This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Purchase Article View Shopping Cart Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Zavaroni, I. Articles by Reaven, G. M. Search for Related Content PubMed PubMed Citation Articles by Zavaroni, I. Articles by Reaven, G. M.

Can Weight Gain in Healthy, Nonobese Volunteers Be Predicted by Differences in Baseline Plasma Insulin Concentration?¹

Istituto Clinica Medica Generale e Terapia Medica (I.Z., A.Z., P.G., P.M., A.B.), Parma University; Stanford University School of Medicine (G.M.R.), Stanford, California 94305; and Shaman Pharmaceuticals, Inc., South San Francisco, California 94080

Address correspondence and requests for reprints to: Ivana Zavaroni, Istituto Clinica Medica Generale e Terapia Medica, via Gramsci 14, 43100 Parma, Italy.

	Abstract

Top Abstract Introduction Subjects and Methods Results Discussion References

 
In this study, we have evaluated the effect, over approximately 14 yr, of differences in baseline degree of hyperinsulinemia on weight gain in 647 healthy, nonobese factory workers. The subjects were divided into 4 quartiles, on the basis of their plasma insulin response to an oral glucose challenge, in 1981. At that time, the mean (±SD) plasma insulin concentration, 2 h after the glucose challenge, varied from 18 ± 5 to 106 ± 42 µU/mL. Despite this approximate 6-fold difference in plasma insulin response at baseline, the weight gain over the period of observation was similar in all quartiles, with mean (±SD) increments (kg) of 1.8 ± 5.1, 1.6 ± 5.3, 2.3 ± 5.2, and 2.3 ± 5.7, going from the lowest quartile to the highest quartile, in terms of insulin concentration. Furthermore, when the population was considered as a whole, there was no correlation between baseline degree of hyperinsulinemia and change in either absolute (r = 0.004) or percent (r = 0.003) weight gain. Finally, there was no difference in the number of individuals who gained more than 4.5 kg, as a function of their baseline insulin response. Consequently, we conclude that 6-fold differences in plasma insulin responses to glucose do not predict weight gain in a healthy, nonobese population.

	Introduction

Top Abstract Introduction Subjects and Methods Results Discussion References

 
DURING THE past few years, conflicting results have been published, as to the impact of differences in insulin-mediated glucose disposal and/or compensatory hyperinsulinemia on changes in body weight over time (1, 2, 3, 4, 5, 6). More specifically, evidence has been published, from prospective studies in several population groups (1, 2, 3, 4), that insulin resistance and/or compensatory hyperinsulinemia, at baseline, predict a decrease in weight gain. However, results in two other studies (5, 6) have come to exactly the opposite conclusion. The situation becomes even more confounded when the results of prospective studies in adult Pima Indians (1, 3) could not be replicated in Pima Indian children (6), in whom weight gain over time was actually greater in those deemed to have fasting hyperinsulinemia. Obviously, the apparent paradox between the effects of insulin resistance and compensatory hyperinsulinemia on weight gain in Pima Indian adults and children needs clarification. However, at a simpler level, the relationship between insulin metabolism and weight gain has often been studied in populations with an increased prevalence of obesity. Consequently, it seemed important to define the relationship between insulin resistance and weight gain over time in a population not known to have a great increase in prevalence of obesity. The data to be presented in this report represent our effort to address this question.

	Subjects and Methods

Top Abstract Introduction Subjects and Methods Results Discussion References

 
In 1981, we surveyed 732 factory workers for a variety of metabolic and hemodynamic risk factors for coronary heart disease (7). A complete medical history was taken, physical examination performed, and the height and weight of each patient determined after removal of shoes and in light clothes. The measurements consisted of height and weight. Body mass index (BMI) was calculated by dividing weight (kg) by height square (m²). In addition, plasma glucose (8) and insulin concentrations (9) were measured before and 1 h and 2 h after a 75-g oral glucose load.

Starting in 1993 and ending in 1996, all subjects were reevaluated, and their height and weight were assessed, as in 1981. This report is based on evaluation of 647 of the original 732 factory workers who were free of any disease at baseline.

