Rheumatology and Rehabilitation, 1976, 5, 10 ORIGINAL PAPER
INHIBITION OF SECONDARY OSTEOPHYTE FORMATION IN RHEUMATOID ARTHRITIS BY ROBERT L. SWEZEY* AND DAVID M. BJARNASONf
IT is generally assumed that osteoarthritic degenerative joint disease (DJD) changes are a response to joint cartilage damage. These changes associated with cartilage damage and cartilage loss are typically proliferative in nature, and are characterized by subchondral bony sclerosis and joint marginal osteophye formation. The paradoxical lack of proliferative changes in those joints whose cartilage has been severely eroded (radiographically narrowed) by rheumatoid arthritis (RA) has been noted since 1909 (Nichols and Richardson, 1909). Several authors have subsequently commented on this apparent lack of degenerative change in the rheumatoid joint, but the extent of the disparity between joint cartilage loss and proliferative osteophyte formation in rheumatoid arthritis has not been adequately studied (Knaggs, 1926; Bennett et ah, 1942; Fletcher and Rowley, 1968; Martel, 1968; Ahlback, 1968). Because of the possible significance of these observations to the pathogenesis of both RA and DJD the authors undertook a study of the extent of proliferative osteoarthritic change in joints afflicted by rheumatoid arthritis compared to the proliferative changes occurring in similar joints affected by degenerative joint disease alone. A comparison of the number and size of osteophytes [as indicators of proliferative changes (DJD)] in relation to the degree of radiographically demonstrable joint narrowing (as an index of joint damage) in comparable joints of patients with either RA or DJD is made. The joints selected for study, both frequently involved in RA and DJD, are the proximal interphalangeal joints (PIP) and the knees (Ahlback, 1948; Stecher, 1950). METHODS with RA and DJD and involvement of knees, and/or proximal interphalangeal joints, were collected from the LAC/USC Medical Center Clinics and from the files of the clinical faculty of the Rheumatology Section of the USC School of Medicine. Criteria for the diagnosis of DJD were those of Kellgren and Lawrence (1952). For PATIENTS
•Professor of Medicine, Chief, Division of Rehabilitation Medicine UCLA School of Medicine, California, U.S.A. t401 E. Forest Street, Marshfield, Wisconsin. Accepted for publication March, 1975. 10
Downloaded from http://rheumatology.oxfordjournals.org/ at University of California, Santa Barbara on July 6, 2015
SUMMARY Using radiographically demonstrated joint narrowing as an index of joint damage, a reduced incidence of osteophyte size and number as a measure of osteoarthritic change in the PEP and knee joints of patients with RA as compared to DJD has been demonstrated. The possible pathogenetic factors discussed include reduced joint use in RA, increased resilience of juxtaarticular bone associated with osteoporosis in RA and the possibility that products of inflammation in the RA joint act locally or via juxta-articular vascular mechanisms to inhibit the secondary changes of DJD. Further study of the possible pathogenic mechanisms including the roles of joint stress, juxta-articular osteoporosis, and possible effects of the inflammatory reaction of the rheumatoid joint on intra-articular structures and juxta-articular vascular phenomena are needed.
SWEZEY AND BJARNASON: INHffilTION OF OSTEOPHYTES
11
RESULTS COMPARISON OF OSTEOPHYTE SIZE AND JOINT NARROWING IN PIP JOINTS
Fifty-six per cent (88/158) of DJD minimally narrowed PIP joints had osteophytes, compared to 10% (28/293) of the RA group. Eighty-nine per cent (31/35) of the DJD markedly narrowed PIP joints had osteophytes, compared to 27 % (75/268) of the RA group. Thirty-three per cent of DJD joints had large osteophytes (grade 3-4) compared with 3 % of the RA group (see Table I).
Downloaded from http://rheumatology.oxfordjournals.org/ at University of California, Santa Barbara on July 6, 2015
RA, the definite or classical categories of the ARA criteria (Ropes et ah, 1956) were used. No patient was accepted if a question of diagnosis occurred or if there was evidence of co-existence of both RA and DJD, or if the patient had taken systemic steroids or received more than occasional local steroid injections. The age range of the DJD patients was 23 to 93 years (mean age 65 years) and the RA patients from 25-83 years (mean age 54 years). One hundred and ninety-three PIP joints (28 cases) and 62 knee joints (37 cases) with DJD were compared with 564 PIP joints (76 cases) and 158 knee joints (84 cases) with RA. Twenty-two per cent of the patients with DJD knees were males and 18 % of the RA knees were from males. Fifteen per cent of the patients with DJD PIP joints were males and 12% of the patients with RA involvement of the PIP joints were males. Of actual joints studied (more joints than patients) 16% of the knee joints of the degenerative disease group and 14% of the rheumatoid arthritic knee joints were from male patients. Standard 40-inch radiographic views of the hands (PA) and knees (AP) were evaluated. All of the X-ray films were reviewed simultaneously by both investigators but scored individually. Although an attempt to read the films 'blind' was made, the diagnosis radiographically was usually apparent, and in the case of some of the osteoarthritic patients a deliberate search for cases in the younger age groups was made. Some of these selected cases were known as such by the authors. When the classifications of osteophytes and/or joint narrowing were not mutually agreed upon they were graded so as to record the greater osteophyte size and the lesser degree of joint narrowing, e.g. medial compartment right knee read by observer S as 3 mm and by observer B as 2 mm would be recorded as 3 mm- Discrepancies in inter-observer grading were not tabulated but were estimated retrospectively to have occurred in approximately 15% of joints studied. The discrepancies occurred primarily in the RA joints because of difficulty distinguishing residual marginal bony spicules from osteophytes. Joints with clinical flexion or extension deformities greater than 10 degrees were not studied because of the difficulty in determining joint width in the presence of these deformities. Joint narrowing in the knee was measured in both the medial and lateral compartments with the measurements in millimetres taken from the midpoint of the inferior surface of each femoral condyle to the corresponding line of the anterior surface of the tibial plateau. Where possible, both weight-bearing and non-weight-bearing films were studied. Joint narrowing in the PIP joints was recorded on a zero to four scale. Grade 1 represents perceptible narrowing (25 % or less of the joint space); grade 2 represents further narrowing without perceptible obliteration of any portion of the joint space; grade 3 represents narrowing with partial apposition of articular bony surfaces; and grade 4 represents complete loss of joint cartilage, articular bony fusion or severe destruction and disorganization of articular surfaces. Osteophyte size in both the knee and PIP joints were graded on a zero to four scale according to standardized criteria, the grading being a function of the largest osteophyte present in the area in question (Epidemiology of Chronic Rheumatism, 1963).
