- Email: [email protected]
Elderly patients are at increased risk for mortality undergoing surgical repair of dens fractures
Clinical Neurology and Neurosurgery 115 (2013) 2056–2061
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
Elderly patients are at increased risk for mortality undergoing surgical repair of dens fractures Ardeshir Ardeshiri a,∗ , Siamak Asgari b , Elias Lemonas a , Neriman Oezkan a , Marc Schlamann c , Ulrich Sure a , I. Erol Sandalcioglu a a
Department of Neurosurgery, Universitätsklinikum Essen, Essen, Germany Department of Neurosurgery, Klinikum Ingolstadt, Ingolstadt, Germany c Department of Neuroradiology, Universitätsklinikum Essen, Essen, Germany b
a r t i c l e
i n f o
Article history: Received 24 April 2013 Received in revised form 2 June 2013 Accepted 6 July 2013 Available online 31 July 2013 Keywords: Cervical fracture Dens fractures Elderly patients Odontoid fracture Ventral screw fixation
a b s t r a c t Objective: Dens fractures are common cervical injuries in advanced aged patients. The presented study was undertaken to analyze the clinical results and risks of surgically treated patients with dens fractures over 70 years. Methods: Data of 28 patients (17 female, 11 male) over 70 years treated from September 2004 to October 2009 were recorded. Clinical and radiological parameters were obtained including type of fracture, associated cervical and/or other injuries, comorbidities, symptoms, neurological condition, surgical strategy, postoperative course and complications. Results: 89% were in a good neurological condition before surgery (ASIA E or D). In most cases, surgery was performed at an early stage after trauma (21 patients within 5 days). Ventral screw fixation was the preferred surgical strategy (64%). A slight worsening of neurological functions immediately after operation was only seen in one patient. Five patients died in the early and 2 in the late postoperative course which means a treatment mortality of 25%. Among the surviving patients two had general medical complications. Conclusion: Type II dens fractures are a common fracture of elderly patients. Our results are good concerning the neurological functions. Surgical and general medical complications were acceptable. However, the study also underlines that mortality rate is high and therefore treatment options should be wellconsidered in this high risk group. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Dens fractures count to nearly 20% of all cervical fractures. Among these 65–74% are type II-fractures according to Anderson and D’Alonzo [1]. They are the most common cervical fractures of patients over 70 years. In patients over 80 years they are more frequent than all cervical fractures together [2]. Typical trauma mechanism in the elderly patients is falls or pitching of the head leading to extension injuries and dorsal dislocation [2]. Instability might injure directly or decelerated the spinal cord [3]. Nonunion rate of type II-dens fractures is significant [4]. There are some factors which are known to be associated with an increased risk of nonunion such as advanced age, delayed
∗ Corresponding author at: Department of Neurosurgery, Universitätsklinikum Essen, Hufelandstraße 55, 45122 Essen, Germany. Tel.: +49 201 7231201; fax: +49 201 7235909. E-mail address: [email protected] (A. Ardeshiri). 0303-8467/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.clineuro.2013.07.006
treatment, anterior dislocation more than 4 mm and posterior dislocation [5] as well as reduced halo vest tolerance in nonoperatively treated elderly patients [5]. Furthermore, a mortality rate of 26–47% has been reported for these nonoperatively treated patients due to respiratory-related complications caused by prolonged periods of immobilization [6]. Based on the existing comorbidity in elderly patients, the morbidity and mortality rates are high for both, operative and nonoperative treated patients [7]. Surgical techniques include posterior atlantoaxial fusion with different devices (transarticular screws, laminar clamps, sublaminar wiring) and anterior screw fixation [6]. Posterior fixation has a high rate of union (80–100%) [8], but results in a decrease of the rotational range of the cervical spine [9]. Anterior screw fixation has a high fusion rate (84–96%) and additionally preserves axial rotation [6]. Further advantages are shorter surgery time and no graft harvesting is needed. However, this technique can be limited by anatomic contraindications including a disrupted transverse ligament, in cases of fractures with cranial-posterior to caudalanterior direction and in patients with emphysema of the thorax [10]. Problems of surgical fixation in advanced aged patients are
A. Ardeshiri et al. / Clinical Neurology and Neurosurgery 115 (2013) 2056–2061 Table 1 Number of patients in the different ASIA groups (modified) pre- and postoperatively. Overall mean score was 4.57 and 4.55, respectively. Modified ASIA score A = complete: No motor or sensory function is preserved (1 point) B = incomplete: Sensory but no motor functions preserved below the level of injury (2 points) C = incomplete: Motor function is preserved below the level of injury and more than half of key muscles below the neurological level have muscle grade less than 3 (3 points) D = incomplete: Motor function is preserved below the level of injury, and at least half of key muscles below the neurological level have a muscle grade of 3 or more (4 Points) E = normal: Motor and sensory functions are normal (5 points) Overall mean score
Preoperative
Postoperative
2
2
0
0
1
1
2
3
23
22
4.57
4.55
