Epidemiology of traumatic spinal cord injury in Tianjin, China: An 18-year retrospective study of 735 cases

Abstract

Study Design: Hospital-based retrospective study

Objectives: To evaluate the pathogenetic features of traumatic spinal cord injury (TSCI) during 1999–2016 according to changed injury etiology with time, explore different characteristics of patients suffered a TSCI during 1999–2007 and 2008–2016 in Tianjin, China.

Setting: Tianjin Medical University General Hospital

Methods: In this study, the medical records of TSCI patients were obtained from Tianjin Medical University General Hospital (TMUGH) from 1st January 1999 to 31th December 2016. Variables were recorded, including age, gender occupation, etiology, the level of injury, America Spinal Injury Association (ASIA) impairment scale, the severity, concomitant injuries, death and its cause. To explore the differences in characteristics by etiology and by two periods, related statistical methods were used to calculate the correlation of some variables. Differences in etiology of TSCI during 1999–2016 were evaluated and differences in epidemiological characteristics were separately compared and analyzed between the 1999–2007 period and the 2008–2016 period.

Results: From 1999–2016, 831 TSCI cases were identified and 96 cases were excluded from analyses. The male-to-female ratio was 2.9:1 and the mean age was 49.7±15.2 years, which changed significantly between 1999–2007 (45.1±14.2) and 2008–2016 (51.6±15.2). Traffic accidents (45.8%) were the leading cause of TSCI during the 1999–2007 period, followed by low falls (30.7%). However, the opposite result was observed during the 2008–2016 period. Significant difference was observed compared with thoracic, lumbar and sacral levels, cervical level was the most commonly affected levels and the percentage decreased to a certain degree between 1999–2007 and 2008–2016 (from 84.4% to 68.9%). The proportions of ASIA grades A, B, C, and D were 20.5%, 10.3%, 23.3%, and 45.9%, respectively. The percentage of complete tetraplegia decreased from 22.9% in 1999–2007 to 13.2% in 2008–2016, and the percentage of incomplete paraplegia increased from 9.7% to 27.9%.

Conclusion: According to the changes in the epidemiological characteristics of TSCI, relevant health service, laws and regulations, preventative strategies should be readjusted to follow up the changing situation and epidemiological characteristics of TSCI.

Introduction

Traumatic spinal cord injury (TSCI) is known worldwide as a life-disrupting condition that leads to variable degrees of motor, sensory and autonomic deficits. It causes huge burden not only for patients and their families but also for society as a whole.^1–3 Currently, drug treatments, surgery and cell transplantation strategies are promising treatments for spinal cord injury, but there are no regenerative treatments clinically available that provide pre-injury life satisfaction.^4–6 Under these circumstance, primary and secondary TSCI preventions are important. In the past 40 years, a great deal of studies about TSCI epidemiology have been published, and researchers worldwide have learnt much more about the characteristics of TSCI. However, in several recent studies, researchers have focused on the changing trends of TSCI over time and some different characteristics were found. In Galicia, Spain, changes in incidence, the mean age and cervical levels were observed from 1995–2014.⁷ In Scotland, the demographic profiles in TSCI were subject to change from 1994–2013.⁸ The changing demographics of TSCI were reported in Québec, Canada from 2000–2011.⁹ In Spain, a similar changes in etiology and average age of patients was reported in Valencia from 2001–2013.¹⁰ In view of the above-mentioned facts, reliable, objective and valid data on basic epidemiological research is imperative in developing prevention strategies of TSCI.

Tianjin is one of the earliest municipalities in China, with a considerably high population density, and the incidence of TSCI in Tianjin has increased year by year. As the international scientific and technology cooperation base of Spinal Cord Injury and the largest general medical center in Tianjin, Tianjin Medical University General Hospital –(TMUGH)- is one of the few hospitals that can provide proper treatment for patients with TSCI. Most TSCI patients are admitted to TMUGH, and patient information is documented in detail. Although Feng et al. and Zhou et al. have reported the epidemiological profile of TSCI in Tianjin, China, from 1998–2009 and from 2009–2014, respectively,^11,12 a complete epidemiological description about the recent two decades is also meaningful. The purposes of this study were to evaluate the characteristics of TSCI patients by injury etiology from 1999–2016, and differences in the characteristics of patients who sustained a TSCI during1999–2007 and 2008–2016 in Tianjin, China were also determined.

