DEMETRIOS STYL. KORRES
Hon Prof in Orthopaedics
National and Kapodistrian University of Athens
When in 1975, nearly 40 years ago, I started studying the injuries of the Cervical Spine, I was impressed by two events which attracted my attention; both, were related to the odontoid process, this peculiar anatomical element which characterize the Axis vertebra. The first was the multiple pattern of fractures which this element was subjected to. The second one, their evolution and their final outcome
The incidence of these fractures is around 14% (1) among the cervical spine fractures and may occur at any age, with a higher risk for patients older than 65years old but, to note that in this age group the fracture of the odontoid is more likely to be missed at the initial examination.
The cause of injury is mainly a road traffic accident or a fall, but, other causes are, also, implicated to these injuries. However, the exact prevalence is not known as a certain number of patients who sustained such an injury did not survive and were transported dead to the hospital. The fracture of the odontoid is due often to high-energy trauma with the implication of a combination of forces and the major loading path that cause the lesion is not well established. According to experimental data the causes of the fractures of the odontoid process are a combination of vertical compression and horizontal shear which acting in a different angle create different patterns of fracture . In practice we can assume that the main forces responsible for a fracture of the odontoid process are the hyperextension, the flexion and the lateral bending, but is certain that combined forces could also be responsible for a fractured odontoid.
The clinical picture of this injury varies from a mild to a severe one (quadriplegia or even death) (1). The patient is coming in the emergencies holding his head or is transferred on stretcher complaining of pain in the cervical region; is keeping his head unable to move it because of pain. A thorough clinical examination to exclude symptoms corresponding to a neurological damage is needed although there are usually missing. The severity of the neurological disorder corresponds to the degree of displacement and the consequent instability of the odontoid process. Older people have a higher rate of mortality (1).
It is also important to look for an involvement of the vertebral arteries which if damaged could induce symptoms not only at an early stage, but, also, days after the accident.
The fracture of the odontoid process represents a separate entity and special attention is needed in order to recognize the fracture, but also, to apply the appropriated treatment. Now a days, it is clear that the pattern of the fracture contributes in the favorable or poor evolution, that is, in the appearance of complications, and particularly of pseudarthrosis ; at the same time, it may indicates the way these lesions could be managed. So, there is a necessity for a complete and precise clinical and radiological approach, which will permit us to put the correct diagnosis. Problems could be arise in children (the presence of congenital malformations, the immature skeleton etc) and in old people ( Degenerative disease, pathological condition like tumor etc).
Radiological investigation is of importance. This must include apart the conventional x-ray views (Anterolateral (open mouth view), lateral and in certain instances dynamic views), also, CT-Scan (with reconstruction imaging), M.R.I and in certain instances too, 3-D imaging. Of course this will be done in a ‘step by step’ manner. If the patient is unconscious, then, the entire spine has to be investigated. The raison for such a meticulous radiologic investigation is due to the fact that the fracture pattern is unforeseen, as multiple forces can be applied to the upper cervical spine simultaneously or in continuity.
Two basic classification categories according to : a. The position of the fracture and b. to the direction of the line fracture are recognized.
In the first category there are four different classifications described :
1.The Schatzker’s classification (2) in which two types are recognized, (a) the low fracture at the level or below the level of the attachment of the accessory ligaments and (b) the high fracture above the attachment of the accessory ligaments.
2.The Mourgues’ Classification (3) in which there are two types of fractures. The first is reported to fractures of the base of the odontoid process with a 100% union prognosis, and the second is reported to fractures at the neck of the odontoid process with a union prognosis of less than 50 %.
3.The Anderson-D’Alonzo’s classification (4) who proposed the division of the fractures into three types. In the first instance, the authors are reporting to fractures at the top of the odontoid process with a favorable prognosis for union. In the second instance they are referring to fractures of the neck with a mean rate of pseudarthrosis of about 36 %. The last type the authors are referring is the fractures at the base of the dens. This fracture has a 90 % mean rate of union.
4.The Althoof’s classification (5) who proposed 4 types of fractures. The first type with the lesion situated above the neck and having a rate of pseudarthrosis around 64 %. The second type concerns fractures through the neck with a rate of pseudarthrosis of about 55 %. The third type is the one where one of the lateral masses is involved; this lesion has a mean rate of union around 50% and the fourth type in which the fracture is situated at the body of the axis vertebra with a high rate of union as 93 %.