For analytical purposes, subjects were divided into quartiles, on the basis of their plasma insulin concentrations 2 h after the glucose load as determined in 1981: quartile I represents the group with the lowest insulin response, and quartile IV those with the highest responses.

Results are expressed as mean ± SD. Differences in mean values among the four quartiles were assessed by one-way ANOVA, for all variables except weight gain, which was evaluated by covariance analysis (adjusting for differences in age, gender, and BMI).

	Results

Top Abstract Introduction Subjects and Methods Results Discussion References

 
The changes in weight of the 647 healthy individuals studied in 1981 and approximately 14 yr later are shown in Table 1. The volunteers have been subdivided on the basis of their insulin concentration 2 h after an oral glucose challenge, as determined in 1981. It can be seen that the quartiles were similar, in terms of age and gender distribution. These results indicate that quartile IV subjects, the 25% of the population with the highest insulin concentrations 2 h after the oral glucose challenge, had insulin approximately 6-fold greater than in the 25% with the lowest insulin levels (quartile I). Despite this dramatic difference in baseline plasma insulin concentration, there was only a difference of 2 kg in baseline body weight between individuals in quartile IV and those in quartile I. The results in Table 1 also demonstrate that weight gain of the group with the lowest insulin concentration (quartile I), over the (approximately) 14 yr of observation, was only 0.5 kg less than weight gain in the most insulin-resistant subjects (quartile IV). Furthermore, none of the small differences in weight gain among the four quartiles was significant. These data emphasize the relative stability of the weight in this population of Italian factory workers, observed over the follow-up period. Specifically, mean net weight in the four groups varied from 1.8–2.3 kg, with a percent change in weight, over the 14-yr period, from 2.8–3.4%.

Another way to address the issue at hand was to define the relationship between the plasma insulin concentration 2 h after the glucose load, and weight gain over the 14-yr period of observation, in the population as a whole. When this was done, it was apparent that there was no correlation between plasma insulin concentration and either absolute (r = 0.004) or percent (r = 0.003) weight gain from 1981 to 1993. As before, this finding was independent of the insulin measurement used to establish the quartiles.

Finally, we ascertained the number of individuals in each quartile who gained more than 4.5 kg during the period of observation. Again, we could not discern any effect of the baseline insulin response to glucose affecting this measurement (with 35, 36, 40, and 42 individuals from quartiles I–IV) gaining more than 4.5 kg over the 14-yr period of observation.

	Discussion

Top Abstract Introduction Subjects and Methods Results Discussion References

 
The results presented show that weight change in healthy volunteers, over a period of approximately 14 yr, was independent of profound differences in baseline plasma insulin concentrations and, presumably, degrees of insulin resistance. These findings are somewhat different from previously published studies, in adults, showing that hyperinsulinemia was associated with either less or more weight gain. Unfortunately, for many reasons, it is not possible to reconcile these disparate results. For example, the index of hyperinsulinemia varied from study to study. Thus, the finding by Sigal et al. (5), that hyperinsulinemia predicts more weight gain, was based on the acute insulin response to iv glucose; fasting plasma insulin concentration did not predict weight gain in the same study. However, a difference in assessment of hyperinsulinemia cannot explain why Schwartz et al. (3) found that hyperinsulinemia predicted less weight gain, irrespective of whether hyperinsulinemia was estimated by determining the plasma insulin response to meals, to an oral glucose challenge, or to an acute iv glucose bolus.

Schwartz et al. (3) and Sigal et al. (5) also came to equally disparate views, as to the power of insulin resistance to predict weight gain. Schwartz et al. (3) reported that the more insulin sensitive an individual, the more likely to gain weight. In contrast, Sigal et al. (5) found no effect of insulin sensitivity on weight gain, but qualified this observation by pointing out that the greatest risk of gaining weight was in those who had both a high acute insulin response to glucose and high insulin sensitivity. The combination of increased insulin sensitivity and hyperinsulinemia seems an unlikely combination, given extensive evidence of a direct relationship between hyperinsulinemia and insulin resistance (not insulin sensitivity), assessed as fasting insulin concentration, as insulin response to an oral glucose challenge, or as the acute insulin response to iv glucose (10, 11, 12). The most likely explanation for the disparity between the results of Sigal et al. (5) and other studies in adults is that they only examined offspring of two parents with type 2 diabetes; a population that may not be characteristic of the world at large.