Materials and methods

Study setting

A hospital-based retrospective study from 1999 to 2016 was carried out. The International Classification of Diseases Version 10 (ICD-10), diagnostic code of T09.302, was used in our hospital to represent the injury during this period. According to this standard, the medical records of patients were included: Persons with TSCI that were admitted to TMUGH between 1st January 1999 and 31th December 2016. Some of the medical records matching the following criteria were excluded: (i) Less than 15 years old (ii) Patients with ASIA grade E ; and (iii) Incomplete medical records or uncertain diagnosis.

Study characteristics

In this study, characteristics of each patient were recorded, including: age, gender, occupation, etiology, the level of injury, America Spinal Injury Association (ASIA) grade, severity, concomitant injuries, death and its cause. Persons were categories into 5 groups based on their age: 15–29, 30–44, 45–59, 60–74, and >= 75 years. Occupations included worker, peasant, office clerk, driver, unemployed, retired, and other occupation (civil servant, seaman, teacher, student and doctor). The etiology of injury included traffic accidents, falls (including low fall: < 1 m and high fall: ≥ 1 m), struck by falling objects, collision of the head against objects, injuries involving machinery, sports, roller coasters, massage, and injuries caused by another person. Because the number of some etiologies was not significant, we recorded collision of the head against objects, injuries involving machinery, sports, roller coaster, massage and injuries caused by another person in one category as other traumatic cause. The neurological level of injury was classified as cervical, thoracic, lumbar and sacral. The period of 18 years was divided into six sub-periods: 1999–2001, 2002–2004, 2005–2007, 2008–2010, 2011–2013 and 2014–2016; it was then subdivided into 1999–2007 and 2008–2016 for convenient comparison. According to the 2011 revision of the International Standards for Neurological Classification of Spinal Cord Injury, the neurological functions below injury site were evaluated by the ASIA grade, and the severity was also classified as complete or incomplete tetraplegia and complete or incomplete paraplegia according to the ASIA scale.¹³ Concomitant injuries were categorized as spine fracture or dislocation, limb fracture rib fracture, brain or head injuries and fracture of the pelvis. Patients were categorized as having or not having a concomitant injury.

Statistical analysis

SPSS 19.0 (IBM, Armonk, NY, USA) and Microsoft Excel (Microsoft Corporation, Redmond, WA, USA) were used by two researchers to manage and analyze the data. The characteristics listed in Tables were expressed as numbers or mean value ± s.d. To explore the differences in characteristics by etiology and by two periods, a descriptive study of relevant variables was carried out. One-way analysis of variance was used to calculate continuous variables (age) between-group, and categorical variables (gender, level of injury, ASIA scale, severity, concomitant injury, occupation, and etiology) were calculated with the chi-square test between-group. Bonferroni method and the chi-square method were used for the within-group comparisons. Kolmogorov–Smirnov test was used to calculate the normality of the variables (age) and t test was used to compare the differences between the two periods. A p value < 0.05 was defined as statistically significant.

Results

Characteristics of the cases

From 1999–2016, 831 TSCI cases were identified and 96 were excluded from analyses (Less than 15 years: 55; ASIA grade E: 29; Incomplete medical records or uncertain diagnosis: 13). As shown in Table 1, the number of TSCI patients increased rapidly from 1999 to 2016. The male-to-female ratio was 2.9:1 and the mean age was 49.7±15.2 years. The occupation characteristics with percentages are as follows: peasant, worker, unemployed, retired, office-clerks, driver and other occupation were 26.0%, 21.0%, 27.1%, 12.8%, 5.6%, 3.5% and 4.0%, respectively. Falls were the most common cause and traffic accidents were the second most frequent cause. Cervical level was the most commonly affected segments, with a percentage of 72.4%. Regarding the proportions of ASIA grade, grade A was 20.5%, grade B was 10.3%, grade C was 23.3%, and grade D was 45.9%. For the severity of injury, the most prevalent category of TSCI was incomplete tetraplegia. A total of 13 patients died during hospitalization: 5 of them died from multiple organ failure, 5 patients died from respiratory failure, 2 patients died from cardiac failure and 1 patient died from acute pulmonary embolism.

