Between August 1999 and June 2002, 60 consecutive patients were referred to our angiography laboratory for suspected acute pulmonary thromboembolism. Among them, 25 patients with hemodynamic impairment due to acute massive pulmonary thromboembolism were prospectively selected for interventional radiology treatment. Right ventricular afterload was confirmed on transthoracic echocardiography in all patients. The criteria for inclusion in our study were an angiographically confirmed acute massive pulmonary thromboembolism and involvement of the central pulmonary arteries and hemodynamic impairment (mean pulmonary artery pressure, ≥ 25 mm Hg). Twenty hemodynamically stable patients (mean pulmonary artery pressure, < 25 mm Hg) were excluded. Fourteen patients with chronic pulmonary thromboembolism and one patient with clinically severe hemodynamic impairment who was receiving percutaneous cardiopulmonary support and who had to be returned to the coronary care unit during angiography were also excluded from this study.

Approval for our study was obtained from the local university ethics committee, and depending on the patient's condition, either written or oral informed consent was obtained. If a patient was unconscious and receiving respiratory therapy at the time of inclusion in the study, the ethics committee was notified of the inclusion, and consent was initially obtained from the family and then from the patient after recovery.

Our final study population was composed of 17 women and eight men, ranging in age from 35 to 77 years old. Patient data and details are shown in Table 1. The mean ± SD shock index (heart rate / systolic blood pressure) was 0.77 ± 0.4. Supplementary mechanical ventilation and positive inotropic support were added, according to the severity of the patient's hemodynamic impairment.

View Larger Version

TABLE 1 Demographic and Clinical Data for 25 Patients Who Underwent Hybrid Treatment for Acute Massive Pulmonary Thromboembolism

Pulmonary angiography was initially performed in all patients using a combination of digital subtraction angiography and rotational digital angiography [6]. The extent of thrombus formation found on angiography was independently assessed by three radiology specialists using the method described by Miller et al. [7]. In this method, the right and left main pulmonary arteries are considered to have nine and seven major branches, respectively, and an embolus in any of these branches is scored as 1 point. Each lung is considered to have an upper, middle, and lower zone, and in each of these three zones, the absence of pulmonary artery flow is scored as 3 points; severely reduced flow, 2 points; mildly reduced flow, 1 point; and normal flow, 0 points. Thus, the Miller score has a range of 0–34.

A 6-French short sheath was inserted in the femoral vein, and a 6-French curved pigtail catheter for pulmonary angiography (K-PA catheter, Medi-kit) was advanced into the pulmonary artery.

Through the catheter in the pulmonary artery, a 260-cm guidewire (Radifocus, Terumo) was placed in the peripheral pulmonary artery. The curved pigtail catheter was then withdrawn, leaving the guidewire in the peripheral pulmonary artery. The 6-French sheath was exchanged over the guidewire for an adapted 95-cm-long 8-French sheath with a curved tip initially designed for cardiac biopsy. The 8-French sheath was placed in either the right or the left main pulmonary artery. The guidewire was left in a peripheral part of the pulmonary artery.

To allow the pigtail catheter to move freely over the guidewire in the pulmonary artery, we inserted the trailing tip of the guidewire in the most proximal side hole of the curved pigtail catheter (Figs. 1A and 1B). Once the trailing tip of the guidewire had passed through the catheter opening, the catheter was inserted over the guidewire and through the 8-French sheath placed in the patient.

View larger version (41K)

Fig. 1A. —Catheter system for manual fragmentation of pulmonary thromboembolism. Photograph shows 6-French curved pigtail catheter for pulmonary angiography (K-PA catheter, Medi-kit) that is used during procedure.

View larger version (56K)

Fig. 1B. —Catheter system for manual fragmentation of pulmonary thromboembolism. Photograph shows 260-cm guidewire (Radifocus, Terumo), proximal tip of which is inserted into most proximal side hole of curved pigtail catheter. Catheter is first inserted over guidewire and through pulmonary artery sheath and is then withdrawn, leaving guidewire in peripheral pulmonary artery.

With the guidewire remaining in the peripheral pulmonary artery, the pigtail catheter was advanced to a central portion of the pulmonary artery. Leaving the 8-French pulmonary artery sheath, the curve of the pigtail catheter deployed itself in the pulmonary artery, and the curve of the pigtail catheter became sufficiently free to rotate over the guidewire in the pulmonary artery.

