Abstract

Background:

As formula diets have demonstrated to be effective in reducing weight, we hypothesised that in patients with overweight or obesity and accompanied cardiovascular risk factors, combining a liquid formula diet with a lifestyle intervention is superior in reducing weight and improving cardiovascular risk factors than lifestyle intervention alone.

Methods:

In this multicenter RCT 463 participants with overweight or obesity (BMI: 27–35 kg/m²; at least one additional comorbidity of the metabolic syndrome) were randomised (1:2) into either a control group with lifestyle intervention only (CON, n = 155) or a lifestyle intervention group including a liquid meal replacement (INT, n = 308). Both groups used telemonitoring devices (scales and pedometers), received information on healthy diet and were instructed to increase physical activity. Telemonitoring devices automatically transferred data into a personalised online portal and acquired data were discussed. INT obtained a liquid meal replacement substituting three meals/day (~1200 kcal) within the first week. During weeks 2–4, participants replaced two meals/day and during weeks 5–26 only one meal/day was substituted (1300–1500 kcal/day). Follow-up was conducted after 52 weeks. Intention-to-treat analyses were performed. Primary outcome was weight change. Secondary outcomes comprised changes in cardiometabolic risk factors including body composition and laboratory parameters.

Results:

From the starting cohort 360 (78%, INT: n = 244; CON: n = 116) and 317 (68%, INT: n = 216; CON: n = 101) participants completed the 26-weeks intervention phase and the 52-weeks follow-up. The estimated treatment difference (ETD) between both groups was −3.2 kg [−4.0; −2.5] (P < 0.001) after 12 weeks and −1.8 kg [−2.8; −0.8] (P < 0.001) after 52 weeks.

Conclusions:

A low-intensity lifestyle intervention combined with a liquid meal replacement is superior regarding weight reduction and improvement of cardiovascular risk factors than lifestyle intervention alone.

Introduction

A high energy intake combined with low physical activity are major determinants for overweight and obesity and contribute to the overall increase of non-communicable diseases [1]. Although lifestyle interventions have been shown to induce clinically relevant effects, adherence to these approaches remains low overall. Therefore, alternative treatment strategies need to be considered [2, 3]. In this context, liquid meal replacements have been shown to be an useful treatment option to manage obesity and diseases such as type 2 diabetes [4–6], leading to improvements in fat mass, blood pressure, HbA1c or insulin [7, 8]. Furthermore, there is a positive association between partial and complete meal replacement with weight reduction which was shown in favour of complete meal replacement in patients with type 2 diabetes [9]. Based on their positive effects in the management of patients with type 2 diabetes, liquid meal replacements have been included into current guidelines for baseline treatment of type 2 diabetes [10–12], but not uniformly for the routine management of overweight and obesity [3]. In this regard, there is still uncertainty about weight maintenance and long-term effectivity of formula diets [13, 14] and whether there is a beneficial effect of adding a formula diet to an lifestyle intervention and/or nutrition counselling alone in patients with overweight and obesity [12]. Hence, an international and multicenter RCT, the Almased Concept against Overweight and Obesity and Related Health Risk (ACOORH)-study, was conducted to examine the impact of a liquid meal replacement together with a low-intensity lifestyle intervention compared to a low-intensity lifestyle intervention alone on weight loss in patients with overweight or obesity and accompanied cardiovascular risk factors.

Materials/subjects and methods

Study design and population

Participating volunteers (n = 463) were randomly allocated with a ratio of 1:2 into either a lifestyle intervention group (CON, n = 155) or a meal replacement-based lifestyle intervention group (INT, n = 308). The lifestyle intervention was characterised by a 26-week intervention phase and a follow-up phase until week 52 and the study design has been described in detail previously in a predefined subanalysis of the ACOORH study focusing solely on patients with prediabetes [15]. This multicenter RCT received ethical approval (registered at drks.de; ID: DRKS00006811) for each participating centre and the study reporting adheres to CONSORT guidelines. Informed consent was obtained from all participating volunteers. Study participants were recruited in all study centres either through direct contacting based on existing patient files, (2) proactive study enquiry by the participants via the study centre homepages, or (3) advertisements in newspapers. Inclusion and exclusion criteria have been described in detail previously [15].

