How did you become involved in researching reverse epidemiology, and what particular successes or failures did you encounter along the way?

Our ongoing quest to explore the causes of the enormous cardiovascular epidemic in dialysis patients, in order to find an effective solution for it, was indeed the main reason for our intense involvement in this line of research. Dialysis patients have an unacceptably high mortality rate, currently almost 20% per year. Over half of these deaths are related to cardiovascular disease. Moreover, some cardiovascular risks factors, such as high blood pressure or high serum homocysteine, are quite common among dialysis patients.

Hence, for years, many researchers blindly followed the Framingham paradigms, in which they took it for granted that the etiology of cardiovascular disease and death in dialysis patients is related to the high prevalence of such cardiovascular risk factors as hypertension or hyperhomocysteinemia. However, some data indicated that malnutrition and/or inflammation were much more strongly associated with death risk in dialysis. Indeed, hypoalbuminemia has been identified as the strongest risk factor for death in these individuals⁷.

My first major funding came from a five-year award granted by the National Institutes of Health (K23 DK61162), entitled “Nutrition and Inflammation Evaluation in Dialysis Patients” (NIED) Study. The NIED Study has been performed in DaVita dialysis facilities in Los Angeles area since mid 2001. We found that anorexia (diminished appetite) was strongly associated with higher death rate ⁸, that gain in total body fat over time conferred survival advantages ⁹, and that a low, rather than a high, homocysteine level was associated with higher death risk in dialysis patients¹⁰.

The National Kidney Foundation, too, has supported my epidemiologic research projects to explore the true nature of the reverse epidemiology phenomenon by conducting time-varying survival analyses of a large national database of 60,000 dialysis patients in the USA from the second-largest dialysis care provider, i.e., DaVita.

Because many respected opinion leaders have opposed our theory of reverse epidemiology and denounced it as spurious or irrelevant, we have encountered many challenges including publication bias at many peer-reviewed journals and major hardship in obtaining more grant funds for the continuation of our research projects in the field of reverse epidemiology¹¹.

However, we continue to strive to form collaborative and multi-disciplinary research groups with investigators from other fields who are pursuing the same ideas—such as cardiologists, geriatricians, oncologists, nutritionists, and epidemiologists, who are engaged in cachexia and inflammation research, survival in heart failure, etc.—hoping that the scientific community can be more tolerant and more receptive of this new concept.

As an example, a recent study known as the “4D Trial” showed that cholesterol-lowering treatment did not improve survival in diabetic dialysis patients¹². We had previously shown that most dialysis patients have low, rather than high, serum lipid levels¹³ and had predicted such negative results based on the theory of reverse epidemiology. Indeed, some cardiologist colleagues had also advanced the “lipoprotein-endotoxin hypothesis”¹⁴, according to which circulating serum lipoproteins (cholesterols) are an effective defense mechanism against the harmful endotoxins in heart failure patients.

These and similar hypotheses are usually denounced by traditional investigators as “counterintuitive,” “controversial,” “flawed,” etc. Nevertheless, the advancement of science is indeed based upon the constructive confrontation of opposing ideas. Hence, no matter how challenging this path is, we welcome all ideas, especially those that question the validity of the reverse epidemiology hypotheses. This, we believe, is an effective way for the concept of reverse epidemiology to unfold and further evolve, while we recognize, and hope that we are prepared for, the ongoing hardship that we will have to encounter in this long way to the truth.

  I see that the work in your 2005 Hypertension paper was supported by a “young investigator award from the National Kidney Foundation.” Would you tell us a little about this award?

The two-year Young Investigator Award was granted to me for a research proposal that I submitted in late 2003 to the National Kidney Foundation to explore the epidemiologic nature of the reverse epidemiology phenomena in over 50,000 dialysis patients across the nation. These individuals have been undergoing maintenance dialysis treatment in DaVita.

In general, dialysis patients have an unacceptably high mortality, currently approximately 20% per year in the USA . By conducting this research project, we hope that we better understand the link between the risk factor paradoxes and the high mortality in the fast growing dialysis patient population as well as in other similar populations.

  Would you give us some background on your 2003 Kidney International paper and your 2005 Blood Purification paper?

In our 2003 Kidney International paper¹ , we advanced the concept of reverse epidemiology for dialysis patients for the first time as a conclusive hypothesis to explain the observed paradoxes such as obesity, hypertension, hypercholesterolemia, and hyperhomocysteinemia as they relate to better survival in this group of patients.

In our 2005 Blood Purification paper² we explained that the concept of reverse epidemiology is not restricted to dialysis patients but goes beyond this population of half a million and includes several even larger populations such as those with chronic heart failure, cancer, AIDS, etc.