The diameter of the pigtail curve varies between the 5- and 6-French catheters. The 6-French catheter, having a 10-mm diameter, was used in the central portion of the pulmonary artery, whereas the 5-French catheter, having a 6-mm diameter, was used in the peripheral part of the pulmonary vascular tree. Emboli were fragmented by the mechanical action of the rotating pigtail catheter. The catheter was rotated manually around the axis of the stationary guidewire and was advanced or withdrawn over the guidewire.

After the emboli were fragmented with the rotating pigtail catheter, all patients received an intrapulmonary injection of recombinant human-tissue plasminogen activator (rt-PA) (640 × 10⁴ IU, equivalent to 12.8 mg/64 min) followed by manual clot aspiration using a large-lumen percutaneous transluminal coronary angioplasty (PTCA) guide catheter (8-French Guider Softip XF, Boston Scientific, SciMed Life Systems). Strong manual aspiration was created through a regular Luer-Lok 20-mL syringe plunger (Becton Dickinson) while the catheter was slowly withdrawn through the introducer sheath. During thrombolysis, all patients received heparin sodium (initial dose, 5,000 IU; maintenance dose, activated partial thromboplastin time ratios of 2). After these hybrid treatments, the 8-French long sheath was exchanged for a standard 9-French short sheath, and a temporary filter (Antheor, Boston Scientific) was inserted at the infrarenal inferior vena cava as protection against recurrent pulmonary thromboembolism. Additional systemic urokinase infusion was administered in the ICU for the residual thrombi if a patient's condition required it. The dosing regimen originally was 24–48 × 10⁴ IU per day for 3 days. The femoral vein sheath was removed after administration of the systemic urokinase therapy. We used the paired Student's t test for statistical analysis.

We experienced one case of cardiac arrest during pigtail catheter rotation (patient 4). The patient was intubated, received cardiopulmonary resuscitation, and underwent catheter thrombolysis and thrombectomy. Her heartbeat was restarted 30 min after the event, and her pulmonary artery pressure decreased. The length of the patient's stay in the angiography laboratory was 167 min. She was discharged from hospital 70 days after the procedure with no subsequent problems.

In one patient (patient 9), the catheter shaft fragmented in the sheath during catheter rotation but was easily pulled out with Péan's forceps. No other complications were encountered during or after the procedure.

The rotational movement of the pigtail portion of the catheter acts directly on the clots in the pulmonary arteries, causing fragmentation and distal migration of the smaller fragments. The volume of the peripheral pulmonary arteries is approximately twice that of the central pulmonary arteries. Therefore, redistribution of a large central clot may acutely improve cardiopulmonary hemodynamics [2, 8, 9].

A previous phase I multicenter clinical study used a special fragmentation catheter device [9]. We used a conventional curved pigtail catheter for pulmonary angiography. The long guidewire was left in a peripheral site of the pulmonary artery, and the proximal tip of the guidewire was inserted into the most proximal side hole of the curved pigtail catheter. The catheter was then inserted over the guidewire and through the pulmonary artery sheath. This modified technique is easy to implement and requires no greater skill than that needed for right heart catheterization. In the previous study, three catheter mismatches occurred among 10 patients [9]. The pigtail diameter used was too small for a large embolus in the dilated main artery. In our series, the diameter of the pigtail catheter varied between 6 and 10 mm, and therefore we had no catheter-related mismatch. If the diameter of the right or left main pulmonary arteries was greater than 10 mm, we could enlarge the diameter of the pigtail catheter by inserting the guidewire. We encountered a minor complication—a case of the catheter shaft fragmenting during catheter rotation. We have since strengthened the shaft.

Our technique offers a possible synergistic effect with concurrent thrombolytic therapy because the resulting clot fragments have a greater surface area exposed to the thrombolytic agent, thus improving the results of lytic activity and allowing a reduction of dose and infusion time [2].

The few reports of treatment of massive pulmonary thromboembolism using mechanical fragmentation combined with thrombolytic therapy have all indicated that the treatment results in rapid hemodynamic improvement [4, 8, 10, 11]. However, the mean infusion time has been reported to be 18–48 hr, and infusion was continued in the ICU. In the treatment of acute myocardial infarction using percutaneous transluminal coronary recanalization, fast high-dose administration of rt-PA has been used [4]. In our series, 24 patients received an intrapulmonary injection of rt-PA (640 × 10⁴ IU, equivalent to 12.8 mg/64 min) in the angiography laboratory. Only one patient (patient 10) received 480 × 10⁴ IU of rt-PA (equivalent to 9.6 mg/48 min) because she had received systemic administration of urokinase at the city hospital before undergoing the procedure. The dose administered was the approved maximum dose in Japan for percutaneous transluminal coronary recanalization, and no bleeding complications occurred. The superiority of rt-PA over urokinase for treatment of massive pulmonary thromboembolism has yet to be confirmed [11].

