The use of thrombolytic therapy in pregnancy

Abstract

The relative hypercoagulable state of pregnancy leads to an increased risk of thrombotic complications, of which some may be life-threatening or medically devastating. In the non-pregnant patient, the current guidelines suggest thrombolysis as the primary treatment in acute ischemic stroke, myocardial infarction when percutaneous intervention is unavailable, certain cases of mechanical valve thrombosis, and pulmonary embolism with hemodynamic compromise or shock. Given that clinical trial data regarding thrombolytic use in pregnant women are absent due to exclusion, the goal of this review is to summarize the available published data regarding the use of thrombolytic agents and subsequent outcomes and complications in pregnant women. Overall, the use of thrombolytic agents in pregnancy is associated with a relatively low reported complication rate, especially given the severe medical conditions for which they are indicated. The data would suggest that thrombolysis should be considered for appropriate indications similar to that of non-pregnant patients. However, caution should be exercised when drawing conclusions regarding maternal and fetal safety, given the lack of controlled clinical trials including pregnant women and the nature of the weak evidence level of the cumulative data presented in this review.

Coagulation in pregnancy and risk of thrombotic complications

Many significant physiologic changes occur in hemostasis during pregnancy that prevent substantial hemorrhage during delivery and allow for expansion of the maternal–fetal circulations at the utero-placental interface. Unfortunately, these necessary changes also are responsible for an increased risk of thrombotic complications during and after the pregnancy. During the pregnancy, and peaking at term and the immediate peri-partum period, there is a state of relative hypercoagulability – marked by an increase in the majority of pro-clotting factors, a reduction in natural thrombolytic activity, and a reduction in anti-coagulant protein activity (e.g. activated protein C, protein S). Following delivery, these coagulation changes cease and normal hemostasis generally returns 3–4 weeks following delivery.^1–3

Thrombotic complications during pregnancy are among the leading causes of maternal mortality worldwide, and are the main cause in many developed countries.^4,5 The hypercoagulable state of pregnancy augments the risk of certain complications more than others, with the most marked effect seen with venous-thromboembolic events (VTE) (Table 1).

Table 1.

Risk of thrombotic complications in pregnancy.

Thrombotic complication	Incidence or relative risk in pregnancy	Incidence in non-pregnant women	References
Deep venous  thrombosis	0.615/1000 (age <35)	0.11–0.41/1000	6,7
	1.216/1000 (age >35)
Pulmonary  embolism	0.33–1/1000	0–0.2/1000a	7,8
Cerebral  infarction	RR 0.7 (antepartum)	0.11/1000	9,10
	RR 8.7 (post-partum)
Artificial valve  thrombosis	39/1000 (warfarin)b	3–13/1000b	11,12
	92/1000 (heparin)b
Myocardial  infarction	0.1/1000	0.05/1000	10,13

^aAge-dependent incidence (essentially zero incidence in women <18).

^bIncidence expressed in those with an artificial cardiac valve.

Therapeutic standards of care and the use of thrombolysis in non-pregnant patients

The range of therapies available for thrombotic diseases has evolved greatly over the last several decades, and many strategies now include the use of systemic or locally directed thrombolytic agents. Given the devastating hemorrhagic complications that can accompany the administration of thrombolytics, the decision to employ them requires careful weighing of the known risks and benefits. In some conditions, there exists strong evidence to support usage as the primary treatment, while in others thrombolysis is reserved for those patients with hemodynamic compromise or imminent risk of death.

