Radovan Malý, MD, PhD, First Department of Internal Medicine, University Hospital, Sokolská 581, 500 05 Hradec Králové, Czech Republic. Email: firstname.lastname@example.org
Aims: The aim of the present study was to compile a specific algorithm of prevention of venous thromboembolism in hospitalized psychiatric patients because this specific issue has not been addressed sufficiently in the literature.
Methods: The computer database MEDLINE was searched using key words (schizophrenia OR depression OR bipolar) AND (antipsychotic OR antidepressant) AND (venous thromboembolism OR pulmonary embolism) AND (prevention OR prophylaxis) in 2006.
Results: Based on the literature regarding non-surgical and surgical patients with respect to specificities in mental disorders (obesity induced with psychotropic drugs, possible catatonia, physical restraint, potential dehydration, antipsychotic treatment), a scoring system and a synoptic algorithm of prevention of venous thromboembolism modified for hospitalized psychiatric patients, were suggested.
Conclusions: According to the authors' knowledge this is the first attempt to establish such guidelines exclusively in psychiatry. Individual preventative clinical measures are suggested, ranging from regular physical exercise of lower extremities to repeated parenteral application of high doses of heparin tailored to every patient's risk for venous thromboembolism. Economic data support implementation of a proposed decision procedure into psychiatric clinical practice. Prospective discussion of its international applicability would be beneficial from both the clinical and the scientific points of view.
VENOUS THROMBOEMBOLISM (VTE) is one of the main causes of morbidity and mortality in hospitalized non-surgical patients. Increasing age, prolonged immobilization, history of VTE, cancer, surgery, trauma, acute ischemic stroke, obesity, pregnancy, puerperium, hormone therapy, varicose veins, and significant heart failure represent important clinical risk factors for VTE. In contrast, mutation in factor V Leiden or prothrombingenes, deficiency of antithrombin, protein C or S, hyperhomocysteinemia, high concentrations of factors VIII, IX or XI, antiphospholipid syndrome, disfibrinogenemia, and abnormal fibrinolysis are prominent laboratory risk factors. Combination of several factors increasing the risk of VTE in one patient at the same time is common.1,2 Risk of VTE is increased in psychiatric patients as compared to mentally healthy people,3,4 especially in schizophrenia and bipolar disorder.5 Meier-Ewert et al. found a fivefold higher incidence of VTE in patients treated with antipsychotics (AP) or antidepressants (AD) as compared to subjects without this medication.6 Zornberg and Jick documented a significantly increased risk of VTE in individuals receiving first-generation AP, particularly in the first months of the treatment.7 The Food and Drug Administration ascertained the total yearly death risk due to pulmonary embolism at 1:3450 during clozapine treatment.8 Modern second-generation AP such as risperidone or olanzapine may also be associated with an increased risk of VTE.9–12 Antipsychotics are presumed to induce pathological blood clotting via sedation of the patient as well as reduced locomotion, metabolic syndrome including obesity, hyperprolactinemia13,14 or antiphospholipid syndrome.15–18 Increased plasma concentration of antiphospholipid antibodies may even be present in mentally ill people primarily.19,20 Some authors also described hyperhomocysteinemia in schizophrenia21 or depression.22 Hyperhomocysteinemia >18.5 μmol/L increases the risk of VTE 2–2.5-fold.23
VTE diagnosis in the mentally ill is troublesome because of their sedation, immobilization or psychopathological symptoms. This is why VTE may not always be recognized, hence its fatal consequences of pulmonary embolism. In addition, adequate antithrombotic treatment is not always possible because of the patient's serious mental clinical state and non-compliance. A specific procedure for VTE prevention in psychiatry has not been described.24,25
The aim of the present study was to compile a specific algorithm of prevention of VTE in hospitalized psychiatric patients.
The computer database medline was searched using key words (schizophrenia OR depression OR bipolar) AND (antipsychotic OR antidepressant) AND (venous thromboembolism OR pulmonary embolism) AND (prevention OR prophylaxis) in 2006. We selected the articles most relevant for assessment of risk of VTE and its prevention in psychiatry.1,2,26–30
Based on the literature regarding non-surgical and surgical patients with respect to specificities in mental disorders (obesity induced with psychotropic drugs, possible catatonia, physical restraint, potential dehydration, antipsychotic treatment), we suggest a scoring system for VTE risk factors in hospitalized psychiatric patients with reduced mobility (Table 1). We defined reduced mobility as inability to walk 10 m at least for 1–2 weeks after admission to hospital. This is in accordance with other related studies.27 Scientific evidence-based factors score 2 points while 1 point was assigned to items based on consensus.
Table 1. Score of risk factors for VTE in hospitalized psychiatric patients with reduced mobility
Score 2 risk factors
Score 1 risk factors
Great saphenous vein and small saphenous vein, swelling and redness;
‡ dehydration can be defined as a clinically significant loss of body water (a loss of >10% is considered severe), e.g. due to diarrhea, hyperthermia, vomiting, burns, gastroenteritis, malnutrition with electrolyte imbalance, severe hyperglycemia especially in diabetes mellitus etc.
Mutation in factor V Leiden gene, mutation in 20210 prothrombin gene, deficiency of antithrombin, deficiency of protein C or protein S, antiphospholipid syndrome, hyperhomocysteinemia, high concentration of factor VIII, disfibrinogenemia.
Other factors (treatment with antidepressants, parkinsonism or smoking) could also be considered, but we did not find conclusive data on their significance in the MEDLINE database. Central venous catheter may represent a significant risk factor for VTE but it is not routinely used in mental health care in the Czech Republic.
