Eur Spine J (2015) 24:203–208 DOI 10.1007/s00586-014-3534-1
ORIGINAL ARTICLE
Incidence and the risk factors of spinal deformity in adult patient after spinal cord injury: a single center cohort study Mitsuru Yagi • Atsushi Hasegawa • Masakazu Takemitsu Yoshiyuki Yato • Masafumi Machida • Takashi Asazuma
•
Received: 15 May 2014 / Revised: 20 August 2014 / Accepted: 21 August 2014 / Published online: 24 August 2014 Ó Springer-Verlag Berlin Heidelberg 2014
Abstract Study design A retrospective consecutive case series of adult spinal cord injuries (SCIs) patients. Objective To assess the incidence and risk factors of spinal deformity in a large sample of patients with SCIs. Summary of background data Post-traumatic spinal deformities are well-recognized sequelae of SCIs. Despite the devastating complications for SCI patients with trunk imbalance, the incidence, clinical outcomes, and independent risk factors of scoliosis after SCI remain controversial. Materials and methods We assessed 214 consecutive adult compressive SCI patients who were hospitalized in our hospital. We compared patients who developed spinal deformities with those who did not. Univariate and multivariate analyses to determine the independent risk factors were performed. Age, gender, etiology, ASIA grade (American Spinal Injury Association) surgery, and other demographic data were analyzed to determine the risk factors for developing a spinal deformity. Results The average patient age was 58.3 years (20–86 years). The etiology was trauma (n = 158), ossification of ligament (n = 22), infectious (n = 17), and others. One hundred fifty-two patients had cervical spine involved, 62 had thoracic spine involved. 26 patients classified as ASIA A, 54 were ASIA B, 96 were ASIA C, and 42 were ASIA D 4. One hundred thirty-five patients had either decompression or decompression and fusion M. Yagi and A. Hasegawa are equally contributed to this work. M. Yagi (&) A. Hasegawa M. Takemitsu Y. Yato M. Machida T. Asazuma Department of Orthopedic Surgery, National Center for Musculoskeletal Disorders Murayama Medical Center, 2-37-1 Gakuen, Musashi-Murayama City, Tokyo, Japan e-mail: [email protected]
surgery. The incidence of spinal deformities was 21 % (44/ 214). The mean Cobb angle was 28.9 degrees (13–38°). ASIA grade and surgery predicted the occurrence of spinal deformity in both the univariate model (ASIA grade, OR: 1.59 [95 % CI: 1.04–2.44; P = 0.032]; Surgery, OR: 4.47 [95 % CI: 1.89–10.06; P = 0.0007]) and the multivariate model (ASIA grade, OR: 1.63 [95 % CI: 1.04–2.57; P = 0.033]; Surgery, OR: 4.59 [95 % CI: 1.91–11.04; P = 0.0006]), whereas surgery was the most important risk factor in the Cox model (HR: 3.50 [95 % CI: 1.56–7.88; P = 0.0025]). Conclusions The SCI patients with high ASIA grades and those who had undergone surgery had a higher likelihood of developing a spinal deformity. Of these risk factors, surgery was the stronger risk factor. Keywords
Spinal cord injury Scoliosis Deformity
Introduction Because of the concomitant sequelae of spinal cord injuries (SCIs), providing comprehensive treatment for SCI patients is often challenging. Post-traumatic spinal deformities are well-recognized sequelae of SCIs. Despite the devastating complications in SCI patients with trunk imbalances, the incidence, clinical outcomes, and independent risk factors for scoliosis after an SCI are still to be debatable. Previous studies have demonstrated the high incidence of post-SCI spinal deformities in pediatric patients [1–9]. Minimizing complications and optimizing outcomes are some of the essential goals of patient treatment. Dearolf et al. [2] reported a high rate of post-traumatic spinal deformities in pediatric patients with SCIs. They reported that the overall rate of post-SCI spinal
123
204
deformities was 97 % in pediatric patients who had not yet experienced a growth spurt. Bergstrom et al. [1] reported on late spinal deformities after childhood-onset SCI and their risk factors, which include age at onset and the severity of the SCI. There is a paucity of information about the incidence and risk factors of post-SCI spinal deformities that specifically addresses independent risk factors in relation to the rates of post-SCI spinal deformities in adult patients. To our knowledge, no studies have reported the incidence of and quantitative information about the risk factors for post-SCI spinal deformities in adult patients based on independent variables. The purpose of this study was to assess the incidence of and risk factors for spinal deformities in a large sample of adult patients with SCIs, as well as to relate these results to patient complications.
Materials and methods This retrospective cohort study enrolled 336 consecutive adult patients with compressive SCIs who were hospitalized for extensive rehabilitation. The inclusion criteria were as follows: (1) age over 20 years at the onset of injury, (2) compressive SCI, and (3) periodical radiographs. The patients who had SCIs of ASIA grade (American Spinal Injury Association) E, spinal cord infarction and other non-compressive SCIs, as well as patients who had a previous history of scoliosis, were excluded from this study [10]. A total of 214 met the inclusion criteria. Radiographic measurements and demographic data were reviewed. A minimum of 1 year of postinjury follow-up were required. Radiographic follow-up included radiographs taken at five different time points: immediately after the SCI, 1 month post-SCI, 2 months post-SCI, 6 months post-SCI, 1 year post-SCI, and at a final follow-up visit. Ambulatory patients were asked to stand up straight and look straight ahead with their knees locked, their feet shoulder-width apart, and their arms straight at their sides. Non-ambulatory patients were asked to sit straight and to look straight ahead. Patients who could not maintain a seated posture in a chair were excluded from the study. If the radiographic images of a patient were too poor to enable an accurate Cobb measurement, we did not include the patient in the present study. The diagnostic criteria for scoliosis were a physical examination consistent with spinal curvature in the standing anteroposterior radiographs exhibiting a minimum of 10° of lateral curve in the coronal plane based on the Cobb method. The average patient age was 58.3 years (20–86 years). The etiology was trauma (n = 158), ossification of ligament (n = 22), infectious (n = 17), and others. One Hundred fifty-two patients had cervical spine involved, 62 had thoracic spine involved. Twenty six patients classified
123
Eur Spine J (2015) 24:203–208
as ASIA A, 54 were ASIA B, 96 were ASIA C, and 42 were ASIA D 4. One hundred thirty-five patients had either decompression or decompression and fusion surgery (90 patients had a fusion and 45 had a decompression alone). We compared the patients who developed post-injury spinal deformities with those who did not. Age, gender, etiology, level of injury, ASIA grade, history of surgery, and other demographic data were analyzed to identify the risk factors for developing a spinal deformity. Two of the authors (MY and AH) collected all of the radiographic measurements in collaboration with a senior spinal surgeon who was independent of the treatment team. Statistical analysis The outcomes of interest were (1) age, (2) gender, (3) etiology, (4) level of injury, (5) ASIA grade, and (6) surgery. We calculated the following summary statistics: means and standard deviations for continuous variables and frequencies and percentages for categorical variables. After performing the descriptive analysis, we performed univariate comparisons to determine the independent associations between potential risk factors and the development of a post-SCI spinal deformity. To assess the magnitude of the association, we calculated unadjusted odds ratios and their respective 95 % confidence intervals (CIs). Next, we created a multivariate binary logistic regression model to evaluate the adjusted predictive ability of each potential explanatory variable for the development of a post-SCI spinal deformity. We included variables with a univariate significance level of 0.20 or less and variables that we thought were clinically relevant to the analysis performed using the multivariate logistic regression model. Using a forward stepwise procedure, we removed the variables that failed to achieve a P value of 0.15 or less in the final model. Due to the explanatory nature of these analyses, we chose P = 0.15 as the threshold for retention in the multivariate logistic regression model. To assess the risk factors for developing a post-SCI spinal deformity over time, while minimizing the influence of confounding variables, for the period beyond the follow-up time, we used a Cox proportional hazards regression model to estimate the hazard ratios and their 95 % CIs. For all regression models, a P \ 0.05 with a CI of 95 % was considered significant. All of the analyses were performed with the Statistical Package for the Social Sciences (SPSS, Chicago, Illinois).
Results Among the 214 patients evaluated, the incidence of spinal deformities was 44 of 214 (21 %) (Table 1). The mean Cobb angle was 28.9° (13–38°). There were 29 single and 9
Eur Spine J (2015) 24:203–208
205
Table 1 Descriptive summary of patient cohort No. of patient
214
Age
58.3 ± 17.6 (20–86)
Gender (F:M)
49:165
Follow-up (years)
1.4 ± 1.1 (0.5–8.5)
Etiology SCI without bone lesion
124
SCI with bone lesion
34
OPLL/OYL
22
Pyogenic Spondylitis/Potts
17
Others
17
Level of injury Cervical
152
Proximal thoracic
8
Middle thoracic
21
Distal thoracic ASIA grade
33
A
26
B
54
C
96
D
42
SCI with scoliosis
44
SCI without scoliosis
170
Mean and SD. Range in parenthesis
Table 2 Descriptive summary of SCI patient with scoliosis No. of patient
44
Age
58.1 ± 17.8 (19–84)
Gender (F:M)
11:33
Term (years)
1.5 ± 0.9 (1–6)
Cobb angle (degrees) Curve direction (RT:LT)
18.1?/4.9 (13–28) 32:12
Curve type (single:double)
33:11
Curve progression (5°)
38/44
Range in parenthesis
double curves. Among the 29 single curves, 22 patients had apes at thoracic lesion and 7 patients had apex at thoracolumbar lesion. Among the group of patients with spinal deformities, the average age was 58.1 years. Twenty-eight patients had a major curve directed to the right, and 10 had a major curve to the left. Thirty-nine of 44 (89 %) patients who developed spinal deformities during follow-up did so within the first 3 months post-injury (Table 2). In 38 of these 44 patients, the curve progressed more than 5° during follow-up (Figs. 1, 2). One patient required additional corrective spinal surgery who had ASIA grade B SCI at C5 level and developed post-SCI scoliosis Fig. 3. Our univariate analysis suggested that there are many risk factors associated with developing spinal deformities,
including age, ASIA grade, OPLL/OYL, and surgery, but only ASIA grade and surgery were statistically significant (ASIA grade, OR: 1.59 [95 % CI: 1.04–2.44; P = 0.032]; Surgery, OR: 4.47 [95 % CI: 1.89–10.06; P = 0.0007]; Table 3). Our multivariate analysis confirmed that these two risk factors were independently associated with the development of spinal deformities (Table 4). ASIA grade and surgery were the most significant independent risk factors for developing a spinal deformity (ASIA grade, OR: 1.63 [95 % CI: 1.04–2.57; P = 0.033]; Surgery, OR: 4.59 [95 % CI: 1.91–11.04; P = 0.0006]; Table 4). Interestingly, the etiology of the injury, level of injury, age, and gender were not independent risk factors according to our multivariate analysis. Having surgery was the most important risk factor (HR: 3.50 [95 % CI: 1.56–7.88; P = 0.0025]) in our provisional hazard model. The patients who underwent surgery were 3.5 times more likely to develop a spinal deformity (Fig. 1). ASIA grade was a non-significant risk factor in our provisional hazard model (Table 5).
Discussion In the present study, we found that the incidence of postSCI spinal deformities in an adult population was 21 %. The incidence of post-SCI spinal deformities has been reported in previous publications [1–9] Previous studies have reported a wide range (50–100 %) for the incidence of post-SCI spinal deformities in pediatric patients [1, 2, 4, 5]. However, these reports address only adolescent populations [1, 2, 4, 5]. To the best of our knowledge, there are no studies of post-SCI spine deformity in adult populations. The risk factors for post-SCI spinal deformities include a younger age at the onset of injury, higher ASIA grade, and spasticity [1, 2, 4, 5, 7]. Some spinal deformities have a multi-factorial etiology that can be related to patient-oriented factors [1–3, 5]. Mayfield et al. reported that the overall incidence of post-SCI spinal deformities was 64 % in a pediatric population, and most of these children required spinal fusion to maintain a seated posture [11]. In contrast, Mehta et al. evaluated 123 pediatric SCI patients and demonstrated that bracing patients with spinal curvature of less than 10 degrees prevented the need for surgical intervention. However, the relative risk for post-SCI spinal deformities associated with clinical factors has never been quantified. In the present study, we identified two potential risk factors, ASIA grade and surgery. Our finding is particularly important because previous studies did not include multivariate analyses of data, and it is possible that there are confounding variables associated with the risk factors that have been reported. The incidence of post-SCI spinal deformities in our sample was limited to the incidence among adult patients.
123
206
Eur Spine J (2015) 24:203–208
Fig. 1 Forty nine year-old male had ASIA grade B SCI due to burst fracture at C5 and had a combined anterior/posterior spine fusion. The scoliosis significantly progressed from 8 degrees to 27°. [a Immediate SCI Cobb angle 8°, b 2 months after SCI Cobb angle 20°, c 6 months after SCI Cobb angle 27°]
A
B
C
8
20
P.I. 2 month 16
P.I. 1 month
A
27
P.I. 321 month
B
C
8
P.I. 1 month
21
16
P.I. 2 month
31
P.I. 6 month
Fig. 2 30 year-old male had ASIA grade C SCI due to burst fracture at T8. The scoliosis significantly progressed from 8° to 31°. [a Immediate SCI Cobb angle 8°, b 2 months after SCI Cobb angle 21°, c 6 months after SCI Cobb angle 31°]
A post-SCI spinal deformity is a complication that can occur in any age after an SCI; therefore, we cannot report that our incidence rate is accurate for all ages. Our incidence rate of post-SCI spinal deformities (21 %) is lower than the rate given in previous reports. Previous studies focused on the incidence of post-SCI spinal deformities in pediatric populations, and most of the patients studied in previous research had complete paralysis or tetraplegia. In addition, the etiology of injury included in most of the previous reports varied from the compressive spinal cord injury to spinal infarction. This heterogeneity in the
123
population led to diverse types of scoliosis and weakened the studies’ conclusions. We recognize the limitations of the present study. First, our retrospective study design precludes us from making any strong conclusions about the independent risk factors we identified. However, our multivariate logistic analysis was conducted with a consecutive patient sample, which is the most feasible method for identifying and quantifying risk factors in clinical research when randomized controlled trials are not possible. Second, the follow-up period varied from 1 year to 5 years. In the present study, we
Eur Spine J (2015) 24:203–208
207 Table 5 Hazard ratio for scoliosis after SCI
1.00
Valuables
P value
Hazard ratio
Scoliosis
0.80
95 % CI Lower limit
Model 1
0.60
Model 2
0.40
Upper limit
Surgery
0.0025*
3.5033
1.5566
7.8849
ASIA grade
0.0623
1.4422
0.9813
2.1196
* Indicated statistically significant
0.20
0.00 0
1
2
3
4
5
6
Yrs
Fig. 3 The graph shows a cumulative hazard plot for the development of post-SCI scoliosis with stratification based on ASIA grade and application of surgery. According to this model, those with higher ASIA grade and had a surgery have a higher risk at all time points for developing post-SCI scoliosis than those with lower ASIA grade and did not have a surgery. Model 1: Patient with ASIA D SCI and did not have a surgery. Model 2: Patient with ASIA A SCI and had a surgery
Table 3 Uni-variate predictors of scoliosis after SCI Valuables
P value
Odds ratio
95 % CI Lower limit
Upper Limit
Age
0.3700
0.9916
0.9735
Gender (female)
0.7097
1.1579
0.5351
1.0101 2.5055
ASIA grade
0.0322*
1.5936
1.0403
2.4413
Level of SCI
0.5764
1.0800
0.8245
1.4148
SCI without bone lesion
0.6559
1.1669
0.5919
2.3005
SCI with bone lesion
0.6409
1.2308
0.5143
2.9452
OPLL/OYL
0.4012
0.5815
0.1640
2.0616
Spondylitis
0.8523
0.8837
0.2404
3.2476
Others Surgery
0.7571 0.0007*
0.8153 4.4725
0.2236 1.8870
2.9724 10.6007
* Indicated statistically significant
Table 4 Adjusted odds ratio for scoliosis after SCI Valuables
P value
Odds ratio
95 % CI Lower limit
Upper limit
OPLL/OYL
0.5331
0.7324
0.2751
1.9499
ASIA Grade Surgery
0.0331* 0.0006*
1.6344 4.5966
1.0403 1.9136
2.5679 11.0413
* Indicated statistically significant
found that 90 % of spinal deformities occurred within 3 months after SCIs, and 86 % progressed significantly. This result suggests that a minimum of 1 year is acceptable
for measuring the incidence of spinal deformities after SCIs. Moreover, our provisional hazard model was based on our multivariate analysis from a relatively large sample of patients, which is often the most feasible method for identifying and quantifying risk factors in the medical literature when randomized controlled trials are not possible. To our knowledge, there are no studies currently available in the literature that addresses the differences in the deformities of adult post-SCI conditions. We have identified two independent risk factors, ASIA grade and surgery, for post-SCI spinal deformities in a uniform adult population. The wide variety of ASIA grades observed in the present study and the differences in the ages of the patients in the cohort may have played a role in the lower incidence of post-SCI spinal deformities observed in the present study. Previous studies have identified non-independent risk factors associated with post-SCI spinal deformities in adolescent populations [1–7]. In the present study, we identified multiple factors, including ASIA grade and surgery (Table 4). Both of these factors were independently associated with post-SCI spinal deformities. Our finding is particularly important because previous studies did not perform multivariate analyses, and it is possible that there were confounding variables associated with the risk factors identified in those studies. To date, there is no study that provides a quantitative assessment of risk factors for the development of post-SCI spinal deformities based on independent variables. Surgery was specifically identified in our Cox proportional hazards model. This result may be explained by the relationship that many of the risk factors, including ASIA grade and surgery, have with each other. One could argue that ASIA grade and surgery are inherently related. However, we did not find evidence of this relationship in our data analysis, suggesting that the ASIA grade and surgery independently increased the risk of post-SCI spinal deformities in the adult patients in the present study. We do not have a clear understanding of the effect that surgery has on the development of post-SCI spinal deformities, but one possible explanation exists. The comparisons of the incidence of post-SCI spinal deformities between decompression alone and decompression and fusion indicated no statistical difference (Table 6). Intuitively, this risk factor could increase the stress concentration at the proximal and distal
123
208
Eur Spine J (2015) 24:203–208
Table 6 The incidence of scoliosis in surgically treated patients with SCI Decompression alone
Decompression and fusion
Total
Scoliosis
14
18
32
Normal
31
72
103
Total
45
90
135
P value 0.24
end of the fusion, theoretically increasing the risk for developing a junctional deformity [12–14]. A recent study also reported that post-surgical trunk imbalance after spine surgery affects iatrogenic spinal deformity, but this finding has not been uniform in studies assessing the risk factors for post-SCI spinal deformities [12–14]. Our findings suggest that the etiology of post-SCI spinal deformities is uncertain in adult populations. Various studies have proposed that the development of spinal deformities is related to a combination of age at onset, spasticity, severity of the SCI, and other factors [1–7]. Of these contributing factors, age at onset, skeletal maturity, the severity of the SCI, and injury level have all been implicated in the development of post-SCI spinal deformities. Although these results may be accurate, our analysis only suggests that ASIA grade and surgery are associated with post-SCI spinal deformities. While we identified risk factors similar to those reported in the literature, they did not independently predict post-SCI spinal deformities. It remains unclear why surgery had an impact on the development of post-SCI spinal deformities, but our data suggest that patients with severe SCIs who undergo surgery should be monitored closely for the development of post-SCI spinal deformities.
Conclusion SCI patients with higher ASIA grades who underwent surgery had a higher likelihood of developing spinal deformities. Of these risk factors, surgery was the stronger risk factor.
123
Conflict of interest
None.
References 1. Bergstro¨m EM, Short DJ, ASIA HL, Henderson NJ, Jones PR (1999) The effect of childhood spinal cord injury on skeletal development: a retrospective study. Spinal Cord 37(12):838–846 2. Dearolf WW 3rd, Betz RR, Vogel LC, Levin J, Clancy M, Steel HH (1990) Scoliosis in pediatric spinal cord-injured patients. J Pediatr Orthop 10(2):214–218 3. Denis F, Armstrong GW, Searls K, Matta L (1984) Acute thoracolumbar burst fractures in the absence of neurologic deficit. A comparison between operative and nonoperative treatment. Clin Orthop Relat Res 189:142–149 4. Hamilton MG, Myles ST (1992) Pediatric spinal injury: review of 61 deaths. J Neurosurg 77(5):705–708 5. Lancourt JE, Dickson JH, Carter RE (1981) Paralytic spinal deformity following traumatic spinal-cord injury in children and adolescents. J Bone Joint Surg Am 63(1):47–53 6. Malcolm BW, Bradford DS, Winter RB, Chou SN (1981) Posttraumatic kyphosis. A review of forty-eight surgically treated patients. J Bone Joint Surg Am 63(6):891–899 7. Malcolm BW (1979) Spinal deformity secondary to spinal injury. Orthop Clin North Am 10(4):943–952 8. Mehta S, Betz RR, Mulcahey MJ, McDonald C, Vogel LC, Anderson C (2004) Effect of bracing on paralytic scoliosis secondary to spinal cord injury. J Spinal Cord Med 27:S88–S92 9. Vaccaro AR, Silber JS (2001) Post-traumatic spinal deformity. Spine 26:S111–S118 (Review) 10. Marino RJ, Barros