Progression, incidence, and risk factors for intervertebral disc degeneration in a longitudinal population-based cohort: the Wakayama Spine Study

Accepted Manuscript Progression, incidence, and risk factors for intervertebral disc degeneration in a longitudinal population-based cohort: the Wakayama Spine Study Masatoshi Teraguchi, MD, PhD, Noriko Yoshimura, MD, PhD, Hiroshi Hashizume, MD, PhD, Hiroshi Yamada, MD, PhD, Hiroyuki Oka, MD, Akihito Minamide, MD, PhD, Keiji Nagata, MD, PhD, Yuyu Ishimoto, MD, PhD, Ryohei Kagotani, MD, PhD, Hiroshi Kawaguchi, MD, PhD, Sakae Tanaka, MD, PhD, Toru Akune, MD, PhD, Kozo Nakamura, MD, PhD, Shigeyuki Muraki, MD, PhD, Munehito Yoshida, MD, PhD PII:

S1063-4584(17)30007-9

DOI:

10.1016/j.joca.2017.01.001

Reference:

YJOCA 3933

To appear in:

Osteoarthritis and Cartilage

Received Date: 25 April 2016 Revised Date:

23 December 2016

Accepted Date: 4 January 2017

Please cite this article as: Teraguchi M, Yoshimura N, Hashizume H, Yamada H, Oka H, Minamide A, Nagata K, Ishimoto Y, Kagotani R, Kawaguchi H, Tanaka S, Akune T, Nakamura K, Muraki S, Yoshida M, Progression, incidence, and risk factors for intervertebral disc degeneration in a longitudinal population-based cohort: the Wakayama Spine Study, Osteoarthritis and Cartilage (2017), doi: 10.1016/ j.joca.2017.01.001. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT 1

Running Head: Progression, incidence, and risk factors for disc degeneration

2 3

Progression, incidence, and risk factors for intervertebral disc degeneration in a

4

longitudinal population-based cohort: the Wakayama Spine Study

RI PT

5

Masatoshi Teraguchi, MD, PhD1; Noriko Yoshimura, MD, PhD2; Hiroshi Hashizume, MD,

7

PhD1; Hiroshi Yamada, MD, PhD1; Hiroyuki Oka, MD3; Akihito Minamide, MD, PhD1; Keiji

8

Nagata, MD, PhD1; Yuyu Ishimoto, MD, PhD1; Ryohei Kagotani, MD, PhD1; Hiroshi

9

Kawaguchi, MD, PhD5; Sakae Tanaka, MD, PhD4; Toru Akune, MD, PhD6; Kozo Nakamura, MD, PhD6; Shigeyuki Muraki, MD, PhD2; Munehito Yoshida, MD, PhD1

M AN U

10

SC

6

11 1

Department of Orthopaedic Surgery, Wakayama Medical University, Wakayama, Japan

13

2

Department of Joint Disease Research, 22nd Century Medical & Research Center, Faculty of

14

Medicine, The University of Tokyo, Tokyo, Japan

15

3

16

Medical and Research Center, Faculty of Medicine, The University of Tokyo, Tokyo, Japan

17

4

18

Japan

19

5

20

6

21

Tokorozawa City, Saitama, Japan

22

*Corresponding author: Hiroshi Hashizume, MD, PhD

23

Wakayama Medical University

24

811-1 Kimiidera 641, Japan

25

Tel: 81-073-447-2300; Fax: 81-073- 448-3008

TE D

12

Department of Medical Research and Management for Musculoskeletal Pain, 22nd Century

AC C

EP

Department of Orthopaedic Surgery, Faculty of Medicine, The University of Tokyo, Tokyo,

Japan Community Healthcare Organization Tokyo Shinjuku Medical Center, Tokyo, Japan Rehabilitation Services Bureau, National Rehabilitation Center for Persons with Disabilities,

1

ACCEPTED MANUSCRIPT 26

E-mail address: [email protected]

27

Grants and Funding Sources

29

This study was supported by: H23-Choujyu-002 (Director, Toru Akune), H-25-Choujyu-007

30

(Director, Noriko Yoshimura), H25-Nanchitou (Men)-005 (Director, Sakae Tanaka),

31

201417014A (Director, Noriko Yoshimura), and H22-Choujyu-Wakate-007 (Director,

32

Shigeyuki Muraki) from the Ministry of Health, Labour and Welfare;

33

a Grant-in-Aid for Scientific Research (B26293139, B23390172 to Noriko Yoshimura,

34

B2629333, C20591774 to Shigeyuki Muraki, C26462249 to Hiroshi Hashizume, C25462305

35

to Hiroshi Yamada); a Grant-in-Aid for Young Researchers (B25860448 to Yuyu Ishimoto,

36

B26861286 to Masatoshi Teraguchi, B26860419 to Ryohei Kagotani, B15K20013 to Hiroki

37

Iwahashi); a Grant-in-Aid for Challenging Exploratory Research (15K15219 to Noriko

38

Yoshimura, 26670307 to Shigeyuki Muraki, 24659666 to Hiroyuki Oka, 25670293 to Toru

39

Akune) of JSPS KAKENHI grant; a Grant from the Japanese Orthopaedics and Traumatology

40

Foundation, Inc. (No. 287) to Masatoshi Teraguchi; and Collaborating Research with NSF

41

08033011- 00262 (Director, Noriko Yoshimura) from the Ministry of Education, Culture,

42

Sports, Science and Technology in Japan. This study also was supported by grants from the

43

Japan Osteoporosis Society (Noriko Yoshimura, Shigeyuki Muraki, Hiroyuki Oka, and Toru

44

Akune), JA Kyosai Research Institute (Hiroyuki Oka), Mitsui Sumitomo Insurance Welfare

45

Foundation (Shigeyuki Muraki), and research aid from the Japanese Orthopaedic Association

46

(JOA-Subsidized Science Project Research 2006-1 & 2010-2; Director, Hiroshi Kawaguchi).

47

The study sponsors played no role in the study design, collection, analysis, and interpretation

48

of data, writing of the report, or the decision to submit the manuscript for publication. The

49

corresponding author had full access to all the data and had the final decision to submit for

50

publication.

AC C

EP

TE D

M AN U

SC

RI PT

28

2

ACCEPTED MANUSCRIPT 51

Abstract

53

Objective: The present study examined the progression, incidence, and risk factors for

54

intervertebral disc degeneration (DD) throughout the lumbar spine using magnetic resonance

55

imaging (MRI) in a large population-based cohort.

56

Methods: We followed up 617 subjects for more than 4 years as part of the Wakayama Spine

57

Study. 1) “Progression of DD” in each of the entire, upper (L1/2 to L3/4) and lower (L4/5 and

58

L5/S1) lumbar spine was defined as Pfirrmann grade progression at follow-up in at least one

59

disc in the affected region. 2) “Incidence of DD” in each of these regions was defined if all

60

discs were grade 3 or lower (white disc) at baseline, and at least one disc had progressed to

61

grade 4 or higher (black disc) at follow-up. Logistic regression analyses were used to

62

determine the risk factors for progression and incidence of DD.

63

Results: DD progression and incidence in the entire lumbar spine were 52.0% and 31.6% in

64

men, and 60.4% and 44.7 % in women, respectively. Women was associated with DD

65

progression in the upper lumbar spine (odds ratio [OR]=1.68, 95% confidence interval

66

[CI]=1.18–2.42). Aging was associated with the incidence of DD in each region (entire:

67

OR=1.14, CI=1.06–1.14; upper: OR=1.10, CI=1.05–1.15; lower: OR=1.11, CI=1.05–1.19).

68

Diabetes mellitus was associated with the incidence of DD in the upper lumbar spine

69

(OR=6.83, CI=1.07–133.7).

70

Conclusion: This 4-year longitudinal study is the first to demonstrate DD progression and

71

incidence in the lumbar spine and their risk factors in a large population-based cohort.

SC

M AN U

TE D

EP

AC C

72

RI PT

52

73 74 75 76 3

ACCEPTED MANUSCRIPT Introduction

78

Low back pain causes functional impairment, diminished quality of life, loss of working

79

ability, and increased health care costs (1-4). Intervertebral disc degeneration (DD) in the

80

lumbar spine is one of the causes of low back pain (3, 4). Although many studies have been

81

directed at identifying risk factors for DD, aging remains the only established risk factor for

82

DD progression (2-7). Factors, such as smoking, obesity, diabetes mellitus (DM),

83

hypertension (HT), and physical activity, such as driving and lifting weight, might enhance

84

DD progression (2-15); however, these associations are still unclear. This may be due to

85

limitations of previous studies, such as insufficient sample size, variability in subject age,

86

ethnicity, and radiological acquisition, and use of a cross-sectional study design (2-15).

87

Moreover, to our knowledge, no longitudinal study with a large population-based cohort has

88

investigated the progression, incidence, and risk factors for DD in the lumbar spine.

M AN U

SC

RI PT

77

DD etiology in the lumbar spine is also unclear. Importantly, some investigators have

89

emphasized the importance of examining the upper and lower lumbar spine separately, as

91

they are differentially influenced by genetic, environmental, and metabolic factors (16, 17).

92

However, previous studies have mostly focused on the lumbar spine as a whole, with DD

93

regarded a result of aging and mechanical injuries throughout the entire lumbar spine (2-15).

94

Thus, the present study aimed to examine the progression, incidence, and risk factors for DD

95

in the entire lumbar spine, in the upper and lower lumbar spine separately, and at each

96

intervertebral level by using a large-scale, population-based study: the Wakayama Spine

97

Study.

AC C

EP

TE D

90

98 99

Patients and Methods

100

Participants for the Wakayama Spine Study

101

Our study was based on the Wakayama Spine Study (3, 14, 18) which was a sub-cohort of the 4

ACCEPTED MANUSCRIPT Research on Osteoarthritis/Osteoporosis Against Disability (ROAD) study (19-21). ROAD

103

participants were recruited from resident registration listings in three communities with

104

varying geographical characteristics: an urban region in I town (Tokyo), a mountainous

105

region in H town (Wakayama), and a coastal region in T town (Wakayama). Upon the second

106

visit of the ROAD study to H and T towns (conducted between 2008 and 2010), 1063

107

volunteers were recruited for MRI. Among the 1063 volunteers, 52 declined to attend the

108

examination; therefore, 1011 inhabitants were recruited for the baseline survey of the

109

Wakayama Spine Study (Figure 1). The second survey of the Wakayama Spine Study was

110

conducted 4 years after the baseline, and consisted of the same interview, examination,

111

biochemical measurements, and radiographic assessment performed at baseline. Among the

112

1011 participants who participated in the baseline survey, 275 and 816 were mountainous and

113

coastal region inhabitants, respectively. In this follow-up study, however, we recruited only

114

the 816 coastal region participants because MRI was not performed in the mountainous

115

region during the second survey owing to cost and time.

SC

M AN U

TE D

116

RI PT

102

Inclusion criteria were the ability to walk to the survey site, report data, and provide informed consent. Participants with known spine tumors, infections, chronic inflammatory

118

conditions, posterior spinal fusion operations, MRI-sensitive implanted devices (e.g.,

119

pacemakers), and/or other disqualifiers (e.g., pregnancy) were excluded. The Wakayama

120

Spine Study was approved by the local ethics committee of the University of Tokyo, the

121

Tokyo Metropolitan Institute of Gerontology, and Wakayama Medical University. All

122

participants provided written informed consent.

AC C

EP

117

123 124 125 126

Eligible subjects We attempted to trace and review all 816 coastal participants in the Wakayama Spine Study by inviting them to attend a follow-up interview and undergo repeated whole-spine 5

ACCEPTED MANUSCRIPT MRI. Among them, 23 participants (2.8 %) had died by the time of review 4 years later, 6

128

(0.7 %) did not participate in the follow-up study due to poor health, 13 (1.6 %) had moved,

129

and 32 (3.9 %) did not participate for unknown reasons. Therefore, 755 participants attended

130

the second survey of the Wakayama Spine Study. We also identified participants who

131

attended the second visit but declined MRI because they were not qualified to participate (94

132

participants). Among the 661 individuals who participated in the follow-up study, we

133

excluded 1 participant (0.2%) who underwent a posterior spinal fusion surgery and 43 (6.5%)

134

who had incomplete lumbar spine MRI at baseline or follow-up (Figure 1).

SC

RI PT

127

We therefore enrolled 617 participants (75.6% of baseline participants; 178 men) 4

135

years after the baseline study. The mean age at follow-up was 65.4 ± 12.0 years. The

137

participants completed an interviewer-administered questionnaire of 400 items that included

138

lifestyle information (e.g., smoking and drinking alcohol at least more than once a month),

139

family history, past medical history, and lifetime occupational activity history (i.e., driving

140

≥4 h/day, lifting loads weighing ≥10 kg at least once a week). Anthropometric measurements

141

were also obtained and included height and body mass index (BMI = weight [kg]/height

142

[m]2). An experienced public health nurse measured systolic and diastolic blood pressure

143

(BP), with hypertension (HT) diagnosed as systolic BP ≥130 mmHg and/or diastolic BP ≥85

144

mmHg.

AC C

145

EP

TE D

M AN U

136

146

MRI

147

A mobile MRI unit (Excelart 1.5 T; Toshiba, Tokyo, Japan) was used for the baseline study;

148

another mobile MRI unit (Achieva 1.5 T; Philips Medical Systems, Best, the Netherlands)

149

was used for the follow-up study. Whole-spine MRI was performed for all participants on the

150

same day as the questionnaire and anthropometric examination. The participants were supine

151

during MRI, and those with rounded backs used triangular pillows under their head and 6

ACCEPTED MANUSCRIPT knees. The imaging protocol included sagittal T2-weighted fast-spin echo (FSE) (repetition

153

time [TR]: 4000 ms/echo, echo time [TE]: 120 ms, field of view [FOV]: 300 × 320 mm) at

154

baseline. Sagittal T1-weighted images were not obtained owing to cost and time limitations;

155

only T2-weighted images were obtained at baseline. However, both T1- and T2-weighted

156

images were obtained at follow-up. The imaging protocols at follow-up were T2-weighted

157

fast-spin echo (FSE) (TR: 3000 ms/echo, TE: 120 ms, FOV: 270 × 270 mm) and T1-

158

weighted FSE (TR: 540 ms/echo, TE: 10 ms, FOV: 270 × 270 mm).

