Diastolic function and cardiovascular risk among patients with severe obesity referred to a lifestyle-program – a pilot study

Abstract

Objectives. Severe obesity is associated with a high risk of comorbidities and alterations of cardiac structure and function. The primary aim of the study was to investigate the proportion of diastolic dysfunction (DD) at baseline, and changes in cardiac function from baseline (T1) to 6 months follow-up (T2) among participants with severe obesity attending a lifestyle-intervention. The secondary aim was to explore changes in body mass index (BMI), physical fitness (VO_2peak) and cardiovascular risk from T1 to T2 and 12 months follow-up (T3).

Design. This was an open single-site prospective observational study. Patients were recruited from an obesity clinic to a lifestyle-intervention consisting of three 3-weeks intermittent stays over 12-months period. Echocardiography was performed at T1 and T2 and BMI, VO_2peak and cardiovascular risk measured at T1, T2 and T3.

Results. Fifty-six patients were included (mean age 45.1 years; BMI 41.9). Six of 52 patients (12%) had grade 1 DD at T1, while five subjects had DD at T2. E/A ratio (11%, p = .005) and mitral deceleration time (9%, p = .014) were improved at T2. A reduction in BMI (–1.8, p < .001) and improvement in VO_2peak (1.6 mL/kg min, p = .026) were assessed at T2 and this improvement persisted at T3. The total cardiovascular risk score was not significantly changed.

Conclusion. The patients with severe obesity had low prevalence of DD. For all participants, an improvement in diastolic parameters, and an important initial weight loss was observed.

Clinical Trial number: NCT02826122

Keywords:

• Severe obesity
• left ventricular diastolic function
• cardiovascular risk
• life-style intervention

Introduction

Obesity has become a serious health problem worldwide. In Norway, several population surveys show a dramatic increase in prevalence of obesity (body mass index (BMI) ≥30 kg/m²) from the middle of 1980s (8%) until 2016 (24%) among both genders [1,2]. The proportion of people with severe obesity with BMI ≥40 kg/m² is high and increasing [1,2]. Severe obesity is characterised as a chronic disease and is a complex, multifactorial condition, with genetic, epigenetic, behavioural, socioeconomic and environmental origins [3]. Living with severe obesity is associated with a high risk of comorbidities such as hypertension and type 2 diabetes mellitus (T2D) and alterations of cardiac structure and function, especially diastolic dysfunction (DD) and heart failure [4–6]. DD is associated with hypertension and T2D, but also seen in patients with severe obesity without hypertension or T2D. Heart failure with preserved ejection fraction (HFpEF) starts with DD, resulting in reduced compliance mostly because of left ventricle (LV) hypertrophy. The mildest form of DD (grade 1) is a reduced relaxation of the LV, which is a common finding in normal ageing, early grades of hypertension and in T2D. In grade 3 DD, the filling of the ventricle is not adequate, leading to an elevated filling pressure of the LV and potentially heart failure [4,6]. It is therefore important to treat comorbidities and stop further weight gain by medication and lifestyle changes to avoid these compensatory mechanisms over time [4].

Gastric bypass surgery can result in significant weight-loss in patients with severe obesity in the long term (5-years follow-up) [7]. Lifestyle interventions can also result in clinically relevant weight loss (>10%) in the long-term and one study showed that approximately 25% of patients with severe obesity were able to maintain weight-loss 5 years after a lifestyle intervention [8]. However, weight loss and improvements in risk factors, T2D and hypertension resolution are larger for those having bariatric surgery [7,8].

Beside weight-loss, conservative lifestyle interventions including physical exercise and dietary counselling has the potential to influence several cardiovascular risk-factors such as physical fitness, hypertension, cholesterol, triglycerides and fasting glucose [9]. Physical inactivity is also an independent risk factor and a strong predictor for future cardiovascular risk [10,11].

Massive weight-loss following bariatric surgery seems to reverse some of the left ventricular dysfunction seen in severe obese [12,13]. Similar modifications are found in moderately obese men following conservative interventions such as physical exercise [14]. However, whether similar effects are seen following conservative lifestyle interventions when a smaller weight-loss is expected are debated.

In Norway, several institutions offer conservative lifestyle interventions for patients with severe obesity with different types and lengths of follow-up. A follow-up time between 2 and 5 years is usual. In this prospective observational study, we have investigated patients with severe obesity referred to a conservative intermittent lifestyle intervention and followed them from program start through 12 months.

