Differences in multijoint radiographic osteoarthritis phenotypes among African Americans and Caucasians: The Johnston County Osteoarthritis Project

Authors


  • The findings and conclusions reported herein are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.

Abstract

Objective

To define and contrast multiple joint radiographic osteoarthritis (OA) phenotypes describing hand and whole-body radiographic OA among African Americans and Caucasians.

Methods

We conducted a cross-sectional analysis in the Johnston County Osteoarthritis Project, using radiographic data for the hands, tibiofemoral (TF) joints, patellofemoral joints, hips, and lumbosacral (LS) spine. Radiographs were read for OA by a single radiologist using standard atlases. Fisher's exact test, with correction for multiple comparisons, was used to compare phenotype frequencies by race and sex. Logistic regression was used to provide odds ratios, which were adjusted for sex, age, and body mass index (BMI).

Results

Sixteen mutually exclusive hand (n = 2,083) and 32 whole-body (n = 1,419) radiographic OA phenotypes were identified. We found that in comparison to Caucasians, African Americans had significantly less frequent radiographic OA of the distal interphalangeal joints, both in isolation and in combination with other hand joint sites, but had comparable frequencies of radiographic OA for other hand joint sites. Moreover, African Americans had less frequent radiographic OA of the hand, both in isolation and in combination with other joint sites, as compared to Caucasians. In contrast, African Americans had more than twice the odds of isolated OA of the TF joint and 77% higher odds of radiographic OA of the TF joint and LS spine together as compared to Caucasians.

Conclusion

Even after adjustment for sex, age, and BMI, African Americans were less likely than Caucasians to have hand radiographic OA phenotypes, but more likely to have knee radiographic OA phenotypes involving the TF joint. African Americans may have a higher burden of multiple large-joint OA involvement not captured by most definitions of “generalized OA.”

Osteoarthritis (OA) commonly affects multiple joints, although a universally accepted definition of “generalized” OA has yet to be established. A variety of methods for defining generalized OA, such as counting the number of affected joints or summing radiographic grades across multiple joints, have been used in various studies (1–6). These definitions often result in a sum score or cutoff point that defines individuals with generalized OA. While this is useful for determining case status and overall OA burden, it does not provide information about the full multijoint OA phenotype of the individual. For prognosis and treatment planning, it may be of use to know which joints are most often involved together and which joints are rarely simultaneously involved in a given individual.

A few studies have reported on patterns of radiographic or symptomatic OA involvement using various combinations of joint sites (5, 7–9). Clinically identified nodal changes in the hands have been associated with an increased risk of undergoing knee or hip joint replacement (9) and with an increased predisposition to knee OA after meniscectomy (6, 10) and have been variably associated with hip OA in other studies (11, 12). OA at one site has also been found to predict the development and/or progression of OA at distant sites (13, 14). The Genetics of Generalized Osteoarthritis study, which enrolled affected sibling pairs based on clinical evidence of hand OA, found that a substantial proportion of individuals had radiographic OA of the knee, hip, or knee and hip in addition to hand OA (4).

Multiple joint involvement in OA has been most frequently studied among women, and almost exclusively in Caucasians. Among African populations, lower frequencies of nodal hand OA have been seen as compared to the frequencies in Caucasian populations (15). Approximately 10% of men and women in a black South African population had radiographic OA in 3 or more joint groups, which is less than that reported in contemporary studies of Caucasians (15). A comparison study of individuals in Jamaica showed a higher frequency of distal interphalangeal (DIP) joint and knee radiographic OA and a lower frequency of nodes and of first metatarsophalangeal joint radiographic OA than UK subjects, as well as a reduced frequency of lumbar spine radiographic OA among Jamaican men (16). Sowers et al (17), in the only study of radiographic OA patterns among African Americans, found an increased frequency of both hand and knee OA among African American women, along with an increased frequency of knee OA alone and of metacarpophalangeal (MCP) joint OA, as compared to Caucasian women.

Our group of investigators has evaluated differences in the knee and hip joints between African Americans and Caucasians in the Johnston County Osteoarthritis (JoCo OA) Project, a prospective, community-based cohort of individuals in rural North Carolina. Compared to Caucasians, African Americans in this population have a higher prevalence of radiographic and symptomatic knee OA and a similar to slightly increased prevalence of radiographic and symptomatic hip OA (18, 19). The purpose of the current analysis was to examine potential differences in mutually exclusive multijoint radiographic OA phenotypes among African American and Caucasian men and women at the joint sites most frequently affected by OA: the hands, hips, tibiofemoral (TF) and patellofemoral (PF) joints, and the lumbosacral (LS) spine.