RI PT

152

SC

159

Radiographic assessment

161

Sagittal T2-weighted images were used to assess DD at all intervertebral levels from L1/2 to

162

L5/S1 at both baseline and follow-up. DD grading at both time-points was performed by the

163

same board-certified orthopedic surgeon (MT) blinded to baseline and follow-up status. DD

164

degree assessed by MRI was classified based on the 5-grade Pfirrmann system (22), with

165

grades 4 and 5 indicating DD (3, 14, 18). The signal intensity for grade 4 was intermediate to

166

hypointense for the cerebrospinal fluid (dark gray), whereas the structure was

167

inhomogeneous. Meanwhile, the signal intensity was hypointense for the cerebrospinal fluid

168

(black) for grade 5, and the structure was inhomogeneous. Furthermore, the disc space was

169

collapsed (Figure 2). To evaluate intra- and inter-observer variabilities, 100 randomly

170

selected lumbar spine MRI scans were rescored by the same observer (MT) more than 1

171

month after the first reading at baseline. Two orthopedic surgeons (MT and RK) similarly

172

scored the MRI scans. The intra- and inter-observer variabilities for DD, evaluated by kappa

173

analysis, were 0.94 and 0.94, respectively, in the baseline study. Furthermore, the intra- and

174

inter-observer variabilities for DD grading, evaluated by kappa analysis, were 0.86 and 0.89,

175

respectively, in the follow-up study (MT and RK).

AC C

EP

TE D

M AN U

160

176 7

ACCEPTED MANUSCRIPT Blood examination

178

All blood and urine samples were extracted between 9:00 AM and 3:00 PM, with some

179

extracted under fasting conditions. After blood sample centrifugation, sera were immediately

180

placed onto dry ice, and transferred to a deep freezer within 24 h. These samples were stored

181

at −80°C until assayed. In the baseline study, the following were measured: blood counts,

182

hemoglobin, hemoglobin A1c (HbA1c), blood sugar, total protein, aspartate

183

aminotransferase, alanine aminotransferase, γ-glutamyl transpeptidase, high-density

184

lipoprotein cholesterol, total cholesterol, triglycerides, blood urea nitrogen, uric acid, and

185

creatinine. These analyses were performed at the same laboratory within 24 h of extraction

186

(Osaka Kessei Research Laboratories Inc., Osaka, Japan).

SC

M AN U

187

RI PT

177

We did not use data of serum glucose levels due to their large variation dependent on hours after eating. Instead, we used serum HbA1c level ≥ 6.1% (JDS) (HbA1c [NGSP]

189

[%] is estimated as an NGSP equivalent value calculated by the formula HbA1c [%] =

190

HbA1c [JDS] [%] + 0.4%) to indicate DM. We decided to use BMI ≥ 25 kg/m2 as an

191

indicator of obesity based on criteria of the Japan Society for the Study of Obesity (23).

192

These are indices used in the National Health and Nutrition Survey in Japan (24).

EP

193

TE D

188

Interview-administered questionnaire

195

Information collected regarding occupational activity included a lifetime occupational history

196

with details of seven types of specific workplace physical activities, including sitting on a

197

chair, kneeling, squatting, standing, walking, climbing, and heavy lifting. Participants were

198

asked whether they engaged in the following activities: sitting on a chair ≥2 h/day, kneeling

199

≥1 h/day, squatting ≥1 h/day, standing ≥2 h/day, walking ≥3 km/day, climbing slopes or steps

200

≥1 h/day, and lifting weight ≥10 kg at least once a week. Information on these activities was

201

obtained for the principal job, defined as the job wherein the participant worked the longest.

AC C

194

8

ACCEPTED MANUSCRIPT 202

Definition of DD progression and incidence

204

For this study, 1) a participant was defined as showing DD progression in the entire lumbar

205

(L1/2 to L5/S1), upper lumbar (L1/2 to L3/4), and lower lumbar spine (L4/5 and L5/S1), and

206

at each level if at least one intervertebral disc showed an increase in Pfirrmann grade,

207

regardless of the grade at baseline (Figure 3). We excluded participants with a baseline score

208

of grade 5 for all intervertebral discs of the affected region, as DD could not show any further

209

progression. 2) A participant was also defined as exhibiting DD in the entire, upper, and

210

lower lumbar spine, and at each intervertebral level, if all intervertebral discs had a score of

211

grade 3 or less at baseline (white disc), and at least one intervertebral disc had progressed to

212

grade 4 or higher (dark or black disc) at follow-up in the affected region (Figure 4).

M AN U

SC

RI PT

203

213

Statistical analysis

215

All statistical analyses were performed using JMP version 9 (SAS Institute Japan, Tokyo,

216

Japan). First, DD prevalence was examined by sex in the entire, upper, and lower lumbar

217

spine, and at every level, followed by DD progression and incidence in these regions. We determined risk factors for DD progression and incidence in the entire, upper, and

EP

218

TE D

214

lower lumbar spine by using univariable and multivariable logistic regression analyses after

220

adjustment for age and sex. DD progression and incidence were considered the objective

221

variables; all other evaluated items (i.e., age, sex, obesity, smoking, DM, HT, driving, and

222

lifting weight) reported as risk factors for DD (2-15) were considered explanatory variables.

223

We further examined significant risk factors with multivariable regression analysis, using the

224

explanatory variables that were significantly associated on univariable regression analysis.

225

The threshold for statistical significance was a p-value <0.05. Furthermore, we included the

226

data of possible risk factors with no statistically significant trends.

AC C

219

9

ACCEPTED MANUSCRIPT 227

Results

229

Table 1 shows selected baseline characteristics of the participants. Those participating in the

230

follow-up survey were younger than those who did not survive or who did not participate for

231

other reasons (responders, 61.4 years; non-responders, 70.7 years; p < 0.0001). The

232

participants in the follow-up survey were also more likely to be women (responders, 70.7%

233

women; non-responders, 56.8% women; p < 0.001). However, the height and weight were

234

similar (height: responders [mean ± standard deviation, SD] 157.3 ± 9.1 cm; non-responders,

235

157.2 ± 8.5 cm/weight: responders, 57.1 ± 11.2 kg; non-responders, 56.6 ± 12.3 kg).

SC

RI PT

228

DD prevalence in the entire lumbar spine was 89.3% and 91.3% in men and women

237

at baseline, respectively, and 91.0% and 94.5% in men and women at follow-up, respectively.

238

DD progression over 4 years in the entire lumbar spine was observed in 52.0% (95% CI;

239

37.3–66.3) and 60.4% (51.4–69.4) of men and women, respectively; in the upper lumbar

240

spine in 35.6% (21.7–49.5) and 48.2% (39.0–57.4) of men and women, respectively; and in

241

the lower lumbar spine in 25.7% (12.8–38.6) and 25.4% (17.2–33.6) of men and women,

242

respectively. DD progression at L1/2, L2/3, L3/4, L4/5, and L5/S1 was 19.3% (7.9–30.7),

243

22.9% (10.4–35.4), 14.9% (4.3–25.5), 14.0% (3.4–24.6), and 16.7% (5.3–28.1), respectively,

244

in men, and 30.3% (21.7–38.9), 28.3% (19.9–36.7), 22.8% (10.7–34.9), 15.7% (8.6–22.8),

245

and 16.5% (9.1–25.6), respectively, in women.

TE D

EP

AC C

246

M AN U

236

Table 2 shows DD incidence in the entire, upper, and lower lumbar spine, as well as

247

at each level. For men, DD incidence was the highest at L5/S1, followed by L4/5. For

248

women, DD incidence was the highest at L4/5, followed by L3/4.

