PATIENTS WITH SPINAL DEFORMITY may have increased

Transcription

1 335 Pulmonary Function and Spinal Characteristics: Their Relationships in Persons With Idiopathic and Postpoliomyelitic Scoliosis Meng-Chih Lin, MD, Mei-Yun Liaw, MD, Wen-Jer Chen, MD, Pao-Tsai Cheng, MD, Alice May-Kuen Wong, MD, Wen-Ko Chiou, PhD ABSTRACT. Lin M-C, Liaw M-Y, Chen W-J, Cheng P-T, Wong AM-K, Chiou W-K. Pulmonary function and spinal characteristics: their relationships in persons with idiopathic and postpoliomyelitic scoliosis. Arch Phys Med Rehabil 2001; 82: Objective: To identify what influence the various features of spinal deformity have on pulmonary function in persons with idiopathic and the postpoliomyelitic scoliosis. Design: Prospective, cohort, observational study with clinical and radiologic evaluations. Setting: Hospital-based rehabilitation units. Patients: Forty-four patients with idiopathic scoliotis and 16 with postpoliomyelitic scoliotis. Each group was divided into subgroups: normal and abnormal pulmonary function. Intervention: Clinical and radiologic evaluation of spinal deformity, full pulmonary functional test and respiratory muscle strength were performed. Presence of dyspnea on exertion and low back pain (LBP) was recorded. Main Outcome Measures: Pulmonary function: spirometry, lung volume test, and diffusing capacity. Respiratory muscle strength: maximal inspiratory pressure (MIP) and maximal expiratory pressure (MEP). Presence of dyspnea on exertion (DOE), and LBP were also recorded. Characteristics of spinal deformity: direction of convexity, uppermost vertebra, number of involved vertebrae, Cobb angle, the apical vertebra, degree of rotation at the apical vertebra, type of scoliotic curve, and presence of balanced spine and pelvic obliquity. Chi-square analyses and Mann-Whitney U test for between-groups comparisons. Spearman s rho correlation coefficient to determine the existence and magnitude of a relationship. Results: We found significant differences between the idiopathic and postpoliomyelitis groups in the degree of rotation at the apical vertebra, MIP, average percentage of predicted vital capacity, residual volume/total lung capacity, presence of double or triple curves, pelvic obliquity, and DOE. Between the idiopathic scoliotis subgroups we found significant differences in the uppermost vertebra and number of vertebrae in the scoliotic curve. Between the subgroups of the postpoliomyelitis From the Division of Pulmonary and Critical Care Medicine (Lin), Departments of Rehabilitation Medicine (Liaw, Cheng, Wong), of Orthopedic Surgery (Chen), of Industrial Design (Chiou), Chang Gung Memorial Hospital, Chang Gung University, Taipei, Taiwan; and Division of Pulmonary and Critical Care Medicine, Ton Yen General Hospital (Lin), Taipei, Taiwan. Accepted in revised form June 20, Supported by National (Taiwan) Medicine Research Program 610 (grant no. NSC B ). An organization with which one or more of the authors is associated has received or will receive financial benefits from a commercial party having a direct financial interest in the results of the research supporting this article. Reprint requests to Mei-Yun Liaw, MD, Dept of Rehabilitation Medicine, Chang Gung Memorial Hospital, 199 Tung Hwa N Rd, Taipei, Taiwan, myliaw@adm.cgmh.org.tw /01/ $35.00/0 doi: /apmr group were significant differences in the location of the apical vertebra and the uppermost vertebral body of scoliotic curve. In the idiopathic group, pulmonary function was mostly related to scoliotic angle, number of vertebrae in the scoliotic curve, location of the uppermost vertebra, and the patients age; MIP and MEP were negatively related to the scoliotic angle and degree of rotation of apical vertebra. In the postpoliomyelitis group, pulmonary function was mostly related to scoliotic angle, kyphotic angle, location of the uppermost vertebra of the scoliotic curve, and age. Conclusion: No single factor can predict the severity of impairment in scoliotic patients pulmonary function. In both groups, severity of pulmonary impairment was related to the combined features of the spinal deformity. However, uppermost vertebra, scoliotic angle, and patient s age may play important roles influencing pulmonary function in both groups. Key Words: Breathing exercises; Postpoliomyelitis syndrome; Rehabilitation; Respiratory function tests; Scoliosis by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation PATIENTS WITH SPINAL DEFORMITY may have increased incidence of dizziness, insomnia, dyspnea on exertion (DOE), reduction of vital capacity (VC), and pulmonary hypertension, sometimes leading to severe pulmonary impairment and unexpected death from a chest infection attributable to the lack of respiratory reserve. 