T, Biering-Sorensen F et al (2003) ASIA Neurological Standards Committee 2002. International standards for neurological classification of spinal cord injury. J Spinal Cord Med 26:S50–S56 11. Mayfield JK, Erkkila JC, Winter RB (1981) Spine deformity subsequent to acquired childhood spinal cord injury. J Bone Joint Surg Am 63(9):1401–1411 12. Kim YJ, Bridwell KH, Lenke LG, Glattes CR, Rhim S, Cheh G (2008) Proximal junctional kyphosis in adult spinal deformity after segmental posterior spinal instrumentation and fusion: minimum five-year follow-up. Spine 33(20):2179–2184 13. Parent S, Dimar J, Dekutoski M, Roy-Beaudry M (2010) Unique features of pediatric spinal cord injury. Spine 35:S202–S208 14. Arun R, Srinivas S, Mehdian SM (2010) Scoliosis in Duchenne’s muscular dystrophy: a changing trend in surgical management: a historical surgical outcome study comparing sublaminar, hybrid and pedicle screw instrumentation systems. Eur Spine J 19(3):376–383
ORIGINAL ARTICLE
Incidence and the risk factors of spinal deformity in adult patient after spinal cord injury: a single center cohort study Mitsuru Yagi • Atsushi Hasegawa • Masakazu Takemitsu Yoshiyuki Yato • Masafumi Machida • Takashi Asazuma
•
Received: 15 May 2014 / Revised: 20 August 2014 / Accepted: 21 August 2014 / Published online: 24 August 2014 Ó Springer-Verlag Berlin Heidelberg 2014
Abstract Study design A retrospective consecutive case series of adult spinal cord injuries (SCIs) patients. Objective To assess the incidence and risk factors of spinal deformity in a large sample of patients with SCIs. Summary of background data Post-traumatic spinal deformities are well-recognized sequelae of SCIs. Despite the devastating complications for SCI patients with trunk imbalance, the incidence, clinical outcomes, and independent risk factors of scoliosis after SCI remain controversial. Materials and methods We assessed 214 consecutive adult compressive SCI patients who were hospitalized in our hospital. We compared patients who developed spinal deformities with those who did not. Univariate and multivariate analyses to determine the independent risk factors were performed. Age, gender, etiology, ASIA grade (American Spinal Injury Association) surgery, and other demographic data were analyzed to determine the risk factors for developing a spinal deformity. Results The average patient age was 58.3 years (20–86 years). The etiology was trauma (n = 158), ossification of ligament (n = 22), infectious (n = 17), and others. One hundred fifty-two patients had cervical spine involved, 62 had thoracic spine involved. 26 patients classified as ASIA A, 54 were ASIA B, 96 were ASIA C, and 42 were ASIA D 4. One hundred thirty-five patients had either decompression or decompression and fusion M. Yagi and A. Hasegawa are equally contributed to this work. M. Yagi (&) A. Hasegawa M. Takemitsu Y. Yato M. Machida T. Asazuma Department of Orthopedic Surgery, National Center for Musculoskeletal Disorders Murayama Medical Center, 2-37-1 Gakuen, Musashi-Murayama City, Tokyo, Japan e-mail: [email protected]
surgery. The incidence of spinal deformities was 21 % (44/ 214). The mean Cobb angle was 28.9 degrees (13–38°). ASIA grade and surgery predicted the occurrence of spinal deformity in both the univariate model (ASIA grade, OR: 1.59 [95 % CI: 1.04–2.44; P = 0.032]; Surgery, OR: 4.47 [95 % CI: 1.89–10.06; P = 0.0007]) and the multivariate model (ASIA grade, OR: 1.63 [95 % CI: 1.04–2.57; P = 0.033]; Surgery, OR: 4.59 [95 % CI: 1.91–11.04; P = 0.0006]), whereas surgery was the most important risk factor in the Cox model (HR: 3.50 [95 % CI: 1.56–7.88; P = 0.0025]). Conclusions The SCI patients with high ASIA grades and those who had undergone surgery had a higher likelihood of developing a spinal deformity. Of these risk factors, surgery was the stronger risk factor. Keywords
Spinal cord injury Scoliosis Deformity
Introduction Because of the concomitant sequelae of spinal cord injuries (SCIs), providing comprehensive treatment for SCI patients is often challenging. Post-traumatic spinal deformities are well-recognized sequelae of SCIs. Despite the devastating complications in SCI patients with trunk imbalances, the incidence, clinical outcomes, and independent risk factors for scoliosis after an SCI are still to be debatable. Previous studies have demonstrated the high incidence of post-SCI spinal deformities in pediatric patients [1–9]. Minimizing complications and optimizing outcomes are some of the essential goals of patient treatment. Dearolf et al. [2] reported a high rate of post-traumatic spinal deformities in pediatric patients with SCIs. They reported that the overall rate of post-SCI spinal
123
204
deformities was 97 % in pediatric patients who had not yet experienced a growth spurt. Bergstrom et al. [1] reported on late spinal deformities after childhood-onset SCI and their risk factors, which include age at onset and the severity of the SCI. There is a paucity of information about the incidence and risk factors of post-SCI spinal deformities that specifically addresses independent risk factors in relation to the rates of post-SCI spinal deformities in adult patients. To our knowledge, no studies have reported the incidence of and quantitative information about the risk factors for post-SCI spinal deformities in adult patients based on independent variables. The purpose of this study was to assess the incidence of and risk factors for spinal deformities in a large sample of adult patients with SCIs, as well as to relate these results to patient complications.
