Assessment of the hand osteoarthritis activity in real clinical practice: possibilities and opportunities

Introduction

For a long period, osteoarthritis (OA) was considered a degenerative disease causing the negative acceleration of reparative processes in affected cartilage as a result of biomechanical and biochemical changes in the joint [1]. While studying the disease pathogenesis and implementing new methods of diagnostics, it was shown that, in OA, chronic inflammation existed, which involved all the joint components, including synovial lining, joint capsule, ligaments, sinews, and subchondral bone [2]. It has been this reason that at the moment has led to a change of the previously accepted name of this pathology "osteoarthrosis" to "osteoarthritis". Anti-inflammatory cytokines, in particular, interleukin-1 beta (IL-1β) and tumor necrosis factor alpha (TNFα), are of great importance in developing low-intensive ("low-grade") inflammation and thus synovitis in OA. Furthermore, in OA, there is insufficient production of inflammatory cytokines, including the IL-1β antagonist.

One of the OA versions – the affection of small joints of hand (HOA) – is currently under a magnifying glass of investigators. A wide spectrum of serum biomarkers of the inflammatory process, such as C-reactive protein (CRP), cartilage and bone degradation markers, was studied. However, up to now, no biomarkers of the HOA development and progression risk, possessing a diagnostic and prognostic value, have not been revealed [3]. Also, the role of cytokine profile in developing and maintaining HOA is actively studied.

Objective: to examine the relationship of functional disorders, joint structural changes with acute phase parameters (erythrocyte sedimentation rate (ESR) and CRP) and interleukin-1 beta (IL-1β) in patients with HOA.

Materials and methods

In the study, 52 women (39–87 years old) affected by HOA were enrolled. The mean age of examined persons was 63.4 (10.0) years. The HOA diagnosis was made on the basis of the criteria of the American College of Rheumatology [4]. The study did not include patients with secondary OA. In all patients, the pain level in hand joints at rest, in motion, and at night was determined using the visual analog scale (VAS). Additionally, the severity of the hand joint stiffness in the morning was assessed. The assessment of the hand joint pain and stiffness in the morning was performed in millimeters (mm), where 0 is the general absence of pain, 100 is the pain of the maximum severity. The degree of functional disorders and the patient’s aesthetic dissatisfaction with the general appearance of hands were assessed in points using the author’s questionnaire (Table 1).

Table 1

To perform laboratory tests, blood was taken in a plastic test tube without a stabilizer. In a laboratory, the blood was centrifuged at 1500 rpm for 10 min, and then the supernatant was poured in plastic single-use Eppendorf test tubes. Before the study, samples were stored at the temperature of -70 °C. The determination of the IL-1β level was conducted by the method of enzyme-linked immunosorbent assay (ELISA) using the eBioscience diagnostic set of reagents for determining IL-1β (Bender MedSystems, USA) according to the enclosed instruction for use. Washing of plates was performed by means of the ImmunoChem 2600 automatic microplate washer (High Technology, USA). Optical density was read using the ImmunoChem 2100 semi-automatic microplate reader (High Technology, USA). In the SoftMax® Pro computer application (Molecular Devices, Inc., USA), a calibration graph was plotted. Using this graph, the measurable indicator concentration was calculated. To determine ESR by the Panchenkov method [5], blood was taken into a plastic test tube with ethylenediaminetetraacetic acid (EDTA). To determine the CRP level, blood was taken into a plastic test tube without a stabilizer; the CRP level was determined by an immunoturbidimetric method using the Vital set of reagents (the Russian Federation).

In all patients, hand joint X-ray and ultrasound (US) examinations were performed. In twenty women, hand joint magnetic resonance imaging (MRI) was additionally performed. Changes in the carpometacarpal joint I (CMCJ I), interphalangeal joint I (IPJ I), distal interphalangeal joint (DIPJ), and proximal interphalangeal joint (PIPJ) of fingers II–V of both hands were assessed. The X-ray examination was carried out in the anteroposterior plane using standard modes on the DIRA-RG X-ray unit (Germany). X-rays were described in accordance with Kosinskaya’s classification [6]. A stage was determined in accordance with the severity of changes in the most affected joint. The OA X-ray stage, according to Kosinskaya’s classification, was met to the first one in 10 patients, the second one in 24 patients, and the third one in 18 patients. US was conducted on a GE Vivid 4 apparatus (USA) using a multi-frequency linear transducer with a frequency of 5–12 mHz. The joint visualization was realized in the sagittal plane from the back of the hand; the assessment of synovitis presence was performed by the gray scale. MRI was conducted on the GE Signa 1.5 tesla (USA). The study was performed in coronary, sagittal, and axial planes; the slice thickness was 2 mm. The following impulse sequences were used: Cor PD FatSat frFSE, Cor STIR fast IR, Cor T1 FSE, Cor T2 FSE FatSat, Cor PD frFSE, Sag PD FatSat frFSE, Ax PD FatSat frFSE. The assessment of tenosynovitis was performed without contrasting in the STIR fast IR and FatSat frFSE mode; the sign of tenosynovitis was considered to be the enhancement of the intensity of linear-form MR signal of the next sinew along the fiber direction. The assessment and comparison of examination methods were performed by the following parameters: joint space width, presence/absence of erosion, size of osteophytes. Additionally, by the US and MRI data, the synovitis presence was assessed. The assessment by the "joint space width" parameter was performed in points from 0 to 2 (where 0 is a lack of the joint space contraction, 1 is an insignificant contraction, and 2 is a significant contraction of the joint space). The size of osteophytes was estimated in millimeters. The assessment of the presence or absence of erosions or synovitis was performed in the following way: 0 is absence of erosion/synovitis, 1 is the presence of erosion/synovitis.

The analysis of the obtained data was conducted using the IBM SPSS 20 statistics software. Descriptive statistics for quantitative signs included arithmetic mean (M) and root-mean-square deviation (σ) for signs with normal distribution; the data were represented in the M(σ) format. In the case of the non-Gaussian distribution, signs were assessed by means of the median (Me), the upper limit of the first quartile (Q1), and upper limit of the third quartile (Q3). To assess the interconnection of values, the following methods of correlation analysis were used: Pearson’s coefficient of correlation (r) in case of normal distribution of study parameters and Spearman’s rank coefficient of correlation (rs) in case of non-Gaussian distribution. The correlation was considered to be satisfactory, if the corresponding coefficient was not less than 0.5 in absolute magnitude and was statistically significant (p < 0.05).

Results

The pain severity at rest, in motion, and at night by VAS was, on average, 28.3 (25.0) mm, 34.4 (26.0) mm, and 27.2 (26.6) mm, respectively; and the severity of stiffness in the morning was 39.7 (29.9) mm. The average scoring of functional disorders and aesthetic dissatisfaction of general appearance of hands was 2.3 (4.0) and 4.7 (3.1) points, respectively. The average level of ESR, CRP, and IL-1β in the study group was 14.9 (9.7) mm/h, 3.6 (2.3) mg/L, and 1.7 (3.6) pg/mL, respectively. The group general characteristics are given in Table 2.

Table 2

The X-ray, US, and MRI data were originally summed up in the joint score in line with the parameter to be assessed and method of diagnostics separately to DIPJ, PIPJ, CMCJ I, and IPJ I (Table 3).

Table 3

When analyzing the obtained data by means of correlation analysis, it was stated that there was no correlation between IL-1β, ESR, and CRP. Also, the researchers did not reveal any reliable dependence between the levels of ESR, CRP, and IL-1β and the severity of pain, stiffness in the morning by VAS, degree of functional disorders, and dissatisfaction with the general appearance of hands (r < 0.5; rs < 0.5) (Table 4).

Table 4

The results of the joint score for each of the parameters (size of joint space, size of osteophytes, presence of synovitis and erosions) and the method of diagnostics were similar to the results of laboratory testing (Table 5): no regularities between the severity of morphological changes and acute phase parameters were obtained (r < 0.5; rs < 0.5).

Table 5

Discussion

In accordance with the data of the present study, the ESR and CRP levels did not reflect the objective information on the degree of severity of inflammatory changes in HOA. It should be taken into account that ESR is a highly sensitive, but non-specific marker of systemic inflammation. The ESR determination has a limited value for diagnostics of rheumatologic diseases [7]. Some authors in their works determined the CRP level by a highly sensitive method, which the authors did not use. Literature data on how functional disorders, the severity of pain, and structural changes in hands correspond to the severity of systemic inflammation, which is assessed by acute phase parameters determined by means of various procedures, have contradictory character. So, Assirelli et al., in 2011, did not reveal a dependence between highly selective CRP (hsCRP) when comparing a group of healthy persons and patients with various versions of erosive and non-erosive HOA [8]. However, in another study performed in 2018 by Wen et al., the hsCRP level was correlated with the degree of severity of X-ray changes in HOA and OA of the knee joints and appeared to be higher in the erosive version of HOA [9]. It should be taken into account that both authors revealed the dependence of CRP on body mass index (BMI), smoking, and alcohol abuse. Some other studies [10][11] revealed a rise of both acute phase parameters in some patients with OA. In the study of Punzi et al. [12], a more significant increase in hsCRP was noted in patients with erosive HOA as compared to its non-erosive version. Furthermore, the hsCRP level correlated with the degree of X-ray changes and quantity of affected joints.

The obtained results do not reflect the dependence between the IL-1β level and degree of functional disorders, and the severity of structural changes in accordance with the data of X-ray, US, and MRI examinations. It corresponds to the data of other investigators present in the literature. So, Ma et al. (2019) declared that one of the problems of the IL-1β study was its low concentration in peripheral blood, which is difficult to quantify. In a study that the authors performed while examining the knee joint OA, IL-1β was found only in 16% of all samples despite the use of a highly sensitive set (ELISA), which the authors used in their study. Highly sensitive CRB (hsCRB) and IL-1β were associated with the severity of the disease state determined by the X-ray examination in the one-factor analysis and were not statistically significant after correction by BMI [13]. However, Roux et al. demonstrated in the study performed in 2016 that the degree of functional disorders and erosive affection in HOA were similar to the serum level of IL-1β. Nevertheless, they along with the authors did not obtain any relationship with the severity of synovitis according to the MRI data [14]. In another study performed in 2019, Liu et al. revealed a correlation between the blood and urine level of IL-1β and the stage in the knee joint OA [15].

Conclusion

Although the function of IL-1β in the pathogenesis of OA in a whole and HOA, in particular, is significant, the present study did not provide any data having a diagnostic value for the basic determination of IL-1β in the blood of patients with HOA. The determination of the IL-1β level in dynamics may be useful to assess the efficiency of treatment, in particular by the IL-1β inhibitors, and determine the prognosis for the disease. However, it requires an additional examination. HOA, as well as a majority of rheumatologic diseases, is characterized by the level of integral activity summing up from acute phase parameters and a degree of functional disorders. However, the results of the ESR and CRP data obtained by the authors using routine methods appeared to be non-informative to assess the HOA integral activity as well (in contrast to some other rheumatologic diseases). The use of highly sensitive methods to determine CRP may allow finding an interconnection of this parameter with the HOA integral activity, which will be important for real clinical practice for sure.