249

Table 3 shows the risk factors for DD progression in the entire, upper, and lower

250

lumbar spine. Female sex was significantly associated with DD progression in the upper

251

lumbar spine (odds ratio [OR], 1.68; 95% CI, 1.18–2.42) and was also significantly 10

ACCEPTED MANUSCRIPT 252

associated with DD progression in the upper lumbar spine after adjustment for age (OR, 1.69;

253

95% CI, 1.17–2.43). Furthermore, female sex and smoking habit also were possible risk

254

factors for DD progression in the entire lumbar spine after adjustment for age and sex (OR:

255

female sex, 1.41; 95% CI, 0.99–2.01; smoking, 0.83; 95% CI, 0.47–1.46). Table 4 shows the risk factors for DD incidence in the entire, upper, and lower lumbar

RI PT

256

spine. Aging was a significant risk factor in the entire, upper, and lower lumbar spine after

258

adjustment for sex (entire, OR, 1.14; 95% CI, 1.06–1.25; upper, OR, 1.10; 95% CI, 1.05–

259

1.15; lower, OR, 1.11; 95% CI, 1.05–1.19). Moreover, DM was a significant risk factor for

260

the DD incidence in the upper lumbar spine after adjustment for all other significant variables

261

in Crude analysis (OR, 6.83; 95% CI, 1.07–133.7). Furthermore, HT was a possible risk

262

factor for DD progression in the entire lumbar spine after adjustment for age and sex (OR,

263

3.37; 95% CI, 0.66–26.0); HT and lifting weight were also possible risk factors for DD

264

progression in the upper lumbar spine after adjustment for age and sex (HT, OR, 1.12; 95%

265

CI, 0.53–2.34; lifting weight, OR, 0.82; 95% CI, 0.40–1.67).

TE D

M AN U

SC

257

266

Discussion

268

The present longitudinal study is the first to determine the progression, incidence, and risk

269

factors for DD in the lumbar spine using MRI in a large-scale population-based cohort. We

270

followed up participants from the coastal region of the Wakayama Spine Study 4 years after

271

the baseline study, achieving a 75.6% participation rate in our follow-up survey. First, we

272

found high rates of DD progression and incidence in Japanese elderly people. We also

273

elucidated that aging was significantly associated with incidence, but not with DD

274

progression. Moreover, female sex was significantly associated with DD progression in the

275

upper lumbar spine. Furthermore, DM was significantly associated with DD incidence in the

276

upper lumbar spine.

AC C

EP

267

11

ACCEPTED MANUSCRIPT 277

Among the few existing population-based epidemiological studies regarding DD progression, reported rates of annual progression in adulthood have varied from 0.42% to

279

76% (6, 7, 8, 15, 26, 27). This may be due to limitations, including inconsistencies in study

280

samples used (e.g., regarding subjects’ age and nationality), and differences in the definition

281

of DD as assessed by radiography or MRI. The present study, which used a uniform cohort

282

for age and nationality, clarified that DD progression per year (based on MRI) in the entire

283

lumbar spine was 13.0 % and 15.1% in Japanese men and women, respectively.

One population-based study examined lumbar spondylosis incidence and

SC

284

RI PT

278

determined DD incidence (28). Symmons et al. found that 4.2% of individuals showed

286

degenerative change in the lumbar spine per year in a radiographic survey of Dutch women

287

(mean age, 54 years) (28). However, to our knowledge, no detailed MRI study has provided

288

qualitative information, such as disc space narrowing and signal intensity loss of the

289

intervertebral discs. Thus, the present study is the first large-scale longitudinal study using

290

MRI to examine the lumbar spine, and found that DD incidence in the entire lumbar spine per

291

year was 7.9 % and 11.2 % in men and women, respectively.

TE D

M AN U

285

We also examined risk factors for DD. Over the past three decades, DD has been

293

commonly thought to occur at a single level, predominantly in the lower lumbar spine, and

294

that any further degeneration occurs consecutively in the adjacent intervertebral levels.

295

However, “skipped” level DD and “dysgenerated” discs occur in the upper lumbar spine, with

296

their occurrence being influenced by genetic, environmental, and endogenous factors in

297

addition to aging and physical loading (29-31). Furthermore, Battie et al. emphasized the

298

importance of examining the upper and lower lumbar spine separately in the study of DD risk

299

factors (16, 17). The present study therefore examined the potential risk factors, including

300

aging, lifestyle, and environmental factor, and metabolic factors, for DD in the upper and

301

lower lumbar spine separately. DD progression in the upper lumbar spine was significantly

AC C

EP

292

12

ACCEPTED MANUSCRIPT associated with female sex. This may be partly explained by the difference in environmental

303

factors affecting men and women besides the sex difference. For example, some factors, such

304

as physical loading, endogenous factors, and genetics, have also been reported to play a role

305

in DD, as women show degenerative change 10 years later compared with men (32, 33).

306

However, it was difficult to clarify the factor for the difference of upper lumbar DD between

307

men and women in the present study; therefore, future investigations should include

308

continued follow-up surveys for shedding light on the pathology and etiology of lumbar DD.

309

Aging was also significantly associated with DD incidence, but not with its progression in the

310

lumbar spine. This is in agreement with previous cross-sectional studies, including our own

311

(2-7) and another longitudinal study (8). Our results are also consistent with that of a UK

312

twin study of the spine using MRI, which found that aging had no detectable effect on lumbar

313

DD progression (15). This may be because aging results in the initial degenerative change as

314

dark or black nucleus pulposus in the intervertebral discs, but affects the later progression of

315

degenerative change less strongly. To our knowledge, no study has examined the influence of

316

aging on DD incidence and progression.

SC

M AN U

TE D

317

RI PT

302

DM was significantly associated with DD incidence in the upper lumbar spine. We also found that intervertebral discs in the entire and lower lumbar spine of all subjects with

319

DM underwent degeneration. In contrast to these results, our previous cross-sectional study

320

found an association between DM and thoracic DD, but not with lumbar DD (14). We

321

speculate that as the lumbar spine comprises mobile segments, intervertebral discs in that

322

region are easily affected by mechanical and motion stress. Concurrently, the effect of

323

metabolic factors, such as DM on the lumbar spine, may be masked, but can eventually be

324

uncovered in a more detailed longitudinal study. However, our results should be interpreted

325

with caution, as the sample size used was small and the results were not significant. With a

326

larger sample and a longer follow-up period, DM may be proved a significant contributor to

AC C

EP

318

13

ACCEPTED MANUSCRIPT 327 328

DD. In previous studies, occupational activities, such as driving and lifting weight, have been commonly thought to be associated with DD in the lumbar spine (2, 5). Although we

330

could not identify this association in the present study, our result is in agreement with a

331

Finnish twin cohort (9) and Chinford studies (7). However, the limited scope of our

332

questionnaire on physical occupation might have reduced our ability to detect a small effect

333

of occupation on DD, if it exists. Further investigation and continued longitudinal survey are

334

needed to clarify whether physical/occupational activities are risk factors for DD.

SC

335

RI PT

329

We have assessed low back pain in both baseline and follow-up studies. The prevalence of low back pain was 38.1% and 31.7% in the baseline and follow-up studies,

337

respectively. Furthermore, we have examined the association between DD incidence and

338

progression in the lumbar spine and low back pain; however, we could not find a significant

339

associations on the chi-square test (DD incidence, p = 0.7; DD progression, p = 0.6). In a

340

future continuous longitudinal study, we will also determine low back pain with the DD

341

incidence and progression.

