Concomitant Lower-Extremity Deep Vein Thrombosis in Patients With Pulmonary Embolism Undergoing Catheter-Directed Therapy

Background. Deep vein thrombosis (DVT) is often seen in patients with acute pulmonary embolism (PE). Risk stratification of PE patients is useful in predicting mortality risk and hospital course. However, rates or predictors of DVT or proximal DVT (popliteal, femoral, common femoral, or iliac thrombosis) have not been studied in the highest-risk patients who receive catheter-directed therapy (CDT) for their PE. A single-center retrospective analysis of patients referred for CDT for confirmed PE was conducted to evaluate rates and predictors of DVT or proximal DVT and the impact on short-term outcomes. In 137 consecutive patients undergoing CDT for PE with available lower-extremity ultrasound, the rates of DVT and proximal DVT in PE patients receiving CDT were 76.6% and 65.0%, respectively. Rates of DVT (P=.68) and proximal DVT (P=.72) did not differ between high-risk or non-high risk PE patients. The only significant factor associated with presence of concomitant DVT was previous DVT (P=.045). The presence of a concomitant DVT or proximal DVT was not associated with an increase in all-cause mortality or hospitalization at 30 days or 1 year compared with an absence of concomitant DVT or proximal DVT. The results of this study suggest that patients with PE clinically requiring CDT have high rates of concomitant DVT and proximal DVT, prior DVT predicts concomitant DVT, and the presence of DVT is not associated with additional risk in this already high-risk population of patients.

J INVASIVE CARDIOL 2021;33(11):E910-E915.

Key words: catheter-directed therapy, deep vein thrombosis, pulmonary embolism

Venous thromboembolism (VTE) represents pathological thrombosis inclusive of deep vein thrombosis (DVT) and pulmonary embolism (PE). Previous research has shown that more than 90% of PE originates from a lower-extremity DVT.¹ Although DVT or PE can occur separately, they often occur concomitantly, with DVT preceding PE. The rate of concomitant PE and residual DVT is variable, and is estimated to be between 40%-70%.^2-5 Reported risk factors for concomitant PE and DVT include age,¹ active malignancy,^6,7 and male gender,^2,6,7 although controversy exists.^3-8 So far, no studies have examined rates of DVT or proximal DVT or factors associated with concomitant DVT in patients deemed to be high enough risk to receive catheter-directed therapy (CDT) for PE. CDT included Angiojet embolectomy (Boston Scientific), EKOS thrombolysis (Boston Scientific), Penumbra aspiration (Penumbra), and Cragg-McNamara thrombolysis (Medtronic).

Classification of PE severity in patients is important to guide clinical decision making. The European Society of Cardiology (ESC) set out guidelines in 2019 to classify PE severity based on 4 parameters — hemodynamic instability, pulmonary embolism severity index (PESI) class, right ventricular (RV) dysfunction, and elevated cardiac troponin) — where higher-risk PE is associated with a greater mortality risk, and non-high risk PE is an indication for early discharge and home anticoagulation treatment.⁹ Recent advances in invasive therapies have shown improvement in RV function in intermediate-high and high-risk PE patients who undergo invasive therapy for PE.^10-13 In a retrospective analysis of normotensive PE patients, concomitant DVT identified intermediate-low risk PE patients who have outcomes more similar to intermediate-high risk PE patients.¹⁴ In all-comer registries, PE patients with concomitant DVT have been shown to have higher 30-day all-cause mortality,¹⁵ as well as higher recurrent VTE than patients with PE alone.⁶ It is unclear whether intervention with inferior vena cava (IVC) filter placement at the time of CDT for PE would improve outcomes in these patients. This study sought to determine the prevalence and risk factors of concomitant DVT in patients with PE requiring CDT.

The objective of this study was to determine the prevalence and risk factors of concomitant DVT and proximal DVT (popliteal, femoral, or iliac vein) in patients with PE who underwent CDT for their PE. To achieve this, a single-center retrospective chart review of patients with PE and a lower-extremity Doppler scan over a 10-year period (July 2010 to July 2020) was conducted. Identifying risk factors can help to identify patients who may benefit from further protection from recurrent PE with an IVC filter

Study design. A retrospective chart analysis over a 10-year period of patients (age ≥18 years) at a tertiary medical center who received CDT for their PE and a lower-extremity Doppler scan was performed from July 2010 to July 2020 (not all patients who received CDT for their PE received a lower-extremity Doppler scan). Study design was approved by the hospital’s institutional review board.

