Enchondromatosis is a rare disorder characterized by anomalous enchondral ossification leading to multifocal hyperplastic cartilaginous enchondromas that usually affect the diaphysis or metaphysis of long bones. Although previously known by various names, it is perhaps best known as Ollier disease, named for the physician who first described it in 18981,2. The understanding of this disease has evolved, so that there are now numerous subtypes that differ according to the various clinical and radiographic findings associated with multiple enchondromas. A review of the literature suggests that enchondromatosis may not have a typical uniform pattern of clinical appearance3-12.
We present two cases of patients with extreme enchondromatosis characterized by the development of massive cartilage tumors. We believe that these cases represent a new variant of generalized enchondromatosis because of the slow progressive growth of the cartilage tumors. Both patients developed severe bilateral extremity and spine involvement. The clinical presentation and imaging characteristics are reviewed. A strategy for clinical surveillance of these patients is presented. The patients and their families were informed that data concerning their cases would be submitted for publication and they consented.
Case 1. A girl was seen at the age of five months because of skeletal abnormalities noticed on a screening pelvic radiograph. At birth, she was noted to have a depression on the left side of the chest and respiratory distress requiring intubation with positive pressure ventilation. Examination also revealed decreased motion of the left hip, and pelvic radiographs revealed irregularities of the proximal femoral metaphyses and epiphyses. A radiographic skeletal survey revealed abnormalities throughout the skeleton with a predisposition for the long bones. The spine had platyspondyly with beaking of the lumbar vertebral bodies. Involvement of the extremities was greater in the proximal metaphyses. A diagnosis of "enchondromatosis with spinal involvement" was established.
The patient was followed closely with observation and imaging. At the age of seven years, she had hemangiomas of the left hand in addition to the multiple large enchondromas, and the diagnosis was changed to Maffucci syndrome variant. At the age of ten, symmetric phalangeal deformity of both hands (including the characteristic "light-bulb" deformity of the digits) due to the ongoing expansion of multiple enchondromas had progressed so that the hands had become functionally impaired (Fig. 1). The upper extremities had multiple large lesions. The lower extremities were deformed by massive proximal femoral lesions that made hip motion impossible, and superficial wounds occurred (Figs. 2 and 3).
Spinal involvement was noted when the patient was six months of age, but radiographs of the spine demonstrated only mild platyspondyly and no scoliosis. By the age of two years and three months, the patient had scoliosis with a left thoracic curve of 11° and a right lumbar curve of 15° as well as thoracic kyphosis of 45°. By the age of four years, this had progressed to a 70° left thoracic scoliosis with a compensatory 45° to 50° lumbar curve, despite the use of a thoracolumbosacral orthosis. A vertical expandable prosthetic titanium rib (VEPTR; Synthes, West Chester, Pennsylvania) was implanted at the age of five years and was expanded five times over the next three years, after which time L3 and L4 pedicle instrumentation was included (Fig. 4).
At the time of writing, the patient required the use of a wheelchair for mobility because of a progressive decline in walking ability secondary to proximal femoral involvement. She has normal intelligence and mental development. She was undergoing regular surveillance imaging and physical examination for potential malignant degeneration of the enchondromas.
Case 2. A girl was first referred for evaluation at the age of five years. A product of a full-term pregnancy without complications, she was noted shortly after birth to have short extremities. Radiographs demonstrated that she had generalized bilateral enchondromatosis with involvement of the spine.
The patient was managed with serial observation and imaging until the age of twelve years, by which time enlarging bilateral proximal humeral lesions were restricting shoulder motion (Fig. 5). By the age of thirteen years, abduction of the left shoulder was limited to 50°, with no internal or external rotation because of progressive enlargement of the proximal humeral enchondroma (Fig. 6). Surgical debridement of a portion of the 20-cm tumor from the proximal part of the humerus to improve shoulder motion and rule out malignant degeneration was done. Shoulder motion was markedly improved postoperatively, with abduction to 95° and internal and external rotation to 40° and 60°, respectively. Tissue from the lesion revealed large nodules of hyaline cartilage separated by bands of loose fibrovascular connective tissue and fragments of hematopoietic marrow (Fig. 7), with minimal mitotic activity. Pathologic examination did not support the diagnosis of malignant degeneration of this enchondroma.
The patient also had substantial deformity of both hands and both femora. Spinal involvement initially included platyspondyly and vertebral stippled calcifications without scoliosis. By the age of seven years, she had developed a left thoracolumbar curvature of 40°, which was initially treated with a brace but ultimately was managed with posterior spinal fusion (Fig. 8). At the time of writing, the patient relied on a wheelchair for mobility. She had normal intelligence.
The recent clinical course has been complicated by the development of an enlarging lesion of the left parotid gland. Computed tomography examination also demonstrated a large enchondromatous lesion of the sphenoid with compression of the subarachnoid cistern (Fig. 9). The findings of a biopsy of the parotid mass were consistent with a low-grade mucoepidermoid carcinoma, which was treated with proton-beam radiation therapy.
In the first description in the literature, enchondromatosis was named dyschondroplasie by Louis Xavier Édouard Léopold Ollier in 18981,2. At that time, the distinction among enchondromatosis, multiple hereditary exostoses, and various other irregularities of enchondral ossification was not clearly defined. The original description tended toward unilateral involvement. Hunter and Wiles were the first to attempt to define Ollier disease as its own entity, in 1935, identifying key distinguishing features such as a lack of heredity, childhood onset, and unilateral involvement13. More recently, Spranger et al. argued that unilateral enchondromatosis is in fact simply "a particular topographic form of enchondromatosis and not a separate entity."12 The name Ollier disease survives for largely historical reasons.
The pathogenesis of enchondromatosis is still not clearly understood. A genetic basis for some cases of enchondromatosis has been suggested, although a mendelian inheritance pattern has not been identified8,11,13-15. The predominantly unilateral distribution that is most commonly seen suggests a post-zygotic, somatic mutation in a mesenchymal cell line early during embryogenesis that is expressed in a mosaic pattern. This form of mosaicism has been described mostly in the dermatologic literature involving unilateral cutaneous conditions and has not been exclusively evaluated in studies of enchondromatosis.
The true incidence of enchondromatosis is not known, although it has been estimated to be one in 100,000 births. Maffucci syndrome, a condition with osseous manifestations that are identical to those in enchondromatosis but with the additional feature of numerous hemangiomas predominantly in the subcutaneous tissue and occasionally in the viscera, is even less common5,15-18. "Extreme" enchondromatosis appears to be the rarest of all of the variants. Although both of our patients were female, there is no evidence to support any sex predilection. In previous reports, the youngest age at which any of the enchondromatosis variants was diagnosed was nine months10,11. The age at diagnosis of five months in our Case 1 is the youngest reported to date.
Attempts have been made to classify the various forms of enchondromatosis. Spranger et al. defined six subtypes19. Halal and Azouz later added three subtypes to this classification system on the basis of case reports of enchondromatosis with various other associated findings9. There are, however, several other rare subtypes that do not fall into this classification system and for which the nomenclature is highly variable in the literature8,12,14,19,20. This modified Spranger classification system is widely used to categorize cases of enchondromatosis, and the common subtypes are listed in Table I.
The subtype cheirospondyloenchondromatosis is distinguished by a particularly severe and diffuse involvement of the appendicular skeleton, especially in the hands and feet (the root cheiro- comes from the Greek word meaning hand). Onset is usually in the first year of life. The characteristic finding in the spine is platyspondyly, although scoliosis is also seen. Asymmetric growth of the extremities relative to the torso can be observed. Delayed mental development has been noted12. This subtype appears to most closely fit the two cases described in the present report. To our knowledge, a case of cheirospondyloenchondromatosis with such extreme and diffuse axial and appendicular deformities has never been reported. Neither of our patients demonstrated delayed mental development. If these cases do not represent a distinct clinical entity, they must be recognized as an extreme subvariant of a previously described subtype.
Case 1 is remarkable for the combination of severe deformity of the appendicular skeleton, spinal involvement that is characteristic of cheirospondyloenchondromatosis, and soft-tissue hemangiomas. The last finding is classically associated with Maffucci syndrome, a condition in which the spine is typically not affected. While the current literature seems to describe cheirospondyloenchondromatosis and Maffucci syndrome as two distinct clinical entities, our Case 1 seems to show them as existing on a continuum. Interestingly, a review of the literature did reveal a case described as Maffucci syndrome with substantial spinal involvement, although neither detailed descriptions of the spine nor radiographic images were provided15.
Sarcomatous degeneration of enchondromas is a known complication in patients with enchondromatosis, with rates reported in the literature ranging from 25% to 50% in patients with Ollier disease and up to 100% in those with Maffucci syndrome10,21,22. Sixteen cases of sarcoma (including twelve secondary chondrosarcomas) were reported in fifty-five patients with Ollier disease in a Mayo Clinic series, a prevalence of 29%10. It has been suggested that the actual cartilaginous volume in these patients may predict the potential for malignant change, although this has not been substantiated22. The clinical picture of enlarging painful enchondromas should prompt an evaluation for possible malignant degeneration.
The definition of proper surveillance of these patients deserves discussion. Although theoretical, we believe that the likelihood of malignant degeneration or development of other neoplasms in these patients is high and warrants close surveillance. We recommend annual physical examination and additional imaging of painful lesions or a new deformity, particularly in the spine. As these patients are skeletally immature, growth of the enchondromas would be expected until skeletal maturity, as was demonstrated by our cases. Using increasing size as an indication for biopsy in a skeletally immature patient would result in multiple surgical procedures with a low diagnostic yield. Surgery in an immature patient should be considered for palliative attempts to improve function, and the decision whether to operate should be made on an individual basis. The cartilage lesions should not enlarge once the patient is finished growing, so an enlarging lesion, especially with pain, should be considered possibly malignant. A biopsy should be performed to evaluate for high-grade malignant lesions, but it will often be nondiagnostic when it is done in an attempt to differentiate a benign enchondroma from a low-grade malignant chondrosarcoma and should not be used for treatment decisions in this scenario. This dilemma was illustrated by Case 2, in which a rapidly enlarging mass was biopsied because of concerns about malignant degeneration, and no histological evidence of malignant change was identified.
Other neoplasms that have been associated with enchondromatosis include osteogenic sarcoma, chondroid chordoma, intracranial glioma, and ovarian mesenchymal tumors10,22-26. The discovery of a parotid tumor in Case 2 is an example of this association, although one that has not been previously described, to our knowledge. Although we did not observe visceral malignant lesions in our patients, the highly aggressive nature of the disease could suggest a propensity toward visceral malignant lesions, as is seen in Maffucci syndrome. For these reason, we perform yearly abdominal ultrasound examinations to screen the viscera for malignant change while avoiding the radiation exposure of annual computed tomography scans. Any suspicious finding would then be evaluated with computed tomography or magnetic resonance imaging as needed. In addition, spinal involvement requires serial examination and imaging to monitor for progressive scoliosis.