Summary
The prevalence of adolescent idiopathic scoliosis (AIS) has come from epidemiological and school screening studies. This chapter explores the benefits and limitations of extrapolation of prevalence data from these studies and summarizes them based on country, age group, gender, and curve types. This chapter describes the natural history of untreated AIS in terms of curve progression and the impact of untreated scoliosis on back pain, cardiopulmonary function, and psychosocial issues from long-term follow-up studies.
Key words
adolescent idiopathic scoliosis – prevalence – natural history – curve progression – back pain – psychosocial issues3 Prevalence and Natural History
3.1 Introduction
The prevalence of any condition is defined as the number of patients in an existing normal population, either during a period of time (period prevalence) or at a particular date in time (point prevalence). This should not be used interchangeably with incidence, which refers to the number of new occurrences of a condition in a population over a period of time. The presence of scoliosis is defined by the Scoliosis Research Society (SRS) as a lateral curvature of 10 degrees or more, and our knowledge of the prevalence of scoliosis has come from both epidemiological studies and school screening reports around the world.
However, these types of studies may not accurately inform us the true prevalence of scoliosis. First, different class intervals of curve magnitude are not directly comparable. For example, studies that use intervals of 0–5, 6–10, 11–15 degrees, and so forth will have slightly different results than studies that use 0–4, 5–9, and 10–14 degrees as their interval class. Second, the aim of school screening for scoliosis is for early detection of adolescent idiopathic scoliosis (AIS) and prescription of spinal bracing to prevent curve progression. Typically, students aged 10 years or above are deemed eligible for screening, but the upper age limit varies in different countries. This “selective screening” process allows the greatest proportion of at-risk individuals to be screened and is more effective in any healthcare system where resources are finite. However, this represents a source of inaccuracy when one extrapolates information regarding the prevalence of scoliosis from these studies. Third, there is a large diversity in study design, screening tests used, frequency of screening, and duration of follow-up. In particular, most studies do not have all screened students followed up until skeletal maturity. Insufficient follow-up may affect our understanding of the prevalence of scoliosis, as some referred students with nonsignificant curves may later progress, while scoliosis may develop during adolescence in other screened students who did not have the condition at the time of first screening. With this background, we can evaluate the known prevalence of idiopathic scoliosis in the population.
3.2 School Screening
Data from school screening studies conducted in Germany, Korea, Brazil, Singapore, Hong Kong, China, Turkey, and Greece have contributed to our understanding of the overall prevalence of AIS. The German study conducted by the Robert Koch Institute consisted of 17,641 children aged between 0 and 17 years and examined for chronic diseases. 1 The study in Brazil included 1,340 children with a mean age of 12.7 years. 2 In Singapore, 72,699 children aged 6 to 14 years were included in a 1-year prospective epidemiologic study. 3 A population-based cohort study of 394,401 children aged between 10 and 19 years entered into the screening program in Hong Kong was analyzed. 4 An epidemiological study conducted in eastern China included 79,122 students aged between 10 and 16 years. 5 A total of 3,175 children aged 10 to 15 years were enrolled in the study conducted in Turkey. 6 An analysis of 17 school screening programs in Greece that included 130,689 children between 1975 and 1999 was performed. 7 The combined data from these studies indicate an overall prevalence of 0.47 to 5.2% for AIS. Table 3‑1 provides a summary of all the studies detailed earlier.
