論文ID: CJ-21-0987
Background: The prevalence of deep venous thrombosis (DVT) among hospitalized psychiatric patients after coronavirus disease 2019 (COVID-19) infection remains unclear.
Methods and Results: We retrospectively investigated the prevalence of proximal DVT after COVID-19 infection among 50 hospitalized patients in a Japanese psychiatric hospital that in which a COVID-19 cluster developed between August and September 2020. The prevalence of proximal DVT was 10.0%. Patients with proximal DVT had a lower body weight and higher maximum D-dimer levels and International Medical Prevention Registry on Venous Thromboembolism (IMPROVE) VTE scores.
Conclusions: After COVID-19 infection, hospitalized psychiatric patients are at high risk of DVT and should be carefully followed up.
Since coronavirus disease 2019 (COVID-19) was first reported in December 2019 in Wuhan, China, COVID-19 has rapidly spread all over the world.1 Many studies reported that patients with COVID-19 developed coagulopathy, leading to thromboembolic complications, especially venous thromboembolism (VTE).2 In Japan, some recent studies reported COVID-19-associated VTE, although the incidence was lower than in other countries.3 Furthermore, a previous study reported that, following COVID-19 infection, patients were at high risk of VTE, including fatal pulmonary embolism (PE) and proximal deep venous thrombosis (DVT), even after discharge from hospital.4
The frequency of VTE has been reported to be higher in psychiatric than non-psychiatric patients because of a hypercoagulable state due to the administration of antipsychotic agents, physical immobilization, and underlying inherent coagulant dysfunction.5 Thus, the incidence of VTE in patients with COVID-19 is likely to be higher among psychiatric patients. In the present study, we used portable ultrasonography to screen for proximal DVT among psychiatric patients after a high-density COVID-19 cluster in a Japanese psychiatric hospital, and investigated the prevalence and clinical features of proximal DVT.
This study was a retrospective single-center cohort study that enrolled patients in a ward of a psychiatric hospital in Mie prefecture, Japan, that developed a COVID-19 cluster between August and September 2020.
Assessment of Proximal DVTEach patient was referred to an acute care medical hospital for the treatment of COVID-19. The patients who survived returned to the psychiatric hospital and underwent 3-point compression ultrasound (3PCUS). Patients were assessed by a multidisciplinary team that consisted of a cardiologist, a psychologist, and a clinical psychotherapist. The 3PCUS was performed using a V scan dual probe with a 3.4- to 8.0-MHz linear probe and a V scan Extend R2 with a 3.3- to 8.0-MHz linear probe (GE Healthcare, Chicago, IL, USA). The 3 points consisted of the bilateral common femoral veins, femoral veins, and popliteal veins, which were checked more than once with adequate compression by the probe until the veins were fully compressed. Patients were placed in the sitting position when feasible; otherwise, patients were placed in the supine position.6 Proximal DVT was defined as non-compressibility of the vein or visualization of an intraluminal clot.
Patient data were collected from the hospital charts. Additional patient information during COVID-19 treatment in the acute care medical hospitals was collected by questionnaires sent to individual physicians.
COVID-19 SeverityCOVID-19 severity was categorized as mild, moderate, or severe. Patients who did not require oxygen were defined as having mild COVID-19, those requiring oxygen were defined as having moderate COVID-19, and those requiring mechanical ventilation were defined as having severe COVID-19.3
Physical Function and Nutritional StatusThe controlling nutritional status (CONUT) score and Clinical Frailty Scale (CFS) were calculated at the time of the 3PCUS. The CONUT score reflects nutritional status and immunological characteristics by classifying patients into 3 groups: normal nutritional status (CONUT 0–1), mild malnutrition (CONUT 2–4), and moderate-severe malnutrition (CONUT ≥5).7 The CFS evaluates specific domains, including comorbidity, function, and cognition, generating a frailty score ranging from 1 (very fit) to 9 (terminally ill).8
VTE Risk AssessmentsThe maximum D-dimer level was defined as the highest D-dimer level between the onset of COVID-19 and the time of 3PCUS. The International Medical Prevention Registry on Venous Thromboembolism (IMPROVE) VTE score was used to assess the risk of VTE in hospitalized medical patients at the time of 3PCUS.9
Definition of Bleeding and Thrombotic EventsMajor bleeding events were defined according to the International Society on Thrombosis and Haemostasis definitions.10 Thrombotic events were defined as symptomatic arterial and/or venous thrombotic events diagnosed between the onset of COVID-19 and the time of 3PCUS.
