¹Obesity Unit, 3rd Department of Internal Medicine, The First Medical Faculty, Charles University, 128 08 Prague, Czech Republic

²Department of Psychiatry, University of Pennsylvania, Philadelphia, PA 19104, USA

³Obesity Research Center, St Luke's/Roosevelt Hospital, New York, NY 10025, USA

Correspondence to: V Hainer, Obesity Management Centre, General Faculty Hospital, 3rd Department of Medicine, The First Faculty of Medicine, Charles University, Ke Karlovu 2, 128 08 Prague 2, Czech Republic. vhai@lf1.cuni.cz

OBJECTIVE: To assess intrapair resemblance in changes of body weight, total body fat, fat distribution, resting metabolic rate, fasting respiratory quotient and cardiovascular disease risk factors in response to therapeutic weight loss in female obese identical twins.

DESIGN: Patients stayed for 40 days on an inpatient metabolic unit under careful supervision. The stay was divided into three parts: an initial period of 7 days for adjustment to the hospital environment and for baseline measurements, 28 days of the weight reduction regimen when negative energy balance was achieved mainly by a very low calorie diet (1.6 MJ per day) and 5 days of testing after weight reduction.

SUBJECTS: Fourteen pairs of premenopausal female obese identical twins (age: 39.0±1.7 y; body weight (BW): 93.9±21.2 kg; body mass index (BMI): 34.2±7.8 kg/m²) participated in the study.

MEASUREMENTS:. Before and after weight loss, the following measurements were made: body composition by anthropometry and hydrodensitometry, intra-abdominal fat by ultrasonography, resting metabolic rate by indirect calorimetry. Total cholesterol, high-density lipoprotein-cholesterol, triglycerides and uric acid were determined by standard laboratory procedures. Blood pressure was measured in the morning in the recumbent position.

RESULTS: Subjects lost 8.8±1.9 kg of weight, from 93.9±21.2 to 85.1±10.9 kg (P<0.0001) and 6.5±2.3 kg of body fat (P<0.001). Weight losses varied widely among subjects, with a high correlation between losses of members of twin pairs for body weight (r=0.85; P<0.001) and for body fat (r=0.88; P<0.0001). Changes in uric acid resulting from weight loss were also correlated among members of twin pairs whereas changes in blood pressure, cholesterol and triglycerides were not.

CONCLUSION: The great intrapair resemblance observed in very low calorie diet-induced weight and fat losses in female obese identical twins suggests an important role of genetic factors in response to the weight reduction regimen.

International Journal of Obesity (2000) 24, 1051-1057

obesity; twins; weight reduction; fat loss; resting metabolic rate; cardiovascular risk factors; genetic factors

Introduction

New discoveries are revolutionizing our understanding of obesity. The identification of genes coding for obesity in mice¹ has led to a search for human homologues.^2,3,4 Single gene mutations, however, may account for only a small number of cases of human obesity.^2,3,4 Quantitative genetic studies suggest that the heritability of human obesity is due to many distinct genetic loci. Accordingly, new research strategies are seeking to move upstream from the study of obesity as a phenotype to the study of the determinants of the disorder, so-called 'intermediate phenotypes'. These strategies involve assessment of genetic contributions to energy metabolism¾energy intake^5,6 and energy expenditure.^7,8,9 These studies have, however, assessed obese persons only when their obesity was in a steady state. More promising may be experimental approaches to the study of energy regulation, as in the two landmark reports by Bouchard and his group. The first study utilized overfeeding to cause weight gain in male identical twins of normal weight;¹⁰ the second utilized exercise to cause weight loss in male identical twins with modestly elevated weight.¹¹ These studies^10,11 as well as postoverfeeding results¹² suggested a genetic contribution to the regulation of body weight, body fat and their determinants. Thus, there was a great similarity in changes in these variables within twin pairs despite large differences among twin pairs. There has been, until now, no study on the role of genetic similarity vs genetic difference in the regulation of the response of these variables to the weight loss protocol in obese persons.

The present study assessed the intrapair similarity in regulation of body weight, body fat and their determinants in 14 pairs of female obese identical twins undergoing therapeutic dieting for a period of 28 days.