TE D

342

M AN U

336

Despite the strengths, this study has some limitations. First, the participants included in the Wakayama Spine Study may not represent the general population because they were

344

recruited from only two local areas and were restricted to those who were available for MRI

345

evaluation. To confirm whether the participants of the Wakayama Spine Study at baseline are

346

representative of the Japanese population surveyed by the National Health and Nutrition

347

Survey in Japan (24), we compared anthropometric measurements and frequencies of

348

smoking and alcohol consumption between the general Japanese population and study

349

participants. No significant differences in BMI were observed (men: 24.0 in the Japanese

350

population and 23.7 in study participants, p = 0.33; women: 23.5 in the Japanese population

351

and 23.1 in study participants, p = 0.07). In contrast, the proportion of current male smokers

AC C

EP

343

14

ACCEPTED MANUSCRIPT and drinkers (defined as those who regularly smoke or consume alcohol more than once per

353

month) and the proportion of current female drinkers were significantly higher in the general

354

Japanese population than in the study population (male smokers: 32.6% in the Japanese

355

population and 25.2% in study participants, p = 0.015; male drinkers: 73.9% in the Japanese

356

population and 56.8% in study participants, p < 0.0001; female drinkers: 28.1% in the

357

Japanese population and 18.8% in study participants, p < 0.0001). However, no significant

358

difference was found in the proportion of current female smokers (4.9% in the Japanese

359

population and 4.1% in study participants, p = 0.50). These results suggest that the subjects in

360

this cohort likely had healthier lifestyles than the general Japanese population. This “healthy”

361

selection bias should be taken into consideration when evaluating potential risk factors in the

362

Wakayama Spine Study. Second, this longitudinal study had only a 4-year follow-up,

363

potentially limiting our ability to detect risk factors with a small effect on DD. Third, the

364

present study could not examine the detail association between DD incidence and progression

365

and low back pain owing to the limited pain profile assessment available. Furthermore, the

366

prevalence of DD was high at baseline and follow-up surveys; therefore, the association

367

between DD and low back pain may be difficult to show in this elderly cohort. However, a

368

discrepancy is often found between the clinical profile and DD on MRI owing to the natural

369

history of aging in the intervertebral disc, as shown in a previous study (3). Therefore, such

370

information should be used in the future to assess in-depth clinical relevance and utility.

371

Fourth, we used the traditional multivariable models to adjust the risk estimates for potential

372

confounding bias. However, excluding one variable may possibly lead to residual

373

confounding or collider stratification bias (34). Furthermore, the risk factor estimation has

374

been performed using logistic regression, and the results are presented for odds ratios in the

375

present study. Odds ratios do not approximate well to the relative risk when the outcome was

376

common (incidence ≥ 10%) (35, 36). The OR results in this study calculated with logistic

AC C

EP

TE D

M AN U

SC

RI PT

352

15

ACCEPTED MANUSCRIPT regression analysis cannot be directly translated to relative risk. Finally, the cohort assessed

378

consisted only of elderly subjects. We therefore may have been unable to identify some risk

379

factors for DD in the lumbar spine, particularly in the lower lumbar spine, due to the small

380

participant number. Importantly, however, the present study maintained a very high

381

participation rate at follow-up (75.6%), a considerable strength.

382

RI PT

377

This longitudinal study clarified the progression, incidence, and risk factors for DD throughout the lumbar spine in a large-scale population-based cohort. For men, DD

384

progression and incidence in the entire lumbar spine were 52.0% and 31.6% respectively, and

385

60.4% and 44.7% in women, respectively. The female sex was also a risk factor for DD

386

progression in the upper lumbar spine. Moreover, aging was a risk factor for the incidence,

387

but not DD progression in the entire lumbar spine, and DM was a significant risk factor for

388

DD incidence in the upper lumbar spine. On uncovering the reasons for the differential

389

effects of the risk factors we identified, we might shed light on the pathology and etiology of

390

lumbar DD. Furthermore, large-scale longitudinal studies are needed to further validate our

391

findings and address their clinical impact.

TE D

M AN U

SC

383

393

EP

392

Author contributions

395

All authors worked collectively to develop the protocols and method described in this paper.

396

MT, NY, HH, HY, HO, KN, YI, RK, TA, and SM were principal investigators responsible for

397

the fieldwork in the Wakayama Spine study. MT, HH, and HO performed the statistical

398

analysis. All authors contributed to the analysis and interpretation of results. MT wrote the

399

report. All authors read and approved the final manuscript.

AC C

394

400 401

Role of the funding source 16

ACCEPTED MANUSCRIPT 402

The study sponsors played no role in the study design, the collection, analysis, and

403

interpretation of data, writing of the report, or the decision to submit the paper for

404

publication. The corresponding author had full access to all the data and had the final

405

decision to submit for publication.

RI PT

406

Acknowledgments

408

The authors wish to thank Mrs. Tamako Tsutsumi, Mrs. Kanami Maeda, and other members

409

of the Public Office in Taiji Town for their assistance in the location and scheduling of

410

participants for examinations. Furthermore, the authors appreciate Prof. Toshio Shimokawa in

411

Clinical Research Center of Wakayama Medical University for his supports in the statistical

412

analysis.

413

M AN U

SC

407

Conflict of interest statement

415

The authors declare that we have no conflicts of interest.

418 419 420 421 422

EP

417

AC C

416

TE D

414

423 424 425 426 17

ACCEPTED MANUSCRIPT 427 428

References 1. Dagenais S, Caro J, Haldeman S. A systematic review of low back pain cost of illness studies in the United States and internationally. Spine J 2008;8:8–20.

429 430

2.

Luoma K, Riihimaki H, Raininko R, Luukkonen R, Lamminen A, Vii- kari-Juntura E. Lumbar disc degeneration in relation to occupation. Scand J Work Environ Health

432

1998;24:358–66.

433

3.

RI PT

431

Teraguchi M, Yoshimura N, Hashizume H, Muraki S, Yamada H, Minamide A, et al. Prevalence and distribution of intervertebral disc degeneration over the entire spine in a

435

population-based cohort: the Wakayama Spine Study. Osteoarthritis Cartilage

436

2014;22:104–10. 4.

M AN U

437

SC

434

Samartzis D, Karppinen J, Mok F, Fong DY, Luk KD, Cheung KM. A population-based study of juvenile disc degeneration and its association with

439

overweight and obesity, low back pain, and diminished functional status. J Bone Joint

440

Surg Am 2011;93:662–70.

441

5.

TE D

438

Hanagai M, Kaneoke K, Kuno S, Hinotsu S, Sakane M, Mamizuka N, et al. Factors associated with lumbar intervertebral disc degeneration in the elderly. Spine J

443

2008;8:732–40. 6.

Progression of lumbar disc degeneration over a decade: a heritability study. Ann

445

Rheum Dis 2011; 70:1203–7.

446 447

7.

Hassett G, Hart DJ, Manek NJ, Doyle DV, Spector TD. Risk factors for progression of

lumbar spine disc degeneration: the Chingford Study. Arthritis Rheum 2003;48:3112–7.

448 449

Williams FM, Popham M, Sambrook PN, Jones AF, Spetor TD, MacGregor AJ.

AC C

444

EP

442

8.

Schoenfeld AJ, Nelson JH, Burk R, Belmont PJ Jr. Incidence and risk factors for

450

lumbar degenerative disc disease in the United States Military 1999–2008. Military

451

Medicine 2011;176:1320–4.

452

9.

