Bleeding Complications: Talking to a Gastroenterologist

Dr. Bhala was the senior author of a July 2011 British Medical Journal clinical review article, “Anticipating and managing bleeding complications in patients with coronary stents who are receiving dual antiplatelet treatment.” We talk with him to find out the non-interventional cardiologist’s view of post stent bleeding complications.

You are a gastroenterologist. Why do a clinical review on post-stent bleeding complications?

As a physician and a gastroenterologist, I have always been interested in bleeding as an outcome. Although I am more on the treatment side, bleeding complications are potentially preventable in many cases, particularly those that occur in the gastrointestinal tract. During my research experience, I have been working in the clinical trial service unit in Oxford, which has done some of the studies looking at the vascular benefits of anti-thrombotic treatments. So I have seen both sides of the equation and that was one of the reasons why we did this study, to talk about trying to get a balance of benefit and harms. Also, we highlighted the need for different specialists to work together in order to both prevent bleeding complications and if they do occur, treat them.

What did you find?

We did a narrative review, particularly focusing on some of the large-scale trials and some of the issues related to bleeding such as thrombotics, stents, and also clinical practice guidelines. Although for the particular question of anticipating and managing bleeding complications, sometimes you cannot get all your questions answered by the randomized trial evidence. Still, what we compiled is relevant to quite a wide patient population — especially those who had stents and are on antiplatelets.

There are a number of clinical bleed scores which have looked at the risk of bleeding in people who have been stented, but broadly, when we talk about bleeding, it falls into three categories: gastrointestinal bleeding, procedural-related bleeds and/or the bleeds into vascular access sites, and intracranial hemorrhage. There are other types of bleeding, which are fewer in number, but some of which are also fatal, so there is a spectrum of severity. Most bleeding scores probably focus too much on the overall outcomes, lumping these bleeding types together. We tried to give some precision and some overall clinically relevant guidance in this topic.

Physicians often do not feel comfortable predicting which patients might suffer a bleed.

With risk, it can often be difficult. Even if you risk score someone, it does not necessarily mean they will not have a bleed, simply that this patient is low risk according to that score. It can be slightly akin to trying to predict the weather in the longer term. But broadly speaking, there are commonplace signs. Within my own specialty, gastrointestinal bleeding, there are some fairly clear risk factors. Far and away the biggest one is if someone had a previous ulcer. But it can often get a bit specialty-focused, meaning that people focus on answering the questions relevant to the cardiovascular risk, and forget about some of the commonplace questions that are equally relevant to the bleeding risk. These are often things that can be found in the past medical history. Also, as people get older, their bleeding and vascular risks increase. Appropriate risk stratification for bleeding risk, not just cardiovascular risk, is a key issue. Hindsight is a wonderful thing and often when we go back, we can see that there were one or two pertinent risk factors that were potentially treatable and/or preventable.

Can you give us an overview of the bleeding complications you discuss in your review?

In terms of bleeding from vascular access sites, your readership will be well versed in handling this particular complication. There have been trials looking at the transradial or transfemoral approach, for example. In terms of the common vascular access site, complications include hematomas, pseudoaneurysms, infection, ischemia or bleeding. There is clearly a spectrum between a small bleeding hematoma that can be managed conservatively, and a large or enlarging hematoma which requires investigation. Retroperitoneal bleeding is rare, but the classic example might be if someone comes into an emergency department with abdominal pain and hypotension post-procedure. Here is a case where we need to make sure there is good communication in place. The ED physician needs to phone the interventionalist who would then say, we should rule out retroperitoneal bleed.

