Bleeding is the main CDT complication and occurred in 5% to 10% of cases with most located at the venous access site [34,35].
The review discusses the potential factors that may improve safety and efficacy during thrombolysis procedures.
We divided these factors into eight factors; patient selection criteria, fibrinolytic drugs, mode of fibrinolytic drug injection, biochemical markers, timing of intervention, intermittent pneumatic calf compression (IPCC), ward monitoring and thrombolysis imaging assessment.
1) Patient selection criteria
Most of the bleeding complications probably occur as a result of action of thrombolytic drugs at sites of vascular injury or malformation. This is why most of the studies agree that strict exclusion criteria for local thrombolysis can improve safety and avoid bleeding complication.
Society of Intervention Radiology (SIR) [36] produced its standards for endovascular thrombus removal of lower limb DVT and divided the contraindication to pharmacologic CDT into absolute contraindication and relative contraindications:
➀ Absolute contraindicationsActive internal bleeding or disseminated intravascular coagulation.
Recent cerebrovascular event (including transient ischemic attacks), neurosurgery (intracranial, spinal), or intracranial trauma (less than three months).
Absolute contraindication to anticoagulation.
➁ Relative contraindicationsRecent cardiopulmonary resuscitation, major surgery, obstetrical delivery, organ biopsy, major trauma, or recent eye surgery (less than seven to ten days), intracranial tumour, other intracranial lesion, or seizure disorder.
Uncontrolled hypertension: systolic >180 mmHg, diastolic >110 mmHg.
Recent major gastrointestinal bleeding (less than three months).
Serious allergic or other reaction to thrombolytic agent, anticoagulant, or contrast media (not controlled by steroid/antihistamine pretreatment).
Severe thrombocytopenia.
Known right-to-left cardiac or pulmonary shunt or left heart thrombus, massive PE with hemodynamic compromise.
Suspicion for infected venous thrombus.
➂ Other relative contraindicationsRenal failure (estimated glomerular filtration rate <60 mL/min), pregnancy or lactation, severe hepatic dysfunction, bacterial endocarditis and diabetic hemorrhagic retinopathy.
Most of the trials advice similar contraindication, however, Cavent and adjunctive catheter-directed thrombolysis trials [8] added others exclusions criteria in addition to the one mentioned by SIR like life expectancy less than 2 years, chronic non-ambulatory status and haemoglobin less than 9; international normalized ratio more than 1.6 before warfarin.
2) Types of fibrinolytic drugs
All of the plasminogen activators share the potential of inducing plasmin action on fibrin, with an associated greater or lesser effect on plasma fibrinogenolysis (lytic state).
Currently approved fibrinolytic drugs include streptokinase, anistreplase, urokinase, recombinant tPA, two recombinant derivatives of tPA, reteplase [37] and tenecteplase [38].
There are small numbers of studies to compare different fibrinolytic drugs regarding safety and efficacy. Grünewald and Hofmann [11] is a retrospective single center study of 72 patients (82 limbs) study comparing alteplase, reteplase and urokinase found no difference between safety and efficacy of the 3 drugs. However, alteplase and reteplase were significantly less expensive than urokinase (P<0.001 and P<0.01, respectively), the same result has been confirmed by Sugimoto et al. [39] which found no statistical difference between altepase and urokinase thrombolysis success rates. However, tPA was significantly (P<0.05) less expensive and faster than urokinase.
The use of tenecteplase in the peripheral system is few, and these studies are not well controlled. Assent-2 trial [41] is a double-blinded RCT comparing between single bolus tenecteplase (30–50 mg according to bodyweight) and less than 6 hour duration rapid infusion of alteplase (≥100 mg) for treatment of acute myocardial infarction showed that tenecteplase has fewer non-cerebral bleeding (26.43% vs. 28.95%, P=0.0003) and less need for blood transfusion (4.25% vs. 5.49%, P=0.0002) than those treated with altepase.
