JFS Special: The latest research on actual deaths seen after surgery for weight loss

Will bariatric surgery for weight loss extend or shorten a patient’s life? After more than 40 years of these medical interventions being practiced, surprisingly, this basic question has remained unanswered. The largest and strongest study to date examining death rates after bariatric surgeries has just been published in the journal Archives of Surgery.

Researchers, led by Dr. Lewis H. Kuller, M.D., DrPH, at the Department of Epidemiology, University of Pittsburgh, compiled data on every bariatric surgery done in Pennsylvania on residents during the entire decade from 1995 through 2004, and compared them with mortality data. The Pennsylvania Health Care Cost and Containment Council collects data on all hospital and outpatient surgical procedures done in the state of Pennsylvania. The death certificates on all bariatric patients who died during that decade were identified and obtained from the Division of Vital Records and examined. Actual deaths are the most accurate figures to measure outcomes and the least amenable to statistical manipulation.

There were 16,683 bariatric surgeries done in Pennsylvania during that decade; 82.3% on women and 17.7% on men. The average age of the patients was 48 years (median age 49 years).

Most of the deaths (82.7%) occurred in patients who’d had their surgeries done at the hospitals performing most of the surgeries (90%), adding strength to this evaluation being of the procedures themselves, not surgical centers or surgeons.

During the first 30 days, about 1% of all of the bariatric patients had died. The highest post-op deaths were among those over age 55, with those over age 65 having more than a three-fold increased risk.

Thereafter, among the average age patients, annual death rates were about 1.31% for the women and 4.09% for the men. Confirming other studies, men, older people and blacks had higher mortality rates.

Looking at the cumulative deaths according to the time after surgery, they found that nearly 3% overall had died after the first year and 6.4% of the patients were dead by the end of the fourth year after their surgeries. They also looked at long-term risks, reporting:

We also estimated the long-term mortality for individuals who had undergone surgery many years ago. For the 1995 cohort who had at least 9 years of follow-up, 13.0% had died. From the 1996 cohort with 8 years of follow-up, 15.8% had died, and from the 1997 cohort with 7 years of follow-up, 10.5% had died. For the 1998-1999 cohorts with 5 to 6 years of follow-up, the total mortality was 7.0% to 2004.

Putting these figures into perspective

It may be hard to place any meaning to these figures without knowing what the death rates of the people might have been had they not had the surgeries. The authors stated that their study wasn’t able to make those direct comparisons as no figures are available and it is “unlikely, for now, that a true randomized trial of bariatric surgery versus nonsurgical treatment will be performed.”

We can, however, look at the death rates among Americans of the same average age and weight of the bariatric patients for comparison. Ignore, for a moment, the popular observation made by obesity organizations that if over 6% of fat people were dropping dead every four years, there would be no fat people left.

The U.S. National Center for Health Statistics of the Centers for Disease Control and Prevention data reports that the overall death rates among Americans of the same age is 0.352% — for men it is 0.44% while for women this age it’s 0.26%.

These are minuscule compared to the numbers dying after bariatric surgery. In fact, there is no data to suggest that fat people die at ten times the rate of most Americans to even approach these death rates. The most available figures are the latest estimates of deaths according to BMI led by Dr. Katherine Flegal, Ph.D., at the CDC, National Center for Health Statistics. Using NHANES I - III data, they found that among American adults of this age, those with BMIs of 35 and higher (qualifying for bariatric surgeries) had increased relative risks of dying of 25% for nonsmokers or 83% overall. (These rates are slightly higher than calculating actual deaths and estimated numbers of Americans with BMIs >35 in NHANES III. As Flegal and colleagues noted, the latest NHANES data indicates mortality rates associated with high BMIs have continued to drop since then with each NHANES.)

This equates to actual mortality rates for the most “morbidly” obese of about 0.44% to 0.64% — still not appreciably higher than the rest of the population and nowhere close to the bariatric surgical patients of comparable ages and weights in this study.

So, by best estimates, bariatric surgery likely increased the actual mortality risks for these patients in the first four years by 363% to 250%.

The body of evidence — distinctions between studies

This study provides more reliable figures, and differs in two very distinct ways, from other studies that have claimed to find bariatric surgeries improve long-term survival for ‘obese’ patients. First, it looked objectively at an entire population of patients — not just those who surgeons selectively chose to report on or didn’t lose in follow-up. This strength was emphasized in an editorial by Dr Edward H. Livingston, M.D., of the Department of Gastrointestinal/Endocrine Surgery at the University of Texas Southwestern Medical Center in Dallas. In the same issue of the Archives of Surgery, he wrote:

The power of these findings emanates from their being collected from all hospitals in one state where a great deal of bariatric surgical procedures are performed. Because these are state-level data, there is a greater likelihood that most, if not all, patients undergoing bariatric surgery are included in the analysis. This avoids the potential selection bias and incomplete follow-up that can occur in single-institution studies. Population studies from other states have found similar mortality statistics.

As Dr Kuller and colleagues noted, there have been two earlier published population-wide studies examining mortalities of bariatric surgical patients, both finding disturbingly high mortality rates. The strongest was the study led by Dr. David R. Flum, M.D., published in the Journal of the American Medical Association which examined all of the bariatric surgeries done nationwide from 1997 to 2002 covered by Medicare. As reviewed previously, of the 16,000 patients, with an average age of only 35-54 years, death rates at one year averaged 4.6%, but among patients 65 to 74 years old, nearly 13% of the men and about 6% of the women died.

A second study by Flum and colleagues examined 30-day mortality rates among bariatric patients throughout Washington state. They reported 1.9% died within the first month.

As Dr. Livingston wrote in his editorial review, the high mortality rates seen in these two objective studies “called into question the risk-benefit ratio for operations.”

While the Washington state figures are similar to these other population studies, any conclusions this Washington study made that the surgery may lower risks of dying among obese people were unsound, according to these authors, because they compared the bariatric patients to hospitalized sick people, not comparable obese people in the general population. As Dr. Kuller and colleagues said: “The obese nonsurgical sample, however, was substantially biased by selection for hospitalization, the Berksonian bias.”

The Kuller study spent some time explaining this bias because it is the second most common flaw seen in studies seeming to find bariatric surgeries (or diets or other weight loss interventions) improve outcomes for obese people. The use of hospitalized obese patients or sicker people as controls “introduces a substantial bias referred to as Berksonian bias” they wrote, because these controls are not representative of all obese individuals of that obesity class and are more likely to have higher mortality rates compared with obese individuals in the community.

Surprising findings

There was an especially disturbing and unexpected finding in this study. Dr. Kuller and associates found that among the bariatric surgical patients there had been “45 deaths from traumatic causes including 16 deaths (4%) due to suicide and 14 due to drug overdoses (3%) that were not classified as suicide.” To help readers realize how extraordinarily high these suicide death rates were, the authors wrote that according to the U.S. Vital Statistics, there are approximately 7 suicides for white women and 25 for men for every 100,000 people in the population of the same age as these bariatric surgical patients. Translating these figures, the researchers had anticipated an estimated two suicide deaths would have occurred among the women and one suicide among the men during this study.

Instead they saw more than five times the expected numbers of suicides, not even counting the drug overdoses. “There is a substantial excess of suicide deaths, even excluding those listed only as drug overdose,” they wrote. “The large number of deaths due to suicide and drug overdose, in excess of what we expected, is also a cause for concern. Most of them occurred at least one year after surgery.”

Dr Livingston’s review also noted that the frequency of suicide and drug overdoses “was unexpected.” The higher suicide rates were especially significant among those 25 to 34 years of age, with death rates of 13.8 and 5.0 per 1000 persons per year for men and women, respectively, nearly ten times the rates of 1.3 and 0.6 per 1000 persons per year in the general population.

While some have alleged that fat people are more likely to suffer from depression, this has certainly not been seen in studies of the general population of fat people. In fact, a study earlier this year in Archives of Internal Medicine, found suicide risk dropped by 11% for every 1 unit increase in BMI among the men. These researchers noted that most of the strongest studies have found similar associations and also pointed out that “increased prevalence of overweight and obesity in the United States in the 1990s was accompanied by a 6% decrease in suicide mortality rates.”

Such high suicide rates seen among bariatric surgery survivors abrogate claims that the quality of life is as improved for these patients after surgeries as much of the marketing appears to suggest.

We now have several large studies objectively examining the actual deaths associated with bariatric surgeries, all of which suggest the risks are considerably higher than they are among people of similar ages and weights. The vast majority of these people, as research is continuing to show, would not have seen higher death rates based on the health risk factors used for surgical eligibility, and with current clinical management of actual health problems. Hopefully, healthcare professionals and prospective patients and their loved ones will learn about this evidence and consider it carefully before making any decisions.

The American Society for Bariatric Surgery estimated 180,000 people in the U.S. had bariatric surgery last year and this trade group is working to ensure that the 15 million more who are eligible get the surgeries. Based on the most objective available evidence to date, this would equate to nearly one million deaths four years later.

© 2007 Sandy Szwarc