1-7 Pehrsson et al 6 investigated the mortality and causes of death in 115 patients (80 women, 35 men), born from 1902 to 1937, with untreated scoliosis, and found that the risk of mortality significantly increased at the age of 40 to 50 years. Furthermore, mortality was significantly higher in infantile, juvenile, and postpoliomyelitic scoliosis. The factors responsible for pulmonary impairment in scoliotic patients are not clearly understood. Researchers 1,2,4,8,9 have examined the relations between pulmonary function and the angle of scoliosis, establishing that pulmonary impairment increased with the angle of scoliosis. However, considering that subjects with the same angle of scoliosis may have a different degree of pulmonary impairment, we cannot infer that the angle of scoliosis accounts for much of the variability in the relationship. 1,9 It is more likely that factors other than the scoliotic angle, such as the kyphotic angle, the number of vertebrae included in the scoliotic curve, the degree of rotation of the scoliotic spine, the presence of unbalanced spine and pelvic obliquity, age at onset, duration of deformity, concurrent respiratory disorders, and respiratory muscle strength also contribute to pulmonary impairment. In the present study, we sought to determine the correlation or differences among pulmonary parameters, respiratory muscle strength, and various spinal characteristics between idiopathic and postpoliomyelitic scoliotic patients and between the subgroups with normal and abnormal pulmonary function within each group. Our hope was that with simple anteropos-

2 336 PULMONARY FUNCTION AND SCOLIOTIC CHARACTERISTICS, Lin terior (AP) and lateral views of the whole spine, we could make an earlier and better prediction of possible deterioration of pulmonary function, allowing patients to start more aggressive rehabilitative intervention such as wearing bracing, 1-3,10-14 performing exercise, 15 using electric stimulation, 1,3,16 and even undergoing surgery to prevent deterioration from progression of the curve. 1-3,16-20 METHODS Subjects The present study included the idiopathic and postpoliomyelitic (postpolio) scoliotic patients who were seen at the orthopedic, pulmonary, or rehabilitation departments of Chang Gung Memorial Hospital. The patients with concurrent respiratory disorders and previous spinal surgery were excluded. No patient was receiving medical treatment at the time of the study. The patients with more severe spinal deformity were first seen in the orthopedic department. Patients with respiratory impairment were seen in the pulmonary department. Patients who had mild scoliosis or needed a spinal brace for scoliosis were seen in the rehabilitation department. All subjects received an explanation of all procedures and signed informed consent forms. The patients scoliosis was categorized by the age at which it was first noted: infantile-type (birth 3yr), juvenile-type (3 10yr), adolescent-type (10 yr maturity), and adult-type (maturity old age). 1,21 All patients underwent clinical and radiologic evaluation of their spinal deformity and a full pulmonary function test (PFT). The measurement of respiratory muscle strength, subjective presence of DOE, and low back pain (LBP) were also recorded. Measurements Radiologic measurements. Spinal curvatures were measured, by the Cobb method, using the standard AP and lateral views of the whole spine that were taken at the time of first evaluation. The anomalies were classified using criteria described in Moe s Textbook of Scoliosis and Other Spinal Deformities 21 and adopted by the Scoliosis Research Society. 1 These features of deformity included the single, double, or triple curve; the side of the convexity, the location of the apical vertebra and uppermost vertebra, and scoliotic and kyphotic angles; the involved number of vertebrae (defined as the number of vertebrae between the uppermost end and lowest end of the scoliotic curve); the degree of rotation at the apical vertebra (evaluated by the Nash and Moe 21 technique, 4 grades of rotation according to the position of the pedicles); the presence of unbalanced spine; and the pelvic obliquity of the scoliotic curve. 22 Mild scoliosis was defined as a curvature of less than 30, moderate as 30 to 50, and severe as more than 50. PFT and respiratory muscle strength. pulmonary function was defined as forced vital capacity (FVC) 80% of predicted, forced expiratory volume in 1 second (FEV 1 ) 80% of predicted, FEV 1 /FVC 75% of predicted, and diffusing capacity (DLCO) 80% of predicted. Mild restrictive impairment was defined as FVC 60% to 79% of predicted values, moderate impairment as 51% to 59% of predicted, and severe impairment as lower than 50% of predicted. 23 The spirometry and lung volume test were performed with the SensorMedics 2800 a device, and the diffusing capacity tests were performed with the SensorMedics 2400 a device using the American Thoracic Society 24 standard. The patient s arm span was used to calculate corrected height when he/she could not stand. 