THE
INFLUENCE
OF AGE R. NAKAMURA,
From the Department
of Orthopaedic
AND
Y. TANAKA,
SEX T. IMAEDA
Surgery (Division of Hand Surgery),
ON
ULNAR
VARIANCE
and T. MIURA Nagoya University School of Medicine, Japan
Ulnar variance was measured in 325 normal wrists and 41 wrists with Kienbiick’s disease. A positive correlation between ulnar variance and age was confirmed in normal wrists, and it was lower in males than in females. Therefore, studies comparing ulnar variance in abnormal and normal wrists require carefully selected age- and sex-matched controls. There was no significant difference in ulnar variance between Japanese with normal wrists and those affected by KienbSck’s disease, when the effects of sex and age were taken into account. Based on these results, we believe that ulnar variance is highly unlikely to be an important predisposing factor in KienbBck’s disease. Journal of Hand Surgery (British Volume, 1991) 16B: 84-88
In 1928, Hulten found that 17 out of 23 patients (74%) with Kienbiick’s disease (Kienbock, 1910) had an ulna which was shorter than the radius at the wrist by more than 1 mm and this incidence was remarkably high compared to 8% in normal wrists. Based on this finding, he proposed that negative ulnar variance is a predisposing factor to the development of Kinebock’s disease. Since that time, Hulten’s finding has been confirmed by many investigators (Axelsson, 1973; Beckenbauch et al., 1980; Gelberman et al., 1975; Mirabello et al., 1987; Persson, 1950; Stahl and Reis, 1986; Steinhauser and Posival, 1981; Watson et al., 1985) but has been questioned by others (Chan et al., 1977; Fisk, 1984; Kristensen et al., 1986; and Nathan et al., 1987). To date, this controversy has not been resolved. Ulnar variance also has attracted attention in conditions other than Kienbiick’s disease. Czitrom et al. (1987) reported that post-traumatic scapho-lunate dissociation is associated with a high incidence of negative ulnar variance, and ulno-carpal impingement syndrome has been reported to occur in wrists with positive ulnar variance (HultCn, 1928; Palmer and Werner, 1984). Ulnar variance must also be measured to evaluate residual deformity resulting from a distal radius fracture and to plan surgical radial shortening, ulnar lengthening or Sauve-Kapandji’s procedure (Kapandji, 1986). However, the accurate measurement of ulnar variance is not simple. As pointed out by Epner et al. (1982) and Palmer et al. (1982), the apparent variance changes with the position of arm during roentgenography, and Kristensen et al. (1987) have reported a pseudo-lengthening resulting from arthrosis of the radius as a consequence rather than a cause of Kienbiick’s disease. It is also possible that the ulnar variance is related to age and sex. To assess this, we studied age- and sexrelated changes in ulnar variance among normal Japanese. These values were compared to the amount of ulnar variance in patients with Kienbock’s disease. Material and methods 325 roentgenograms of wrists without abnormalities of the bony structure taken between 1986 and 1989 were 84
used as samples of normal wrists. Films from patients younger than 14 years or with a history of injury to the elbow, forearm, wrist or fingers or with a history of joint disease were excluded from the study. The normal group consisted of 203 males (62.5%) and 122 females (37.5x), ranging in age from 14 to 79 years. A standard postero-anterior projection was obtained (Palmer et al., 1982) with a film focus distance of 125 cm and the wrist fixed in neutral position by a supporter (Nakamura, 1989). The ulnar variance was measured using a template (Palmer et al., 1982), and results were recorded in 0.5 mm increments. Roentgenograms of wrists from 41 patients with Kienbock’s disease who visited our hospital during the same period also were obtained. There were 29 males (71%) and 12 females (29%) ranging in age 14 to 75 years. Roentgenography and the measurement of ulnar variance were performed under the same conditions as in the control group. 41 films of the affected side and 36 films of the unaffected side were available for study. The measurements of ulnar variance were analysed to establish the following relationships : 1. Correlation coefficient between ulnar variance and age in the normal wrists and wrists with Kienbiick’s disease. 2. Sex-related differences in ulnar variance in normal wrists. 3. The difference in ulnar variance between affected and unaffected wrists in patients with Kienbiick’s disease. 4. The difference in ulnar variance between the normal wrists and wrists with Kienbiick’s disease. Correlations and differences were considered statistically significant when p was less than 0.05. Results Ulnar variance in 325 normal wrists ranged from + 5 mm to - 4 mm with a mean + SD of 0.20 f 1.39 mm. Positive ulnar variance (more than 1 mm) was present in 108 cases (32.2x), no variance in 128 (39.3x), and negative variance (less than - 1 mm) in 89 cases (27.4%). The ulnar variance of the 41 wrists affected with THE
JOURNAL
OF HAND
SURGERY
THE
INFLUENCE
OF AGE
AND
Kienbiick’s disease ranged from + 4 mm to -4 mm. Positive ulnar variance was seen in nine cases (22x), no variance in 17 cases (41x), and negative variance in 15 cases (37%).
SEX ON
ULNAR
VARIANCE
that of females, we considered the possibility that this difference influenced in our result. To clarify this, the ulnar variance between sexes was compared in four age groups: 14 to 30, 31 to 40, 41 to 50 and 51 to 79 years. Males had a lower mean value in all groups, with statistically significant difference in two groups at the 5% level, 31 to 40, and 41 to 50. The greatest mean difference (1.61 mm) was between the youngest males and the oldest females (Table 1).
Correlation between ulnar variance and age
The correlation coefficient between ulnar variance and age in normal wrists was 0.250 in males, 0.315 in females and 0.364 overall. All coefficients were highly significant (t-test, p < O.OOl), confirming a positive correlation between ulnar variance and age (Figs 1 and 2). In wrists with Kienbock’s disease, the correlation coefficient between ulnar variance and age was 0.266 which was not significant at 5% level (t-test, 0.05 < p < 0.1; Fig. 3).
Comparison of ulnar variance between affected and unafected wrists inpatients with Kienbiick’s disease
The mean ulnar variance in affected wrists with Kienbock’s disease was lower than in unaffected wrists; 0.18 mm. This difference was not statistically significant (t-test, p > 0.5, Table 2).
Sex-related differences of ulnar variance in normal wrists Comparison of ulnar variance between normal wrists and wrists with Kienbijck’s disease
Negative ulnar variance was present in 73 males (36.1%) and 16 females (13.1%); this difference was significant (x2-test, p ~0.001). Overall, the mean .value also was lower in males than females (t-test, p< 0.01) but since the mean age among males was 12.2 years younger than
The incidence of negative ulnar variance in affected wrists was not significantly different from that in normal wrists (x2-test, 0.1
INFLUENCE
OF AGE R. NAKAMURA,
From the Department
of Orthopaedic
AND
Y. TANAKA,
SEX T. IMAEDA
Surgery (Division of Hand Surgery),
ON
ULNAR
VARIANCE
and T. MIURA Nagoya University School of Medicine, Japan
Ulnar variance was measured in 325 normal wrists and 41 wrists with Kienbiick’s disease. A positive correlation between ulnar variance and age was confirmed in normal wrists, and it was lower in males than in females. Therefore, studies comparing ulnar variance in abnormal and normal wrists require carefully selected age- and sex-matched controls. There was no significant difference in ulnar variance between Japanese with normal wrists and those affected by KienbSck’s disease, when the effects of sex and age were taken into account. Based on these results, we believe that ulnar variance is highly unlikely to be an important predisposing factor in KienbBck’s disease. Journal of Hand Surgery (British Volume, 1991) 16B: 84-88
In 1928, Hulten found that 17 out of 23 patients (74%) with Kienbiick’s disease (Kienbock, 1910) had an ulna which was shorter than the radius at the wrist by more than 1 mm and this incidence was remarkably high compared to 8% in normal wrists. Based on this finding, he proposed that negative ulnar variance is a predisposing factor to the development of Kinebock’s disease. Since that time, Hulten’s finding has been confirmed by many investigators (Axelsson, 1973; Beckenbauch et al., 1980; Gelberman et al., 1975; Mirabello et al., 1987; Persson, 1950; Stahl and Reis, 1986; Steinhauser and Posival, 1981; Watson et al., 1985) but has been questioned by others (Chan et al., 1977; Fisk, 1984; Kristensen et al., 1986; and Nathan et al., 1987). To date, this controversy has not been resolved. Ulnar variance also has attracted attention in conditions other than Kienbiick’s disease. Czitrom et al. (1987) reported that post-traumatic scapho-lunate dissociation is associated with a high incidence of negative ulnar variance, and ulno-carpal impingement syndrome has been reported to occur in wrists with positive ulnar variance (HultCn, 1928; Palmer and Werner, 1984). Ulnar variance must also be measured to evaluate residual deformity resulting from a distal radius fracture and to plan surgical radial shortening, ulnar lengthening or Sauve-Kapandji’s procedure (Kapandji, 1986). However, the accurate measurement of ulnar variance is not simple. As pointed out by Epner et al. (1982) and Palmer et al. (1982), the apparent variance changes with the position of arm during roentgenography, and Kristensen et al. (1987) have reported a pseudo-lengthening resulting from arthrosis of the radius as a consequence rather than a cause of Kienbiick’s disease. It is also possible that the ulnar variance is related to age and sex. To assess this, we studied age- and sexrelated changes in ulnar variance among normal Japanese. These values were compared to the amount of ulnar variance in patients with Kienbock’s disease. Material and methods 325 roentgenograms of wrists without abnormalities of the bony structure taken between 1986 and 1989 were 84
