Risk factors and treatment interventions associated with incomplete thrombus resolution and pulmonary hypertension after pulmonary embolism

Abstract

Background

Residual pulmonary vascular occlusion (RPVO) affects one half of patients after a pulmonary embolism (PE). The relationship between the risk factors and therapeutic interventions for the development of RPVO and chronic thromboembolic pulmonary hypertension is unknown.

Methods

This retrospective review included PE patients within a 26-month period who had baseline and follow-up imaging studies (ie, computed tomography [CT], ventilation/perfusion scans, transthoracic echocardiography) available. We collected the incidence of RPVO, percentage of pulmonary artery occlusion (%PAO), baseline CT %PAO, most recent CT %PAO, and difference between the baseline and most recent %PAO on CT (Δ%PAO).

Results

A total of 354 patients had imaging reports available; 197 with CT and 315 with transthoracic echocardiography. The median follow-up time was 144 days (interquartile range [IQR], 102-186 days). RPVO was present in 38.9% of the 354 patients. The median Δ%PAO was −10.0% (IQR, −32% to −1.2%). Fewer patients with a provoked PE developed RPVO (P ≤ .01), and the initial troponin level was lower in patients who developed RPVO (P = .03). The initial thrombus was larger in the patients who received advanced intervention vs anticoagulation (baseline CT %PAO: median, 61.2%; [IQR, 27.5%-75.0%] vs median, 12.5% [IQR, 2.5%-40.0%]; P ≤ .0001). Catheter-directed thrombolysis (CDT; median Δ%PAO, −47.5%; IQR, −63.7% to −8.7%) and surgical pulmonary embolectomy (SPE; median Δ%PAO, −42.5; IQR, −68.1% to −18.7%) had the largest thrombus reduction compared with anticoagulation (P = .01). Of the 354 patients, 76 developed pulmonary hypertension; however, only 14 received pulmonary hypertension medications and 12 underwent pulmonary thromboendarterectomy. Cancer (odds ratio [OR], 1.7) and planned prolonged anticoagulation (>1 year; OR, 2.20) increased the risk of RPVO. In contrast, the risk was lower for men (OR, 0.61), patients with recent surgery (OR, 0.33), and patients treated with SPE (OR, 0.42). A larger Δ%PAO was found in men (coefficient, −8.94), patients with a lower body mass index (coefficient, −0.66), patients treated with CDT (coefficient, −18.12), and patients treated with SPE (coefficient, −21.69). A lower Δ%PAO was found in African-American patients (coefficient, 7.31).

Conclusions

The use of CDT and SPE showed long-term benefit in thrombus reduction.

Article Highlights.

• •Type of Research: A single-center, retrospective cohort study
• •Key Findings: In our study, the incidence of residual pulmonary vascular occlusion was 39% after a pulmonary embolism. Catheter-directed thrombolysis (median, −47.5%; interquartile range, −63.7% to −8.7%) and pulmonary embolectomy (median, −42.5%; interquartile range, −68.1% to −18.7%) showed greater long-term thrombus reduction. A provoked pulmonary embolism and male sex were associated with a higher rate of thrombus reduction. Malignancy, female sex, and African-American race were associated with a lower rate of thrombus reduction.
• •Take Home Message: The use of catheter-directed thrombolysis and surgical pulmonary embolectomy showed long-term benefits in thrombus reduction.

After the diagnosis of pulmonary embolism (PE), despite the use of anticoagulation therapy, complete resolution of the pulmonary thrombus does not occur in all the patients. Nearly one half of the patients still develop residual pulmonary artery thrombosis. This condition has been described as residual pulmonary vascular occlusion (RPVO).¹ Previous reports have defined RPVO as persistent pulmonary vascular obstruction found on imaging studies after 3 to 6 months of anticoagulation monotherapy. The presence of RPVO is not always associated with symptoms such as dyspnea or exercise limitations. However, the presence of RPVO is also associated with an increased risk of recurrent PE and chronic thromboembolic pulmonary hypertension (CTEPH), which negatively affects patients' quality of life.² Multiple factors can contribute to the development of RPVO, such as thrombus age at anticoagulation initiation, the presence of comorbidities that affect cellular immunity, thrombus reabsorption (ie, malignancy), and pulmonary artery vasoreactivity. However, no direct link has been found between these mechanisms and the development of RPVO. Although older age and a history of unprovoked PE are known risk factors, a more definitive mechanism of RPVO has not yet been established.²

Little to no evidence is available to suggest that catheter-based interventions or surgical pulmonary embolectomy (SPE) can reduce the incidence of RPVO or CTEPH in the long term. Furthermore, the type and duration of anticoagulation have not been shown to influence these outcomes. The purpose of the present study is to evaluate the risk factors and therapeutic approaches in the development of RPVO and CTEPH after acute PE.