Table 1. Characteristics of patients with TSCI in Tianjin in six periods studied.

Variables	1999–2001	2002–2004	2005–2007	2008–2010	2011–2013	2014–2016	Overall
Age
15–29	9	8	6	17	20	22	82
30–44	21	23	15	31	40	49	179
45–59	20	22	20	50	78	98	288
60–74	4	4	8	30	37	62	145
≥75	1	1	4	3	9	23	41
Mean age±s.d.	44.0±13.9	42.5±12.4	49.2±15.7	48.9±15.5	50.0±15.0	53.1±15.1	49.7±15.2
Gender
Male	46	44	43	104	126	184	547
Female	9	14	10	27	58	70	188
Occupation
Peasant	15	9	10	34	58	65	191
Worker	9	6	10	26	50	53	154
Unemployed	24	38	20	31	31	55	199
retired	2	1	3	17	19	52	94
office-clerks	0	0	1	6	16	18	41
driver	2	2	3	7	4	8	26
Other occupation	3	2	6	7	6	6	30
Etiology
Traffic accidents	26	27	23	47	75	73	271
Falls	26	22	27	75	92	164	406
Struck by falling objects	3	6	1	3	5	11	29
Other traumatic cause	0	3	2	6	12	6	29
Level of injury
cervical	46	50	44	97	127	168	532
Thoracic	1	7	5	19	30	43	105
Lumbar and sacral	8	1	4	15	27	43	98
ASIA scale
A	19	19	16	30	40	27	151
B	9	7	5	12	25	18	76
C	12	6	9	30	32	82	171
D	15	26	23	59	87	127	337
Severity
Complete tetraplegia	14	14	10	22	27	26	113
Incomplete tetraplegia	19	36	38	78	100	115	386
Complete paraplegia	10	5	4	10	13	19	61
Incomplete paraplegia	12	3	1	21	44	94	175

ASIA, American Spinal Injury Association; s.d., Standard Deviation; TSCI, traumatic spinal cord injury.

Differences in characteristics by etiology

As Table 2 shows, statistically significant differences in age of TSCI patients were found in the etiology (P < 0.001). Traffic accidents, low falls and other traumatic causes were the main reasons for TSCI in the 45–59 group. In terms of low falls, these usually occurred in the older group (>= 75 years). No significant difference was observed between sex and etiology (P = 0.052). A statistically significant difference in the level of injury was observed in the etiology (P < 0.001). There was obvious difference compared with thoracic, lumbar and sacral levels, cervical level was the most commonly affected, with a percentage of 72.4%.

Table 2. Characteristics of patients with TSCI in Tianjin according to etiology (1999–2016).

Variables	Traffic accidents	Low falls	High falls	Struck by falling objects	Other traumatic cause	P-value
Age						<0.001
15–29	34	8	28	5	7
30–44	81	28	53	13	4
45–59 ab	106	114	48	7	13
60–74 abc	42	87	7	4	4
≥75 abc	8	31	2	0	1
Mean age±s.d.	41.2±15.6	58.5±16.0	35±14	24±0	37±18.0	0.003
Overall	271	268	138	29	29
Gender						0.052
Male	195	199	111	25	17
Female	76	69	27	4	12
Overall	271	268	138	29	29
Level of injury						<0.001
Cervical	209	229	53	18	23
Thoracic a	31	24	42	5	3
Lumbar and sacral a	31	15	43	6	3
Overall	271	268	138	29	29
ASIA scale						0.004
A	67	25	44	8	7
B a	30	32	12	0	2
C a	52	79	29	7	4
D a	122	132	53	14	16
Overall	271	268	138	29	29
Severity						<0.001
Complete tetraplegia	57	24	24	3	5
Incomplete tetraplegia a	145	184	27	11	19
Complete paraplegia ab	20	7	28	4	2
Incomplete paraplegiaab	49	53	59	11	3
Overall	271	268	138	29	29
Concomitant injury						<0.001
Having	218	61	73	13	10
Not having	53	207	65	16	19

ASIA, American Spinal Injury Association; TSCI, traumatic spinal cord injury.

