Schizophrenia Research 162 (2015) 175–182 Contents lists available at ScienceDirect Schizophrenia Research journal homepage: www.elsevier.com/locate/schres Global hypercoagulability in patients with schizophrenia receiving long-term antipsychotic therapy Vincent Chow a,b,c,CarolineReddela,b,c, Gabrielle Pennings a,b,c, Elizabeth Scott d,TundraPasqualone, Austin C.C. Ng b,c,ThomasYeohb, Jennifer Curnow a,c,f, Leonard Kritharides a,b,c,⁎ a ANZAC Research Institute, Sydney, Australia b Department of Cardiology, Concord Repatriation General Hospital, Sydney Local Health District, Sydney, Australia c University of Sydney, Australia d Brain & Mind Research Institute, University of Sydney, Australia e Department of Psychiatry, Croydon Health Centre, Sydney, Australia f Department of Haematology, Concord Repatriation General Hospital, Sydney Local Health District, Sydney, Australia article info abstract Article history: Background: Patients with schizophrenia are at increased risk of venous thromboembolism. The mechanisms Received 21 January 2014 underlying this association are poorly understood. Received in revised form 4 December 2014 Aims: We investigated whether there is a global hypercoagulable state in patients with schizophrenia utilising the Accepted 6 December 2014 overall haemostatic potential (OHP) assay which assesses overall coagulation potential (OCP), haemostatic Available online 27 January 2015 potential (OHP) and fibrinolytic potential (OFP). Method: Citrated plasma was collected for OHP assays from patients with schizophrenia on long-term antipsy- Keywords: chotic treatment and compared with healthy age- and sex-matched controls. Time courses of fibrin formation Schizophrenia Blood coagulation test and degradation were measured by spectrophotometry (absorption of 405 nm) after the addition of tissue factor Coagulation and tissue plasminogen activator to plasma. Thrombosis Results: Ninety patients with schizophrenia (antipsychotic treatment-15.9 ± 9.7 years) and 30 controls were Fibrinolysis recruited. Patients with schizophrenia had higher rates of smoking and levels of inflammatory markers (high- sensitivity C-reactive protein and neutrophil-to-lymphocyte ratio) than controls. Whilst D-dimer, fibrinogen and platelet count did not differ between patients with schizophrenia and controls, the OCP (54.0 ± 12.6 vs 45.9 ± 9.1, p = 0.002) and OHP (12.6 ± 5.8 vs 7.2 ± 3.7, p b 0.001) were higher, and OFP was lower (76.6 ± 9.8% vs 84.9 ± 6.4%, p b 0.001) in patients with schizophrenia, implying both a hypercoagulable and hypofibrinolytic state in these patients. Importantly, abnormalities in overall coagulation were independently predicted by levels of plasminogen-activator-inhibitor-1, fibrinogen, platelet count, inflammatory markers and plasma triglycerides, suggesting a multifactorial aetiology. Conclusion: Patients with schizophrenia have evidence of a global hypercoagulable and hypofibrinolytic state which may contribute to their increased risk of venous thromboembolism. © 2015 Elsevier B.V. All rights reserved. 1. Introduction reports of increased incidence of VTE amongst patients receiving con- ventional antipsychotic medications (Varia et al., 1983; Roche-Bayard Patients with schizophrenia have an increased risk of venous throm- et al., 1990)wereconfirmed by larger case–control studies which boembolic disease and other cardiovascular diseases. This increased reported a seven-fold increase in risk of VTE (Zornberg and Jick, 2000) risk is attributed to multiple factors including conventional risk factors and 13-fold increase risk of fatal pulmonary embolism (PE) (Parkin for cardiovascular disease such as smoking, obesity, the metabolic et al., 2003). There is some evidence of higher risk amongst new users syndrome and systemic inflammation. Other specific factors may and those receiving atypical antipsychotic medications (Walker et al., include immobility, catatonia during severe illness and the use of anti- 1997; Hagg et al., 2009; Parker et al., 2010). VTE has a reported early psychotic medications (both conventional and atypical). Early case case fatality rate of 7–11% (Stein et al., 2004), and a reported 5-year cumulative mortality rate of up to 32% (Ng et al., 2011). An increased risk of VTE occurs when at least one of Virchow's triad ⁎ Corresponding author at: Department of Cardiology, Concord Repatriation General is present (i.e. vascular endothelial damage, stasis of blood flow, and/or Hospital, Level 3 West, Hospital Road, Concord, NSW 2139, Australia. Tel.: +61 2 9767 7359; fax: +61 2 9767 6994. hypercoagulability of blood) (Anderson and Spencer, 2003). In the most E-mail address: [email protected] (L. Kritharides). recent VTE guidelines and current literature, hypercoagulability has http://dx.doi.org/10.1016/j.schres.2014.12.042 0920-9964/© 2015 Elsevier B.V. All rights reserved. 