Methods

Study design and participants

Patients with a diagnosis of acute PE by computed tomography (CT), ventilation/perfusion (V/Q) scan, or pulmonary angiography and with at least one imaging follow-up study (CT of the chest or V/Q scan) to assess for RPVO after discharge were included. A history of a prior PE did not exclude the patient from the study. A qualitative analysis was performed using study reports (no images) to categorize patients by the presence or absence of RPVO. In addition, a quantitative analysis using the percentage of pulmonary artery occlusion (%PAO) was performed of patients with both baseline and follow-up CT images available. An analysis of pulmonary artery pressure was performed for all patients with follow-up transthoracic echocardiography (TTE) and a reported right ventricular (RV) systolic pressure (RVSP).

Pulmonary artery thrombus burden

On the admission or baseline CT scan, the PE clot burden was described as saddle, segmental, or subsegmental.³ The presence of RV strain was also noted on the CT scans.⁴ V/P scan findings were considered positive if they showed a high probability for PE.

To identify RPVO, the reports of the most recent CT or V/Q scans were reviewed. Any residual thrombosis on the most recent CT scan or any mismatch defect on the most recent V/Q scan was considered positive for RPVO.

The Qanadli index (QI) was used to quantify the %PAO. The QI divides each pulmonary artery (right and left) into 10 segments. Thrombus in the most proximal arterial level is scored with a value equal to the number of segmental arteries arising from that artery. The degree of obstruction is scored as follows: 0, no thrombus; 1, partially occlusive thrombus; and 2, totally occlusive thrombus. The maximum possible QI is 40. The percentage of obstruction is calculated using the following formula: %PAO = score/40 × 100.⁵ The following definitions were established: %PAO on the baseline CT is identified as α%PAO, %PAO in the most recent CT is identified as Ω%PAO, and the difference between the Ω%PAO and α%PAO is identified as Δ%PAO. The CT and V/Q scans were read by the on-call radiologists. The QI was calculated using both the initial and the follow-up CT scans. Two independent reviewers evaluated the CT images and reports for corresponding accuracy and consistency. The first reading of the CT scans was performed by the radiologist on-call at the time of hospitalization or outpatient follow-up. The second reading was performed by the study radiologist (A.F.), who confirmed the findings of the first reading and calculated the QI for each CT scan.

TEE scans

TTE scans were performed during admission after the diagnosis of PE and at the follow-up appointments. The protocol included parasternal long-axis and short-axis views to visualize the aortic, pulmonic, tricuspid, and mitral valves. The RV inflow view was used to interrogate the tricuspid valve and document tricuspid regurgitation for estimation of the RVSP. We obtained tricuspid annular plane systolic excursion measurements to evaluate RV systolic function. Each echocardiogram was performed by a registered echocardiography technician and interpreted by a board-certified physician in echocardiography. The interpreting cardiologist estimated the RV size (normal or mildly, moderately, or severely dilated) and function (normal or mildly, moderately, or severely decreased). RVSP is reported in mm Hg, and pulmonary pressure is defined as normal (<35 mm Hg), mildly elevated (35-45 mm Hg), moderately elevated (46-60 mm Hg), or severely elevated (>60 mm Hg).

Catheter-based interventions and SPE

The catheter-based intervention (CBI) group included all patients who underwent catheter-directed thrombolysis (CDT) as the main therapy for acute PE. All the procedures in the CBI group were performed using the EkoSonic Endovascular System (Boston Scientific). No patients who underwent endovascular mechanical thrombectomy were included in this study (owing to a lack of follow-up images). SPE was performed via median sternotomy, and all the patients were placed under cardiopulmonary bypass with mild hypothermia. Central aortic and bicaval venous cannulation was used. Incisions were made in the right and left pulmonary arteries. Thrombus was removed in its entirety, up to the subsegmental level using a combination of forceps and suction thrombectomy. Intermittent, 15- to 30-second circulatory arrest periods were initiated as necessary. The operation was routinely performed on a beating heart without placement of an aortic cross-clamp.

