Reduced induction of anti-PF4/heparin antibody in RA patients after total knee arthroplasty

Heparin-induced thrombocytopenia (HIT) is an immune-mediated disorder that can develop during anticoagulant therapy with heparin [1]. HIT is associated with antibodies (Abs) that recognize the complexes formed between platelet factor 4 (PF4) and heparin [2]. There is increasing evidence that immune complexes formed by IgG Abs and the PF4/heparin complex bind and activate platelets resulting in acceleration of the coagulation pathway [3]. Recently, anti-PF4/heparin Abs have also been demonstrated in individuals who have not received heparin treatment [4]. PF4 could be an antigenic target in autoimmune diseases [5]. Additionally, anti-PF4/heparin Abs can be induced in patients who have undergone major surgery, even without any exposure of unfractionated heparin [6]. The development of HIT has been associated with postoperative thromboprophylaxis with fondaparinux, a factor Xa inhibitor [7]. Other examples of so-called “spontaneous” HIT preceding inflammatory or infectious triggers have been reported [8]. Krauel et al. [9] proposed that bacterial infections could trigger anti-PF4/heparin Ab formation. Anti-PF4/heparin IgG Abs are generated in approximately 20 % of patients undergoing total knee arthroplasty (TKA) [10]. In addition to the administration of heparin for anticoagulation during thromboprophylaxis, TKA poses a substantial challenge to the immune system because of mechanical damage to articular connective tissue [11].

PF4 is rapidly released after platelet activation and forms tetramers that bind to polyanions such as heparin, forming PF4/heparin complexes [12]. The resultant immune complexes induce the immune response resulting in the production of anti-PF4/heparin Abs [13]. However, the immunobiology of the anti-PF4/heparin Ab response is not well understood. Among surgical procedures, joint replacements result in extremely high rates of postoperative anti-PF4 Abs. The joint replacement procedure itself might be capable of triggering anti-PF4 Ab formation in both patients with osteoarthritis (OA) and those with rheumatoid arthritis (RA), which is an autoimmune disease; however, the frequency of and risks associated with anti-PF4/heparin Ab formation in RA patients who undergo joint replacement surgery without heparin exposure remain unclear.

In this study we compared the rate of seroconversion with anti-PF4/heparin Ab positive status between patients with OA and with RA during the postoperative period after TKA.

In this study we examined the differences in the rates of seroconversion to anti-PF4/heparin Ab positive status between patients with OA and with RA who underwent TKA. Previous studies have characterized the presence of anti-PF4/heparin Ab in patients with systemic lupus erythematosus (SLE) or anti-phospholipid Ab syndrome [5, 18]. Anti-PF4/heparin Abs have been reported to occur significantly more often in SLE patients than in healthy subjects [5]. However, the prevalence of anti-PF4/heparin Abs has not been investigated in other rheumatic diseases, including RA. To our knowledge, this is the first report documenting the frequency of seroconversion to anti-PF4/heparin Ab positive status after major joint surgery in RA patients. The most significant finding of this study was that TKA can induce seroconversion to anti-PF/heparin Ab positive status without exposure to heparin; however, in RA patients the induction of anti-PF4/heparin Ab production occurred less frequently than in OA patients. Of the OA patients who underwent TKA, 25.8 % became seropositive for anti-PF4/heparin Ab. These findings are consistent with the findings of previous studies, in which the seroconversion rates in similar patient populations that were not exposed to heparin were reported to be 14–24 % [6]. However, a significantly lower percentage of RA patients became seropositive for anti-PF4/heparin Abs after TKA than OA patients (4.5 % vs 25.8 %) in our study.

It is generally accepted that HIT is caused by antibodies which recognize complexes of PF4 and heparin [2]. The PF4 tetramer binds to heparin and glycosaminoglycans, an interaction that is central to the pathogenesis of HIT [19]. It has also been demonstrated that the binding of PF4 to certain bacteria or nucleic acids can initiate the generation of HIT-like Abs [20, 21]. Heparin has high avidity for PF4, and thus provides antigenic stimulation for HIT Abs [21]; however, a substantial number of patients who undergo major orthopedic surgery subsequently produce anti-PF4/heparin Abs without exposure to heparin. Perioperative inflammatory stimuli seem to disrupt PF4/heparin-specific B-cell tolerance [22]. In addition to the presence of heparin, the formation of immunogenic complexes that provoke HIT Abs depends on the availability of PF4. Upon release from activated platelets, PF4 rapidly associates with heparin sulfate in endothelial cells and can be brought back into circulation [23]. The availability of PF4 is therefore influenced by acute platelet activation, and logically plays a role in the risk for generation of PF4/heparin Abs during major joint surgery. However, preoperative serum PF4 levels did not differ between patients with RA and with OA.