The primary aim of the study was to investigate the proportion of the sample having DD at baseline, and then changes in cardiac function from baseline to 6 months. The secondary aim was to explore changes in physical fitness (VO_2peak), BMI, physical activity level and cardiovascular risk profile from baseline to 6- and 12 months follow-up.

Methods

Study design, study population and recruitment

This is a pragmatic prospective observational “real world” pilot study among patients with severe obesity referred to a conservative intermittent lifestyle program at the Norwegian Heart and Lung Association rehabilitation clinic at Røros (LHL-R) with three intermittent stays over 12 months. The patients enrolled in the study was referred from the Obesity clinic at St. Olavs University Hospital, Trondheim, Norway. According to the national guidelines in Norway, patients with reduced health-related quality of life and BMI > 40 kg/m² or a BMI > 35 kg/m² with weight-related comorbidities should be referred to the specialist health care service if the patient has not achieved treatment goals in the primary health service within a reasonable time [15]. Primary treatment choice is lifestyle treatment with focus on physical activity, dietary guidance and coping psychology.

Therefore, criteria for inclusion to the conservative intermittent lifestyle program in this study were BMI > 40 kg/m² or a BMI > 35 kg/m² with weight-related comorbidities. All patients referred to the program in the period between August 2016 and February 2017 were asked to participate in the study (T1). Six-month (T2) and one-year (T3) follow-up were completed between June 2017 and February 2018, respectively.

The lifestyle intervention

The patients had three intermittent stays during 12 months at LHL-R. Each stay lasted for three weeks. Data were collected at T1, T2 and T3.

A multidisciplinary team consisting of a physician, dietician, physiotherapist, occupational therapist, psychiatric nurse trained in behavioural cognitive therapy (BCT) were responsible for the program, which was based on current guidelines for the management of overweight and obesity in adults [16].

The weight-loss program included physical activity, dietary sessions and BCT on five weekdays during each stay. The physical activity part included two hours of theory in plenary and 31 h of practical focusing on being physically active through endurance and strength exercise. Dietary advice was offered in a one-hour lecture and four hours of group work, focussing on calories and nutrition, two hours of examining the ingredients and nutrition of groceries, and four hours of cooking. Dietary sessions focussed on planning and preparing meals and ways to modify eating behaviour. Personalised meal plans and dietary advice targeted a calorie reduction of approximately 600 Kilocalories a day. Individual calorie restricted diets were based on the given estimate for resting metabolic rate. BCT included three hours of plenary sessions, four hours group work and individual sessions. BCT was aimed at goal setting and increased awareness of maladaptive cognitions that contribute to the maintenance of emotional distress and problematic eating behaviour [17]. Topics covered were expectations towards attending the weight-loss program, motivation and conflicting interests, excuses, compensatory strategies and foundation for change.

Each patient had a primary contact who performed an individual health conversation at the beginning and at the end of each intermittent stay where goals were discussed, put together and written down. Between the stays, the participants were encouraged to apply their acquired knowledge from the stay at the clinic and use their individual exercise schedule at home.

Endpoints and assessment

Outcome measures were assessed at baseline (T1), during the second (T2) and third (T3) stay. Body weight, height, resting blood pressure, cardiopulmonary exercise testing (CPET), echocardiography and blood samples were collected at T1 and T2. Blood samples were taken at T3 together with CPET assessments, body weight and resting blood pressure. Self-reported questionnaires were filled in at T1, T2 and T3.

Echocardiography

Echocardiography was performed to assess ventricular structure and function according to standard methods [18,19] by two operators using Vivid E9 scanner (GE Healthcare, Horten, Norway) with offline analyses (EchoPac version 201, GE Healthcare) equipped with a 1.4 − 4.6 MHz transducer (GE-M5Sc-D XDClear, GE Medical Systems). Pathological findings were recorded.