3.2.1 Prevalence According to Genetic Factors
There is some evidence that genetic factors and race can influence the prevalence and incidence of AIS, although no definitive cause and effect has been demonstrated, and it is likely to have a polygenic background. 8 , 9 Epidemiological and screening studies from different continents have demonstrated that prevalence of AIS may be influenced by race. Kamtsiuris et al found a higher prevalence of scoliosis in German children (5.5%) than in immigrant children (3.5%). 1 In an epidemiological study from New Zealand, Ratahi et al found that Polynesian children (Maori and Pacific Islander) had a significantly lower incidence of idiopathic scoliosis than European children. 10 In Singapore, Chinese girls in the 11- to 12-year and 16- to 17-year age groups were found to have a significantly higher rate of scoliosis than Malay and Indian girls. 11 Apart from race, Grivas et al reported a significant positive association between prevalence of AIS and latitude for girls (p < 0.001) and a curvilinear trend but not significant positive association for boys (p < 0.111). 12
3.2.2 Prevalence by Age
Studies have also found different age-related prevalence for AIS. Daruwalla et al, using a definition of a curve showing a Cobb angle of 5 degrees or more, found a higher prevalence for each higher age group: 0.12% in age 6 to 7 years, 1.0% in 11 to 12 years, and 3.12% in 16 to 17 years. 11 Using a curve size of Cobb angle ≥ 10 degrees, Yong et al reported that the prevalence of AIS increases from 0.27% at age 9 years to 2.49% at age 13 years. 13 A large population-based study in mainland China using a Cobb angle ≥ 10 degrees found that the prevalence of AIS increases from 0.97% at age 11 to 12 years, peaks at age 13 to 14 years with 8.22% and age 14 to 15 years with 8.8%, and gradually decreases to 2.92% at age 19 to 20 years. 14
3.2.3 Prevalence by Gender
From all the aforementioned studies, the female to male prevalence ratio is 2:1 that increases with age. Kamtsiuris reported a prevalence ratio of 1.5:1, with a slight increase with age. 1 Nery et al found a prevalence ratio of 2:1 without differentiation into different age groups. 2 , 6 , 7 Wong et al and Zheng et al found an increasing prevalence in girls with increasing age. 3 , 5
Moreover, several studies have reported larger Cobb angles in girls than in boys, indicating that scoliosis may progress more severely in girls. 15 , 16 , 17 , 18 It is unclear why that is the case, and it may be a genetic component in the etiology of scoliosis. Monozygous twins have a higher rate of concordance than dizygous twins, and the curves develop and progress together. 19 A nongenetic cause that may explain a predilection for girls can be a result from dysfunction of the sympathetic nervous system. 20 Lower body mass index (BMI) in girls with AIS is associated with decreased circulating leptin levels. Leptin, secreted by adipocytes, is a master hormone with many regulatory functions for growth and reproduction. It is postulated in AIS-susceptible girls that altered sensitivity to leptin results in increased sympathetic nervous activity and contributes to the development of scoliosis through earlier growth and skeletal maturation.
3.2.4 Prevalence by Curves
In terms of curve magnitudes, smaller curves are more prevalent than large curves in AIS. The prevalence decreases from 4.5% for curves of 6 degrees or more to only 0.29% for curves of 21 degrees or more. 17 This prevalence does not change with race and is similar in the modern era. A recent study in China showed that the highest overall prevalence (1.16%, 95% confidence interval [CI]: 1.09–1.24%) was found in students with Cobb angles ranging between 10 and 24 degrees, which decreased to 0.33% (95% CI: 0.30–0.38%) when the curve magnitude became moderate (25–40 degrees), while those who were diagnosed with severe AIS had the lowest prevalence of 0.02% (95% CI: 0.01–0.03%). 5 The pooled prevalence of spinal curves greater than 20 degrees was 0.22% (95% CI: 0.15–0.30%) from 36 studies on AIS. 21 In terms of curve types, thoracic curves are the most common (48%), followed by thoracolumbar/lumbar curves (40%). Double curves (9%) and double thoracic curves (3%) are less common. 18 Eight percent of all children have thoracic or thoracolumbar curves.
