Very low energy ketogenic therapy: an anti-inflammatory medical nutritional approach for MASLD in obesity Authors: Luigi Barrea, Ludovica Verde, Martina Galasso, Silvia Savastano, Annamaria Colao, Giovanna Muscogiuri Published: 2025 ABSTRACT AbstractBackgroundChronic inflammation, oxidative stress, and high CRP are linked to metabolic dysfunction-associated steatotic liver disease (MASLD); Very low energy ketogenic therapy (VLEKT) shows anti-inflammatory effects in obesity. The primary aim was to evaluate the effects of ketogenic phase of VLEKT on MASLD and the secondary aim was to identify the underlying mechanisms in women with obesity.MethodsThis prospective outpatient study included 80 women with obesity (mean age 39.25 ± 4.89 years; BMI 37.27 ± 4.58 kg/m2) and MASLD (FLI ≥ 60) who followed ketogenic phase of VLEKT (< 800 kcal/day, with macronutrient composition comprising 13% carbohydrates, 43% proteins and 44% fats at the University Hospital Federico II of Naples. Assessments at baseline and after ketogenic phase of VLEKT included anthropometric measurements, body composition (via bioelectrical impedance analysis), biochemical analyses, high-sensitivity C-reactive protein (hs-CRP) levels and the Fatty Liver Index (FLI).ResultsAfter ketogenic phase, significant reductions were observed in BMI, waist circumference, fat mass, total and LDL cholesterol, triglycerides, gamma-glutamyl transferase (γGT), aspartate aminotransferase (AST), alanine aminotransferase (ALT), fasting plasma glucose and Hs-CRP levels (p < 0.001, for all). Weight loss (%Δ weight) showed significant positive correlations with %Δ fat mass (p < 0.001), %Δ total cholesterol (p < 0.001), %Δ LDL cholesterol (p < 0.001), and %Δ γGT (p < 0.001), while %Δ HDL cholesterol showed a significant negative correlation (p < 0.001). Multiple regression analysis identified %Δ hs-CRP as the main predictor of MASLD improvement, explaining 65.9% of the variance (p < 0.001).ConclusionsMASLD improved post–ketogenic-phase of VLEKT in women with obesity, primarily via reduced chronic inflammation.Keywords: Obesity, Very low-energy ketogenic therapy, VLEKT, MASLD, Phase angle, High sensitivity C reactive protein levels Introduction Metabolic dysfunction-associated steatotic liver disease (MASLD) is recognized as one of the most common chronic liver conditions globally [1]. Its increasing prevalence is closely associated with the rising rates of obesity, metabolic syndrome, and type 2 diabetes mellitus, positioning it as a significant public health concern [2]. Characterized by the excessive accumulation of fat in the liver in the absence of significant alcohol consumption, MASLD encompasses a spectrum of liver damage ranging from simple steatosis to non-alcoholic steatohepatitis (NASH), fibrosis, and eventually cirrhosis [3]. The multifactorial nature of MASLD pathogenesis involves a complex interplay of oxidative stress, inflammation, and metabolic dysfunction, which together drive disease progression [4]. The cornerstone of MASLD management in obesity currently involves weight loss achieved through lifestyle modifications, including dietary interventions and performing regular physical activity. Among dietary approaches, Very low-calorie ketogenic diet (VLCKD), recently named by a panel of experts from the scientific societies Italian Association of Dietetics and Clinical Nutrition (ADI) and Italian Society of Nutraceuticals (SINut) as very low-energy ketogenic therapy, with the acronym VLEKT [5], has garnered significant attention for its ability to induce rapid weight loss and improve metabolic health parameters [6]. VLEKT, a highly restrictive form of ketogenic diet, promotes a metabolic shift from glucose utilization to fat oxidation, leading to reduced hepatic fat deposition[7]. This protocol is divided into distinct phases. The initial ketogenic phases involve a very low-calorie diet (650–800 kcal/day), with minimal carbohydrates (< 30 g/day) and fat, ensuring adequate protein intake to preserve lean mass. Clinical monitoring is essential during these stages. Subsequent non-ketogenic phases gradually reintroduce food groups, starting with low-glycemic-index carbohydrates, such as fruits and vegetables, followed by low-fat dairy