Weight loss and metabolic profiles in obese individuals using two different approaches

Abstract

Intentional weight loss with a reduction in adipose tissue is associated with an improvement in medical complications linked to obesity. The aim of this study was to compare the effect of two different weight loss diets on obese individuals (BMI 30–40 kg m⁻²) for improvements in anthropometric measurements and blood biomarkers. Study 1 comprised a low energy diet (LED) of 5000–6000 kJ d⁻¹ for a slow but steady weight loss over 12 weeks. Study 2 comprised a very low energy diet (VLED) of 3000 kJ d⁻¹ using meal replacements for rapid weight loss over a shorter period of 4 weeks followed by 10 weeks of weight stabilisation to prevent rebound after rapid weight loss. Nutrition information sessions were given to both groups. Fasting blood samples, anthropometric measurements and 3-day food diaries were collected at baseline and again at completion of weight loss, at 12 weeks for LED group and 4 weeks for VLED group. Mean weight loss in the LED group (n = 18) was −3.17 kg (−3.7%) compared to a −6.54 kg (−7%) loss in the VLED group (n = 14) (p < 0.001). The VLED group experienced significantly greater reductions in fat mass, −13.9% compared to −8.9% for the LED group (p < 0.05). Significantly greater reductions in blood glucose (p < 0.05), cholesterol and LDL-C (p < 0.01 for both), and waist circumference (p ≤ 0.05) were noted in VLED compared to those for LED diet. Short-term, rapid weight loss produced the desired ≥5% weight loss suggested to substantially reduce metabolic abnormalities associated with obesity, and to reduce health risks.

---

1. Introduction

Obesity is a chronic metabolic disorder linked to the development of cardiovascular disease (CVD),¹ and associated with increased risk of morbidity and mortality,² including the development of insulin resistance and type 2 diabetes mellitus (T2DM), disturbances in blood lipids and dyslipidemia,³ and increased levels of inflammatory biomarkers.⁴

Intentional weight loss with a reduction in adipose tissue, particularly central or visceral adipose tissue, improves many of these medical complications by decreasing levels of inflammatory biomarkers^5–7 and reducing metabolic abnormalities.⁸ It has been suggested that a weight loss of between 5 and 10% is needed to substantially reduce these risks,^9,10 and a number of strategies are available to assist with weight loss. Popular weight loss programs focus on reducing energy intake, either by decreasing total fat intake¹¹ (e.g. the Ornish diet), reducing fat intake and increasing consumption of low GI foods¹² (e.g. the Low GI Diet), extreme carbohydrate restriction (e.g. the Atkins diet) or reducing carbohydrate and increasing protein intake¹³ (e.g. the Zone diet). With the exception of low fat diets, the resulting weight loss with these diets is less than the energy level of the diets would indicate,¹⁴ most likely as a consequence of poor dietary compliance,^15,16 but it is unrealistic to expect that dieters can restrict their food intake in the face of hunger. The use of commercial food replacements (MRs) for a very low energy diet (VLED) reduces feelings of hunger and improves compliance^17,18 resulting in significantly greater weight loss than with low energy diets (LEDs).¹⁹

The aim of this study was to compare two different dietary weight loss strategies, a rapid, short-term VLED using MRs and a slow, longer-term healthy eating LED, for weight and fat loss, improvements in blood lipid levels and improvements in levels of inflammatory biomarkers, namely tumor necrosis factor-α (TNF-α), Interleukin-6 (IL-6), C-reactive protein (CRP), leptin and adiponectin.

2. Method

2.1 Participants

Both male and female participants, aged 18–55 years and with a BMI of between 30–40 kg m⁻², were recruited from the university campus and the general public in Newcastle, Australia. Exclusion criteria were diagnosed diabetes mellitus, a chronic inflammatory condition, already following a kilojoule-restricted diet, or women who were pregnant or lactating. Forty healthy, obese individuals met the criteria to participate in one of two studies which were conducted according to the guidelines laid down in the Declaration of Helsinki and approved by the Human Research Ethics Committee of the University of Newcastle, Australia. Written, informed consent was obtained from participants prior to commencement.