Clinical variables. A retrospective analysis of the electronic medical records was performed to determine demographic and clinical characteristics. Laboratory findings of serum creatinine, troponin, and N-terminal B-type natriuretic peptide were recorded. Measurements of left ventricular (LV)/right ventricular (RV) ratio on the computed tomography (CT) scan and preprocedural echocardiogram were performed. Procedural characteristics of the CDT were determined by the procedural report and corresponding angiograms. All-cause mortality and time to death were determined through chart review. If information was not available in the chart, patients or their family were contacted.

PE risk classification. Patients were classified into high, intermediate-high, intermediate-low, or low PE risk groups based on 2019 ESC guidelines.⁹Hemodynamic instability was defined as per the 2019 ESC guidelines. RV dysfunction was defined by either RV/LV ratio >1.0 on CT (or echocardiography) and/or RV dysfunction as per transthoracic echocardiogram. Elevated troponin levels were defined by troponin T >0.03 ng/mL or a high-sensitivity troponin T >14 ng/L. PESI was calculated according to a validated scoring system.¹⁶ For analyses, high and intermediate-high risk PEs were combined into the high-risk PE subgroup, and intermediate-low and low were combined into the non-high risk PE subgroup, as seen in Figure 1. This grouping was performed with deference to the clinical and therapeutic similarities between these respective patient clusters.

Statistical analysis. Descriptive statistics were used to summarize baseline clinical characteristics. The Chi-squared or Fisher’s exact test was used for comparison between categorical variables when comparing patients with presence of concomitant DVT vs absence of concomitant DVT and patients with presence of concomitant proximal DVT vs absence of concomitant proximal DVT. P-values <.05 were considered significant. For survival analysis, the Kaplan-Meier estimator was used to determine time-to-event and the log-rank test was used to compare probability of the combined endpoint of all-cause mortality or hospitalization between high-risk PE and non-high risk PE or concomitant DVT and no concomitant DVT. Statistical analyses were performed using Stata/MP, version 15.1 (StataCorp LP).

Of the 189 PE patients who underwent CDT at our institution during the study period, 137 underwent both lower-extremity Doppler scan and CDT. Mean age was 58.1 ± 14.9 years and 46.0% were men. A total of 94 patients (68.6%) had a high-risk PE. Table 1 shows other study population characteristics. Of these patients, 105 (76.6%) had a DVT and 89 (65.0%) had a proximal DVT. High-risk PE was not significantly associated with a higher rate of DVT (P=.68) or proximal DVT (P=.72). Factors associated with either a higher prevalence of concomitant DVT or proximal DVT are summarized in Table 2 and Table 3. Only a history of DVT was significantly associated with a higher rate of concomitant DVT in PE patients. No factors analyzed were significantly associated with a higher rate of concomitant proximal DVT in PE patients.

Kaplan-Meier analysis comparing the primary outcome of all-cause mortality or hospitalization at 30 days or 1 year between high-risk and non-high risk PE patients (Figure 2A and Figure 3A), patients with concomitant DVT or no concomitant DVT (Figure 2B and Figure 3B), and patients with concomitant proximal DVT or no concomitant proximal DVT (Figure 2C and Figure 3C) was performed. High-risk PE was found to be associated with higher mortality or hospitalization at 30 days (P<.01) and at 1 year (P=.01) vs non-high risk PE. Outcomes at 30 days or 1 year were not statistically different in patients with concomitant DVT vs patients without concomitant DVT (P=.14 at 30 days, P=.31 at 1 year) or patients with concomitant proximal DVT vs patients without concomitant proximal DVT (P=.12 at 30 days, P=.08 at 1 year).

In the study population, 59 patients (43.1%) received an IVC filter during the same presentation as their PE. The indications for IVC filter placement were known DVT in the setting of higher-risk PE in 57 patients (96.6%) and empiric therapy for probable PE on higher-risk PE in 2 patients (3.4%). Of the patients who received an IVC filter, 27 (45.7%) had their IVC filter removed. The average time to removal of their IVC filter placement was 337 ± 339 days. Reasons for inability to have the filter removed are summarized in Table 4.

To our knowledge, this is the first study to examine the rates of concomitant DVT in PE patients receiving CDT. In this selected population with 68.6% classified as high-risk PE, rates of concomitant DVT and proximal DVT were 76.6% and 65.0%, respectively. Only prior DVT is a predictor of concomitant DVT and PE, and although there is higher 30-day and 1-year primary outcome of all-cause death or all-cause hospitalization in patients with high-risk PE, 30-day and 1-year outcome of patients with concomitant DVT and PE is not statistically different compared with patients with PE alone.

A review by Becatinni et al¹⁵ found that only 56% of all PE patients of varying risk had concomitant DVT. The higher percentage of patients presenting with concomitant DVT in our study could be due to the fact that our patients were high enough risk to be deemed candidates for CDT. These patients may have higher overall DVT clot burden that partially embolize compared with patients with smaller or lower-risk PE. Since only patients with a lower-extremity ultrasound were included in the present study, it is possible that the true rate of concomitant DVT in PE patients who receive CDT is lower than what was found, as the patients in this study had a clinical indication for a lower-extremity Doppler ultrasound.

Among the factors analyzed, only a history of prior DVT was significantly associated with a higher rate of concomitant DVT in PE patients (Table 2). None of the factors analyzed were associated with a higher rate of concomitant proximal DVT (Table 3). This study might have been under-powered to determine other risk factors associated with DVT; however, it is also possible that there are no significant factors associated with concomitant proximal DVT in the high-risk cohort of patients who receive CDT. In this population, high-risk PE was not associated with a higher rate of concomitant DVT or proximal DVT, in contrast with a prospective study by Cordeanu et al,¹⁷ which found that concomitant DVT and proximal DVT were associated with higher-risk PE in a population. Unlike Cordeanu et al,¹⁷ this study included selected patients who received CDT.

High-risk PE was associated with higher all-cause mortality and hospitalization at 30 days (Figure 2A) and 1 year (Figure 3A).This is expected, as high-risk PE patients have been shown to have higher mortality rates than non-high risk PE patients.¹⁸ Additionally, neither concomitant DVT nor concomitant proximal DVT was associated with higher all-cause mortality and hospitalization at 30 days (Figure 2B and Figure 2C) and 1 year (Figure 3B and Figure 3C). In our study population, there was a trend toward improved outcomes in patients without concomitant DVT, but this result was not significant. Previous research has been conflicted on whether concomitant DVT increases mortality. A study by Cordeanu et al¹⁷ did not find a significant increase in mortality associated with concomitant DVT, while Becatinni et al¹⁵ found that concomitant DVT was associated with a significantly associated mortality at 30 days. An explanation for why concomitant DVT is not associated with higher mortality in patients who receive CDT for their PE is that these patients are higher risk from the outset. Any added mortality risk that might be associated with concomitant DVT is small compared with the factors that predispose someone to receive CDT rather than oral anticoagulants. Additionally, people who receive CDT might have had more of their initial thrombus that embolized to the lung. This could explain why patients with no concomitant DVT have lower survival rates, as the clot that embolizes to their lungs could be larger than in patients who have a concomitant DVT with their PE.

IVC filter placement can easily be performed as a concomitant procedure along with CDT as a way to prevent recurrent PE due to DVT; however, IVC filter usage in patients with concomitant PE and DVT is controversial.^19-24 Despite the fact that there was a high rate of high-risk PE in patients undergoing CDT in our analysis, and that most of these patients had DVT, there was no difference in mortality between those with and without DVT. This raises the question about the necessity of IVC filters in these patients. Our findings are in contrast with a recent meta-analysis, which found that mortality in all-comer PE patients in clinical trials and case series receiving IVC filter was significantly lower than PE patients who do not receive an IVC filter; however, in this meta-analysis, only 43.8% of patients underwent CDT.²⁵ In addition, in our study population, only 45.7% of patients who received an IVC filter had it subsequently removed at an average of 337 days later. Despite the fact that our institution has a well-developed tracking system for patients receiving IVC filters, the actual rate of and time to IVC retrieval in these patients is not ideal. The unclear benefit of preventing DVT mobilization as well as the long time to IVC filter removal suggest that IVC filter placement in these patients should only be used in selected cases.

Study limitations. This study has several limitations. Since the study is retrospective, it is possible that not all of the relevant information was available for analysis. The single-center nature of the study could lead to institutional bias in deciding which individuals should receive CDT for their PE, or which patients should receive a lower-extremity Doppler ultrasound. Finally, the small size of the population limits the statistical power of the study.

In patients receiving CDT for their PE, only a history of DVT is associated with a higher frequency of concomitant DVT. No factors were found that had a significant relationship with a higher rate of proximal DVT. The presence of concomitant DVT or proximal DVT did not affect all-cause mortality or all-cause hospitalization at 30 days or 1 year.