Low-dose, short course alteplase treatment of submassive pulmonary embolism: a case series from the National Institutes of Health Clinical Center

Abstract

Guidelines-recommend thrombolytic therapy for pulmonary embolism in patients with severe hemodynamic compromise and low risk of bleeding. Thrombolytics in submassive pulmonary embolism have an unfavorable risk/benefit ratio and remain controversial. Based on our experience with extensive, lower extremity thrombi, nine patients with symptomatic, submassive pulmonary embolisms (five medical, four surgical) were treated with low-dose alteplase (<10 mg/day, infused over 6 h per treatment). Alteplase was delivered by pulse spray and/or directed or undirected central venous catheters depending on clot size and location. All patients improved symptomatically and as determined objectively by pulmonary artery pressures and/or imaging, though acute benefits ranged from substantial to modest. One surgical patient required reexploration for bleeding at the site of a recent retroperitoneal lymph node dissection. This experience may help guide the design of a randomized controlled trial to determine the safety and efficacy of low-dose alteplase for submassive pulmonary embolism.

Introduction

Current guidelines recommend fibrinolytic therapy with tissue plasminogen activator (alteplase) for massive pulmonary embolism with hemodynamic collapse (e.g., alteplase 100 mg over 2 h) [1,2]. In submassive pulmonary embolism, the potential benefits of clot lysis are offset by bleeding complications [3–7] and accordingly fibrinolytic therapy has been advocated only if these patients develop hemodynamic compromise [8,9]. However, the evidence supporting this ‘watch and wait’ approach only applies to upfront, full-dose systemic fibrinolytic therapy for which the risk of hemorrhagic stroke is 10 times higher than anticoagulation alone [7]. Somewhat lower dose alteplase regimens (0.6 mg/kg and 50 mg) and shorter duration infusions have also been investigated and appear to be effective, but still engender serious bleeding complications [10–15]. In our experience, alteplase doses of 10 mg/day or less, for 1–3 days can lyse large volume, deep vein thrombi involving an entire leg [16], so we empirically tried this low-dose approach in submassive pulmonary embolism cases not requiring rapid clot lysis with ‘standard’ doses of alteplase.

The Office of Human Subjects Research Protections of the NIH Clinical Center declined to consider our case series to be research and waived the need for Institutional Review Board (IRB) approval as they considered our activity to be the practice of medicine.

Case reports

Between 2011 and 2016, nine patients enrolled in various clinical research protocols at the NIH Clinical Center received low-dose alteplase treatment in addition to anticoagulation for submassive pulmonary embolism (Table 1). All of these patients, ranging from 29 to 86 years of age, presented with severe dyspnea, large embolic burden, evidence of cardiac strain, and/or limitations in cardiopulmonary reserve. Four of the patients had undergone surgery within the prior 30 days. N-terminal prohormone of brain natriuretic peptide and/or troponin levels performed in eight patients showed evidence of cardiac strain and/or injury. Echocardiography demonstrated various degrees of right ventricular (RV) dilatation and/or dysfunction in all nine patients (refer to Supplemental Table 1, http://links.lww.com/BCF/A54 for detailed echocardiographic results).

Table 1.

Clinical summary of patients treated with low-dose alteplase for submassive pulmonary embolism

