Is atlantoaxial instability the cause of type 3 Klippel-Feil anomaly? Outcome analysis of 99 patients treated by atlantoaxial fixation: A case series

ABSTRACT

Aim:

The authors analyze their experience with 99 patients having Type 3 Klippel-Feil anomaly (KFA), signifying fusion of cervical vertebrae at and below the level of C3 vertebra.

Materials and Methods:

During the period January 2004–November 2024, 99 patients presented with clinical symptoms and radiological characteristics that indicated the presence of Type 3 KFA and associated musculoskeletal and neural alterations. There were 57 males and 42 females, and their ages ranged from 4 to 54 years (average 26 years). Apart from symptoms that included longstanding pain in the nape of the neck, spasm of neck muscles, and restricted neck movement, all patients had a range of spastic quadriparesis and sensory dysfunction. Associated entities included Chiari formation (30 cases), syringomyelia (17 cases), Group A (58 cases) and Group B basilar invagination (12 cases), bifid arch of atlas (17 cases), assimilation of atlas (49 cases), C2–3 fusion (44 cases), os-odontoideum (3 cases), torticollis (29 cases), and dorsal kyphoscoliosis (3 cases).

Results:

All patients were treated by atlantoaxial fixation. No bone decompression was done at any spinal level. The follow-up ranged from 3 to 189 months. Clinical improvement was observed in all patients.

Conclusions:

Vertebral fusion/s in KFA and other associated or related clinically obvious musculoskeletal and radiologically obvious neural compressive, neural deforming, or neural altering and otherwise considered “pathological” entities are secondary to chronic atlantoaxial instability, are naturally protective or adaptive, and are potentially or manifestly reversible following atlantoaxial stabilization. Successful atlantoaxial fixation can lead to a gratifying “life-changing” clinical outcome.

INTRODUCTION

The clinical entity of Klippel-Feil anomaly (KFA) or cervical vertebral fusion syndrome has been under discussion since its first description by Klippel and Feil in 1912.[1,2] Fusion of cervical vertebrae, short neck, restricted neck movements, and low hairline are hallmarks of presenting clinical features. Other than these primary issues, a significant majority of these patients harbor several cranial and spinal neural and nonneural and systemic clinical symptoms and musculoskeletal alterations. Although most cases are sporadic, meaning that there is no family history, the general existing opinion is that failure of segmentation during embryonic development, probably related to genetic abnormality, is the cause of this entity.

Depending on the levels of cervical vertebral fusions, KFA has been subclassified earlier by several authors.[2,3,4,5,6] We incorporated the gist of earlier classifications and classified KFA into 3 types. Type 1 KFA was where there was only assimilation of the atlas. Type 2 KFA was when there was C2–3 fusion. Type 3 KFA was when there was cervical vertebral fusion at and below the C3 vertebral level. In the presented study, patients having only Type 3 KFA have been studied.

In this report, we analyze the management of 99 cases having Type 3 KFA, all presenting with a clinical syndrome suggestive of moderate-to-severe myelopathy. Our previously published concept that identifies chronic or longstanding atlantoaxial instability that is more often of central or axial variety (CAAD) as the nodal point of pathogenesis of several clinical entities that include KFA is elaborated and analyzed.[7,8,9]

MATERIALS AND METHODS

During the study period January 2004–November 2024, 99 patients having Type 3 KFA who were surgically treated in our departments of neurosurgery were retrospectively analyzed. All procedures were performed in accordance with ethical standards and in accordance with the principles of the Declaration of Helsinki.

There were 57 males and 42 females, and their ages ranged from 4 to 54 years (average 26 years). The follow-up ranged from 3 to 189 months (average 29 months). Table 1 summarizes the major clinical signs and symptoms at the time of presentation. Duration of clinical symptoms ranged from 9 to 36 months. There was no history of major trauma prior to the onset of clinical symptoms in any cases. Patients who had undergone any kind of surgery to the cervical spine were excluded. Also excluded from the study were those patients who had any evidence of infectious spinal disease. Table 2 summarizes the clinical and neurological assessments that were based on parameters of the Goel clinical grading scale,[10] Japanese Orthopedic Association score, and the Visual analog scale. As per the classification of KFA, all patients had multiple-level cervical spinal fusion, short neck, and low hairline. All patients had “long-duration” moderate to severe pain in the nape of the neck that exacerbated on neck flexion, restriction of the neck movements and spasm of the muscles of the nape of neck. All patients had clinical evidence of myelopathy. Apart from other clinical deficits, 11 patients had low volume or hoarse voice, 29 patients had sleep apnea, and 38 patients had breathlessness on moderate activity. At the time of admission for surgery, 17 patients needed heavy support to walk, and 9 patients were wheelchair bound.