Battie MC, Videman T, Gibbons LE, Manninen H, Gill K, Pope M, et al. Occupational 18

ACCEPTED MANUSCRIPT driving and lumbar disc degeneration: a case-control study. Lancet 2002;360:1369–74.

453 454

10.

Battié MC, Videman T, Gill K, Moneta GB, Nyman R, Kaprio J, et al. Volvo award in

455

clinical sciences. Smoking and lumbar intervertebral disc degeneration: an MRI study

456

of identical twins. Spine 1991;16:1015–21. 11.

Samartzis D, Karppinen J, Chan D, Luk KD, Cheung KM. The association of lumbar

RI PT

457 458

intervertebral disc degeneration on magnetic resonance imaging with body mass index

459

in overweight and obese adults. Arthritis Rheum 2012;64:1488–96. 12.

Takatalo J, Karppinen J, Taimela S, Niinimäki J, Laitinen J, Sequeiros RB, et al.

SC

460

Association of abdominal obesity with lumbar disc degeneration--a magnetic resonance

462

imaging study. PLoS One 2013;8:e56244.

463

13.

M AN U

461

Liuke M, Solovieva S, Lamminen A, Luoma K, Leino-Arjas P, Luukkonen R, et al.

464

Disc degeneration of the lumbar spine in relation to overweight. Int J Obes (Lond)

465

2005;29:903–8. 14.

Teraguchi M, Yoshimura N, Hashizume H, Muraki S, Yamada H, Oka H, et al.

TE D

466 467

Metabolic syndrome components are associated with intervertebral disc

468

degeneration: the Wakayama Spine Study. PLoS ONE 2016;11:e0147565.

the progression in lumbar degeneration. Spine 2006;31:671–8.

470 471

16.

17.

Battié MC, Videman T, Parent E. Lumbar disc degeneration: epidemiology and

genetic influences. Spine 2004;29:2679–90.

474 475

Battie MC, Videman T, Levalahti E, Gill K, Kaprio J. Genetic and environmental

effects on disc degeneration by phenotype and spinal level. Spine 2008;33:2801–8.

472 473

Videman T, Battié MC, Ripatti S, Gill K, Manninnen H, Kaprio J. Determinants of

EP

15.

AC C

469

18.

Teraguchi M, Yoshimura N, Hashizume H, Muraki S, Yamada H, Minamide A, et al.

476

The association of combination of disc degeneration, end plate signal change, and

477

Schmorl node with low back pain in a large population study: the Wakayama Spine 19

ACCEPTED MANUSCRIPT Study. The Spine Journal 2015;15:622–8.

478 479

19.

Muraki S, Akune T, Oka H, Ishimoto Y, Nagata K, Yoshida M, et al. Incidence and

480

risk factors for radiographic lumbar spondylosis and lower back pain in Japanese

481

men and women: the ROAD study. Osteoarthritis Cartilage 2012;20:712–8. 20.

Yoshimura N, Muraki S, Oka H, Kawaguchi H, Nakamura K, Akune T. Cohort

RI PT

482 483

profile: Research on Osteoarthritis/Osteoporosis Against Disability (ROAD) study. Int

484

J Epidemiol 2010;39:988–95. 21.

Yoshimura N, Muraki S, Oka H, Mabuchi A, En-Yo Y, Yoshida M, et al. Prevalence of

SC

485

knee osteoarthritis, lumbar spondylosis, and osteoporosis in Japanese men and

487

women: the research on osteoarthritis/osteoporosis against disability study. J Bone

488

Miner Metab 2009;27:620–8. 22.

classification of lumbar intervertebral disc degeneration. Spine 2001;26:1873–8.

490 491

Pfirrmann CW, Metzdorf A, Zanetti M, Hodler J, Boos N. Magnetic resonance

23.

Examination Committee of Criteria for ‘Obesity Disease’ in Japan; Japan Society for

TE D

489

M AN U

486

492

the Study of Obesity. New criteria for ‘obesity disease’ in Japan. Circ J

493

2002;66:987–92.

and Nutrition Survey 2008 [in Japanese]. Available at:

495

http://www.mhlw.go.jp/houdou/2008/04/dl/h0430-2c.pdf.

496 497

Ministry of Health, Labour and Welfare. The outline of the results of National Health

EP

24.

AC C

494

25.

Videman T, Battié MC, Parent E, Gibbons LE, Vainio P, Kaprio J. Progression and determinants of quantitative magnetic resonance imaging measures of lumbar disc

498 499

degeneration: a five-year follow-up of adult male monozygotic twins. Spine

500

2008;33:1484–90.

501 502

26

Elfering A, Semmer N, Birkhofer D, Zanetti M, Hodler J, Boos N. Risk factors for lumbar disc degeneration: a 5-year prospective MRI study in asymptomatic 20

ACCEPTED MANUSCRIPT individuals. Spine 2002;27:125–34.

503 504

27.

Jarvik JG, Hollingworth W, Heagerty PJ, Haynor DR, Boyko EJ, Deyo RA.

505

Three-year incidence of low back pain in an initially asymptomatic cohort: clinical

506

and imaging risk factors. Spine 2005;30:1541–8. 28.

Symmons DPM, van Hemert AM, Vandenbroucke JP, Valkenburg HA. A longitudinal

RI PT

507 508

study of back pain and radiological changes in the lumbar spines of middle aged

509

women. II. Radiographic findings. Ann Rheum Dis 1991;50:162–6. 29.

Cheung KM, Samartzis D, Karppinen J, Mok FP, Ho DW, Fong DY, et al.

SC

510

Intervertebral disc degeneration: new insights based on “skipped” level disc

512

pathology. Arthritis Rheum 2010;62:2392–400.

513

30.

M AN U

511

Cheung KM1, Samartzis D, Karppinen J, Luk KD. Are "patterns" of lumbar disc

514

degeneration associated with low back pain? New insights based on skipped level

515

disc pathology. Spine 2012;37:E430–8.

8.

517

32.

sex, and spine level in 600 autopsy specimens. Spine 1988;13:173–8.

519 520

33.

34.

35.

527

Davies HT, Crombie IK, Tavakoli M. When can odds ratios mislead? BMJ 1998;316:989–91.

525 526

Shrier I, Platt RW. Reducing bias through directed acyclic graphs. BMC Med Res Methodol 2008;8:70.

523 524

Friberg S, Hirsch C. Anatomical and clinical studies on lumbar disc degeneration. Acta Orthop Scand 1949;19:222–42.

521 522

Miller JA, Schmatz C, Schultz AB. Lumbar disc degeneration: correlation with age,

EP

518

Luk KDK, Samartzis D. Intervertebral disc “dysgeneration”. Spine J 2015;15:1915–

TE D

31.

AC C

516

36.

McNutt LA, Wu C, Xue X, Hafner JP. Estimating the relative risk in cohort studies and clinical trials of common outcomes. Am J Epidemiol 2003;157:940–3. 21

ACCEPTED MANUSCRIPT 528

Tables

529

Table 1. Baseline characteristics of subjects in the present study.

530

Table 2. Incidence of disc degeneration in the entire, upper, and lower lumbar spine as well

532

as at each level for 4 years.

RI PT

531

533

Table 3. Risk factors for the progression of disc degeneration in the entire, upper, and lower

535

lumbar spine.

SC

534

536

Table 4. Risk factors for the incidence of disc degeneration in the entire, upper, and lower

538

lumbar spine.

M AN U

537

539

Figure legends

541

Figure 1. Flow diagram for the present study.

