Pulmonary embolism (PE) involves the partial or complete occlusion of the pulmonary arteries and/or their branches, with compromised ventilation and severe hemodynamic consequences determined by the size and location of the embolus, as well as any pre-existing cardiopulmonary disease [1]. In the majority of cases, lower-extremity deep vein thrombosis (DVT) is the cause of embolism [1]. PE is the third leading cause of cardiovascular mortality, with over 100,000 deaths per year in the United States, accounting for 5%-10% of in-hospital deaths [2]. The clinical presentation varies from asymptomatic (incidentally diagnosed) to severe or fatal [2]. Silent PE is prevalent in patients with DVT [3]. Previous epidemiological studies have reported that the incidence of silent PE in DVT patients ranges from 11% to 46% [4,5]. However, an incidence of up to 66%-72% has been reported when computed tomography pulmonary angiography (CTPA) is used for routine screening [6].

Routine screening for occult PE in patients with DVT has been advocated considering the high incidence, potential for severe cardiopulmonary consequences, and benefits of baseline imaging for individualizing subsequent treatment options [1,4-8]. However, this practice is controversial due to increased costs, radiation exposure, and the use of contrast media [9,10]. Moreover, the absence of high-quality evidence demonstrating the clinical and economic benefits of routine investigations has led the European Society for Vascular Surgery (ESVS) to recommend against routine screening for silent PE in patients newly diagnosed with DVT in its recent guidelines (class III, level C) [3]. Instead, selective screening in high-risk DVT patients has been proposed as a reasonable alternative [11]. However, specific DVT characteristics that could serve as risk factors for silent PE have not been well documented or elucidated.

This study aimed to evaluate the incidence and extent of occult PE in patients with symptomatic DVT, and its relationship with DVT characteristics and potential risk factors. The ultimate goal was to investigate whether routine or selective screening for silent PE in DVT patients can be justified.

1) Inclusion and exclusion criteria

We performed a retrospective analysis of prospectively collected data from symptomatic DVT patients from the vascular surgery department of a single tertiary university center. The study period covered 5 years, from 2019 to 2023. The study protocol was approved by institutional review board of “G. Gennimatas” General Hospital of Thessaloniki (approval No. 13/2019) and was conducted in accordance with the Declaration of Helsinki (2013 amendment). Written informed consent was obtained from all participants. Two independent reviewers searched the electronic medical records and picture archiving and communication system of our hospital in order to identify eligible patients. A total of 256 consecutive patients with lower-extremity DVT were admitted to our department. All patients with chronic DVT and symptoms suggestive of PE, such as dyspnea, cyanosis, chest pain, cough, hemoptysis, tachycardia, hypotension, dizziness, fainting, excessive sweating, or low oxygen saturation, were excluded from the analysis.

2) Clinical and imaging methodology

Before admission, all patients were examined in the emergency department and assessed using the Wells score to determine the probability of DVT and PE. A detailed medical history was obtained, and laboratory workup included a complete blood count, a basic metabolic panel, and basic coagulation tests. Plasma D-dimer levels were measured prior to admission. A D-dimer value greater than 0.5 mg/L was considered abnormal. The diagnosis of DVT was made by means of B-mode 3-point compression venous ultrasound scanning, after which the patient was admitted to our department. Chest radiography (CXR; frontal and lateral views) and 12-lead electrocardiography (ECG) were performed upon admission. On day 1, we performed a detailed bilateral whole-leg colored duplex ultrasound covering the iliac, common femoral, femoral, popliteal, and calf veins. According to our routine protocol, a combined CTPA and iodine contrast-enhanced abdominal and pelvic CT scan (after per os Gastrografin administration) was performed on day 2 to detect occult PE and/or any subclinical, previously undiagnosed neoplasm. In all patients, 60 mL of iodine contrast medium (iopromide, 300 mg/mL) was administered at a rate of 4 mL/s for the CTPA. After the CTPA examination, an additional 80 mL of contrast medium was administered for the abdominal and pelvic CT scans. PE was confirmed if one or more intraluminal filling defects were present in the trunk or branches of the pulmonary arteries. PE was categorized based on the location of the embolus: main pulmonary, lobar, segmental, or subsegmental arteries. Patients diagnosed with PE underwent transthoracic echocardiography and were monitored in a high-dependency unit.

