Archive for November, 2010

FDA Weighs Lowering the Bar for Bariatric Surgery

November 23rd, 2010

On December 3, 2010, an Food and Drug Administration Advisory Committee will consider a change affecting which patients would be eligible for Lap-Band(R) surgery. The current standard is a Body Mass Index of 35 with comorbid conditions (See Health section). The proposal from Allergan Inc. would lower it a BMI of 30 with comorbid conditions. This is a good idea. See my written comments below for why it should be approved.

 “I would like to submit the following comments for the Advisory Committee reviewing the application of Allergan Inc. to expand the indication for Lap-Band surgery to a lower Body Mass Index. I am the Editor and Publisher of the DowneyObesityReport.com and have worked in obesity advocacy since 1996 as executive director of the American Obesity Association and later as the executive vice president of The Obesity Society. I consult with a number of organizations, including the sponsor, on obesity policy and patient issues.

A. Background

The original Food and Drug Administration (FDA) labeling for patient selection for the Lap-Band of a Body Mass Index (BMI) 35 with cormorbid conditions criteria was taken from a 1991 National Institutes of Health consensus statement which is out-of-date and was not, and is not, based on scientific evidence. BMI is one of several alternative tools to measure obesity. However, it lacks reliability in key populations, such as women, African-Americans, Hispanics and persons of Asian ancestry. The BMI cutoff points are subject to change according to scientific evidence.  For most people, weight gain is incremental over many years. By the time a person reaches a BMI of 30, many adverse co morbid conditions are already well established. Weight loss resolves most but not all comorbid conditions. Surgery achieves the highest degree of weight loss compared to pharmaceutical products and lifestyle interventions. Changing the FDA labeling would allow physicians and patients to assess the appropriateness of LAGB without unnecessary and unsupported constraints which discriminate against a significant number of Americans.

A national goal has been established to reduce the prevalence and health effects of obesity. For certain patients at a BMI of 30 and above  with co morbidities, a LAGB procedure will allow earlier and more effective protection against future weight gain, adverse health effects and increased costs. Conversely, maintaining the current labeling will force patients wishing to have bariatric surgery to undergo gastric bypass (open or laparoscopic) since such surgeries are not limited by FDA labeling. Requiring obese patients to take a higher risk procedure when a safer and equally effective alternative is available cannot be considered sound medical practice nor ethically justified.

Leading medical opinion seeks ways to intervene earlier and more aggressively given the limitations in contemporary treatments for cancer, heart disease, type 2 diabetes and mental illness.

B. Origin of the current patient selection criteria

When the Lap-Band was approved by the Food and Drug Administration, bariatric surgery was regarded as a dangerous operation which should be reserved only for the most severe cases of obesity. There being no professionally recognized standards for patient selection of Lap-Band at the time (2001), the FDA and the sponsor company (then Inamed Inc.) agreed to follow the recommendations of the National Institutes of Health promulgated in 1991. The NIH guidelines (hereafter “Guidelines)” were the product of a consensus development conference and were titled, “Gastrointestinal Surgery for Severe Obesity.”

The Guidelines now appear on the NIH website with the following prominent disclaimer:

This statement is more than five years old and is provided solely for historical purposes. Due to the cumulative nature of medical research, new knowledge has inevitably accumulated in this subject area in the time since the statement was initially prepared. Thus some of the material is likely to be out of date, and at worst simply wrong.

(Accessed June 9, 2010 at http://consensus.nih.gov/1991/1991GISurgeryObesity084html.htm)

A review of the Guidelines (now nearly 20 years old) quickly brings out several points:

a. the purpose of the consensus conference was standards for surgery for severe obesity, a BMI of 40 or greater. The lower BMI discussion is almost an afterthought;

b. there was no scientific evidence discussed which weighed the BMI of 35 criteria against a BMI of 30;

c. there was no anticipation of the radical changes coming to bariatric surgery within ten years, namely the introduction of laparoscopic procedures in general and adjustable gastric bypass in particular; and

d. the description of the limitations of non-surgical interventions for severe obesity is not different from a description offered today.

The Guidelines’ discussion of these operations at BMI levels of 40 follows:

In certain instances less severely obese patients (with BMI’s between 35 and 40) also may be considered for surgery. Included in this category are patients with high-risk comorbid conditions such as life-threatening cardiopulmonary problems (e.g., severe sleep apnea, Pickwickian syndrome, and obesity-related cardiomyopathy) or severe diabetes mellitus. Other possible indications for patients with BMI’s between 35 and 40 include obesity-induced physical problems interfering with lifestyle (e.g., joint disease treatable but for the obesity, or body size problems precluding or severely interfering with employment, family function, and ambulation).