2.2 Study design

Two weight loss studies were designed to target a weight loss of 5% of initial body weight using two very different dietary weight loss strategies. This was a non-randomised intervention. Initially participants were recruited to take part in Study 1 and this was then followed using the same process to recruit participants for Study 2. Study 1 comprised of a slow and steady weight loss program of 5000 kJ d⁻¹ for females and 6000 kJ d⁻¹ for males. To accommodate the slow weight loss, the study took place over 12 weeks. This healthy eating LED was based on the Australian Guide to Healthy Eating (AGHE)²⁰ and enabled participants to eat a variety of healthy foods, albeit a reduced amount. Study 2 comprised a fast VLED of 3000 kJ d⁻¹ using MRs. Because of the very low energy intake and rapid weight loss, the study was conducted over a shorter time frame of 4 weeks. For the first 2 weeks, participants received Optifast® bars and shakes to replace meals according to the Intensive Phase of the Optifast Very Low Calorie Diet Program (Novartis, Australia), supplemented daily with raw and cooked vegetables and 2 L of drinking water. In weeks three and four MRs were gradually phased out and healthy low energy meals, based on the AGHE, were phased in so that by the end of week 4 participants no longer consumed MRs. After rapid weight loss, the potential is there for weight rebound unless new manageable and effective dietary habits are established. Thus, at the end of the weight loss phase, the VLED participants were observed for a further 10 weeks while they established healthy eating behaviours which stabilised their weight.

Both of the studies commenced with individual nutrition education sessions, all of which were conducted by the same educator to ensure uniformity of information presented. The topics covered included the energy density of foods, understanding and using food labels, appropriate portion sizes and the number of portions to be consumed daily from the different food groups. Both studies were conducted during the summer season to ensure a similar availability of food for both groups.

Prior to commencing the weight loss program, participants recorded their food intake over three days (2 weekdays and 1 day at the weekend). Participants were shown how to use weights given on food labels and handy measures to record their dietary intake which was then analysed using the dietary software program, Foodworks® Professional 2009, version 6 (Xyris Software (Australia) Pty Ltd). Participants in LED group again completed a 3-day food diary in the final week of weight loss; the VLED group again completed a 3-day food diary after 10 weeks of dietary stabilisation to measure dietary compliance.

2.3 Anthropometric measurements

On the first day of the weight loss study, anthropometric measurements were taken in the morning following a 10 h overnight fast with no alcohol consumption, with all participants dressed in light clothing and without shoes. Standing height was measured to the nearest 0.1 cm using a stadiometer. Body weight was measured to the nearest 0.1 kg using a calibrated balance beam scale (PCS Measurement, NSW, Australia). Waist circumference (WC) was measured at the mid-point between the lowest rib and the top of the hipbone; the hip measurement was taken at the fullest point of the hip, as viewed from the side. Fat mass (FM) and fat-free mass (FFM) were measured using single-frequency bioelectrical impedance (Maltron International, Rayleigh, Essex, UK). Under identical conditions, the anthropometric measurements were repeated at the end of the study.

2.4 Biochemical analyses

Fasting (≥10 h) blood samples were collected into tubes, EDTA, Lithium Heparin and Sodium Fluoride, by venipuncture at baseline and again at the end of the weight loss phase. The samples were prepared by centrifuging (Heraeus Biofuge Stratos) for 10 min at 3000 g at 4 °C. Plasma samples from the EDTA tubes were collected and stored at −80 °C awaiting further analysis. The Lithium Heparin and Sodium Fluoride tubes were taken to Hunter Area Pathology Services (Newcastle) for plasma analysis for blood lipids and blood glucose, respectively.

High sensitivity enzyme-linked immunosorbent assay (ELISA) kits (R&D Systems, Minneapolis, MN, USA) were used to determine levels of tumor necrosis factor-α (hs-TNF-α) and interleukin-6 (hs-IL-6). Minimal detectable concentration of the kits was 0.106 pg mL⁻¹ and 0.039 pg mL⁻¹ respectively with an intra- and inter-assay coefficients of variation (CV) of <9%. Analysis of high-sensitivity C-reactive protein (hs-CRP) was conducted using an immunoturbidimetric method (Hunter Area Pathology Service, Newcastle, NSW, Australia) with a minimal detection level of 0.15 mg L⁻¹. Plasma leptin levels were quantified with a commercial double-antibody enzyme immunometric assay (EIA) (Cayman Chemical Company, Ann Arbor, MI, USA) with a detection limit of 1.0 ng mL⁻¹ and inter- and intra-assay CVs <9%. Adiponectin levels were determined using an ELISA kit (SPI-bio, Montigney le Bretonneux, France). Intra- and inter-assay CV were 6.4% and 7.3% respectively with a detection range of 0.1–10.0 μg mL⁻¹.