From the data above, we can see that there are not enough studies to compare between fibrinolytic agents. However, the high specificity of alteplase and tenecteplase [40,41] to fibrin make them theoretically more safe and efficient than urokinase and streptokinase as we avoid systemic lytic effect.
The longer half-life of tenecteplase, high resistant to inactivation by PAI-1 and the less affinity to DD may lead to improving safety by avoiding systemic lytic effect and at the same time theoretically we can give it as a single bolus that will be more comfortable as the patient is not bed bound and can start mobilizing after the end of thrombolysis. However, there is a lack of the studies comparing between tenecteplase and other tPA in the management of acute lower limb DVT, although, it shows promising results in the management of acute myocardial infarction.
3) Mode of fibrinolytic drug injection
There are various techniques for fibrinolytic drug administration including systemic thrombolysis, continuous CDT and single bolus CDT.
In this review, we compare the three different techniques in view of safety and efficacy.
➀ Systemic thrombolysisInitial attempts to treat acute DVT with thrombolytic therapy were by the peripheral administration.
Camerota and Kagan [35] has reported thirteen studies since 1968, pool analysis shows that there is complete or significant lysis in 45% of cases, partial lysis in 18% None or worse in 37%.
Out of the thirteen studies, 11 studies discussed bleeding complication with systemic thrombolysis and showed that minor bleeding complication happened in 28.5% of cases and major bleeding occur in 26.23%.
➁ CDT (continuous infusion)Pool analysis of 14 studies [12–22,24] (440 patients) including Cavent trial [8] for patients who received only CDT using either urokinase and tPA from 1994 for acute lower limb DVT shows complete lysis in 64% of limbs treated, partial lysis in 29.55% and no lysis in 4% of cases. Minor bleeding and major bleeding complication in 5.7% and 4% of cases respectively.
➂ Single bolus CDTThe strong fibrin affinity of recombinant tissue plasminogen activator tPA theoretically obviates continuous infusion or replacement of tPA after direct intrathrombic injection.
Chang et al. [25] is a cohort (non randomized) pilot study of 12 patients with acute DVT ranging from inferior vena cava thrombosis to bilateral calf vein thrombosis to unilateral popliteal and calf vein thrombosis, which evaluate single daily catheter-directed injection of tPA as a thrombolytic treatment for acute DVT of the lower extremity. Significant or complete lysis was achieved in 11 of the 12 extremities, and one has 75% lysis. Although the average total dose of altepase was 106 mg, bleeding complications were minor. No patient had a decrease in hematocrit of greater than 2% or decrease in hemoglobin of more than 1 g, and no patient required blood transfusion.
In 2011, Lozier et al. [26] conducted a prospective study of 30 patients using single bolus tPA (maximum dose used was 10 mg of alteplase per dose), and inherited thrombophilic traits were identified in 13 patients (43%) and 7 patients has iliofemoral DVT, and the remaining 23 patients (77%) had femoral popliteal DVT.
Venograms performed the day following last thrombolytic treatment showed that antegrade venous flow was restored in 29 of 30 patients (97%) using an average total dose of 19.7 mg (range, 8–38 mg) of alteplase over an average of 2.7 treatments (range, 1 to 4 treatments or days) for an average dose of 7.3 mg tPA/day. There were no major bleeding complication and minor bleeding in 3/30 (10%) had hematoma at catheter insertion site.
Analysis of the previous data shows that, CDT has better clinical outcome than systemic thrombolysis with significant decrease in risk of both minor bleeding (RR, 0.38; OR, 0.34) and major bleeding (RR, 0.05; OR, 0.12) (Fig. 3).
Single bolus CDT shows promising results in acute iliofemoral DVT, as it can help to reduce the cost of treatment with same efficacy and also more comfortable as the patient is not bed bound and can start mobilizing one hour after the end of thrombolysis; however, these studies may be biased (moderate risk of bias) as good outcome may be related to including cases with calf vein and popliteal vein thrombosis (low thrombus burden).
4) Biochemical markers
The role of haemostatic parameters measurement in the prediction of bleeding complication and treatment outcome during thrombolysis is not clear yet.
Endogenous fibrinolysis is regulated at two levels. PAIs, particularly the type 1 form (PAI-1), which prevent excessive plasminogen activation by regulating the activity of tPA [42–44].
PTTs, fibrinogen and DD level are the main haemostatic parameters measured during thrombolysis in most of the studies.
There are few other studies discussing the role of others haemostatic parameters like PAI-1, systematic tPA, plasminogen and alpha2–anti plasmin activity.
➀ D-dimerD-dimer (DD) is one of the Fibrin degradation products and resulted from the action of tPA on fibrin [45–47].
DD has been identified in the blood of patients with various thrombotic or thrombolytic disorders [46].
DD level during thrombolysis procedure has 3 phases:
Phase 0: DD level rises abruptly by the time tPA administration is complete.
Phase 1: DD reaches a plateau level that indicates continuous thrombolysis. Phase 1 duration in continuous CDT is up to 2 days while in single bolus CDT is 8 hours [26].
Phase 3: DD level significantly decrease at the third day of continuous CDT in most of the studies and after 8 hours in single bolus thrombolysis [26].
Failure to achieve the initial peak (phase 0) of DD level may indicate failing of thrombolysis therapy. However, coordination with venogram finding is required.
Grünewald et al. [9] discussed the correlation between clinical events and hemostatic parameters during systemic thrombolysis in lower limb DVT and shows that persistent high level or plateau of DD and fibrinogen degradation products (FDPS) during thrombolysis indicate either;
Increase bleeding tendency and it is supported by the biochemical fact that DD has affinity as potent as fibrin as a stimulator of plasminogen activation by alteplase. So persistent high DD leads to increase plasminogen and alteplase activity and in turn increase bleeding tendency, this finding is supported by Bovill et al. [10] that correlate between bleeding incidence and haemostatic parameters and found that there were a correlation between high peak DD level and bleeding tendency (P=0.007) during systemic thrombolysis in acute myocardial infarction.
Resistant thrombus as the patient most probably will be procoagulant and interpretation with venogram result needed to confirm it.
➁ FibrinogenThrombolysis agent has local and systemic fibrinolytic activity, although CDT act locally on fibrin but also thrombolytic agent can escape systemically and cause lysis of soluble fibrinogen.
There are a small number of retrospective and prospective non-randomized studies trying to link between fibrinogen levels and bleeding complication.
Grünewald et al. [9] found that there is no relationship between fibrinogen level and bleeding incidence or treatment success (P=0.06).
Bovill et al. [11] found that that low nadir fibrinogen levels associated with bleeding complication (P=0.005) and recommended to keep fibrinogen level between 100–150 mg/dL.
Vandelli et al. [48] discussed the correlation between increase of the risk of intracerebral hemorrhage after intravenous thrombolysis for acute ischemic stroke and low fibrinogen level. Fibrinogen levels were determined at 2 hours after therapy: patients were classified as belonging to ‘low fibrinogen group’ if levels decreased to less than 2 g/L and/or by 25% or more. Bleeding rate in the low fibrinogen group was significantly higher (43.9%) than that in the normal fibrinogen group (9.5%; OR, 7.43; P<0.001). The result of this study may not correlate with the DVT thrombolysis and intracerebral bleeding risk may be high due to nature of the disease (ischemic stroke).
In view of the previous studies, there is no clear evidence to correlate between fibrinogen level and bleeding incidence or treatment success in management of iliofemoral DVT.
➂ Plasminogen activator inhibitor-1Experimental study was conducted by Carmeliet et al. [49] who observed PAI-1 level on mice blood after PAI-1 gene alteration and relation to resolving jugular vein induced thrombus. He found that disruption of the PAI-1 gene in mice (low PAI-1 level) appears to induce a mild hyperfibrinolytic state and a greater resistance to venous thrombosis but not to impair hemostasis.
Grünewald et al. [9] observed that during continuous systemic tPA infusion there was significant decrease of PAI-1 from day 3 to day 5 and PAI-1 level was significantly lower in patients with bleeding comparing to those without (P=0.01), suggesting that serial measurements of PAI-1 might help to predict bleeding and prevent it (level 2 evidence).
The effect of single bolus tPA on PAI-1 level was discussed by Lozier et al. [26] and shows that PAI-1 fell to essentially undetectable levels immediately after completion of alteplase administration. As tPA activity decreased over the following one to two hours, mean PAI-1 levels rose rapidly and by the eight-hour time point were three times the baseline values (Fig. 3).
Other study conducted by Grünewald et al. [24] comparing locoregional to systematic thrombolysis observed that there is a rapid decrease in PAI level to immeasurable value at the time of thrombolytic agent injection (1st therapeutic cycle) with gradual increase in value of PAI-1 over the following cycles to reach normal or elevated level at therapeutic cycle three.
PAI-1 level remains low through the treatment in continuous tPA infusion which according to Carmeliet et al. [49] can induce a mild hyperfibrinolytic state and a greater resistance to venous thrombosis but not to impair hemostasis.
However, if PAI-1 levels persistently decrease significantly (reach zero level), bleeding is likely to happen.
In single bolus altepase, the rebound increase in PAI-1 level either locoreginally or systemically to normal or three fold of normal level may lead to increase safety by opposing the active tPA but also may affect negatively the thrombolysis process.
So, monitoring and maintaining PAI-1 level at low level (above zero level) may improve both safety and efficacy of CDT (level 3 evidence).
➃ Activated partial thromboplastin timeSIR [36] recommended blood draws for hematocrit, platelet count and PTT at least every 12 hours.
During CDT, the intravenous heparin dose should be adjusted to keep aPTT at 1.2 to 1.7 times prolongation, that is, at 40s to 60s [8].
Proper matching of the anticoagulation level to each patient according to bleeding risk should be considered. For example, young, healthier patients can tolerate more robust heparin and rTPA than elderly or debilitated patient. However, we have to make sure that aPTT level values are not above therapeutic level during thrombolysis to avoid unnecessary bleeding [36].
5) Ward monitoring
CDT is a safe procedure but ward monitoring is one of the important factors that can help to improve safety and avoid the need of post lysis HDU admission, which in turn will help to decrease the cost of the procedure.
SIR [36] recommended the following measures to prevent bleeding complication such as complete bed rest with immobility of the catheter-bearing extremity, frequent contact with nursing staff, blood draws for hematocrit, platelet count, fibrinogen and aPTT at least every 12 hours, consider potential markers of impending bleeding like pericatheter oozing, minor sentinel bleeds (e.g., epistaxis), and elevated aPTT, and it should be confirmed that arterial punctures and intramuscular injections did not occur during thrombolysis.
Bækgaared et al. [50] is a retrospective analysis of Copenhagen experience (89 patients with 91 limbs) suggest that in addition to same recommendations by society of interventional radiology absolute bed rest with IPC and careful observation of vital signs and bleeding signs every eight hours by dedicated nurse (Table 1).
6) Timing of intervention
There is no formal definition of an acute or chronic DVT.
The speed of intervention in acute thrombotic events is of clinical relevance as there is an increase possibility to reverse the occlusion, relief of symptoms, and preservation of valve function, which helps to decrease the risk of post-thrombotic syndrome (Fig. 4).
It is known that acute thrombi respond better to thrombolysis compared to established DVTs (86% vs. 68%, significant grade II or III lysis; 34% vs. 19%, grade III lysis) due to thrombus organisation over time several studies and international guidelines recommended that CDT should be performed within 14 days of onset of symptoms [26,35]. However, the quality of evidence supporting early thrombus removal strategies is very low (level 3 evidence) because of the methodological limitations of the relevant studies (lack of randomization, incomparability of study groups, loss to follow-up).
National Venous Registry suggested that patients with symptoms more than ten days in duration had significantly worse outcomes than those with a first episode of acute iliofemoral DVT of less than ten days in duration. However, the 10-day interval of symptoms was arbitrary as symptom duration among those symptomatic for 10 days varied from days to many months [13]. Cavent trial treated patients with symptoms less than 21 day [8].
Arnoldussen et al. [28]; a prospective study with a total of 53 cases of DVT to identify DVT characteristics with contrast-enhanced MRV (Fig. 4) comparing it with average duration of complaints and was able to classify DVT into acute 2–13 days, subacute 8–18 days, chronic 15–32 days.
Bækgaard et al. [29] has treated the 53 cases with thrombolysis and found that acute thrombus will lyse quickly, subacute thrombus will lyse within reasonable time and chronic will either lyse little or not at all, even in an extended time frame.
We recommend treating the iliofemoral DVT within 14 days of diagnosis. However, the earlier start of treatment the better as it helps to decrease total tPA dose needed to treat the DVT which in turn reflect on procedure safety and efficacy.
7) Intermittent pneumatic calf compression
During CDT, patient has to be immobilised during management to prevent moving of the catheter from thrombus site and avoid bleeding from puncture site. Immobilisation lead to static flow due to decrease venous return, which in turn may lead to further thrombosis despite adequate thrombolytic therapy.
IPCC can improve venous blood flow during continuous thrombolysis, which may help to improve CDT outcome and decrease amount of thrombolytic agent needed for thrombolysis.
Ogawa et al. [27] evaluated the effects and safety of CDT with IPC for acute proximal DVT compared with CDT alone revealed that adding IPC to CDT using low-dose urokinase for DVT treatment of the leg resulted in better early and late outcomes compared with CDT alone and was not associated with an increased risk of symptomatic PE. The overall effect of thrombolysis for CDT with IPC was better than for CDT alone (P=0.0037).
There are no other studies discuss the effect of IPC on CDT from safety and efficacy.
However, the principle of IPC to improve venous outflow and prevent stasis in immobile patient during thrombolysis may help to improve outcome and decrease total tPA dose needed for thrombolysis (level 3 evidence).
8) Thrombolysis imaging assessment
Clot lysis may be quantified and stratified according to the percentage of venous luminal patency restored.
The difference between the pre- and post-lysis thrombus scores divided by the pre-lysis score gave the grade of thrombolysis; grade I ≤50%; grade II=50%–90%, and grade III=complete thrombolysis [14].
Venography is the main assessment image for thrombolysis procedure to detect any stenotic or occlusive iliac lesions that need stenting.
It is known that IVUS in chronic patients is an excellent imaging modality to identify the extension of the intra-luminal and mural lesions, which can be missed on a single-plane venography. However, it has not been recommended in any guidelines in the management of acute iliofemoral DVT, probably due to the lack of data on the use of IVUS in existing publications in patients with CDT.
Neglén and Raju [30] is a cohort study comparing between IVUS and standard, single-plane, transfemoral venography were performed in 304 consecutive limbs during balloon dilation and stenting of an obstructed iliac venous segment and showed that venography had poor sensitivity (45%) and negative predictive value (49%) in the detection of a venous area stenosis of >70% when compared to IVUS as the standard, the actual stenotic area was more severe when measured directly with IVUS (P<0.001), probably as a result of the non-circular lumen geometry of the stenosis.
IVUS has many other potential advantages than can help to provide more information before stenting such landing zone for the distal part of the stent plus ensure the optimal lumen restoration.
Also it provides more adequate morphological information, which can help to identify post-thrombolysis residual thrombus and underlying cause of acute DVT, that can not be identified by single plan venography, like trabeculation, frozen valves, mural thickness, and outside compression.
This can help to identify the early need for stenting of this lesions to prevent recurrence and improve safety by avoid unnecessary thrombolytic drugs injection.
PDF
Standard view
Export citation
Share 
Download
Download
Download
Download
Download
