Changes in body composition, physical performance and ...

European Journal of Clinical Nutrition (2005) 59, 685–694

& 2005 Nature Publishing Group All rights reserved 0954-3007/05 $30.00 www.nature.com/ejcn

ORIGINAL COMMUNICATION Changes in body composition, physical performance and cardiovascular risk factors after a 3-week integrated body weight reduction program and after 1-y follow-up in severely obese men and women NA Maffiuletti1,2*, F Agosti1, PG Marinone1, G Silvestri3, CL Lafortuna4 and A Sartorio1,3 1 Experimental Laboratory for Endocrinological Research, Italian Institute for Auxology, IRCCS, Milan, Italy; 2INSERM Erm 207 Laboratory, University of Burgundy, Dijon, France; 33rd Division of Metabolic Diseases, Italian Institute for Auxology, IRCCS, Piancavallo - Verbania, Italy; and 4Institute of Molecular Bioimaging and Physiology, CNR, Milan, Italy

Objective: To investigate the short- and medium-term (ie, at 1-y follow-up) effects of a hospital-based body weight reduction (BWR) program lasting 3 weeks in severely obese individuals. Design: Longitudinal, clinical intervention study entailing energy-restricted diet, tailored aerobic-strength exercise, psychological counselling and nutritional education during a 3-week period followed by 49 weeks of indirect supervision at home. Subjects: In all, 45 women and 19 men aged 30.277.2 y (mean7s.d.) with severe obesity (BMI: 41.374.3 kg/m2), classified as weight losers (finaloinitial weight) and regainers (finalZinitial weight). Interventions: Body composition, physical performance and cardiovascular risk factors before and after the BWR program and at follow-up. Results: The large majority of the experimental subjects obtained a clinical success after the 1-y period, and the rate was higher in females (n ¼ 37/45, ie, 82.2%) than in males (n ¼ 11/19, ie, 57.9%). At follow-up, weight losers had higher percent fat-free mass, muscle strength, HDL-cholesterol and self-reported physical activity level and lower total cholesterol and glucose levels than weight regainers (Po0.05–0.01). Males displayed significantly larger fat-free mass losses than females at post-BWR and at follow-up. Conclusions: A 3-week hospital-based integrated BWR program resulted in a high rate of clinical success in severely obese individuals at 1-y follow-up (particularly in females), that was associated with increased levels of self-reported physical activity and improved muscle strength and lipid profile. Further research is needed to establish the long-term effects (ie, at 2–5 y) associated to this treatment.

European Journal of Clinical Nutrition (2005) 59, 685–694. doi:10.1038/sj.ejcn.1602130 Published online 16 March 2005 Keywords: follow-up; fat-free mass; muscle strength; physical activity rating score; blood pressure; cholesterol

Introduction Severe obesity (ie, BMI440 kg/m2) is frequently characterized by low physical fitness and high cardiovascular risk due *Correspondence: NA Maffiuletti, INSERM Erm 207 Laboratory (Motricite´-Plasticite´), University of Burgundy, BP 27877 - 21078 Dijon cedex, France. E-mail: [email protected] Guarantors: NA Maffiuletti and A Sartorio. Contributors: AS, CLF and NAM designed the study. FA, GS and PGM undertook the protocol and collected the data. NAM and AS interpreted the data and drafted the manuscript. NAM revised and edited the manuscript. Received 23 June 2004; revised 18 November 2004; accepted 7 January 2005; published online 16 March 2005

to a widespread sedentary lifestyle. However, the majority of the obesity-linked complications can be reversed with a number of intervention strategies including diet, exercise or a combination of both, aimed at reducing body weight and more particularly fat mass. In their meta-analysis, Miller et al (1997) concluded that either a diet or diet plus exercise program would be more beneficial on a short-term basis than exercise alone. In the same way, better long-term maintenance of weight loss is obtained through a combination of diet and exercise, due to the compensatory effects associated with each individual treatment, that is, reduced resting metabolic rate with diet and increase in energy intake with exercise (Zachwieja, 1996). Hills and Byrne (1998)

Medium-term changes in health outcomes in obese NA Maffiuletti et al

686 recently suggested that the combined approach of diet, exercise and behaviour modification in weight management programs gives the best and often permanent results. For an obese individual, one way to adhere rapidly to lifestyle modifications is represented by a short-term inpatient treatment supervised on a daily basis by medical doctors, dieticians, therapists and psychologists. It has indeed been demonstrated that a 3-week body weight reduction (BWR) program entailing energy-restricted diet, physical activity, education and psychological counselling resulted in marked decreases of body weight (more likely fat mass) and common cardiovascular disease risk factors (Sartorio et al, 2001a) as well as improved muscle function (Sartorio et al, 2001b, 2004a) and aerobic fitness (Sartorio et al, 2003) in severely obese men and women. However, whether these positive outcomes acquired after a few weeks in such a high-risk population are maintained during medium-term (eg, after 1 y) follow-up is not known, since the majority of follow-up studies considered treatments lasting more than 1 month (Pronk & Wing, 1994) and overweight or moderately obese subjects (Miller et al, 1997). Moreover, the short- and medium-term changes in body composition, physical performance and cardiovascular risk factors are typically analyzed separately, without considering the strong interdependence between these groups of variables. The main aim of this study was therefore to investigate if the short-term effects of a hospital-based BWR program entailing energy-restricted diet, exercise, psychological

counselling and nutritional education on body composition, physical performance and cardiovascular risk factors were still visible after medium-term follow-up in a population of severely obese men and women. For this reason, 64 obese patients were tested before and after a 3-week BWR intervention and once again 1 y later. A secondary aim was to explore whether subjects that exhibited body weight regains between baseline and 1-y follow-up (ie, a clinical failure) were different at the three stages of the experiment with respect to their successful counterparts (ie, weight losers) in body composition, physical performance and cardiovascular risk variables.