3.3 Natural History
Although the etiology and pathogenesis of AIS are not completely understood, many large-scale natural history studies have given clinicians a clear understanding of the sequelae of untreated scoliosis. This knowledge defines the clinical goals in managing AIS and guides clinicians on the rationale and timing of interventions. It also helps the shared decision-making process in this modern era of medicine with patients and their caregivers. The University of Iowa has published one of the longest natural history study series on AIS patients. Weinstein and Ponseti followed up 219 patients with untreated AIS and reported factors related to curve progression between 1932 and 1948. 22 Subsequently, they have reported the outcomes of 117 untreated patients at an average follow-up of 51 years compared with 62 age- and sex-matched volunteers in terms of mortality, back pain, pulmonary symptoms, general function, depression, and body image. 23 In Gothenburg, Sweden, Nachemson first reported the outcome of 117 patients with untreated scoliosis. 24 Subsequent studies of this cohort have examined the long-term sequelae in terms of back pain, psychological effects, respiratory function, and survival. 24 , 25 , 26 , 27 , 28 , 29 , 30 These studies have formed the foundation of our understanding of long-term sequelae of untreated AIS and will be discussed in the remainder of this chapter.
3.4 Curve Progression
Curve progression in AIS is probably the most important factor in the decision for interventions, as progressive curves are thought to be the major cause for back pain, cardiopulmonary dysfunction, cosmetic concerns, and psychosocial impacts. Hence, the ability to predict whether AIS curves will progress or not can affect the ensuing treatment.
The risks of curve progression can be divided broadly into two categories: (1) curve characteristics and (2) stage of skeletal growth.
3.4.1 Curve Characteristics
Curve magnitude at the time of presentation influences curve progression, with larger magnitude curves tending to have higher risks of progression. In the series reported by Rogala et al, 17 10.3% of curves between 10 and 20 degrees at presentation progressed at least by 5 degrees into the range of 20 degrees or more, while 21.2% progressed if the Cobb angles were between 20 and 29 degrees. Lonstein and Carlson 31 also reported that only 17% of curves between 15 and 19 degrees at presentation progressed, while 34% of curves between 20 and 29 degrees progressed. For patients who are skeletally mature, curves that reach 50 degrees are likely to continue to progress at a rate of 1 degree per year. 22 , 23 , 32 , 33 This forms the threshold and basis for offering surgical correction and spinal fusion in order to prevent curve progression and functional deterioration in early adulthood.
Curve location is another important factor in determining curve progression. Thoracic curves tend to increase more. 22 , 23 , 33 , 34 Edgar reported 28 patients with thoracic curves with mean Cobb angle of 73 degrees (range, 36–118 degrees) that progressed at a mean of 8 degrees over a 17-year follow-up (0.47 degree/year). 33 The detailed information from the University of Iowa series reported that after 40 years, single thoracic curves between 50 and 75 degrees progressed to 29.4 degrees with a mean of 0.73 degrees per year, although it is important to note that only 11 patients in this series had curves between 50 and 75 degrees. 22 It was also noted in this series that curves greater than 30 degrees at maturity did not experience progression and tended to have an apical vertebral rotation of less than 20 degrees and Mehta angles of less than 20 degrees. Pesenti et al conducted a retrospective multicenter review of a French cohort with a minimum 20-year follow-up consisting of 62 Lenke 1 patients, of whom 17 had a Cobb angle greater than 45 degrees in adolescence. 35 At last follow-up, 18 patients showed greater than 20 degrees of progression, 8 of whom had Cobb angle greater than 45 degrees in adolescence. The increased risk of greater than 20 degrees of progression of thoracic scoliosis with initial Cobb angle greater than 45 degrees was statistically nonsignificant.