Table 1.

The presenting clinical symptoms

Presenting symptom	Number of patients
Neck pain	99
Restriction of neck movements	99
Weakness in limbs	82
Spasticity	99
Paresthesia	74
Bowel/bladder symptoms	7
Sleep apnea	29
Hoarseness/nasal regurgitation	11

Table 2.

The pre- and postoperative clinical grades of the patients

Clinical grading	Preoperative (99 patients)	Postoperative (at last follow-up -patients)
Goel’s clinical grading
Grade 1- Independent and normally functioning	-	68
Grade 2- Walks on own but needs support/help to carry out routine household activities	38	21
Grade 3 - Walks with minimal support and requires help to carry out household activities	35	10
Grade 4 - Walks with heavy support and is unable to carry out household activities	17	-
Grade 5 - Unable to walk and dependent for all activities (L after the grade indicates lower cranial nerve weakness)	9	-
JOA score
<7	26	-
8–12	34	8
13–15	30	23
16–17	9	68
VAS score
Neck pain	0–9 (6.6)	0–1 (0.2)

VAS: Visual Analogue Scale, JOA: Japanese Orthopaedic Association

All patients underwent dynamic plain radiography and computed tomography (CT) scan with the head in flexion, extension, and neutral position, and magnetic resonance imaging. Table 3 lists the major associated so-called “pathological” clinical entities. This case series has been reported in line with the PROCESS Guideline.[11]

Table 3.

The associated radiological abnormalities

	Number of patients
Chiari	30
Basilar invagination group A	58
Basilar invagination group B	12
Type A atlantoaxial instability	82
CAAD (type B and C instability)	17
Syringomyelia	17
External syringomyelia	41
Platybasia	6
Assimilation of atlas	49
C2–3 fusion	44
Bifid C1 arch	17
Bifid C2 arch	3
Os - odontoideum	3
Short neck	99
Torticollis	29
Dorsal Kyphoscoliosis	3
Omohyoid bone	3
Subaxial spinal cord signal indicating compression	9

CAAD: Central or axial atlantoaxial dislocation

Surgical technique

All patients underwent atlantoaxial fixation using the technique described by the first author in 1994.[12,13] The atlantoaxial joint was widely opened, articular cartilage was denuded, bone graft pieces were impacted into the articulation and plate, and screw fixation was then done. A bone graft was harvested from the iliac crest. No decompression by removal of the bone or soft tissues of the foramen magnum was done in any case. No manipulation of bones or soft tissues of the subaxial spine was done.

Special surgical issues

Short neck and multiple bone fusions in several cases that included assimilation of atlas (49 cases), platybasia (6 cases), Group A basilar invagination (58 cases), and Group B basilar invagination (12 cases) made the location of the atlantoaxial articulation significantly rostral, making the exposure during surgery technically challenging. In two cases, the fixation could be done only on one side due to technical difficulties in exposure of the contralateral articulation. In one patient, due to difficulty in the placement of a screw in the pedicle and superior facet of C2 on both sides, screws were inserted on the inferior facet of C2 vertebra. In two patients, on one side our technique of C1–C2 fixation and on the other side Magerl’s technique of C1–C2 transarticular screw fixation[14] was performed. In two patients, on one side, transcranial insertion of a C1 screw[15] and on the other side, transarticular screw fixation was performed. In none of the cases was the occipital bone or subaxial spinal segments included in the fixation construct.

Clinical outcome

The follow-up ranged from 3 to 189 months (average 29 months). Table 2 summarizes the clinical outcome. In addition to the other parameters, patient self-assessment and patient interviews, both before and after surgery, and at the last follow-up, were used to analyze the clinical outcome. Wherever available, video recordings were additionally used to assess clinical condition both prior to and after surgery and at the time of follow-up. All patients improved in their breathing, in their voice quality, and were completely relieved of sleep apnoeas. Although no measurements were made, visual inspection and comparisons with available photographs indicated that the neck size increased in all patients. All patients walked unaided. No patient underwent re-exploration for a failed surgical procedure or for recurrence of symptoms, failure of implant, or infection [Figures 1-4].

Figure 1.