and, eventually, legumes and whole grains. Caloric intake is progressively increased across phases to support weight maintenance [7]. Emerging evidence suggests that VLEKT is also effective in mitigating oxidative stress and systemic inflammation, key processes implicated in MASLD pathogenesis [8]. In addition, VLEKT has been demonstrated to improve MASLD outcomes by the end of the nutritional protocol [9]. However, no specific information is available regarding its effects at the end of the ketogenic phases; in addition, there is scarce evidence on the underlying mechanisms through which VLEKT might exert a beneficial effect on MASLD. This study aimed to investigate the effects of VLEKT on MASLD at the end of the ketogenic phase and to explore potential mechanisms underlying its action. Materials and methods Design and setting This prospective outpatient study was carried out between December 2023 and October 2024 at the Endocrinology and Obesity Unit (Centro Italiano per la Cura e il Benessere del Paziente con Obesità, “C.I.B.O.”), recognized as a Collaborating Centre for Obesity Management by the European Association for the Study of Obesity (EASO-COMs). The research was conducted within the Clinical Medicine and Surgery Department of the University of Naples Federico II, located in Naples, Italy. The study received ethical approval from the Federico II Ethics Committee (protocol number 114/2023). Prior to participation, all women provided informed consent after being fully informed about the study’s aims and methodology. Study population A total of 80 women with obesity, aged between 30 and 48 years, were consecutively enrolled. Inclusion criteria included a body mass index (BMI) ≥ 30.0 kg/m² and MASLD. All women had a history of unsuccessful dietary attempts and expressed a desire to lose excess weight. Exclusion criteria were based on contraindications to VLEKT reported in guidelines from the Obesity Management Task Force (OMTF) of the EASO [10] and included the following: type 1 diabetes, latent autoimmune diabetes in adults, β-cell failure in type 2 diabetes, use of sodium/glucose cotransporter 2 (SGLT2) inhibitors, pregnancy and breastfeeding, severe renal impairment, hepatic failure, heart failure (NYHA III-IV), respiratory failure, unstable angina, recent myocardial infarction or stroke (< 12 months), cardiac arrhythmias, eating disorders or serious mental illnesses, alcohol or substance abuse, active infections, frail elderly patients, perioperative periods, and rare metabolic disorders (e.g., carnitine deficiency, mitochondrial fatty acid β-oxidation disorders). Study protocol At baseline (T0), every woman underwent a comprehensive medical and nutritional assessment to confirm that they met the inclusion and exclusion criteria. An endocrinologist conducted a comprehensive clinical history evaluation and excluded any VLEKT contraindications in accordance with EASO guidelines [10]. All women subsequently enrolled underwent anthropometric and lifestyle assessments, body composition evaluations, biochemical analyses, and guidance on the VLEKT protocol. A follow-up assessment (T1) was scheduled at the end of ketogenic phase, during which women underwent anthropometric assessments, body composition evaluations, and biochemical analyses. During the protocol each week, an experienced nutritionist conducted telephone check-in to monitor adherence to the VLEKT protocols. During these calls, β-hydroxybutyrate levels were measured from capillary blood using test strips (Optium Xceed Blood Glucose and Ketone Monitoring System; Abbott Laboratories, Chicago, IL, USA). Ketosis status was recorded as either present or absent (yes/no). VLEKT protocol According to the last consensus statement from the working group of the Club of the Italian Society of Endocrinology (SIE), women meeting eligibility criteria underwent a VLEKT protocol using a commercial meal replacement program [7]. Ketogenic phase of VLEKT was characterized by an energy intake of < 800 kcal/day, with macronutrient composition comprising 13% carbohydrates (< 30 g/day), ~ 43% proteins (1.3 g/kg ideal body weight), and ~ 44% fats. High biological value protein sources (whey, soy, eggs, and peas) were used for meal replacements. Supplementation