Graduated compression stockings providing 30 to 40 mmHg were applied from day 1, and an adjusted therapeutic dose of low molecular weight heparin was administered daily for the duration of hospitalization. Post-discharge anticoagulant therapy consisted of a direct oral anticoagulant at a therapeutic dosage, such as 20 mg of rivaroxaban once daily or 5 mg of apixaban twice daily. Anticoagulant therapy lasted for at least 3 to 6 months, based on the assessment of recurrence risk factors and bleeding risk.

3) Data collection and statistical analysis

We reviewed the patient files and recorded demographics (age and sex), the delay from symptom onset to admission, clinical manifestations, comorbidities, and medications with a particular emphasis on antiplatelet or anticoagulant treatments. Additionally, we noted risk factors for DVT such as smoking, thrombophilia, immobilization, recent surgery, obesity, cancer, the peripartum period, and previous history of venous thromboembolism (VTE). Moreover, imaging information regarding DVT characteristics (involved extremities, location and extent of thrombus) and the location and extent of PE were recorded.

The data was entered into a computerized database, and statistical analysis was performed using the SPSS Statistics program, version 22.0, for macOS (IBM Corp.). Categorical variables are presented as counts and percentages, analyzed using the chi-square test. Continuous variables following a normal distribution are presented as mean±standard deviation and were analyzed using Student t-test, whereas those not following a normal distribution are presented as median and interquartile range (IQR), analyzed using Mann–Whitney U-test. The diagnostic value of D-dimer as a marker for occult PE was further analyzed using a receiver operating characteristic (ROC) curve. Finally, the effects of various risk factors on the presence of PE were assessed using logistic regression with both univariate and multivariate analyses. The significance level for the statistical tests was set at a P-value <0.050.

A total of 256 consecutive patients with symptomatic lower-extremity DVT were admitted to our department. Fifty-five of them met the exclusion criteria and were subsequently excluded. Finally, 201 patients were included in this analysis. CTPA was performed in all patients to diagnose occult PE. Table 1 presents the demographics, basic characteristics, and clinical presentations of the study sample. The overall incidence of silent PE in our sample of DVT patients was 48.8% (98/201). Patients were grouped according to the presence or absence of silent PE as detected on CTPA. A higher proportion of males was observed in the PE group (63.3% vs. 40.8%, P<0.001). Factors such as smoking, recent surgery, cancer, and a history of VTE were significantly more frequent in the PE group (P<0.050 for each factor). The median delay from symptom onset to hospital admission was 3 days (IQR=2-4), with no significant differences between the two groups. A PE Wells score >4 on the initial clinical evaluation was marginally associated with a higher incidence of silent PE (11.2% vs. 7.8%, P=0.045).

Table 1 . Demographics, clinical characteristics, comorbidities, and risk factors of patients with lower-extremity DVT.

Variable	Patients with lower-extremity DVT			P-value
	All patients (n=201)	Silent PE (n=98)	No PE (n=103)
Demographics
Age (y)	66.2±8.3	64.6±8.2	68.9±8.7	0.256
Male sex	104 (51.7)	62 (63.3)	42 (40.8)	<0.001*
Comorbidities
Smoking	130 (64.7)	68 (69.4)	62 (60.2)	0.034*
Hypertension	180 (89.6)	87 (88.8)	93 (90.3)	0.201
Diabetes	78 (38.8)	37 (37.8)	41 (39.8)	0.243
Dyslipidemia	176 (87.6)	85 (86.7)	91 (88.3)	0.309
CVD	18 (9.0)	9 (9.2)	9 (8.7)	0.299
CAD	21 (10.4)	10 (10.2)	11 (10.7)	0.301
COPD	15 (7.5)	7 (7.1)	8 (7.8)	0.290
CKD	13 (6.5)	6 (6.1)	7 (6.8)	0.286
Antiplateleta	17 (8.5)	8 (8.2)	9 (8.7)	0.306
Anticoagulantb	5 (2.5)	2 (2.0)	3 (2.9)	0.212
Risk factors
Immobility	44 (21.9)	22 (22.4)	22 (21.4)	0.189
Recent surgery	23 (11.4)	13 (13.3)	10 (7.8)	0.039*
Recent trauma	10 (5.0)	6 (6.1)	4 (3.9)	0.089
Varicose veins	18 (9.0)	8 (8.2)	10 (9.7)	0.101
Obesity	40 (19.9)	21 (21.4)	19 (18.4)	0.098
Pregnancy	1 (0.5)	1 (1.0)	0 (0.0)	NA
Puerperium	6 (3.0)	4 (4.1)	2 (1.9)	0.076
Thrombophilia	3 (1.5)	2 (2.0)	1 (1.0)	0.069
Cancer	21 (10.4)	13 (13.5)	8 (7.8)	0.011*
Hormone therapy	4 (2.0)	2 (2.0)	2 (1.9)	0.202
Previous VTE history	33 (16.4)	18 (18.4)	15 (14.6)	0.018*
Unprovoked DVT	53 (26.3)	27 (27.5)	26 (25.2)	0.082
Clinical characteristics
Delay from symptom to admission (d)	3 (2-4)	4 (2-5)	2 (1-4)	0.066
DVT Wells score ≥2	176 (87.6)	86 (87.7)	90 (87.3)	0.334
PE Wells score >4	19 (9.5)	11 (11.2)	8 (7.8)	0.045*
D-dimer (mg/L)	6.79 (2.9-12.7)	9.60 (3.2-13.6)	5.51 (2.4-11.2)	0.001*
D-dimer ≥5.14 mg/L	130 (64.7)	87 (88.8)	43 (41.7)	<0.001*

