Intentional weight loss improved performance in obese ischaemic heart patients: A two centre intervention trial
Obesity is an important risk factor for the development of cardiovascular disease, including the development of heart failure (HF) . The risk of developing HF increases both with the severity of obesity (in the Framingham Heart Study the risk of HF increased by 7% in women and 5% in men for each 1 kg/m2 increment in BMI ) and its duration . Despite this strong association between obesity and HF, several studies have found that lower body weight predicts increased mortality in people with diagnosed HF . A meta-analysis of the relationship between body mass index (BMI) and mortality in patients with HF found that overweight and obesity were associated with lower all-cause and cardiovascular mortality . Classifying by BMI class revealed a U-shaped relationship, with lowest risk of mortality at BMI 25-35 kg/m2 . In a systematic review and meta-analysis including 2.88 million individuals Flegal et al. found a J-shaped relation between BMI and all-cause mortality . The lowest all-cause mortality was at BMI 25-30 kg/m2. Relative to normal weight, obesity was associated with significantly higher all-cause mortality . Kapoor and Heidenreich found the lowest risk of mortality in 542 obese HF patients over 60 years of age was at BMI 36-40 kg/m2 . In contrast a study in nearly 8000 subjects with chronic mild to moderate HF and diabetes mellitus (DM) found that obesity conferred no survival benefit . The studies linking low weight to mortality have all been observational and their design, in particular their failure or inability to differentiate between the effect of intentional and unintentional low weight, undermines their validity [10,11]. We hypothesized that intentional weight loss would improve the physical performance of obese patients with HF or ischaemic heart disease (IHD). This was investigated at two centres, one in Denmark (DK) and one in Britain (UK). Both studies were performed as pilot studies to assess the feasibility, safety and size of effects of therapeutic weight loss induced by low energy diet (LED) in obese patients with moderate-to-severe HF or IHD. The primary outcome at both centres was change in physical performance, measured in the form of six-minute walk test (6-MWT) at the Danish centre and in the form of cardiac performance measured by maximum oxygen uptake (VO2max) at the UK centre. Changes in 6-MWT predict the changes in VO2max and survival in patients with severe HF [12,13]. Secondary outcomes were feasibility and safety of LED induced weight loss, and changes in metabolic parameters.
The DK centre performed a non-blinded, randomized, controlled trial conducted in two rounds. Subjects participated in intensive 12-week supervision in accordance with randomization. Interim visits to monitor and assess safety were conducted weekly from weeks 0 through 3, and biweekly in weeks 4 through 12. All subjects gave written consent after having received verbal and written information about the study. The trial was carried out at Copenhagen University Hospital, Gentofte, and was approved by the Scientific Ethical Committee of Capital Region Denmark (H-4-2010-087).Design – UK centre
The UK Centre performed a prospective intervention trial, designed as a pilot study to assess feasibility and safety. Interim visits to monitor and assess safety were conducted weekly between weeks 0 through 6, and biweekly in weeks 7 through 16. All subjects gave written consent after having received verbal and written information about the study. The study was carried out at the Clinical Research Facility, Addenbrooke’s (Hospital) Centre of Clinical Investigation, UK, and approved by South Cambridgeshire Ethics Committees (REC 06/Q0108/37).Subjects – DK centre
Subjects were identified from patient records at the Department of Cardiology at Copenhagen University Hospital, Gentofte, during autumn 2010. A total of 37 subjects, all diagnosed moderate-to-severe HF, New York Heart Association (NYHA) II-III, were eligible for inclusion and invited to take part, 11 declined, 26 were screened, four failed the screening (could not complete the 6-MWT). Twenty-two subjects were included in the study and randomized – opaque concealed envelopes with patients’ names inside were drawn by an impartial person to either intervention (N=11) or control (N=11). One subject in the intervention group fell ill and with-drew prior to baseline measurement. Four subjects in the control group withdrew consent prior to baseline because they were dissatisfied with randomisation; the remaining seven completed the program. In order to encourage control subjects to complete the trial they were promised an opportunity to take part in the intervention arm of the study, if they met the inclusion criteria, subsequent to completion of the control program. Four subjects accepted this and crossed over to the intervention program after completing their period as controls. Hence a total of 14 subjects (10 from the original group and a subsequent group of 4 cross-over from the control group) completed the intervention.
