Abstract
Background
The incidence and clinical significance of pulmonary residual thrombosis 6 months after an acute pulmonary embolism (PE) are still not well-known.
Purpose
To evaluate the association between residual vascular obstruction and the risk of venous thromboembolism (VTE) recurrence or death.
Material and Methods
Computed tomography pulmonary angiography (CTPA) was repeated in 97 consecutive patients 6 months after an acute episode of hemodynamically stable pulmonary embolism. We assessed the long-term consequences of residual thrombosis on vital status and incidence of recurrent VTE.
Results
Six patients were lost for follow-up. The remaining 91 patients were classified according to the presence (Group 1: 18 cases) or absence (Group 2: 73 cases) of residual pulmonary vascular obstruction. After a mean ± SD of 2.91 ± 0.99 years, there were eight (8.8%) deaths and 11 (12.1%) VTE recurrences. Groups 1 and 2 did not differ in the incidence of death or VTE recurrence.
Conclusion
Persistent pulmonary vascular obstruction on 6-month CTPA did not predict long-term adverse outcome events.
Pulmonary embolism (PE) remains a major health problem. Computed tomography pulmonary angiography (CTPA) has a central role in the diagnosis of the disease (1). After acute PE, intraluminal vascular filling defects are believed to resolve in the majority of cases (2), although they might persist in some patients 6 months after the acute event (3). CTPA has been used not only as a diagnostic technique, but also as a predictor of short-term outcome in these patients (4). The prognostic significance of incomplete resolution of pulmonary embolism after anticoagulation therapy remains largely unknown. The objective of this study was to assess if persistence of residual vascular obstruction 6 months after PE might predict long-term adverse events. The patient population in the present study was based on the population in a previous study that assessed the prevalence and 6-month evolution of right ventricle dysfunction and pulmonary hypertension, and described its relation to CTPA findings (3).
Material and Methods
Patients
From April 2005 to August 2008, 110 consecutive, hemodynamically stable patients presenting to our hospital with PE were prospectively recruited. Inclusion criteria were: age ≥ 18 years, and CTPA-confirmed PE. Exclusion criteria were: massive PE (defined as the presence of two consecutive systolic blood pressure measurements < 90 mmHg measured > 15 min apart); co-morbidity that predicted a 6-month mortality rate > 50% (e.g. metastatic cancer or end-stage respiratory disease); creatinine clearance < 35 mL/min or a history of allergy to contrast media. Initial patient examinations included clinical history, physical examination, electrocardiography, arterial blood gas analysis, echocardiography, CTPA, and lower-limb venous ultrasonography. CTPA and echocardiography were repeated at 6-month follow-up in all patients. Lower-limb venous ultrasonography was repeated at 6-month follow-up if the initial study disclosed deep vein thrombosis (DVT).
The present study evaluated the vital status and the incidence of PE-related adverse events in the patients included in the original study. We obtained information about the present vital status, hospital admissions, treatments, primary care physician visits, and emergency-room visits from a centralized clinical data management program. When information about the vital status was not available through this program, we tried to contact directly with the patient or relatives. The cause of death and diagnosis of adverse events were obtained from medical reports. When medical reports were not available, efforts were made to collect information concerning the cause of death. The present, non-interventional study was approved by our institution's review board.
Computed tomography pulmonary angiography
All CTPAs were performed within 6 h of patient admission to the hospital and repeated after 6 months, as previously described (3). A total of 100 mL of Ioversol at a concentration of 240 mg/mL (Optiray Ultraject; Mallinckrodt Inc, Hazelwood, MO, USA) was administered intravenously at an injection rate of 2.5 mL/s for 40 s. Images were obtained on the same scanner (Somatron PQ 2000S; Picker International, Cleveland, OH, USA). If possible, scanning was performed during a 32-s breath hold. If patients were dyspnoic, spiral CT angiography was performed during shallow breathing. The scans were obtained using 125 mA and 130 kV. A 5 mm/s table feed was used to scan a 16-cm volume in the craniocaudal direction (pitch of 1.5). An imaging delay of 20 s was used, and overlapping images were reconstructed every 1.5 mm and viewed at lung and mediastinal settings. Images were read in axial views. PE was diagnosed if an intraluminal filling defect was seen. Patients were classified into two groups depending on the presence (Group 1) or absence (Group 2) of residual pulmonary vascular obstruction at 6-month CTPA.
Management strategies
All the patients included in the original study were treated with anticoagulation, as previously described. Thrombolytic treatment was not used in any case (3). After 6-month follow-up, the duration of anticoagulation was decided by the attending physician according to standard clinical criteria. Continuation or discontinuation of anticoagulation therapy was recommended regardless of the 6-month CTPA findings. At hospital discharge, patients were instructed to attend unscheduled visits in case of sudden-onset dyspnea, hemoptysis, chest pain, syncope, new-onset or worsening of swelling, redness, pain or warmth of an extremity.
Definition of study endpoints
The primary study outcomes were all-cause mortality and VTE recurrence. The criterion for diagnosis of recurrence of PE was a new filling defect at CTPA. The criteria for diagnosis of DVT as a recurrence of VTE in patients without DVT at baseline were the presence of a non-compressible proximal vein at ultrasonography or an intraluminal filling defect on venography. In cases with DVT at baseline, criteria for recurrence of DVT were either results on ultrasonography or venography suggestive of thrombosis in a different venous segment, as described above, an extension of a previous intraluminal filling on venography, or a 4 mm or greater increase in thrombus diameter on ultrasonography.
