The geriatric demographic is one of the fastest growing in North America. The baby boomer generation may now be considered “elderly” as the median age for this group reaches 65 in 2010.1 This population will increase annually by 2.8% until 2030, at which time 1 in 5 Americans will be at least 65 years of age. Spinal injuries are relatively common in the older demographic. Injuries related to falls from standing are of increased prevalence.2 This may be related to cardiovascular disease, peripheral neuropathy, central neuroaxis conditions, low bone mass, and musculoskeletal disability including reduced lean muscle mass that results in more frequent injuries from low energy trauma. One of the most common anatomic regions of injury in the geriatric population is the upper cervical spine. As the population ages, these spinal fractures are likely to become increasingly relevant to clinical practice. Aside from immediate care concerns, the burden of spinal trauma in the elderly population also has significant social, economic, and ethical implications.
The prevalence of upper cervical spine fractures in patients over age 65 is between 2.4% and 4.7%.3,4 Odontoid fractures (Figure 1) account for 10% to 15% of all cervical spine fractures, and the majority of these fractures occur in elderly patients.5 Falls from standing are the most common mechanism of injury, associated with 62% of isolated cervical spine fractures in one study of 177 elderly patents6 and 68.9% of cervical spine injuries in another study of 193 elderly patients.7 Falls from standing were the cause of 70% of cervical spine injuries reported in another series.8 Odontoid fracture morbidity may be considered analogous to hip fracture morbidity, wherein a significant fraction of patients never return to their preinjury functional status.
The treatment of odontoid fractures in elderly patients remains controversial. Conservative treatment, including halo vest immobilization and rigid cervical orthosis, is generally poorly tolerated in this population and is associated with elevated rates of nonunion, morbidity, and mortality.9,10 Nonunion may lead to chronic pain and neurologic deterioration, and prolonged immobilization has been associated with loss of fracture alignment as well as pulmonary complications.11 Surgical management has also been associated with significant complications. Increased perioperative medical risks and specific surgical risks occur with advanced age and decreased bone mass.2 With an increasing emphasis on informed patient choice, access to quality information regarding surgical and nonsurgical risks gains relevance. Enhanced patient and family understanding of shortcomings related to both external immobilization and surgical options may allow for enhanced exchange and improved shared decision-making. This may lead to improved expectations with selection of surgical or orthosis treatments. In the high-risk patient population, this may also allow for well informed and reasonable selection of palliative treatments in an “informed benevolent neglect approach.”
With recognition of the prevalence of odontoid fractures and the current lack of consensus regarding management, the primary focus of this study was to define the morbidity and mortality rates for surgical treatment of odontoid fractures in the elderly population. This was conducted to define findings and limitations of the literature and to identify the areas of future investigation most likely to have a significant impact on this population.
This review addresses the following questions:
- What is the morbidity and mortality following surgery for odontoid fractures in the elderly?
- Among elderly patients treated surgically for odontoid fractures, are there factors that are associated with a greater risk of morbidity or mortality?
Materials and Methods
Electronic Literature Database
The literature search is outlined in detail elsewhere.11a A systematic search was conducted in Medline, EMBASE, and the Cochrane Collaboration Library for literature published through June 2009. The search results were limited to human studies published in the English language. Reference lists of key articles were also systematically checked.
All articles in which upper cervical (C1 and C2 vertebrae) surgery was performed on elderly patients (mean age ≥65 years) that reported on morbidity (any complication) or mortality were identified. We excluded articles by title or abstract that clearly indicated that patients received surgery on vertebrae other than C1 or C2. Other exclusions are described below. Full-text articles of these studies were then obtained and reviewed. Outcomes of interest were mortality and morbidity, including, but not limited to nonunion, infection, deep vein thrombosis, pulmonary embolism, cardiac arrest/myocardial infarction, stroke, pneumonia, skin decubiti, adverse reaction to anesthesia, optic neuritis, loss of fixation, neurologic injuries, vascular injuries, iliac crest bone harvest complications including hematoma, chronic pain, nerve and vessel injuries. Other exclusions included mean patient age less than 65 years (unless the authors presented results separately for the elderly vs. the young patients), vertebral levels not identified, spinal surgery not performed (unless the authors presented results separately for surgical vs. nonsurgical patients), cadaver studies, fewer than 10 patients evaluated per treatment group, reviews, diagnostic studies, editorials, case reports, non-English written studies, and animal studies (Figure 2). All manuscripts were evaluated as prognostic studies (Supplemental Digital Content, Table 1, available at: http://links.lww.com/BRS/A412). We did not place a time limit on the occurrence of adverse events following surgery. Regarding the second question, all above articles that evaluated potential risk factors for increased morbidity or mortality following surgery of odontoid fractures, including (but not limited to) surgical approach, patient age, and postoperative treatment were identified.
