SPINE Volume 39, Number 2, pp 149-152 ©2014, Lippincott Williams & Wilkins

DEFORMITY

Thoracic Scoliosis Prevalence in Patients 50 Years or Older and Its Relationship With Age, Sex, and Thoracic Kyphosis Julio Urrutia, MD, Tomas Zamora, MD, and laniv Klaber, MD

Study Design. Cross-sectional study. Objective. To determine the prevalence of thoracic scoliosis in patients aged 50 years or older and to investigate the association of adult thoracic scoliosis with age, sex, and thoracic sagittal curve. Summary of Background Data. The prevalence of adult thoracic scoliosis has not been clearly determined. In addition, limited data are available on the correlation of adult thoracic scoliosis to age, sex, and thoracic kyphosis. Methods. We studied 750 patients aged 50 years or older (380 males and 380 females) who were evaluated using standing chest plain radiographs. The thoracic curvatures in the coronal and sagittal planes were measured using the Cobb method. Scoliosis was defined by the presence of a coronal curvature 10° or more. We performed a correlation analysis of the coronal curve with age and sagittal curve; in addition, a linear regression analysis was carried out to evaluate age, sex, and sagittal curve as independent predictors of the coronal Cobb angle of the thoracic spine. Results. The prevalence of thoracic scoliosis was 24.2% (184 cases); 160 patients (21.1%) had curves 10° or more but less than 20°; 20 patients (2.6%) had curves 20° or more but less than 30°; and 4 patients (0.5%) had curves 30° or more. Females exhibited a higher prevalence of scoliosis (28.9%) than did males (19.4%), P < 0.01. The older patients exhibited increased scoliosis, but no differences were observed in thoracic kyphosis with increasing scoliosis. Age and sex were independent predictors of the coronal Cobb angle; however, the sagittal angle was not. Conclusion. We found a 24.2% prevalence of thoracic scoliosis in patients 50 years or older; most curves were less than 20°. Thoracic scoliosis was more common in females and in older patients.

From the Department of Orthopaedic Surgery, School of Medicine, Pontificia Universidad Católica de Chile, Santiago, Chile. Acknowledgment date: August 26, 2013. Revision date: Septemher 29, 2013. Acceptance date: October 14, 2013. The manuscript submitted does not contain information about medical device(s)/drug(s). No funds were received in support of this work. No relevant financial activities outside the submitted work. Address correspondence and reprint requests to Julio Urrutia, MD, Department of Orthopaedic Surgery, School of Medicine, Pontificia Universidad Católica de Chile, Marcoleta 352, Santiago, Chile; F-mail: jurrutia®med.puc.cl DOI: 10.1097/BRS.0000000000000095 Spine

Key words: adult scoliosis, prevalence, aging spine, thoracic scoliosis. Level of Evidence: 3 Spine 2014;39:149-152

A

dult scoliosis is an important condition that affects the aging spine. It is important to determine the prevalence of adult scoliosis to establish the overall burden of scoliosis on society. Because most adults with scoliosis are asymptomatic,' it is essential to determine the epidemiology of this disorder using a screening tool that is independent of the presence of spinal symptoms. Recent studies have evaluated the prevalence of adult lumbar scoliosis using widely applied screening tools to minimize the selection bias of the cohort studied; this strategy attempts to better reflect the true prevalence of this condition'^; however, only limited data are available on the epidemiology of thoracic scoliosis in adults. Previous studies of adult lumbar scoliosis have reported that scoliosis is more common in females and in older patients^-^; however, only limited information is available on the associations of sex and age with thoracic scoliosis in adults. Moreover, despite descriptions in the literature of the association of thoracic scoliosis with a decreased thoracic kyphosis or even thoracic lordosis in adolescent idiopathic scoliosis,*^"** scant data are available on this association in adult patients with thoracic scoliosis who are not seeking medical care for spinal symptoms. Chest plain radiographs are frequently obtained to evaluate various conditions, including fever, respiratory symptoms, and as preoperative evaluations in smokers and patients older than 70 years old. Despite the fact that chest radiographs were not designed to study the thoracic spine, current electronic tools used to visualize radiographs allow adjustments in gray scales that result in an adequate visualization of the spine. The purpose of this study is to determine the prevalence of thoracic scoliosis in patients 50 years or older, and to define the relationship of adult thoracic scoliosis to age, sex, and thoracic sagittal plane curves in the same population, using chest plain radiographs as a screening tool. www.spinejournal.com

