An Osteological Study on the Prevalence of Osteochondromas

Abstract

Background:

Osteochondromas are frequently reported to be the most common benign tumor of bone. However, there is not definitive data on their prevalence rate as they are often clinically silent, with previous studies estimating 1-3%.

Methods:

We examined a large osteologic collection of 2954 skeletons to identify likely osteochondromas. One author examined all bones excluding the skull in each skeleton for atypical cortical projections. These candidates were then evaluated by an orthopedic resident and then a fellowship trained pediatric orthopedic surgeon for final inclusion using strict criteria.

Results:

13 cases of osteochondroma were identified from 2954 skeletons to yield a prevalence of 0.44% (95% confidence interval 0.20% – 0.68%) in the study population. All were located on long bones: three in the humerus, six in the femur and four in the tibia.

Conclusions:

We found a prevalence of 0.44% for osteochondromas, which is approximately half the lowest value reported in previous literature.

Introduction

Osteochondromas are defined according to WHO classification as a cartilage-capped bony projection arising on the external surface of bone and containing a marrow cavity that is continuous with that of the underlying bone.¹ The lesions can be pedunculated or sessile, and they have been reported as the most common benign tumor of bone based on pathological and radiographic series, although the true relative prevalence is unclear.^1-5 Osteochondromas typically arise in children at the metaphysis of long bones, especially the distal femur, proximal humerus, proximal tibia and fibula. It is rare to find involvement of the carpal or tarsal bones, patella, sternum, skull or spine, although any bone can be affected.^{1, 2} There is no sexual predilection for this disorder.¹

Osteochondromas are slow growing lesions with rare malignant transformation but can raise cosmetic concerns and impinge on local structures, including nerves, vessels and tendons, leading to symptomatology.^2,6,7 The majority of lesions are solitary, but they can occur as multiple lesions as part of the syndrome hereditary multiple exostosis (HME), which affects approximately 15% of patients with osteochondromas and is an autosomal dominant inherited disorder.^4,9,10

Earlier studies of osteochondromas have estimated prevalence at 1-3%. These are based on estimates of patients undergoing detailed radiographical evaluation, and pathology series.^7,10 Pedunculated osteochondromas are considered to be more common. However, to our knowledge, true prevalence data does not exist for osteochondromas, with estimation difficult given that clinical prevalence studies are highly subject to bias and as many of these lesions are discovered incidentally. Osteochondroma incidence is reported as 35% of benign and 8% of all bone tumors, though this is considered an underestimate as most are asymptomatic.¹ Most cases are diagnosed within the first three decades of life, commonly in children or adolescents between 10 and 15 years of age.⁸ This study looked at a large osteological collection to estimate prevalence of osteochondromas.

Methods

This study examined 2954 skeletons in the Hamann-Todd Osteological Collection for evidence of osteochondromas. This collection is maintained at the Cleveland Museum of Natural History and complete demographic information is available for 2499 of the specimens. The skeletons were from individuals that ranged between less than one year and 80 years of age, with an average age of 51 ± 13 years. There were 83 specimens of children aged between one year of age and 17 years. The sexual distribution was 2121 males and 378 females. The racial distribution was 1631 Caucasian, 859 black, five American Indian, and four other. The majority of the skeletons were complete with all of the bones except the skull examined.

One author (GG) screened all bones for potential osteochondromas. Special emphasis was placed on the long bones of the arm and leg. Bones with abnormal boney outgrowths then examined by an orthopedic resident (DW) to help rule in and out specimens before they were examined a final time by an attending fellowship-trained pediatric orthopedic surgeon (RL) for inclusion to the final count of osteochondromas.

Figure 1.

A. Anterior lateral proximal tibia B. Anterior medial distal femur (bone cut during processing) C. Posterior medial proximal tibia D. Anterior lateral distal femur. E. Posterior lateral proximal tibia. F. Anterior medial proximal humerus. G. Posterior lateral midshaft femur. H. Posterior lateral proximal tibia. I. Anterior lateral proximal femur. J. Posterior midshaft femur. K. Anterior proximal humerus L. Anterior lateral proximal femur. M. Posterior medial midshaft humerus.

Table I.

