Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Long-term impact of the metabolic status on weight loss-induced health benefits

Long-term impact of the metabolic status on weight loss-induced health benefits Background: While short‑term effects of weight loss on quality of life and metabolic aspects appear to be different in metabolically healthy (MHO) and metabolically unhealthy obese (MUO), respective long‑term data is still missing. Given the high relevance of long‑term changes, we aimed to address these in this post ‑hoc analysis of the MAINTAIN trial. Methods: We analyzed 143 overweight/obese subjects (BMI ≥ 27 kg/m , age ≥ 18 years) before and after a 3‑month weight loss program (≥ 8% weight loss), after a 12‑month period of a randomized weight maintenance intervention (n = 121), and after another 6 months without intervention (n = 112). Subjects were retrospectively grouped into MHO and MUO by the presence of metabolic syndrome and secondarily by estimates of insulin sensitivity (HOMA‑ IR and ISI ). Quality of life (QoL), blood pressure, lipids, HOMA‑IR, and ISI were assessed and evaluated using Clamp Clamp mixed model analyses. Results: Despite similar short‑ and long‑term weight loss, weight loss‑induced improvement of HOMA‑IR was more pronounced in MUO than MHO after 3 months (MHO: 2.4[95%‑ CI: 1.9–2.9] vs. 1.6[1.1–2.1], p = 0.004; MUO: 3.6[3.2–4.0] vs. 2.0[1.6–2.4], p < 0.001; p = 0.03 for inter‑ group comparison). After 21 months, the beneficial effect was no longer seen in MHO (2.0[1.5–2.6], p = 1.0), while it remained partially preserved in MUO (2.9[2.4–3.3], p = 0.002). QueryShort‑ term improvements of lipid parameters were similar in both groups. However, long‑term improvements of HDL ‑ cholesterol and triglycerides were only seen in MUO (44.4[41.5–47.4] vs. 49.3[46.2, 52.3] mg/dl, p < 0.001; 176.8[158.9– 194.8] vs. 138.8[119.4–158.3] mg/dl, p < 0.001, respectively) but not in MHO. Weight loss‑induced improvements in the QoL and particularly the physical health status were maintained in MUO until the end of the trial, while benefits disappeared over time in MHO. Group allocation by HOMA‑IR and ISI revealed higher benefits for MUO mainly in Clamp parameters of the glucose metabolism and QoL. Conclusions: Our data demonstrates stronger and longer‑lasting improvements of metabolism and QoL in MUO after weight loss. Trial registration (ClinicalTrials.gov): NCT00850629. Registered 25 February 2009, https:// clini caltr ials. gov/ ct2/ show/ NCT00 850629. Keywords: Healthy obesity, Weight loss, Insulin sensitivity, Quality of life Introduction Obesity is a chronic, systemic, and multifactorial disease *Correspondence: knut.mai@charite.de and is defined—according to the world health organiza - Department of Endocrinology and Metabolism, Charité – 2 tion (WHO)—by a body mass index (BMI) of 30  kg/m Universitätsmedizin Berlin, corporate member of Freie Universität or more. Despite its preventable nature, the prevalence Berlin, Humboldt‑Universität Zu Berlin, and Berlin Institute of Health, Chariteplatz 1, 10117 Berlin, Germany of obesity has been increasing over the last decades [1]. Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Soll et al. Nutrition & Metabolism (2022) 19:25 Page 2 of 10 Obesity is associated with several musculoskeletal, meta- and special health care providers in clinical practice. To bolic, and cardiovascular diseases, amongst others [2]. evaluate the impact of alternative definitions of metabolic Weight loss has been shown to improve the obesity- health, we performed comparable analyses after classifi - related metabolic disturbances and is therefore recom- cation of MUO and MHO using the HOMA-IR and the mended in the European guideline for the management hyperinsulinemic-euglycemic clamp as an easily estima- of obesity, among others [3]. In general, positive effects ble and a more precise determinant of insulin resistance, on metabolic diseases, such as lower incidence of type respectively. 2 diabetes and increased insulin sensitivity, have been shown for a weight loss of 5–8% [4, 5]. Given the ris- Study design and methods ing number of obese subjects and limited resources, the Participants and study design identification of subjects who will benefit most from The study (MAINTAIN trial, ClinicalTrials.gov registry weight loss interventions may help to optimize current number NCT00850629) was performed between 2010 treatment strategies. and 2016 in Berlin, Germany. Detailed trial informa- More than 15 years ago, a subset of obese patients has tion has been reported previously [10, 16, 17]. The flow been identified that is not affected by the mentioned of participants is shown in Additional file  1: Figure S1. In metabolic changes that frequently go along with obesity. short, 156 overweight and obese subjects (BMI ≥ 27  kg/ Primarily, they remain insulin sensitive despite being m , age ≥ 18  years) underwent a 3-month-weight loss obese [6]. Current data indicates that acute metabolic program, realized by caloric restriction using a very low benefits of weight loss were more pronounced [7] or energy diet for 8 weeks (replacement of all meals by a for- did only occur in metabolically unhealthy obese (MUO) mula diet with 800  kcal per day) and an energy-reduced [8, 9], but not in metabolically healthy obese (MHO). diet (healthy food diet with approximately 1500 kcal per Although this data supports a metabolic improvement day) for the following 4  weeks, both accompanied by by dietary weight loss primarily in MUO, long-term data nutritional counseling, physical exercises, and psycho- on the different or comparable effectiveness of tempo - logical advice. Subjects that experienced a prior weight rary lifestyle interventions for MHO and MUO is not yet loss of more than 5  kg in the last 2  months, changed available. This is crucial as the impact of temporary life - their smoking habits or diets within the last 3  months, style interventions on weight loss is often frustrating due were pregnant, or suffered from other endocrine disor - to frequently observed weight regain [10]. ders, eating disorders, or severe chronic diseases were Although the classification of MHO and MUO is fre - not included in this trial. All participants who succeeded quently based on the presence or absence of the meta- to lose at least 8% of their body weight (n = 143) were bolic syndrome [7, 11, 12], this still remains inconsistent then randomized into an intervention or control group and is a matter of debate. The International Diabetes Fed - for 12  months of weight maintenance. The intervention eration (IDF) defines the metabolic syndrome as a spe - comprised a continuous multimodal counseling focus- cific cluster of medical conditions with central obesity, ing on caloric restriction, nutritional counseling, physi- insulin resistance, hypertension, and dyslipidemia [13]. cal exercises, and psychological support. In contrast, the Nevertheless, the presence of insulin resistance in fast- control group was under free living conditions without ing state or during an oral glucose tolerance test, or the further counseling. Following the randomized interven- results of a hyperinsulinemic-euglycemic clamp are also tion period, both groups underwent a 6-month-follow- used to define metabolic health [7, 8, 14, 15]. The added up period without any further intervention. More details value of these definitions of metabolic health is currently regarding the lifestyle intervention can be found in the still unknown. Additional file 1. Given this lack of evidence, we report short- and long- Retrospectively, all randomized subjects were allo- term data from a 21-month weight loss/weight mainte- cated to be MHO or MUO depending on the absence nance trial focusing on metabolic healthy and unhealthy or presence of metabolic disturbances. In this regard, obese. In detail, we intend to elucidate whether a life- three different definitions were used. First, subjects were style-based weight loss intervention demonstrates differ - categorized in accordance to the criteria of the meta- ent efficacies regarding body weight reduction, metabolic bolic syndrome as defined by the IDF [18]: central obe - improvement, and quality of life (QoL) in obese subjects sity defined as waist circumference ≥ 94  cm for men differing in their metabolic health status. The efficacy and ≥ 80  cm for women plus any two of the following was especially assessed in the long term. In order to do four factors: (1) elevated triglycerides level > 150  mg/dl so, we used the IDF criteria from 2005 for metabolic syn- or intake of lipid-lowering drugs, (2) reduced HDL cho- drome to separate MUO from MHO individuals as they lesterol < 40  mg/dl for men and < 50  mg/dl for women, are well-established and easily applicable for primary (3) raised blood pressure ≥ 130  mmHg for systolic S oll et al. Nutrition & Metabolism (2022) 19:25 Page 3 of 10 or ≥ 85  mmHg for diastolic blood pressure or intake of surface per minute and of glucose at an individual rate antihypertensive drugs, (4) raised fasting plasma glu- to fix plasma glucose levels at 80  mg/dl ± 8  mg/dl. The cose ≥ 100 mg/dl or previously diagnosed type 2 diabetes. steady state was defined as plasma glucose levels of Next, subjects were allocated to be MHO/MUO by their 80  mg/dl ± 8  mg/dl for at least 30  min. The ISI was Clamp respective HOMA-IR before weight loss. Here, subjects calculated by dividing the glucose infusion rate at steady with a HOMA-IR ≥ median HOMA-IR (~ 2.19) were dis- state (mg/min) per body weight (kg) by plasma insulin at tributed to MUO, subjects with a HOMA-IR < median steady state (mU/l) [20]. HOMA-IR to MHO. Last, subjects were allocated by their respective insulin sensitivity index (ISI ) before Statistical analysis Clamp weight loss. Here, subjects with an I SI ≥ median Data of 143 subjects who lost at least 8% of their initial Clamp −1 −1 −1 ISI (~ 0.058 mg  kg  min/(mU  l )) were defined body weight was analyzed. The primary analysis was per - Clamp as MHO, subjects with a I SI < median ISI were formed using group allocation according to IDF criteria. Clamp Clamp classified as MUO. Due to the lack of reliable cut-offs for Statistical analysis was performed using RStudio Version a “pathological” HOMA-IR or I SI , the group was 1.2.1335 (RStudio Inc., Boston, MA) and the R software Clamp divided by the respective median. package. We performed a linear mixed effects analy - sis of the relationship between anthropometric/meta- Procedures bolic parameters and time using the nlme package. For Fasting blood sampling as well as measurements of blood group comparisons, additional analysis of the relation- pressure, waist circumference, body weight, height, and ship between anthropometric/metabolic parameters and assessment of QoL were performed before ( T ) and after the interaction of time and group affiliation was done. -3 (T ) weight loss, 12  months ( T ) after randomization We entered age, sex, and randomization state (interven- 0 12 and after another 6  months without active intervention tion group or control group) as fixed effects and poten - (T ). At T, T , and T , subjects underwent a hyperin- tial confounders. In order to adjust for BMI change in 18 -3 0 12 sulinemic-euglycemic clamp. the course of the study, we repeated this analysis after Blood samples were centrifuged; plasma and serum additional inclusion of either BMI, waist circumference, samples were frozen immediately at − 80  °C. Glucose or body fat mass at all time-points as a fixed effect. Inter - was measured using the glucose oxidase method (Dr. cepts for subjects were considered as a random effect. Müller Super GL, Freital, Germany). HbA1c was meas- Results were considered to be significant, if the two-sided ured by high-performance liquid chromatography using α was below 0.05. Presented data represents estimated the VARIANT II (Bio-Rad, Hercules, US). Lipids were marginal mean and 95%-confidence interval (CI). P val - measured by standard laboratory methods using Cobas ues were obtained from comparisons using the emmeans ISE direct and c111 Analyzer (Roche Diagnostics, Man- package. Adjustment for multiple testing was performed nheim, Germany). Serum insulin was measured by by Bonferroni correction. fluoroimmunometric assay (AutoDelfia; Perkin Elmer, Rodgau, Germany). Fat mass was assessed by bioelectri- Results cal impedance analysis using the AKERN BIA 101 (SMT Clinical and metabolic characteristics at baseline medical GmbH & Co. KG, Würzburg, Germany). General data on the trial population has already been In order to assess potential changes in quality of life published previously [10, 16, 21]. Respective relevant data and patient-reported health status that might accompany can be found in Additional file  1: Table  S1. To eliminate weight loss and maintenance, participants were asked to the effect of interfering factors, age, sex, and randomi - fill in the Short Form (36) Health Survey (SF-36). It con - zation state were included as fixed effects in the mixed tains 36 items in eight subscale scores: general health effect model analysis for group comparisons. Using IDF perceptions, physical functioning, role limitations due to criteria, there were significant differences between both physical problems, bodily pain, mental health, role limi- MHO and MUO at T for waist circumference (Δ(MHO- -3 tations due to emotional problems, vitality, and social MUO) − 6.5 [95% CI: − 10.4, − 2.6] cm, p = 0.001) and functioning. A physical component summary score (PCS) BMI (− 2.2 [− 4.2, − 0.2] kg/m , p = 0.03). Consistent is derived from the first four, a mental component sum - with the parameters for allocation, both groups also dif- mary score (MCS) from the latter four components [19]. fered in triacylglycerol levels (− 57.7 [− 84.3, − 31.1] mg/ The insulin resistance index HOMA-IR was calculated dl, p < 0.001), HDL cholesterol (11.1 [6.8, 15.3] mg/dl, as previously described: fasting glucose (mg/dl) × fasting p < 0.001), fasting glucose (− 12.6 [− 16.9, − 8.2] mg/dl, insulin (mU/l)/405 [20]. p < 0.001), and HbA1c (− 0.42 [− 0.65, − 0.18], p < 0.001). The hyperinsulinemic-euglycemic clamp involved Additionally, MHO and MUO differed in the estimates infusion of insulin at the rate of 40  mU per m of body for insulin resistance HOMA-IR (− 1.3 [− 1.9, − 0.7], Soll et al. Nutrition & Metabolism (2022) 19:25 Page 4 of 10 −1 −1 Improved quality of life persists in MUO but not in MHO p < 0.001) and ISI (0.03 [0.02, 0.04] mg  kg  min / Clamp −1 At baseline, MHO had a higher physical health status (mU  l ), p < 0.001), but in none of the other presented than MUO (Δ(MHO-MUO) 4.8 [− 0.5, 10.2], p = 0.005), parameters. (Table 1). while mental health status did not differ (− 1.5 [− 7.5, 4.6], p = 0.4). After initial weight loss, the MUO reported Short‑term effects of weight loss on metabolism an improved physical and mental health status, while During the initial 3-month weight reduction period, the MHO indicated an improved physical but not men- weight loss was similar in MHO and MUO (− 4.6 [− 5.4, 2 2 tal health (Fig.  1A, B). Physical health status remained − 3.7] kg/m vs. − 4.7 [− 5.4, − 4.0] kg/m , p = 1.0). How- improved in MUO in the long term while the improve- ever, triglycerides and fasting glucose declined only in ment of mental health disappeared after 12  months. In MUO, but not in MHO. (Table  1) While HOMA-IR was contrast, the short-term increase of physical health in significantly more reduced in MUO than in MHO (− 1.7 MHO was already reversed after 12 months (Fig. 1A, B). [− 2.2, − 1.1] vs. − 0.8 [− 1.4, − 0.2], p = 0.03), HDL cho- When groups are separated by HOMA-IR, the results lesterol was less modified by weight loss in MUO (− 2.7 regarding quality of life are almost similar (Fig.  1C, D), [− 5.2, − 0.1] vs. -6.6 [− 9.6, − 3.5] mg/dl, p = 0.06), but although improvement of physical health persisted up the difference did not reach significance. In contrast, to 12  months after weight loss in MHO. Comparable comparable improvements were detected in both groups changes and differences were found for group allocation in ISI , total cholesterol, and LDL cholesterol, as well Clamp by ISI (Fig. 1E, F). Clamp as systolic and diastolic blood pressure. Overall, group allocation based on HOMA-IR showed a stronger improvement in MUO only regarding fast- Improved insulin resistance in MUO is not exclusively ing glucose and HOMA-IR (which was not modified in explained by weight loss MHO) but not in lipids during the weight loss period Considering the differential development of insulin (Additional file  1: Table  S2). Comparably, group alloca- resistance in MHO and MUO despite comparable weight tion according to ISI revealed almost similar find - loss, we aimed to further investigate the interrelation of Clamp ings with different developments between groups only in improved obesity and insulin resistance. Hence, to elimi- HbA1c and HOMA-IR (Additional file 1: Table S3). nate the impact of body weight changes, we addition- ally adjusted the improvement of HOMA-IR for BMI. Long‑term effects of weight loss on metabolism Interestingly, while the MHO did not present changes After initial weight loss, participants showed a mod- in HOMA-IR anymore, it still was signifi - BMI-adjusted erate increase in BMI, but still maintained a signifi - cantly declined by weight loss in MUO in the short- and cantly lower body weight until the end of the study after medium-term (Fig.  2). Likewise, adjustment for waist 18 months. There was no difference in the weight course circumference or body fat mass gave comparable results between both, MUO and MHO (Table  1). In analogy, (Additional file 1: Figure S2). trajectory of waist circumference—as an established estimate of visceral obesity—was equal in both groups. Discussion Similarly, changes of ISI were comparable in MUO Clamp In the presented study, we aimed to answer the ques- and MHO. In contrast, in comparison to baseline MUO tion whether MHO and MUO subjects gain metabolic demonstrated improved HOMA-IR, triglycerides, and and health benefits to different extents from weight loss/ HDL cholesterol for up to 18 months and improved fast- weight maintenance by a multimodal lifestyle inter- ing glucose, total cholesterol, and LDL cholesterol until vention. There are few studies that already addressed 12  months after weight loss (Table  1). However, these this issue in a short-term context, but were partially parameters were no longer improved in MHO at T or not consistent in their results [7–9, 11, 12, 15]. On the T compared to baseline. Nevertheless, between-group one hand, Dalzill et  al. and Ruiz et  al. found a similar comparisons for ΔT T did not reveal any significant 3 18 improvement in HOMA-IR and several other meta- differences in these parameters. bolic parameters in MUO and MHO investigating 124 Group allocation depending on HOMA-IR revealed a patients before and after a 9-month intensive lifestyle long-term improvement of HOMA-IR in MUO but not modification program and 78 postmenopausal women in MHO (p < 0.001 for inter-group comparison) (Addi- before and after a 3-month energy-restricted treatment, tional file  1: Table S2). A similar development of HOMA- respectively [11, 12]. On the other hand, several previ- IR in both groups is found if groups are separated by ous studies described beneficial effects preferentially for ISI (Additional file  1: Table  S3). However, this clas- Clamp MUO. While Janiszewski et al. studied 106 adults before sification revealed additional differences in the long-term and after various 3-to-6-month lifestyle interventions courses of BMI and ISI between both groups. Clamp S oll et al. Nutrition & Metabolism (2022) 19:25 Page 5 of 10 Table 1 Estimates of clinical parameters of MHO/MUO following the IDF definition of metabolic syndrome Metabolically Healthy Obese (MHO) Metabolically Unhealthy Obese (MUO) p value (MHO vs MUO) T T T T T T T T T → T T → T T → T 3 0 12 18 3 0 12 18 3 0 3 12 3 18 n: m/w 59: 9/50 59: 9/50 46: 7/39 44: 6/38 84: 22/62 84: 22/62 75: 19/56 68: 19/49 Randomization: 29/30 42/42 control/interven tion Age [years] 48.6 [45.2,52.0] 51.9 [49.2,54.5] BMI [kg/m2] 36.0 [34.2,37.7] 31.4 32.5 33.5 38.2 [36.8,39.6] 33.5 34.1 35.2 1 1 1 [29.7,33.1]*** [30.8,34.3]*** [31.7,35.2]*** [32.1,34.9]*** [32.7,35.6]*** [33.8,36.6]*** Waist circumfer 107.6 99.4 100.9 101 114 103.8 104.4 106 0.6 0.14 1 ence [cm] [104.2,110.9] [96,102.7]*** [97.5,104.4]*** [97.5,104.5]*** [111.3,116.8] [101.1,106.5]*** [101.6,107.2]*** [103.2,108.8]*** Serum glucose 82.4 [78.8,86] 80.8 [77.2,84.4] 84.3 [80.4,88.2] 86.9 [83.0,90.9] 95.0 [92.0,97.9] 86.1 89.7 [86.6,92.8]** 93.2 [90.1,96.3] 0.01 0.02 0.06 [mg/dl] [83.1,89.0]*** HbA1c [%] 5.5 [5.3,5.7] 5.9 [5.7,6]** 5.4 [5.2,5.6] 5.5 [5.3,5.7] 5.9 [5.8,6.1] 5.9 [5.8,6.1] 5.6 [5.4,5.8]*** 5.7 [5.6,5.9] 0.02 0.66 1 HOMAIR 2.4 [1.9,2.9] 1.6 [1.1,2.1]** 1.9 [1.3,2.4] 2.0 [1.5,2.6] 3.6 [3.2,4] 2.0 [1.6,2.4]*** 2.3 [1.9,2.7]*** 2.9 [2.4,3.3]** 0.03 0.06 0.91 ISI 0.08 [0.07,0.09] 0.10 0.09 [0.08,0.10]* 0.05 [0.04,0.06] 0.08 0.07 0.31 0.23 Clamp −1 −1 [mg kg min / [0.09,0.11]*** [0.07,0.09]*** [0.07,0.08]*** −1 (mU l )] Total cholesterol 200.4 [190,210.9] 171.2 192.5 192.4 198.1 167.5 188.8 191 [182.1,199.8] 1 1 1 [mg/dl] [160.8,181.6]*** [181.4,203.7] [181.3,203.6] [189.7,206.5] [159.1,175.9]*** [180.1,197.5]* LDL Cholesterol 121.2 102.1 114.6 [105,124.1] 117.2 121.8 103.4 113.7 118.4 [110.8,126] 1 1 1 [mg/dl] [112.2,130.2] [93.1,111.1]*** [107.6,126.7] [114.5,129] [96.1,110.6]*** [106.2,121.2]* HDL Cholesterol 55.5 [51.9,59.2] 49 [45.3,52.6]*** 58 [54.2,61.8] 58 [54.2,61.8] 44.4 [41.5,47.4] 41.8 [38.9,44.7]* 49.9 49.3 0.06 0.43 0.99 [mg/dl] [46.9,52.9]*** [46.2,52.3]*** Triglycerides 119.1 [97.0,141.2] 96.1 [74.1,118.2] 95.5 [70.9,120.0] 101.9 176.8 110.1 149.1 138.8 0.01 1 1 [mg/dl] [77.3,126.5] [158.9,194.8] [92.2,128.1]*** [130.2,168.1]* [119.4,158.3]*** Systolic RR 127.6 118.7 121 [115.2,126.7] 121 [115.0,126.9] 132.4 121.4 125 [120.5,129.5] 127.6 1 1 1 [mmHg] [122.4,132.8] [113.6,123.9]* [128.0,136.7] [117.1,125.6]*** [122.9,132.3] Diastolic RR 80.6 [77.8,83.4] 74.1 77.9 [74.8,80.9] 79.8 [76.6,82.9] 80.5 [78.1,82.8] 74.8 79 [76.6,81.4] 78.1 [75.6,80.6] 1 1 1 [mmHg] [71.2,76.9]*** [72.5,77.2]*** Values shown (except for n, randomization, and age) are estimated marginal means with 95% confidence intervals from model with adjustment to sex, age, and randomization p < 0.05, ** p < 0.01, *** p < 0.001 vs. T in respective group -3 Soll et al. Nutrition & Metabolism (2022) 19:25 Page 6 of 10 † C A E 60 60 60 ††† 55 55 55 ** *** * *** *** *** 50 50 50 *** *** *** *** *** ** *** *** 45 45 40 40 40 35 35 35 -3 0 12 18 -3 0 12 18 -3 0 12 18 Time [month] Time [month] Time [month] MHO D F 60 60 60 MUO 55 55 55 ** 50 50 50 *** ** 45 45 45 (MUO) 40 40 40 35 35 35 -3 0 12 18 -3 0 12 18 -3 0 12 18 Time [month] Time [month] Time [month] Fig. 1 Eec ff t of weight loss and maintenance on patient ‑reported quality of life in MHO/MUO. A, B MHO + MUO were assigned following the IDF criteria for metabolic syndrome. C, D MHO + MUO were assigned on the basis of HOMA‑IR. E, F MHO + MUO were assigned on the basis of ISI . Clamp Values represent estimated marginal means with 95% confidence intervals from model with adjustment to sex, age, and randomization. *p < 0.05, **p < 0.01, ***p < 0.001 vs. T in respective group. Where indicated, †/††/††† describe between‑ group comparisons, respectively. SF-36 