Alex Nowbar’s research reviews—26 March 2019

Alex Nowbar reviews the latest research from the top medical journals


Atrial Fibrillation ablation and outcomes

The CABANA trial looked at whether ablation for atrial fibrillation reduces mortality compared with drug therapy. It didn’t. In two simultaneous reports on the trial, which randomised over 2000 patients, ablation improved quality of life (using quality of life scores specific for atrial fibrillation) and reduced cardiovascular hospitalisations. But with these two outcomes comes the warning that the trial was unblinded. Staff and patients knew what intervention the patients had, and this has an undeniable impact. The absence of blinding skews those outcomes, and thus the true effect of ablation on those outcomes has not been measured. The lesson is that interventional trials, like drug trials, require placebo control with double blinding.


3D fetal hearts

These researchers developed an open source, motion correction technique for imaging the fetal cardiovascular system with magnetic resonance imaging (MRI) and assessed it in 85 cases. It is intended for use in pregnancies with a fetus with known or suspected congenital heart disease. The researchers found it was reliable (using echocardiography as the standard), and the visual and diagnostic quality were better than the two dimensional, uncorrected MRI. In particular, the vascular structures could be seen better, and in 10 cases the new technique identified anatomical features that weren’t previously known about. This could provide useful information in preparation for the birth and postnatal care.


Antithrombotic regimens for patients with atrial fibrillation

For treatment of cardiovascular disease, the base drug is usually aspirin. For acute coronary syndrome or stenting, this usually gets topped up with clopidogrel, ticagrelor, or, less commonly, prasugrel. But if the patient has atrial fibrillation, they will also need anticoagulation. Should they be given warfarin or the xabans? Dual or triple therapy? The AUGUSTUS trial elegantly used 2×2 factorial design to address both questions. Over 4000 patients were randomised to apixaban or a vitamin K antagonist (which was open label) and to either aspirin or matched placebo (which was double blinded). And they all received a P2Y12 inhibitor such as clopidogrel. The strategy with the best outcomes was a P2Y12 inhibitor and apixaban without aspirin.

Angiography for arrests

Percutaneous coronary intervention for STEMI is one of the few medical interventions nearing parachutism (a parachute being a metaphor for a treatment without which the chance of death is near 100%). No one honestly thinks PCI will be that critical for any other condition but, of course, it can still be a good treatment. Since ischaemic heart disease is the commonest cause of cardiac arrest, Lemkes et al investigated whether immediate versus delayed angiography for patients with an out-of-hospital cardiac arrest with an initial shockable rhythm. They randomised 552 patients to either immediate angiography (with PCI if needed) or angiography delayed until neurologic recovery. There was no statistically significant difference in mortality at 90 days between groups, nor in any of the other endpoints other than it taking longer for the immediate angiography patients to reach target temperature. Over a third of patients had no clinically significant coronary disease.  

Cardioversion for recent-onset symptomatic atrial fibrillation

Pluymaekers et al randomised 437 patients to early or delayed cardioversion in a non-inferiority trial. The delayed approach was non-inferior for restoration of sinus rhythm. This is a really useful practical result for the medical community. Spontaneous conversion was common and if it happens then electrical cardioversion will have been avoided which is preferably. This study allays fears that not cardioverting early on and thus remaining longer in atrial fibrillation might be bad because electrical cardioversion might be less effective at restoring sinus rhythm at this later stage. It was just as effective at the later stage in terms of the proportion of patients achieving sinus rhythm. The authors also explain that the theoretical risk of stroke in the period of delay to cardioversion is probably mitigated by the use of anticoagulant therapy.

A new era for aortic stenosis

Surgical aortic valve replacement has been trumped by transcatheter valve replacement for low-risk patients in the PARTNER-3 trial and the Evolut Low Risk Trial. 

PARTNER-3 looked at the balloon-expandable valve. In this non-inferiority randomised trial of 1000 patients, the headline finding was the much lower rate of death, stroke, or rehospitalisation (which together formed the primary endpoint) in the transcatheter group. The transcatheter group also had less atrial fibrillation and shorter stays. Follow-up was for a year which is reasonable for a mortality assessment but patients will also be interested in the longevity of their new valve.

The Evolut Low Risk trial looked at the self-expanding valve. The primary endpoint was death or disabling stroke. Again, it was non-inferior to surgical valve replacement. In the context of being able to avoid these major outcomes through the transcatheter approach, the outcome of being more likely to have a pacemaker implanted (as was shown for the self-expanding valve) does not seem hugely problematic but will be relevant to patients.


Game of Troponins

For this study, a Southampton-based team of researchers got ethical approval to test the blood of 20,000 patients without their consent. The blood had been taken from patients for biochemistry testing for a clinical reason (for a minority this was chest pain). The researchers put these samples through their troponin assay to work out the “true” normal range of troponins in patients (as opposed to the healthy people used by the assay manufacturer).

They found that the 99th centile of troponin level of the cohort of 20,000 was several times higher than that reported by the manufacturer for 300 healthy men and 300 healthy women. Well, it was a different population, a population selected for the unique bundle of things that made people in the summer of 2017 get into different hospital areas and end up with a blood test. They repeated their analysis after excluding actual myocardial infarctions. But the amount of and reliability of the data on what else was going on with these patients was limited by the observational methodology which is a shame.

If centiles aren’t your preferred concept, their results also showed that 5% of people with the unique bundle of things had a higher troponin level than 99% of the healthy people the manufacturer will have tested. It would have been weird if only 1% of their cohort had a higher troponin level than 99% of healthy people because then they would have matched the group of healthy people. But sometimes it’s good for research to confirm obvious things, i.e. don’t use a troponin in isolation to diagnose myocardial infarction.

Risk of miscarriage

Over 400,000 women in Norway were studied to better understand the association between age and miscarriage. The lowest risk was in the 25-29 age group at 10% but aged 45 and over the risk was 53%. Risk of recurrence after one miscarriage was high, higher after 2 miscarriages and even higher after 3. The aetiological factors influencing these associations remain fairly obscure but they did find that “other pregnancy outcomes cluster with the risk of miscarriage, suggesting that these outcomes might share underlying causes”. The relatively high risk of miscarriage in the youngest women might suggest “reproductive immaturity”. The mechanism of increased miscarriage risk could also be socially-mediated.

Alex Nowbar is a clinical research fellow at Imperial College London.

Competing interests: None declared