Obviously, what is said of patients between BMI of 35 and 40 can also be true of patients between BMI of 30 to 40. There is no further discussion of why a BMI of 35 was chosen as a cutoff. This is unfortunate because this Guideline is the only guideline which uses a BMI of 35. The United States Government and the World Health Organization use cutoffs points of 30 for obesity and 40 for severe obesity. The Food and Drug Administration uses a BMI of 30 or BMI of 27 with comorbid conditions for reviewing pharmaceutical products for the treatment of obesity.

It would appear that the NIH Guidelines, on which the FDA relied for its initial approval of the Lap-Band, were sui generis, not based on scientific evidence and not the objective of the consensus conference which produced them. It seems reasonable, then, to ask if the same measurement tool (the BMI) and the same cutoff (35) would be proposed today for LAGB.

The rapid technological changes in bariatric surgery have preceded changes in treatment guidelines. The original NIH Guidelines for bariatric surgery were promulgated in 1991. In 1994, Clark and Wittgrove showed that gastric bypass could be performed safely laproscopically. Laparoscopic Gastric Bypass, Roux-en-Y: Preliminar… [Obes Surg. 1994] – PubMed result The Lap-Band was approved by the FDA in 2001. In 1995, Poires et al showed bariatric surgery provided durable weight loss and significant improvement in comorbidities, including type 2 diabetes and an overall reduction in mortality over 14 years. Who would have thought it? An operation proves to … [Ann Surg. 1995] – PubMed result

Reflecting on the technological changes, as well as on studies by Sjostrom and Adams on reductions in mortality for patients undergoing bariatric surgery (see below), Dr. George Bray wrote in a New England Journal of Medicine editorial in 2007, “Has the time come to reconsider BMI guidelines for bariatric surgery? In addition to the improvement in the risk of diabetes, the reduction in deaths from cancer may also argue in this direction. Sjostrom et al, and Adams et al show that weight loss saves lives in obese patients. Thus, the question as to whether intentional weight loss improves life span has been answered, and the answer appears to be a resounding yes.” The missing link – lose weight, live longer. [N Engl J Med. 2007] – PubMed result

C. The Body Mass Index as the Measurement for Obesity

Obesity is universally recognized as a major health problem because of its impact on mortality and morbidity. But what exactly is “obesity.” Obesity refers to excess body fat mass or adipose tissue. Adipose tissue is a normal component of the human body. So, the concern is about “excess” adipose tissue. Body fat can be measured in many ways including skinfold thickness, hydrostatic weighing, air displacement, dual energy X-ray absorptiometry and bioelectrical impedance.

The Body Mass Index does not measure excess adipose tissue. It is a formula combining height and weight (weight in kg/height in meters, squared). The Body Mass Index has been shown to correlate very highly with excess adipose tissue, especially at a population level. At the level of specific groups, more variation is seen. At the individual level, reliance on the BMI alone may be unwarranted. The BMI can overestimate excess adipose tissue in younger or more muscular persons as well as in persons with edema. The BMI can also underestimate excess adipose tissue, especially in the elderly who have lost lean body mass.

 Waist Circumference has been discussed alone or in connection with the BMI to identify obesity related health risks in general, Waist circumference and not body mass index explai… [Am J Clin Nutr. 2004] – PubMed result   and for type 2 diabetes and cardiovascular disease risk in particular. Does Waist Circumference Predict Diabetes and Cardiovascular Disease Beyond Commonly Evaluated Cardiometabolic Risk Factors? — Diabetes Care

BMI in specific populations

Not all adipose is created equal. Visceral adipose tissue or VAT is more metabolically active than other adipose tissue sites and appears to contribute to many metabolic abnormalities associated with excess body weight. VAT is measured by waist circumference or waist-hip ratio. Some research suggests that BMI and waist circumference do not adequately measure visceral fat in different racial and ethnic groups. Visceral fat, waist circumference, and BMI: impact… [Obesity (Silver Spring). 2008] – PubMed result