Methods Subjects and experimental procedures The subjects were 70 obese volunteers (50 women and 20 men, Table 1) referred to the 3rd Division of Metabolic Diseases (Italian Institute for Auxology, Piancavallo, Verbania, Italy) by physicians for in-patient treatment because of obesity. Subjects were selected on the basis of the following criteria: age Z18 y; BMI Z35 kg/m2; absence of hearth, kidney, liver, endocrine and other major disease; premenopausal state for women; no previous participation in an intensive BWR program during the year preceding the study and motivation to take part in a structured weight loss program. For different reasons, two women dropped out during the in-patient BWR program and three other women and one man dropped out during the 1-y follow-up, leading

Table 1 Baseline characteristics of the experimental subjects and body composition data before and after the 3-week BWR program and after the 1-y follow-up

Age (y) Height (m) BMI (kg/m2)a Fat-free mass (kg)b,d,e Fat-free mass (%)b,d,e Fat mass (kg)a

Females (n ¼ 45) Males (n ¼ 19) Females Males All subjects Females Males All subjects Females Males All subjects Females Males All subjects Females Males

Mean7s.d. a significant time effect (Po0.001). b significant time effect (Po0.05). c significantly different from baseline (Po0.001). d significant gender effect (Po0.001). e significant time gender interaction (Po0.05). f significantly different from baseline (Po0.05). g significantly different from baseline (Po0.01). h significantly different from post-BWR (Po0.05).

European Journal of Clinical Nutrition

Pre-BWR

Post-BWR

Follow-up

30.977.1 28.577.2 1.6170.07 1.7570.07 41.374.3 40.773.4 42.675.7 59.7714.6 51.574.8 79.2710.8 52.476.4 48.974.0 60.771.5 53.7710.5 54.6710.8 51.679.8

— — — — 39.373.9c 38.973.2 40.375.2 58.5713.2f 51.174.1 76.0710.5 53.976.1g 50.774.1 61.471.9 50.0710.3c 50.8710.8 48.178.9

— — — — 38.874.9c 38.075.2 40.773.9 58.8713.3 51.274.9 76.878.8 55.276.2c,h 52.475.3 61.771.7 48.0712.1c 47.8713.7 48.577.5


687 to a total of 64 patients definitively included in the analyses (45 women and 19 men). The baseline characteristics of these subjects are given in Table 1. The majority of them has been obese in the last 5–10 y and has received unsuccessful out patient treatments for body weight reduction. Of these 64 subjects, nine were taking medications for hypertension, six for diabetes and only two for dyslipidaemia. All the patients gave their signed informed consent to participate in the study, which was approved by the Ethics Committee of the Italian Institute for Auxology. All the procedures were conducted according to the principles expressed in the Declaration of Helsinki. A series of examinations (see below) was performed before and at the end of the 3-week BWR program and once again during a day hospital organized 1 y after the beginning of the treatment. After the intensive intervention, subjects were left on their own and were encouraged to control their diet and to exercise to avoid body weight gain. A table listing the energy content of popular foods was provided and the procedure for calculating daily energy intake was explained to every patient at the end of the BWR program. Individuals were also encouraged to maintain the type and quantity of physical activity performed at the hospital (see below) during the B11-month period, but they did not try any additional diet/activity treatment during this period. Patients were strongly encouraged to contact our dieticians, therapists and psychologists for counselling at any time during the free living period. This occurred one time for 25 (39%) subjects and 2–3 times for eight (12%) subjects (mainly for nutritional counselling). The 64 patients were contacted through telephone 6 months from baseline, according to Jakicic (2003) suggestions. The main aim of this contact was to quantify the amount of physical activity performed following the BWR program by using the physical activity rating questionnaire (Jackson et al, 1990). The phone call also comprised a discussion of diet and behavioural strategies, but adherence to the nutritional instructions was not directly verified.