Each retrieved citation was reviewed by 2 independent reviewers (D.C.N. and R.H.). Most articles were excluded on the basis of information provided by the title or abstract. Citations that appeared to be appropriate or those that could not be excluded unequivocally from the title and abstract were identified, and the corresponding full text reports were reviewed by the 2 reviewers. Any disagreement between them was resolved by consensus (D.C.N. and R.H.). From the included articles, the following data were extracted: patient demographics, diagnosis, and type of surgical intervention, morbidity, and mortality.
Level of evidence ratings were assigned to each article independently by 2 reviewers using criteria set by The Journal of Bone and Joint Surgery, American Volume12 for prognostic studies and modified to delineate criteria associated with methodologic quality and described elsewhere (Supplemental Digital Content, Table 1, individual study ratings, available at: http://links.lww.com/BRS/A412).
Complication rates were reported as the number of events divided by the number of patients or the number of vertebral levels. Mortality rates were reported as the number of events divided by the number of patients. We reported an overall mortality rate, an in-hospital mortality rate, and a postdischarge mortality rate if the authors reported these separately. We combined individual study raw data to arrive at a pooled complication rate for each complication and mortality rate. These were weighted by study sample size. To evaluate the effect of specific risk factors on morbidity and mortality, we used the pooled risks to calculate unadjusted relative risks (RR) with corresponding confidence intervals and P values using Stata 9.0.3
Data were summarized in Tables 1–7 and qualitative analysis13 was performed considering the following three domains: quality of studies (level of evidence), quantity of studies (the number of published studies similar in patient population, condition treated, and outcome assessed), and consistency of results across studies (whether the results of the different studies lead to a similar conclusion).14 The body of literature was evaluated to determine if it represented a minimum standard for each of the 3 domains using the following criteria: for study quality, at least 80% of the studies reported needed to be rated as a level I or II evidence; for study quantity, at least 3 published studies were needed which were adequately powered to answer the study question; for study consistency, at least 70% of the studies had to have consistent results. The overall strength of the body of literature was expressed in terms of the impact that further research may have on the results. An overall strength of “HIGH” means that further research is very unlikely to change our confidence in the estimate of effect. The overall strength of “MODERATE” is interpreted as further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. A grade of “LOW” means that further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate, whereas “VERY LOW” means that any estimate of effect is very uncertain.11a
The initial literature search criteria identified 247 articles potentially reporting morbidity or mortality following surgery of odontoid fractures in the elderly. From these potential articles, 48 underwent full text review. All articles reported outcomes after odontoid fracture surgery. After full text review, 34 articles were excluded for the following reasons: less than 10 patients per relevant treatment group (n = 10); mean age less than 65 (and outcomes not reported separately for elderly vs. younger patients) (n = 17); not limited to upper cervical spine (n = 2); outcomes not reported separately for surgical versus nonsurgical treatment (n = 2); morbidity or mortality not reported (n = 1); not limited to patients with fractures (n = 1); and 1 article was a review. Of the remaining 14 articles, all were retrospective cohort studies. The selection process is summarized in Figure 3.
Because of limitations in available data, we were not able to accurately compile an overall complication rate for this study population. Limitations in the data available in the published literature included variable definitions of complications. Although the currently available data are not sufficient to project as per admission or as per fracture overall complication rate, for example, individual rates for each complication are reported.
Tables 1–3 summarize the rates of major, site-specific, and other minor complications following surgery, respectively, for odontoid fractures in elderly patients. Brief study details are reported in Table 6, whereas more thorough details for each study can be found in Supplemental Digital Content, Table 2, available at http://links.lww.com/BRS/A412. Studies included here use a variety of surgical approaches, and results for both anterior and posterior fixation are included.