Short Form ‑3 (36) Health Survey, PCS physical component summary score, MCS mental component summary score from 44 patients before and after a 6-month dietary intervention, described enhanced myocellular insulin sensitivity in MUO, but not in MHO[9]. Similar to our MHO study, most previous studies reported data from second- MUO ** ary analyses. While weight loss ranged from 3 to 9% in *** the aforementioned studies, our patients lost more than 12% during weight loss. The patients’ age ranged from 45 to 60  years in most previous studies, but Ruiz et  al. and Shin et  al. had markedly younger participants of about 0 35 to 40  years [11, 15]. However, there is no single, eas- -3 0 12 18 ily distinguishable parameter—like age, sex, number of Time [month] participants, group definition, extent of weight loss, or Fig. 2 Eec ff t of weight loss and maintenance on insulin resistance duration of intervention—in previous studies explaining (adjusted for BMI) in MHO/MUO. MHO + MUO were assigned why some did find different developments between MHO following the IDF criteria for metabolic syndrome. Values represent and MUO and others did not. Nevertheless, we were able estimated marginal means with 95% confidence intervals from model with adjustment to sex, age, randomization, and BMI at all to address several factors that might have influenced the time‑points. * p < 0.05, ** p < 0.01, *** p < 0.001 vs. T in respective ‑3 discrepancies in the existing results: these include time group points of measurements, group definition, and tech - niques of measuring metabolic changes. They will be dis - cussed in the following paragraphs. and found a larger increase of insulin sensitivity in MUO First, it is important to notice that in all previous stud- after weight loss [7], Kantartzis and colleagues demon- ies only short-term effects were addressed, as the over - strated improved HOMA-IR and glucose levels during all observation periods spanned from 3 to 9 months and an OGTT only in MUO, but not in MHO in a report on measurements took place directly after the intervention, 103 patients before and after a 9-month lifestyle inter- while we were able to track participants in our study vention [8]. Similarly, Karelis et  al., who performed for up to 18  months after weight loss (i.e. 21  months hyperinsulinemic-euglycemic clamps and reported data after study begin). Our study design allows separation SF-36 PCS SF-36 MCS HOMA-IR BMI-adjusted SF-36 PCS SF-36 MCS SF-36 PCS SF-36 MCS S oll et al. Nutrition & Metabolism (2022) 19:25 Page 7 of 10 of short-term and long-term effects of weight loss by Last, most previous studies (including both studies that repeated measurements after the weight loss inter- found no difference between the groups) reported meta - vention, which has not been done before. Thereby, we bolic changes in the lipid profile and the HOMA-IR as a could confirm the studies demonstrating stronger short- measure of insulin sensitivity [8, 11, 12, 15] while a few term effects in MUO in several metabolic parameters, reported results from the hyperinsulinemic-euglycemic although both subgroups did benefit from weight loss. clamp [7, 9] or an oral glucose tolerance test [8]. We The between-group difference partly persisted over time are the first to report data from the hyperinsulinemic- during weight maintenance, which has not been reported euglycemic clamp as well as HOMA when comparing before. Actually, most metabolic improvements dissolved MHO and MUO regarding differences after weight loss. in MHO while they were largely preserved in MUO. It Interestingly, our data from the hyperinsulinemic-eug- cannot be ruled out that the disappearance of the met- lycemic clamp indicated a comparable improvement of abolic benefits in MHO in the long term after weight the ISI in both groups up to 12 months after weight Clamp loss results from ‘healthier’ baseline values compared to loss. Currently, we have no definite explanation for this MUO. In a study on obese subjects, Klöting et  al. found observation. However, the I SI preferentially reflects Clamp that insulin sensitive obese had a lower visceral fat area, myocellular insulin sensitivity. Hence, this might indicate reduced macrophage infiltration into the omental adipose that the differences in the weight loss-induced changes tissue relevantly changing its structure as well as lower of HOMA-IR between MUO and MHO could rather be plasma levels of inflammatory parameters compared to caused by changes in other body compartments than insulin resistant obese subjects [7]. Hence, diverging con- muscle, such as liver or subcutaneous fatty tissue. The stitutions of the adipose tissue might account for differ - use of labelled glucose for the hyperinsulinemic-euglyce- ences in the metabolic improvement between MHO and mic clamp in future studies could target and potentially MUO and might be in the focus of future research. answer this question [22]. For the first time, we report Second, different definitions of metabolic health have data on differences in QoL between MHO and MUO been used in previous studies: in both studies that found after weight loss. no difference in the magnitude of metabolic change MUO demonstrated a stronger and prolonged benefit between MHO and MUO, group allocation followed the in QoL, particularly in physical components of health. By absence or presence of metabolic syndrome, respectively definition, MUO have a higher number of comorbidities [11, 12]. In the other studies, group definition was done than MHO—at least when it comes to metabolic disor- according to one of the following: absence or presence of ders. It is likely that the strong improvement of metabolic metabolic syndrome [7, 15], presence of insulin resistance parameters in MUO contributes not only to ameliora- measured by oral glucose tolerance test [8] or hyperinsu- tion of comorbidities but consequently also to improved linemic-euglycemic clamp [7, 9]. However, it remained physical health. The fact that MHO showed only short- unclear, whether the differential findings in previous term improvements in metabolism and QoL, while MUO studies are caused by the specific MHO/MUO definition demonstrated long-term improvements in both, allows used in each trial [7–9, 11, 12, 15]. We present data with the interpretation that improvement of metabolic param- three different group allocation definitions (by presence eters may be associated with improved physical health. of metabolic syndrome (IDF definition), by HOMA-IR, Our data suggests that mental health is not as closely and by ISI ). On the whole, the results of all used clas- linked to weight and weight loss as physical health. In the Clamp sifications correspond well, confirming the findings of long term, mental health was not improved irrespective Janiszewski et al. who also used and compared two defi - of metabolic status. The predominant effect of weight nition systems [7]. Nevertheless, in our study, differences loss on physical health is in line with results from other between MUO and MHO appear to be partially greater lifestyle intervention trials. In the Look AHEAD trial, in those parameters that were used for group allocation. the lifestyle intervention led to an improved physical but While MUO classified by HOMA-IR or ISI showed not mental function in the SF-36 over the course of the Clamp benefits preferentially in the glucose metabolism, the IDF study [23]. Actually, the Look AHEAD trial included only classification revealed different benefits in MUO within overweight and obese subjects already diagnosed with the glucose and lipid metabolism as well. Overall how- type 2 diabetes. It makes this study population to some ever, the similarities in the results of both classifications extent comparable to our metabolically ill MUO sub- outweigh the differences—especially regarding QoL, the group, although a preexisting diabetes was not manda- groupings generally allow the same conclusion. So, group tory in our MUO subjects. However, data regarding the allocation by the highly sensitive, but vastly impractical effect of an identical weight loss/weight maintenance hyperinsulinemic-euglycemic clamp is not necessary in intervention on estimates of QoL in MUO and MHO clinical practice. has currently not yet been available, even if QoL has Soll et al. Nutrition & Metabolism (2022) 19:25 Page 8 of 10 developed into an important outcome measure of weight were predominantly female, a distribution frequently loss therapies in general [24, 25]. It is of particular rele- found in lifestyle intervention trials. Although we statisti- vance for real-world settings, where changes in physical cally considered sex as an interfering factor in our linear function strongly affect daily life and will therefore result model, we cannot rule out a bias by the sex imbalance. in improved well-being. Moreover, intended improve- Furthermore, behavioral factors were not considered ment of QoL can be easily communicated to patients in this analysis, even though they are known to influ - and will represent a substantial motivation to participate ence the long-term outcome of weight loss interventions in weight loss programs. Therefore, our data gives new [29]. Moreover, our trial suffered from a limited number insight in the weight loss-induced long-term benefits in of dropouts so that after 21  months only 112 from the different obesity subgroups. Interestingly, these effects original 143 participants could be analyzed. Unfortu- could apparently not be explained by differences in body nately, only general reasons for dropout from this study weight changes, as both groups demonstrated a compara- were enquired. However, dropouts from a lifestyle inter- ble weight course throughout the study. vention trial including a placebo group are not unex- Taken together, our data implies that MUO will ben- pected. A comparison of weight loss induced changes efit stronger in short- and long-term from weight loss of metabolic and anthropometric parameters revealed interventions regarding several health-related outcomes. comparable improvement in dropouts and subjects who uTh s, identification of obese patients with metabolic completed the study (data not shown). We present data impairments might be a crucial strategy to increase the regarding obesity, metabolism, and clinical relevant out- general efficacy of weight loss interventions in the con - comes like QoL. Nevertheless, effects on cardiovascular text of health benefits. Although this was not directly morbidity and mortality were out of scope of this study addressed by our study, this assumption is supported as it was primarily conceived to analyze anthropometric by data from a previous meta-analysis investigating the and metabolic outcomes after a lifestyle intervention. The effect of weight loss on all-cause mortality: Harrington strengths of our study include the large sample size with et  al. have shown that unhealthy obese lower their all- a long observation period, especially compared to previ- cause mortality by intentional weight loss while healthy ous studies. On top of that, we present data from com- obese appear to have an unaltered all-cause mortality prehensive phenotyping at several time-points including after weight loss [26]. It is also in line with results of a the assessment of insulin sensitivity by the hyperinsuline- large French cohort study that recently demonstrated mic-euglycemic clamp as well as highly relevant patient- a substantially higher risk for cardiovascular events in reported outcomes of QoL. Additionally, we present data MUO compared to MHO [27]. However, most of the employing and comparing three different definitions for beneficial metabolic long-term effects in