Aug 21, 2017
Dr Carolyn Lam: Welcome to Circulation on the Run, your weekly podcast summary and backstage pass to the journal and its editors. And Dr. Carolyn Lam, Associate Editor from the National Heart Center, and Duke National University at Singapore.
What is the effect of obesity and underweight status on perioperative outcomes of congenital heart operations?
Our feature paper this week sheds light from the Society of Thoracic Surgeons Database. More soon, right after these summaries.
The first original paper highlights the role of micro RNAs in metabolic remodeling and heart failure. As a reminder, micro RNAs are small, noncoding RNAs important in post transcriptional modification and influencing many cellular processes simultaneously.
First author, Dr. Heggermont, corresponding author, Dr. Heymans, and colleagues from Maastricht University in the Netherlands use mice subjected to pressure overload by means of endotension to infusion or transverse aortic constriction. They show that micro RNA 146A was up regulated in whole-heart tissues in these murine pressure overload models, as well in left ventricular biopsies of aortic stenosis patients. Over expression of micro RNA 146A in cardio cardiomyocytes provoked cardiac hypertrophy and left ventricular dysfunction in vivo, whereas genetic knockdown or pharmacological blockade of micro RNA 146A blunted the hypertrophic response and attenuated cardiac dysfunction in Vivo.
Mechanistically, micro RNA 146A reduced its target dihydrolipoyl succinyltransferase or DLST, a mitochondrial protein that functions as a TCA cycle transferase. DLST protein levels were reduced in pressure overload mice, while they were partially maintained in micro RNA 146A knockout mice. Furthermore, overexpression of DLST in wild type mice, protected against cardiac hypertrophy and dysfunction in Vivo. Thus, micro RNA 146A and its target DLST are important metabolic players in LV dysfunction. These results also opened the door to novel therapies to treat metabolic disturbances and improve energy efficiency of a failing heart.
Program cell death is critically involved in ischemic cardiac injury, pathologic cardiac remodeling, and heart failure progression. Our next paper sheds light on the regulatory mechanisms of necroptosis and its significance in the pathogenesis of heart failure. Using genetic mouse models, first authors Dr. Guo and Yin, corresponding author Dr. Liu, and colleagues from University of Washington in Seattle, identified a critical role for a tumor necrosis factor receptor associated factor 2 or TRAF2 in myocardial survival and homeostasis by suppressing necroptosis.
The authors delineated an important TRAF2 mediated NF-KB independent pro-survival pathway in the heart by suppressing necroptotic signaling. They identified novel molecular mechanisms whereby TRAF2 suppressed TNF receptor 1 mediated, receptor interacting protein 3 dependent necroptosis, which is critical for myocardial survival and homeostasis. Thus, this finding suggests that the necroptosis suppressing TRAF2 signaling pathway and its effectors may serve as novel therapeutic targets for pathologic cardiac remodeling and heart failure.
Our next paper tells us that cerebral hyperperfusion may be associated with accelerated cognitive decline and an increased risk of dementia in the general population. First author Dr. Walters, corresponding authors Dr. Ikram, and colleagues from Erasmus University Medical Center in Rotterdam, The Netherlands, measured cerebral blood flow by 2D phase contrast MRI in non-demented participants of the population based Rotterdam study. A 4,759 participants with a median age of 61 years, and a median follow up of 6.9 years, 123 participants developed dementia.
Lower cerebral perfusion was associated with higher risk of dementia and this risk was even higher with increasing severity of white matter hyperintensities on MRI. At cognitive reexamination after an average of 5.7 years, lower baseline perfusion was associated with accelerated decline in cognition, which was similar after excluding those with incident dementia, and again, most pronounced in individuals with higher volumes of white matter hyperintensities.
Thus, lower cerebral perfusion was associated with accelerated cognitive decline and increased risk of dementia in the general population. This association was modified by hypertension and cerebral small vessel disease, possibly reflecting impaired arteriola and capillary function. This paper calls for further long term study and evaluation of optimizing cerebral perfusion as a means to prevent cognitive deterioration, for example, in patients with heart failure or carotid artery stenosis.
Well, that wraps it up for our summaries. Now for our feature discussion. For today's feature discussion, we will be looking at data from the Society of Thoracic Surgeons Database. This time looking at the effect of body mass index on perioperative outcomes of congenital heart operations in children, adolescents, and young adults. To discuss this, we have none other than the first and corresponding author, Dr. Michael O’Byrne from Children's National Medical Center in Washington D.C., as well as Dr. Naveed Sattar, Associate Editor from University of Oxford. Welcome gentlemen.
Dr Michael O’Byrne: Good morning.
Dr Naveed Sattar: Good morning.
Dr Carolyn Lam: Michael, we know that extreme body mass indices, very high or very low, has been associated with increased risk of at first, perioperative outcomes in mainly older adults undergoing cardiac surgery. We also know about the obesity paradox in conditions like heart failure, so why was it important to look at this specific group of patients? Congenital heart patients and children, adolescents, and young adults?
Dr Michael O’Byrne: Yeah, I think that as a pediatric cardiologist, a lot of the data that we use to guide our management is extrapolated from adult studies. However, in this particular case, it wasn't clear necessarily that adult data would necessarily be applicable to children and adolescents and young adults. We are aware that there are epidemiologic trends that congenital heart disease population ages and there are also in increasing problems of obesity among children in the United States.
The convention wisdom among surgeons in the United States is that obesity would increase perioperative risk and the thought is that some combination of exposure to hypertension and diabetes and peripheral vascular disease might impede wound healing and that body habit as itself might be a risk for the technical approach in wound healing. Acknowledging that there's a lot of evidence both for extreme BMI being a risk in surgical patients and adults, but also the idea that obesity paradox might be important in children because the biological mechanisms might be different.
Children themselves are exposed, their sort of dose response or dose exposure is less, they're younger, and so haven't been obese for a prolonged period of time, so that the integrated effect of having diabetes, hypertension, and obesity might be less. At the same time, we also acknowledge that in children with heart disease, we have congenital cardiac disease, the same issues with cachexia and frailty are present. i.e. that children with very low body mass index might be assigned to their own medical frailty, or a part of a heart failure cachexia syndrome.
One of the challenges in dealing with children with congenital heart disease, however, as you know is that its rarer than cardiac disease of the aging and additionally, that the population is very heterogenous in terms of the actual defects that are present and the surgeries that are performed. It was relevant to look and see over a wide range of sort of technical complexity surgeries with a wide range of sort of intrinsic preoperative risk of perioperative outcome, whether or not BMI would be associated with an adverse outcome. Either operative mortality in this case, or a composite outcome of mortality, major adverse events, and wound infection.
Dr Carolyn Lam: Wow, that makes a lot of sense and congratulations. This is not just the first, it's huge and really comprehensive. Could you just tell us a little bit more about what you did and what you found?
Dr Michael O’Byrne: I think as this point, I'd have to acknowledge that the challenges that we described in terms of both a sample size and in terms of getting a representative sample, is a constant challenge in our field and we have to give credit to my co-authors Marshall and Jeff Jacobs for their work in developing the collaboration that allowed for the STS Congenital Heart Surgery Database to exist. Also, on top of shepherding the database, their research, along with the people at Duke Clinical Research Institute, they've developed a robust risk stratification model for mortality that we utilize as part of this study. Without that, this would be really be very challenging.
What we did is performed an observational cohort study using the STS Congenital Heart Surgery Database to look at the risk of perioperative mortality and composite outcome in patients undergoing surgery in the United States between 2010 and 2015. We looked at both the actual events, the sort of observed events, in terms of mortality and adverse events, and then created multivariate models to adjust for the known covariance.
We hypothesized that extreme BMI, either very high or very low, would be associated with increased risk of mortality and increased risk of that composite outcome. What we found that operative mortality and that perioperative adverse events occurred more frequently in obese and severely underweight subjects. However, because they have an unequal distribution of potentially important covariance, we used multivariate modeling to adjust for those covariance.
Our multivariate models for death, however, the severely underweight subjects had an odds ratio of 1.4 and obese subjects had an odds ratio of 1.3, but neither was specifically significant in that context. We sort of anticipated that with a possibility given the very low event rate. That's the reason we've used a composite outcome, a higher event rate.
For that composite outcome, in both different versions of the multivariate model that we used, the severely underweight subjects had an odds ratio of 1.5, underweight subjects had an odds ratio of 1.3, and obese subjects had an odds ratio of 1.2. An increased risk in all three of those populations of interest relative to normal weight or just overweight subjects.
Dr Carolyn Lam: We're always saying that at circulation we do want to publish papers that have direct and important clinical implications, so Naveed, could you share some thoughts on what this means clinically?
Dr Naveed Sattar: Yeah, I think they went through the review process and I think the paper was very well written. I think Michael and his colleagues clearly understood the strength and the limitations of the data so that you can only ever itself prove associations here and therefore, clinically when we push them on trying to make clinical inferences, I think clearly they recognize that once they find associations between obesity and adverse outcomes and underweight.
What they need to do next, now this is a paper that then leads you to think, "Well actually, I need to do some clinical trials to prove that module ..." You're preventing these outcomes or in very under knowledge where they're actually increasing the BMI but improving their nutrition, cannot also improve outcomes following surgery. Now those are tough things to do. Michael, what do you think from some of the clinical inference? My inferences were the associations were there, particularly for the normal [inaudible 00:12:35] outcomes, but actually to prove that, to make a difference, you probably might need to do some intervention trials or is that how you take it as well?
Dr Michael O’Byrne: I agree with you 100%. I think that as an epidemiologist, I think that what we see in an observational study like this is an association. The two next levels of research that are necessary at this point are to see whether or not in this population BMI is a modifiable factor in the short run before surgery, or even in the long run. And the second question to answer is whether those adjustments in BMI, if they are achievable, affect outcome with surgery. Absolutely.
It's a tremendous challenge, both logistically in organizing a study, and honestly, in terms of capturing a cohort that would be large enough, given that this is almost 100% of the surgeries that occurred over a six year period in the United States.
Dr Naveed Sattar: I looked at it and thought, "Well, the mortality association once you adjusted were not quite significant but are there any individuals you would not do surgery on based on their BMI based on these results?
Dr Michael O’Byrne: The motivation for the study is exactly to try to begin to shed light on that kind of question. I think that it might be what I would call a tiebreaker potentially, if you have a situation where a patient is near meeting criteria but isn't quite at a place where you need to do surgery at that point. It might dissuade you from proceeding immediately potentially pursuing a course that might adjust their BMI in the correct direction.
At the same time also, in a patient who's underweight particularly to evaluate whether their medical regimen has been optimized and if there are other residual lesions that can be addressed in a non-surgical or medical fashion.
Dr Naveed Sattar: I suppose the other trick with this type of research research is always trying to make sure that people understand these are the associations and not trying to attribute causality because it's always physical, isn't it? But I think you and your team did that very well and I'm sure we had a back and forth with review but I think your discussion section, your limitation section, is beautifully written and covers those kinds of caveats, which I think is important as well.
Dr Michael O’Byrne: I thank you for that. That's very complimentary and we certainly strived for that, but I think that you as an editor, and also in terms of the reviewers also, were very helpful in that sort of collaborative process to try to make sure that we're communicating it. It's not always clear in a project that takes months and years to finish when you're writing it necessarily, you may be constantly aware of trying to be clear in your communication but it's also helpful to have a reviewer from the outside carefully read the study.
Dr Carolyn Lam: That's wonderful and Michael, may I just join Naveed in congratulating you on beautiful paper? And maybe just one other little question, did you have any insights into the mechanisms of increased risk for composite events in the extremes of BMI?
Dr Michael O’Byrne: I think it's an important question. There's been a tremendous amount of research in adult cardiac disease about whether it is the BMI as a steady state or BMI changes immediately before and after surgery that are relevant in this case. From this kind of observational study, it's very hard and very challenging to try to make any sort of inferences about the causes. It would be an important part of any study moving forward to include ways to investigate that, and honestly, as an interventional cardiologist and epidemiologist, I probably would defer to Naveed, he might have more cogent and logical ideas about that than I do.
Dr Naveed Sattar: We've had lots of research from a whole variety of researchers. We all understand it's finally serious but recognize it's difficult, so one of the ways moving forward and I think Michael and his colleagues have this is if you have serial BMI data prior to surgery, that could try and inform on reverse causality because of the low BMI, but in terms of the mechanisms, remember these are associations, but I think mechanisms are well covered if you are obese and clearly you have risk factors for death, across the vasculature, across the cardiac functions, across the whole variety of things.
We know those mechanisms, question is, to what extent are they actually operating and causing increased risk in the surgical arena and that's a really tough ask. I think people can come up with a multitude of mechanisms. I think the key things, like this particular paper, is that there are potential mechanisms but these are associations ... Look, this is what we found, and clinically now we need to try and address this within the following types of interventions or at least provide some guidance to colleagues and clinicians.
Exactly as Michael says, if there is somebody who is approaching surgery whose quite obese, perhaps they should try and intervene to try and lessen their weight for a short period of time prior to [inaudible 00:17:07], you know what happens. It would be nice to do some big trials but I think doing trials in this area is going to be really tough, but with imagination, with good collaboration across centers, trials are not impossible. I think they can be done.
Dr Michael O’Byrne: Naveed, I think, actually articulated what I think is both the difficulty of doing that trial but also the importance of it. I think that looking at ... In these databases, we don't have a serial BMI and I think that's an important missing piece of information that we tried to address in our discussion and I think it's something that would be really valuable moving forward. And certainly testing interventions, whether they're medical, interventional, or surgical, to help these patients who are obese either lose or maintain an appropriate weight is the next step.
On the converse side, this research highlighted to me the prevalence of chachectic or underweight patients in our population and it's something that outside of the infant period, we don't necessarily think about tremendously and we don't think about it as a modifiable factor. I think that's another group of patients who also deserve some attention.
Dr Carolyn Lam: Listeners, you've been listening to Circulation on the Run. I'm sure you learned a lot as I did. Don't forget to tune in again next week.