2.5 Statistical analyses

Data are presented as mean values and standard deviations, with p ≤ 0.05 indicating significance. A power calculation for a 5% reduction for the primary outcome, weight loss, at the 0.05 level of significance at 80% power indicates that a sample of 13 subjects per group would be required to complete the study. Allowing for a 20% drop-out rate, recruitment of 16 subjects per group is needed. One-way ANOVA was used to test group mean differences, and within group changes from baseline were determined using paired samples t-tests. Pearson product-normal correlations (r) were used to show relationships. All statistical analyses were carried out with SPSS software (version 17.0, SPSS Inc., Chicago, IL, USA).

3. Results

Forty healthy, obese participants commenced the study, with 20 participants in each of the two groups. Thirty two of these participants completed one of the two weight loss diets, 18 participants (6 males and 12 females) in the LED group and 14 participants (3 males and 11 females) in the VLED group. Although the completion rates for each group were not the same, this difference in the number of participants in each group was not significant (p = 0.85). At the commencement of the two studies, daily energy intake for each group was calculated and was found to be similar, 8022 kJ d⁻¹ (8127 kJ d⁻¹ males and 7966 kJ d⁻¹ females) for LED group and 7846 kJ d⁻¹ (9684 kJ d⁻¹ males and 7345 kJ d⁻¹ females) for VLED group (p = 0.76). At baseline there were no significant differences between the two groups for age (p = 0.64), gender (p = 0.47), or blood biomarkers, apart from HDL levels which were higher for VLED group 1.37 ± 0.28 mmol L⁻¹, compared to LED group 1.16 ± 0.24 mmol L⁻¹ (p = 0.03). There were also no significant differences in baseline anthropometric measurements, weight (p = 0.97), height (p = 0.74), BMI (p = 0.68), FM (p = 0.53), FFM (p = 0.81), WC (p = 0.99) or hip (p = 0.60).

At the end of the weight loss period, that is after 12 weeks for the LED group and after 4 weeks for the VLED group, there was a significant reduction in weight, BMI, FM and hip circumference within both groups (p < 0.001), and WC (p < 0.001) and (p < 0.05) for VLED group and LED group, respectively (Table 1). By the end of 4 weeks the VLED group reduced weight by approximately twice as much as the LED group did by the end of 12 weeks, −6.54 kg (−7%) and −3.17 kg (−3.7%) respectively (p < 0.001). Individually, only two (out of 14) VLED participants reduced their weight by less than 5%, while 15 (out of 18) LED participants had a weight reduction of less than 5%. The reduction in FM was significantly greater for VLED group, −13.9% (−5.54 kg) compared to LED group, −8.9% (−3.43 kg) (p < 0.05), which corresponded with the significantly greater reduction in WC for VLED group (p = 0.05) (Table 1).

Table 1 Baseline values and changes from baseline for anthropometric measurements^a

	LED (n = 18)			VLED (n = 14)			P value
	BL	PI	Δ	BL	PI	Δ
a Mean value ± standard deviation. Significant changes within groups: * p < 0.05; ** p < 0.01; *** p < 0.001. P shows the significant differences between LED and VLED.
Age/y	41 ± 3.0			42 ± 2.0			0.64
Weight/kg	93.95 ± 16.40	90.78 ± 16.12	−3.17 ± 3.09***	94.16 ± 16.64	87.63 ± 15.41	−6.54 ± 2.08***	0.001
BMI/kg m^−2	32.53 ± 3.58	31.42 ± 3.70	−1.11 ± 0.99***	33.04 ± 3.17	30.77 ± 3.16	−2.27 ± 0.16***	0.001
FM/kg	38.98 ± 7.49	35.55 ± 8.06	−3.43 ± 3.51***	40.69 ± 7.65	35.15 ± 7.28	−5.54 ± 1.34***	0.04
FFM/kg	54.40 ± 10.43	54.69 ± 10.02	0.29 ± 1.77	53.50 ± 10.82	52.50 ± 10.53	−1.00 ± 1.29	0.03
WC/cm	103.11 ± 12.11	100.86 ± 11.65	−2.25 ± 3.93*	103.11 ± 11.19	98.14 ± 11.10	−4.96 ± 3.54***	0.05
Hip/cm	119.67 ± 9.78	117.14 ± 10.43	−2.53 ± 2.46***	117.99 ± 7.49	113.75 ± 6.47	−4.24 ± 2.38***	0.058
Waist : hip	0.86 ± 0.08	0.86 ± 0.08	0.0 ± 0.03	0.87 ± 0.08	0.86 ± 0.08	−0.01 ± 0.02	0.175