MUO shown metabolic health. in our study marginally failed to be significant in inter- group comparisons to MUO. This might be caused by the limited sample size and the high variability of individual Conclusions long-term changes. Overall, our data demonstrate that MUO have a greater Interestingly, adjustment for improvement of differ - benefit in health-related QoL and several metabolic ent estimates of obesity during weight loss indicated parameters than MHO not only in the short-term but that additional mechanisms apart from changes in body also in the long-term course after weight loss. However, it weight appear to play a relevant role for improvement of is necessary to remember that MHO still have increased insulin sensitivity during weight loss in MUO. This might rates of incident diabetes, heart failure, and mortal- include inflammatory, metabolic, or hormonal factors ity compared to metabolically healthy non-obese [30]. such as leptin [28], adiponectin [28], or atrial natriuretic There might still be benefits from weight loss interven - peptides [16]. Future research investigating the underly- tions regarding effects on the musculoskeletal system, ing mechanism of the stronger improvement of insulin among others, which were not addressed in this study. A resistance in MUO is warranted. recent review on this subject also points out that meta- Our study comes with obvious limitations. First, the bolically healthy obesity might represent a transitional primary aim of the presented MAINTAIN trial was the state for some patients that might turn into metabolically comparison of different treatment strategies to achieve unhealthy obesity at one point [31]. u Th s, lifestyle modi - body weight maintenance. Thus, we defined the groups fication should still be recommended for obese patients for this study retrospectively. However, MUO and MHO in general. In settings of limited resources the MUO underwent the same procedures and we implemented might however become the preferential target group for potential relevant confounders including the treatment lifestyle interventions. group in our statistical analyses. Participants in our trial S oll et al. Nutrition & Metabolism (2022) 19:25 Page 9 of 10 Abbreviations References BMI: Body mass index; HbA1c: Hemoglobin A1c; HDL: High‑ density lipopro‑ 1. World Health Organization. Obesity and overweight. 2021. Available tein; HOMA‑IR: Homeostatic model assessment for insulin resistance; IDF: from: https:// www. who. int/ news‑ room/ fact‑ sheets/ detail/ obesi ty‑ and‑ Internationale Diabetes Federation; ISI : Insulin sensitivity index (hyperin‑overw eight Clamp sulinemic‑ euglycemic clamp); LDL: Low‑ density lipoprotein; MHO: Metaboli‑ 2. Menke A, Casagrande S, Geiss L, Cowie CC. Prevalence of and trends cally healthy obese; MUO: Metabolically unhealthy obese; OGTT : Oral glucose in diabetes among adults in the United States, 1988–2012. JAMA. tolerance test; QoL: Quality of life; SF‑36: Short Form (36) Health Survey; WHO: 2015;314(10):1021. World Health Organization. 3. Yumuk V, Tsigos C, Fried M, Schindler K, Busetto L, Micic D, et al. European guidelines for obesity management in adults. Obes Facts. 2015;8(6):402–24. Supplementary Information 4. Tuomilehto J, Lindström J, Eriksson JG, Valle TT, Hämäläinen H, Ilanne‑ The online version contains supplementary material available at https:// doi. Parikka P, et al. Prevention of type 2 diabetes mellitus by changes in org/ 10. 1186/ s12986‑ 022‑ 00660‑w. lifestyle among subjects with impaired glucose tolerance. N Engl J Med. 2001;344(18):1343–50. 5. Stefan N, Häring H‑U, Schulze MB. Metabolically healthy obesity: the Additional file 1. Electronic supplementary material. low‑hanging fruit in obesity treatment? Lancet Diabetes Endocrinol. 2018;6(3):249–58. 6. Karelis AD, Faraj M, Bastard J‑P, St ‑Pierre DH, Brochu M, Prud’homme D, Acknowledgements et al. The Metabolically healthy but obese individual presents a favorable We thank K. Simon, B. Horchler, N. Huckauf, and C. Kalischke for excellent tech‑ inflammation profile. J Clin Endocrinol Metab. 2005;90(7):4145–50. nical assistance as well as A. Reisshauer for the support regarding to physical 7. Janiszewski PM, Ross R. Eec ff ts of weight loss among metabolically activity intervention. We thank Nestlé HealthCare Nutrition GmbH, Frankfurt healthy obese men and women. Diabetes Care. 2010;33(9):1957–9. am Main, Germany for the opportunity to purchase the Optifast 2 diet at a 8. Kantartzis K, Machann J, Schick F, Rittig K, Machicao F, Fritsche A, et al. reduced price. Eec ff ts of a lifestyle intervention in metabolically benign and malign obesity. Diabetologia. 2011;54(4):864–8. Authors’ contributions 9. Karelis AD, Messier V, Brochu M, Rabasa‑Lhoret R. Metabolically healthy K.M., D.S., and J.S. researched data and wrote the manuscript. D.S., K.M., and but obese women: effect of an energy‑restricted diet. Diabetologia. J.G. were responsible for data analysis. All authors contributed to interpreta‑ 2008;51(9):1752–4. tion of the results. All authors critically read and edited several drafts before 10. Mai K, Brachs M, Leupelt V, Jumpertz‑ von Schwartzenberg R, Maurer submission. All authors read and approved the final manuscript. L, Grüters‑Kieslich A, et al. Eec ff ts of a combined dietary, exercise and behavioral intervention and sympathetic system on body weight mainte‑ Funding nance after intended weight loss: results of a randomized controlled trial. Open Access funding enabled and organized by Projekt DEAL. This research Metabolism. 2018;83:60–7. was supported by the Deutsche Forschungsgemeinschaft (DFG KFO 218/1), 11. Ruiz JR, Ortega FB, Labayen I. A weight loss diet intervention has a similar the German Diabetes Society (DDG) and the German Ministry for Education beneficial effect on both metabolically abnormal obese and meta‑ and Research (BMBF) by support of the German Centre for Cardiovascular bolically healthy but obese premenopausal women. Ann Nutr Metab. Research (DZHK, BER5.1). 2013;62(3):223–30. 12. Dalzill C, Nigam A, Juneau M, Guilbeault V, Latour E, Mauriège P, et al. Data availability Intensive lifestyle intervention improves cardiometabolic and exercise The datasets generated during and/or analyzed during the current study are parameters in metabolically healthy obese and metabolically unhealthy available from the corresponding author on reasonable request. obese individuals. Can J Cardiol. 2014;30(4):434–40. 13. Ninomiya JK, L’italien G, Criqui MH, Whyte JL, Gamst A, Chen RS,. Associa‑ tion of the metabolic syndrome with history of myocardial infarction Declarations and stroke in the third national health and nutrition examination survey. Circulation. 2004;109(1):42–6. Ethics approval and consent to participate 14. Klöting N, Fasshauer M, Dietrich A, Kovacs P, Schön MR, Kern M, et al. Insu‑ The study protocols were approved by the Institutional Review Board of the lin‑sensitive obesity. Am J Physiol Endocrinol Metab. 2010;299(3):E506–15. Charité Medical School (EA1/140/12) and all subjects gave written informed 15. Shin M‑ J, Hyun YJ, Kim OY, Kim JY, Jang Y, Lee JH. Weight loss effect on consent. All analyses were conducted in accordance with the Declaration of inflammation and LDL oxidation in metabolically healthy but obese Helsinki. The trial was registered at ClinicalTrials.gov (NCT00850629). (MHO) individuals: low inflammation and LDL oxidation in MHO women. Int J Obes. 2006;30(10):1529–34. Consent for publication 16. Brachs M, Wiegand S, Leupelt V, Ernert A, Kintscher U, von Schwarzenberg Not applicable. RJ, et al. ANP system activity predicts variability of fat mass reduction and insulin sensitivity during weight loss. Metabolism. 2016;65(6):935–43. Competing interests 17. Mai K, Li L, Wiegand S, Brachs M, Leupelt V, Ernert A, et al. An Integrated We declare that there is no conflict of interest that could be perceived as understanding of the molecular mechanisms of how adipose tissue prejudicing the impartiality of the research reported. metabolism affects long‑term body weight maintenance. Diabetes. 2019;68(1):57–65. Author details 1 18. International Diabetes Federation. The IDF consensus worldwide defini‑ Department of Endocrinology and Metabolism, Charité – Universitätsmedizin tion of the metabolic syndrome. 2021. Available from: https:// www. idf. Berlin, corporate member of Freie Universität Berlin, Humboldt‑Universität Zu org/e‑ libra ry/ conse nsus‑ state ments/ 60‑ idfco nsens us w‑orld wide‑ defin Berlin, and Berlin Institute of Health, Chariteplatz 1, 10117 Berlin, Germany. 2ition of‑ the‑ metab olic‑ syndr ome. html Charité Center for Cardiovascular Research, Charité – Universitätsmedizin 19. Ellert U, Kurth B. Methodische Betrachtungen zu den Summenscores Berlin, corporate member of Freie Universität Berlin, Humboldt‑Universität 3 des SF‑36 anhand der erwachsenen bundesdeutschen Bevölkerung. Zu Berlin, and Berlin Institute of Health, 10117 Berlin, Germany. DZHK (Ger‑ Bundesgesundheitsblatt ‑ Gesundheitsforschung ‑ Gesundheitsschutz. man Centre for Cardiovascular Research), partner site Berlin, Berlin, Germany. 4 2004. NutriAct‑ Competence Cluster Nutrition Research Berlin‑Potsdam, Nuthetal, 20. Gutch M, Kumar S, Razi SM, Gupta KK, Gupta A. Assessment of insulin Germany. sensitivity/resistance. Indian J Endocrinol Metab. 2015;19(1):160–4. 21. Li L, Spranger L, Soll D, Beer F, Brachs M, Spranger J, et al. Metabolic Received: 15 November 2021 Accepted: 16 March 2022 impact of weight loss induced reduction of adipose ACE‑2 ‑ Potential implication in COVID‑19 infections? Metabolism. 2020;113:154401. Soll et al. Nutrition & Metabolism (2022) 19:25 Page 10 of 10 22. Haigh JW, Johnston DG, McCulloch AJ, Laker MF, Welby J, Evans S. Assess‑ ment of glucose turnover in normal man with the use of a non‑radioac‑ tive isotopically labelled preparation, [6,6–2H]glucose, as tracer. Clin Sci (Lond). 1982;63(5):437–40. 23. Rubin RR, Wadden TA, Bahnson JL, Blackburn GL, Brancati FL, Bray GA, et al. Impact of intensive lifestyle intervention on depression and health‑ related quality of life in type 2 diabetes: the look AHEAD trial. Diabetes Care. 2014;37(6):1544–53. 24. Omotosho P, Mor A, Shantavasinkul PC, Corsino L, Torquati A. Gastric bypass significantly improves quality of life in morbidly obese patients with type 2 diabetes. Surg Endosc. 2016;30(7):2857–64. 25. Poelemeijer YQM, van der Knaap ETW, Marang‑ van de Mheen PJ, Demirkiran A, Wiezer MJ, Hazebroek EJ, et al. Measuring quality of life in bariatric surgery: a multicentre study. Surg Endosc. 2020;34(12):5522–32. 26. Harrington M, Gibson S, Cottrell RC. A review and meta‑analysis of the effect of weight loss on all‑ cause mortality risk. Nutr Res Rev. 2009;22(1):93–108. 27. Fauchier G, Bisson A, Bodin A, Herbert J, Semaan C, Angoulvant D, et al. Metabolically healthy obesity and cardiovascular events: a nationwide cohort study. Diabetes Obesity Metab. 2021. https:// doi. org/ 10. 1111/ dom. 14492 28. Fasshauer M, Blüher M. Adipokines in health and disease. Trends Pharma‑ col Sci. 2015;36(7):461–70. 29. Del Corral P, Bryan DR, Garvey WT, Gower BA, Hunter GR. Dietary adher‑ ence during weight loss predicts weight regain. Obesity (Silver Spring). 2011;19(6):1177–81. 30. Zhou Z, Macpherson J, Gray SR, Gill JMR, Welsh P, Celis‑Morales C, et al. Are people with metabolically healthy obesity really healthy? A prospec‑ tive cohort study of 381,363 UK Biobank participants. Diabetologia. 2021;64(9):1963–72. 31. Blüher M. Metabolically healthy obesity. Endocr Rev. 2020;41(3):405–20. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nutrition & Metabolism Springer Journals

Long-term impact of the metabolic status on weight loss-induced health benefits

Loading next page...
