The controversial relationship between osteoarthritis and osteoporosis ...

International Journal of Rheumatic Diseases 2015

ORIGINAL ARTICLE

The controversial relationship between osteoarthritis and osteoporosis: an update on hand subtypes Roberta RAMONDA,1 Leonardo SARTORI,2 Augusta ORTOLAN,1 Paola FRALLONARDO,1 Mariagrazia LORENZIN,1 Leonardo PUNZI1 and Estella MUSACCHIO2 1

Rheumatology Unit, Department of Medicine DIMED, University of Padova, and 2Clinica Medica I, Department of Medicine DIMED, University of Padova, Padova, Italy

Abstract Aim: To compare hand osteoarthritis (HOA) subtypes and to examine possible links with local bone mineral density (BMD). Method: Fifty-five patients with erosive hand osteoarthritis (EHOA) and 21 patients with nodal hand osteoarthritis (NOA) fulfilling American College of Rheumatology criteria for HOA were evaluated. Subjects showing at least two erosions of the interphalangeal joints were assigned to the EHOA group; the others were considered NOA. Disease duration, number of active joints and radiological scores were assessed. All patients and 174 controls underwent phalangeal radiographic absorptiometry (pRA) of the middle phalanges of the non-dominant hand to assess BMD, T- and Z-scores. Results: BMD was higher in EHOA with respect to NOA and controls (P = 0.05); T- and Z-scores were significantly higher in EHOA (P = 0.01 and P < 0.01). Values suggestive of osteopenia were found in 32% of EHOA and 22% of NOA patients, and in 44% of controls (P = 0.05 EHOA vs. NOA); a T-score < 2.5 standard deviations was present in 15% of EHOA and 28% of NOA patients, and in 21% of controls (P = 0.05 EHOA vs. NOA). Conclusion: Phalangeal BMD was higher in EHOA compared to NOA and controls. This characteristic could potentially be exploited to differentiate the two hand OA subtypes. Key words: bone mineral density, erosive hand osteoarthritis, osteoarthritis, osteoporosis, phalangeal radioabsorptiometry.

INTRODUCTION Osteoarthritis (OA) and osteoporosis (OP) are two important disorders affecting mainly post-menopausal women, but their associations are not entirely understood.1 OA has a chronic evolution characterized by cartilage degeneration of diarthrodial joints. Mostly involving the hand followed by the knee, foot, spine

Correspondence: Dr Roberta Ramonda, Rheumatology Unit, Department of Medicine DIMED, University of Padova, Via Giustiniani 2, 35128 Padova, Italy. Email: [email protected]

and shoulder joints, OA leads to degradation of articular cartilage often accompanied by subchondral bone changes and osteophyte formation at the joint surfaces and margins leading to joint failure. All of these structures may be involved at any of the disease’s stages, contributing to its development and evolution.2 Age, sex, race, genetics, mechanical stress, obesity, metabolic diseases and inflammation are modifiable and non-modifiable factors that can also influence OA onset. Several hand OA subsets, such as erosive hand OA (EHOA), nodal interphalangeal (NOA) and thumb base OA have been distinguished. EHOA, defined by radiographic erosive lesions, abrupt onset, high clinical burden, an aggressive clinical

© 2015 Asia Pacific League of Associations for Rheumatology and Wiley Publishing Asia Pty Ltd