Intracranial bleeding is fortunately a very rare outcome, but catastrophic. In people who present with any sudden onset focal neurology after a stent insertion and on antiplatelets, there needs to be a high index of suspicion for this. It requires radiological and neurological modalities of management, as well as up-to-date guidelines. A comprehensive review was done in the BMJ in 2009 on this topic.²

Finally, gastrointestinal bleeding is one of the commonest causes of bleeding post stent, particularly in people who are taking dual antiplatelet therapy. Again, this is a case where someone will come in as an emergency not to the cardiologist, but as an emergency to a gastroenterologist or the emergency department, highlighting the need for multidisciplinary management. You would want to institute the appropriate treatment, such as proton pump inhibitors, and you would also want to do a very prompt endoscopy. But then it is very much a question of trying to balance the gastrointestinal bleeding risks and the cardiovascular risks, and that requires good communication between both the gastroenterologist and the cardiologist.

What about the appropriateness of proton pump inhibitors for someone who is on clopidogrel?

There has been some uncertainty about the use of proton pump inhibitors with clopidogrel, on the basis of some warnings by the FDA, because of the proposed pharmacodynamic interaction between clopidogrel and omeprazole.  Since that time, there has been a large, randomized trial, the COGENT trial, carried out in the U.S. and published in the NEJM last year.³ In this trial, there was a definite effect of proton pump inhibitors in reducing gastrointestinal outcomes, as one would expect, but there isn’t any reliable evidence of a difference in cardiovascular outcomes when a patient is put on a proton pump inhibitor and clopidogrel compared to a proton pump inhibitor alone. Some of the previous studies that initially put this hypothesis of a vascular interaction were small and potentially biased. I would prefer to look towards the randomized evidence, such as the COGENT trial, for a reliable evaluation, but there are also other proton pump inhibitors apart from omeprazole, so there are other treatment options available.

Can you discuss the importance of a multidisciplinary effort in managing these patients?

The challenge in terms of modern medicine is that often we can become very specialty-focused. I am happy to put my hand up and say that I too am guilty as a gastroenterologist. But while the cardiologists will often focus on the stent, the gastroenterologist focuses on the upper gastrointestinal tract, and both forget about the rest of the patient. The best management requires communication between the relevant specialists. The example we gave in our review is of the gastroenterologist and the cardiologist, and a consideration of the different risk factors regarding the longer-term management of the patient, even though clinical practice varies. People feel very cautious about giving antithrombotics, but clearly there is a proven evidence base that they will be beneficial in such patients. There is a real need for multidisciplinary management and thinking holistically about the patient overall, particularly considering the vascular risks.

The biggest issues arise when patients have an intervention, then go home and maybe a few weeks after, come into a different physician. We need to break down existing specialty silos and make sure that the overall healthcare organization fosters communication, both for interventional cardiologists in this setting and the relevant specialties where patients may present with relevant outcomes such as bleeding.

You anticipate that in the future, “bleeding complication management will become increasingly complex.” Why?

We focused in the narrative review on dual antiplatelets and stents because that is probably the most common combination that we currently see. There are other drugs that are highly relevant to bleeding, including anticoagulants and non-steroidal anti-inflammatory drugs, and all of these affect both bleeding risks, but also cardiovascular risks. We are seeing older patients with more comorbidities who are getting stents and requiring more complex management, which also applies for ensuing bleeding complications. The truth is, some of these issues are going to become progressively more complicated, so getting the basics right is important.

Dr. Bhala can be contacted at nijbhala@doctors.org.uk.

References

1. Bhala N, Taggar JS, Rajasekhar P, Banerjee A. Anticipating and managing bleeding complications in patients with coronary stents who are receiving dual antiplatelet treatment. BMJ 2011 Jul 21;343:d4264. doi: 10.1136/bmj.d4264.
2. Al-Shahi SR, Labovitz DL, Stapf C. Spontaneous intracerebral haemorrhage. BMJ 2009 Jul 24;339:b2586. doi: 10.1136/bmj.b2586.
3. Bhatt DL, Cryer BL, Contant CF, et al. Clopidogrel with or without omeprazole in coronary artery disease. N Engl J Med 2010 November 11;363(20):1909-1917.