Outcomes

The primary outcome is to identify the risk factors and therapeutic interventions associated with the presence of RPVO (qualitative), Δ%PAO (quantitative), and the development of CTEPH.

Statistical analysis

Categorical variables are presented as frequencies and percentages. Continuous variables are presented as the mean ± standard deviation for normally distributed variables and the median and interquartile range (IQR) for non-normally distributed variables. To assess the differences between the groups, the Pearson χ² test, Fisher exact test, Student t test, or Mann-Whitney U test was used.

A multivariable analysis was performed to evaluate the risk factors (ie, age, sex, body mass index [BMI], race, history of cancer, recent surgery) and therapeutic approaches (ie, time between diagnosis and follow-up images, use of CDT, systemic thrombolytic agents, SPE, extracorporeal membrane oxygenation [ECMO], duration of anticoagulation [>1 year vs <1 year]) associated with the presence of RPVO and Δ%PAO. Logistic regression was used to evaluate the relationship of risk factors and therapeutic interventions associated with the presence of RPVO and CTEPH. Linear regression was used for Δ%PAO. A two-tailed P value of < .05 was considered statistically significant.

To create a Kaplan-Meier curve for thrombus resolution over time, we used the date of the acute PE diagnosis and the date of the most recent CT scan or VQ scan. The primary outcome was defined as the absence of RPVO. Statistical analysis was performed using MedCalc software.

Results

A total of 760 patients were diagnosed with acute PE during the study period. Only 354 patients had at least one follow-up imaging report available, including 6 VQ scans and 348 CT scans. The median time between the diagnosis and the most recent image was 144 days (IQR, 102-186 days). Per the radiology reports, RPVO was present in 134 patients (38.9%), identified on 3 VQ scans and 131 CT scans. Of the 354 patients, only 197 had both baseline and recent CT images available for review. Of these 197 patients, the median α%PAO was 20% (IQR, 5%-52%) and the median Ω%PAO was 0% (IQR, 0%-15%; P = .0001; Fig 1). The median Δ%PAO was −10.0% (IQR, −32% to −1.2%; range, −92.5 to 67.5). Complete thrombus resolution (Ω%PAO, 0%) was observed in 135 patients (68.5%), with Ω%PAO ≥1% in 62 patients (31.4%).

Fig 1.

Box and whisker graph of the difference in the median percentage of obstruction for the 197 patients with baseline and follow-up computed tomography (CT) pulmonary angiography data available. The percentage of pulmonary artery occlusion (%PAO) on baseline CT scan was 20% (interquartile range [IQR], 5%-52%) vs %PAO on the most recent CT scan (median, 0%; IQR, 0%-15%; P = .0001).

The demographic characteristics of the 354 patients (220 patients with no RPVO vs 134 with RPVO) are similar, although the proportion of RPVO was lower in patients with recent trauma (13.2% vs 5.3%; P = .01) or surgery within 90 days (36.8% vs 17.2%; P = .0001; Table I).

Table I.

Baseline patient characteristics

Demographic and clinical characteristics	Complete resolution (n = 220)	Incomplete resolution (n = 134)	P value
Mean age, years	53.5 ± 17.2	54 ± 15.8	.6
Male sex	108 (49.1)	73 (54.5)	.32
Race
White	118 (53.6)	60 (44.8)	.1
African American	96 (43.6)	68 (50.7)	.19
Other	6 (2.7)	6 (4.5)	.37
BMI, kg/m2	31.0 ± 8.1	30.7 ± 8.0	.69
Coronary artery disease	30 (13.7)	17 (12.8)	.8
Hypertension	123 (55.9)	67 (50.0)	.28
Diabetes mellitus types 1 and 2	51 (24.8)	22 (16.9)	.09
Chronic kidney diseasea	14 (6.5)	10 (7.6)	.71
Current smoker	44 (20.0)	29 (21.6)	.47
Pregnancy	3 (1.4)	1 (0.7)	.56
Trauma within 90 days	29 (13.2)	7 (5.3)	.01
Thrombophiliab	14 (6.4)	13 (9.7)	.25
Surgery within 90 days	81 (36.8)	23 (17.2)	.0001
Family history of VTE	12 (5.5)	7 (5.2)	.92
History or active cancer	49 (22.4)	41 (30.8)	.07

BMI, Body mass index; VTE, venous thromboembolism.

