Prognosis of Cervical Diffuse Idiopathic Skeletal Hyperostosis-Related Spine Injuries in Elderly Patients: Analyses of Both Fracture and Spinal Cord Injury Without Fracture

Abstract

Study Design

Retrospective multicenter study.

Objective

The purpose of this study was to compare the prognosis of elderly patients with injuries related to cervical diffuse idiopathic skeletal hyperostosis (cDISH) to matched control for each group, with and without fractures.

Methods

The current multicenter study was a retrospective analysis of 140 patients aged 65 years or older with cDISH-related cervical spine injuries; 106 fractures and 34 spinal cord injuries without fracture were identified. Propensity score–matched cohorts from 1363 patients without cDISH were generated and compared. Logistic regression analysis was performed to determine the risk of early mortality for patients with cDISH-related injury.

Results

Patients with cDISH-related injuries with fracture did not differ significantly in the incidence of each complication and ambulation or severity of paralysis compared to matched controls. In patients with cDISH-related injury without fracture, those who were nonambulatory at discharge comprised 55% vs 34% of controls, indicating significantly poorer ambulation in those with cDISH-related injuries (P = .023). There was no significant difference in the incidence of complications and ambulation or paralysis severity at 6 months as compared with controls. Fourteen patients died within 3 months. Logistic regression analysis identified complete paralysis (odds ratio [OR] 36.99) and age (OR 1.24) as significant risk factors for mortality.

Conclusions

The current study showed no significant differences in the incidence of complications, ambulation outcomes between patients with cDISH-related injury with fracture and matched controls, and that the ambulation at discharge for patients with cDISH-related injury without fractures were significantly inferior to those of matched controls.

Keywords

elderly diffuse idiopathic skeletal hyperostosis cervical diffuse idiopathic skeletal hyperostosis cervical spinal cord injury spinal cord injury without fracture

Introduction

The catastrophic consequences of diffuse idiopathic skeletal hyperplasia (DISH)–based spinal fractures have come to the forefront of attention in recent years because of the increasing prevalence of DISH secondary to the aging of the general population worldwide.^1,2 DISH can occur in the entire spine, and ossification occurs primarily in the anterior longitudinal ligament.³ The resulting fused spinal segment becomes a long lever arm that is subject to traumatic forces. Thus, not only do traumatic impacts that would not result in injury in a normal flexible spine cause complex disability for the DISH spine, but fractures at the DISH fusion site also have greater instability as long bone fractures. As a result, this injury is prone to neurologic deficits,^4–6 and even if paralysis is avoided immediately after injury, there is a high risk of delayed neurologic problems.² Thus, DISH-related spinal injuries have a distinctive character.

The subgroup of cervical DISH (cDISH) fractures is difficult to treat. Patients are not only older by nature but also have more medical comorbidities.⁷ Conservative treatment carries a risk of delayed palsy, and perioperative complications are common.⁷ Furthermore, mortality is high in patients with DISH cervical spine injuries.^4,6,8,9 In a previous report, Ozpinar et al² performed surgical treatment for a DISH fracture and reported the largest surgical cohort of 48 patients. They showed that preoperative American Spinal Injury Association (ASIA) grade and American Society of Anesthesiologists score were risk factors for inability to walk at discharge, and the subaxial injury classification score was a risk factor for increased 30-day mortality. Katoh et al⁵ compared 84 cDISH fractures with 201 thoracolumbar DISH fractures and found that more cDISH fracture cases were associated with ossification of the posterior longitudinal ligament (OPLL), disc-level fracture, a higher rate of spinal cord injury (SCI), and a higher mortality rate. Thus, patients with cDISH injuries represent a high-risk population for treatment.

However, it is unclear to what extent older adults with cDISH-related injuries are at higher risk compared with the general elderly population. In addition, previous studies have focused on fractures of the DISH mass itself, ignoring SCI without fractures. SCI without fracture is a clinically important subgroup of cervical spinal cord injuries, especially in elderly individuals. DISH-based SCI without fracture can be as severe as that with fractures because of the instability resulting from a long lever arm. However, there have been no reports examining non–bone-injury cervical spinal cord injuries associated with cDISH. Therefore, we investigated not only patients with cDISH fractures but also SCI without fractures and compared them with the general elderly population. The purpose of the current study was to provide an overview of the characteristics and clinical course of elderly patients with cDISH-related injuries compared with a matched cohort of patients with non–cDISH-related injuries. The analyses were performed on groups of patients with and without fractures, respectively.

Methods

Patient Population

The current multicenter study was based on a retrospective database of patients aged 65 years or older hospitalized for cervical spinal cord and/or cervical spine injuries between 2010 and 2020, conducted by the Japan Association of Spine Surgeons with Ambition (JASA) group.¹⁰ Thirty-three medical institutions participated in data collection. The study was approved by the Institutional Review Board (No. 3352-1) and each of the 25 representative institutions. Because this was a retrospective study, informed consent was not required for submission. The optout of this study was posted on a website (https://web.sapmed.ac.jp/orsurg/guide/hj0g2h00000007ax-att/pgsps60000000g3l.pdf), and we did not receive any inquiries.

A total of 1512 patients were entered into the database, and 149 patients with cDISH were identified. Excluding 9 of these patients, 140 cDISH-related injuries were included in the analysis; 106 cDISH-related injuries with fracture and 34 injuries without fracture were identified. Propensity score–matched cohorts from 1363 patients without cDISH for each control were generated and compared (Figure 1). Because of the relatively small number of patients with cDISH-related injury without fracture, corresponding controls were selected following a 1:2 ratio. In addition to these, logistic regression analysis was performed to determine the risk of early mortality for patients with cDISH-related injury.

Figure 1.

Flowchart of patient inclusion.

Cervical Diffuse Idiopathic Skeletal Hyperostosis

DISH was diagnosed as the presence of calcification or ossification in at least 4 consecutive vertebrae on computed tomography (CT) images, relatively preserved disc space in the affected region, and no findings suggestive of disc degeneration such as osteosclerosis of the vertebral margins. cDISH was defined as the presence of DISH involving the cervical level. Therefore, patients with DISH in the thoracolumbar spine but with no fused vertebrae in the cervical spine were excluded from the group of cDISH patients.

cDISH-Related Injury

We defined a cDISH-related injury as a fracture in the fused segment due to cDISH and an SCI between the cDISH segment and the next level. Fractures were identified by CT or magnetic resonance imaging (MRI). The SCI level was confirmed by MRI. Fractures occurred in segments that were fused due to DISH. Nine patients with cDISH whose injury levels were distant from cDISH were excluded, as previously described. This population included 1 Hangman fracture, 1 Anderson type III odontoid fracture, and 7 cervical spinal cord injuries without fractures, all more than 1 level away from cDISH.

Patient Demographic and Operative Data

Age, sex, comorbidities, ambulation before injury, cause of injury, presence of OPLL/fracture/dislocation/motor paralysis, and ASIA Impairment Scale (AIS) were recorded, and the 5-Item modified Frailty Index (mFI-5) was determined. In addition, we recorded whether surgery was performed, the surgical procedure, perioperative and postoperative complications during hospitalization, and ambulatory ability at discharge and 6 months later as well as AIS score 6 months later.

Statistical Analysis

Data are presented as means ± standard deviations for continuous variables and as number and percentage for categorical data. We performed statistical analyses using R version 4.2.1 (http://www.R-project.org) on Wilcoxon rank-sum test, Fisher exact test, and Pearson chi-squared test. A P value <.05 was considered a significant difference. We performed propensity score matching using the MatchIt package (https://cran.r-project.org/web/packages/MatchIt). The following items were selected to adjust for patient background: age, sex, and presence of OPLL. Patients with upper cervical fractures were excluded from the control group because there were no patients with upper cervical fractures in patients with cDISH-related injuries with fractures. Paralysis severity (complete paralysis or not) was added to the adjusting items for the group with cDISH-related injury without fracture.

Results

Compared to patients without cDISH, patients with cDISH-related injury were significantly more likely to be male (89% vs 64%, P < .001), had a higher rate of OPLL (54% vs 18%, P < .001), had more fractures (76% vs 53%, P < .001), had fracture levels more commonly in the middle and lower cervical spine (0% vs 41% for upper cervical fractures, P < .001), and had more complete motor paralysis (AIS A/B; 27% vs 18%, P = .035). In addition, the proportion of causes of injury was different, with 52% of patients in the cDISH group suffering from falls from standing or sitting positions. This was in contrast to falls from high places and traffic accidents, which accounted for more than half of patients without cDISH. Thus, the injury profiles of the 2 groups were found to be significantly different (Table 1). Of note, at least 86% of patients with cDISH-related injury had DISH in the thoracolumbar region, which was significantly greater than the 15% of patients without cDISH at most (P < .001). The number of autofused levels at the cervical spine (C-2 to T-1) was as follows: 6 patients had 1 level, 17 patients had 2 levels, 27 patients had 3 levels, 23 patients had 4 levels, and 67 patients had more levels. Additionally, Table 2 shows the range of cDISH (Figure 2).