Body weight reduction program The BWR program combined an energy-restricted diet, nutritional education, psychological counselling and moderate physical activity during a 3-week period. The laboratory’s trained team of dieticians individually planned an energy-restricted diet (5023–7535 kJ/day, ie, 1200–1800 kcal/day) containing 21% proteins, 53% carbohydrates and 26% lipids. The amount of energy to be given with the diet was calculated by subtracting approximately 2093 kJ (500 kcal) from the measurement of basal energy expenditure, as assessed by indirect calorimetry (Vmax 29; SensorMedics Corporation, Yorba Linda, CA, USA) for a total duration of 20 min while the subjects rested in bed. Owing to time-scheduling constraints, basal energy expenditure was measured exclusively before the BWR program. The esti-

mated H2O content of the food was 1000 ml/day and the estimated salt content was 1560 mg/day of Na þ , 3600 mg/ day of K þ and 900 mg/day of Ca2 þ . A fluid intake of at least 2000 ml/day was encouraged. Dietary compliance was evaluated every day by a dietitian. Nutritional education included lectures, demonstrations and group discussions with and without a supervisor, which took place every day throughout the experimental period (duration 60 min). Subjects were shown how to improve their diet (eg, through increasing fruit and vegetable consumption, and decreasing energy intake from fats). Nutritional education also included information about nutrition and physical activity as well as discussions of various aspects of the food consumption process. Sessions were taught by endocrinologists, nutritionists, psychologists and physical educators/exercise physiologists who are specialized in the treatment of obesity and experienced in providing nutritional education to obese individuals. Psychological counselling sessions were conducted by clinical psychologists 2–3 times/week (duration 60 min) and were based on individual or cognitive-behavioural strategies, for example, stimulus control procedures, problem solving training, stress management skills, development of healthy eating habits, assertiveness training, facilitation of social supports, cognitive restructuring of negative maladaptive thoughts and relapse prevention training. Physical activity included five training sessions/week individually supervised by our laboratory’s trained team (exercise physiologists, physical educators and physicians). Each session included 30–40-min aerobic activities (typically 10 min on an ergocycle, 20 min on a treadmill and 5 min on an arm-ergometer) performed at 40–70% of individual maximal oxygen consumption (VO2 max), followed by 15 repetitions at 40–70% of one repetition-maximum (1-RM) load on three isotonic machines for the upper and lower limb muscles (Technogym, Gambettola, Italy). For both aerobic and strength conditioning, the training intensity was progressively increased ( þ 10% every week) from the first to the last week and the initial level (40 or 50%) was individually selected according to the age and to the general fitness level of the subjects.

Body composition assessment Fat-free mass and fat mass were assessed by bioelectric impedance analysis, performed in the early morning after an overnight fast, according to a conventional standard technique (Lukaski et al, 1986). Whole-body resistance to an applied current (at 1, 5, 10, 50 and 100 kHz, 0.8 mA) was measured with a tetrapolar device (Human IM, Dietosystem, Milan, Italy). The electrodes were placed on the right wrist and ankle of the subjects while they lay comfortably supine in a bed with the limbs abducted from the body. Fat-free mass was calculated with the fat-specific equations derived by Gray et al (1989). Fat mass was derived as the difference between total body weight and fat-free mass. European Journal of Clinical Nutrition


688 Physical performance assessment Stair climbing test (SCT). SCT consists in a modification of the test proposed by Margaria et al (1966) and has recently been applied in a large population of severely obese individuals (Sartorio et al, 2004a, b) to estimate the maximal anaerobic power of the lower limb muscles. At the moment of the first execution, 2–3 practice trials were allowed so that the subjects gained a good control of the performing technique. Subjects were invited to climb up ordinary stairs at the highest possible speed, according to their capabilities. The stairs consisted of 13 steps of 15.3 cm each, thus covering a total vertical distance of 1.99 m. An experimenter measured the time employed to cover the test with a digital stopwatch. SCT repeatability in adult obese individuals has been previously assessed in our laboratory and the coefficient of variation between measurements has been found to be lower than 5% (Sartorio et al, 2001b). Muscle strength test. 1-RM load was quantified on a leg press isotonic machine (Technogym, Gambettola, Italy) during closed-kinetic chain eccentric–concentric actions, with a range of motion at the knee joint of about 901. Subjects completed 3–4 series of 15–20 repetitions with submaximal loads, and 1-RM was estimated by using the formula proposed by Brzycki (see Mayhew et al, 1995). Maximal oxygen consumption test. VO2 max was predicted from a graded submaximal test performed on a cycle ergometer (Technogym, Gambettola, Italy). After warming up for 6 min at 50 W, subjects were exposed to a workload of 60, 70 and 80 W during which they were instructed to maintain B60 rpm. Each step lasted 4 min to ensure steady state. The test was stopped if heart rate, consistently measured by means of a telemetric system (Polar, Kempele, Finland), exceeded 75% of the individual maximal heart rate, estimated as 220age (y). Heart rate recorded during the last minute of each step was used to estimate VO2 max, according to the formula proposed by the American College of Sports Medicine (2000). Perceived exertion was also graded by the subjects at the end of every step using a modified Borg scale (Borg, 1998). Physical activity rating questionnaire. The physical activity rating scale provides an assessment score of 0–7 on an individual’s level of physical activity (Jackson et al, 1990). The questionnaire consists in a series of eight statements about routine physical activity and subjects have to select only one response that best describes their physical activity level. Obese patients were administered the questionnaire at baseline, after 6 months (by telephone call) and after 1 y.