R. Ramonda et al.

course and poor outcome,3 is a relatively uncommon disease that afflicts more often middle-aged women. It is characteristically a degenerative disorder affecting the interphalangeal (IP), proximal (P) and distal (D) joints, showing more severe inflammatory signs. NOA, instead, is a classic, less severe, non-erosive and more common form of hand OA with respect to EHOA. Clinical and radiographic characteristics of the two subtypes were described in a previous paper.4 As bone architecture primarily reflects the integration of bone density and quality, OP is defined as a metabolic bone disorder characterized by compromised architecture predisposing to an increased risk of fracture. Low bone mineral density (BMD) is linked to many factors, such as genetics, lifestyle, drug consumption and menopausal status.5,6 The BMD in subchondral bone may be altered in patients with OA due to the stress experienced by articular cartilage during loading and physical activity. OP is a frequent finding in patients with disabling chronic rheumatic diseases primarily involving the joints, such as rheumatoid arthritis (RA) and ankylosing spondylitis, and its pathogenesis appears largely related to the same mechanisms that sustain the underlying disease.7 There are, conversely, no systemic manifestations in OA and pathological changes are limited to the joints. The object of this study was to evaluate BMD levels in EHOA and NOA patients, as well as in healthy subjects and to briefly review the studies on this topic outlined in the literature and summarized in Table 1.7–11

PATIENTS AND METHODS BMD levels in EHOA and NOA subjects were evaluated in consecutive outpatients attending the Hand Clinic of the Rheumatology Unit, Department of Medicine (DIMED), University of Padova, (Italy) Medical Center between 2010 and 2013. A total of 76 patients fulfilling the American College of Rheumatology criteria for hand OA were enrolled in the study. Patients reporting trauma or who had a history of traumatic joint injuries or of other arthropathies including RA, psoriatic arthritis, gout, or chondrocalcinosis, were excluded. Subjects who had consumed steroids or other medications interfering with bone metabolism (bisphosphonates) in the 6 months preceding enrolment were also excluded. All of the patients underwent clinical and X-ray assessment, as well as phalangeal radiographic absorptiometry (pRA) evaluations. Standard hand X-rays were assessed in accordance with Altman by specifically trained, experienced readers.12 One hundred and seventy-four (156 female) individuals without clinical and radiological signs of OA were considered as controls and consequently assessed. Patients showing X-ray central erosions in IP joints were included in the EHOA group (n = 55, 48 female), patients with no erosions were classified as NOA (n = 21, all female); those with radiographic erosions in the metacarpophalangeal (MCP) joints were excluded. Anthropometric measurements were taken by trained physicians using standard methods.

Table 1 Studies that examined the association between osteoporosis (OP) and osteoarthitis (OA) of the hand Author

Pub. year

No. of study patients

Bone and joint focus

Key points

References

El-Sherif HE

2008

40 F

2006

55 F

EHOA points are at risk to develop hand OP EHOA points are at risk to develop OP

7

Zoli A

Hochberg MC

2004

436 (Baltimore study) (M: 298; F: 139)

Higher bone mass develops radiological knee OA

9

Yahata Y

2002

567 F

Relationship between hand RS-KL and hand BMD (DEXA) Relationship between hand RS-KL and hip and lumbar spine BMD X-ray of hands and knees, and DEXA of lumbar spine and femur Appendicular BMD and RS-KL of knees and hands

Pogrund H

1986

NA

To evaluate hand OA and OP

Appendicular BMD increases with hand OA, in particular with osteophytes Indifferently negative and positive correlations

8

10

11

BMD, bone mineral density; DEXA, dual emission X-ray absorptiometry; EHOA, erosive hand osteoarthritis; NA, not available; RS-KL, Kellgren and Lawrence radiographic score.

2


Hand osteaoarthritis and osteoporosis

Body weight was measured using a calibrated physician’s scale to the nearest 0.1 kg. Height was measured using a wall-mounted stadiometer to the nearest 0.5 cm. BMI was calculated as body weight in kilograms divided by the square of height in meters. All the patients were evaluated for disease duration (self-reported and confirmed by medical records), the number of affected active (swollen and painful or tender) joints (NAAJ), radiographic score according to Kellgren and Lawrence (RS-KL).13 BMD and quality were assessed by pRA scanning using a 60 kV X-ray system (Metriscan, Alara Metric, Hayward, CA, USA, short-term precision 1.1%, Fig. 1).14 BMD was calculated for the three middle phalanges of the second, third and fourth fingers of the nondominant hand, excluding the articular surface, and the average was expressed as bone mineral content per unit of projected bone area (g/cm2). The instrument provides T- and Z-scores, that is, the number of standard deviations (SD) compared to young adults and to an age- and sex-matched populations, respectively. However, these scores are not equal to those provided by the dual emission X-ray absorptiometry (DEXA) at spine and hip level, but have the same meaning and are automatically calculated by the machine using an internal database and were used here to describe bone density and to compare

individuals. Patients and controls were exposed to less than 0.012 mSv at each examination. With respect to a traditional DEXA, the pRA instrument is portable and the test is extremely rapid (exposure is for 300 msec). Figure 2 shows the output of the pRA test with a phantom comparison. The study protocol was approved by the local Ethics Committee and all the patients and controls gave written informed consent to participate in the study.

Statistical analysis Quantitative variables were summarized as means SD and qualitative ones as frequency distributions. Analysis of variance or the t-test was used to compare mean values between the two groups (EHOA vs. NOA) for normally distributed variables, and the non-parametric Mann–Whitney test was used for nonnormal variables. Frequency comparisons were analyzed using the v2 test (chi-square) or Fisher’s exact test in the event of non-parametric variables. A P-level < 0.05 was considered significant. The results of the EHOA and NOA subjects were compared and analyzed, and the control group represented the average condition of an age-matched population in particular with regard to BMD. Statistical analyses were performed using a SPSS program (version 14, SPSS Inc., Chicago, IL, USA).

RESULTS

Figure 1 MetriScan Rx digital radiographic absorptiometry (RA).


The patients’ characteristics are outlined in Table 2. The two patient groups and the controls were found comparable for age and BMI. With respect to the NOA group, disease duration was longer (P < 0.01) and NAAJ was higher in the EHOA subjects (P < 0.01). The RS-KL score was worse in the EHOA patients (P = 0.0001). Needless to say, erosions were present only in the EHOA cohort (5.4 3.0). With regard to the female population (Table 3a), the BMD was higher in the EHOA with respect to the NOA patients and controls (P = 0.05). Accordingly, both the T- and Z-scores were worse in the controls, but again higher in EHOA compared to NOA (P = 0.01 and P < 0.01, respectively). Results overlapped when the males were included in the analysis (Table 3b). The prevalence of subjects presenting with a T-score between 2.5 and 1 SD was higher in the EHOA patients, while the prevalence of T–score < 2.5 was higher in the NOA patients (both P = 0.05). The same pattern was observed in the total study population (P = 0.025 and P = 0.03, respectively).

3

R. Ramonda et al.

Figure 2 The output of the radio absorptiometry test with the phantom comparison.

Table 2 The main characteristics of the study population Variable Age (years) Males (%) BMI (kg/m2) Disease duration (years) NAAJ Erosions RS-KL

EHOA n = 55 62.7 13 24.1 10.5 9.4 5.4 73.2

6.4

3.9 6.7 4.6 3.0 25.7

NOA n = 21 62.7 0 24.6 7.7 7.1 0 42.4

8.3 4.0 5.4 4.5 25.8

Controls n = 174

P EHOA vs. NOA

62.2 7.4 18 24.7 3.9 – – – –

ns – ns < 0.01 < 0.01 – 0.0001

BMI, body mass index; NAAJ, number of affected active (swollen and painful or tender) joints; RS-KL, radiological score Kellgren–Lawrence.

Table 3 (a) Main bone mass parameters in only the female population (EHOA, NOA and controls). (b) The same parameters in the entire population (a) Variable BMD (g/cm2) T-score Z-score T score< 2.5 (%) 2.5 < T-score < 1 (%) (b) Variable BMD (g/cm2) T-score Z-score T-score < 2.5 (%) 2.5 < T-score < 1 (%)

