Authors: David M. Bennett
Categories: Current Concept Review, Adolescent kyphosis, Scheuermann's kyphosis, Non-surgical treatment, Spinal health, Long-term prognosis
Source: Journal of the Pediatric Orthopaedic Society of North America
Authors: David M. Bennett
Adolescent idiopathic kyphosis, also known as Scheuermann's kyphosis (SK), is a prominent form of thoracic or thoracolumbar spine hyperkyphosis affecting up to 8% of adolescents. This condition is characterized by structural vertebral changes and a kyphotic Cobb angle greater than 45°. This review examines the long-term progression and outcomes of SK when untreated, integrating findings from multiple decades of research, including landmark studies. By engaging with this review, we hope to convey a consolidated understanding of how Scheuermann's kyphosis behaves across the lifespan when untreated; from the modest rate of post-maturity progression to the clinical signals that separate stable curves from those likely to cause functional difficulty. The information can provide clarity when advising families, planning follow-up, and determining when continued non-operative care is reasonable or when surgical intervention might be considered.
(1)Scheuermann's kyphosis affects roughly eight percent of adolescents and is defined by structural vertebral wedging that produces a thoracic Cobb angle greater than 45°.(2)Longitudinal cohorts show that after skeletal maturity, the curve enlarges only about a third to half a degree each year, with very few patients ever exceeding 100°.(3)Pain and everyday function are usually acceptable, yet individuals whose curves pass 75° and develop forward sagittal imbalance report noticeably greater disability.(4)Radiographic size alone does not predict symptoms; the combination of deformity and loss of global alignment is a more reliable signal for clinical vigilance.(5)Conservative measures such as physiotherapy and bracing during growth suffice for most patients, while operative consultation is reserved for those with progressive imbalance or persistent discomfort.
Adolescent kyphosis, most often referring to Scheuermann's kyphosis (SK), is a rigid hyperkyphosis of the thoracic or thoracolumbar spine that arises during adolescence due to anterior wedging of vertebral bodies. It is distinct from postural kyphosis in that SK features structural vertebral changes (Schmorl's nodes, endplate irregularities, etc.) and a kyphotic Cobb angle greater than 45°. The incidence of SK is reported to be as high as approximately 8% in adolescents. A question is how such kyphotic deformities evolve throughout the patient's life if managed non-surgically [1]. For the past 60 years, numerous studies, ranging from classic long-term follow-ups to recent cohort analyses, have examined the natural history of untreated adolescent kyphosis and the risk of progression into adulthood [2].
The radiographic criteria for Scheuermann's disease were established by Sørensen in 1964, requiring at least 3 adjacent vertebrae with ≥5° of wedging. Early observations indicated that SK often causes pain during the teenage growth spurt, but this pain typically diminishes after skeletal maturity. In Sørensen's original follow-up, about 50% of adolescents with SK reported thoracic back pain, while only ∼25% experienced pain after reaching adulthood. This pain was usually mild and “not incapacitating”. Such findings led to the notion that the natural history of Scheuermann's kyphosis might be relatively benign once growth ceases [1]. However, other foundational work hinted at less optimistic outcomes for some patients. Bradford (1977) reported that adults with a history of Scheuermann's kyphosis had a higher incidence of significant back pain than individuals without SK [3]. Likewise, Lowe (2004) noted that patients with severe deformities (>75°) of untreated SK can develop chronic thoracic pain from degenerative changes and experience substantial functional limitations [4]. These contrasting views called into focus the idea that long-term consequences might depend on the magnitude of the kyphosis and prompted further systematic research.
One of the first rigorous long-term studies of untreated Scheuermann's kyphosis was conducted at the University of Iowa by Murray, Weinstein, and Spratt (1993). They followed 67 SK patients (average kyphosis ∼71° at diagnosis) for an average of 32 years and compared outcomes with 34 age-matched controls. This landmark study found that, decades after adolescence, patients with SK reported more frequent and intense back pain than controls and experienced slightly reduced trunk extension strength and range of motion.
Importantly, however, major aspects of quality of life were not significantly there were no major differences in educational level, days of work missed due to back issues, interference from back pain with daily activities, or social and self-image measures between the kyphosis group and controls. Patients with SK were no more likely to use pain medication than controls and were “little preoccupied” with their physical appearance. Murray et al. concluded that while untreated SK can cause some functional limitations and chronic back pain, these typically do not result in severe disability or major life interference. In their cohort, kyphotic curves generally stabilized after growth, and progression was not a significant finding in most patients (few, if any, progressed beyond the extreme 100° curvature range). Notably, only patients who had kyphosis >100° (which was rare) showed restrictive lung deficits; those with curves <100° had normal pulmonary function on long-term follow-up. Mild scoliosis was a common incidental finding, but no cases of spondylolisthesis were observed. Murray's influential 1993 study painted the natural history of Scheuermann's kyphosis as fairly benign in terms of life impact, an interpretation consistent with the Pediatric Orthopaedic Society's summary that kyphosis <100° has a benign course (pain being the main issue).
In the past 2 decades, researchers have revisited the natural history of SK with larger cohorts, control comparisons, and modern health metrics. Ristolainen et al. (2012, Finland) conducted a 37-year follow-up of patients diagnosed with Scheuermann's in their youth. They surveyed 49 untreated SK patients (mean age 59 at follow-up) alongside approximately 3,800 population controls, focusing on back pain and functional status. This study found that Scheuermann's patients had significantly higher odds of chronic back pain and activity limitations in adulthood compared to the controls. For instance, the SK group had an odds ratio of about 2.5 for chronic back pain and around 3.7 for experiencing back pain in the past month, relative to pain rates in the general Finnish population. They were also more than 5 times more likely to report difficulty climbing stairs and nearly 7 times more likely to have trouble carrying a modest load (5 kg for 100 m). Notably, however, Ristolainen et al. found no correlation between the degree of kyphosis and the presence or severity of pain or disability. In other words, having Scheuermann's disease increased the risk of back pain, but within the SK group, a larger curve did not predict worse pain. Their conclusion was that untreated SK patients do have a higher long-term risk of back pain and some disability in daily activities, but the kyphosis angle itself was not a reliable predictor of quality of life or health status [5].
Ristolainen et al. (2017, Finland) extended this work by examining radiographic progression in a subset of these patients. They tracked 19 untreated SK individuals who had standing lateral spine X-rays in their youth and again approximately 46 years later (mean age approximately 65 at follow-up). This study, published in the Journal of Orthopaedic Science, provides the first direct quantification of kyphosis progression over several decades. The results showed that the mean thoracic kyphosis increased from 46° at baseline to 60° at follow-up – an average increase of approximately 14° over 46 years (P < .001). This corresponds to a very slow progression rate of roughly 0.3° per year. Importantly, about two-thirds of the patients experienced at least a 5° increase in kyphosis over time, indicating that progression, albeit slight, occurred in the majority of cases. The wedging of vertebrae also progressed modestly (mean wedge angle per vertebra increased from 8.8° to 9.9°, P = .046). Consistent with their 2012 findings, Ristolainen et al. reported no significant predictors of progression – neither the initial curve size, patient's sex, age at baseline, nor length of follow-up correlated with the amount by which the kyphosis increased.
Moreover, the extent of curve progression did not correlate with clinical patients who progressed more were not functionally worse off than those who progressed less. This study concluded that “the degree of radiographic deformity progressed slightly during long-term follow-up” and that progression magnitude “did not predict symptoms.” Still, in comparing these patients to population norms, the authors noted that SK patients (especially males) had higher rates of back pain and disability than their peers (e.g., adjusted odds of back pain in the last 5 years were ∼5.4 times higher for male SK patients). Thus, even moderate SK tends to worsen a bit radiographically in adulthood, but this mild progression does not necessarily translate into proportionately worse pain or functional decline [6].
Ragborg et al. (2020, Denmark) reported on a 39-year follow-up of non-operated Scheuermann's patients originally seen in a pediatric clinic between 1972 and 1982. Out of 55 identified cases, 38 participated in this study (mean age 53 at follow-up), with 34 undergoing new radiographs. This cohort was analyzed for health-related quality of life (HRQoL) outcomes using the SRS-22r and SF-36 questionnaires, and results were compared to normative data. The location of the kyphosis apex was also considered (thoracic versus thoracolumbar). Similar to other studies, Ragborg et al. found that untreated SK patients in mid-adulthood have slightly lower HRQoL scores than the general population in specific domains. In particular, the SK group scored significantly worse on the SRS-22r Pain and Self-Image domains (P = .049 and .006, respectively), indicating more pain and lower satisfaction with appearance.
Other domains, such as SRS Function and Mental Health, and the total SRS-22r score showed a trend toward lower values in SK patients, but did not reach statistical significance. Pulmonary function testing in a subset indicated no significant differences between those with thoracic versus thoracolumbar kyphosis apex, and there was no evidence of major respiratory compromise in this mid-life group, consistent with prior findings that only extreme kyphosis greater than 100° causes pulmonary restriction. Interestingly, the location of the kyphosis (high thoracic versus lower thoracolumbar) did not significantly affect outcomes in this series. Both apex groups exhibited comparable spinopelvic alignment parameters and HRQoL scores, except for a slightly lower SRS-22r Function score in the thoracolumbar group. Although detailed radiographic progression data were not fully quantified in the publication, the authors illustrated that kyphosis generally persists into middle age with no spontaneous correction; in fact, all patients who had serial radiographs displayed some degree of progression over the decades, even if modest. The conclusion was that untreated Scheuermann's leads to mild long-term decrements in pain and self-image measures, but global quality-of-life is only subtly affected, and significant functional impairment is not the norm [7].
Garrido et al. (2021, UK) provided one of the most data-rich analyses of long-term curve progression and its clinical correlates. In this study, 113 patients with Scheuermann's kyphosis who had never undergone surgical treatment were identified through a national spine deformity database. Of these, 66 patients (mean age 45 years) were evaluated approximately 27 years after their initial diagnosis (which occurred around the end of growth in most cases); 47 consented to new spinal radiographs for progression analysis. The radiographic results confirmed that untreated kyphosis often continues to increase slightly even long after skeletal maturity. The mean thoracic kyphosis in this group was 66° at skeletal maturity and progressed to 78° by the time of follow-up, representing a significant 12° increase. This equates to approximately 0.45° of progression per year over adult life, which is very much in line with the slow rate found by Ristolainen.
Every patient in the radiographic subset experienced an increase in kyphosis (P < .001 for the cohort's change). Clinically, Garrido et al. found that untreated SK patients reported worse outcomes than age-matched norms, corroborating findings from other studies. Both the SRS-22 questionnaire (which measures spine-specific health status) and the SF-36 (general health survey) indicated significantly lower scores in SK patients for pain and functional domains, along with higher scores on the Oswestry Disability Index (ODI, which indicates more disability). Importantly, this study further identified predictive factors linking radiographic alignment to symptoms. Using multivariate regression, they discovered that greater sagittal imbalance – reflected as an increased sagittal vertical axis (SVA) – was moderately correlated with poorer ODI scores (r = 0.59) and lower SRS-22 pain and function scores (P = .001). In other words, patients with kyphosis accompanied by a forward shift in posture (imbalance) tended to have more pain and functional difficulties.
Similarly, larger kyphosis curves and compensatory cervical alignment changes were linked to a poorer self-image on the SRS-22. The authors summarized that the long-term progression of SK averaged approximately 0.5° per year, and that sagittal alignment parameters such as SVA and total kyphosis can predict which patients experience worse outcomes in terms of pain, function, and self-image. This suggests that not only does untreated kyphosis tend to slowly worsen, but if it worsens in a way that the body cannot fully compensate (leading to forward imbalance), patients are more likely to experience significant symptoms [8].
A unique perspective stems from a recent population-based study of older adults. Jönsson et al. (2023, Sweden) evaluated spinal X-rays and pain reports in a cohort of 1,417 men (all in their 80s) to identify those with radiologic signs of past Scheuermann's disease. Interestingly, they found a 5.3–7.9% prevalence of definitive Scheuermann's changes in these elderly men, consistent with lifetime incidence estimates. Crucially, the presence of Scheuermann's in these octogenarians was not associated with increased back or neck pain compared to those without it. About 51% of men with radiographic SK had back pain, which was slightly less than the 55% in those without SK (difference not significant). None of the SK-identified men reported severe back pain; most classified their pain as mild or moderate, a distribution very similar to controls. This suggests that a subset of individuals with Scheuermann's kyphosis can reach old age without any greater pain burden than their peers – indicating a variable long-term trajectory. It is possible that those with milder deformities or better compensatory adaptation continue to cope well, while others (perhaps with larger curves in mid-life) might have faced more issues or even undergone surgery earlier, thus not represented in this elderly cohort. Nonetheless, the Swedish study supports the notion that untreated Scheuermann's does not invariably cause late-life pain or disability [9].
Across these studies, which span various populations and follow-up durations, several consistent themes emerge regarding the lifelong implications of untreated adolescent
Unlike adolescent idiopathic scoliosis (where post-maturity progression tends to occur mainly in larger curves), Scheuermann's kyphosis displays a tendency for slow, continuous progression even after skeletal maturity. Multiple cohort studies with serial radiographs (Finland 2017, UK 2021) documented an average increase in kyphosis of about 0.3–0.5° per year during adulthood. For a patient, this might imply that a thoracic curve measuring, say, 60° at age 18 could be approximately 70–75° by their 50s. Not every patient will progress to that extent. Still, these studies indicate that the majority (well over 50%) of untreated SK patients experienced measurable curve increases over time and the progression rate varies, and predictors are limited.
The natural history of Scheuermann's kyphosis appears to be less disabling than one might expect from the radiographic deformity. Across multiple long-term follow-ups, most patients – even those with substantial kyphosis – were found to be leading essentially normal lives in terms of work status, daily activities, and social function. Back pain is the most common long-term complaint, and studies consistently show a higher prevalence of back pain in adults who had SK as teens compared to those who didn't [10]. However, this pain is often described as intermittent or manageable (“mild and not incapacitating” in many cases). Physical function tends to be only mildly reduced in untreated SK. It's often only the subset with the most extreme curves that shows notably inferior physical function or needs significant lifestyle modifications [8,11].
In line with earlier clinical teaching, cardiopulmonary compromise is generally not a concern in untreated Scheuermann's kyphosis unless the curve exceeds approximately 100°. Neurologic issues are uncommon; the kyphotic deformity, even when progressive, typically does not encroach on the spinal cord—unlike some cases of untreated congenital kyphosis that carry a risk of paralysis.
David M. Bennett: Writing – review & editing, Writing – original draft, Supervision, Project administration, Investigation, Formal analysis, Data curation, Conceptualization.
The author declares that no patient consent was necessary as no images or identifying information are included in the article.
The author affirms that all possible conflicts of interest, including financial, consultant, institutional, and other relationships that might lead to bias, have been disclosed above. If no additional conflicts exist, this is explicitly stated as none declared.
Funding from organizations such as the National Institutes of Health, Wellcome Trust, Howard Hughes Medical Institute, or others was not received for the work reported in this manuscript.
During the preparation of this work the author(s) used ChatGPT models o3 and 4.5 (OpenAI) in order to refine, coalesce, and synthesize the research into a more readable format. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the publication.
None.
Dr Bennett holds a patent for a spinal hook implant; serves as a consultant for 2nd MD (secondary educational consultations); collaborated with Johnson & Johnson on USS spinal implants, which concluded more than 2 years ago.