Cardiovascular risk factors assessment Two blood pressure determinations were performed manually by using an aneroid sphygmomanometer and appropriate cuff bladders (Maxwell et al, 1982) after the patients had been sitting at least 5 min, and the mean systolic blood pressure value was retained. Blood samples were collected after an overnight fast in standard tubes for European Journal of Clinical Nutrition

serum. Total cholesterol, HDL cholesterol and glucose levels were immediately measured with enzymatic-colorimetric methods (Hitachi Instruments, Tokyo, Japan). Blood pressure and cholesterol levels were considered without regard to the use of antihypertensive or lipid-lowering medications. Statistical procedures First, data were analyzed for significant differences using a two-way analysis of variance with repeated measures on time. The factors to be analyzed included time (pre- and post-BWR program and follow-up) and gender. A Tukey post hoc test was conducted if significant main effects or interactions were present. Subsequently, differences between weight losers vs weight regainers at baseline, post-BWR and follow-up were analyzed by using a two-tailed Student’s t-test for independent samples. In each case the level of significance was established at Pr0.05. Data are presented as mean7s.d. in figures, tables and the text. Statistical analyses were performed using Statistica software (StatSoft, Tulsa, OK, USA).

Results The BWR program resulted in significant reductions of body weight and BMI (4.771.2%; mean7s.d.) and kg of both fat mass (7.173.5%) and fat-free mass (1.574.0%) at the end of the 3-week period (Table 1). In the same way, significant improvements at this stage were found for all the physical performance (Table 2) and cardiovascular risk variables (Table 3). The largest relative change was found for muscle strength ( þ 37.4723.9%), while the smallest variation was the 4.979.4% reduction in systolic blood pressure. At follow-up, mean body weight and fat mass were still significantly lower than baseline, respectively, 5.879.1 and 10.3715.7%. A significant time gender interaction was found for fat-free mass (in kg), since males displayed larger losses than females both at post-BWR and at follow-up (Figure 1). Physical performance variables slightly worsened between post-BWR and follow-up, but subjects even performed better at follow-up than at baseline. On the other hand, all cardiovascular risk variables significantly increased from post-BWR to follow-up and attained values equal to or significantly higher than baseline (HDL-cholesterol and glucose). The physical activity rating score almost doubled from baseline to 6-month follow-up (Po0.001) but then decreased at 1-y follow-up (Po0.001), even though it was still higher than baseline (Po0.01; Table 2). For each individual, final (at follow-up) vs initial body weight was then compared to study the rate of clinical success (ie, finaloinitial weight: weight losers) vs failure (ie, finalZinitial weight:weight regainers) at follow-up. The large majority (n ¼ 48/64, ie, 75%) of the obese individuals involved in the present study obtained a clinical success after the 1-y period, and the rate was higher in females (n ¼ 37/45, ie, 82.2%) than in males (n ¼ 11/19, ie, 57.9%).


689 Table 2 Physical performance data before and after the 3-week BWR program and after the 1-y follow-up

PAR scorea 1-RM leg pressa,e VO2 max (ml/min/kg)a,f Stair climbing time (s)a,g

All subjects Females Males All subjects Females Males All subjects Females Males All subjects Females Males

Pre-BWR

Post-BWR

Follow-up

2.372.2 2.472.2 2.372.1 262.1773.2 233.0752.8 331.0769.7 23.876.2 23.275.8 25.377.1 3.670.7 3.770.6 3.470.8

4.872.2b 4.972.3 4.572.2 357.67105.3 317.6785.9 452.4785.3 27.376.7b 26.276.6 30.076.2 3.370.5b 3.470.5 3.070.4

3.472.6c,d 3.472.6 3.372.5 308.5788.5b,d 278.5765.8 378.2796.5 26.076.8b 24.774.8 29.379.4 3.470.5c 3.670.5 3.170.4

Mean7s.d.; PAR, physical activity rating; 1-RM, one repetition maximum. a significant time effect (Po0.001). b significantly different from baseline (Po0.001). c significantly different from baseline (Po0.01). d significantly different from post-BWR (Po0.001). e significant gender effect (Po0.001). f significant gender effect (Po0.05). g significant gender effect (Po0.01). Note: post-BWR for physical activity rating is post 6 months.

Table 3 Cardiovascular risk data before and after the 3-week BWR program and after the 1-y follow-up Pre-BWR SBP (mmHg)

a,b

Total-C (mg/dl)a,e HDL-C (mg/dl)a,b Glucose (mg/dl)a

All subjects Females Males All subjects Females Males All subjects Females Males All subjects Females Males

126.9711.6 124.4710.7 132.6711.9 191.9739.9 197.5740.2 178.5736.9 43.679.0 45.579.2 39.176.6 81.9713.0 82.5714.7 80.778.1

Post-BWR

Follow-up c

120.2710.7 117.3710.1 126.879.2 159.0732.1f 165.5730.9 143.6730.5 35.977.2f 38.077.3 31.173.9 74.178.4f 74.079.2 74.276.4

128.8716.1d 127.2716.1 132.6715.9 192.3741.8d 200.4740.2 173.1740.1 50.4712.0f,d 52.9712.3 44.679.2 93.1712.2f,d 93.0712.8 93.5711.0

Mean7s.d.; SBP, systolic blood pressure; C, cholesterol. a significant time effect (Po0.001). b significant gender effect (Po0.01). c significantly different from baseline (Po0.01). d significantly different from post-BWR (Po0.001). e significant gender effect (Po0.05). f significantly different from baseline (Po0.001).