Table 1.

Patient Demographics.

	Overall, N = 1503	cDISH-Related Injury, N = 140	Others, N = 1363	P-value
Age, year	75.8 ± 6.9	76.2 ± 6.8	75.8 ± 7.0	.39
Sex, (male%)	999 (66%)	125 (89%)	874 (64%)	<.001
BMI, kg/m²	22.0 (3.6)	22.8 (3.7)	22.0 (3.6)	.011
Thoracolumbar DISH				<.001
DISH	189 (13%)	120 (86%)	69 (5.1%)
No DISH	1168 (78%)	14 (10%)	1154 (85%)
Unknown	146 (9.7%)	6 (4.3%)	140 (10%)
Cervical OPLL	326 (22%)	75 (54%)	251 (18%)	<.001
Ambulation before injury				.76
Walk w/o assistance	1422 (96%)	134 (96%)	1288 (95%)
Walk w/assistance	52 (3.5%)	4 (2.9%)	48 (3.6%)
Nonambulatory	15 (1.0%)	2 (1.4%)	13 (1.0%)
Unknown	14	0	14
mFI-5 ≥2	848 (57%)	89 (64%)	759 (56%)	.10
Unknown	14	0	14
Cause of injury				.001
Fall from standing/sitting	572 (38%)	73 (52%)	499 (37%)
Fall from height	572 (38%)	40 (29%)	532 (39%)
Traffic accident	287 (19%)	18 (13%)	269 (20%)
Others	72 (4.8%)	9 (6.4%)	63 (4.6%)
Fracture	832 (55%)	106 (76%)	726 (53%)	<.001
Fracture level				<.001
Upper (O-C2)	298 (36%)	0 (0%)	298 (41%)
Middle (C2/3-5/6)	275 (33%)	47 (44%)	228 (31%)
Lower (C6-)	238 (29%)	58 (55%)	180 (25%)
Posterior	20 (2.4%)	1 (.9%)	19 (2.6%)
Motor paralysis	1048 (70%)	104 (74%)	944 (69%)	.22
AIS				.035
A/B	200 (19%)	28 (27%)	172 (18%)
C/D	841 (81%)	76 (73%)	765 (82%)

BMI, body mass index; cDISH, cervical diffuse idiopathic skeletal hyperostosis; OPLL, ossification of the posterior longitudinal ligament; mFI-5, 5-factor modified frailty index; AIS, ASIA (American Spinal Injury Association) impairment scale.

Table 2.

cDISH Range.

	cDISH-Related Injury with Fracture, N = 106	cDISH-Related Injury without Fracture, N = 34
Cervical	32 (30%)	9 (26%)
Cervicothoracic	49 (46%)	7 (21%)
Other	25 (24%)	18 (53%)

Figure 2.

cDISH range. (A) cervical; (B) cervicothoracic; (C) others.

cDISH-Related Injury With Fracture

After performing propensity score matching, we found no significant differences in gender, OPLL complication rates, fracture level, or severity of paralysis in the matched cohort (Tables 3 and 4). Of the patients with cDISH-related injury, 54% were injured by falls from a standing or sitting position, indicating that minor trauma was a significantly more common cause of injury in these patients than in the control group. Delayed diagnosis was present in 16% of patients with cDISH-related injury, significantly more than in the control group (P = .007). Many patients were managed surgically (79% vs 72%, P = .20), with the majority of both surgical approaches being posterior. Complications occurred in about half of both of the groups, and the mortality rate for patients with cDISH-related injury was approximately double that of the control group (12% vs 6.6%, P = .16). However, there was no significant difference in the incidence of each complication, including breathing and swallowing problems. Furthermore, there were no significant differences in ambulation or paralysis severity at discharge.

Table 3.

Propensity Score Matching for Patients With Fracture.

