Death rates after surgery: an interview with Dr Rupert Pearse

Interview conducted by April Cashin-Garbutt, MA (Editor)Oct 1 2012

How many people were previously estimated to die following general surgery?

There are very few published estimates describing the number of deaths following surgery at a national level and none at an international level.

Previous estimates have suggested between 1 in a 100 and 2 in a 100 patients will die within 30 days of surgery.

How different were recent estimates of death rates following surgery?

The estimate from this most recent work suggests that the mortality rate could be as high as 4 in a 100 patients who have surgery, which would be as much as double the previous estimates.

However, the design of the study is very pragmatic and although it does show that there is more work to do to improve outcomes of surgical patients, we should bear in mind that this is only another estimate.

How were these death rates calculated?

We engaged with investigators throughout Europe at a national level and at a local hospital level and asked them to collect data on every single patient who had surgery within a 7 day period. We also asked them to follow those patients until they left hospital to tell us who were admitted to critical care and who died.

They then reported this data to us via the internet using an online data capture system. We then performed statistical analysis to create a scientific report.

Why do you think previous estimates were so optimistic?

I’m not sure the previous estimates were optimistic. Lots of us have been concerned for some time about the need to improve outcomes for surgical patients.

In particular, mortality rates may be a lot higher for a high-risk sub-group of patients who tend to be older, have other medical problems and undergo major surgery. This high-risk group needs a lot of work to improve patient outcomes.

What are the main causes of death following surgery?

Very few of the deaths amongst surgical patients happen during the surgery itself and even fewer are related to any kind of error that a doctor may make. The great majority of deaths following surgery happen because of medical problems such as hearts attacks, pneumonia and stroke. These are more likely, partly because the patient is pre-disposed to them because of their medical problems and partly because the surgery itself makes them more likely.

How do death rates vary between countries?

Our study does suggest some differences between countries across Europe although this observation needs to be interpreted with care, in particular for some outlying nations. However, in countries where there are lots of patients, like the UK and Germany, it does appear that there are differences in outcomes between health care systems that are operating to a similar standard.

This variation suggests that there are preventable deaths occurring and that is the real importance of this work. It gives a strong argument that improving the quality of patient care, by applying what we know works well, may reduce the number of preventable deaths after surgery.

Did your research adjust for confounding variables?

Yes, we adjusted extensively for confounding variables. These included:

• age
• the number of medical problems a patient might have
• whether the surgery was an emergency
• what type of surgery it was

We also adjusted for other factors that might differ between hospitals and countries.

Did your research look at how death rates varied following different types of surgery?

We do have data showing that the mortality rates differ between categories of surgical procedure. However, we are inclined to interpret these cautiously, because when you divide the patient sample into smaller categories, the margins of error become more of an issue.

However, what we have learnt from previous studies and this recent work is that patients having abdominal surgery, in particular abdominal gut surgery, are a particular concern.

Are there any plans in place to try to improve death rates following surgery?

Yes, absolutely. We have been working for some time with research colleagues, with national societies, with government organisations to improve the quality of patient care for surgical patients. The findings of this study show that we must continue to increase this important work to improve the number of preventable deaths that occur after surgery.

What impact do you think your research will have?

I hope the research will stimulate a discussion among people involved in the care of surgical patients and that doctors will think more carefully about the needs of individual patients.

I hope also that it will allow us to focus more on measuring the rates of death after surgery on a more regular basis. In particular, robust, public audit of patient deaths will really help to improve outcomes. We wouldn’t then need to rely on studies like this one, where we just capture data for a short period of seven days to give us an estimate of what happens.

Do you have any plans for further research into this area?

Yes, we are conducting various studies at the moment as well as proposing new ones. A major area of focus is on how we can identify the patients that are at high risk before they have surgery allowing us to better plan the care patients receive.

We are also looking at ways of improving the quality of patient care for patients in the highest risk groups. For example, patients who undergo an emergency laparotomy - an emergency procedure to correct a problem in the abdomen. This is a patient group we know have particularly high mortality rates.

There are also many other initiatives in the UK and internationally looking for similar solutions.

Where can readers find more information?

They can read our paper here: http://www.thelancet.com/journals/lancet/article/PIIS0140-6736%2812%2961148-9/abstract

About Dr Rupert Pearse

Rupert Pearse graduated in 1996 from St George's Hospital Medical School, London, UK. After time working in general medicine and anaesthesia, he returned to St George's Hospital where he developed many of his current research interests and completed his training in Intensive Care Medicine.

In 2006, he was appointed Senior Lecturer in Intensive Care Medicine at Barts and The London School of Medicine and Dentistry and was promoted to Reader in 2011. He has now given up anaesthesia to concentrate on his clinical duties on the intensive care unit at The Royal London Hospital and his research interests in improving outcomes following major surgery and the cardiovascular pathophysiology of critical illness. Rupert plays a leading role in a number of large multi-centre studies in peri-operative medicine.