Materials and methods This retrospective cohort study enrolled 336 consecutive adult patients with compressive SCIs who were hospitalized for extensive rehabilitation. The inclusion criteria were as follows: (1) age over 20 years at the onset of injury, (2) compressive SCI, and (3) periodical radiographs. The patients who had SCIs of ASIA grade (American Spinal Injury Association) E, spinal cord infarction and other non-compressive SCIs, as well as patients who had a previous history of scoliosis, were excluded from this study [10]. A total of 214 met the inclusion criteria. Radiographic measurements and demographic data were reviewed. A minimum of 1 year of postinjury follow-up were required. Radiographic follow-up included radiographs taken at five different time points: immediately after the SCI, 1 month post-SCI, 2 months post-SCI, 6 months post-SCI, 1 year post-SCI, and at a final follow-up visit. Ambulatory patients were asked to stand up straight and look straight ahead with their knees locked, their feet shoulder-width apart, and their arms straight at their sides. Non-ambulatory patients were asked to sit straight and to look straight ahead. Patients who could not maintain a seated posture in a chair were excluded from the study. If the radiographic images of a patient were too poor to enable an accurate Cobb measurement, we did not include the patient in the present study. The diagnostic criteria for scoliosis were a physical examination consistent with spinal curvature in the standing anteroposterior radiographs exhibiting a minimum of 10° of lateral curve in the coronal plane based on the Cobb method. The average patient age was 58.3 years (20–86 years). The etiology was trauma (n = 158), ossification of ligament (n = 22), infectious (n = 17), and others. One Hundred fifty-two patients had cervical spine involved, 62 had thoracic spine involved. Twenty six patients classified
123
Eur Spine J (2015) 24:203–208
as ASIA A, 54 were ASIA B, 96 were ASIA C, and 42 were ASIA D 4. One hundred thirty-five patients had either decompression or decompression and fusion surgery (90 patients had a fusion and 45 had a decompression alone). We compared the patients who developed post-injury spinal deformities with those who did not. Age, gender, etiology, level of injury, ASIA grade, history of surgery, and other demographic data were analyzed to identify the risk factors for developing a spinal deformity. Two of the authors (MY and AH) collected all of the radiographic measurements in collaboration with a senior spinal surgeon who was independent of the treatment team. Statistical analysis The outcomes of interest were (1) age, (2) gender, (3) etiology, (4) level of injury, (5) ASIA grade, and (6) surgery. We calculated the following summary statistics: means and standard deviations for continuous variables and frequencies and percentages for categorical variables. After performing the descriptive analysis, we performed univariate comparisons to determine the independent associations between potential risk factors and the development of a post-SCI spinal deformity. To assess the magnitude of the association, we calculated unadjusted odds ratios and their respective 95 % confidence intervals (CIs). Next, we created a multivariate binary logistic regression model to evaluate the adjusted predictive ability of each potential explanatory variable for the development of a post-SCI spinal deformity. We included variables with a univariate significance level of 0.20 or less and variables that we thought were clinically relevant to the analysis performed using the multivariate logistic regression model. Using a forward stepwise procedure, we removed the variables that failed to achieve a P value of 0.15 or less in the final model. Due to the explanatory nature of these analyses, we chose P = 0.15 as the threshold for retention in the multivariate logistic regression model. To assess the risk factors for developing a post-SCI spinal deformity over time, while minimizing the influence of confounding variables, for the period beyond the follow-up time, we used a Cox proportional hazards regression model to estimate the hazard ratios and their 95 % CIs. For all regression models, a P \ 0.05 with a CI of 95 % was considered significant. All of the analyses were performed with the Statistical Package for the Social Sciences (SPSS, Chicago, Illinois).
Results Among the 214 patients evaluated, the incidence of spinal deformities was 44 of 214 (21 %) (Table 1). The mean Cobb angle was 28.9° (13–38°). There were 29 single and 9
Eur Spine J (2015) 24:203–208
205
Table 1 Descriptive summary of patient cohort No. of patient
214
Age
58.3 ± 17.6 (20–86)
Gender (F:M)
49:165
Follow-up (years)
1.4 ± 1.1 (0.5–8.5)
Etiology SCI without bone lesion
124
SCI with bone lesion
34
OPLL/OYL
22
Pyogenic Spondylitis/Potts
17
Others
17
Level of injury Cervical
152
Proximal thoracic
8
Middle thoracic
21
Distal thoracic ASIA grade
33
A
26
B
54
C
96
D
42
SCI with scoliosis
44
SCI without scoliosis
170
Mean and SD. Range in parenthesis
Table 2 Descriptive summary of SCI patient with scoliosis No. of patient
44
Age
58.1 ± 17.8 (19–84)
Gender (F:M)
11:33
Term (years)
1.5 ± 0.9 (1–6)
Cobb angle (degrees) Curve direction (RT:LT)
18.1?/4.9 (13–28) 32:12
Curve type (single:double)
33:11
Curve progression (5°)
38/44
Range in parenthesis
double curves. Among the 29 single curves, 22 patients had apes at thoracic lesion and 7 patients had apex at thoracolumbar lesion. Among the group of patients with spinal deformities, the average age was 58.1 years. Twenty-eight patients had a major curve directed to the right, and 10 had a major curve to the left. Thirty-nine of 44 (89 %) patients who developed spinal deformities during follow-up did so within the first 3 months post-injury (Table 2). In 38 of these 44 patients, the curve progressed more than 5° during follow-up (Figs. 1, 2). One patient required additional corrective spinal surgery who had ASIA grade B SCI at C5 level and developed post-SCI scoliosis Fig. 3. Our univariate analysis suggested that there are many risk factors associated with developing spinal deformities,
including age, ASIA grade, OPLL/OYL, and surgery, but only ASIA grade and surgery were statistically significant (ASIA grade, OR: 1.59 [95 % CI: 1.04–2.44; P = 0.032]; Surgery, OR: 4.47 [95 % CI: 1.89–10.06; P = 0.0007]; Table 3). Our multivariate analysis confirmed that these two risk factors were independently associated with the development of spinal deformities (Table 4). ASIA grade and surgery were the most significant independent risk factors for developing a spinal deformity (ASIA grade, OR: 1.63 [95 % CI: 1.04–2.57; P = 0.033]; Surgery, OR: 4.59 [95 % CI: 1.91–11.04; P = 0.0006]; Table 4). Interestingly, the etiology of the injury, level of injury, age, and gender were not independent risk factors according to our multivariate analysis. Having surgery was the most important risk factor (HR: 3.50 [95 % CI: 1.56–7.88; P = 0.0025]) in our provisional hazard model. The patients who underwent surgery were 3.5 times more likely to develop a spinal deformity (Fig. 1). ASIA grade was a non-significant risk factor in our provisional hazard model (Table 5).