Superscript symbols a, b, c, d, and e represent the first, second, third, fourth, and fifth groups respectively; There were no statistical differences between subgroups when no superscripts were marked.

Superscript symbols a, b, c, d, and e marked on each item means statistical differences were observed when comparing with different groups.

Compared with the first group, ^aP < 0.05; Compared with the second group, ^bP < 0.05; Compared with the third group, ^cP < 0.05; Compared with the forth group, ^dP < 0.05; Compared with the fifth group, ^eP < 0.05.

Statistically difference was observed according to ASIA grade (P = 0.004) and severity (P < 0.001) classified by etiology. ASIA grade D was the most widespread grade for all the etiologies of TSCI. Moreover, there was obvious difference compared with other subgroups, incomplete tetraplegia was usually caused by traffic accidents, low falls and other traumas. High falls were the main etiology of incomplete paraplegia. Statistically significant differences of TSCI patients were easily observed between the etiology and concomitant injury (P < 0.001). Patients, who got injured by traffic accidents, were usually associated with concomitant injury, accounting for 80.4%. However, the percentage of concomitant injury caused by low falls was very low.

Differences in characteristics by two periods

From Table 3, differences in age were found by two periods (P < 0.001). The percentage of the patients in the 60–74 group increased rapidly from 9.6% in 1999–2007 to 22.7% in 2008–2016, while the corresponding percentage in the 30–44 group decreased from approximately 35.6% to 21.1% during the same period. The mean age changed significantly between 1999–2007 (45.1 ±14.2) and 2008–2016 (51.6±15.2) (P < 0.001). A statistically significant difference in the occupation of TSCI patients was observed (P < 0.001). The number of retired people and office-clerks increased remarkably compared with other occupations (3.6% in 1999–2007 and 15.5% in 2008–2016, 0.6% in the former and 7.0% in the latter). However, the overall rate of unemployed people decreased from 49.4% to 20.6%. A statistically difference in the etiology was also observed in two periods (P = 0.021). Traffic accidents (45.8%) were the leading cause of TSCI during the 1999–2007 period, followed by low falls (30.7%). However, the result was the opposite during the 2008–2016 period.

Table 3. Characteristics of TSCI in Tianjin in two periods studied.

Variables	1999–2007	2008–2016	Overall	P-value
Age				<0.001
15–29	23	59	82
30–44	59	120	179
45–59 b	62	226	288
60–74 abc	16	129	145
≥75 b	6	35	4
Mean age±s.d.	45.1±14.2	51.6±15.2	49.7±15.2	<0.001
Gender				0.056
Male	133	414	547
Female	33	155	188
Occupation				<0.001
Peasant	34	157	191
Worker	25	129	154
Unemployed ab	82	117	199
Retired abc	6	88	94
office-clerk abc	1	40	41
Driver de	7	19	26
Other occupation abde	11	19	30
Etiology				0.021
Traffic accidents	76	195	271
Low fall a	51	217	268
High fall a	24	114	138
Struck by falling objects c	10	19	29
Other traumatic cause	5	24	29
Level of injury				<0.001
Cervical	140	392	532
Thoracic a	13	92	105
Lumbar and sacral a	13	85	98
ASIA scale				<0.001
A	54	97	151
B	21	55	76
C ab	27	144	171
D a	64	273	337
Severity				<0.001
Complete tetraplegia	38	75	113
Incomplete tetraplegia a	93	293	386
Complete paraplegia	19	42	61
Incomplete paraplegia abc	16	159	175

ASIA, American Spinal Injury Association; s.d., Standard Deviation; TSCI, traumatic spinal cord injury.

Superscript symbols a, b, c, d, and e represent the first, second, third, fourth, and fifth groups respectively.

There were no statistical differences between subgroups when no superscripts were marked; Superscript symbols a, b, c, d, and e marked on each item means statistical differences were observed when comparing with different groups.