12
RHEUMATOLOGY AND REHABILITATION VOL. XV NO. 1 TABLE I COMPARISON OF OSTEOPHYTE SIZE AND JOINT NARROWING IN PIP JOINTS
Minimal narrowing (0-1)
Marked narrowing (2-4)
Osteophyte size
RA
DJD
Grade
265 24 3 1 0
70 55 30 3 0
191 55 14 8 0
4 9 11 5 6
0 1 2 3 4
10.0%
56.0%
27.0%
89.0%
293
158
268
35
Percentage of joints with osteophytes Total number of joints
COMPARISON OF MINIMALLY NARROWED PIP JOINTS WITH OSTEOPHYTES IN MALES TO ALL PATIENTS
Four out of 28 of all DJD and 11 out of 74 of all RA patients whose PEP joints were studied were males. In the male patients with DJD, 7 out of 17 joints with minimal narrowing had osteophytes while in the male patients with RA only 12 out of 64 joints with minimal narrowing had osteophytes. This compares to 111 out of 196 of all DJD minimally narrowed joints and 57 out of 392 of all RA minimally narrowed joints respectively. COMPARISON OF OSTEOPHYTE SIZE AND PIP JOINT NARROWING IN CASES OF RA DJD (AGED 50-59 YEARS)*
AND
Of 45 DJD minimally narrowed PD? joints aged 50-59, 10 (24%) compared with 56% of all DJD PIP joints (mean age 62) had osteophytes. Fifteen per cent (15/98) of RA minimally narrowed PEP joints aged 50-59, compared with 9 % of all RA PIP joints (mean age 55), had osteophytes. Twelve out of twelve DJD markedly narrowed PIP joints (aged 50-59), compared with 89 % of all DJD joints (mean age 62), had osteophytes. Thirty-eight per cent (27770) of RA markedly narrowed PIP joints aged 50-59, compared with 29% of all RA joints (mean age 55), had osteophytes. Of DJD markedly narrowed PIP joints, six out of 12 had osteophytes of grade 3 or 4, compared to only four out of 70 of correspondingly narrowed RA PIP joints. COMPARISON OF OSTEOPHYTE PREVALENCE AND KNEE-JOINT NARROWING IN CASES OF RA AND DJD WITH JOINT 'SPACE' OF 4 MM OR GREATER (AGED 50-59 YEARS)
In the medial femoral compartments with 'joint spaces' measuring over 4 mm, eight out of 23 (65%) of RA compared to 12 out of 12 DJD knees, had osteophytes. In the lateral compartments seven out of 23 (32 %) of RA had osteophytes compared to eight out of 13 (61 %) with DJD. Only two DJD knee medial and one DJD lateral compartments measured less than 3 mm and this small number precluded valid comparisons (see Table H). •Inadequate sample size precluded comparison of other age groups.
Downloaded from http://rheumatology.oxfordjournals.org/ at University of California, Santa Barbara on July 6, 2015
DJD
RA
SWEZEY AND BJARNASON: INfflBITION OF OSTEOPHYTES
13
TABLE H COMPARISON OF OSTEOPHYTE SIZE AND KNEE-JOINT NARROWING*
Joint 'space' Compartments
Lateral
3 mm or less
RA
DJD
RA
37 5 7 5 0
11 10 12 13 2
54
DJD
Grade
39 38 14 7 6
2 1 4 4 3
0 1 2 3 4
48
104
14
Total joints
34 11 4 1 2
17 14 17 6 3
48 26 16 14 2
3 1 1 0 0
0 1 2 3 4
52
57
106
5
Total joints
*The lower limit of normal knee-joint space in the adult is 4 mm (Ahlback, 1968). OSTEOPHYTE PREVALENCE IN MALE KNEE JOINTS COMPARED TO FEMALES
When the medial tibio-femoral compartment interspace measured 4 mm or greater, three of 19 males and 21 of 36 females with RA had osteophytes, while eight out of eight males and 29 out of 40 females with DJD had osteophytes. With corresponding measures of the lateral compartment, four out of 19 males and 20 out of 33 females with RA had osteophytes and 8 out of 10 males and 37 out of 50 females with DJD had osteophytes. With further tibio-femoral j oint-space narrowing (less than 4 mm) in the medial compartment, 11 out of 15 RA males, 49 out of 73 RA females and two out of three DJD males and 10 out of 11 DJD females had osteophytes. In the similarly narrowed lateral compartments, 11 out of 15 RA males, 49 out of 95 RA females and no (0/1) DJD male and three out of five DJD females had osteophytes. STATISTICAL ANALYSIS This study was designed to test the hypothesis that for a given measure of PIP or knee-joint narrowing no difference in the number of osteophytes in patients with RA and DJD would be detected. Analysis of the data utilized the chi-squared test of independence and f-test for significant differences between percentages. Using a 9 5 % confidence level, the following statements can be made: 1. In the proximal interphalangeal joints, osteophyte formation develops differently in RA from DJD in all joints with minimal and with marked narrowing. This held true for the joints of patients aged 50-59 in which marked joint space narrowing was present. 2. In the PEP joints of both RA and DJD patients, osteophyte development was not significantly different with regard to sex. 3. In the medial knee joint compartment at all degrees of joint narrowing osteophyte formation was significantly more prevalent in the DJD joints. In the lateral knee joint compartment this held true only when the joint space was 4 mm or more in width.