Subjects

Subjects consisted of 14 pairs of female obese identical twins who responded to a story about the proposed study which appeared in a popular tabloid newspaper. They were 39.0+1.7 y of age (range 24-48 y), with a BMI of 34.2±7.8 (mean±s.d.) and a weight of 93.9±21.2 kg. Zygosity was determined by history, physical appearance and identity of blood groups, red cell antigens, HLA-antigen system of the A, B and C loci and apolipoprotein B₃ hypervariable region. Thirteen of the pairs had been living apart for at least 10 y. All subjects were premenopausal and none reported any menstrual irregularities. All individuals reported that the weight fluctuations during the 4 month period preceding the study did not exceed 2 kg. Clinical, hormonal and ultrasound examinations excluded polycystic ovary syndrome. Only two pairs smoked and only one pair was nulliparous. No subjects had a history of recent illness except for one pair of twins, each of whom had a 4 y history of non-insulin-dependent diabetes mellitus and hypertension. Their diabetes was treated by diet only and their hypertension was well controlled by small doses of an ACE inhibitor (Trandolapril 0.5 mg once a day). The study was approved by the Ethics Committee of the Charles University Hospital in Prague and by the Institutional Review Board of the University of Pennsylvania School of Medicine. It was fully explained to the subjects and their informed voluntary consent was obtained.

Methods

Initial examination

Initial examinations were performed before admission to the study on an outpatient basis in the Obesity Unit of the 4th Department of Internal Medicine, of the General Faculty Hospital, Charles University, Prague. A thorough work-up included a medical history, a physical examination, routine blood tests and an electrocardiogram. Neither the results of these tests nor those of a battery of psychological tests gave reason for exclusion of any subjects. After being cleared for participation in the study, subjects were admitted to the hospital in groups of two or four twin pairs at a time.

Study protocol

The study protocol was the same as that utilized on the Obesity Unit in the treatment of 450 obese patients during the past 5 y.^13,14 It consisted of 40 days in a metabolic unit, divided into three parts, an initial period of 7 days for adjustment to the hospital and for baseline measurements, 28 days of the weight reduction regimen and 5 days of testing after weight reduction. During the first 7 days subjects received a diet specially prepared in the hospital kitchen that provided an average daily energy intake 8.558±3.058 MJ/day (the energy content of current dietary recommendation for middle aged women performing light physical activity¹⁵), which achieved the weight stability during the week preceding the study.

Tests

During the initial week and during the five final days of the inpatient stay the following examinations were performed. Body composition was assessed by whole body densitometry employing underwater weighing with residual lung volume measured by the helium dilution method (Jaeger Pneumoscreen, Wurzburg, Germany) according to Meneely and Kaltreider.¹⁶ Total body energy was estimated from the energy content of body fat and fat-free mass. The energy content of body fat was assumed to be 39.3 kJ per g, whereas that of fat-free mass was assumed to be 4.3 kJ per g.¹⁷ Anthropometric measurements were conducted by a single examiner using flexible tape measures and Best and Harpenden calipers. Anthropometric measurements were carried out according to recommendations of the Airlie Conference¹⁸ and WHO.¹⁹ Ten skinfolds, waist circumferences and sagittal diameter at L4.5 were measured. The sum of 10 skinfolds according to Parizkova assessed subcutaneous fat²⁰ while five trunk and four extremity skinfolds assessed the trunk/extremity ratio. A Toshiba Ultrasound Scanner was used to assess intra-abdominal fat by ultrasonography, which, according to several reports,^21,22,23 is highly correlated with visceral fat measured by computerized tomography. Resting metabolic rate (RMR) was measured after a 12 h overnight fast in the recumbent position for 30 minues using a modified Oxyscreen apparatus (Jaeger GmbH, Wurzburg, Germany) equipped with paramagnetic and infrared analyzers. The mean intra-individual coefficient of variation from replicated measurements was 2.9%. Patients were instructed to avoid smoking, caffeine intake and vigorous activity for 24 h prior to the indirect calorimetry. Blood pressure was measured each morning in the recumbent position. The lowest of three measurements was recorded and those on the first and last day of the very low calorie diet (VLCD) were used in the analysis. Total cholesterol, high-density lipoprotein (HDL)-cholesterol, triglycerides and uric acid were determined by standard laboratory procedures.

Weight loss protocol

For 28 days subjects received a VLCD ReditaÒ (Promil Ltd, Nov Bydzov, Czech Republic) developed at the Obesity Unit and used in the program for the past 5 y. It provides 1.6 MJ/day with 37 g of protein, 50 g of carbohydrate and 3.8 g of fat.¹³ Daily programmed exercise was supervised by an exercise physiologist and remained at the same level throughout the study. Subjects were carefully monitored for adverse responses by daily clinical examinations and weekly laboratory tests; none were reported. Adherence to diet was monitored by daily determination of urinary ketones and adherence to activity by daily pedometer measurements.