Compared with the first group, ^aP < 0.05; Compared with the second group , ^bP < 0.05; Compared with the third group, ^cP < 0.05; Compared with the forth group, ^dP < 0.05; Compared with the fifth group, ^eP < 0.05.

Statistically significant difference was observed between level of injury and two periods (P < 0.001). The ratio of the cervical level patients has been decreasing since 2008, while the ratio of other levels increased. A similar statistically significant difference in the ASIA grade was found between 1999–2007 and 2008–2016 (P < 0.001), where the percentage of grades A and B had decreased over time (32.5% and12.7% in 1999–2007, and 17.0% and 9.7% in 2008–2016, respectively). In contrast, the percentage of grades C and D had increased (16.2% and 38.6% in 1999–2007, and 25.3% and 48.0% in 2008–2016, respectively). The difference in severity between 1999–2007 and 2008–2016 were found (P < 0.001). Complete tetraplegia and incomplete paraplegia are the most representative in the group. Furthermore, the percentage of complete tetraplegia decreased from 22.9% to 13.2%, and the percentage of incomplete paraplegia increased from 9.7% to 27.9%.

Discussion

This study is a hospital-based retrospective research study that described the epidemiological differences and change trends of TSCI in Tianjin, China, from 1999 to 2016. The relevant epidemiological characteristics of TSCI in Tianjin, China were reported by our research team.^11,12,14,15 However, those articles only reported a general epidemiological profile of TSCI over a short period of time. This study described the epidemiological characteristics of 18 years from 1999 and compared the epidemiological changes between the 1999–2007 and 2008–2016 periods using 735 cases. As the newest and most comprehensive analysis of epidemiological characteristics in Tianjin, China. The results can provide guidance for precautions and treatment for TSCI in Tianjin, China.

Among all patients, the male-to-female ratio was 2.9:1, which was similar to the distribution in other countries (Spain, Iceland, Austria),^7,16,17 and the change of gender ratio was not obvious between the 1999–2007 and 2008–2016 periods. This result indicates that men remain the main bearers of high-risk outdoor occupations and provide financial support for their families. In contrast, women were mostly engaged in domestic work and outdoor occupations with low risk of injury.¹⁸ With regard to the age variation, patients 30–59 years old were represented most, accounting for 58.9%. The mean age was 49.7±15.2 years, and it had increased significantly from the period of 1999–2007 to the period of 2008–2016, which was similar to other reports.^7,19,20 However, the increase of the mean age of the TSCI patients, especially in the 60–74 year-old group, was consistent with previous studies. With the rapid aging of Chinese society, more elderly people are at risk of getting injured, especially from low falls, which became the leading cause of spinal cord injury for the period of 2008–2016, rather than traffic accidents for the period of 1999–2007, this shift in cause also contributed to the increase of the mean age at the time of injury. It was also observed that compared with the elderly, young and middle-aged people are more likely to become injured in traffic accidents. The 0–14-year-old group was excluded in this study.

Regarding the occupations of TSCI patients, the unemployed, peasants and workers are the most susceptible populations. A remarkable growth in the proportion of retired people and office-clerks was observed in this research between the 1999–2007 period and the 2008–2016 period. Moreover, the overall rate of unemployed people decreased from 49.4% to 20.6%. The change in occupations of patients could be interpreted as: (1) With the economic development and the modernization of society in China, increasing numbers of unemployed people have begun to engage in a job. (2) The aging society and prolonged life expectancy have increased the proportion of retirees.

