Obesity, Mortality, and Bariatric Surgery Death Rates

Despite widespread recognition that obesity is one of the most important threats to good health and longevity, its treatment remains remarkably stubborn. This is especially true for morbidly obese individuals. Diets, exercise, and medications usually only lead to modest and transient amounts of weight loss.¹ Only bariatric surgery results in substantial, long-lasting weight control. Despite their successes, bariatric operations have not been uniformly accepted and many insurers will not provide coverage for them. This results from uncertainty regarding the safety of these procedures, with the principal concern being perioperative mortality. Bariatric operations are elective and even though the resultant weight loss confers many benefits in terms of comorbidity control and quality of life improvements, perioperative death rates should not exceed those that might be expected from medically treated obesity.

Estimates of obesity-related mortality have been inconsistent. Although it is generally agreed that obesity is a risk factor for premature mortality (with relative risk ratios in the 1-2 range),² several large-scale epidemiological studies did not find an association between obesity and the risk of premature death.^{3 - 6} Obesity is a heterogeneous disease. Obese individuals with large collections of fat in the lower body have a lower risk for cardiovascular disease than those with predominantly upper body (ie, central) obesity.⁷ Any population-level obesity-related mortality risk will depend on many factors including race, socioeconomic status, comorbidity control, and genetics. In the United States, it is generally agreed that obesity results in increased risk for mortality; however, risk estimates have been downgraded,² most likely due to improved comorbidity control that has been observed in recent years.⁸ The reduced importance of obesity as a cause of death⁹ places increased pressure on bariatric surgery to have an exceedingly low mortality rate to remain acceptable as a form of therapy.

Bariatric surgery has been on the defensive because of several studies showing high mortality rates.^{10 - 12} Two articles^{10 ,12} reported high death rates for Medicare patients, who are inherently high-risk operative candidates either because of their advanced age or disability status. More than 90% of Medicare bariatric surgery patients receive Medicare because they are disabled. These individuals have a high operative risk because of their large body size, greater number of comorbidies, and lower socioeconomic status.

One study¹¹ reported on a relatively small number of patients over 15 years in Washington State. During most of the years of that study, bariatric surgery was not readily available and insurance companies tended to only approve surgeries for patients with serious obesity-related complications who were also at high risk for perioperative complications and deaths.

What then is the expected population-level bariatric surgery mortality rate and is that rate lower than the risk of nonsurgically treated obesity? Findings that help answer these questions are reported in an important article published in the October issue of the Archives of Surgery. Omalu and colleagues¹³ reported results for bariatric surgery mortality from the Pennsylvania Health Care Cost and Containment program. Their results are similar to others found in the literature, suggesting that the expected perioperative mortality rates for bariatric surgery can be defined.

The in-hospital mortality rate for bariatric procedures as assessed in population-representative databases is less than 0.6%.^{14 - 21} Exceptions to this rate only include the special populations summarized above. Several studies have attempted to compare survival following weight loss surgery with that of obese control groups.^{11 ,22 - 24} In general, the closer the match between the surgical group and the control group, the smaller the magnitude of survival difference between the groups. Because none of the comparative studies were randomized controlled trials, they all have the same potential for selection bias. Patients who were highly motivated to lose weight and undergo surgery were compared with control groups who chose not to pursue aggressive obesity treatment. It is likely that patients willing to undergo bariatric surgery to control their weight are more invested in their health compared with those who do not have surgery and remain obese. The small survival advantage observed in matched cohort studies for bariatric surgery patients (1.3%-1.4%) might be reduced if the selection bias could be eliminated.

The report by Omalu et al helps place this risk in context. The 0.9% 30-day mortality rate is consistent with most other population-based studies, suggesting that the expected 30-day mortality rate for these operations should be less than 1%. The 6.4% mortality rate at 5 years suggests that the expected postbariatric surgery death rate is about 1% per year. These numbers have been consistent in the literature and should serve as a benchmark against which to compare other studies. For example, a Canadian postbariatric surgery study²³ reported a 0.68% mortality rate at 9 years for patients who had undergone bariatric surgery, more than 10-fold less than expected. The Swedish Obesity Study²⁴ reported a 5.0% mortality rate at 11 years but the preponderance of their patients underwent banded gastroplasties, which would be expected to have a lower perioperative mortality rate than gastric bypass operations. In contrast, the comparative study from Washington State reported an 11.8% mortality rate 15 years after bariatric surgery and a 16.3% mortality rate at 16 years for those who were obese but did not have surgery.¹¹ These latter numbers are consistent with the approximately 1% per year expected death rate and demonstrate a survival benefit for bariatric surgery.

Bariatric surgery has come a long way toward acceptance. Just a decade ago these operations were mired in controversy, although some controversy still exists over which patients should receive bariatric surgery and why some do not.²⁵ However, obesity is becoming more common and nonsurgical treatments generally only result in modest, transient weight loss. Surgery effectively controls body weight and obesity-related comorbid conditions¹ with the caveat that bariatric operations can be performed with low complication rates and low death rates. The study by Omalu et al and others like it demonstrate that this is the case.