ROUTINE POSTERIOR LONGITUDINAL LIGAMENT (PLL) RESECTION IS NOT NECESSARY FOR SURGICAL TREATMENT OF RADICULOPATHY
The need for PLL resection during decompression for cervical radiculopathy has been debated in the literature, with conclusions that the ultimate decision is surgeon and case-dependent. We agree that the PLL should be resected in cases in which the ligament is obviously causing neural compression (ie, ossification of the PLL) or if a subligamentous disk herniation is present. However, we contend that routine resection of the PLL for radiculopathy is not necessary and offers minimal additional clinical benefit to the patient.
Biochemical stains performed on PLLs of rat models identified that 2 distinctive systems of nociceptive fibers using calcitonin gene-related peptide and tyrosine hydroxylase, which have been considered markers of nociceptive and noradrenergic fibers respectively. A study investigating the source of axial pain found that of 88 patients who underwent single-level PLL resection, no patients had aggravated intraoperative pain responses except during resection of the PLL, implying that the intervertebral PLL could be the site of origination of axial pain.1 However, when this practice is implemented into the surgical technique, the clinical benefit of routine PLL resection is unclear. A literature review performed by Avila et al2 of 4 articles comprising 108 patients divided into groups that had their PLL resected versus retained found that both had improvements in clinical Japanese Orthopaedic Society scores during follow-up without a demonstrable clinical difference. Similarly, Bai and colleagues performed a retrospective study on 130 patients who underwent anterior cervical discectomy and fusion (ACDF) with 62 patients undergoing PLL resection and 68 patients having retained PLL. Despite better Japanese Orthopaedic Society scores and better recovery rate for spinal cord neurological function for patients with PLL resection at 36 months follow-up, the authors noted the operating time was significantly longer (P=0.002), required greater technical skill and more experienced surgeons in the PLL resection group.3
Retaining the PLL is associated with biomechanical advantages and may aid in providing constraints to overdistraction of the disk space and facets during placement of an interbody cage. Akaishi4 found that failure load, failure stress, and dynamic stiffness of the PLL tend to be higher than those of the anterior longitudinal ligament (ALL), implying that the PLL has better biomechanical properties as a stabilizer for cervical motion than the ALL in the lower cervical spine. Anterior cervical decompression significantly increases the instability and alters the kinematics between cervical motions segments. An in vitro cadaveric study by Chen et al5 demonstrated that PLL resection resulted in a significant range of motion increase in flexion-extension, lateral bending, and rotation, as well as shifts in the location of the axis of rotation. Another group developed a finite element model of C3–C7 and found that in models with a resected PLL there was a greater range of motion and stresses in the facet joint and flaval and capsular ligaments.6 This effectively transfers the burden of stability onto the implant and remaining soft tissue structures, which could potentially have short-term and long-term negative outcomes that have not yet been studied. This alteration in cervical kinematics becomes pertinent when utilizing cervical disk arthroplasty for the treatment of cervical radiculopathy. Current literature highlights the pivotal role the PLL plays in the biomechanical stability of the cervical spine after cervical disk arthroplasty. Animal model testing has shown significantly increased stability in anterior discectomy alone compared with anterior discectomy+PLL resection.7 These findings cannot be overlooked in the discussion of routine PLL resection during the treatment of cervical radiculopathy.
Routine resection of the PLL may create unnecessary perioperative risks and complications due to the intimate association of the PLL with dura mater. Injury to the spinal cord, neurological deficits, cerebrospinal fluid leaks, and epidural hematoma secondary to disruption of the epidural vascular plexus can lead to significant injury and disability. In the series by Wang et al,8 1 patient in the PLL removal group had a cerebrospinal fluid leak compared with none in which the PLL was retained. Nassr and colleagues performed a retrospective study on 459 patients who underwent cervical corpectomy and evaluated for risk factors for C5 palsy. The study noted that partial or complete resection of the PLL was associated with a 4× higher odds of sustaining a C5 palsy injury.9 Proponents of PLL resection argue it continues to have compressive effects if left intact. However, Chin et al10 found that in a series of 33 patients who continued to have symptoms after ACDF for cervical myeloradiculopathy, there were no cases of the retained PLL interpreted as compressing the spinal cord on the magnetic resonance image (MRI) postoperatively.