Morbidity and Mortality Following Surgery for Odontoid Fractures in the Elderly
Andersson et al in 2000 retrospectively reported outcomes for a series of 29 consecutive patients over the age of 65 who presented with odontoid fractures.2 The mean age was 78 years (range, 66–99). Eighteen patients were managed surgically, with either anterior screw fixation (n = 11) or posterior C1–C2 fusion (n = 7); of these patients, 18 had type II or III (Anderson and D'Alonzo) odontoid fractures, and 1 had a C1 arch fracture. In addition, 11 patients received nonsurgical treatment. The time from injury to treatment was not reported. No details were given on postoperative immobilization. Only in-hospital mortality was reported separately for surgical and nonsurgical patients; therefore, the patients from this study did not contribute to the overall mortality rate in this review but were included in the in-hospital analysis.
In 1995, Bednar et al conducted a prospective study of 11 geriatric patients with type II odontoid fractures without neurologic deficit.15 The mean patient age was 74 years (range, 58–85). Patients underwent posterior atlantoaxial arthrodesis at a mean of 4 days (range, 0–9 days) postinjury followed by at least 3 months of Philadelphia collar immobilization until radiographically confirmed union was achieved.
In 1997, Berlemann and Schwarzenbach reviewed 19 elderly patients who underwent anterior screw fixation for the treatment of type II dens fractures.16 Four patients also sustained fractures of the C1 arch. Patients were treated after surgery with a soft or hard collar or a Minerva type of plaster for at least 6 weeks. The mean patient age was 75 (range, 65–87) years.
Borm et al in 2003 reported results from a case-control study of 27 elderly patients with type II odontoid fractures.17 Although the mean age was 66 years (range, 20–98 years), patients were subdivided into 2 age-based groups: group 1 patients were 70 years and older (n = 15, mean age: 81 years); group 2 patients were younger than 70 years of age (n = 12; mean age: 49 years). Patients at high risk for anesthetic complications were excluded. All patients received anterior odontoid screw fixation; 2 threaded screws were used per patient. The mean time from injury to surgery was 10 days. Patients wore a semirigid collar for 6 weeks following the procedure.
Collins and Min in 2008 retrospectively evaluated 15 consecutive patients over the age of 60 who presented with Grauer type II B odontoid fractures.18 Patients were treated with anterior screw fixation followed by a soft neck collar for a minimum of 6 weeks after surgery. One patient had an associated C4–C5 distraction-flexion injury; no other patients had any associated spinal fractures. The mean patient age was 68.9 (range, 61–78) years.
Frangen et al in 2007 performed a retrospective analysis of 27 elderly patients with isolated, acute type II odontoid fractures that occurred as a result of low-energy falls.19 The average patient age was 85.5 (range, 63–98) years. Three patients had preoperative neurologic deficits; patients with coinjuries were excluded. All patients underwent a posterior C1/C2 fusion with cannulated screws and subsequently wore a cervical collar for 6–12 weeks.
In 2000, Harrop et al reported ten patients over the age of 65 with type II odontoid fractures treated with anterior odontoid screw fixation.20 Four patients had associated C1 fractures, one of whom also had a C6 fracture. One patient presented with complete tetraplegia because of a diving accident, and the remaining patients were neurologically intact at presentation. Six patients had osteopenia. Patients were placed in a rigid external orthosis (halo vests or collars) following surgery.
In 2000, Kuntz et al conducted a retrospective review of elderly patients with type II odontoid fractures.11 Although 20 patients were included, only 11 underwent surgical treatment [mean age of surgically treated patients: 76 years (range, 66–92)]. Nine patients received posterior C1/C2 transarticular screw fixation with a modified Gallie fusion, whereas 2 patients underwent anterior odontoid screw fixation. All patients were stabilized after surgery with a Philadelphia or Miami J collar for 4 to 6 weeks. Two patients had associated C1 fractures, 1 of whom also sustained fractures at C6, C7, and T5.
Omeis et al published a retrospective study of 29 patients 70 years of age and older who underwent surgical treatment of odontoid fractures.21 Two patients had a central cord syndrome, and 5 patients had accompanying C1 fractures. Odontoid screw fixation was performed in 16 patients (1 patient required revision, in which C1–C2 lateral mass screw fixation was performed), C1–C2 lateral mass screw fixation alone was done for 7 patients, C1–C3 lateral mass screw fixation was performed in 2 patients, cervical laminectomy with occipital cervical lateral mass screw fixation was done in 2 patients, and transarticular screw fixation with modified Gallie fusion was performed in 2 patients. Patients wore a Miami J or soft collar after surgery.