osteopenia- and respiratory-related morbidity [6]. Thus, despite a normal neurological condition without deficits in many cases the treatment of these elderly patients bears an increased risk, surgically and medically. This study was conducted to evaluate morbidity and mortality of patients older than 70 years undergoing anterior or posterior surgery for dens fractures at our department. 2. Materials and methods 2.1. Patients For this analysis medical reports of patients with dens fractures treated surgically at our hospital from October 2004 to September 2009 were reviewed retrospectively. A total number of 28 patients over 70 years were included into this study. 17 (61%) were female and 11 (39%) male with a mean age of 81.1 years (range 71–94 years). In most cases trauma mechanism was fall. In 25 patients (89.3%) diagnosis was made within 2 days after trauma, among these in 19 patients at day 1 after trauma. In the remaining three cases diagnosis was made 1 month, 6 weeks and 4 months after trauma. A type II-fracture (Anderson and D’Alonzo) was seen in 27 patients (96%) and a mix fracture of type II/III was diagnosed in one patient (4%). Associated cervical injuries or fractures were present in 11 (39%) patients, mostly affecting C1. Additional injuries were nose fracture (n = 2), radius fracture (n = 1), L3-fracture (n = 1), TH7-fracture (n = 1) and rib-fracture (n = 1). 11 (39%) patients had cardiopulmonary comorbidities, in 5 (18%) patients osteoporosis was documented and in 8 (29%) patients no significant comorbidities were recorded. Concerning the anaesthesiological risk in 22 patients an ASA (American Society of Anesthesiologists) score 3, in 5 patients ASA score 2 and in one patient ASA score 4 was documented. ASA 1 and 2 were classified to low-risk and ASA 3–5 to high-risk patients. 2.2. Diagnostics and surgical procedure Clinical symptoms and the neurological condition were classified according to a modified ASIA-classification (American Spinal Injury Association). We assigned each grade points with 5 points for the best neurological function (ASIA E) and 1 point for the worst grade (ASIA A) (Table 1). Preoperative neuroimaging procedures included computed tomography (CT) with coronal and sagittal reconstructions in all patients in order to classify the type and extension of fracture and position of the facet joints. In 18 patients X-rays were performed and four patients additionally underwent
2057
preoperative magnetic resonance imaging (MRI) to evaluate ligamental lesions, epidural haemorrhages and spinal cord lesions. Preoperative extension therapy (Gardner-Wells) was performed in 12 cases (43%) in order to achieve a better reposition of the fracture. Anterior screw fixation [11] with a single screw was performed in 18 patients (Fig. 1). In the remaining 10 cases posterior stabilization techniques were performed. Posterior stabilization was used in cases when anatomic conditions made an anterior approach not possible or when the patient was admitted to our hospital 3 days or later after the trauma. Dependant on the complexity of the fracture, anatomical conditions and the bone quality dorsal fixation was achieved either with wiring and bone graft/PMMA (Figs. 2 and 3) or with Ransford loop (C0–C3/4). 2.3. Follow up During follow up patients were evaluated clinically and radiologically. Patients were seen prospectively either at our department or by external physicians. Long-term radiological follow up was possible for only 7 patients with a mean of 41.4 months (range 24–66 months). In order to determine fusion rate of the fracture X-rays were performed in these 7 patients. An experienced neuroradiologist evaluated X-rays. Bony and fibrous union of the fracture were classified as stable and non-union as instable. The other patients who could not be evaluated physically for follow up examination were questioned by telephone. 3. Results 3.1. Clinical symptoms and neurological state Predominant preoperative symptom was pain in 24 patients (86%), 5 (18%) had neurological deficits and one patient (4%) had no symptoms. Preoperative neurological score according to the ASIAclassification was class E in 23 patients, class D in 2 patients, C in one patient and class A in 2 cases. Ventral screw fixation was performed in 18 patients (64.3%). 10 patients were operated by a dorsal approach (reasons see above). Among these, in 7 cases C1/2 fixation with wiring and bone graft (n = 5) or PMMA (n = 2) was performed and in one of these 2 patients (PMMA-group) with additional transarticular screw fixation. In 3 patients dorsal C0–C3/4 fixation with a Ransford loop was performed. Early postoperative neurological state was the same in all patients except in one case who had a worsening from ASIA E to D. Overall mean pre- and postoperative ASIA score was 4.57 and 4.55, respectively (Table 1). 3.2. Clinical and radiological follow up On the whole 7 patients were examined clinically and radiologically for long-term follow up (mean 41.4 months, range 24–66 months). All patients had no new neurological deficit. X-rays showed stable bony union in 3 patients and stable fibrous union in 4 patients. Further 7 patients were questioned by telephone. They were clinically unchanged and neurologically stable without any signs for instability. From the remaining 14 patients 7 died in the postoperative course (see below), 4 died to other reasons and 3 were lost to follow up. 3.3. Complications, mortality Surgical complications included dislocation of the dorsal wiring in one patient and break of the ventral screw in another case (Fig. 4). Both patients underwent reoperation and were treated by dorsal C0–C3/4 fixation. One patient (4%) had wound healing problems
2058
A. Ardeshiri et al. / Clinical Neurology and Neurosurgery 115 (2013) 2056–2061
Fig. 1. Sagittal CT scan of a 79-year-old female patient with typical type II dens fracture (left side). After ventral screw fixation a good surgical result with satisfactory reposition of the dens axis was achieved (right side).
Fig. 2. 81-year-old female patient with type II dens fracture (left side). Dorsal approach with posterior C1/2 fixation using wiring and bone graft was performed as fracture was treated 4 months after trauma (right side).
Fig. 3. Another 78-year-old female patient with type II fracture (left) undergoing transarticular screw fixation and dorsal wiring. Note that in this case PMMA was used in order to avoid additional bone harvesting (right).
A. Ardeshiri et al. / Clinical Neurology and Neurosurgery 115 (2013) 2056–2061
2059
Fig. 4. Initial surgery with ventral screw fixation (left) failed due to screw failure (middle) (sagittal CT scans). Therefore, posterior C0–C4 fixation with Ransford loop was performed. Multisegmental fixation was done due to the low bony quality and osteoporosis. The ventral screw was left in situ (right).