In addition, because Sigal et al. (5) studied a population primarily of European origin, whereas Schwartz et al. (2) evaluated Pima Indians, there is the possibility that differences in racial background might account for the polar opposite results of the two studies. However, this seems unlikely, given our results that differences in plasma insulin concentrations have nothing do to with subsequent weight gain in a population similar in ethnicity to the one observed by Sigal et al. (5). Furthermore, both Valdez et al. (2) and Hoag et al. (4) indicated that baseline hyperinsulinemia predicted less weight gain in both non-Hispanic whites and Hispanics (2, 4), with no apparent effect from differences in ethnicity.

Finally, the disparity in results of studies performed in adults becomes even more perplexing when focus is shifted to the one study in children. In this instance, evidence was presented (6) that fasting hyperinsulinemia, and presumably insulin resistance, predicted greater weight gain in Pima children, whereas previous results in Pima adults (1, 3) demonstrated that the rate of weight gain was diminished by either hyperinsulinemia or insulin resistance. One explanation for these discordant results in adults and children from the same population, as suggested by Odeleye (6), is that a common factor could promote both hyperinsulinemia and obesity and that the established hyperinsulinemia/insulin resistance can then protect against weight gain.

Although an unequivocal answer to the question: can weight gain in healthy volunteers be predicted by differences in plasma insulin concentration? is not possible, focusing only on data from the adult observations permits us to arrive at some generalizations. The most divergent results among the five studies are those of Sigal et al. (5). In addition to being the most discrepant in terms of the link between insulin resistance, hyperinsulinemia, and weight gain, the results of Sigal et al. were based upon study of relatively few individuals (107), all of whom had two parents with type 2 diabetes. If we now turn to the remaining four large population-based studies, the results are more similar than they seem initially. Thus, the conclusion by Valdez et al. (2) that hyperinsulinemia was a predictor of weight gain was only true of the tertile of their population that was most obese; in two-thirds of their population, their results paralleled ours; differences in plasma insulin concentration did not predict subsequent weight gain. On the basis of these results, they concluded that hyperinsulinemia, and presumable insulin resistance, prevents additional weight gain in individuals who are already obese. The Pima Indians, studied by Swinburn et al. (1) and Schwartz et al. (3) were quite heavy, with mean values for BMI of approximately 35 kg/m²; and the link between insulin resistance, hyperinsulinemia, and decreased weight gain observed in these two studies resembles that of the heaviest subjects observed by Valdez et al. (2). The propensity to obesity, as well as the short period of observation, may help explain the disparity between our results and those of Hoag et al. (2). Thus, Hoag et al. (4) only followed their population for 4.3 yr; and during this period, subjects gained approximately 1 kg/yr, and 17% of the subjects gained more than 5 kg over this short time period. In contrast, we followed patients for approximately 14 yr, and the mean weight gain per year was approximately one-sixth of the yearly weight gain observed by Hoag et al. (3). On the other hand, our results were similar when we divided individually into two groups, obese and nonobese, using a BMI of 27 kg/m² as the cutoff point. Thus, although we cannot rule out the possibility that insulin resistance and/or compensatory hyperinsulinemia may act to limit weight gain over time in individuals who are obese, wide variations in insulin resistance and plasma insulin concentration had little, if any, effect on predicting weight gain in healthy, nonobese volunteers.

	Footnotes

 
¹ This work was supported by research grants from MURST (University Ministry of Scientific Research and Technology) and the National Institutes of Health (HL-08506 and RR-00070).

Received March 9, 1998.

Revised June 25, 1998.

Accepted July 6, 1998.

