JBJS | Kienböck Disease and Negative Ulnar Variance*

The Journal of Bone & Joint Surgery, Volume 80, Issue 8

Articles | August 01, 1998

Kienböck Disease and Negative Ulnar Variance*

MARIUSZ BONZAR, M.D.†; JOHN C. FIRRELL, PH.D.†; MEG HAINER, M.D.†; EDWARD T. MAH, B.M., B.S., M.D., F.R.A.C.S.‡; STEVEN J. MCCABE, M.D., F.R.C.S.†, LOUISVILLE, KENTUCKY

View Disclosures and Other Information

Investigation performed at the Christine M. Kleinert Institute for Hand and Micro Surgery, Louisville

The Journal of Bone & Joint Surgery. 1998; 80:1154-57

5 Recommendations (Recommend) | 3 Comments | Saved by 3 Users Save Case

text A A A

Abstract

We compared the degree of ulnar variance, measured on standardized radiographs of the wrist, in forty-four patients who had Kienböck disease with that in ninety-nine control subjects who had been selected from a general clinic population and had radiographs of the wrist. The purpose of our study was to determine if there is a true relationship between negative ulnar variance and the development of Kienböck disease. Gender was not found to influence the degree of ulnar variance, but an association was found between age and negative ulnar variance in both the control subjects and the patients who had Kienböck disease. The findings of the present study confirmed an association between negative ulnar variance and the development of Kienböck disease even after correction for the influence of age on the measurement of ulnar variance.

Figures in this Article

Introduction

Introduction | Materials and Methods | Results | Discussion | References

The history and features of Kienböck disease were reviewed recently in studies by Gerwin and De Smet. Despite extensive investigation, the etiology of Kienböck disease remains unknown. Alterations in the relationship between the distal aspects of the ulna and radius as well as in the pattern of vascularity of the lunate bone may be predisposing factors. The association between negative ulnar variance and Kienböck disease was described by Hultén in 1928. Subsequently, negative ulnar variance has been considered to be a predisposing factor^1,15, and treatment based on this concept has included shortening of the radius.

Nakamura et al. reported that ulnar variance was related to age and gender. As ulnar variance tends to become more positive with increasing age, a control group drawn from the general population would tend to have more positive ulnar variance than would patients who have Kienböck disease, as the control group would usually be older. Thus, a potentially spurious association could be made between age and ulnar variance. D'Hoore et al., in a study of a European population, compared individuals who had Kienböck disease with age-matched control subjects and found no such association.

The relationship between ulnar variance and Kienböck disease appears to vary in different regions of the world. Nakamura et al. reported the presence of the disease in Japanese patients who had positive ulnar variance. Chan and Huang, in a study of a Chinese population (none of whom had Kienböck disease), reported a predominance of patients who had positive ulnar variance. Gelberman et al.⁷ reported differences between populations with respect to the prevalence of negative ulnar variance; they also suggested that negative ulnar variance is at most a predisposing factor for Kienböck disease.

The purpose of the present study was to evaluate the relationship, controlled for the effect of age, between negative ulnar variance and the development of Kienböck disease in a North American population.

*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.

†Christine M. Kleinert Institute for Hand and Micro Surgery, 225 Abraham Flexner Way, Louisville, Kentucky 40202-3840.

‡Department of Orthopaedics and Trauma, The Queen Elizabeth Hospital, Woodville Road, Woodville 5011, South Australia, Australia.

*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.

†Christine M. Kleinert Institute for Hand and Micro Surgery, 225 Abraham Flexner Way, Louisville, Kentucky 40202-3840.

‡Department of Orthopaedics and Trauma, The Queen Elizabeth Hospital, Woodville Road, Woodville 5011, South Australia, Australia.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 1 Graph showing the distribution of ages among the ninety-nine randomly selected control subjects and the forty-four patients who had Kienböck disease.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2 Graph of the separate best-fit linear regression lines plotted for the forty-four patients who had Kienböck disease (open circles) and the ninety-nine control subjects (closed circles), showing the relationship between age and ulnar variance. Each set of data demonstrates a significant association between increasing (more positive) ulnar variance and increasing age. The correlation coefficient was 0.33 for the patients who had Kienböck disease (p = 0.02) and 0.21 for the control subjects (p = 0.04).