25,26 Respiratory muscle strength was measured with the Micro Medical Mouth Pressure Meter b while each subject was standing and wearing a nose-clip. The maximal expiratory pressure (MEP) was measured near total lung capacity (TLC) after a maximal inspiration; the maximal inspiratory pressure (MIP) was measured near residual volume (RV) after a maximal expiration. The pressures measured were maintained for at least 1 second and the tests were repeated until 2 technically satisfactory measurements were recorded; the highest value was used in subsequent calculations. 27 Data Analysis Chi-square analysis was used to compare differences in categorical data between groups. A Mann-Whitney U test was used to compare interval data when it was not normally distributed or when there was inhomogeneity of variance between groups. The Mantel-Haenszel test was used to determine linear association among ordinal variables. Bonferroni s test, the most conservative test for multiple comparison, was used to determine where significant differences lay after a significant F ratio for the analysis of variance (ANOVA). A Spearman s rho correlation coefficient was used to determine the existence and magnitude of a relationship between interval or ratio data. A p value less than.05 was considered significant. Data were analyzed using SAS, c version 6.11 for Windows, and SPSS, d version 8.01 for Windows. RESULTS Clinical and Spinal Characteristics Sixty-six patients were eligible for the study, but 6 of them did not complete the whole examination; therefore, 60 patients (53 women, 7 men) were analyzed. Of them, 44 patients (41 women, 3 men; mean age standard deviation [SD], yr) had idiopathic scoliosis, and 16 subjects (12 women, 4 men; mean age, yr) had scoliosis from poliomyelitis. The only significant differences in clinical characteristics were the presence of the double or triple scoliotic curve, pelvic obliquity, and DOE between the 2 groups (table 1). In the idiopathic group, 14 (31.8%) cases were mild scoliosis, 14 (31.8%) were moderate, and 16 (36.4%) were severe. In Table 1: Clinical Characteristics of Sample, Idiopathic and Postpoliomyelitic Scoliotic Patients Idiopathic (n 44) Postpoliomyelitic (n 16) Gender (F:M) 41:3 12:4 Age (yr) (mean SD) Body height (cm) (mean SD) Body weight (kg) (mean SD) Age of observation (A:B:C:D) 0:9:29:6 1:12:3:0 Degree of scoliosis n (Cobb angle) Mild ( 30 ) 14 (31.8%) 4 (25%) Moderate (30 50 ) 14 (31.8%) 3 (18.8%) Severe ( 50 ) 16 (36.4%) 9 (56.3%) Scoliotic curve 13:31* 10:6* (Single:Double/Triple) (29.5%:70.5%) (62.5%:37.5%) Pelvis obliquity 15:29* 15:1* (Yes:No) (%) (34.1%:65.9%) (93.8%;6.3%) Convex of curve (Right:Left) 43:1 14:2 Back pain n (%) 26 (59.1%) 11 (68.8%) Dyspnea on exertion n (%) 17* (38.6%) 9* (56.3%) Abbreviations: A, birth to 3yr; B, 3 10yr; C, 10yr to maturity; D, after maturity. * Mann-Whitney U test, p.05; chi-square test, p.05.

3 PULMONARY FUNCTION AND SCOLIOTIC CHARACTERISTICS, Lin 337 Table 2: Subjects Characteristics of Spinal Deformity, Respiratory Muscle Strength, and Pulmonary Function Parameters Idiopathic (n 44) Postpoliomyelitic (n 16) Mean SD Mean SD Characteristics of spinal deformity Apical vertebra Uppermost vertebra Involved no. of vertebrae Scoliotic angle (deg) Rotation (deg) 1.8* 1.3* 2.6* 1.3* Kyphotic angle Respiratory muscle strength MIP (cmh 2 O) 58.1* 24.1* 82.7* 35.9* MEP (cmh 2 O) Pulmonary function test VC (L) VC (%pred) 95.2* 31.5* 69.8* 29.9* FVC (L) FVC (%pred) FEV 1 (L) FEV 1 (%pred) FEV 1 /FVC MMEF (L/s) TLC (L) RV (L) FRC (L) RV/TLC (%) 34.0* 11.8* 44.0* 15.1* DLCO (ml/min/mmhg) DLCO (%pred) Abbreviation: MMEF, mid-maximal expiratory flow. * Mann-Whitney U test, p.05. the postpolio group, only 4 (25%) cases were mild, 3 (18.8%) were moderate, and 9 (56.3%) were severe scoliosis. Only 1 case in the idiopathic group and 2 cases in the postpolio group had left convex scoliosis (see table 1). Using the Mann-Whitney U test to analyze the differences in the characteristics of the scoliosis, respiratory muscle strength, and pulmonary function between the idiopathic and postpolio groups, we found significant differences (table 2) only in the degree of apical rotation, MIP, average percentage of predicted (%pred) VC and RV/TLC. Respectively, the degree of apical rotation, MIP, VC %pred, and RV/TLC were , cmH 2 O, 95.2% 31.5%, 34% 11.8% in the idiopathic group versus , cmH 2 O, 69.8% 29.9%, 44% 15.1% in the postpolio group (see table 2). Kruskal-Wallis 1-way ANOVA showed the pulmonary function in subgroups within each group (table 3) to have statistically significant differences in age, location of the uppermost vertebra, involved number of vertebrae, scoliotic angle, presence of double or triple scoliotic curves, and kyphosis (table 4). By using Bonferroni s test to determine the significant differences between subgroups in each group, we found significant differences in the location of the uppermost vertebra and the involved number of scoliotic vertebrae