PF4 binds to heparin-like molecules and PF4/heparin complexes stimulate anti-PF4/heparin Abs [24]. Thus, we reasoned that the increase in anti-PF4/heparin Abs after TKA might be part of the immune activation that occurs during postoperative inflammatory processes. In support of this hypothesis, we observed the appearance of anti-PF4/heparin Abs during the postoperative period after arthroplasty. However, the B-cell-mediated immune responses to PF4/heparin in RA patients were shown to be lower compared with those in OA patients. Assuming a causal relationship, we hypothesized that prescribed immunosuppressive medications, such as steroids or methotrexate, might reduce the likelihood of seroconversion to anti-PF4/heparin Ab positive status in RA patients. However, our results indicate that these medications did not affect anti-PF4/heparin Ab induction. Additionally, our data suggest that the heparin-binding affinity of circulating PF4 was not different between RA patients and OA patients.

More recently, Ohyama et al. [25] demonstrated that immune complexes containing PF4 are present specifically in the sera of RA patients, and these immune complexes could be a biomarker for the diagnosis of RA. These findings suggest that circulating PF4 is bound to IgG in established RA patients, which may contribute to the reduced likelihood of anti-PF4/heparin Abs in these patients. Although there is no statistical significance, the titers of RF were lower in RA patients with postoperative ant-PF4/heparin Ab compared with those without postoperative ant-PF4/heparin Ab. Furthermore, titers of anti-CCP Ab were significantly lower in RA patients with postoperative ant-PF4/heparin Ab compared with those without postoperative ant-PF4/heparin Ab. Immune complexes containing PF4 were found in the serum of 52 % of RA patients with anti-CCP Ab [26, 27]. It is possible that RA patients with high titers of RF or anti-CCP Ab could be preimmunized with PF4 which may be related to the abortive induction of anti-PF4/heparin Ab [27]. Our findings indicate that the circulating PF4 isolated from RA patients had a greater ability to bind IgG than the circulating PF4 in OA patients, which suggests preexisting immune responses against PF4 during the chronic immune-mediated inflammatory processes of RA. These responses may interfere with anti-PF4/heparin plasma cell differentiation through either regulatory or immune-tolerance mechanisms. The affinity of circulating PF4 for IgG in RA patients suggests that the immature B-cell immune response against PF4 occurs more frequently in RA patients than in OA patients. Our results suggest that certain immune reactions against PF4 are preexisting in RA patients. These preexisting immune responses may explain the reduced postoperative production of anti-PF4/heparin Abs in RA patients. Previous studies have demonstrated that an encounter between an immature B cell and its cognate antigen can lead to B-cell anergy [28]. This suggests a possible mechanism of B-cell tolerance for PF4/heparin complexes in patients with chronic inflammatory disorders such as RA. A recent study of anti-PF4/ heparin immunization in patients after cardiac surgery suggested that perioperative inflammation affects the immune response [29]. Alternatively, chronic rheumatoid inflammation may downregulate these perioperative inflammations that contribute to occurrence of anti-PF4/ heparin Ab in RA patients.

This study had several limitations. One major limitation was the potential for several biases in this quasi-randomized controlled trial. We were not able to obtain sufficient numbers of patients (particularly RA patients) to achieve good statistical power, so there is the undeniable possibility that the study may have been underpowered. Also, the J-PSVT does not provide information related to the RA severity score or data regarding treatments, such as antirheumatic drugs and glucocorticosteroids, so we were not able to adjust for these in our statistical models.

This study revealed a lower rate of seroconversion to anti-PF4/heparin IgG positive status after TKA in RA patients than in OA patients. This suggests that the autoimmune-like features of anti-PF4/heparin Ab production cannot be explained by the postoperative induction of this antibody. Our data provide further evidence for the unique profile of anti-PF4/heparin Ab induced during the postarthroplasty phase without exposure of heparin.