We attempted to contact all of the patients who were listed in the outpatient files at our institution with a recorded diagnosis of hereditary multiple exostoses. Information was available for twenty-eight patients (fifty-six extremities), and they were asked to return for a complete evaluation. Twelve patients were female and sixteen were male. The ages of the patients at the time of the evaluation ranged from five to fifty years (average, twenty-one years) (Fig. 1). A detailed history was recorded, and a general physical examination was performed with specific attention to the upper extremities. The range of motion of the upper extremities (dorsiflexion and volar extension of the wrist and of the elbow, radial and ulnar deviation of the wrist, and pronation and supination of the forearm) was measured with use of a goniometer.
Radiographs were made of the upper extremity, and the radial inclination (the angle between a line drawn perpendicular to the axis of the radius and the articular orientation of the distal aspect of the radius) was measured on the anteroposterior radiograph (Fig. 2). We attempted to measure radial tilt (the volar inclination of the radial articular surface with respect to the long axis of the radius) on the lateral radiographs. However, due to the severe distortion of the osseous landmarks, reproducible measurements were not possible. Ulnar variance (the difference between the lengths of the radius and ulna on an anteroposterior radiograph of the wrist) was measured with use of the method of Hafner et al. (Fig. 3). This method allows meaningful measurements despite severe alteration of the contour of the distal aspect of the ulna and radius. Radiographic evidence of degeneration of the joints was evaluated as being either present or absent. If evidence of narrowing of the joint space, sclerosis of adjacent subchondral bone, or osteophytes was noted, the joint was classified as degenerative. The articulation between the radial head and the capitellum was graded as entirely normal, subluxated (when more than 25 per cent but less than 50 per cent of the articular surface was not in contact with the capitellum), or dislocated (when more than 50 per cent of the articular surface was not in contact with the capitellum).
Each patient (or, if appropriate, the parent) was asked to rate the function of the extremity according to specific criteria (Table I). An extremity that had totally normal function received a score of 5 points, so that a patient who had normal function of both upper extremities had a total score of 10 points.
The function of the hand was evaluated with use of the hand test of Jebsen et al.7,16. The test consists of seven subtests designed to evaluate the strength and dexterity of the hand in the completion of specific tasks. The tests include (1) writing a short sentence, (2) turning over three-by-five-inch (7.6-by-12.7-centimeter) cards, (3) picking up small objects and placing them into a container, (4) stacking checkers, (5) simulating eating, (6) moving large empty cans, and (7) moving large weighted cans. The time needed to perform the tasks was recorded for each hand. The test was administered by an occupational therapist on our staff who was experienced in its use. Previous reports have established the reliability of this test for both adults and children7,16. Age-grouped normal values are available with standard deviations for each subtest. Grip strength was also measured and was compared with published normal values1,9. With use of the Guidelines to Evaluation of Permanent Impairment, published by the American Medical Association, physical impairment (as distinguished from disability) was calculated for each extremity on the basis of the loss of motion of the upper extremity only. By combining the scores for both upper extremities (again ignoring any impairment attributable to involvement of a lower extremity), each patient was assigned a rating for so-called whole-person impairment.
Our philosophy for the treatment of hereditary multiple exostoses has been to offer excision only to patients who have painful exostoses. No patient in this study group had had an excision, a corrective osteotomy, or a lengthening procedure.
Radiographs revealed abnormalities of the radial head in ten (18 per cent) of the fifty-six extremities. Subluxation of the radial head was noted in nine extremities (16 per cent) and dislocation was noted in one (2 per cent), although only two (4 per cent) of the fifty-six extremities had symptoms at the elbow. Three (5 per cent) of the fifty-six extremities had evidence of degenerative joint disease.
Functional Assessment
The grip strength was compared with known age-matched-control values. The strength on the side of the dominant upper extremity in seven (25 per cent) of the twenty-eight patients was below the fifth percentile for their age. The strength on the side of the non-dominant extremity in ten (36 per cent) of the patients was below the fifth percentile for their age.