---

With the greater weight reduction there was a significantly greater reduction in TC for the VLED group than for the LED group, −0.86 ± 0.84 mmol L⁻¹ compared to −0.10 ± 0.56 mmol L⁻¹, respectively (p = 0.005); also a greater reduction in LDL-C for the VLED group than for the LED group, −0.54 ± 0.72 mmol L⁻¹ compared to −0.06 ± 0.47 mmol L⁻¹, respectively (p = 0.008). The reduction in leptin levels was significantly greater for the VLED group compared to the LED group, −19.67 ± 11.82 mmol L⁻¹ and −0.26 ± 13.82 mmol L⁻¹, respectively (p < 0.001). There was also a significant difference in the change to glucose levels with a decrease of −0.05 ± 0.33 mmol L⁻¹ for the VLED group and an increase of 0.31 ± 0.50 mmol L⁻¹ for the LED group (p = 0.03) (Table 2). Changes to inflammatory markers, which are shown in Table 2, were not significantly different between the two groups.

Table 2 Baseline values and changes from baseline for blood biomarkers^a

	LED (n = 18)			VLED (n = 14)			P value
	BL	PI	Δ	BL	PI	Δ
a Mean values ± standard deviation. Significant changes within groups: ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001. P shows the significant differences between LED and VLED.
	(week 12)			(week 4)
Trigs/mmol L^−1	1.33 ± 0.71	1.07 ± 0.40	−0.25 ± 0.59	1.28 ± 0.51	1.02 ± 0.48	−0.26 ± 0.49	0.98
TC/mmol L^−1	4.84 ± 0.93	4.74 ± 0.73	−0.10 ± 0.56	5.4 ± 0.77	4.54 ± 1.18	−0.86 ± 0.84**	0.005
HDL-C/mmol L^−1	1.16 ± 0.24	1.12 ± 0.27	−0.04 ± 0.18	1.37 ± 0.28	1.19 ± 0.20	−0.17 ± 0.17	0.11
LDL-C/mmol L^−1	3.07 ± 0.81	3.13 ± 0.66	0.06 ± 0.47	3.43 ± 0.79	2.90 ± 0.96	−0.54 ± 0.72*	0.008
Glucose/mmol L^−1	4.71 ± 0.51	5.02 ± 0.57	0.31 ± 0.50	5.01 ± 0.47	4.96 ± 0.56	−0.05 ± 0.33	0.027
Leptin/ng mL^−1	35.21 ± 13.60	34.95 ± 19.16	−0.26 ± 13.82	32.18 ± 14.42	12.51 ± 8.77	−19.67 ± 11.82***	0.001
Adiponectin/mg mL^−1	11.92 ± 6.32	13.83 ± 7.38	1.91 ± 6.48	10.21 ± 5.76	10.41 ± 5.11	0.20 ± 2.73	0.364
hs-CRP/mg mL^−1	5.95 ± 6.46	4.79 ± 5.43	−1.16 ± 3.72	5.79 ± 5.47	5.48 ± 6.38	−0.30 ± 5.37	0.598
hs-IL-6/pg mL^−1	2.75 ± 2.39	2.95 ± 3.69	0.19 ± 1.58	2.07 ± 0.95	1.60 ± 0.77	−0.48 ± 0.78*	0.157
hs-TNF-α/pg mL^−1	1.34 ± 1.13	1.01 ± 1.11	−0.33 ± 0.80	1.13 ± 0.98	0.93 ± 0.89	−0.20 ± 1.18	0.705