542

TE D

540

Figure 2. Mid-sagittal view on T2-weighted images of the spine magnetic resonance

544

imaging.

545

Grade was described according to Pfirrmann classification. Grade 1 was bright hyperintense

546

white signal intensity disc to the cerebrospinal fluid with homogenous structure. Grade 2 was

547

hyperintense white signal disc with inhomogenous strucuture, and clear distinction between

548

nucleus and annulus with or without horizontal gray band. Grade 3 was intermediate gray

549

signal intensity disc with inhomogenous structure, and unclear distinction between nucleus

550

and annulus with horizontal gray band. The signal intensity for grade 4 was intermediate to

551

hypointense (dark gray), whereas the structure is inhomogeneous. Meanwhile, for grade 5,

AC C

EP

543

22

ACCEPTED MANUSCRIPT 552

the signal intensity is hypointense (black), and the structure is inhomogeneous. Moreover, the

553

disc space is collapsed. Grades 4 and 5 were considered “degenerated” in this study.

554

Figure 3. Progression of disc degeneration in the lumbar spine.

556

The subject had progression of disc degeneration at L1/2 in the upper lumbar spine (L1/2 to

557

L3/4) because at least one intervertebral disc showed an increase in Pfirrmann grade,

558

regardless of the grade at baseline in the affected region.

RI PT

555

SC

559

Figure 4. Incidence of disc degeneration in the lumbar spine.

561

The subject shows incidence of disc degeneration at L2/3 in the upper lumbar spine (L1/2 to

562

L3/4) because all intervertebral discs had Pfirrmann grade 3 or less (white disc) at baseline

563

and at least one intervertebral disc progressed to grade 4 or higher (black disc) at follow-up.

AC C

EP

TE D

M AN U

560

23

ACCEPTED MANUSCRIPT

Women 439

Age at baseline, years

61.5±12.1

61.6±13.1

61.5±11.7

Height at baseline, cm

157.4±8.5

166.6±6.4

153.6±6.0

Weight at baseline, kg

58.1±11.1

66.9±11.2

54.6±8.9

BMI at baseline, kg/m2

23.4±3.6

24.1±3.5

23.1±3.6

Smoking habit, %

11

27.8

4.4

RI PT

Table 1. Baseline characteristics of subjects in the present study. Overall Men No. of participants 617 178 Demographic characteristics

30.6

Diabetes mellitus

8.1

Hypertension

36.5

28.3

13.5

6

M AN U

Obestiy

SC

Prevalence of each metabolic component at baseline, %

71.1

78

68.4

Percentages of subjects with occupation activity at baseline, % Driving Lifting weight

12.4

16.2

10.7

47.2

62.8

41

Values are the means ± standard deviation. BMI means body mass index.

TE D

Smoking habit was defined as regularly smoking more than once per month. Obesity was diagnosed as BMI ≥ 25, Diabetes mellitus was diagnosed as serum HbA1c level ≥ 6.1 %, Hypertension was diagnosed as systolic blood pressure ≥130mmHg and/or diastolic blood pressure ≥85mmHg.

AC C

EP

Driving; driving for ≥6 hours/day, Lifting; lifting loads weighting ≥10kg at least once a week.

ACCEPTED MANUSCRIPT

Table 2. Incidence of disc degeneration in the entire, upper, and lower lumbar spine as well as at each level for four years. Entire lumbar spine

Upper lumbar spine

Lower lumbar spine

L1/2

L2/3

L3/4

L4/5

L5/S1

No. at risk N (mean % [95%CI]) No. at risk N (mean % [95%CI])

No. at risk

N (mean % [95%CI])

No. at risk

N (mean % [95%CI])

No. at risk

N (mean % [95%CI])

No. at risk

N (mean % [95%CI])

No. at risk

N (mean % [95%CI])

Overall

57

23 (40.4% [37.9-42.9])

195

69 (35.4% [34.0-36.8]) 69

27 (39.1% [37.1-41.1])

444

94 (21.1% [13.5-28.7])

320

109 (34.1% [23.7-44.5])

237

82 (34.6% [22.2-47.0])

142

53 (37.3% [21.1-53.5])

171

51 (30.0% [16.2-43.8])

Men

19

6 (31.6% [27.6-35.6])

53

15 (28.3% [24.7-31.9]) 24

7 (29.2% [25.4-33.0])

125

24 (19.2% [5.2-33.2])

92

26 (28.3% [9.9-46.7])

68

17 (25.0% [4.2-45.8])

49

14 (28.6% [2.8-54.4])

57

17 (29.8% [5.6-54.0])

Women

38

17 (44.7% [41.6-47.8])

142

54 (38.0% [36.4-39.6]) 45

20 (44.4% [41.4-47.4])

319

70 (21.9% [12.7-31.1])

228

83 (36.4% [23.8-49.0])

167

65 (38.5% [23.7-53.3])

93

39 (41.9% [21.5-62.3])

114

34 (29.8% [12.8-46.8])

RI PT

No. at risk N (mean % [95%CI])

Incidence of disc degeneration in the lumbar region was defined as all intervertebral disc having Pfirrmann grade 3 or less at baseline, and at least one intervertebral disc was grade 4 or higher at follow-up.

Incidence of disc degeneration at each level in the lumbar region was defined as each intervertebral disc having Pfirrmann grade 3 or less at baseline, and relevant intervertebral disc was grade 4 or higher at follow-up.

AC C

EP

TE D

M AN U

SC

95%CI: 95% confidence interval

ACCEPTED MANUSCRIPT

Table 3. Risk factors for the progression of disc degeneration in the entire, upper, and lower lumbar spine, respectively. Upper lumbar spine

Hypertension

Driving

Lifting weight

63/177 (25.6%)

Women

265/439 (60.4%)

1.41 (0.99-2.00) 1.41 (0.99-2.01)

211/438 (48.2%)

<25

239/427 (56.0%)

1

≥25

118/189 (62.4%)

No

Crude OR (95%CI)

Adjusted OR 1 (95%CI)

1.00 (0.97-1.02)

1.01 (0.99-1.02)

1

1.68 (1.18-2.42)*

M AN U

1

181/426 (42.5%)

1

1.31 (0.92-1.86) 1.35 (0.95-1.92)

93/189 (49.2%)

1.31 (0.93-1.85)

323/546 (59.2%)

1

249/545 (45.7%)

1

Yes

33/68 (48.5%)

0.65 (0.39-1.08) 0.83 (0.47-1.46)

24/68 (35.3%)

0.65 (0.38-1.08)

No

323/564 (57.3%)

1

Yes

34/50 (68.0%)

No

TE D

Diabetes mellitus

92/177 (52.0%)

252/563 (44.8%)

1

1.59 (0.87-3.01) 1.61 (0.87-3.11)

22/50 (44.0%)

0.97 (0.54-1.73)

100/170 (58.8%)

1

73/170 (42.9%)

1

Yes

246/421 (58.4%)

0.98 (0.68-1.41) 0.94 (0.64-1.38)

193/420 (46.0%)

1.13 (0.79-1.62)

No

305/529 (57.7%)

1

236/529 (44.6%)

1

Yes

45/75 (60.0%)

1.10 (0.68-1.82) 1.22 (0.74-2.03)

33/75 (44.0%)

0.98 (0.60-1.58)

No

189/320 (59.1%)

1

146/320 (45.6%)

1

Yes

162/285 (56.8%)

0.91 (0.66-1.26) 0.99 (0.71-1.39)

124/285 (43.5%)

0.92 (0.67-1.27)

EP

Smoking habit

Men

AC C

Obesity

No. with the progression of DD/ No.at risk (%)

1.01 (0.99-1.02) 1.01 (0.99-1.02)

Age Sex

Adjusted OR 1 (95%CI)

SC

No. with the progression Crude OR of DD/ No.at risk (%) (95%CI)

RI PT

Entire lumbar spine

The crude OR were calculated by univariable logistic regression analysis without adjustment. The adjusted OR 1 were calculated by multivariable logistic regression analysis after adjustment for age and sex. Upper lumbar spine means L1/2 to L3/4, and lower lumbar spine means L4/5 and L5/S1.