The scores from the hand test of Jebsen et al.7,16 were converted to z-scores ([score of hand test—mean]/standard deviation) to allow statistical analysis of results from patients of different ages and to allow comparison of results from dominant and non-dominant extremities. Despite the physical deformity, the study population scored below two standard deviations from the mean in only one of the subtests of Jebsen et al.—simulated eating with use of the non-dominant extremity. Over-all, the average score was in the forty-seventh percentile for the dominant upper extremity and in the twenty-eighth percentile for the non-dominant extremity. With the numbers available, there was no significant difference between the function of the extremities that had a normal radial head and those that had an abnormal radial head in any of the seven subtests of the hand test of Jebsen et al.7,16.
Age
The effect of the age of the patient on deformity, function, and impairment was evaluated in two ways. First, the Pearson correlation coefficients were used to test age as a variable throughout the entire range of ages represented in the study population. Radial inclination was found to increase with increasing age (p = 0.004, r = 0.38). None of the other variables were found to be correlated with age when tested in this way. Second, we postulated that deformity progresses with age but becomes stable at skeletal maturity (Table II). As there were few patients in our series who were in late adolescence, the patients were divided into two groups: those who were ten years old or less and those who were more than ten years old. Pronation decreased from an average of 87 degrees in the younger patients to an average of 66 degrees in the older patients (p = 0.001). Supination decreased from an average of 90 degrees to an average of 84 degrees (p = 0.001). The subjective-assessment score decreased from an average of 9.7 points to an average of 8.1 points (p = 0.001), and the impairment rating increased from an average of 2.3 per cent to an average of 5.1 per cent (p = 0.005). The status of the radial head was significantly different between the age-groups (p = 0.03). Ulnar variance; dorsiflexion, volar flexion, and radial deviation of wrist; and radial inclination were not found to be related to age with use of either method of analysis.
Status of the Radial Head
The status of the radial head (normal, subluxated, or dislocated) was significantly related to radial deviation (p = 0.001), ulnar deviation (p = 0.009), pronation (p = 0.001), extension of the elbow (p = 0.04), radial inclination (p = 0.005), negative ulnar variance (p = 0.008), and impairment rating (p = 0.001) (Table III). With the numbers available, the status of the radial head was not significantly related to the age or sex of the patient, supination, flexion of the elbow, dorsiflexion, volar flexion, grip strength, or the subjective-assessment score. The status of the radial head had no effect on the performance of the patients in any of the seven subtests of the hand test of Jebsen et al.7,16.
Physical Impairment
The ratings for physical impairment were determined by measuring the range of motion of the extremity2, as described. We wanted to determine if physical impairment measured with this method was actually associated with the functional assessment as reflected by a performance evaluation (the hand test of Jebsen et al.7,16).
The average rating for impairment was 7 per cent (range, 0 to 28 per cent) for the dominant extremity and 8 per cent (range, 0 to 25 per cent) for the non-dominant extremity. The so-called whole-person impairment rating averaged 5 per cent (range, 0 to 17 per cent). With the numbers available, the impairment rating was not significantly associated with the sex of the patient or with the grip strength of the extremity, but it was associated with age when grouped (p = 0.005), negative ulnar variance (p = 0.001, r = 0.56), the status of the radial head (p = 0.001, r = 0.47), the subjective rating score (p = 0.005, r = 0.37), and subtests 2 (p = 0.03, r = 0.28), 3 (p = 0.04, r = 0.27), 4 (p = 0.006, r = 0.36), 5 (p = 0.005, r = 0.37), and 6 (p = 0.009, r = 0.35) of the hand test of Jebsen et al.7,16. Subtests 1 and 7 (writing a short sentence and moving large weighted cans) were not associated with the impairment rating.
