key: cord-0720289-l9txc67d
authors: Gong, Xiaowei; Yuan, Boyun; Yuan, Yadong
title: Incidence and prognostic value of pulmonary embolism in COVID-19: A systematic review and meta-analysis
date: 2022-03-14
journal: PLoS One
DOI: 10.1371/journal.pone.0263580
sha: a3b5932e5d30435d7371aff297bac74fc54f699c
doc_id: 720289
cord_uid: l9txc67d

BACKGROUND: Pulmonary embolisms are frequently and prognostically in individuals infected by coronavirus disease 2019 (COVID-19); the incidence of pulmonary embolisms is varied across numerous studies. This study aimed to assess the pooled incidence of pulmonary embolic events and the prognostic value of such events in intensive care unit (ICU) admissions of patients with COVID-19. METHODS: The Cochrane Library, PubMed, and EmBase were systematically searched for eligible studies published on or before October 20, 2021. The pooled incidence of pulmonary embolism was calculated using the random-effects model. Moreover, the prognostic value was assessed by measuring the sensitivity, specificity, positive and negative likelihood ratio (PLR and NLR), diagnostic odds ratio (DOR), and the area under the receiver operating characteristic curve (AUC). RESULTS: Thirty-six studies involving 10,367 COVID-19 patients were selected for the final meta-analysis. The cumulative incidence of pulmonary embolism in patients with COVID-19 was 21% (95% confidence interval [95%CI]: 18−24%; P<0.001), and the incidence of pulmonary embolism in ICU and non-ICU patients was 26% (95%CI: 22−31%; P<0.001) and 17% (95%CI: 14−20%; P<0.001), respectively. The predictive role of pulmonary embolism in ICU admission was also assessed, and the sensitivity, specificity, PLR, NLR, DOR, and AUC were 0.31 (95%CI: 0.21−0.42), 0.84 (95%CI: 0.75−0.90), 1.88 (95%CI: 1.45−2.45), 0.83 (95%CI: 0.75−0.91), 2.25 (95%CI: 1.64−3.08), and 0.61 (95%CI: 0.57−0.65), respectively. CONCLUSION: This study found that the incidence of pulmonary embolism was relatively high in COVID-19 patients, and the incidence of pulmonary embolism in ICU patients was higher than that in non-ICU patients.

The severe acute respiratory syndrome caused by the novel coronavirus (SARS-CoV-2) was first reported as a cluster of cases in December 2019 [1] . The disease grew into the scale of a global pandemic rapidly, with 93,194,922 confirmed cases, and more than 2 million deaths have been reported as of January 2021 [2] . The clinical presentation of the coronavirus disease 2019 (COVID- 19) varies, creating challenges in determining the optimal management of patients infected with COVID-19. The primary cause of fatality in patients with COVID-19 was atypical acute respiratory distress syndrome (ARDS) due to the dissociation between wellconserved lung compliance and severe hypoxemia caused by pulmonary vasoregulation disruption and local thrombogenesis [3, 4] . Moreover, coagulopathy was reported in COVID-19 patients presenting with high levels of D-dimer, which causes inflammation (cytokine storm) and coagulation activation, and is associated with an increased risk of mortality [5] [6] [7] . However, the incidence of pulmonary embolism in patients in the intensive care unit (ICU) versus the general ward remains uncertain.