included B-complex vitamins, vitamins C and E, potassium, sodium, magnesium, calcium, and omega-3 fatty acids to maintain physiological balance. Ketogenic phase of VLEKT lasted 45 (± 2) days, after which participants underwent a follow-up assessment (T1). Baseline assessments (T0) were conducted immediately before the start of the ketogenic phase, ensuring that all outcomes could be directly attributed to this intervention period. Lifestyle assessment Physical activity assessment was performed at T0 and T1. Women who consistently exercised for at least 30 min per day (yes/no) were categorized as physically active, as detailed in our previous studies [11, 12]. One of the exclusion criteria was a medical history of alcohol abuse or a suspicion of it. The threshold was determined at 20 g/day for women [13, 14]. Anthropometric measurements To maintain consistency, the same nutritionist evaluated anthropometric measurements at T0 and T1. Following an overnight fast, women wore light clothing and no shoes for early morning measurements. A wall-mounted stadiometer with an accuracy of 0.5 cm was used to measure height, and a calibrated beam scale (Seca 711; Seca, Hamburg, Germany) was used to measure weight to within 0.1 kg. Weight (kg) divided by height squared (m2) was used to compute BMI, and subjects were categorized into overweight (25.0–29.9 kg/m2) obesity grade I (30.0–34.9 kg/m2),obesity grade II (35.0–39.9 kg/m2), and obesity grade III (≥ 40.0 kg/m2) in accordance with WHO criteria [15]. Using a non-elastic tape that was accurate to 0.1 cm, the waist circumference (WC) was measured at the midpoint between the lowest rib and the iliac crest, or at the umbilicus if the narrowest point could not be reached. Bioelectrical impedance analysis At T0 and T1, body composition was assessed by the same experienced nutritionist using a phase-sensitive bioelectrical impedance analysis (BIA) device (BIA 101 RJL, Akern Bioresearch, Florence) under standardized conditions to minimize inter-observer and inter-device variability. Following ESPEN guidelines [16], assessments were performed after a 6-hour fast, with participants instructed to abstain from physical activity and alcohol for 24 h prior and to empty the bladder 30 min before measurement [17]. Women were measured barefoot and positioned supine, with electrodes applied to the right hand and foot according to standard procedures [18]. The device was calibrated daily, and under 50 kHz conditions, the phase angle (PhA) was calculated as: [19]. This analysis allowed for the evaluation of the following parameters: fat mass (FM, kg and %), fat-free mass (FFM, kg and %), total body water (TBW, liters and %), intracellular water (ICW, liters and %), and extracellular water (ECW, liters and %). Metabolic dysfunction-associated steatotic liver disease (MASLD) At T0 and T1 the presence of MASLD was determined, as previously reported [20], using the ‘fatty liver index’ (FLI), which is known to correlate with both abdominal ultrasound and Fibroscan assessments of MASLD [21, 22]. Based on Bedogni’s validated formula [23], FLI value of ≥ 60 was used as the threshold to indicate the presence of MASLD. The formula for FLI was calculated as follows: Blood pressure assessments At T0 and T1 all women had their systolic blood pressure (SBP) and diastolic blood pressure (DBP) measured three times using a random zero sphygmomanometer (Gelman Hawksley Ltd., Sussex, UK) following ten minutes of sitting at rest. The average of the final two readings was then reported. Laboratory assessments Laboratory evaluations were performed at T0 and T1. Blood samples were collected via venipuncture after an overnight fast, between 8:00 and 10:00 a.m., and processed by standard protocols. Measurements included total cholesterol (mg/dL), LDL cholesterol (mg/dL), HDL cholesterol (mg/dL), triglycerides, AST (U/L), ALT (U/L), γGT (U/L), and fasting plasma glucose (mg/dL). High-sensitivity C reactive protein (hs-CRP) level measurement At T0 and T1 venous blood samples were collected in the morning after an overnight fast of at least 8 h, between 9:00 and 11:00. High-sensitivity C-reactive protein (hs-CRP) levels were measured using a high-sensitivity nephelometric analysis (CardioPhase hsCRP kit, Siemens Healthcare Diagnostics, Marburg, Germany). The assay had a lower detection limit of 0.01 mg/L, with intra- and inter-assay coefficients of variability below 7%. Women were categorized into three cardiovascular risk (CVR) groups based on hs-CRP levels, in accordance with guidelines from the Centers for Disease Control and Prevention and the American Heart Association: low CVR (< 1.0 mg/L), intermediate CVR (1.0–3.0 mg/L), and high CVR (≥ 3.0 mg/L) [24]. Power analysis The sample power was determined by comparing the means ± standard deviation (SD) of the FLI before and after ketogenic phase of VLEKT (85.64 ± 16.25 and 71.61 ± 24.22, respectively). The required minimum sample size was 75 individuals. The calculated power was 95%, with a type I (α) error of 0.05 (95% confidence) and a type II (β) error of 0.05. These sample size and power calculations were performed using the Clinical Calc sample size calculator (https://clincalc.com/stats/samplesize.aspx, accessed on 10 November 2022). Statistical analysis All women completed ketogenic phase of VLEKT protocol and underwent to baseline and follow – up assessments. Data were analyzed using SPSS. Continuous variables were presented as mean ± standard deviation (SD), while categorical variables were expressed as number and percentage (n, %). Normality of data distribution was assessed using the Kolmogorov–Smirnov test. Logarithmic transformation was applied to skewed variables. Paired Student’s t-tests were used for continuous variables to compare T0 and T1 parameters, while the chi-square (χ2) test was employed for categorical variables. The correlations between study variables were performed using Pearson r correlation coefficients. Linear regression analysis was performed with ∆FLI as the dependent variable to evaluate the predictive value of baseline characteristics and changes in relevant parameters. A p-value < 0.05 was considered statistically significant. Results Eighty women were included in the analysis, of whom 38.3% reported engaging in physical activity, while 61.7% were sedentary at baseline (T0). None of the women changed their physical activity levels at the end of VLEKT dietary intervention. Table 1 reports anthropometrical measurements, body composition parameters, blood pressure, biochemical parameters, FLI, and hs-CRP levels for the entire population at T0 and after ketogenic phase of VLEKT (T1). Table 1Anthropometrical measurements, body composition parameters, blood pressure, biochemical parameters, FLI and hs-CRP levels for the entire population at baseline (T0) and after ketogenic phase of VLEKTParametersBaseline(N = 80)After ketogenic phase of VLEKT (N = 80)p-value Anthropometrical measurements Weight (kg)99.65 ± 14.4392.53 ± 13.71 < 0.001 BMI (kg/m2)37.27 ± 4.5834.61 ± 4.30 < 0.001 Overweight (n, %)-14, 17.3%χ2 = 13.23, < 0.001Obesity I class (n, %)28, 35.0%34, 42.5%χ2 = 0.66, p = 0.417Obesity II class (n, %)29, 36.3%22, 27.5%χ2 = 1.04, p = 0.309Obesity III class (n, %)23, 28.8%10, 12.5%χ2 = 5.50, p = 0.019 WC (cm) 109.04 ± 13.78102.43 ± 12.65 < 0.001 < 88 cm (n, %)3, 3.8%10, 12.5%χ2 = 3.01, p = 0.083> 88 cm (n, %)77, 96.3%70, 86.4% Body composition R (Ω)466.12 ± 77.53474.25 ± 72.110.092Xc (Ω)45.76 ± 8.7749.96 ± 9.86 < 0.001 TBW (lt)41.63 ± 5.5040.96 ± 5.51 < 0.001 ECW (lt)19.86 ± 2.9918.58 ± 2.79 < 0.001 FM (kg)44.22 ± 11.8337.91 ± 10.51 < 0.001 FFM (kg)55.42 ± 6.1854.61 ± 6.52 0.012 PhA (°)5.63 ± 0.806.02 ± 0.81 < 0.001 Blood pressure SBP (mmHg)140.68 ± 9.40122.56 ± 9.44 < 0.001 DBP (mmHg)89.62 ± 6.8779.81 ± 6.43 < 0.001 Biochemical parameters Total cholesterol (mg/dL)232.15 ± 21.39201.52 ± 18.56 < 0.001 HDL cholesterol (mg/dL)46.36 ± 7.3750.78 ± 6.51 < 0.001 LDL cholesterol (mg/dL)154.11 ± 21.13124.91 ± 17.80 < 0.001 TG (mg/dL)158.37 ± 52.28129.13 ± 48.29 < 0.001 AST (U/L)31.13 ± 13.3124.93 ± 10.15 < 0.001 ALT (U/L)36.58 ± 19.7727.91 ± 11.98 < 0.001 γGT (U/L)46.61 ± 19.1030.72 ± 14.31 < 0.001 Fasting plasma glucose (mg/dL)102.51 ± 10.6186.50 ± 9.89 < 0.001 FLI 85.64 ± 16.2571.61 ± 24.22 < 0.001 < 60 (n, %)11, 13.8%26, 32.5%χ2 = 6.89, p = 0.009≥ 60 (n, %)69, 86.3%54, 67.5% Inflammatory parameter hs-CRP (mg/L)3.72 ± 2.662.23 ± 1.99 < 0.001 Low risk (n, %)10, 12.5%19, 23.8%χ2 = 2.70, p = 0.101Intermediate risk (n, %)30, 37.5%41, 51.3%χ2 = 2.53, p = 0.112High risk (n, %)40, 50.0%20, 25.0%χ2 = 9.63, p = 0.002A p-value in bold type denotes a significant difference (p < 0.05). Data are expressed as mean ± standard deviation or n (%). one-way ANOVA for continuous variables and χ2 test for categorical variablesBMI, body mass index; WC, waist circumference; R, resistance; Xc, reactance; TBW, total body water; ; ECW, Extracellular water total cholesterol; FM, fat mass; FFM, Fatt-free mass; PhA, phase angle; SBP, systolic blood pressure; DBP, diastolic blood pressure; HDL cholesterol, high-density lipoprotein cholesterol; LDL cholesterol, low-density lipoprotein cholesterol; TG, triglycerides; AST, aspartate aminotransferase; ALT, alanine aminotransferase; γGT, gamma-glutamyl transferase; FLI, fatty liver index; hs-CRP, high-sensitivity C-reactive protein; VLEKT, Very low energy ketogenic therapy Anthropometrical measurements, body composition parameters, blood pressure, biochemical parameters, FLI and hs-CRP levels for the entire population at baseline (T0) and after ketogenic phase of VLEKT A p-value in bold type denotes a significant difference (p < 0.05). Data are expressed as mean ± standard deviation or n (%). one-way ANOVA for continuous variables and χ2 test for categorical variables BMI, body mass index; WC, waist circumference; R, resistance; Xc, reactance; TBW, total body water; ; ECW, Extracellular water total cholesterol; FM, fat mass; FFM, Fatt-free mass; PhA, phase angle; SBP, systolic blood pressure; DBP, diastolic blood pressure; HDL cholesterol, high-density lipoprotein cholesterol; LDL cholesterol, low-density lipoprotein cholesterol; TG, triglycerides; AST, aspartate aminotransferase; ALT, alanine aminotransferase; γGT, gamma-glutamyl transferase; FLI, fatty liver index; hs-CRP, high-sensitivity C-reactive protein; VLEKT, Very low energy ketogenic therapy Grade I obesity was detected in 28 women (35.0%), grade II obesity in 29 women (36.3%), and grade III obesity in 23 women (28.8%) at baseline. Significant reductions were observed in weight (99.65 ± 14.43 vs. 92.53 ± 13.71 kg; p < 0.001) and BMI (37.27 ± 4.58 vs. 34.61 ± 4.39 kg/m²; p < 0.001) from T0 to T1. As expected, a shift in obesity classifications was noted, with a decline in the prevalence of women belonging to higher obesity classes (23, 28.8% vs. 10, 12.5%; p < 0.001). WC also significantly decreased (109.04 ± 13.78 vs. 102.43 ± 12.65 cm; p < 0.001) from T0 to T1, with a trend toward an increase in the proportion of women with WC < 88 cm (3.8% vs. 12.5%; p = 0.083). A significant reduction in FM (44.22 ± 11.83 vs. 37.91 ± 10.51 kg; p < 0.001) and an improvement in PhA (5.63 ± 0.80 vs. 6.02 ± 0.80°; p < 0.001) were observed at T1 compared to T0. Similarly, significant reductions were found in total cholesterol (232.15 ± 21.39 vs. 201.52 ± 18.56 mg/dL; p < 0.001), LDL cholesterol (154.11 ± 21.13 vs. 124.91 ± 17.80 mg/dL; p < 0.001), triglycerides (158.37 ± 52.28 vs. 129.13 ± 48.29 mg/dL; p < 0.001), and markers of liver function, including AST (31.13 ± 13.31 vs. 24.93 ± 10.15 U/L; p < 0.001), ALT (27.91 ± 11.98 vs. 23.56 ± 9.17 U/L; p < 0.001), and γGT (46.61 ± 19.10 vs. 30.72 ± 14.31 U/L; p < 0.001). Fasting plasma glucose also decreased significantly (102.51 ± 10.61 vs. 86.50 ± 9.89 mg/dL; p < 0.001), coupled with an increase in HDL cholesterol (46.36 ± 7.37 vs. 50.78 ± 6.51 mg/dL; p < 0.001) from T0 to T1. Notably, both SBP and DBP showed considerable improvements at T1 compared to T0 (140.68 ± 9.40 vs. 122.56 ± 9.44 mmHg and 89.62 ± 6.87 vs. 79.81 ± 6.43 mmHg, respectively; p < 0.001). The FLI substantially improved with a significant decrease from T0 to T1 (85.64 ± 16.25 vs. 71.61 ± 24.22; p < 0.001), with an increase in the number of women with normalized FLI (< 60 FLI; 11, 13.6% vs. 26, 32.1%; p = 0.009). Also, hs-CRP levels showed a significant reduction (3.72 ± 2.66 vs. 2.23 ± 1.99; p < 0.001) from T0 to T1. Consequently, the proportion of women at high inflammatory risk (hs-CRP ≥ 3.0 mg/L) dropped from 50.0% (n = 40) at T0 to 25.0% (n = 20) at T1. Table 2 shows the correlations of ∆% weight with ∆% of anthropometrical measurements, body composition parameters, blood pressure, biochemical