 
/lp/springer-journals/long-term-impact-of-the-metabolic-status-on-weight-loss-induced-health-inRI3QJeuJ

References (43)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2022
eISSN
1743-7075
DOI
10.1186/s12986-022-00660-w
Publisher site
See Article on Publisher Site

Abstract

Background: While short‑term effects of weight loss on quality of life and metabolic aspects appear to be different in metabolically healthy (MHO) and metabolically unhealthy obese (MUO), respective long‑term data is still missing. Given the high relevance of long‑term changes, we aimed to address these in this post ‑hoc analysis of the MAINTAIN trial. Methods: We analyzed 143 overweight/obese subjects (BMI ≥ 27 kg/m , age ≥ 18 years) before and after a 3‑month weight loss program (≥ 8% weight loss), after a 12‑month period of a randomized weight maintenance intervention (n = 121), and after another 6 months without intervention (n = 112). Subjects were retrospectively grouped into MHO and MUO by the presence of metabolic syndrome and secondarily by estimates of insulin sensitivity (HOMA‑ IR and ISI ). Quality of life (QoL), blood pressure, lipids, HOMA‑IR, and ISI were assessed and evaluated using Clamp Clamp mixed model analyses. Results: Despite similar short‑ and long‑term weight loss, weight loss‑induced improvement of HOMA‑IR was more pronounced in MUO than MHO after 3 months (MHO: 2.4[95%‑ CI: 1.9–2.9] vs. 1.6[1.1–2.1], p = 0.004; MUO: 3.6[3.2–4.0] vs. 2.0[1.6–2.4], p < 0.001; p = 0.03 for inter‑ group comparison). After 21 months, the beneficial effect was no longer seen in MHO (2.0[1.5–2.6], p = 1.0), while it remained partially preserved in MUO (2.9[2.4–3.3], p = 0.002). QueryShort‑ term improvements of lipid parameters were similar in both groups. However, long‑term improvements of HDL ‑ cholesterol and triglycerides were only seen in MUO (44.4[41.5–47.4] vs. 49.3[46.2, 52.3] mg/dl, p < 0.001; 176.8[158.9– 194.8] vs. 138.8[119.4–158.3] mg/dl, p < 0.001, respectively) but not in MHO. Weight loss‑induced improvements in the QoL and particularly the physical health status were maintained in MUO until the end of the trial, while benefits disappeared over time in MHO. Group allocation by HOMA‑IR and ISI revealed higher benefits for MUO mainly in Clamp parameters of the glucose metabolism and QoL. Conclusions: Our data demonstrates stronger and longer‑lasting improvements of metabolism and QoL in MUO after weight loss. Trial registration (ClinicalTrials.gov): NCT00850629. Registered 25 February 2009, https:// clini caltr ials. gov/ ct2/ show/ NCT00 850629. Keywords: Healthy obesity, Weight loss, Insulin sensitivity, Quality of life Introduction Obesity is a chronic, systemic, and multifactorial disease *Correspondence: knut.mai@charite.de and is defined—according to the world health organiza - Department of Endocrinology and Metabolism, Charité – 2 tion (WHO)—by a body mass index (BMI) of 30  kg/m Universitätsmedizin Berlin, corporate member of Freie Universität or more. Despite its preventable nature, the prevalence Berlin, Humboldt‑Universität Zu Berlin, and Berlin Institute of Health, Chariteplatz 1, 10117 Berlin, Germany of obesity has been increasing over the last decades [1]. Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Soll et al. Nutrition & Metabolism (2022) 19:25 Page 2 of 10 Obesity is associated with several musculoskeletal, meta- and special health care providers in clinical practice. To bolic, and cardiovascular diseases, amongst others [2]. evaluate the impact of alternative definitions of metabolic Weight loss has been shown to improve the obesity- health, we performed comparable analyses after classifi - related metabolic disturbances and is therefore recom- cation of MUO and MHO using the HOMA-IR and the mended in the European guideline for the management hyperinsulinemic-euglycemic clamp as an easily estima- of obesity, among others [3]. In general, positive effects ble and a more precise determinant of insulin resistance, on metabolic diseases, such as lower incidence of type respectively. 2 diabetes and increased insulin sensitivity, have been shown for a weight loss of 5–8% [4, 5]. Given the ris- Study design and methods ing number of obese subjects and limited resources, the Participants and study design identification of subjects who will benefit most from The study (MAINTAIN trial, ClinicalTrials.gov registry weight loss interventions may help to optimize current number NCT00850629) was performed between 2010 treatment strategies. and 2016 in Berlin, Germany. Detailed trial informa- More than 15 years ago, a subset of obese patients has tion has been reported previously [10, 16, 17]. The flow been identified that is not affected by the mentioned of participants is shown in Additional file  1: Figure S1. In metabolic changes that frequently go along with obesity. short, 156 overweight and obese subjects (BMI ≥ 27  kg/ Primarily, they remain insulin sensitive despite being m , age ≥ 18  years) underwent a 3-month-weight loss obese [6]. Current data indicates that acute metabolic program, realized by caloric restriction using a very low benefits of weight loss were more pronounced [7] or energy diet for 8 weeks (replacement of all meals by a for- did only occur in metabolically unhealthy obese (MUO) mula diet with 800  kcal per day) and an energy-reduced [8, 9], but not in metabolically healthy obese (MHO). diet (healthy food diet with approximately 1500 kcal per Although this data supports a metabolic improvement day) for the following 4  weeks, both accompanied by by dietary weight loss primarily in MUO, long-term data nutritional counseling, physical exercises, and psycho- on the different or comparable effectiveness of tempo - logical advice. Subjects that experienced a prior weight rary lifestyle interventions for MHO and MUO is not yet loss of more than 5  kg in the last 2  months, changed available. This is crucial as the impact of temporary life - their smoking habits or diets within the last 3  months, style interventions on weight loss is often frustrating due were pregnant, or suffered from other endocrine disor - to frequently observed weight regain [10]. ders, eating disorders, or severe chronic diseases were Although the classification of MHO and MUO is fre - not included in this trial. All participants who succeeded quently based on the presence or absence of the meta- to lose at least 8% of their body weight (n = 143) were bolic syndrome [7, 11, 12], this still remains inconsistent then randomized into an intervention or control group and is a matter of debate. The International Diabetes Fed - for 12  months of weight maintenance. The intervention eration (IDF) defines the metabolic syndrome as a spe - comprised a continuous multimodal counseling focus- cific cluster of medical conditions with central obesity, ing on caloric restriction, nutritional counseling, physi- insulin resistance, hypertension, and dyslipidemia [13]. cal exercises, and psychological support. In contrast, the Nevertheless, the presence of insulin resistance in fast- control group was under free living conditions without ing state or during an oral glucose tolerance test, or the further counseling. Following the randomized interven- results of a hyperinsulinemic-euglycemic clamp are also tion period, both groups underwent a 6-month-follow- used to define metabolic health [7, 8, 14, 15]. The added up period without any further intervention. More details value of these definitions of metabolic health is currently regarding the lifestyle intervention can be found in the still unknown. Additional file 1. Given this lack of evidence, we report short- and long- Retrospectively, all randomized subjects were allo- term data from a 21-month weight loss/weight mainte- cated to be MHO or MUO depending on the absence nance trial focusing on metabolic healthy and unhealthy or presence of metabolic disturbances. In this regard, obese. In detail, we intend to elucidate whether a life- three different definitions were used. First, subjects were style-based weight loss intervention demonstrates differ - categorized in accordance to the criteria of the meta- ent efficacies regarding body weight reduction, metabolic bolic syndrome as defined by the IDF [18]: central obe - improvement, and quality of life (QoL) in obese subjects sity defined as waist circumference ≥ 94  cm for men differing in their metabolic health status. The efficacy and ≥ 80  cm for women plus any two of the following was especially assessed in the long term. In order to do four factors: (1) elevated triglycerides level > 150  mg/dl so, we used the IDF criteria from 2005 for metabolic syn- or intake of lipid-lowering drugs, (2) reduced HDL cho- drome to separate MUO from MHO individuals as they lesterol < 40  mg/dl for men and < 50  mg/dl for women, are well-established and easily applicable for primary (3) raised blood pressure ≥ 130  mmHg for systolic S oll et al. Nutrition & Metabolism (2022) 19:25 Page 3 of 10 or ≥ 85  mmHg for diastolic blood pressure or intake of surface per minute and of glucose at an individual rate antihypertensive drugs, (4) raised fasting plasma glu- to fix plasma glucose levels at 80  mg/dl ± 8  mg/dl. The cose ≥ 100 mg/dl or previously diagnosed type 2 diabetes. steady state was defined as plasma glucose levels of Next, subjects were allocated to be MHO/MUO by their 80  mg/dl ± 8  mg/dl for at least 30  min. The ISI was Clamp respective HOMA-IR before weight loss. Here, subjects calculated by dividing the glucose infusion rate at steady with a HOMA-IR ≥ median HOMA-IR (~ 2.19) were dis- state (mg/min) per body weight (kg) by plasma insulin at tributed to MUO, subjects with a HOMA-IR < median steady state (mU/l) [20]. HOMA-IR to MHO. Last, subjects were allocated by their respective insulin sensitivity index (ISI ) before Statistical analysis Clamp weight loss. Here, subjects with an I SI ≥ median Data of 143 subjects who lost at least 8% of their initial Clamp −1 −1 −1 ISI (~ 0.058 mg  kg  min/(mU  l )) were defined body weight was analyzed. The primary analysis was per - Clamp as MHO, subjects with a I SI < median ISI were formed using group allocation according to IDF criteria. Clamp Clamp classified as MUO. Due to the lack of reliable cut-offs for Statistical analysis was performed using RStudio Version a “pathological” HOMA-IR or I SI , the group was 1.2.1335 (RStudio Inc., Boston, MA) and the R software Clamp divided by the respective median. package. We performed a linear mixed effects analy - sis of the relationship between anthropometric/meta- Procedures bolic parameters and time using the nlme package. For Fasting blood sampling as well as measurements of blood group comparisons, additional analysis of the relation- pressure, waist circumference, body weight, height, and ship between anthropometric/metabolic parameters and assessment of QoL were performed before ( T ) and after the interaction of time and group affiliation was done. -3 (T ) weight loss, 12  months ( T ) after randomization We entered age, sex, and randomization state (interven- 0 12 and after another 6  months without active intervention tion group or control group) as fixed effects and poten - (T ). At T, T , and T , subjects underwent a hyperin- tial confounders. In order to adjust for BMI change in 18 -3 0 12 sulinemic-euglycemic clamp. the course of the study, we repeated this analysis after Blood samples were centrifuged; plasma and serum additional inclusion of either BMI, waist circumference, samples were frozen immediately at − 80  °C. Glucose or body fat mass at all time-points as a fixed effect. Inter - was measured using the glucose oxidase method (Dr. cepts for subjects were considered as a random effect. Müller Super GL, Freital, Germany). HbA1c