SUBJECTS AND METHODS

Study population.

The analyses in this study used data from the JoCo OA Project, a community-based prospective cohort study of noninstitutionalized African American and Caucasian men and women ages 45 years and older, both with and without OA, living in rural North Carolina, as previously described (18). All participants signed informed consent forms and completed 2 home interviews and 1 clinic visit with physical examination, including functional measures and radiography, administered by trained study personnel. Hand and spine radiographs were added to the study (previously collecting only knee and hip radiographs) at the cohort-enrichment followup (in 2003–2004) and at the second followup (in 2006–2010), so data from these time points were used for the current analysis.

Starting from a combined total of 2,121 subjects, 2 subsamples were analyzed: 2,083 individuals with complete data for all hand joints (the hand radiographic OA phenotypes group), and a subset of 1,419 subjects who also had complete data for the other 4 joint sites (the whole-body radiographic OA phenotypes group) (Figure 1).

Figure 1.

Flow chart of the inclusion of individuals with radiographic data for assessments of hand and whole-body radiographic osteoarthritis (OA). TFJs = tibiofemoral joints; CMC = carpometacarpal; IP = interphalangeal; PFJ = patellofemoral joint.

Self-reported age, sex, and race data were obtained from interviewer-administered questionnaires, while the body mass index (BMI; in kg/m2) was calculated from the height (in cm) and the weight (in kg), which were measured during the clinic examination by trained study examiners. This cross-sectional analysis included demographic, clinical, and radiographic data collected at the same time for each participant (during either 2003–2004 or 2006–2010). The JoCo OA Project has been continuously approved by the Institutional Review Boards of the University of North Carolina and the Centers for Disease Control and Prevention.

Radiographs.

Bilateral radiographs of the hands, knees, and hips, as well as radiographs of the spine were obtained at a single clinic visit for each participant and were interpreted as follows. Posteroanterior radiographs of the hands were graded according to the Kellgren/Lawrence (K/L) scale (20) at each of 30 joints for each hand (the DIP, proximal interphalangeal [PIP], MCP, carpometacarpal [CMC], and thumb IP and MCP joints). Fixed-flexion, weight-bearing posteroanterior views of the TF joints obtained using a SynaFlexer device (CCBR-Synarc) were graded according to the K/L scale. Sunrise views of the PF joints were graded for osteophytes using the Burnett atlas of OA (21). Radiographs of the PF joints were added later in the study, and only ∼70% of them had been read at the time of this analysis. Anteroposterior supine views of the pelvis were graded according to the K/L scale. Lateral view radiographs of the LS spine (patient lying on his or her left side) were graded for osteophytes and disc narrowing at 5 levels (L1/2 through L5/S1) using the Burnett atlas of OA. LS spine and hip radiographs were not obtained on women under the age of 50 years. All radiographs were read by a single experienced musculoskeletal radiologist (JBR) and were previously shown to have high intra- and interrater reliability (κ = 0.89 and κ = 0.86, respectively) (22).

We defined radiographic OA of the TF joint or the hip as a K/L grade of ≥2 in either TF joint or either hip joint, respectively. Replaced knees were categorized as having OA if the participant reported OA as the reason for the replacement or if radiographic OA was present in the contralateral knee. Replaced hips were categorized as having OA only if the participant reported OA as the reason for the joint replacement, since hips are more often subject to replacement due to fracture. For the PF joint, any osteophyte graded ≥2 was considered to indicate radiographic OA of the PF joint. For radiographic OA of the hand, we used a composite definition requiring bilateral involvement, at least 1 DIP joint with a K/L grade of ≥2, and at least 3 joints (DIP, PIP, or CMC) involved (4). OA of a hand joint group (e.g., OA of the DIP joints) was defined as a K/L grade of ≥2 in any joint in the group. Radiographic OA of the LS spine was recorded if both osteophytes and disc narrowing of grade ≥1 were simultaneously present in at least 1 vertebral level.

Statistical analysis.

We did not use population-based weightings, since this analysis used a sample from the JoCo OA Project that included both the population-based original followup sample and the cohort-enrichment sample. The unit of analysis is the person throughout. Descriptive statistics were calculated for age, BMI, race, and sex for the sample with complete data for hand phenotypes (n = 2,083) as well as for the subsample with complete data for whole-body phenotypes (n = 1,419). Frequencies of radiographic OA were determined using the above definitions for each subsample. Then, 16 mutually exclusive phenotypes were defined for hand joint groups and 32 mutually exclusive phenotypes for whole-body phenotypes, representing all possible combinations of multijoint radiographic OA. Since each phenotype was mutually exclusive, the referent group for a given phenotype was the combination of all others; therefore, the sample size remained constant. Frequencies were calculated for each phenotype by race and by sex, and comparisons were made using Fisher's exact test because of the small cell sizes in the contingency tables. We used the Hochberg method (23) of correction for multiple comparisons via the MultProc procedure in Stata (24).

To allow for adjusted analysis using 4 explanatory variables (race, sex, age, and BMI), only the phenotypes seen in at least 40 individuals (∼10 events per covariate) were assessed using logistic regression models (25). For the hand phenotypes sample (n = 2,083), this represented 8 of 16 phenotypes (those occurring in at least 2% of this sample): no hand OA, DIP joint only, PIP joint only, CMC joint only, DIP and PIP joints only, DIP and CMC joints only, DIP/PIP/CMC joints, and DIP/PIP/MCP/CMC joints. For the whole-body radiographic OA phenotypes (n = 1,419), this comprised 12 of 32 phenotypes (those occurring in at least 3% of this sample): no OA, hand only, LS spine only, TF joint only, hip only, hand and LS spine, TF joint and LS spine, hip and LS spine, hand/TF joint/LS spine, hip/TF joint/LS spine, hand/LS spine/hip, and hand/TF joint/hip/LS spine. Race-by-sex interactions were determined to be significant at a P value of less than 0.1; thus, analyses stratified by race and sex were performed where there were significant interactions. The regression models without interactions included terms for race, sex, age, and BMI.

RESULTS

Two subsamples were evaluated in the analysis, as described above and as shown in Table 1 and Figure 1. Of the 2,121 subjects with multijoint radiographic data in the total sample, 16 had evidence of inflammatory arthritis and were excluded. In another 22, radiographic data for at least 1 hand joint were missing, leaving 2,083 subjects for the analysis of hand radiographic OA phenotypes. For the analysis of whole-body radiographic OA phenotypes, 606 individuals were missing PF joint readings (because of knee replacement or interpretation not available), 35 were women under the age of 50 years who did not undergo hip or spine radiography, and 23 were missing data on at least 1 joint site, leaving 1,419 subjects for the analysis (Figure 1). Selected characteristics of each sample are detailed in Table 1.

Table 1. Characteristics of the study sample with data for hand radiographic OA phenotypes and the subsample with data for whole-body radiographic OA phenotypes*
CharacteristicRadiographic OA phenotype
Hand sample (n = 2,083)Whole-body subsample (n = 1,419)
  • *

    TF = tibiofemoral; PF = patellofemoral; LS = lumbosacral.

  • Sample size for radiographic osteoarthritis (OA) of the hand was 2,081 for age and 2,079 for the body mass index (BMI).

  • Sample size for whole–body radiographic OA was 1,418 for the BMI and for the proximal interphalangeal (PIP), metacarpophalangeal (MCP), and carpometacarpal (CMC) joints.

  • §

    For radiographic OA of the hand, we used a composite definition requiring bilateral involvement, at least 1 distal interphalangeal (DIP) joint with a Kellgren/Lawrence grade of ≥2, and at least 3 joints (DIP, PIP, or CMC) involved (4).

Age, mean ± SD years65.1 ± 10.967.4 ± 9.7
BMI, mean ± SD kg/m231.3 ± 7.130.9 ± 6.3
No. (%) African American705 (33.9)456 (32.1)
No. (%) male689 (33.1)490 (34.5)
Radiographic feature, no. (%) of subjects  
 Any TF joint OA589 (41.5)
 Any PF joint with osteophyte grade ≥2172 (12.1)
 Any hip OA503 (35.5)
 Any LS spine OA877 (61.8)
 Any hand OA§600 (28.8)453 (31.9)
 Any DIP joint OA899 (43.2)671 (47.3)
 Any PIP joint OA593 (28.5)447 (31.5)
 Any MCP joint OA195 (9.4)150 (10.6)
 Any CMC joint OA576 (27.7)446 (31.5)

Comparable to many OA samples, the mean age of the study subjects was >65 years, and the mean BMI was in the obese category (∼31 kg/m2). About one-third of the sample was male and one-third African American. Overall, about 42% had radiographic OA of the TF joint, 12% the PF joint, 36% the hip, and 32% the hand. Radiographic OA of the LS spine was very common, occurring in 62% of the sample (Table 1).