Downloaded from http://rheumatology.oxfordjournals.org/ at University of California, Santa Barbara on July 6, 2015
Medial
4 mm or more
Osteophyte size
14
RHEUMATOLOGY AND REHABILITATION VOL. XV NO. 1
DISCUSSION It has been shown that osteophyte formation occurs in the absence of radiographically demonstrable progressive knee articular damage (joint narrowing or subchrondral bone sclerosis), but with the exception of rheumatoid arthritis the converse (joint narrowing without osteophyte production) has not been described (Danielsson and Hernborg, 1970a, b). Bennett et al. (1942) have shown that degenerative changes in cartilage occur essentially universally by the second decade. This study has demonstrated that for the PIP and knee joints the extent of osteophyte formation in the RA joints is less than that found in comparably damaged (narrowed) degenerative joint disease joints. This is true for the proximal interphalangeal and the knee joints when corrections for age and sex are made. At the time of tabulation, data on the prior occupations and on race were not available and therefore the impact of these factors cannot be assessed. Even on comparison of weight-bearing and nonweight-bearing radiographs of the knees where severe degrees of narrowing not appreciated on routine non-weight-bearing films can be brought to light, the relative sparing of DJD changes in the rheumatoid knee was still observed (Ahlback, 1968). It should be noted in this connection that the nature of the rheumatoid process tends to produce diffuse joint-cartilage destruction, and thus asymmetrical narrowing seen on weightbearing films in DJD is frequently absent in RA. Further, in contrast to uncomplicated DJD, there is often little difference in the 'joint space' between weight-bearing and nonweight-bearing films of the rheumatoid knees. In order to minimize bias, wherever interobserver discrepancy in reading osteophytes occurs the larger size was chosen. In many of the RA joints, erosive thinning of articular margins occurred in such a manner as to project small bony 'spikes' at the joint margins. Although these did not appear to be proliferative osteophytic changes,
Downloaded from http://rheumatology.oxfordjournals.org/ at University of California, Santa Barbara on July 6, 2015
4. A statistically significant independence of disease (RA versus DJD) and osteophyte formation was found in patients aged 50-59 with a medial knee-joint compartment of less than 3 mm and where the joint space was greater than 4 mm in both the medial and lateral compartments. In the same age-group in those patients with less than 3 mm joint space, the lateral compartment showed no independence of disease and osteophyte formation. 5. Sex and osteophyte size were independent (osteophyte formation was more prevalent in females than in males) in the rheumatoid arthritic patients when the kneejoint medial and lateral compartment spaces were greater than 4 mm. This difference was not found when the joint spaces were 3 mm or less in the rheumatoids nor was there a statistically significant independence of sex and osteophyte formation at any degree of narrowing in either knee-joint compartment in the cases with DJD. In all the instances where there were differences between sexes, the males had fewer osteophytes than females, and since the vast majority of patients in this study were females, the bias of this data would therefore tend to minimize the probability of finding the significance of the conclusions drawn in this study that disease and osteophyte formation are independent, i.e. that osteophyte formation is less prevalent at any degree of joint narrowing in the rheumatoid arthritics than in osteoarthritis. 6. In all cases above where the statistical significance of the differences between osteophyte formation and disease could not be established, the trend was in the same direction as the results that were of statistical significance, i.e. at corresponding degrees of joint narrowing, osteophytes were less prevalent in RA than in DJD. Inadequate sample size appears to be the reason that these tests did not obtain significance.
SWEZEY AND BJARNASON: INHIBITION OF OSTEOPHYTES
15
A third, perhaps most intriguing possibility, is that some factor or factors of the inflammatory process of the RA joint play an inhibitory role in the development of secondary DJD. It has been demonstrated that the addition of thawed leucocytes to cultured monolayers of human synovial fibroblasts increases fibroblastic synthesis of hyaluronic acid and that this can be inhibited by anti-inflammatory drugs (Yaron et al., 1971). The possibility that metabolic products of leucocytes, synoviocytes, or antigenantibody interactions in the RA joint may result in the formation of inhibitors to ths proliferative DJD process merits further study. Because of the persistent nature of the inflammatory process in RA, the effects of the inflammatory process on both intra-articular and juxta-articular structures is prolonged. Thus, vascular alterations as a consequence of the inflammation may not abate and both intra and extra-articular perturbations resulting from the RA inflammation persist and may play an important role in creating the environment in a damaged joint which is not conducive to the emergence of DJD. CONCLUSION Using radiographically demonstrated joint narrowing as an index of joint damage, a reduced incidence of osteophyte size and number as a measure of osteoarthritic change in the PIP and knee joints of patients with RA as compared to DJD has been demonstrated. The possible pathogenetic factors discussed include reduced joint use in RA,
Downloaded from http://rheumatology.oxfordjournals.org/ at University of California, Santa Barbara on July 6, 2015
they were so interpreted for the purpose of this study. Therefore the actual incidence of osteophytic formation in the rheumatoid joints may have been somewhat lower than recorded. Whereas the association of hypertrophic osteoarthritic changes occurring as a primary process or secondary to a variety of joint traumata—direct joint injury, obesity, haemorrhage, joint instability in neuropathic joints—and secondary to joint sepsis is well recognized; the relative sparing of the rheumatoid joint by osteoarthritic changes has not been adequately appreciated (Traut, 1956). Documentation of an apparent inhibiting effect by the RA process on the development of secondary degenerative joint disease exclusive of the influence of age or sex raises several questions. Does the presence of RA with its accompanying pain and inhibition of joint use cause a relative sparing of the joint from the mechanical stresses which may act as a stimulus to the hypertrophic degenerative joint disease changes ? The role of repeated joint trauma in DJD is well known and the sparing effect of paralysis on Heberden's node formation has also been documented (Stecher, 1950; Kellgren, 1961). Further, it is the authors' impression that degenerative changes in RA are most commonly observed in those patients who seem to have great tolerance for pain and who use their joints despite the obvious inflammatory involvement, and in those RA patients whose previous inflammatory disease has abated and who resume active function on previously damaged joints (Swezey et al., 1972-3). Although these observations would suggest that an age factor (time for disease remission and further joint abuse) might play a role, comparison of age-matched groups did not support this possibility. A second consideration requiring further study is the effect of osteoporosis on DJD. Both generalized and juxta-articular bone demineralization are common factors in RA. Foss and Byers (1972) documented the remarkable lack of DJD of the hip in the presence of diminished femoral bone density. Radin and Paul have recently shown (1970) that resilience of subchrondral bone plays a role in joint degeneration. The possibility that a change in resilience of subchondral bone as a consequence of osteoporosis might cause a greater cushioning effect on joint surfaces has been suggested as a mechanism in the sparing of DJD changes of the hip and may apply to the RA joint as well.