149

Spine

DEFORMITY

Prevalence of Thoracic Scoliosis in Older Patients • Urrutia et al

MATERIALS AND METHODS Institutional review board approval was obtained to conduct this study. We studied 760 patients 50 years or older (380 males and 380 females) who were consecutively evaluated using standing anteroposterior and lateral chest plain radiographs at a tertiary care University Hospital and also at the network of medical centers located in the same city, from April to June 2013. The radiographs were requested for a variety of indications, including fever, respiratory symptoms, as a preoperative evaluation in smokers or patients older than 70, among other causes. The exclusion criteria were chest radiographs obtained in decubitus, the presence of instrumentation in the thoracic spine and being studied using a spine radiograph (of any segment) at the same time (to avoid patients being evaluated for back symptoms). Radiographs were evaluated using the Impax Web 1000 program (Agfa-Gevaert, Mortsel, Belgium), which is available in our institution. The thoracic curvature magnitude in the coronal plane was measured using the Cobb method if a curve was present on the anteroposterior view of the thoracic spine or between T l and T12 if no obvious curve was observed, after adjusting the scale of grays to allow an adequate visualization of the spine. Scoliosis was defined by the presence of a curvature equal to or greater than 10°. In addition, the thoracic kyphosis was also measured from T5 to T12 using the Cobb method. The curvatures were measured by 2 of the authors (T.Z. and I.K.). In addition, 20 of the cases were randomly chosen to determine the intra- and interobserver variability of the coronal Cobb angle measurement; the definition of the end vertebrae and measurement of the Cobb angle were repeated after at least a 1-month interval to reduce the effect of memory on the measurements. The sample size estimation was performed assuming a condition with an unknown prevalence; using an error margin of 5% and a confidence level of 95%, the necessary sample size was 377 cases per sex. Prevalence was expressed as a proportion of patietits, with a 95% confidence interval. A correlation analysis (Pearson) of the thoracic coronal curve and predictor variables (age and thoracic kyphosis) was performed. A lineal regression analysis was performed to determine the impact of age, sex, and sagittal curve on the presence and magnitude of thoracic scoliosis.

In addition, we aggregated the patients with coronal curves less than 10° into 1 subgroup, and created 3 additional subgroups (scoliosis, 10"-19°; 20°-29°; and >30°) for analyses. An analysis of variance was performed to detect differences in age and thoracic kyphosis among these 4 subgroups. A value of P < 0.05 was considered to indicate a statistically significant difference. Data were analyzed using SPSS version 18 (SPSS, Chicago, IL).

RESULTS We studied 380 male and 380 female patients 50 years or older, with a mean age of 66.64 ± 11.02 years. The prevalence of thoracic scoliosis was 24.2%, 95% confidence interval: 21.3% to 27.2% (184 patients). Among the patients who presented scoliosis, 160 (21.1% of the sample) exhibited curves greater than or equal to 10° but less than 20°, 20 patients (2.6%) had curves greater than or equal to 20° but less than 30°, and 4 patients (0.5%) exhibited a scoliosis curve greater than or equal to 30°. Fernales exhibited a 28.9% prevalence of scoliosis (110 patients), whereas males exhibited a prevalence of 19.4% (74 patients), P < 0.01. The intraobserver variability of the coronal Cobb angle measurement was 3.49° ± 3.61°; the interobserver variability was 3.24° ± 1.63°. The effects of age and sagittal curve were evaluated by performing a correlation analysis, which revealed a small positive correlation of the Cobb angle with age {r = 0.2; P < 0.01) and an even smaller correlation with the sagittal curve (r = 0.08, P = 0.02). A subgroup analysis by sex indicated that age was correlated with the coronal curve in both sexes (females: r = 0.196; P < 0.01 and males: r = 0.195; P < 0.01); however, the same subgroup analysis revealed that the sagittal curve was not significantly correlated with the coronal curve in either females or males as separate groups (females: r = 0.06; P = 0.24 and males: r = 0.09; P = 0.06). The analyses of the subgroups aggregated by coronal curve magnitude revealed that patients were older as their scoliosis became more severe, but no differences were observed in thoracic kyphosis as scoliosis increased (Table 1). Lineal regression analysis was also performed to evaluate whether age, sex, and the sagittal curve independently affected the Cobb angle in the coronal plane. Age (ß coefficient = 0.108; P < 0.01) and sex (ß coefficient = 1.171; P < 0.01) infiuenced the coronal Cobb angle in the lineal

muB

TABLE 1 .

30°

Total

576 (75.8%)

160(21.1%)

20 (2.6%)

4 (0.5%)

760

Ma^

306 (80.5%)

68(17.9%)

6(1.6%)

0

380

Female

270(71.1%)

92 (24.2%)

14(3.7%)

4(1.1%)

380

65.6 ± 10.6

68.9 ±11.4

75.8 ±11.1

74.5 ± 12.2

66.6 ± 11

Thoracic scoliosis prevalence in patients 50 years or older and its relationship with age, sex, and thoracic kyphosis.

Cross-sectional study...
5MB Sizes 0 Downloads 0 Views