Friday, December 7, 2007

Twenty stomach surgeries may save one life at best

stomach surgery
But in smaller clinics without star surgeons, it may take up to a hundred gastric bands to prevent one death of an obese patient in five years. These are the very mixed results of three recent studies on the long-term effect of bariatric surgery.

First, let's take a comparing look at everyday routine cases like a sore appendix where surgery is necessary to prevent a risk of death within a few days. No question, appendix surgery is simple and immediately life-saving.

Gastric banding, on the other hand, is complicated, and the question if it is life-saving must remain open, without clear-cut answers. Too much depends on the details. Three studies have been published recently, telling us that bariatric surgery (gastric banding) may reduce the relative risk of death in the following five years by 64 percent in Italy, 72 percent in Australia or even 89 percent in Canada, compared to obese patients without surgery.

Translating relative risks to real figures

The 89 percent lower risk reported in the Canadian study sounds impressive. But what does it mean? In a group of 1035 surgical patients, seven (0.68 %) have died, and in a larger group of 5746 patients without surgery, three hundred fifty-four (6.17 %) have died. Without surgery, 64 of the 1035 patients may have died within five years. That is, for every life saved, 18 surgeries have been performed, about twenty given the statistical uncertainty.

Applying the same calculation, the numbers of surgeries needed for saving one life are about 58 in Italy and about 96 in Australia.

Much depends on the surgeon

It is interesting to see how the best performance in Canada has been reached. All 1035 surgeries have been done at one high volume clinic at McGill University Health Centre in Montreal, Canada. One single star surgeon performed 49 percent of all cases, another 48 percent have been performed by three other surgeons, and the tiny rest by another three surgeons.

Now let's have a look at Germany where no world's best teams are available. In a low-volume clinic at W├╝rzburg, only eighty-five obese patients within six years have received a gastric band. The study reports only 37 percent success compared to 63 percent failures where either the bands had to be removed or the weight loss goal has not been reached. Long-term mortality has not been assessed in this study, but looking at Australia or Italy, one may guess that much more than a hundred gastric bands may be required to save one life in Germany. With a life-saving success of less than one percent, we get close to the world-wide mortality rate of 0.25 percent after bariatric surgery.

Is it the weight loss, anyway?

Looking again at the best possible conditions in Canada, we see that a stomach band may lead to weight loss and may reduce mortality. But weight loss, a clearly visible sign of long-term success, is not the most likely cause of better health but rather a visible sign of it, too.

Many of the very obese patients are diabetics, and after bariatric surgery, diabetes improves within weeks, long before any weight loss can be seen. Thus, a very strong positive effect of gastric banding is independent of weight loss. A number of favourable changes in hormones and metabolism are caused by the banded stomach. These changes influence the risk of, for instance, a heart attack on the one hand and lead to a reduction of body fat on the other hand. In other words, reduced risk as well as reduced body fat may both be caused by gastric banding.

Conclusions: If not done in a high-volume clinic by top surgeons, gastric banding may have a poor success. Weight loss is a visible long-term outcome of successful gastric banding (bariatric surgery), but immediate and ongoing changes in metabolism are the most likely causes if health improves after surgery.

Photo credit:

No comments: