Background and Rationale
In patients with acute pulmonary embolism (PE), the risk of early death or haemodynamic
collapse ranges from less than 1 to over 50%, depending on the severity of the acute
episode as well as the cardiopulmonary reserves and comorbidity of the patient.[1]
[2]
[3] Current European clinical practice guidelines place particular emphasis on early
risk stratification to determine the optimal risk-adjusted treatment strategy. Patients
with haemodynamic compromise or frank cardiogenic shock at presentation have at least
20% risk of early adverse events and should receive immediate reperfusion therapy,
while those with low-risk PE, as defined in the 2014 Guidelines of the European Society
of Cardiology (ESC[2]), and no other indication for hospitalization should be considered for early discharge
and home treatment.[4]
[5]
[6]
[7]
[8] The remaining patients with acute PE are classified into the intermediate-risk category
and depending on the method and the cut-off values used for definition, they may account
for up to 60% of all PE patients and have up to 7.7% risk of early mortality.[2]
[9]
The prognostic impact and therapeutic implications of intermediate-risk PE, and particularly
the ‘upper zone’ of severity within this category, were highlighted by the results
of the Pulmonary Embolism International Thrombolysis (PEITHO) trial.[10] In the placebo (heparin-only) arm of PEITHO, the upper limit of the 95% confidence
interval (CI) for the risk of haemodynamic decompensation was as high as 8% (mean
5.6%) in normotensive patients who presented both with right ventricular dysfunction
on echocardiography or computed tomographic pulmonary angiography (CTPA), and with
evidence of myocardial injury indicated by elevated cardiac troponin levels.[10]
Historically, the standard therapeutic regimen for patients with acute PE has consisted
of parenteral anticoagulation (intravenous unfractionated heparin, or subcutaneous
low-molecular-weight heparin [LMWH] or fondaparinux) with overlapping administration
of a vitamin K antagonist (VKA) until reaching a therapeutic range of the international
normalized ratio (INR).[2] More recently, four non-vitamin K–dependent direct oral anticoagulants (NOACs),
that is, one direct thrombin inhibitor (dabigatran) and three factor Xa inhibitors
(rivaroxaban, apixaban and edoxaban) were demonstrated to be non-inferior and at least
as safe as traditional treatment with LMWH and VKA, and subsequently approved for
the treatment and secondary prevention of venous thromboembolism (VTE).[11]
[12]
[13]
[14] These drugs, which do not require laboratory monitoring, were developed to overcome
the limitations and risks that characterize the initial overlap of parenteral and
oral anticoagulants and the subsequent chronic treatment with VKAs, notably the bleeding
or thrombotic complications caused by unstable INR values. In the large phase III
trials in patients with acute PE and/or deep vein thrombosis (DVT), NOAC treatment
was associated with a significantly lower risk of major (relative risk [RR]: 0.60;
95% CI: 0.41–0.88) and fatal (RR: 0.36; 95% CI: 0.15–0.87) haemorrhage compared with
VKAs, while having comparable efficacy (RR: 0.88; 95% CI: 0.74–1.05). Importantly,
however, the safety and efficacy of NOAC in patients with intermediate-risk PE have
not been systematically addressed in a dedicated trial thus far, and it is largely
unknown what the overall number of patients with acute intermediate-risk PE was in
all but one of the phase III studies mentioned earlier.[11]
[12]
[13]
[14]
The Hokusai-VTE phase III trial reported data for the subgroup of patients with acute
PE and right ventricular (RV) dysfunction, defined as right-to-left ventricular (RV/LV)
diameter ratio ≥0.9 on CT angiography, and for patients with N-terminal pro-brain
natriuretic peptid (NT-proBNP) levels greater than 500 pg/mL.[12]
[15] Recurrent VTE occurred in 14 of 465 patients with elevated NT-proBNP treated with
LMWH followed by edoxaban and in 30 of 507 of their counterpart in the LMWH-warfarin
group for a hazard ratio of 0.50 (95% CI: 0.26–0.94).[15] Similar differences were observed in the subgroup of patients with both an elevated
NT-proBNP level and a right-to-left ventricular ratio ≥0.9 on CTPA (hazard ratio:
0.44 [95% CI: 0.14–1.36]). When viewed together with the substantial risk of early
decompensation and the possible need for rescue thrombolysis (or, alteratively, interventional
or surgical therapy) in the intermediate-risk category, these results generate the
hypothesis that an initial parenteral anticoagulation with subcutaneous LMWH, followed
by an NOAC may be the safest approach for this patient group. In this regard, however,
the minimum period after which treatment may be safely switched to oral anticoagulation
remains to be defined.[16] It is thus necessary to show whether patients may safely switch from parenteral
heparin treatment to an NOAC already after the first 72 hours, that is, earlier than
in the regimens tested in previous trials, and whether there is a subgroup of intermediate-risk
patients that might need a more conservative approach with a longer heparin period
before switching to an NOAC.[11]
[12]
[17] Finally, a trial focusing on intermediate-risk PE should provide insights, using
prospectively defined follow-up intervals and standardized outcome parameters, into
the rate of recovery of RV function under anticoagulation for acute PE.
Study Overview
Study Design and Objectives
PEITHO-2 is a prospective, multicentre, multinational, single-arm management trial
(NCT02596555, EudraCT 2015–001830–12). The main characteristics of the trial design
are summarized in [Table 1] against the background of recent trials using a heparin lead-in phase. PEITHO-2
is an investigator-initiated trial with an academic sponsor, the University Medical
Center Mainz. The primary objective is to determine whether treatment of acute intermediate-risk
PE with parenteral anticoagulation for 72 hours (rather than 5 days or longer currently
required before the switch to dabigatran or edoxaban), followed by dabigatran over
6 months, is effective and safe. The secondary objectives are to assess the safety
of the studied treatment regimen as well as the recovery of RV function assessed by
serial echocardiograms, and the importance of this recovery for the 6-month prognosis
of patients with acute intermediate-risk PE.
Table 1
Overview of trials on non-vitamin K–dependent direct oral anticoagulants with a heparin
lead-in phase
|
RE-COVER[11]
|
HOKUSAI-VTE[12]
|
PEITHO-2 (present study)
|
|
Design
|
Double-blind, double-dummy, non-inferiority phase III study
|
Double-blind, double-dummy, non-inferiority phase III study
|
Single-arm phase IV management study
|
|
Treatment and dosage
|
LMWH for minimum of 10 d followed by either dabigatran etexilate 150 mg bid or INR-adjusted
warfarin
|
LMWH for a minimum of 7 d followed by either edoxaban 60 mg od or INR-adjusted warfarin
|
LMWH for 72 h, followed by dabigatran etexilate 150 mg bid[a]
|
|
Duration (primary endpoint)
|
6 mo
|
3–12 mo
|
6 mo
|
|
Patients enrolled in the original trial
|
5,128 patients with acute VTE
|
8,292 patients with acute VTE
|
700 patients with acute intermediate-risk PE
|
|
Number of patients with PE in the trial population
|
1,602 patients with acute PE
|
972 patients with acute PE and NT-proBNP ≥500 pg/mL
|
All study patients have acute intermediate-risk PE
|
|
Primary efficacy outcome
|
Recurrent VTE or death related to PE
|
Recurrent VTE or death related to PE
|
Recurrent VTE or death related to PE
|
|
Efficacy outcome: results (PE patients only)
|
2.9% (dabigatran) versus 3.1% (warfarin; HR: 0.93; 95% CI: 0.53–1.64)
|
3.0% (edoxaban) versus 5.9% (warfarin; HR: 0.50; 95% CI: 0.26–0.94)
|
Trial ongoing
|
|
Primary safety outcome
|
Major bleeding
|
Major bleeding
|
Major bleeding
|
|
Safety outcome: results (all patients)
|
1.2% (dabigatran) versus 1.7% (warfarin; HR: 0.69; 95% CI: 0.36–1.32)
|
1.4% (edoxaban) versus 1.6% (warfarin; HR 0.84; 95% CI: 0.59–1.21)
|
Trial ongoing
|
Abbreviations: bid, twice daily; CI, confidence interval; HR, hazard ratio; LMWH,
low-molecular-weight heparin; NT-proBNP, N-terminal pro-brain natriuretic peptide;
od, once daily; PE, pulmonary embolism; PEITHO, Pulmonary Embolism International Thrombolysis;
VTE, venous thromboembolism.
a A reduced dose of 110 mg twice daily is recommended for patients aged 80 years or
older, and for those under concomitant treatment with verapamil, according to the
European Summary of Product Characteristics (SmPC) for dabigatran.
Rationale for the Study Design
PEITHO-2 has prospectively defined and is evaluating a ‘dynamic’ strategy of risk-adjusted
anticoagulation. This strategy encompasses (1) inclusion of patients with intermediate-risk
PE, and further differentiation into intermediate-high and intermediate-low risk;
(2) application of clinical and haemodynamic criteria for determining the earliest
safe time point for the switch from parenteral to oral treatment and (3) analysis
of the temporal pattern of RV recovery after acute intermediate-risk PE. To reach
its objectives, PEITHO-2 was designed as a prospective management (cohort) trial without
a control treatment arm. This was done for the following reasons.
Dabigatran is an already approved agent for the treatment of patients with acute PE,
without excluding those with intermediate-risk PE. Although it is unknown how many
of the patients included in the dabigatran phase III RECOVER trials belonged to the
latter risk category, the members of the PEITHO-2 steering committee agreed that it
would be inappropriate to repeat the randomized design of the RECOVER trials (dabigatran
vs. VKA) within this focused study population. Instead, it is planned to analyse the
data from PEITHO-2, upon completion, against the patient-level data from either treatment
arm (dabigatran vs. warfarin) of the pooled RECOVER population.[11] This will allow us to compare the efficacy and safety of dabigatran in the present
trial with (1) the efficacy and safety of the same drug in the ‘unselected’ (with
regard to severity) PE subpopulation of RECOVER (dabigatran arm) and (2) the efficacy
and safety of VKA (warfarin arm) in the PE patients of RECOVER.
In further analyses, the 6-month clinical course of the intermediate-high-risk subgroup
(defined as simplified Pulmonary Embolism Severity Index [sPESI] of 1 or higher, along
with an imaging test indicating RV dysfunction and elevated troponin or natriuretic
peptide levels[2]) within the present trial will be determined. The data will then be compared with
the 6-month follow-up data from a corresponding subgroup, which will be selected on
a patient-level basis from the placebo (heparin followed by VKA) arm of the PEITHO
randomized controlled trial (unpublished data).
Considering all the above, we believe that the single-arm design of PEITHO-2 will
allow us to successfully put the results of PEITHO-2 into perspective, while also
fulfilling an important ethical requirement, namely keeping the number of patients
and the complexity as well as the duration of the trial to the minimum required for
adequately answering the study questions. This rationale was endorsed by the trial's
central ethics committee in Germany and by the ethics committees of the countries
in which the trial is already recruiting.
Patient Population and Eligibility
The key inclusion and exclusion criteria are summarized in [Table 2]. Patients with symptomatic acute, intermediate-risk PE, diagnosed by multidetector
CT angiography, ventilation/perfusion lung scan or selective invasive pulmonary angiography,
with or without symptomatic DVT, who are haemodynamically stable at presentation,
are eligible for enrolment in the study. The diagnosis of intermediate-risk PE is
confirmed at each participating site, based on standardized diagnostic and risk stratification
criteria and as recommended in current guidelines.[2]
Table 2
Key inclusion and exclusion criteria
|
Inclusion criteria
|
Exclusion criteria
|
|
1. Age ≥ 18 y
2. Objectively confirmed diagnosis of acute PE, with or without DVT[1]
3. No haemodynamic decompensation or collapse at presentation (none of the following):
a. Need for cardiopulmonary resuscitation
b. Systolic Bp < 90 mm Hg, or drop by ≥40 mm Hg, for at least 15 min, with clinical
signs of end-organ hypoperfusion (cold extremities, urinary output <30 mL/h, mental
confusion)
c. Need for catecholamines to maintain adequate organ perfusion and a systolic BP
of >90 mm Hg
4. Intermediate-risk PE (≥1 of the a, b or c criteria):
a . Elevated troponin levels[a]
b. NT-proBNP levels >600 pg/mL
c. RV pressure overload/dysfunction on CT angiography or echocardiography:
i. CTPA: RV/LV end-diastolic diameter ratio >1.0
ii. Echocardiography (any of the following):
– RV/LV end-diastolic diameter ratio >1.0 (apical or subcostal 4-chamber view)
– RV end-diastolic diameter >30 mm (parasternal long-axis or short-axis view)
– RV free wall hypokinesis (any view)
– Tricuspid regurgitant jet velocity >2.6 m/s
– Absence of inspiratory collapse of the inferior vena cava
|
1. Any medical or psychological condition that would not permit signing of informed
consent or completion of the trial; unwillingness or inability to adhere to treatment
or to the follow-up visits
2. Pregnancy or lactation (or women of childbearing potential not practising a medically
accepted contraception during the trial)
3. History of hypersensitivity to dabigatran
4. Use of a fibrinolytic agent, surgical embolectomy, interventional (catheter-directed)
thrombus aspiration or lysis, or use of a vena cava filter
5. Active bleeding or known significant bleeding risk
6. Need for long-term treatment with any anticoagulant, or need for antiplatelet
agents except acetylsalicylic acid ≤100 mg/d
7. Artificial heart valves requiring treatment with an anticoagulant
8. Renal insufficiency with estimated creatinine clearance <30 mL/min/1.73 m2
9. Chronic liver disease with aminotransferase levels two times or more above the
local upper limit of normal range
10. Concomitant administration of strong inhibitors of P-glycoprotein like ketoconazole,
cyclosporin, itraconazole or dronedarone
11. Life expectancy less than 6 mo
|
Abbreviations: BP, blood pressure; CTPA, computed tomographic pulmonary angiography;
DVT, deep vein thrombosis; LV, left ventricular; NT-proBNP, N-terminal pro brain natriuretic
peptide; PE, pulmonary embolism; RV, right ventricular; VTE, venous thromboembolism.
a Troponin elevation is defined as an abnormal result of any validated troponin test
based on the reference values determined by the local Department of Clinical Chemistry
at each participating site.
In contrast to Hokusai-VTE,[12]
[15] PEITHO-2 uses the more practical and ‘intuitive’ 1.0 threshold for defining an abnormally
high RV/LV ratio, in agreement with other interventional trials[10]
[18] and cohort studies[19]
[20]
[21] in haemodynamically stable patients with acute PE. Furthermore, the threshold of
600 pg/mL is used in the inclusion criterion involving the biomarker NT-proBNP, as
we have previously reported that this may be the most appropriate value for defining
intermediate risk.[22]
[23]
[24]
Current guidelines include the Pulmonary Embolism Severity Index (PESI) in the definition
of intermediate-risk PE.[2] However, despite the prognostic value of the PESI,[25] it should be kept in mind that it was primarily designed as an epidemiological tool
and not as a direct guide to PE management. Moreover, it has been reported that (severe)
RV dysfunction may be present despite a simplified PESI of 0.[26] If reproduced, this would be a relevant finding that could change future risk stratification
algorithms. For these reasons, the present study does not use the simplified PESI
as an inclusion criterion. Importantly, however, the index is prospectively calculated
and documented in all patients, and it will thus be possible to determine if and how
many study patients may exhibit a discordance between the clinical score and their
imaging or biochemical parameters at presentation.
Treatment Regimen
In the RECOVER trials, the median duration of initial parenteral treatment with LMWH
was 9 days.[11]
[17] This was longer than the current average duration of hospitalization of patients
with PE in Europe and North America, which generally does not exceed 4 to 6 days.[27]
[28] Longer treatment periods with a parenteral anticoagulant before switching to an
NOAC might delay the discharge of stable patients and thus increase treatment-related
costs. Therefore, in this study, we are investigating the approach of an earlier switch
from parenteral anticoagulants to dabigatran in clinically stabilized intermediate-risk
PE patients.
The study flow is summarized in [Fig. 1]. After enrolment, patients start or continue to receive LMWH at a therapeutic dosage
for a period of 72 hours from the moment of PE diagnosis. If LMWH is not available,
treatment with unfractionated heparin given by intravenous infusion is acceptable.
The 72-hour duration of parenteral anticoagulation was chosen based on our previous
PEITHO trial,[10] in which the vast majority of early adverse outcomes (and thus cases of possible
need for switch to rescue reperfusion therapy) in patients with intermediate-high-risk
PE occurred within the first 3 days of randomization (unpublished data). Thereafter,
and following clinical assessment of the patient's condition, treatment is switched
to oral dabigatran at the dosage of 150 mg twice daily. A reduced dose of 110 mg twice
daily is recommended for patients aged 80 years or older, and for those under concomitant
treatment with verapamil, according to the European Summary of Product Characteristics
(SmPC) for dabigatran. The switch from LMWH to dabigatran is postponed by the investigator
if (1) the patient is in shock, or haemodynamic collapse is considered imminent in
view of deteriorating clinical parameters such as tachycardia, hypotension or dyspnoea
and hypoxaemia; (2) the patient has received thrombolytic, surgical or catheter-directed
treatment, or underwent cava filter implantation within the past 48 hours; (3) there
is active bleeding or (4) the patient is, for any other reason, unable to receive
oral anticoagulation. If at least one of these criteria is present, LMWH is continued
and the eligibility for switching to dabigatran treatment is reassessed 48 hours later.
Of note, the diagnostic workup and management of patients with early haemodynamic
deterioration is based on local hospital protocols and the recommendations of current
guidelines, and not dictated by the trial protocol.
Fig. 1 Overview of design of the PEITHO-2 (Pulmonary Embolism International Trial 2) study.
PE, pulmonary embolism; I, inclusion; CE, clinical evaluation; T, time; LMWH, low-molecular-weight
heparin.
Oral anticoagulation with dabigatran is continued for 6 months. An overview of the
tests performed and the parameters collected upon enrolment and at the follow-up visits
is provided in [Table 3]. After the end of the study period, the decision on whether to continue anticoagulation
treatment, and which anticoagulant drug to administer, is left to the discretion of
the physician(s) caring for the patient.
Table 3
Trial visit plan and data collection schedule
|
Visit 1: enrolment (screening and baseline)
|
Visit 2: 72 (+12) h after PE diagnosis or 48 (±4) h later
|
Visit 3: discharge
|
Visit 4: FU 30 (±7) d after discharge
|
Visit 5: FU 180 (±14) d after discharge
|
Visit 6: FU 30 (±7) d after the end of treatment
|
|
In hospital
|
Outpatient follow-up
|
|
Medical history
|
X
|
|
|
|
|
|
|
Demographic data
|
X
|
|
|
|
|
|
|
Clinical examination
|
X
|
X
|
|
X
|
|
|
|
ECG
|
X
|
X
|
|
|
|
|
|
sPESI
|
X
|
|
|
|
|
|
|
VTE-BLEED
|
X
|
|
|
X
|
|
|
|
CTPA: RV/LV ratio
|
X
|
|
|
|
|
|
|
Echocardiography
|
X
|
|
X[a]
|
|
X
|
|
|
Blood gas analysis
|
X
|
|
|
|
|
|
|
Troponin I and/or t-test
|
X
|
|
|
|
|
|
|
Further laboratory tests (NT-proBNP, renal and liver function)
|
X
|
|
X
|
|
X
|
|
|
Check inclusion and exclusion criteria
|
X
|
|
|
|
|
|
|
Obtain informed consent
|
X
|
|
|
|
|
|
|
Pregnancy test (for women of childbearing age)
|
X
|
X
|
X
|
X
|
X
|
X
|
|
Check criteria for switch to oral therapy
|
|
X
|
|
|
|
|
|
Documentation of
|
|
• Adverse events[b]
|
X
|
X
|
X
|
X
|
X
|
X
|
|
• Co-medications and therapies
|
X
|
X
|
X
|
X
|
X
|
|
|
• Rehospitalization[c]
|
|
|
|
X
|
X
|
X
|
|
End of trial
|
|
|
|
|
|
X
|
Abbreviations: CTPA, computed tomographic pulmonary angiography; DVT, deep vein thrombosis;
ECG, electrocardiogram; FU, follow-up; LV, left ventricular; NT-proBNP, N-terminal
pro brain natriuretic peptide; PE, pulmonary embolism; RV, right ventricular; sPESI,
simplified Pulmonary Embolism Severity Index.
a Must be done at 6 ± 1 days after PE diagnosis or upon discharge, whichever comes
first.
b Particularly death (and cause of death), haemodynamic deterioration, symptomatic
recurrent VTE, major bleeding and clinically relevant non-major bleeding.
c For venous thromboembolism or bleeding.