in the idiopathic group and in the apical vertebra and the location of the uppermost vertebra in the postpolio group (table 4). Pulmonary Function Comparing the pulmonary function of both groups to predicted values, such as the average percentages of predicted TLC, FVC, FEV 1, functional residual capacity (FRC), and RV, we found no significant differences (see table 2), although the idiopathic group had better ventilation than the postpolio group. In the idiopathic group, 17 (38.6%) patients had a FVC %pred lower than 80%; of them, 10 (22.7%) had mild impairment, 3 (6.8%) had moderate impairment, and 4 (9%) had severe impairment. Only 3 (of 39) patients had a TLC %pred of less than 80%. Eighteen (40.9%) patients had abnormal diffusion; of them, 13 had mild and 5 had moderately impaired pulmonary function. Two subjects had severe ventilatory impairment and could not perform the diffusing capacity test. Five subjects with normal FVC %pred were found to have mild to moderate diffusing impairment. In the postpolio scoliotic group, 9 (56.3%) patients had a FVC %pred lower than 80%; of them, 4 (25%) had mild impairment, 3 (18.8%) had moderate impairment, and 2 (12.5%) had severe impairment. Seven (43.8%) patients in the group had a TLC %pred less than 80%, Table 3: Pulmonary Function in Subgroups Within the Sample Idiopathic (n 44) Postpoliomyelitis (n 16) Pulmonary Function (n 27) (n 17) (n 7) (n 9) p VC (L) * VC (%pred) * FVC (L) * FVC (%pred) * FEV 1 (L) * FEV 1 (%pred) * MMEF (L/s) * TLC (L) * TLC (%pred) * RV (L) RV/TLC FRC (L) * DLCO (ml/min/mmhg) * DLCO (%pred) * NOTE. Data presented as mean standard deviation. * Kruskal-Wallis 1-way ANOVA, p.05.

4 338 PULMONARY FUNCTION AND SCOLIOTIC CHARACTERISTICS, Lin Table 4: Characteristics of Spinal Deformity and Respiratory Muscle Strength of Subgroups with and Pulmonary Function Pulmonary Function (n 27) Idiopathic (n 44) Postpoliomyelitic (n 16) (n 17) (n 7) (n 9) Age (yr) * Characteristics of spinal deformity Apical vertebra Uppermost vertebra * Involved no. of vertebrae * Degree of rotation Scoliotic angle (deg) * Kyphotic angle (deg) Respiratory muscle strength MIP (cmh 2 O) MEP (cmh 2 O) Double or triple curve 18 (69.2%) 13 (72.2%) 1 (14.2%) 5 (55.6%) * Kyphosis 6 (23.1%) 2 (11.1%) 0 (0%) 5 (55.6%) * NOTE. Data presented as mean SD. * Kruskal-Wallis 1-way ANOVA, p.05. p and all of them had mild TLC %pred impairment. Two subjects had ventilation so severe they could not perform the diffusing capacity test; in addition, 4 (25%) patients had an abnormal diffusing capacity, among them, 1 had mild and 3 had moderate impairment. In those patients with an abnormal FVC %pred, 9 cases (52.9%) in the idiopathic and 6 cases (66.7%) in the postpolio group had a scoliotic angle greater than 60. The abnormal PFT, such as the average FVC %pred, the average TLC %pred, and the average scoliotic angle SD were 62.5% 21.4% versus 52.8% 16.9%, 91.1% 14.6% versus 74.6% 4.6%, and versus in idiopathic and postpolio groups, respectively (see tables 3, 4). On the other hand, for idiopathic scoliosis, 9 (75%) of 12 patients with a scoliotic angle more than 60 had an abnormal PFT, and 6 (66.7%) of 9 postpoliomyelitis patients had an abnormal PFT. Three patients in the idiopathic group and 7 patients in postpolio group had abnormal RV/TLC (abnormal, 50%). The average RV/TLC was 34% 11.8% in the idiopathic group versus 44% 15.1% in the postpolio group. It was even higher, up to 69%, in some patients from both groups. Among the 17 idiopathic patients with an abnormal pulmonary function, 13 (76.5%) had a double major curve, and 5 (55.6%) of 9 postpolio patients with an abnormal PFT had a double curve. In the idiopathic group, 26 cases (59.1%) had LBP, and 17 cases (38.6%) had DOE. In the postpolio group, 11 cases (68.8%) had LBP, and 9 cases (56.3%) had DOE. We found no significant difference in back pain between the 2 groups. There was a statistically significant difference in DOE between the 2 groups. Analyzing the data by Spearman s rho correlation analysis, we found pulmonary impairment in the idiopathic group to be most related to scoliotic angle, the involved number of vertebrae, the patient s age at the time of the study, and the uppermost vertebra of the scoliotic curve. The MIP and MEP correlated negatively to the scoliotic angle and the degree of rotation of the scoliotic apical vertebra (table 5). In the post- Spinal Characteristics Pulmonary Parameters Table 5: The Relationships Among Age, Spinal Characteristics, and Pulmonary Function in Idiopathic Age Apical Uppermost Curve Length Scoliotic Kyphotic Degree of Rotation FVC (L) FVC (%pred) FEV 1 (L) FEV 1 (%pred) MMEF (L/s) VC (L) VC (%pred) TLC (L) RV (L) FRC (L) MIP (cmh 2 O) MEP (cmh 2 O) NOTE. Spearman s rho correlation analysis (p.05). Abbreviations:, positive correlation between spinal characteristics and pulmonary parameters;, negative correlation; or, p.05; or, p.01; or, p.001.