---

Following weight loss, 12 of the 14 participants in the VLED group continued for 10 weeks of weight stabilisation using whole foods after relying on MRs to assist their weight loss. During this time, they lost a small amount of weight −1.57 kg ± 3.70, which was accompanied by small, non-significant reductions in FM, FFM, WC and hip circumference, which are shown in Table 3. There was a significant rebound in levels of triglycerides, TC (p < 0.05 for both) and leptin (p < 0.01). There was also an increase in levels of LDL-C and hs-CRP, but the differences were not significant. The level of HDL-C, which had decreased with weight loss, increased significantly (p < 0.01) and was now higher than the baseline level had been. Levels of adiponectin increased and IL-6 and TNFα decreased but the changes were not significant.

Table 3 Changes from the end of weight loss to the end of weight stabilisation for VLED^a

	VLED (n = 12)
	End weight loss	End weight stabilisation	Δ	P value
a Mean values ± standard deviation.
Weight/kg	83.55 ± 10.01	81.98 ± 10.28	−1.57 ± 3.70	0.170
BMI/kg m^−2	30.46 ± 3.11	29.85 ± 3.37	−0.61 ± 1.40	0.161
FM/kg	33.45 ± 5.90	32.46 ± 5.68	−0.99 ± 3.45	0.340
FFM/kg	50.13 ± 8.00	49.67 ± 7.95	−0.46 ± 0.77	0.063
WC/cm	95.83 ± 9.11	93.54 ± 10.69	−2.29 ± 4.35	0.095
Hip/cm	112.42 ± 5.95	110.70 ± 6.78	−1.71 ± 3.70	0.138
Trigs/mmol L^−1	1.06 ± 0.52	1.20 ± 0.52	0.14 ± 0.21	0.043
TC/mmol L^−1	4.73 ± 1.17	5.27 ± 1.15	0.55 ± 0.76	0.028
HDL-C/mmol L^−1	1.23 ± 0.18	1.44 ± 0.28	0.21 ± 0.21	0.005
LDL-C/mmol L^−1	3.03 ± 0.97	3.29 ± 1.05	0.26 ± 0.71	0.230
Glucose/mmol L^−1	4.97 ± 0.57	4.88 ± 1.06	−0.08 ± 1.06	0.791
Leptin/ng mL^−1	12.22 ± 8.85	20.92 ± 9.84	8.69 ± 7.69	0.002
Adiponectin/mg mL^−1	10.98 ± 5.08	12.00 ± 8.27	1.02 ± 4.24	0.423
hs-CRP/mg mL^−1	4.51 ± 4.40	5.01 ± 4.14	0.50 ± 2.81	0.550
hs-IL-6/pg mL^−1	1.48 ± 0.73	1.46 ± 0.63	−0.03 ± 0.69	0.897
hs-TNF-α/pg mL^−1	0.74 ± 0.56	0.31 ± 0.42	−0.42 ± 0.70	0.062

---

4. Discussion

There are differences of opinion on the optimal diet for weight loss and weight management. Registered dietitians typically recommend portion control to reduce energy intake as a weight loss strategy.²¹ All of the participants in this study were instructed on portion control for weight management, and it was also the primary strategy used for weight loss with the LED group. Commencing the study with a mean baseline energy intake of 8022 kJ d⁻¹ (7966 kJ d⁻¹ females and 8127 kJ d⁻¹ males), the target energy intake for the LED group was 5000 kJ d⁻¹ for females and 6000 kJ d⁻¹ for males. Reducing energy intake by 2000 kJ d⁻¹ should have resulted in a weight loss of 0.5 kg week⁻¹¹⁵ with an anticipated loss of 6 kg over the 12 weeks, but this did not happen. The mean weight loss was 3.17 kg, with some participants achieving a much better weight loss than others; for example, a reduction of 11.5 kg (10%) and 9.5 kg (11.6%) for two participants while two other participants did not lose any weight at all. The participants were responsible for choosing their food and limiting the amount that they consumed, which would not have been easy when they had to manage feelings of hunger. Dietary compliance was challenging, as evidenced by the mean daily intake of 7552 kJ d⁻¹ for males and 6701 kJ d⁻¹ for females, recorded in the 3-day food diaries completed at the end of weight loss.