	Before Matching			After Propensity Score Matching
	Means Treated	Means Control	Std. Mean Diff	Means Treated	Means Control	Std. Mean Diff
Distance	.723	.818	-.853	.723	.744	-.185
Age, year	77.387	74.922	.374	77.387	77.491	-.016
Sex, male	.887	.664	.705	.887	.802	.268
Presence of OPLL	.462	.102	.723	.462	.406	.114

Table 4.

Matched Cohort Analysis for Patients With Fracture.

	cDISH-Related Injury with Fracture, N = 106	Matched Cohort, N = 106	P-value
Age, year	77.4 ± 6.6	77.5 ± 7.6	.88
Sex, male	94 (89%)	85 (80%)	.088
BMI, kg/m²	22.6 ± 3.7	22.2 ± 3.3	.48
Cervical OPLL	49 (46%)	43 (41%)	.41
Ambulation before injury			.58
Walk w/o assistance	100 (94%)	103 (97%)
Walk w/assistance	4 (3.8%)	2 (1.9%)
Nonambulatory	2 (1.9%)	1 (.9%)
mFI-5 ≥2	67 (63%)	61 (58%)	.40
Cause of injury			<.001
Fall from standing/sitting	57 (54%)	27 (25%)
Fall from height	29 (27%)	46 (43%)
Traffic accident	16 (15%)	29 (27%)
Others	4 (3.8%)	4 (3.8%)
Fracture level			.11
Middle (C2/3-5/6)	47 (44%)	59 (56%)
Lower (C6-)	58 (55%)	44 (42%)
Posterior	1 (.9%)	3 (2.8%)
Motor paralysis	70 (66%)	78 (74%)	.23
AIS			.46
A/B	24 (34%)	22 (29%)
C/D	46 (66%)	55 (71%)
Delayed diagnosis	17 (16%)	5 (4.7%)	.007
Dislocation	20 (19%)	27 (25%)	.25
Surgical treatment	84 (79%)	76 (72%)	.20
Surgical procedure			.87
Posterior	80 (95%)	71 (93%)
AP combined	2 (2.4%)	2 (2.6%)
Anterior	2 (2.4%)	3 (3.9%)
Postoperative complication	55 (52%)	53 (50%)	.34
Death	13 (12%)	7 (6.6%)	.16
Worsening of paralysis	4 (3.8%)	2 (1.9%)	.68
C5 palsy/unilateral upper limb	2	1
Quadriplegia/Tetraplegia	2	1
Instrumentation failure	1 (.9%)	2 (1.9%)	>.99
Deep SSI	3 (2.8%)	1 (.9%)	.62
Reoperation	7 (6.6%)	4 (3.8%)	.35
Debridement for SSI/wound failure	3	2
Additional decompression	1	2
Others^a	3	0
Respiratory failure	21 (20%)	18 (17%)	.59
Dysphagia	17 (16%)	20 (19%)	.59
DVT	2 (1.9%)	3 (2.8%)	>.99
PE	1 (.9%)	0 (0%)	>.99
Pneumonia	25 (24%)	18 (17%)	.23
Delirium	16 (15%)	13 (12%)	.55
Ambulation at discharge			.32
Walk w/o assistance	30 (32%)	41 (42%)
Walk w/assistance	23 (25%)	24 (24%)
Nonambulatory	40 (43%)	33 (34%)
Unknown^b	13	8
AIS at discharge			.17
A/B	12 (21%)	19 (29%)
C/D	43 (74%)	39 (59%)
E	3 (5.2%)	8 (12%)
Unknown^b	48	40
Ambulation 6 months after			.42
Walk w/o assistance	46 (68%)	43 (61%)
Walk w/assistance	5 (7.4%)	10 (14%)
Nonambulatory	17 (25%)	17 (24%)
Unknown^b	38	36
AIS 6 months after			.53
A/B	5 (14%)	9 (19%)
C/D	26 (72%)	29 (60%)
E	5 (14%)	10 (21%)
Unknown^b	70	58

^aRealignment for dysphagia, repair of spinal fluid leak, reinsertion of misplaced screws.

^bIncluding deceased patients.

cDISH-Related Injury Without Fracture

There were no significant differences in patient background or treatment (Tables 5 and 6). All patients had motor paralysis, with 12% of patients with cDISH-related injury and 13% of controls having complete motor paralysis; most patients had incomplete motor paralysis. Many patients were managed surgically (74% vs 62%, P = .24), and in all but 1 case, surgery was performed via the posterior approach. Although there was no significant difference in complication rates, 55% of patients with cDISH-related injuries were nonambulatory at discharge compared with 34% of the controls, and 18% of those with cDISH-related injuries were able to ambulate without assistance vs 46% of controls, indicating that walking ability at discharge was significantly worse for individuals with cDISH-related injuries (P = .023). Among the patients who could be examined, we found no significant differences in either ambulation or severity of paralysis after 6 months.

Table 5.

Propensity Score Matching for Patients Without Fracture.

	Before Matching			After Propensity Score Matching
	Means Treated	Means Control	Std. Mean Diff	Means Treated	Means Control	Std. Mean Diff
Distance	.880	.951	-.994	.880	.880	.003
Age, year	72.353	75.386	-.487	72.353	72.221	.021
Sex, male	.912	.710	.711	.912	.912	.000
Presence of OPLL	.765	.298	1.100	.765	.765	.000
AIS A/B	.118	.134	-.049	.118	.132	-.046

Table 6.

Matched Cohort Analysis for Patients Without Fracture.

	cDISH-Related Injury without Fracture, N = 34	Matched Cohort, N = 68	P-value
Age, year	72.4 ± 6.2	72.2 ± 5.8	.95
Sex, male	31 (91%)	62 (91%)	>.99
BMI, kg/m²	23.5 ± 3.5	23.0 ± 3.5	.51
Cervical OPLL	26 (76%)	52 (76%)	>.99
Ambulation before injury			>.99
Walk w/o assistance	34 (100%)	67 (99%)
Walk w/assistance	0 (0%)	1 (1.5%)
mFI-5 ≥2	22 (65%)	36 (53%)	.26
Cause of injury			.28
Fall from standing/sitting	16 (47%)	32 (47%)
Fall from height	11 (32%)	21 (31%)
Traffic accident	2 (5.9%)	11 (16%)
Others	5 (15%)	4 (5.9%)
Motor paralysis	34 (100%)	68 (100%)	>.99
AIS			>.99
A/B	4 (12%)	9 (13%)
C/D	30 (88%)	59 (87%)
Delayed diagnosis	0 (0%)	3 (4.4%)	.55
Dislocation	2 (5.9%)	5 (7.4%)	>.99
Surgical treatment	25 (74%)	42 (62%)	.24
Surgical procedure			>.99
Posterior	25 (100%)	41 (98%)
AP combined	0 (0%)	1 (2.4%)
Postoperative complication	12 (35%)	17 (25%)	.28
Death	1 (2.9%)	0 (0%)	.33
Worsening of paralysis	0 (0%)	2 (2.9%)	.55
C5 palsy/unilateral upper limb	0	2
Deep SSI	0 (0%)	1 (1.5%)	>.99
Reoperation	0 (0%)	1 (1.5%)	>.99
Debridement for SSI	0	1
Respiratory failure	3 (8.8%)	3 (4.4%)	.40
Dysphagia	2 (5.9%)	3 (4.4%)	>.99
DVT	1 (2.9%)	2 (2.9%)	>.99
PE	0 (0%)	1 (1.5%)	>.99
Pneumonia	5 (15%)	5 (7.4%)	.29
Delirium	2 (5.9%)	3 (4.4%)	>.99
Ambulation at discharge			.023
Walk w/o assistance	6 (18%)	31 (46%)
Walk w/assistance	9 (27%)	13 (19%)
Nonambulatory	18 (55%)	23 (34%)
Unknown^a	1	1
AIS at discharge			.45
A/B	2 (6.2%)	5 (7.9%)
C/D	30 (94%)	54 (86%)
E	0 (0%)	4 (6.3%)
Unknown^a	2	5
Ambulation 6 months after			.23
Walk w/o assistance	13 (52%)	34 (72%)
Walk w/assistance	4 (16%)	4 (8.5%)
Nonambulatory	8 (32%)	9 (19%)
Unknown^a	9	21
AIS 6 months after			.45
A/B	0 (0%)	3 (6.5%)
C/D	22 (88%)	40 (87%)
E	3 (12%)	3 (6.5%)
Unknown^a	9	22

^aIncluding deceased patients.

Of the 140 patients with cDISH-related injury, 14 (10%) died within 3 months. Univariate analysis comparing deaths to others met significance at P < .1 for age, mFI-5, presence of dislocation, and complete paralysis (ASIA A/B; Table 7). Logistic regression analysis performed with these items identified complete paralysis (odds ratio [OR] 36.99) and age (OR 1.24) as significant increased risk factors for mortality (Table 8).

Table 7.

Univariate Analysis for Mortality of cDISH-Related Injury.

	Death, N = 14	Alive, N = 126	P-value
Age, year	81.1 ± 4.7	75.6 ± 6.8	.004
Sex, male	14 (100%)	111 (88%)	.36
mFI-5 ≥2	12 (86%)	77 (61%)	.070
Fracture	13 (93%)	93 (74%)	.19
Dislocation	6 (43%)	16 (13%)	.010
Spinal cord injury	13 (93%)	91 (72%)	.12
AIS A/B	10 (71%)	18 (14%)	<.001

cDISH, cervical diffuse idiopathic skeletal hyperostosis; mFI-5, 5-factor modified frailty index; AIS, ASIA (American Spinal Injury Association) impairment scale.

Table 8.

Multivariate Analysis for Mortality of cDISH-Related Injury.

	OR	95% CI	P-value
Age	1.241	1.093-1.451	.010
AIS A/B	36.992	7.989-255.971	<.0001
mFI-5 ≥2	4.016	.771-33.674	.135

Discussion

This study is the largest-scale retrospective study of elderly patients with cDISH. We analyzed a total of 140 patients with cDISH-related injuries, focusing not only on 106 fractures but also on 34 SCI without fracture. In contrast to previous reports, we found that elderly patients with cDISH-related injury with fractures did not differ significantly from matched controls in the incidence of complications, ambulatory outcomes, or severity of paralysis. Furthermore, patients with cDISH-related injuries without fractures mainly had incomplete motor paraplegia, and although the incidence of complications was not significantly different compared with the controls, these patients had significantly worse ambulation at discharge than the matched controls did.

We used a large database not only to analyze the largest ever group of elderly cDISH patients but also to compare the results with those of a well-matched group of subjects selected by matching from a concurrently collected group of elderly patients with cervical spine and cervical SCI. Westerveld et al⁴ compared patients with DISH (40 patients) and AS (14 patients) injury to 111 other patients. They found that patients with DISH or AS had significantly higher complication and mortality rates. However, the study had weaknesses in that the patient population was diverse, including thoracolumbar as well as cervical spine injuries, and the controls were not matched. Schoenfeld et al¹¹ also compared patients with DISH (27 patients) and AS (16 patients) with matched controls. They found no significant difference in the risk of mortality between the DISH group and the controls, but their sample size was limited. Therefore, when limited to elderly patients with cervical spine injuries, the prognostic impact of the presence or absence of DISH is unknown.

In the present study, there were no statistically significant differences in neurologic or life outcomes of elderly patients with cDISH-related injuries as compared with patients without cDISH. Originally, the elderly patients were more frail,^12–14 and recovery in the group of elderly patients with SCI was poor and mortality was high compared with younger patients.^15,16 Therefore, when it comes to cervical spinal cord injuries in older adults, the influence of the presence or absence of DISH on the fate of patients may be limited. It should be noted, however, that the mortality rate of patients with cDISH-related injury with fracture was almost double that of controls. Furthermore, DISH not only leads to osteoporosis, but also increases the risk of falls due to impaired trunk flexibility and compromised balance. In the present database, patients with cDISH suffered the same degree of spinal injury as the group of patients without DISH from a minor fall. DISH is a systemic disease and is associated with abnormal conditions such as metabolic syndrome,¹⁷ which adversely affects life expectancy. We cannot rule out the possibility that the still insufficient number of patients in the present study might have led to erroneous statistical determinations; cDISH fractures may remain a potential factor in worsening life outcome.⁸

Next, we focused our attention not only on fractures but also on SCI without fractures. SCI without fracture is common in patients with preexisting conditions such as cervical spondylosis, OPLL, and canal stenosis (ie, older adults) when they experience minor trauma without bone damage.^18,19 Most cDISH-related injuries without fracture occurred at the margin of the cDISH. Most patients had incomplete motor paralysis, and there was no significant difference in complication rates compared with matched controls. On the other hand, interestingly, patients with cDISH-related SCI without fracture had significantly poorer ambulatory ability at discharge (the short-term outcome) compared with controls; however, neurologic outcomes at 6 months were not significantly different between the groups. Because ankylosing spine negatively affects trunk stability and balance,²⁰ patients with DISH have poor activity. This might have led to differences in the results of short-term rehabilitation.