Specimen	Age	Sex	Race	Location of Lesion
HTH 0088	35	Male	White	Tibia
HTH 0186	68	Male	White	Femur
HTH 0452	38	Male	White	Tibia
HTH 1086	21	Male	Black	Femur
HTH 1831	26	Male	Black	Tibia
HTH 1925	30	Female	Black	Humerus
HTH 2379	67	Male	Black	Femu
HTH 2388	45	Male	White	Tibia
HTH 2678	40	Male	Black	Femur
HTH 3115	57	Male	White	Femur
HTH 3250	42	Male	Black	Humerus
HTH 3261	58	Male	White	Femur
HTH 3385	44	Male	Black	Humerus

Criteria for inclusion included location on the metadiaphyseal side of the growth plate, direction of growth away from the growth plate and the shape of the ossification. We excluded florid reactive ossifications, which were identified as large disorganized growths of osseous material, most often around joints; traction exostoses; and possible sessile osteochondromas as they could not be adequately identified or differentiated from other normal boney variants.

Results

Out of 2954 skeletons examined, >100 examples of abnormal ossification were found. These included normal variants, infections, florid reactive periostitis, traction exostoses and osteochondromas. Careful review produced 13 highly likely pedunculated osteochondromas (Figure 1, Table 1). We therefore report a prevalence of 4.4 per 1000 (0.44%, 95% confidence interval 0.20% – 0.68%) in the sample population. All of the lesions were located on long bones: four (31%) were located on the tibia, six (46%) on the femur and three (23%) on the humerus. We did not locate any definite lesions on the flat bones in this study. Ten (77%) were located near the metaphysis of the bones, while three (23%) were located in the diaphysis.

All of the osteochondromas were found on skeletons that had complete demographic information. We compared the demographic information of affected individuals to the subset of the museum collection that had complete demographic information (2499 of the 2954 total subjects). The average age of skeleton that included an osteochondroma was 44 years, while the average age of the specimens in the collection was 51 years. One female specimen had an osteochondroma, for a prevalence of 0.26%. The male prevalence was 0.56%. There were 5 lesions in black specimens, for a prevalence of 0.81%. The remaining 8 tumors were in white specimens, for a prevalence of 0.37%.

Discussion

In our evaluation of a large osteologic collection we report a prevalence value of 4.4 per 1000 (0.44%). This is lower than estimated by other reports. This is one of the few direct estimates of this lesion on a random large population, and the only one that uses an osteological collection to evaluate this lesion. We believe our results are lower than the literature values for several reasons. First, although the bones are well maintained, it is possible that some pedunculated tumors could have been lost through handling and processing of the collection. We believe that this is unlikely to be a major source of error though, as damage would likely have left behind remnants of the stalk or a hole in the cortex of the bone that could have been identified. Second, it is possible that additional tumors on smaller and irregular bones, like the bones of the carpus or the vertebrae, were missed. Third, we only included pedunculated tumors and none of the sessile osteochondroma variants. Pedunculated lesions are considered more common than sessile, but earlier reports we reviewed did not give an analysis of the relative rates. Sessile osteochondromas could not be adequately differentiated from the numerous traction exostoses that were identified in the collection. Fourth, there have been reports of spontaneous resorption of osteochondromas, although this is felt to be fairly uncommon.^11,12 Finally, our screen included pediatric skeletons where a potential future osteochondroma might not have formed yet, although the proportion of immature skeletons in the collection is proportionally small. Despite all these possibilities, it is also possible that given that previous reports likely were influenced by a higher capture rate secondary to symptomatic or prominent osteochondromas, our data is a more realistic representation of the true disease prevalence.

Evaluating the demographics of specimens with osteochondromas did not reveal any obvious factors correlated with development of a lesion. The prevalence amongst females was 0.26% and amongst males was 0.56%. However, with the smaller number of females represented within the collection there was just one female osteochondroma, making statistical comparison impossible. The prevalence in whites was 0.37% and in blacks it was 0.81%, and were similarly at low enough prevalence that meaningful analysis was not possible. Finally, the average age of the specimens with an osteochondroma was 44 years, versus an overall average of 50 years for the collection, which suggests that the lesions did not significantly alter lifespan in this population.

Summary

In this study we found an incidence of osteochondromas of 0.44%. We did not find a significant sexual or racial predilection for development of solitary lesions in this cohort. Lesions were most likely to develop on long bones, with no observed lesions on flat bones. This data suggests a lower prevalence of osteochondroma than previously thought.