The MRs for the VLED group initially removed the responsibility of managing food intake from a variety of foods, and this appears to have facilitated dietary compliance. A study comparing two weight loss groups, LED and VLED, found that increasing the variety of food choice was associated with greater food consumption and that a less varied diet resulted in the greatest decrease in food cravings, possibly mediated by reduced hunger.²² The study found that the LED group (n = 19) reduced weight by 6.51 kg over 12 weeks while consuming 5024 kJ d⁻¹ and the VLED (n = 22) reduced weight by 11.97 kg over 6 weeks while consuming 3349 kJ d⁻¹ with MRs²² showing greater weight reduction in a shorter period for VLED, which is similar to the findings in our study. Other studies using MRs for VLED of between 2520–3570 kJ d⁻¹ for weight loss have also reported considerable weight loss over short periods of time, for example Haugaard (2007) reported a mean weight reduction of 9.3 kg (8.8%) over 8 weeks,²³ Harder (2004) reported a mean weight loss of 10.9 kg (11%) over 8 weeks,²⁴ while Clement (2004) reported a mean weight loss of 6 kg (6.4%) over 4 weeks.²⁵ Using MRs has been criticized because they obviate the need to choose from a variety of foods and to control portions when replacing meals and do not encourage the development of improved eating habits.²⁶ Concerns have also have also been raised about fast and/or large decreases in weight and the potential for weight regain. However, a study by Nackers et al. (2010) found that losing weight slowly does not lead to greater long term weight loss and smaller regain,²⁷ although a large decrease in weight, considered to be >30%, has been associated with a greater weight regain.²⁸ The weight reduction in the VLED group in this study was 7% (−6.54 kg) over 4 weeks which was twice the amount lost by the LED group over 12 weeks, but to address these concerns, the VLED group did not continue with their weight loss unchecked. To minimize the potential for weight rebound and to fully implement the development of healthy eating behaviours after weight loss, the participants followed a healthy balanced diet for 10 weeks to stabilise their weight and to not intentionally continue to reduce weight. However, while they were trying to establish the optimal energy intake for maintenance, there was a further mean weight reduction of 1.85% (−1.57 kg).

The greater weight loss with the VLED group was unsurprising since energy intake of that group was 60% of the energy intake for the LED group. However the reduced energy intake for the LED group did occur over a longer time frame, that is 12 weeks compared to 4 weeks for the VLED group, so they had a longer time in which to lose weight. Despite the significant differences between the two groups in the amount of weight lost, the reduction in triglyceride levels was very similar for both groups. With weight loss, levels of LDL-C and glucose increased for the LED group while, for the VLED group, levels of LDL-C and glucose reduced. There was a greater reduction in TC for the VLED group relative to weight lost (0.13 mmol L⁻¹ per kg of weight lost) compared to the LED group (0.03 mmol L⁻¹ per kg of weight lost) but the difference was not significant.

HDL-C also decreased and again the reduction was greater for the VLED group (0.03 mmol L⁻¹ per kg of weight lost) compared to the LED group (0.01 mmol L⁻¹ per kg of weight lost) but the difference between the two groups was not significant. While there are clinical benefits of a reduction in levels of TC and LDL-C, it is well recognised that HDL-C has a cardioprotective effect²⁹ and a reduction is undesirable. Previous studies have reported different effects of weight loss on HDL-C, some showing small increases, others a small decrease and some no change at all.³⁰ It would appear that HDL-C levels are reduced during active weight loss and levels subsequently increase when a stabilized, reduced weight is attained.^19,31 Dattilo (1999) suggests that this occurs because lipoprotein lipase decreases during active weight loss, reducing triglyceride rich lipoprotein synthesis which impairs VLDL-C catabolism and the transfer of lipids to HDL-C, thereby reducing HDL-C concentrations. When weight then stabilises at a reduced level, lipoprotein lipase increases as does hydrolysis of VLDL-C and the transfer of lipids to HDL-C resumes.³¹ With the 10-week weight stabilisation/maintenance phase in the VLED group in our study, there was a significant increase in HDL-C to above baseline levels. Triglyceride, TC and LDL-C levels also increased during the 10-week weight stabilisation phase but they remained lower than baseline levels. The reason for the increases could be attributed to the change from MRs to resuming meal consumption which contains higher levels of saturated fats.

