According to the Scoliosis Research Society, scoliosis is defined as a lateral curvature of the spine greater than 10 degrees, as measured by the Cobb method on a standing radiograph¹⁾. For some patients, an underlying cause can be determined, including congenital changes, secondary deformities related to a neuropathic or myopathic condition, or degenerative spondylosis later in life. However, the cause of most scoliosis cases is not known, and such patients are diagnosed with idiopathic scoliosis²⁾. Idiopathic scoliosis can be classified into three categories according to the age at which the deformity is detected: (1) infant: before three years of age; (2) juvenile: between the age of three and 10 years (or at the onset of puberty); and (3) adolescent: when it appears after 10 years of age or after the onset of puberty. Idiopathic scoliosis is one of the most frequent deformities involving the spine, with a worldwide incidence rate ranging from 0.5% to 10%³⁾.

A 50-year-long follow-up cohort study of 332 untreated idiopathic patents in the United States was reported in 2003⁴⁾. There was no significant difference in estimated probability of survival between the patient group and the healthy population group. It was also reported that untreated idiopathic scoliosis causes little physical impairment other than back pain and cosmetic concerns⁴⁾. Curves measuring form 30 to 50 degrees progress an average of 10 to 15 degrees over a lifetime. Curves greater than 50 degrees at maturity would progress steadily at a rate of 1 degree per year^1,5). The curvature of a patient diagnosed with a curve over 50 degrees in the adolescent period will progress severely over their life compared to an individual with a less severe case of idiopathic scoliosis.

The thoracic spine is an important structure consisting of the thoracic cage in which major organs, such as the heart and lungs, are located. Structural changes of the thoracic cage can influence the functions of these vital organs, and several reports have identified a relationship between respiratory symptoms and scoliosis severity^6,8,9). Also, structural anomalies of the heart, especially mitral valve prolapse, are frequently found in patients with scoliosis^7,10). However, only a few studies have investigated the relationship between cardiac function and scoliosis severity^6,11,12).

In the current study, 90 patients who were diagnosed with idiopathic scoliosis and were undergoing correction of this disorder at a single institute (Spine center in Gangnam Severance Hospital) from September 2010 to January 2014 were retrospectively reviewed. Therefore, the purpose of this study was to find an association between preoperative cardiopulmonary function and scoliosis severity before surgery. We performed tissue Doppler imaging in addition to conventional echocardiograph measurements to evaluate diastolic and systolic myocardial function.

The clinical demographic characteristics and the results of pulmonary function and echocardiographic measurements are listed in Table 1.

The male-to-female ratio was 15:75 (16.7% vs. 83.3%) in the 90 adolescent patients. The mean ages of the patients at the time of diagnosis were 14.8±2.2 years (range, 10.5-19.9 years). In total, 78 patients (86.7%) were classified into group A, and there were no significant differences in sex (12:66 vs. 3:9, P=0.405), age (177.1 months vs. 182.8 months, P=0.502), BSA (1.46 m² vs. 1.52 m², P=0.284), body weight (48.4 kg vs. 52.5 kg, P=0.227), BMI (19.0 kg/m² vs. 20.3 kg/m², P=0.207), or Cobb angle (53.8 degrees vs. 53.4 degrees, P=0.924) between the two groups (Table 1). None of the clinical characteristics were statistically correlated with Cobb angle in each group. In group A, 72 patients (92.3%) underwent the preoperative PFT, while in group B, 9 patients (75%) underwent the same tests. Only the FEV₁ values (72.1% vs. 83.9%, P=0.003) were significantly different between the two groups. In groups A and B, preoperative echocardiographic examinations were performed in 41 cases (52.6%) and 8 cases (66.7%), respectively. There were no significant differences in the main pulmonary artery velocity (90.4 m/sec vs. 97.6 m/sec, P=0.658), left ventricular ejection fraction (66.9% vs. 67.8%, P=0.751), mitral E/A (2.07 vs. 2.41, P=0.090), tissue Doppler E/E' (6.40 vs. 6.65, P=0.644), or tissue Doppler E'/A' (2.11 vs. 1.91, P=0.435), between groups (Table 1).

Scoliosis is the most common abnormality of the spine in adolescents¹⁴⁾. The prevalence of adolescent idiopathic scoliosis using a Cobb angle cutoff point of 10 or more is approximately 2% to 2.5%²⁾. Idiopathic scoliosis is generally more common in female children²⁾. The ratio of girls to boys with low curvatures is equal but increases to a ratio of 5.4 girls to 1 boy for curves of 21 degrees or greater²⁾ or 10 girls for every one boy with curves greater than 30 degrees¹⁾. In this study the overall range of Cobb angles was 28.5-94.0 degrees (53.8±15.0 degrees), of which, 83.3% were female patients. The ratio of girls to boys was about 5:1, which is comparable to previously published data.