View Larger

TABLE I DATA ON THE PATIENTS

	No. of Patients	Mean Age (yrs.)	Mean Ulnar Variance (mm)
Control subjects	99	40.1	-0.7

Patients who had Kienböck disease
All	44	33.3	-2.9
Stage I	5	40.2	-4.2
Stage II	13	33.0	-2.8
Stage III	26	32.1	-2.6

Add to My JBJS

TABLE I DATA ON THE PATIENTS

	No. of Patients	Mean Age (yrs.)	Mean Ulnar Variance (mm)
Control subjects	99	40.1	-0.7

Patients who had Kienböck disease
All	44	33.3	-2.9
Stage I	5	40.2	-4.2
Stage II	13	33.0	-2.8
Stage III	26	32.1	-2.6

Materials and Methods

Introduction | Materials and Methods | Results | Discussion | References

All patients with Kienböck disease who were managed at our clinic between 1982 and 1993 were included in the study. The diagnosis was made on the basis of the history, physical examination, and radiographic findings. Bone-scanning and magnetic resonance imaging were used as additional diagnostic tools for patients who had early changes of Kienböck disease at the time of presentation.

The stage of the Kienböck disease was classified with use of the system of Lichtman et al., recorded in the medical record, and confirmed on radiographic examination. Ulnar variance was measured on standard anteroposterior radiographs, with use of the concentric ring method described by Palmer et al. and Epner et al., in forty-four patients who had Kienböck disease and in ninety-nine control subjects who had normal radiographic findings. The radiographs of the control subjects had been made for conditions unrelated to Kienböck disease. The twenty-six female and eighteen male patients with Kienböck disease had a median age of thirty-one years (range, seventeen to seventy-four years). The thirty-one female and sixty-eight male control subjects had a median age of forty years (range, twelve to seventy-nine years). The control subjects were randomly chosen from the general clinic population, as is required for a valid analysis of covariance. The demographic data for the two groups were recorded.

The relationship between age and ulnar variance was determined by linear regression analysis with use of SPSS statistical software (SPSS, Chicago, Illinois). An analysis of covariance was performed with use of age as the covariant, ulnar variance as the dependent variable, and the stages of Kienböck disease as the factors. The analysis was performed with use of Helmert contrast¹⁷, which allowed the control subjects to be compared with the patients as a group and also as subgroups with respect to the stage of the disease. Gender was included as a factor in the analysis in order to assess its relationship with ulnar variance.

Results

Introduction | Materials and Methods | Results | Discussion | References

The ranges of age among the patients who had Kienböck disease and the control subjects were similar, but the group with Kienböck disease had proportionally more patients in the younger age-groups (Fig. 1). The effect of age was adjusted to eliminate it as a confounding factor. The variation in ulnar variance with respect to age was seen in both the control group and the group with Kienböck disease (Fig. 2). With use of Spearman regression coefficients, ulnar variance was found to become significantly less negative with increasing age in both the control group and the group with Kienböck disease. The correlation coefficient was 0.33 for the group with Kienböck disease (p = 0.02) and 0.21 for the control group (p = 0.04). When the control group and the group with Kienböck disease were combined (n = 143), ulnar variance demonstrated a highly significant shift as it became more positive with increasing age (correlation coefficient, 0.29; p = 0.0003).

As age was shown to have an important influence on ulnar variance in both the control group and the group with Kienböck disease, it was used as a covariant in the analysis. The difference between the mean non-adjusted ulnar variance for all patients who had Kienböck disease (-2.9 millimeters) and that for all control subjects (-0.7 millimeter) was significant (p < 0.005, Table I). This difference existed even when the ulnar variance was corrected for the effect of age. The patients who had stage-I disease had a mean ulnar variance of -4.2 millimeters, which was more negative than that associated with the other stages. The mean age of the five patients who had stage-I disease was greater (p < 0.05) than that of the patients who had the other stages of the disease, and, when adjusted for age, the mean ulnar variance was marginally increased to -4.3 millimeters (range, 0 to -10 millimeters). This extreme value did not qualify as an outlier because a value of -9 millimeters was found in the group that had stage-III disease.

Regression analysis showed that gender did not influence ulnar variance. The mean ulnar variance (and standard deviation) in the control group was -0.6 ± 1.6 millimeters for the female subjects and -0.9 ± 1.2 millimeters for the male subjects.

Discussion

Introduction | Materials and Methods | Results | Discussion | References

Elucidation of the true relationship between Kienböck disease and ulnar variance is critically important, as most operative treatment is based on the premise of a causal relationship. Our results confirmed a strong association between negative ulnar variance and Kienböck disease. Although this theory has been widely accepted since it was proposed by Hultén in 1928, recent studies showing the effect of age and radiographic technique on the measurement of ulnar variance have cast doubt on the relationship^4,12. Kristensen et al. suggested that Kienböck disease can cause subchondral thickening of the radius, which may alter measurements and give the appearance of negative ulnar variance.