The BMI is highly useful for population studies but it does not measure fat mass or percentage of fat mass for which there are no clearly defined cutoffs. In a recently published study, Okorodudu et al performed a meta-analysis of the correlation of BMI with body fat percentage. They found that commonly used BMI cutoffs to diagnose obesity have high specificity but low sensitivity to identify excess adiposity as they failed to identify half of the people with excess body fat percentage. Diagnostic performance of body mass index to ident… [Int J Obes (Lond). 2010] – PubMed result

Indeed, there is a population of normal weight but metabolically obese individuals, mainly women, who have normal BMIs but whose metabolic parameters are similar to that of persons with obesity, referred to a MONW, metabolically obese normal weight. They display high insulin sensitivity, high abdominal and visceral fat, higher blood pressure and lower physical activity. Characteristics of metabolically obese normal-weig… [Appl Physiol Nutr Metab. 2007] – PubMed result

Women

The BMI failed to identify nearly half of the reproductive-age women who met the criteria for obesity by percent body fat. Accuracy of current body mass index obesity classi… [Obstet Gynecol. 2010] – PubMed result

DeLorenzo and colleagues found in a study comparing BMI with percent of body fat and found a significant of both males and females with excess body fat percentage would not be considered obese by BMI alone. How fat is obese? [Acta Diabetol. 2003] – PubMed result

Minority Populations

The reliability of the BMI to identify cases of excess adipose tissue has been disputed in studies involving Mexicans (Diabetes and hypertension increases in a society w… [Public Health Nutr. 2005] – PubMed result), Hispanic-Americans (Interethnic differences in the accuracy of anthrop… [Int J Obes (Lond). 2009] – PubMed result (in which the BMI was described as being “almost uninformative.)”.

Also, the BMI’s reliability is questionable among several Asian heritage populations Are Asians at greater mortality risks for being ov… [Public Health Nutr. 2009] – PubMed result. Particular populations affected include: 

Alternatives

Many clinicians and researchers recommend evaluating overweight not solely by BMI but including other the presence of other diseases, smoking, blood pressure, glucose intolerance and fat distribution. Human variation and body mass index: a review of t… [J Physiol Anthropol. 2007] – PubMed result or elevated fasting triglyceride concentration. Abdominal obesity and the metabolic syndrome: cont… [Arterioscler Thromb Vasc Biol. 2008] – PubMed result

While the BMI has well-known limitations, it continues to be widely used because of its convenience, overall reliability and the lack of a widely-accepted, validated alternative. However, its limitations indicate that it is one of several tools to evaluate excess adipose tissue and it should be employed for clinical decision-making with some flexibility. Its employment as a barrier to access to a safe and effective treatment, such as LAGB, should therefore be subject to significant scrutiny.

It should be noted that the FDA’s use of the BMI threshold for bariatric surgery has created a Catch-22 for many surgical candidates. Many insurers require that candidates for bariatric surgery undergo 6-12 months of non-surgical weight loss program before being approved for surgery. Those who fall below a BMI of 35 are then rejected for surgery because the insurer claims it is not necessary, even though weight regain is very common. Those who do not fall below a BMI of 35 are also not approved because they are deemed “non-compliant.”

D. Adverse Health Effects begin in Overweight and continue into Obese BMI Categories

Despite the limitations of the BMI, there is a close relationship between it and the incidence of several chronic diseases caused by excess fat, including type 2 diabetes, hypertension, coronary heart disease and cholelithiasis. All risks are greatly increased for subjects with a BMI of over 29, independent of gender. Guidelines for healthy weight. [N Engl J Med. 1999] – PubMed result 

In a recent meta-analysis, Guh et al found overweight to be statistically significant for type 2 diabetes, breast cancer, endometrial cancer, ovarian cancer, colorectal cancer, esophageal cancer, kidney cancer, pancreatic cancer, prostate cancer, hypertension, stroke, coronary artery disease (but not congestive heart failure), asthma, chronic back pain, osteoarthritis, pulmonary embolism, and gallbladder disease. The incidence of co-morbidities related to obesity… [BMC Public Health. 2009] – PubMed result   (NB: review did not include sleep apnea, dyslipidemia,)

In a recent article in The Lancet, researchers examining 57 studies encompassing 900,000 adults found a BMI 22.5-25 was optimal for longevity. At a BMI of 30-35, median survival is reduced by 2-4 years. Body-mass index and cause-specific mortality in 90… [Lancet. 2009] – PubMed result. Adams et al found in an analysis of BMI at age 50, among non-smokers, the risk of death increased 20% to 40% among overweight persons and by 2 to at least 3 times among obese persons. Overweight, obesity, and mortality in a large pros… [N Engl J Med. 2006] – PubMed result