The second category, in which the direction of the line is considered, includes the classification of Roy-Camille (6) with three types of fractures :
(a) Anterior oblique fractures with posterior-superior towards anterior-inferior direction of the fracture line originating at the level of the neck and directed through the body to the anterior surface of the axis vertebra. This fracture is the result of an impact of the strong transverse ligament on the odontoid process at the time of the applied flexion momentum, and represent the commonest type.
(b) Posterior oblique fractures with anterior-superior towards posterior-inferior direction of the fracture’s line. These fractures are caused by an applied extension momentum, when the anterior rim of the foramen magnum strikes the apex of the odontoid process, and have the tendency to displace posterior.
(c) horizontal fractures with the line of fracture through the neck only; these fractures are the result of the application of a lateral force and are the most unstable. In some cases the fracture’s line extends to the axis body and becomes trans-articular representing the “English Polishmen’s helmet”.
The evolution in the diagnostic methods revealed the existence of other types of fractures like the vertical one, and the complex fractures . The vertical fracture is characterized by the division of the odontoid process into two parts with the line fracture extending from the apex to the base, while the complex fractures are characterized by a diversity of fractures lines creating the compound or double level fractures. (Fig 2E)
The fractures of the odontoid process represent the mechanical failure of this particular anatomic element following the application of forces (1). The direction of these forces, the internal architecture, the mechanical strength of the bone trabeculae, the proportion of the cortical and cancellous bone, the magnitude of the odontoid process displacement, the vascular supply of the odontoid process and the age of the patients are the most important factors in the creation of specific fracture types and the prognosis of these injuries. In line, but not, well documented, in the literature, radiographic, and histomorphometric studies outlined the structural difference between the odontoid process and the body of the axis. Data from these studies could distinguish the fractures at the base of the odontoid process and the underlying body of the axis. This was also shown in a recent study using peripheral quantitative computed tomography (pQCT) in cadaveric specimens of the axis (7). Moreover, this study showed the difference of the internal architecture of the axis between young and older patients; in subjects more than 40-year-old a large void of thin trabecular bone has been identified extending from anterior-inferior to superior- posterior to the base of the odontoid process indicating a mechanically weak region that may predispose to specific fracture patterns (7).
The Anderson - D’Alonzo classification offers a simple and topographic approach to odontoid process fractures. However, it does not contribute to the thorough understanding of the mechanism of the fracture, nor it incorporates any bio- mechanical characteristics or specific characteristics of the internal architecture of the odontoid process and has limitations .
In addition, the many attempts made for its improvement or replacement , the existence of alternative classifications based on the direction of the fracture’s line, the heterogenity of the reported pseudarthrosis rates at type II fractures and the presence of various unclassified fracture types such as some vertical or oblique fractures may suggest the inadequacy of the aforementioned classification schemes.
The classification proposed by Anderson and D’Alonzo is misleading and contributes to confusion regarding fracture location . The Anderson and D’Alonzo type III fracture is not a fracture of the odontoid process, but rather a horizontal rostral fracture through the upper aspect of the body of the axis (8). Further adding of subtypes, such as type IIA(9), type IIB (10), type IIC, type II 1-5 (11) and type IIIA (12) perpetuate the confusion. At the same time, Koller stressed the point of a lack of comprehensive classification for fractures of the body and the odontoid process (13)
In the literature there are fractures not corresponding to the already existing classifications and there is a confusion to be present, so, it is clear that a more appropriate one would be obligatory (14).
The proposed herein classification is based on the structural, anatomical and bio- mechanical properties of the odontoid process (7).
The anatomical classification we propose, accepts four types of fractures and recognizes a zone where practically no fractures are noted (the neutral zone), which is found at the level of transverse ligament, an area of phylogenically strong bone.
Type A fractures are rare. It is an avulsion fracture at the points of insertion of the alars or apical ligaments with an incidence of 2,3 % in our series ; It’s stability is questionable, but they responded favorable to a conservative treatment.
Type B fractures represent the most common fractures of the dens fracture in the literature, and particularly in the elderly population although in our series were second in incidence with 44,1 % . They are the result of lateral force which initiates a rotational movement. In the presence of osteoarthritic changes this leads easier to a type B or C fracture seeing most frequently in older persons. These fractures represent an unstable lesions with a tendency to pseudarthrosis, so, they need a careful evaluation and appropriate treatment, conservative or surgical, particularly in the old patients .
Type C fractures were found to represent 46,6 % ; they have a more favorable outcome than type B , responding to a less aggressive management, finally,
Type D fractures are not so uncommon (7 % in our series). The axial loading, while the neck is in extension, as well as, a combination of applied forces, which act either simultaneously or not, are probably responsible for this injury. Although it is an unstable lesion, it seems to respond better to a non-surgical treatment.
