1968 but was not the thecal sac and joining to form a confluent root (L4-L4*) distally. This included in the Neidre and Macnab classification system. root then exits caudal to the L-4 pedicle. The dura and epidural tissue Keon-Cohen described nerve root fibers emerging from overlying the L4–5 disc space are noted for orientation. the L-5 spinal level that joined with fibers emerging from the S-1 spinal level.16 The joined fibers formed a confluent and are the most common anomaly reported in the litera- nerve root, which then further branched distally. Since ture.3,5,14,21,28,29,32,34,35 According to the Neidre and Macnab Keon-Cohen’s original description, 1 case of a cervi- classification system, all conjoined roots eventually - di cal confluent root has been reported,13 but our literature vide and exit via separate foramina. Conjoined nerve search did not find another case of a lumbosacral con- roots are further subdivided into those that arise from a fluent nerve root anomaly. Despite the apparent rarity of common dural sheath (Type 1A) and those that have ori- confluent nerve roots, it is still important for surgeons to gins very close together on the thecal sac but do not share be aware of this variant in the event it is encountered dur- a common dural sheath (Type 1B) (Fig. 6). The anomaly ing surgery. Therefore, we propose adding this as a Type encountered in Case 1 was a Type 1B anomaly. 4 variant in the current Neidre and Macnab classification Type 2. Type 2 anomalies involve redundant or system. “twinned” nerve roots, a situation in which 2 nerve roots Management of Nerve Root Anomalies During TLIF exit through 1 intervertebral foramen. Type 2 nerve roots are further subdivided according to the presence of an ex- Satisfactory bone removal is an important part of the tra root. If an extra nerve root is not present, the twinning TLIF procedure. Removal of the entire facet joint with in the intervertebral foramen will leave 1 foramen with no completion of pedicle-to-pedicle decompression creates nerve roots passing through it (Type 2A). If an extra nerve a large working space, or safety zone, lateral to the the- root is present, it occupies the would-be empty foramen cal sac and between the exiting nerve root and the cau- so that all the foramina are occupied (Type 2B) (Fig. 6). dal pedicle.23 In patients with normal anatomy, working

Fig. 6. Artistic rendering of the 3 classifications described by Neidre and Macnab (the areas of interest are highlighted): Type 1A, conjoined nerve root originating from the rostral nerve root itself rather than from the thecal sac; Type 1B, conjoined nerve root arising from the thecal sac near the caudal nerve root; Type 2A, 2 nerve roots exiting from the same neural foramen, leaving 1 unoccupied neural foramen; Type 2B, 2 nerve roots exiting from 1 neural foramen with all the other foramina occupied; and Type 3, 2 adjacent nerve roots with an anastomosis between them. Copyright Shane Burke. Published with permission.

4 Neurosurg Focus / Volume 35 / August 2013

Unauthenticated | Downloaded 10/07/21 05:33 PM UTC Nerve root anomalies: implications for TLIF