used as samples of normal wrists. Films from patients younger than 14 years or with a history of injury to the elbow, forearm, wrist or fingers or with a history of joint disease were excluded from the study. The normal group consisted of 203 males (62.5%) and 122 females (37.5x), ranging in age from 14 to 79 years. A standard postero-anterior projection was obtained (Palmer et al., 1982) with a film focus distance of 125 cm and the wrist fixed in neutral position by a supporter (Nakamura, 1989). The ulnar variance was measured using a template (Palmer et al., 1982), and results were recorded in 0.5 mm increments. Roentgenograms of wrists from 41 patients with Kienbock’s disease who visited our hospital during the same period also were obtained. There were 29 males (71%) and 12 females (29%) ranging in age 14 to 75 years. Roentgenography and the measurement of ulnar variance were performed under the same conditions as in the control group. 41 films of the affected side and 36 films of the unaffected side were available for study. The measurements of ulnar variance were analysed to establish the following relationships : 1. Correlation coefficient between ulnar variance and age in the normal wrists and wrists with Kienbiick’s disease. 2. Sex-related differences in ulnar variance in normal wrists. 3. The difference in ulnar variance between affected and unaffected wrists in patients with Kienbiick’s disease. 4. The difference in ulnar variance between the normal wrists and wrists with Kienbiick’s disease. Correlations and differences were considered statistically significant when p was less than 0.05. Results Ulnar variance in 325 normal wrists ranged from + 5 mm to - 4 mm with a mean + SD of 0.20 f 1.39 mm. Positive ulnar variance (more than 1 mm) was present in 108 cases (32.2x), no variance in 128 (39.3x), and negative variance (less than - 1 mm) in 89 cases (27.4%). The ulnar variance of the 41 wrists affected with THE
JOURNAL
OF HAND
SURGERY
THE
INFLUENCE
OF AGE
AND
Kienbiick’s disease ranged from + 4 mm to -4 mm. Positive ulnar variance was seen in nine cases (22x), no variance in 17 cases (41x), and negative variance in 15 cases (37%).
SEX ON
ULNAR
VARIANCE
that of females, we considered the possibility that this difference influenced in our result. To clarify this, the ulnar variance between sexes was compared in four age groups: 14 to 30, 31 to 40, 41 to 50 and 51 to 79 years. Males had a lower mean value in all groups, with statistically significant difference in two groups at the 5% level, 31 to 40, and 41 to 50. The greatest mean difference (1.61 mm) was between the youngest males and the oldest females (Table 1).
Correlation between ulnar variance and age
The correlation coefficient between ulnar variance and age in normal wrists was 0.250 in males, 0.315 in females and 0.364 overall. All coefficients were highly significant (t-test, p < O.OOl), confirming a positive correlation between ulnar variance and age (Figs 1 and 2). In wrists with Kienbock’s disease, the correlation coefficient between ulnar variance and age was 0.266 which was not significant at 5% level (t-test, 0.05 < p < 0.1; Fig. 3).
Comparison of ulnar variance between affected and unafected wrists inpatients with Kienbiick’s disease
The mean ulnar variance in affected wrists with Kienbock’s disease was lower than in unaffected wrists; 0.18 mm. This difference was not statistically significant (t-test, p > 0.5, Table 2).
Sex-related differences of ulnar variance in normal wrists Comparison of ulnar variance between normal wrists and wrists with Kienbijck’s disease
Negative ulnar variance was present in 73 males (36.1%) and 16 females (13.1%); this difference was significant (x2-test, p ~0.001). Overall, the mean .value also was lower in males than females (t-test, p< 0.01) but since the mean age among males was 12.2 years younger than
The incidence of negative ulnar variance in affected wrists was not significantly different from that in normal wrists (x2-test, 0.1