Subsequently, dependent variables for subjects from the weight regainers group (n ¼ 16; eight males and eight females; age: 30.276.8 y; BMI: 41.473.0 kg/m2) were compared to those of 16 weight losers, matched for gender, age (28.278.0 y) and BMI (42.776.3 kg/m2). It was indeed demonstrated that BMI and fat mass were significantly higher and percent fat-free mass was significantly lower at follow-up in weight regainers (Po0.01–0.001; Figure 2). Physical activity rating score was significantly higher at 6 and 12 months in weight losers (Po0.01) and the same result was observed for maximal muscle strength at follow-up (Po0.05; Figure 3). Total- and HDL-cholesterol were sig-

nificantly different at post-BWR and follow-up between the two subgroups (Po0.05; Figure 4). Finally, glucose levels were significantly higher in weight regainers than in weight losers at the end of the experimental period (Po0.01; Figure 4d).

Discussion The results of this study confirm that a 3-week hospital-based BWR program entailing energy-restricted diet, physical activity, psychological counselling and nutritional education improved significantly body composition, physical European Journal of Clinical Nutrition


690 0

a

50

***

45

-1 BMI (kg/m²)

-1.5 -2 -2.5 -3 -3.5

40 35 30 Losers Regainer

25

Post-BWR

Females Males

*

Figure 1 Mean (7s.d.) absolute changes for females (n ¼ 45) and males (n ¼ 19) in fat-free mass at post-BWR and at follow-up with respect to baseline. *: Significantly lower in males than in females (Po0.05); ***: Significantly lower in males than in females (Po0.001).

performance and cardiovascular risk factors in severely obese subjects of both genders. An original finding is that male individuals lost approximately 3 kg of fat-free mass after the treatment, while reductions were negligible in their female counterparts. For the first time in our laboratory, the medium-term effects associated to the treatment were also examined and it was found that 75% of the original subjects (82% of the women and 58% of the men) maintained a body weight lower than baseline B11 months after the end of the in-patient BWR program. Interestingly, clinical success at follow-up was associated with higher levels of self-reported physical activity with respect to clinical failure, and as a consequence, weight losers had higher muscle strength and lower cardiovascular risk factors than regainers at the end of the study. Changes in body composition, physical performance and cardiovascular risk factors induced by a short-term hospitalbased treatment in severely obese individuals were studied concomitantly in the present investigation. Different body composition changes were observed between genders, where women mainly reduced the proportion of fat mass and only to a lesser extent fat-free mass, while the opposite outcome was observed in men. These findings are in accordance with the study of Goodpaster et al (1999), who reported that obese men lost more fat-free mass than obese women (4.6 vs 2.1 kg) following a 16-week caloric restriction-induced weight loss program. These authors also demonstrated that glucose levels were almost identical between obese males and females both before and after the intervention, therefore confirming that in our study the absence of sex effect in

Figure 2 Mean (7s.d.) BMI (a), fat-free mass in kg (b) and in % (c) and fat mass (d) before and after the 3-week BWR program and after the 1-y follow-up, for weight losers and regainers (n ¼ 16 for both groups). **: Significantly higher in losers (Po0.01); ***: Significantly higher in regainers (Po0.001).


20 Pre-BWR

Follow-up

b

Post-BWR

Follow-up

90 80

Fat-free mass (kg)

***

70 60 50 40 30

Pre-BWR

Post-BWR

Follow-up

c 70

Fat-free mass (%)

-4 -4.5

**

60

50

40

30

Pre-BWR

Post-BWR

Follow-up

***

d 70

60 Fat mass (kg)

∆Fat-free mass (kg)

-0.5

50

40

30 Pre-BWR

Post-BWR

Follow-up


691

a

8

Physical activity ratio.

7

Losers Regainers

** **

6 5 4 3 2 1 0

b

500

1-RM leg press (kg)

Pre-BWR

400

Post-6 mo

Follow-up *

300

200

100 Pre-BWR

VO2 max (ml/min/kg) .

c

Post-BWR

Follow-up

40

30

20

10

d

5

Stair climbing time (s)

Pre-BWR

4

Post-BWR

Follow-up

3

2

1 Pre-BWR

Post-BWR

Follow-up

Figure 3 Mean (7s.d.) score at the physical activity rating questionnaire (a), 1-RM strength at the leg press machine (b), maximal oxygen consumption (c) and stair climbing time (d) before and after the 3-week BWR program and after the 1-y follow-up, for weight losers and regainers (n ¼ 16 for both groups). *: Significantly higher in losers (Po0.05); **: Significantly higher in losers (Po0.01).