raminal dissection just rostral to the caudal pedicle. We then identify the exiting nerve root, paying close attention to the axilla of the nerve root. Next we explore the tissue between the axilla of the nerve root and the previously identified caudal disc. A nerve root anomaly, if present, would most likely be identified within this tissue. Typi- cally, however, this area consists only of epidural veins, which we coagulate and cut, to allow ample visualization of the entire disc space.19 At this time, it is safe to proceed with discectomy and placing the interbody graft. There are 3 more clues that should alert the surgeon to look closely for a nerve root anomaly during the operation: 1) an atypical location of a nerve root, as in Case 1, in which the L-4 nerve root exited just rostral to the L-5 pedicle instead of just caudal to the L-4 pedicle; 2) a nerve root that exits the thecal sac at an atypical angle; and 3) difficulty mobilizing a nerve root despite satisfactory pedicle-to-pedicle decompression, as in Case 2. Other authors who have identified nerve root anomalies during surgery have confirmed that these intraoperative findings are diagnostic.5,6,14,15,21 Another clue, although not confirmed by other authors, might be the absence of normal ligamentum flavum, as seen in Case 1. In some instances of nerve root anomalies, placing an interbody graft is not possible. There are multiple ways to address this situation. First, one must decide whether increasing the height of the intervertebral foramen is necessary to alleviate the patient’s symptoms. If so, this can be addressed by performing a contralateral facetectomy and placing an interbody graft on the contralateral side. One can also consider distracting the pedicle screws away from each other during rod placement to further open the foramen. This procedure may be suboptimal, however, Fig. 7. Artistic rendering of Case 2 showing a confluent nerve root because distraction is a kyphogenic maneuver. If increas- at the left L4–5 level. The area of interest is highlighted. This proposed ing the height of the intervertebral foramen is not neces- Type 4 variant illustrates a confluent nerve root with 2 contributions aris- sary, one could consider using a bilateral pedicle screw ing from the thecal sac coalescing to form a single L-4 nerve root that construct without an interbody graft.6 exits via the L4–5 intervertebral foramen. Specifically note how the caudal contribution is within the TLIF safety zone. Copyright Shane Burke. Published with permission. Conclusions Nerve root anomalies present unique surgical chal- within this safety zone minimizes the risk for injury to lenges and can increase the risk for iatrogenic nerve inju- the thecal sac or to the exiting nerve root during discec- ry. It is important to use a foraminal dissection technique tomy and placement of the interbody graft. However, as that enables recognition of nerve root anomalies before our experience demonstrates, patients with nerve root the nerves are iatrogenically injured. Of the 2 cases pre- anomalies are at risk for neural injury during TLIF if the sented here, 1 patient had a nerve root anomaly not pre- safety zone is not clear of anomalous roots.7,15,20,27 Thus, it viously classified: a confluent variant. We propose that is important to use a foraminal dissection technique that this variant be classified as a Neidre and Macnab Type 4 enables nerve root anomalies to be recognized before in- anomaly. juries occur. To minimize the likelihood of neural injury and to Disclosure identify any occult nerve root anomalies intraoperatively, The authors report no conflict of interest concerning the mate- we adhere to the following technique during all TLIF sur- rials or methods used in this study or the findings specified in this geries. First, as mentioned, we perform a complete fac- paper. etectomy to create a large working area. This can be done Author contributions to the study and manuscript preparation in several ways, and our method can be seen in Video 1. include the following. Conception and design: Riesenburger, Burke, Additional bone is removed until there is complete pedi- Safain. Acquisition of data: Riesenburger, Burke, Safain. Analy- sis and interpretation of data: all authors. Drafting the article: all cle-to-pedicle decompression. We then remove ligamen- authors. Critically revising the article: Riesenburger, Burke, Safain. tum flavum until we are able to identify the lateral border Reviewed submitted version of manuscript: all authors. Approved of the thecal sac. Then, to identify the caudal aspect of the final version of the manuscript on behalf of all authors: Riesen- the disc in the typical safety zone, we always start fo- burger. Study supervision: Riesenburger.

Neurosurg Focus / Volume 35 / August 2013 5

Unauthenticated | Downloaded 10/07/21 05:33 PM UTC S. M. Burke et al.