Hand radiographic OA phenotypes (n = 2,083).

The frequency of each of the 16 mutually exclusive phenotypes by race and by sex in this population of 2,083 subjects is shown in Table 2. Many of the phenotypes were uncommon in the sample, occurring in fewer than 30 individuals (MCP only, DIP and MCP only, PIP and MCP only, PIP and CMC only, MCP and CMC only, DIP/PIP/MCP, DIP/MCP/CMC, PIP/MCP/CMC). The remainder of the phenotypes occurred in at least 40 individuals, and these were included in the adjusted analyses described below. The most common outcome in this community-based sample was no hand radiographic OA (45%; n = 929), followed by OA in the DIP joint only (13%; n = 269), the DIP/PIP/CMC joints (9%; n = 189), and the DIP and PIP joints only (8%; n = 166).

Table 2. Frequencies and adjusted ORs for the 16 mutually exclusive hand radiographic OA phenotypes in 2,083 study subjects, by race and sex*
Hand radiographic OA phenotypeComparison by raceComparison by sex
No. (%) Caucasian (n = 1,378)No. (%) AA (n = 705)P, by unadjusted Fisher's exact testAdjusted OR (95% CI) for AANo. (%) of women (n = 1,394)No. (%) of men (n = 689)P, by unadjusted Fisher's exact testAdjusted OR (95% CI) for men
  • *

    Adjusted odds ratios (ORs) were estimated only for phenotypes occurring in at least 40 persons. Models included dichotomous race (reference group Caucasian) and sex (reference group female), as well as continuous age (in years) and body mass index (in kg/m2). OA = osteoarthritis; AA = African American; DIP = distal interphalangeal; PIP = proximal interphalangeal; MCP = metacarpophalangeal; CMC = carpometacarpal.

  • Significant after adjustment for multiple comparisons using the Hochberg method.

  • Significant interaction by race and sex (see text for details).

  • §

    Significant adjusted result (95% confidence interval [95% CI] does not include 1).

No radiographic OA of the hand495 (36)434 (62)<0.001See text590 (42)339 (49)0.003See text
DIP joint only200 (15)69 (10)0.0020.66 (0.49–0.89)§187 (13)82 (12)0.3670.86 (0.65–1.13)
PIP joint only45 (3)31 (4)0.2171.35 (0.84–2.18)46 (3)30 (4)0.2631.39 (0.87–2.23)
MCP joint only10 (<1)8 (1)0.3305 (<1)13 (2)0.001
CMC joint only81 (6)41 (6)1.0000.99 (0.67–1.47)82 (6)40 (6)1.0001.01 (0.68–1.49)
DIP and PIP joints138 (10)28 (4)<0.0010.42 (0.27–0.64)§122 (9)44 (6)0.0710.69 (0.48–1.00)
DIP and MCP joints9 (<1)7 (1)0.43110 (1)6 (1)0.791
PIP and MCP joints2 (<1)2 (<1)0.608 3 (<1)1 (<1)1.000
DIP and CMC joints97 (7)19 (3)<0.0010.40 (0.24–0.66)§80 (6)36 (5)0.6850.89 (0.59–1.34)
PIP and CMC joints11 (<1)10 (1)0.24513 (1)8 (1)0.644
MCP and CMC joints6 (<1)3 (<1)1.0000 (0)9 (1)<0.001
DIP, PIP, and MCP joints19 (1)10 (1)1.00014 (1)15 (2)0.045
DIP, MCP, and CMC joints10 (<1)1 (<1)0.1116 (<1)5 (1)0.521
PIP, MCP, and CMC joints1 (<1)4 (<1)0.0484 (<1)1 (<1)1.000
DIP, PIP, and CMC joints167 (12)22 (3)<0.0010.24 (0.15–0.39)§151 (11)38 (6)<0.0010.44 (0.30–0.65)§
DIP, PIP, MCP, and CMC joints87 (6)16 (2)<0.0010.35 (0.20–0.62)§81 (6)22 (3)0.0100.51 (0.31–0.85)§

Significant differences in the unadjusted frequencies of hand radiographic OA phenotypes by race and sex are shown in Table 2, and for race in Figure 2. African Americans were more likely than Caucasians to have no radiographic OA in the hands, and were, in general, less likely to have any of the phenotypes that involved the DIPs. This difference was statistically significant (after adjusting for multiple comparisons) for DIP only, DIP and PIP, DIP and CMC, DIP/PIP/CMC, and DIP/PIP/MCP/CMC (P ≤ 0.002 for each comparison). Involvement of the PIP or CMC joints alone was similar among African Americans and Caucasians. Fewer significant differences were seen by sex, with men more likely to have no hand radiographic OA, slightly more likely to have OA of the MCP joint only or of the MCP and CMC joints (but these were very infrequent), and less likely to have OA of the DIP/PIP/CMC joints (P < 0.001) or involvement of all of the joints of the hand (P = 0.010).

Figure 2.

Distribution of mutually exclusive hand radiographic osteoarthritis (OA) phenotypes by race. The frequency of each joint combination is shown for African Americans (AAs) at the top and for Caucasians at the bottom, sorted by frequency among Caucasians, illustrating the overall shift in pattern for hand joint involvement among African Americans. HOA = hand OA; CMC = carpometacarpal; DIP = distal interphalangeal; MCP = metacarpophalangeal; PIP = proximal interphalangeal.

Adjusted results are shown in Table 2, with the exception of the no hand radiographic OA outcome, which demonstrated an interaction between race and sex (P for interaction = 0.007) and is summarized here. Compared to Caucasian women, Caucasian men had 76% higher odds of having no hand radiographic OA (adjusted OR [ORAdj] 1.76 [95% confidence interval (95% CI) 1.36–2.29]), while African American men and women had more than 2 times higher odds of having no hand radiographic OA (ORAdj 3.11 [95% CI 2.18–4.43] and ORAdj 3.36 [95% CI 2.57–4.41], respectively). The results after adjustment for age and BMI again showed that African Americans were less likely to have any phenotype, including the DIP joints (ORAdj 0.24–0.66) (Table 2). Also after adjustment, the odds of men having involvement of multiple hand joints (DIP/PIP/CMC joints or DIP/PIP/MCP/CMC joints) were reduced by ∼50% compared to women. As expected, increasing age was associated most strongly with outcomes, including multiple joints, and not with single-joint involvement except at the CMC; increasing BMI was significantly associated with higher odds of having radiographic OA of the DIP/PIP/CMC joints and the DIP/PIP/MCP/CMC joints (covariate ORAdj not shown).

Whole body radiographic OA phenotypes (n = 1,419).

The frequencies of the 32 mutually exclusive whole body radiographic OA phenotypes are shown in Table 3. As noted for the hand radiographic OA phenotypes, several of the whole-body phenotypes were very infrequent, occurring in ≤20 individuals (14 of 32 phenotypes) (see Table 3). Six of the combinations were seen in 20–40 individuals each (Table 3). The remaining 12 phenotypes (no OA, hand only, LS spine only, TF joint only, hip only, hand and LS spine, TF joint and LS spine, hip and LS spine, hand/TF joint/LS spine, hip/TF joint/LS spine, hand/LS spine/hip, and hand/TF joint/hip/LS spine) were seen in at least 40 individuals and are included in the adjusted results described below. Again, the most common outcome in the sample was no radiographic OA at any site (17%; n = 237), followed by the LS spine only (16%; n = 231) and the TF joint and LS spine (7%; n = 104).

Table 3. Frequencies and adjusted ORs for the 32 mutually exclusive whole-body radiographic OA phenotypes in 1,419 study subjects, by race and sex*
Whole-body radiographic OA phenotypeComparison by raceComparison by sex
No. (%) Caucasian (n = 963)No. (%) AA (n = 456)P, by Fisher's exact testAdjusted OR (95% CI) for AANo. (%) of women (n = 929)No. (%) of men (n = 490)P, by Fisher's exact testAdjusted OR (95% CI) for men
  • *

    Adjusted odds ratios (ORs) were estimated only for phenotypes occurring in at least 40 persons. Models included dichotomous race (reference group Caucasian) and sex (reference group female), as well as continuous age (in years) and body mass index (BMI; kg/m2). AA = African American; LS = lumbosacral; PF = patellofemoral; TF = tibiofemoral.

  • Significant interaction by race and sex (see text for details).

  • For radiographic osteoarthritis (OA) of the hand, we used a composite definition requiring bilateral involvement, at least 1 distal interphalangeal (DIP) joint with a Kellgren/Lawrence grade of ≥2, and at least 3 joints (DIP, proximal interphalangeal, or carpometacarpal joint) involved (4).