parameters, FLI, and hs-CRP levels. The ∆% weight exhibited significant positive correlations with ∆% Fat Mass (r = 0.620; p < 0.001) and ∆% total cholesterol (r = 0.777; p < 0.001). Furthermore, ∆% weight showed significant positive correlations with ∆% LDL cholesterol (r = 0.721; p < 0.001) and ∆% γGT (r = 0.388; p < 0.001). Notably, ∆% HDL-cholesterol demonstrated a significant negative correlation with ∆% weight (r = -0.473; p < 0.001). In contrast, no significant correlation was found between ∆% weight and ∆% in resistance R or PhA. A weak negative correlation was observed between ∆% weight and ∆% Xc (r = -0.053), but this was not statistically significant (p = 0.641). Table 2Correlations of ∆% weight loss with ∆% of anthropometrical measurements, body composition parameters, blood pressure, biochemical parameters, FLI, and hs-CRP levelsParameters(∆%)∆% weightN = 80rp-value Anthropometrical measurements BMI (kg/m2)1.000 < 0.001 WC (cm)0.471 < 0.001 Body composition R (Ω)0.0000.999Xc (Ω)-0.0590.602FFM (kg)0.0710.530FM (kg)0.620 < 0.001 PhA(°)-0.1350.232TBW (lt)0.0730.522ECW (lt)0.2060.067 Blood pressure SBP (mmHg)0.400 < 0.001 DBP (mmHg)0.1890.093 Biochemical parameters Total cholesterol (mg/dL)0.777 < 0.001 HDL cholesterol (mg/dL)-0.473 < 0.001 LDL cholesterol (mg/dL)0.721 < 0.001 TG (mg/dL)0.269 0.016 γGT (U/L)0.388 < 0.001 Fasting plasma glucose (mg/dL)-0.0880.440 FLI 0.479 < 0.001 Inflammatory parameter hs-CRP (mg/L) 0.481 < 0.001 A p-value in bol type denotes a significant difference (p < 0.05). BMI, body mass index; WC, waist circumference; R, resistance; Xc, reactance; FM, fat mass; FFM, Fat-Free Mass; PhA, phase angle; TBW, total body water; ECW, Extracellular water; SBP, systolic blood pressure; DBP, diastolic blood pressure; HDL cholesterol, high-density lipoprotein cholesterol; LDL cholesterol, low-density lipoprotein cholesterol; TG, triglycerides; γGT, gamma-glutamyl transferase; FLI, fatty liver index; hs-CRP, high-sensitivity C reactive protein Correlations of ∆% weight loss with ∆% of anthropometrical measurements, body composition parameters, blood pressure, biochemical parameters, FLI, and hs-CRP levels A p-value in bol type denotes a significant difference (p < 0.05). BMI, body mass index; WC, waist circumference; R, resistance; Xc, reactance; FM, fat mass; FFM, Fat-Free Mass; PhA, phase angle; TBW, total body water; ECW, Extracellular water; SBP, systolic blood pressure; DBP, diastolic blood pressure; HDL cholesterol, high-density lipoprotein cholesterol; LDL cholesterol, low-density lipoprotein cholesterol; TG, triglycerides; γGT, gamma-glutamyl transferase; FLI, fatty liver index; hs-CRP, high-sensitivity C reactive protein Table 3 displays the linear regression analysis with ∆%FLI as the dependent variable. In Model 1, ∆% hs-CRP was the main predictor, explaining 65.9% of the variance (R² = 0.659). The association between ∆% hs-CRP and ∆% FLI was strong and statistically significant (β = 0.812; p < 0.001). Other variables, including ∆% BMI, ∆% WC, ∆% FM, ∆% total cholesterol, ∆% TG, ∆% LDL cholesterol, and ∆% γGT, were excluded from the model. The overall model fit was highly significant (p < 0.001). Table 3Linear regression analysis with ∆% FLI as dependent variable (outcome) to estimate the predictive value of ∆% BMI, ∆% waist circumference, ∆% fatt mass, ∆% total cholesterol, ∆% TG, ∆% LDL cholesterol, ∆% γGT and ∆% hs-CRPOutcome ∆% FLI R 2 βCI (95%)p-valueOverall p-value∆% hs-CRP0.6590.8120.49–0.68 < 0.001 < 0.001 Variables excluded: ∆% BMI, ∆% Waist circumference, ∆% Fat mass, ∆% total cholesterol, ∆% TG, ∆% LDL-cholesterol, ∆% γGTFLI, fatty liver index; BMI, body mass index; TG, triglycerides; LDL, low density lipoprotein; γGT, gamma-glutamyl transferase; hs-CRP, high-sensitivity C reactive protein Linear regression analysis with ∆% FLI as dependent variable (outcome) to estimate the predictive value of ∆% BMI, ∆% waist circumference, ∆% fatt mass, ∆% total cholesterol, ∆% TG, ∆% LDL cholesterol, ∆% γGT and ∆% hs-CRP Variables excluded: ∆% BMI, ∆% Waist circumference, ∆% Fat mass, ∆% total cholesterol, ∆% TG, ∆% LDL-cholesterol, ∆% γGT FLI, fatty liver index; BMI, body mass index; TG, triglycerides; LDL, low density lipoprotein; γGT, gamma-glutamyl