Case	Age (years) sex	Underlying conditions	Signs and symptoms	Imaging	Echocardiogram prior to tPA	Cardiac Biomarkersa	Thrombolytic therapy	Anticoagulantb	Outcome
1	42 M	Metastatic thymoma s/p left pneumonectomy in remote past and chemotherapy-induced cardiomyopathy; recent (<1 month) debridement diabetic leg ulcer	Dyspnea and SpO2 of 85% on room air	CTA: Large right main PA clot with numerous rightsided segmental and subsegmental emboli	Severe RV dilation with RV dysfunction present but not worse compared with baseline	NT-proBNP: 354 pg/ ml (0–92pg/ml); troponin I: normal	6 mg of tPA per day for 2 days via a central venous catheter (1 mg/h for 6 h)	UFH	48 h following the first dose of tPA, repeat CTA demonstrated a significant reduction in the right main PA clot that no longer extended into the upper and lower lobe branches. Discharged home on enoxaparin and did not require supplemental oxygen
2	86 F	Heparin-induced thrombocytopenia	Dyspnea and SpO2 of 86% on room air	V/Q: Mismatched perfusion defect of the entire LLL and 2 seqments of the RUL	Moderate RV dilation with RV dysfunction	NT-proBNP: 5889 pg/ml (0–623 pg/ml); troponin I: 0.506 ng/ml	6 mg of tPA per day for 2 days via a central venous catheter (1 mg/h for 6 h)	Argatroban	48 h following the first dose of tPA, repeat V/Q scan demonstrated improvement in perfusion of the RUL and one segment of the LLL; repeat TTE 12 days after tPA revealed decreased RV dilation, normal RV function, and a decrease in estimated RVSP from 81 to 58 mmHg. Discharged home on warfarin and did not require supplemental oxygen
3	75 F	Metastatic endometrial cancer, bilateral malignant pleural effusions on home oxygen (3 l/min)	Worsening dyspnea and SpO2 of 85% on 5 l/min O2	CTA: Large left main PA embolus with multiple bilateral filling defects	Mild RV dilatation with normal RV function	Not tested	6mg of tPA per day for 2 days via a central venous catheter (1 mg/h for 6 h)	UFH	48 h following the first dose of tPA, repeat CTA demonstrated reduction of RLL thrombus and multiple RUL thrombi; partial lysis of large left main PA thrombus with residual clot embolization of the LLL branch. Discharged home on enoxaparin and an increased amount of supplemental oxygen (SpO2 94% on 4 l/min)
4	41 F	Sickle cell anemia s/p failed HSCT, PH, and prior PE on enoxaparin at presentation	Dyspnea, pleuritic chest pain; SpO2 87% on room air	CTA: Extensive bilateral filling defects	Severe RV dilation with RV dysfunction	NT-proBNP: 4033 pg/ml (0–124 pg/ml); troponin: not tested	6 mg of tPA per day for 2 days via a central venous catheter (1 mg/h for 6 h)	UFH	48 h following first dose of tPA, repeat TTE showed restoration of RV function, but no change in RV dilation and estimated RVSP remained markedly elevated at 93 mmHg. TTE at 1 week showed slight improvement in RV size with preserved function, and a decrease in estimated RVSP to 48 mmHg. Discharged home on apixaban and did not require supplemental oxygen
5	46 M	Von Hippel–Lindau disease with a history of multiple surgeries for CNS hemangioblastomas and renal tumors; 3 weeks s/p ventriculoperitoneal shunt revision complicated by heparin-induced thrombocytopenia	Sudden onset tachypnea and SpO2 75% on room air	CTA: Large saddle embolus and multiple, bilateral filling defects	Moderate RV dilation with RV dysfunction	NT-proBNP: 62 pg/ml (0–124 pg/ml); troponin T: 0.039 ng/ml	2 mg of tPA administered by pulse spray catheter injection to right and left main PA followed by 6 mg of tPA via PA pigtail catheter (1 mg/ h for 6 h); interatrial defect noted incidentally during PA catheterization	Argatroban	Hypoxia resolved overnight. Repeat TTE demonstrated improved RV function 15 h following start of tPA and bubble study confirmed right to left interatrial shunting. Estimated RVSP was not measurable. Discharged home on rivaroxaban and did not require supplemental oxygen
