If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Patients with esophageal atresia (EA) are at risk of scoliosis, musculoskeletal asymmetries, reduced physical fitness and motor skills.
•
What is new information.
•
In adolescents with EA, scoliosis and musculoskeletal asymmetries are related to reduced physical activity and impaired motor skills.
•
Most of the EA adolescents with scoliosis did not have vertebral anomalies.
Abstract
Background
Increased risk of scoliosis and musculoskeletal abnormalities in adolescents with esophageal atresia (EA) is reported, but the impact of these abnormalities on physical fitness and motor skills are not known.
Methods
Scoliosis was assessed radiographically and shoulder and chest abnormalities by a standardized protocol. Physical fitness was evaluated with Grippit, Six-minute walk test, and International Physical Activity Questionnaire and motor skills by Motor Assessment Battery for Children.
Results
Sixty-seven EA adolescents median 16 (13–20) years participated. The prevalence of significant scoliosis (≥ 20º) was 12% (8/67) whereas 22% (15/67) had mild scoliosis (10–19º). Vertebral anomalies occurred in 18/67 (27%), eight of them (44%) had scoliosis. The majority of adolescents (15/23) with scoliosis did not have vertebral anomalies. Musculoskeletal abnormalities were detected in 22–78%. Balance problems occurred three times more frequently than expected (44% vs. 15%, p = 0.004). Submaximal exercise capacity was significantly reduced compared to reference values (p < 0.001). Scoliosis ≥ 20º was related to reduced physical activity (p = 0.008), and musculoskeletal abnormalities to reduced physical activity and impaired motor skills (p = 0.042 and p < 0.038, respectively).
Conclusions
Significant scoliosis was diagnosed in 12% of the EA adolescents and related to reduced physical activity. Musculoskeletal abnormalities identified in more than half of the patients, were related to reduced physical activity and impaired motor skills, and exercise capacity was significantly below reference group. EA patients with and without vertebral anomalies need health-promoting guidance to prevent impaired motor skills and consequences of reduced physical activity.
Children with esophageal atresia (EA) are at risk of developing scoliosis, musculoskeletal abnormalities, impaired motor skills and reduced exercise capacity [
]. Motor skills and exercise capacity are vital in keeping up with peers in daily activities during childhood and in adolescence. The prevalence of scoliosis and musculoskeletal abnormalities in children with EA and impact on physical fitness at different ages are poorly understood.
A recent systematic review reports scoliosis after open EA repair with a prevalence ranging from 3% to 67% [
]. EA patients have increased risk of developing both primary scoliosis related to vertebral anomalies and secondary scoliosis without vertebral anomalies. Scoliosis may cause pain, reduced physical fitness and may affect daily activities and quality of life. Scoliosis and musculoskeletal abnormalities are well-recognized complications after neonatal thoracotomy [
We speculated that EA adolescents with scoliosis and musculoskeletal abnormalities have more difficulties with muscle strength, functional exercise capacity, daily life physical activities and motor skills.
The primary aim of our study was to assess the prevalence of scoliosis, vertebral anomalies, and musculoskeletal asymmetries in the shoulder and chest wall in adolescents with EA.
Our secondary aim was to present physical fitness (muscle strength, functional exercise capacity and daily physical activities) and motor skills in EA adolescents, and to explore if scoliosis and musculoskeletal abnormalities are risk factors for reduced physical fitness and impaired motor skills.
2. Material and methods
2.1 Participants
All survivors with EA born between January 1996 and December 2002 (n = 125) and operated in one of the three tertiary hospitals in Norway were eligible. We excluded (n = 7) patients with neuromuscular or cognitive impairments, e.g. cerebral palsy, muscular dystrophy and trisomy 21. All assessments were performed between 2015 and 2017.
2.2 Design
This cross-sectional study was conducted in the out-patient clinic during interdisciplinary follow-up at our tertiary level university hospital. In addition to clinical and motor assessments the EA adolescents filled out a questionnaire mapping physical activity.
2.3 Medical characteristics
Demographics and clinical data were retrieved from medical records;
(1)
Demographic data: sex, gestational age (GA), prematurity and birth weight.
(2)
Clinical data: EA classification according to Gross [
X-ray examinations of the spine were evaluated by two experienced radiologists. Vertebral anomalies were recorded and Cobb angle was evaluated for scoliosis. Scoliosis was divided into Cobb angle 10–19°, 20–45° and > 45° [
All standardized assessments were performed following a written procedure by an experienced pediatric physical therapist (UIM). We examined the adolescents in the upright position.