To note that is easy to recognize a type B from a type C fracture in the lateral view. In type B fracture the Harris ring is intact as the fracture’s line is above this ring. In type C the fracture line is projected at the upper part of the Harris ring.
This classification is considered more realistic since it is simple, it includes the whole spectrum of fractures, it refers only to one anatomical structure ( the odontoid process), it correlates to the bio-mechanical characteristics of the axis, it indicates the prognosis of the different fracture types, it suggests the adequate evaluation and management of the fractures. Fractures not involving the odontoid process, like the type III in the Anderson - D’ Alonzo classification, could not be called odontoid fractures as they are, anatomically speaking, fractures of the axis’ body.
Fractures of the odontoid process should be divided into (a) stable and (b) unstable(6) This is necessary in order to proceed with the correct therapeutic plan. Stable injuries are
managed conservatively, while unstable injuries are treated surgically. Fracture’s instability depends mostly on the presence of associated lesions, the type of fracture and the initial displacement, particularly if there is vertical displacement
Type A fractures are treated conservatively by using external mobilization, like a rigid cervical collar or a halo vest ; rarely an operation should be required.
In type B fractures, although there are unstable, a controversy is still on-going among different authors regarding the surgical approach and management of this lesion. To us an initially conservative treatment should be attempted unless a score of more than 10 points is found.
In type C, either conservative or operative management is applied in relation to the instability these injuries shown.
In type D, the appropriated conservative management is offering excellent results.
Several treatment modalities are proposed in the presence of this injury, but, before take any decision, it is wise, particularly in the elderly patients, to look for any clinical co- morbidities that may affect the management, and to rule out any concomitant or double level fracture of the cervical spine.
Conservative treatment is suggested by the use of traction with the application of a Crutchfield skull tongs, in order to reduce and stabilize the fracture or with the use of external immobilization as custom-made orthosis, halo vest or cervical collar.
In case of application of traction, the weight applied should not exceed 2-3 Kg, in order to avoid distract the fractured fragments. Special care must be given, not only to the application of traction, but, also, in it’s direction. Attention must be paid when traction in flexion is applied, because of the potential danger of a vertebral artery lesion or of a neurological injury which may induce respiratory compromise or other neurological symptoms .
Stabilization must be kept for at least 4 to 6 weeks during which radiographic control of the position of the fracture is necessary, as well as careful examination for avoidance of complications related to the traction and prolonged bed rest. After this period the patient may be mobilized using a four-point support brace for a period of four to six more weeks. At the end of 10 to 12 weeks, dynamic x-ray views in flexion and extension are taken in order to detect any sign of instability. If instability is proven, then, surgical treatment must be considered. If no instability is detected and the fusion is complete, the patient must use a soft collar for a short period of time.
The use of a Halo vest (fig 3) is not always recommended as the rate of complication reported is as high as 26 % with older persons suffering more from discomfort. Even more, the traction obtained at the beginning, is slowly turned into compression, in the mobilized patient, resulting in mal-union, if the reduction of the fracture should be lost due to sliding.
The fusion rate in the conservatively treated patients is reported to be from 35 % as high as 85 %, but this is related to the age and other parameters, like the time the treatment was applied, the type of the fracture and the initial displacement; the later is being correlated to the direction (anterior or posterior) the fracture is displaced off.
Also, often, the traction applied leads to distraction at the fracture site; this leads to development of late instability and pseudarthrosis.
Vieweg and Schultheib in contrast to Wolter and Reimann(1), advocate the use of Halo vest in type II fractures, as the percentage of healing is as high as 85%, particularly in the no-displaced ones ; they conclude, also, that in type III the Halo vest is the treatment of choice as the rate of healing is about 97%.
The fractures of the odontoid process are prone to complications either at the trauma scene or later in the hospital.
Two of the complications appearing in a late period are of importance and have to be well clarified : pseudarthrosis and mal-union.
Of importance is the pseudarthrosis these fractures may develop and the potential danger for late myelopathy or for direct injury of the spinal cord.
According to the literature the percentage of pseudarthrosis related to the applied treatment, is between 4% to 100 % for all types of fractures and consequently complicating their treatment. The predisposing factors which have been accused for pseudarthrosis include age, mechanism of injury, displacement, the blood supply of the odontoid process, the direction of fracture’s line and type of fracture. Also, the application of excessive traction, the stability of immobilization, the timing of immobilization, as well as the co-existence of an other fracture, either in the atlas or even in the axis itself. The internal architecture of the axis seems to play an important role in the development of pseudarthrosis. Finally, the possibility of interference of the transverse ligament between the fragments may, also, cause difficulties in the reduction of the fracture, resulting in the development of pseudarthrosis. Some authors may disagree as to the importance of some of the above factors. To our experience all the above mentioned factors play a certain role and contribute to the development of pseudarthrosis resulting in instability at the fracture site.
Age has an important role in pseudarthrosis since it has been proved that nonunion or pseudarthrosis is found at a higher incidence in patients over the age of 50 years. Fracture displacement according to Blokey and Purser (1) should not be correlated to pseudarthrosis. However, other authors did not accept this opinion. Appuzo et al (1) described that a shown displacement of more than 4 mm is of significance. However, it is extremely difficult to be aware of the original displacement that was present at the time of the accident.
The traction applied, as well as its direction, play a significant role according to Ryan and Taylor(1). This is due to the possibility of creating a greater wedging at the fracture level than the one desired, and to lead to a deformed odontoid process.
The type of fracture is also implicated in the development of pseudarthrosis since in fractures involving the neck of the dens as well as the direction of the fracture’s line this was in a rate of 32 % and as high as 90 % . In our experience the types A, C and D are prone to unite, while type B fracture have a high rate of pseudarthrosis, particularly the one that shown a posterior displacement of more than 5mm or an angulation of more than 10 degrees.
There is no doubt that the presence of co-existing injuries, as well as a delay in the diagnosis or if the immobilization is not the proper one, then this will increase the instability of the fracture.
The knowledge of these parameters as factors influencing the development of a pseudarthrosis, obliged us to proceed in their grading, according to the role each one plays. In this manner, the evaluation of a fracture is more easier, as we will predict the risk for pseudarthrosis. It has been estimated that if the sum of the graded factors is greater than 10 points, then the fracture must be characterized as being at risk of pseudarthrosis and surgical treatment must be considered (1,15).
The blood supply of the odontoid process is not considered to participate in the development of pseudarthrosis. Although a recent experimental study of ours revealed decreased blood supply at the area where type B fractures happen, which is the area where pseudarthrosis often develop (16).
Pseudarthrosis in the elderly is not always a major problem as it was proven ; this is not accompanied always by a clinically significant instability due to the development of fibrous tissue at the fracture site. However operation has to be considered if instability persists or if signs of myelopathy are present.
Operative treatment of dens non-unions is not without risk, taking into consideration the age of the patients the high co-morbidity those patients have and, also, knowing that the success rate is low.
The second most frequent and severe complication of the fractures of the odontoid process concerns their mal-union(1). Special care so must be given to this complication as may result in spinal canal stenosis which may induce, in long term, cervical myelopathy.
The greater the displacement and wedging of the fracture, the less the width of the spinal canal. This results in chronic compression or friction of the dura matter and the spinal cord on the upper posterior corner of the body of the axis and hence cervical myelopathy. The treatment of this complication is difficult. It requires anterior or posterior decompression combined with posterior fusion extending from the occipital bone to C1,C2 or C3 and even lower.
A number of minor complications may accompany a fractured odontoid. Stiffness, decreased range of movements, discomfort or even mild pain are easy managed with physiotherapy or other conservative methods. Younger patients response better and an uneventful outcome is usually achieved.
Failure to treat conservatively a fractured odontoid is an indication for surgical intervention.
Instability must be treated operatively as soon as possible. The operative approach is suggested by many authors. There are several methods that allow the safe management of the unstable fractures. This is done by closed or open reduction, the use of osseous graft, and stabilization by means of a wire or nylon or use of implant(s). Auto-graft is the most suitable material to be used for achievement of a stable fusion. The approach used is either anterior, lateral, posterior or combined.
Posterior stabilization includes :
Posterior C1-C2 wiring technique with Gallie’s or Brooks’ techniques or other methods (fig4).
Anterior stabilization includes:
a. Application of a plate between the anterior arch of the atlas and the body of the axis or the vertebral body of C3
b. Internal fixation with screw(s). This technique is gaining popularity, but indications have to be set very carefully as osteoporosis, fracture of the anterior wall of the body of C2, posterior displacement, comminuted or type D fractures or even a
narrow diameter of the spinal canal are among the contraindications. The osteosynthesis of the odontoid process permits a nearly normal function of the C1-C2 level (5).
Lateral stabilization includes :
Fusion using the Dutoit technique. The lateral approach is used for a C1- C2arthrodesis in case of traumatic instability, tumors or infections and for cases in whic in which an other, previously used approach, or a counter-indication is present.
The surgical results seems to be better than these achieved with a non-operative management as the reported fusion rate is high between 80-100% . However, there is not unanimous acceptance of an appropriate treatment of patients with these fractures, particularly in the elderly population. (Table I summarizes the indications for the appropriated management of an odontoid fracture). Physiotherapy in both instances, has an important role to play in the final outcome those patients should have.
We can easily appreciate the importance of dens fractures ; they need correct diagnosis and appropriated treatment trying to minimize the side effects to the patient and to the society. In the later, the financial cost is very important as in the last decade, the cost for the treatment of these fractures is very high with the increased hospitalization and the increased number of patients treated surgically (17)
B Surgical or conservative
C1 Conservative or surgical
Fractures of the odontoid process represent a particular entity.; they need a careful handling us their behavior is not easy predictable. Diagnosis of the correct type of fracture is very important. Complications due to multiple factors, are not rare and have to be carefully managed and treated. Pseudarthrosis is the most common complication related to many factors the most important being the instability at the fracture site.
The proposed new classification has been proven successful lasting in time and is now considered more realistic as it is simple, is including the whole spectrum of fractures, it refers to one anatomical structure (the odontoid process), it correlates to the bio-mechanical characteristics of the axis, it indicates the prognosis and it suggests the management of the fracture.
The study of our cases revealed some directions for the development and treatment of these fractures. The evaluation of our cases showed that type A and D fractures can and should be treated conservatively ; they usually have a good prognosis. Type B fractures are prone to further complications and hence should be treated - in their majority - surgically. Type C, need to be followed closely and in case of instability and/or late displacement have to be operated
1. Korres D.S. The Axis Vertebra. Ed by Springer-Verlag Italy. 2013
2. Schatzker J et al. (1971) Fractures of thedens (odontoid process): an analysis of thirty-seven cases. J Bone Jt Surg – Br. 53 : 392-405
3. De Mourgues Fischer LP et al. (1981) Fractures of the odontoid process (in French). Rev Chir Orthop Reparatrice App Mot 67 : 783-790
4. Anderson LD, D;Alonzo (1974) Fractures of the Odontoid process of the axis. J Bone Jt Surg –Am 56: 1663-1674
5. Althoff B, Bardholm H (1979) Fractures of the odontoid process. An xperimental and clinical study. Acta Orthop Scan Suppl 177 1-91
6. Roy-Camille R, Saillant G, et al. (1983) Fractures recentes de l’odontoide. Presse Med. 12, 2233-2236
7. Korres DS, Karachalios Th et al. (2004) The structural properties of the Axis studied in cadaveric speciments Clin Orthop Rel Research. 418 ; 134-140
8. Benzel EC, Hart BL et al (1994) Fractures of the C2 vertebral body. J Neurosurgery 81 : 206-212
9. Hadley MN, Brown CM et al. (1988) New subtype of acute odontoid fractures (type IIA) Neurosurgery 22 : 67-71
10. Gauer JN, Shaft B et al (2005) Proposal of a modified, treatment-oriented classification of odontoid fractures Spine L 5 : 123-129
11. Reihold M, Bellabarba C et al (2011) Radiographic analysis of type II odontoid fractures in a geriatric patient population :description and pathomechanism of the “Geier”–deformity. Eur Spine J 20:1928-1939
12. Jea A, Tatsui CI et al (2006) Vertically unstable type III odontoid fractures:case report. Neurosurgery 58 : E 797
13. Koller H, Kammermeier V et al.(2006) Spinal stenosis C1-C2 following redo surgery for failed odontoid screw fixation-scrutinizing the odontoid fracture classification. Internet J Spine Surgery 3;1-11
14. Korres DS, Mavrogenis FA, et al. (2008) It is time to reconsider the classification of dens fractures. Eur J Orthop Surg Traumatol 18 : 189-195
15. Korres DS, Stamos KG,Andreakos AG et al (1989) Fractures of the dens and risk of pseudarthrosis. Arch Orthop Trauma Surg 108 :373-376
16. Γκανταΐφης Νικολαος (2014) Η μελέτη του περιοστέου και της περιοστικής αιμάτωσης της οδοντοειδούς απόφυσης του δευτέρου αυχενικού σπονδύλου ως πιθανού παράγοντα ψευδάρθρωσης διαφόρων τύπων καταγματων του οδόντος.
Διδακτορική Διατριβή. Αθήνα 2014
17. Daniels, Alan H.; Arthur, Melanie; Esmende, Sean M.et al. (2014) Incidence and Cost of Treating Axis Fractures in the United States From 2000 to 2010 Spine: 15 August 2014 - Volume 39 - Issue 18 - p 1498–1505