During the investigation, falls appeared to be the most common cause of TSCI, comprising up to 55.2% of all patients, followed by traffic accidents. An obvious statistically significant difference in the etiology was found, in that traffic accidents (45.8%) were the leading cause of TSCI during the 1999–2007 period, followed by low falls (30.7%). However, the opposite result was observed during the 2008–2016 period. Montoto-Marques et al.⁷ and McCaughey et al.⁸ reported a similar finding to our study in which the percentage of falls increased distinctly in different periods. The decrease in traffic accidents was consistent with studies in other countries.^7,19,21 In addition, low falls were the predominant cause among the old aged group, ²¹ while high falls and traffic accidents were more likely to occur in youth.²² This could be explained because, compared with the past, old people are more willing to go outside to enrich their lives rather than staying at home after retirement. Another possible state that the increasing incidence of TSCI among elder people was the degenerative joint disease and prevalence of osteoporosis among them.^23,24 This discrepancy could be partly explained by traffic safety legislation, management and people’s increased safety awareness. This result revealed that more attention should be paid to the aged, especially the ones living alone. Infrastructure construction should be improved by the government to ensure safety while walking and strict legislation and regulation should be continued to decrease the incidence of traffic accidents.

Changes in injury level of TSCI in two different periods were observed. The ratio of the cervical level has been decreasing since 2008, whereas other levels increased. The cervical level is the most vulnerable part of the spine, according to this study, which was consistent with the investigations in many other countries.^7,16,17,20 The high proportion of cervical TSCI might be attributed to the peculiar structures and functions of the cervical spine. In addition, the preventive measures for TSCI in Tianjin were not as sufficient as in other developed countries before 2010, which also increases the number of cervical level injuries. With the increasing number of the elderly, more TSCI at the thoracic or lumbar and sacral levels occurred because more patients suffered degenerative diseases of thoracic, lumbar and sacral levels, as well as low falls. These factors could help to explain the change in injury level between the 1999–2007 period and the 2008–2016 period. Moreover, the degree of severity also changed between the 1999–2007 period and the 2008–2016 period, more incomplete paraplegia was observed instead of complete tetraplegia, which may be the result of increasing incidence of low falls among the elderly. ^25,26

Our study highlighted that, a similar statistically significant difference in the ASIA grade was found between 1999–2007 and 2008–2016, where the percentage of grades A and B decreased and grades C and D increased. ASIA grade D was most commonly observed in all TSCI patients, regardless of etiology. The ever-increasing proportion of ASIA grade D in patients demonstrated that the condition of TSCI patients is not as severe as before due to changed etiologies, injury level and improved treatment measures in recent years. The reasons for the low mortality in this study can be listed as follows: (1) patients who died on the way to hospital were not included; (2) According to Chinese traditional customs, patients who could not avoid death usually went home from the hospital to spend their last time with their families.

Limitations also existed in our study. First, there was no electronic TSCI registry system in our hospital and it was inevitable that some data might be lost. Moreover, some detailed information about occupation and etiology were not obtained. Second, this research focused only on survivors of TSCI, while patients who died on the way to the hospital were excluded. Third, a single-center descriptive study cannot fully estimate the incidence rate of TSCI in Tianjin, China. Fourth, patients younger than 14 years old were not included. Fifth, some groups had low numbers of person.

Conclusions

The epidemiological characteristics of TSCI in Tianjin have changed significantly during this period. People with TSCI are getting older in Tianjin, China and people older than 60 years or retired people became a new vulnerable group. Along with age-related changes in the epidemiological characteristics of TSCI, falls became the most significant cause. ASIA grade D became the most common grade in TSCI patients regardless of etiology, and incomplete paraplegia was the most relevant changes in severity. According to the changes in the epidemiological characteristics of TSCI, more attention should be paid to the aged, especially the ones living alone, and assistant facilities should be set up in public venues to prevent falls. Relevant health service, laws and regulations, preventative strategies also should be readjusted to follow up the changing situation and epidemiological characteristics of TSCI.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Funding Statement

This work was supported by National Natural Science Foundation of China (81330042), Ministry of Science and Technology of the People Republic of China (2014DFR31210), National Natural Science Foundation of China (81620108018), Tianjin Science and Technology Committee, China (14ZCZDSY00044, 13RCGFSY19000), National Natural Science Foundation of China (81472070), and National Science Foundation of China (81501889).

Acknowledgements

We are deeply grateful to the colleagues of Tianjin Medical University General Hospital’s medical record statistics room for their supporting of collection the medic record.

Disclaimer statements

Contributors None.

Conflict of interest None.

Ethics approval None.

Declaration of interest The authors report no declarations of interest.