6	78 M	Prostate cancer and recent LLE DVT	Sudden onset of dyspnea, SpO2 97% on 2–3 l/min O2	CTA: Extensive bilateral filling defects	Severe RV dilation with RV dysfunction	NT-proBNP: 8573 pg/ml (0–449 pg/ml); troponin T: 0.024 ng/ml	Days 1 and 2: 2 mg of tPA administered by pulse spray catheter injection to right and left main PA followed by 6 mg of tPA via PA pigtail catheter (1 mg/ h for 6 h); Day 3: 2 mg of tPA by pulse spray catheter injection to right and left main PA followed by 6 mg of tPA to LLE by pulse spray catheter injection; Day 4: 4 mg of tPA to DVT LLE via a LLE venous catheter (1 mg/h for 4 h)	UFH	6 h following the initiation of tPA, PA pressures decreased from 50/23 to 37/17 mmHg. After the first dose of tPA, CTA showed modest decrease in the extent of emboli. Repeat CTA 6 d following first tPA treatment showed marked improvement in right sided emboli and decreased emboli on the left side, but to a lesser extent. TTE 1 week after first tPA treatment showed reduced RV dilation and improved RV function, estimated RVSP decreased from 60 to 32 mmHg. Discharged home on enoxaparin and did not require supplemental oxygen
7	65 F	Sickle cell anemia, PH, history of catheter-associated DVT on warfarin, and on home oxygen (2 I/m in)	Dyspnea on exertion, SpO2 87% on 2 l/min O2; right scapular pain	V/Q: Mismatched segmental perfusion defects bilaterally and global decrease in lung perfusion	Severe RV dilation with RV dysfunction, both worse compared with baseline	NT-proBNP: 6239 pg/ml (0–124 pg/ml); troponin T: 0.021 ng/ml	6 mg/day of tPA for 3 days via central venous catheter (1 mg/h for 6 h)	UFH	72 h following presentation, repeat V/Q scan showed improvement in segmental defects bilaterally as well as improvement of global right lung perfusion. Discharged home on enoxaparin and baseline amount of supplemental oxygen (SpO2 95–100% on NC at 2 l/min O2)
8	52 F	Cushing’s syndrome POD 22 s/p laparoscopic right adrenalectomy	Dyspnea, SpO2 78% on room air and difficulty ambulating	V/Q: Numerous, bilateral mismatched segmental and subsegmental perfusion defects; Doppler US: Left LE DVT	Moderate RV dilation with RV dysfunction; McConnell’s sign noted	NT-proBNP: 2390 pg/ml (0–124 pg/ml); troponin T: 0.048 ng/ml	Day 1: 2 mg of tPA administered by pulse spray catheter injection to right and left main PA followed by 6 mg of tPA via PA pigtail catheter (1 mg/h for 6 h); Day 2: 0.5 mg of tPA to right PA and 0.5 mg tPA to left PA followed by 9 mg of tPA to LLE DVT by pulse spray catheter injections	UFH	6 h following initiation of tPA, PA pressures decreased from 59/29 to 39/18 mmHg. 72 h following presentation, repeat TTE demonstrated improved RV function. Discharged home on warfarin and did not require supplemental oxygen
9	29 M	Metastatic papillary renal cell carcinoma POD 2 s/p left total nephrectomy with RP and pelvic lymph node dissection complicated by aortic injury	Sudden onset of dyspnea, tachycardia, and SpO2 89% on 6 l/min O2	CTA: Right-sided segmental PE in the middle and lower lobes	Mild RV dilation with RV dysfunction	NT-proBNP: 1254 pg/ml (0–124 pg/ml); troponin T: 0.01 7 ng/ml	Day 1: 2 mg of tPA administered by pulse spray catheter injection to right PA followed by 4 mg of tPA via right PA pigtail catheter (1 mg/h for 4 h); Day 2: 6 mg of tPA via central venous catheter (1 mg/h for 6 h)	UFH	10 h following 1st tPA dose, PA pressures decreased from 58/20 to 39/22 mmHg and symptoms greatly improved; repeat TTE showed persistent RV dysfunction. 4 h following infusion of 2nd tPA dose, patient developed hypotension and abdominal distension/pain requiring emergent surgery for hemorrhage at the site of RP lymph node dissection. UFH was reversed with protamine. Subsequently received prophylactic dose UFH with no recurrence of PE and uneventful recovery. Discharged home on treatment dose enoxaparin and did not require supplemental oxygen