Subjects were identified from patient records at Addenbrooke’s Hospital, Cambridge, UK. Recruitment was extended to general practitioners, and cardiology departments at Luton and Dunstable, Bedford, and West Suffolk Hospitals, UK. A total of 34 individuals were screened, 14 met the inclusion criteria and were included in the study. Three dropped out, one due to cancer, one was intolerant to the diet, and one was allocated to control (group subsequently terminated due to insufficient number of subjects). A total of 11 subjects completed the intervention, four had moderate-to-severe HF, NYHA II or III; the remaining subjects had major cardiovascular risk factors (Table S1 in supplemental material).
All subjects were advised to adhere to dietary advice according to the European Society of Cardiology . The diet during week 0 through 8 (Table 2) provided the intervention subjects with BMI below 40 kg/m2 with 800 kcal/day, and subjects with BMI above 40 kg/m2 with 1000 kcal/day. The diet was a mainly liquid LED of 6 sachets Nupo®, and one Nupo® snack bar or 200 g of specified vegetables, daily (Nupo®, Denmark). Six sachets (equivalent to ≈750 kcal) of the LED products supplied subjects with the recommended daily amount of vitamins and minerals. At the baseline visit the intervention group received a package containing all the various flavours of Nupo® shakes and soups. At the following interim visits through week six the subjects were supplied with the product flavours they preferred, e.g. one subject chose only soup and another chose only two different flavours of shakes. From week eight through 12 the intervention group followed plan of energy restricted reintroduction to regular foods supplying 1200 kcal/day. The diet included two daily Nupo® LED products (one sachet and one meal replacement bar) in combination with regular foods based on recipes with high amounts of protein and low amounts of carbohydrate. Control group subjects attended the same number of interim visits as the intervention group throughout the trial and were instructed to adhere to a conventional diet according to the Nordic Nutrition Recommendations .
Anthropometric data were collected at all visits prior to giving dietary advice. Height was measured at baseline to nearest 0.5 cm using a wall-mounted stadiometer (Seca, Hamburg, Germany). Subjects, wearing only undergarments and without shoes, were weighed to nearest 0.1 kg using a calibrated professional scale (Tanita HD-351, Tanita Corporation of America, Illinois, USA). Body composition was estimated by bioelectric impedance (Tanita BC-418, Tanita Corporation of America, Illinois, USA). Physical performance was assessed at baseline and the end of the trial by measuring 6-MWT – subjects walked six minutes at a comfortable pace on a 30 meter long track in a basement corridor. The 6-MWT serves as an indicator for exercise tolerance and is validated to reflect change in cardiac capacity and exertion in concordance with change in symptoms in HF patients [13,16,17]. Information on background cardiovascular risk factors including left ventricular ejection fraction and pharmacotherapy was retrieved from medical records.
Anthropometric measures and functional status – UK centre
Height was measured to the nearest 0.5 cm using a wall-mounted stadiometer (Seca, Hamburg, Germany) at baseline. Anthropometric measures were obtained at baseline, week six and week sixteen. Body composition, FM and FFM were measured
using a four-component model described elsewhere . VO¬- 2max served as an indicator of cardiac performance and was assessed by cardiopulmonary exercise testing.
Blood analysis – DK centre
Venous blood samples were drawn at the start of each visit by the subject to the department, always in the morning after an overnight fast. Blood samples for routine plasma analysis were collected in tubes with lithium heparin for quantification of potassium, lipids and triglycerides (TG), and in sodium fluoride tubes for glucose. Lithium heparin plasma was aliquoted and stored at –80°C until batch analysis for C-reactive protein (CRP), insulin, leptin, adiponectin (total) and TNF-α. Plasma glucose was measured by a colorimetric assay, potassium by a potentiometric assay and CRP by a high sensitivity immunoassay (Ortho-Clinical Diagnostics, Johnson & Johnson Medical, Birkeroed, Denmark) for the Vitros 5.1 FS analyzer. Between batch coefficients of variations (CVs) were 0.7% for glucose at 4.6 and 6.5 mmol/L, 0.7% for potassium at 3.6 and 4.9 mmol/L, and 1.3% for CRP at 3.18 mg/L. Plasma total cholesterol, high-density lipoprotein (HDL) cholesterol, and TG, were measured by enzyme assays (Ortho-Clinical Diagnostics) for the Vitros 5.1 FS analyzer. Between batch CVs were 1.4% at 5.4 mmol/L, 2.0% at 1.66 mmol/L, and 1.5% at 1.44 mmol/L. Very-low-density lipoprotein (VLDL) and low-density lipoprotein (LDL) were calculated from Friedewald’s formula. Insulin and N-terminal prohormone of brain natriuretic peptide (NT-proBNP) were measured by chemiluminescent immunometric assays (Siemens Healthcare, Ballerup, Denmark) with use of ADVIA Centaur XP and IMMULITE 2000, respectively. Leptin, adiponectin and TNF-α were measured in duplicate by ELISA immunoassays (R&D Systems Europe, Abingdon, UK) with use of a Bio-Tek EL808 microplate reader. Between batch CVs were 2.5% at 89 pmol/L for insulin, 3.8% at 19263 pg/mL leptin, 4.1% at 6589 ng/mL adiponectin and 7.3% at 1.34 pg/ mL for TNF-α.
Venous blood was drawn in the fasting state at weeks 0, 6 and 16, and plasma was immediately obtained and stored at -20 C. Routine biochemical analyses were performed after every visit to quantify plasma glucose, TG, total cholesterol, and HDL. All other biochemical analyses were performed in one batch. Both VLDL and LDL were calculated from Friedewald’s formula. CRP was measured by using high-sensitivity, two-site enzyme linked immunoassay (ELISA) as described elsewhere . Adiponectin
and leptin were assayed by two-site microtitre platebased DELFIA (Perkin Elmer Life Sciences, Wallac Oy, Turku, Finland). Antibodies and standards reagents were supplied by R&D Systems (R&D Systems Europe, Abingdon, UK). Between batch CVs for adiponectin were 5.4% at 3.6 μg/mL, 5.2% at 9.2 μg/mL, and 5.8% at 15.5 μg/mL. Between batch CVs for leptin were 7.1% at 2.7 ng/mL, 3.9% at 14.9 ng/mL, and 5.7% at 54.9 ng/mL. Insulin was assayed in singleton on a 1235 Auto DELFIA (Perkin Elmer Life Sciences, Wallac Oy, Turku, Finland) automatic immunoassay system using a two-step time resolved fluorometric assay (Kit No. B080-101). Between batch CVs were 3.1% at 29 pmol/L, 2.1% at 79.4 pmol/L, 1.9% at 277 pmol/L, and 2.0% at 705 pmol/L. Plasma BNP concentration was assessed by Triage® BNP test kid (Biosite® Incorporated, California). TNF-α was assayed by quantitative ELISA technique with the reagent supplied by R&D Systems (R&D Systems Europe) and between batch CVs were 7% at 5.97 pg/mL, 8.3% at 37.4 pg/mL, 6.4% at 282 pg/mL, and 6.3% at 3869 pg/mL
The intervention group in the DK centre (N=14) and the UK centre are presented in Tables S7 and S8 were pooled with the completers from UK Centre (N=11). Data from a total of 25 patients were included in the analyses, and data from all seven control subjects were included unless otherwise stated. As two different, but comparable, methods were used to assess physical performance capacity in the two centres the effect of treatment is calculated by the change in percentage. Plasma lipids and cardiovascular markers from the two centres are analysed differently and changes are therefore calculated on the true values, as well as the percentage of change from baseline to end of study.For baseline data normality was assessed using the Shapiro- Wilk test  (Table S5 and S6 in supplemental material). Baseline data are presented as medians and inter quartile ranges (IQR), and differences at baseline were evaluated using two-sample t-tests. Effects of treatment were assessed by means of linear mixed model analyses. The models included adjustment for interaction between treatment and visit, intervention subjects with earlier control participation (subject- specific random effects), and differences between the DK/UK centres. The models included baseline, age, height, and visit. Results of the linear mixed model analyses are presented as mean ± standard error (SE). Results are interpreted using a significance level of 0.05. All statistical analyses were performed using Microsoft Excel and SPSS 21.0 for Windows. Results from the DK centre (only including the initial intervention subjects (N=11)) and the UK centre are presented in Tables S7 and S8 (in supplemental material).