glucose levels at post-BWR or follow-up was not strictly related to the significant gender differences in fat-free mass during the same time frames. Other studies failed to demonstrate such a sex difference in the composition of weight loss (Ballor & Poehlman, 1994; Garrow & Summerbell, 1995; Dionne et al, 1999; Doucet et al, 2002), since changes in fat-free mass associated to diet and/or exercise were comparable between men and women. These conflicting findings are mainly attributable to the heterogeneity in experimental subjects, study design, exercise intensity and type, as well as intervention length between the present and the above-cited studies. In their meta-analysis, Garrow and Summerbell (1995) indicated that, at any given weight loss, the loss of fat-free mass is likely to be less in exercising subjects than in sedentary ones. It is indeed possible that our male obese subjects were more sedentary than their female counterparts, since baseline maximal oxygen consumption normalized to the fat-free mass was higher in the latter group with respect to the former (47.8 vs 41.8 ml/min/kg). The fat-free mass reductions reported here and in our previous study (Sartorio et al, 2004a) for male subjects (B3 kg) are very similar to the losses observed in obeseoverweight men following 12–16 weeks of diet (Kraemer et al, 1999; Cox et al, 2003) or diet plus aerobic exercise (Hagan et al, 1986). In obese children, fat-free mass reduction during a 3-week BWR program similar to that adopted here has been recognized as a major factor for later regain in body weight (Schwingshandl & Borkenstein, 1995). In this context, it is noteworthy that consistent fat-free mass losses observed in our males after the short-term treatment were associated with a lower rate of medium-term success compared to their female counterparts. Therefore, one could conclude that 3 weeks of diet plus exercise performed at moderate intensity for five sessions per week may not have been sufficient enough to maintain fat-free mass in severely obese men. It is indeed recommended that the physical training considered in the present study must be implemented, at least for male individuals, with specific exercises for the development of muscle power and/or strength (eg, resistance training), in order to preserve the amount of fatfree mass during body weight reduction (Donnelly et al, 2003). In the whole group of subjects, the level of self-reported physical activity at month 6 was twice with respect to baseline, therefore indicating that severely obese subjects modified their lifestyle in the first months consecutive to the completion of the BWR program. These effects were however less evident after 12 months, since the physical activity rating score significantly decreased with respect to month 6, even though it remained still higher than baseline. The more interesting finding was however observed when physical activity rating score was studied in clinical success vs failure groups. Indeed, the former subjects almost saturated the questionnaire at month 6 and then maintained the same score at follow-up, while the self-reported level of physical activity in weight regainers did not change significantly European Journal of Clinical Nutrition


692 during the course of the study, therefore confirming that low levels of physical activity are predictive of subsequent weight regain in both men and women (Kuczmarski et al, 1994).

a

160 150

SBP (mmHg)

140 130 120 110 100

Losers Regainers

90 80

b

Pre-BWR

Post-BWR

Follow-up

270

*

Cholesterol (mg/dl) .

240 *

210 180 150 120 90

c

Pre-BWR

Post-BWR

70

Follow-up **

HDL (mg/dl) .

60 *

50 40 30 20 Pre-BWR

Glucose (mg/dl) .

d

Post-BWR

Follow-up **

120

100

These findings are also reinforced by the fact that higher muscle strength and lower cardiovascular risk factors (discussed below) were observed in weight losers at followup. The ensemble of these results allows to conclude that physical exercise plays a crucial role in weight loss and in maintenance of weight loss in severely obese individuals from both genders. The rate of clinical success observed in the current investigation is particularly high (75%) compared to previous research on medium-term follow-up (Hoiberg et al, 1984; McGuire et al, 1999). The fact that success rate was higher in female (82%) than in male subjects (58%) is difficult to explain based on the present findings, but it is nevertheless interesting to speculate that the consistent fatfree mass reductions observed in the latter subjects both after the BWR program and at follow-up would have contributed to this occurrence. Based on fat-free mass findings, one could also conclude that the present 3-week intervention was too aggressive for our male obese individuals and therefore it would not be sustainable in the long term. Further research using larger sample sizes is needed in this direction. Long-term elevated levels of high blood pressure, high cholesterol and overweight are risk factors for disease (eg, renal failure, heart failure, atherosclerosis, some cancers, etc). Abundant data in the scientific literature have shown weight loss to be associated with decreases in blood pressure, total cholesterol and glucose levels as well as increases in HDL cholesterol (Dattilo & Kris-Etherton, 1992; Svendsen et al, 1994; Fox et al, 1996; Kraemer et al, 1999; Sartorio et al, 2001a; Neter et al, 2003), which are very similar to those observed at the completion of the present BWR program. Different factors, alone or in combination, are responsible for these changes, although their independent role is at present difficult to determine. In fact, both weight loss and exercise have been shown to independently promote a shift in the balance between parasympathetic and sympathetic activity in obese subjects (Arrone et al, 1995), responsible for a significant decrease in resting systolic blood pressure. Moreover, data from the meta-analysis of Dattilo and KrisEtherton (1992) confirm that weight reduction is associated with a significant decrease in total cholesterol, and a biphasic trend for HDL-cholesterol, that is, a significant decrease during the phase of active weight loss followed by an increase at a stabilized reduced weight, in keeping with our results. The more fascinating results were, however, observed when cardiovascular risk factors were compared between weight losers and weight regainers. This was possible because

80

60

40 Pre-BWR


Post-BWR

Follow-up

Figure 4 Mean (7s.d.) systolic blood pressure (a), total- (b) and HDL-cholesterol (c) and glucose level (d) before and after the 3-week BWR program and after the 1-y follow-up, for weight losers and regainers (n ¼ 16 for both groups). *: Significantly different between losers and regainers (Po0.05); **: Significantly different between losers and regainers (Po0.01).