Nakamura et al. showed a relationship between ulnar variance (or its measurement) and the age of the patient. This finding was confirmed by D'Hoore et al. in a study of a European population and by our study of a North American population. In the current analysis, age was a confounder, as it was unrelated to the question that we were attempting to answer, and it was associated with both ulnar variance (or its measurement) and the presence of Kienböck disease. To control for the effect of age on ulnar variance, D'Hoore et al. compared a group of patients who had Kienböck disease and a group of individuals with no disease who had the same mean age as the patients. The difference between the ulnar variance in the patients who had the disease and that in the individuals who had no disease was not significant (p = 0.63). In the present study, we analyzed the effect of age on ulnar variance in a different manner. The control subjects were randomly chosen from a population of patients who had had standardized radiographs of the wrist made in our clinic. The effect of age on the measurement of ulnar variance was controlled statistically. The technique that was used for the measurement of ulnar variance in our study has been shown to have high interobserver and intraobserver reliability¹⁶. Still, we found a strong association between negative ulnar variance and Kienböck disease, an observation that differs from the conclusions of Nakamura et al. and D'Hoore et al. If it is assumed that the results of both of those studies were accurate and that the concern expressed by Weiss with regard to the analysis of D'Hoore et al. was not valid, a possible explanation for the discrepancy in the results is the variability of the disease itself. In the present study, none of the patients who had the disease had positive ulnar variance. However, Tsuge and Nakamura identified positive ulnar variance in patients who had Kienböck disease. The variations among different populations of patients with respect to the prevalence of positive ulnar variance seems to suggest that other factors may influence the development and the expression of the disease.

An interaction between the circulation of the lunate and ulnar variance may be a factor in the genesis of the disease. If so, Kienböck disease may develop in patients who have negative ulnar variance and an at-risk vascular pattern of the lunate⁶. Alternatively, negative ulnar variance may not be a causative factor for the disease but simply a developmental anatomical marker for a population of patients who may have a higher proportion of alterations in the vascular anatomy or other unknown predisposing factors. If this explanation is correct, then operative realignment of the radius and ulna would be inappropriate and unwarranted. Unknown factors may influence the development of the disease.

We examined the relationship between the degree of ulnar variance and the severity of Kienböck disease. At the time of the diagnosis, the patients who had stage-I disease had a greater degree of negative ulnar variance than did those who had a later stage of the disease. However, this finding was due to one patient who had ulnar variance of -10 millimeters. Therefore, the results should be interpreted with caution. It also could be that patients who had pain in the wrist and a large degree of negative ulnar variance were evaluated more extensively, making it possible to diagnose the disease early.

An increasing number of patients have been managed with shortening of the ulna for the treatment of disorders on the ulnar side of the wrist. If there is a causal relationship between negative ulnar variance and Kienböck disease, the disease may develop in some patients who have had shortening of the ulna. It is difficult to predict when this might occur, as it depends on the attributable risk and the time-period before the onset of the disease, both of which are unknown. It should be noted that a case report¹³ on a patient who had negative ulnar variance without Kienböck disease is extremely weak evidence against the association. The successful results reported after the treatment of Kienböck disease with a procedure to level the joint suggest but do not prove that negative ulnar variance is the cause of the disease^1,15.

We found that the relationship between the distal aspects of the ulna and radius changes with age. Although the ulnar variance can become more positive after excision of the radial head, it is difficult to understand why this happens when there has been no traumatic injury or operative intervention. One could argue that changes in the cartilage or ligamentous support, or a change in the shape of the bones, may alter the relationship between the radius and ulna at the wrist. It is also possible that the measurement of ulnar variance may be influenced by age.

The present study confirms the association between negative ulnar variance and Kienböck disease, but this finding should not be interpreted as proof that negative ulnar variance is the cause of the disease. Additional studies are needed for further evaluation of this relationship and the possible interactions of negative ulnar variance with other anatomical features.