Data presented as mean ± standard deviation or number (%).

Boldface P values represent statistical significance.

^aDefined as glomerular filtration rate <60 mL/min.

^bFactor V Leiden; prothrombin gene mutation; deficit of protein C, S, or antithrombin; antiphospholipid antibody syndrome.

On initial presentation, no differences were seen in the clinical classification of PE (ie, massive, submassive, low risk) or thrombus location (ie, saddle, segmental, subsegmental) between the patients with and without RPVO. The troponin level on admission was higher in the patients without RPVO than in those with RPVO (median, 0.05 ng/mL; [IQR, 0.02-0.17 ng/mL] vs median, 0.02 ng/mL [IQR, 0.02-0.11 ng/mL]; P = .03). Among the 197 patients with follow-up images available, the median Ω%PAO was larger in the patients with RPVO than in those without RPVO (median, 16.2% [IQR, 0%-47.5%] vs median, 0% [IQR, 0%-0%]; P ≤ .0001); the Δ%PAO was larger in the patients without RPVO (median, −10.0 [IQR, −40.0 to −2.5] vs median, −7.5 [IQR, −20 to 0]; P = .05; Table II).

Table II.

Initial clinical, radiologic, and biomarker characteristics (n = 354)

Characteristic	Complete resolution (n = 220)	Incomplete resolution (n = 134)	P value
Clinicala
Massive	24 (10.9)	15 (11.2)	.93
Submassive	94 (42.7)	50 (37.3)	.31
Low risk	102 (46.4)	69 (51.5)	.34
Radiologic (CTPA)
Saddle	29 (13.2)	24 (17.9)	.22
Segmental	114 (52.3)	76 (56.7)	.41
Subsegmental	75 (34.4)	34 (25.4)	.07
RV strain on CTPA	54 (30.2)	42 (37.2)	.21
α%PAOb	10.0 (2.5-51.2)	23.7 (5.0-50.0)	.11
Ω%PAOb	0 (0-0)	16.2 (0-47.5)	< .0001
Δ%PAOb	−10.0 (−40.0 to −2.5)	−7.5 (−20 to 0)	.05
Biomarker
Troponin, ng/mL	0.05 (0.02-0.17)	0.02 (0.02-0.11)	.03
Lactate, mmol/L	1.8 (1.3-2.6)	1.7 (1.3-2.8)	.96
ProBNP, pg/mL	1730 (433-3720)	658 (144-5202)	.18

α%PAO, Baseline percentage of pulmonary artery occlusion on computed tomography; Δ%PAO, difference between baseline and most recent percentage of pulmonary artery occlusion on computed tomography; Ω%PAO, most recent percentage of pulmonary artery occlusion on computed tomography; CTPA, computed tomography pulmonary angiography; ProBNP, pro B-type brain natriuretic peptide; RV, right ventricular.

Data presented as number (%) or median (interquartile range).

Boldface P values represent statistical significance.

^aDefined by the American Heart Association as massive, submassive, and low risk.

^bFrom 195 computed tomography images only.

Only 66 of the 354 patients required either one or a combination of advanced therapies: 22 patients received ECMO, 35 received SPE, 7 received systemic thrombolytic agents, and 14 received CDT. Among the 14 patients treated with CDT, the average thrombolytic dose infused through the catheters was 28.87 ± 17.67 mg of tissue plasminogen activator, and the average infusion time was 37.74 ± 17.88 hours (range, 21-72 hours). No differences were found in the presence of RPVO based on advanced treatment modality. Patients prescribed anticoagulation for <1 year had less frequently developed RPVO (P = .01) than those prescribed anticoagulation for >1 year (P = .0004; Table III). Among the 195 patients with CT images available, those who received advanced therapies had a larger α%PAO than those treated with anticoagulation alone (median, 61.2% [IQR, 27.5%-75.0%] vs median, 12.5% [IQR, 2.5%-40.0%]; P ≤ .0001). Therefore, the Δ%PAO was calculated to be greater in the patients who received advanced therapies compared with those treated with anticoagulation only (median, −57.5% [IQR, −65.0% to −7.5%] vs median, −7.5% [IQR, −20.0% to 0%]; P = .0003). The median α%PAO, Ω%PAO, and Δ%PAO for each treatment modality is shown in Table IV. The rate of RPVO occurrence over time is described in Fig 2.

Table III.

Initial and long-te rm therapy (n = 354)

Variable	Complete resolution (n = 220)	Incomplete resolution (n = 134)	P value
Initial therapy
Any advanced therapy	42 (19.1)	24 (17.9)	.78
ECMO	11 (5.0)	11 (8.2)	.22
SPE	25 (11.4)	10 (7.5)	.23
Systemic thrombolysis	2 (0.9)	5 (3.7)	.06
CDT	8 (3.6)	6 (4.5)	.69
Anticoagulation only	161 (73.2)	96 (71.6)	.75
No therapya	18 (8.2)	14 (10.4)	.47
Anticoagulation duration
<1 year	54 (24.5)	19 (14.2)	.01
>1 year	47 (21.4)	52 (38.8)	.0004
Not described	108 (49.1)	57 (42.5)	.23
No anticoagulation	11 (5.0)	6 (4.5)	.82

CDT, Catheter-directed thrombolysis; ECMO, extracorporeal membrane oxygenation; SPE, surgical pulmonary embolectomy.

Data presented as number (%).

Boldface P values represent statistical significance.

^aThe patient was not eligible for anticoagulation or any other advanced therapy because of absolute contraindications.

Table IV.

Percentage of pulmonary artery occlusion at baseline and most recent computed tomography stratified by treatment modality (n = 197)

Treatment modality	α%PAO	Ω%PAO	Δ%PAO	P Value
Anticoagulation alone	12.5 (2.5-40.0)	0 (0-5.0)	−7.5 (−20 to 0)	<.0001
All advanced interventions	61.2 (27.5-75.0)	0 (0-33.7)	−38.7 (−65.0 to −7.5)	.0001
CDT	60.0 (35.0-67.5)	10 (0-27.5)	−47.5 (−63.7 to −8.7)	.01
SPE	57.5 (0-75.0)	0 (0-33.1)	−42.5 (−68.1 to −18.7)	.01
Systemic thrombolysis	65.0 (52.5-81.8)	41.2 (0-86.2)	−31.2 (−78.7 to 12.5)	.56
VA-ECMO	72.5 (54.3-79.3)	17.5 (0-70.0)	−32.5 (−65.0 to −5.0)	.01

α%PAO, Baseline percentage of pulmonary artery occlusion on computed tomography; Δ%PAO, difference between baseline and most recent percentage of pulmonary artery occlusion on computed tomography; Ω%PAO, most recent percentage of pulmonary artery occlusion on computed tomography; CDT, catheter-directed thrombolysis; SPE, surgical pulmonary embolectomy; VA-ECMO, venous-arterial extracorporeal membrane oxygenation.

Data presented as median (interquartile range).

Boldface P values represent statistical significance.

Fig 2.

Kaplan-Meier curve of thrombus resolution over time.

Of the 354 patients, 315 had undergone follow-up TTE. The median time between discharge and the most recent follow-up TTE was 204 days (IQR, 32-914 days). Of the 315 patients, 76 (24.1%) were found to have an elevated estimated RVSP on their most recent TTE. Only 14 patients (4.4%) were prescribed pulmonary vasodilator medications at any point after discharge (9 received sildenafil or tadalafil, and 5 riociguat or macitentan). Twelve patients (3.8%) later underwent pulmonary thromboendarterectomy, generally performed shortly after discharge (median, 16 days; IQR, 8-34 days).

Among the 354 patients with imaging reports, logistic regression analysis found that the diagnosis of cancer increased the risk of RPVO (odds ratio [OR], 1.74; 95% confidence interval [CI], 1.02-2.97; P = .04). However, this risk was lower for men (OR, 0.61; 95% CI, 0.38-0.99; P = .04) and those with a history of surgery within 90 days (OR, 0.36; 95% CI, 0.20-0.63; P = .0001). Logistic regression analysis also found that both a longer time between the baseline and most recent images (OR, 0.99; 95% CI, 0.998-0.999; P = .02) and SPE (OR, 0.42; 95% CI, 0.18-0.99; P = .04) were associated with a lower risk of RPVO. Patients prescribed with anticoagulation for >1 year had a higher risk of RPVO (OR, 2.20; 95% CI, 1.29-3.75; P = .03). The results for the logistic regression analyses are reported in Table V.

Table V.

Logistic regression for presence of residual pulmonary vascular occlusion stratified by risk factors and treatment modality (n = 354)

Variable	OR	95% CI	P value
Risk factor
Age	0.99	0.98-1.01	.98
Male sex	0.61	0.38-0.99	.04
Body mass index	1.00	0.97-1.03	.70
African-American race	1.34	0.85-2.13	.20
History or active cancer	1.74	1.02-2.97	.04
Recent surgery (≤90 days)	0.36	0.20-0.63	.0001
Recent trauma (≤90 days)	0.48	0.19-1.19	.11
Treatment modality
Time between images	0.99	0.99-0.99	.02
CDT	1.28	0.40-4.05	.66
Systemic thrombolytic agents	2.79	0.47-16.53	.25
SPE	0.42	0.18-0.99	.04
VA-ECMO	1.62	0.61-4.26	.32
Anticoagulation <1 year	0.57	0.30-1.07	.08
Anticoagulation >1 year	2.20	1.29-3.75	.003

CDT, Catheter-directed thrombolysis; CI, confidence interval; OR, odds ratio; SPE, surgical pulmonary embolectomy; VA-ECMO, venous-arterial extracorporeal membrane oxygenation.

Boldface P values represent statistical significance.

Linear regression analysis of the 197 patients with CT images found that male sex (coefficient, −8.94; P = .02) and lower BMI (coefficient, −0.66; P = .003) were associated with a higher Δ%PAO. In this analysis, African-American patients (coefficient, 7.31; P = .05) were found to have a lower Δ%PAO. Linear regression analysis also showed that the use of CDT (coefficient, −18.12; P = .02) and SPE (coefficient, −21.69; P = .008) were associated with a higher Δ%PAO. The results for linear regression analyses are reported in Table VI.

Table VI.

Linear regression for presence of residual pulmonary vascular occlusion stratified by risk factors and treatment modalities (n = 197)

Variable	Coefficient	P value
Risk factor
Age	0.10	.36
Male sex	−8.94	.02
Body mass index	−0.68	.002
African-American race	7.31	.05
History or active cancer	−2.02	.65
Recent surgery (≤90 days)	3.75	.36
Recent trauma (≤90 days)	8.82	.14
Treatment modality
Time between images	−0.0009	.83
CDT	−18.12	.02
Systemic thrombolytic agents	−11.47	.42
SPE	−21.69	.008
VA-ECMO	−9.41	.22
Anticoagulation <1 year	−6.36	.18
Anticoagulation >1 year	−0.23	.95

CDT, Catheter-directed thrombolysis; SPE, surgical pulmonary embolectomy; VA-ECMO, Venous-arterial extracorporeal membrane oxygenation.

Boldface P values represent statistical significance.

Among the 315 patients with follow-up TTE, multivariate analysis did not show any significant correlation between the α%PAO and the risk of developing an elevated estimated RVSP (or subsequent treatment of pulmonary hypertension with pulmonary thromboendarterectomy or a pulmonary vasodilator). A second analysis also did not demonstrate a significant correlation between the use of advanced therapies (ie, CDT, SPE, systemic thrombolysis, ECMO) and the risk of developing an elevated RVSP or the need for pulmonary hypertension therapy.

Discussion

In this study, the prevalence of RPVO was 31% to 38%, and the median Δ%PAO was approximately −10 points. We found that patients with possible provoked PE from recent trauma or surgery were less likely to develop RPVO. Patients who developed RPVO had lower troponin levels at their initial presentation. Patients requiring advanced therapies typically had a larger clot burden on the initial CT scan compared with those treated with anticoagulation. On qualitative analysis, the presence of RPVO was not associated with any specific intervention. However, on quantitative analysis, advanced therapies, specifically CDT and SPE, were associated with a greater reduction in the thrombus burden. Furthermore, male sex and a lower BMI were associated with higher thrombus resolution. In the long term, male sex, recent surgery, SPE, and anticoagulation for <1 year were associated with a lower risk of RPVO. In contrast, African-American patients and patients with a history of cancer or active cancer were more likely to develop RPVO.

The prevalence of RPVO has been shown to vary across different clinical trials. In the SCOPE (study on the clinical course of pulmonary embolism) trial, Pesavento et al⁶ reported the prevalence of RPVO to be 50.1%. However, Planquette et al⁷ reported a prevalence of 19%. In our study, we found an RPVO prevalence of 31% to 38%. These studies primarily differ in the time between imaging studies, definition of RPVO, and treatment modalities involved. The average length of follow-up in the SCOPE trial⁶ and our study was 6 months and ∼3.5 months, respectively. In contrast, Planquette et al⁷ reported an average follow-up of 51 months. This suggests that a longer time between imaging studies might be associated with a lower reported incidence of RPVO. Furthermore, the definition of RPVO is not uniform across studies. Our study defined RPVO as any residual thrombosis in the pulmonary artery found on CT or V/Q scan or a Ω%PAO >1. Regarding the treatment modalities, in the SCOPE trial, 8% of patients received systemic thrombolysis, with no other reperfusion therapies. In our study, 19% of patients received advanced therapies.

Thrombus resolution is significantly influenced by the age of the thrombus at the diagnosis. Typically, after day 9, collagen begins to deposit in the thrombus, increasing resistance to thrombolysis.⁸ In our study, patients with a provoked PE exhibited a lower incidence of RPVO and a higher Δ%PAO, likely due to the early initiation of anticoagulation. This finding is supported by those from other studies suggesting that unprovoked PE is associated with a higher RPVO incidence.^2,6 Additionally, patients with provoked PE receive <1 year of anticoagulation but those with unprovoked PE receive lifelong anticoagulation therapy.

Biomarkers such as pro-brain natriuretic peptide and troponin are used to stratify the severity of acute PE. Troponin levels increase in response to acute dilatation of the right ventricle and is, therefore, associated with a worse prognosis.^9,10 However, a normal troponin level in the setting of a large clot burden and RV strain suggests that the PE could have occurred ≥1 week before the diagnosis. In a previous study from our institution, up to one third of patients referred for SPE had acute on chronic thrombus in the pathology report.¹¹ We suspect that most of the patients who developed RPVO had some degree of chronic thrombus at the diagnosis and, hence, had presented with lower troponin levels.

Clot burden itself is not typically a part of risk scoring systems because it is a poor predictor of short-term mortality.12, 13, 14 However, the clot burden size correlates with RV dilation, which differentiates low-risk from intermediate-risk PE. The presence of RV dysfunction strongly correlates with short-term mortality, significantly influencing the decision to pursue advanced therapies.¹⁵ In our study, the median α%PAO for those who received advanced interventions was significantly greater than that of patients treated with anticoagulation. This finding suggests that the clot burden is crucial to the clinical assessment of disease severity, because it correlates with RV dysfunction and hemodynamic instability.

Previous literature has shown that catheter-based interventions for PE result in a significant reduction in the thrombus burden by 24 to 72 hours16, 17, 18, 19, 20; however, long-term data on outcomes have not been previously available. In our study, a larger Δ%PAO was observed on imaging analysis for patients who received advanced interventions, especially CDT and SPE. Prior studies have not shown a risk reduction of CTEPH with systemic thrombolysis.²¹ However, these studies did not investigate catheter-based therapies or SPE. Further studies, including the ongoing randomized PE-TRACT (pulmonary embolism – thrombus removal with catheter-directed therapy) and HI-PEITHO (higher risk pulmonary embolism thrombolysis) trials, can hopefully determine whether catheter-based therapies reduce the CTEPH risk.

In our study, no significant correlation was found between the use of advanced interventions for acute PE and an eventual need for pulmonary hypertension therapy. This result might reflect an overall underdiagnosis of CTEPH and a lack of statistical power to detect such a difference, given the smaller number of patients with elevated pulmonary pressure on follow-up echocardiography. However, intimal hyperplasia and inflammation have been demonstrated as part of the component of RPVO in animal models with untreated PE.²² This could explain why some patients develop pulmonary hypertension and other patients do not. No human studies are available yet. Studying the pathways of endothelial dysfunction and vascular remodeling and fibrinolytic activity could help answer this question.

We found that patients with known malignancy had a significantly elevated risk of RPVO at follow-up, consistent with the known association between cancer and CTEPH. Active malignancy is a well-established risk factor for venous thromboembolism (VTE) overall and has been identified as a risk factor for the subsequent development of CTEPH, increasing its incidence by more than threefold.^23,24

Previous studies have reported a higher risk of VTE for female patients than for male patients, which has been attributed to several factors, including pregnancies, hormones, contraception treatment, and so forth.25, 26, 27 In our study, women were found to have an increased risk of both RPVO and lower thrombus resolution. Estrogen has been described as cardioprotective in premenopausal women, because it is associated with increased activity of endothelial tissue plasminogen activator.28, 29, 30 However, it is unclear whether this cardioprotective effect also applies to venous thrombus resolution. The presence of gender biases across the medical field that could delay the diagnosis for female patients is undeniable,31, 32, 33 and a delay in the diagnosis and initiation of anticoagulation could have contributed to the development of RPVO.

A similar phenomenon can be observed in obese patients, because obesity itself increases the risk of VTE.³⁴ Symptoms such as shortness of breath and leg edema are more likely to be missed in obese patients because of the body habitus, affecting the ability to promptly identify VTE and initiate anticoagulation therapy. However, chronic inflammation and enhanced activity of plasminogen activator inhibitor-1 has been observed in obese patients and patients with metabolic syndrome, increasing the risk of impaired thrombolysis and RPVO.^35,36

Finally, previous studies have demonstrated that African-American patients have an increased risk of venous thromboembolism and complications from chronic disease.^37,38 In a previous study, we reported that African-American patients were more likely to exhibit RV dilation and dysfunction with higher pulmonary pressures at the time of an acute PE diagnosis, strongly suggesting the presence of underlying chronic disease.³⁹ Although African-American patients are more likely to have hypercoagulable diseases (ie, sickle cell trait, sickle cell disease, elevated factor VIII, decreased protein C), these conditions typically do not affect thrombus resolution. No known specific diseases affect thrombolysis in African-American patients.⁴⁰

The findings from the present study show that a better understanding of the risk factors that can predict for RPVO can help identify at-risk patient populations and potential candidates for advanced therapies. We found that patients with unprovoked PE (anticoagulation >1 year), female patients, African-American patients, obese patients, and cancer patients had an increased risk of RPVO. These patients might benefit the most from CBI and SPE, which have demonstrated the greatest degree of thrombus reduction. The selection between CBI and SPE should be determined by the physician's skills, evidence of chronic thrombus, and patient's comorbidities, because the presence of chronic PE could limit the success of CBI.

Study limitations

The present study has several limitations, including its retrospective study design. A standard protocol to follow-up PE patients at regular intervals is needed. Regular screening for RPVO is not supported by the current guidelines, and only patients with long-term symptoms presented for follow-up imaging, which could have led to a self-selected patient population. Patients who are asymptomatic or have less severe symptoms and residual thromboembolic disease could have been missed. A larger sample size would also be required to demonstrate a clear relationship between RPVO and CTEPH, which could be addressed through the creation of post-PE clinics that could identify early cases of chronic thromboembolic pulmonary disease and CTEPH. Finally, because a standard definition for RPVO has yet to be established, the field must work toward creating universally accepted diagnostic criteria that account for the imaging modality, percentage of occlusion, and time to follow-up.

Conclusions

The present study is one of the first to provide qualitative and quantitative data about long-term thrombus resolution after PE and its relationship to the development of pulmonary hypertension. Our findings underscore the importance of early diagnosis and treatment and identified multiple risk factors associated with the presence of chronic residual thrombus in the pulmonary artery. Advanced therapies such as CBIs and SPE, might confer long-term benefits in thrombus resolution. Routine screening for RPVO (imaging studies 3-6 months after the index PE) would help us to understand the dimension of the disease, and further studies are needed to investigate thrombus resolution and chronic thromboembolic disease in high-risk populations, such as female patients, obese patients, and African-American patients.

Author contributions

Conception and design: AF, SC, VE, KJ, KN, ST, JJ, DA, BG, JS, RS, BL, RC

Analysis and interpretation: MM, JS, RC

Data collection: AF, PH, TJ, AL, RC

Writing the article: AF, PH, SC, RC

Critical revision of the article: AF, MM, PH, SC, TJ, AL, VE, KJ, KN, ST, JJ, DA, BG, JS, RS, BL, RC

Final approval of the article: AF, MM, PH, SC, TJ, AL, VE, KJ, KN, ST, JJ, DA, BG, JS, RS, BL, RC

Statistical analysis: MM, JS, RC

Obtained funding: Not applicable

Overall responsibility: RC

From the American Venous Forum