5 PULMONARY FUNCTION AND SCOLIOTIC CHARACTERISTICS, Lin 339 Spinal Character Pulmonary Parameter Table 6: The Relationships Among Age, Spinal Characteristics, and Pulmonary Function in Postpolio Age Apical Uppermost Curve Length Scoliotic Kyphotic Degree of Rotation FVC (L) FVC (%pred) FEV 1 (L) FEV 1 (%pred) MMEF (L/s) VC (L) VC (%pred) TLC (L) RV (L) FRC (L) MIP (cmh 2 O) MEP (cmh 2 O) NOTE. Spearman s rho correlation analysis (p.05). Abbreviations:, positive correlation between spinal characteristics and pulmonary parameters;, negative correlation; or, p.05; or, p.01; or, p.001. polio group, pulmonary impairment was most related to the scoliotic angle, kyphotic angle, the uppermost vertebra of the scoliotic curve, and patient s age at the time of study (table 6). DISCUSSION The abnormal lung function found in scoliotic patients was mostly restrictive in type (ie, it was characterized by a decreased VC with relatively normal expiratory airflow). Because of reduced inspiratory capacity, FEV 1 and FVC were decreased, but the FEV 1 /FVC remained normal. Decreased single breath diffusing capacity was also found in some other studies. 8,28-32 In idiopathic scoliotic patients, VC and TLC may be reduced, but RV is usually within normal limits; the RV/TLC can be as high as 50%. 8,9,26,28-32 The PFT patterns of the postpolio scoliotic group were different from those of the idiopathic group. Because many of the postpolio patients had spinal deformity earlier than the idiopathic scoliotic patients, their spinal deformity could start earlier in their life and progress faster, especially before skeletal maturity and during the period of growth spurts. 1 The postpolio patients not only had a larger Cobb angle, but also were more likely to have abnormal pulmonary function. In this present study, patients with abnormal PFT in both groups, the VC, FVC, and TLC were decreased, the RV was increased, and the RV/TLC also was high (as high as 69% in some cases). The average VC, FVC, and TLC of the postpoliomyelitic scoliotic group were lower, and the average RV/TLC was higher than those of the idiopathic scoliotic group. These results are similar to the results found in previous reports. 8,9,26,28,32 Several reports 1,2,4,8-33 showed a correlation between the degree of thoracic curvature and the reduction in VC in severe scoliosis. Gagnon et al 19 and Jackson et al 34 reported that patients with curves of 50 or more had VC reduction. Jackson 34 also reported that scoliosis greater than 40 and kyphosis greater than 50 correlated with increasing pain and decreasing FVC. In moderate (Cobb angle, ) to severe (Cobb angle, 50 ) scoliosis, the reduction in VC is related to the measured degree of spinal curvature. 26,31 Adult scoliosis patients with a Cobb angle of more than 60 will have a compromised TLC. 3,5,20,32,33,35 Research 4,5 suggests that spinal surgery should be performed before the Cobb angle exceeds 65, because possible irreversible pulmonary change could occur and no significant postoperative improvement in regional lung perfusion would be obtained. Although, after surgical correction of the kyphoscoliosis, the physiological dead space would be decreased by 40%. Researchers 4,5,31,32,36 have suggested that surgery should be performed before the angle reaches 70. Sakic et al, 5 studying the relations between deteriorating cardiorespiratory function and the degree and locality of the spinal curvature s apex, found that scoliosis affected pulmonary function only in upper thoracic curves exceeding 70. In such cases, a direct correlation existed between the VC and increased curve severity. Sakic 5 also reported that patients with an upper thoracic curve (idiopathic T5 T8) with a Cobb angle greater than 70 had an associated restrictive ventilation disorder, increased airway resistance, decreased FEV 1, and latent hypoxemia during the exercise tolerance test. As the curve approached 100, a significant reduction in VC, PaO 2, and FEV 1 occurred all because of restrictive lung disease with alveolar hypoventilation and consequent arterovenous shunting. Cor pulmonale also occurred in these severe cases. 