Patients with cDISH-related injury with fracture are often overlooked, leading to the subsequent occurrence of paralysis or more severe paralysis.^4,21–23 We found that the increased risk of mortality in patients with cDISH-related injuries was influenced by older age and the occurrence of complete paralysis. Therefore, to reduce the risk of mortality, the delay in diagnosis must be reduced. Our results showed a significantly higher diagnostic delay (16%) in patients with cDISH-related injuries compared with the control group. However, the total mortality rate was 10%, which is comparable or lower than in previous reports,²¹ considering that the mean age of the patient population in previous reports was younger than that in the current study. Although recent improvements in the knowledge of DISH might have reduced diagnostic delays and indirectly contributed to the lower mortality rate, the reduction is insufficient. Therefore, there is a need for continued efforts to educate health care providers about the diagnostic difficulties of this injury.

The current study had several limitations. First, the retrospective nature of this research might have led to selection bias. Second, although cDISH was determined by skilled spine surgeons based on imaging findings, it is not clear whether the ossification extends to the posterior elements and is a complete fusion or only ossification of the anterior longitudinal ligament. Therefore, it is unclear whether the results of the present study represented cDISH as an ankylosed spine, and future analysis based on the severity of ossification of DISH is needed. Third, the findings regarding the sacroiliac joint used for a rigorous DISH diagnosis were not confirmed in all cases. However, because the number of AS patients in Japan is small, it can be assumed that the present data set consists of DISH patients. Thus, this study included data from a wide area and allowed for the evaluation of outcomes in elderly patients with cDISH.

Conclusion

In this study, which is the largest retrospective study of elderly patients with cDISH-related injuries, there were no significant differences in the incidence of complications, ambulation outcomes, or severity of paralysis in patients with cDISH-related injury with fracture as compared with matched controls, and although most patients with cDISH-related injury without fractures had incomplete paralysis, their ambulation outcomes at discharge were significantly inferior to those of the matched control group.

Footnotes

Acknowledgments

We thank all members of the Japan Association of Spine Surgeons with Ambition (JASA) for collecting the data.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical Approval

Thirty-three health care facilities were included in the database. The study protocol was approved by the Institutional Review Board (no. 3352-1) of each of the 25 representative institutions.

ORCID iDs

Naoki Segi

Hiroaki Nakashima

Takeshi Sasagawa

Kota Watanabe

Atsushi Yunde

Tomohiro Yamada

Hidenori Suzuki

Yasuaki Imajo

Shota Ikegami

Masashi Uehara

Ko Hashimoto

Hidetomi Terai

Koji Tamai

Gen Inoue

Yasushi Oshima

Toshitaka Yoshii

Tetsuro Ohba

Masayuki Ishihara

Shiro Imagama

References

World Population Prospects - Population Division - United Nations . https://population.un.org/wpp/. Accessed May 3, 2022.

Ozpinar

Perez

Hacker

, et al. Operative outcomes after cervical diffuse idiopathic skeletal hyperostosis fracture in the elderly. Internet J Spine Surg. 2022;16(3):435-441. doi:10.14444/8252

Olivieri

D’Angelo

Palazzi

Padula

Mader

Khan

. Diffuse idiopathic skeletal hyperostosis: Differentiation from ankylosing spondylitis. Curr Rheumatol Rep. 2009;11(5):321-328. doi:10.1007/s11926-009-0046-9

Westerveld

van Bemmel

Dhert

WJA

Oner

Verlaan

. Clinical outcome after traumatic spinal fractures in patients with ankylosing spinal disorders compared with control patients. Spine J. 2014;14(5):729-740. doi:10.1016/j.spinee.2013.06.038

Katoh

Okada

Yoshii

, et al. A comparison of cervical and thoracolumbar fractures associated with diffuse idiopathic skeletal hyperostosis—A nationwide multicenter Study. J Clin Med. 2020;9(1):208. doi:10.3390/jcm9010208

Caron

Bransford

Nguyen

Agel

Chapman

Bellabarba

. Spine fractures in patients with ankylosing spinal disorders. Spine. 2010;35(11):E458-E464. doi:10.1097/BRS.0b013e3181cc764f