Acknowledgment 20. Maiuri F, Gambardella A: Anomalies of the lumbosacral nerve roots. Neurol Res 11:130–135, 1989 The authors thank Walter Dent for video and audio support. 21. Maiuri F, Gangemi M, Gambardella A: Anatomical and ra- diological variations of the lumbar dural sac and nerve root References sheaths. Clin Neurol Neurosurg 92:203–213, 1990 22. McCormick CC: Developmental asymmetry of roots of the cau- 1. Aota Y, Saito Y, Yoshikawa K, Asada T, Kondo S, Watanabe da equina at metrizamide myelography: report of seven cases K: Presurgical identification of extradural nerve root anoma- with a review of the literature. Clin Radiol 33:427–434, 1982 lies by coronal fat-suppressed magnetic resonance imaging: 23. Moskowitz A: Transforaminal lumbar interbody fusion. Or- a report of six cases and a review of the literature. J Spinal thop Clin North Am 33:359–366, 2002 Disord 10:167–175, 1997 24. Mummaneni PV, Rodts GE Jr: The mini-open transforaminal 2. Artico M, Carloia S, Piacentini M, Ferretti G, Dazzi M, lumbar interbody fusion. Neurosurgery 57 (4 Suppl):256–261, Franchitto S, et al: Conjoined lumbosacral nerve roots: obser- 2005 vations on three cases and review of the literature. Neurociru- 25. Neidre A, MacNab I: Anomalies of the lumbosacral nerve gia (Astur) 17:54–59, 2006 roots. Review of 16 cases and classification. Spine (Phila Pa 3. Böttcher J, Petrovitch A, Sörös P, Malich A, Hussein S, Kaiser 1976) 8:294–299, 1983 WA: Conjoined lumbosacral nerve roots: current aspects of 26. Pamir MN, Ozek MM, Ozer AF, Keleş GE, Erzen C: Surgical diagnosis. Eur Spine J 13:147–151, 2004 considerations in patients with lumbar spinal root anomalies. 4. Bouchard JM, Copty M, Langelier R: Preoperative diagnosis Paraplegia 30:370–375, 1992 of conjoined roots anomaly with herniated lumbar disks. Surg 27. Peyster RG, Teplick JG, Haskin ME: Computed tomography Neurol 10:229–231, 1978 of lumbosacral conjoined nerve root anomalies. Potential 5. Cannon BW, Hunter SE, Picaza JA: Nerve-root anomalies in cause of false-positive reading for herniated nucleus pulposus. lumbar-disc surgery. J Neurosurg 19:208–214, 1962 Spine (Phila Pa 1976) 10:331–337, 1985 6. Davidson D, Rowan R, Reilly C: Lumbosacral nerve root 28. Postacchini F, Urso S, Ferro L: Lumbosacral nerve-root anom anomaly associated with spondylolisthesis in an adolescent: alies. J Bone Joint Surg Am 64:721–729, 1982 a case report and review of the literature. Spine (Phila Pa 29. Prestar FJ: Anomalies and malformations of lumbar spinal 1976) 31:E718–E721, 2006 nerve roots. Minim Invasive Neurosurg 39:133–137, 1996 7. Epstein JA, Carras R, Ferrar J, Hyman RA, Khan A: Con- 30. Rask MR: Anomalous lumbosacral nerve roots associated joined lumbosacral nerve roots. Management of herniated with spondylolisthesis. Surg Neurol 8:139–140, 1977 discs and lateral recess stenosis in patients with this anomaly. 31. Salehi SA, Tawk R, Ganju A, LaMarca F, Liu JC, Ondra SL: J Neurosurg 55:585–589, 1981 Transforaminal lumbar interbody fusion: surgical technique 8. Ethelberg S, Riishede J: Malformation of lumbar spinal roots and results in 24 patients. Neurosurgery 54:368–374, 2004 and sheaths in the causation of low backache and sciatica. J 32. Savas R, Calli C, Yünten N, Alper H: Hypoplastic lumbar ped- Bone Joint Surg Br 34-B:442–446, 1952 icle in association with conjoined nerve root MRI demonstra- 9. Goldstein B: Anatomic issues related to