Impact of atrial remodeling on AF outcomes

Prashanthan Sanders

Prashanthan Sanders

If an individual has a few risk factors, the incidence of persistent atrial fibrillation is much lower than if you have a greater number of risk factors, in which case it is more likely that these individuals will progress to persistent atrial fibrillation.

Marking the very first presentation of the 2015 AF Symposium, Prashanthan Sanders from the Royal Adelaide Hospital, Australia, kick-started proceedings on Thursday morning with an exploration of the mechanisms of atrial remodeling and their impact of AF outcomes.

“We know that AF is a progressive disease that most patients go from paroxysmal to persistent, and then to permanent atrial fibrillation,” he began. “Data [has] indicated that this may be related to the associated cardiovascular risk factors. If an individual has a few risk factors, the incidence of persistent atrial fibrillation is much lower than if you have a greater number of risk factors, in which case it is more likely that these individuals will progress to persistent atrial fibrillation.”

“Dr. Nattel’s work has kindly demonstrated the central role of atrial fibrosis in the mechanistic method of this substrate for atrial fibrillation. Whether it is with ageing, heart failure, hypertension or mitral regurgitation, these result in atrial stretch, ionic remodeling and atrial fibrosis. Atrial fibrosis itself adds back to this, and results in persistent atrial fibrillation. There is clear evidence to suggest that AF feeds back on itself to remodel this process, and there is even a suggestion that it may induce further atrial fibrosis.”

Referring to work by Dr Jeffrey Garrett looking at aggressive cardioversion to try and maintain sinus rhythm, Professor Sanders relayed how they were able to show that most patients’ AF recurred, suggesting that there is another factor in the promotion of AF. Going back to Dr. Nattel’s work with substrates in AF, he continued: “This is the ventricular pacing model of heart failure, where they were able to show that in the atria there is an increased amount of atrial fibrosis associated with conduction heterogeneity, and increase of the duration of atrial fibrillation.

“Many years ago we were able to look at this in the clinic, using patients with an ejection fraction of less than 35%. Voltage maps of the atria demonstrated large areas of low voltage, increased areas of scarring and a more complex electrogram in a patient with heart failure, when compared to control… This has become what is a common feature for the substrate of atrial fibrillation.”

Professor Sanders has been able to extend this for a number of other models. For instance, using sheep models with progressive hypertension from baseline to 15 weeks, his group were able to show activation maps with a progressive delay and increased curvature, and isochronal crowding, suggesting that induction is progressively becoming more abnormal. “From baseline to 15 weeks there was an increased amount of fibrosis in these animals,” he said.

In other work looking at sleep apnea patients (apnea-hypopnea index, or AHI, of 38), compared to an age- and gender-matched control (AHI of 6), when looking at voltage maps in the left and right atrium, large areas of low voltage and fractionated points were apparent for the sleep-apneic patient. “Recently this data has been extended again by Dr Nattel’s group, where they looked at the impact of repetitive obstructions in an animal model,” said Professor Sanders. “They were able to show some of the mechanistic features of this: an increase in atrial size; slowed atrial conduction; increased fibrosis of the atrial myocardium; reduction in connex-in-43 and, no surprise, induction of atrial fibrillation was greater in the animals who had repetitive obstruction mechanically.”

He added: “We have now evaluated a number of conditions, both in the clinic and in the preclinical scenario, and it seems fairly uniform that the substrate that predisposes to atrial fibrillation is structural changes, be they interstitial fibrosis,low-voltage and scarring – the result is abnormal conduction.”

Moving on to discuss obesity models, Professor Sanders remarked that when taking an animal from a weight of around 50-60 kg up to around 100-110 kg, you will see the increase in fibrosis, and rate of damage to the atria in terms of fibrous replacement. “We see in our activation maps – from baseline to four months when they are overweight, and eight months when they are obese – increased delay in activation and isochronal crowding, suggesting delayed conduction,” he said. “In addition, there is a greater inducibility of atrial fibrillation.”

This has been extended by more work Professor Sander’s team, in collaboration with a Parisian group, in which they have been able to show that the pericardial fat volumes increase in this model as the animal puts on weight. “Where the pericardial fat is adjacent to myocardium, we see infiltration of the myocardium with fat cells,” he said. “And this may be a unique substrate in obesity on its own, compared to the other substrates that we have evaluated. We’d like to know if this process is reversible, [but] unfortunately studies to date have suggested this is not the case.”

Delving deeper into the obesity animal studies led by his group, Professor Sanders described a particular setup involving sheep with sustained obesity for 72 weeks, i.e. chronic obesity, who underwent diet control to get them to lose weight over an eight-month period, with two distinct 15% and 30% weight reduction measurements being taken along the course of weight loss. Weight of the animals rose from an average of 58 kg to 111 kg, and was then reduced to 94 kg and 77 kg respectively for the 15 and 30% weight loss measurement points. Fat increased immensely from 7 kg to an average of 40 kg for the obese group, but halved again by the time of the 30% weight stage. Left atrial pressure and pulmonary artery pressure increased, and left atrial size increased, although the animals were too big to fit into scanners to measure this exactly.

“We undertook a battery of investigations to really characterize the atrial tissue itself,” said Professor Sanders, adding: “With respect to the conduction abnormalities, and focusing on the obese versus lean, control animals, we saw a longer time to activate the atria, also isochronal crowning, and indeed areas of conduction that we’d never seen.

“With 30% weight reduction we were back to what we’d seen in the lean controls, i.e. removal of the conduction abnormalities. To our surprise, when we looked at atrial fibrosis … we saw that in the obese atrium, there were areas with increased fibrosis, but with 30% weight reduction the fibrosis was no longer there, and in fact it was back to the levels that we’d see in the lean controls. Similarly, connexin-43 reduced with obesity and then normalized with 30% weight reduction.”

Summarizing the major changes seen in the investigation, Professor Sanders stressed that there were improvements in weight, conduction heterogeneity, left atrial pressure, atrial TGf beta 1, inflammatory cell counts and fibrosis, alongside normalized conduction velocities and reversal of connexin-43 and ETR-beta.

Offering his conclusions, he said: “Perhaps modifying cardiovascular diseases can modify our substrate for AF, and this is an area in which we hope to see some incredible changes in the next few years.”

Tammy Griffin-Kumpey