From Obesity to type 2 diabetes to cardiovascular disease

Many components of the physiological processes leading from excess adipose tissue to type 2 diabetes to cardiovascular diseases are still topics of active research. However, the progression from overweight to obese to diabetic is clear. The linkage between excess body fat (and central obesity) is probably high concentrations of free fatty acids, altered adipokine expression and low grade inflammation which contribute to B-cell failure leading to insulin resistance and type 2 diabetes. Persons with obesity with depressed insulin-mediated glucose can recover after weight loss. The road from obesity to type 2 diabetes. [Angiology. 2008 Apr-May] – PubMed result

E. Bariatric Surgery achieves superior weight loss and duration

Buchwald et al published a review and meta-analysis of type 2 diabetes after bariatric surgery covering 134 studies including 22, 094 patients from January 1990 to June 5, 2003. They found that bariatric surgery is followed by resolution of type 2 diabetes in 48% of patients who underwent laparoscopic adjustable gastric banding, 84% of patients who underwent gastric bypass, and 98% of patients with biliopancreatic diversion/duodenal switch. Caution is in order in assuming a causal link. 10% of diabetic patients are normal weight and ¾ of morbidly obese patients are not diabetic. Evidence has been mounting that changes in gut hormones may be important factors. Weight and type 2 diabetes after bariatric surgery… [Am J Med. 2009] – PubMed result

An earlier meta-analysis by Buchwald found bariatric surgery successful in resolving wholly or partially hypertension, type 2 diabetes, hyperlipidemia, and, obstructive sleep apnea. Bariatric surgery: a systematic review and meta-an… [JAMA. 2004] – PubMed result

The positive impact of bariatric surgery on survival of persons with morbid obesity has been documented in the Swedish Obesity Study Effects of bariatric surgery on mortality in Swedi… [N Engl J Med. 2007] – PubMed result and by other researchers Long-term mortality after gastric bypass surgery. [N Engl J Med. 2007] – PubMed result.             

LAGB is an elective procedure. Nearly 80% of patients pay out-of-pocket as health insurance reimbursement is limited. (See AHRQ HCUP Statistical Brief #23).  So which patients might select a hospital-based surgical procedure with some risks? There are several medical categories of patients who have failed at lifestyle modifications and are at elevated risks if their weight is not brought under control.

A. Cancer patients who are overweight and obese have increased risk of developing one of several cancers. Less recognized is that obesity is also associated with poorer outcomes in resected colon cancer patients and in prostrate cancer patients. Obese cancer patients are at increased risk for problems following cancer surgery, including would complication, lymphedema, second cancers as well as obesity-related comorbidities. Weight loss has been recommended for obese cancer survivors who are otherwise healthy. Obesity and cancer: the risks, science, and potent… [Oncology (Williston Park). 2005] – PubMed result 

B. Type 2 diabetics with high insulin resistance who are unresponsive to other treatment.

C. Coronary heart disease patients who are overweight or  obese patients may demonstrate increases in weight, not decreases.   Management of overweight and obese patients with c… [Eur J Cardiovasc Prev Rehabil. 2010] – PubMed result For such patients, more intensive interventions may be necessary to achieve an optimal weight. LAGB as well as LGB have been shown to be effective in reducing biochemical markers of coronary heart disease. One year improvements in cardiovascular risk facto… [Obes Surg. 2010] – PubMed result.  Livingston et al have suggested changing the NIH obesity surgery patient selection criteria if patients have risk factors for cardiovascular disease. Do current body mass index criteria for obesity su… [Surg Obes Relat Dis. 2007 Nov-Dec] – PubMed result

D. Smokers

It is well known that cessation of smoking has an effect of increasing body weight. This is often a deterrent to smokers to discontinue their habit.  The French Observational Cohort of Usual Smokers (… [BMC Public Health. 2010] – PubMed resultPsychosocial factors associated with weight contro… [J Natl Med Assoc. 2009] – PubMed result  Assuming a smoker might have also failed at attempts to change lifestyle factors such as diet and exercise, an alternative intervention of LAGB could encourage smoking cessation and achieve significant weight loss, thereby meeting two important public health goals.

F. Modern Medicine intervenes too late with too little

While medicine has improved in a great many areas, obesity has seen little in the way of the development of new therapeutic approaches, save for LABG. The frustration in related areas of medicine with contemporary approaches to type 2 diabetes and cardiovascular disease is apparent.