	References

Top Abstract Introduction Subjects and Methods Results Discussion References

 

1. Swinburn BA, Nyomba BL, Saad MF, et al. 1991 Insulin resistance associated with lower rates of weight gain in Pima Indians. J Clin Invest. 88:168–173.
2. Valdez R, Mitchell BD, Haffner SM, et al. 1994 Predictors of weight change in a bi-ethnic population. The San Antonio Heart Study. Int J Obes. 18:85–91.
3. Schwartz MW, Boyko EJ, Kahn SE, Ravussin E, Bogardus C. 1995 Reduced insulin secretion: an independent predictor of body weight gain. J Clin Endocrinol Metab. 80:1571–1576.[Abstract/Free Full Text]
4. Hoag S, Marshall JA, Jones RH, Hamman RF. 1995 High fasting insulin levels associated with lower rates of weight gain in persons with normal glucose tolerance: The San Luis Valley Diabetes Study. Int J Obes. 19:175–180.[Medline]
5. Sigal RJ, El-Hashimy M, Martin BC, Soeldner JS, Krolewski AS, Warram JH. 1997 Acute postchallenge hyperinsulinemia predicts weight gain. Diabetes. 46:1025–1029.[Abstract]
6. Odeleye OE, de Courten M, Pettitt DJ, Ravussin E. 1997 Fasting hyperinsulinemia is a predictor of increased body weight gain and obesity in Pima Indian children. Diabetes. 46:1341–1345.[Abstract]
7. Zavaroni I, Bonora E, Pagliara M, et al. 1989 Risk factors for coronary artery disease in healthy persons with hyperinsulinemia and normal glucose tolerance. N Engl J Med. 320:702–706.[Abstract]
8. Kadish AH, Litle RL, Sterberg JC. 1968 A new and rapid method for determination of glucose by measurement of rate of oxygen consumption. Clin Chem. 14:116–131.[Abstract]
9. Hales CN, Randle PJ. 1968 Immunoassay of insulin with insulin antibody precipitate. Biochem J. 88:137–146.
10. Olefsky J, Farquhar JW, Reaven GM. 1973 Relationship between fasting plasma insulin level and resistance to insulin-mediated glucose uptake in normal and diabetic subjects. Diabetes. 22:507–513.[Medline]
11. Hollenbeck CB, Chen Y-DI, Reaven GM. 1984 A comparison of the relative effects of obesity and non-insulin-dependent diabetes mellitus on in vivo insulin-stimulated glucose utilization. Diabetes. 33:622–626.[Abstract]
12. Kahn SE, Prigeon RL, Mcculloch DK, et al. 1993 Quantification of the relationship between insulin sensitivity, and ß-cell function in human subjects. Diabetes. 42:1663–1672.[Abstract]

This article has been cited by other articles:

				Z. T. Bloomgarden Gut Hormones, Obesity, Polycystic Ovarian Syndrome, Malignancy, and Lipodystrophy Syndromes Diabetes Care, July 1, 2007; 30(7): 1934 - 1939. [Full Text] [PDF]

				Z. T. Bloomgarden Third Annual World Congress on the Insulin Resistance Syndrome: Mediators, antecedents, and measurement Diabetes Care, July 1, 2006; 29(7): 1700 - 1706. [Full Text] [PDF]

				A. G. Pittas, S. K. Das, C. L. Hajduk, J. Golden, E. Saltzman, P. C. Stark, A. S. Greenberg, and S. B. Roberts A Low-Glycemic Load Diet Facilitates Greater Weight Loss in Overweight Adults With High Insulin Secretion but Not in Overweight Adults With Low Insulin Secretion in the CALERIE Trial Diabetes Care, December 1, 2005; 28(12): 2939 - 2941. [Full Text] [PDF]

				E. J. Mayer-Davis, G. J. Kirkner, A. J. Karter, and D. J. Zaccaro Metabolic Predictors of 5-Year Change in Weight and Waist Circumference in a Triethnic Population: The Insulin Resistance Atherosclerosis Study Am. J. Epidemiol., April 1, 2003; 157(7): 592 - 601. [Abstract] [Full Text] [PDF]

				Z. T. Bloomgarden American Association of Clinical Endocrinologists (AACE) Consensus Conference on the Insulin Resistance Syndrome: 25-26 August 2002, Washington, DC Diabetes Care, March 1, 2003; 26(3): 933 - 939. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Purchase Article View Shopping Cart Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Zavaroni, I. Articles by Reaven, G. M. Search for Related Content PubMed PubMed Citation Articles by Zavaroni, I. Articles by Reaven, G. M.