that resolved completely with conservative treatment. Another patient had to be decompressed at C6/7 due to a spinal epidural haematoma, which occurred most likely due to the primary trauma and not associated to the initial surgical procedure. Postoperative general medical complications were pulmonary congestion and dysphagia in one patient and laryngeal swelling and tracheotomy in another case. In the early postoperative course 5 patients died due to cardiopulmonary insufficiency. Among these patients one had a cardiopulmonary resuscitation after trauma and surgery was performed very early in order to stabilize the instable dens fracture. CT scans performed after surgery showed a hypoxic brain damage due to resuscitation. Two further patients died in the late postoperative course, one due to pulmonary insufficiency two months, and another one 1 month after operation due to myocardial infarction. Concerning the ASA-score 85.7% of these patients were high-risk patients whereas among the surviving patients 81.0% were highrisk. Thus, the overall morbidity rate was 7% (2 patients), mortality rate was 25% (7 patients). 4. Discussion 4.1. Treatment options and strategy Dens fractures are common fractures in the elderly patient, type II being the most frequent in patients older than 70 years [12]. As the population is getting older surgeons are more faced with the treatment of these types of fractures in aged patients. There is still no agreement in the best treatment of dens fractures in the elderly people [13]. As these patients often have severe comorbidities like cardiopulmonary diseases and take anticoagulant medication treatment is associated with an increased risk for both, surgical and nonoperative therapy. Thus, morbidity and mortality rates are significantly higher when comparing with younger patients [14]. Nonsurgical treatment of dens fractures consists of external rigid mobilization (halo-thoracic vest, Minerva brace, Miami J Collar) while surgical procedures include anterior or posterior fixation techniques. Anterior fixation is achieved by ventral screw fixation. Posterior fixation includes different methods like transarticular screws, laminar clamps or sublaminar wiring. Rigid orthoses have a high risk of pseudarthrosis ranging from 15% to 85% [15]. Posterior fusion has a higher rate of union but has the disadvantage of reducing the rotational movement range of the head up to 40% and flexion–extension can be reduced by 10% [16]. Ventral screw fixation also leads to high union rates (84–96%) [6] without the handicap of decreasing the range of movement. It is
less traumatic because dissection of muscles is almost not necessary. Patients can be mobilized at the very early stage after surgery and hospitalisation can be reduced [3]. Therefore, anterior screw fixation of dens fractures has become a popular treatment option for these lesions. However, anatomical conditions or the direction of the fracture can make an anterior approach impossible. Thus, further surgical strategies are mandatory. 4.2. Clinical results There are only a few studies reporting the clinical and radiological results of surgically treated dens fractures in elderly patients. In the study by Harrop et al. [6] 10 patients over 65 years treated surgically were analyzed. Among them, only one patient had neurological deficits and was ASIA A but postoperative neurological data are missing. Agrillo et al. [16] presented 9 elderly patients with type II remote fractures of the dens treated with anterior screw fixation. None of them showed neurological worsening at the last clinical follow-up. Recently, Platzer et al. [13] presented in a large series 56 patients over 65 years who were treated surgically either with anterior approach (n = 37) or with posterior arthrodesis (n = 19). In their group, 6 patients were ASIA D and 3 were ASIA C who recovered fully after surgery. In the postoperative course three patients developed neurological deficits (in two cases with anterior and in one case with posterior fixation). One year after surgery 80% of these patients were at the same activity level than before trauma. Berlemann and Schwarzenbach [9] presented in their series 19 patients over 65 years treated with anterior screw fixation. 14 of them had no initial neurological deficit, 3 had a short loss of consciousness, another one a period of dysesthesia in the extremities directly after trauma and one patient suffered of C6 hypesthesia on both sides but recovered after treatment. Mean follow-up was 4.5 years. 17 patients (89.5%) were free of pain at follow-up while 2 patients had occasional pain. No neurological worsening was seen and all patients could perform normal activities. In our series 89% (n = 25) were in a good initial neurological condition. We achieved good results concerning the neurological status early after surgery as we saw only a minor neurological worsening in one case. Clinical results during follow-up were also good. Patients who could be evaluated for long-term follow up showed no new neurological deficit. Fusion rates are known to correlate with age and other factors. In the study by Dailey et al. [3] an overall union rate of 81% was reported for anterior screw fixation with a follow up of 3–62 months (mean 15 months). Berlemann and Schwarzenbach [9] had 84% solid fusion for direct odontoid fixation (clinical follow up 1–11 years, mean 4.5 years; radiological follow up 0.3–11 years, mean
2060
A. Ardeshiri et al. / Clinical Neurology and Neurosurgery 115 (2013) 2056–2061
Table 2 Surgical studies on dens fractures in elderly patients (ant.: anterior; post.: posterior; n.m.: not mentioned; morb.: morbidity; mort.: mortality). Study
Age (years)
n
Approach
Fusion rate
Fracture healing
Morb.
Mort.
Agrillo et al. [16] Berlemann and Schwarzenbach [9] Dailey et al. [3] Platzer et al. [13] Present study
>65 >65 >70 >65 >70
9 19 57 56 28
Anterior Anterior Anterior Ant./post. Ant./post.
88% 84% 81% n.m. n.m.