Discussion In the present study, we found that the incidence of postSCI spinal deformities in an adult population was 21 %. The incidence of post-SCI spinal deformities has been reported in previous publications [1–9] Previous studies have reported a wide range (50–100 %) for the incidence of post-SCI spinal deformities in pediatric patients [1, 2, 4, 5]. However, these reports address only adolescent populations [1, 2, 4, 5]. To the best of our knowledge, there are no studies of post-SCI spine deformity in adult populations. The risk factors for post-SCI spinal deformities include a younger age at the onset of injury, higher ASIA grade, and spasticity [1, 2, 4, 5, 7]. Some spinal deformities have a multi-factorial etiology that can be related to patient-oriented factors [1–3, 5]. Mayfield et al. reported that the overall incidence of post-SCI spinal deformities was 64 % in a pediatric population, and most of these children required spinal fusion to maintain a seated posture [11]. In contrast, Mehta et al. evaluated 123 pediatric SCI patients and demonstrated that bracing patients with spinal curvature of less than 10 degrees prevented the need for surgical intervention. However, the relative risk for post-SCI spinal deformities associated with clinical factors has never been quantified. In the present study, we identified two potential risk factors, ASIA grade and surgery. Our finding is particularly important because previous studies did not include multivariate analyses of data, and it is possible that there are confounding variables associated with the risk factors that have been reported. The incidence of post-SCI spinal deformities in our sample was limited to the incidence among adult patients.
123
206
Eur Spine J (2015) 24:203–208
Fig. 1 Forty nine year-old male had ASIA grade B SCI due to burst fracture at C5 and had a combined anterior/posterior spine fusion. The scoliosis significantly progressed from 8 degrees to 27°. [a Immediate SCI Cobb angle 8°, b 2 months after SCI Cobb angle 20°, c 6 months after SCI Cobb angle 27°]
A
B
C
8
20
P.I. 2 month 16
P.I. 1 month
A
27
P.I. 321 month
B
C
8
P.I. 1 month
21
16
P.I. 2 month
31
P.I. 6 month
Fig. 2 30 year-old male had ASIA grade C SCI due to burst fracture at T8. The scoliosis significantly progressed from 8° to 31°. [a Immediate SCI Cobb angle 8°, b 2 months after SCI Cobb angle 21°, c 6 months after SCI Cobb angle 31°]
A post-SCI spinal deformity is a complication that can occur in any age after an SCI; therefore, we cannot report that our incidence rate is accurate for all ages. Our incidence rate of post-SCI spinal deformities (21 %) is lower than the rate given in previous reports. Previous studies focused on the incidence of post-SCI spinal deformities in pediatric populations, and most of the patients studied in previous research had complete paralysis or tetraplegia. In addition, the etiology of injury included in most of the previous reports varied from the compressive spinal cord injury to spinal infarction. This heterogeneity in the
123
population led to diverse types of scoliosis and weakened the studies’ conclusions. We recognize the limitations of the present study. First, our retrospective study design precludes us from making any strong conclusions about the independent risk factors we identified. However, our multivariate logistic analysis was conducted with a consecutive patient sample, which is the most feasible method for identifying and quantifying risk factors in clinical research when randomized controlled trials are not possible. Second, the follow-up period varied from 1 year to 5 years. In the present study, we
Eur Spine J (2015) 24:203–208
207 Table 5 Hazard ratio for scoliosis after SCI
1.00
Valuables
P value
Hazard ratio
Scoliosis
0.80
95 % CI Lower limit
Model 1
0.60
Model 2
0.40
Upper limit
Surgery
0.0025*
3.5033
1.5566
7.8849
ASIA grade
0.0623
1.4422
0.9813
2.1196
* Indicated statistically significant
0.20
0.00 0
1
2
3
4
5
6
Yrs
Fig. 3 The graph shows a cumulative hazard plot for the development of post-SCI scoliosis with stratification based on ASIA grade and application of surgery. According to this model, those with higher ASIA grade and had a surgery have a higher risk at all time points for developing post-SCI scoliosis than those with lower ASIA grade and did not have a surgery. Model 1: Patient with ASIA D SCI and did not have a surgery. Model 2: Patient with ASIA A SCI and had a surgery
Table 3 Uni-variate predictors of scoliosis after SCI Valuables
P value
Odds ratio
95 % CI Lower limit
Upper Limit
Age
0.3700
0.9916
0.9735
Gender (female)
0.7097
1.1579
0.5351
1.0101 2.5055
ASIA grade
0.0322*
1.5936
1.0403
2.4413
Level of SCI
0.5764
1.0800
0.8245
1.4148
SCI without bone lesion
0.6559
1.1669
0.5919
2.3005
SCI with bone lesion
0.6409
1.2308
0.5143
2.9452
OPLL/OYL
0.4012
0.5815
0.1640
2.0616
Spondylitis
0.8523
0.8837
0.2404
3.2476
Others Surgery
0.7571 0.0007*
0.8153 4.4725
0.2236 1.8870
2.9724 10.6007
* Indicated statistically significant
Table 4 Adjusted odds ratio for scoliosis after SCI Valuables
P value
Odds ratio
95 % CI Lower limit
Upper limit
OPLL/OYL
0.5331
0.7324
0.2751
1.9499
ASIA Grade Surgery
0.0331* 0.0006*
1.6344 4.5966
1.0403 1.9136
2.5679 11.0413
* Indicated statistically significant
found that 90 % of spinal deformities occurred within 3 months after SCIs, and 86 % progressed significantly. This result suggests that a minimum of 1 year is acceptable