In a multi-center study of 5, 535 patients with coronary heart disease, De Bacquer and others observed,

Overweight and obese patients had more frequently raised blood pressure and elevated cholesterol after adjustment for age, gender, education, diabetes, and centre. In patients using blood pressure lowering agents, 56% of obese and 51% of overweight patients were still having raised blood pressure compared to 42% in normal weight patients. A similar result was observed for the therapeutic control of total cholesterol. In the period between coronary event and interview, body weight had increase with at least five kilograms in a quarter of all patients. Overweight and obesity in patients with established coronary heart disease: Are we meeting the challenge? — Eur Heart J

Dr. Thomas A. Buchanan of the Keck School of Medicine at the University of Southern California, after reviewing the limitations of current therapies to address the tidal wave of type 2 diabetes and accompanying cardiovascular disease, wrote:

So where does all this leave us regarding prevention and early treatment of type 2 diabetes? Basically, we need to get real about what we are up against. Losing a little weight or taking a relatively weak medication such as metformin will do little more than buy some time. Even the use of more potent medications, e.g. thiazolidinediones, will arrest disease progression in only a subset of individuals. We must apply some “big guns,” and they should be pointed at obesity. Those guns can be big in the public health sense: aiming at the development of population-based approaches to improve nutrition and reduce obesity. They must also be big in the individual sense – they must be truly effective in reducing obesity among individuals who become obese and begin to manifest B-cell failure. The implementation of lifestyle interventions in such individuals  – with the intensity used in the DPP and the Finnish Diabetes Prevention Study – is definitely a good idea, but we must keep in mind that most of these individuals will continue to progress toward diabetes even if they “lose a little weight.”

 

People who continue to progress in the face of whatever lifestyle changes they can make are logical candidates for medications that either promote weight loss or change fat biology. It seems prudent at present to use those medications very early in the course of diabetes, before the B-cell “horse” is completely out of the bar. But again, many people will continue to manifest rising glucose levels, reflecting failing B-cell function. This is where I believe we need to change our views about the use of bariatric surgery for hyperglycemia. Currently, the most effective and impressive surgeries involve both restrictive and malabsorptive components. They are very invasive and are generally reserved for people who are already at the end stages of obesity and B-cell failure. That approach seems appropriate to me. But I believe that less invasive restrictive surgeries, such as gastric banding, may have an important role earlier in the evolution of obesity and hyperglycemia.

 

The steepest dose-response curve between obesity and insulin resistance occurs across the range of BMI that encompasses overweight and moderate obesity. This is the range that is characteristic of many individuals with prediabetes and early type 2 diabetes. Although their glucose levels and risk of long-term diabetic complications are not yet high, their insulin resistance and B-cell function are already very bad. More important, the only real hope of preventing further deterioration is through is through potent weight loss. Gastric banding and other restrictive bariatric approaches may be well suited for this stage of disease.  These procedures lead to weight loss that is greater and longer-lasting than can be achieved with lifestyle interventions alone. The resulting unloading of B-cells is likely to be much more than we can achieve with lifestyle changes or available medications alone. Theoretically, such weight loss could provide long-term stabilization or even reversal of the B-cell disease that underlies prediabetes and type 2 diabetes.  

 

The epidemic of hyperglycemia that is currently sweeping across the globe is very clearly related to obesity. The most logical direct approach to stemming the epidemic is to strike hard at obesity. This should include public health measures to minimize the development of obesity in everyone. It should include more focused lifestyle interventions and targeted pharmacological approaches in individuals who manifest rising glycemia. But if we stop there, the available evidence is pretty clear: we will lost the battle against type 2 diabetes. We need very potent tools to reverse obesity and its metabolic effects if we are to win this battle. Modestly invasive bariatric approaches must be studies in this context to determine their impact on insulin resistance and the B-cell disease that leads to type 2 diabetes. Stemming the tide of type 2 diabetes: bring on the… [Obesity (Silver Spring). 2010] – PubMed result 

 

G. Conclusion

The current BMI threshold for patient selection for Lap-Band is not based on a sound scientific basis. The limitations of the BMI itself should limit its use as a “hard” measure of excess adiposity because several populations are not well identified using the BMI. This has led to a phenomenon of treating the BMI, not the patient. Reducing the BMI threshold will give patients and physicians more freedom to decide appropriate courses of action for long term weight reduction.

Sincerely,

Morgan Downey

Editor and Publisher