16
RHEUMATOLOGY AND REHABILITATION VOL. XV NO. 1
increased resilience of juxta-articular bone associated with osteoporosis in RA and the possibility that products of inflammation in the RA joint act locally or via juxta-articular vascular mechanisms to inhibit the secondary changes of DJD. Further study of the possible pathogenic mechanisms including the roles of joint stress, juxta-articular osteoporosis and possible effects of the inflammatory reaction of the rheumatoid joint on intra-articular structures and juxta-articular vascular phenomena are needed.
REFERENCES
(1968) "Osteoarthritis of the Knee: A Radiographic Investigation". Acta Radiol. Suppl. 277, 7-72. BENNETT, G. A., WAINE, H. and BAUER, W. (1942) Changes in the Knee Joint at Various Ages. New York: The Commonwealth Fund, p. 21. DANIELSSON, L. and HERNBORO, J. (1970) "Morbidity and Mortality of Osteoarthritis of the Knee (Gonarthrosis) in Malmo, Sweden". Clin. Orthop., 69, 224-6. and (1970) "Clinical and Rocntgenologic Study of Knee Joints with Osteophytes". Clin. Orthop., 69, 302-12. Epidemiology of Chronic Rheumatism, Vol. 2 (1963) Atlas of Standard Radiographs of Arthritis. Philadelphia: F. A. Davis. FLETCHER, D. E. and ROWLEY, K. A. (1965) "The Radiological Features of Rheumatoid Arthritis". Br. J. Radiol., 25, 663. Foss, M. V. L. and BYERS, P. D. (1972) "Bone Density, Osteoarthrosis of the Hips and Fracture of the Upper End of the Femur". Ann. Rheum. Dis., 31, 259-64. KELLGREN, J. H. and LAWRENCE, F. A. (1952) "Radiological Assessment of Osteoarthritis". Br. J. Radiol., 25, 282-95. (1961) "Osteoarthrosis in Patients and Populations". Br. Med. J., 2, 1-6. KNAGGS, R. LAWFORD (1926) The Inflammatory and Toxic Diseases of Bone. London: Sompkin, Marshall, Hamilton, Kent and Co. Ltd., pp. 165-6. MARTEL, WILLIAM (1968) "Radiologic Manifestations of Rheumatoid Arthritis with Particular Reference to the Hand, Wrist and Foot". Med. Clin. N. Am., 52, 663. NICHOLS, E. H. and RICHARDSON, F. L. (1909) "Arthritis Deformans". / . Med. Res. (New Series, Vol. 16), 21, No. 2, 149-222. RADIN, E. L. and PAUL, I. L. (1970) "Does Cartilage Compliance Reduce Skeletal Impact Loads T' Arthritis Rheum., 13, 139-44. ROPES, M. W., BENNETT, G. A., COBB, S., JACOX, R. and VESSAR, R. A. (1956) "A Committee of the American Rheumatism Association Proposed Diagnostic Criteria for Rheumatoid Arthritis, No. 4". Bull. Rheum. Dis., 7, 121-4. STECHER, R. M. (1950) "Heberden's Nodes: A Clinical Description of Osteoarthritis of the Finger Joints". Ann. Rheum. Dis., 14, 1-10. SWEZEY, R. L., BJARNASON, D., ALEXANDER, S. and FORRESTER, D. B. (1972-3) "Resorptive Arthropathy and the Opera Glass Hand Syndrome". Semin. Arthritis Rheum., 2, 191-243. TRAXTT, E. F. (1956) "Degenerative Arthritis: Its Causes and Management". Med. Clin. N. Am., 40, 63-78. YARON, M., YARON, I. and ALLALOUF, D. (1971) "Effect of Some Anti-inflammatory Drugs on Fibroblast-Leukocyte Interaction in vitro". Ann. Rheum. Dis., 30, 613-18. AHLBACK, SVEN
Downloaded from http://rheumatology.oxfordjournals.org/ at University of California, Santa Barbara on July 6, 2015
ACKNOWLEDGMENTS
I would like to thank Steven Salzberg, M.S., Senior Statistician, Los Angeles County/ U.S.C. Medical Center, for his assistance in obtaining tho statistical information in this report. This work was supported in part by a grant from the Southern California Chapter of the Arthritis Foundation.