  Why do you think your 2005 Blood Purification paper is highly cited?

We advanced the theory of "reverse epidemiology" for the first time in our articles in the journal Kidney International in March 2003 for dialysis patients¹ and in the Journal of American College of Cardiology in April 2004 for chronic heart failure patients³. According to the reverse epidemiology hypothesis, conventional risk factors of cardiovascular disease and death in the general population such as serum cholesterol, blood pressure, and body weight relate to outcome in maintenance dialysis patients and heart failure patients in an opposite direction.

Obesity, hypercholesterolemia, and hypertension are paradoxically protective features that are associated with a greater survival among dialysis or heart failure patients. These paradoxical findings are in sharp contrast to the well-known associations between surrogates of over-nutrition and poor outcome in the general population. Because the concept of reverse epidemiology is against the contemporary cardiovascular principles, it may be considered counterintuitive and has provoked a great deal of debate and rebuttals.

There are both enthusiastic proponents and opponents of the reverse epidemiology. According to some opinion leaders, the reverse epidemiology findings are spurious and not consistent with the truth. In one of our position papers in Blood Purification in January 2005², we explained that these paradoxical findings appear robust and highly consistent in both maintenance hemodialysis patients and heart failure patients; we referred to our ongoing studies in 40,000 to 60,000 hemodialysis patients across the USA, in whom we have shown that hypertension (systolic blood pressure as high as 180 mmHg)⁴ or morbid obesity (body mass index >35 kg/m2)⁴ are associated with better survival in dialysis patients, whereas the so-called “normal range” blood pressure or body mass index groups have significantly higher death risks.

We have estimated that there are at least 20 to 30 million individuals in the USA, including those with chronic disease states or advanced age, in whom the reverse epidemiology hypothesis holds and in whom lowering weight, blood pressure or serum cholesterol might indeed be deleterious².

  Does this paper describe a new discovery or a new methodology that's useful to others?

The concept of reverse epidemiology leads to emerging paradigms that put the very foundation and generalizability of the modern cardiovascular epidemiology into question. Until the advancement of the reverse epidemiology hypothesis in 2003, the robustness of Framingham-based paradigms have not been seriously questioned. There were some previous data indicating paradoxical associations between cardiovascular risks factors and survival in dialysis patients.

However, even the investigators who had reported these data did not feel comfortable to offer stand-alone and inclusive hypotheses and insisted that such observations were spurious and likely due to confounding factors that were clinically irrelevant. We were the first group to announce systematically the possibility of true associations between such cardiovascular risk factors as obesity or hypertension and better survival chance among dialysis patients¹.

We have advanced the hypothesis that “malnutrition-inflammation-cachexia syndrome” (MICS)⁶ is the most likely etiology of the reverse epidemiology and that the time discrepancy between the two sets of competitive risk factors, i.e., over-nutrition (which needs years to decades to exert its deleterious cardiovascular effect) and under-nutrition (which can be fatal within a much shorter period of time), can lead to the reverse epidemiology phenomenon as a true entity in dialysis and heart failure patients^1-3.

The emergence of our hypotheses has overcome (or at least mitigated) the publication bias against paradoxical data and led to the exponential emergence of reverse epidemiology data in the past two to three years from not only dialysis patients, but also from data of individuals with heart failure, cancer, AIDS, rheumatoid arthritis, and chronic pulmonary disease, as well as elderly individuals².

  As you’ve said in your papers, reverse epidemiology is a counterintuitive phenomenon. Since you have identified it in such a large patient population, how can physicians today know which of their at-risk patients to treat for hypertension, obesity, and high cholesterol? Are new standards for these individuals in the works, or are you still working to convince the medical community at large of the validity of reverse epidemiology?

Reverse epidemiology research is still in its infancy, and I believe we still have a long way to go here. At this early stage, it is not clear how clinically and scientifically valid the hypotheses related to the reverse epidemiology are. Many opinion leaders have denounced the concept of reverse epidemiology as “statistical fallacy” that must be “explained away” using proper statistical models.

For many scientists and clinicians it is hard to believe that obesity or high cholesterol levels may confer survival advantages in distinct groups of people. The clinical and public health mindset of the late 20^th and early 21^st century has been deeply rooted in the Framingham paradigms, according to which elements of over-nutrition such as obesity or hypercholesterolemia are the foundations of the cardiovascular disease and death in our society and are intuitively related to inferior survival. Hence, stating that the Framingham principles may have significant exceptions is widely considered counterintuitive, provocative, and unwelcome at this point in time.