In acute ischemic stroke, intravenous thrombolysis is indicated as the primary therapy within a prescribed symptom-onset window, and has been shown to significantly increase the chances of complete or near-complete recovery at three months.¹⁴ For acute myocardial infarction (MI), the use of systemic thrombolysis is still common in regions without ready access to percutaneous intervention technology, and its expedient use is associated with a very significant reduction in mortality.¹⁵ In cases of thrombotic obstruction of prosthetic cardiac valves, due to the appreciable surgical mortality associated with this condition, the use of systemic thrombolysis is often chosen – especially in those who are at high risk for surgery or have right-sided prosthetic valves.¹⁶

Unfortunately, the evidence for the benefit of thrombolysis in acute VTE has not been demonstrated as clearly. Despite the paucity of solid evidence, all guidelines agree that thrombolysis should be given in acute pulmonary embolism presenting with significant hypotension or evidence of shock.^17,18 When the much more common presentation of hemodynamically stable VTE is encountered, the evidence to support the use of thrombolytics remains uncertain. A meta-analysis of available trials showed a non-statistically significant improvement in mortality and a significant increase in major bleeding events with the use of thrombolytics plus anticoagulation versus anticoagulation alone – which led to the recommendation against their use in patients without evidence of hypotension or shock.¹⁸ It should be noted, however, that there was a large difference in major bleeding events depending on bleeding risk (6.2% versus 0.1%, high versus low-risk) – with high risk defined by a point system scoring for recent bleeding, anemia, renal insufficiency, prior pulmonary embolism (PE), cancer, and age >75.¹⁹ Finally, there is debate regarding whether thrombolysis improves right ventricular function and pulmonary hemodynamics in a sustained fashion, and some have argued for this consideration to prevent the development of possible long-term altered cardio-pulmonary hemodynamics.^20,21

Pharmacology and teratogenicity of thrombolytic agents

Given the significant complications that potentially can occur following the administration of thrombolytic agents, there understandably is concern regarding their effects on fetal outcomes. There are no formal studies examining the effects of thrombolytics in human pregnancy outcomes, and pregnancy is a universal exclusion criterion in their clinical trials. Therefore, the basis for fetal safety must rely on what is known regarding drug entry into the placenta. Several factors determine transfer of molecules to the placenta – lipid solubility, pH, protein binding, and most importantly, molecular weight – with drugs with molecular weights greater than 1000 Da crossing poorly into the placenta.²² As such, it is thought that all thrombolytic agents do not pass into the placenta (Table 2).

Table 2.

Molecular weights of thrombolytic agents.²³

Thrombolytic agent	Molecular weight (Da)
Alteplase (rtPA)	59042.3
Urokinase	31126.5
Streptokinase	47286.7
Tenecteplase	58951.2
Reteplase	39589.6

Da: Daltons; rtPA: recombinant tissue plasminogen activator.

There is a similar dearth of quality data surrounding the issue of potential teratogenicity of these agents, although similar arguments as above suggest there is not a potential for effects. The known fetal outcomes reported in case reports and case series are presented below – and overwhelmingly show no obvious teratogenetic effects. Likewise, there is no evidence in animal studies to suggest adverse fetal effects due to these agents.²⁴

Clinical experience with thrombolytic use in pregnancy

Search strategy

As stated previously, pregnant women are uniformly excluded from participation in clinical trials involving thrombolytics. As such, the information below represents the known reports and series discussing the use of thrombolysis in various conditions. A systematic review of the use of thrombolytics in pregnant women was performed by searching the electronic databases MEDLINE (from 1946) and the Cochrane Library up to March 2013. The search included MeSH terms ‘thrombolytic therapy’, ‘pregnancy’, and ‘tissue plasminogen activator’; and included also were the search terms ‘pregnan*’, ‘streptokinase’, ‘urokinase’, ‘tPA’, ‘tenecteplase’, ‘alteplase’, ‘thrombolytic’, ‘thrombolysis’, and ‘tissue plasminogen activator’. The references listed in the full text articles were reviewed to ensure additional citations were not missed. All publication types that reported the use of a thrombolytic agent for any indication in a pregnant woman, and described the outcome, were included in the review. In sum, a total of 56 articles were found, and included a cumulative total of 231 patients (Table 3). There have been no randomized or prospective trials involving the use of thrombolytics in pregnant women.

Table 3.

Thrombolytic use in pregnancy by clinical indication.