  • §

    Significant adjusted result (95% confidence interval [95% CI] does not include 1).

  • Significant after adjustment for multiple comparisons using the Hochberg method.

No radiographic OA at any site140 (15)97 (21)0.002See text145 (16)92 (19)0.135See text
Hand only37 (4)5 (1)0.0040.31 (0.12–0.79)§36 (4)6 (1)0.0050.30 (0.13–0.73)§
LS spine only160 (17)71 (16)0.6450.88 (0.65–1.21)137 (15)94 (19)0.0341.29 (0.96–1.72)
PF joint only2 (<1)1 (<1)1.0002 (<1)1 (<1)1.000
TF joint only27 (3)34 (7)<0.0012.51 (1.48–4.25)§40 (4)21 (4)1.0001.14 (0.65–1.99)
Hip only43 (4)28 (6)0.1931.41 (0.86–2.31)50 (5)21 (4)0.4420.73 (0.43–1.24)
TF joint and PF joint only6 (<1)3 (<1)1.0007 (1)2 (<1)0.727
Hand and TF joint only15 (2)5 (1)0.63216 (2)4 (1)0.236
Hand and PF joint only1 (<1)0 (0)1.0000 (0)1 (<1)0.345
Hip and TF joint only17 (2)12 (3)0.31617 (2)12 (2)0.435
Hip and PF joint only0 (0)1 (<1)0.3210 (0)1 (<1)0.345
Hand and hip only17 (2)3 (<1)0.14617 (2)3 (1)0.095
Hand and LS spine only55 (6)8 (2)<0.0010.31 (0.14–0.66)§46 (5)17 (3)0.2240.65 (0.37–1.15)
TF joint and LS spine only59 (6)45 (10)0.0161.77 (1.17–2.67)§59 (6)45 (9)0.0541.61 (1.07–2.43)§
PF joint and LS spine only2 (<1)2 (<1)0.5983 (<1)1 (<1)1.000
Hip and LS spine only54 (6)27 (6)0.8071.13 (0.70–1.82)50 (5)31 (6)0.4721.20 (0.75–1.91)
Hand, TF joint, and hip only23 (2)2 (<1)0.00814 (2)11 (2)0.396
Hand, PF joint, and hip only1 (<1)1 (<1)0.5402 (<1)0 (0)0.548
Hand, TF joint, and LS spine only63 (7)10 (2)<0.0010.32 (0.16–0.63)§50 (5)23 (5)0.6150.82 (0.49–1.37)
Hand, PF joint, and LS spine only1 (<1)0 (0)1.0000 (0)1 (<1)0.345
Hand, TF joint, and PF joint only4 (<1)3 (<1)0.6874 (<1)3 (1)0.698
PF joint, TF joint, and LS spine only15 (2)19 (4)0.00524 (3)10 (2)0.588
PF joint, TF joint, and hip only3 (<1)4 (<1)0.2217 (1)0 (0)0.103
Hip, TF joint, and LS spine only40 (4)24 (5)0.3411.32 (0.78–2.24)40 (4)24 (5)0.5941.23 (0.73–2.08)
Hand, LS spine, and hip only58 (6)5 (1)<0.0010.18 (0.07–0.46)§45 (5)18 (4)0.3450.69 (0.39–1.22)
PF joint, LS spine, and hip only4 (<1)2 (<1)1.0003 (<1)3 (1)0.422
Hand, TF joint, hip, and LS spine only50 (5)13 (3)0.0530.53 (0.28–1.01)43 (5)20 (4)0.6860.87 (0.50–1.53)
Hand, PF joint, hip, and LS spine only3 (<1)1 (<1)1.0003 (<1)1 (<1)1.000
Hand, PF joint, TF joint, and LS spine only23 (2)3 (<1)0.03220 (2)6 (1)0.298
PF joint, TF joint, hip, and LS spine only10 (1)14 (3)0.00813 (1)11 (2)0.280
PF joint, TF joint, hip, and hand5 (<1)2 (<1)1.0006 (1)1 (<1)0.433
Hand, TF joint, PF joint, hip, and LS spine25 (3)11 (2)1.00030 (3)6 (1)0.021

Differences in frequencies by race and sex are shown in Table 3. No sex differences were significant after adjustment for multiple comparisons. However, several of the differences by race were statistically significant. African Americans were more likely to have radiographic OA of the TF joint in isolation (P < 0.001). Caucasians were more likely to have phenotypes that included hand radiographic OA, with significant differences for the hand and LS spine, the hand/TF joint/LS spine, and the hand/hip/LS spine (P ≤ 0.001).