Several systematic reviews and meta-analyses have addressed the incidence of pulmonary embolism in patients with COVID-19 [8] [9] [10] [11] [12] . Suh et al. identified studies published before June 15, 2020, and found that the pooled incidence of pulmonary embolism was 16.5% and was significantly higher in ICU patients than in the non-ICU patients (24.7% versus 10.5%) [8] . Malas et al. found that the pooled incidence of pulmonary embolism was 13% in non-ICU and 19% in ICU patients [9] . Liao et al. reviewed 19 studies and found the pooled incidence of pulmonary embolism in COVID-19 patients was 15.3% [10] . Roncon et al. identified 23 studies and found that the pooled in-hospital incidence of pulmonary embolism was 23.4% in COVID-19 patients in the ICU and 14.7% in those in the general wards [11] . Lu et al. found that the pooled incidence of pulmonary embolism was 15% but did not find any significant effect on mortality rate when using anticoagulation therapy in hospitalized COVID-19 patients [12] . However, the pooled incidence for pulmonary embolism varied due to the varying severity of the included COVID-19 cases. Moreover, the most recent data were not included, and the results need to be updated. Therefore, this study was conducted to update the pooled incidence of pulmonary embolism in ICU and non-ICU patients infected with COVID-19. Furthermore, the predictive value of pulmonary embolism for ICU admission was also assessed.

This systematic review and meta-analysis were performed and reported following the Preferred Reporting Items for Systematic Reviews and Meta-Analysis Statement (PRISMA Checklist) [13] . The eligible studies reported the incidence of pulmonary embolism in patients infected by COVID-19 and included whether the patients were admitted to the ICU. PubMed, EmBase, and the Cochrane Library were systematically searched for eligible studies published on or before October 20, 2021, and the following search terms were applied: ((("COVID- )))))) OR (("severe acute respiratory syndrome coronavirus 2"[Supplementary Concept] OR "severe acute respiratory syndrome coronavirus 2"[All Fields]) OR "2019 ncov"[All Fields])) OR (("severe acute respiratory syndrome coronavirus 2"[Supplementary Concept] OR "severe acute respiratory syndrome coronavirus 2"[All Fields]) OR "sars cov 2"[All Fields]))) AND (("embol � "[All Fields] OR "thromb � "[All Fields]) OR ("Embolism"[MeSH Terms] OR "Thrombosis"[MeSH Terms])). The reference lists of the relevant review and original articles were also reviewed manually to identify any new eligible study that met the inclusion criteria.

The literature search and study selection were independently performed by two reviewers, and conflicts between the reviewers were resolved by mutual discussion. The inclusion criteria were as follows: (1) study design: no restrictions placed on study design, including prospective cohort, retrospective cohort, and case series; (2) participants: patients with COVID-19; and (3) exposure and outcomes: pulmonary embolism incidence after admission in ICU or non-ICU patients. Case reports and matched case-control studies were excluded since the incidence of pulmonary embolism was potentially biased.

The data items were independently extracted by two authors and included the names of the first authors of the article, publication year, study design, country, sample size, mean age, male proportion, setting, body mass index (BMI) of the patients, diabetes mellitus status (DM), hypertension status, cardiovascular disease status, and the number of pulmonary embolisms. Subsequently, the two authors independently assessed the quality of each individual study using the Newcastle-Ottawa Scale (NOS), which is based on selection (4 items), comparability (1 item), and outcome (3 items) [14] . Any disagreement between the authors regarding data abstracted and/or quality assessment was settled by mutual discussion until a consensus was reached.

The incidence of pulmonary embolism was based on the occurrence of the event and the total sample size in the cohort. Pooled incidence with 95% confidence intervals (CIs) was calculated using the random-effects model, which considered the underlying variability among the included studies [15, 16] . The sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), diagnostic odds ratio (DOR), and area under the receiver operating characteristic curve (AUC) for the role of pulmonary embolism in subsequent ICU admission were calculated, based on the number of ICU admissions in the patient group with a pulmonary embolism and the group without a pulmonary embolism. The pooled analysis was based on the bivariate generalized linear mixed model and random-effects model [15] [16] [17] . The heterogeneity across the included studies was assessed using I 2 and Q statistics, and significant heterogeneity was defined as I 2 >50.0% or P<0.10 [18, 19] . Subgroup analyses were performed for the pooled incidence of pulmonary embolism based on ICU admission or non-admission. Moreover, subgroup analyses were also conducted for the DOR based on the study design, mean age, male proportion, and study quality. Publication biases were assessed by using the funnel plot and Egger and Begg tests [20, 21] . All reported inspection levels were two-sided, and P<0.05 was regarded as statistically significant. All statistical analyses in this study were performed using STATA (version 10.0; Stata Corporation, College Station, TX, USA) and Meta-DiSc (version 1.4; Universidad Complutense, Spain) software.