DVT/PEa	N	Thrombolytic	Maternal/fetal complications	Comments
Pfeifer25	12	SK	NA	Cases were DVT
Hall26	1	SK	No long-term complications	Significant uterine hemorrhage; however, SK given for 41 h (including 3 hours after delivery)
Ludwig27	24	SK	No complications	DVT – 21 cases, PE – 3 cases
McTaggart28	1	SK	Mother alive, fetal death in utero	Mother presented in shock
Ludwig and Genz29	122b	SK	2 severe maternal bleeding events; 1 fetal death	All cases were treatment of DVT
Delclos and Davila30	1	UK	No complications
Fagher et al.31	1	SK	Severe maternal bleeding; no fetal complications	SK given directly before and after delivery – total 29 hours.
Baudo et al.32	1	rtPA	No complications
Flossdorf et al.33	1	rtPA	No complications
Seifried et al.34	1	rtPA	No complications
De Stefano et al.35	1	rtPA	No complications	Hepatic and portal vein thrombosis
Mazeika and Oakley36	1	SK	Moderate cervical hemorrhage, normal fetus	Utilized mechanical disruption of PE following SK infusion
Kramer et al.37	1	UK	No complications
La Valleur et al.38	2	UK	No complications	Cases were DVT only
Grand et al.39	1	rtPA	No complications	Acute ileo-femoral vein thrombosis
Saviotti et al.40	1	rtPA	Minor uterine bleeding, no fetal complications
Krishnamurthy et al.41	3	UK	No complications	DVT – 2 cases, PE -- 1 case. UK delivered via directed catheter
Sofocleous et al.42	1	rtPA	Mother alive, fetal death in utero	rtPA administered via catheter after failed catheter embolectomy
Henrich et al.43	1	SK	No complications	Acute severe iliac vein thrombosis
Ahearn et al.44	1	rtPA	No complications
Yap et al.45	1	rtPA	No complications
Patel et al.46	1	rtPA	No complications
Stefanovic et al.47	1	SK	Mild vaginal bleeding, no fetal complications	SK given for PE 12 hours after caesarean delivery
Trukhacheva et al.48	1	rtPA	No complications
Bechtel et al.49	1	rtPA	No complications	rtPA administered via catheter after mechanical fragmentation
te Raa et al.50	1	SK	No complications
Fasullo et al.51	1	rtPA	No complications
Holden et al.52	3	SK/rtPA	No complications	Followed out for 2 years
Lonjaret et al.53	1	rtPA	No complications
Neurologic usesc	N	Thrombolytic	Maternal/fetal complications	Comments
Niwa et al.54	1	rtPA	No complications	Direct treatment of superior sagittal sinus thrombosis via select venography
Elford et al.55	1	rtPA	Mild neurologic deficits, no fetal complications	rtPA administered via directed intra-arterial catheter
Dapprich and Boessenecker56	1	rtPA	Radiographic hemorrhagic transformation of infarction, almost complete recovery at 4 weeks, no fetal complications	No indication of timing of onset of symptoms prior to rtPA administration
Weatherby et al.57	1	rtPA	No complications	Direct treatment of cerebral venous sinus thrombosis
Johnson et al.58	1	rtPA	No complications	rtPA administered via directed intra-arterial catheter