References

Introduction | Materials and Methods | Results | Discussion | References

Almquist, E. E., and Burns, J. F., Jr.: Radial shortening for the treatment of Kienböck's disease—a 5- to 10-year follow-up. J. Hand Surg,7: 348-352, 1982.7348 1982

Chan, K. P., and Huang, P.: Anatomic variations in radial and ulnar lengths in the wrists of Chinese. Clin. Orthop.,80: 17-20, 1971.8017 1971 [PubMed]

De Smet, L.: Ulnar variance: facts and fiction review article. Acta Orthop. Belgica,60: 1-9, 1994.601 1994

D'Hoore, K.; De Smet, L.; Verellen, K.; Vral, J.; and Fabry, G.: Negative ulnar variance is not a risk factor for Kienböck's disease. J. Hand Surg.,19A: 229-231, 1994.19A229 1994

Epner, R. A.; Bowers, W. H.; and Guilford, W. B.: Ulnar variance—the effect of wrist positioning and roentgen filming technique. J. Hand Surg.,7: 298-305, 1982.7298 1982

Gelberman, R. H.; Bauman, T. D.; Menon, J.; and Akeson, W. H.: The vascularity of the lunate bone and Kienböck's disease. J. Hand Surg.,5: 272-278, 1980.5272 1980

Gelberman, R. H.; Salamon, P. B.; Jurist, J. M.; and Posch, J. L.: Ulnar variance in Kienböck's disease. J. Bone and Joint Surg.,57-A: 674-676, July 1975.57-A674 1975

Gerwin, M.: The history of Kienböck's disease. Hand Clin.,9: 385-390, 1993.9385 1993 [PubMed]

Hultén, O.: Über anatomische Variationen der Handgelenkknochen. Ein Beitrag zur Kenntnis der Genese zwei verschiedener Mondbeinveränderungen. Acta Radiol.,9: 155-168, 1928.9155 1928

Kristensen, S. S.; Thomassen, E.; and Christensen, F.: Ulnar variance in Kienböck's disease. J. Hand Surg.,11-B: 258-260, 1986.11-B258 1986

Lichtman, D. M.; Mack, G. R.; MacDonald, R.I.; Gunther, S. F.; and Wilson, J. N.: Kienböck's disease: the role of silicone replacement arthroplasty. J. Bone and Joint Surg.,59-A: 899-908, Oct 1977.59-A899 1977

Nakamura, R.; Tanaka, Y.; Imaeda, T.; and Miura, T.: The influence of age and sex on ulnar variance. J. Hand Surg.,16-B: 84-88, 1991.16-B84 1991

Nathan, P. A., and Meadows, K. D.: Ulna-minus variance and Kienböck's disease. J. Hand Surg.,12A: 777-778, 1987.12A777 1987

Palmer, A. K.; Glisson, R. R.; and Werner F. W.: Ulnar variance determination. J. Hand Surg.,7: 376-379, 1982.7376 1982

Rock, M. G.; Roth, J. H.; and Martin, L.: Radial shortening osteotomy for treatment of Kienböck's disease. J. Hand Surg.,16A: 454-460, 1991.16A454 1991

Steyers, C.M., and Blair, W. F.: Measuring ulnar variance: a comparison of techniques. J. Hand Surg.,14A: 607-612, 1989.14A607 1989

Tamhane, A. C.; Hochberg, Y.; and Dunnett, C. W.: Multiple test procedures for dose finding. Biometrics,52: 21-37, 1996.5221 1996 [PubMed]

Tsuge, S., and Nakamura, R.: Anatomical risk factors for Kienböck's disease. J. Hand Surg.,18-B: 70-75, 1993.18-B70 1993

Weiss, A. P.: Letter to the editor. J. Hand Surg.,19A: 1057-1058, 1994.19A1057 1994

Topics

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 1 Graph showing the distribution of ages among the ninety-nine randomly selected control subjects and the forty-four patients who had Kienböck disease.

View Large | Download Slide(.ppt)
Add to My JBJS

View Larger

TABLE I DATA ON THE PATIENTS

	No. of Patients	Mean Age (yrs.)	Mean Ulnar Variance (mm)
Control subjects	99	40.1	-0.7

Patients who had Kienböck disease
All	44	33.3	-2.9
Stage I	5	40.2	-4.2
Stage II	13	33.0	-2.8
Stage III	26	32.1	-2.6

Add to My JBJS

TABLE I DATA ON THE PATIENTS

	No. of Patients	Mean Age (yrs.)	Mean Ulnar Variance (mm)
Control subjects	99	40.1	-0.7

Patients who had Kienböck disease
All	44	33.3	-2.9
Stage I	5	40.2	-4.2
Stage II	13	33.0	-2.8
Stage III	26	32.1	-2.6