In 2007, Platzer et al published 2 reports that summarized outcomes for elderly patients who underwent surgery for odontoid fractures. For both of these studies, patients who died or had incomplete follow-up were excluded from all calculations except the overall morbidity (which only included cardiac failure, respiratory, pneumonia, severe infection, and thromboembolism) and mortality rates. Platzer et al compared outcomes for older and younger patients with type II odontoid fractures.22 Patients at high risk for complications because of anesthesia and those with fractures unsuited for anterior screw fixation were excluded. A total of 7 of 117 patients were excluded from the author's analyses unless otherwise noted (6 of 47 in the older group; 1 of 70 in the younger group): 5 died during the first 32 days following surgery, and 2 did not attend follow-up examinations. After these exclusions, the older group consisted of 41 patients with a mean age of 71 years (range, 65–83) and the younger group was made up of 69 patients with a mean age of 37 years (range, 6–62). All patients received anterior screw fixation using 2 screws, and were subsequently immobilized for at least 6 weeks following surgery by use of a soft or hard collar. Platzer et al also reviewed the outcomes of 62 elderly patients with type II or type III dens fractures.26 Six of the 62 patients were excluded from the author's analyses unless otherwise noted: 4 died within 32 days postoperation, and 2 were lost to follow-up. Patients were treated with either anterior screw fixation (n = 37) or posterior cervical arthrodesis (n = 19). The mean patient age was 71.4. (66–83).
Smith et al reported in-hospital outcomes in a retrospective study of octogenerians with a mean age of 85.5 years.23 Patients presented with type II odontoid fractures and were surgically managed via an anterior (n = 10) or posterior (n = 32) approach. No other surgical details were provided. Those with associated cervical spine fractures or neurologic deficits were excluded.
Smith et al also reviewed in-hospital complications in patients at least 80 years of age with isolated type II odontoid fractures treated over a 20-year period at a single institution.24 The mean patient age was 82.3 years, and patients were observed during the length of their hospital stay, which was 20.3 days on average. All patients were neurologically intact. Twenty percent of patients underwent anterior odontoid screw fixation (n = 15), 24% C1 lateral mass and C2 pedicle/isthmus constructs (n = 18), 13% Brooks fusion (n = 10), 17% Magerl technique (n = 13), and 25% received occipitocervical fusion (n = 19). Postoperative treatment was not reported.
Tashjian et al reported a 0% in-hospital mortality rate in a nonconsecutive case series of 13 selected elderly patients who received surgical fixation for type II odontoid fractures.25 These patients did not wear a halo vest; the mortality rate for the 4 patients who were treated with surgery but did wear a halo vest was not disclosed. The mean age of all 78 patients, including the 61 who did not undergo operative treatment, was 80.7 years. This study however, excluded all patients who died within 48 hours of hospital admission; the authors did not disclose how many patients were excluded for this reason or whether patients underwent surgery. In addition, those who initially presented with a Glasgow Coma Score of 3 were excluded from the study. Complications were reported but were not separated for surgically- and nonsurgically treated patients.