The recanalization rate achieved after 2 hr by thrombolysis alone was reported as a reduction in the angiographic score of 17.8% (IV urokinase) and 22.4% (IV rt-PA) in one multi-center study using acute dosing regimens [12] and of 12.0% (intrapulmonary rt-PA) and 15.4% (IV rt-PA) in another multicenter study also using acute dosing regimens [13].

Percutaneous pulmonary thrombectomy has not achieved widespread use. Possible reasons include the need for a venotomy or a large introducer sheath (16- to 24-French), special skills for steering and pulmonary placement, and only partial removal of the embolus, necessitating repositioning and subsequent passes [3, 14, 15].

Percutaneous pulmonary clot aspiration using a large-lumen 8-French PTCA catheter has been described [16, 17]. The advantages of this technique are that it is less vessel-invasive and more convenient to use in a standard angiography laboratory because a small 8-French introducer sheath and a conventional PTCA guiding catheter are used. In this series, the central pulmonary thrombus was fragmented using a modified pigtail catheter, which decreased the clot volume and facilitated clot aspiration. There was a risk of blood depletion, and we observed a mean decrease in the hemoglobin level of 1–2 g/dL, but no patient required blood transfusion because of this procedure.

The end point of this hybrid procedure was to improve the hemodynamic situation in the patient, and the treatment was generally successful in all cases. With an average total procedure time of 124.6 min, a high recanalization rate was rapidly accomplished. Angiography in all patients after treatment showed improvement of pulmonary perfusion, with the mean Miller score decreasing from 22.2 to 13.6 (p < 0.01). The results were significantly better than those of the multicenter studies using thrombolysis alone, either catheter-directed or peripherally injected [12, 13]. In the phase I multicenter clinical study, no significant decrease was noted in the mean arterial pulmonary pressure immediately after fragmentation with the pigtail catheter [9]. This finding was believed to have been caused by pulmonary vasoconstriction secondary to the local release of neurohumoral factors such as endothelin.

A prospective multicenter study of fragmentation using rotation of a pigtail catheter found a significant decrease in the mean pulmonary artery pressure from 31 mm Hg before fragmentation to 28 mm Hg after fragmentation (n = 15 patients; real fragmentation time, 17 min) [18]. Therefore, because the effect of fragmentation by a pigtail catheter alone may not be sufficient, we decided to use a hybrid approach. In our series, mean pulmonary artery pressure decreased from 32.6 to 23.4 mm Hg (p < 0.01) immediately after mechanical fragmentation, local fibrinolysis, and manual clot aspiration. The hemo-dynamic condition of the patients improved significantly, so we decided to continue to use systemic urokinase therapy instead of catheter-directed thrombolysis.

In unfavorable situations, catheter fragmentation may initially worsen the situation by dislodging pulmonary emboli in the still-perfused major pulmonary artery branches, producing an additional occlusion there [19]. This event can be viewed as a complication of the fragmentation technique that we could successfully treat by subsequent fragmentation, thrombolysis, and clot aspiration.

A mean of 4.4 days of additional systemic urokinase infusion in the ICU was required for the fibrinolytic therapy of residual thrombi after the hybrid therapy, which is longer than the 48 hr of thrombolysis reported in the phase I clinical data [9]. We could have minimized the dose of urokinase by careful evaluation of each patient's hemodynamic condition. We selected urokinase for the additional systemic fibrinolytic therapy because it is the only fibrinolytic drug accepted by the Ministry of Welfare in Japan for the treatment of pulmonary thromboembolism. We were able to use rt-PA only for selected situations because of financial constraints.

In conclusion, fragmentation using a modified rotating pigtail catheter with intrapulmonary fibrinolysis and manual clot aspiration can improve severe hemodynamic impairment rapidly and safely in patients with acute massive pulmonary thromboembolism. This hybrid treatment appears to be especially useful in patients at high risk of right ventricular failure and is a minimally invasive alternative to surgical embolectomy. These promising results emphasize the need for a prospective randomized trial that compares results of this hybrid treatment with those achieved with systemic rt-PA, which should define the role of hybrid therapy more clearly.