The characteristic deformity of the upper extremity that is seen in hereditary multiple exostoses consists of severe shortening of the ulna with angular deformity of the distal aspect of the radius. The wrist drifts into ulnar deviation, and often a gradual bowing of the entire radial shaft is apparent. The radial head may be subluxated or dislocated. Motion may be restricted, especially pronation and supination of the forearm. Burgess and Cates evaluated the specific deformities of the forearm in thirty-five patients who had multiple cartilaginous exostoses. Detailed evaluation of serial radiographs revealed a correlation between radial and ulnar shortening (r2 = 0.86). They found that the forearm as a unit showed over-all retardation of growth, and the theory that the shortening of the ulna results in the subsequent shortening of the radius was not supported by their data. Angular deformity of the distal aspect of the radius was not correlated with negative variance of the ulna (r2 = 0.30) or with carpal slip. Although their series was small, it appeared that severe shortening of the ulna of more than 8 per cent was significantly associated (Fisher exact test, p = 0.002) with dislocation of the radial head. The function of the forearm was not evaluated in their study.
The operative treatment of deformities of the forearm in patients who have hereditary multiple exostoses has been described in many reports3,8,10,11,13,17. All have addressed the excision of the exostoses in an attempt to avoid deformity or the correction of existing deformity with use of osteotomy or stapling of the radial physis, or both. Although improved function was discussed in each article, an objective measurement of function, either before or after the operation, was not included.
In our patients, a dislocated radial head was found to be significantly associated with a loss of pronation (p = 0.001), a higher impairment rating (p = 0.001), and greater ulnar variance (p = 0.008). However, ulnar variance alone cannot be the single greatest factor causing the radial head to dislocate. The three patients who had the greatest negative ulnar variance all had a normal radial head. Ulnar variance was not associated with age. The severity of the retardation of growth at the distal aspect of the ulna may be genetically determined and, thus, in a sense, preprogrammed to occur. The suggestion that excision of an exostosis at the distal aspect of the ulna will allow the retarded growth of the distal ulnar physis to accelerate and approach normal is purely conjectural. The loss of pronation was significantly associated with an increased impairment rating (as this rating is derived from measuring loss of range of motion) and with a worse score for the status of the radial head, but several patients who had severe loss of pronation had a normal score for the status of the radial head. Osteochondromas impinging on the interosseous membrane may be responsible for the loss of pronation, but this possibility was not evaluated in the present study. If other factors are responsible for the loss of function, maintenance of the radial head in its correct location may not stop or reverse the loss of function in these patients.
The loss of function of the forearm is probably related to a very complex relationship of multiple factors. The shortening of the ulna (perhaps as a percentage of the length of the forearm), the degree and direction of radial bowing, impingement on the interosseous membrane, the status of the radial head, and other factors may all interact in a way that we cannot predict at present. Whether operative attempts to correct these progressive deformities actually result in measurable improvement in function remains to be studied.
It is clear that deformities in the upper extremity due to hereditary multiple exostoses are variable but well tolerated by most patients. Even when rated solely on the basis of loss of range of motion, the average physical impairment was slight. The objective measurements showed that the measurable clinical function of the hand is largely preserved within the degree of deformity that is customary for these patients. This attests to the adaptability of growing children when confronted with a slowly developing deformity. The subjective-assessment scores were high, but we made no attempt to evaluate the patient's perception of the cosmetic deformity.
Previous reports on hereditary multiple exostoses have focused on various operative methods to address the physical deformities of the forearm4,8,10,13,17. Although the combination of ulnar lengthening and stapling of the radial physis appears to be effective for the treatment of the cosmetic deformity, we know of no data that support the supposition that this approach leads to an appreciable increase in function. As the status of the radial head was not associated either with the subjective-assessment score or with the performance on the hand tests of Jebsen et al.7,16, operative attempts to preserve this articulation may not lead to any predictable improvement in function. We think, as do most of our patients and their parents, that operative treatment of the deformity of the forearm is justified, if only for cosmetic improvement. However, reports in the orthopaedic literature on hereditary multiple exostoses have suggested that operative treatment is intended to improve function, and it is clear that this goal has not yet been documented4,8,10,13,17. The present report provides an objective scheme for the assessment of function that may be applied before and after operations on patients who have hereditary multiple exostoses, thus allowing any actual functional benefit from operative procedures to be documented.