In summary of the pooled data from these studies, major complications (Table 1) included cardiac failure in a mean of 6.8% (range, 2%–10%) of patients,2,11,15,19–24,26 respiratory failure in 7.7% (range, 3%–20%) of patients,17,19,21,22,26 pneumonia in 9.9% (range, 5%–20%) of patients,16,19,20,22–24,26 deep vein thrombosis or thromboembolism in a mean of 3.2% (range, 2%–6%) of patients,22–24,26 stroke in 2.8% (range, 2.7%–3.1%) of patients,24,25 severe infection in 3.2% (range, 2%–6%) of patients,22–24,26 and liver failure in 6.7% of patients.17
Site-specific complications (Table 2) included nonunion in a mean of 8.9% (range, 0%–27%) of patients,2,11,22–26 failure of reduction or fixation in 9.9% (range, 3%–14%) of patients,2,19,22,26 neck pain that varied from mild to severe at follow-up in 10.8% (range, 0%–15%) of patients,2,16,18,19,22,26 new or unimproved neurologic deficits in 3.6% (range, 0%–9%) of patients,11,15,16,18,19,26 significant loss of neck rotation in 25.3% (range, 10%–53%) of patients,16,18 and limited neck movement in other planes in 20.6% (range, 13%–26%) of patients.16,18,22
Other minor complications (Table 3) reported by at least 2 studies included respiratory insufficiency or respiratory distress in a mean of 6.7% (range, 0%–8%) of patients,17,24 tracheotomy in 14.0% (range, 3%–25%) of patients,21,23,24 feeding tube in a 13.3% (range, 3%–19%) of patients,21,23,24 pulmonary edema in 3.7% (range, 3%–4%) of patients,23,24 dysphagia in 9.8% (range, 7%–13%) of patients,18,23,24 pneumothorax in 1.9% (range, 1%–3%) of patients,23,24 arrythmia in 17.8% (range, 9%–38%) of patients,23,24 urinary tract infection in 3.2% (range, 3%–5%) of patients,16,23,24 wound infection in 3.0% (range, 0%–4%) of patients,19,23,24 and cancer in 9.1% (no range) of patients.11,15 All other complications occurred in less than 10% of patients and were reported only by 1 study. Details of each study can be found in the Supplemental Digital Content, Table 2, available at http://links.lww.com/BRS/A412.
Table 4 summarizes the rates of mortality following surgery for odontoid fractures in the elderly. Overall, a mean of 10.1% (range, 0%–22%) of patients died during the study period as reported by 13 studies.2,11,15–26 The reported in-hospital mortality rate was a mean of 6.2% (range, 0%–13%) of patients,2,11,15–21,23–25 and the time between surgery and death was not consistently reported in all studies. The mean postdischarge mortality rate was 8.8% (range, 0%–18%) of patients.11,15–21 The mortality rates at 6- and 12- months were 7.4% (range, 0%–12.5%) and 7.7% (range, 0%–18%), respectively as reported by twelve studies.11,15–18,20–26 The two studies by Platzer et al did not separate in-house from postdischarge deaths, so they are included in the overall mortality rate; however, are not included in-house and postdischarge rates which explains why these individual rates are less than the overall rate.
Risk Factors for Morbidity and Mortality
The type of surgical approach (anterior vs. posterior) was the only treatment variable which we could consistently extract from the included studies to allow for quantitative evaluation. Six studies2,11,21,23–25 reported on patients treated with either anterior fixation or posterior fusion (Table 5). There was no difference in the pooled mortality rates between these approaches. The in-hospital mortality rate was 7% (3 of 44) following anterior fixation and 8% (7 of 90) following posterior fusion as reported by four studies,2,11,21,24 whereas the overall mortality rate was 9% for both approaches (anterior, 6 of 70; posterior, 9 of 102) as reported by 4 studies.11,21,24,26
More patients developed fracture nonunion following anterior (8%, 5 of 66) as compared with posterior surgery (0%, 0 of 49) as reported by 4 studies.2,11,21,26 Similarly, anterior fixation patients had higher rates of technical complications (including loosening of screws or implant, implant cut-out, secondary loss of reduction, incorrect reduction, malpositioning of implants, or intraoperatively abandoned technique) as well as revision surgery than did patients who underwent posterior fusion: technical failures (anterior: 17%, 11 of 64); posterior: 8%, 3 of 40)2,21,26; revision surgery (anterior: 11%, 4 of 37); posterior (2%, 1 of 42).11,21,26 These differences were not found to be statistically significant, however, which is likely related to the low sample size.
There was no significant difference in the rate of major airway complications (including tracheotomy because of the inability to wean from ventilator, reintubation after surgery, pneumonia, or swallowing/vocal cord dysfunction) between the anterior and posterior approaches (17%, 13 of 78; 18%, 21 of 115), respectively, as reported by 4 studies.2,11,21,24,26 One study23 evaluated impairment in activities of daily living following surgery, and found that patients treated with posterior fusion had higher rates of “impairment in nearly all activities” (26%, 5 of 19 vs. 0%, 0 of 37 for anterior fixation) as well as “impairment in certain activities” (74%, 14 of 19 vs. 30%, 11 of 37) for anterior fixation; RR = 0.40, 95% CI: 0.23–0.71; P = 0.002). Although these patients also had a higher rate of mild pain (posterior: 21%, 4 of 19); anterior: 8%, 3 of 37), the percentage of patients experiencing severe pain at 1-year follow-up was low in both groups [posterior, 5% (1 of 19); anterior, 3% (1 of 37)]. The patient numbers for these studies were relatively low, and larger studies should be conducted to evaluate patient or surgical risk factors for complications and mortality.