DD was measured according to the recommendations from 2009 [18], because our study started before the new recommendations from 2016 [20]. Achieving enough parameters was difficult because of poor acquisition in many of the study participants. Therefore we used the parameters we had in most of the participants according to the 2009 recommendations to describe DD: (1) E wave velocity (maximum early diastolic E wave velocity at transmitral blood flow Doppler); (2) MDT (mitral E wave deceleration time); (3) septal e′ velocity (maximum E wave velocity from tissue Doppler imaging); (4) E/A ratio (early to late mitral wave velocity at transmitral blood flow Doppler) and (5) E/e′ (the ratio between early transmitral filling (E) and the corresponding myocardial tissue velocity (e′)) measured in basal septum was used as an index of LV filling pressure. DD was pragmatically described with at least three out of four parameters describing different grades: grade I; E/A-ratio < 0.8, DT > 200 ms, e′ < 8, E/e′<8, grade II; E/A-ratio 0.8–1.5, MDT 160–200 ms, e′ 9–12, E/e′ 9–12, or grade III; E/A-ratio ≥ 2.0, DT < 160 ms, e′< 8; E/e′ > 13).

LV mass (LVM) was estimated using the Devereaux formula (LVM = 0.8 × 1.04 × [(IVSd + LVIDd + PWTd)3 – LVIDd3] + 0.6 normalised for height powered by ^2.7 [21,22], to achieve LVM index in subjects with obesity (LVMI) [18]. The Simpson 's biplane method was used to calculate left ventricular systolic function (ejection fraction = EF) from apical two and four chamber projections of ventricular volumes.

Cardiopulmonary exercise testing

Aerobic capacity was assessed by direct measurement of oxygen consumption walking or running on a treadmill. Prior to entering the treadmill (Woodway Inc., WI, USA), patients were equipped with a heart rate monitor (Polar WearLink; Polar Electro Oy, Kempele, Finland), and a fitted face mask (Hans Rudolph, Shawnee, KS, USA) or mouthpiece (Hans Rudolph, Shawnee, KS, USA) if having beard/moustache. Oxygen uptake and heart rate were measured continuously during an incremental, individualised protocol chosen by experienced test personal until exhaustion. All VO₂ kinetics were measured using the Jaeger Masterscreen CPX (CareFusion, Hoechberg, Germany). A test was considered maximal (VO_2max) if the VO₂ did not increase more than 2 mL × 1/kg × 1/min despite increased workload, combined with a respiratory exchange ratio (RER) at or above 1.1. Since few patients reached both VO_2max criteria, the term VO_2peak was used. The VO_2peak was registered as the mean of the three successively highest 30 s VO₂ with simultaneous RER ≥ 1.05.

Self-reported physical activity and psychological status

The participants completed a three-item questionnaire on leisure time physical activity habits developed for use in the Nord Trondelag Health Survey (HUNT1-PAQ). Frequency was assessed as “How often do you exercise?”, with five response alternatives (scoring): never (0), less than once a week (0.5), once a week (1), 2–3 times a week (2.5), almost every day (5). Intensity was assessed as "How hard do you push yourself?", with three response alternatives: no sweat or heavy breathing (1), heavy breath and sweat (2) and near exhaustion (3). Finally, duration had four response alternatives: <15 min (0.1), 15–29 min (0.38), 30–60 min (0.75) and >60 min (1.0). Each participants’ physical activity index (PA Index) score was calculated multiplying frequency, intensity and duration (0–15). An index score ranging 0 − 1.50 considered to signify low levels of activity, 1.51 − 3.75 as medium level of activity and equal or above 3.76 signified high level of activity. The PA Index is previously established as valid and reliable [23].

They completed the Hospital Anxiety and Depression Scale (HADS), a 14 items checklist in which seven items relate to anxiety and seven items to depression. Scoring for each item ranges from 0 to 3, with 3 as highest anxiety or depression level. A total subscale score of >8 points out of a possible 21 mean considerable symptoms of anxiety or depression. Original cut-offs for anxiety and depression have been defined as 0–7: no relevant symptoms, 8–10: possible anxiety and/depression and 11–21: significant anxiety/depression [24,25].

Blood samples and cardiovascular disease risk factors

Fasting venous blood samples were analysed for triglycerides (TG), low-density lipoprotein (LDL) (analysed directly), high-density lipoprotein (HDL), total cholesterol, glycosylated haemoglobin (HbA1c), haemoglobin (Hb), serum C-reactive protein (CRP) and N-terminal B-type natriuretic peptide (NT-proBNP) at department of Clinical Chemistry at St. Olavs University Hospital, Trondheim, Norway.