Images of a 23-year-old male patient. (a) T2-weighted magnetic resonance imaging image showing Group A basilar invagination, assimilation of atlas, C2–3 and C4–5 fusions. Severe compression of the spinomedullary cord is seen opposite the odontoid process. External syringomyelia can be observed in the subaxial spine. (b) Sagittal cut of computed tomography (CT) scan in flexion showing Group A basilar invagination, atlantoaxial dislocation, assimilation of atlas and C2–3 and C4–5 fusions. (c) Sagittal cut of the CT scan in extension showing incomplete reduction of the dislocation. (d) Sagittal cut of CT scan passing through the facets showing listhesis of the facet of atlas over the facet of axis (Type A facetal instability). (e) Three-dimensional reconstructed CT image showing the C2–3 and C4–5 fusions. (f) Postoperative CT image showing reduction of the atlantoaxial dislocation. (g) Postoperative CT image showing the implants

Figure 4.

Images of a 17-year-old female patient. (a) T2-weighted magnetic resonance imaging (MRI) image showing Chiari formation, syringomyelia, C2-3-4 and C5-6-7 – D1 fusions. (b) Sagittal cut of computed tomography (CT) scan in flexion showing vertical atlantoaxial dislocation and C2-3-4 and C5-6-7 – D1 fusions. (c) Sagittal cut of CT scan in extension showing reduction of the vertical atlantoaxial dislocation. (d) Sagittal cut of CT scan passing through the facets showing facets of atlas and axis in alignment (Type C facet instability). (e) Three-dimensional reconstructed CT image showing the C2-3, 4, and C5-6-7 – D1 fusions. (f) Postoperative CT image. (g) Postoperative CT image showing the implants for atlantoaxial fixation. (h) Postoperative T2-weighted MRI showing reduction in the syringomyelia and resolution of the Chiari formation

Figure 2.

Images of a 36-year-old male patient. (a) T2-weighted magnetic resonance imaging image showing Group B basilar invagination, platybasia, Chiari formation, and C2–3 and C7 D1 fusions. (b) Sagittal cut of computed tomography (CT) scan in flexion showing Group B basilar invagination, platybasia, assimilation of atlas and C2–3 and C7 D1 fusions. The atlantodental interval is normal. (c) Sagittal cut of CT scan passing through the facets showing that the facets of C1 and C2 are in alignment (Type C facetal instability). (d) Postoperative CT image. (e) Postoperative CT image showing the implants. (f) Reconstructed CT image showing Goel’s C1–C2 fixation on one side and transarticular fixation on the other side

Figure 3.

Images of a 22-year-old female patient. (a) T2-weighted magnetic resonance imaging image showing atlantoaxial dislocation, C2–3, C5–6, and D1-2-3 fusions. Cord compression is seen opposite odontoid process. Subaxial external syringomyelia is seen. (b) Sagittal cut of computed tomography (CT) scan in flexion showing atlantoaxial dislocation, os odontoideum, C2–3, C5–6, and D1-2-3 fusions. (c) Sagittal cut of the CT scan in extension showing reduction of the atlantoaxial dislocation. (d) Sagittal cut of CT scan passing through the facets showing Type A facet instability. (e) Three-dimensional reconstructed CT angiography image showing the abnormal course of the vertebral artery on the left side. The artery travels posterior to the atlantoaxial articulation. The bifid arch of atlas is visualized. (f) Postoperative CT image showing alignment of the craniovertebral junction. (g) Postoperative reconstructed CT image showing the implants

DISCUSSION

Our 40-year experience in dealing with craniovertebral junction instability-related issues suggests that atlantoaxial instability is the most underrecognized and under-treated clinical entity.[7] While “acute” atlantoaxial instability has specific and more commonly recognized clinical signs and related symptoms and parallel diagnostic parameters, chronic or long-standing atlantoaxial instability manifests itself in a number of formats that are currently recognized to be “pathological” clinical entities. Our articles identify a number of musculoskeletal alterations like basilar invagination, short neck, torticollis, platybasia, Klippel-Feil alteration, bifid arches of atlas, os-odontoideum, assimilation of atlas, and C2–3 fusion and neural alterations like Chiari formation and syringomyelia to be consequences of chronic atlantoaxial instability.[7] It was observed that all these secondary manifestations of chronic atlantoaxial instability are naturally protective or adaptive in their function and are reversible following stabilization of the unstable spinal segment. We discussed that all these secondary alterations can be present discretely or in a cohort and still suggest the presence of atlantoaxial instability.

KFS is a relatively rare clinical syndrome. The clinical analysis of the subject is essentially based on isolated case reports and relatively few case series.[16] Although exact epidemiological data are not available, the general observation is that KFA happens in about 1 in 40,000–42,000 newborns worldwide, and there is a female preponderance by approximately 3:2.[17,18,19] Although the exact incidence and its correlation with Western data is not possible, it does appear that the incidence of KFA and of basilar invagination is higher in India and other South Asian countries. Our earlier articles speculate that apart from injury to the neck in infants by improper neck handling, poor child delivery practices that prefer home deliveries by half or nontrained health personal and involves use of forceps in protein calorie malnourished mother and child initiates injury to the neck that leads to potential or manifest atlantoaxial instability that forms a nodal point of pathogenesis of a host of clinical syndromes that include basilar invagination, Chiari malformation and KFA.[9,10,20,21,22,23] Perinatal history was not appropriately taken or scientifically analyzed in our patient cohort, and such a correlation can be considered only at the level of speculation. In this context, our observation is that KFA is not an embryological dysgenesis or error in segmentation or somatogenesis, but it is due to mechanical injury at some stage of development and growth, most probably in the late fetal or early neonatal life, that results in atlantoaxial instability that initiates and propels spinal and cranial alterations. Genetic alterations or mutations have been implicated as a cause of KFA. Such studies were not appropriately or scientifically done in any case in our case series.

Clinical analysis of the subject is based on our earlier published concept that KFA is one of the several secondary alterations in the presence of chronic or longstanding atlantoaxial instability. All our patients with KFA had one or several other musculoskeletal or neural alterations that characterize chronic atlantoaxial instability.[7]

A significant subgroup of patients with KFA is diagnosed based on physical deformities and has marginal or no significant neurological symptoms. Our impression based on a large series of cases is that the clinical symptoms and neurological deficits in cases with KFA are relentlessly progressive, are initially marginal or troubling, then disabling, then crippling, and ultimately have the potential to lead to slow and miserable death.

Chronic pain, muscle spasm, and restricted neck movements seem to be the nodal point of pathogenesis of shortening of the neck, torticollis, and secondary bone fusions and related symptoms that formed the complex of KFA. There was a short neck and a low hairline in all cases. Although morphometric parameters to measure neck length and to identify the baseline normal hairline are available, such an assessment was not consistently done in our cases.[24] There was mild-to-severe torticollis in 29 patients. Dorsal kyphoscoliosis as a major presenting symptom was observed in three patients.

Major clinical symptoms included tingling paresthesiae in hands (74 patients), weakness of the limbs (82 patients), breathing disturbances (29 patients), and several such symptoms [Table 2]. Seventeen patients were unable to stand without support, and 9 were wheelchair bound. As can be observed from Table 2, a significant majority of our patients presented in a neurologically devastating clinical condition. Analysis of articles published in the literature and review of clinical material discussed in the presented article, it appears that the patient cohort treated by us had “significantly” more profound clinical and neurological symptoms and related deficits. Furthermore, the duration of clinical symptoms was longer when compared to similar studies. This appeared to be related primarily to a relatively poor and illiterate patient population treated by the authors in their charitable hospital/s that provides free medical treatment. Ignoring early symptoms and difficulty in easy and quick accessibility of specialized medical care were obvious reasons.

Fifty-eight patients had Group A basilar invagination. Twelve patients had Group B basilar invagination.[19] Eighty-two patients had atlantoaxial instability on dynamic flexion-extension imaging. The atlantoaxial instability was completely reducible in 21 patients and incompletely reducible in 61 patients on complete head extension. In 10 patients, there was mobile vertical atlantoaxial instability, meaning that on flexion of the head, the odontoid process migrated rostrally in the foramen magnum and on extension of the head, the odontoid process returned to its normal position in relationship with the arch of atlas.[25] The atlantodental interval is not abnormally increased in such “vertical” atlantoaxial instability.

We recently classified atlantoaxial instability based on facet alignment on lateral profile imaging with the head in neutral position.[26,27] Type A facet atlantoaxial instability was when the facet of the atlas was dislocated anterior to the facet of axis, or there was listhesis of the facet of atlas over the facet of axis. Such instability is usually associated with Group A basilar invagination; there is an abnormal increase in the atlantodental interval, and there is evidence of dural or neural compression by the tip of the odontoid process. In the presented series, 82 patients had Type A facet atlantoaxial instability. Type B facet atlantoaxial instability was when the facet of the atlas was dislocated posterior to the facet of the axis, or there was retrolisthesis of the facets. Type C atlantoaxial facet instability was when the facets of atlas and axis were in alignment, and atlantoaxial instability was diagnosed only based on a high degree of clinical suspicion, telltale clinical and radiological evidence, and was finally confirmed by direct intraoperative manipulation of bones during surgery. In both Type B and C facet atlantoaxial instability, the atlantodental interval might not be abnormal, and there may not be any dural or neural compression by the tip of the odontoid process. Instability in Type B and Type C atlantoaxial facetal instability is referred to as central or axial atlantoaxial dislocation/ instability (CAAD).[26,27,28] Twelve patients had Type B, and 5 had Type C atlantoaxial instability CAAD is associated with chronic or longstanding atlantoaxial instability that is more often seen in KFA. Essentially, our observation is that KFA is one of the telltale pieces of evidence of chronic atlantoaxial dislocation. More recently, we have adopted the use of dynamic head rotation and head tilt CT imaging to further sub-classify CAAD.[29,30] However, such an investigation could not be comprehensively evaluated for the presented case series.