The initial electronic search yielded 8,346 articles, with 5,642 remaining after the duplicate articles were removed. Subsequently, 5,483 articles were excluded because of irrelevant topics. The remaining 159 studies were retrieved for further full-text evaluations, and 123 studies were excluded as they did not report pulmonary embolism (n = 47), had a case-control or case report structure (n = 42), and had no ICU admission information included (n = 34). The cited references included in the articles were reviewed; however, no additional eligible studies were found. Thus, the remaining 36 studies were selected for the final meta-analysis (Fig 1) .

The baseline characteristics of the identified and selected studies are shown in Table 1 . Of the 36 included studies, six used a prospective design, while the remaining 30 were retrospective studies or case series. All studies were conducted in Europe, and the sample size ranged from 23 to 2,233, with a total of 10,367 patients with COVID-19. Thirty studies reported the incidence of pulmonary embolism in patients with COVID-19 admitted to the ICU, and 25 studies reported the incidence of pulmonary embolism in COVID-19 patients admitted to the general ward. Nine studies were categorized as having a relatively high quality with an NOS of seven stars, while 12 studies were rated as six stars, 11 as five stars, and four as four stars.

After pooling all the included studies, we noted that the pooled incidence of pulmonary embolism in patients with COVID-19 was 21% (95%CI: 18−24%; P<0.001; Fig 2) , and significant heterogeneity was observed among the included studies (I 2 = 95.0%; P<0.001). Subgroup analyses revealed that the incidence of pulmonary embolism in patients with COVID-19 admitted to the ICU was 26% (95%CI: 22−31%; P<0.001), while that in patients with COVID-19 admitted to the general ward was 17% (95%CI: 14−20%; P<0.001).

Seventeen studies reported the predictive value of pulmonary embolism for ICU admission. The pooled sensitivity and specificity were 0.31 (95%CI: 0.21−0.42) and 0.84 (95%CI: 0.75 −0.90), respectively (Fig 3) . There was significant heterogeneity in sensitivity (I 2 = 92.9%; P<0.01) and specificity (I 2 = 96.4%; P<0.01). Moreover, the pooled PLR and NLR were 1.88 (95%CI: 1.45−2.45) and 0.83 (95%CI: 0.75−0.91), respectively (Fig 4) . We noted potential significant heterogeneity in the PLR (I 2 = 36.8%; P<0.01) and NLR (I 2 = 62.4%; P<0.01). Furthermore, the DOR for the predictive value of pulmonary embolism for ICU admission was 2.25 (95%CI: 1.64−3.08; P<0.001; Fig 5) , and there was significant heterogeneity in the DOR (I 2 = 51.7%; P = 0.007). Lastly, the AUC for the role of pulmonary embolism in ICU admission was 0.61 (95%CI: 0.57−0.65; Fig 6) .

The results of subgroup analyses performed for the DOR are shown in Table 2 . Although the DOR of pulmonary embolism for ICU admission was statistically significant in most subgroups, this predictive value was not statistically significant when prospective studies were pooled. Moreover, we noted that the DOR of pulmonary embolism for ICU admission could be affected by the study design (P = 0.046) and male proportion (P = 0.030).

The publication biases for the incidence of pulmonary embolism and the predictive value of pulmonary embolism for ICU admission are listed in Bompard 2020

Hekimian 2020

Marone 2020

Middeldorp 2020

Leonard−Lorant 2020

Alonso−Fernandez 2020

Badr 2021

Filippi 2021

Leonard−Lorant 2020

Artifoni 2020

Nordberg 2021

Subtotal (I−squared = 95.0%, p = 0.000)

Lodigiani 2020

Grillet 2020

Helms 2020

Longchamp 2020

Soumagne 2020

Lodigiani 2020

Badr 2021

Valle 2021

Llitjos 2020

Garcia−Ortega 2021

Bruggemann 2020

Cerda 2020 bias for the incidence of pulmonary embolism (P-value for Egger: <0.001; P-value for Begg: <0.001); after adjusting for potential publication bias, the pooled incidence changed to 8% (95%CI: 6%-11%; P<0.001). Moreover, there was no significant publication bias for the predictive value of pulmonary embolism for ICU admission (P = 0.61).

The incidence of pulmonary embolism in patients with COVID-19 varies, and whether pulmonary embolism could predict ICU admission remains inconclusive. A total of 36 studies involving 10,367 COVID-19 patients was selected in this quantitative analysis, and the characteristics of these studies and patients were broad. This study found that the pooled incidence of pulmonary embolism was 21% (95%CI: 18−24%), and the incidence of pulmonary embolism in ICU patients was higher than that in non-ICU patients with COVID-19. Moreover, we noted a mild predictive value of pulmonary embolism for subsequent ICU admission in patients infected by COVID-19. Lastly, the predictive value of pulmonary embolism for subsequent ICU admission could be affected by the study design and male proportion. Several systematic reviews and meta-analyses have already illustrated the pooled incidence of pulmonary embolism in COVID-19 patients. However, these previous reviews combined the pooled incidence of pulmonary embolism according to ICU status, while the predictive role of pulmonary embolism for ICU admission was not illustrated [8] [9] [10] [11] [12] . Moreover, while the initial phases of COVID-19 yielded numerous new studies, recent meta-analyses and the pooled incidence of pulmonary embolism in this patient group should be updated. Therefore, Grillet 2020

Bompard 2020

Lodigiani 2020

Whyte 2020

Fauvel 2020

Marone 2020

Middeldorp 2020

Benito 2020

Alonso−Fernandez 2020

Scudiero 2020

Cerda 2020

Garc¨ªa this systematic review and meta-analysis were performed to update the pooled incidence of pulmonary embolism in patients with COVID-19 patients and assess the predictive role of pulmonary embolism for ICU admission in patients with COVID-19. This study found that the pooled incidence of pulmonary embolism for COVID-19 patients was 21%, and the incidence in each study ranged from 2−79%. Sensitivity analysis found that the pooled incidence of pulmonary embolism in COVID-19 patients ranged from 20.0−21.9%. The variability of the pooled incidence of pulmonary embolism in individual studies could be explained by the differences in the disease severity and frequency of diagnostic imaging. Several mechanisms could explain the elevated incidence of pulmonary embolism in patients with COVID-19, including that these patients present with abnormally elevated levels of proinflammatory cytokines [5] . The elevated systemic inflammation could be related to endothelial injury by attachment of the virus to the angiotensin-2 receptor of the endothelial cells and viral replication, which could cause prothrombotic endothelial dysfunction [58, 59] . Moreover, the use of central venous catheters, mechanical ventilation, platelet activation, immobilization, and the other characteristics of COVID-19 therapies could play an important role in the prothrombotic state. In addition, these factors could explain the significant heterogeneity in the pooled incidence of pulmonary embolism in patients infected with COVID-19. Grillet 2020

Bompard 2020

Lodigiani 2020

Whyte 2020

Fauvel 2020

Marone 2020

Middeldorp 2020

Benito 2020

Alonso−Fernandez 2020

Scudiero 2020

Cerda 2020

Garc¨ªa−Ortega 2021

Valle 2021

Nordberg 2021

Badr 2021 This study found that the pooled incidence of pulmonary embolism in patients with COVID-19 in the ICU was higher than that in non-ICU patients. Moreover, we noted that the predictive value of pulmonary embolism for ICU admission was statistically significant, although the predictive value of pulmonary embolism for ICU admission was mild. The higher incidence of pulmonary embolism in ICU patients could be explained by the severe procoagulant state in patients with COVID-19 presenting in a critical condition [60, 61] . However, we noted significant heterogeneity among the included studies in the relationship between pulmonary embolism and ICU admission, which could be explained by the variability in disease status, inflammatory status, and the diagnostic imaging modality used.

