Non-contact radiofrequency-induced reduction of subcutaneous abdominal fat correlates with initial cardiovascular autonomic balance and fat tissue hormones: safety analysis

Background and objective: The non-invasive reduction of subcutaneous abdominal fat became popular in the last decade. Radiofrequency (RF), non-contact, selective-field device Vanquish® has been developed to selectively induce deep fat tissue heating to reduce waist circumference. Our analysis evaluates immediate and sustained effects of this treatment on cardiovascular autonomic function and on selected metabolic parameters. Study design/ patients and methods: A retrospective proof-of-concept analysis of RF treatment effects was conducted in 20 individuals with metabolic syndrome, to reduce the subcutaneous abdominal fat. Four 30-minutes treatment sessions (manufacturer´s standard protocol) were performed in 1-week intervals. Vital signs, ECG, lab screening, body composition, subcutaneous fat thickness and spectral analysis of heart rate variability (HRV) have been examined before, after the 1 st and 4 th treatment, and at follow-up visits 1 month and 3 months after the treatment. Results: The RF treatment led to a significant reduction of abdominal circumference after the 4 th session (p<0.001), and during follow-up after 1 and 3 months (p<0.001 and p<0.02, resp.). There was a significant correlation (r=-0.58, p=0.007) between reduction of abdominal circumference and initial very-low frequency (VLF) spectral power at 1 month follow-up. A significant increase of cumulative spectral power in low frequency (p=0.02) and reduction in high frequency (p=0.05) band have been observed immediately (20 +14 minutes) after the treatment. On the contrary, no sustained impact on autonomic balance has been recorded 39 +18 days after the treatment. A significant correlation between the initial adiponectin values and immediate autonomic response to one treatment was observed in VLF and total spectral bands (r>0.59, p<0.04). Conclusions: Our analysis shows that the selective-field RF treatment is safe and efficient for reduction of subcutaneous abdominal fat. While the treatment increases the immediate sympathetic response of the body to deep tissue heating, no sustained change in autonomic function could be recorded at 1 month follow-up. The observed correlation between initial VLF spectral power and waist circumference reduction at follow-up, as well as the association of initial adiponectin values and immediate autonomic response to the treatment might be instrumental for decisions on body contouring strategies.

Obesity considerably impairs individual health and aesthetic appearance. In addition to its known health impacts --reduction of life expectancy and quality of life 1 --obesity leads to numerous problems including disadvantages in employment 2 , in social interactions and decreased satisfaction with own body image 3 . These aspects lead to social pressure and subsequently to increased demand for effective procedures for weight reduction, body contouring and beauty enhancement.
Central obesity is associated with insulin resistance and related components of metabolic syndrome that can be typically treated by nutritional, behavioural and lifestyle changes 1 . Although the reduction of subcutaneous fat alone does not lead directly to reduction of cardiovascular risk in obese subjects 4 , there is some evidence that the large-volume liposuction might positively influence the insulinemia 5 and thus insulin sensitivity. Furthermore, clinical experience shows that aesthetic procedures leading to improved patient's self esteem 6 often significantly enhance the motivation to further lifestyle changes towards healthier goals.
Various invasive and particularly non-invasive body contouring procedures for reduction of subcutaneous fat layers have been introduced in the last decade. While the surgical liposuction still counts for the most effective gold-standard procedure in this respect 7 , due to its invasiveness, downtime and side effects, a bunch of non-or semi-invasive procedures became available as its indirect alternative on the quickly growing (often called "lunch-time-procedure") market 8 . However, despite many individual -often only anecdotal -user reports, only a minority of methods is proven according to the evidence-based medicine standards. Such evidence is available for efficacy of chemical lipolysis, based on injection of phosphatidylcholine and deoxycholic acid 9 , and for selected energy-based technologies using focused ultrasound 10 , cryolipolysis 11 and/or thermal/radiofrequency for lipolysis 12 . Despite the broad use in the practice, clinical safety data of these aesthetic procedures are scarce, with only a few publications available 13 . Although these intensive procedures might have a significant impact on autonomic homeostasis and individual health, to our knowledge, no immediate or sustained effects of such treatments on autonomic function have been investigated.
Analysis of beat-to-beat fluctuations of heart rate (heart rate variability, HRV) is an established tool to non-invasively quantify cardiac autonomic function 14 . The frequency (spectral) decomposition and quantification of irregular course of heart rate into three main frequency bands allows a detailed view of different domains of the cardiovascular control. The short-term HRV spectral analysis is proven useful for assessment of impact of various physiological stimuli on the body such as food restriction 15 , endurance physical training 16 or guided breathing 17 . In particular, the very low frequency (VLF) spectral band has been shown to reflect thermoregulatory vasomotor mechanisms, changes in peripheral chemoreceptor activity and fluctuations in renin-angiotensin systems 14,18 . In this respect, analysis of the VLF band enables the quantification of sympathothermogenic autonomic responses related to energy metabolic control, as it has been demonstrated e.g. by an acute cold exposure, spicy food containing capsaicin and green tea extract or low-calorie diet [19][20][21][22] .
The selective-field radiofrequency device Vanquish ® , using electromagnetically induced rapid oscillations of electrical dipoles to heat up the fatty tissue 23 , is increasingly being used for reduction of subcutaneous abdominal fat. Its efficacy has already been demonstrated 24 . However, although the reported patient acceptance of these treatments was well to superlative 24 , no metabolic and/or safety data have been published yet. Our aim was therefore to evaluate the safety and efficacy of this novel technology in a proofof-concept retrospective data analysis of all clients who attended our clinic and were subjected to the treatment including follow-up within the 5-months time period. This paper focuses on the immediate and sustained effects of the treatment on the autonomic balance of the body and related metabolic values. The analysis is thought as a preparation for a further controlled prospective observation.