BNP, brain natriuretic peptide; CNS, central nervous system; CTA, computed tomography angiography; DVT, deep vein thrombosis; HSCT, hematopoietic stem cell transplant; LE, lower extremity; LLE, left lower extremity; LLL, left lower lobe; NC, nasal cannula; NT, not tested; PA, pulmonary artery; PE, pulmonary embolism; PH, pulmonary hypertension; POD, postoperative day; RLL, right lower lobe; RP, retroperitoneal; RUL, right upper lobe; RV, right ventricle; RVSP, right ventricular systolic pressure; SpO₂, pulse oximeter oxygen saturation; tPA, tissue plasminogen activator (alteplase); TTE, transthoracic echocardiogram; UFH, unfractionated heparin; V/Q, ventilation/perfusion scan.

^aNormal range for NT-proBNP changed with change in assay, normal range shown in (). The normal range for troponin I is 0–0.045 ng/ml and for troponin T is 0–0.009 ng/ml. The assay for cardiac troponin was changed from troponin I subtype to troponin T during the timeframe of these case presentations.

^bUnfractionated heparin was administered concomitant with alteplase with a target activated partial thromboplastin time (apt) of 50–70 s or a target anti-Xa level of 0.2–4 IU/ml. Argatroban was administered concomitant with alteplase with a target aPTT of 50–70 s.

Each patient received 10 mg/day or less of alteplase (reconstituted and diluted to 0.1 mg/ml in normal saline) administered through central venous (five patients) or pulmonary artery catheters (four patients). The latter patients underwent pulmonary artery catheterization and direct intraclot injection of small starting doses (i.e., 2 mg) through four French pulse spray catheters, followed by infusion of the remaining alteplase dose (up to 6 mg) through a pigtail catheter positioned in the affected pulmonary artery. All patients received concomitant anticoagulation with heparin or argatroban.

All patients demonstrated modest to substantial improvement in symptoms, pulmonary perfusion, pulmonary artery pressures, and/or RV function within 72 h of initiation of treatment (4/9, ≤24 h after one infusion; 8/9, ≤48 h after one or two infusions). For case 4 with only slight echocardiographic improvement at 48 h, estimated RV systolic pressure decreased from 101 to 48 mmHg at 1 week, and computerized tomography (CT) angiogram showed complete radiographic resolution of pulmonary embolism after 1 month. Case 5 with a large ‘saddle’ embolus needed only one treatment of alteplase (total dose 10 mg) with marked improvement overnight and no need for supplemental oxygen by 36 h. In two cases with both pulmonary embolism and acute deep vein thrombosis (DVT), pulse spray catheters were used to treat large pulmonary emboli and then redirected to inject alteplase into large DVTs of the left leg (case 6, Day 2; case 8, Day 2).

Although the majority of patients obtained some symptomatic relief within 24 h, ventilation/perfusion scans, and/or echocardiography examinations were not repeated frequently enough to closely associate these clinical benefits with decreased clot burden and/or improved RV function. Repeat computed tomography angiography was not routinely employed and was clinically indicated for only three of our patients (1, 3, and 6), usually to assess concomitant problems such as pleural effusions or cancer progression. However, pulmonary artery catheter systolic pressure measurements obtained in three cases (#6, #8, and #9) treated with directed alteplase infusions demonstrated reductions of 13 mmHg at 6 h, 20 mmHg at 6 h, and 19 mmHg at 10 h, respectively. All seven patients not requiring supplemental oxygen before their pulmonary embolism were discharged home on room air.

Alteplase was well tolerated without bleeding complications, except case 9. This patient, only 2 days after left nephrectomy, improved symptomatically with 2 mg of alteplase administered by pulse spray catheter injection of clot in the right pulmonary artery, followed by 4 mg via a right pulmonary artery pigtail catheter (1 mg/h for 4 h). However, a repeat echocardiogram still demonstrated persistent RV dysfunction and only a minimal decrease in estimated RV systolic pressure. Therefore, it was decided to administer a second infusion of low-dose alteplase, but the pulmonary artery pigtail developed a leak and was replaced by a central venous line. Four hours after completion of this undirected alteplase infusion, the patient developed hypotension and abdominal distension. A large hematoma was evacuated at a lymph node dissection site without evidence of active bleeding.

Discussion

An ideal treatment for pulmonary embolism would be safe, lyse emboli quickly, and be simple to administer. High-dose alteplase (100 mg/2 h) is effective, and indicated for massive pulmonary embolism, but not submassive pulmonary embolism, for which the risk of bleeding exceeds the more modest benefits. Although lower doses of alteplase might provide a more favorable risk/benefit ratio, alternative very low dose regimens for submassive pulmonary embolism have not been investigated adequately. Effective fibrinolysis (especially with low alteplase doses) should be guided by the biochemical properties of alteplase and the effect emboli have had on pulmonary artery blood flow. Alteplase binds clot surfaces due to its strong affinity for fibrin. When thrombi are small, a simple brief infusion may be highly effective because the surface area to volume ratio is high. As the thrombus becomes larger, this ratio becomes less favorable and pulse spray catheter injection may be needed to drive alteplase into the thrombus and thereby potentiate lysis.