Because of study designs and low patient numbers in exposure groups, no other potential risk factors were identified that could undergo quantitative evaluation in this systematic analysis. Factors which were considered, however, but not found to be reported adequately for analysis included age, dementia, and preinjury functional status.
The overall strength of the evidence for studies evaluating morbidity or mortality and their potential risk factors following surgery for odontoid fractures in elderly patients is “low”, that is, future research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate (Table 6).
The overall strength of the evidence for studies evaluating potential risk factors following surgery for odontoid fractures in elderly patients is “very low,” that is, any estimate of effect is very uncertain (Table 7).
This study aimed to determine the literature reported morbidity and mortality following surgery for odontoid fractures in elderly patients. Fourteen studies met the inclusion criteria. The most commonly reported major complications after odontoid fracture surgery in the elderly include cardiac failure (6.8%), DVT (3.2%), stroke (3.2%), pneumonia (9.9%), respiratory failure (7.7%), liver failure (6.7%), and severe infection (3.2%). Other site specific and minor complications are also reported fairly consistently in the literature. The overall mortality rate after surgery is 10.1% (in-hospital, 6.2%; postdischarge, 8.8%); the 6- and 12- month mortality rates were 7.4% and 7.7%, respectively.
Significant shortcomings were identified in this analysis of the literature. None of the studies reported an overall morbidity rate that included each complication of interest. Additionally, these studies reported complications over a variable range of time after surgery, from 0 days to 11 years following surgery. Because the specific time between surgery and each complication was not typically reported, it was not possible to analyze negative outcomes at a specific length of time following surgery or to consider the consistency of the data between studies.
Because of the study design and the low number of patients in potential exposure groups (if reported), the only risk factor that could be quantitatively evaluated was surgical approach. Six studies compared outcomes in patients treated with anterior surgery with those treated with posterior surgery. There were no differences found in the in-hospital or overall pooled mortality rates between patients who underwent anterior compared with posterior surgery. The rates of major airway complications were also similar in these 2 groups. The pooled rate of site-specific complications such as instrumentation failure, nonunion, or revision surgery was higher, however, in patients who received anterior surgery. These differences were not statistically significant, which is likely related to the small sample size, but patients treated with anterior surgery had better clinical success than their posterior counterparts as reported in one study (P = 0.002).
Although a full comparison of surgical versus conservative treatment is beyond the scope of this review, 4 of the analyzed studies did report outcomes for nonsurgical patients2,11,24,26 and these are summarized here. A total of 124 (range, 11–61) patients were managed nonsurgically with halo vests or rigid cervical orthosis. The overall mortality rate was 27.4% (31 of 113 patients) (range, 8.3%–39.3%) as reported by three studies.11,24,26 Deaths occurred as a result of respiratory arrest (pneumonia) in 11 patients and myocardial infarction in 12 patients, whereas the causes of the remaining 8 deaths were not disclosed. Nonunion or inadequate stabilization of the fracture occurred in 47.8% (11 of 23) (1 had stable nonunion) of patients as reported by 2 studies. One study noted significant airway complications (tracheotomy because of inability to wean from ventilator, repeat postoperative intubation, or pneumonia) in 18% of patients treated nonsurgically compared with 31% of surgical patients, a difference which was not found to be statistically significant (P = 0.1). In addition, there was no significant difference in the rate of feeding tube placement in patients managed nonsurgically (5%) versus surgically (18.8%).24 Other complications were reported in only one or two studies, including pneumonia (3 of 52, or 5.8%),11,24 recurrent fracture displacement (1 of 11 or 9%),2 skin blisters from the collar (1 of 11 or 9%),2 failure to thrive (depression and self-induced starvation) (1 of 12 or 8.3%),11 pin site infection (1 of 12 or 8.3%),11 mental status change (not described) (3 of 40, or 7.5%),24 arrhythmia (3 of 40, or 7.5%),24 myocardial infarction (1 of 51, or 2.0%),2,24 bacteremia (1 of 40, or 2.5%),24 subdural hematoma (1 of 40, or 2.5%),24 thrombocytopenia (1 of 40, or 2.5%),24 and metabolic abnormality (1 of 40, or 2.5%).24
The studies reviewed here did not assess whether patients over 80 years of age were at higher risk of complications compared with patients over 65 years of age, but this has been hypothesized to be an important distinction. This factor should be addressed in future studies. One study by Smith et al24 reported early outcomes for 32 octogenarians following odontoid surgery. The in-hospital mortality rate was 13% (4 of 32). This rate is higher than the mean in-hospital mortality rate of 6.2% as reported by 12 studies, and accounts for the high range outlier (range, 0%–13%). If Smith et al were excluded from this range, then the range would be 0%–11%. Patients were not followed past discharge. Rates of complications tended to be higher in octogenarians compared with the overall rates for all elderly patients over age 65 (Tables 1 and 3), including those that occurred in at least two patients: cardiac arrest [9.4% (vs. 6.8% overall)], deep vein thrombosis [6.2% (vs. 3.2% overall)], systemic infection [6.2% (vs. 3.2% overall)], pneumonia [15.6% (vs. 9.9% overall)], tracheotomy [25% (vs. 14% overall)], feeding tube [19% (vs. 13% overall)], reintubation [13% (vs. 9% overall)], dysphagia without feeding tube [13% (vs. 10% overall)], arrhythmia [13% (vs. 18% overall)], and metabolic abnormality [9% (vs. 9% overall)].24 Other risk factors such as dementia were not reported and could not be assessed in this review.
The pooled information presented here enables a more realistic discussion of risks and expected outcomes following surgical treatment of odontoid fractures in the elderly. Future high quality data are likely to further enhance informed health care decisions. Because geriatric spine injuries are becoming increasingly more common, determining preferred treatments has attained a certain acuity.
Specific patient factors should be considered in future studies that evaluate geriatric odontoid fracture treatment risks. These factors include age over 80 years, dementia, presence of neck deformity, presence of medical comorbidities and preoperative functional status. Future research reporting time to complications with assessment of mortality using a survivorship analysis would be helpful in augmenting our current understanding of outcomes. Perhaps future data could be used to develop a scoring system to predict treatment risks and significantly impact the currently complex and arduous decision-making process of determining the most optimal treatments for our patients.
The morbidity and mortality following surgical treatment of odontoid fractures in elderly patients is significant. Based on low-quality evidence, the most commonly reported major complications after odontoid fracture surgery in the elderly include cardiac failure, DVT, stroke, pneumonia, respiratory failure, liver failure, and severe infection. These all occur at a mean rate of less than 10%. The overall mean mortality rate after surgery is 10.1% (in-hospital, 6.9%; postdischarge, 8.8%).
Although the overall level of evidence in the current literature is low, and significant selection bias may exist, pooled data indicate that posterior surgical treatment is advantageous compared with anterior surgical treatment with regard to complications such as instrumentation failure, nonunion, and need for revision surgery.
Although this analysis did not directly compare surgical and nonsurgical treatment, the significant surgical morbidity and mortality determined here must be carefully considered when making odontoid fracture treatment decisions. Given the significant risks, when treating patients considered relatively high risk for surgery (and for formal immobilization), “benign neglect” may be a reasonable treatment choice.
- There exists no consensus regarding the role of surgical treatment. No quality data exist regarding complications for odontoid fractures in the elderly. Additionally, there is significant variability in clinical practice.
- The most commonly reported major complications after odontoid fracture surgery in the elderly include cardiac failure, DVT, stroke, pneumonia, respiratory failure, liver failure, and severe infection at a mean rate of less than 10%.
- The overall mortality rate after surgery to treat odontoid fractures in the elderly is 10.1%.
- There was no difference found between anterior and posterior surgery regarding mortality or airway complication rates. There was however, a higher rate of site-specific complications following anterior compared with posterior surgery. Clinical outcomes were better in the patients treated with posterior surgery.
- Future research will likely have a significant impact on our understanding of complications associated with surgery for odontoid fractures in the elderly.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Web site (www.spinejournal.com).
The authors are indebted to Ms. Nancy Holmes, RN, for her administrative assistance. The authors thank Mark Dekustowski, MD, for reviewing the manuscript. The authors also thank Jens Chapman, MD, for reviewing the manuscript and for providing leadership of this project.