Study design
A retrospective, uncontrolled, single site proof-of-concept analysis of the impact of selective-field radiofrequency treatment on cardiovascular autonomic control and on selected metabolic data (insulin resistance parameters and fat tissue hormones) was conducted in overweight individuals with components of metabolic syndrome and visually detectable excess of subcutaneous fat who wished to reduce the abdominal circumference. This study was a retrospective proof-of-concept trial and the data have been retrospectively evaluated for the future preparation of a controlled prospective trial, which will require a submission to the local Ethics Committee. For the retrospective data evaluation no approval of IRB was required.
Data have been routinely acquired before, at visits immediately after the 1 st and 4 th treatments, and at follow-up visits in 1-month and 3-months after the last treatment. For assessment of metabolic data, blood sampling was performed before, on the next morning after the 1 st and 4 th treatments, and 1 and 3 months after the last treatment. Assessment of the intervention effect on the autonomic balance, using the standardized analysis of short-term heart rate variability as obtained during the modified orthostatic load 18,25 , was performed before, immediately (acute effect) after the 1 st treatment, and 1 month (sustained effect) after the last treatment. These data have been acquired during routine services of the clinic.

Inclusion and exclusion criteria
The selective-field RF treatment protocol has been offered to all individuals with visually excessive subcutaneous fat wishing to reduce their waist circumference. In the retrospective evaluation of efficacy and safety all patients have been included who accepted the necessity of follow-up investigations. Attendance of follow-up visits was a prerequisite for waiving their treatment fees. No reimbursement or coverage of travel expenses have been offered to these patients.
The following standard routine criteria of our clinic for exposition to RF treatment were applied: Inclusion criteria were age 20-70 years, both genders, BMI over 25kg/m 2 , abdominal circumference over 80 and 94 cm in women and men, respectively, with at least 20 mm of abdominal subcutaneous adipose tissue (as measured by calliper at predefined locations), stable weight over the last 6 months and signed informed consent on treatment. Exclusion criteria were pregnancy or insufficient contraceptive methods, surgical liposuction within the last 12 months, insufficiently controlled metabolic disease including diabetes mellitus of both types, untreated hypo-or hyperthyroidism, uncontrolled liver, kidney or cardiovascular disease, implanted pacemaker or metal implant, acute or feverish disease, history of thrombophlebitis, any haematological disease, chronic medication of corticosteroids, beta-blockers, anticoagulants, insufficient treatment adherence or any other clinical or biochemical condition bearing potential to interfere with the treatment targets. Females in child-bearing age were educated about necessary contraceptive methods, and those planning pregnancy in the following 12 months were not subjected to the RF treatment.

Patients
The study population consisted of n=20 Further details can be found in Table 1.