693 of the similar number of individuals using medications for diabetes (n ¼ 1 in each group) and hypertension (n ¼ 2 and 3, respectively, in weight losers and regainers) in the two groups (n ¼ 0 in both groups for lipid-lowering medications). At baseline, cardiovascular risk factors were not significantly different between the two groups, even if a trend was observed for total cholesterol to be higher (B20 mg/dl) and for HDL-cholesterol to be lower (B5 mg/dl) in regainers. Surprisingly, these between-group differences became significant at the end of the BWR program and more particularly after the follow-up. This would imply that low levels of total cholesterol and high levels of HDL-cholesterol could be associated in some measure to medium-term clinical success in severely obese individuals. This hypothesis remains to be checked carefully in the future. Our study was unique in that it was designed to include energy-restriction diet, ‘tailored’ physical activity, education and psychological counselling in an intensive 3-week treatment that was intended to reduce body weight in severely obese individuals of both genders. On the other hand, one limitation of the current investigation is represented by the lack of control exerted on subjects’ lifestyle after the BWR program. Indeed, participants were only contacted once during the free-living period and consequently they did not have a constant reminder to control nutritional intake and physical activity until these behaviors were incorporated into their lifestyle. Physical activity rating questionnaire allowed to control, at least partially, the achievement of these goals even if no attempt was made to investigate the adherence to nutritional instructions. However, if one considers that obese patients were unsupported during the 49-week follow-up period and therefore subject to a range of factors likely to influence their ability to maintain their reduced body weight, the present results can be considered encouraging. In conclusion, the present investigation demonstrated that a 3-week BWR program resulted in a high rate of clinical success in severely obese individuals at 1-y follow-up (particularly in females), which was associated with increased levels of self-reported physical activity. It was also shown that weight loss had not persistent beneficial effect on physical performance and cardiovascular risk factors after regaining weight. On the other hand, weight reduction, when maintained at medium term, had positive effects on muscle strength and lipid profile of severely obese individuals. Further investigation to establish the long-term effects (ie, at 2–5 y) associated to such an intervention for severely obese subjects is necessary.

Acknowledgements This work is partially supported by Progetto di Ricerca Corrente, Istituto Auxologico Italiano, IRCCS, Milan, Italy. The technical assistance of Mr S Ottolini and M Proietti was greatly appreciated. We are also indebted to Drs A Tibaldi and M Resnik and to Ms F Pera (head nurse) of the 3rd

Division of Metabolic Diseases for their qualified assistance and to Dr A Martin (University of Burgundy) for statistical advice and recommendations. References American College of Sports Medicine (2000): American College of Sports Medicine’s Resource Manual for Guidelines for Exercise Testing and Prescription, ed. JL Roitman. Baltimore: Lippincott, Williams and Wilkins. Arrone LJ, Mackintosh R, Rosenbaum M, Liebel RL & Hirsch J (1995): Autonomic nervous system activity in weight gain and weight loss. Am. J. Physiol. 269, R222–R225. Ballor DL & Poehlman ET (1994): Exercise-training enhances fat-free mass preservation during diet-induced weight loss: a metaanalytical finding. Int. J. Obes. Relat. Metab. Disord. 18, 35–40. Borg GAV (1998): Borg’s Perceived Exertion and Pain Scales, ed. G Borg. Leeds: Human Kinetics. Cox KL, Burke V, Morton AR, Beilin LJ & Puddey IB (2003): The independent and combined effects of 16 weeks of vigorous exercise and energy restriction on body mass and composition in free-living overweight men—a randomized controlled trial. Metabolism 52, 107–115. Dattilo AM & Kris-Etherton PM (1992): Effects of weight reduction on blood lipids and lipoproteins: a meta-analysis. Am. J. Clin. Nutr. 56, 320–328. Dionne I, Despre´s JP, Bouchard C & Tremblay A (1999): Gender difference in the effect of body composition on energy metabolism. Int. J. Obes. Relat. Metab. Disord. 23, 312–319. Donnelly JE, Jakicic JM, Pronk N, Smith BK, Kirk EP, Jacobsen DJ & Washburn R (2003): Is resistance training effective for weight management? Evidence-Based Prev. Med. 1, 21–29. Doucet E, St-Pierre S, Alme´ras N, Imbeault P, Maurie`ge P, Pascot A, Despre´s JP & Tremblay A (2002): Reduction of visceral adipose tissue during weight loss. Eur. J. Clin. Nutr. 56, 297–304. Fox AA, Thompson JL, Butterfield GE, Gylfadottir U, Moynihan S & Spiller G (1996): Effects of diet and exercise on common cardiovascular disease risk factors in moderately obese older women. Am. J. Clin. Nutr. 63, 225–233. Garrow JS & Summerbell CD (1995): Meta-analysis: effect of exercise, with or without dieting, on the body composition of overweight subjects. Eur. J. Clin. Nutr. 49, 1–10. Goodpaster BH, Kelley DE, Wing RR, Meier A & Thaete FL (1999): Effects of weight loss on regional fat distribution and insulin sensitivity in obesity. Diabetes 48, 839–847. Gray DS, Bray GA, Gemayel N & Kaplan K (1989): Effects of obesity on bioelectrical impedance. Am. J. Clin. Nutr. 50, 255–260. Hagan RD, Upton SJ, Wong L & Wittam J (1986): The effect of aerobic conditioning and/or caloric restriction in overweight men and women. Med. Sci. Sports Exerc. 18, 87–94. Hills AP & Byrne NM (1998): Exercise prescription for weight management. Proc. Nutr. Soc. 57, 93–103. Hoiberg A, Berard S, Watten RH & Caine C (1984): Correlates of weight loss in treatment and at follow-up. Int. J. Obes. Relat. Metab. Disord. 8, 457–465. Jackson AS, Blair SN, Mahar SN, Wier MT, Ross RM & Stuteville JE (1990): Prediction of functional aerobic capacity without exercise testing. Med. Sci. Sports Exerc. 22, 863–870. Jakicic JM (2003): Exercise in the treatment of obesity. Endocrinol. Metab. Clin. N. Am. 32, 967–980. Kraemer WJ, Volek JS, Clark KL, Gordon SE, Puhl SM, Koziris LP, McBride JM, Triplett-McBride NT, Putukian M, Newton RU, Ha¨kkinen K, Bush JA & Sebastianelli WJ (1999): Influence of exercise training on physiological and performance changes with weight loss in men. Med. Sci. Sports Exerc. 31, 1320–1329. Kuczmarski RJ, Flegal KM, Campbell SM & Johnson CL (1994): Increasing prevalence of overweight among US adults. The National Health and Nutrition Examination Surveys, 1960 to 1991. J. Am. Med. Assoc. 22, 205–211.



