The association between depressive state and inflammation-related substances in serum in patients with knee OA

Purpose: Symptoms of knee OA patients, such as pain and functional disturbances, are associated with the psychosocial factors. In particular, the association between depressive state and symptoms of patients with knee OA has been focused. It has been reported that, for instance, depressive state and the disease progression are independently associated with the pain severity. In addition, the outcomes of therapeutic interventions were affected by depressive state of the knee OA patients. While the pathophysiology of depressive state in patients with knee OA has not yet been fully elucidated, the general population survey reported that the depressive states were associated with the serum levels of inflammatory cytokines in the patients, suggesting the involvement of inflammation in the depressive state of the patients. We have also reported that the severity of depressive state was associated with the serum IL-6 levels regardless of either the severity of the disease or the severity of pain. Based on these data, it is suggested that the depressive state of the patients with knee OA is associated with inflammation, although further study is needed to clarify. In this study, we examined whether there were any associations between the severity of depressive state and serum levels of the inflammation-related substances, and also examined the associations between the serum levels of the inflammation-related substances in the knee OA patients. Methods: Postmenopausal women with knee OA, who had visited our outpatient clinic due to knee pain and provided written informed consent, were enrolled in this study. Patients who met any of the following exclusion criteria were excluded from the study: 1) a history of drug therapy for knee OA or the administration of an intra-articular injection within 2 weeks before visit; 2) confirmed or suspected infectious/traumatic arthritis or rheumatoid arthritis; 3) a history of joint replacement surgery; and 4) a history of psychiatric disease or a history of consulting a psychiatrist. The severity of knee OA was classified according to the Kellgren-Lawrence (K/L) grading scale. All patients had knee OA with K/L grade 2 or more. Serum levels of IL-6, hyaluronic acid (HA), and hs-CRP were measured to monitor inflammation. The Zung self-rating depression scale (SDS) was used to measure the mental state. The SDS scores range from 20 to 80 and indicate a depressive state when these are 40 or more. Pain was evaluated by a pain and stiffness sub-category of the Japanese Knee Osteoarthritis Measure (JKOM-pain), which is a patient-oriented outcome measure with sufficient reliability and validity, similar to WOMAC. Results: While 160 patients with knee OA were registered, 146 (K/L grade 2/3/4: 64/53/29) was enrolled in the present study. As 72 (K/L grade 2/3/4: 35/22/15) of 146 patients (49.3%) showed 40 or more of SDS, they were included in the depressive state group (group D). The remaining 74 (K/L grade 2/3/4: 29/31/14) patients (50.7%) showed less than 40 of SDS, they were included in the non-depressive group (group ND). There was no significant differences of the severity distribution of knee OA between D and ND groups (p = 0.350). The serum IL-6 levels (2.6 ± 1.7 pg/ml), SDS (45.9 ± 4.7 point), and the JKOM-pain score (16.1 ± 6.6 point) in Group D were significantly higher than those in Group ND (sIL-6: 2.0 ± 1.1 pg/ml, p = 0.004; SDS: 32.3 ± 5.2 point, p = 0.001; JKOM-pain score: 12.9 ± 6.4 point, p = 0.004). sIL-6 levels were significantly associated with SDS of the all patients (r = 0.313, p = 0.001). sIL-6 levels were also significantly associated with SDS of the patients in both group D (r = 0.222, p = 0.061) and group ND (r = 0.308, p = 0.008), respectively. Serum HA levels (sHA) were significantly associated with SDS of the patients in group D (r = 0.351, p = 0.003), while those were not in patients with group ND (r = −0.200, p = 0.091). sIL-6 levels were significantly associated with sHA levels in both the all patients (r = 0.330, p = 0.001) and the patients in group D (r = 0.413, p = 0.003). On the other hand, no association between sIL-6 and sHA levels was observed in the patients with group ND (r = 0.077, p = 0.516). sIL-6 levels were significantly associated with hs-CRP levels in both the all patients (r = 0.355, p = 0.001) and the patients in group D (r = 0.597, p = 0.001). Conclusions: Almost half of the patients with painful knee OA who visited the outpatient clinic of our university hospital were in depressive state regardless of the OA severity. Both sIL-6 and sHA were associated with SDS of the patients in group D. Although further study is need, these data may suggest the involvement of synovitis occurred in knee OA for the depressive state in patients with knee OA.