However, if the theory of reverse epidemiology shows clinical and public health plausibility, then the recommended norms in the healthy general population and the Framingham-based guidelines may be revised for millions of people with chronic heart failure, chronic kidney disease, or other chronic disease states or the elderly (octogenarians and nonagenarians). According to our conservative estimates, these populations comprise currently 20 to 30 million Americans and are fast growing as the life expectancy has continued to increase in most nations of our planet. Hence, reverse epidemiology will become even more relevant as the populations with advanced age or chronic disease states continue to comprise larger proportions of the societies in the 21^st century.

  How, precisely, does the malnutrition-inflammation complex syndrome cause reverse epidemiology? What other mechanisms might be involved?

We hypothesize that in populations with chronic disease states such as chronic kidney disease or heart failure, the malnutrition-inflammation complex (or cachexia) syndrome overwhelms the influence of the traditional associations on outcomes, so that a “bad-gone-good” phenomenon is observed. In highly industrialized, affluent nations, malnutrition is an uncommon cause of poor outcome, where instead over-nutrition is associated with a greater risk of cardiovascular disease and shortened survival. In contrast, in dialysis patients under-nutrition (due to the malnutrition-inflammation complex and resultant anorexia) is a common risk factor for adverse cardiovascular events and death.

Hence, certain markers, such as decreased body mass index and lower serum cholesterol, which predict a low likelihood of cardiovascular events and an improved survival in the general population, become strong risk factors for increased morbidity and death in hemodialysis patients. Moreover, some indicators of over-nutrition such as obesity or even morbid obesity actually predict improved outcome in hemodialysis patients.

  Could you summarize the significance of your work in layman's terms?

In the healthy general population, there are known risk factors for heart disease such as obesity, high serum cholesterol, and high blood pressure. However, in individuals that are very old or in those people who have chronic diseases such as kidney failure or heart failure, these known risk factors for heart disease may have the opposite effect.

As an example, obese individuals who have chronic kidney disease and need hemodialysis treatment to survive may live longer than those dialysis patients who are not obese. The same goes for high serum cholesterol, in that dialysis patients or heart failure patients with high total serum cholesterol may live longer than those who have a normal-to-low serum cholesterol.

This phenomenon is called reverse epidemiology, because the traditional epidemiology of heart disease (in that high cholesterol is bad for the heart and vessels) appears to be in the opposite direction in these groups of individuals.

  Are there any social or political implications to your research?

A major socio-political obstacle for the concept of reverse epidemiology is that it does not appear politically correct or time-appropriate for our current environment of the early 21^st century. Currently, over-nutrition appears to be a more serious challenge for our society than under-nutrition. Indeed, the advancement of the reverse epidemiology research may be conceived as a major handicap for the current efforts to fight against the obesity epidemic in our society.

While we wholeheartedly advocate the anti-obesity campaign among children, teenagers and most young adults, we do not believe that obesity should categorically be demonized, nor do we believe that a parsimonious black-and-white phenomenon exists to justify any campaign for or against obesity in ALL members of our society.

The concept of reverse epidemiology advances the notion that there are distinct populations in whom obesity is not only harmless but indeed advantageous and may lead to greater survival. There may be at least 20 to 30 million individuals in the USA in whom obesity should not be targeted blindly as a deleterious characteristic if the theory of reverse epidemiology should have clinical validity.

Certain branches of the pharmaceutical industry may become defensive when the generalizability of targeting hypercholesterolemia, hypertension, or hyperhomocysteinemia is questioned by the concept of reverse epidemiology. However, a genuine and valid concept in science will eventually lead to the emergence of new paradigms and will reshape the future, if and only if the concept is valid. In no way do we insist on the validity of our hypotheses. We just mean to seek the truth, hopefully without any bias towards or against any direction. The inevitable and ongoing scientific evolution will find its way towards the truth no matter what.

  Where do you plan to take your work in reverse epidemiology next? Where do you see this work in 5, 10 years?

At this point in time it is difficult to predict exactly where we are going. The most important question that has not yet been answered is the issue of causality. In other words, should we recommend that dialysis patients or heart failure patients gain weight in order to live longer? Well-designed randomized clinical trials would be the ultimate tools to answer this important question. I do hope that at some point we have the funds and resources to conduct more studies including randomized clinical trials.

However, due to the provocative and counterintuitive nature of the reverse epidemiology hypotheses, it may be somewhat challenging to secure adequate funds and support to conduct such clinical trials at this juncture. Nevertheless, I am optimistic that in 5 to 10 years from now the scientific community will be inclined to view our hypotheses with more constructive skepticism, as it has already been happening among cardiologists and nutritionists.