for measuring the incidence of spinal deformities after SCIs. Moreover, our provisional hazard model was based on our multivariate analysis from a relatively large sample of patients, which is often the most feasible method for identifying and quantifying risk factors in the medical literature when randomized controlled trials are not possible. To our knowledge, there are no studies currently available in the literature that addresses the differences in the deformities of adult post-SCI conditions. We have identified two independent risk factors, ASIA grade and surgery, for post-SCI spinal deformities in a uniform adult population. The wide variety of ASIA grades observed in the present study and the differences in the ages of the patients in the cohort may have played a role in the lower incidence of post-SCI spinal deformities observed in the present study. Previous studies have identified non-independent risk factors associated with post-SCI spinal deformities in adolescent populations [1–7]. In the present study, we identified multiple factors, including ASIA grade and surgery (Table 4). Both of these factors were independently associated with post-SCI spinal deformities. Our finding is particularly important because previous studies did not perform multivariate analyses, and it is possible that there were confounding variables associated with the risk factors identified in those studies. To date, there is no study that provides a quantitative assessment of risk factors for the development of post-SCI spinal deformities based on independent variables. Surgery was specifically identified in our Cox proportional hazards model. This result may be explained by the relationship that many of the risk factors, including ASIA grade and surgery, have with each other. One could argue that ASIA grade and surgery are inherently related. However, we did not find evidence of this relationship in our data analysis, suggesting that the ASIA grade and surgery independently increased the risk of post-SCI spinal deformities in the adult patients in the present study. We do not have a clear understanding of the effect that surgery has on the development of post-SCI spinal deformities, but one possible explanation exists. The comparisons of the incidence of post-SCI spinal deformities between decompression alone and decompression and fusion indicated no statistical difference (Table 6). Intuitively, this risk factor could increase the stress concentration at the proximal and distal
123
208
Eur Spine J (2015) 24:203–208
Table 6 The incidence of scoliosis in surgically treated patients with SCI Decompression alone
Decompression and fusion
Total
Scoliosis
14
18
32
Normal
31
72
103
Total
45
90
135
P value 0.24
end of the fusion, theoretically increasing the risk for developing a junctional deformity [12–14]. A recent study also reported that post-surgical trunk imbalance after spine surgery affects iatrogenic spinal deformity, but this finding has not been uniform in studies assessing the risk factors for post-SCI spinal deformities [12–14]. Our findings suggest that the etiology of post-SCI spinal deformities is uncertain in adult populations. Various studies have proposed that the development of spinal deformities is related to a combination of age at onset, spasticity, severity of the SCI, and other factors [1–7]. Of these contributing factors, age at onset, skeletal maturity, the severity of the SCI, and injury level have all been implicated in the development of post-SCI spinal deformities. Although these results may be accurate, our analysis only suggests that ASIA grade and surgery are associated with post-SCI spinal deformities. While we identified risk factors similar to those reported in the literature, they did not independently predict post-SCI spinal deformities. It remains unclear why surgery had an impact on the development of post-SCI spinal deformities, but our data suggest that patients with severe SCIs who undergo surgery should be monitored closely for the development of post-SCI spinal deformities.
Conclusion SCI patients with higher ASIA grades who underwent surgery had a higher likelihood of developing spinal deformities. Of these risk factors, surgery was the stronger risk factor.
123
Conflict of interest
None.
References 1. Bergstro¨m EM, Short DJ, ASIA HL, Henderson NJ, Jones PR (1999) The effect of childhood spinal cord injury on skeletal development: a retrospective study. Spinal Cord 37(12):838–846 2. Dearolf WW 3rd, Betz RR, Vogel LC, Levin J, Clancy M, Steel HH (1990) Scoliosis in pediatric spinal cord-injured patients. J Pediatr Orthop 10(2):214–218 3. Denis F, Armstrong GW, Searls K, Matta L (1984) Acute thoracolumbar burst fractures in the absence of neurologic deficit. A comparison between operative and nonoperative treatment. Clin Orthop Relat Res 189:142–149 4. Hamilton MG, Myles ST (1992) Pediatric spinal injury: review of 61 deaths. J Neurosurg 77(5):705–708 5. Lancourt JE, Dickson JH, Carter RE (1981) Paralytic spinal deformity following traumatic spinal-cord injury in children and adolescents. J Bone Joint Surg Am 63(1):47–53 6. Malcolm BW, Bradford DS, Winter RB, Chou SN (1981) Posttraumatic kyphosis. A review of forty-eight surgically treated patients. J Bone Joint Surg Am 63(6):891–899 7. Malcolm BW (1979) Spinal deformity secondary to spinal injury. Orthop Clin North Am 10(4):943–952 8. Mehta S, Betz RR, Mulcahey MJ, McDonald C, Vogel LC, Anderson C (2004) Effect of bracing on paralytic scoliosis secondary to spinal cord injury. J Spinal Cord Med 27:S88–S92 9. Vaccaro AR, Silber JS (2001) Post-traumatic spinal deformity. Spine 26:S111–S118 (Review) 10. Marino RJ, Barros T, Biering-Sorensen F et al (2003) ASIA Neurological Standards Committee 2002. International standards for neurological classification of spinal cord injury. J Spinal Cord Med 26:S50–S56 11. Mayfield JK, Erkkila JC, Winter RB (1981) Spine deformity subsequent to acquired childhood spinal cord injury. J Bone Joint Surg Am 63(9):1401–1411 12. Kim YJ, Bridwell KH, Lenke LG, Glattes CR, Rhim S, Cheh G (2008) Proximal junctional kyphosis in adult spinal deformity after segmental posterior spinal instrumentation and fusion: minimum five-year follow-up. Spine 33(20):2179–2184 13. Parent S, Dimar J, Dekutoski M, Roy-Beaudry M (2010) Unique features of pediatric spinal cord injury. Spine 35:S202–S208 14. Arun R, Srinivas S, Mehdian SM (2010) Scoliosis in Duchenne’s muscular dystrophy: a changing trend in surgical management: a historical surgical outcome study comparing sublaminar, hybrid and pedicle screw instrumentation systems. Eur Spine J 19(3):376–383