176 V. Chow et al. / Schizophrenia Research 162 (2015) 175–182 been shown to increase the risk of VTE by 2–9 fold in the general popu- cardiac evaluation and screening of patients with schizophrenia receiv- lation (Rogers et al., 2012; Authors/Task Force et al., 2014). ing antipsychotic treatment were recruited. The control groups com- Appropriate risk stratification for VTE risk and development of prised of healthy volunteers of similar age and gender with no prior prevention strategies requires a clear understanding of the mechanisms history of schizophrenia or VTE or ischaemic heart disease and those involved, and very few studies have directly quantified or addressed the not receiving antipsychotic or long-term anticoagulation medications. mechanisms of increased thrombotic risk in patients with schizophrenia Amongst the 90 patients with schizophrenia, there were no patients receiving long-term antipsychotic treatment. Overall coagulation and receiving conventional neuroleptics. 68/90 (76%) of patients were fibrinolysis are determined by the interplay between pro-coagulant receiving clozapine treatment, and the remaining 22 patients received factors such as factor VIII (FVIII) and plasminogen activator inhibitor the following treatment: olanzapine (6), aripiprazole (4), quetiapine type-1 (PAI-1), and anti-coagulant factors such as protein C and protein (4), paliperidone (4), risperidone (2), amisulpride (1) and zuclopenthixol S. FVIII is an important component of the intrinsic coagulation pathway (1). Patients were excluded if they: 1) were prescribed an antipsychotic and is responsible for the amplification of the coagulation cascade after medication for less than one year; 2) deemed noncompliant with initial thrombin generation (Furie and Furie, 2008). PAI-1 is one of the antipsychotic medications based on psychiatry reviews and serum most important inhibitors of the plasma fibrinolytic activity and is clozapine levels for patients receiving this drug; 3) had previous raised in inflammatory states, atherosclerosis and obesity (Cesari et al., history of VTE (PE and/or deep vein thrombosis [DVT]); 4) had symp- 2010). Elevated levels of FVIII and PAI-1 have been shown to increase toms suggestive of ischaemic heart disease and/or were found to the risk of VTE by 4.8-fold (Koster et al., 1995a; Jenkins et al., 2012b) have ischaemic heart disease on functional testing or 5) receiving and 1.6-fold (Meltzer et al., 2010) respectively. Proteins C and S exert long-term oral anticoagulation. All patients with schizophrenia met anticoagulant effects by inactivating coagulation factors V and VIII. DSM-IV diagnostic criteria for a schizophrenic disorder according Patients with protein C and S deficiencies have a 7.3–8.5 fold lifetime to the guidelines of the American Psychiatric Association (1994). increased risk of VTE (Koster et al., 1995b). This cross-sectional study was approved by the institutional Human Previous studies have found inconsistent associations between FVIII Ethics Committees. levels (Bank et al., 2007; Jenkins et al., 2012a), protein C and S levels (Mahmoodi et al., 2010; Bucciarelli et al., 2012), PAI-1 levels (Yukizawa et al., 2012), direct effects of antipsychotics (Carrizo et al., 2.2. Overall haemostatic potential (OHP) assay 2008) and VTE risk in these patients (Reitter-Pfoertner et al., 2013). Conventional testing for thrombophilia involves screening for specific Blood samples were collected by venipuncture from the cubital vein individual abnormalities in haemostasis which can be costly and with a 21 gauge butterfly needle. The tourniquet was removed and the prone to misinterpretation (Jennings et al., 2005). Consequently, cur- first 3 ml of blood was discarded. Blood was collected into BD rent guidelines do not recommend routine screening for heritable Vacutainer 0.109 M sodium citrate tubes (Becton-Dickinson, Franklin thrombophilia in asymptomatic individuals (Baglin et al., 2010; Lakes, NJ, USA) then centrifuged for 10 minutes (min) at 23 °C and Middeldorp, 2011a). 2500 g (Eppendorf 5804 R, USA). The plasma supernatant was centri- As the underlying biological mechanisms by which patients with fuged under the same conditions to produce platelet-poor plasma. We schizophrenia develop VTE are likely multifactorial, these may lend used a modified OHP assay based on that described by Blomback (He themselves to investigation by global coagulation assays rather et al., 1999, 2001; Curnow et al., 2007). Fibrin time curves were gener- than by measuring individual factors. The overall haemostatic ated in microtiter plate wells and plasmas were tested in duplicate. potential (OHP) assay is a simple, validated and inexpensive