With this in mind, I can see more observational studies, which will both endorse and disprove the concept of reverse epidemiology, and more clinical trials to that end. Eventually the Framingham norms will be revised for distinct populations such as elderly individuals or those with chronic disease states. I think in the near future we will see that the definition of obesity will become dynamic and population specific, so that we will not see one single mechanistic definition for “the norm” as it is now.

Similar population-specific modifications will happen in other cardiovascular risk factors such as serum cholesterol level or blood pressure value as a consequence of ongoing studies. Defining the “normal weight” as a body mass index between 18.5 and 25 kg/m2 may then become an inappropriate recommendation for dialysis patients in the USA .

Sometimes I wonder if, as a practicing physician, I am doing the right thing to recommend to an 86-year-old lady with stable clinical conditions and a body mass index above 30 kg/m2 to lose weight only because she is labeled as “obese” according to the current World Health Organization (WHO) guidelines. I do not have any answer for this question, at least at this point in time.

Kamyar Kalantar-Zadeh, MD, PhD, MPH
FAAP, FACP, FASN, FAHA
Associate Professor of Medicine & Pediatrics
Director, Dialysis Expansion & Epidemiology
Harbor-UCLA Division of Nephrology & Hypertension
Los Angeles Biomedical Research Institute, and
UCLA David Geffen School of Medicine, Los Angeles, CA

References:

1.       Kalantar-Zadeh K, Block G, Humphreys MH, Kopple JD: Reverse epidemiology of cardiovascular risk factors in maintenance dialysis patients. Kidney Int 63:793-808, 2003

2.       Kalantar-Zadeh K, Kilpatrick RD, Kuwae N, Wu DY: Reverse epidemiology: a spurious hypothesis or a hardcore reality? Blood Purif 23:57-63, 2005

3.       Kalantar-Zadeh K, Block G, Horwich T, Fonarow GC: Reverse epidemiology of conventional cardiovascular risk factors in patients with chronic heart failure. J Am Coll Cardiol 43:1439-1444, 2004

4.       Kalantar-Zadeh K, Kilpatrick RD, McAllister CJ, Greenland S, Kopple JD: Reverse epidemiology of hypertension and cardiovascular death in the hemodialysis population: the 58th annual fall conference and scientific sessions. Hypertension 45:811-817, 2005

5.       Kalantar-Zadeh K, Kopple JD, Kilpatrick RD, McAllister CJ, Shinaberger CS, Gjertson DW, Greenland S: Association of morbid obesity and weight change over time with cardiovascular survival in hemodialysis population. Am J Kidney Dis 46:489-500, 2005

6.       Kalantar-Zadeh K, Ikizler TA, Block G, Avram MM, Kopple JD: Malnutrition-inflammation complex syndrome in dialysis patients: Causes and consequences. Am J Kidney Dis 42:864-881, 2003

7.       Kalantar-Zadeh K, Kilpatrick RD, Kuwae N, McAllister CJ, Alcorn Jr H, Kopple JD, Greenland S: Revisiting mortality predictability of serum albumin in the dialysis population: time dependency, longitudinal changes and population-attributable fraction. Nephrol Dial Transplant 20:1880-1889, 2005

8.       Kalantar-Zadeh K, Block G, McAllister CJ, Humphreys MH, Kopple JD: Appetite and inflammation, nutrition, anemia and clinical outcome in hemodialysis patients. Am J Clin Nutr 80:299-307, 2004

9.       Kalantar-Zadeh K, Kuwae N, Wu DY, Shantouf RS, Fouque D, Anker SD, Block G, Kopple JD: Associations of body fat and its changes over time with quality of life and prospective mortality in hemodialysis patients. Am J Clin Nutrition 83:202-210, 2006

10.     Kalantar-Zadeh K, Block G, Humphreys MH, McAllister CJ, Kopple JD: A low, rather than a high, total plasma homocysteine is an indicator of poor outcome in hemodialysis patients. J Am Soc Nephrol 15:442-453, 2004

11.     Chumlea WM, Sun SS: The availability of body composition reference data for the elderly. J Nutr Health Aging 8:76-82, 2004

12.     Wanner C, Krane V, Marz W, Olschewski M, Mann JF, Ruf G, Ritz E: Atorvastatin in patients with type 2 diabetes mellitus undergoing hemodialysis. N Engl J Med 353:238-248, 2005

13.     Kalantar-Zadeh K, Kilpatrick RD, Kopple JD, Stringer WW: A matched comparison of serum lipids between hemodialysis patients and nondialysis morbid controls. Hemodial Int 9:314-324, 2005

14.     Rauchhaus M, Coats AJ, Anker SD: The endotoxin-lipoprotein hypothesis. Lancet 356:930-933, 2000