INHIBITION OF SECONDARY OSTEOPHYTE FORMATION IN RHEUMATOID ARTHRITIS BY ROBERT L. SWEZEY* AND DAVID M. BJARNASONf
IT is generally assumed that osteoarthritic degenerative joint disease (DJD) changes are a response to joint cartilage damage. These changes associated with cartilage damage and cartilage loss are typically proliferative in nature, and are characterized by subchondral bony sclerosis and joint marginal osteophye formation. The paradoxical lack of proliferative changes in those joints whose cartilage has been severely eroded (radiographically narrowed) by rheumatoid arthritis (RA) has been noted since 1909 (Nichols and Richardson, 1909). Several authors have subsequently commented on this apparent lack of degenerative change in the rheumatoid joint, but the extent of the disparity between joint cartilage loss and proliferative osteophyte formation in rheumatoid arthritis has not been adequately studied (Knaggs, 1926; Bennett et ah, 1942; Fletcher and Rowley, 1968; Martel, 1968; Ahlback, 1968). Because of the possible significance of these observations to the pathogenesis of both RA and DJD the authors undertook a study of the extent of proliferative osteoarthritic change in joints afflicted by rheumatoid arthritis compared to the proliferative changes occurring in similar joints affected by degenerative joint disease alone. A comparison of the number and size of osteophytes [as indicators of proliferative changes (DJD)] in relation to the degree of radiographically demonstrable joint narrowing (as an index of joint damage) in comparable joints of patients with either RA or DJD is made. The joints selected for study, both frequently involved in RA and DJD, are the proximal interphalangeal joints (PIP) and the knees (Ahlback, 1948; Stecher, 1950). METHODS with RA and DJD and involvement of knees, and/or proximal interphalangeal joints, were collected from the LAC/USC Medical Center Clinics and from the files of the clinical faculty of the Rheumatology Section of the USC School of Medicine. Criteria for the diagnosis of DJD were those of Kellgren and Lawrence (1952). For PATIENTS
•Professor of Medicine, Chief, Division of Rehabilitation Medicine UCLA School of Medicine, California, U.S.A. t401 E. Forest Street, Marshfield, Wisconsin. Accepted for publication March, 1975. 10
Downloaded from http://rheumatology.oxfordjournals.org/ at University of California, Santa Barbara on July 6, 2015
SUMMARY Using radiographically demonstrated joint narrowing as an index of joint damage, a reduced incidence of osteophyte size and number as a measure of osteoarthritic change in the PEP and knee joints of patients with RA as compared to DJD has been demonstrated. The possible pathogenetic factors discussed include reduced joint use in RA, increased resilience of juxtaarticular bone associated with osteoporosis in RA and the possibility that products of inflammation in the RA joint act locally or via juxta-articular vascular mechanisms to inhibit the secondary changes of DJD. Further study of the possible pathogenic mechanisms including the roles of joint stress, juxta-articular osteoporosis, and possible effects of the inflammatory reaction of the rheumatoid joint on intra-articular structures and juxta-articular vascular phenomena are needed.
SWEZEY AND BJARNASON: INHffilTION OF OSTEOPHYTES
11
RESULTS COMPARISON OF OSTEOPHYTE SIZE AND JOINT NARROWING IN PIP JOINTS
Fifty-six per cent (88/158) of DJD minimally narrowed PIP joints had osteophytes, compared to 10% (28/293) of the RA group. Eighty-nine per cent (31/35) of the DJD markedly narrowed PIP joints had osteophytes, compared to 27 % (75/268) of the RA group. Thirty-three per cent of DJD joints had large osteophytes (grade 3-4) compared with 3 % of the RA group (see Table I).
Downloaded from http://rheumatology.oxfordjournals.org/ at University of California, Santa Barbara on July 6, 2015
RA, the definite or classical categories of the ARA criteria (Ropes et ah, 1956) were used. No patient was accepted if a question of diagnosis occurred or if there was evidence of co-existence of both RA and DJD, or if the patient had taken systemic steroids or received more than occasional local steroid injections. The age range of the DJD patients was 23 to 93 years (mean age 65 years) and the RA patients from 25-83 years (mean age 54 years). One hundred and ninety-three PIP joints (28 cases) and 62 knee joints (37 cases) with DJD were compared with 564 PIP joints (76 cases) and 158 knee joints (84 cases) with RA. Twenty-two per cent of the patients with DJD knees were males and 18 % of the RA knees were from males. Fifteen per cent of the patients with DJD PIP joints were males and 12% of the patients with RA involvement of the PIP joints were males. Of actual joints studied (more joints than patients) 16% of the knee joints of the degenerative disease group and 14% of the rheumatoid arthritic knee joints were from male patients. Standard 40-inch radiographic views of the hands (PA) and knees (AP) were evaluated. All of the X-ray films were reviewed simultaneously by both investigators but scored individually. Although an attempt to read the films 'blind' was made, the diagnosis radiographically was usually apparent, and in the case of some of the osteoarthritic patients a deliberate search for cases in the younger age groups was made. Some of these selected cases were known as such by the authors. When the classifications of osteophytes and/or joint narrowing were not mutually agreed upon they were graded so as to record the greater osteophyte size and the lesser degree of joint narrowing, e.g. medial compartment right knee read by observer S as 3 mm and by observer B as 2 mm would be recorded as 3 mm- Discrepancies in inter-observer grading were not tabulated but were estimated retrospectively to have occurred in approximately 15% of joints studied. The discrepancies occurred primarily in the RA joints because of difficulty distinguishing residual marginal bony spicules from osteophytes. Joints with clinical flexion or extension deformities greater than 10 degrees were not studied because of the difficulty in determining joint width in the presence of these deformities. Joint narrowing in the knee was measured in both the medial and lateral compartments with the measurements in millimetres taken from the midpoint of the inferior surface of each femoral condyle to the corresponding line of the anterior surface of the tibial plateau. Where possible, both weight-bearing and non-weight-bearing films were studied. Joint narrowing in the PIP joints was recorded on a zero to four scale. Grade 1 represents perceptible narrowing (25 % or less of the joint space); grade 2 represents further narrowing without perceptible obliteration of any portion of the joint space; grade 3 represents narrowing with partial apposition of articular bony surfaces; and grade 4 represents complete loss of joint cartilage, articular bony fusion or severe destruction and disorganization of articular surfaces. Osteophyte size in both the knee and PIP joints were graded on a zero to four scale according to standardized criteria, the grading being a function of the largest osteophyte present in the area in question (Epidemiology of Chronic Rheumatism, 1963).