Intervention
The non-contact, selective-field radiofrequency system Vanquish ® (BTL Industries) has been used for treatment of subcutaneous fat layers 23 . All subjects underwent four 30 minutes treatment sessions in the abdominal area, with one week break between the sessions, as recommended in the standard treatment protocol by the manufacturer. Flat multipolar applicator panel was used for emitting the radiofrequency (27.12 MHz) energy for selective generation of deep tissue thermal heating of adipose tissue layers. The unit adjusts the parameters of the emitted energy in real time and shows the instantaneous value on display. This electromagnetic radiation is heating up the adipose tissue much more effectively then surrounding tissues, while limiting potential side effects due to minimized exposition of skin, muscles, or internal organs to this energy 23 . The treatment procedure consists of placing the emitting panel over abdomen and flanks close to the skin using a spacer which standardizes distance between the panel and the body surface. Once it is in a proper position, treatment can be started while the intensity of the emitted energy is set according to the protocol and to tuning efficiency of the system. The skin temperature is measured before, in 10, 20 and 30 minutes during the treatment, while the subject is frequently asked to give feedback on subjective thermal perception and to immediately report any pain or unpleasant sensations. Operator adjusts the emitted energy intensity close to the tolerable level according to client's feedback during the treatment procedure and to the measured skin temperature while the safety threshold is set to 42°C.

Routine clinical assessments
Blood sampling, autonomic balance evaluated by heart rate variability analysis, clinical assessment including vital signs, casual blood pressure, electrocardiogram (ECG), body composition evaluated by bioelectrical impedance measurements, abdominal circumference in three predefined points and anthropometric assessment by calliper were evaluated at predefined time points as indicated elsewhere.

Measurement of vital signs
Casual blood pressure has been measured in accordance with standard recommendations 26 in sitting position, using validated oscillometric automated monitor Omron M6 (Omron, Japan). The average from three measurements has been used for data analysis.
A calibrated, computer-assisted system ECG Seiva (Seiva, Czech Republic) has been used for 12-lead surface ECG recordings. Data have been electronically stored and evaluated by a single specialist experienced in ECG readings.
Height has been measured by validated ultrasound height measuring unit ADE MZ10020 (ADE, Germany) within standardized conditions as set by manufacturer.
Temperature before and during the treatment sessions has been measured by a calibrated non-contact infrared skin thermometer BaseTech IRT-350 (BaseTech, Germany) at predefined locations (around umbilicus and at upper and lower abdominal wall on both sides).
Weight has been measured within the body composition assessment as described below.

Assessment of weight and body composition
Body composition has been assessed by a calibrated scale, Omron BF 511 (Omron, Japan), a 8-sensor, one-frequency (50 kHz, 500 uA) bioelectrical body impedance analysis device, under strictly standardized conditions as set by the manufacturer. The device delivers along with weight and BMI also gender-specific percentage of body fat and muscle mass, basal metabolic rate (in kcal) and amount of visceral fat (arbitrary units). The declared weight measurement accuracy is 1% 27 .

Assessment of waist circumference
Waist circumference was measured using a measuring tape with a spring handle (www.netzwerk-lipolyse.de), in order to control for the pressure exerted on the patient's abdomen. Three measurements in different locations have been performed at the end of gentle expiration, in the standing position: horizontally around the patient's abdomen at its narrowest part (under the rib cage), at the level of the umbilicus, and 5 cm below the umbilicus. Data were recorded to the nearest millimetre.
Assessment of subcutaneous fat layer using calliper (skinfold thickness measurements) All measurements were performed with the subject in standing position. The measurement points were selected as follows: above the iliac crest in the mid-axillary line, right and left, paraumbilically at 1/3 distance between the iliac crest and umbilicus, right and left, and 5 cm below umbilicus. The skinfold was pinched up firmly between the thumb and forefinger and pulled away from the underlying tissues. The measurements were performed with calibrated calliper of Harpenden type, ie. with a constant measuring pressure 10p/mm 2 , in accordance with established guidelines 28 . The results are presented in mm, as average of five subsequent measurements per one point.
A standard blood sampling has been performed in the morning by venipuncture after an overnight 10 hours fasting. After clotting, the serum was separated and immediately explored for most analyses. For fat hormones, the serum was stored at -20°C until analysed. Insulin resistance was evaluated using the HOMA2 calculations based on fasting glycemia and C-peptide values (Homeostatic model assessment as described by Levy et al. 29 ).

Statistical analysis
Statistical analysis was performed using standard statistical packages (SPSS, Statistical Package for the Social Sciences V10.0, SPSS Inc., Chicago, USA). Normality of data distribution was verified by Kolmogoroff-Smirnoff test. A two-tailed paired Student's t-test was applied to estimate differences between groups in case of normal data distribution. Relations among variables were assessed using Pearson's correlation analysis. Data are presented as means ± SD, unless indicated otherwise. The significance level was set a priori at p<0.05.

Treatment intensity
During all four sessions, the average skin temperature values before, at 10, 20 and 30 minutes of treatment were 31.8±1.1, 39.8±0.7, 39.6±0.6 and 39.2±1.0°C respectively, while the delivered total average maximum energy was 158.5±13.0 W and the total average effective energy was 156.2±13.1 W. While starting the treatment session at 160 W energy level as suggested by manufacturer, in 25 out of 84 (29,8%) sessions the energy intensity could be increased -in accordance with subject's heat sensation --to 170-200 W within the first 10 minutes of treatment, and in 13 out of 84 (15.5%) sessions the energy intensity had to be reduced to 100-150 W due to excess heat perception. The average effective emitted energy in each of four treatment session was therefore 156±14, 160±17, 160±19, and 153±16 W respectively. A significant correlation between the averaged skin temperature after 30 minutes of treatment and reduction in abdominal circumference was observed 1 month after the last treatment (r=-0.49, p=0.03).

Vital signs
When compared with initial values, the average casual blood pressure was significantly lower after the 4 th treatment session (134±12 vs. 127±10 mmHg, p=0.003) and raised to 129±9 mmHg (p=0.04 vs. initial value) 1 month after the treatment. The average heart rate has changed from 69±12 to 67±9/min (p=0.04) after 4 th treatment session, and raised to 69±11/min (p=0.33, both p vs. initial value) after 1 month. No other significant changes have been observed in ECG.

General effects
The radiofrequency selective-field treatment lead to a significant reduction of abdominal circumference as measured at 3 different locations after the 4 th session (

Autonomic balance
Regarding the immediate effects of the treatment on autonomic balance, a significant increase in low frequency (p=0.02) and reduction in high frequency (p=0.05) band cumulative spectral powers have been observed in HRV 20±14 minutes after the treatment.
No sustained effects on autonomic balance, however, have been observed during the follow-up period after the treatment. Figure 1 and Figure 2 summarize the impact of the treatment on autonomic balance immediately after one treatment and 39±18 days (sustained effect) after the last treatment, respectively.
At follow-up after 1 month, there was a significant correlation between the reduction of abdominal circumference and the initial very-low frequency band cumulative spectral power (r=-0.58, p=0.007, Figure 3). Moreover, in a subgroup comparison, subjects   with a higher initial cumulative VLF power (6.4±0.4 LN ms 2 ) demonstrated a significantly bigger drop in abdominal circumference after the 4 th treatment (4.1±1.9 vs. 2.6±0.9 cm, p=0.045) than those with lower initial cumulative VLF spectral power (5,1±0.5 LN ms 2 ).
Interrelationships between treatment effects and metabolic parameters As expected, a significant correlation between weight and insulin resistance index based on HOMA2 calculations has been observed before the treatment (p=-0.53, p=0.016). Change of body weight correlated significantly with the initial HOMA2 indices after the 4 th treatment (r=-0.54, p=0.014 for HOMA2, and r=-0.47, p=0.036 for % beta function) and 1 month after the last treatment (p=-0.57, p=0.009 for % beta function). There was a significant correlation between the initial adiponectin values and deltas of total body fat percentage (r=-0.45, p=0.05) and body weight (r=-0.52, p=0.02) observed after the 4 th treatment.

Metabolic and autonomic interrelationships
The immediate autonomic response to one treatment (Figure 4), as observed in the VLF band (r=0.60, p=0.005) and in the total spectral power (r=0.45, p=0.04) correlated significantly with the initial adiponectin values. Furthermore, a subgroup analysis based on initial adiponectin values (cut-off value 13.0 ng/ml) revealed a significantly stronger acute autonomic response to one treatment in those with higher initial adiponectin level (15.8±1.8 ng/ml) than with a lower one (10.3±1.8 ng/ml). Similarly, in respect to sustained effects, there was a significant correlation between delta of

Side effects and drop-outs
Overall, two drop-outs have been recorded. In two cases the local skin irritation led to interruption of the protocol after the 2 nd treatment. These subjects were not included in the analysis. One subject underwent elective surgery at 1 month of follow-up and did not attend the planned visit. Four subjects did not attend the last followup visit after 3 months.
After first two treatments, one subject reported abdominal discomfort, and another one a hyperesthesia around umbilicus. These symptoms resolved within 1 week after the treatment session. No more adverse reactions have been observed after exchange of spacer used for proper positioning of the energy emitting panel at the 3 rd treatment session. The heart rate variability data are provided.

Discussion
The principal findings of our study are threefold: the efficacy of selective-field radiofrequency treatment in terms of reduction of waist circumference up to 3 months after the treatment series was confirmed. The treatment is safe, as no clinically relevant side-effects were observed. The impact on autonomic cardiovascular balance is significant but transient, while being limited to an increased sympathetic response immediately after this energybased treatment in abdominal area. No sustained effect of the intervention on autonomic balance has been observed 1 month after the last treatment. The treatment efficacy is inversely associated with insulin resistance and other features of metabolic syndrome and may be explained by the inhibitory effect of higher insulin levels on the (treatment-induced) lipolysis 32 . Whether the treatment efficacy could be better predictable using assessment of VLF spectral power and insulin resistance profile, and/or it could be further supported e.g. by pharmacological agents (such as metformin or insulin sensitisers 33 ) or other means, this should be investigated in further prospective trials.
There is a pathogenic link among autonomic imbalance, insulin resistance and obesity. In addition to genetic background and lifestyle factors, autonomic imbalance could be a common root of obesity, hypertension and/or type 2 diabetes mellitus. At early stages of obesity/metabolic syndrome development, parasympathetic control is decreased while sympathetic overactivity usually occurs 34 . This dysfunction increases cardiovascular workload, hemodynamic stress and induces potentially significant cardiac pathology leading to serious arrhythmias. It remains open, however, whether elevated sympathetic tone is a primary feature that contributes to the development of obesity and metabolic syndrome or if it develops and/or changes secondary to the obese state.
It has been shown that sympathetic overactivity precedes the development of insulin resistance and type 2 diabetes mellitus 35 . Studies with genetically predisposed humans with insulin resistance have shown that early insulin resistance is already associated with increased sympathetic control, and it has been suggested that hyperinsulinemia is the initiating factor leading to increase of sympathetic neural activity 36 . Subsequently, adrenoceptor down-regulation and/or reduced sensitivity are likely to develop which situation results in a secondary reduction of sympathetic responsiveness. As adrenoceptors are involved in control of energy expenditure, their down-regulation leads further to impaired food-induced thermogenesis and post-prandial fat oxidation, promoting the accumulation of body fat. In this way, the development of obesity can be seen as a consequence of inappropriate/insufficient sympathetic control, energy dissipation, gaining weight and then insulin resistance 37 . This theory also confirms the earlier popular Bray's MONALISA hypothesis, stating that "Most Obesities kNown Are Low In Sympathetic Activity" 38 . It is also consistent with findings from population studies, e.g. in observation of 7000 individuals without hypertension at baseline, low heart rate variability predicted greater risk of incident hypertension over 9 years of follow-up 39 . Similarly, in almost 2000 participants of Framingham Offspring Study, LF power and LF/HF ratio were lower in diabetic subjects than in those with normal fasting glucose. HRV was inversely associated with plasma glucose levels and was reduced in diabetic individuals as well as in subjects with impaired fasting glucose levels 40 .
Heart rate variability measurement is an established tool for the assessment of impact of intervention on autonomic balance 15-17, 41,42 . While the HRV LF and HF frequency bands have been sufficiently studied in short-and long-term ECG recordings, interpretation of the VLF band --particularly in short-term recordings -is less explored. Along with influences coming from sympatho-thermoregulation, renin-angiotensin system and chemoreceptors, a clear VLF response to excessive temperature variations has been demonstrated 19 . Further on, significant impacts of a spicy food 20 or capsaicin 21 on VLF spectral power have been reported. These findings are consistent with our results where a significant correlation has been observed between the initial adiponectin level and the immediate VLF band autonomic response to a single treatment, as well as between the initial adiponectin and reduction of percentage of body fat after the treatment series ( Figure 4). Additionally to fat percentage, the initial VLF spectral power significantly correlated with change in waist circumference seen after the treatment series ( Figure 3). These observations raise a question whether individuals with higher VLF spectral power and higher adiponectinemia/lower insulin resistance might enjoy a better sympatho-thermogenic capability to "burn" the available energy while more readily inducing lipolysis processes, than those individuals with lower VLF tone. This hypothesis might have clinical implications in weight management programs and/or body contouring treatments for subcutaneous fat layers reduction.
At present, there are some new therapeutic targets and procedures taking into account autonomic imbalance in obesity as an independent and sensitive marker of health. Autonomic dysfunction is reversible with lifestyle changes such as hypocaloric nutrition or fasting 15 and physical endurance training 16 . Recently, the approach of cold-induced, facultative thermogenesis aiming for sympathoadrenergically-mediated weight reduction by stimulation of brown fat tissue has been introduced 43,44 . A sympathetic stimulation of brown fat tissue leading to increased daily energy expenditure by 200-400 kcal 45 has been suggested as the main mechanism in this successful model, and there is some hope for success of this approach on the reduction of obesity 46 . The conversion of white adipose tissue to the highly thermogenic beige adipose tissue has been shown to be influenced by acute sympathetic activation, as well 47 .
Since the autonomic imbalance is a marker of adverse risk 14,18 , its improvement resulting from weight loss should be beneficial for the health of obese/diabetic individuals.
Our relatively short and limited observation could not deliver sufficient evidence on whether subcutaneous abdominal fat reduction using selective-field RF treatment improves obesity-related cardiovascular risk. However, despite even only little changes in body weight, patients with significantly reduced waist circumference are reported to have an improved metabolic profile 48 . It has been shown that the waist circumference is a proven marker of higher total mortality risk 49 as well as of a cardiovascular risk 50 . Therefore, reducing waist circumference and percentage of fat in body composition may represent a useful and clinically relevant target 48 . Moreover, the initial successful waist reduction may play a significant role in further stimulating the motivation of patients with metabolic syndrome for long-term adaptation and adherence to "healthier" lifestyle habits. As this is a proof-of-concept uncontrolled retrospective analysis, we cannot fully exclude external confounding factors that might have contributed to our observations. As a logical step, a randomized, controlled trial verifying the results in an appropriate patient sample would contribute to a better understanding of our findings.
In conclusion, our analysis provided a proof of concept for safety and efficacy of selective field RF treatment using the standard 4 × 30 minutes protocol for moderate reduction of subcutaneous fat tissue. Only transient and non-sustained effects on autonomic balance have been found during the follow-up after the treatment series. The efficacy of Vanquish RF treatment in terms of waist circumference reduction was shown and it was significantly related to initial VLF spectral power and adiponectin levels. This implicates that less insulin resistance may offer better conditions for lipolytic action of the treatment. This body contouring procedure was the most efficient in moderate abdominal overweight with lower insulin resistance, and as such can well complement other, systemic clinical measures for weight reduction based on lifestyle and nutritional changes. As the measurements of HRV and fat hormone status are easily performed, we suggest considering the inclusion of these parameters into the clinical prescreening armamentarium in order to enhance the outcomes of aesthetic body contouring methods.

Consent
Written informed consent for publication of their anonymised clinical details was obtained from all patients.

Data availability
F1000Research: Dataset 1. Data of non-contact radiofrequencyinduced reduction of abdominal fat HRV, 10.5256/f1000research.5708. d38309 51 Author contributions JP and KH conceived the analysis. JP and KH designed the treatment protocol following the standard recommendation of the manufacturer. JP, ZK and KH carried out the research.
ES contributed to the analysis and provided expertise in cardiology. JP and KH prepared the first draft of the manuscript. ES contributed to the preparation of the manuscript. All authors were involved in the revision of the draft manuscript and have agreed to the final content.