On the basis of a review of the published literature, there is no significant difference in clinical outcomes for PLL resection during cervical decompression. However, given the potential risk in complications and destabilizing effect of PLL resection, we contend that routine PLL resection should not be performed until further randomized prospective trials are performed to further delineate the risks and benefits of this procedure.
ROUTINE PLL RESECTION IS NECESSARY FOR THE SURGICAL TREATMENT OF RADICULOPATHY
Patients undergoing operative treatment of cervical radiculopathy often have several significant degenerative changes contributing to nerve root compression at the involved level, including herniated disks, endplate changes, arthritic uncovertebral joints with protruding osteophytes, and a hypertrophied PLL. All of these may simultaneously contribute to decreasing the space available for the exiting nerve root in the intervertebral neuroforamen. Although the majority of surgical treatment focuses on the bony work involved to resect the disk space and remove disk osteophyte complexes, controversy still exists on whether the PLL requires resection. We contend that adequate treatment for radiculopathy should involve complete resection of the PLL to visualize the exiting nerve root and ensure that no further compressive lesions exist.
While the PLL is present along the posterior aspect of the vertebral bodies in all areas of the spine, it is particularly wider at the level of the intervertebral disk in the cervical spine.11 The deep layer of the PLL runs along the posterior margin of the disk to eventually merge with the ALL at the anterior aspect of the foramen.12 Normally, it is agreed that the PLL provides a stabilizing role in the cervical spine as well as a protective role, by preventing disk or bony material from entering the spinal canal.2 However, when ligamentous integrity is compromised, disk or bony material may enter the canal and foramen. Therefore, visualization of any herniated or sequestered disk fragments posterior to the PLL requires either palpation via instruments or PLL resection to directly assess whether neural elements are being compressed. Elayouty et al13 retrospectively identified 50 patients that underwent ACDF with routine microscopic PLL resection and found satisfactory clinical outcomes with significant improvement in arm and neck pain as well as motor grades without an additional increase in complications.
Some authors have postulated that MRI is useful in predicting whether disk herniation material is anterior or posterior to the PLL and thus help guide the need for surgical resection of the PLL.14 The annulus fibrosus of the intervertebral disk and the PLL are comprised primarily of collagen and confer a lower intensity signal on T1-weighted and T2-weighted sequences. Therefore, as the posterior annulus fibrosus merges with the PLL, differentiation between the 2 structures on MRI may be difficult. An increase in T2 signal in the PLL may signify a discontinuity in the ligament and increase suspicion for disk material posterior to the PLL. Humphreys et al14 analyzed preoperative MRIs in 54 patients undergoing surgical treatment for clinical symptoms of radiculopathy. In 26 patients, the authors found intraoperative disk material posterior to the PLL, however, MRI failed to predict this in 14 of the cases. They concluded that because of the low sensitivity, MRI should be used cautiously in predicting disk material posterior to the PLL and direct intraoperative examination is required for an accurate assessment. In the study by Chin et al,10 where 33 patients were reimaged with MRIs for persistent symptoms after undergoing ACDF with PLL retention, the PLL was not identified as causing significant compression. However, the authors also suggested that MRI should be used cautiously for predicting free disk material posterior to the PLL due to its low sensitivity.
Risk of complications and increased operative time are 2 additional concerns in routine PLL resection. The close proximity of the dura mater to the posterior aspect of the PLL increases the risk for a dural tear and cerebrospinal fluid leak. However, the interface between these 2 structures involves a loose attachment that can be easily separated with blunt dissection techniques.2,11 Microcurettes and blunt-ended hooks can be used to gently separate the epidural membrane from the PLL. One exception where dissection may be difficult is in the case of ossification of the PLL, where the PLL is calcified and adhered onto the dura. In these cases, adherent islands of PLL can be left on the dura and adjacent areas can be safely removed. Proper technique can prevent dural tears during manipulation and PLL resection.