A reduction in adipose tissue with weight loss leads to a decrease in circulation levels of inflammatory biomarkers³² and decreased metabolic abnormalities. Although there was a significantly greater reduction in FM in the VLED group, the increase in adiponectin levels, and the reduction in hs-CRP and hs-TNF-α levels, were greater in the LED group but not significantly. Levels of hs-IL-6 reduced in the VLED group, but not significantly. With 10 weeks of dietary stabilisation for the VLED group, there was a further increase in levels of adiponectin, and further decreases of hs-IL-6 and hs-TNF-α, with total changes now exceeding those in the LED group. It is possible that time, as well as weight loss, is a contributing factor to changes in inflammatory biomarkers.

There was a significant 60.6% decrease in leptin after 4 weeks of weight loss in the VLED group (p < 0.001). The magnitude of this reduction is similar to those reported in a number of studies. Wisse et al., observed a 61% reduction in fasting leptin after 14 days of energy restriction of 1900 kJ d⁻¹, a 52% reduction after 14 days of a low energy balanced-deficit diet of 5500 kJ d⁻¹ and a 76% reduction in leptin after 14 days of fasting.³³ Mars et al., observed a 39.4% reduction in fasting leptin after a 4-day, 65% energy restricted diet.³⁴ Dubuc et al., reported a 35–65% reduction in fasting leptin after 7 days of an energy restricted MR diet of either 2646 kJ d⁻¹ for females or 3528 kJ d⁻¹ for males.³⁵ These authors contend that the decrease in leptin does not correspond to the changes in body fat.^33,35 It has been suggested that decreasing leptin levels are associated with self-perceived increases in hunger,^34,36 a mechanism in place to regulate food intake to prevent starvation. Controlling that hunger response can be difficult and one of the reasons for the high failure rate with weight loss diets. However, after the 10 weeks of stabilisation, leptin levels had significantly increased (p < 0.01) although they did not return to baseline levels.

In summary, the very low energy, rapid weight loss diet produced the desired ≥5% weight loss that had been suggested for a reduction of metabolic abnormalities, and for the 4-week period an improvement in blood biomarkers was observed. The concerns for weight rebound following rapid weight loss have been addressed in this study with a short rapid weight loss phase followed by a recovery/stabilisation phase to reinforce balanced eating behaviours. This very low energy diet, followed by a recovery cycle to stabilise weight, can be followed until the desired weight is achieved, at which point the weight stabilisation diet can be followed for weight maintenance.

5. Limitations

A limitation of this study is the small sample size with its low completion rate. More studies with a larger sample size, and which extend the dietary weight loss periods, are needed to further investigate these findings.

6. Acknowledgements

Nestle Nutrition, Australia, supplied the Optifast® bars and shakes for meal replacements.