Imaging studies showed the associated spinal and neural alterations [Table 3]. Other morphological alterations included bifid arches of the atlas (17 cases), os-odontoideum (3 cases), bifid arch of axis (3 cases), platybasia (6 cases), and omohyoid bone (3 cases). Neural alterations included Chiari 1 formation in 30 cases and syringomyelia in 17 cases. External syringomyelia, an entity described earlier, was identified in 41 cases.[31,32] There was physical and radiological evidence of spinal cord compression at one or more levels of fused and adjacent spinal segments in nine patients.

The neck deformities included torticollis. In three cases, there was a hump in the lower half of the neck related to the omohyoid bone. In our earlier article, we likened the omohyoid bone to a cycle stand, and it is a natural protection meant to support an unstable neck.[33] Significant dorsolumbar kyphosis that formed a part of the presenting complaint was seen in 3 patients. We identified the presence of short head, short neck, short spine, and short body height as secondary and protective adaptations or formations in the presence of chronic atlantoaxial instability.[34] Although renal anomalies, congenital heart disease, and cleft palate have been identified in association with KFA, in none of our patients were such anomalies identified.

As chronic atlantoaxial instability was identified to be the primary initiator and propeller of all the secondary alterations, patients underwent only atlantoaxial stabilization, and the surgery was aimed to achieve solid arthrodesis of this joint. The atlantoaxial joint was not “fused” in any case and was identified during surgery to be pathological and hypermobile. Despite the evidence of neural compression and deformation at several levels other than the atlantoaxial joint, only atlantoaxial fixation was done. Although occipitoaxial fixation is technically relatively easier in such a clinical situation, we resorted to and accomplished atlantoaxial fixation in all cases.

Remarkable and gratifying clinical recovery was observed in all patients following surgery. The recovery in clinical symptoms began in the immediate postoperative period as soon as the patient was reversed from anesthesia. All patients continued to improve progressively during the follow-up. During the period of follow-up, all patients who were unable to sit or stand without support could walk unaided. Voice quality improved in all patients, and the improvement was observed in the immediate postoperative period. All patients recovered completely in their sleep apnea. This critical clinical symptom never recurred even once following surgery in the 29 patients who had this symptom prior to surgery.

Whilst short neck and other musculoskeletal alterations have been incriminated to embryonic dysgenesis, our speculation is that chronic atlantoaxial instability-related pain in the nape of the neck and muscle spasm, which probably aim to reduce harmful and dangerous neck movements, result in shortening of the neck size and associated morphological alterations in the musculoskeletal system. The very fact that neck size visibly increased following atlantoaxial stabilization is testimony to this speculation.

CONCLUSIONS

Understanding the fact that chronic atlantoaxial instability forms the nodal point of origin of KFS and other associated musculoskeletal and neural alterations can have therapeutic implications. Atlantoaxial fixation results in gratifying clinical recovery.

Author contribution statement

Atul Goel: Conceptualization, Validation, Formal analysis, Writing – Original draft, Writing – Review and Editing, Supervision, Project Administration. Nasser El-Ghandour: Methodology, Validation, Writing – Review and Editing. Abhidha Shah: Conceptualization, Methodology, Validation, Formal analysis, Writing – Original draft, Writing – Review and Editing, Supervision. Apurva Prasad: Methodology, Validation, Writing – Review and Editing. Jash Doshi: Methodology, Validation, Writing – Review and Editing. Siddharth Gautam: Methodology, Validation, Writing – Review and Editing. Ravikiran Vutha: Methodology, Validation, Writing – Review and Editing. Tejas Vaja: Methodology, Validation, Writing – Review and Editing.

Highlights

• Chronic atlantoaxial instability is the nodal point of the pathogenesis of KFS

• Several musculoskeletal and neural alterations associated with KFS are secondary manifestations of atlantoaxial instability

• Atlantoaxial stabilization is the treatment.

Declaration of patient consent

The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient has given his consent for his images and other clinical information to be reported in the journal. The patient understands that his name and initials will not be published and due efforts will be made to conceal his identity, but anonymity cannot be guaranteed.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.