Biomechanical studies suggest that removal of the PLL causes increasing instability.5 However, in surgical cases where the operative segment is stabilized with instrumentation and plate fixation, the structural benefits of retaining the PLL are questionable.10 The rigidity provided by plate fixation is substantially more than that provided by the native ligament. Even in cases of cervical disk arthroplasty, PLL resection does not seem to significantly affect motion segment kinematics if a prothesis with sufficient stiffness is used. In a cadaveric study with 8 specimens, Voronov et al15 found that PLL resection at 1 or 2 levels did not significantly affect motion segment kinematics including flexion and extension, lateral bending, and axial rotation when using an arthroplasty design that compensates by providing resistance to rotational forces.
Some surgeons believe that discectomy with intervertebral distraction and graft placement sufficiently increases the area in the neuroforamen to relieve radiculopathy symptoms. Others opt to make the intraoperative decision to resect the PLL if there is evidence of a ligamentous compromise to search for any disk fragments posterior to the PLL. We assert that routine takedown of the PLL in all patients undergoing surgery for radiculopathy symptoms ensures that no compressive fragments are retained in the canal or neuroforamen.
1. Lin W, Xue Y, Zhao Y, et al. Disc associating axial pain were indicated by PLL resection in ACDF surgery. Eur Spine J. 2017;26:1211–1216.
2. Avila MJ, Skoch J, Sattarov K, et al. Posterior longitudinal ligament resection or preservation in anterior cervical decompression surgery. J Clin Neurosci. 2015;22:1088–1090.
3. Bai C-R, Wang B-Q, Li K-H, et al. Benefit of degenerative posterior longitudinal ligament removal during anterior decompression in cervical spondylotic myelopathy. Orthopedics. 2015;38:e54–e61.
4. Akaishi F. Biomechanical properties of the anterior and posterior longitudinal ligament in the cervical spine [in Japanese]. Nihon Ika Daigaku Zasshi. 1995;62:360–368.
5. Chen TY, Crawford NR, Sonntag VK, et al. Biomechanical effects of progressive anterior cervical decompression. Spine (Phila Pa 1976). 2001;26:6–13; discussion 14.
6. Qi L, Zhongjun M, Fan Y. Biomechanical comparison of cervical arthroplasty with and without posterior longitudinal ligament resection: a finite element study. J Med Imaging Health Inform. 2016;6:1559–1565.
7. McAfee PC, Cunningham B, Dmitriev A, et al. Cervical disc replacement-porous coated motion prosthesis: a comparative biomechanical analysis showing the key role of the posterior longitudinal ligament. Spine (Phila Pa 1976). 2003;28:S176–S185.
8. Wang X, Chen Y, Chen D, et al. Removal of posterior longitudinal ligament in anterior decompression for cervical spondylotic myelopathy. J Spinal Disord Tech. 2009;22:404–407.
9. Nassr A, Aleem IS, Eck JC, et al. Does resection of the posterior longitudinal ligament impact the incidence of C5 palsy after cervical corpectomy procedures?: a review of 459 consecutive cases. Spine (Phila Pa 1976). 2017;42:E392–E397.
10. Chin KR, Ghiselli G, Cumming V, et al. Postoperative magnetic resonance imaging assessment for potential compressive effects of retained posterior longitudinal ligament after anterior cervical fusions: a cross-sectional study. Spine (Phila Pa 1976). 2013;38:253–256.
11. Loughenbury PR, Wadhwani S, Soames RW. The posterior longitudinal ligament and peridural (epidural) membrane. Clin Anat. 2006;19:487–492.
12. Hayashi K, Yabuki T, Kurokawa T, et al. The anterior and the posterior longitudinal ligaments of the lower cervical spine. J Anat. 1977;124:633–636.
13. Elayouty AED, Mashaly HA, Sabry HA. Posterior longitudinal ligament resection in anterior cervical discectomy. Egypt J Neurosurg. 2018;33:3.
14. Humphreys SC, Hodges SD, Fisher DL, et al. Reliability of magnetic resonance imaging in predicting disc material posterior to the posterior longitudinal ligament in the cervical spine: a prospective study. Spine (Phila Pa 1976). 1998;23:2468–2471.
15. Voronov LI, Havey RM, Tsitsopoulos PP, et al. Does resection of the posterior longitudinal ligament affect the stability of cervical disc arthroplasty? Int J Spine Surg. 2018;12:285–294.