transferase; hs-CRP, high-sensitivity C reactive protein Discussion The main finding of this prospective outpatient study is that ketogenic phase of VLEKT protocol significantly improved MASLD in women with obesity, mainly through a reduction in low-grade inflammation, as reflected by a significant decrease in FLI. Additionally, significant reductions in hs-CRP, BMI, WC, FM, and markers of liver function (AST, ALT, and γGT) were observed, alongside improvements in lipid profiles, including decreased total cholesterol, LDL cholesterol, and triglycerides, and increased HDL cholesterol. These results were consistent with existing literature and further support the efficacy of VLEKT in addressing MASLD [6, 25]. Similarly to our findings, Rinaldi R. et al. reported a reduction in the degree of MASLD, associated with decreases in weight, BMI, WC, total cholesterol, LDL cholesterol, triglycerides, AST, ALT, and γGT after 8 weeks of VLEKT [9]. However, their study did not provide specific data on the changes occurring at the end of ketogenic phase of the protocol [9]. This distinction is critical, as it represents a unique window to evaluate the specific effects of ketosis itself, rather than the broader impact of weight loss alone. Understanding this phase may offer insight into the mechanisms through which ketosis influences MASLD, beyond the simple reduction in adiposity. In our study, the reduction of FLI tightly correlated with the reduction in hs-CRP protein, a key marker of systemic inflammation and a known driver of MASLD progression [26]. The role of low-grade chronic inflammation in the pathogenesis and progression of MASLD is well established [26]. As demonstrated by Tarantino et al. [27], low-grade inflammation exacerbates hepatic steatosis through increased cytokine production, insulin resistance, and oxidative stress [27]. Indeed, ketone bodies have been known to exert anti-inflammatory effects [28], and thus, we speculate that the reduction of FLI in our study could recognize the anti-inflammatory property of ketone bodies as one of the mechanisms involved in VLEKT-related MASLD improvements. From a clinical perspective, the decrease in hs-CRP observed here supports its potential role as a target and biomarker for dietary interventions aimed at reducing hepatic inflammation in obesity-related MASLD. Beyond this pathway, other mechanisms may also contribute to the observed benefits. In particular, improvements in hepatic insulin sensitivity and gut microbiome modulation, both of which are strongly influenced by carbohydrate restriction and dietary ketosis, could synergistically reduce hepatic fat accumulation and systemic inflammation. Additionally, we observed a significant increase in PhA following dietary intervention. PhA, which was not evaluated in the study by Rinaldi R et al. [9], serves as a marker of cellular health and membrane integrity and is known to be lower in patients with MASLD, as has been demonstrated by Guangya Chen et al. [29]. PhA has been shown to be a predictor of inflammation [17]. Specifically, in that study, PhA was found to be a reliable predictor of inflammation in cases of low 25(OH)-D levels, further highlighting its role as a biomarker of both cellular health and inflammatory status [17]. Furthermore, it has been demonstrated that the state of ketosis itself mitigates inflammation by inhibiting NLRP3 inflammasome activation, with β-hydroxybutyrate (the primary ketone body) acting in a manner similar to anti-inflammatory medications but without adverse effects [28]. This improvement further supports the notion that VLEKT may mitigate inflammation, which is a central feature of MASLD. Clinically, the observed increase in PhA may offer physicians a non-invasive and cost-effective tool to track both inflammatory status and overall cellular health during dietary management of MASLD. Finally, the regression analysis confirmed the pivotal role of reducing low-grade inflammation, as assessed by hs-CRP, in improving FLI. In fact, the linear regression model demonstrated that ∆% hs-CRP was the sole predictor of ∆% FLI, explaining 65.9% of the variance (R² = 0.659). These findings emphasize the importance of targeting inflammation reduction as a central mechanism in