As the scoliosis patients presented the lower compliance of the respiratory system, in particular the decreased chest wall compliance plays an important role in the impairment of lung volumes, and close correlation with the scoliosis severity and the decrease of FVC⁸⁾. Scoliosis has generally been associated with the development of restrictive lung disease, which could result in decreased lung volume as manifested by a decrement of total lung capacity^8,15). Cobb angle had a negative correlation with FVC and FEV₁ values in group A, which implies that thoracic restriction associated with scoliosis decreases total lung capacity and subsequently FVC and FEV₁¹⁵⁾. Expiratory flow rates are diminished in proportion to the restricted lung volume, while the FEV₁/FVC ratio is within normal ranges⁸⁾. Although the FEV₁/FVC ratio is the most common measure of obstructive lung disease, there was no statistical correlation between FEV₁/FVC and Cobb angle in the current study. This suggests that scoliosis patients do not develop obstructive lung disease. The proportions of patients with an FVC<70% and FEV₁<70% were 25% (18) and 37% (29), respectively. The most scoliosis patient had normal PFT ranges, even though in severe scoliosis children showed the decreased PFT values with statistically significance. Linear regression analysis between FVC, FEV₁, and Cobb angle in group A is shown in Fig. 2. Using the equations developed from the linear regression analysis, the Cobb angles were predicted to be 77.8 degrees and 59.0 degrees when the FVC and FEV₁ were 70% respectively.

Thoracic-dominant scoliosis has an impact on the thoracic cage, which may change the pulmonary artery resistance or flow pattern. In this study, the exact measurement of pulmonary artery flow revealed a lower velocity in thoracic-dominant scoliosis (group A) compared to lumbar scoliosis (group B). This suggests that changes to the thoracic structure interrupt the laminar flow velocity in the main pulmonary artery. Although, the statistical data showed that the flow velocity in the main pulmonary artery was not significantly affected by the level of scoliosis, further study should be considered.

Tissue Doppler imaging in the longitudinal direction showed that the apex is generally immobile, whereas the base moves toward the apex in systole and away from the apex towards the annulus in diastole. This differential motion between base and apex results in a velocity gradient along the myocardial wall, with the highest velocities at the base and low or zero velocity at the apex^16,17). In patients with impaired diastolic cardiac function, the velocity of the E' wave decreased and the E'/A' ratio is usually below 1¹⁷⁾.

Distortion of the thoracic cavity caused by scoliosis could affect the subtle functions of the heart and large vessels by altering their position and their relation to each other¹⁵⁾. Dhuper et al.¹⁰⁾ reported that mitral valve prolapse is four times more common in patients with severe scoliosis than in the normal adolescent population. Li et al.¹¹⁾ have reported that tricuspid annular displacement decreases in patients with severe idiopathic scoliosis. Meanwhile, Liang et al.⁶⁾ showed there were no significant differences in left ventricle ejection fraction and fractional shortening between patients with congenital scoliosis according to the Cobb angles. In the current study, the mitral E/A ratio and tissue Doppler E'/A' ratio in patients with thoracic-dominant scoliosis showed a significant negative correlation with the Cobb angle. Thoracic cage deformity could limit cardiac diastolic movement, as demonstrated in restrictive lung capacity of PFT, and also pulmonary venous return impairment may influence decreased early phase atrioventricular flow and/or atrial contraction. However, the other echocardiographic parameters, including the peak velocity of the main pulmonary artery, left ventricular ejection fraction, and tissue Doppler E/E' ratio, were not statistically correlated with Cobb angles in group A. Mostly, the Doppler measurement including tissue Doppler E/E' and E'/A' and left ventricular ejection fraction were within reasonable ranges in both groups (Table 1). When each patient's values were analyzed in detail, the tissue Doppler E/E' measurements were lower in the thoracic-dominant group compared to the lumbar-dominant group, as measured by the pulmonary artery flow velocity.

It has been reported that the estimated probability of survival of idiopathic scoliosis patients does not differ significantly from a matched general population group⁵⁾. Furthermore, since it is difficult to define the precise relationship between cardiac functional change by echocardiography and scoliosis severity, further meticulous investigations should be performed to clarify the subtle cardiopulmonary functional changes in scoliosis patients.

Therefore, even though various parameters of the echocardiographic and Doppler examinations were within reasonable ranges, tissue Doppler E'/A' showed a negative correlation with Cobb angles in thoracic scoliosis, implying that scoliosis deteriorates diastolic myocardial function.

In conclusion, children with idiopathic scoliosis should undergo an evaluation of cardiopulmonary function regardless of symptoms, especially patients with larger Cobb angles. In the future, the idiopathic scoliosis children should be followed for their cardiopulmonary function after the spine correction operation.