EHOA n = 48

NOA n = 21

Controls n = 156

53.8 6.5 0.74 1.5 0.27 1.10 15 32

50.5 5.8 1.54 1.4 0.13 0.99 28 22

50.7 5.3 1.49 1.3† 0.02 0.83† 21 44‡

EHOA n = 55

NOA n = 21

Controls n = 174

57.8 20.9 0.37 1.4 1.01 3.6 20 35

50.5 5.8 1.54 1.4 0.13 0.99 28 22

51.0 5.5 1.41 1.3 0.02 0.80 20 40

P EHOA vs. NOA 0.05 0.01 < 0.01 0.05 0.05 P EHOA vs. NOA 0.025 0.01 < 0.01 0.025 0.03

†P = 0.05 vs. EHOA; ‡P = 0.05 vs. EHOA and vs. NOA. BMD, bone mineral density; EHOA, erosive hand osteoarthritis; NOA, nodal hand osteoarthritis.

4



DISCUSSION EHOA is more aggressive, according to several studies, with respect to NOA.15 Radiographic features make it possible to differentiate the two forms on the basis of the presence/absence of subchondral central erosions, cortical destruction and subsequent reparative responses, leading to ankylosis in the former.16,17 While the relationship between OA and OP has been largely debated, no consensus has as yet been reached.1 In the present study a higher BMD was found in EHOA with respect to that in NOA and in controls. This finding, which could be explained by increased vascularization and consequent osteo-production associated with florid local inflammation, was quite unexpected considering the inflammatory nature of EHOA and its longer disease duration. Local inflammation together with subchondral bone erosions, thus do not appear to determine sufficient bone loss to cause impairment in EHOA with respect to NOA patients. El Sherif7 found instead lower local BMD associated with HOA. However, he did not specifically address the discrimination between EHOA and NOA and analyzed a population with a much higher BMI (34 vs. 24 kg/m2). Moreover, in his study a different instrument was employed to assess BMD. Consequently, the analysis of a different region of interest (ROI) may also account for the discrepancy. Evidence for a direct correlation between cartilage damage and bone density has been reported in the first metatarsal.18 In accordance with studies showing an increased turnover of markers of the cartilage matrix in OA,19,20 our previous data on HOA confirmed that an inflammatory process is linked to cartilage matrix loss which inevitably alters joint biomechanics, leading to the collapse and thickening of the underlying subchondral bone trabeculae.16 Although OA at all sites (hand, hip, knee, spine . . .) is well defined, OP is usually assessed only at the femur and spine, even though a recent study seems to suggest a good correlation between BMD values obtained at the femur and spine with those obtained at the hands, at least with the DEXA method.21 The correlation between the two diseases has for the most part been studied using cross-sectional data accumulated over the last 40 years and since the time Foss and Byers first reported the relationship between OA and OP and noted that the former is rarely found in patients with femoral neck fractures.22 Most observations made since then have referred to elderly populations being assessed for spine or femoral OP.7–11,23 The existence of genetic factors in


both OA and OP has been recognized by family studies.24 Peak bone mass in the hip has been found to be increased in daughters of women with OA,25 and hip fracture risk has been found to be reduced in the mothers of patients with hip OA.26 Twins with hip osteophytes were found to have a higher femoral neck BMD than their unaffected co-twin.27 Some studies have subsequently found significant associations between OA and polymorphic markers for collagen genes, vitamin D, and estrogen receptor-a, transforming growth factorb1 and interleukins and so on, as well as between OP and DNA polymorphisms in the same genes.28–30 Other investigations have, conversely, found increased bone mass with radiographic alterations in OA not linked to a reduced risk of fractures.31,32 The differences in results outlined in the literature confirm the view that the relationship between OP and OA is hard to pin down and may vary depending on the disease site and stage.33,34 Limitations of the present study are linked to its observational design and the inclusion of a relatively small number of patients who were assessed only with regard to their phalangeal BMD. This, of course, can be explained by the restrictive inclusion criteria that were utilized and the study’s focus, which was the hand, making phalangeal BMD indispensable. In conclusion, few reports are available in the literature concerning BMD levels in patients affected with HOA, and the data regarding hand studies continue to be vigorously debated. Our study found a better bone status with higher T-scores in EHOA patients with respect to NOA ones. As the articular surface that is directly affected by erosions, osteophytes and sclerosis was not included in the ROI analyzed by pRA, the discrepancy highlighted in our study may refer to a structural diversity in the repaired tissue. This should not be interpreted as a more or less severe form of OP, but rather as a difference in bone architecture that requires further evaluation by appropriate instruments which will be able to elucidate the elusive relationships between peripheral OP and OA.

ACKNOWLEDGMENT The authors would like to express their appreciation to Linda Inverso Moretti (native English teacher) for her help in preparing the English version of this manuscript.

CONFLICT OF INTEREST None.

5

R. Ramonda et al.

FUNDING INFORMATION This study was not supported by any specific grant.

CONTRIBUTIONS RR: study design, data analysis, manuscript preparation. LS: study design, data collection and statistical analysis. AO, PF, ML: data collection and analysis. LP: study design, critical revision. EM: study design, data analysis, manuscript preparation.

REFERENCES 1 Im GI, Kim MK (2014) The relationship between osteoarthritis and osteoporosis. J Bone Miner Metab 32 (2), 101–9. 2 Ramonda R, Frallonardo P, Musacchio E, Vio S, Punzi L (2014) Joint and bone assessment in hand osteoarthritis. Clin Rheumatol 33, 11–9. 3 Belhorn LR (1993) Erosive osteoarthritis. Semin Arthritis Rheum 5, 298–306. 4 Punzi L, Ramonda R, Sfriso P (2004) Erosive osteoarthritis. Best Pract Res Clin Rheumatol 5, 739–58. 5 Garnero P, Piperno M, Gineyts E, Christgau S, Delmas PD, Vignon E (2001) Cross sectional evaluation of biochemical markers of bone, cartilage and synovial tissue metabolism in patients with knee osteoarthritis: relation with disease activity and joint damage. Ann Rheum Dis 60, 619–26. 6 Sinigaglia L, Varenna M, Girasole G, Bianchi G (2006) Epidemiology of osteoporosis in rheumatic diseases. Rheum Dis Clin North Am 32, 631–58. 7 El-Sherif HE, Kamal R, Moawyah O (2008) Hand osteoarthritis and bone mineral density in postmenopausal women; clinical relevance to hand function, pain and disability. Osteoarthritis Cartilage 16, 12–7. 8 Zoli A, Lizzio MM, Capuano A, Massafra U, Barini A, Ferraccioli G (2006) Osteoporosis and bone metabolism in postmenopausal women with osteoarthritis of the hand. Menopause 13, 462–6. 9 Hochberg MC, Lethbridge-Cejku M, Tobin JD (2004) Bone mineral density and osteoarthritis: data from the Baltimore Longitudinal Study of Aging. Osteoarthritis Cartilage 12, S45–8. 10 Yahata Y, Aoyagi K, Yoshida S et al. (2002) Appendicular bone mass and knee and hand osteoarthritis in Japanese women: a cross-sectional study. BMC Musculoskelet Disord 3, 24. 11 Pogrund H, Rutenberg M, Makin M, Robin GC, Steinberg R, Bloom R (1986) Osteoarthritis of the hand and osteoporosis. Clin Orthop Relat Res 203, 239–43. 12 Altman RD, Hochberg M, Murphy WA Jr, Wolfe F, Lequesne M (1995) Atlas of individual radiographic features in osteoarthritis. Osteoarthritis Cartilage 3 (Suppl A), 3–70.