Shoulder asymmetry was identified as a difference of ≥ 2 cm between the left and the right acromion height (yes /no) [
Winged scapula was defined as a prominent medial border of the scapula towards the chest wall. Assessment was done in a resting position and during movement with full flexion and abduction of the arms (yes/no) [
]. The highest peak strength values were recorded. As normative values we used the age-appropriate peak grip strength for the right and left hand reported in Swedish children [
The validated six-minute walk test (6MWT) was used according to the manual to assess the submaximal level of functional exercise capacity reflecting activities of daily living [
. We used the 95% confidential interval for gender and age categories 12–15 years and ≥ 16 years obtained from a reference group of healthy Austrian children [
We evaluated self-reported physical activity with the validated short Norwegian version of the International Physical Activity Questionnaire (IPAQ), which is a questionnaire of physical activity related to daily life during the past seven consecutive days [
]. The results were calculated into Metabolic Equivalent Task minutes per week scores (METs/week), and then categorized into a level of physical activity: high, moderate, or low [
]. A total MABC-2 test-score consists of the sub-scores: manual dexterity (fine motor skills) and gross motor skills (ball and balance skills). We used the 11–16 years age band, with the reference data from the United Kingdom to classify motor performance as normal (percentile score > 15), at risk for motor delay (percentile score 6–15), or motor delay (percentile score < 6) [
Data are summarized as numbers (%), mean (SD or range), and median (range) as appropriate. MABC-2 results are presented as categorical data. For comparisons, we have used Mann Whitney tests as appropriate. Correlation analyses were calculated with Spearman's rho. For comparing outcome data in 6MWT with reference data, the results in 6MWT and reference data were converted to z-scores [
]. Normative data values for specific percentiles (1, 2.5, 5, 10, 25, 50, 75, 90, 95, 97.5, 99) were Box-Cox transformed to remove skewness and meet the requirements for a Gaussian distribution. From these a normality plot (x = transformed value, y = the z-value in the standard normal distribution associated with the percentile) was constructed and estimates for the expectation value μ and standard deviations were calculated. Patient data values (X) were treated likewise. From the value Y = Box-Cox transformed X, the z-score = (Y- μ)/s was calculated and incorporated in normality plot. To be biologically relevant, the mean z-score difference between the study group and the reference group must supersede the Cohen's d = 0.2 criterium. Statistically significant values were accepted at the level p = 0.05. Data are analyzed with the use of SPSS Statistics version 25 (IBM, Armonk, NY).
2.8 Ethics
Informed written consent was obtained from all parents and patients. The study has obtained approval from The Norwegian Regional Ethics Committee for Medical Research (2014/1224/REK).
3. Results
3.1 Patients
Among the 125 EA adolescents identified 16 patients (13%) died because of associated major anomalies or serious complications. Of the 109 eligible, seven patients were excluded (Fig. 1). Thirty-four patients (33%) declined participation and 68 adolescents attended follow-up. As one adolescent refused radiographic examination, we were able to include 67 (66%) patients median 16 (13–20) years of age.
Fig. 1Flowchart of the included and not included patients. A total sample of 67 adolescents with esophageal atresia (EA) included in the data analysis.
Medical characteristics are listed in Table 1. There were no statistical differences in basic data between the 67 participants and 35 non-participants.
Table 1Demographics and clinical evaluations of participants and non-participants.
At follow-up 44/67 (66%) had no scoliosis (< 10°). Scoliosis between 10° and 19° was observed in 15 (22%) patients, 20–45° in seven (10%) and > 45° in one (1.5%) patient (Fig 2). One patient with Cobb angle 37° at follow-up had several serious comorbidities with vertebral/costal anomalies and severe early onset scoliosis > 120° This patient had been operated on several times with rib based distraction and vertebral column resection. None of the other patients with scoliosis needed correctional surgery and these patients (12 girls and 10 boys) had a mean Cobb angle of 19° (SD 11) at mean 16.5 years of age (SD 2.2). There was a significant correlation between scoliosis (≥ 10º) and age (r = 0.452, p = 0.03, Fig. 2). All patients with > 20° curvature were 16 years or older and were considered mature.
Fig. 2Scatterplot of adolescents with esophageal atresia (EA) with Cobb angle ≥10° and age (years). Significant correlation between scoliosis ≥ 10º and age (years) (r = 0.452, p = 0.03).