EndpointsThe primary outcome was the prevalence of proximal DVT detected by 3PCUS. In addition, the entire population was divided into 2 groups based on the diagnosis of proximal DVT (i.e., with or without proximal DVT), and clinical information and patient data were compared between the 2 groups.
Ethical ConsiderationsAll procedures were in performed in accordance with the Declaration of Helsinki. The research protocol was approved by the ethics committees of Mie University Hospital and Suzuka Kosei Hospital after DVT screening had been performed for all patients (Reference no. H2021-066). In this study, because we used clinical information obtained as part of routine clinical practice, informed consent was obtained in the form of an opt-out option on the hospital’s website and bulletin boards in the hospital. This method is concordant with the guidelines for epidemiological studies issued by the Ministry of Health, Labor, and Welfare in Japan.
Statistical AnalysisCategorical variables are presented as numbers and percentages. Continuous variables are presented as the mean±SD or as the median and interquartile range (IQR) depending on data distribution. Categorical variables were compared with the Chi-squared test when appropriate; otherwise, Fisher’s exact test was used. Continuous variables were compared using Student’s t-test or the Mann-Whitney U test depending on data distribution. All statistical analyses were performed using Medcalc version 20.015. All reported P values are 2-tailed, and P<0.05 was considered significant.
There were 59 patients in the ward of the psychiatric hospital that recorded the COVID-19 cluster. Of these patients, 55 developed COVID-19. Five patients died because of COVID-19 with no sign of symptomatic PE. The remaining 50 patients underwent 3PCUS 38.6±8.7 days after the onset of COVID-19 and were enrolled in the study (Figure 1). The timing of 3PCUS in the mild, moderate, and severe COVID-19 groups was 39.8±9.1, 35.5±7.0, and 36.7±3.8 days after COVID-19 diagnosis, respectively. Maximum D-dimer levels were recorded 15 days (IQR 6–24 days) after COVID-19 onset. Maximum D-dimer levels in the mild, moderate, and severe COVID-19 groups were recorded 15 days (IQR 7–32 days), 13 days (IQR 4–17 days), and 16 days (IQR 15–17 days) after COVID-19 onset, respectively. Across the entire study population, the mean age was 68.9±9.7 years, 50.0% were male, and the mean body weight (BW) and body mass index (BMI) were 49.4±9.0 kg and 19.8±3.2 kg/m2, respectively. The median CONUT score and CFS were 4 and 6 points, respectively, indicating mild malnutrition and a moderately frail status (Table). Antipsychotic agents were prescribed in 43 (86.0%) patients, including 36 (72.0%) patients with schizophrenia.
Study flow chart. 3PCUS, 3-point compression ultrasound; DVT, deep venous thrombosis.
| Total (n=50) |
Patients with DVT (n=5) |
Patients without DVT (n=45) |
P value | |
|---|---|---|---|---|
| Age (years) | 68.9±9.7 | 75.6±5.6 | 68.2±9.8 | 0.06 |
| Male sex | 25 (50.0) | 2 (40.0) | 23 (51.1) | 0.67 |
| Body weight (kg) | 49.4±9.0 | 40.7±7.1 | 50.4±8.6 | 0.049 |
| Height (cm) | 158.1±7.5 | 155±6.4 | 158.4±7.6 | 0.68 |
| BMI (kg/m2) | 19.8±3.2 | 16.9±2.5 | 20.1±3.1 | 0.06 |
| CONUT score (points) | 4 [2.25–6] | 6 [5–8] | 4 [2–6] | 0.11 |
| CFS (points) | 6 [5–7] | 7 [7–7] | 6 [5–7] | 0.11 |
| Schizophrenia | 36 (72.0) | 4 (80.0) | 32 (71.1) | 1.00 |
| Administered antipsychotic agents | 43 (86.0) | 4 (80.0) | 39 (86.7) | 0.55 |
| Comorbidities | ||||
| Hypertension | 10 (20.0) | 0 (0) | 10 (22.2) | 0.57 |
| Diabetes | 5 (10.0) | 1 (20.0) | 4 (8.9) | 0.42 |
| Dyslipidemia | 4 (8.0) | 1 (20.0) | 3 (6.7) | 0.35 |
| Heart disease | 2 (4.0) | 0 (0) | 2 (4.4) | 1.00 |
| History of VTE | 0 (0) | 0 (0) | 0 (0) | NA |
| During COVID-19 treatment | ||||
| Hospital ward | ||||