694 Lukaski HC, Bolonchuk WW, Hall CB & Siders WA (1986): Validation of tetrapolar bioelectrical measurements to assess human body composition. J. Appl. Physiol. 60, 1327–1332. Margaria R, Aghemo P & Rovelli E (1966): Measurement of muscular power (anaerobic) in man. J. Appl. Physiol. 21, 1662–1664. Maxwell MH, Waks AU, Schroth PC, Karam M & Dornfeld LP (1982): Error in blood-pressure measurements due to incorrect cuff size in obese patients. Lancet 2, 33–36. Mayhew JL, Prinster JL, Ware JS, Zimmer DL, Arabas JR & Bemben MG (1995): Muscular endurance repetitions to predict bench press strength in men of different training levels. J. Sport Med. Phys. Fitness 35, 108–113. McGuire MT, Wing RR, Klem ML, Lang W & Hill JO (1999): What predicts weight regain in a group of successful weight losers? J. Consult. Clin. Psychol. 67, 177–185. Miller WC, Koceja DM & Hamilton EJ (1997): A meta-analysis of the past 25 years of weight loss research using diet, exercise or diet plus exercise intervention. Int. J. Obes. Relat. Metab. Disord. 21, 941–947. Neter JE, Stam BE, Kok FJ, Grobbee DE & Geleijnse JM (2003): Influence of weight reduction on blood pressure. A meta-analysis of randomized controlled trials. Hypertension 42, 878–884. Pronk NP & Wing RR (1994): Physical activity and long-term maintenance of weight loss. Obes. Res. 2, 587–599. Sartorio A, Lafortuna CL, Agosti F, Proietti M & Maffiuletti NA (2004a): Elderly obese women display the greatest improvement in stair climbing performance after a 3-wk body mass reduction program. Int. J. Obes. Relat. Metab. Disord. 28, 1097–1104.


Sartorio A, Lafortuna CL, Conte G, Faglia G & Narici MV (2001b): Changes in motor control and performance after a short term body mass reduction program in obese subjects. J. Endocrinol. Invest. 24, 393–398. Sartorio A, Lafortuna CL, Massarini M & Galvani C (2003): Effects of different training protocols on exercise performance during a short-term body weight reduction programme in severely obese patients. Eating Weight Disord. 8, 36–43. Sartorio A, Lafortuna CL, Vangeli V, Tavani A, Bosetti C & La Vecchia C (2001a): Short-term changes of cardiovascular risk factors after a non-pharmacological body weight reduction program. Eur. J. Clin. Nutr. 55, 865–869. Sartorio A, Proietti M, Marinone PG, Agosti F, Adorni F & Lafortuna CL (2004b): Influence of gender, age and BMI on lower limb muscular power output in a large population of obese men and women. Int. J. Obes. Relat. Metab. Disord. 28, 91–98. Schwingshandl J & Borkenstein M (1995): Changes in lean body mass in obese children during a weight reduction program: effect on short term and long term outcome. Int. J. Obes. Relat. Metab. Disord. 19, 752–755. Svendsen OL, Hassager C & Christiansen C (1994): Six months’ follow-up on exercise added to a short-term diet in overweight postmenopausal women—effects on body composition, resting metabolic rate, cardiovascular risk factors and bone. Int. J. Obes. Relat. Metab. Disord. 18, 692–698. Zachwieja JJ (1996): Exercise as treatment for obesity. Endocrinol. Metab. Clin. N. Am. 25, 965–988.