1 Examining previous reports, 32 reviewers found a measurable deficit of pulmonary function even in mild and moderate scoliosis. Smyth et al 8 measured 44 adolescents with mild idiopathic scoliosis (Cobb angle 30 ); all were symptom free, but 6 (13.6%) showed a restrictive defect, with their FVC less than 80% of the predicted value. Smyth 8 also showed that adolescents with mild idiopathic scoliosis might have reduced lung volumes in the presence of minimal lateral curvatures. In the present study, we found that 25% of idiopathic scoliotic patients and 33.3% of postpolio patients with a Cobb angle of more than 60 had normal lung function. Two of 3 cases in the idiopathic group with a scoliotic angle greater than 90 had severe pulmonary function impairment, and the third only had mild restrictive lung impairment. In the postpolio group, in which 4 subjects had a Cobb angle greater than 90, 2 had severe FVC impairment, 1 had moderate impairment, and 1, who had lower body weight, was shorter, and had a lower uppermost vertebra of the scoliosis, had normal lung function. Therefore, the correlation between the severity of the scoliosis and the degree of change in certain indices of respiratory function was not strict. Gagnon et al, 19 investigating 42 women with idiopathic scoliosis (mean age, 20.8yr; mean major thoracic curve, 58.2 ), found that the women with double-major curves had more severe pulmonary functional impairment but no significant

6 340 PULMONARY FUNCTION AND SCOLIOTIC CHARACTERISTICS, Lin difference compared with those with single thoracic curves. Curves over 50 had significantly lower VC. In the present study, we found that in the postpolio group the double or triple curve and the kyphosis were related to abnormal pulmonary function. Considering a 40 kyphotic angle as abnormal, 22 among the idiopathic group only 11.8% (2/17) of patients with abnormal FVC %pred had kyphosis, and only 16.7% (2/12) of patients with kyphosis had abnormal pulmonary function. Among the postpolio group, 55.6% (5/9) of patients with abnormal pulmonary function had kyphosis. And all of the patients with kyphosis had abnormal pulmonary function impairment. Therefore, kyphosis seemed to have a stronger association with abnormal pulmonary function in the postpolio group than in the idiopathic group. Kearon et al 9 found that angle of scoliosis was just 1 of 4 features of spinal deformity associated with reduced VC %pred; greater number of vertebrae involved, cephalad location of the curve, and loss of normal thoracic kyphosis made an equal and additive contribution to pulmonary function impairment. The scoliotic curve usually extends to and includes T4, T5, or T6 at its upper end and T11, T12, or L1 at its lower limit. 3 Sakic et al 5 found that increased airway resistance and decreased FEV 1 existed only in cases involving upper thoracic scoliosis (T5 T8). Most patients with a higher uppermost end of the scoliotic curve had greater pulmonary impairment. In the present study, the locations for the uppermost vertebra in the idiopathic patients with abnormal FVC %pred were distributed mainly between thoracic vertebrae (T1 T5), a location that was higher than the normal subgroup ( ); in the postpolio group with abnormal FVC %pred, it was thoracic vertebrae (T1 T5), a location that was also higher than the normal subgroup of the same group ( ). Scoliosis is often associated with pain, especially back pain. Evaluating 101 adult patients with painful idiopathic scoliosis, none of whom had had prior surgical treatment, Jackson et al 34 found that scoliosis greater than 40 and kyphosis greater than 50 correlated with increasing pain and decreasing FVC. Reduction in FVC also correlated with curve rigidity, and rotation correlated closely with the degree of scoliosis and had the highest correlation with pain. In the present study, we found that the restrictive lung impairment of some patients probably was caused by a painful chest wall, although the spinal deformity was not very severe. The right thoracic curve is among the most common scoliotic patterns. Left thoracic curves and S curves were more at risk of progression. 3 However, we did not find that scoliotic patients with a left scoliotic convex had more severe pulmonary functional impairment (p.4043, see table 1). Whether reduced respiratory muscle strength is a primary mechanism in the development of idiopathic scoliosis or merely secondary to other pathogenetic factors is unknown. 8 Respiratory muscle strength is a more important determinant of this impairment than the radiologically determined degree of spinal curvature. 