6

13 Kellgren JH, Lawrence JS (1957) Radiological assessment of osteo-arthrosis. Ann Rheum Dis 16, 494–502. 14 Thorpe JA, Steel SA (2008) The Alara Metriscan phalangeal densitometer: evaluation and triage thresholds. Br J Radiol 81, 778–83. 15 Marshall M, Nicholls E, Kwok WY et al. (2015) Erosive osteoarthritis: a more severe form of radiographic hand osteoarthritis rather than a distinct entity? Ann Rheum Dis 74, 136–41. 16 Punzi L, Ramonda R, Deberg M et al. (2012) Coll2–1, Coll2–1NO2 and myeloperoxidase serum levels in erosive and non-erosive osteoarthritis of the hands. Osteoarthritis Cartilage 20, 557–61. 17 Kellgren JH, Moore R (1952) Generalised osteoarthritis and Heberden’s nodes. Br Med J 1, 181–7. 18 Bareither D, Manion BL, Sumner DR et al. (1998) Relationship between articular cartilage damage and bone density in the first metatarsal. J Foot Ankle Surg 37, 401–9. 19 Okada Y (2000) Matrix-degrading metalloproteinases and their roles in joint destruction. Mod Rheumatol 10, 121–8. 20 Henrotin Y, Deberg M, Dubuc JE, Quettier E, Christgau S, Reginster JY (2004) Type II collagen peptides for measuring cartilage degradation. Biorheology 41, 543–7. 21 Alioglu K, Dogu B, Sirzai H, Yilmaz F, Kuran B (2013) Hand bone mineral density reference values in a Turkish healthy female population. Int J Rheum Dis. doi: 10.1111/ 1756–185X.12159 [Epub ahead of print]. 22 Foss MV, Byers PD (1972) Bone density, osteoarthrosis of the hip, and fracture of the upper end of the femur. Ann Rheum Dis 31, 259–64. 23 Unger K, Rahimi F, Bareither D, Muehleman C (2000) The relationship between articular cartilage degeneration and bone changes of the first metatarsophalangeal joint. J Foot Ankle Surg 39, 24–33. 24 Ramonda R, Musacchio E, Campana C et al. (2011) Immunogenetic aspects of erosive osteoarthritis of the hand in patients from northern Italy. Scand J Rheumatol 40, 139–44. 25 Naganathan V, Zochling J, March L, Sambrook PN (2002) Peak bone mass is increased in the hip in daughters of women with osteoarthritis. Bone 30, 287–92. 26 Astrom J, Beertema J (1992) Reduced risk of hip fracture in the mothers of patients with osteoarthritis of the hip. J Bone Joint Surg Br 74, 270–1. 27 Antoniades L, MacGregor AJ, Matson M, Spector TD (2000) A cotwin control study of the relationship between hip osteoarthritis and bone mineral density. Arthritis Rheum 43, 1450–5. 28 Rodriguez-Fontenla C, Calaza M, Evangelou E et al. (2014) Assessment of osteoarthritis candidate genes in a meta-analysis of nine genome-wide association studies. Arthritis Rheumatol 66, 940–9. 29 Aerssens J, Dequeker J, Peeters J, Breemans S, Broos P, Boonen S (2000) Polymorphisms of the VDR, ER and



COLIA1 genes and osteoporotic hip fracture in elderly postmenopausal women. Osteoporos Int 11, 583– 91. 30 Blumenfeld O, Williams FM, Valdes A et al. (2014) Association of interleukin-6 gene polymorphisms with hand osteoarthritis and hand osteoporosis. Cytokine 69, 94–101. 31 Dequeker J, Aerssens J, Luyten FP (2003) Osteoarthritis and osteoporosis: clinical and research evidence of inverse relationship. Aging Clin Exp Res 15, 426–39.


32 Arden NK, Nevitt MC, Lane NE et al. (1999) Osteoarthritis and risk of falls, rates of bone loss, and osteoporotic fractures. Study of Osteoporotic Fractures Research Group. Arthritis Rheum 42, 1378–85. 33 Cooper C, Snow S, McAlindon TE et al. (2000) Risk factors for the incidence and progression of radiographic knee osteoarthritis. Arthritis Rheum 43, 995–1000. 34 Felson DT, Nevitt MC (2004) Epidemiologic studies for osteoarthritis: new versus conventional study design approaches. Rheum Dis Clin North Am 30, 783–97.

7