12
RHEUMATOLOGY AND REHABILITATION VOL. XV NO. 1 TABLE I COMPARISON OF OSTEOPHYTE SIZE AND JOINT NARROWING IN PIP JOINTS
Minimal narrowing (0-1)
Marked narrowing (2-4)
Osteophyte size
RA
DJD
Grade
265 24 3 1 0
70 55 30 3 0
191 55 14 8 0
4 9 11 5 6
0 1 2 3 4
10.0%
56.0%
27.0%
89.0%
293
158
268
35
Percentage of joints with osteophytes Total number of joints
COMPARISON OF MINIMALLY NARROWED PIP JOINTS WITH OSTEOPHYTES IN MALES TO ALL PATIENTS
Four out of 28 of all DJD and 11 out of 74 of all RA patients whose PEP joints were studied were males. In the male patients with DJD, 7 out of 17 joints with minimal narrowing had osteophytes while in the male patients with RA only 12 out of 64 joints with minimal narrowing had osteophytes. This compares to 111 out of 196 of all DJD minimally narrowed joints and 57 out of 392 of all RA minimally narrowed joints respectively. COMPARISON OF OSTEOPHYTE SIZE AND PIP JOINT NARROWING IN CASES OF RA DJD (AGED 50-59 YEARS)*
AND
Of 45 DJD minimally narrowed PD? joints aged 50-59, 10 (24%) compared with 56% of all DJD PIP joints (mean age 62) had osteophytes. Fifteen per cent (15/98) of RA minimally narrowed PEP joints aged 50-59, compared with 9 % of all RA PIP joints (mean age 55), had osteophytes. Twelve out of twelve DJD markedly narrowed PIP joints (aged 50-59), compared with 89 % of all DJD joints (mean age 62), had osteophytes. Thirty-eight per cent (27770) of RA markedly narrowed PIP joints aged 50-59, compared with 29% of all RA joints (mean age 55), had osteophytes. Of DJD markedly narrowed PIP joints, six out of 12 had osteophytes of grade 3 or 4, compared to only four out of 70 of correspondingly narrowed RA PIP joints. COMPARISON OF OSTEOPHYTE PREVALENCE AND KNEE-JOINT NARROWING IN CASES OF RA AND DJD WITH JOINT 'SPACE' OF 4 MM OR GREATER (AGED 50-59 YEARS)
In the medial femoral compartments with 'joint spaces' measuring over 4 mm, eight out of 23 (65%) of RA compared to 12 out of 12 DJD knees, had osteophytes. In the lateral compartments seven out of 23 (32 %) of RA had osteophytes compared to eight out of 13 (61 %) with DJD. Only two DJD knee medial and one DJD lateral compartments measured less than 3 mm and this small number precluded valid comparisons (see Table H). •Inadequate sample size precluded comparison of other age groups.
Downloaded from http://rheumatology.oxfordjournals.org/ at University of California, Santa Barbara on July 6, 2015
DJD
RA
SWEZEY AND BJARNASON: INfflBITION OF OSTEOPHYTES
13
TABLE H COMPARISON OF OSTEOPHYTE SIZE AND KNEE-JOINT NARROWING*
Joint 'space' Compartments
Lateral
3 mm or less
RA
DJD
RA
37 5 7 5 0
11 10 12 13 2
54
DJD
Grade
39 38 14 7 6
2 1 4 4 3
0 1 2 3 4
48
104
14
Total joints
34 11 4 1 2
17 14 17 6 3
48 26 16 14 2
3 1 1 0 0
0 1 2 3 4
52
57
106
5
Total joints
*The lower limit of normal knee-joint space in the adult is 4 mm (Ahlback, 1968). OSTEOPHYTE PREVALENCE IN MALE KNEE JOINTS COMPARED TO FEMALES
When the medial tibio-femoral compartment interspace measured 4 mm or greater, three of 19 males and 21 of 36 females with RA had osteophytes, while eight out of eight males and 29 out of 40 females with DJD had osteophytes. With corresponding measures of the lateral compartment, four out of 19 males and 20 out of 33 females with RA had osteophytes and 8 out of 10 males and 37 out of 50 females with DJD had osteophytes. With further tibio-femoral j oint-space narrowing (less than 4 mm) in the medial compartment, 11 out of 15 RA males, 49 out of 73 RA females and two out of three DJD males and 10 out of 11 DJD females had osteophytes. In the similarly narrowed lateral compartments, 11 out of 15 RA males, 49 out of 95 RA females and no (0/1) DJD male and three out of five DJD females had osteophytes. STATISTICAL ANALYSIS This study was designed to test the hypothesis that for a given measure of PIP or knee-joint narrowing no difference in the number of osteophytes in patients with RA and DJD would be detected. Analysis of the data utilized the chi-squared test of independence and f-test for significant differences between percentages. Using a 9 5 % confidence level, the following statements can be made: 1. In the proximal interphalangeal joints, osteophyte formation develops differently in RA from DJD in all joints with minimal and with marked narrowing. This held true for the joints of patients aged 50-59 in which marked joint space narrowing was present. 2. In the PEP joints of both RA and DJD patients, osteophyte development was not significantly different with regard to sex. 3. In the medial knee joint compartment at all degrees of joint narrowing osteophyte formation was significantly more prevalent in the DJD joints. In the lateral knee joint compartment this held true only when the joint space was 4 mm or more in width.