Using an established cardiovascular disease (CVD)-risk prediction model (NORRISK 2), we predicted 10-year CVD-risk (non-fatal myocardial infarction, fatal coronary heart disease and non-fatal or fatal stroke) based on established risk factors (age, smoking, systolic blood pressure, anti-hypertensives, total and HDL-cholesterol, and family history of myocardial infarction). CVD risk profile in NORRISK 2 is defined as low if <4%, <8% and <12% in the different age groups 45–54 years, 55–64 years and 65–74 years, respectively. Cardiac risk profile using NORRISK 2 was calculated individually in percent risk for each participant from a computerised calculator [26].

Statistical analyses

IBM SPSS Statistics (version 25/26) was used for statistical analysis. Descriptive data are presented as mean ± standard deviation (SD) for continuous variables and percentages for categorical variables. All analyses are presented for the total sample and shown for both genders according to socio-demographic data and comorbidities at baseline. The data set was examined for erroneous outliners and each outcome variable was tested for normality distribution (skewness, kurtosis and plots). Continuous normally distributed baseline data were analysed with an independent t-test to test for differences between groups. Linear mixed models (LMM) were used for analyses of level and change over time in outcome variables [27]. First, for variables measured over three points of time unconditional models were estimated to decide whether change was linear or not. The results showed that two slope factors improved model fit compared to linear models, thus representing different changes over two intervals [28]. In order to make these models possible to estimate, the slope factors were fixed. Thus, the final analyses were random intercept fixed slope models [28]. The standard model tested homogenous residuals. In contrast to the standard t-test of change, LMM partialize out some change as arbitrary residual variance and estimate true underlying latent change. In addition, LMM uses all available data, even if only one or two of the three measurements were present for a case. Therefore, LMM assumes Missing at Random (MAR), and not Missing Completely at Random (MCAR) [29]. Due to the pilot nature of the study, no adjustment for multiple comparisons were made. This strategy will increase the probability for false positive findings (type I error) but at the same time keep the risk for rejecting true findings (type II error) at a minimum level.

The statistical significance was set to p < .05.

Ethical issues

The study was approved by the regional committee for medical and health research ethics in Central Norway (REK Midt 2016/833). The study is registered in the ClinicalTrials.gov-registry under the unique trial number NCT02826122. Written informed consent was obtained from all the participants. All standard safety measures and processes for the ethical handling of human subjects were adhered to in this study.

Results

A total of 56 subjects (61% women) out of 61 eligible agreed to participate and were included in the study. Withdrawal from attending the second stay (T2) resulted in 47 subjects (84%) at T2 and 38 subjects (70%) in the last 3-week stay (T3). Reasons for not attending at T2 and T3 are shown in Figure 1. There were no differences in baseline characteristics such as age, sex, years with obesity, blood pressure, body weight, BMI, VO_2peak, PA Index or NORRISK 2 between those who completed all stays (38 subjects) and dropouts at T2 and T3 (data not shown). However, we found a significantly higher score in both anxiety (10 ± 4 vs 5 ± 4, p < .001) and depression (7 ± 4 vs 5 ± 3, p = .036) among the dropouts compared to the completers.

Figure 1. Flow-chart of patient recruitment to follow-up.

Baseline characteristics are presented in Tables 1 and 2. Hypertension was the most common comorbidity with 45% of all, 37% reported sleep apnoea while 25% had T2D.

Table 1. Socio-demographic data and comorbidities at baseline.