Baseline characteristics are presented in Table 3. The only differences found between groups were in %FM and TNF-α.There was no difference between intervention and control groups at baseline in 6-MWT (465 (327-524) meter (N=14) vs. 416 (374-514) meter (N=7) (P=0.908)) (DK centre only).
All results from anthropometric and plasma variables are presented in Table 4. Body weight in the LED subjects was reduced by 13.9 kg ± 6.5 versus 1.2 ± 1.8 in the control group (P=0.000). The weight lost by LED consisted mostly of FM (68%), while almost none of the metabolized tissue in the control group was FM (<1%) (P=0.001).
Of the plasma lipids, triglyceride (TG) and very-low-density lipoprotein (VLDL) were reduced by LED compared to control group (P=0.000 and P=0.001, respectively). At end of study mean TG was within the reference value for patients with heart disease (2.0 mmol/L).
Of the measured inflammatory markers only CRP changed due to treatment (-16.50% ± 57.83 versus 33.07% ± 136.81 (P=0.010)). There was no effect of treatment on plasma insulin or glucose, but insulin requirement fell by at least 60% from baseline to end of study in patients treated with insulin. There was no difference between groups in N-terminal prohormone of brain natriuretic peptide (NT-proBNP) baseline concentrations (DK centre only) or in effect of treatment in either NT-proBNP or BNP (data not shown).
There were no changes in renal parameters, or cardiac ejection fraction (only measured in UK center).
The weight loss induced by LED in obese patients with moderate-to-severe HF and IHD in the present study is similar to other reports [21,22]. The degree of weight loss of ≈14 kg, of which 9.4 kg was FM associated with improved in physical and cardiac performance are likely to predict improved survival .
Earlier studies investigating the effects of therapeutic weight loss in obese HF patients all found positive effects of intentional weight loss on physical performance and cardiovascular risk markers [3,17]. Alpert et al. monitored the effect of gastric surgery in morbidly obese younger patients (≈38 years) with or without HF . They observed the same improvements in cardiovascular risk markers in both groups, and no adverse effects of the rapid and large weight loss in either group. A single case study found HF and left ventricular ejection fraction to be improved after extreme weight loss achieved with LED diet over three years . A small scale controlled intervention study in HF patients found 10 kg weight loss after a 12 week high protein diet, and improvements in cardiovascular risk factors and physical performance, measured as 6-MWT and VO¬2max .
In the present study change in BMI was predictive of the improvements in 6-MWT, as found previously . Two other studies have investigated the effect of weight loss on physical performance [25,26]. One was a retrospective study that monitored the effect on obese HF out-patients of a rehabilitation program including weight loss . They found that the 45 patients who complied with the program lost 5 kg in weight and improved exercise capacity. Those who did not lose weight (N=81) showed no change in exercise capacity. The other study found that an intervention of at least four weekly low-level exercise sessions (60% aerobic) in moderate-to-severe overweight HF patients only improved cardiopulmonary fitness in those who lost 5% weight or more (26). In another study of exercise, overweight HF patients increased their VO2max regardless of weight loss. Overall, the suggests that the positive effects of exercise on physical performance are increased by weight loss .
In the present study we found that treatment of patients with HF and IHD with LED reduced the level of CRP. Higher levels of CRP are associated with more severe heart failure, and are independently associated with mortality and morbidity. Weight loss has previously been shown to reduce CRP, but to a lesser extent than found in the present study , though the reduction in CRP is similar to that seen in subjects compliant with a low glycaemic diet after LED .
During the present study the Danish control group lost 1.2 kg ± 1.8 (P=0.002), and at the same time their physical performance deteriorated by 22.1% ± 25.6. Evangelista et al. found a similar reduction in 6-MWT in a control group that lost 1.5 kg over 12 weeks . It is likely that this deterioration in 6-MWT reflects the natural history of HF, supported by the fact that the weight loss in the control group was unintentional and primarily consisted of FFM.
There are no reports of adverse events associated with intentional weight loss in overweight HF, not even in association with weight loss of 33-50% of initial body weight [3,17,23,25].
On the contrary, intentional weight loss and improvement of physical performance has been shown to reduce rate of re-admittance to hospital, improve quality of life, and reduce risk of all-cause mortality [26,29]. Experience from bariatric surgery, cardiovascular risks and impaired cardiac performance are common in patients, has shown that weight loss reduces cardiovascular events . Despite these findings some consider that weight loss in obese HF patients “may even be potentially harmful” , and that “severe calorie restriction in patients with severe HF has the potential to worsen cardiac muscle function” . The study referred to is by Alden et al., who found significant cardiac atrophy after intentional weight loss . However, this was an animal study conducted in 11 dogs with results analysed after three weeks of acute protein-calorie restriction resulting in a 20-25% weight loss. These findings cannot be compared with the effect of an intentional weight loss achieved with a nutritionally complete (except for energy) diet of 750-1200 kcal per day provided by a well-formulated LED or energy restricted reintroduction.
In the present study treatment significantly reduced TG and VLDL. At the end of the study only two subjects had TG levels above the reference value for patients with cardiovascular disease, the median value being within the reference. The small changes seen probably reflect that 14 of the subjects in the intervention group were treated with statins.
The weaknesses of this study include the small sample size that itself was composed by combining two separate studies and this could have caused type I error. Recruitment at both DK and UK centres proved difficult: only few patients at the sites met inclusion criterion and of those many declined in part due to difficulty of transport to and from the centres; since both studies had similar objectives and interventions we felt it appropriate to merge the data. The UK centre was uncontrolled (it was designed as a pilot for a fully powered trial). The use of two different assessment methods of body composition might have weakened our results, as the correlation between the four-component model and bioelectric impedance is moderate , though both methods have been found to be capable of detecting change in body composition during weight loss . The two centres analysed blood samples separately, and hence there may be inconsistency in the crude data. We have attempted to correct for bias, at least to some extent, by calculating the crude changes and the percentages, and including both in the statistical. Despite these limitations, robust and clinically important outcomes were observed.
The changes found in physical performance and cardiovascular risk markers indicate a beneficial effect of using LED in the treatment of moderate-to-severe HF and IHD patients. While this study was of short duration, it does demonstrate the safety of prescribing a high protein LED that is nutritionally complete apart from reduced energy for these patients, and the considerable benefits obtained by a substantial weight loss. Longterm outcome studies are needed to investigate if these shortterm benefits translate into reduced cardiovascular events and mortality.
Statement of Authorship
The trial performed at the DK centre was designed by NRWG, SMHL, CTP, SS and AA, performed by NRWG and SMHL, and MRA was in charge of all biochemical analyses. The trial performed at the UK centre was designed by KSM and NF, performed by KSM, and PH performed echocardiogram. KSM, NF and KD interpreted the results in the UK centre. NRWG performed all the statistical work and prepared the data for the present manuscript. All authors participated in the interpretation of the results and critically reviewed and approved the manuscript.
All LED products used in DK centre were kindly supplied by Nupo®, Denmark, who had no influence on either study design or data handling. The UK centre was sponsored by Research and Development Department, Cambridge University Hospital NHS Foundation Trust (R&D A090622). Christian Ritz assisted in the statistical analyses and Tina Cuthbertson proofread the manuscript (both from Department of Nutrition, Exercise and Sports, University of Copenhagen). Thank you to all the study participants in Denmark and UK.
Supplementary information is available at the journal’s website.
6-MWT: six minute walk test; BMI: body mass index; BNP: brain natriuretic peptide; CRP: C-reactive protein; HC: hip circumference; HDL: high density lipoprotein; IHD: ischaemic heart disease; FFM: fat free mass; FM: fat mass; LDL: low density
lipoprotein; IQR: inter quartile range; LED: low energy diet; NT-proBNP: N-terminal prohormone of brain natriuretic peptide; TG: triglycerides; VLDV: very low density lipoprotein, VO2max: maximum oxygen uptake; WC: waist circumference.
1.Kumanyika SK, Obarzanek E, Stettler N, Bell R, Field AE, et al. Population-Based Prevention of Obesity The Need for Comprehensive Promotion of Healthful Eating, Physical Activity, and Energy Balance: A Scientific Statement From American Heart Association Council on Epidemiology and Prevention, Interdisciplinary Committee for Prevention (Formerly the Expert Pan-el on Population and Prevention Science). Circulation. 2008, 118(4):428-464.
3.Alpert MA, Terry BE, Mulekar M, Cohen MV, Massey CV, et al. Cardiac morphology and left ventricular function in normotensive morbidly obese patients with and without congestive heart failure, and effect of weight loss. Am J Cardiol. 1997, 80(6):736-740.
6.Davos CH, Doehner W, Rauchhaus M, Cicoira M, Francis DP, et al. Body mass and survival in patients with chronic heart failure without cachexia: the importance of obesity. Journal of cardiac failure. 2003, 9(1): 29-35.
7.Flegal KM, Kit BK, Orpana H, Graubard BI. Association of All-Cause Mortality With Overweight and Obesity Using Standard Body Mass Index CategoriesA Systematic Review and Meta-analysisAll-Cause Mortality Using BMI Categories. JAMA. 2013, 309(1): 71-82.
9.Adamopoulos C, Meyer P, Desai RV, Karatzidou K, Ovalle F, et al. Absence of obesity paradox in patients with chronic heart failure and diabetes mellitus: a propensity-matched study. Eur J Heart Fail. 2011, 13(2):200-206.
14.Swedberg K, Cleland J, Dargie H, Drexler H, Follath F, et al. Guidelines for the diagnosis and treatment of chronic heart failure: executive summary (update 2005) The Task Force for the Diagnosis and Treatment of Chronic Heart Failure of the European Society of Cardiology. Eur Heart J. 2005, 26(11):1115- 1140.
16.Olsson LG, Swedberg K, Clark AL, Witte KK, Cleland JGF. Six minute corridor walk test as an outcome measure for the assessment of treatment in randomized, blinded intervention trials of chronic heart failure: a systematic review. Eur Heart J. 2005, 26(8):778-793.
17.Evangelista LS, Heber D, Li Z, Bowerman S, Hamilton MA, et al. Reduced body weight and adiposity with a high-protein diet improves functional status, lipid profiles, glycemic control, and quality of life in patients with heart failure: a feasibility study. The Journal of cardiovascular nursing. 2009, 24(3):207.
18.Packianathan I, Fuller N, Peterson D, Wright A, Coward W et al. Use of a reference four-component model to define the effects of insulin treatment on body composition in type 2 diabetes: the ‘Darwin study’. Diabetologia. 2005, 48(2): 222-229.
19.Packard CJ, O’Reilly DS, Caslake MJ, McMahon AD, Ford I, et al. Lipoprotein-associated phospholipase A2 as an independent predictor of coronary heart disease. N Engl J Med. 2000, 343(16):1148-1155.
24.Horwich TB, Leifer ES, Brawner CA, Fitz-Gerald MB, Fonarow GC, et al. The relationship between body mass index and cardiopulmonary exercise testing in chronic systolic heart failure. Am Heart J. 2009 Oct;158(4):S31-S6.
25.Lavie CJ, Milani RV. Effects of cardiac rehabilitation, exercise training, and weight reduction on exercise capacity, coronary risk factors, behavioral characteristics, and quality of life in obese coronary patients. Am J Cardiol. 1997, 79(4):397-401.
26.Evangelista LS, Doering LV, Lennie T, Moser DK, Hamilton MA, et al. Usefulness of a home-based exercise program for overweight and obese patients with advanced heart failure. Am J Cardiol. 2006, 97(6):886-890.
28.Gögebakan Ö, Kohl A, Osterhoff MA, van Baak MA, Jebb SA, et al. Effects of Weight Loss and Long-Term Weight Maintenance With Diets Varying in Protein and Glycemic Index on Cardiovascular Risk FactorsClinical Perspective The Diet, Obesity, and Genes (DiOGenes) Study: A Randomized, Controlled Trial. Circulation. 2011, 124(25):2829-2838.
29.Horwich TB, Broderick S, Chen L, McCullough PA, Strzelczyk T, et al. Relation among body mass index, exercise training, and outcomes in chronic systolic heart failure. The American journal of cardiology. 2011, 108(12):1754-1759.
33.Packianathan I, Fuller N, Peterson D, Wright A, Coward W, et al. Use of reference four-component model to evaluate the ability of alternative methods and prediction techniques to estimate body composition in Type 2 diabetes and its changes following insulin treatment. International Journal of Body Composition Research. 2004, 2(4):141-148.