In the Iowa series, similar to the thoracic group, thoracolumbar curves between 50 and 75 degrees at skeletal maturity progressed most, increasing on an average of 22.3 degrees during the 40-year study period. 22 Ascani et al 36 reported 14 patients with thoracolumbar curves of greater than 40 degrees and found a mean progression of 18.7 degrees after a mean 34-year follow-up (0.55 degree/year). Edgar 33 reported that 18 patients with a mean thoracolumbar curve of 54 degrees had a mean progression of 10 degrees over 17.6 years of follow-up. However, this group of patients tended to develop translator shifts with lateral listhesis during adulthood, which may give rise to back pain, and these curves become more rigid as degeneration develops, making them more technically difficult for surgical treatment in later life. In the French cohort, among 96 patients with Lenke 5 curves, of whom 55 had a Cobb angle greater than 35 degrees in adolescence, 36 patients showed greater than 20 degrees of progression. The increased risk of greater than 20 degrees of progression with an initial Cobb angle greater than 35 degrees was significant (odds ratio = 4.728; p = 0.002). 35
Lumbar curves tend to progress at a slower rate compared with thoracic and thoracolumbar curves. In the Iowa series, 17 patients with lumbar curves greater than 30 degrees had a mean progression of 16.2 degrees over 29 years. 22 A similar finding was reported by Ascani et al. 36 In these curves, factors associated with progression included right-sided curves, apical vertebral rotation of greater than 33%, and a high-riding fifth lumbar vertebra, 22 , 23 , 34 while curves less than 30 degrees, deep-seated lumbar vertebrae, sacralized fifth lumbar vertebra, and absence of translatory shift development protected against progression.
Lastly, loss of thoracic kyphosis has been shown to influence the likelihood of curve progression, 37 , 38 although the exact correlation has not yet been determined.
3.4.2 Stage of Skeletal Growth
It is well established that the age of diagnosis is associated with progression, with younger age posing the greatest risk. 22 , 31 , 36 , 39 In girls, progression risk is greatest before the onset of menarche. In general, the greater the amount of skeletal growth remaining, the greater the risk for progression. Curve acceleration begins during Risser stage 0 of the iliac apophysis. Sanders et al 40 showed that skeletal maturation scores derived with the use of the Tanner and Whitehouse (Tanner–Whitehouse 3) method using the radius, ulna, and small bone of the hand 41 were highly correlated with the curve-acceleration phase. In particular, radiographically, the timing of the curve-acceleration phase corresponded with the change from a covered (Tanner–Whitehouse 3 stage F) to a capped (Tanner–Whitehouse 3 stage G) phalangeal epiphysis.
Recently, it has been shown that at initial presentation, progressive curves show different three-dimensional morphological changes from reconstructed spinal radiographs. 42 There is some evidence to show that peak curve progression lags behind peak skeletal growth rate by approximately 7 months and, overall, the period of potential curve progression extends 1.5 years beyond the peak growth phase until skeletal maturity. 43
3.5 Back Pain
Back pain and related disabilities are not common complaints during adolescence, but back pain may develop in adulthood due to asymmetric disc degeneration, facet degeneration, and osteoporosis with asymmetric fracture collapse. However, the precise cause of back pain is relatively unknown, and it does not appear to correlate with curve severity or the presence of radiographic osteoarthritis. 23 , 34 In a 50-year natural history study, 61% of AIS patients reported chronic back pain at any level of the spine compared with 35% of controls. Current back pain was found in 77% of patients with scoliosis compared with 37% of controls. This indicated that acute and chronic back pain was more prevalent in patients with untreated scoliosis. Foraminal stenosis can develop in the concavity of the curve contributing to radicular pain. A 20-year follow-up on the Swedish patients treated with bracing found that lumbar spinal motion and muscle endurance were reduced, which in turn was correlated with more lumbar back pain than in controls. 44 An earlier study of 34 patients treated between 1935 and 1975 with thoracolumbar, lumbar, or double major curves with magnitude between 20 and 55 degrees were followed for an average of 22 years. 45 Sixty-five percent of patients reported back pain in the low back or buttocks compared with 32% in the control group, and 24% reported radicular symptoms compared with 16% in the control group. However, none of the patients underwent surgical intervention for these symptoms. A more recent minimum 20-year follow-up in the multicenter French study showed that there was no statistical significant difference in terms of back pain and quality of life for patients who had been treated by operation in adolescence or adulthood, or by observation. 35
Patients with AIS do not suffer from significant disability and are able to work and perform everyday activities. In an epidemiological study, AIS was not found to represent a disproportionate number of disability pensions. 46 The number of AIS patients requiring surgery for back pain in adulthood was also similar to that of the general population. 28
3.6 Cardiopulmonary Function
Early studies reporting mortality due to cardiopulmonary diseases largely involved a heterogeneous group of scoliosis patients and not only patients with AIS. 28 , 29 , 47 In a 50-year follow-up study of untreated scoliosis patients, the mortality rate was double that of the general population, and the majority died from kyphoscoliotic cardiopathy with cor pulmonale. 24 However, most scoliosis that develop in adolescence do not lead to respiratory and cardiac failure in the middle age. 48 There are specific consequences to the respiratory physiology that are correlated with curve sizes. Reduced exercise tolerance, diffusion capacity, and lower maximal oxygen uptake have been found even in adolescents with moderate to severe curves. 49 , 50 , 51 In a study in which patients’ curves were reconstructed in three dimensions from EOS images (large thoracic deformities with a coronal Cobb >80 degrees, thoracic lordosis >20 degrees, and apical rotation >25 degrees), moderate-to-severe impairments of pulmonary function were documented. 52 The 50-year follow-up of the Iowa patients showed that 22% of patients had shortness of breath in daily activities compared with 15% of controls. 23 This correlated with the size of the curvature, especially if the Cobb angle was greater than 80 degrees and there was a large angle of rotation.