cervical and lumbo- tion. Comput Med Imaging Graph 22:77–79, 1998 sacral radiculopathy. Phys Med Rehabil Clin N Am 13:423– 33. Scuderi GJ, Vaccaro AR, Brusovanik GV, Kwon BK, Berta SC: 437, 2002 Conjoined lumbar nerve roots: a frequently underappreciated 10. Gomez JG, Dickey JW, Bachow TB: Conjoined lumbosacral congenital abnormality. J Spinal Disord Tech 17:86–93, 2004 nerve roots. Acta Neurochir (Wien) 120:155–158, 1993 34. Song SJ, Lee JW, Choi JY, Hong SH, Kim NR, Kim KJ, et 11. Heary RF, Vaicys C, Grigorian AA: “Twinning” of a cervical al: Imaging features suggestive of a conjoined nerve root on nerve root mimicking a neoplasm. Case illustration. J Neuro- routine axial MRI. Skeletal Radiol 37:133–138, 2008 surg 97 (3 Suppl):398, 2002 35. Stambough JL, Balderston RA, Booth RE, Rothman RH: Sur- 12. Helms CA, Dorwart RH, Gray M: The CT appearance of con- gical management of sciatica involving anomalous lumbar joined nerve roots and differentiation from a herniated nucle- nerve roots. J Spinal Disord 1:111–115, 1988 us pulposus. Radiology 144:803–807, 1982 36. White JG III, Strait TA, Binkley JR, Hunter SE: Surgical 13. Higo S, Koizumi M, Kawai K, Honma S, Tokiyoshi A, Tama- treatment of 63 cases of conjoined nerve roots. J Neurosurg maki N, et al: Anomaly with no right ventral root at the sev- 56:114–117, 1982 enth cervical segment in humans: gross anatomical and neu- roanatomical study. Anat Sci Int 82:133–138, 2007 14. Kadish LJ, Simmons EH: Anomalies of the lumbosacral nerve Manuscript submitted February 7, 2013. roots. An anatomical investigation and myelographic study. J Accepted February 26, 2013. Bone Joint Surg Br 66:411–416, 1984 Please include this information when citing this paper: DOI: 15. Kang CH, Shin MJ, Kim SM, Lee SH, Kim HK, Ryu JA, et al: 10.3171/2013.2.FOCUS1349. Conjoined lumbosacral nerve roots compromised by disk her- Supplemental online information: niation: sagittal shoulder sign for the preoperative diagnosis. Video 1: http://mfile.akamai.com/21490/wmv/digitalwbc.download. Skeletal Radiol 37:225–231, 2008 akamai.com/21492/wm.digitalsource-na-regional/focus13-49_ 16. Keon-Cohen B: Abnormal arrangement of the lower lumbar video_1.asx (Media Player). and first sacral nerves within the spinal canal. J Bone Joint http://mfile.akamai.com/21488/mov/digitalwbc.download.akamai. Surg Br 50:261–265, 1968 com/21492/qt.digitalsource-global/focus13-49_video_1.mov 17. Kern M, Lee GYF: A rare anatomical variation of the C7 (Quicktime). pedicle and intraspinal course of the C7 nerve root. J Clin Video 2: http://mfile.akamai.com/21490/wmv/digitalwbc.download. Neurosci 15:1146–1148, 2008 akamai.com/21492/wm.digitalsource-na-regional/focus13-49_ 18. Kikuchi S, Hasue M, Nishiyama K, Ito T: Anatomic and clini- video_2.asx (Media Player). cal studies of radicular symptoms. Spine (Phila Pa 1976) 9: http://mfile.akamai.com/21488/mov/digitalwbc.download.akamai. 23–30, 1984 com/21492/qt.digitalsource-global/focus13-49_video_2.mov 19. Kraemer R, Wild A, Haak H, Herdmann J, Krauspe R, Krae- (Quicktime). mer J: Classification and management of early complications Address correspondence to: Ron Riesenburger, M.D., Depart in open lumbar microdiscectomy. Eur Spine J 12:239–246, ment of Neurosurgery, Proger 7, 800 Washington St., Boston, MA 2003 02110. email: [email protected].

6 Neurosurg Focus / Volume 35 / August 2013

Unauthenticated | Downloaded 10/07/21 05:33 PM UTC