Adjusted results for those phenotypes seen in at least 40 participants are shown in Table 3 and Figure 3. As seen for the hand phenotypes, there was a significant interaction between race and sex for no radiographic OA at any site (P for interaction = 0.01). Compared to Caucasian women, African American women had nearly twice the odds of having no OA at any site (ORAdj 1.93 [95% CI 1.30–2.88]), while the odds for men were not significantly different (ORAdj 1.48 [95% CI 0.99–2.19] in Caucasian men versus ORAdj 1.19 [95% CI 0.70–2.00] in African American men). The results after adjustment again showed that African Americans were less likely to have phenotypes that included hand radiographic OA (ORAdj 0.18–0.53) (Table 3). African Americans were more likely to have phenotypes involving the TF joint, with 2.5 times the odds of having radiographic OA of the TF joint alone and 77% higher odds of radiographic OA of the TF joint and LS spine as compared to Caucasians. African Americans also had 30–40% higher odds of having only radiographic OA of the hip or the hip/TF joint/LS spine, although these differences were not statistically significant. Men had 70% reduced odds of having radiographic OA of the hand only as compared to women and 61% increased odds of radiographic OA of the TF joint and LS spine. Again, increasing age was associated with higher odds of multiple joint involvement, but not with single-joint involvement. Increasing BMI was inversely associated with radiographic OA of the hand only, the LS spine only, the hip only, and the hand/hip/LS spine, while the BMI was associated with higher odds of radiographic OA of the TF joint alone (covariate ORAdj not shown).

Figure 3.

Adjusted odds ratios (ORAdj) and 95% confidence intervals (95% CIs) for whole-body radiographic osteoarthritis (OA) phenotypes among African Americans as compared to Caucasians. Shown is the spectrum of phenotypes that are least likely in African Americans (left) to those that are most likely (right). Values are the point estimate of the ORAdj, with 95% CIs. Horizontal line indicates an ORAdj of 1 (no difference). Data are represented by a logarithmic scale because of the logarithmic nature of the ratio measures. LS = lumbosacral (spine); TFJ = tibiofemoral joint.

DISCUSSION

We have identified several differences in mutually exclusive multijoint radiographic OA phenotypes among African American and Caucasian men and women in this community-based sample. Considering first the hand radiographic OA phenotypes, African Americans had less frequent radiographic OA of the DIP joints in any combination, but similar involvement of the PIP and CMC joints and their combinations. For the whole-body radiographic OA phenotypes, significant differences by race included more frequent involvement of the TF joint and the combination of TF joint and LS spine radiographic OA in African Americans, while Caucasians had more frequent hand radiographic OA and its combinations. Interestingly, the racial differences we found in radiographic OA phenotypes were, in general, more significant than the sex differences.

The frequency of radiographic OA in this community-based population was quite high. Previous reports from the JoCo OA Project concerning OA prevalence in the knee and hip (18) used the baseline weighted sample; since the current study included the second followup (roughly 10 years later) of those individuals, it is reasonable to think that more of them would have developed radiographic OA, resulting in a higher frequency of radiographic OA in the current analysis. Older estimates of radiographic OA prevalence were lower (26–28) than those reported more recently (18, 19, 29), likely due to increased obesity and aging of the populations under study, in addition to methodologic differences.

Patterns of hand joint involvement have been assessed in other studies, primarily in Caucasians, and have shown that radiographic OA of the hand joints tends to group by row rather than by ray, and symmetric involvement is common (30–33). Egger et al (33), reporting on patterns of radiographic OA of the hand in the Chingford Study, found that although the strongest associations among hand joints were for other joints in the same group (e.g., DIP with DIP), there were also substantial associations between DIP and PIP involvement. In a family study selected on the basis of clinical hand OA, 3 radiographic OA combinations (DIP and PIP joints [29%], DIP, PIP, MCP, and CMC1 joints [29%], and DIP, PIP, and CMC1 joints [35%]) were found to be the most common (4). In our community study, which included individuals with and without evidence of OA, we also found these combinations to be among the most common (8%, 5%, and 9%, respectively), but we additionally identified a high frequency of individual joint involvement (particularly of the DIP [13%] and CMC1 [6%] joints).