Values are presented as number (%) or mean±standard deviation. Delay from symptom to admission and D-dimer are presented as median (interquartile range)..

BMI, body mass index; CAD, coronary artery disease; CKD, chronic kidney disease; CVD, cerebrovascular disease; COPD, chronic obstructive pulmonary disease; DVT, deep vein thrombosis; NA, not applicable; PE, pulmonary embolism; VTE, venous thromboembolism..

^aAspirin or clopidogrel, ^bdabigatran, rivaroxaban, apixaban, or acenocumarol..

*Statistically significant differences at P<0.05..

The median admission D-dimer value in our sample was 6.79 mg/L (IQR=1.98-11.78) and was significantly higher in patients with silent PE than in those without PE (9.60 vs. 5.51 mg/L, P=0.001). The diagnostic value of D-dimer as a marker for occult PE was further analyzed using a ROC curve (Fig. 1). The area under the curve was 0.661 (P<0.001). A D-dimer value >5.14 mg/L was found to be discriminative in predicting silent PE in patients with DVT, with a sensitivity (Sn) of 68.2% and specificity (Sp) of 59.3%. The positive predictive value (PPV) was 61.4% and the negative predictive value (NPV) was 66.2%.

Figure 1. ROC analysis was conducted to assess the predictive value of D-dimer testing for silent PE in patients with lower extremity DVT. ROC, receiver operating characteristic; AUC, area under the curve; PE, pulmonary embolism; DVT, deep vein thrombosis.

Table 2 summarizes the imaging and anatomical data of DVT for the study sample. Unilateral DVT was observed in 181 patients (90.0%), while bilateral DVT was present in 20 patients (10.0%). The left leg was more commonly affected (69.2% vs. 30.8%, P<0.001). In most cases, the DVT was proximal (90.5%), with the iliofemoral type accounting for 60.7% of all cases. In 11 patients (5.5%), the thrombus extended to the inferior vena cava (IVC). Silent PE was significantly more common on the right side, affecting the right lung (46.6% vs. 33.7%, P=0.039), and bilateral PE was observed in 19.7% of patients. The embolus locations were as follows: main pulmonary artery in 29.6%, lobar arteries in 32.7%, segmental arteries in 20.4%, and subsegmental arteries in 17.3%. There was no statistically significant difference in the incidence of silent PE with respect to the number or side of the affected limb(s). However, a higher incidence of occult PE was observed in patients with iliofemoral DVT (65/122, 53.3%) than those without iliofemoral DVT (33/79, 41.8%; P=0.037), particularly when the thrombus extended to the IVC (90.9% vs. 46.3%, P=0.003) compared with those without extension to IVC. Moreover, iliofemoral DVT was associated with a larger and more proximally located (main pulmonary or lobar) embolus, whereas distal DVT primarily involved, the segmental and subsegmental arteries (P=0.041; Table 3). Other factors associated with proximal silent PE, identified through univariate analysis, included male sex, recent surgery, cancer, previous VTE history, D-dimer value ≥5.14 mg/L, and thrombus extension to the IVC (P<0.050 for each).

Table 2 . Imaging and anatomical data of the study sample for lower-extremity DVT.

Variable	Patients with lower-extremity DVT			P-value
	All patients (n=201)	Silent PE (n=98)	No PE (n=103)
Unilateral	181 (90.0)	87 (88.8)	94 (91.3)	0.198
Left side	139 (69.2)	67 (68.4)	72 (69.9)	0.248
Right side	62 (30.8)	31 (31.6)	31 (30.1)	0.239
Bilateral	20 (10.0)	11 (11.2)	9 (8.7)	0.184
Iliofemoral	122 (60.7)	65 (66.3)	57 (55.3)	0.037*
Femoropopliteal	60 (29.9)	28 (28.6)	32 (31.1)	0.203
Proximal DVT (iliofemoral and femoropopliteal)	182 (90.5)	93 (94.9)	89 (86.4)	0.012*
Distal DVT (calf)	19 (9.5)	5 (5.1)	14 (13.6)	0.009*
IVC thrombus extension	11 (5.5)	10 (10.2)	1 (1.0)	0.003*

Variables are presented as number (%)..

DVT, deep vein thrombosis; IVC, inferior vena cava; PE, pulmonary embolism..

*Statistically significant differences at P<0.05..

Table 3 . PE location by DVT extent.

DVT extent	PE location
	Main pulmonary	Lobar	Segmental	Subsegmental	Total
Iliofemoral	21	24	12	8	65
Femoropopliteal	8	7	6	7	28
Distal	0	1	2	2	5
Total	29	32	20	17	98

DVT, deep vein thrombosis; PE, pulmonary embolism..

Finally, to assess the predictive value of the various risk factors identified through above univariate analysis for the incidence of silent PE in patients with DVT, we performed a multivariate logistic regression with the occurrence of silent PE as the dependent dichotomous variable. The analysis revealed that male sex (odds ratio [OR]=2.46, 95% confidence interval [CI]: 1.39-3.53; P=0.026), cancer (OR=2.76, 95% CI: 1.45-4.07; P=0.017), previous VTE history (OR=2.67, 95% CI: 1.33-4.01; P=0.022), a D-dimer value ≥5.14 mg/L (OR=2.24, 95% CI: 1.10-3.38; P=0.033), iliofemoral DVT (OR=2.13, 95% CI: 1.19-3.07; P=0.041), and thrombus extension to the IVC (OR=2.95, 95% CI: 1.43-4.47; P=0.009) as independent predictors for silent PE in patients with lower-extremity DVT (Table 4). More specifically, multivariate logistic regression analyzing the risk of proximal silent PE identified cancer (OR=1.99, 95% CI: 1.22-2.76; P=0.045), previous VTE history (OR=2.13, 95% CI: 1.29-2.97; P=0.042), D-dimer value ≥5.14 mg/L (OR=2.69, 95% CI: 1.55-3.83; P=0.030), iliofemoral DVT (OR=2.72, 95% CI: 1.54-3.90; P=0.035), and thrombus extension to the IVC (OR=3.09, 95% CI: 1.78-4.40; P=0.007) as independent predictors for proximal silent PE in patients with lower-extremity DVT (Table 5).

Table 4 . Multivariate, binomial logistic regression analysis of risk factors for silent PE identified through univariate analysis in DVT patients.

Variable	Exp (B)	95% CI		P-value
		Lower	Upper
Male sex	2.46	1.39	3.53	0.026*
Smoking	1.89	0.89	2.89	0.101
Recent surgery	1.52	0.77	2.27	0.184
Cancer	2.76	1.45	4.07	0.017*
Previous VTE history	2.67	1.33	4.01	0.022*
PE Wells score >4	1.21	0.68	1.74	0.201
D-dimer ≥5.14 mg/L	2.24	1.10	3.38	0.033*
Iliofemoral DVT	2.13	1.19	3.07	0.041*
IVC thrombus extension	2.95	1.43	4.47	0.009*

CI, confidence interval; VTE, venous thromboembolism; PE, pulmonary embolism; DVT, deep vein thrombosis; IVC, inferior vena cava..

*Statistically significant differences at P<0.05..

Table 5 . Multivariate, binomial logistic regression analysis of risk factors for proximal silent PE (main pulmonary and lobar) identified through univariate analysis in patients with DVT.

Variable	Exp (B)	95% CI		P-value
		Lower	Upper
Male sex	1.23	0.46	2.00	0.081
Recent surgery	1.08	0.32	1.84	0.128
Cancer	1.99	1.22	2.76	0.045*
Previous VTE history	2.13	1.29	2.97	0.042*
D-dimer ≥5.14 mg/L	2.69	1.55	3.83	0.030*
Iliofemoral DVT	2.72	1.54	3.90	0.035*
IVC thrombus extension	3.09	1.78	4.40	0.007*

CI, confidence interval; VTE, venous thromboembolism; DVT, deep vein thrombosis; IVC, inferior vena cava..

*Statistically significant differences at P<0.05..

Occult PE is prevalent in patients with lower-extremity DVT [3]. Previous studies have reported that the incidence of silent PE in these patients ranges from 11% to 46% [4,5]. However, incidences as high as 66% to 72% have been reported when CTPA is used for routine screening [6]. In our study, the overall incidence of silent PE was 48.8%, indicating that nearly one in two patients was affected by PE at the time of DVT diagnosis. Moreover, the emboli were centrally located in the main pulmonary and lobar arteries in 62.2% of the cases. This non-negligible proportion of patients is at risk of developing chronic thromboembolic pulmonary hypertension (CTEPH) over the long term, which can have devastating consequences [12]. Of note, approximately 25% of patients with CTEPH do not report any episodes of symptomatic PE, highlighting the importance of silent PE as a risk factor [12].

Some studies suggested that untreated PE is associated with an in-hospital mortality rate of 30% [13]. This rate was even higher in patients with known pulmonary, cardiac, or oncological diseases [8,12]. Furthermore, studies have suggested that patients with DVT and silent PE are more likely to experience recurrent PE than those without silent PE [4]. More specifically, these patients face an increased risk of symptomatic PE during the initial 2 weeks of treatment, although no significant difference is noted after 3 months of treatment [14]. This applies also to cases of unprovoked proximal DVT, in which the estimated rate of recurrent VTE was 11% in the first year [15].

Finally, the knowledge of the presence of silent PE in DVT patients is crucial because subsequent pulmonary symptoms might be misdiagnosed as new PE episodes, despite adequate anticoagulation. This misdiagnosis can lead to unnecessary therapeutic measures, such as IVC filter insertion [16,17]. Although IVC filter placement is considered an alternative therapy for recurrent DVT/PE under full anticoagulation, most international guidelines remain skeptical about the benefits of IVC filter [3,18,19].

Therefore, routine screening for occult PE in newly diagnosed DVT patients has been advocated, considering the high incidence, potential severe cardiopulmonary consequences, and benefits of baseline imaging for individualizing treatment options if the patient subsequently develops respiratory symptoms [4,5,9]. However, this strategy remains controversial and is still under debate due to concerns over raised costs, radiation, and exposure to contrast media [9,10]. While CTPA is considered a non-invasive examination, the risks contrast-induced nephropathy and radiation exposure should be considered [6,8]. Alternatively, pulmonary ventilation and perfusion scintigraphy can be performed, but it offers lower sensitivity and specificity [6,8]. Moreover, the absence of high-quality evidence demonstrating the clinical and health-economic benefits of routine screenings has led the ESVS to recommend against routine screening for silent PE in newly diagnosed DVT patients in its recent guidelines (class III, level C) [3]. Despite this consensus recommendation, we maintain that the benefits of routine screening for silent PE in DVT patients outweigh the potential disadvantages.

Selected screening in high-risk DVT patients has been advocated as a reasonable alternative approach [11]. However, DVT characteristics that could serve as risk factors for silent PE have not been well documented or elucidated. In patients diagnosed with DVT, the incidence of silent PE increases with age and is higher in male patients and those with proximal DVT compared to those with calf DVT [20-22]. Moreover, previous studies have found that right-sided DVT is associated with an increased risk of silent PE, a finding attributed to possible left iliac vein compression [8,23]. In our study, a higher proportion of men was observed in the silent PE group, but there was no difference in mean age between the two groups. Smoking, recent surgery, cancer, and a previous history of VTE were more frequent in the silent PE group [24]. Unilateral lower-extremity DVT was observed in the majority of patients. However, there was no statistically significant difference in the incidence of silent PE according to the number or side of the affected limb.

In our study, silent PE was significantly more common on the right side, with the emboli located at the main pulmonary, lobar, segmental, and subsegmental arteries, accounting for 29.6%, 32.7%, 20.4%, and 17.3% of cases, respectively. A higher incidence of occult PE was observed in patients with iliofemoral DVT than in those with distal calf DVT, which is consistent with previous studies [4,8,25]. Moreover, iliofemoral DVT was associated with a larger size and a more proximal location (main pulmonary or lobar) of the embolus, whereas in cases with distal DVT, the segmental and subsegmental arteries were more commonly involved. Thus, in patients with iliofemoral DVT, recognizing the presence of silent PE through CTPA imaging is crucial, as these patients may benefit from individualized, long-term anticoagulation to mitigate the risk of long-term CTEPH [4,5,9]. Furthermore, the extension of the thrombus into the IVC was associated with an even higher incidence of occult PE than isolated lower-extremity DVT, confirming the results of previous studies and highlighting the significant risk faced by this patient subgroup [4,8,11]. Other factors associated with proximal silent PE, identified through univariate analysis, included male sex, recent surgery, cancer, a previous history of VTE, and elevated D-dimer levels.

Early induction of proper anticoagulant therapy prevents thrombus extension, thus reducing the incidence of silent PE in DVT patients [26]. This means that delayed admission and initiation of anticoagulant therapy could result in an increased incidence of PE [11,26]. However, our study showed that the delay from symptom onset to admission was not related to the incidence of silent PE, as the median delay was similar between the two groups. During clinical evaluations, screening for PE may be beneficial in certain subgroups of DVT patients, such as those presenting with ECG or CXR abnormalities, free-floating thrombi, or cardiac biomarkers indicative of potential pulmonary involvement [4,8,14]. In our study, a PE Wells score >4 on the initial clinical evaluation was marginally associated with a higher incidence of silent PE. D-dimer testing in the emergency department is associated with a high NPV and low PPV for the diagnosis of PE [3]. Our study demonstrate that patients with DVT who had silent PE exhibited significantly higher D-dimer levels than those without PE, consistent with previous studies [4,8]. Shi et al. [11] found a D-dimer level of 3.82 μg/mL (Sn: 76.5%, Sp: 42.1%) was discriminative of PE occurrence. In our study, a D-dimer value ≥5.14 mg/L (Sn: 68.2%, Sp: 59.3%) was discriminant in predicting silent PE in DVT patients, with a PPV of 61.4% and a NPV of 66.2%. The different D-dimer assay methods and corresponding cut-offs used in each study may explain these differences [11].

Finally, our multivariate logistic regression analysis identified male sex, cancer, previous VTE history, D-dimer value ≥5.14 mg/L, iliofemoral DVT, and thrombus extension to the IVC as significant independent predictors of silent PE in patients with lower-extremity DVT. More specifically, cancer, previous VTE history, D-dimer value ≥5.14 mg/L, iliofemoral DVT, and thrombus extension to the IVC were also found, according to the multivariate analysis, to serve as independent predictors of a proximal PE location. Therefore, we believe that selective screening for silent PE in high-risk DVT patients may be worthwhile, given the high incidence of occult PE and the potential for severe cardiopulmonary consequences, despite the additional cost, radiation exposure, and risk of contrast-induced nephropathy. However, even with this selective approach, Level I evidence is still lacking. Consequently, we cannot recommend reformulating ESVS guidelines for routine or selective silent PE screening in DVT patients simply based on this study. Nevertheless, we firmly believe that our study provides important insights into this issue and contributes to shifting the current trend in favor of this diagnostic approach.

Our study has several limitations. This was a single-center, retrospective, observational study with a relatively small sample size. Moreover, this study was limited by its retrospective design, selection bias, and confounding factors, which may have compromised the results. A multicenter prospective cohort study with a larger sample size and longer follow-up duration is needed to fully assess the clinical impact of silent PE on subsequent cardiopulmonary function. Furthermore, we did not analyze the cost-effectiveness of CTPA screening for silent PE. A cost-effectiveness analysis is essential for evaluating the clinical and health-economic importance of routine screening of all DVT patients, in order to detect possible underlying occult PE.

There is a high incidence of asymptomatic silent PE in patients with lower-extremity DVT. Male sex, cancer, previous VTE history, D-dimer levels, iliofemoral DVT, and thrombus extension into the IVC were significant independent predictors of occult PE. Screening DVT patients who exhibit these predictive risk factors with CTPA may be necessary and justified for efficient management of VTE and its long-term complications. Further multicenter, prospective studies on clinical outcomes and cost-effectiveness are essential to justify this screening practice.

None.

The authors have nothing to disclose.

Concept and design: DAC, GAP. Analysis and interpretation: DAC, Apostolos G. Pitoulias, GAP. Data collection: DAC, TNZ, GVT. Writing the article: DAC. Critical revision of the article: all authors. Final approval of the article: all authors. Statistical analysis: DAC. Obtained funding: none. Overall responsibility: GAP.