9,34,37 MIP and MEP as the indices of respiratory muscle strength are reduced in severe idiopathic scoliosis. 8,38 Smyth et al 8 found that FVC correlated significantly with MIP and MEP but was not related to the degree of thoracic curvature. Cooper et al 37 examined 29 patients with typical thoracic curvatures of mild to moderate degree ( 60 ) who had similar results. Respiratory muscle strength should be considered when developing treatment strategies to improve or prevent deterioration of pulmonary function in scoliosis. An isolated lateral curvature of the spine rarely occurs; rotation of the spinal column with or without changes in the normal thoracic kyphosis is invariably part of the deformity of idiopathic scoliosis. In the present study, the MIP, MEP, and FVC of both groups were decreased and negatively correlated to the scoliotic angle and the degree of rotation of scoliosis at the apex. The correlation between VC and MIP is more physiologically appealing in view of the known importance of MIP in the determination of VC. 8 Kafer 33 reported that possible mechanisms for the effects of scoliosis on lung volume include abnormal development of the thorax, the increased elastic force of the respiratory system (which opposes the muscle forces during maximum inspiration and expiration), and the development of inspiratory or expiratory muscle force. Therefore, pulmonary impairment may be related to the combined influence of different specific features of spinal deformity in scoliotic patients. CONCLUSIONS The present study produced these findings: (1) the changes of abnormal pulmonary function in idiopathic and postpoliomyelitic scoliosis were restrictive in type, with a reduction in VC and FVC and an increase in RV; (2) the patients with poliomyelitic scoliosis had a higher frequency of pulmonary function impairment than patients with idiopathic scoliosis; (3) predictive factors of patients abnormal lung function included not only the scoliotic angle but also the uppermost vertebra or the number of vertebrae involved in the scoliotic curve, the degree of rotation, the kyphotic angle, and the patient s age; (4) in patients with idiopathic scoliosis, the major determining factors for abnormal lung function were scoliotic angle, number of vertebrae in the scoliotic curve, the location of the uppermost vertebra, and patient s age; (5) in patients with poliomyelitic scoliosis, the scoliotic angle, kyphotic angle, the uppermost vertebral body of the scoliosis, and patient s age contribute to change in lung function; and (6) multiple factors lead to pulmonary function impairment, in both idiopathic and postpoliomyelitic scoliotic patients. The angle of the scoliosis is unlikely to account for much of the relationship. Other features of scoliosis, such as patient s age, the position of the curve along the thoracic spine, the extension of the spinal deformity, the degree of rotation, kind of scoliotic curve, the presence of the pelvic obliquity and kyphosis, may also influence pulmonary function. Although the present study provided insight into the relations between scoliotic deformity and pulmonary function, one cannot predict pulmonary function from radiography alone. References 1. Graham JJ. Medical management of scoliosis. In: Goodgold J, editor. Rehabilitation medicine. St Louis: Mosby; p Engler GL. Surgical management of scoliosis. In: Goodgold J, editor. Rehabilitation medicine. St Louis: Mosby; p Netter FH. The Ciba collection of medical illustration. Musculoskeletal system, part II: developmental disorders, tumors, rheumatic diseases, and joint replacement. Vol 8. New York: Ciba- Geigy; p Sakic K, Pecina M, Pavicic F. Pulmonary function in adolescents with idiopathic scoliosis. Int Orthop 1992;16: Sakic K, Pecina M, Pavicic F. Cardiorespiratory function in surgically treated thoracic scoliosis with respect to degree and apex of scoliotic curve. Respiration 1992;59: Pehrsson K, Larsson S, Anders O, Nachemson A. Long-term follow-up of patients with untreated scoliosis. A study of mortality, causes of death, and symptoms. Spine 1992;17: Primiano FP, Nussbaum E, Hirschfeld SS, Nash CL, Horowitz JG, Lough MD, et al. Early echocardiographic and pulmonary function findings in idiopathic scoliosis. J Pediatr Orthop 1983;3: Smyth RJ, Chapman KR, Wright TA, Crawford JS, Rebuck AS.