was meas- Results were considered to be significant, if the two-sided ured by high-performance liquid chromatography using α was below 0.05. Presented data represents estimated the VARIANT II (Bio-Rad, Hercules, US). Lipids were marginal mean and 95%-confidence interval (CI). P val - measured by standard laboratory methods using Cobas ues were obtained from comparisons using the emmeans ISE direct and c111 Analyzer (Roche Diagnostics, Man- package. Adjustment for multiple testing was performed nheim, Germany). Serum insulin was measured by by Bonferroni correction. fluoroimmunometric assay (AutoDelfia; Perkin Elmer, Rodgau, Germany). Fat mass was assessed by bioelectri- Results cal impedance analysis using the AKERN BIA 101 (SMT Clinical and metabolic characteristics at baseline medical GmbH & Co. KG, Würzburg, Germany). General data on the trial population has already been In order to assess potential changes in quality of life published previously [10, 16, 21]. Respective relevant data and patient-reported health status that might accompany can be found in Additional file  1: Table  S1. To eliminate weight loss and maintenance, participants were asked to the effect of interfering factors, age, sex, and randomi - fill in the Short Form (36) Health Survey (SF-36). It con - zation state were included as fixed effects in the mixed tains 36 items in eight subscale scores: general health effect model analysis for group comparisons. Using IDF perceptions, physical functioning, role limitations due to criteria, there were significant differences between both physical problems, bodily pain, mental health, role limi- MHO and MUO at T for waist circumference (Δ(MHO- -3 tations due to emotional problems, vitality, and social MUO) − 6.5 [95% CI: − 10.4, − 2.6] cm, p = 0.001) and functioning. A physical component summary score (PCS) BMI (− 2.2 [− 4.2, − 0.2] kg/m , p = 0.03). Consistent is derived from the first four, a mental component sum - with the parameters for allocation, both groups also dif- mary score (MCS) from the latter four components [19]. fered in triacylglycerol levels (− 57.7 [− 84.3, − 31.1] mg/ The insulin resistance index HOMA-IR was calculated dl, p < 0.001), HDL cholesterol (11.1 [6.8, 15.3] mg/dl, as previously described: fasting glucose (mg/dl) × fasting p < 0.001), fasting glucose (− 12.6 [− 16.9, − 8.2] mg/dl, insulin (mU/l)/405 [20]. p < 0.001), and HbA1c (− 0.42 [− 0.65, − 0.18], p < 0.001). The hyperinsulinemic-euglycemic clamp involved Additionally, MHO and MUO differed in the estimates infusion of insulin at the rate of 40  mU per m of body for insulin resistance HOMA-IR (− 1.3 [− 1.9, − 0.7], Soll et al. Nutrition & Metabolism (2022) 19:25 Page 4 of 10 −1 −1 Improved quality of life persists in MUO but not in MHO p < 0.001) and ISI (0.03 [0.02, 0.04] mg  kg  min / Clamp −1 At baseline, MHO had a higher physical health status (mU  l ), p < 0.001), but in none of the other presented than MUO (Δ(MHO-MUO) 4.8 [− 0.5, 10.2], p = 0.005), parameters. (Table 1). while mental health status did not differ (− 1.5 [− 7.5, 4.6], p = 0.4). After initial weight loss, the MUO reported Short‑term effects of weight loss on metabolism an improved physical and mental health status, while During the initial 3-month weight reduction period, the MHO indicated an improved physical but not men- weight loss was similar in MHO and MUO (− 4.6 [− 5.4, 2 2 tal health (Fig.  1A, B). Physical health status remained − 3.7] kg/m vs. − 4.7 [− 5.4, − 4.0] kg/m , p = 1.0). How- improved in MUO in the long term while the improve- ever, triglycerides and fasting glucose declined only in ment of mental health disappeared after 12  months. In MUO, but not in MHO. (Table  1) While HOMA-IR was contrast, the short-term increase of physical health in significantly more reduced in MUO than in MHO (− 1.7 MHO was already reversed after 12 months (Fig. 1A, B). [− 2.2, − 1.1] vs. − 0.8 [− 1.4, − 0.2], p = 0.03), HDL cho- When groups are separated by HOMA-IR, the results lesterol was less modified by weight loss in MUO (− 2.7 regarding quality of life are almost similar (Fig.  1C, D), [− 5.2, − 0.1] vs. -6.6 [− 9.6, − 3.5] mg/dl, p = 0.06), but although improvement of physical health persisted up the difference did not reach significance. In contrast, to 12  months after weight loss in MHO. Comparable comparable improvements were detected in both groups changes and differences were found for group allocation in ISI , total cholesterol, and LDL cholesterol, as well Clamp by ISI (Fig. 1E, F). Clamp as systolic and diastolic blood pressure. Overall, group allocation based on HOMA-IR showed a stronger improvement in MUO only regarding fast- Improved insulin resistance in MUO is not exclusively ing glucose and HOMA-IR (which was not modified in explained by weight loss MHO) but not in lipids during the weight loss period Considering the differential development of insulin (Additional file  1: Table  S2). Comparably, group alloca- resistance in MHO and MUO despite comparable weight tion according to ISI revealed almost similar find - loss, we aimed to further investigate the interrelation of Clamp ings with different developments between groups only in improved obesity and insulin resistance. Hence, to elimi- HbA1c and HOMA-IR (Additional file 1: Table S3). nate the impact of body weight changes, we addition- ally adjusted the improvement of HOMA-IR for BMI. Long‑term effects of weight loss on metabolism Interestingly, while the MHO did not present changes After initial weight loss, participants showed a mod- in HOMA-IR anymore, it still was signifi - BMI-adjusted erate increase in BMI, but still maintained a signifi - cantly declined by weight loss in MUO in the short- and cantly lower body weight until the end of the study after medium-term (Fig.  2). Likewise, adjustment for waist 18 months. There was no difference in the weight course circumference or body fat mass gave comparable results between both, MUO and MHO (Table  1). In analogy, (Additional file 1: Figure S2). trajectory of waist circumference—as an established estimate of visceral obesity—was equal in both groups. Discussion Similarly, changes of ISI were comparable in MUO Clamp In the presented study, we aimed to answer the ques- and MHO. In contrast, in comparison to baseline MUO tion whether MHO and MUO subjects gain metabolic demonstrated improved HOMA-IR, triglycerides, and and health benefits to different extents from weight loss/ HDL cholesterol for up to 18 months and improved fast- weight maintenance by a multimodal lifestyle inter- ing glucose, total cholesterol, and LDL cholesterol until vention. There are few studies that already addressed 12  months after weight loss (Table  1). However, these this issue in a short-term context, but were partially parameters were no longer improved in MHO at T or not consistent in their results [7–9, 11, 12, 15]. On the T compared to baseline. Nevertheless, between-group one hand, Dalzill et  al. and Ruiz et  al. found a similar comparisons for ΔT T did not reveal any significant 3 18 improvement in HOMA-IR and several other meta- differences in these parameters. bolic parameters in MUO and MHO investigating 124 Group allocation depending on HOMA-IR revealed a patients before and after a 9-month intensive lifestyle long-term improvement of HOMA-IR in MUO but not modification program and 78 postmenopausal women in MHO (p < 0.001 for inter-group comparison) (Addi- before and after a 3-month energy-restricted treatment, tional file  1: Table S2). A similar development of HOMA- respectively [11, 12]. On the other hand, several previ- IR in both groups is found if groups are separated by ous studies described beneficial effects preferentially for ISI (Additional file  1: Table  S3). However, this clas- Clamp MUO. While Janiszewski et al. studied 106 adults before sification revealed additional differences in the long-term and after various 3-to-6-month lifestyle interventions courses of BMI and ISI between both groups. Clamp S oll et al. Nutrition & Metabolism (2022) 19:25 Page 5 of 10 Table 1 Estimates of clinical parameters of MHO/MUO following the IDF definition of metabolic syndrome Metabolically Healthy Obese (MHO) Metabolically Unhealthy Obese (MUO) p value (MHO vs MUO) T T T T T T T T T → T T → T T → T 3 0 12 18 3 0 12 18 3 0 3 12 3 18 n: m/w 59: 9/50 59: 9/50 46: 7/39 44: 6/38 84: 22/62 84: 22/62 75: 19/56 68: 19/49 Randomization: 29/30 42/42 control/interven tion Age [years] 48.6 [45.2,52.0] 51.9 [49.2,54.5] BMI [kg/m2] 36.0 [34.2,37.7] 31.4 32.5 33.5 38.2 [36.8,39.6] 33.5 34.1 35.2 1 1 1 [29.7,33.1]*** [30.8,34.3]*** [31.7,35.2]*** [32.1,34.9]*** [32.7,35.6]*** [33.8,36.6]*** Waist circumfer 107.6 99.4 100.9 101 114 103.8 104.4 106 0.6 0.14 1 ence [cm] [104.2,110.9] [96,102.7]*** [97.5,104.4]*** [97.5,104.5]*** [111.3,116.8] [101.1,106.5]*** [101.6,107.2]*** [103.2,108.8]*** Serum glucose 82.4 [78.8,86] 80.8 [77.2,84.4] 84.3 [80.4,88.2] 86.9 [83.0,90.9] 95.0 [92.0,97.9] 86.1 89.7 [86.6,92.8]** 93.2 [90.1,96.3] 0.01 0.02 0.06 [mg/dl] [83.1,89.0]*** HbA1c [%] 5.5 [5.3,5.7] 5.9 [5.7,6]** 5.4 [5.2,5.6] 5.5 [5.3,5.7] 5.9 [5.8,6.1] 5.9 [5.8,6.1] 5.6 [5.4,5.8]*** 5.7 [5.6,5.9] 0.02 0.66 1 HOMAIR 2.4 [1.9,2.9] 1.6 [1.1,2.1]** 1.9 [1.3,2.4] 2.0 [1.5,2.6] 3.6 [3.2,4] 2.0 [1.6,2.4]*** 2.3 [1.9,2.7]*** 2.9 [2.4,3.3]** 0.03 0.06 0.91 ISI 0.08 [0.07,0.09] 0.10 0.09 [0.08,0.10]* 0.05 [0.04,0.06] 0.08 0.07 0.31 0.23 Clamp −1 −1 [mg kg min / [0.09,0.11]*** [0.07,0.09]*** [0.07,0.08]*** −1 (mU l )] Total cholesterol 200.4 [190,210.9] 171.2 192.5 192.4 198.1 167.5 188.8 191 [182.1,199.8] 1 1 1 [mg/dl] [160.8,181.6]*** [181.4,203.7] [181.3,203.6] [189.7,206.5] [159.1,175.9]*** [180.1,197.5]* LDL Cholesterol 121.2 102.1 114.6 [105,124.1] 117.2 121.8 103.4 113.7 118.4 [110.8,126] 1 1 1 [mg/dl] [112.2,130.2] [93.1,111.1]*** [107.6,126.7] [114.5,129] [96.1,110.6]*** [106.2,121.2]* HDL Cholesterol 55.5 [51.9,59.2] 49 [45.3,52.6]*** 58 [54.2,61.8] 58 [54.2,61.8] 44.4 [41.5,47.4] 41.8 [38.9,44.7]* 49.9 49.3 0.06 0.43 0.99 [mg/dl] [46.9,52.9]*** [46.2,52.3]*** Triglycerides 119.1 [97.0,141.2] 96.1 [74.1,118.2] 95.5 [70.9,120.0] 101.9 176.8 110.1 149.1 138.8 0.01 1 1 [mg/dl] [77.3,126.5] [158.9,194.8] [92.2,128.1]*** [130.2,168.1]* [119.4,158.3]*** Systolic RR 127.6 118.7 121 [115.2,126.7] 121 [115.0,126.9] 132.4 121.4 125 [120.5,129.5] 127.6 1 1 1 [mmHg] [122.4,132.8] [113.6,123.9]* [128.0,136.7] [117.1,125.6]*** [122.9,132.3] Diastolic RR 80.6 [77.8,83.4] 74.1 77.9 [74.8,80.9] 79.8 [76.6,82.9] 80.5 [78.1,82.8] 74.8 79 [76.6,81.4] 78.1 [75.6,80.6] 1 1 1 [mmHg] [71.2,76.9]*** [72.5,77.2]*** Values shown (except for n, randomization, and age) are estimated marginal means with 95% confidence intervals from model with adjustment to sex, age, and randomization p < 0.05, ** p < 0.01, *** p < 0.001 vs. T in respective group -3 Soll et al. Nutrition & Metabolism (2022) 19:25 Page 6 of 10 † C A E 60 60 60 ††† 55 55 55 ** *** * *** *** *** 50 50 50 *** *** *** *** *** ** *** *** 45 45 40 40 40 35 35 35 -3 0 12 18 -3 0 12 18 -3 0 12 18 Time [month] Time [month] Time [month] MHO D F 60 60 60 MUO 55 55 55 ** 50 50 50 *** ** 45 45 45 (MUO) 40 40 40 35 35 35 -3 0 12 18 -3 0 12 18 -3 0 12 18 Time [month] Time [month] Time [month] Fig. 1 Eec ff t of weight loss and maintenance on patient ‑reported quality of life in MHO/MUO. A, B MHO + MUO were assigned following the IDF criteria for metabolic syndrome. C, D MHO + MUO were assigned on the basis of HOMA‑IR. E, F MHO + MUO were assigned on the basis of ISI . Clamp Values represent estimated marginal means with 95% confidence intervals from model with adjustment to sex, age, and randomization. *p < 0.05, **p < 0.01, ***p < 0.001 vs. T in respective group. Where indicated, †/††/††† describe between‑ group comparisons, respectively. SF-36 