History of deep-vein thrombosis or pulmonary embolism
In Table 2 we divided the quantity of assessed risk for VTE in psychiatry into three categories, similarly to general surgery.28 We preferred an exponential allocation of a risk magnitude to an even one because the risk of VTE significantly increases if several causative factors are present at the same time. In this case, the individual odds ratios are not simply added but are multiplied by each other.1,2 We recommend adequate clinical prevention steps pursuant to every risk range.
Table 2. Recommended prevention of VTE based on risk level
LMWH < 3400 U daily s.c. or UFH 5000 U every 12 h s.c. (in presence of 1 of evidenced risk factors – see Fig. 1, and/or physical restraint ≥8 h till full mobilization)
LMWH > 3400 U daily s.c. or UFH 5000 U every 8 h s.c. till full mobilization
Finally, we provide a synoptic algorithm of prevention of VTE27 modified for hospitalized psychiatric patients (Fig. 1).
Approximately 10 million people live in the Czech Republic, and 99% of them are Caucasians. According to the data on the same ethnic group from a community-based study conducted in western France,31 the overall incidence of deep vein thrombosis and pulmonary embolism in the Czech Republic is estimated at 14 000 per year and 6000 per year, respectively. The occurrence of VTE rises markedly with increasing age, the annual incidence is assessed as 1 per 100 subjects >75 years. The Epidemiology Groupe d'Etude de la Thrombose de Bretagne Occidentale (EPI-GETBO)32 study found that a large majority of VTE cases (70%) manifested during hospitalization of the patient in a medical ward. Hospitalizations for mental disorders were associated with 2.7% of VTE events.
Reports on VTE mortality are scarce,33,34 and the estimates differentiate considerably. The 30-day mortality rates fluctuate between 10 and 30%, and reports on 1-year case-fatality rates are even more inconsistent. More than one-third of deaths occur immediately after an unrecognized VTE arises. The 30-day mortality is more than twofold higher in patients with pulmonary embolism as compared to subjects with deep vein thrombosis.
Medical care of VTE patients in a Czech mental hospital is fully comparable to the treatment of mentally healthy people with VTE at an internal medicine ward in a general hospital. Every Czech mental hospital employs at least one physician specialized in internal medicine, and an internal medicine ward is an integral part of most mental hospitals in the country. The situation at psychiatric wards of general hospitals is even better.
We created an original algorithm of prevention of VTE in hospitalized psychiatric patients based on data on non-surgical and surgical patients27 modified in compliance with specificities of mental disorders and their treatment. According to our knowledge, this is the first attempt to establish such guidelines exclusively in psychiatry. Medical evidence-based data on a unique association of a risk for VTE with mental disorders and their therapy3–6 support this. We suggest individual preventative clinical measures ranging from regular physical exercise of lower extremities, to a repeated parenteral application of high doses of heparin tailored to every patient's risk for VTE.
Applicability of the present results is limited by missing scientific evidence. For example, a prospective observation of occurrence of VTE in a sample of mentally ill people untreated with psychotropic drugs as compared to a group of sex- and age-matched healthy controls would clarify the role of mental illness in VTE etiopathogenesis. Whether or not VTE is induced with antipsychotic treatment could be assessed in a prospective randomized double-blind trial comparing schizophrenia patients treated with antipsychotics versus subjects with schizophrenia without neuroleptic drugs. It is obvious that such research would be unacceptable from an ethical point of view.35
Other limitations result from pharmacological characteristics of proposed therapy. Pharmacological prophylaxis with low-molecular-weight heparin (LMWH) or unfractionated heparin (UFH) can be established only if no serious contraindications are present. Heparin-induced thrombocytopenia, hypersensitivity to heparin, active or uncontrollable bleeding, acute gastric/duodenal ulcer with a risk of bleeding, hemorrhagic stroke, serious liver disease with coagulopathy (international normalized ratio ≥1.5) or significant thrombocytopenia are the main contraindications of heparin prophylaxis.36
Caution during treatment with heparin is necessary in combined antiplatelet therapy, serious renal insufficiency (creatinine clearance <30 mL/min), uncontrolled arterial hypertension (blood pressure >180/110 mmHg), and in patients weighing <40 kg.37 In the case of serious renal insufficiency, doses of heparin can be monitored using plasmatic anti-Xa antibodies. If LMWH is not accessible, it is possible to carry on prophylaxis with unfractionated heparin 2–3 times s.c. a day. The safety profile of LMWH is better, and the LMWH dosage scheme is simpler as compared to UFH.38
Economic data support implementation of a proposed algorithm of VTE prevention into psychiatric clinical practice. The cost of 1 day of hospitalization for a patient suffering from VTE in an intensive care unit is Euro 62.3, and this cost comes up to Euro 82.2 in a case of pulmonary embolism (data provided by the Department of Economic Analyses, University Hospital Hradec Králové; Euro 1 = Czech Crowns 28 on 28 November 2006). The 1-day cost of VTE prophylaxis with LMWH (nadroparin 3800 U) is only Euro 2.9.39
The algorithm presented in this paper is designed to assist clinicians in deciding whether and to what extent an individual psychiatric patient should receive thromboprophylaxis. We make use of the proposed preventative measures at the Department of Psychiatry, University Hospital, Hradec Králové, and suggest the same practice for other psychiatric inpatient facilities. The following discussion on its international applicability would be beneficial from both the clinical and the scientific points of view.
Supported by the Research Projects MZO 00179906 and MSM 0021620816.