Neurologic usesc	N	Thrombolytic	Maternal/fetal complications	Comments
Murugappan et al.59	8	rtPA/UK	Maternal: 1 death from dissection during angioplasty, 4 minor bleeding events/asymptomatic ICH; 3 fetal deaths/SAB (1 due to maternal death), 3 MTP, 2 healthy infants	2 SABs (6wk in 40yo mother with PCV/ET, first trimester mother with bacterial endocarditis)
Wiese et al.60	1	rtPA	No complications
Leonhardt et al.61	1	rtPA	Maternal: residual effects of CVA; preterm delivery at 32 + 6 weeks – healthy infant at 1 year follow-up
Yamaguchi et al.62	1	rtPA	No complications
Li et al.63	1	rtPA	No complications	rtPA administered via directed intra-arterial catheter
Tassi et al.64	1	rtPA	No complications
Cardiac usesd	N	Thrombolytic	Maternal/fetal complications	Comments
Witchitz et al.65	1	SK	Minor uterine hemorrhage, no other maternal/fetal complications	Mitral valve
Ramamurthy et al.66	1	SK	No complications	Mitral valve
Azzano et al.67	1	rtPA	Uterine hemorrhage and rethrombosis of valve leading to interruption of pregnancy and need for cardiac surgical intervention	Tricuspid valve
Schumacher et al.68	1	rtPA	No complications	Acute myocardial infarction
Fleyfel et al.69	1	rtPA	No complications	Mitral valve
Rinaldi et al.70	1	rtPA	No complications	Aortic valve
Abbadi71	1	SK	No complications	Mitral valve
Anbarasan et al.72	1	SK	No complications	Mitral valve
Nanas et al.73	1	rtPA	No complications	Mitral valve
Behrendt et al.74	1	rtPA	No complications	Aortic valve
Nassar et al.75	1	rtPA/SK	No complications	rtPA administered after presumed rethrombosis occurred after SK use
Sahnoun-Trabelsi et al.76	7	rtPA	Two maternal deaths, one severe non-fatal bleeding complication, three completed healthy pregnancies	Mix of mitral and aortic valve cases. Two deaths presented in shock and had failure of treatment.
Wei et al.77	1	rtPA	No complications	Pulmonic valve
Kaya et al.78	1	rtPA	No complications	Mitral valve
Ozer et al.79	3	rtPA	TIA – complete resolution, transient epistaxis; no fetal complications	Mitral valve
Srinivas et al.80	1	SK	No complications	Mitral valve
Totals (DVT/PE, Neurologic, Cardiac)	231 (189/18/24)	No maternal death from hemorrhage; six major (four SK, two rtPA), 10 minor bleeding events (six SK/UK, four rtPA)

DVT: deep vein thrombosis; PE: pulmonary embolism: SK: streptokinase: UK: urokinase; rtPA: recombinant tissue plasminogen activator; ICH: intracranial hemorrhage; MTP: medical termination of pregnancy; CVA: cerebral vascular accident; SAB: spontaneous abortion; PCV/ET: polycythemia vera/essential thrombocytosis.

^aCases represent PE unless stated in comments.

^bUnclear if this report included population from 1973 reference.

^cTreatment of acute stroke via systemic intravenous thrombolysis unless otherwise stated in comments.

^dTreatment of acute prosthetic valve thrombosis unless otherwise indicated in comments.

Deep vein thrombosis (DVT) and PE

Twenty-nine articles were found discussing the role of various thrombolytic agents in DVT and PE, with a total case population of 189. There exists significant heterogeneity in the thrombolytic agent used, the method of administration (e.g. systemic versus directed therapy), and clinical indication for which they were used. The largest series comprised 122 patients with DVT,²⁹ which found a major bleeding complication rate of 1.6%, no maternal deaths, and one fetal death. In the remaining 67 patients with VTE, there were no maternal deaths, three major bleeding events (all associated with streptokinase), two minor bleeding events (one rtPA, one streptokinase), and two deaths in utero (one after mother presented in shock, one after failed surgical embolectomy and subsequent rtPA administration).^{25–28,30–53}

Acute stroke and other neurologic indications

Eleven articles have been published describing outcomes in 18 pregnant patients who received thrombolytics (majority rtPA) for various neurologic thrombotic complications (acute stroke, venous sinus thrombosis).^54–64 There were no maternal deaths attributed to thrombolytic administration (one death reported due to arterial dissection during angioplasty), no major bleeding events, four minor bleeding events, one preterm delivery with good outcome, and three fetal deaths/spontaneous abortions (one due to maternal death, one at six weeks gestation in a 40-year-old mother with polycythemia vera and essential thrombocytosis, one in first trimester in a mother with bacterial endocarditis leading to acute stroke).

