Elevated erythrocyte sedimentation rate and high-sensitivity C-reactive protein in osteoarthritis of the knee: relationship with clinical findings and radiographic severity

Abstract

Purpose

We assessed erythrocyte sedimentation rate and high-sensitivity C-reactive protein concentration in knee osteoarthritis and non-knee osteoarthritis. In addition, we investigated potential relationship between the levels of erythrocyte sedimentation rate and high-sensitivity C-reactive protein with clinical findings and radiographic severity.

Methods

We compared erythrocyte sedimentation rate and high-sensitivity C-reactive protein concentration between 104 patients with knee osteoarthritis (knee osteoarthritis group; 25 males, 79 females; mean age, 73 y) and 50 patients without knee osteoarthritis (non-knee osteoarthritis group; 16 males, 34 females; mean age, 64 y) excluding any patients with comorbid joint osteoarthritis, rheumatoid arthritis, malignant tumours or inflammatory diseases. In the knee osteoarthritis group, we assessed whether erythrocyte sedimentation rate and high-sensitivity C-reactive protein concentration differed in clinical features and Kellgren-Lawrence (KL) grades.

Results

Erythrocyte sedimentation rate and high-sensitivity C-reactive protein were significantly higher in the knee osteoarthritis group than in the non-knee osteoarthritis group (P = 0.0013 and 0.00010, respectively). In the knee osteoarthritis group, erythrocyte sedimentation rate was significantly elevated in patients with tenderness and patellar ballottement (P = 0.032 and 0.038, respectively), and high-sensitivity C-reactive protein concentration was significantly elevated in patients with tenderness, swelling and patellar ballottement (P = 0.0042, 0.00030 and 0.019, respectively). Erythrocyte sedimentation rate in KL-I was lower than erythrocyte sedimentation rate in KL-III and -IV (P = 0.012 and 0.037, respectively). Erythrocyte sedimentation rate in KL-II did not significantly differ from erythrocyte sedimentation rate in the other groups. High-sensitivity C-reactive protein concentration was lower in grade I than in KL-II, -III and -IV (P = 0.044, 0.0085 and 0.049, respectively).

Conclusions

Erythrocyte sedimentation rate and high-sensitivity C-reactive protein concentration were higher in patients with knee osteoarthritis and were related to clinical features. In knee osteoarthritis, high-sensitivity C-reactive protein concentration may increase in early-stage KL-II.

Keywords

Knee osteoarthritis erythrocyte sedimentation rate high-sensitivity C-reactive protein Kellgren-Lawrence grade

Introduction

The degrees of osteoarthritis (OA) progression and severity are often evaluated by radiographic findings. However, in the early stages of OA, the features of OA in radiographic findings may be poor. Therefore, it would be beneficial to use serum and/or urinary biochemical markers to assess arthritis in its earlier stages and to monitor the effect of any treatment. There are several reports that local inflammation can affect the pathogenesis of OA.^1–4 The erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) concentration are the most common laboratory markers of systemic inflammatory diseases. Rheumatoid arthritis (RA) is a severe inflammatory disease, and ESR and CRP are elevated in patients with RA. In contrast, OA is a common joint disease, and it was previously considered that OA was not an inflammatory disease and that serum makers of inflammation were not elevated in patients with OA. However, there are several recent reports that ESR and CRP are slightly elevated in OA.^5–9 Furthermore, Montagne et al.¹⁰ introduced an immunoassay to measure high-sensitivity CRP (hsCRP) concentration,¹⁰ after which Sipe¹¹ reported that hsCRP concentration was elevated in OA. Therefore, the recent availability of a test to measure hsCRP concentration, which detects low-grade inflammation, could improve the determination of inflammation in OA.^12–15

The purpose of the current study was to determine whether hsCRP concentration and/or ESR are elevated in OA of the knee. We compared the values of ESR and hsCRP concentration in patients with knee OA (KOA) with those in patients without KOA. Additionally, we also evaluated the relationships between ESR, hsCRP concentration and clinical findings of KOA.

Materials and methods

We selected subjects for this study among patients in whom ESR and hsCRP concentration were measured preoperatively from April 2004, when hsCRP concentration was first measured in our hospital, until the present. hsCRP was assayed by the latex photometric immunoassay in our hospital laboratory. Patients who underwent arthroscopic treatment, high tibial osteotomy (HTO) and total knee arthroplasty (TKA) for OA of the knee were included in the KOA group. Patients who had previously undergone HTO and TKA were excluded from the KOA group. There were 104 patients (25 males, 79 females) with a mean age of 73 y (range: 48–85 y) in the KOA group. We evaluated the degree of KOA with radiographic findings using the Kellgren-Lawrence (KL) grade system.¹⁶ There were 10 patients (3 males, 7 females), with a mean age of 66 y (range: 49–83 y), in KL grade I; 21 patients (6 males, 15 females), with a mean age of 69 y (range: 52–85 y), in KL grade II; 31 patients (5 males, 26 females), with a mean age of 71 y (range: 53–81 y), in KL grade III; and 42 patients (11 males, 31 females), with a mean age of 72 y (range: 48–84 y), in KL grade IV. If the KL grades of the right and left knees were different, we chose the worse side. All patients were administered non-steroidal anti-inflammatory drugs (NSAIDs).

Patients with preoperative ESR and hsCRP concentration measurements with carpal tunnel syndrome, cubital tunnel syndrome, lumbar vertebrae disk herniation and soft tissue tumours were included in the non-KOA group. We excluded patients with any co-morbid joint OA (except the knee), RA, malignant tumours or inflammatory diseases (e.g. immune disorders or infection) from the non-KOA group. There were 50 patients (16 males, 34 females), with a mean age of 64 y (range: 43–83 y), in the non-KOA group.

We first evaluated whether age was statistically correlated with ESR and hsCRP concentration. We then compared ESR and hsCRP concentration between the KOA and non-KOA groups to determine whether ESR and hsCRP concentration were elevated in OA of the knee. In addition, in the KOA group, we assessed the relationships between ESR and hsCRP concentration and the following features of KOA: tenderness, swelling, patellar ballottement, flexion contracture and decreased flexion. Furthermore, the differences in ESR and hsCRP concentration by each KL grade were compared in the KOA group.

Statistical analysis

The correlations between age and ESR and hsCRP concentration were analysed using the Spearman’s correlation coefficient by rank test. Differences in ESR and hsCRP concentration between the KOA and non-KOA groups were analysed using the Mann-Whitney U-test. In the KOA group, differences in ESR and hsCRP concentration for each clinical finding were also analysed using the Mann-Whitney U-test. Differences in ESR and hsCRP concentration by KL grade were analysed using the Kruskal-Wallis test and the Mann-Whitney U-test. SPSS version 21 (IBM Corporation, Armonk, New York, USA) was used for statistical analysis. A P value of 0.05 was considered to be statistically significant.

Results

The correlations between age and ESR and hsCRP concentration are shown in Figure 1. Age was not statistically correlated with ESR and/or hsCRP concentration in either the KOA group or the non-KOA group. The comparison of ESR and hsCRP concentration between the KOA group and the non-KOA group is shown in Figure 2. Both ESR and hsCRP concentration were significantly higher in the KOA group than in the non-KOA group.

Figure 1.

The correlations between age and ESR and hsCRP concentration. (a) ESR in the KOA group. R = 0.168, P = 0.11. (b) hsCRP concentration in the KOA group. R = 0.112, P = 0.27. (c) ESR in the non-KOA group. R = 0.312, P = 0.088. (d) hsCRP concentration in the non-KOA group. R = 0.147, P = 0.32. Age was not correlated with ESR or hsCRP concentration in either group.

Figure 2.

ESR and hsCRP concentration in the KOA and non-KOA groups. Minimum, quartile 1 (Q1), median, quartile 3 (Q3) and maximum values are shown. (a) Comparisons of ESR between the KOA and non-KOA groups. ESR was significantly higher in the KOA group (median, 18.5; Q1, 10.0; Q3, 28.0 mm/h) than in the non-KOA group (median, 12.0; Q1, 5.0; Q3, 19.0 mm/h) (P = 0.0013). (b) Comparisons of hsCRP concentration between the KOA and non-KOA groups.

The relationships between ESR and OA features of the knee in the KOA group are shown in Table 1. ESR was significantly higher in patients with tenderness and/or patellar ballottement than those without these findings. The relationships between hsCRP concentration and OA features of the knee are shown in Table 2. The hsCRP concentration in patients with tenderness, swelling and patellar ballottement was significantly higher than that in patients without swelling, tenderness and patellar ballottement. ESR and hsCRP concentration were not significantly different in patients with and without flexion contracture (>5°) and decreased flexion (<120°).

Table 1.

The relationships between ESR and the OA features of the knee.

Clinical findings	ESR in the group with findings		ESR in the group without findings
Clinical findings	No.	Median (interquartile range)	No.	Median (interquartile range)	P ^a
Tenderness	71	20 mm/h (12–32)	33	15 mm/h (9.5–22)	0.032
Swelling	51	21.5 mm/h (10–30)	53	17 mm/h (10.5–23)	0.12
Patellar ballottement	28	22 mm/h (13–38)	76	17.5 mm/h (10–24.5)	0.038
Flexion contracture (>5°)	26	23.5 mm/h (12.5–32)	76	17 mm/h (10–26)	0.068
Decreased flexion (<120°)	34	21 mm/h (12–28)	70	17.5 mm/h (10–27)	0.17

ESR: erythrocyte sedimentation rate; OA: osteoarthritis.

ESR in patients with tenderness and patellar ballottement is significantly higher than that in patients without tenderness and patellar ballottement.

For two patients, findings of flexion contracture were unknown.

Mann-Whitney U-test (group with findings vs. group without findings).

Table 2.

The relationships between hsCRP concentration and OA feature of the knee.

Clinical findings	hsCRP in the group with findings		hsCRP in the group without findings
Clinical findings	No.	Median (interquartile range)	No.	Median (interquartile range)	P ^a
Tenderness	71	0.12 mg/dl (0.05–0.27)	33	0.05 mg/dl (0.04–0.085)	0.00030
Swelling	51	0.13 mg/dl (0.06–0.27)	53	0.06 mg/dl (0.035–0.125)	0.0042
Patellar ballottement	28	0.135 mg/dl (0.085–0.28)	76	0.06 mg/dl (0.04–0.16)	0.019
Flexion contracture (>5°)	26	0.055 mg/dl (0.02–0.44)	76	0.10 mg/dl (0.05–0.165)	0.53
Decreased flexion (<120°)	34	0.09 mg/dl (0.03–0.22)	70	0.10 mg/dl (0.05–0.16)	0.84

hsCRP: high-sensitivity C-reactive protein; OA: osteoarthritis.

The hsCRP concentration in patients with swelling, tenderness and patellar ballottement was significantly higher than that in patients without swelling, tenderness and patellar ballottement.

In findings of flexion contracture, as for two patients, the evaluation was unknown.

Mann-Whitney U-test (Group with findings vs. Group without findings).

The comparisons of ESR and hsCRP concentration at each KL grade are shown in Figure 3. ESR levels were not different between KL grades by the Kruskal-Wallis test (P = 0.107). ESR levels in grade I were significantly lower than those in grades III and IV by the Mann-Whitney U-test (Figure 3(a)). The hsCRP concentrations were different between KL grades by the Kruskal-Wallis test (P = 0.03). The hsCRP concentration in grade I was significantly lower than that in grades II, III and IV by the Mann-Whitney U-test (Figure 3(b)).

Figure 3.

ESR and hsCRP concentration at each KL grade.

Discussion

In the current study, hsCRP concentration and ESR were not correlated with age in either the KOA or the non-KOA group, although we hypothesized that hsCRP concentration might be. Kraus et al.¹⁷ previously reported that age was not correlated with CRP concentration. Therefore, we considered that adjusting for age was not necessary in this study. hsCRP concentration and ESR were higher in the KOA group than in the non-KOA group, in accordance with previous reports.^9,11,15 Although estimating the cut-off values of ESR and hsCRP concentration that detect patients with KOA would be useful in diagnosing OA, we were unable to do so here because of the large overlap of ESR and hsCRP values between the KOA and non-KOA groups.

In the KOA group, the clinical findings of tenderness, swelling and patellar ballottement correlated with hsCRP concentration and ESR. These findings are similar to previous reports in which elevated hsCRP concentration and ESR reflected clinical findings of KOA.^7,13 Thus, it is not clear whether the elevation of the values of ESR and hsCRP related to only degenerative severity of KOA in radiographic findings. ESR and hsCRP may rise by clinical symptoms such as tenderness, swelling and patellar ballottement.

In this study, we evaluated hsCRP and ESR according to each KL grade I–IV in detail. Regarding KL grades, hsCRP concentration in KL grades II, III and IV was significantly higher than that in KL grade I, and ESR in KL grades III and IV was significantly higher than that in KL grade I. However, there were no significant differences for hsCRP concentration and ESR between KL grades II, III and IV. Therefore, it appears that hsCRP concentration might predict OA progression in earlier stages of KOA, such as KL grade II. There are several reports that show no relationship between CRP and OA grades. However, in those studies, the newer method to determine hsCRP concentration was not used, and the outcome of progression of KOA was established as a 2-mm or more reduction of the joint space as evidenced on radiography or during operation for total knee arthroplasty; additionally, detailed severity of KL grade has not been considered in longitudinal studies.^9,14,18 He et al.¹⁹ reported that hsCRP concentration was above normal hsCRP in KL grade II. A more detailed evaluation of KL grade may be needed because there was no difference in hsCRP concentration among KL grades II, III and IV, which indicates severe KOA, as in our study. To our knowledge, there are few studies with patients with only early-stage KOA. In the early progression from KL grade I to II, hsCRP concentration may predict KOA.

There are limitations in the current study. First, patient characteristics, such as body mass index (BMI) and the type of NSAIDs used, were not investigated. There are controversial reports on the relationship between CRP and BMI. There are several studies reporting a relationship between CRP and the prevalence of KOA following adjustment for BMI,^6,18,20 whereas several other studies have reported that elevated CRP was not associated with OA following adjustment for BMI.^21,22 We consider that BMI is more related to OA than to CRP concentration, and future longitudinal studies should investigate BMI because several studies have reported an association between obesity and KOA.^23,24 The use of NSAIDs was not an important consideration, because most patients took them to treat knee pain associated with OA. Second, the design of this study was not longitudinal but cross-sectional. Because a cross-sectional analysis does not allow assessment of prognostic values, the cut-off value of hsCRP for OA progression could not be determined in the current study. However, we believe that hsCRP concentration is useful for assessing KOA in early stages (KL stages I and II) but not in severe stages (KL stages III and IV), and a longitudinal study of the low-level increases in hsCRP during the early stage of KOA is needed.

Conclusion

Age was not statistically correlated with ESR or hsCRP concentration. ESR and hsCRP concentrations were higher in patients with KOA than in patients without KOA and were related to the swelling and tenderness features of knee OA. Finally, hsCRP concentration may become elevated in earlier stages of KOA compared to ESR, potentially making it a more useful biomarker of OA progression.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical approval

Not required. This study was retrospective.

Guarantor

YK.

Contributorship

MH, HF and HK researched literature and conceived the study. MH and MT were involved in protocol development. MH wrote the first draft of the manuscript. All authors reviewed and edited the manuscript and approved the final version of the manuscript.

References

Smith

Triantafillou

Parker

. Synovial membrane inflammation and cytokine production in patients with early osteoarthritis. J Rheumatol 1997; 24: 365–371.

Saxne

Lindell

Månsson

. Inflammation is a feature of the disease process in early knee joint osteoarthritis. Rheumatology (Oxford) 2003; 42: 903–904.

Konttinen

Sillat

Barreto

. Osteoarthritis as an autoinflammatory disease caused by chondrocyte-mediated inflammatory responses. Arthritis Rheum 2012; 64: 613–616.

Rainbow

Ren

Zeng

. Inflammation and joint tissue interactions in OA: implications for potential therapeutic approaches. Arthritis 2012; 2012: 741582–741582.

Sharif

Elson

Dieppe

. Elevated serum C-reactive protein levels in osteoarthritis. Br J Rheumatol 1997; 36: 140–141.

Spector

Hart

Nandra

. Low-level increases in serum C-reactive protein are present in early osteoarthritis of the knee and predict progressive disease. Arthritis Rheum 1997; 40: 723–727.

Conrozier

Chappuis-Cellier

Richard

. Increased serum C-reactive protein levels by immunonephelometry in patients with rapidly destructive hip osteoarthritis. Rev Rhum Engl Ed 1998; 65: 759–765.

Sowers

Jannausch

Stein

. C reactive protein as a biomarker of emergent osteoarthritis. Osteoarthritis Cartilage 2002; 10: 595–601.

Takahashi

Naito

Abe

. Relationship between radiographic grading of osteoarthritis and the biochemical markers for arthritis in knee osteoarthritis. Arthritis Res Ther 2004; 6: R208–R212.

10.

Montagne

Laroche

Cuillière

. Microparticle-enhanced nephelometric immunoassay for human C-reactive protein. J Clin Lab Anal 1992; 6: 24–29.

11.

Sipe

. Acute-phase proteins in osteoarthritis. Semin Arthritis Rheum 1995; 25: 75–86.

12.

Stürmer

Brenner

Koenig

. Severity and extent of osteoarthritis and low grade systemic inflammation as assessed by high sensitivity C reactive protein. Ann Rheum Dis 2004; 63: 200–205.

13.

Punzi

Ramonda

Oliviero

. Value of C reactive protein in the assessment of erosive osteoarthritis of the hand. Ann Rheum Dis 2005; 64: 955–957.

14.

Pearle

Scanzello

George

. Elevated high-sensitivity C-reactive protein levels are associated with local inflammatory findings in patients with osteoarthritis. Osteoarthritis Cartilage 2007; 15: 516–523.

15.

Chen

Shah

Stabler

. Biomarkers associated with clinical phenotypes of hand osteoarthritis in a large multigenerational family: the CARRIAGE family study. Osteoarthritis Cartilage 2008; 16: 1054–1059.

16.

Kellgren

Lawrence

. Radiological assessment of osteo-arthrosis. Ann Rheum Dis 1957; 16: 494–502.

17.

Kraus

Stabler

Luta

. Interpretation of serum C-reactive protein (CRP) levels for cardiovascular disease risk is complicated by race, pulmonary disease, body mass index, gender, and osteoarthritis. Osteoarthritis Cartilage 2007; 15: 966–971.

18.

Brenner

Klotz

Alscher

. Osteoarthritis of the knee – clinical assessments and inflammatory markers. Osteoarthritis Cartilage 2004; 12: 469–475.

19.

Siebuhr

Brandt-Hansen

. Type X collagen levels are elevated in serum from human osteoarthritis patients and associated with biomarkers of cartilage degradation and inflammation. BMC Musculoskelet Disord 2014; 15: 309–309.

20.

Sharif

Shepstone

Elson

. Increased serum C reactive protein may reflect events that precede radiographic progression in osteoarthritis of the knee. Ann Rheum Dis 2000; 59: 71–74.

21.

Kerkhof

Bierma-Zeinstra

Castano-Betancourt

. Serum C reactive protein levels and genetic variation in the CRP gene are not associated with the prevalence, incidence or progression of osteoarthritis independent of body mass index. Ann Rheum Dis 2010; 69: 1976–1982.

22.

Jin

Beguerie

Zhang

. Circulating C reactive protein in osteoarthritis: a systematic review and meta-analysis. Ann Rheum Dis 2015; 74: 703–710.

23.

Coggon

Reading

Croft

. Knee osteoarthritis and obesity. Int J Obes Relat Metab Disord 2001; 25: 622–627.

24.

Reijman

Pols

Bergink

. Body mass index associated with onset and progression of osteoarthritis of the knee but not of the hip: the Rotterdam Study. Ann Rheum Dis 2007; 66: 158–162.