Outcomes
[Table 4] summarizes the outcomes of the PEITHO-2 trial. The primary efficacy outcome is the
occurrence of recurrent symptomatic VTE or death related to PE within 6 months of
therapy. Recurrent symptomatic DVT is confirmed by a new non-compressible venous segment
on compression ultrasonography or by a new filling defect on CT venography. Recurrent
symptomatic PE is diagnosed by a new filling defect on CT pulmonary angiography, or
by a more proximal obstruction of a previously occluded pulmonary artery, or by a
new segmental mismatch on ventilation–perfusion lung scan. Death related to PE is
defined as death following acute, objectively confirmed symptomatic recurrent PE or
as sudden death not explained by another cause. The safety outcomes include major
bleeding,[29] non-major clinically relevant bleeding and occurrence of serious adverse events
([Table 4]). In addition, the recently developed and validated VTE-BLEED score will be prospectively
evaluated with respect to its ability to predict bleeding events.[30]
[31]
[32]
Table 4
Primary and secondary outcomes
|
Primary outcome
|
Recurrent symptomatic VTE or death related to PE within the first 6 mo
|
|
Safety outcomes
|
Major bleeding according to the criteria of the International Society on Thrombosis
and Haemostasis[30]
|
|
Non-major clinically relevant bleeding defined as:
• Spontaneous skin haematoma of at least 25 cm2
• Spontaneous nose bleeding of more than 5 min
• Macroscopic haematuria, either spontaneous or, if associated with intervention,
lasting more than 24 h
• Spontaneous rectal bleeding
• Gingival bleeding for more than 5 min
• Bleeding leading to hospitalization and/or requiring surgical treatment
• Bleeding leading to transfusion of less than 2 units of whole blood or red cells
• Any other bleeding event considered clinically relevant by the investigator
|
|
Occurrence of serious adverse events
|
|
Secondary outcomes
|
Death from any cause, or haemodynamic collapse or decompensation, within the first
30 d
|
|
PE-related death, or PE-related haemodynamic collapse or decompensation, within the
first 30 d
|
|
Recovery of RV function at 6 ± 1 d after PE diagnosis or upon discharge (whichever
comes first) and at 6-mo follow-up
|
|
Normalization of NT-proBNP levels during follow-up
|
|
Overall duration of hospital stay (index event and repeated hospitalizations due to
PE [or to a bleeding event]) within 6 mo
|
|
Death from any cause within 6 mo
|
Abbreviations: NT-proBNP, N-terminal pro brain natriuretic peptide; PE, pulmonary
embolism; RV, right ventricular; VTE, venous thromboembolism.
The management of patients with severe haemorrhagic complications under anticoagulant
treatment is based on local hospital protocols and the recommendations of current
guidelines and consensus statements.[33]
[34] It is not dictated by the protocol of the present trial.
All primary and secondary outcomes will be adjudicated by an independent committee,
the members of which are mentioned in the Appendix A.
Sample Size Calculation and Statistical Analysis Plan
The expected incidence of the primary outcome of the present study is derived from
data provided by Hokusai-VTE and PEITHO.[10]
[12] Considering the definition of RV dysfunction in Hokusai-VTE, and the inclusion criteria
of PEITHO (focusing on intermediate-high risk), we assumed that, in PEITHO-2, the
incidence rate of the primary outcome in the total study population of ‘unselected’
intermediate-risk PE will be at least as high and possibly higher than that of the
Hokusai study (3.3%), but it will not exceed the point estimate of that in the 6-month
follow-up of the ‘intermediate-high-risk’ PEITHO trial (6.1%). Thus, assuming a 3.3%
incidence of the primary outcome, we will test the null hypothesis (H0; probability [p] ≥ 0.061; 6.1%) against the alternative hypothesis (H1; p < 0.061) using a one-sided exact binomial test in a two-stage Pocock group sequential
design. The significance level was chosen as α = 5%, leading to a nominal α = 0.003037 at the first stage. It is expected that 650 patients will provide a power
of 90% for rejecting H0 (p ≥ 0.061) at the final stage if p is indeed 0.033. Taking into account a loss to follow-up of 7% maximum, a total number
of 700 patients will have to be enrolled in PEITHO-2.
All analyses will be performed in the intention-to-treat population. We will consider
our hypothesis that treatment with dabigatran preceded by at least 72 hours of parental
anticoagulation is an effective treatment for patients with acute intermediate-risk
PE as proven if the null hypothesis can be rejected, that is if 27 or less events
are observed among 650 analysed patients.
A formal interim analysis is planned after 20% of the targeted patient population
has been included and followed for 30 days. We will consider more than 11 events as
an indication that it is no longer realistic to reject H0 in the further course of the trial, as the conditional power to reject the null hypothesis
would then be below 50%. In this case, the Data Safety Monitoring Board may propose
a protocol amendment that dictates that the switch from parenteral anticoagulants
to dabigatran should be delayed to 5 ± 1 days in all patients.
Subgroups will be analysed with regard to the following outcomes: (1) primary outcome;
(2) death or haemodynamic collapse or decompensation from any cause, PE-related death
or haemodynamic collapse or decompensation, within the first 30 days; (3) overall
duration of hospital stay within 6 months; (4) death from any cause within 6 months
and (5) all safety outcomes ([Table 4]). The prespecified variables and cut-offs for defining patient subgroups are the
following: age (<75 vs. ≥75 years), sex, provoked versus unprovoked PE, patients with versus patients without cancer, presence versus absence of symptomatic confirmed DVT, patients with sPESI of 0 versus 1 or higher at baseline, intermediate-low-risk versus intermediate-high-risk PE, baseline NT-proBNP (≤600 vs. >600 pg/mL), switch to dabigatran
after 72 hours versus switch after 5 days or later, echocardiographically assessed cardiac recovery versus no recovery and NT-proBNP levels (normalization vs. improvement vs. no improvement
vs. deterioration). Finally, the rates of the primary efficacy outcome and major bleeding
in the study population will be compared with those in either treatment arm (dabigatran
or warfarin) of the pooled RECOVER population.[11] In addition, the rates of the primary outcome and major bleeding in the predefined
intermediate-high-risk subgroup of patients enrolled in the present trial will be
compared with those in the placebo (heparin followed by VKA) arm of the PEITHO randomized
controlled trial at 6-month follow-up.
Implications of PEITHO-2
PEITHO-2 is the first clinical trial to apply a contemporary risk-tailored treatment
strategy for acute PE, implementing and, possibly, further advancing the recommendations
of current European guidelines.[2] The first PEITHO study published in 2014 showed that patients at the ‘upper zone’
of the severity spectrum of intermediate-risk PE appeared to have an overall larger
clinical benefit from a watchful waiting strategy. This consisted of initial anticoagulation,
monitoring and rescue reperfusion treatment (only) in case of haemodynamic decompensation,
rather than routine primary thrombolysis.[11] PEITHO-2 continues to develop the risk-adjusted management strategy where PEITHO
stopped. It investigates the strategy of parenteral anticoagulation followed by an
early switch to oral treatment within the large intermediate-risk group of PE patients,
and follows that group prospectively over a 6-month period (in PEITHO, follow-up beyond
the first 30 days focused primarily on overall mortality[35]). Comparison of the results of PEITHO-2 with the already available data from PEITHO
will help put the efficacy and safety of the PEITHO-2 strategy into perspective. With
the same objective, the results of PEITHO-2 will be compared with those from the phase
III RECOVER trials with dabigatran, although possible differences in the clinical
severity of PE, and the (expected) shorter duration of initial heparin anticoagulation
in the present study, might pose some limitations in this regard.
If the hypothesis of PEITHO-2 is confirmed, patients with intermediate-risk PE will
need only 72 hours of parenteral heparin treatment before switching to oral anticoagulation.
This may offer the potential to safely reduce the duration of hospitalization for
PE, with an impact on hospital-related complication rates, morbidity and costs. On
the other hand, the PEITHO-2 study will also be able to determine which patients within
the broad spectrum covered by intermediate-risk PE might need a more conservative
approach with longer periods of parenteral heparin treatment. Finally, with its comprehensive
follow-up strategy, the PEITHO-2 study will provide a deeper insight into RV recovery
after acute PE, and its implications for the long-term prognosis of the patients.