It is possible to conclude that patients with AIS do not face an increased mortality risk from cardiopulmonary compromise, but larger curves will give rise to pulmonary symptoms.
3.7 Psychosocial Issues and Cosmesis
While most health care professionals are aware of the psychological issues faced by patients with AIS, there have been very few studies on this matter. It was reported that only 5% of studies on scoliosis included a measure of esthetic appearance, 1.48% a measure of health-related quality of life (HRQoL), and 1% disability. 53 The SRS developed an AIS-specific HRQoL questionnaire (SRS-22) that consists of function, pain, self-image and mental health domains, and a subtotal score, 54 , 55 and more information regarding the quality of life in AIS is known. Nonetheless, this instrument is inadequate to detect a wide spectrum of psychosocial issues. 56
Bengtsson et al reported on the psychological adjustment of 26 women with idiopathic scoliosis of a high degree (average curvature of 105 degrees). 25 Although most patients seemed to have good superficial psychosocial adjustment, the group was characterized by hypersensitivity and insecurity, with a tendency to dysphoric mood. Recently, increasing evidence suggests that AIS treatment has a significant negative impact on quality of life and psychological functioning. 57 , 58 Payne et al found that AIS was a risk factor for the development of depression no matter what treatment the patient has received. 59 Patients with AIS have an increased incidence of suicidal ideation and rate of alcohol abuse. Findings from other studies have indicated that AIS may have a negative effect on psychological state, development of friendships, and ability to adapt socially. 60 Scoliosis patients tend to experience negative self-suggestions (i.e., self-stimulation through negative subjective imagination) and various degrees of inferiority complex. Matsunaga et al 61 and Danielsson et al 62 have reported that approximately 40% of patients with AIS experienced feelings of loneliness and depression both during and after treatment. Edgar 33 reported that 10% of the patients received treatment for depression, and Ascani et al 36 noted that 19% had psychological disturbances.
Using a validated questionnaire, the Iowa 50-year follow-up study found that 32% of patients felt that their scoliosis had limited their life, particularly with regard to buying clothes, decreased physical activity, and increased self-consciousness. 23 However, this study did not find any mental problems needing psychiatric treatment, and there was no difference compared with the normal population. Long-term follow-up study (average of 32 years) with the Swedish cohort with moderate scoliosis (mean Cobb angle of 30 degrees) also found they had a good quality of life measured by SRS-22 and Short Form-36 (SF-36) and did not display any difference compared with the age-matched norm groups. 63
A 22-year follow-up study looking at childbearing, curve progression, and sexual function in women with scoliosis found no difference in marital status and number of children compared with a control group. 64 However, limitations of sexual function from the back were reported, largely because of participating in activities or self-consciousness about appearance. There was no correlation between curve progression and number of pregnancies, or age at first pregnancy.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