Statistical analysis

The initial statistical analysis used a 2´2 (time and twins) repeated measures analysis of variance to compute the statistical significance of the changes from baseline to postintervention for each of the variables of interest¾body weight, fat mass, fat-free mass, hip and thigh circumferences, the sum of 10 skinfold thicknesses as well as 5 trunk skinfolds (subscapular, suprailiac, abdominal, upper thorax and lower thorax) and four extremity skinfolds (triceps, biceps, upper thigh, mid calf), individual skinfold thicknesses, sagittal diameter, ultrasound, RMR and fasting respiratory quotient (RQ). The second step was to estimate the genetic contribution to these changes by computing intraclass correlations (3, 1), as defined by Shrout and Fliess, between members of a twin pair.²⁴ These and all subsequent correlations were also computed after adjusting for age and initial values. The third step was to determine the changes in selected cardiovascular measures¾serum lipids, blood pressure and uric acid¾in response to the weight loss. We then calculated the intrapair correlations for these changes, both adjusted and unadjusted for weight loss.

Results

The 4 weeks of VLCD resulted in an 8.8±1.9 kg loss of body weight, from 93.9±21.2 kg to 85.1±19.9 kg (P<0.0001). Body composition also changed, with a greater loss of fat mass (FM) 6.5±2.3 kg (14.7%) than of fat-free mass (FEM) 2.1±1.5 kg (4.2%), reflected in a decrease in the FM/FFM ratio from 0.88 to 0.78 (P<0.0001). FFM accounted for 23.9% of weight loss. Body energy content decreased by 265 MJ or 13.6% (see Table 1).

Variability in changes in body composition

The changes in body composition in response to VLCD showed great variability Thus, although the mean weight loss was 8.8 kg, losses ranged from 5.9 to 12.4 kg. These changes were not, however, randomly distributed among the 28 subjects but were highly correlated within the twin pairs. There was 12.8 times (F ratio) more variability for changes in body weight between pairs than there was within pairs, and the F ratio for fat mass was even greater; 17.0. These findings are illustrated in Figures 1 and 2, which show the similarity in the decreases in body weight and fat mass of members of the twin pairs. For body weight and fat mass both the unadjusted intraclass correlations (ICC) and those adjusted for age and initial values were highly significant (P<0.0005). As would be expected from these values, the decrease in body energy stores within twin pairs was also highly correlated (ICC=0.90; P<0.0005).

RMR fell from 4.66±1.22 to 4.29±1.17 kJ/min (P=0.099), and RQ fell from 0.777±0.055 to 0.736±0.051 (P<0.005). No within-pair resemblances were revealed for decreases in both RMR and RQ (Table 1).

Changes in circumferences, sagittal diameter, skinfolds and intraabdominal fat are shown in Table 2. Waist and hip circumference decreased significantly and showed evidence of preferential mobilization of fat from the trunk. Thus the waist-to-hip ratio decreased, from 0.89 to 0.84 (P<0.0005). Further evidence of a decrease in the critical abdominal visceral fat was provided by the ultrasound measurement of intraabdominal thickness, which fell by 29.5% (P<0.0005), twice that of total body fat, while sagittal diameter decreased by 13.3% (P<0.0005), similar to the decrease in fat mass. Pearson correlations for these measures were modest: for waist 0.51 (P=0.06), for sagittal diameter 0.53 (P<0.05), and for ultrasound 0.47 (P=0.09). However, the intrapair similarity in the amount of intra-abdominal fat lost with the protocol was significant after adjustment for age and initial value (r=0.59, P=0.03).

The decrease in subcutaneous fat, as measured by the sum of 10 skinfolds, was 27.3%, from 265.3 to 192.7 mm (P<0.0005), The similarity between twin pairs was also comparable to that for body weight and body fat, with an F ratio of 11.0 and a correlation of 0.75 (P<0.001). There was similar within-pair resemblance for decreases in trunk and extremity skinfolds and in the trunk-to-extremity ratio. Note that the correlations adjusted for age and initial values differed only modestly from the unadjusted correlations.

Influence of weight loss on cardiovascular risk factors

Weight losses led to favorable changes in risk factors for cardiovascular disease¾lipids, blood pressure and uric acid. As was true of body composition, the baseline values of these indices were highly correlated within twin pairs (data not shown). Reduction in uric acid in response to weight loss was correlated within twin pairs, suggesting a genetic influence on it, as on reductions in body composition. By contrast, reductions in the other risk factors were not significantly correlated within twin pairs (Table 3). After controlling for weight loss only the correlation for uric acid remained marginally significant.