Mechanical cardiac valve thrombosis and acute MI

Sixteen articles were found describing the outcomes in 24 cases of patients receiving thrombolytics for cardiac indications (23 mechanical valve thromboses, one acute MI).^65–80 In addition to considering the relatively high mortality associated with these conditions, the indications for choosing thrombolysis versus operative intervention were not clear in the majority of the reports, and thus may have played a role in some of the observed outcomes. There were two maternal deaths, five fetal deaths (two due to maternal death), two serious bleeding complications, and two minor bleeding complications.

Discussion

The relative hypercoagulable state of pregnancy that is necessary to allow for normal growth of a fetus and arrest hemorrhage during delivery unfortunately carries with it increased risk of thrombotic complications. Some of these complications occur more frequently than others, with some approaching relative risks 6–10 times above the non-pregnant state. While some of thrombotic events are medically manageable, occasionally these conditions present in a life-threatening state or one that may carry lifelong debility. As such, the use of active thrombolysis may be warranted regardless of the pregnancy to ensure the safety of the mother and the health of the child. That stated, given their potential to induce significant hemorrhage, thrombolytic agents only should be considered in appropriate clinical scenarios and in the safest manner possible.

This review described the known reports and series regarding the use of thrombolytic drugs in pregnant patients for a variety of serious clinical scenarios. When examining the group as a whole, it is difficult to ascertain the maternal and fetal complication rates that would be strictly attributable to the use of thrombolysis alone, given that many of these patients clinically presented in shock, after surgical attempts at treatment, and with large burden of medical comorbidity (both before and after their thrombotic complication). It would seem reasonable to expect that both maternal and fetal complications will occur in a cohort of critically and/or chronically ill women, irrespective of the additional burden of thrombotic complications and thrombolytic treatment. That stated, the observed complication rates in these articles is low and in proportion or less than reports in non-pregnant patients who were generally much older, especially given the conditions that were being treated carry a high mortality even without the addition of the pregnancy.^14,81 There were only a few maternal deaths reported – with none being reported due to the thrombolytic treatment in the VTE or neurologic indications. There were 10 fetal losses reported, with the majority associated with maternal death, maternal presentation in shock, and significant maternal medical comorbidity. A mixture of thrombolytic agents was used in the articles described, with rtPA and streptokinase being the most common. There were low rates of bleeding complications overall, with 10 events occurring with streptokinase and urokinase (six minor, four major) and six events occurring with rtPA (four minor, two major). No deaths were reported due to bleeding.

These data are suggestive that the use of thrombolytic agents in pregnant women should be considered as similar as when they would be indicated in a non-pregnant patient. Given the comparison of complication rates presented, rtPA likely would be selected as the agent of choice for several reasons; and regardless, this generally is the only agent now used and widely available. There may be several reasons that rtPA demonstrates a lower complication profile. First, it is administered over a short time period, versus some of the reports with streptokinase being given over a period of days. Next, it has been reported that rtPA’s mechanism of action is ideal for acute thrombotic events in which one would desire the pathologic clot to be dissolved, and not induce a systemic ‘lytic’ effect. It has been shown that rtPA has high affinity for plasminogen only in the presence of fibrin, thus affording this theoretical benefit of less systemic hemorrhage risk.⁸² Lastly, compared to other thrombolytics, rtPA is purported not to induce antigenicity, thus allowing for repeated administration – which is particularly important in the cases of mechanical valve thrombosis.⁸²

The available published data suggest the safety of these agents in pregnancy and that the observed complication rates are seemingly on-par with their use in non-pregnant patients. However, it must be stated that there is a lack of scientific rigor leading to these conclusions given the poor quality of evidence (i.e. case reports and case series). There is an inherent publication bias at this level of evidence, especially when the subject is surrounding a potentially life-threatening scenario. It is reasonable to assume that a clinician would be much less likely to offer for publication a case of thrombolysis used in pregnancy that led to catastrophic bleeding and maternal–fetal death. With that understood, however, all available evidence suggest that one should approach the treatment of these thrombotic complications in pregnant women in an identical fashion to the general population.

Funding

This review received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.