n.m. n.m. n.m. 93% 100%
0% 26% n.m. 16% 7%
0% 0% 9% 6% 25%
2.5 years). Platzer et al. [13] had a fracture-healing rate of 93% in his patients. A stable healing rate was achieved in 100% of our patients who could be evaluated for long-term follow up. Compared to these studies, our data are favourable, although it has to be mentioned that the number of surviving patients undergoing radiological long-term follow up in our series is low (n = 7). However, a stable condition was achieved in all other patients without clinical signs indicating instability (n = 7). Table 2 gives an overview of some of the surgical studies on dens fractures in elderly patients. 4.3. Complications, mortality Cervical orthoses are often not well tolerated by the elderly patient and can lead to cardiac or pulmonary compromise [16]. Kuntz et al. [17] reported of failures of cervical orthoses in 6 of 12 treated patients. All these 6 patients were immobilized either in a halo-thoracic vest or Minerva brace. According to Lind et al. [18] halo-thoracic orthoses might lead to reduced ventilatory capacity. In the series by Tashjihan et al. [19] cardiopulmonary complications for patients treated with a halo vest older than 65 years were 2-fold increased and mortality rate was over 30%. Halo vest is associated with a 4-fold greater mortality rate in patients over 65 when comparing with younger patients [19]. Hanigan et al. [14] reported about 5 deaths (26%) among 19 patients over 80 years with dens fractures during the hospital course. All were treated nonsurgically and none of the 5 operated patients died. The authors also pointed out that there was no late neurological worsening in patients with a stable non-union. They summarized that bed rest and rigid immobilization showed significant morbidity and mortality. Philadelphia collar with development of a stable fibrous union may be adequate in aged patients. In another series the in-hospital-mortality for 33 nonsurgically treated elderly patients with type II fractures was 46% while none of the 11 surgically patient died [20]. Morbidity ranges between 10% and 25% for conservative management in aged patients [13]. Our general complications (7%) are in accordance with the literature. Platzer et al. had a complication rate of 16% [13]. In the study by Dailey et al. [3] 35% had dysphagia postoperatively and in 25% a temporary feeding tube was necessary. 19% had perioperative pneumonia, one had to be reintubated due to laryngeal spasm and in 5% myocardial infarctions were seen perioperatively. The mortality for operated elderly patients with odontoid fractures varies in the literature ranging from 0% to 57% [9,13]. Our study showed a perioperative mortality rate of 25% (n = 7) which is high, however in accordance with the literature. But is has to be mentioned that 6 of these 7 patients were at high anesthesiological risk (ASA 3–5). Therefore, the general medical situation of elderly patients including cardiopulmonary diseases should thoroughly be taken into consideration in order to minimize treatment risks. 5. Conclusions The presented study was undertaken to reveal our clinical results, morbidity and mortality of surgically treated patients over
70 years. Anterior screw fixation is the preferred surgical technique at our department because the approach is less traumatic, the mobility of the neck preserved and the fusion rate is high. If an anterior approach is not possible, a posterior fixation is recommended. The technique for posterior approach is dependant on the anatomical conditions, associated bony and ligamentous lesions, bone quality and expertise of the surgeon. Neurological function after surgery is favourable, general medical and surgical complications were acceptable. However, our study also underlines that these advanced aged patients are at a high risk for mortality. A more thorough evaluation of the general medical situation and comorbidities might help to further shrink down the mortality rates. However, a conservative treatment strategy of this patient group bears also a poor prognosis. Therefore, a prospective study of operatively and nonoperatively treated elderly patients is recommended. Acknowledgements We thank Bernd Otto Hütter, PhD, for help in preparing the manuscript and Tobias Breyer, MD, for radiological help. References [1] Hart R, Saterbak A, Rapp T, Clark C. Nonoperative management of dens fracture nonunion in elderly patients without myelopathy. Spine (Phila Pa 1976) 2000;25:1339–43. [2] Blauth M, Lange UF, Knop C, Bastian L. Spinal fractures in the elderly and their treatment. Orthopade 2000;29:302–17. [3] Dailey AT, Hart D, Finn MA, Schmidt MH, Apfelbaum RI. Anterior fixation of odontoid fractures in an elderly population. Journal of Neurosurgery Spine 2010;12:1–8. [4] Anderson LD, D‘Alonzo RT. Fractures of the odontoid process of the axis. Journal of Bone and Joint Surgery 1974;56:1663–74. [5] Dunn ME, Seljeskog EL. Experience in the management of odontoid process injuries: an analysis of 128 cases. Neurosurgery 1986;18:306–10. [6] Harrop JS, Przybylski GJ, Vaccaro AR, Yalamanchili K. Efficacy of anterior odontoid screw fixation in elderly patients with Type II odontoid fractures. Neurosurgical Focus 2000;8:e6. [7] Pepin JW, Bourne RB, Hawkins RJ. Odontoid fractures, with special reference to the elderly patient. Clinical Orthopaedics and Related Research 1985;17:8–18, 3. [8] Pal D, Sell P, Grevitt M. Type II odontoid fractures in the elderly: an evidence-based narrative review of management. European Spine Journal 2011;20:195–204. [9] Berlemann U, Schwarzenbach O. Dens fractures in the elderly. Results of anterior screw fixation in 19 elderly patients. Acta Orthopaedica Scandinavica 1997;68:319–24. [10] Apfelbaum RI, Lonser RR, Veres R, Casey A. Direct anterior screw fixation for recent and remote odontoid fractures. Journal of Neurosurgery 2000;93:227–36. [11] Bohler J. Anterior stabilization for acute fractures and non-unions of the dens. Journal of Bone and Joint Surgery 1982;64:18–27. [12] Muller EJ, Wick M, Russe O, Muhr G. Management of odontoid fractures in the elderly. European Spine Journal 1999;8:360–5. [13] Platzer P, Thalhammer G, Oberleitner G, Schuster R, Vecsei V, Gaebler C. Surgical treatment of dens fractures in elderly patients. Journal of Bone and Joint Surgery 2007;89:1716–22. [14] Hanigan WC, Powell FC, Elwood PW, Henderson JP. Odontoid fractures in elderly patients. Journal of Neurosurgery 1993;78:32–5. [15] Clark CR, White 3rd AA. Fractures of the dens. A multicenter study. Journal of Bone and Joint Surgery 1985;67:1340–8. [16] Agrillo A, Russo N, Marotta N, Delfini R. Treatment of remote type ii axis fractures in the elderly: feasibility of anterior odontoid screw fixation. Neurosurgery 2008;63:1145–50, discussion 1150–1141.
A. Ardeshiri et al. / Clinical Neurology and Neurosurgery 115 (2013) 2056–2061 [17] Kuntz IV C, Mirza SK, Jarell AD, Chapman JR, Shaffrey CI, Newell DW. Type II odontoid fractures in the elderly: early failure of nonsurgical treatment. Neurosurgical Focus 2000;8:e7. [18] Lind B, Bake B, Lundqvist C, Nordwall A. Influence of halo vest treatment on vital capacity. Spine (Phila Pa 1976) 1987;12: 449–52.
2061
[19] Tashjian RZ, Majercik S, Biffl WL, Palumbo MA, Cioffi WG. Halo-vest immobilization increases early morbidity and mortality in elderly odontoid fractures. Journal of Trauma 2006;60:199–203. [20] Bednar DA, Parikh J, Hummel J. Management of type II odontoid process fractures in geriatric patients; a prospective study of sequential cohorts with attention to survivorship. Journal of Spinal Disorders 1995;8:166–9.