Bransford

Koller

Caron

, et al. Cervical spine trauma in diffuse idiopathic skeletal hyperostosis: Injury characteristics and outcome with surgical treatment. Spine. 2012;37(23):1923-1932. doi:10.1097/BRS.0b013e31825b17fc

Romero-Muñoz

Tipper

Segura-Fragoso

Barriga-Martín

. Outcomes of spinal cord injury following cervical fracture in ankylosing spondylitis and diffuse idiopathic skeletal hyperostosis (DISH): A prospective cohort study. Neurocirugia. 2022;33(6):275-283. doi:10.1016/j.neucir.2021.06.004

Teunissen

Verbeek

Cha

Schwab

. Spinal cord injury after traumatic spine fracture in patients with ankylosing spinal disorders. J Neurosurg Spine. 2017;27(6):709-716. doi:10.3171/2017.5.SPINE1722

10.

Yokogawa

Kato

Sasagawa

, et al. Differences in clinical characteristics of cervical spine injuries in older adults by external causes: A multicenter study of 1512 cases. Sci Rep. 2022;12(1):15867. doi:10.1038/s41598-022-19789-y

11.

Schoenfeld

Harris

McGuire

Warholic

Wood

Bono

. Mortality in elderly patients with hyperostotic disease of the cervical spine after fracture: An age- and sex-matched study. Spine J. 2011;11(4):257-264. doi:10.1016/j.spinee.2011.01.018

12.

Kobayashi

Imagama

Sato

, et al. Postoperative complications associated with spine surgery in patients older than 90 years: A multicenter retrospective study. Global Spine J. 2018;8(8):887-891. doi:10.1177/2192568218767430

13.

Kobayashi

Imagama

Ando

, et al. Complications associated with spine surgery in patients aged 80 years or older: Japan association of spine surgeons with ambition (JASA) multicenter study. Global Spine J. 2017;7(7):636-641. doi:10.1177/2192568217716144

14.

Kobayashi

Imagama

Ando

, et al. Risk factors for delirium after spine surgery in extremely elderly patients aged 80 years or older and review of the literature: Japan association of spine surgeons with ambition multicenter study. Global Spine J. 2017;7(6):560-566. doi:10.1177/2192568217700115

15.

Inglis

Banaszek

Rivers

, et al. In-hospital mortality for the elderly with acute traumatic spinal cord injury. J Neurotrauma. 2020;37(21):2332-2342. doi:10.1089/neu.2019.6912

16.

Johansson

Luoto

Vainionpää

, et al. Epidemiology of traumatic spinal cord injury in Finland. Spinal Cord. 2021;59(7):761-768. doi:10.1038/s41393-020-00575-4

17.

Kuperus

Mohamed Hoesein

FAA

de Jong

Verlaan

. Diffuse idiopathic skeletal hyperostosis: Etiology and clinical relevance. Best Pract Res Clin Rheumatol. 2020;34(3):101527. doi:10.1016/j.berh.2020.101527

18.

Chikuda

Seichi

Takeshita

, et al. Acute cervical spinal cord injury complicated by preexisting ossification of the posterior longitudinal ligament: A multicenter study. Spine. 2011;36(18):1453-1458. doi:10.1097/BRS.0b013e3181f49718

19.

Machino

Yukawa

Ito

, et al.

Can magnetic resonance imaging reflect the prognosis in patients of cervical spinal cord injury without radiographic abnormality?

Spine. 2011;36(24):E1568-E1572. doi:10.1097/BRS.0b013e31821273c0

20.

Acar

Ilçin

Gürpinar

Can

. Core stability and balance in patients with ankylosing spondylitis. Rheumatol Int. 2019;39(8):1389-1396. doi:10.1007/s00296-019-04341-5

21.

Rustagi

Drazin

Oner

, et al. Fractures in spinal ankylosing disorders: A narrative review of disease and injury types, treatment techniques, and outcomes. J Orthop Trauma. 2017;31:S57-S74. doi:10.1097/BOT.0000000000000953

22.

Hartmann

Tschugg

Wipplinger

Thomé

. Analysis of the literature on cervical spine fractures in ankylosing spinal disorders. Global Spine J. 2017;7(5):469-481. doi:10.1177/2192568217700108

23.

Hendrix

Melany

Miller

Rogers

. Fracture of the spine in patients with ankylosis due to diffuse skeletal hyperostosis: Clinical and imaging findings. Am J Roentgenol. 1994;162(4):899-904. doi:10.2214/ajr.162.4.8141015