Statistical analysis
Data are expressed as mean ± SD or as percentages. The D'Agostino-Pearson test was used to assess the normal distribution of data. For continuous variables, comparisons between groups were made using unpaired t-test or Mann-Whitney rank-sum test, as appropriate. Fisher's exact test was used for categorical variables. Kaplan-Meyer time-to-event curves that represent cumulative event rates were obtained, and compared by the log-rank test. A P value of <0.05 was considered significant.
Results
Patients and management
One hundred and ten patients (mean age 69 ± 15 years) were considered for inclusion in the original study, as previously reported (3). Complete follow-up, including 6-month CTPA was achieved in 97 cases. Fig. 1 shows the flow diagram for the present study. Time from the inclusion in the original study until the last contact was 2.91 ± 0.99 years. We were not able to obtain information on the evolution of six patients. Thus, 91 patients completed the present study. Seventeen (18.7%) of these patients were treated with long-term anticoagulation. The reasons for prolonging anticoagulation beyond 6 months were: diagnosis of atrial fibrillation (6 cases), inferior vena cava filter placement (1 patient), diagnosis of other procoagulant conditions (4 cases), and the fact that the PE at the inclusion in the original study was an episode of recurrent VTE (6 patients). The percentages of patients who were anticoagulated in the long term were not significantly different in patients with resolution or persistence of vascular obstruction (16.4% vs. 27.7%, respectively; P = 0.31).
Flow chart of the study protocol. CTPA, computed tomography pulmonary angiography; IPF, idiopathic pulmonary fibrosis; PE, pulmonary embolism; VTE, venous thromboembolism
Long-term evolution
Characteristics of patients categorized according to 6-month CTPA findings
Results are expressed as mean ± standard deviation or as number of cases (percentages)
Clinical outcomes of patients with and without persistent vascular obstruction
Results are expressed as: number of cases (percentages)
VTE, venous thromboembolism
Clinical outcomes of patients with and without persistent vascular obstruction (excluding cases on long-term anticoagulation therapy)
Results area expressed as number of cases (percentages)
VTE, venous thromboembolism
Figs. 2–4 show the Kaplan-Meier curves for Groups 1 and 2 regarding death, VTE recurrence, and a composite of all adverse events. There were no significant differences between groups.
Kaplan-Meier curve for all-cause death (P = 0.57). Solid line, cases without residual vascular obstruction; dashed line, patients with residual obstruction Kaplan-Meier curve for VTE recurrence (P = 0.47). Solid line: cases without residual vascular obstruction; dashed line, patients with residual obstruction Kaplan-Meier curve for all adverse events (P = 0.57). Solid line, cases without residual vascular obstruction; dashed line, patients with residual obstruction
Discussion
This study has found residual intraluminal filling defects in the pulmonary vasculature of 19.7% of patients, 6 months after an acute PE. Persistence of vascular obstruction did not correlate with a higher incidence of adverse outcome events. Routine performance of CTPA after termination of anticoagulant therapy in PE is not a standardized procedure (2). However, it may offer valuable information for the management of these patients. It would help to recognize residual, unresolved pulmonary vascular obstruction after PE, and to avoid falsely diagnosing these cases as having recurrent PE. This is a relevant clinical problem, as previous studies have shown that up to 50% of patients will still have persistent vascular obstruction after 6 months of anticoagulation therapy. Therefore, some authors have suggested that routine re-imaging should be considered after anticoagulation withdrawal, with the intention to establish a new baseline (2).
Our hypothesis was that control CTPA performed 6 months after the episode of PE would offer additional clinical information. We have speculated that residual filling defects would correlate with a higher rate of subsequent adverse events through three different mechanisms: first, patients with residual pulmonary vascular obstruction might have a reduced cardiorespiratory reserve. Thus, they would be at a higher risk of death if they suffered another disease that could impose a further strain on the cardiovascular system. Second, these patients would be at a higher risk of developing chronic thromboembolic pulmonary hypertension, a disease with a bad prognosis (5). Finally, residual obstruction of the pulmonary vascular bed might associate with an increased risk for recurrent thrombosis. Previous studies have demonstrated an association between residual venous thrombosis and recurrent VTE (6). It has been suggested that this link reflects an underlying prothrombotic state in patients with persistent venous thrombosis that puts them at higher risk for recurrence of VTE (6). We theorized that persistent pulmonary vascular obstruction might have similar implications and could therefore help to identify high-risk patients that may benefit from extended anticoagulation.
The results of the study do not support our hypothesis. There were no differences in the rate of adverse events in cases with or without residual vascular obstruction. Also, patients with persistent intraluminal filling defects did not have a higher prevalence of idiopathic PE. This is not consistent with the hypothesis of an underlying hypercoagulable state in these cases.
The present study might be underpowered to detect statistically significant differences between both groups. It must be noted that 18.7% of our patients received long-term anti-coagulation therapy and this might have reduced the incidence of recurrent VTE and, consequently, the statistical power of the study. Also, it must be remarked that the percentages of patients who had a normal CTPA at the sixth month were higher than those found in previous studies (7, 8). There are some possible explanations. Remy-Jardin et al. (7) recruited patients with massive PE that were referred to an intensive care unit. By contrast, patients with hemodynamic instability were excluded from the present study. Thus, it seems plausible than patients with more severe PE would have lower rates of resolution of thrombi at the sixth month. On the other hand, Van Rossum et al. (8) studied 19 patients 6 weeks after PE. It is possible that the rate of normalization of PE in their study (32%) would have been higher if the follow-up would have been longer.
In conclusion, this study suggests that control CTPA 6 months after PE is not useful to identify patients at a higher risk of suffering subsequent adverse outcomes. Therefore, our results do not support the need of performing follow-up CTPA after PE. However, due to the limitations of size and design of the study, these results must be considered preliminary.