In total, 18/67 (27%) adolescents were diagnosed with vertebral anomalies, located at the cervical-thoracic level in eight patients, at the lumbar-sacral level in six, and in four adolescents the vertebral anomalies were located in both the upper and lower spine. Among the patients with vertebral anomalies 8/18 (44%) had scoliosis. Fifteen of the 23 (65%) with scoliosis did not have vertebral anomalies. In the patients without vertebral anomalies the curvature was in the upper thorax and convex to the left in 12 out of 15 patients. The median Cobb angle of the eight patients with scoliosis and vertebral anomalies was 27° (10–37°) compared to median 12° (10–48°) in the patients with scoliosis but no vertebral anomalies (p = 0.185). Fifteen of 49 patients (31%) without vertebral anomalies developed scoliosis (≥ 10°). In the patients with minimal curvature (10–19°), 3/15 (20%) had vertebral anomalies compared to 5/8 (63%) in the patients with large curvatures (≥ 20°).
3.4 Assessment of physical fitness
3.4.1 Muscle strength
When using the Grippit instrument, a significant difference in grip strength between males and females for right and left hand was detected, p = 0.004 and p < 0.001 respectively, with all mean peak grip strength results lower and outside the reference range values [
]. But peak grip strength between right and left hand did not differ significantly in either males or females (p = 0.574 and p = 0.762, respectively).
3.4.2 Functional exercise capacity
One patient interrupted the test because of pain in both legs. The median walking distance assessed with 6MWT was 643.5 (473–780) meters. The mean z-score of the study group was - 0.87 (SD 1.20), being significantly different from that of the reference group with a mean z-score = 0 (p < 0.001) (Fig. 3). Twelve adolescents (18%) had lower functional exercise capacity than the lowest reference limit [
Fig. 3Z-score plot of six-minute walk test in adolescents with esophageal atresia (EA) (n = 66, one missing data). The blue dotted line shows the EA adolescents and the red dotted line shows the reference group [
]. In addition to be statistically significant, the mean z-score difference = - 0.865 also superseded the Cohen's d = 0.2 criterion, thus being of biological importance.
Forty adolescents (60%) did not meet the national and international recommendation of 60 min of daily physical activity during one week for adolescents (calculated to at least consisting of 1764 METs/week) [
. IPAQ total score did not differ between males and females (p = 0.922), with self-reported daily physical activity levels classified as low in 38%, moderate in 35% and high in 27%.
3.5 Assessment of motor skills
Forty-one adolescents (61%) met the appropriate age-criteria for evaluation with MABC-2 scores. Twelve patients (29%) were classified with definite motor function delay (Fig. 4). Balance skills were most frequently affected with only 56% scoring within normal range (Fig. 4).
Fig. 4Motor skill score results (normal, at risk, and delayed) evaluated with Motor Assessment Battery for Children, Second Edition (MABC-2) in esophageal atresia (EA) adolescents (n = 41). Reference data (percentile rank), total motor score, fine-, and gross (ball and balance) motor skill scores. The EA adolescents scored significantly worse in balance skills compared to reference data (p = 0.004).
Self-reported physical activity was significantly positively correlated to functional exercise capacity (r = 0.300, p = 0.014).
Low self-reported physical activity was correlated to scoliosis ≥ 20º and to identified winged scapula (r = −0.537, p = 0.008 and r = −0.251, p = 0.042, respectively). Total motor skills, with balance skills most frequently affected, were negatively correlated to atrophy of the right pectoralis muscle (r = −0.529, p < 0.001), and fine motor skills correlated to shoulder asymmetry and atrophy of right pectoralis major muscle (r = −0.361, p = 0.022 and r = −0.330, p = 0.038, respectively).
4. Discussion
In this study we confirmed our suspicion that scoliosis and musculoskeletal abnormalities are related to physical fitness and motor skills in adolescents born with EA.
One third of the EA adolescents had scoliosis which was correlated to reduced daily physical activity. Interestingly, the majority of patients with scoliosis, 15/23 (65%), did not have vertebral anomalies. This suggests that EA patients without vertebral anomalies also need regular follow-up visits into adolescence or adulthood to evaluate scoliosis. Exercise capacity in EA adolescents was, compared to reference values, significantly reduced and shoulder asymmetry and atrophy of the right pectoralis major muscle, were related to impaired motor skills, of which balance skills were the most affected.
Development of scoliosis after EA surgery and other thoracic procedures in children is well known [
Prevalence and natural history of scoliosis and associated congenital vertebral anomalies in patients operated for esophageal atresia with or without tracheoesophageal fistula.
. The EA patients in our study developed scoliosis 10 times more frequently than the normal population. This corresponds with other studies reporting a scoliosis incidence of 18–67% in EA patients [
Prevalence and natural history of scoliosis and associated congenital vertebral anomalies in patients operated for esophageal atresia with or without tracheoesophageal fistula.
Table 2Algorithm for relevant examinations by a pediatric physical therapist in a multidisciplinary follow-up program for children born with esophageal atresia, from infancy throughout childhood into adolescence.
Domains
Assessments
Relevance / intervention
Infancy (0–1 years)
Motor development
Age-appropriate locally available formal tests
Early referral pediatric physical therapist and provide guidance in follow-up procedure
Musculoskeletal development
Structured physical examination
Early referral pediatric physical therapist and orthopedic surgeon if indicated
X-ray spine if indicated
Preschool age (2–5 years)
Motor development
Age-appropriate locally available formal tests
Referral pediatric physical therapist Motor activity through play
Musculoskeletal development
Structured physical examination
Referral pediatric physical therapist and orthopedic surgeon if indicated
X-ray spine if indicated
School age
(10–12 years, before puberty)
Motor development
Age-appropriate locally available formal tests
Referral pediatric physical therapist
Musculoskeletal development
Structured physical examination
Referral pediatric physical therapist and orthopedic surgeon if indicated
X-ray spine
Physical fitness
Age-appropriate locally available tests
Encourage to regular physical activity Sports participation / exercise training
- Exercise capacity
- Muscle strength
Adolescence
≥ 13 years into adulthood
Motor development
Age-appropriate locally available formal tests
Referral pediatric physical therapist
Musculoskeletal development
Structured physical examination
Referral pediatric physical therapist and orthopedic surgeon if indicated
Associated congenital vertebral anomalies are frequently seen in patients with EA and the incidence varies between reports (4–45%). This is in accordance with the 27% found in our study [
Prevalence and natural history of scoliosis and associated congenital vertebral anomalies in patients operated for esophageal atresia with or without tracheoesophageal fistula.
]. Patients with EA may develop primary scoliosis because of vertebral anomalies, but in our study patients with vertebral anomalies and scoliosis did not have significantly greater spinal curvatures than patients with scoliosis without vertebral anomalies.
The age of the child is of great importance when diagnosing and treating scoliosis. Sistonen et al. reported scoliosis in 56% of 100 adult EA patients after muscle cutting posterolateral thoracotomy, of which 44% had mild scoliosis (10–20°) and 12% with more severe curvature (> 20°) [
]. They reported a greater number of patients with mild scoliosis, and we therefore cannot ignore the possibility of mild scoliosis becoming overt during adolescence and into adulthood. This would also be suggested by the significant correlation between scoliosis ≥ 10º and age in EA adolescents in our study.
As we did a cross-sectional study of our cohort, the data do not allow determination of age of onset of scoliosis. However, it could be assumed that this happens during puberty when skeletal growth is most rapid [
]. This implies that EA children, irrespective of vertebral anomalies, need prolonged follow-up to identify those with Cobb angle > 20° for referral to an orthopedic surgeon.
A systematic review by Mishra et al. reported a 13% prevalence of scoliosis in EA children without vertebral anomalies (n = 937) [
]. Most patients in Mishra´s review were younger than the patients in our study, as scoliosis increases with age this may be one reason for the apparent higher prevalence of scoliosis in our study.
In our patients as in most EA patients with thoracic scoliosis without vertebral anomalies, the spinal curvature is convex to the left, i.e. opposite the right posterolateral thoracotomy [
Prevalence and natural history of scoliosis and associated congenital vertebral anomalies in patients operated for esophageal atresia with or without tracheoesophageal fistula.
]. Thus, there may be multiple factors at play in the development of spinal curvatures in EA patients.
With regards to thoracotomy techniques, acquired rib fusion related to tight intercostal closure has been discussed as the cause of the development of scoliosis [
Prevalence and natural history of scoliosis and associated congenital vertebral anomalies in patients operated for esophageal atresia with or without tracheoesophageal fistula.
. Sistonen et al. found rib fusions in 30% of adults after thoracotomy, and 60% of these patients developed scoliosis, concluding that rib fusion was a possible risk factor for scoliosis [
]. In contradiction to this, Soliman et al. reported 17% rib fusion in 106 EA children and found no significant relationship between rib fusion and the development of scoliosis [
Prevalence and natural history of scoliosis and associated congenital vertebral anomalies in patients operated for esophageal atresia with or without tracheoesophageal fistula.
]. Unfortunately, we were unable to examine postoperative chest X-rays, so we cannot conclude on the frequency of tight intercostal closure and rib fusions in our cohort.
Atrophy of right pectoralis major muscle was registered in more than two thirds of our EA adolescents. This corroborated by a previous report by Cherup et al. after posterolateral thoracotomies in children [
]. We also found impaired grip strength in both hands and no difference between the right and the left hand. Previous studies have demonstrated that in case of right-handed dominance, the strength in the right hand is better than the left hand [
. Lack of strength in a dominant right hand, in combination with atrophy of the right pectoralis major muscle, may pose extra challenges for children and adolescents with EA.