The limitations of this study should be acknowledged. This study was both a prospective and retrospective study. Thus, the results may have been affected by the uncontrolled biases from each study design. The diagnostic modality and frequency also differed amongst the studies, which could have affected the number of pulmonary embolisms diagnosed. In addition, the significant heterogeneity was not further analyzed because the number of studies reporting the predictive value of pulmonary embolism for ICU admission was small. Moreover, publication bias was inevitable because of the nature of the review and meta-analysis, which was based on the currently published articles. This study was not registered in PROSPERO, and the transparency of this study could be affected; the analysis used pooled data, and the stratified analyses according to the characteristics of the patients were restricted.

This study found that the pooled incidence of pulmonary embolism was 21% (95%CI: 18 −24%). Moreover, the incidence of pulmonary embolism in the ICU patients with COVID-19 was higher than that in the non-ICU patients with COVID-19. Moreover, the predictive value of pulmonary embolism for ICU admission was mild. Further research should be conducted to assess the role of pulmonary embolism in the prognosis of patients with COVID-19.

Supporting information S1 Checklist. PRISMA checklist. (DOC) 

Clinical Characteristics of Coronavirus Disease 2019 in China. The New England journal of medicine

Assessment of complications after pancreatic surgery: A novel grading system applied to 633 patients undergoing pancreaticoduodenectomy

Management of COVID-19 Respiratory Distress

Respiratory mechanics and gas exchanges in the early course of COVID-19 ARDS: a hypothesis-generating study

Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study

Abnormal coagulation parameters are associated with poor prognosis in patients with novel coronavirus pneumonia

Clinical Characteristics of 138 Hospitalized Patients With 2019 Novel Coronavirus-Infected Pneumonia in Wuhan, China

Pulmonary Embolism and Deep Vein Thrombosis in COVID-19: A Systematic Review and Meta-Analysis

Thromboembolism risk of COVID-19 is high and associated with a higher risk of mortality: A systematic review and meta-analysis. EClini-calMedicine

Incidence and mortality of pulmonary embolism in COVID-19: a systematic review and meta-analysis

Incidence of acute pulmonary embolism in COVID-19 patients: Systematic review and meta-analysis. European journal of internal medicine

A meta-analysis of the incidence of venous thromboembolic events and impact of anticoagulation on mortality in patients with COVID-19. International journal of infectious diseases: IJID: official publication of the International Society for Infectious Diseases

Preferred reporting items for systematic reviews and metaanalyses: the PRISMA statement

The HYSLAR trial: a prospective randomized controlled trial of the use of a restrictive fluid regimen with 3% hypertonic saline versus lactated Ringers in patients undergoing pancreaticoduodenectomy

Meta-analysis in clinical trials

Medical decision making: an international journal of the Society for Medical Decision Making

Properties of the summary receiver operating characteristic (SROC) curve for diagnostic test data

Cochrane Handbook for Systematic Reviews of Interventions 501

Measuring inconsistency in meta-analyses

Bias in meta-analysis detected by a simple, graphical test

Operating characteristics of a rank correlation test for publication bias. Biometrics

Acute Pulmonary Embolism in Patients with COVID-19 at CT Angiography and Relationship to d-Dimer Levels. Radiology

Acute Pulmonary Embolism Associated with COVID-19 Pneumonia Detected with Pulmonary CT Angiography

High risk of thrombosis in patients with severe SARS-CoV-2 infection: a multicenter prospective cohort study

Severe pulmonary embolism in COVID-19 patients: a call for increased awareness

Thrombotic and hemorrhagic events in critically ill COVID-19 patients: a French monocenter retrospective study

More on clinical and computed tomography characteristics of COVID-19 associated acute pulmonary embolism

Acute pulmonary embolism in non-hospitalized COVID-19 patients referred to CTPA by emergency department

Factors Associated With Pulmonary Embolism Among Coronavirus Disease 2019 Acute Respiratory Distress Syndrome: A Multicenter Study Among 375 Patients

Venous thromboembolism in critically Ill patients with COVID-19: Results of a screening study for deep vein thrombosis. Research and practice in thrombosis and haemostasis

Confirmation of the high cumulative incidence of thrombotic complications in critically ill ICU patients with COVID-19: An updated analysis

High incidence of venous thromboembolic events in anticoagulated severe COVID-19 patients

Pulmonary embolism in patients with COVID-19 pneumonia. The European respiratory journal

Systematic assessment of venous thromboembolism in COVID-19 patients receiving thromboprophylaxis: incidence and role of D-dimer as predictive factors

Thrombotic complications of patients admitted to intensive care with COVID-19 at a teaching hospital in the United Kingdom

Venous and arterial thromboembolic complications in COVID-19 patients admitted to an academic hospital in

Pulmonary embolism in hospitalised patients with COVID-19

Pulmonary embolism in COVID-19 patients: a French multicentre cohort study

Pulmonary Embolism in Patients With COVID-19: Awareness of an Increased Prevalence

Characteristics of Venous Thromboembolism in COVID-19 Patients: A Multicenter Experience from Northern Italy. Annals of vascular surgery

Thrombotic risk in COVID-19: a case series and case-control study

Incidence of pulmonary embolism in non-critically ill COVID-19 patients. Predicting factors for a challenging diagnosis

Incidence of venous thromboembolism in hospitalized patients with COVID-19

Antiphospholipid antibodies are not elevated in patients with severe COVID-19 pneumonia and venous thromboembolism

Pulmonary Thrombosis or Embolism in a Large Cohort of Hospitalized Patients With Covid-19. Frontiers in medicine

Prevalence of pulmonary embolism in patients with COVID-19 pneumonia and high D-dimer values: A prospective study

Pulmonary embolism or thrombosis in ARDS COVID-19 patients: A French monocenter retrospective study

Association Between Pulmonary Embolism and COVID-19 in Emergency Department Patients Undergoing Computed Tomography Pulmonary Angiogram: The PEPCOV International Retrospective Study. Academic emergency medicine: official journal of the Society for Academic Emergency Medicine

Pulmonary embolism in COVID-19 patients: prevalence, predictors and clinical outcome

The prevalence of pulmonary embolism in patients with COVID-19 and respiratory decline: A three-setting comparison

Blood test dynamics in hospitalized COVID-19 patients: Potential utility of D-dimer for pulmonary embolism diagnosis

Clinical characteristics and day-90 outcomes of 4244 critically ill adults with COVID-19: a prospective cohort study

Incidence, risk factors, and thrombotic load of pulmonary embolism in patients hospitalized for COVID-19 infection

Association between pulmonary embolism and COVID-19 severe pneumonia: Experience from two centers in the core of the infection Italian peak

Central pulmonary embolism in patients with Covid-19

Incidence and Outcomes of Pulmonary Embolism among Hospitalized COVID-19 Patients

Pulmonary embolism in patients with COVID-19 pneumonia: When we have to search for it?

Hyperinflammation and derangement of renin-angiotensin-aldosterone system in COVID-19: A novel hypothesis for clinically suspected hypercoagulopathy and microvascular immunothrombosis

SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor

Pulmonary embolism in the mechanically-ventilated critically ill patient: is it different?

COVID-19-Related Severe Hypercoagulability in Patients Admitted to Intensive Care Unit for Acute Respiratory Failure. Thrombosis and haemostasis