Case Report | Open Access
Surgical Management of Intracanal Rib Head Dislocation in Neurofibromatosis Type 1 Dystrophic Kyphoscoliosis: Report of Two Cases and Literature Review
There is still no consensus on the management of severe intracanal RH dislocation in neurofibromatosis type 1 dystrophic kyphoscoliosis. This study notes the early cord function impairment signs, reports a serious complication in a susceptible cord, identifies possible mechanisms of injury, and discusses the management of intracanal RH dislocation presented in the literature. First report is as follows: a 12-year-old female with cord compromise and preoperative neurology that underwent thoracotomy and anterior release. The RH was left in situ following a rib excision. During the posterior stage of the procedure she presented with complete loss of all IOM traces prior to any correction manoeuvres. The neurology recovered 72 h postop and the final correction and instrumented fusion were uneventfully completed 15 days postop. Second report is as follows: a 10-year-old male, whose only neurology was a provoked shock-like sensation to the lower limbs following direct pressure on the rib cage. He underwent an uneventful posterior RH excision and instrumented correction and posterior spinal fusion. In conclusion, any possible cord dysfunction sign should be sought during examination. Decompression of the spinal cord by resecting the impinging bony part, even in the absence of neurological symptoms, is advised before any attempt to release or correct the deformity.
Neurofibromatosis type 1 (NF1), also known as von Recklinghausen’s disease, is a single gene hamartomatous disease inherited by the autosomal dominant trait [1, 2]. The relentless deterioration of the short dystrophic curves, which leads to acute kyphosis and possible vertebral subluxation, mandates surgical stabilization [1, 3]. Intracanal rib head (RH) dislocation at the convex of the dystrophic curve may impinge on the cord and constitute another cause of neurology [3–9]. The “painful rib hump” sign caused by the RH dislocation has recently been described .
Even though there are recent reports of retraction of the RH away from the cord along with curve correction [2, 10, 11], there is still no consensus among authors that the excision of the dislocated RHs is indicated in symptomatic NF1-dystrophic curves [5, 8, 12, 13]. The preoperative identification of a “cord at risk,” the susceptibility of the cord to intraoperative manipulations, and the sequence of an iatrogenic injury are not fully described yet.
We report two selected cases: a case of intraoperative neurological injury in a previously symptomatic patient with cord changes and a case of “hidden” neurology in an otherwise asymptomatic patient. The purpose of this study is to report a serious complication, identify a possible mechanism of injury, highlight the importance of early neurology, and discuss management of intraspinal RH dislocation.
2. Case 1
A 12-year-old premenarchal girl presented with back pain due to dystrophic NF1 left thoracic kyphoscoliosis (Figure 1(a)). She had bilateral brisk patellar reflexes and a left four beat ankle clonus. The patient reported shooting pain into her legs on deep forward bending. Preoperative MRI and CT scans revealed a flattened cord at the apex of the kyphosis with a penetrating left 6th RH adjacent to the cord without compression (Figures 1(b)–1(d)).
A combined fusion and posterior RH excision were planned with the aid of multimodal spinal cord monitoring. Following a thoracotomy the left 6th rib was excised, leaving the neck and head in situ. During the T4–T9 discectomies a transient loss of Motor Evoked Potentials (MEPs) occurred bilaterally in lower limbs, which responded to an increase in the mean arterial blood pressure. Somatosensory Evoked Potentials (SSEPs) remained normal.
Normal reference traces were present at the beginning of the second stage: posterior instrumented spinal fusion. After completion of the instrumentation and prior to correction manoeuvres all MEPs and SSEPs were lost completely. Following a laminectomy, the dislocated RH, which was not adherent to the dura but was impinging on the cord, was excised. Wake-up test showed no spontaneous movement in the lower limbs with good upper limbs movement.
Postoperative neurological examination showed grade 3 muscle power (MRC grading) in all muscle groups of the left lower limb. Right lower limb was normal. Fine touch and proprioception remained intact bilaterally. 48 h postop MRI scan showed no evidence of cord signal changes. 72 h postop the patient regained normal muscle power and urinary continence.
Fifteen days postoperatively the patient underwent posterior correction of the deformity. The MEPs remained stable during the procedure. Postoperative radiographs evidenced a main thoracic curve of 45° (51% correction). Lateral radiographs and CT scan confirmed a correction of the thoracic kyphosis to 48° (36% correction). At 2-year follow-up (Figure 1(e)) there has been no significant loss of correction and the patient remains asymptomatic.
3. Case 2
A 10-year-old boy presented with a rapidly progressive spinal deformity and scapulae asymmetry, due to dystrophic NF1 sharp angular proximal thoracic kyphoscoliosis (Figure 2(a)). There was no neurologic deficit. Interestingly, the patient complained of discomfort only when lying on his right side with shock-like sensations to his lower limbs bilaterally. The CT/MRI imaging shows intracanal dislocation of the right 4, 5, and 6th ribs, with the 5th being in contact with the spinal cord and no cord-substance high signal present (Figures 2(b)–2(d)).
He underwent posterior resection of the 5th RH through a hemilaminectomy approach. Following alar/coronal ligament release, the RH was extracted with stable spinal cord monitoring traces. The posterior instrumented spinal fusion was then completed uneventfully (Figure 2(e)). A CTLSO was used postoperatively.
The canal expansion due to dural ectasia is probably protective, regarding the early development of neurological symptoms . Apart from possible neurofibromas, the acute kyphotic deformity and instability may lead to neurology compromise due to secondary cord injury [1–7, 9, 12, 14, 15].
The dislocation of the RH into the canal is caused by the insufficiency of the costovertebral/costotransverse articulation and medial-ward pressure by the thoracic cage [1–3]. It takes place through the enlarged neural foramina at the apex of the convexity of the curve . The RH may impinge on the cord [3–9] and may cause neurology [3, 7–9]. Gkiokas et al.  described clearly the mobile RH, which causes neurology and pain in an otherwise neurologically intact patient, introducing the “painful rib hump” sign. Lhermitte’s type phenomenon might represent early sign of cord irritation, manifesting as shock-like sensation to lower limbs [4, 12]. This was the only clinical finding on our 2nd case, indicating an unstable RH and possible early cord injury.
MRI with contrast is the imaging modality of choice for soft tissue lesions in NF and the T2W sequences may demonstrate the intraspinal RH dislocation . However, there are reports where preoperative MRI failed to diagnose the intraspinal RH dislocation leading to postoperative neurological injury [5, 9, 15]. The CT scan can reliably demonstrate the intracanal RH dislocation because of better delineation of bony anatomy [3, 13]. The preoperative evaluation of the imaging should be cautious, as the RH and spine positions without gravitational forces may underestimate the degree of canal intrusion.
Surgical treatment is difficult due to excessive bleeding and the distorted anatomy. High pseudoarthrosis rates have been reported with posterior-only fusion in dystrophic scoliosis and therefore generous anteroposterior fusion has been advocated for curves exceeding 50° of kyphosis .
No consensus exists in the literature regarding the surgical sequence for the treatment of intraspinal RH dislocation and scoliosis correction.
A literature review reveals 16 publications with a total of 49 patients with NF1 and intraspinal dislocation of the rib (Table 1). Most of the patients were teenagers with a mean age of 13 years (average: years, range: 6–41 y) with almost 1 : 1 male to female ratio (25f : 23m). Neurological status was reported in 24 cases. Neurological deficits were present in 50% of the reported cases. Two or more ribs penetrating into the canal were presented into 43% of the nonsymptomatic patients and in 42% of the symptomatic patients. In 47% (9/19) of the reported cases the RH was in close proximity or was impinging on the spinal cord. Six of the nine patients with a RH in close proximity to the spinal cord underwent resection. Of the three patients in whom the RH was not removed during scoliosis fusion, two patients developed delayed neurological deficits requiring subsequent decompression. The intraoperative monitoring traces were lost in the patient who had his RH left in situ and the distal rib resected .
|This table shows all published studies in the English literature to date, which are reporting on the management of intracanal rib head dislocation in neurofibromatosis type 1 dystrophic curves; level of evidence (LoE) V, case series: (LoE) IV, PSF: posterior spinal fusion, RH: rib heads, and LL: lower limbs. Op: operation.|
Among all the reported cases with intracanal dislocation, 32 patients received PSF only; one had noninstrumented in situ PSF  and one had instrumented in situ PSF following previous correction by halo traction . The in situ noninstrumented PSF led to lower limb weakness and paraparesis 6 weeks postoperatively with the authors not clarifying the cause of the neurology deterioration . There were 19 staged operations with 11 of them being combined anterior and posterior. Of the 11 patients, 8 had ASF and 6 underwent the anterior procedure first.
Mao et al.  and Sun et al.  showed that spontaneous RH reduction occurred, following curve correction. Yalcin et al.  also observed under direct vision how the RH migrated out of the canal during scoliosis correction. They concluded that, in the presence of neurological symptoms or evidence of compression, resection of the rib prior to any surgical manipulation (release or correction) is necessary . In asymptomatic patients with no evidence of spinal cord compression, RH excision was considered questionable. Yalcin et al.  suggested direct visualisation of the RH via hemilaminectomy during correction manoeuvres.
Abdulian et al.  offer a different point of view, advocating resection of every intracanal RH dislocation. In addition, the authors recommended rib shaft osteotomy in cases where the RH remains unresectable because of cord adhesions.
Similar to our case 1, in the report by Mukthar et al.  the spinal cord monitoring traces were lost following resection of the rib shaft and leaving the RH in situ. According to Leung et al.  a spinal cord at risk is more likely to demonstrate intraoperative monitoring changes and those changes are twice likely to be associated with postoperative neurological deficit. Cheh et al.  reported loss of MEPs in 21% of paediatric kyphosis correction which was attributed mainly to hypotension, overcorrection, or combination of the two and was completely reversed by increasing the mean arterial pressure or reducing the magnitude of correction. Shimizu et al.  in their recent animal study showed that severe kyphosis causes demyelination, reduced blood supply, and neuronal loss of anterior horn cells.
Several factors may have contributed to the transient neurological injury to herein reported case 1. Vascular insufficiency during the anterior discectomy stage may have caused the drop of the IOM traces, as they recovered after an increase in the mean arterial blood pressure. The RH was not removed during the anterior stage as this would result into greater haemorrhage and possible further secondary vascular insult to the cord. Also, there is still no strong evidence in the literature to suggest a direct injury to the cord by a remaining RH stump. The subsequent prone position and the posterior facetectomies may have contributed to a degree of spinal instability and alteration of the kyphoscoliotic angle. Also, the change in alignment with translation of the spinal cord towards the convexity may have produced compression by the penetrating RH. Furthermore, the RH, having lost its lateral stabilizers due to the thoracotomy and rib resection, might have adopted a new medial position further impinging on the cord. Similar to our second case, it may be safer to relieve the cord from the RH compression, prior to any release or correction, which will change the relationships in the acute kyphos.
We postulate that the flail RH, left in the spinal canal following a rib-only resection, may be risky, because it loses the stability provided by the rib cage and because it may potentially change position. Secondly the loose RH will not reduce by ligamentotaxis during translational correction manoeuvres. In cases where the RH remains unresectable, because of cord adherence, translational correction manoeuvres should probably be avoided. We also believe that even asymptomatic cases without gross MRI cord signal changes should still be investigated for subtle signs of cord impairment, which might render it vulnerable during correction manoeuvres.
Even though the intracanal RH dislocation is a well-documented manifestation of the NF kyphoscoliosis, its severe form is not frequent enough for any conclusion to be supported by large number of cases. However, the conclusion is reasonable and is in line with the experience presented in all reports on the same topic in the literature. Preoperatively, we would suggest a thorough clinical investigation for cord impairment signs (“painful rib hump,” Lhermitte’s-like, etc.) and precise imaging for the intracanal RH position in relation to the cord. We would recommend excision of the RH, if in close proximity to the spinal cord, prior to attempting anterior spinal release or posterior correction manoeuvres. This sequence will also aid in the correct identification of the cause of a possible intraoperative IOM event.
Rib head intracanal dislocation is a dystrophic feature of patients with NF1 scoliotic curves. The protruding part of the rib although usually asymptomatic may cause neurological impairment by impinging on the spinal cord.
Provoked neurological signs should be sought during clinical examination in order to identify any cord dysfunction. CT and MRI scans should be performed to diagnose the extent of rib head penetration or cord involvement and to assist in surgical planning.
Decompression of the spinal cord by resecting the impinging bony part, even in the absence of neurological symptoms, is advised before any attempt to release or correct the deformity. This strategy seems to be the safest and will aid the surgeon and the neurophysiologist in discriminating the cause of possible positive IOM events during surgery.
The authors did not receive grants or outside funding in support of their research or preparation of this paper. They did not receive payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity. They have declared that no commercial entity paid, or directed, or agreed to pay or direct, any benefits to any research fund, foundation, educational institution, or other charitable or nonprofit organizations with which the authors are affiliated or associated.
- A. I. Tsirikos, A. Saifuddin, and M. H. Noordeen, “Spinal deformity in neurofibromatosis type-1: diagnosis and treatment,” European Spine Journal, vol. 14, no. 5, pp. 427–439, 2005.
- N. Yalcin, E. Bar-on, and M. Yazici, “Impingement of spinal cord by dislocated rib in dystrophic scoliosis secondary to neurofibromatosis type 1: radiological signs and management strategies,” Spine, vol. 33, no. 23, pp. E881–E886, 2008.
- M. Cappella, N. Bettini, E. Dema, M. Girardo, and S. Cervellati, “Late post-operative paraparesis after rib penetration of the spinal canal in a patient with neurofibromatous scoliosis,” Journal of Orthopaedics and Traumatology, vol. 9, no. 3, pp. 163–166, 2008.
- J. Ton, R. Stein-Wexler, P. Yen, and M. Gupta, “Rib head protrusion into the central canal in type 1 neurofibromatosis,” Pediatric Radiology, vol. 40, no. 12, pp. 1902–1909, 2010.
- K. I. Khoshhal and R. D. Ellis, “Paraparesis after posterior spinal fusion in neurofibromatosis secondary to rib displacement: case report and literature review,” Journal of Pediatric Orthopaedics, vol. 20, no. 6, pp. 799–801, 2000.
- M. H. Abdulian, R. W. Liu, J. P. Son-Hing, G. H. Thompson, and D. G. Armstrong, “Double rib penetration of the spinal canal in a patient with neurofibromatosis,” Journal of Pediatric Orthopaedics, vol. 31, no. 1, pp. 6–10, 2011.
- I. A. Mukhtar, M. Letts, and K. Kontio, “Spinal cord impingement by a displaced rib in scoliosis due to neurofibromatosis,” Canadian Journal of Surgery, vol. 48, no. 5, pp. 414–415, 2005.
- A. Gkiokas, S. Hadzimichalis, E. Vasiliadis, M. Katsalouli, and G. Kannas, “Painful rib hump: a new clinical sign for detecting intraspinal rib displacement in scoliosis due to neurofibromatosis,” Scoliosis, vol. 1, no. 1, article 10, pp. 1–4, 2006.
- M. Deguchi, N. Kawakami, H. Saito, K. Arao, K. Mimatsu, and H. Iwata, “Paraparesis after rib penetration of the spinal canal in neurofibromatous scoliosis,” Journal of Spinal Disorders, vol. 8, no. 5, pp. 363–367, 1995.
- S. Mao, B. Shi, S. Wang et al., “Migration of the penetrated rib head following deformity correction surgery without rib head excision in dystrophic scoliosis secondary to type 1 Neurofibromatosis,” European Spine Journal, vol. 24, no. 7, pp. 1502–1509, 2015.
- D. Sun, F. Dai, Y. Y. Liu, and J.-Z. Xu, “Posterior-only spinal fusion without rib head resection for treating type I neurofibromatosis with intracanal rib head dislocation,” Clinics, vol. 68, no. 12, pp. 1521–1527, 2013.
- B. M. Flood, W. P. Butt, and R. A. Dickson, “Rib penetration of the intervertebral foraminae in neurofibromatosis,” Spine, vol. 11, no. 2, pp. 172–174, 1986.
- M. R. Major and B. A. Huizenga, “Spinal cord compression by displaced ribs in neurofibromatosis. A report of three cases,” Journal of Bone and Joint Surgery—Series A, vol. 70, no. 7, pp. 1100–1102, 1988.
- M. G. Lykissas, E. K. Schorry, A. H. Crawford, S. Gaines, M. Rieley, and V. V. Jain, “Does the presence of dystrophic features in patients with type 1 neurofibromatosis and spinal deformities increase the risk of surgery?” Spine, vol. 38, no. 18, pp. 1595–1601, 2013.
- J. N. Dacher, S. Zakine, M. Monroc, D. Eurin, J. Lechevallier, and P. Le Dosseur, “Rib displacement threatening the spinal cord in a scoliotic child with neurofibromatosis,” Pediatric Radiology, vol. 25, no. 1, pp. 58–59, 1995.
- A. H. Crawford, “Neurofibromatosis,” in The Pediatric Spine: Principles and Practice, S. L. Weinstein, Ed., pp. 471–490, Lippincott Williams & Wilkins, Philadelphia, Pa, USA, 2001.
- Y. L. Leung, M. Grevitt, L. Henderson, and J. Smith, “Cord monitoring changes and segmental vessel ligation in the ‘at risk’ cord during anterior spinal deformity surgery,” Spine, vol. 30, no. 16, pp. 1870–1874, 2005.
- G. Cheh, L. G. Lenke, A. M. Padberg et al., “Loss of spinal cord monitoring signals in children during thoracic kyphosis correction with spinal osteotomy: why does it occur and what should you do?” Spine, vol. 33, no. 10, pp. 1093–1099, 2008.
- K. Shimizu, M. Nakamura, Y. Nishikawa, S. Hijikata, K. Chiba, and Y. Toyama, “Spinal kyphosis causes demyelination and neuronal loss in the spinal cord: a new model of kyphotic deformity using juvenile Japanese small game fowls,” Spine, vol. 30, no. 21, pp. 2388–2392, 2005.
- S. V. Kamath, P. K. Kleinman, R. L. Ragland et al., “Intraspinal dislocation of the rib in neurofibromatosis: a case report,” Pediatric Radiology, vol. 25, no. 7, pp. 538–539, 1995.
- B. Legrand, G. Filipe, A. Blamoutier, N. Khouri, and P. Mary, “Intraspinal rib penetration in four patients in neurofibromatosis vertebral deformities,” Revue de Chirurgie Orthopédique et Réparatrice de L'appareil Moteu, vol. 89, no. 1, pp. 57–61, 2003.
- R. Krishnakumar and J. Renjitkumar, “Teaching NeuroImages: rib penciling and intraspinal dislocation of rib heads in type 1 neurofibromatosis,” Neurology, vol. 78, no. 13, article e85, 2012.
Copyright © 2016 George I. Mataliotakis et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.