Safa et al. identified in a cohort of 104 children, including 51 EA patients after thoracic procedures including thoracoscopy, musculoskeletal deformities in 39% and scoliosis in 16%. Division of the serratus anterior muscle was in Safa´s study the only significant predictor of developing musculoskeletal deformities, most notably scoliosis. They concluded that the risk of musculoskeletal deformities might be reduced by thoracotomy with muscle sparing technique [
]. Interestingly, in a prospective follow-up study, wherein 94 out of 106 EA patients had a muscle sparing technique, Soliman reported a 49% prevalence of scoliosis [
Prevalence and natural history of scoliosis and associated congenital vertebral anomalies in patients operated for esophageal atresia with or without tracheoesophageal fistula.
]. This may also confirm that EA children, despite minimal traumatic thoracic approach, should be followed-up throughout childhood with the awareness of musculoskeletal abnormalities and scoliosis.
Daily physical activities below national recommendation were reported by more than half of our patients, and reduced daily physical activity and impaired motor skills were related to scoliosis and musculoskeletal abnormalities. We speculate that musculoskeletal asymmetries after thoracotomy may affect trunk stability, leading to impaired gross motor performance, and particularly impaired balance skills. The reasons why the EA adolescents seem to exercise less than reference population are probably multifactorial. We suggest that lack of muscle strength and reduced submaximal exercise capacity are important factors and may discourage physical activity and hence lead to impaired motor performance and reduced physical activity. For a young EA patient physical activity and exercise capacity are important when it comes to socializing with peers, promoting somatic health along with improving cognitive, emotional, and psychosocial development [
]. Earlier studies on quality of life have reported low scores in physical functioning in EA adolescents, which may also be explained by reduced physical performance as registered in our study [
. Therefore, we advocate a multidisciplinary follow-up protocol including standardized examinations by a dedicated pediatric physical therapist for all children with EA throughout childhood into adolescence, to be aware of the risk of developing scoliosis and to provide tailor-made lifestyle counselling (Table 2).
Our study has some limitations: A limited number of patients were assessed for motor development, because the widely used MABC-2 is only validated up to 16 years of age. Validated tests for clinical assessments of muscular asymmetries are missing. The evaluations were, however, performed by standardized assessments following a written procedure, and therefore, we think the observations may contribute to the knowledge on long-term musculoskeletal abnormalities in adolescents with EA.
Closure of the intercostal space may impact on development of scoliosis [
, and our study is limited in that we were not able to evaluate the closure of the intercostal space in our patients. On the other hand, a systematic review which showed a prevalence of scoliosis in EA adolescents of 13%, found no effect from rib fusion [
]. The impact of intercostal space closure on scoliosis is therefore still unknown.
Strengths of the current study are the relatively large sample of EA adolescents included, particularly concerning the scoliosis evaluations, and that all participants were evaluated by an experienced pediatric physical therapist in one single tertiary level clinic. The instruments used for the evaluations of physical fitness and motor skills are all well validated tests. All assessments (both clinical examinations and questionnaires) were performed during the same day.
5. Conclusion
Scoliosis and musculoskeletal asymmetries were diagnosed in more than one third of the EA adolescents and the prevalence of scoliosis of ≥ 20° is high (12%). The recognition of the frequent occurrence of scoliosis in patients without vertebral anomalies, that may not become apparent until the child is older, also provides a warning and possible need for closer follow-up to identify patients needing timely referral for orthopedic evaluation.
Exercise capacity in EA adolescents was significantly lower compared to the reference population and the prevalence of scoliosis and musculoskeletal abnormalities were significantly related to reduced physical activity and impaired motor skills. These sequelae may affect daily life activities with peers and thus affect social life in young EA patients.
Declaration of interest
None
Acknowledgment
The authors would like to thank all adolescents born with EA, along with their parents, for participation and positive response which made this cross-sectional study possible.
This research is generated within the European Reference Network for rare Inherited and Congenital Anomalies (ERNICA) - Project ID No 739544 (not financially supported).
Prevalence and natural history of scoliosis and associated congenital vertebral anomalies in patients operated for esophageal atresia with or without tracheoesophageal fistula.
VACTERL, Vertebral defect, Anorectal malformations, Cardiac defect, Trachea-Esophageal fistula, Renal abnormalities and Limb abnormalities.
00658-3&id=gr1.jpg){kind=link}
00658-3&id=gr2.jpg){kind=link}
00658-3&id=gr3.jpg){kind=link}
00658-3&id=gr4.jpg){kind=link}