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
Elderly patients are at increased risk for mortality undergoing surgical repair of dens fractures Ardeshir Ardeshiri a,∗ , Siamak Asgari b , Elias Lemonas a , Neriman Oezkan a , Marc Schlamann c , Ulrich Sure a , I. Erol Sandalcioglu a a
Department of Neurosurgery, Universitätsklinikum Essen, Essen, Germany Department of Neurosurgery, Klinikum Ingolstadt, Ingolstadt, Germany c Department of Neuroradiology, Universitätsklinikum Essen, Essen, Germany b
a r t i c l e
i n f o
Article history: Received 24 April 2013 Received in revised form 2 June 2013 Accepted 6 July 2013 Available online 31 July 2013 Keywords: Cervical fracture Dens fractures Elderly patients Odontoid fracture Ventral screw fixation
a b s t r a c t Objective: Dens fractures are common cervical injuries in advanced aged patients. The presented study was undertaken to analyze the clinical results and risks of surgically treated patients with dens fractures over 70 years. Methods: Data of 28 patients (17 female, 11 male) over 70 years treated from September 2004 to October 2009 were recorded. Clinical and radiological parameters were obtained including type of fracture, associated cervical and/or other injuries, comorbidities, symptoms, neurological condition, surgical strategy, postoperative course and complications. Results: 89% were in a good neurological condition before surgery (ASIA E or D). In most cases, surgery was performed at an early stage after trauma (21 patients within 5 days). Ventral screw fixation was the preferred surgical strategy (64%). A slight worsening of neurological functions immediately after operation was only seen in one patient. Five patients died in the early and 2 in the late postoperative course which means a treatment mortality of 25%. Among the surviving patients two had general medical complications. Conclusion: Type II dens fractures are a common fracture of elderly patients. Our results are good concerning the neurological functions. Surgical and general medical complications were acceptable. However, the study also underlines that mortality rate is high and therefore treatment options should be wellconsidered in this high risk group. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Dens fractures count to nearly 20% of all cervical fractures. Among these 65–74% are type II-fractures according to Anderson and D’Alonzo [1]. They are the most common cervical fractures of patients over 70 years. In patients over 80 years they are more frequent than all cervical fractures together [2]. Typical trauma mechanism in the elderly patients is falls or pitching of the head leading to extension injuries and dorsal dislocation [2]. Instability might injure directly or decelerated the spinal cord [3]. Nonunion rate of type II-dens fractures is significant [4]. There are some factors which are known to be associated with an increased risk of nonunion such as advanced age, delayed
∗ Corresponding author at: Department of Neurosurgery, Universitätsklinikum Essen, Hufelandstraße 55, 45122 Essen, Germany. Tel.: +49 201 7231201; fax: +49 201 7235909. E-mail address: [email protected] (A. Ardeshiri). 0303-8467/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.clineuro.2013.07.006
treatment, anterior dislocation more than 4 mm and posterior dislocation [5] as well as reduced halo vest tolerance in nonoperatively treated elderly patients [5]. Furthermore, a mortality rate of 26–47% has been reported for these nonoperatively treated patients due to respiratory-related complications caused by prolonged periods of immobilization [6]. Based on the existing comorbidity in elderly patients, the morbidity and mortality rates are high for both, operative and nonoperative treated patients [7]. Surgical techniques include posterior atlantoaxial fusion with different devices (transarticular screws, laminar clamps, sublaminar wiring) and anterior screw fixation [6]. Posterior fixation has a high rate of union (80–100%) [8], but results in a decrease of the rotational range of the cervical spine [9]. Anterior screw fixation has a high fusion rate (84–96%) and additionally preserves axial rotation [6]. Further advantages are shorter surgery time and no graft harvesting is needed. However, this technique can be limited by anatomic contraindications including a disrupted transverse ligament, in cases of fractures with cranial-posterior to caudalanterior direction and in patients with emphysema of the thorax [10]. Problems of surgical fixation in advanced aged patients are
A. Ardeshiri et al. / Clinical Neurology and Neurosurgery 115 (2013) 2056–2061 Table 1 Number of patients in the different ASIA groups (modified) pre- and postoperatively. Overall mean score was 4.57 and 4.55, respectively. Modified ASIA score A = complete: No motor or sensory function is preserved (1 point) B = incomplete: Sensory but no motor functions preserved below the level of injury (2 points) C = incomplete: Motor function is preserved below the level of injury and more than half of key muscles below the neurological level have muscle grade less than 3 (3 points) D = incomplete: Motor function is preserved below the level of injury, and at least half of key muscles below the neurological level have a muscle grade of 3 or more (4 Points) E = normal: Motor and sensory functions are normal (5 points) Overall mean score
Preoperative
Postoperative
2
2
0
0
1
1
2
3
23
22
4.57
4.55
osteopenia- and respiratory-related morbidity [6]. Thus, despite a normal neurological condition without deficits in many cases the treatment of these elderly patients bears an increased risk, surgically and medically. This study was conducted to evaluate morbidity and mortality of patients older than 70 years undergoing anterior or posterior surgery for dens fractures at our department. 2. Materials and methods 2.1. Patients For this analysis medical reports of patients with dens fractures treated surgically at our hospital from October 2004 to September 2009 were reviewed retrospectively. A total number of 28 patients over 70 years were included into this study. 17 (61%) were female and 11 (39%) male with a mean age of 81.1 years (range 71–94 years). In most cases trauma mechanism was fall. In 25 patients (89.3%) diagnosis was made within 2 days after trauma, among these in 19 patients at day 1 after trauma. In the remaining three cases diagnosis was made 1 month, 6 weeks and 4 months after trauma. A type II-fracture (Anderson and D’Alonzo) was seen in 27 patients (96%) and a mix fracture of type II/III was diagnosed in one patient (4%). Associated cervical injuries or fractures were present in 11 (39%) patients, mostly affecting C1. Additional injuries were nose fracture (n = 2), radius fracture (n = 1), L3-fracture (n = 1), TH7-fracture (n = 1) and rib-fracture (n = 1). 11 (39%) patients had cardiopulmonary comorbidities, in 5 (18%) patients osteoporosis was documented and in 8 (29%) patients no significant comorbidities were recorded. Concerning the anaesthesiological risk in 22 patients an ASA (American Society of Anesthesiologists) score 3, in 5 patients ASA score 2 and in one patient ASA score 4 was documented. ASA 1 and 2 were classified to low-risk and ASA 3–5 to high-risk patients. 2.2. Diagnostics and surgical procedure Clinical symptoms and the neurological condition were classified according to a modified ASIA-classification (American Spinal Injury Association). We assigned each grade points with 5 points for the best neurological function (ASIA E) and 1 point for the worst grade (ASIA A) (Table 1). Preoperative neuroimaging procedures included computed tomography (CT) with coronal and sagittal reconstructions in all patients in order to classify the type and extension of fracture and position of the facet joints. In 18 patients X-rays were performed and four patients additionally underwent
2057
preoperative magnetic resonance imaging (MRI) to evaluate ligamental lesions, epidural haemorrhages and spinal cord lesions. Preoperative extension therapy (Gardner-Wells) was performed in 12 cases (43%) in order to achieve a better reposition of the fracture. Anterior screw fixation [11] with a single screw was performed in 18 patients (Fig. 1). In the remaining 10 cases posterior stabilization techniques were performed. Posterior stabilization was used in cases when anatomic conditions made an anterior approach not possible or when the patient was admitted to our hospital 3 days or later after the trauma. Dependant on the complexity of the fracture, anatomical conditions and the bone quality dorsal fixation was achieved either with wiring and bone graft/PMMA (Figs. 2 and 3) or with Ransford loop (C0–C3/4). 2.3. Follow up During follow up patients were evaluated clinically and radiologically. Patients were seen prospectively either at our department or by external physicians. Long-term radiological follow up was possible for only 7 patients with a mean of 41.4 months (range 24–66 months). In order to determine fusion rate of the fracture X-rays were performed in these 7 patients. An experienced neuroradiologist evaluated X-rays. Bony and fibrous union of the fracture were classified as stable and non-union as instable. The other patients who could not be evaluated physically for follow up examination were questioned by telephone. 3. Results 3.1. Clinical symptoms and neurological state Predominant preoperative symptom was pain in 24 patients (86%), 5 (18%) had neurological deficits and one patient (4%) had no symptoms. Preoperative neurological score according to the ASIAclassification was class E in 23 patients, class D in 2 patients, C in one patient and class A in 2 cases. Ventral screw fixation was performed in 18 patients (64.3%). 10 patients were operated by a dorsal approach (reasons see above). Among these, in 7 cases C1/2 fixation with wiring and bone graft (n = 5) or PMMA (n = 2) was performed and in one of these 2 patients (PMMA-group) with additional transarticular screw fixation. In 3 patients dorsal C0–C3/4 fixation with a Ransford loop was performed. Early postoperative neurological state was the same in all patients except in one case who had a worsening from ASIA E to D. Overall mean pre- and postoperative ASIA score was 4.57 and 4.55, respectively (Table 1). 3.2. Clinical and radiological follow up On the whole 7 patients were examined clinically and radiologically for long-term follow up (mean 41.4 months, range 24–66 months). All patients had no new neurological deficit. X-rays showed stable bony union in 3 patients and stable fibrous union in 4 patients. Further 7 patients were questioned by telephone. They were clinically unchanged and neurologically stable without any signs for instability. From the remaining 14 patients 7 died in the postoperative course (see below), 4 died to other reasons and 3 were lost to follow up. 3.3. Complications, mortality Surgical complications included dislocation of the dorsal wiring in one patient and break of the ventral screw in another case (Fig. 4). Both patients underwent reoperation and were treated by dorsal C0–C3/4 fixation. One patient (4%) had wound healing problems
2058
A. Ardeshiri et al. / Clinical Neurology and Neurosurgery 115 (2013) 2056–2061
Fig. 1. Sagittal CT scan of a 79-year-old female patient with typical type II dens fracture (left side). After ventral screw fixation a good surgical result with satisfactory reposition of the dens axis was achieved (right side).
Fig. 2. 81-year-old female patient with type II dens fracture (left side). Dorsal approach with posterior C1/2 fixation using wiring and bone graft was performed as fracture was treated 4 months after trauma (right side).
Fig. 3. Another 78-year-old female patient with type II fracture (left) undergoing transarticular screw fixation and dorsal wiring. Note that in this case PMMA was used in order to avoid additional bone harvesting (right).
A. Ardeshiri et al. / Clinical Neurology and Neurosurgery 115 (2013) 2056–2061
2059
Fig. 4. Initial surgery with ventral screw fixation (left) failed due to screw failure (middle) (sagittal CT scans). Therefore, posterior C0–C4 fixation with Ransford loop was performed. Multisegmental fixation was done due to the low bony quality and osteoporosis. The ventral screw was left in situ (right).
that resolved completely with conservative treatment. Another patient had to be decompressed at C6/7 due to a spinal epidural haematoma, which occurred most likely due to the primary trauma and not associated to the initial surgical procedure. Postoperative general medical complications were pulmonary congestion and dysphagia in one patient and laryngeal swelling and tracheotomy in another case. In the early postoperative course 5 patients died due to cardiopulmonary insufficiency. Among these patients one had a cardiopulmonary resuscitation after trauma and surgery was performed very early in order to stabilize the instable dens fracture. CT scans performed after surgery showed a hypoxic brain damage due to resuscitation. Two further patients died in the late postoperative course, one due to pulmonary insufficiency two months, and another one 1 month after operation due to myocardial infarction. Concerning the ASA-score 85.7% of these patients were high-risk patients whereas among the surviving patients 81.0% were highrisk. Thus, the overall morbidity rate was 7% (2 patients), mortality rate was 25% (7 patients). 4. Discussion 4.1. Treatment options and strategy Dens fractures are common fractures in the elderly patient, type II being the most frequent in patients older than 70 years [12]. As the population is getting older surgeons are more faced with the treatment of these types of fractures in aged patients. There is still no agreement in the best treatment of dens fractures in the elderly people [13]. As these patients often have severe comorbidities like cardiopulmonary diseases and take anticoagulant medication treatment is associated with an increased risk for both, surgical and nonoperative therapy. Thus, morbidity and mortality rates are significantly higher when comparing with younger patients [14]. Nonsurgical treatment of dens fractures consists of external rigid mobilization (halo-thoracic vest, Minerva brace, Miami J Collar) while surgical procedures include anterior or posterior fixation techniques. Anterior fixation is achieved by ventral screw fixation. Posterior fixation includes different methods like transarticular screws, laminar clamps or sublaminar wiring. Rigid orthoses have a high risk of pseudarthrosis ranging from 15% to 85% [15]. Posterior fusion has a higher rate of union but has the disadvantage of reducing the rotational movement range of the head up to 40% and flexion–extension can be reduced by 10% [16]. Ventral screw fixation also leads to high union rates (84–96%) [6] without the handicap of decreasing the range of movement. It is
less traumatic because dissection of muscles is almost not necessary. Patients can be mobilized at the very early stage after surgery and hospitalisation can be reduced [3]. Therefore, anterior screw fixation of dens fractures has become a popular treatment option for these lesions. However, anatomical conditions or the direction of the fracture can make an anterior approach impossible. Thus, further surgical strategies are mandatory. 4.2. Clinical results There are only a few studies reporting the clinical and radiological results of surgically treated dens fractures in elderly patients. In the study by Harrop et al. [6] 10 patients over 65 years treated surgically were analyzed. Among them, only one patient had neurological deficits and was ASIA A but postoperative neurological data are missing. Agrillo et al. [16] presented 9 elderly patients with type II remote fractures of the dens treated with anterior screw fixation. None of them showed neurological worsening at the last clinical follow-up. Recently, Platzer et al. [13] presented in a large series 56 patients over 65 years who were treated surgically either with anterior approach (n = 37) or with posterior arthrodesis (n = 19). In their group, 6 patients were ASIA D and 3 were ASIA C who recovered fully after surgery. In the postoperative course three patients developed neurological deficits (in two cases with anterior and in one case with posterior fixation). One year after surgery 80% of these patients were at the same activity level than before trauma. Berlemann and Schwarzenbach [9] presented in their series 19 patients over 65 years treated with anterior screw fixation. 14 of them had no initial neurological deficit, 3 had a short loss of consciousness, another one a period of dysesthesia in the extremities directly after trauma and one patient suffered of C6 hypesthesia on both sides but recovered after treatment. Mean follow-up was 4.5 years. 17 patients (89.5%) were free of pain at follow-up while 2 patients had occasional pain. No neurological worsening was seen and all patients could perform normal activities. In our series 89% (n = 25) were in a good initial neurological condition. We achieved good results concerning the neurological status early after surgery as we saw only a minor neurological worsening in one case. Clinical results during follow-up were also good. Patients who could be evaluated for long-term follow up showed no new neurological deficit. Fusion rates are known to correlate with age and other factors. In the study by Dailey et al. [3] an overall union rate of 81% was reported for anterior screw fixation with a follow up of 3–62 months (mean 15 months). Berlemann and Schwarzenbach [9] had 84% solid fusion for direct odontoid fixation (clinical follow up 1–11 years, mean 4.5 years; radiological follow up 0.3–11 years, mean
2060
A. Ardeshiri et al. / Clinical Neurology and Neurosurgery 115 (2013) 2056–2061
Table 2 Surgical studies on dens fractures in elderly patients (ant.: anterior; post.: posterior; n.m.: not mentioned; morb.: morbidity; mort.: mortality). Study
Age (years)
n
Approach
Fusion rate
Fracture healing
Morb.
Mort.
Agrillo et al. [16] Berlemann and Schwarzenbach [9] Dailey et al. [3] Platzer et al. [13] Present study
>65 >65 >70 >65 >70
9 19 57 56 28
Anterior Anterior Anterior Ant./post. Ant./post.
88% 84% 81% n.m. n.m.
n.m. n.m. n.m. 93% 100%
0% 26% n.m. 16% 7%
0% 0% 9% 6% 25%
2.5 years). Platzer et al. [13] had a fracture-healing rate of 93% in his patients. A stable healing rate was achieved in 100% of our patients who could be evaluated for long-term follow up. Compared to these studies, our data are favourable, although it has to be mentioned that the number of surviving patients undergoing radiological long-term follow up in our series is low (n = 7). However, a stable condition was achieved in all other patients without clinical signs indicating instability (n = 7). Table 2 gives an overview of some of the surgical studies on dens fractures in elderly patients. 4.3. Complications, mortality Cervical orthoses are often not well tolerated by the elderly patient and can lead to cardiac or pulmonary compromise [16]. Kuntz et al. [17] reported of failures of cervical orthoses in 6 of 12 treated patients. All these 6 patients were immobilized either in a halo-thoracic vest or Minerva brace. According to Lind et al. [18] halo-thoracic orthoses might lead to reduced ventilatory capacity. In the series by Tashjihan et al. [19] cardiopulmonary complications for patients treated with a halo vest older than 65 years were 2-fold increased and mortality rate was over 30%. Halo vest is associated with a 4-fold greater mortality rate in patients over 65 when comparing with younger patients [19]. Hanigan et al. [14] reported about 5 deaths (26%) among 19 patients over 80 years with dens fractures during the hospital course. All were treated nonsurgically and none of the 5 operated patients died. The authors also pointed out that there was no late neurological worsening in patients with a stable non-union. They summarized that bed rest and rigid immobilization showed significant morbidity and mortality. Philadelphia collar with development of a stable fibrous union may be adequate in aged patients. In another series the in-hospital-mortality for 33 nonsurgically treated elderly patients with type II fractures was 46% while none of the 11 surgically patient died [20]. Morbidity ranges between 10% and 25% for conservative management in aged patients [13]. Our general complications (7%) are in accordance with the literature. Platzer et al. had a complication rate of 16% [13]. In the study by Dailey et al. [3] 35% had dysphagia postoperatively and in 25% a temporary feeding tube was necessary. 19% had perioperative pneumonia, one had to be reintubated due to laryngeal spasm and in 5% myocardial infarctions were seen perioperatively. The mortality for operated elderly patients with odontoid fractures varies in the literature ranging from 0% to 57% [9,13]. Our study showed a perioperative mortality rate of 25% (n = 7) which is high, however in accordance with the literature. But is has to be mentioned that 6 of these 7 patients were at high anesthesiological risk (ASA 3–5). Therefore, the general medical situation of elderly patients including cardiopulmonary diseases should thoroughly be taken into consideration in order to minimize treatment risks. 5. Conclusions The presented study was undertaken to reveal our clinical results, morbidity and mortality of surgically treated patients over
70 years. Anterior screw fixation is the preferred surgical technique at our department because the approach is less traumatic, the mobility of the neck preserved and the fusion rate is high. If an anterior approach is not possible, a posterior fixation is recommended. The technique for posterior approach is dependant on the anatomical conditions, associated bony and ligamentous lesions, bone quality and expertise of the surgeon. Neurological function after surgery is favourable, general medical and surgical complications were acceptable. However, our study also underlines that these advanced aged patients are at a high risk for mortality. A more thorough evaluation of the general medical situation and comorbidities might help to further shrink down the mortality rates. However, a conservative treatment strategy of this patient group bears also a poor prognosis. Therefore, a prospective study of operatively and nonoperatively treated elderly patients is recommended. Acknowledgements We thank Bernd Otto Hütter, PhD, for help in preparing the manuscript and Tobias Breyer, MD, for radiological help. References [1] Hart R, Saterbak A, Rapp T, Clark C. Nonoperative management of dens fracture nonunion in elderly patients without myelopathy. Spine (Phila Pa 1976) 2000;25:1339–43. [2] Blauth M, Lange UF, Knop C, Bastian L. Spinal fractures in the elderly and their treatment. Orthopade 2000;29:302–17. [3] Dailey AT, Hart D, Finn MA, Schmidt MH, Apfelbaum RI. Anterior fixation of odontoid fractures in an elderly population. Journal of Neurosurgery Spine 2010;12:1–8. [4] Anderson LD, D‘Alonzo RT. Fractures of the odontoid process of the axis. Journal of Bone and Joint Surgery 1974;56:1663–74. [5] Dunn ME, Seljeskog EL. Experience in the management of odontoid process injuries: an analysis of 128 cases. Neurosurgery 1986;18:306–10. [6] Harrop JS, Przybylski GJ, Vaccaro AR, Yalamanchili K. Efficacy of anterior odontoid screw fixation in elderly patients with Type II odontoid fractures. Neurosurgical Focus 2000;8:e6. [7] Pepin JW, Bourne RB, Hawkins RJ. Odontoid fractures, with special reference to the elderly patient. Clinical Orthopaedics and Related Research 1985;17:8–18, 3. [8] Pal D, Sell P, Grevitt M. Type II odontoid fractures in the elderly: an evidence-based narrative review of management. European Spine Journal 2011;20:195–204. [9] Berlemann U, Schwarzenbach O. Dens fractures in the elderly. Results of anterior screw fixation in 19 elderly patients. Acta Orthopaedica Scandinavica 1997;68:319–24. [10] Apfelbaum RI, Lonser RR, Veres R, Casey A. Direct anterior screw fixation for recent and remote odontoid fractures. Journal of Neurosurgery 2000;93:227–36. [11] Bohler J. Anterior stabilization for acute fractures and non-unions of the dens. Journal of Bone and Joint Surgery 1982;64:18–27. [12] Muller EJ, Wick M, Russe O, Muhr G. Management of odontoid fractures in the elderly. European Spine Journal 1999;8:360–5. [13] Platzer P, Thalhammer G, Oberleitner G, Schuster R, Vecsei V, Gaebler C. Surgical treatment of dens fractures in elderly patients. Journal of Bone and Joint Surgery 2007;89:1716–22. [14] Hanigan WC, Powell FC, Elwood PW, Henderson JP. Odontoid fractures in elderly patients. Journal of Neurosurgery 1993;78:32–5. [15] Clark CR, White 3rd AA. Fractures of the dens. A multicenter study. Journal of Bone and Joint Surgery 1985;67:1340–8. [16] Agrillo A, Russo N, Marotta N, Delfini R. Treatment of remote type ii axis fractures in the elderly: feasibility of anterior odontoid screw fixation. Neurosurgery 2008;63:1145–50, discussion 1150–1141.
A. Ardeshiri et al. / Clinical Neurology and Neurosurgery 115 (2013) 2056–2061 [17] Kuntz IV C, Mirza SK, Jarell AD, Chapman JR, Shaffrey CI, Newell DW. Type II odontoid fractures in the elderly: early failure of nonsurgical treatment. Neurosurgical Focus 2000;8:e7. [18] Lind B, Bake B, Lundqvist C, Nordwall A. Influence of halo vest treatment on vital capacity. Spine (Phila Pa 1976) 1987;12: 449–52.
2061
[19] Tashjian RZ, Majercik S, Biffl WL, Palumbo MA, Cioffi WG. Halo-vest immobilization increases early morbidity and mortality in elderly odontoid fractures. Journal of Trauma 2006;60:199–203. [20] Bednar DA, Parikh J, Hummel J. Management of type II odontoid process fractures in geriatric patients; a prospective study of sequential cohorts with attention to survivorship. Journal of Spinal Disorders 1995;8:166–9.