improving MASLD outcomes during the initial phase of the VLEKT protocol. In addition, we hypothesized that by significantly reducing monosaccharides and overall carbohydrate intake, VLEKT minimizes glucose and insulin spikes, thus improving insulin sensitivity, a key factor in MASLD pathogenesis [30]. VLEKT is known to eliminate the negative impact of excessive fructose consumption, a known contributor to hepatic fat accumulation, by minimizing or eliminating dietary sources of mono-, di-, and polysaccharides [31]. Additionally, it improves gut microbiome health, which plays a critical role in the progression of MASLD [25, 32]. Taken together, these complementary mechanisms, inflammation reduction, enhanced insulin sensitivity, and microbiome modulation, may explain the significant clinical improvements observed, positioning VLEKT as a multifaceted dietary approach to MASLD management. These combined effects position VLEKT as a comprehensive dietary approach to managing MASLD, with benefits that extend beyond weight loss alone. The main limit of this study was the observational design, which limits our ability to infer causal relationships between metabolic changes and liver steatosis improvement. The multivariable analysis was conducted to identify the strongest statistical predictor of changes in FLI rather than to establish causality. While ∆% hs-CRP emerged as the main significant predictor, this finding does not imply a direct causal effect but rather highlights a potential mechanistic link between inflammation and liver steatosis reduction, warranting further investigation. However, a strength of this study was the isolation of ketogenic phase of the VLEKT, which allowed us to specifically evaluate the effects of ketone body production, providing a clear understanding of its impact on MASLD. Conclusions In conclusion, our study reinforces the role of VLEKT as a highly effective dietary intervention for MASLD. The reduction in FLI is mainly driven by decreases in chronic low-grade inflammation. Future studies should aim to elucidate the causal mechanisms underlying the beneficial effects of VLEKT on MASLD. Abbreviations Very Low-Energy Ketogenic Therapy Metabolic dysfunction-associated steatotic liver disease To non-alcoholic steatohepatitis Body Mass Index Waist Circumference Fat Mass Fat-Free Mass Total Body Water Extracellular Water Phase Angle Resistance Reactance Systolic Blood Pressure Diastolic Blood Pressure Aspartate Aminotransferase Alanine Aminotransferase Gamma-Glutamyl Transferase Fatty Liver Index High-sensitivity C-Reactive Protein Cardiovascular Risk Triglycerides High-Density Lipoprotein Low-Density Lipoprotein Very Low-Calorie Ketogenic Diet Italian Association of Dietetics and Clinical Nutrition Italian Society of Nutraceuticals Italian Society of Endocrinology European Association for the Study of Obesity Obesity Management Task Force World Health Organization European Society of Parenteral and Enteral Nutrition Author contributions Study conception and design: GM and LB; Data collection: MG and LV; Analysis and interpretation of results: LB; Draft manuscript preparation: LB, LV, MG, SS, AC and GM; Study supervision: AC and SS; All authors reviewed the results and approved the final version of the manuscript. Funding The authors declare that financial support was received for the research, authorship and/or publication of this article. The research leading to these results has received funding from PRIN n.2022BASPYN. Data availability The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate The study has been approved by the Local Ethical Committee and carried out in accordance with the Code of Ethics of the World Medical Association (Declaration of Helsinki) for experiments that involved humans. The aim of the study was clearly explained to all the study participants and a written informed consent was obtained. Consent for publication Not applicable. Competing interests The authors declare that they have no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Luigi Barrea and Ludovica Verde have contributed equally to this work as co-first authors. References Associated Data Data Availability Statement The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.