The cellular components of blood are quantitatively and morphologically normal in uncomplicated primary OA. The platelet count may rise slightly as an acute-phase reactant during certain flares but remains within the normal range.

Although the erythrocyte sedimentation rate (ESR) is most often normal, modest elevations in the ESR may be observed transiently during clinical exacerbations of OA. More persistent elevations may be related to generalized polyarticular OA.¹ To a lesser extent, other acute-phase reactants² may be elevated transiently. There is a nonspecific modest increase in ESR with increasing age. Because the prevalence of OA also increases with age, any modest increase in ESR in OA should be addressed with caution. Hence, marked ESR elevations (above 50 Gmm/hr) should alert the physician to an unrelated, coexistentdisease.

Studies have linked C-reactive protein (CRP) elevations but not ESR with clinical severity of OA of the hip and knee.³ CRP was significantly associated with disability (as measured by the Health Assessment Questionnaire scale), joint tenderness, pain, fatigue, global severity, and depression. Median CRP was 5.9 μg/mL. ESR was associated with only functional disability. CRP levels were higher in those with erosive (mean 4.7 mg/l), in contrast with nonerosive (mean 2.1 mg/l; P = 0.001), interphalangeal OA.⁴ Elevated CRP was associated with generalized versus nongeneralized OA.⁵ Increased CRP has been correlated with hip pain.⁶

In a longitudinal population-based study of knee OA, levels of CRP were higher in 105 women with radiographic OA compared with 740 women without OA. In the 4-year follow-up, median levels of CRP were higher in the 31 women whose disease progressed at least one Kellgren-Lawrence grade (median, 2.6 μg/mL) than in the ones whose disease did not progress (median, 1.3 μg/mL).⁷

No increase in CRP or relation to activity was noted in 274 patients with knee OA, where high levels of soluble receptors of tumor necrosis factor (TNF)-1 were associated with lower physical function, increased knee symptoms, and higher radiographic scores.⁸

In conclusion, it appears that there is a modest increase in ESR with age that is not related to OA. Sensitive techniques of determining CRP most often demonstrate a modest increase in OA. No studies of CRP elevation in OA have corrected for the presence of cardiovascular disease as another reason for an increase in CRP. The significance in the CRP for diagnostic purposes or in monitoring activity of OA is yet to be determined.

Sensitive and specific assays for cartilage proteoglycan components and degradation products¹⁷ have been developed. Although of interest, serum sampling appears of limited value because of several metabolic factors, including dilution in serum from a single joint dysfunction and unknown influences of renal or liver function.^18,19

Keratan sulfate in humans is a distinct and unique sugar derived primarily (95%) from articular and intervertebral disk cartilage. Elevations of serum keratan sulfate levels have been found in patients with OA, but there is a considerable variation in values in both cross-sectional and longitudinal studies.

Plasma levels of hyaluronate in OA were twice that in rheumatoid arthritis (RA) and seven times that in control subjects.²⁰ Elevated hyaluronate was found to correlate with the patients’ functional capacity. Another study of 94 patients with tibiofemoral OA showed that serum hyaluronate values at entry correlated with disease duration, minimum joint space, and previous surgery.²¹ In this retrospective review, radiographic progression or knee surgery correlated with higher baseline hyaluronate levels.
Levels of serum hyaluronate failed to show a correlation with the Lequesne algofunctional index, duration of symptoms, CRP, or severity of radiographic changes. In a cross-sectional study, serum hyaluronan levels correlated with radiographic OA, ethnicity, sex, and age even after adjustment for multiple variables (P <0.0045).²² In a 2-year follow-up, those with knee OA and higher hyaluronan levels at baseline had faster radiographic progression (P <0.005).²³ Serum hyaluronan was followed in a 3-year study of knee OA.²³ Baseline hyaluronan levels did not correlate with progression. However, reduction in hyaluronan found at the first year correlated with radiographic progression at 3 years (r = 0.27, P = 0.02) (similar findings were present for an increase in serum osteocalcin).

In knee OA, elevation of cartilage oligomeric matrix protein (COMP) related to progression.²⁴ COMP increases may directly relate to physical exercise, even walking.^25,26 Pain of hip OA has been correlated with increased COMP.²⁷ Of 81 patients observed for 5 years, progression was defined as those having a decrease of 2 mm or more in joint space on radiographic examination or those requiring knee surgery during the 5-year follow-up.²⁸ Serum COMP levels increased by a mean of 6.42 μg/mL in patients whose OA progressed compared with a mean of 0.07 μg/mL in those whose OA did not progress. In another study of 48 with hip OA and observed prospectively with radiographs and serum samples, levels of COMP correlated with rapidly progressing OA of the hip.²⁹ In the same study, serum levels of bone sialoprotein (BSP) correlated inversely with osteophyte grade and sclerosis grade in OA of the hip.

COMP related to progression of knee OA in a 5-year follow-up and patterns of progression suggested that knee OA progression is episodic or phasic.³⁰ The authors suggested that large individual variation precludes the use of COMP for predicting progression, but that sequential COMP measurements may help identify OA progression.

Stromelysin (matrix metalloproteinase 3 [MMP-3]) has been found to be elevated in the serum of patients with OA and correlated strongly with the articular index.³¹ Levels of collagenase (MMP-1) or of tissue inhibitor of matrix metalloproteinases (TIMP-1) were within normal limits. In a study of 36 patients, MMP-3 and MMP-9 were significantly increased in patients with rapidly destructive hip OA versus OA in patients awaiting total hip prosthesis.³²

Serum sex hormones have been measured in patients with OA. An association has not been found between endogenous estrogen levels and OA or its severity.³³

Nonspecific elevation of plasma substance P levels was seen in patients with OA compared with normal individuals, but it did not occur to the degree of elevation in patients with reactive arthritis.³⁴

When there is a reduction of serum iron levels in patients with OA, it is secondary to another illness, such as gastrointestinal blood loss (increased iron-binding capacity) or “chronic disease” (reduced iron-binding capacity). Increased serum iron concentration in OA may be secondary to another illness (e.g., hemochromatosis).

Serum copper and ceruloplasmin levels are normal in patients with primary OA. Increased serum copper concentrations with secondary OA of the large joints may be the result of Wilson disease (Table 11-1). Wilson disease of small joints may be related to chondrocalcinosis.

Studies of the immune system to date have failed to identify aberrant cellular or humoral immunity in the pathogenesis of OA.

The volume of synovial fluid in the knee may vary from normal (0.5 to 1.5 mL)⁵⁰ to greater than 100 mL in OA. In general, smaller increases in synovial fluid may be evident in other joints with OA. Indeed, the mechanisms that control the volume of synovial fluid are poorly understood and can be explained only partially by vascular hydrostatic pressure. Inexplicably, minimal radiographic evidence of OA may be associated with large synovial effusions; conversely, severe radiographic evidence of OA may elicit only minimal or no synovial effusion.

Intra-articular volumes of synovial fluid as determined by radiolabeling were high (109± 35 mL) and were higher than the volume that could be aspirated (59 ± 28 mL).⁵⁰ Mean clearance rate for the synovial fluid was 0.039 ± 0.030 mL (SD) per minute, somewhat slower than fluids from patients with RA.

The normal synovial fluid usually appears pale yellow, dark yellow, or clear. Along with an increased viscosity, it is reminiscent of the derivation of the term synovial (syn, like; ovum, egg “white”). It may infrequently be blood tinged or frankly bloody. Joint bleeding most often occurs in affected glenohumeral and unstable knee joints and is often associated with an acutely painful exacerbation, trivial trauma, or increased activity. Hemarthrosis may reflect “pinching” of synovium between contiguous osteophytes or irregular joint surfaces or, less frequently, a microfracture of subchondral or osteophytic bone or a tear of the rotator cuff (shoulder) or anterior cruciate ligament (knee). In these cases, the bloody fluid is evident throughout the arthrocentesis procedure, and the fluid often fails to clot. In contrast, bloody fluid resulting from traumatic aspiration technique clears during the course of withdrawal, or, alternatively, blood is seen to enter the syringe and mix with initially yellow fluid during joint aspiration. Repeated aspiration of bloody fluid from a single joint should suggest pigmented villonodular synovitis, particularly if the synovial effusion has a “port wine” color.

In primary OA, shed cartilage “shard” fragments are visible as floating white specks and particles. In OA associated with ochronosis, pigmented shards of cartilage may assume the appearance of ground pepper in joint fluid.⁵¹

Synovial fluid in OA is clear and occasionally faintly turbid.

Synovial fluid viscosity is dependent on a protein-hyaluronic acid complex. The hyaluronate complex consists of an unbranched glycosaminoglycan macromolecule of approximately 2000 kDa composed in turn of polymerized disaccharide dimers of glucuronic acid-glucosamine coiled into a spherical or ellipsoid
conformation. This conformation allows the structure to occupy a solvent domain considerably larger than the volume of the polymer chain. Hyaluronate depolymerization or synovial membrane secretion of a poorly polymerized hyaluronate or a hyaluronate complex with altered conformation results in diminished viscosity.

In OA, viscosity is inversely related to clinical evidence of inflammation; fluid from palpably “cool” joints usually has a normal viscosity and produces a “string” sign (Fig. 11-1). Markedly poor viscosity, in which the fluid drops like water from the syringe, is uncommon and may reflect coexistent pseudogout or another inflammatory arthritis. Conversely, extremely thick, viscous fluid should suggest osteochondromatosis or hypothyroidism.⁵² Pseudomucinous synovial cysts, found over the dorsum of osteoarthritic distal interphalangeal joints, contain pale, gelatinous fluid similar to that observed in ganglia.

The precipitation of the protein salt of hyaluronic acid after acidification of joint fluid is the basis of the mucin clot or “Ropes” test.⁴⁹ An aliquot of synovial fluid is added to a beaker containing a four times greater volume of 2% acetic acid and mixed with a glass rod. The resulting mucin clot (hyaluronate protein) reflects the degree of polymerization of hyaluronic acid. In OA, a tight, ropy mass is formed (graded good); whereas in RA and other inflammatory arthritides, the mass shows friable edges (graded poor). The mucin clot is almost invariably good in OA, even when viscosity is significantly diminished.

If it is uncertain whether synovial fluid has been aspirated, mucin clot formation and metachromatic staining are capable of detecting as little as 0.5 μL of synovial fluid.⁵³