To determine whether there is a systemic predisposition to chondrocalcinosis (CC) and to examine the association between CC and osteoarthritis (OA) at distant joints.
To determine whether there is a systemic predisposition to chondrocalcinosis (CC) and to examine the association between CC and osteoarthritis (OA) at distant joints.
We performed a cross-sectional study embedded in the Genetics of Osteoarthritis and Lifestyle (GOAL) database (n = 3,170). All GOAL participants have had radiographs of the knees, hands, and pelvis performed. These were scored for OA at the knee, hip, wrist, and metacarpophalangeal (MCP) joints and for CC at the knee, hip, wrist, and symphysis pubis joints and for MCP joint calcification. Systemic predisposition to CC was established using cluster analysis. Odds ratios (95% confidence intervals) were used to examine the association between CC at the index and distant joints, CC and OA at the same joint, and index joint OA and distant joint CC. We adjusted for age, sex, and body mass index, and for distant joint OA, if required.
Joints with CC clustered together. This was also observed when participants with OA were excluded from the analysis. CC at each joint was associated with CC at distant joints. Knee and wrist OA but not hip OA was associated with CC at the same joint. MCP joint OA was associated with MCP joint calcification. Knee OA was associated with CC at other joints, and this was independent of OA at the distant joint. There was no association between hip OA and distant joint CC.
There is a systemic predisposition to the apparently sporadic CC. OA is associated with CC at the same joint and at distant joints, except hip OA, which is not associated with hip CC or with CC at distant joints.
Chondrocalcinosis (CC), which predominantly results from calcium pyrophosphate dihydrate crystal deposition disease (CPDD), occurs commonly at the knees, wrists, hips, and symphysis pubis ([1-4]). Extensive polyarticular CC, where a systemic predisposition is apparent, has been reported in early-onset familial CC or in several relatively uncommon metabolic conditions, e.g., hemochromatosis, hyperparathyroidism, hypomagnesemia, and hypophosphatasia ([5-16]). However, whether the common, apparently sporadic, CC results from a generalized predisposition or from a local joint-specific abnormality such as osteoarthritis (OA) remains unclear ([17, 18]).
Although the association between OA and CC at the same joint is well established ([19-21]), the association between OA at one joint and CC at specific distant joints has not been systematically examined. That the association between OA and CC may be joint specific is suggested by the findings of a large community study that reported that CC at any knee, at any hip, or at the symphysis pubis is associated with knee OA, but not with hip OA (). However, a joint-specific analysis examining the association between OA and CC at the same joint was not performed (). Similarly, several other small studies that have examined the association between CC at any joint and OA at other joints have not examined the association between CC and OA in specific joint pairs ([23-26]).
The objectives of this study were to examine if there is a generalized articular predisposition to CC and if OA is associated with CC locally, i.e., at the same joint, and at distant joints.
A cross-sectional analysis was undertaken using data from the Genetics of Osteoarthritis and Lifestyle (GOAL) study. GOAL comprises 3,170 individuals, including 1,007 with clinically severe hip OA, 1,042 with clinically severe knee OA, and 1,121 without knee or hip OA, recruited from 2002–2006. Clinically severe OA was defined as OA with symptoms severe enough to require joint replacement surgery in the presence of radiographic changes of OA. Since the GOAL study was originally set up to examine the genetic associations of large joint OA, a group without clinical and radiographic knee or hip OA was also recruited. In order to minimize unnecessary radiographic screening, this group was selected from those patients who had an intravenous urogram performed at the City Hospital Nottingham. This study was approved by the Nottinghamshire Research Ethics Committee, UK.
The details of the GOAL study have been published previously (). Briefly, all participants completed a detailed questionnaire, were examined by a research metrologist, gave urine and blood samples, and underwent standardized radiographic assessments, including posteroanterior weight-bearing semiflexed knee radiographs using the SynaFlexer positioning frame (Synarc), skyline views of the patellofemoral (PF) joints, supine pelvis radiographs for the hips and symphysis pubis, and anteroposterior hand views (including the wrists).
All radiographs were graded for CC at the knees, wrists, hips, and symphysis pubis; for metacarpophalangeal (MCP) joint calcification; and for structural changes of OA by a single experienced senior research metrologist (SD). Tibiofemoral (TF) joints, PF joints, hips, and MCP joints were allocated a Kellgren/Lawrence (K/L) score for the severity of OA (). The radiocarpal, midcarpal, and common carpometacarpal joints (between the distal row of the carpal bones and the proximal end of the second to fourth metacarpal bones) were scored on a 0–3 scale for joint space narrowing (JSN) and osteophytosis (where 0 = absent, 1 = possible, 2 = definite, and 3 = marked), and sclerosis and cysts were scored as present or absent ().