Short Form ‑3 (36) Health Survey, PCS physical component summary score, MCS mental component summary score from 44 patients before and after a 6-month dietary intervention, described enhanced myocellular insulin sensitivity in MUO, but not in MHO[9]. Similar to our MHO study, most previous studies reported data from second- MUO ** ary analyses. While weight loss ranged from 3 to 9% in *** the aforementioned studies, our patients lost more than 12% during weight loss. The patients’ age ranged from 45 to 60  years in most previous studies, but Ruiz et  al. and Shin et  al. had markedly younger participants of about 0 35 to 40  years [11, 15]. However, there is no single, eas- -3 0 12 18 ily distinguishable parameter—like age, sex, number of Time [month] participants, group definition, extent of weight loss, or Fig. 2 Eec ff t of weight loss and maintenance on insulin resistance duration of intervention—in previous studies explaining (adjusted for BMI) in MHO/MUO. MHO + MUO were assigned why some did find different developments between MHO following the IDF criteria for metabolic syndrome. Values represent and MUO and others did not. Nevertheless, we were able estimated marginal means with 95% confidence intervals from model with adjustment to sex, age, randomization, and BMI at all to address several factors that might have influenced the time‑points. * p < 0.05, ** p < 0.01, *** p < 0.001 vs. T in respective ‑3 discrepancies in the existing results: these include time group points of measurements, group definition, and tech - niques of measuring metabolic changes. They will be dis - cussed in the following paragraphs. and found a larger increase of insulin sensitivity in MUO First, it is important to notice that in all previous stud- after weight loss [7], Kantartzis and colleagues demon- ies only short-term effects were addressed, as the over - strated improved HOMA-IR and glucose levels during all observation periods spanned from 3 to 9 months and an OGTT only in MUO, but not in MHO in a report on measurements took place directly after the intervention, 103 patients before and after a 9-month lifestyle inter- while we were able to track participants in our study vention [8]. Similarly, Karelis et  al., who performed for up to 18  months after weight loss (i.e. 21  months hyperinsulinemic-euglycemic clamps and reported data after study begin). Our study design allows separation SF-36 PCS SF-36 MCS HOMA-IR BMI-adjusted SF-36 PCS SF-36 MCS SF-36 PCS SF-36 MCS S oll et al. Nutrition & Metabolism (2022) 19:25 Page 7 of 10 of short-term and long-term effects of weight loss by Last, most previous studies (including both studies that repeated measurements after the weight loss inter- found no difference between the groups) reported meta - vention, which has not been done before. Thereby, we bolic changes in the lipid profile and the HOMA-IR as a could confirm the studies demonstrating stronger short- measure of insulin sensitivity [8, 11, 12, 15] while a few term effects in MUO in several metabolic parameters, reported results from the hyperinsulinemic-euglycemic although both subgroups did benefit from weight loss. clamp [7, 9] or an oral glucose tolerance test [8]. We The between-group difference partly persisted over time are the first to report data from the hyperinsulinemic- during weight maintenance, which has not been reported euglycemic clamp as well as HOMA when comparing before. Actually, most metabolic improvements dissolved MHO and MUO regarding differences after weight loss. in MHO while they were largely preserved in MUO. It Interestingly, our data from the hyperinsulinemic-eug- cannot be ruled out that the disappearance of the met- lycemic clamp indicated a comparable improvement of abolic benefits in MHO in the long term after weight the ISI in both groups up to 12 months after weight Clamp loss results from ‘healthier’ baseline values compared to loss. Currently, we have no definite explanation for this MUO. In a study on obese subjects, Klöting et  al. found observation. However, the I SI preferentially reflects Clamp that insulin sensitive obese had a lower visceral fat area, myocellular insulin sensitivity. Hence, this might indicate reduced macrophage infiltration into the omental adipose that the differences in the weight loss-induced changes tissue relevantly changing its structure as well as lower of HOMA-IR between MUO and MHO could rather be plasma levels of inflammatory parameters compared to caused by changes in other body compartments than insulin resistant obese subjects [7]. Hence, diverging con- muscle, such as liver or subcutaneous fatty tissue. The stitutions of the adipose tissue might account for differ - use of labelled glucose for the hyperinsulinemic-euglyce- ences in the metabolic improvement between MHO and mic clamp in future studies could target and potentially MUO and might be in the focus of future research. answer this question [22]. For the first time, we report Second, different definitions of metabolic health have data on differences in QoL between MHO and MUO been used in previous studies: in both studies that found after weight loss. no difference in the magnitude of metabolic change MUO demonstrated a stronger and prolonged benefit between MHO and MUO, group allocation followed the in QoL, particularly in physical components of health. By absence or presence of metabolic syndrome, respectively definition, MUO have a higher number of comorbidities [11, 12]. In the other studies, group definition was done than MHO—at least when it comes to metabolic disor- according to one of the following: absence or presence of ders. It is likely that the strong improvement of metabolic metabolic syndrome [7, 15], presence of insulin resistance parameters in MUO contributes not only to ameliora- measured by oral glucose tolerance test [8] or hyperinsu- tion of comorbidities but consequently also to improved linemic-euglycemic clamp [7, 9]. However, it remained physical health. The fact that MHO showed only short- unclear, whether the differential findings in previous term improvements in metabolism and QoL, while MUO studies are caused by the specific MHO/MUO definition demonstrated long-term improvements in both, allows used in each trial [7–9, 11, 12, 15]. We present data with the interpretation that improvement of metabolic param- three different group allocation definitions (by presence eters may be associated with improved physical health. of metabolic syndrome (IDF definition), by HOMA-IR, Our data suggests that mental health is not as closely and by ISI ). On the whole, the results of all used clas- linked to weight and weight loss as physical health. In the Clamp sifications correspond well, confirming the findings of long term, mental health was not improved irrespective Janiszewski et al. who also used and compared two defi - of metabolic status. The predominant effect of weight nition systems [7]. Nevertheless, in our study, differences loss on physical health is in line with results from other between MUO and MHO appear to be partially greater lifestyle intervention trials. In the Look AHEAD trial, in those parameters that were used for group allocation. the lifestyle intervention led to an improved physical but While MUO classified by HOMA-IR or ISI showed not mental function in the SF-36 over the course of the Clamp benefits preferentially in the glucose metabolism, the IDF study [23]. Actually, the Look AHEAD trial included only classification revealed different benefits in MUO within overweight and obese subjects already diagnosed with the glucose and lipid metabolism as well. Overall how- type 2 diabetes. It makes this study population to some ever, the similarities in the results of both classifications extent comparable to our metabolically ill MUO sub- outweigh the differences—especially regarding QoL, the group, although a preexisting diabetes was not manda- groupings generally allow the same conclusion. So, group tory in our MUO subjects. However, data regarding the allocation by the highly sensitive, but vastly impractical effect of an identical weight loss/weight maintenance hyperinsulinemic-euglycemic clamp is not necessary in intervention on estimates of QoL in MUO and MHO clinical practice. has currently not yet been available, even if QoL has Soll et al. Nutrition & Metabolism (2022) 19:25 Page 8 of 10 developed into an important outcome measure of weight were predominantly female, a distribution frequently loss therapies in general [24, 25]. It is of particular rele- found in lifestyle intervention trials. Although we statisti- vance for real-world settings, where changes in physical cally considered sex as an interfering factor in our linear function strongly affect daily life and will therefore result model, we cannot rule out a bias by the sex imbalance. in improved well-being. Moreover, intended improve- Furthermore, behavioral factors were not considered ment of QoL can be easily communicated to patients in this analysis, even though they are known to influ - and will represent a substantial motivation to participate ence the long-term outcome of weight loss interventions in weight loss programs. Therefore, our data gives new [29]. Moreover, our trial suffered from a limited number insight in the weight loss-induced long-term benefits in of dropouts so that after 21  months only 112 from the different obesity subgroups. Interestingly, these effects original 143 participants could be analyzed. Unfortu- could apparently not be explained by differences in body nately, only general reasons for dropout from this study weight changes, as both groups demonstrated a compara- were enquired. However, dropouts from a lifestyle inter- ble weight course throughout the study. vention trial including a placebo group are not unex- Taken together, our data implies that MUO will ben- pected. A comparison of weight loss induced changes efit stronger in short- and long-term from weight loss of metabolic and anthropometric parameters revealed interventions regarding several health-related outcomes. comparable improvement in dropouts and subjects who uTh s, identification of obese patients with metabolic completed the study (data not shown). We present data impairments might be a crucial strategy to increase the regarding obesity, metabolism, and clinical relevant out- general efficacy of weight loss interventions in the con - comes like QoL. Nevertheless, effects on cardiovascular text of health benefits. Although this was not directly morbidity and mortality were out of scope of this study addressed by our study, this assumption is supported as it was primarily conceived to analyze anthropometric by data from a previous meta-analysis investigating the and metabolic outcomes after a lifestyle intervention. The effect of weight loss on all-cause mortality: Harrington strengths of our study include the large sample size with et  al. have shown that unhealthy obese lower their all- a long observation period, especially compared to previ- cause mortality by intentional weight loss while healthy ous studies. On top of that, we present data from com- obese appear to have an unaltered all-cause mortality prehensive phenotyping at several time-points including after weight loss [26]. It is also in line with results of a the assessment of insulin sensitivity by the hyperinsuline- large French cohort study that recently demonstrated mic-euglycemic clamp as well as highly relevant patient- a substantially higher risk for cardiovascular events in reported outcomes of QoL. Additionally, we present data MUO compared to MHO [27]. However, most of the employing and comparing three different definitions for beneficial metabolic long-term effects in MUO shown metabolic health. in our study marginally failed to be significant in inter- group comparisons to MUO. This might be caused by the limited sample size and the high variability of individual Conclusions long-term changes. Overall, our data demonstrate that MUO have a greater Interestingly, adjustment for improvement of differ - benefit in health-related QoL and several metabolic ent estimates of obesity during weight loss indicated parameters than MHO not only in the short-term but that additional mechanisms apart from changes in body also in the long-term course after weight loss. However, it weight appear to play a relevant role for improvement of is necessary to remember that MHO still have increased insulin sensitivity during weight loss in MUO. This might rates of incident diabetes, heart failure, and mortal- include inflammatory, metabolic, or hormonal factors ity compared to metabolically healthy non-obese [30]. such as leptin [28], adiponectin [28], or atrial natriuretic There might still be benefits from weight loss interven - peptides [16]. Future research investigating the underly- tions regarding effects on the musculoskeletal system, ing mechanism of the stronger improvement of insulin among others, which were not addressed in this study. A resistance in MUO is warranted. recent review on this subject also points out that meta- Our study comes with obvious limitations. First, the bolically healthy obesity might represent a transitional primary aim of the presented MAINTAIN trial was the state for some patients that might turn into metabolically comparison of different treatment strategies to achieve unhealthy obesity at one point [31]. u Th s, lifestyle modi - body weight maintenance. Thus, we defined the groups fication should still be recommended for obese patients for this study retrospectively. However, MUO and MHO in general. In settings of limited resources the MUO underwent the same procedures and we implemented might however become the preferential target group for potential relevant confounders including the treatment lifestyle interventions. group in our statistical analyses. Participants in our trial S oll et al. Nutrition & Metabolism (2022) 19:25 Page 9 of 10 Abbreviations References BMI: Body mass index; HbA1c: Hemoglobin A1c; HDL: High‑ density lipopro‑ 1. World Health Organization. Obesity and overweight. 2021. Available tein; HOMA‑IR: Homeostatic model assessment for insulin resistance; IDF: from: https:// www. who. int/ news‑ room/ fact‑ sheets/ detail/ obesi ty‑ and‑ Internationale Diabetes Federation; ISI : Insulin sensitivity index (hyperin‑overw eight Clamp sulinemic‑ euglycemic clamp); LDL: Low‑ density lipoprotein; MHO: Metaboli‑ 2. Menke A, Casagrande S, Geiss L, Cowie CC. Prevalence of and trends cally healthy obese; MUO: Metabolically unhealthy obese; OGTT : Oral glucose in diabetes among adults in the United States, 1988–2012. JAMA. tolerance test; QoL: Quality of life; SF‑36: Short Form (36) Health Survey; WHO: 2015;314(10):1021. World Health Organization. 3. Yumuk V, Tsigos C, Fried M, Schindler K, Busetto L, Micic D, et al. European guidelines for obesity management in adults. Obes Facts. 2015;8(6):402–24. Supplementary Information 4. Tuomilehto J, Lindström J, Eriksson JG, Valle TT, Hämäläinen H, Ilanne‑ The online version contains supplementary material available at https:// doi. Parikka P, et al. Prevention of type 2 diabetes mellitus by changes in org/ 10. 1186/ s12986‑ 022‑ 00660‑w. lifestyle among subjects with impaired glucose tolerance. N Engl J Med. 2001;344(18):1343–50. 5. Stefan N, Häring H‑U, Schulze MB. Metabolically healthy obesity: the Additional file 1. Electronic supplementary material. low‑hanging fruit in obesity treatment? Lancet Diabetes Endocrinol. 2018;6(3):249–58. 6. Karelis AD, Faraj M, Bastard J‑P, St ‑Pierre DH, Brochu M, Prud’homme D, Acknowledgements et al. The Metabolically healthy but obese individual presents a favorable We thank K. Simon, B. Horchler, N. Huckauf, and C. Kalischke for excellent tech‑ inflammation profile. J Clin Endocrinol Metab. 2005;90(7):4145–50. nical assistance as well as A. Reisshauer for the support regarding to physical 7. Janiszewski PM, Ross R. Eec ff ts of weight loss among metabolically activity intervention. We thank Nestlé HealthCare Nutrition GmbH, Frankfurt healthy obese men and women. Diabetes Care. 2010;33(9):1957–9. am Main, Germany for the opportunity to purchase the Optifast 2 diet at a 8. Kantartzis K, Machann J, Schick F, Rittig K, Machicao F, Fritsche A, et al. reduced price. Eec ff ts of a lifestyle intervention in metabolically benign and malign obesity. Diabetologia. 2011;54(4):864–8. Authors’ contributions 9. Karelis AD, Messier V, Brochu M, Rabasa‑Lhoret R. Metabolically healthy K.M., D.S., and J.S. researched data and wrote the manuscript. D.S., K.M., and but obese women: effect of an energy‑restricted diet. Diabetologia. J.G. were responsible for data analysis. All authors contributed to interpreta‑ 2008;51(9):1752–4. tion of the results. All authors critically read and edited several drafts before 10. Mai K, Brachs M, Leupelt V, Jumpertz‑ von Schwartzenberg R, Maurer submission. All authors read and approved the final manuscript. L, Grüters‑Kieslich A, et al. Eec ff ts of a combined dietary, exercise and behavioral intervention and sympathetic system on body weight mainte‑ Funding nance after intended weight loss: results of a randomized controlled trial. Open Access funding enabled and organized by Projekt DEAL. This research Metabolism. 2018;83:60–7. was supported by the Deutsche Forschungsgemeinschaft (DFG KFO 218/1), 11. Ruiz JR, Ortega FB, Labayen I. A weight loss diet intervention has a similar the German Diabetes Society (DDG) and the German Ministry for Education beneficial effect on both metabolically abnormal obese and meta‑ and Research (BMBF) by support of the German Centre for Cardiovascular bolically healthy but obese premenopausal women. Ann Nutr Metab. Research (DZHK, BER5.1). 2013;62(3):223–30. 12. Dalzill C, Nigam A, Juneau M, Guilbeault V, Latour E, Mauriège P, et al. Data availability Intensive lifestyle intervention improves cardiometabolic and exercise The datasets generated during and/or analyzed during the current study are parameters in metabolically healthy obese and metabolically unhealthy available from the corresponding author on reasonable request. obese individuals. Can J Cardiol. 2014;30(4):434–40. 13. Ninomiya JK, L’italien G, Criqui MH, Whyte JL, Gamst A, Chen RS,. Associa‑ tion of the metabolic syndrome with history of myocardial infarction Declarations and stroke in the third national health and nutrition examination survey. Circulation. 2004;109(1):42–6. Ethics approval and consent to participate 14. Klöting N, Fasshauer M, Dietrich A, Kovacs P, Schön MR, Kern M, et al. Insu‑ The study protocols were approved by the Institutional Review Board of the lin‑sensitive obesity. Am J Physiol Endocrinol Metab. 2010;299(3):E506–15. Charité Medical School (EA1/140/12) and all subjects gave written informed 15. Shin M‑ J, Hyun YJ, Kim OY, Kim JY, Jang Y, Lee JH. Weight loss effect on consent. All analyses were conducted in accordance with the Declaration of inflammation and LDL oxidation in metabolically healthy but obese Helsinki. The trial was registered at ClinicalTrials.gov (NCT00850629). (MHO) individuals: low inflammation and LDL oxidation in MHO women. Int J Obes. 2006;30(10):1529–34. Consent for publication 16. Brachs M, Wiegand S, Leupelt V, Ernert A, Kintscher U, von Schwarzenberg Not applicable. RJ, et al. ANP system activity predicts variability of fat mass reduction and insulin sensitivity during weight loss. Metabolism. 2016;65(6):935–43. Competing interests 17. Mai K, Li L, Wiegand S, Brachs M, Leupelt V, Ernert A, et al. An Integrated We declare that there is no conflict of interest that could be perceived as understanding of the molecular mechanisms of how adipose tissue prejudicing the impartiality of the research reported. metabolism affects long‑term body weight maintenance. Diabetes. 2019;68(1):57–65. Author details 1 18. International Diabetes Federation. The IDF consensus worldwide defini‑ Department of Endocrinology and Metabolism, Charité – Universitätsmedizin tion of the metabolic syndrome. 2021. Available from: https:// www. idf. Berlin, corporate member of Freie Universität Berlin, Humboldt‑Universität Zu org/e‑ libra ry/ conse nsus‑ state ments/ 60‑ idfco nsens us w‑orld wide‑ defin Berlin, and Berlin Institute of Health, Chariteplatz 1, 10117 Berlin, Germany. 2ition of‑ the‑ metab olic‑ syndr ome. html Charité Center for Cardiovascular Research, Charité – Universitätsmedizin 19. Ellert U, Kurth B. Methodische Betrachtungen zu den Summenscores Berlin, corporate member of Freie Universität Berlin, Humboldt‑Universität 3 des SF‑36 anhand der erwachsenen bundesdeutschen Bevölkerung. Zu Berlin, and Berlin Institute of Health, 10117 Berlin, Germany. DZHK (Ger‑ Bundesgesundheitsblatt ‑ Gesundheitsforschung ‑ Gesundheitsschutz. man Centre for Cardiovascular Research), partner site Berlin, Berlin, Germany. 4 2004. NutriAct‑ Competence Cluster Nutrition Research Berlin‑Potsdam, Nuthetal, 20. Gutch M, Kumar S, Razi SM, Gupta KK, Gupta A. Assessment of insulin Germany. sensitivity/resistance. Indian J Endocrinol Metab. 2015;19(1):160–4. 21. Li L, Spranger L, Soll D, Beer F, Brachs M, Spranger J, et al. Metabolic Received: 15 November 2021 Accepted: 16 March 2022 impact of weight loss induced reduction of adipose ACE‑2 ‑ Potential implication in COVID‑19 infections? Metabolism. 2020;113:154401. Soll et al. Nutrition & Metabolism (2022) 19:25 Page 10 of 10 22. Haigh JW, Johnston DG, McCulloch AJ, Laker MF, Welby J, Evans S. Assess‑ ment of glucose turnover in normal man with the use of a non‑radioac‑ tive isotopically labelled preparation, [6,6–2H]glucose, as tracer. Clin Sci (Lond). 1982;63(5):437–40. 23. Rubin RR, Wadden TA, Bahnson JL, Blackburn GL, Brancati FL, Bray GA, et al. Impact of intensive lifestyle intervention on depression and health‑ related quality of life in type 2 diabetes: the look AHEAD trial. Diabetes Care. 2014;37(6):1544–53. 24. Omotosho P, Mor A, Shantavasinkul PC, Corsino L, Torquati A. Gastric bypass significantly improves quality of life in morbidly obese patients with type 2 diabetes. Surg Endosc. 2016;30(7):2857–64. 25. Poelemeijer YQM, van der Knaap ETW, Marang‑ van de Mheen PJ, Demirkiran A, Wiezer MJ, Hazebroek EJ, et al. Measuring quality of life in bariatric surgery: a multicentre study. Surg Endosc. 2020;34(12):5522–32. 26. Harrington M, Gibson S, Cottrell RC. A review and meta‑analysis of the effect of weight loss on all‑ cause mortality risk. Nutr Res Rev. 2009;22(1):93–108. 27. Fauchier G, Bisson A, Bodin A, Herbert J, Semaan C, Angoulvant D, et al. Metabolically healthy obesity and cardiovascular events: a nationwide cohort study. Diabetes Obesity Metab. 2021. https:// doi. org/ 10. 1111/ dom. 14492 28. Fasshauer M, Blüher M. Adipokines in health and disease. Trends Pharma‑ col Sci. 2015;36(7):461–70. 29. Del Corral P, Bryan DR, Garvey WT, Gower BA, Hunter GR. Dietary adher‑ ence during weight loss predicts weight regain. Obesity (Silver Spring). 2011;19(6):1177–81. 30. Zhou Z, Macpherson J, Gray SR, Gill JMR, Welsh P, Celis‑Morales C, et al. Are people with metabolically healthy obesity really healthy? A prospec‑ tive cohort study of 381,363 UK Biobank participants. Diabetologia. 2021;64(9):1963–72. 31. Blüher M. Metabolically healthy obesity. Endocr Rev. 2020;41(3):405–20. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions

Journal

Nutrition & MetabolismSpringer Journals

Published: Mar 28, 2022

Keywords: Healthy obesity; Weight loss; Insulin sensitivity; Quality of life

There are no references for this article.