	Total sample (n = 56)	Men (n = 22)	Women (n = 34)
Age (years), mean (SD)	45.1 ± 11.6	44.5	45.4 ± 11.5
BMI (kg/m^2) at referral, mean (SD)	41.9 ± 6.0	43 ± 7.4	41.1 ± 4.9
Years with obesity (years), mean (SD)	26.9 ± 14.1 (n = 51)	26.8 ± (n = 20)	27 ± 15.1 (n = 31)
Marital status
Married/cohabitant, n (%)	45 (80.4)	18 (81.8)	27 (79.4)
Single, n (%)	11 (19.6)	4 (18.2)	7 (20.5)
Highest completed education
Primary school, n (%)	7 (12.5)	3 (13.6)	4 (11.6)
Secondary school, n (%)	28 (50)	12 (54.5)	16 (47.1)
College/University, n (%)	21 (37.5)	7 (31.8)	14 (41.2)
Work status
Employed, n (%)	35 (62.5)	11 (50)	24 (70.6)
Student, n (%)	3 (5.4)	2 (9.1)	1 (2.9)
Unemployed/job seeker, n (%)	2 (3.6)	1 (4.5)	1 (2.9)
Sick-leave/work assessment allowance (AAP), n (%)	11 (19.7)	5 (22.7)	6 (17.7)
Other pension or benefit, n (%)	4 (7.2)	2 (9.1)	2 (5.9)
Comorbidities (ICD-10-diagnosis)
Type II diabetes, n (%)	14 (25)	6 (27.3)	8 (23.5)
Sleep apnoea, n (%)	21 (37.5)	10 (54.5)	11 (32.4)
Hyperlipidaemia, n (%)	17 (30)	7 (32)	10 (29)
Hypertension, n (%)	25 (44.6)	13 (59.1)	12 (35.3)
Arthrosis, n (%)	11 (19.6)	4 (18.2)	7 (20.6)
Musculoskeletal pain, n (%)	27 (48.2)	8 (36.4)	19 (55.9)
Smoking habits
Never, n (%)	16 (28.6)	6 (27.3)	10 (29.4)
Previous smoker, n (%)	29 (51.8)	10 (45.5)	19 (55.9)
Smoking, n (%)	6 (10.7)	4 (18.2)	2 (5.9)

Table 2. Clinical examination among all at baseline (n = 56).

BP, sys, mmHg	126 ± 10
BP, dias, mmHg	80 ± 6
Blood tests
Fasting glucose
mmol/L	5.8 ± 1.2
mg/dL	104 ± 22
HbA1c
%	5.8 ± 0.9
mmol/mol	40 ± 10
Total cholesterol mmol/L	4.6 ± 0.9
HDL-cholesterol mmol/L	1.2 ± 0.3
LDL-cholesterol mmol/L	2.9 ± 0.9
Triglycerides mmol/L	1.4 ± 0.4
Hb g/L	142 ± 11
CRP mg/L	11.0 ± 15.9
NT-proBNP ng/L	60.1 ± 70.8
NORRISK 2, %	2.3 ± 2.1
Self-reported physical activity
PAQ frequency	1.4 ± 1.2
PAQ intensity	1.7 ± 0.5
PAQ duration	0.6 ± 0.3
PA Index	1.7 ± 1.9
CR exercise testing	(n = 50)
VO2peak
mL/kg min	25.1 ± 4.5
L/min	3.0 ±  0.8
RER	1.1 ± 0.1
Peak heart rate beats/min	172 ± 16
HADS total	12.2 ± 6.9

BP, blood pressure; chol, cholesterol; CRP, C-reactive protein; VO₂, oxygen uptake; RER, respiratory exchange ratio; HADS, hospital anxiety and depression scale.

The mean baseline echocardiographic data were all within normal limits including four diastolic parameters, LV dimension and mass and LVEF% (Table 3). Six subjects out of 52 (1211%) with eligible echocardiographic findings were having DD and characterised with grade 1 DD at T1, while five subjects were found with DD at T2. All with DD at T1 were older than 53 years and five were women (see Table 4 for details). Two subjects with DD were found with both T2D and hypertension, while four more had either T2D or hypertension (Table 4).

Table 3. Echocardiographic parameters among subjects with severe obesity at T1 and T2.

	T1 (n = 46–56)	T2 (n = 33–41)
BSA, cm^2	2.41 ± 0.25	2.35 ± 0.25
LVEF, %	59 ± 8	64 ± 5
LVEDV, mL	105.1.2 ± 28.4	101.4 ± 29.1
LVSV, mL	43.6 ± 15.7	36.0 ± 11.9
LVIDd, cm	5.2 ± 0.4	5.0 ± 0.5
IVSd, cm	1.0 ± 0.1	0.9 ± 0.1
LVPWd, cm	0.9 ± 0.1	0.9 ± 0.1
LVMI, g/m^2.7	41.4 ± 8.1	40.0 ± 8.6
RWT	0.36 ± 0.06	0.37 ± 0.05
LA diam, cm	3.4 ± 0.6	3.5 ± 0.5
E wave, cm/s	7.5 ± 1.7	7.8 ± 1.8
E/A ratio	1.1 ± 0.3	1.1 ± 0.4
MDT, ms	235 ± 55	218 ± 53
e′, cm/s	10.9 ± 4.5	10.0 ± 2.7
E/e′ ratio	7.2 ± 1.8	8.3 ± 3.1

Values are mean ± SD; BSA, body surface area; LVEF, left ventricle ejection fraction; LVEDV, LV end diastolic volume; LVSV, LV systolic volume; LVIDd, LV internal diameter in diastole; IVSd, interventricular septum in diastole; LVPWd, LV posterior wall thickness in diastole; LVMI, LV mass index relative to BSA (g/cm²) and height in obesity (g/m^2.7); RWT, relative wall thickness; LA, left atrium; E, maximum early diastolic E wave velocity at transmitral blood flow Doppler; A, maximum late diastolic A wave velocity at transmitral blood flow Doppler; MDT, mitral deceleration time; e′, maximum E wave velocity from tissue Doppler imaging.

Table 4. Diastolic assessments in six subjects at T1 and five subjects at T2 with diastolic dysfunction (three out of four parameters).

	E/A ratio	T1	n = 6	E/e′ ratio		T2	n = 5
Age, years (sex)		MDT msec	e′ cm/sec		E/A	MDT	e′	E/e′	CM
66 (man)	0.72	264	6	8	0.80	209	6	7	HT
59 (woman)	0.76	256	7	9					HT, DM
57 (woman)	0.68	272	7	9	0.64	313	5	12	HT
58 (woman)	0.63	183	7	8					HT, DM
56 (woman)	0.68	264	7	9	0.77	222	7	10
53 (woman)	0.74	235	8	8					DM
39 (woman)					0.78	326	5	9	DM
53 (woman)					0.74	230	11	6
All	0.72	261	7	8	0.82	273	8	8

E, maximum early diastolic E wave velocity at transmitral blood flow Doppler; A, maximum late diastolic A wave velocity at transmitral blood flow Doppler; MDT, mitral deceleration time; e′, maximum E wave velocity from tissue Doppler imaging; CM, co-morbidities; DM, type 2 diabetes mellitus, HT, hypertension.

The LMM analyses showed a significant improvement in E/A ratio (11%, p = .005) and MDT (9%, p = .014), but a small increase in E/e′ ratio from mean value of 7 to 8 (p = .023) was observed from T1 to T2 (Table 5). There was an improvement in LVEF%, but no change in left ventricular dimensions or geometry.

Table 5. Linear Mixed Model (LMM) results of level and changes in diastolic echocardiographic data and ventricular function and geometry from T1 to T2.

		Baseline		Change
	N	Mean	SD	Mean	P
E/A Ratio	53	1.12	0.29	0.12	.005
MDT, ms	52	235.40	38.19	–20.81	.014
e′, cm/s	54	0.11	0.03	–0.01	.189
E/e′ Ratio	53	7.42	2.19	0.79	.023
EF Biplan	47	58.57	3.29	5.46	<.001
LVIDd	51	5.17	0.30	–0.13	.092
RWT	51	0.36	0.02	0.01	.359
LVMI-Obese	51	41.37	5.22	–1.39	.366

E, maximum early diastolic E wave velocity at transmitral blood flow Doppler; A, maximum late diastolic A wave velocity at transmitral blood flow Doppler; MDT, mitral deceleration time; e′, maximum E wave velocity from tissue Doppler imaging; EF, ejection fraction; LVIDd, left ventricular internal diameter in diastole; RWT, left ventricular relative wall thickness; LVMI-Obese, left ventricular mass index in obese subjects.

The LMM analyses showed a reduction in body weight (–5.6, p < .01) and BMI (–1.8, p < .001) and improvement in VO_2peak (1.6 mL/kg min, p = .026) after 6 months. This improvement persisted at 12 months follow-up. Together 20 out of 47 subjects (43%) had a weight loss of more than 5% after 12 months (data not shown). The mean cardiovascular risk score was low at baseline (2%) and did not change significantly from baseline to neither 6- nor 12 months follow-up. Physical activity level (PA Index) increased significantly from baseline to 6-month follow-up to medium level of activity. At 12-month follow-up the PA Index remained at the same level (See Table 6 for details).

Table 6. Linear Mixed Model (LMM) results of level and changes in body weight, physical fitness and activity, and cardiovascular risk profile from T1 to T2 and from T2 to T3.

		Baseline (T1)		Change T1–T2		Change T2–T3
	N	Mean	SD	Mean	p	Mean	p
Weight (kg)	56	121.68	21.61	–5.26	<.001	–0.43	.666
BMI	56	40.48	5.52	–1.77	<.001	–0.11	.731
VO2peak (mL/kg min^))	53	25.03	3.15	1.61	.026	–0.64	.437
PA Index	56	1.76	1.30	1.69	<.001	–0.69	.061
NORRISK 2(%)	55	2.27	2.15	0.20	.090	–0.04	.720

BMI: Body mass index; VO_2peak: peak oxygen uptake; PA Index: physical activity index; NORRISK2: cardiovascular risk profile.

Discussion

The main finding was that the relatively young patients with severe obesity referred to an intermittent lifestyle program had a low proportion of DD (12%). All six with DD at baseline were classified with DD grade 1, a mild form of relaxation abnormality [20], and there was an improvement in cardiac diastolic parameters from T1 to T2. In addition, an improvement in self-reported physical activity and fitness, and a significant mean weight loss between baseline and follow-up after 12 months were found.

Cardiac structure and function

The mean age among all participants in this study was 44 years and they had suffered with severe obesity for a mean time of 27 years when they were included. Both T2D and hypertension are known as important risk factors for increased left ventricular mass and concentric remodelling leading to heart failure with preserved ejection fraction (HFpEF) [30], and there is a strong correlation between severe obesity and T2D and/or hypertension [31,32]. Although many of the participants in our study had hypertension (45%) and one out of four had T2D, a low proportion of diverging cardiac structure was found. This could be explained by a young, well-treated population with HbA1c and blood pressure within normal ranges together with a low cardiovascular risk score. All with DD at baseline were above 53 years old at baseline, and all except one were above 53 years old at T2. Also in the general population, relaxation abnormalities are increasing with increasing age [19].

Our finding of a proportion of DD of 12% in patients referred to lifestyle program for severe obese is in line with a study among 100 patients with a mean age of 48 years and BMI ≥35, who found a prevalence of 13% [32]. However, a study among elderly (71 years old) obese patients (BMI 34) the prevalence of DD was as high as 53% [33]. The proportion of participants with DD in our population is difficult to compare to the prevalence in the general population without severe obesity, because the results from the general population differ from as high as 38% in studies described in the 2009 recommendations to as low as between 1% and 2% in studies described in the 2016 recommendations, respectively. The diastolic parameters used to describe DD in different studies are found to strongly influence the reported prevalence [34,35].

The improvement in E/A ratio and MDT with 11% and 9% from T1 to T2, respectively. Still, increased physical activity and reduced body weight may have contributed to an improvement in diastolic function [36]. The small increase in E/e′ is not clinically relevant and within normal ranges.

In contrast to our findings of mostly normal cardiac structure and function, the Northern Manhattan Study in an older population found that changes in cardiac structure and function adjusting for T2D and hypertension may start already among those being overweight as the only factor (BMI > 25) [33]. Another study found that subjects with obesity develop changes in left ventricular structure and function without having known risk factors for CVD [37]. Others have described that severe obesity is associated with different changes in cardiac structure and function, probably because of increased blood volume leading to increased cardiac output and volume overload on the ventricles [4,38]. Cardiac remodelling in obesity is found to be associated with DD, HFpEF or obesity cardiomyopathy, as shown in studies described in the review by Wiling and Jacob [39]. Still, there is a lack of knowledge of the mechanisms leading to cardiac dysfunction in obesity [40]. The term obesity cardiomyopathy has been introduced, which is classified as a subtype of dilated cardiomyopathy, characterised by changes in ventricular structure and function including left ventricular dilatation, eccentric or concentric left ventricular hypertrophy, systolic and DD and right ventricular dysfunction seen in patients with severe obesity [41]. None in our sample fulfilled the criteria for HFpEF, which in the review from Mishra and Kass [42] is characterised as a multisystem disorder, including the adipose tissue described as an organ.

CVD risk factors

CVD risk factors associated with obesity as hypertension, sleep apnoea and T2D were common in our sample. However, they had a low cardiovascular risk score. The lowest age for scoring in NORRISK 2 [26] is 45 years, and some of the participants were younger and set at 45 years old.

Therefore, NORRISK 2 score seems not to be a good indicator for 10-year CVD-risk in this relatively young population with severe obesity, even though many studies have shown a higher risk among subjects with obesity compared to the general population [3]. Older age is a strong predictor of CVD risk together with smoking, high cholesterol and high blood pressure [43], that could explain the low risk in this study. No CVD risk scoring system today includes all other known risk factors among subjects with obesity such as depression/anxiety, BMI, waist and hip measurements, metabolic syndrome, T2D and low physical activity [26,43]. It is debated if higher fitness is associated with lower prevalence of risk factors for CVD in people with obesity. One study found that improved fitness was not associated with lower prevalence of T2D, metabolic syndrome or hypertension [44].

Changes in the cardiovascular risk score were not observed following the conservative intermittent lifestyle intervention in our study. The cardiovascular risk score was low at baseline, and we could therefore not expect large changes after the intervention. In our intervention the focus was on calorie restriction and not glycemic index. According to a recently published review, nutritional components with the low glycemic index or load diet along with intermittent lifestyle should be tested in a future study [45].

Weight loss

A significant weight loss was found from baseline to 6 months follow-up. The weight loss persisted at 12-month follow-up. This is in line with results from earlier studies [45,46], and may reflect that the enthusiasm and motivation among the participant is strongest at the first part of the lifestyle program [46,47]. Forty-three percent of the participants in our study had weight loss of more than 5% after 12 months. In another study attending an intensive-life-style group, 51% maintained 5% weight loss after a longer follow-up time of 24 months, but the largest weight loss occurred at 6 months follow-up [48]. The initial weight loss was the most important predictor for successful weight loss in the study by Chopra et al. [49].

Study strengths and limitations

A strength of the study is the real life setting from a centre with a lifestyle program for people with severe obesity, that is not willing to or found suitable for bariatric surgery. This study was a pilot and feasibility study, it was not designed as an “effect study”, which imply that the study has some limitations. Major limitation is therefore the lack of a randomisation to an active intervention or a control group (care as usual) or comparison of the effect of the life-style intervention with other groups or other interventions. Next, our study is small with a low number of patients included, which do not allow us to look at possible predictors for successful improvements in cardiac and life-style parameters and weight loss. Neither was it possible to examine sub-groups, because of limited participants included. Unfortunately, we have no data on the severity of sleep-disorder breathing except for that 38% of the sample reported to have sleep apnoea.

In addition, we followed the patients for only one year (six months regarding diastolic parameters). Future studies should follow the patients longer, especially concerning the diastolic parameters. A relative high drop-out rate also leads to limitations. Respectively 17% and 32% of the subjects dropped out at 6- and 12 months follow-up, so the data must be considered with caution. Unfortunately, no data were available on describing the dropouts after 12 months. Diastolic parameters like MDT obtained by Doppler flow analysis are load dependent and may vary between the two measurements. Another limitation is the use of the recommendations from 2009 and not from 2016. Still, the use of three different diastolic parameters to classify DD, including not load-dependent tissue Doppler, is supposed to give reasonably good estimation of subjects with DD.

The use of the CVD risk score NORRISK could have been supplemented by another score system including some younger subjects from 30 years old. NORRISK 2 was used following the development of a revised and updated national risk algorithm that was implemented by the Norwegian health authorities from 2017 onward. The Framingham risk score works from age 30 and should preferably have been used in this study together with the NORRISK 2 or alone [50].

The intervention in this study was multimodal including BCT. Thus, the effect of each individual component is not possible to explore in this design. A future study could be more tailored and adapted to the different characteristics of the individuals included for example concerning different racial and ethnic populations.

Conclusion

The proportion of patients with DD was low among the subjects with severe obesity included. This may be due to a relatively young population with well-treated T2D and hypertension, and a low cardiovascular risk score. The lifestyle program resulted in significant and important initial mean weight loss. Those who dropped out of the program had significantly higher anxiety and depression scores, showing that good mental health is important for adherence to the program. Future studies should follow the patients attending lifestyle programs concerning the diastolic function and cardiovascular function for an extended period using a randomised design.

Acknowledgements

The authors thanks to the subjects who volunteered to participate in this trial, for the contribution from the staff at LHL-R, and cardiologist Ivan Popovic for analysing echocardiographic data. The study is financed by grants from the DAM-foundation

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was financed by fundings from the DAM-foundation in Norway.