Articular CC was present if there was either a linear or nummular calcification in the fibrocartilage or hyaline articular cartilage of the knees, wrists, hips, and symphysis pubis, and synovial/capsular calcification or linear hyaline cartilage calcification at any of the 5 MCP joints. OA was present at the knees, hips, and MCP joints if the K/L score was ≥3 at the knee, ≥2 at the hip, and ≥3 at any of the second to fifth MCP joints (i.e., the K/L grade at each joint that requires definite JSN and definite osteophytes). Wrist OA was present if there was definite JSN, i.e., a JSN score of ≥2 on a 0–3 scale in either the radiocarpal, midcarpal, or common carpometacarpal joint. This is unlikely to be due to inflammatory arthritis because patients with preexisting inflammatory arthritis were excluded from the GOAL study.
Data about age at the hospital visit (in years), sex, current weight (in kg), and current height (in cm) were collected at the first visit. Current weight and height were used to calculate the current body mass index (BMI; in kg/m2).
Mean ± SD and number (percentage) were used to describe continuous and categorical variables, respectively. Cluster analysis was used to examine if CC occurs in multiple joints in the same individual more often than by chance. For this, the number of subjects who would be expected to have CC at 0, 1, 2, 3, 4, or 5 joints (expected frequency) was calculated assuming that the presence of CC at one joint is independent of its occurrence at other joints. The number of participants who had CC at 0, 1, 2, 3, 4, or 5 joints (observed frequency) was compared with the expected frequency using a chi-square test. For this analysis, both knees, both hips, both wrists, symphysis pubis, and MCP joints in both hands were considered as one joint. The minimum and maximum number of sites at which CC could be present was 0 and 5, respectively.
Odds ratios (ORs) and 95% confidence intervals (95% CIs) were calculated to examine the associations between CC at any index joint (e.g., any knee) and distant joint CC (e.g., any wrist), and bilateral CC at the index joint (e.g., both knees) and CC at any distant joint (e.g., any wrist). All joint pairs examined are shown in Figure 1A. Binary logistic regression was used to adjust for age (tertiles), sex (male = 0, female = 1), current BMI (tertiles), and OA at the distant joint (absent = 0, present = 1). Similarly, ORs (95% CIs) were calculated to examine the association between OA and CC at the same joint, and were adjusted for age, sex, and BMI. ORs (95% CIs) were calculated to examine the association between OA at one joint and CC at the contralateral joint, and were adjusted for age, sex, BMI, and OA at the contralateral joint. Similarly, ORs (95% CIs) were calculated to examine the association between OA at one joint (e.g., any knee) and CC at the distant joint (e.g., any wrist), and were adjusted for age, sex, BMI, and OA at the distant joint. All joint pairs examined are shown in Figure 1B. All statistical analysis was done with SPSS, version 15. Statistical significance was set at P values less than 0.05.
Four hundred thirty-five (13.7%) of the 3,170 GOAL participants had CC at any joint. The demographic profile of participants in the GOAL study and the prevalence of OA at different joints have been published previously (). However, in brief, the mean ± SD age was 66.6 ± 7.9 years and there were 1,536 women (48.5%). The overall prevalence of knee, hip, and wrist OA in the GOAL participants was 51.0%, 36.9%, and 11.0%, respectively, while the overall prevalence of knee, hip, and wrist CC in the GOAL participants was 8.0%, 5.0%, and 6.9%, respectively. Symphysis pubis CC and MCP joint calcification were present in 3.6% and 1.5% of the GOAL participants, respectively.
Only patients for whom information about CC and MCP joint calcification at all joints was available were included in this analysis (n = 3,118 [98.4%], 428 with CC). Joints with CC were clustered together more often than would be expected by chance alone (Table 1). This clustering was also present when participants who had OA at any knee, hip, or wrist joint were excluded (Table 1). Similar to this, CC at any one joint (e.g., any knee) was associated with CC at distant joints (e.g., any wrist). The association was present for all joint pairs and was independent of age, sex, BMI, and OA at the distant joint (Table 2). Bilateral CC at one joint (e.g., both knees) was associated with CC or MCP joint calcification at distant joints compared to unilateral CC at that joint for all joint pairs, except for the association between CC at both wrists and both hips and symphysis pubis CC (Table 3). Calcification of MCP joints in both hands was associated with hip CC compared to calcification in MCP joints in one hand. However, this association was not significant for other joint pairs (Table 3).
|No. of joints with CCa||All participantsb||Participants without OAc|
|Knee||Wrist||Hip||Symphysis pubis||MCP joints|
|Knee||–||27.23 (19.14–38.74)a||19.63 (13.67–28.19)a||17.60 (11.60–22.60)a||23.97 (12.57–45.70)a|
|Wrist||22.35 (15.89–31.43)a||–||16.88 (11.59–24.57)a||15.56 (10.18–23.79)a||66.95 (30.05–149.14)a|
|Hip||18.87 (13.04–27.30)a||19.67 (13.29–29.10)a||–||37.39 (24.34–57.42)a||18.72 (10.05–34.85)a|
|Symphysis pubis||16.82 (11.01–25.69)a||17.03 (11.00–26.36)a||37.29 (24.29–57.25)a||–||5.40 (2.39–12.21)a|
|Any MCP joint||21.04 (11.04–40.11)a||74.51 (33.10–167.74)a||19.42 (10.56–35.72)a||5.17 (2.33–11.47)a||–|
|Knee||Wrist||Hip||Symphysis pubis||MCP joints|
|Knee||–||2.23 (1.27–3.93)a||3.79 (2.06–6.98)a||2.21 (1.17–4.17)a||2.78 (1.13–6.86)a|
|Wrist||3.38 (1.82–6.24)a||–c||2.15 (1.44–4.04)a||1.86 (0.93–3.72)b||2.58 (1.11–5.99)a|
|Hip||3.15 (1.51–6.56)a||2.69 (1.27–5.73)a||–||1.52 (0.77–2.97)b||4.22 (1.53–11.68)a|
|Any MCP joint||2.37 (0.43–12.96)b||–c||6.91 (1.40–34.26)a||2.39 (0.39–14.82)b||–|
There was an association between OA and CC at the same knee and wrist joint, and between OA at any MCP joint in one hand and MCP joint calcification in the same hand (Table 4). This association was present when TF joints and PF joints were analyzed separately. However, there was no association between CC and OA at the hip (Table 4). Since the lack of an association between OA and CC at the hips may be due to an inability to visualize CC in hips with severe OA, further analysis was carried out after excluding hips with K/L grade 4 OA. Even after this, there was no association between OA and CC at the right hip (OR 1.27, 95% CI 0.84–1.90) and at the left hip, respectively (OR 0.78, 95% CI 0.44–1.38).
|OR (95% CI)a||ORadj (95% CI)b||OR (95% CI)a||ORadj (95% CI)b|
|Knee||2.39 (1.79–3.20)c||2.20 (1.62–2.99)c||2.78 (2.04–3.79)c||2.59 (1.87–3.60)c|
|TF||2.01 (1.52–2.67)c||1.83 (1.36–2.46)c||2.33 (1.73–3.13)c||2.22 (1.62–3.03)c|
|PF||1.75 (1.29–2.37)c||1.59 (1.16–2.19)c||1.84 (1.32–2.54)c||1.65 (1.17–2.32)c|
|Hip||1.08 (0.73–1.59)d||1.05 (0.72–1.55)d||0.72 (0.44–1.20)d||0.68 (0.41–1.13)d|
|Wrist||4.46 (3.24–6.13)c||3.26 (2.33–4.55)c||4.42 (3.26–6.00)c||3.28 (2.36–4.48)c|
|Any MCP||3.37 (2.97–7.38)c||2.42 (1.09–5.41)c||7.76 (3.85–15.64)c||5.58 (2.72–11.47)c|
Left knee OA and TF joint but not PF joint compartment OA in the left knee were associated with right knee CC after adjusting for age, sex, BMI, and right knee OA (Table 5). However, right knee OA, TF joint compartment OA, and PF joint compartment OA in the right knee were not associated with left knee CC after adjusting for age, sex, BMI, and left knee OA. There was no association between OA in one hip and CC in the contralateral hip. However, OA in one wrist was associated with contralateral wrist CC, independent of age, sex, BMI, and OA at the contralateral joint. A similar pattern was present at the MCP joints.
|Joints||Right OA and left CCa||Left OA and right CCa|
|OR (95% CI)b||ORadj (95% CI)c||OR (95% CI)b||ORadj (95% CI)c|
|Knee||2.28 (1.68–3.09)d||1.22 (0.82–1.83)e||2.26 (1.70–3.02)d||1.53 (1.04–2.24)d|
|TF||2.10 (1.56–2.82)d||1.32 (0.91–1.92)e||2.27 (1.71–3.07)d||1.88 (1.31–2.69)d|
|PF||1.46 (1.05–2.03)d||0.94 (0.61–1.54)e||1.61 (1.17–2.21)d||1.13 (0.74–1.72)e|
|Hip||1.13 (0.73–1.76)e||1.39 (0.84–2.31)e||1.05 (0.70–1.58)e||0.98 (0.61–1.59)e|
|Wrist||5.02 (3.68–6.83)d||2.67 (1.81–3.95)d||4.00 (2.92–5.48)d||1.78 (1.19–2.66)d|
|Any MCP||7.41 (3.71–14.82)d||3.18 (1.31–7.46)d||3.42 (1.44–8.12)d||1.54 (1.25–4.53)d|
Knee OA was associated with CC at each of the distant joints examined, and this was independent of age, sex, current BMI, and distant joint OA (Table 6). However, there was no association between hip OA and CC at any of the distant joints. Wrist OA was associated with knee CC and with MCP joint calcification, whereas MCP joint OA was associated with wrist CC. These associations were independent of age, sex, BMI, and OA at distant joints.
|OR (95% CI)a||ORadj (95% CI)b|
|Hip CC||1.99 (1.35–1.67)c||1.99 (1.38–2.85)c|
|Wrist CC||4.45 (3.14–6.30)c||3.27 (2.25–4.74)c|
|Symphysis pubis CC||1.83 (1.27–2.63)c||2.06 (1.40–6.50)c|
|MCP joint CC||4.73 (2.20–10.17)c||2.91 (1.40–6.50)c|
|Knee CC||1.34 (1.04–1.74)c||1.30 (0.99–1.70)d|
|Wrist CC||1.47 (1.11–1.94)c||1.25 (0.92–1.64)d|
|Symphysis pubis CC||1.10 (0.78–1.57)d||1.08 (0.76–1.53)d|
|MCP joint CC||1.06 (0.78–1.57)d||0.89 (0.49–1.62)d|
|Knee CC||3.64 (2.41–5.49)c||2.42 (1.58–3.73)c|
|Hip CC||1.51 (0.80–2.85)d||1.38 (0.72–2.62)d|
|Symphysis pubis CC||1.05 (0.48–2.30)d||1.03 (0.47–2.26)d|
|MCP joint CC||6.38 (3.18–12.81)c||3.86 (1.88–7.93)c|
|MCP joint OA|
|Knee CC||1.79 (1.32–2.41)c||1.14 (0.83–1.57)d|
|Hip CC||1.59 (1.09–2.32)c||1.38 (0.93–2.03)d|
|Wrist CC||2.66 (1.97–3.58)c||1.48 (1.06–2.04)c|
|Symphysis pubis CC||1.19 (0.78–1.85)d||1.10 (0.71–1.72)d|
This is the first study to demonstrate a systemic predisposition to the apparently sporadic CC. It validates the association between knee OA, wrist OA, and CC at the same joint, and is the largest study to show that OA at these joints is associated with CC at distant joints. However, there was no association between hip OA and CC at the hip or at distant joints. Therefore, the association between OA and CC varies according to the joint with OA.
This study provides evidence for a generalized predisposition to CC and MCP joint calcification independent of OA. Moreover, bilateral CC at a single joint area was associated with CC at distant joints compared to unilateral CC at the referent joint area for most joint pairs examined. This suggests a dose-response relationship between the extent of crystal deposition in a single joint area and CC at distant joints. These findings suggest that the apparently sporadic CC at least in part results from a systemic predisposition and is not only a result of structural arthropathy or other local predisposition at the index joint. Such a systemic predisposition to CC could result from either high synovial fluid inorganic pyrophosphate (PPi) levels, cartilage matrix changes that favor CPDD, or both mechanisms occurring in multiple joints in the predisposed individual ([16, 30, 31]). This is not due to a systemic increase in PPi levels, since serum and urinary PPi levels are comparable in those with and without CPDD ().