Discussion

An important demonstration of this study is that the method of studying weight change in identical twins, pioneered by Bouchard and his colleagues^10,11,12 in normal and modestly overweight persons, is applicable also to the study of frankly obese persons.

A major finding was that intrapair resemblances in body weight changes appear similar to those of the two studies by Bouchard's group, of weight gain by overfeeding¹⁰ and of weight loss by increased physical activity.¹¹ This similarity occurred even though those studies were conducted in men whose weight was elevated only modestly (BMI=26.2) or not at all (BMI=19.7). The involvement of genotype in loss of body weight and of fat mass was demonstrated by the correlations of members of twin pairs in the present study, 0.85 for weight loss and 0.88 for fat mass loss. These values are comparable to the values 0,74 for weight loss and 0.87 for fat mass loss reported by Bouchard in the negative energy balance protocol employing exercise.¹¹ These values are higher than those in the Bouchard¹⁰ overfeeding study, with its correlations for weight gain of 0.55 for body weight and 0.50 for fat mass. In the two weight loss studies induced by different methods in subjects who differed in gender, mean age and BMI, the contribution of genetic similarity vs genetic difference in the response pattern of weight change was similar and much higher than that in the study of weight gain.

No intrapair resemblance was revealed in the changes of RMR and RQ in response to negative energy balance induced by VLCD. RMR and fasting RQ were not reported in the long-term exercise study of Bouchard.¹¹ However, a 22-day exercise protocol in six pairs of male identical twins resulted, like in our negative balance study, in a nonsignificant intrapair correlation for RMR changes if one outlying pair was removed from statistical analysis.²⁵

A second finding was that controlling for baseline values had little effect on the twin correlations for changes of scores. Thus the mechanisms affecting intrapair similarities in responses to decreased energy intake appear to be largely independent of those influencing baseline values. How an obese person loses weight appears to be largely independent of how the obesity developed.

Studies conducted in twins revealed substantial heritability of the serum levels of total cholesterol, HDL-cholesterol, apolipoproteins A-I and B, and triglycerides, ranging from 0.28 to 0.78.²⁶ However, only a few twin studies have explored the intrapair resemblance for changes in cardiovascular risk factors in response to either positive or negative energy balance.^11,12

A third finding of our study concerned the influence of weight loss on the assessed risk factors. As in other studies of VLCDs, by others^{27,28,29,30,31} and by us,^13,14,32 these indices were markedly improved by weight loss. However, within pair resemblance of these favourable changes, suggesting genetic contribution, was limited to uric acid. It means that a favourable amelioration of cardiovascular risks in response to a short-term weight reduction regimen is mainly influenced by environmental factors and not by a hereditary predisposition. In Bouchard's study intrapair resemblance of changes in plasma total cholesterol, HDL-cholesterol and triglycerides was significant under the influence of the overfeeding protocol.¹² In contrast, no consistent within-pair resemblance in the changes of these variables could be detected during the recovery period¹² when the twins had lost about 7 of 8 kg gained during the caloric surplus with the exception of the changes in plasma triglycerides (P<0.02). Similarities in the blood pressure changes were not reported in overfeeding and postoverfeeding studies of Bouchard,^10,12 but no within pair resemblance was observed in the blood pressure changes in response to a negative energy balance protocol induced by exercise in young male identical twins.¹¹

We should like to mention that the correlation between monozygotic twins, independent of dizygotic twins, does not estimate the heritability of a trait but rather the upper bound on heritability, since it is a function of both genetic and shared family environmental influences. In our cohort the effect of shared family environment was minimized as 13 of 14 pairs had been living apart for at least 10 y. Extensive research has shown that shared family environment has little influence on obesity-related traits.^33,34,35 A recent study of Austin et al ³⁵ which included 185 monozygotic and 130 dizygotic twin pairs revealed that age-adjusted heritability for the change in BMI over the decade ranged from 0.57 to 0.86. These estimates remained statistically significant after adjustment for environmental factors and ranged from 0.57 to 0.78, although dizygotic intraclass correlations were low after these adjustments. Therefore, we, like others,^10,11,12 believe that studies of the response of monozygotic twins to interventions provide useful information on genetic effects, even without the presence of dizygotic twins.