In the only recent study to assess radiographic OA of the hand in African Americans, Sowers et al (17), found a similar frequency of DIP, PIP, and CMC1 joint involvement in African American and Caucasian women, but a greater frequency of MCP joint involvement in African Americans. Studies in populations of African descent, as compared to Caucasian populations, have found a reduced frequency of Heberden's nodes, a similar frequency of radiographic OA of the PIP, MCP, and CMC1 joints, and a higher prevalence of radiographic OA of the DIP joints (16, 34). We also found that radiographic OA of the PIP and CMC1 joints occurred at a similar frequency in African Americans and Caucasians, but in contrast to Sowers et al, we found that radiographic OA of the MCP joints was also similar between groups, and in contrast to the previous reports concerning populations of African descent (South African and Jamaican), we found that radiographic OA of the DIP joints was much less frequent among African Americans than among Caucasians.

There is significant variability among the joints assessed, the methods of defining OA, and the populations studied among the few studies assessing whole-body phenotypes in OA. Riyazi et al (8), in a sibling study of Dutch individuals recruited based on the presence of symptomatic OA, found that combinations of hand and spine, knee and spine, and hip and spine OA were most common among probands and siblings, all of whom had at least 2 joint sites involved. In common with those investigators, we found LS spine OA and combinations of joints including the LS spine to be common in our population. Several groups of investigators have found associations between 2 joint sites, such as the hand and TF joint (5), hand and hip (35), and TF joint and hip (7). While these phenotypes alone were uncommon in our analysis (2% or less), phenotypes including these pairs of joints along with other joint sites were fairly frequent, such that any combination of hand and TF joint was seen in 18%, hand and hip in 16%, and TF joint and hip in 18%.

Limitations of the current analysis include its cross-sectional nature, which does not allow determination of the timing of joint involvement, although this could be considered in future studies, since the JoCo OA Project is a longitudinal cohort with ongoing followup assessments. As our goal was to describe a comprehensive set of mutually exclusive radiographic OA phenotypes by sex and race, we included only a minimal set of key covariables in the models (age, BMI) and did not include other potentially important variables, such as symptoms, socioeconomic variables, occupation, or physical activity; these data are available and could be used in future analyses. As would be expected, some of the phenotypes had a small cell size, which precluded adjusted analyses for every possible outcome. PF joint films were read only for osteophytes and not for joint space narrowing, and interpretations for a portion of PF joint films were not available at the time of this analysis, thus limiting our sample size, although it is still quite sizeable. We did not have radiographs of the cervical spine or feet, so these sites could not be included, although they are commonly affected by radiographic OA.

This study also has substantial strengths, including the large overall sample size, the inclusion of African American and Caucasian men and women, and the standardized radiographs of multiple joints read by a single musculoskeletal radiologist (JBR) previously shown to have high reliability. Because of the large sample size, we were able to define discrete, mutually exclusive radiographic OA phenotypes for both hand radiographic OA and whole-body radiographic OA that had not previously been reported, as well as differences in these phenotypes by both race and sex.

In conclusion, we have shown in a large, community-based sample that multijoint radiographic OA involvement varies more significantly by race than by sex and that while African Americans were more likely than Caucasians to have radiographic OA of the TF joints and combinations that included the TF joints, they were less likely to have radiographic OA of the hand (particularly of the DIP joints) or combinations that included the hand. These findings remained significant after adjustment for sex, age, and BMI. Therefore, while “generalized OA” as often defined (hand radiographic OA or nodal changes with involvement of other joints) may be less frequent in African Americans, this group may have a higher burden of large joint involvement, particularly of the TF joints and LS spine. Such differences in radiographic patterns of OA, if confirmed in other populations and future studies, may affect the selection of participants for clinical research, particularly for studies of “generalized OA,” and are suggestive of a substantial clinical and public health burden of large-joint OA among African Americans.

AUTHOR CONTRIBUTIONS

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. Nelson 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. Nelson, Kraus, Helmick, Jordan.

Acquisition of data. Renner, Jordan.

Analysis and interpretation of data. Nelson, Renner, Schwartz, Jordan.

Acknowledgements

We would like to thank the staff and participants of the Johnston County Osteoarthritis Project, without whom this work would not have been possible. We would also like to thank Philip G. Conaghan, MBBS, PhD, and Yvonne Golightly, PT, PhD, for input on drafts of the manuscript.

Ancillary