Intervention and diet regime

Both groups were provided with guideline booklets about healthy cooking, received advices regarding physical activity and a healthy lifestyle including encouragement to lose weight, and were equipped with telemetric scales (smartLAB scale W; HMM Holding AG, Dossenheim, Germany) and pedometers (smartLAB walk P+; HMM Holding AG, Dossenheim, Germany). Probands were recommended to note down a 4-day, unweighted diet record at baseline and after 12, and 52 weeks of the study and all records (including steps and body weight) were discussed during the study visits (personal contact time ≈ 1–2 h per visit). A detailed description of the study can be found elsewhere [15] and is illustrated in Fig. 1. In addition, the INT group was provided with the liquid soy–yoghurt–honey-based meal replacement Almased Vitalkost® (protein content: 53.3% (83% soy-protein-isolate, and 17% milk protein), glycemic index: 27, energy per 100 g powder: 1507 kJ (360 kcal), Almased-Wellness GmbH, Bienenbüttel, Germany [16]) for the first 26 weeks and received an accompanying booklet containing information about preparing and applying the liquid formula diet and general advices about low-carbohydrate, low-glycemic and protein-rich meals. The management of the liquid formula diet regime during the study is described in detail elsewhere [15]. All booklet records were evaluated at each visit by study nurses and used for nutritional and lifestyle counselling.

Measurements

Measurements were performed at baseline as well as after 4, 12, 26 and after 52 weeks as described in detail elsewhere [15]. Body composition (Seca medical Body Composition Analyser® (seca-mBCA 115), Hamburg, Germany [17]) and blood pressure (Mobil-O-Graph PWA; I.E.M. GmbH, Stolberg, Germany) were determined by using validated devices. Biochemical blood parameters were determined by venous blood sampling. Adverse and serious adverse events [18] were documented continuously (participant questionnaire) and were reviewed by an external monitor. 

Statistics

Sample size calculation was based on the results of a previous study [19] and its assumptions, including randomisation and number of dropouts, are described in detail elsewhere [15]. Final sample size per group comprised at least 19 participants for each study centre. However, based on previous experiences in all participating centres with dropout rates >50% for long-term adherence to weight management programmes, at least a number of 40 participants per centre was targeted. Primary outcome of the ACOORH study was body weight in kg after 4, 12, 26 and 52 weeks of intervention. Power calculation was performed for the difference of body weight change after 12 weeks of intervention between INT and CON. Secondary outcomes comprised changes in anthropometric (fat mass (FM), fat free mass (FFM), and waist circumference (WC)) and clinical parameters (fasting blood glucose (FBG), systolic blood pressure (SBP), diastolic blood pressure (DBP), total cholesterol, HDL cholesterol (HDL-C), LDL cholesterol (LDL-C), TG) after 4, 12, 26 and 52 weeks of intervention. An independent institute (ACOMED statistik® , Leipzig, Germany) executed the statistical analysis and a detailed description including statistical tests applied (for parametric and non-parametric data) and software used can be found elsewhere [15]. Completer (per-protocol (PP)) and intention to-treat (ITT) analyses were applied. All statistical tests were two-sided and significance was assumed at α < 0.05. Participants who visited all follow-up assessments were integrated into the PP analysis. Primary analysis focused on the ITT approach as these values are of more clinical relevance. Last observation-carried-forward (LOCF) method was applied to replace missing data for the ITT analysis.

Results

Four hundred thirty-nine (95%, INT: n = 299; CON: n = 140) from the starting cohort finished the first 4 weeks of the intervention phase. Follow-up data after 12, 26 and 52 weeks were available from 396 (86%, INT: n = 270; CON: n = 126), 360 (78%, INT: n = 244; CON: n = 116) and 317 participants (68%, INT: n = 216; CON: n = 101). Anthropometric and clinical parameters of INT and CON at baseline are illustrated in Table 1. Dropouts demonstrated no statistical difference in comparison to the non-dropout group (Supplementary Table S1). Participants dropped out because of (1) health issues, (2) work-related issues, (3) personal issues and (4) other reasons. No acute cardiac event, hospitalisation for cardiovascular disease, or other serious adverse events related to the study participation occurred. Compared to CON, INT significantly lost more weight after 4 weeks (−4.0 kg with 95% CI [−4.3;−3.8] vs. −1.4 kg [−1.8;−1.1]; P < 0.001), 12 weeks (−5.8 kg with 95% CI [−6.3;−5.3] vs. −2.7 kg [−3.3;−2.1]; P < 0.001), 26 weeks (−5.9 kg with 95% CI [−6.5;−5.4] vs. −3.0 kg [−3.8;−2.2]; P < 0.001) and 52 weeks (−4.4 kg [−5.0; −3.8] vs. −2.7 kg [−3.0;−2.0]; P < 0.001) in the ITT analysis. The estimated treatment difference (ETD) between both groups was -2.6 kg [−3.5; −1.8] (P < 0.001) after 4 weeks, −3.2 kg [−4.0; −2.5] (P < 0.001) after 12 weeks, −2.9 kg [−3.7; −2.1] (P < 0.001) after 26 weeks and −1.8 kg [−2.8; −0.8] (P < 0.001) after 52 weeks. These differences were even stronger in the PP analysis after 4 weeks (−4.5 kg with 95% CI [−4.8;−4.2] vs. −1.6 kg [−2.0;−1.2] P < 0.001), 12 weeks (−6.3 kg with 95% CI [−6.8;−5.8] vs. −3.2 kg [−3.9;−2.6] P < 0.001), 26 weeks (−6.8 kg with 95% CI [−7.5;−6.2] vs. −3.6 kg [−4.6; −2.7] P < 0.001) and 52 weeks (−5.0 kg [−5.7;−4.2] vs. −3.5 kg [−4.5;−2.5] P = 0.021). Weight reduction was accompanied with changes in WC, FM, FBG, SBP, DBP, total cholesterol, TG and LDL-C in both groups following the intervention, with a particularly pronounced effect within the first 12 weeks (Fig. 2) (ITT analysis). These effects were already evident after 4 weeks of intervention in all parameters in the INT group (all P < 0.001) (ITT analysis), but not in the CON group. Only FM, WC and SBP (all P < 0.001) as well as DBP and total cholesterol (both P < 0.01) significantly changed after 4 weeks in CON (ITT analysis). The aforementioned 12- week changes remained significantly altered after 26 weeks of intervention in the INT group in all parameters (P < 0.001) (ITT analysis). In contrast, only FM, WC, and SBP remained significantly changed after 26 weeks in the CON group (all P < 0.01) (ITT analysis). Compared to CON, INT significantly reduced more WC, FM, FFM, total cholesterol, and LDL-C after 12 weeks of intervention (Table 2). These differences remained significant after 52 weeks in FM, FFM and INT reduced FM by −3.3 kg with 95% CI [−3.9; −2.7] vs. −2.4 kg [−3.2; −1.5] P = 0.020) and) compared to CON after 52 weeks. INT showed a pronounced loss in FFM compared to CON after 52 weeks (−0.9 kg [−1.3; −0.6] vs. −0.3 kg [−0.9; 0.2] P < 0.001).

Discussion

The results of the ACOORH trial show that a low-intensity lifestyle intervention accompanied with a liquid formula diet contributes to larger reductions in body weight in patients with overweight or obesity and accompanied cardiovascular risk factors compared to a low-intensity lifestyle intervention alone and these findings remain significantly superior even after 52 weeks. The weight reduction after 1 year (−5.8 kg [−6.3; −5.3] (ITT analysis)) is comparable to other lifestyle intervention programmes with smaller cohorts (n = 19–167), which have also shown a significant weight loss ranging from −1.43 kg to −12.1 kg [20]. In particular, very intense lifestyle programmes with rigorous meal replacement regimen [21] or intensive support [22] led to mean weight losses >10 kg. Furthermore, study effects and weight loss show a dose-response pattern in relation to programme duration [23] and intensity of support [20]. The longer the intensive intervention phase and the greater the level of support, the greater the weight loss. A recently published systematic review and meta-analysis demonstrated larger weight reductions following either very low (800 kcal/ day) liquid meal replacements (ranging from 8.9 to 15.0 kg) in patients with obesity (BMI: 36–43 kg/m²) [24]. Compared to the present study can be assumed that the weight reduction difference to the studies in the meta-analysis is resulted by a higher calorie intake per day (1300–1500 kcal/ day). In addition, we chose a more moderate daily energy intake target to increase study compliance and adherence as well as to minimise dropout rates. In support of this approach, it has been shown that a moderate and continuous weight loss reduces the risk for adverse outcomes in the long-term compared to a fast and severe weight loss [25]. In the present study, weight reduction was accompanied with further improvements, (predominantly achieved in the INT-group) during the 12-week intervention phase in cardiometabolic parameters, including FM, WC, DBP and LDL-C and TC. Furthermore, after 52 weeks of follow-up there was still a significant difference in FM loss between both groups. These findings are in line with other lifestyle intervention trials with low-calorie diets in patients with prediabetes [7] or type 2 diabetes [26, 27] or lifestyle interventions with physical activity in patients with obesity [28]. 

The ACCORH trial and its strengths are characterised by (1) a comparably large sample size in an (2) international and multicenter design with (3) a randomised controlled trial approach. Moreover, (4) two intervention groups were followed up over a period of 52 weeks and this trial was conducted in a (5) real-world setting in which a low-intensity lifestyle intervention was combined with liquid meal replacement. The intention was to design a practical lifestyle-based intervention programme which could be easily implemented into present health care programmes. Moreover, the (6) inclusion of only high-risk participants with at least one additional co-morbidity of the metabolic syndrome indicates a further strength of the study. There are also limitations in the present trial that have to be considered. We did not constantly (i) controlled the participants for decreased energy intake or for false food compositions (e.g., amount of carbohydrates or proteins) by monitoring diet diaries. As it is well-known that dietary records of patients with obesity are characterised by systematic errors, we, therefore, had purposely chosen not to constantly monitor these records [29]. However, the prepared 4-day diet diaries of the probands were used in each study visit as a resource of information for the lifestyle counselling. Moreover, volunteers of the INT group should record the number of containers and amount of meal replacement consumed. Thus, we were able, at least, to evaluate the intake of liquid meal replacement within the first 12 weeks. A second limitation was the imputation of missing values by the LOCF approach. More sophisticated imputation methods like multiple imputation could have been performed as this imputation technique takes the uncertainty of the imputed values more realistic into account. However, the LOCF procedure was consciously chosen as it is a conservative statistical approach to estimate treatment effects, which might have even underestimated the results. Concomitantly, the ITT analysis method performed prevents the overestimation of data and takes the number of dropouts into account.

In sum, a low-intensity lifestyle intervention accompanied with a liquid meal replacement contributes to a long-term and clinically relevant weight reduction in patients with overweight and obesity and further cardiovascular risk factors. Furthermore, this weight reduction was characterised with improvements in cardiovascular and cardiometabolic risk factors. The present findings underline the efficacy of the liquid formula diet tested in individuals with overweight or obesity and accompanied cardiovascular risk factors when included in a lifestyle intervention programme. This therapy approach should be considered as a valid option for management of overweight and obesity in clinical, community and health care settings.

Compliance with ethical standards

Acknowledgements

The ACOORH group thanks their consortium members for their excellent work (a complete list of all consortium members including those who did not met authorship criteria can be seen in the Supplementary information). Furthermore, the authors thank their study staff for their excellent work and Dr. Thomas Keller (ACOMED statistik® , Leipzig, Germany) for his support in the statistical analysis. We would also like to thank Verena Heinicke, Katrin Esefeld, Nina Schaller and Johannes Scherr for their significant support of performing the trial at the local Munich site. The study centre of Freiburg thanks their colleagues Sadaf Koohan and Andrea Stensitzky for their great support by conducting the study. We, the West-German Centre of Diabetes and Health, Düsseldorf, thank our study nurse Bettina Prete for her excellent work.

ACOORH study group

Martin Halle1,2, Nina Schaller1 , Martin Röhling3 , Kerstin Kempf3 , Stephan Martin3,12, Winfried Banzer4 , Klaus Michael Braumann5 , Jürgen Scholze8 , Dagmar Führer-Sakel9 , Susanne Tan9 , Hans Georg Predel7 , Aloys Berg11, Sadaf Koohkan13, David McCarthy6 , Hermann Toplak10, Michel Pinget14 12Faculty of Medicine, Heinrich Heine University Dusseldorf, Dusseldorf, Germany; 13University of Freiburg, Institute for Sports and Sports Science, Freiburg, Germany; 14Department Endocrinologie, Diabete et Maladies Métaboliques, Faculte de Medicine de‘l University de Strasbourg, Strasbourg, France

Author contributions

AB had the initial idea for the study design and initiated the study. The protocol was designed together with HT and with additional contributions of SM. MH and MR drafted the manuscript. All authors critically revised the manuscript and approved the final version. WB, AB, KMB, DMC, MH, KK, SM, HGP, JS, DF-S and HT collected data at their local sites. A Berg is the guarantor of this work and all co-authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Funding

The study was financially supported by the Almased Wellness-GmbH. The funder had no influence on study hypothesis/ design, data collection, execution, data analysis, interpretation, manuscript preparation and/or publication decisions.

Conflict of interest

WB, AB, KMB, MH, KK, DMC, HGP, JS, DF-S and HT received research support for their departments from the Almased-Wellness-GmbH to perform the study. AB, MH, DMC and HT have also received speakers' honoraria (category: personal financial interests) from Almased-Wellness-GmbH. All four authors declare that their honoraria had no influence on their contribution to the study design, data collection, data analysis, manuscript preparation and/or publication decisions. NS and MR declare no conflict of interest regarding the publication of this article.