Acknowledgements

This study was supported in part by grants IGA MZ CR 6 4311-3 from the Czech Ministry of Health, the Danone Institute and R03-TW-00561-01 from the NIH.

1 Zhang YU, Proenca R, Maffei M, Barone M, Leopold L, Friedman JM. Positional cloning of the mouse OB gene and its human homologue. Nature 1994; 372: 425-432. MEDLINE

2 Pérusse L, Chagnon YC, Weisnagel J, Bouchard C. The human obesity gene map: the 1998 update. Obes Res 1999; 7: 111-129. MEDLINE

3 Montague CT, Farooqui IS, Whitehead JT, Soos MA, Rau H, Wareham NJ, Sewter CP, Digby JE, Mohammed SN, Hurst IA, Cheetham CJ, Earley AR, Barnett AH, Prins JP, O'Rahilly S. Congentital leptin deficiency is associated with severe early-onset obesity in humans. Nature 1997; 387: 903-908. Article MEDLINE

4 Comuzzie AG, Allison DA. The search for human obesity genes. Science 1998; 280: 1374-1377. Article MEDLINE

5 Pérusse L, Tremblay A, Leblanc C, Cloninger CR, Reich T, Rice J, Bouchard C. Familial resemblance in energy intake: contribution of genetic and environmental factors. Am J Clin Nutr 1988; 47: 629-635. MEDLINE

6 Tarusk V, Beaton GH. The nature and individuality of within-subjects variation in energy intake. Am J Clin Nutr 1991; 54: 464-470. MEDLINE

7 Bouchard C, Pérusse L. Genetics of obesity. Ann Rev Nutr 1993; 13: 337-354.

8 Ravussin F, Bogardus C. Relationship of genetics, age, and physical fitness to daily energy expenditure and fuel utilization. Am J Clin Nutr 1989; 49: 968-975. MEDLINE

9 Roberts SB, Savage J, Coward WA, Chow B, Lucas A. Energy expenditure and intake in infants born to lean and overweight mothers. New Engl J Med 1988; 318: 461-466. MEDLINE

10 Bouchard C, Tremblay A, Després JP, Nadeau A, Lupien PJ, Thériault G, Duussuault J, Moorjani S, Pinault S, Fournier G. The response to long-term overfeeding in identical twins. New Engl J Med 1990; 322: 1477-1482. MEDLINE

11 Bouchard C, Tremblay A, Despres JP, Thériault G, Nadeau A, Lupien PJ, Moorjani S, Prudhomme D, Fournier G. The response to exercise with constant energy intake in identical twins. Obes Res 1994; 2: 44-410.

12 Bouchard C, Tremblay A, Després J-P, Nadeau A, Lupien PJ, Moorjani S, Thériault G, Kim SY. Overfeeding in identical twins: 5-year postoverfeeding results. Metabolism 1996; 45: 1042-1050. MEDLINE

13 Hainer V, Kunesova M, Stich V, Parizkova J, Zak A, Wernischova V, Kozich P, Hrabak P, Dedicova L. Very low energy formula diet in the treatment of obesity. Int J Obes 1989; 13: (Suppl 2) 185-188. MEDLINE

14 Hainer V, Stich V, Kunesova M, Parizkova J, Zak A, Wernischova V, Stukavec V. Body fat distribution, metabolic and hormonal indices in obese patients treated initially by very low calorie diet. In: Romsos DR, Himms-Hagen J, Suzuki M (eds). Obesity: diettary factors and control. Japan Scientific Press: Tokyo and Karger: Basel,, 1991, , pp 215-226.

15 Food and Nutrition Board. Recommended Dietary Allowances, 9th edn. National Research Council, NAS: Washington, DC, . 1980;

16 Meneely GR, Kaltreider NL. The volume of the lung determined by helium dilution, description of the method and comparison with other procedures. J Clin Invest 1949; 28: 129-139.

17 Heymsfield SE, Tighe A, Wang ZM. Nutritional assessment by anthropometric and biochemical methods. In: Shils ME, Olson JA, Shike M (eds). Modern nutrition in health and disease. Lea & Febiger: Philadelphia, PA, 1994, , pp 812-841.

18 Lohman TG, Roche AF, Martorell R. Anthropometric standardization reference manual. Human Kinetics: Champaign, IL, . 1991; ,pp 39-70.

19 Measuring obesity¾classification and description of anthropometric data. Report on a WHO consultation on the epidemiology of obesity, Warsaw, 1987,. pp 24-51.

20 Parizkova J. Body fat and physical fitness. Nijhoff M: The Hague. 1977; pp 32-36.

21 Armellini F, Zamboni M, Rigo L, Todesco T, Bergamo-Andreis IA, Procacci C, Bosello O. The contribution of sonography to the measurement of intraabdominal fat. J Clin Ultrasound 1990; 18: 563-567. MEDLINE

22 Tornaghi G, Raiteri R, Pozzato C, Rispoli A, Bramani M, Cipolat M, Craveri A. Anthropometric or ultrasonic measurements in assessment of visceral fat? A comparative study. Int J Obes Relat Metab Disord 1994; 18: 771-775. MEDLINE

23 Kunesova M, Hainer V, Hergetova H, Parizkova J, Horejs J, Stich V. Simple anthropometric measurements¾relation to body fat mass, visceral adipose tissue and risk factors of atherogenesis. Sborn lek 1995; 96: 257-268.

24 Shroutt PE, Fleiss JL. Intraclass correlation: uses in assessing rater ability. Psychol Bull 1979; 86: 420-428.

25 Poehlman ET, Tremblay A, Nadeau A, Dussault J, Thériault G, Bouchard C. Heredity and changes in hormones and metabolic rates with short-term training. Am J Physiol 1986; 250: E711-E717. MEDLINE

26 Heller DA, de Faire U, Pedersen NL, Dahlen G, McClearn GE. Genetic and environmental influences on serum lipid levels in twins. New Engl J Med 1993; 328: 1150-1156. MEDLINE

27 Uusitupa M, Laakso M, Sarlund H, Majander H, Takala J, Penttila I. Long term effects of a very low calorie diet on metabolic control and cardiovascular risk factors in the treatment of obese noninsulin-dependent diabetics. Int J Obes 1989; 13: (Suppl 2) 163-1644. MEDLINE

28 Anderson JW, Hamilton CC, Brinkman-Kaplan V. Benefits and risks of an intensive very-low- calorie diet program for severe obesity. Am J Gastroenterol 1992; 87: 6-15. MEDLINE

29 Reisin E, Frohlich EDS, Messerli FH, Dreslinski GR, Dunn FG, Jones MM, Batson HM Jr. Cardiovascular changes after weight reduction in obesity hypertension. Ann Intern Med 1983; 98: 315-319. MEDLINE

30 Kanders BS, Blackburn GL. Very-low-calorie diets for the treatment of obesity. In: Blackburn OL, Kanders BS (eds) Obesity, pathophysiology, psychology, and treatment. Chapman and Hall: New York, 1993,, pp 197-216.

31 Voliovitch H, Magazanic A, Voliovitch I, Blachar Y, Rudniki C, Zahavi I. Very low calorie diets¾clinical trial. Int J Obes 1989; 13: (Suppl 2) 157. MEDLINE

32 Schotte DE, Stunkard AJ. The effects of weight reduction on blood pressure in 301 obese patients. Arch Intern Med 1990; 150: 1701-1704. MEDLINE

33 Grilo CM, Pogue-Geile MF. The natures of environmental influences on weight and obesity: behavioral genetic analysis. Psychol Bull 1991; 110: 520-537. MEDLINE

34 Maes HM, Neale MC, Eaves LJ. Genetic and environmental factors in relative body weight and human adiposity. Behav Genetics 1997; 27: 325-351.

35 Austin MA, Friedlander Y, Newman B, Edwards K, Mayer-Davis EJ, King MC. Genetic influences on changes in body mass index: a longitudinal analysis of women twins. Obes Res 1997; 5: 326-331. MEDLINE

Figure 1 Similarity within pairs with respect to decrease in body weight in 14 pairs of obese female twins in response to 28 days of very low calorie diet (1.6 MJ). Each point represents one pair of twins (A and B).

Figure 2 Similarity within pairs with respect to decrease in body fat in 14 pairs of obese female twins in response to 28 days of very low calorie diet (1.6 MJ). Each point represents one pair of twins (A and B).

Table 1 Effect of 28 days of very low calorie diet on body mass, body composition, resting metabolic rate (RMR) and respiratory quotient (RQ). Intrapair correlations of changes. All RMR and RQ correlations have been adjusted for age, fat-free mass and fat mass

Table 2 Effect of 28 days of very low calorie diet on circumferences, sagittal diameter, intra-abdominal fat (measured by ultrasound) and skinfold thicknesses. Intrapair correlations for change

Table 3 Effect of 28 days of very low calorie diet on selected cardiovascular measures and intrapair correlations of change