References

1. P. Poirier, T. D. Giles, G. A. Bray, Y. Hong, J. S. Stern, F. X. Pi-Sunyer and R. H. Eckel, Circulation, 2006, 113, 898–918.
2. L. J. Aronne, Obesity Research, 2002, 10, 105S–115S.
3. G. Fruhbeck, in Adipose Tissue and Adipokines in Health and Disease, ed. G. Fantuzzi and T. Mazzone, Human Press Inc., Totowa, 2007, pp. 63–77.
4. P. Trayhurn and I. S. Wood, Biochem. Soc. Trans., 2005, 33, 1078–1081.
5. J. D. Krebs, L. M. Browning, N. K. McLean, J. L. Rothwell, G. D. Mishra, C. S. Moore and S. A. Jebb, Int. J. Obes., 2006, 30, 1535–1544.
6. J. P. Bastard, C. Jardel, E. Bruckert, P. Blondy, J. Capeau, M. Laville, H. Vidal and B. Hainque, J. Clin. Endocrinol. Metab., 2000, 85, 3338–3342.
7. P. Dandona, R. Weinstock, K. Thusu, E. Abdel-Rahman, A. Aljada and T. Wadden, J. Clin. Endocrinol. Metab., 1998, 83, 2907–2910.
8. D. F. Williamson, Int. J. Obes. Relat. Metab. Disord., 1997, 21, S14–S19.
9. G. Blackburn, Obes. Res., 1995, 3, 211s–216s.
10. D. J. Goldstein, Int. J. Obes. Relat. Metab. Disord., 1992, 16, 397–415.
11. B. V. Howard, J. E. Manson, M. L. Stefanick, S. A. Beresford, G. Frank, B. Jones, R. J. Rodabough, L. Snetselaar, C. Thomson, L. Tinker, M. Vitolins and R. Prentice, J. Am. Med. Assoc., 2006, 295, 39–49.
12. J. C. Brand-Miller, S. H. Holt, D. B. Pawlak and J. McMillan, Am. J. Clin. Nutr., 2002, 76, 281S–285S.
13. P. M. Clifton, J. B. Keogh and M. Noakes, Am. J. Clin. Nutr., 2008, 87, 23–29.
14. A. Avenell, T. J. Brown, M. A. McGee, M. K. Campbell, A. M. Grant, J. Broom, R. T. Jung and W. C. Smith, J. Hum. Nutr. Diet., 2004, 17, 317–335.
15. S. B. Heymsfield, J. B. Harp, M. L. Reitman, J. W. Beetsch, D. A. Schoeller, N. Erondu and A. Pietrobelli, Am. J. Clin. Nutr., 2007, 85, 346–354.
16. M. R. Lowe, Obesity Research, 2003, 11, 44S–59S.
17. G. Egger, Med. J. Aust., 2006, 184, 52–53.
18. D. Q. Rothacker and S. Watemberg, Int. J. Food Sci. Nutr., 2004, 55, 223–226.
19. H. H. Ditschuneit, M. Flechtner-Mors, T. D. Johnson and G. Adler, Am. J. Clin. Nutr., 1999, 69, 198–204.
20. E. Kellet, A. E. Smith and Y. Schmerlaib, Australian Guide to Healthy Eating, Australian Government Department of Health and Ageing, Canberra, 1998, pp. 1–22.
21. H. M. Seagle, G. W. Strain, A. Makris and R. S. Reeves, J. Am. Diet. Assoc., 2009, 109, 330–346.
22. C. K. Martin, P. M. O'Neil and L. Pawlow, Obesity Research, 2006, 14, 115–121.
23. S. B. Haugaard, A. Vaag, C. E. Hoy and S. Madsbad, Obesity, 2007, 15, 117–125.
24. H. Harder, B. Dinesen and A. Astrup, Int. J. Obes. Relat. Metab. Disord., 2004, 28, 180–182.
25. K. Clement, N. Viguerie, C. Poitou, C. Carette, V. Pelloux, C. A. Curat, A. Sicard, S. Rome, A. Benis, J. D. Zucker, H. Vidal, M. Laville, G. S. Barsh, A. Basdevant, V. Stich, R. Cancello and D. Langin, FASEB J., 2004, 18, 1657–1669.
26. D. Q. Rothacker, B. A. Staniszewski and P. K. Ellis, J. Am. Diet. Assoc., 2001, 101, 345–347.
27. L. M. Nackers, K. M. Ross and M. G. Perri, Int. J. Behav. Med., 2010, 17, 161–167.
28. M. T. McGuire, R. R. Wing, M. L. Klem, W. Lang and J. O. Hill, J. Consult. Clin. Psychol., 1999, 67, 177–185.
29. W. E. Boden, Am. J. Cardiol., 2000, 86, 19–22.
30. A. Poobalan, L. Aucott, W. C. Smith, A. Avenell, R. Jung, J. Broom and A. M. Grant, Obes. Rev., 2004, 5, 43–50.
31. A. M. Dattilo and P. M. Kris-Etherton, Am. J. Clin. Nutr., 1992, 56, 320–328.
32. A. H. Berg and P. E. Scherer, Circ. Res., 2005, 96, 939–963.
33. B. E. Wisse, L. A. Campfield, E. B. Mariss, J. A. Morais, R. Tenenbaum and R. Gougeion, Am. J. Clin. Nutr., 1999, 70, 321–330.
34. M. Mars, C. de Graaf, C. P. de Groot, C. T. van Rossum and F. J. Kok, Int. J. Obes., 2006, 30, 122–128.
35. G. R. Dubuc, S. D. Phinney, J. S. Stern and P. J. Havel, Metab., Clin. Exp., 1998, 47, 429–434.
36. N. L. Keim, J. S. Stern and P. J. Havel, Am. J. Clin. Nutr., 1998, 68, 794–801.

---