Downloaded from http://rheumatology.oxfordjournals.org/ at University of California, Santa Barbara on July 6, 2015
Medial
4 mm or more
Osteophyte size
14
RHEUMATOLOGY AND REHABILITATION VOL. XV NO. 1
DISCUSSION It has been shown that osteophyte formation occurs in the absence of radiographically demonstrable progressive knee articular damage (joint narrowing or subchrondral bone sclerosis), but with the exception of rheumatoid arthritis the converse (joint narrowing without osteophyte production) has not been described (Danielsson and Hernborg, 1970a, b). Bennett et al. (1942) have shown that degenerative changes in cartilage occur essentially universally by the second decade. This study has demonstrated that for the PIP and knee joints the extent of osteophyte formation in the RA joints is less than that found in comparably damaged (narrowed) degenerative joint disease joints. This is true for the proximal interphalangeal and the knee joints when corrections for age and sex are made. At the time of tabulation, data on the prior occupations and on race were not available and therefore the impact of these factors cannot be assessed. Even on comparison of weight-bearing and nonweight-bearing radiographs of the knees where severe degrees of narrowing not appreciated on routine non-weight-bearing films can be brought to light, the relative sparing of DJD changes in the rheumatoid knee was still observed (Ahlback, 1968). It should be noted in this connection that the nature of the rheumatoid process tends to produce diffuse joint-cartilage destruction, and thus asymmetrical narrowing seen on weightbearing films in DJD is frequently absent in RA. Further, in contrast to uncomplicated DJD, there is often little difference in the 'joint space' between weight-bearing and nonweight-bearing films of the rheumatoid knees. In order to minimize bias, wherever interobserver discrepancy in reading osteophytes occurs the larger size was chosen. In many of the RA joints, erosive thinning of articular margins occurred in such a manner as to project small bony 'spikes' at the joint margins. Although these did not appear to be proliferative osteophytic changes,
Downloaded from http://rheumatology.oxfordjournals.org/ at University of California, Santa Barbara on July 6, 2015
4. A statistically significant independence of disease (RA versus DJD) and osteophyte formation was found in patients aged 50-59 with a medial knee-joint compartment of less than 3 mm and where the joint space was greater than 4 mm in both the medial and lateral compartments. In the same age-group in those patients with less than 3 mm joint space, the lateral compartment showed no independence of disease and osteophyte formation. 5. Sex and osteophyte size were independent (osteophyte formation was more prevalent in females than in males) in the rheumatoid arthritic patients when the kneejoint medial and lateral compartment spaces were greater than 4 mm. This difference was not found when the joint spaces were 3 mm or less in the rheumatoids nor was there a statistically significant independence of sex and osteophyte formation at any degree of narrowing in either knee-joint compartment in the cases with DJD. In all the instances where there were differences between sexes, the males had fewer osteophytes than females, and since the vast majority of patients in this study were females, the bias of this data would therefore tend to minimize the probability of finding the significance of the conclusions drawn in this study that disease and osteophyte formation are independent, i.e. that osteophyte formation is less prevalent at any degree of joint narrowing in the rheumatoid arthritics than in osteoarthritis. 6. In all cases above where the statistical significance of the differences between osteophyte formation and disease could not be established, the trend was in the same direction as the results that were of statistical significance, i.e. at corresponding degrees of joint narrowing, osteophytes were less prevalent in RA than in DJD. Inadequate sample size appears to be the reason that these tests did not obtain significance.
SWEZEY AND BJARNASON: INHIBITION OF OSTEOPHYTES
15
A third, perhaps most intriguing possibility, is that some factor or factors of the inflammatory process of the RA joint play an inhibitory role in the development of secondary DJD. It has been demonstrated that the addition of thawed leucocytes to cultured monolayers of human synovial fibroblasts increases fibroblastic synthesis of hyaluronic acid and that this can be inhibited by anti-inflammatory drugs (Yaron et al., 1971). The possibility that metabolic products of leucocytes, synoviocytes, or antigenantibody interactions in the RA joint may result in the formation of inhibitors to ths proliferative DJD process merits further study. Because of the persistent nature of the inflammatory process in RA, the effects of the inflammatory process on both intra-articular and juxta-articular structures is prolonged. Thus, vascular alterations as a consequence of the inflammation may not abate and both intra and extra-articular perturbations resulting from the RA inflammation persist and may play an important role in creating the environment in a damaged joint which is not conducive to the emergence of DJD. CONCLUSION Using radiographically demonstrated joint narrowing as an index of joint damage, a reduced incidence of osteophyte size and number as a measure of osteoarthritic change in the PIP and knee joints of patients with RA as compared to DJD has been demonstrated. The possible pathogenetic factors discussed include reduced joint use in RA,
Downloaded from http://rheumatology.oxfordjournals.org/ at University of California, Santa Barbara on July 6, 2015
they were so interpreted for the purpose of this study. Therefore the actual incidence of osteophytic formation in the rheumatoid joints may have been somewhat lower than recorded. Whereas the association of hypertrophic osteoarthritic changes occurring as a primary process or secondary to a variety of joint traumata—direct joint injury, obesity, haemorrhage, joint instability in neuropathic joints—and secondary to joint sepsis is well recognized; the relative sparing of the rheumatoid joint by osteoarthritic changes has not been adequately appreciated (Traut, 1956). Documentation of an apparent inhibiting effect by the RA process on the development of secondary degenerative joint disease exclusive of the influence of age or sex raises several questions. Does the presence of RA with its accompanying pain and inhibition of joint use cause a relative sparing of the joint from the mechanical stresses which may act as a stimulus to the hypertrophic degenerative joint disease changes ? The role of repeated joint trauma in DJD is well known and the sparing effect of paralysis on Heberden's node formation has also been documented (Stecher, 1950; Kellgren, 1961). Further, it is the authors' impression that degenerative changes in RA are most commonly observed in those patients who seem to have great tolerance for pain and who use their joints despite the obvious inflammatory involvement, and in those RA patients whose previous inflammatory disease has abated and who resume active function on previously damaged joints (Swezey et al., 1972-3). Although these observations would suggest that an age factor (time for disease remission and further joint abuse) might play a role, comparison of age-matched groups did not support this possibility. A second consideration requiring further study is the effect of osteoporosis on DJD. Both generalized and juxta-articular bone demineralization are common factors in RA. Foss and Byers (1972) documented the remarkable lack of DJD of the hip in the presence of diminished femoral bone density. Radin and Paul have recently shown (1970) that resilience of subchrondral bone plays a role in joint degeneration. The possibility that a change in resilience of subchondral bone as a consequence of osteoporosis might cause a greater cushioning effect on joint surfaces has been suggested as a mechanism in the sparing of DJD changes of the hip and may apply to the RA joint as well.
16
RHEUMATOLOGY AND REHABILITATION VOL. XV NO. 1
increased resilience of juxta-articular bone associated with osteoporosis in RA and the possibility that products of inflammation in the RA joint act locally or via juxta-articular vascular mechanisms to inhibit the secondary changes of DJD. Further study of the possible pathogenic mechanisms including the roles of joint stress, juxta-articular osteoporosis and possible effects of the inflammatory reaction of the rheumatoid joint on intra-articular structures and juxta-articular vascular phenomena are needed.
REFERENCES
(1968) "Osteoarthritis of the Knee: A Radiographic Investigation". Acta Radiol. Suppl. 277, 7-72. BENNETT, G. A., WAINE, H. and BAUER, W. (1942) Changes in the Knee Joint at Various Ages. New York: The Commonwealth Fund, p. 21. DANIELSSON, L. and HERNBORO, J. (1970) "Morbidity and Mortality of Osteoarthritis of the Knee (Gonarthrosis) in Malmo, Sweden". Clin. Orthop., 69, 224-6. and (1970) "Clinical and Rocntgenologic Study of Knee Joints with Osteophytes". Clin. Orthop., 69, 302-12. Epidemiology of Chronic Rheumatism, Vol. 2 (1963) Atlas of Standard Radiographs of Arthritis. Philadelphia: F. A. Davis. FLETCHER, D. E. and ROWLEY, K. A. (1965) "The Radiological Features of Rheumatoid Arthritis". Br. J. Radiol., 25, 663. Foss, M. V. L. and BYERS, P. D. (1972) "Bone Density, Osteoarthrosis of the Hips and Fracture of the Upper End of the Femur". Ann. Rheum. Dis., 31, 259-64. KELLGREN, J. H. and LAWRENCE, F. A. (1952) "Radiological Assessment of Osteoarthritis". Br. J. Radiol., 25, 282-95. (1961) "Osteoarthrosis in Patients and Populations". Br. Med. J., 2, 1-6. KNAGGS, R. LAWFORD (1926) The Inflammatory and Toxic Diseases of Bone. London: Sompkin, Marshall, Hamilton, Kent and Co. Ltd., pp. 165-6. MARTEL, WILLIAM (1968) "Radiologic Manifestations of Rheumatoid Arthritis with Particular Reference to the Hand, Wrist and Foot". Med. Clin. N. Am., 52, 663. NICHOLS, E. H. and RICHARDSON, F. L. (1909) "Arthritis Deformans". / . Med. Res. (New Series, Vol. 16), 21, No. 2, 149-222. RADIN, E. L. and PAUL, I. L. (1970) "Does Cartilage Compliance Reduce Skeletal Impact Loads T' Arthritis Rheum., 13, 139-44. ROPES, M. W., BENNETT, G. A., COBB, S., JACOX, R. and VESSAR, R. A. (1956) "A Committee of the American Rheumatism Association Proposed Diagnostic Criteria for Rheumatoid Arthritis, No. 4". Bull. Rheum. Dis., 7, 121-4. STECHER, R. M. (1950) "Heberden's Nodes: A Clinical Description of Osteoarthritis of the Finger Joints". Ann. Rheum. Dis., 14, 1-10. SWEZEY, R. L., BJARNASON, D., ALEXANDER, S. and FORRESTER, D. B. (1972-3) "Resorptive Arthropathy and the Opera Glass Hand Syndrome". Semin. Arthritis Rheum., 2, 191-243. TRAXTT, E. F. (1956) "Degenerative Arthritis: Its Causes and Management". Med. Clin. N. Am., 40, 63-78. YARON, M., YARON, I. and ALLALOUF, D. (1971) "Effect of Some Anti-inflammatory Drugs on Fibroblast-Leukocyte Interaction in vitro". Ann. Rheum. Dis., 30, 613-18. AHLBACK, SVEN
Downloaded from http://rheumatology.oxfordjournals.org/ at University of California, Santa Barbara on July 6, 2015
ACKNOWLEDGMENTS
I would like to thank Steven Salzberg, M.S., Senior Statistician, Los Angeles County/ U.S.C. Medical Center, for his assistance in obtaining tho statistical information in this report. This work was supported in part by a grant from the Southern California Chapter of the Arthritis Foundation.