The association between OA and CC is well recognized ([1, 2]) and is independent of age at least at the knee ([19-21]). This large study validates the association between OA and CC and also convincingly demonstrates that the joint-specific association between OA and CC where it exists (i.e., at the knee, wrist, and MCP joints) is independent of age, sex, and BMI. There was no association between OA and CC at the hip. Similar findings were reported in a small age-matched, hospital-based, case–control study and in an age- and sex-matched autopsy study ([32, 33]). The reasons underlying this lack of association between hip OA and hip CC is not well understood.
Both TF joint OA and PF joint OA were associated with knee CC, with overlapping 95% CIs. This suggests that there is no difference in the compartmental location of OA in knees with and without CC. The findings of this study are supported by a previous hospital-based study of 300 knees with symptomatic radiographic OA (). However, this is not in keeping with other reports that suggest that OA occurs more commonly at the PF joints than at the TF joints in knees with CC ([26, 35]). Since the current study was adjusted for confounding factors and has a large sample size, it in conjunction with the previous study () suggests that knee CC is not associated with any specific compartmental location of knee OA.
Previous studies have examined the association between CC at any joint () or between CC in the lower extremity joints () and OA at specific joints. The association between CC and OA in specific joint pairs has not been examined before. This study reports that except for hip OA, OA is associated with CC at distant joints. In keeping with these findings, a large community-based study did not find an association between CC at the hip and symphysis pubis and knee and hip OA (). Because there was no association between hip OA and hip CC when hips with K/L grade 4 OA were excluded from the analysis, it is unlikely that the lack of an association between hip OA and hip CC is because hip CC is difficult to visualize with coexistent severe radiographic hip OA. Moreover, the lack of an association between hip OA and CC at distant joints provides additional validity to this finding. The association between CC and OA at the wrist and MCP joints is in keeping with previous studies. For example, a primary care–based study of 26 CC cases and 104 controls showed that CC at any site is associated with MCP joint OA (adjusted OR 3.1, 95% CI 1.1–8.8) (). Several hospital-based studies also suggest that wrist CC is associated with trapezioscaphoid arthropathy ([37-39]).
This is the first study to look for a systemic predisposition to CC and MCP joint calcification and to systematically examine the association between OA and CC. The analysis of the association between OA and CC for specific joint pairs is novel. However, there are many caveats to these findings. First, this is a hospital-based study carried out within a large case–control study that was assembled to examine risk factors of large joint OA. Therefore, the study population does not resemble a community-based sample. Two-thirds of the participants had severe symptomatic hip or knee OA and the remaining participants had no radiographic or clinical features of hip or knee OA. This therefore limits the generalizability of these findings. Moreover, since the participants were selected for severe symptomatic large joint OA, the results from this study may be confounded by the participants' hip or knee OA status. However, in order to minimize the confounding, we adjusted for OA, and the cluster analysis has been replicated in a subset of the GOAL population without OA. In this study, we used plain radiographs, which are relatively insensitive at detecting CC, to define cases with CC. Musculoskeletal ultrasound may have been more sensitive for the detection of CC at the knee and wrist, but was impractical to apply to 7 sites in such a large number of participants. Moreover, it is unknown whether ultrasound has a greater sensitivity for the detection of hip CC compared to plain radiographs. Nevertheless, the findings of this study should be confirmed in a smaller study using ultrasound. A large number of statistical tests were carried out, which may result in a Type 1 error. The results of this study should therefore be interpreted with caution. Finally, CC is difficult to visualize in joints with severe JSN. Therefore, the lack of an association between OA and CC at the hip may be due to a Type 2 error. However, for reasons discussed above, we do not think this is a true lack of an association. However, further research using synovial fluid aspiration is required to confirm the lack of an association between hip OA and hip CPDD. Moreover, we are unable to examine the previously reported lack of an association between knee CC and knee OA progression due to the cross-sectional nature of this study ().
In conclusion, CC occurs due to a systemic predisposition that is independent of OA. There was an association between OA at one joint and CC at the same joints and at distant joints, except for hip OA, which was not associated with CC. This study, which provides validity to previous observations of an association between CC and OA, needs to be confirmed in a community-based population.
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Abhishek had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design. Abhishek, S. Doherty, Muir, Zhang, M. Doherty.
Acquisition of data. S. Doherty, M. Doherty.
Analysis and interpretation of data. Abhishek, Maciewicz, Zhang, M. Doherty.
AstraZeneca had no role in the study design, data collection, data analysis, and writing of the manuscript, as well as approval of the content of the submitted manuscript. Publication of this article was not contingent on the approval of AstraZeneca.