| ICU | 5 (10.0) | 1 (20.0) | 4 (8.9) | 0.42 |
| General ward | 45 (90.0) | 4 (80.0) | 41 (91.1) | 0.42 |
| VTE risk | ||||
| Maximum D-dimer level (μg/mL) | 1.4 [0.7–5.2] | 7.84 [2–14.4] | 1.1 [0.6–4] | 0.03 |
| IMPROVE VTE risk score (points) | 1 [1–2] | 2 [2–2] | 1 [1–2] | 0.03 |
| Physical immobilization | 10 (20.0) | 1 (20.0) | 9 (20.0) | 1.00 |
| Severity of COVID-19 | ||||
| Need oxygen | 14 (28.0) | 3 (60.0) | 11 (24.4) | 0.13 |
| Need mechanical ventilation | 3 (6.0) | 1 (20.0) | 2 (4.4) | 0.28 |
| Need ECMO | 0 (0) | 0 (0) | 0 (0) | NA |
| Thromboprophylaxis management | ||||
| Compression stockings/intermittent pneumatic compression |
6 (12.0) | 1 (20.0) | 5 (11.1) | 0.49 |
| Anticoagulants | 11 (22.0) | 2 (40.0) | 9 (20.0) | 0.30 |
| Unfractionated heparin at prophylactic dose (%) | 0 | 11.1 | 0.42 | |
| Low-molecular-weight heparin (%) | 50.0 | 11.1 | ||
| Direct oral anticoagulants (%) | 50.0 | 77.8 | ||
| Major bleeding or thrombotic events | 0 (0) | 0 (0) | 0 (0) | NA |
Unless indicated otherwise, data are given as the mean±SD, median [interquartile range], or number (percentage). BMI, body mass index; CFS, clinical frailty scale; CONUT score, controlling nutritional status score; DVT, deep venous thrombosis; ECMO, extracorporeal membrane oxygenation; ICU, intensive care unit; IMPROVE, International Medical Prevention Registry on Venous Thromboembolism; NA, not applicable; VTE, venous thromboembolism.
The primary outcome was observed in 5 (10.0%) patients, all of whom had asymptomatic proximal DVT (Figure 1). The incidence of DVT in the mild, moderate, and severe COVID-19 groups was 5.6%, 18.2%, and 33.3%, respectively (Figure 2). Prophylactic anticoagulant therapy was administered in 11 (22.0%) patients during the treatment of COVID-19. The rate of administration of prophylactic anticoagulation in the mild, moderate, and severe COVID-19 groups was 16.7%, 18.2%, and 100%, respectively. In the mild COVID-19 group, 1 of 6 patients who received prophylactic anticoagulant therapy and 1 of 30 patients who did not receive prophylactic anticoagulant therapy had proximal DVT. In the moderate COVID-19 group, 2 of 9 patients who did not receive prophylactic anticoagulant therapy had proximal DVT (Figure 2). The selection of anticoagulant agents was left to the discretion of individual physicians.
Prevalence of deep venous thrombosis (DVT) and prophylactic anticoagulant therapy in each COVID-19 severity category.
Comparing patients with and without proximal DVT, there were no significant differences in patient characteristics except for BW (40.7±7.1 vs. 50.4±8.6 kg, respectively; P=0.049; Table). In addition, patients with proximal DVT tended to have a lower BMI (16.9±2.5 vs. 21.1±3.1 kg/m2; P=0.06), a higher CONUT score (6 vs. 4 points; P=0.11), and higher CFS (7 vs. 6 points; P=0.11). Furthermore, patients with proximal DVT had significantly higher maximum D-dimer levels (7.8 vs. 1.1 μg/mL; P=0.03) and IMPROVE VTE scores (2 points vs. 1 point; P=0.03; Table). No major bleeding or thrombotic events were observed in either group.
Following a diagnosis of proximal DVT, 2 of 5 patients received anticoagulants, whereas the remaining 3 patients only underwent follow-up 3PCUS because of anemia and small thrombus. None of the patients developed symptomatic PE.
The main findings of the present study at that: (1) the prevalence of proximal DVT following COVID-19 infection in hospitalized psychiatric patients was 10.0%; (2) the incidence of proximal DVT was relatively high among patients with severe COVID-19 despite prophylactic anticoagulant therapy; (3) patients with proximal DVT had a lower BW with a trend towards a lower BMI than patients without DVT, and tended to have a higher CONUT score and CFS; and (4) patients with proximal DVT had higher maximum D-dimer levels and IMPROVE VTE scores.
In the present study, proximal DVT was observed in 10.0% of 50 psychiatric patients following COVID-19. This prevalence of proximal DVT is relatively high compared with recent studies of Japanese patients with COVID-19, which reported an incidence rate of 0.6% for VTE and 1.9–2.9% for arterial and venous thrombotic events.11–13 The findings of the present study indicate that hospitalized psychiatric patients could be at higher risk of COVID-19-associated VTE than non-psychiatric patients. Schizophrenia and antipsychotics are known risk factors for VTE due to increased epinephrine release, increased activation of the markers of thrombogenesis, increased levels of antiphospholipid antibodies, and changes in serotonin metabolism in the platelets.14 In the present study, all but one patient with DVT had schizophrenia and had received antipsychotics, supporting the hypothesis that schizophrenia and antipsychotics lead to an increased incidence of COVID-19-associated VTE. However, all enrolled patients in the present study were screened for proximal DVT. A recent review report noted that the rate of COVID-19-associated VTE was high in studies with vs. without ultrasound screening.15 This may explain, in part, the high occurrence of proximal DVT in the present study; however, it is important to note that 10.0% of hospitalized psychiatric patients had proximal DVT after COVID-19 infection that required management.
In the present study, the rate of proximal DVT was high among patients with moderate or severe COVID-19, supporting the findings of other studies in which the incidence of VTE was high in certain patient groups, such as those in intensive care unit, with severe COVID-19.3,13,15 As reported previously,16 fatal coagulation abnormalities in severe cases of COVID-19 may be an important risk factor for the occurrence of VTE.
Obesity is a well-known risk factor for VTE.17 This has also been observed in Japanese patients with COVID-19.3 Contrary to these previous studies, patients with proximal DVT in the present study had a lower BW and tended to have a low BMI and high CONUT scores compared with patients without proximal DVT. This discrepancy may be explained, in part, by the existence and severity of the psychiatric diseases. Patients with proximal DVT also tended to have high CFS, which may have been ascribed to malnutrition, causing prolonged immobilization and stasis of blood flow in the lower extremities and thus leading to an increased rate of DVT.
D-dimer is a reliable negative predictive marker for the screening of acute VTE. In a previous study involving hospitalized COVID-19 patients, high D-dimer levels were associated with VTE.18 In addition, the IMPROVE VTE score is an extensively validated risk assessment model for the identification of medical hospitalized patients with VTE.9,18 Consistent with previous studies, the present study found high maximum D-dimer levels and IMPROVE VTE scores in patients with proximal DVT. Almost all the patients with proximal DVT had VTE risk factors, including immobilization and aging, which make up the IMPROVE VTE score, supporting the association between frailty and the occurrence of VTE. Conversely, the specificity of D-dimer and the IMPROVE VTE score for COVID-19-associated VTE has been reported to be low.18 Therefore, screening by imaging should not be performed based on only high D-dimer levels or IMPROVE VTE scores because of a poor cost-benefit balance. High COVID-19 severity, severe frailty, and high D-dimer levels and IMPROVE VTE scores in a psychiatric hospitalized patient may alert clinicians to the possibility of occurrence of VTE. Further data accumulation from multicenter surveillance is needed to confirm the findings.
Study LimitationsFirst, the present study was a retrospective single-center study based on data from a quite small population; thus, these findings may not accurately reflect all hospitalized psychiatric patients in Japan. Second, pulmonary imaging procedures could not be performed because of the lack of adequate facilities to diagnose PE in the psychiatric hospital. Thus, the primary outcomes did not include the prevalence of PE. Third, most of the patients with prophylactic anticoagulation received direct oral anticoagulants, which may have affected the prevalence of proximal DVT and D-dimer levels. Fourth, no data are available regarding the presence or absence of DVT before COVID-19 infection in the present study. Therefore, it is unknown whether and to what extent COVID-19 leads to an increased rate of VTE in psychiatric patients.
We have demonstrated some clinical features of DVT in hospitalized psychiatric patients in Japan following COVID-19 infection. The prevalence of proximal DVT was high at 10%. In particular, severe infection, low BW, and high D-dimer levels and IMPROVE VTE scores may lead to an increased rate of VTE. Therefore, the management of VTE may be warranted in COVID-19 psychiatric patients.
None.
This study did not receive any specific funding.
This study was approved by the Mie University Hospital Institutional Review Board (Reference no. H2021-066).
The deidentified participant data will not be shared.