Ideally, alteplase would be delivered to each pulmonary artery in proportion to the clot burden. Ventilation/perfusion scans show blood flow is diminished by pulmonary embolism, with greater blood flow diverted to arteries with less emboli. Therefore, central catheter infusions are likely to be most effective in cases with diffuse, equally distributed clot burden. In such patients, tedious pulse spray catheterization of multiple vessels is probably unnecessary and central venous administration of alteplase may suffice. In case 1 with a solitary lung and a central, large embolus, either option would likely be effective and the risk of inducing an arrhythmia in the presence of cardiomyopathy weighed against pulmonary catheterization. In absence of other considerations, when emboli are large and centrally located (cases 5, 7, and 8) or predominantly located in only one lung (case 9), pulmonary catheterization and pulse spray intraclot injection ensures adequate delivery of alteplase to the clot. Direct pulmonary artery administration might also be indicated in patients with large right to left intracardiac shunts, as alteplase given by central venous infusion is likely to bypass the pulmonary circulation. Notably, case 5 was found to have right to left shunting through a patent foramen ovale, which is present in ~25% of the population [17,18], and may open due to pulmonary embolism-induced increases in right atrial pressure.

Safety is a critical factor to consider for fibrinolytic therapy. Improving the safety of thrombolytic therapy requires more than simply reducing the dose. Despite a 50-fold reduction in alteplase infusion rate from 100 mg/2 h for treating massive pulmonary embolism to the 1 mg/h infusions used here and in other low-dose treatment studies, circulating alteplase levels increase substantially and its physiologic inhibitor, plasminogen activator inhibitor 1 (PAI-1) becomes undetectable [19,20]. One advantage of short duration infusions of alteplase may be the ability to reform beneficial clots between treatments, thereby reducing the risk of bleeding. Between infusions, when alteplase disappears and PAI-1 rebounds to levels above baseline, fibrin clots can be restored to some extent, inversely proportional to the level of anticoagulation. Although our alteplase infusion rate (1 mg/h) was similar to the rate used in the Seattle II study, 2 mg/h for 12 h or 1 mg/h for 24 h [21], the duration of infusion used here was significantly shorter. Therefore, even though Seattle II excluded patients who underwent surgery within the prior 7 days, nonintracranial hemorrhages requiring 2 or more units of packed red cells still occurred in 10% of their patients.

Intraclot injection of alteplase accelerates clot lysis at the target site and decreases (but does not eliminate) undesired off-target clot lysis, particularly at low doses (e.g., 2–4 mg) that we employed in select cases. Therefore, in patients who need to maintain beneficial clots at surgical sites, intraclot injections may improve the margin of safety. In patients not needing to maintain clot integrity at noncompressible sites, short course central venous infusions of low-dose alteplase may be safe, effective, and relatively easy to administer. For case 9, with the only major bleeding adverse event in our series, there was no bleeding at incisions or central line or pleural catheter insertion sites. Bleeding was restricted to the lymph node dissection bed, which lacked sutures, staples or devices to provide physical hemostasis by tamponade. Although symptomatically improved after one dose of alteplase, this patient received a second infusion and 4 h thereafter developed bleeding. In retrospect, central venous administration, rather than therapy directed to the side of clot as was done initially, may have contributed to this complication. More importantly, the second infusion may have been unnecessary in this high-risk surgical patient. Starting 24 h after laparotomy for hemorrhage, the patient was treated with prophylactic dose enoxaparin and was eventually transitioned to full-dose enoxaparin before discharge with no further thrombosis or bleeding.

Our report is limited by the lack of an anticoagulation alone control group (standard therapy), like the Seattle II study, so we cannot attribute any benefits to fibrinolysis alone. Also similar to Seattle II and in contrast to the ‘watch and wait’ approach [8], low-dose alteplase was initiated in our patients without evidence of hemodynamic compromise. Our report is also limited by the fact that there was no control group (e.g., anticoagulation alone) and the possibility of selection bias, since most of our cases had cancer. Further, there was variability in how alteplase was administered (systemic vs. catheter directed in lower extremities vs. catheter directed in pulmonary circulation). Other institutions may have different outcomes that are not reported in the literature. Nonetheless, our series suggests that low dose, brief duration infusions of alteplase may improve the efficacy of anticoagulation alone for submassive pulmonary embolism, without conferring a high risk of bleeding, particularly in patients who have not had recent major surgery. We encourage larger, controlled studies that are necessary to determine whether low-dose alteplase for submassive pulmonary embolism has an acceptable risk/benefit ratio and provides net long-term benefits for patients [22,23].