1.69 (1.18-2.43)*

1.37 (0.97-1.95)

0.89 (0.49-1.59)

0.98 (0.54-1.81)

1.15 (0.79-1.69)

0.93 (0.57-1.53)

0.98 (0.70-1.36)

Lower lumbar spine No. with the progression of Crude OR DD/ No.at risk (%) (95%CI)

1.01 (0.99-1.03) 44/171 (25.7%)

1

109/429 (25.4%)

0.98 (0.66-1.49)

106/420 (25.3%)

1

47/180 (26.1%)

1.05 (0.70-1.55)

136/531(25.6%)

1

17/67 (25.4%)

0.99 (0.54-1.74)

137/551 (24.9%)

1

16/47 (34.0%)

1.56 (0.81-2.90)

49/168 (29.2%)

1

97/407 (23.8%)

0.76 (0.51-1.14)

133/514 (25.9%)

1

16/75 (21.3%)

0.78 (0.42-1.37)

78/311 (25.1%)

1

71/279 (25.5%)

1.02 (0.70-1.48)

Adjusted OR 1 (95%CI)

1.01 (0.99-1.03)

0.99 (0.66-1.49)

1.05 (0.70-1.56)

0.83 (0.42-1.57)

1.73 (0.88-3.28)

0.81 (0.53-1.24)

0.72 (0.39-1.29)

1.00 (0.69-1.48)

ACCEPTED MANUSCRIPT

Smoking habit was defined as regularly smoking more than once per month. Obesity was diagnosed as BMI ≥ 25, Diabetes mellitus was diagnosed as serum HbA1c level ≥ 6.1 %, Hypertension was diagnosed as systolic blood pressure ≥130mmHg and/or diastolic blood pressure ≥85mmHg. Driving; driving for ≥6 hours/day, Lifting; lifting loads weighting ≥10kg at least once a week.

AC C

EP

TE D

M AN U

SC

RI PT

OR; odds ratio, 95%CI; 95% confidence interval *p-value<0.05 was statistically significant.

ACCEPTED MANUSCRIPT

Table 4. Risk factors for the incidence of disc degeneration in the entire, upper, and lower lumbar spine, respectively.

Sex

Men

3/19 (15.8%)

1

Women

10/38 (26.3 %)

1.90 (0.50-9.42)

<25

10/44 (22.7 %)

1

≥25

3/13 (23.1%)

1.02 (0.20-4.15)

No

11/41 (26.8%)

1

Yes

2/16 (12.5%)

0.39 (0.06-1.71)

No

11/55 (20.0%)

1

Yes

2/2 (100%)

N.A.

No

2/16 (12.5%)

1

Yes

11/29 (37.9%)

4.28 (0.95-30.6)

No

13/48 (27.1%)

1

Yes

0/9 (0%)

N.A.

No

7/25 (28.0%)

1

Yes

6/32 (18.8%)

0.59 (0.17-2.07)

Diabetes mellitus

Hypertension

Driving

Lifting weight

1.49 (0.30-9.00)

4.55 (0.60-42.2)

0.93 (0.10-10.5)

N.A.

3.37 (0.66-26.0)

N.A.

0.84 (0.19-3.83)

Adjusted OR 2 (95%CI)

No. with the Crude OR incidence of DD/ (95%CI) No.at risk (%)

1.13 (1.08-1.17)** 1.12 (1.08-1.17)** 1.10 (1.05-1.15)** 15/53 (28.3%)

1

54/142 (38.0%)

1.55 (0.79-3.16)

52/146 (35.6%)

1

17/49 (34.7%)

0.96 (0.48-1.88)

65/163 (39.9%)

1

1.39 (0.64-3.13)

0.74 (0.32-1.64)

1

20/45 (44.4%)

1.94 (0.69-5.87)

21/55 (38.2%)

1 1.21 (0.36-3.99)

24/52 (46.2%)

1

3/17 (17.7 %)

0.25 (0.05-0.88)* 0.65 (0.10-3.43)

1

23/65 (35.4%)

1

7/8 (87.5%)

14.1 (2.44-266.8)* 8.80 (1.28-179.9)* 6.83 (1.07-133.7)*

4/4 (100%)

N.A.

21/68 (30.9%)

1

7/21 (33.3%)

1

48/105 (45.7%)

1.88 (0.99-3.63)

19/36 (52.8%)

2.24 (0.75-7.14)

58/164 (35.4%)

1

27/57 (47.4%)

1

0/12 (0 %)

N.A.

14/32 (43.8%)

1

13/37 (35.1%)

0.70 (0.26-1.84)

0.91 (0.39-2.04)

41/101 (40.6%)

1

27/93 (29.0%)

0.60 (0.33-1.08)

0.68 (0.26-1.86)

0.82 (0.40-1.67)

Adjusted OR 2 (95%CI) 1.11 (1.05-1.19)**

2.00 (0.60-7.21)

6/14 (42.9%)

62/187 (33.2%)

1.12 (0.53-2.34)

0.64 (0.17-1.98)

7/24 (29.2%)

0.22 (0.06-0.58)*

10/30 (33.3%)

0.69 (0.17-2.39)

Adjusted OR 1 (95%CI)

1.12 (1.06-1.19)** 1.12 (1.06-1.19)**

4/32 (12.5%)

TE D

Smoking habit

1.14 (1.06-1.14)** 1.14 (1.06-1.25)**

Adjusted OR 1 (95%CI)

M AN U

years

Crude OR (95%CI)

Lower lumbar spine

SC

Adjusted OR 1 (95%CI)

Age

Obesity

Upper lumbar spine No. with the incidence of DD/ No.at risk (%)

RI PT

Entire lumbar spine No. with the Crude OR incidence of DD/ (95%CI) No.at risk (%)

3.38 (0.76-16.7)

0.48 (0.09-2.02)

N.A.

2.26 (0.65-8.45)

N.A.

0.77 (0.22-2.46)

EP

The adjusted OR 1 were calculated by multivariable logistic regression analysis after adjustment for age and sex.

The adjusted OR 2 were calculated by multivariable logistic regression analysis after adjustment for all other significant variables without adjustment. Upper lumbar spine means L1/2 to L3/4, and lower lumbar spine means L4/5 and L5/S1.

AC C

Smoking habit was defined as regularly smoking more than once per month. Obesity was diagnosed as BMI ≥ 25, Diabetes mellitus was diagnosed as serum HbA1c level ≥ 6.1 %, Hypertension was diagnosed as systolic blood pressure ≥130mmHg and/or diastolic blood pressure ≥85mmHg. Driving; driving for ≥6 hours/day, Lifting; lifting loads weighting ≥10kg at least once a week. OR; odds ratio, 95%CI; 95% confidence interval, *p-value<0.05 was statistically significant, **p-value<0.001 was statistically significant. N.A; not applicapable due to small sample.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT