The incidence of CS-Fx in the general Norwegian population was 14.9/100,000 person-years, and a large fraction of the patients were elderly with significant preinjury comorbidities. The two most frequent trauma mechanisms were falls and bicycle accidents. At the time of injury, 16% of the patients were under the influence of ethanol. Treatment was external immobilization with a stiff neck collar alone in 64.9%, open surgical fixation and/or decompression in 25.9%, and no stabilization or decompression in 9.2%. Concomitant cSCI was registered in 12% and was associated with higher admittance rates to a level 1 trauma center, a higher surgical rate, and an increased LOH stay.
The incidence of traumatic CS-Fx in general Western populations is of 4-17/100,000 person-years in Canada, Finland, Germany, Ireland, Norway, the Netherlands, and Sweden [1–8]. In the present study from 2015 to 2019, we estimate the incidence of CS-Fx in the general Norwegian population to be 14.9/100,000 person-years, thus confirming the results of a previous Norwegian publication covering the period of 2009-2012 [2]. Most likely, the “true” incidence of CS-Fx is slightly higher due to some undiagnosed fractures and underreporting of CS-Fx from local hospitals to the registry at the level 1 trauma hospital (OUH). We believe the three most likely reasons for undiagnosed fractures are as follows: 1) trauma victims do not contact the health care system; 2) appropriate radiological examinations are not performed; and 3) trauma victims die at the accidents’ scene. The fraction of missed cases of CS-Fx due to inappropriate radiological examinations is most likely very low at this time, as high-quality cervical CT with reconstruction has become the standard assessment for cervical injuries in all levels of health care services in Norway. The level of underreporting of CS-Fx from local hospitals to the central registry is low. The completeness of this reporting is supported by the equality of our results compared to the nationwide CS-Fx incidence in our previous publication based on data from the Norwegian Patient Registry [2] and our clinical experience. In the last decade, we have virtually never received nonregistered patients for reevaluation after primary conservative treatment has been initiated at a local hospital in our region.
Most studies report a male preponderance of CS-Fx [1–3, 7, 8]. This is in line with the 68% male patients in our present study. The main reason for male overrepresentation is most likely a gender difference in everyday risk-taking behavior [19]. The gender difference diminished with increasing age, and in the >80-year-old age group, there were more women than men. This is probably due to the longer life expectancy in women. In a previous Norwegian publication, we found that the frequency and the relative incidence of CS-Fx were highest among the elderly and rather rare in children [2]. The latter finding was confirmed in the present study and is similar to what has been observed in contemporary series of patients with TBI [10–12].
Increasing age was associated with both more comorbidities and the need for help with ADL. Of patients ≥ 65 years of age (WHO definition of elderly), 64% had pre-injury severe systemic disease, and 29% needed assistance in ADL. The increasing number of old and frail patients with CS-Fx may change the treatment strategy for some fracture types and, for many, result in prolonged acute care and increased mortality. The most common fracture in the elderly is an odontoid fracture. Surgical fixation has previously been recommended for type II odontoid fractures. We have recently documented that most of these patients can be managed with external immobilization alone, thereby avoiding a surgical procedure associated with high risk in frail older people [20–22].
The most common trauma mechanisms for CS-Fx were falls (57%), followed by bicycle accidents (12%) and 4 W-MVAs (10%), reflecting the rather high mean age of the patients. Of the fall injuries, 49% occurred at home, and 8% occurred in nursing homes or hospitals. The two dominating types of falls were falls from the patient’s height (51%) and falls downstairs (20%). Given the aging population worldwide, the incidence of fall-related CS-Fx is expected to continue rising unless effective fall preventive measures are taken. Defined risk factors for falls are age > 80 years, comorbidities, polypharmacy, impaired cognition, impaired hearing and impaired vision [23, 24]. To date, most studies indicate a benefit of interventions to prevent fall injuries in the elderly [25–28].
In Norway, bicycle injuries are a more frequent cause of CS-Fx than 4 W-MVAs. Road safety policies have had great success for reducing 4 W-MVAs, while road safety for bicyclists has lagged [29]. Norwegian authorities encourage people to commute by bicycle to improve public health, decrease rush-hour traffic jams and reduce pollution. However, the increasing numbers of bicyclists, especially during rush-hour traffic, have resulted in a rising number of serious bicycle injuries [29]. Bicyclists have a much higher injury risk per transported km than car occupants [30]. Thus, road safety must be improved for bicyclists.
As for TBI patients, ethanol influence was registered in many of our patients with CS-Fx and was especially associated with falling in stairs [10, 12]. Excessive alcohol consumption impairs cognitive, motor, and sensory functions, leading to increased injury risk. A significant focus in trauma prevention must still be increased awareness of the effects of excessive ethanol use. Hopefully, public education and increasing knowledge can reduce these alcohol-related injuries. Political legislation may also help prevent ethanol-associated injuries.
A detailed description of the CS-Fx was beyond the scope of the present study. The level of CS-Fx was C0 – C2 in 38%, C3 – C7 in 55%, and both C0 – C2 and C3 – C7 in 7%. The most frequent Fx in the upper cervical spine was C2 odontoid Fx (the most common Fx in the elderly), which is in line with other reports [3, 22]. The most frequent C3 –C7 (subaxial) Fx was a facet Fx involving level C6/C7, also in line with previous studies [31, 32].
Concomitant cSCI was seen in 11.6% of the patients with CS-Fx, of whom 85% had incomplete cSCI and 15% had complete cSCI. Factors significantly associated with concomitant cSCI were male sex, dependent living, and subaxial fractures. Age and multiple traumas were not associated with an increased risk of concomitant cSCI. Our rate of 11.6% concomitant cSCI is substantially lower than in the past [33–35], which is in line with more recent reports [3, 36]. This variation is mainly due to incomplete registration of “less severe” cases of CS-Fx in many studies. In addition, many reports represent subpopulations (e.g., patients admitted to trauma centers or military hospitals) and not general populations. The proportion of CS-Fx patients with concomitant cSCI may also decrease due to improvements in diagnostic neuroradiology. Today, with high-quality CT, we are diagnosing many cases of CS-Fx that were previously missed on plain X-rays. In elderly patients, the most common type of incomplete traumatic cSCI is central cord syndrome following hyperextension in patients with preinjury cervical spinal stenosis (CSS) [37–39]. Preinjury CSS was seen in ~40% of our patients with concomitant cSCI (unpublished results). This information was not included in Table 1 since CSS was not registered for patients without concomitant cSCI in our database. CS-Fx patients with cSCI are more resource-demanding for level 1 trauma centers than those without cSCI due to a higher referral rate, higher rate of open surgery, and longer LOH stay. Thus, knowledge of the expected number of these patients per year is important for hospital planning. We have a separate ongoing study to further characterize patients with traumatic cSCI concerning the level of injury, the severity of the injury, patient age and comorbidities, acute management, access to specialized rehabilitation centers, and long-term functional outcome.
Fifty-seven percent of the patients with CS-Fx in the southeastern region of Norway were admitted to OUH (level 1 trauma center) for initial evaluation and treatment. The remaining patients were managed in other hospitals within our health region when the neurosurgical team at OUH determined that conservative treatment could be instituted at the local hospital. All patients in need of cervical spine surgery were admitted to OUH, as were almost all the patients with concomitant cSCI. In total, 43% of the patients were managed at local hospitals close to their homes and without unnecessary and expensive transportation to and from the level 1 trauma center. The collaboration between the trauma center and local hospitals was efficient and smooth since the safe and fast electronic transfer of CT and MR images is available between all hospitals in our health region.
Treatment was made of external immobilization with a stiff neck collar alone in 64.9%, open surgical fixation with or without decompression in 25.9%, and no stabilization or decompression in 9.2%. The majority of patients in the “no stabilization or decompression group” had isolated fractures of a spinous or transverse process. The surgery rate was significantly higher among patients with concomitant cSCI than for those without concomitant cSCI (75% versus 20%). It is difficult to evaluate whether our rate of surgical fixation is in line with that of other countries since there are hardly any published reports of the rate of surgery for CS-Fx in a defined general population. Recent reports estimated the incidence of CS-Fx surgeries in Germany to be 3.24/100,000 person-years [4] and in Finland to be 4.1/100,000 person-years [40]; both were very similar to our incidence of CS-Fx surgeries of 3.7/100,000 person-years.
To perform state-of-the-art surgical fixation for CS-Fx, the surgical team must be experienced and qualified, to manage all the procedures necessary for anterior and posterior decompression/fixation [41–43]. Perioperative neuronavigation is an advantage in complex cases. Given a rate of CS-Fx surgery of 3.7/100,000 person-years, centralization of this kind of surgery is necessary to maintain a competent 24/7 surgical service for these patients. Emergency neurosurgery for traumatic brain injury is also centralized in Norway, and the crude incidence for such procedures in Norway is 3.9/100 000 person-years [44].
The overall 90-day mortality was 7.1%. In the multivariate logistic regression analysis, the following variables remained significantly associated with increased 90-day mortality: raising age, preinjury ASA score ≥3, dependent living, concomitant cSCI, and severity of the head injury. High mortality is closely linked to older patient age and comorbidities. As mentioned above, we have recently documented that most frail old patients with C2 odontoid Fx can be managed with external immobilization alone, thereby avoiding a surgical procedure associated with high risk in frail older people [20–22]. However, injury preventive measures must be implemented to reduce the mortality rate significantly, e.g., fall prevention measures in the elderly.
Strengths of the study
This is a contemporary prospective study from a defined general population.
Limitations of the study
Most likely, the “true” incidence of CS-Fx in Norway is slightly higher due to some undiagnosed fractures and underreporting of CS-Fx from local hospitals to the registry, as discussed above. The external validity of this study is difficult to assess due to the lack of similar studies from other countries. The lifestyle, age distribution in the population, and environmental factors are associated with CS-Fx. Thus, the external validity of this study is limited to countries with a large proportion of elderly individuals where a fall is the dominating injury mechanism.