7 PULMONARY FUNCTION AND SCOLIOTIC CHARACTERISTICS, Lin 341 Pulmonary function in adolescents with mild idiopathic scoliosis. Thorax 1984;39: Kearon C, Viviani GR, Kirkley A, Killian KJ. Factors determining pulmonary function in adolescent idiopathic thoracic scoliosis. Am Rev Respir Dis 1993;148: Korovessis P, Filos KS, Georgopoulos D. Long-term alterations of respiratory function in adolescents wearing a brace for idiopathic scoliosis. Spine 1996;21: Winter RB. The pendulum has swung too far. Bracing for adolescent idiopathic scoliosis in the 1990s. Orthop Clin North Am 1994;25: Willers U, Normelli H, Aaro S, Svensson O, Hedlund R. Longterm results of Boston brace treatment on vertebral rotation in idiopathic scoliosis. Spine 1993;18: Refsum HE, Naess-Andresen CF, Lange JE. Pulmonary function and gas exchange at rest and exercise in adolescent girls with mild idiopathic scoliosis during treatment with Boston thoracic brace. Spine 1990;15: Emans JB, Kaelin A, Bancel P, Hall JE, Miller ME. The Boston Bracing System for idiopathic scoliosis. Follow-up results in 295 patients. Spine 1986;11: Weiss HR. The effect of an exercise program on vital capacity and rib mobility in patients with idiopathic scoliosis. Spine 1991;16: McCollough NC III. Nonoperative treatment of idiopathic scoliosis using surface electrical stimulation. Spine 1986;11: Kinnear WJM, Johnston IDA. Does Harrington Instrumentation improve pulmonary function in adolescents with idiopathic scoliosis? A meta-analysis. Spine 1993;18: Sullivan JA, Davidson R, Renshaw TS, Emans JB, Johnston C, Sussman M. Further evaluation of the scolitron treatment of idiopathic adolescent scoliosis. Spine 1986;11: Gagnon S, Jodoin A, Martin R. Pulmonary function test. Study after spinal fusion in young idiopathic scoliosis. Spine 1989;14: Lenke LG, Bridwell KH, Baldus C, Blanke K. Analysis of pulmonary function and axis rotation in adolescent and young adult idiopathic scoliosis patients treated with Cotrel-Dubousset instrumentation. J Spinal Dis 1992;5(1): Winter RB. Congenital spine deformity. In: Bradford PS, Lonstein JE, Moe JH, Ogilvie JW, Winter RB, editors. Moe s textbook of scoliosis and other spinal deformities. Philadelphia: WB Saunders; p Cailliet R. Glossary of scoliosis terms. In: Scoliosis, diagnosis and management. Philadelphia: FA Davis; p American Thoracic Society. Evaluation of impairment/disability secondary to respiratory disorders. Am Rev Respir Dis 1986;133: American Thoracic Society. Standardization of spirometry: 1994 update. Am Rev Respir Dis 1995;152: Linderholm H, Lindgren U. Prediction of spirometric values in patients with scoliosis. Acta Orthop Scand 1978;49: Hepper NNG, Black LF, Fowler WS. Relationship of lung volumes to height and arm span in normal subjects and in patients with spinal deformity. Am Rev Respir Dis 1965;91: Black LF, Hyatt RE. Maximal respiratory pressures: normal values and relationship to age and sex. Am Rev Respir Dis 1969;99: Weber B, Smith JP, Briscoe WA, Friedman SA, King TKC. Pulmonary function in asymptomatic adolescents with idiopathic scoliosis. Am Rev Respir Dis 1975;3: Gold WM. Pulmonary function testing. In: Nadel M, editor. Textbook of respiratory medicine. Vol 1. 2nd ed. Philadelphia: WB Saunders; p Muirhead A, Conner AN. The assessment of lung function in children with scoliosis. J Bone Joint Surg Br 1985;67: Leatherman KD, Dickson RA. Risk and reward. In: Leatherman KD, Dickson RA, editors. The management of spinal deformities. Woburn (MA): Butterworth-Heinemann; p Mankin HJ, Graham JJ, Schack J. Cardiopulmonary function in mild and moderate idiopathic scoliosis. J Bone Joint Surg Am 1964;46: Kafer ER. Respiratory and cardiovascular functions in scoliosis. Bull Eur Physiopath Resp 1977;13: Jackson RP, Simmons EH, Stripinis D. Coronal and sagittal plane spinal deformities correlating with back pain and pulmonary function in adult idiopathic scoliosis. Spine 1989;14: Shannon DC, Riseborough EJ, Kazemi H. Ventilation perfusion relationships following correction of kyphoscoliosis. JAMA 1971; 217: Dollery CT, Gillam PMS, Hugh-Jones P, Zorab PA. Regional lung function in kyphoscoliosis. Thorax 1965;20: Cooper DM, Rojas JV, Mellins RB, Keim HA, Mansell AL. Respiratory mechanics in adolescents with idiopathic scoliosis. Am Rev Respir Dis 1984;100: Cook CD, Barrie H, DeForest SA, Welliesen PJ. Pulmonary physiology in children. III. Lung volumes, mechanics of respiration and respiratory muscle strength in scoliosis. Pediatrics 1960; 25: Suppliers a. SensorMedics Corp, 1630 S State College Blvd, Anaheim, CA b. Micro Medical Ltd, PO Box 6, Rochester, Kent, ME1 2AZ England. c. SAS Institute Inc, SAS Campus Dr, Cary, NC d. SPSS, Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL