Egypt’s metabolic dysfunction associated steatotic liver disease: Genetic and dietary nexus, shaping artificial intelligence driven healthier future

Abstract

BACKGROUND

Metabolically dysfunction-associated steatotic liver disease (MASLD) has emerged as a significant global health concern; however, its burden and determinants in Egypt remain poorly characterized.

AIM

To determine the prevalence of MASLD, identify risk factors, and evaluate genetic influences in the Egyptian population.

METHODS

In this multicenter observational cross-sectional study (2019-2024), 1060 adults from 22 governorates underwent a comprehensive assessment, including anthropometrics, metabolic profiling, dietary evaluation (validated Food Frequency Questionnaire), abdominal ultrasonography, vibration-controlled transient elastography (FibroScan^®), and PNPLA3 rs738409 genotyping. MASLD was diagnosed according to the European Association for the Study of the Liver 2024 guidelines.

RESULTS

An alarmingly high prevalence of MASLD (67.7%) was revealed. Obesity conferring a 27.5-fold increased risk (P < 0.001), insulin resistance was associated with a 3.4-fold higher risk, metabolic syndrome present in 41.2% of cases. Each 1-gram increase in monounsaturated fatty acid and polyunsaturated fatty acid intake raised the risk by 3% and 10%, respectively. High glycemic load increased the risk by 1% per unit. PNPLA3 gene variant present in 48.1% of MASLD cases compared to 43.6% of healthy individuals. Variant carriers exhibited distinct risk profiles related to age and polyunsaturated fatty acid intake. Demographic patterns were notable, as prevalence increased from 52% in individuals < 38 years to 83% in those ≥ 38 years, while lean individuals still demonstrated a prevalence of 22.2%.

CONCLUSION

Egypt faces an epidemic of MASLD reflecting complex interactions among metabolic, dietary and genetic factors. Implementation of national screening programs and culturally tailored lifestyle interventions is urgently needed.

Key Words: Metabolic-dysfunction associated steatotic liver disease; Egypt; Epidemiology; PNPLA3; Metabolic syndrome; Nutrition

Core Tip: Egypt faces the world’s highest reported metabolic dysfunction associated steatotic liver disease prevalence (67.7%), driven by synergistic metabolic, dietary, and genetic factors. Obesity (27.5 times the risk), insulin resistance, high intake of refined carbohydrates, and PNPLA3 variants (present in 48% of cases) create a perfect metabolic storm. Alarmingly, 22% of lean individuals are affected, challenging traditional risk paradigms. These findings necessitate the implementation of urgent nationwide screening programs, culturally adapted lifestyle interventions, and the integration of genetic risk assessment into preventive strategies to combat this growing public health crisis.

INTRODUCTION

Metabolically dysfunction-associated steatotic liver disease (MASLD) is the most prevalent chronic liver disease globally, affecting over 30% of the population as of 2023. Its incidence is rising in parallel with epidemics of type 2 diabetes, obesity, and cardiometabolic dysfunction. MASLD significantly increases the risk of progressive liver disease, including liver failure and hepatocellular carcinoma, as well as extra-hepatic complications like cardiovascular events and chronic kidney disease[1]. While approximately 10%-30% of isolated steatosis cases progress to steatohepatitis, this risk escalates dramatically to 65% in patients with type 2 diabetes, underscoring MASLD as a critical public health challenge[2].

Globally, MASLD prevalence is estimated at 25%-30%, but significant geographic, ethnic, and socioeconomic disparities exist. Middle Eastern and North African populations, in particular, are frequently underrepresented in large epidemiological studies, obscuring the true burden and unique drivers of the disease in these regions[3]. In Egypt, a perfect storm of dietary, lifestyle, and genetic factors has converged to create a high-risk environment. The primary causes of MASLD are deeply rooted in a shift towards physical inactivity and diets high in refined carbohydrates and added sugars, compounded by a high prevalence of obesity, which currently affects approximately 7.09 million males and 12.7 million females[4]. Specifically, common dietary practices that drive MASLD include the high consumption of refined grains (e.g., Egyptian traditional baladi bread), sugar-sweetened beverages, and traditional sweets, which create a persistent state of positive energy balance and de novo lipogenesis.

This metabolic burden is unevenly distributed. Egypt’s unique ethnic diversity encompassing groups such as Abazas, Turks, Greeks, Bedouin Arabs, and Dom clans suggests a potential role for genetic predisposition. Furthermore, poverty is concentrated in Upper Egypt (27.8% poverty, 28.7% at risk), raising critical questions about how MASLD prevalence varies across governorates, socioeconomic classes, and genetic backgrounds. Key genetic variants in genes such as PNPLA3, TM6SF2, and GCKR are known to modulate hepatic fat accumulation and fibrosis risk[5].

This is where genetic testing holds significant preventive potential. By identifying individuals with high-risk genetic profiles, healthcare providers can implement stratified prevention strategies. For a patient carrying a risky PNPLA3 variant, for instance, even a moderate degree of overweight or sugar consumption may warrant earlier and more intensive lifestyle intervention, personalized dietary counseling, and vigilant monitoring. This proactive approach allows for the precise targeting of resources to those who would benefit most, moving beyond a one-size-fits-all public health model.

Therefore, this study aims to elucidate the genetic-dietary nexus of MASLD in Egypt by mapping prevalent dietary practices against genetic risk factors to provide a foundational dataset for developing artificial intelligence (AI)-driven, personalized prevention strategies to mitigate the disease’s escalating burden.

MATERIALS AND METHODS

Study design and participants

This cross-sectional observational multicenter study was conducted across 22 Egyptian governorates from March 2019 to February 2024, stratified into Upper Egypt, Middle Egypt, and Lower Egypt to ensure socioeconomic diversity. The study enrolled asymptomatic adults (18-70 years) with body mass index (BMI) ≥ 18.5 kg/m², excluding alcohol consumers, pregnant women, and individuals with other chronic liver disease (like Wilson’s, autoimmune hepatitis)[6]. Participants included medical personnel, university staff and students, and their relatives from four university hospitals and Ministry of Health facilities.

Data collection

This study was approved by Cairo University’s Ethics Committee (No. N-41-2019) and registered on ClinicalTrials.gov (No. NCT04081571). Regarding data collection, the collected information included anthropometrics [height, weight, waist circumference (WC), and BMI], demographics (age and gender), lifestyle factors assessed using the International Physical Activity Questionnaire[7], and dietary history evaluated with a validated, semi-quantitative Food Frequency Questionnaire (FFQ) covering the previous 12 months[8]. The FFQ was designed to capture culturally specific Egyptian foods, including traditional staples such as baladi bread, falafel, fried fish, feteer meshaltet, koshary, ful medames, and commonly used oils and fats. Nutritional analysis was performed by converting consumed portions into nutrients using a database that integrated standard United States Department of Agriculture values with compositional data for these local food items. Additionally, data on metabolic syndrome (MetS) components were obtained, and the diagnosis of MASLD was determined according to the European Association for the Study of the Liver 2024 guidelines[9]. A single operator performed an abdominal ultrasound to minimize variability[10]. Participants with steatosis underwent vibration-controlled transient elastography (VCTE) on the same day, dietary history evaluated with the FFQ covering the previous 12 months and analyzed using the United States Department of Agriculture database[10].

VCTE using FibroScan^® Mini+ 430 was employed to assess both liver stiffness and steatosis. Liver stiffness measurement was interpreted according to American Association for the Study of Liver Diseases criteria[11], where values < 8 kPa excluded advanced fibrosis, 8-12 kPa indicated possible fibrotic metabolic dysfunction-associated steatohepatitis, and ≥ 12 kPa signified advanced fibrosis. Steatosis was graded using controlled attenuation parameter (CAP) scores[12], with 238-260 dB/m corresponding to S1 (11%-33% fat), 260-290 dB/m to S2 (34%-66% fat), and ≥ 290 dB/m to S3 (≥ 67% fat). The XL probe was used for participants with a BMI ≥ 30 kg/m². Examinations required at least 10 valid measurements with a minimum 60% success rate and interquartile range/mean ≤ 30%[13].

Blood sampling and analysis

Fasting blood samples (10 mL) were processed for multiple analyses. Serum analysis (4 mL) included metabolic markers [cholesterol, high-density lipoprotein (HDL), triglycerides (TGs), glucose, and insulin for Homeostasis Model Assessment of Insulin Resistance (HOMA-IR)][14], liver enzymes (aspartate aminotransferase, alanine aminotransferase, gamma-glutamyl transferase, and bilirubin), inflammatory marker [C-reactive protein (CRP)], infectious markers (hepatitis C virus antibody and hepatitis B surface antigen), and thyroid hormones [thyrotropin (TSH), free thyroxine (FT) 3, and FT4]. Glycated hemoglobin (HbA1c) was measured from 2 mL EDTA blood using the Tosoh HLC-723G8 HPLC system[15]. Enzyme-linked immunosorbent assay assays were performed on 2 mL serum to determine L-carnitine (Sunlong Biotech) and lipoprotein lipase (LPL) (Sunred Biological Technology) levels. Genomic DNA was extracted from 2 mL EDTA blood using the EasyPure® kit, and purity was confirmed with NanoDrop^™ (260/280 ratio = 1.8)[16]. The PNPLA3 rs738409 (I148M) variant was genotyped, a well-established genetic risk factor for MASLD severity and progression[17].

Statistical analysis

To assess the prevalence of MASLD in Egypt, a prospective cross-sectional observational study was conducted by a certified biomedical statistician. Based on previous studies, the expected prevalence was 37%. For a 99.9% confidence interval with a margin of error of ± 5% from the expected proportion, the required sample size was calculated to be 1010. To account for a 5% non-response rate, the final sample size was increased to 1060. Epi Info software was used for sample size estimation[18].

Statistical Package for the Social Sciences v27 was used for analysis. Numerical data were expressed as mean ± SD or median (range), and categorical data as n (%). Normality was tested via the Kolmogorov-Smirnov and Shapiro-Wilk tests. Comparisons were conducted as follows: (1) Categorical variables were analyzed using the χ² test or Fisher’s exact test; and (2) Continuous variables were analyzed using the t-test or Mann-Whitney U test for two groups and analysis of variance or the Kruskal-Wallis test for more than two groups, with Bonferroni post hoc analysis.

Spearman’s correlation coefficients were calculated to assess the associations between non-normally distributed numerical variables (r value ranges from -1 to 1, indicating the strength and direction of the relationship). Variables with P < 0.10 in univariate analysis were entered into a stepwise logistic regression model to evaluate the independent effect of factors on MASLD risk. Logistic regression was employed to estimate adjusted odds ratios and 95%CI to quantify the association between risk factors and MASLD. The 95%CI not including 1.0 was considered statistically significant. All tests were two-tailed, and a P value ≤ 0.05 was considered significant[19].

RESULTS

The study included 1060 participants (545 females, 515 males; mean age 38 ± 11 years), of whom 718 (67.7%) were diagnosed with MASLD. MASLD patients were older (41 ± 11 years vs 32 ± 9 years, P < 0.001) and had higher BMIs (30.2 ± 6.3 kg/m²) and WCs (96 ± 14 cm) compared with healthy individuals. Overall, 78.3% were overweight or obese, including 7.9% with morbid obesity. The geographic distribution of MASLD across Egypt was not significant (P = 0.809). Both BMI and WC were significantly higher among MASLD participants, as illustrated in Figure 1A.

Figure 1 Comparison of correlation indices between patients and healthy controls. A: Prevalence of various body mass index and waist circumference categories among participants diagnosed with metabolically dysfunction-associated steatotic liver disease compared to healthy controls; B: Walking and exercise habits among participants with metabolically dysfunction-associated steatotic liver disease and healthy controls. MASLD: Metabolically dysfunction-associated steatotic liver disease.

Physical activity was predominantly walking (77.6%), while only 9% engaged in regular sports. Healthy participants were more active (14.3% sports, 42.7% walking ≥ 20 minute/day) than MASLD patients (6.8% sports, 28.3% walking, P < 0.001), as illustrated in Figure 1B.

Metabolic abnormalities included prediabetes/diabetes (27.6%), insulin resistance (36.3%), hypertension (8.3%), and thyroid dysfunction (7.8%). The PNPLA3 gene variant was present in 46% of the participants, and 12.3% of the participants were smokers. Various clinical characteristics, laboratory values, and genetic markers are presented in Figure 2.

Figure 2 Prevalence and percentage distribution of various clinical characteristics, laboratory values, and genetic markers among the study participants. HbA1c: Glycated hemoglobin; HDL: High-density lipoprotein; HOMA-IR: Homeostasis Model Assessment of Insulin Resistance; LDL: Low-density lipoproteins; TGs: Triglycerides.

MetS was present in 41.2% of MASLD participants, and 54.9% had one or two cardiometabolic risk factors. MASLD was highly prevalent among individuals with metabolic abnormalities: (1) 44.4% with insulin resistance (HOMA-IR > 2.5); (2) 84.6% of prediabetics; (3) 94.4% of diabetics; (4) 94.3% of hypertensive individuals; (5) 79.9% with low HDL-cholesterol (HDL-C); and (6) 86.6% with high TGs.

Liver ultrasound confirmed steatosis in 67.7% of participants, including four cases of cirrhosis, while 24 had gallstones and 36 had splenomegaly. VCTE revealed a mean liver stiffness of 6.3 ± 2.5 kPa, as depicted in Figure 3A, with substantially more severe steatosis and fibrosis in MetS patients (P < 0.001), as illustrated in Figure 3B.

Figure 3 Vibration-controlled transient elastography. A: Distribution across different categories of FibroScan (F), controlled attenuation parameter (S) dB/m, and kilopascals; B: Mean and SD of controlled attenuation parameter and kilopascals among participants with metabolic syndrome and those without. CAP: Controlled attenuation parameter; KPA: Kilopascals; MASLD: Metabolically dysfunction-associated steatotic liver disease.

Smoking prevalence was similar between MASLD (12.35%) and healthy (12.28%) participants, with no significant difference in steatosis (P = 0.201). However, smokers demonstrated slightly higher liver stiffness (6.7 ± 2.7 kPa vs 6.2 ± 2.5 kPa, P = 0.06).

Nutritional analysis revealed that MASLD participants consumed more calories (2293.2 ± 818.5 kcal/day vs 2099.2 ± 818.5 kcal/day, P < 0.001), along with a higher glycemic index (952.5 ± 451.3 vs 863.1 ± 401.7, P = 0.002) and glycemic load (166.6 ± 73.9 vs 155.4 ± 82, P = 0.033). All dietary components were significantly elevated in MASLD patients, as illustrated in Figure 4.

Figure 4 Dietary energy consumption among participants with metabolically dysfunction-associated steatotic liver disease and healthy controls. MASLD: Metabolically dysfunction-associated steatotic liver disease; MUFA: Monounsaturated fatty acid; PUFA: Polyunsaturated fatty acid.

MASLD is strongly associated with disturbances in lipid and glucose metabolism, as evidenced by elevated TGs, low density-lipoproteins (LDL), very LDL (VLDL), fasting glucose, HbA1c, insulin, HOMA-IR, liver enzymes, and CRP levels. Healthy participants had higher FT4 levels. Analysis of the PNPLA3 rs738409 genetic variant revealed its significant role in the Egyptian MASLD population. The variant was highly prevalent, present in 46% of the entire cohort. While its frequency was not drastically different between MASLD patients (48.1%) and healthy controls (43.6%), its impact was modified by body weight and specific dietary factors, pointing to important gene-environment interactions as exhibited in Table 1.

Table 1 Different laboratory parameters among participants with metabolically dysfunction-associated steatotic liver disease and healthy controls, n (%)/median (range)/mean ± SD.

Metabolically dysfunction-associated steatotic liver disease	Yes	No	P value
Lipoprotein lipase	15 (0.5-185)	18 (0.5-214)	0.062
L-carnitine	12.3 (0.1-63)	13 (0.6-58)	0.099
C-reactive protein	4 (0.3-117)	2 (0.2-39.2)	< 0.001
Aspartate aminotransferase	20 (4.5-236)	18 (9-103)	< 0.001
Alanine aminotransferase	20 (4-258)	14 (3-82)	< 0.001
Gamma-glutamyl transferase	23 (4-595)	15 (4-137)	< 0.001
Bilirubin	0.4 (0.1-2)	0.4 (0-54)	0.066
International normalized ratio	1 (0.9-10)	1.1 (1-1.1)	0.302
Homeostasis Model Assessment of Insulin Resistance	2.2 (0.1-73)	1.4 (0-20.2)	< 0.001
Fasting insulin	9.9 (0.3-279.4)	7 (0.3-45)	< 0.001
Triglycerides	120 (3.9-1086)	74 (28-409)	< 0.001
Very LDL	24 (3.1-217.2)	14.8 (5.8-81.8)	< 0.001
Thyrotropin	1.6 (0.01-100)	1.5 (0.01-9.7)	0.227
Albumin	4.7 ± 2.4	4.7 ± 0.4	0.876
Fasting blood sugar	98.3 ± 33.6	88.9 ± 23.4	< 0.001
Glycated hemoglobin (%)	5.7 ± 1.1	5.2 ± 0.7	< 0.001
Total cholesterol	205.9 ± 44.7	193.1 ± 40.2	< 0.001
High-density lipoprotein	46.4 ± 11.9	52.1 ± 13	< 0.001
LDL	132.1 ± 39	122.8 ± 34.4	< 0.001
FT3	3.3 ± 0.5	3.4 ± 1.5	0.196
FT4	1.3 ± 0.3	1.4 ± 0.5	0.003
PNPLA gene
CC	373 (65.9)	193 (34.1)	0.649
CG	261 (68.1)	122 (31.9)
GG	43 (70.5)	18 (29.5)
PNPLA alle
C alle	1007 (66.5)	508 (33.5)	0.382
G alle	347 (68.7)	158 (31.3)

FT: Free thyroxine; LDL: Low-density lipoproteins.

Key predictors of MASLD include age, BMI, monounsaturated fatty acid (MUFA) intake, HOMA-IR, and VLDL levels. Overweight individuals have a 4.5-fold higher risk, which escalates to 27.5-fold in obesity. Elevated HOMA-IR increased risk by 3.4 times, independent of BMI. Aging raised MASLD risk by 6% annually in normal-weight individuals and 13% in obese individuals. Predictors varied by BMI: (1) Lean individuals were more affected by male gender (4-fold risk); (2) Low HDL-C; and (3) Insulin resistance. Each gram increase in MUFA intake raised MASLD risk by 3%, while a higher glycemic load increased risk by 1%, particularly in individuals with obesity, as exhibited in Table 2.

Table 2 Logistic regression analysis investigating the association between various factors and metabolically dysfunction-associated steatotic liver disease.

Logistic regression of significant variables	B	SE	OR	95%CI for OR	P value
BMI groups (reference normal BMI)					< 0.001
BMI (overweight vs normal BMI)	1.5	0.3	4.5	2.5-7.9	< 0.001
BMI (obese vs normal BMI)	3.3	0.3	27.5	14.6-51.7	< 0.001
HOMA-IR (elevated)	0.9	0.2	2.7	1.6-4.1	< 0.001
Age	0.1	0.01	1.06	1.03-1.08	< 0.001
Monounsaturated fatty acid	0.03	0.01	1.03	1.01-1.05	0.003
Very low-density lipoproteins	0.02	0.01	1.02	1.01-1.04	0.024
Normal BMI
Sex (males)	1.4	0.4	4.1	1.8-9.1	0.001
HOMA-IR groups (elevated)	1.2	0.4	3.4	1.5-7.6	0.003
High-density lipoprotein groups (low)	1.0	0.4	2.7	1.3-5.6	0.009
Overweight participants
Metabolic syndrome	1.0	0.5	2.7	1.1-6.8	0.033
Patients with obesity
HOMA-IR (elevated)	1.2	0.5	3.4	1.2-9.5	0.017
Age	0.1	0.02	1.13	1.08-1.2	< 0.001
Glycemic load	0.01	0.004	1.01	1.005-1.02	0.001

BMI: Body mass index; HOMA-IR: Homeostasis Model Assessment of Insulin Resistance; OR: Odds ratio; CI: Confidence interval.

Liver steatosis, measured via CAP, significantly increased with BMI (259 ± 20 dB/m in normal weight, 303 ± 36 dB/m in obesity, P < 0.001), correlating with liver stiffness (5.4 ± 1.5 kPa vs 6.6 ± 2.7 kPa, P < 0.001).

The study observed that the PNPLA3 gene variant was more common in overweight (53.41%) and obese (42.7%) MASLD participants compared to lean individuals (35.3%), with a significant P value of 0.019.

The median age of participants was 38 years, with 61.8% of MASLD patients being older and 73.7% of healthy controls being younger. MASLD prevalence was higher in older individuals (83%) compared with younger ones (52%), with younger males demonstrating a significantly higher risk (P = 0.029).

BMI and WC were significantly higher in older participants (P < 0.001), with mean BMIs of 31.4 ± 5.9 kg/m² in younger and 33.4 ± 5.6 kg/m² in older individuals. Low physical activity was a risk factor for younger MASLD patients (P = 0.012), and higher carbohydrate intake was significant in older MASLD participants.

Metabolic markers varied by age: Total cholesterol and LDL-cholesterol were higher in younger MASLD patients, whereas TGs, VLDL, HbA1c, fasting insulin, HOMA-IR, liver enzymes, and CRP were elevated across all MASLD groups. TSH levels were significantly higher only in younger MASLD participants. MetS was more prevalent in older MASLD patients (51.1%) than in younger ones (25.2%). The laboratory results for different age groups are presented in Table 3.

Table 3 Various clinical and laboratory parameters between metabolically dysfunction-associated steatotic liver disease and healthy participants, stratified by age, median (range).

	MASLD (< 38 years)	Non-MASLD (< 38 years)	P value	MASLD (≥ 38 years)	Non-MASLD (≥ 38 years)	P value
Lipoprotein lipase	16 (0.5-185)	19 (2-204)	0.300	13 (0.5-154)	16 (0.5-214)	0.817
L-carnitine	12.4 (1-58)	14 (1.2-58)	0.113	12.2 (0.1-63)	12.8 (0.6-43)	0.899
C-reactive protein	3.6 (0.3-56)	2 (0.3-33)	< 0.001	4.3 (0.3-117)	2.8 (0.2-39.2)	< 0.001
Aspartate aminotransferase	19 (8-114)	18 (9-103)	< 0.001	20 (4.5-236)	19 (10-42)	0.045
Alanine aminotransferase	21 (4-156)	14 (4-82)	< 0.001	20 (5-258)	15 (3-53)	< 0.001
Gamma-glutamyl transferase	21 (7-242)	13.5 (4-99)	< 0.001	24 (4-595)	17.5 (6-137)	< 0.001
Bilirubin	0.4 (0.1-1.9)	0.4 (0.1-54)	0.098	0.4 (0.1-2)	0.4 (0-1.2)	0.149
Homeostasis Model Assessment of Insulin Resistance	2.2 (0.1-73)	1.4 (0-10.7)	< 0.001	2.2 (0.1-31.9)	1.4 (0.1-20.2)	< 0.001
Fasting insulin	9.9 (0.3-279.4)	7 (0.3-45)	< 0.001	9.9 (0.3-109.7)	6.7 (0.3-39.1)	< 0.001
Triglycerides	104 (36-354)	68.5 (28-392)	< 0.001	130 (3.9-1086)	96.5 (43-409)	< 0.001
Very low-density lipoproteins	20.8 (7-70.8)	13.7 (5.8-78.4)	< 0.001	26 (3.1-217.2)	19.6 (8.6-81.8)	< 0.001
Thyrotropin	1.7 (0.01-100)	1.5 (0.01-9.7)	0.007	1.5 (0.01-16)	1.5 (0.01-6.6)	0.650

MASLD: Metabolically dysfunction-associated steatotic liver disease.

PNPLA3 gene analysis revealed comparable wild-type prevalence in both age groups (53.3% in younger and 51.1% in older participants). Although the GG polymorphism was associated with higher MASLD rates (64.7% in younger and 77.8% in older patients), these differences were not statistically significant (P = 0.299 and P = 0.55, respectively).

Age significantly influenced disease severity (P < 0.001), with older MASLD patients exhibiting higher liver steatosis (CAP: 300 ± 41 dB/m vs 283 ± 41 dB/m) and fibrosis (liver stiffness measurement: 6.6 ± 2.8 kPa vs 5.6 ± 1.7 kPa).

Multivariate analysis identified distinct MASLD risk factors across age groups as shown in Table 4. In younger individuals, BMI, HOMA-IR, and WC were significant predictors, whereas HbA1c played a stronger role in older participants. Overweight younger individuals had a 24-fold higher MASLD risk, while older participants required obesity (BMI ≥ 30) to reach a substantially greater 64.4-fold risk. A higher WC (> 80 cm in women and > 94 cm in men) increased the risk of MASLD by 3.4 times in younger patients. Elevated HOMA-IR (> 2.5) increased the risk by 5.6-fold in younger participants and by 3-fold in older participants. Prediabetes (HbA1c ≥ 5.6%) in older adults was associated with a 3.6-fold higher MASLD risk, though this trend was not observed in people with diabetes, possibly due to limited sample size.

Table 4 Logistic regression analysis examining the association between various factors and metabolically dysfunction-associated steatotic liver disease in relation to participants’ age.

Young population < 38 years	B	SE	OR	95%CI for OR	P value
BMI (reference normal BMI)					< 0.001
BMI (overweight vs normal BMI)	3.2	0.9	24	4.2-134	< 0.001
BMI (obese vs normal BMI)	3.5	0.9	32	5.5-183	< 0.001
Waist circumference (> 80 in female, > 94 in males)	1.2	0.6	3.4	1.1-11	0.042
HOMA-IR (elevated)	1.7	0.5	5.6	1.906	0.002
Old population ≥ 38 years
BMI (reference normal BMI)					< 0.001
BMI (overweight vs normal BMI)	1.3	0.5	3.8	1.5-9.8	0.006
BMI (obese vs normal BMI)	4.2	0.6	64.4	21-197.1	< 0.001
Elevated HOMA-IR	1.1	0.4	3.1	1.4-6.8	0.006
HbA1c (reference: Normal HBA1c ≤5.6)					0.045
HbA1c (prediabetic vs normal HbA1c)	1.3	0.5	3.6	1.3-9.9	0.013
HbA1c (diabetics vs normal HbA1c)	0.0	0.9	1.0	0.2-5.6	0.958

BMI: Body mass index; HBA1c: Glycated hemoglobin; HOMA-IR: Homeostasis Model Assessment of Insulin Resistance; OR: Odds ratio; CI: Confidence interval.

BMI and WC were significantly higher in MASLD participants with elevated HbA1c (P < 0.001). Diabetic and prediabetic MASLD patients had dietary patterns like those with normal liver function but exhibited higher CRP, gamma-glutamyl transferase, and TG levels. Cholesterol, liver enzymes, and thyroid function did not differ by HbA1c status. The distribution of the PNPLA3 gene polymorphism was similar across different HbA1c subgroups (P = 0.742).

Liver steatosis and fibrosis were more severe in diabetic (287 ± 39 dB/m, 5.9 ± 2.2 kPa) and prediabetic (304 ± 44 dB/m, 6.9 ± 2.8 kPa) MASLD patients compared with non-diabetics (P < 0.001). BMI, WC, and HOMA-IR were the strongest predictors in prediabetic/diabetic individuals, with obese diabetics facing the highest risk (17.6-fold).

Both wild-type and variant PNPLA3 gene carriers with MASLD exhibited significantly elevated glucose and lipid markers, liver enzymes, CRP, and FT4 compared with healthy controls, with no additional risk attributable to gene type. High glycemic index, glycemic load, and fructose intake were associated with the PNPLA3 variant in patients with MASLD. Although MUFA intake increased the risk of MASLD, this effect was consistent across all gene types (P = 0.272).

Key predictors of MASLD differed by PNPLA3 status. In wild-type carriers, BMI, WC, high HOMA-IR, and low HDL were primary risk factors. In variant carriers, age (11% increased risk per year), MetS (4.6-fold higher risk), and polyunsaturated fatty acid (PUFA) intake (10% increased risk per gram) were the main drivers of MASLD.

Liver steatosis correlated weakly with CRP, glucose markers, and FT3 (positive) and with HDL and LPL (negative), while revealing moderate correlations with age, BMI, WC, liver enzymes, TGs, and VLDL. Liver fibrosis had weak associations with age, CRP, bilirubin, HbA1c, and TGs/VLDL (positive), as well as with LPL, albumin, and HDL (negative). Still, it correlated moderately with BMI, WC, liver enzymes, fasting glucose/insulin, and HOMA-IR, as displayed in Figure 5.

Figure 5 A heatmap visualizing correlation coefficients (r) between various clinical factors and two measurement methods: Controlled attenuation parameter and kilopascals. ^aP < 0.05. ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; BMI: Body mass index; CRP: C-reactive protein; FT: Free thyroxine; γGT: Gamma-glutamyl transferase; HbA1c: Glycated hemoglobin; HDL: High-density lipoprotein; HOMA-IR: Homeostasis Model Assessment of Insulin Resistance; LDL: Low-density lipoproteins; LPL: Lipoprotein lipase; MUFA: Monounsaturated fatty acid; PUFA: Polyunsaturated fatty acid; TSH: Thyrotropin; VLDL: Very low-density lipoproteins.

DISCUSSION

Egypt carries a significantly higher burden of MASLD compared with global rates. The recent study identified a striking prevalence of 67.7% among Egyptian adults, substantially exceeding both the previously reported national rate of 47.5% and the global average of 25%. This alarming trend persists despite limited awareness among both Egyptian physicians and patients regarding the disease’s severity and long-term consequences.

The escalating obesity epidemic in Egypt (affecting 49.51% of women and 29.53% of men according to 2024 data)[20] appears to be a major driver of MASLD development. The present findings demonstrated a clear weight-dependent progression: 59.5% of overweight individuals and 100% of morbidly obese participants had MASLD, while even 22.2% of lean individuals were affected. Obesity conferred a 27.5-fold higher MASLD risk, and this relationship directly correlated with age. Distinct demographic patterns were additionally observed: (1) Men exhibited a fourfold higher risk compared with women; and (2) Prevalence increased markedly from 52% in younger adults (< 38 years) to 83% in older participants.

Physical activity demonstrated protective effects, aligning with the American College of Sports Medicine’s 2009 guidelines, which recommend 150-250 minutes of moderate exercise weekly for MASLD prevention[21]. Healthy controls in the current study significantly exceeded MASLD patients in meeting this threshold (P < 0.001), particularly through more than 140 minutes of weekly walking. Notably, inactivity specifically elevated risk in younger but not older MASLD patients, suggesting age-modified protective mechanisms.

These findings underscore the urgent need for targeted interventions addressing Egypt’s unique MASLD risk profile, with a focus on weight management in youth and metabolic health in older adults.

Evidence suggested that atherogenic dyslipidemia, marked by high TGs, low HDL-C, and elevated LDL-cholesterol, is predominant in MASLD participants[22]. The present results demonstrated that participants with MASLD had significantly elevated CRP, liver enzymes, and metabolic markers, with low HDL-C tripling the MASLD risk, and each unit increase in VLDL raising the risk by 2%. Insulin resistance (HOMA-IR ≥ 2.5) strongly predicted MASLD among Egyptians, raising the risk 3.4-fold per unit rise, even in normoglycemic individuals, highlighting its central role in MASLD pathogenesis, as previously documented[23].

Egyptian studies suggest that there is higher consumption of refined carbohydrates traditional baladi bread, white bread and polished rice, sugar sweetened beverages, as well as frequent consumption of deep-fried dishes with low intake of fruits and vegetables that directly contributed to the higher MASLD prevalence among Egyptians[24].

Dietary analysis revealed that a 1-gram increase in MUFAs raised the risk of MASLD by 3%, while a higher glycemic load increased the risk by 1%. Carbohydrate intake was significantly higher in older MASLD patients, whereas younger patients exhibited elevated levels across all dietary components analyzed.

Hepatic steatosis, the hallmark of MASLD, was observed in 67.7% of participants in the recent study, with 40.1% exhibiting severe steatosis, 15% having significant fibrosis (≥ 8 kPa), and 3% having cirrhosis (≥ 12 kPa). These results align with a prior Egyptian study, which reported 57.9% severe steatosis (S3) and 5% with fibrosis (F2-F3) on elastography[25].

Per European Association for the Study of the Liver-European Association for the Study of Diabetes-European Association for the Study of Obesity guidelines, non-invasive fibrosis assessment is recommended for patients with abnormal liver enzymes, cardiometabolic risks, or radiological steatosis, as these factors promote steatosis via free fatty acids, inflammation, and oxidative stress[9]. In the current cohort, the prevalence of MetS was higher in older MASLD participants (51.1%) than in younger ones (25.2%), supporting aging and metabolic dysfunction as key predictors of MASLD[26]. Additionally, participants fulfilling all MetS criteria had more severe steatosis and fibrosis than those with 1-2 criteria. This corroborates previous findings that CAP scores alone detect MetS with 0.79 accuracy, improving to 0.904 when combined with machine learning (ML)[27].

Younger individuals with MASLD exhibited significantly higher total cholesterol and LDL levels, while other markers of lipid and glucose metabolism were elevated across all age groups. Prior research identified baseline systolic blood pressure, along with increases in TGs and fasting blood glucose, as risk factors for MASLD in individuals aged ≥ 65 years[28]. Additionally, MASLD patients with diabetes or prediabetes had more severe liver steatosis and fibrosis than non-diabetic MASLD patients, irrespective of BMI.

In the present study, healthy participants displayed marginally higher levels of L-carnitine and LPL. LPL, a TG hydrolase, facilitates lipoprotein uptake via LPL receptors and influences adipocyte differentiation. L-carnitine helps mitigate excess mitochondrial acetyl-CoA, promotes fatty acid oxidation, and reduces hepatic free fatty acid accumulation[29]. Liver steatosis and fibrosis demonstrated a weak negative correlation with LPL and L-carnitine (P < 0.001), supporting their potential role as risk factors for MASLD in the Egyptian population.

Studies from 2012 to 2021 reported an inverse relationship between serum free T4 and fatty liver in euthyroid individuals, with no association observed for TSH or free T3[30,31]. The recent findings align with this: TSH was elevated only in younger MASLD patients, while FT4 was higher in healthy participants. Liver steatosis and fibrosis were weakly correlated with FT3, consistent with research by Van den Berg et al[32], which linked higher FT3, lower FT4, and an elevated FT3/FT4 ratio to euthyroid patients with MASLD. This likely reflects central adiposity, as iodothyronine deiodinase activity increases in obese individuals and high-fat diet models, thereby enhancing the conversion of FT4 to FT3.

Even under isocaloric conditions, the type of dietary fat and carbohydrate influences liver fat accumulation through distinct molecular and metabolic pathways that enhance de novo lipogenesis. Saturated fats and fructose, unlike unsaturated fats, drive the most significant increases in intrahepatic TGs, insulin resistance, and ceramide accumulation, thereby exacerbating the risk of MASLD[33].

In the present study, a 1-gram increase in MUFA and PUFA intake was associated with a 3% and 10% higher risk of MASLD, respectively, while a 1-unit rise in glycemic load was only linked to a 1% increased risk. This apparent paradox, where MUFA and PUFA fats typically considered “healthy” are linked to higher disease risk, likely reflects the specific dietary context in Egypt. The primary sources of these fat are not from whole foods like nuts, avocados, or olive oil used in salad dressings, but rather from oils repeatedly used for deep-frying traditional foods (e.g., falafel, fried fish, and feteer meshaltet) and from hydrogenated oils in commercially produced baked goods and sweets. These processing methods, particularly deep-frying at high temperatures, can lead to the formation of toxic lipid peroxidation products and trans fatty acids, which promote hepatic inflammation, oxidative stress, and insulin resistance[34,35]. Furthermore, a high intake of omega-6 PUFAs without a balanced intake of anti-inflammatory omega-3 PUFAs, a common feature of modern diets, can exacerbate pro-inflammatory eicosanoid production. Therefore, the observed association likely signifies a diet high in energy-dense, processed, and fried foods a pattern that overwhelms any potential benefits of MUFA/PUFA and directly contributes to hepatic lipid accumulation and metabolic dysfunction[35].

The PNPLA3 gene mutation was detected in 43.6% of healthy participants and 48.1% of MASLD participants, aligning with previous reports that PNPLA3 polymorphisms (normal, heterozygous, and homozygous) are highly prevalent in MASLD patients[36]. Nearly half of the present cohort carried this variant, with MASLD participants predominantly exhibiting either the wild-type “CC” or variant “CG/GG” genotypes. The variant was significantly more frequent in overweight (53.41%) and obese (42.7%) individuals with MASLD compared to lean MASLD cases.

Dietary and metabolic associations differed by genotype: In wild-type PNPLA3 carriers, BMI, WC, high HOMA-IR, and low HDL predicted MASLD, whereas mutant gene carriers were influenced by age, PUFA intake, and MetS. Earlier research corroborates that the PNPLA3 genotype modulates MASLD progression, with high intake of carbohydrates, isoflavones, and methionine exacerbating the risk of hepatic fibrosis[37]. These findings underscore the importance of investigating the roles of nutrigenomics and epigenetics in the Egyptian population.

These results establish MASLD as a critical public health priority in Egypt, highlighting the complicated relationship between metabolic, genetic, and lifestyle factors in disease pathogenesis. The key strengths of this study include being the most extensive MASLD study in North Africa, featuring comprehensive metabolic and genetic profiling, standardized imaging protocols across centers, and representative geographic sampling.

While this study provides valuable insights into MASLD epidemiology in Egypt, its cross-sectional design limits the ability to make causal inferences. The reliance on imaging rather than histology, potential selection bias, and unmeasured confounders (e.g., environmental toxins and other genetic variants) underscore the need for longitudinal, biopsy-confirmed studies that incorporate broader risk assessments to validate these findings and inform personalized interventions.

The potential of AI for MASLD management and research

The complex interplay of metabolic, genetic, and lifestyle factors in MASLD pathogenesis makes it a prime candidate for advanced analytical approaches. AI and ML have the potential to improve risk stratification by integrating the predictors identified in this study (e.g., PNPLA3 variants, glycemic load, HOMA-IR) with imaging biomarkers to develop personalized prognostic models. For instance, previous studies have demonstrated that ML models can predict fibrosis progression from ultrasound elastography with high accuracy when combined with metabolic data[38]. In resource-constrained settings like Egypt, AI-powered tools could potentially aid in early detection within primary care, while generative AI might help tailor dietary interventions to local food patterns.

However, the development and deployment of such AI solutions face significant challenges. These include the requirement for large, high-quality, and well-annotated datasets for training, and the critical need to ensure generalizability across diverse populations to mitigate algorithmic bias. The comprehensive, multi-center dataset generated by this study represents a valuable initial step toward building such robust, region-specific models. Future work should focus on validating AI algorithms in the Egyptian and other Middle Eastern and North African populations to ensure their clinical utility and equity.

While this study provides comprehensive insights into the MASLD epidemic in Egypt, several limitations should be considered. First, the cross-sectional design precludes the determination of causality or the assessment of disease progression over time. Second, the diagnosis of MASLD and steatosis grading relied on ultrasonography and VCTE rather than liver histology, the gold standard, which may underestimate the prevalence of steatohepatitis (metabolic dysfunction-associated steatohepatitis) and early fibrosis. Third, although geographically diverse, our sample was one of convenience, potentially limiting its generalizability to all Egyptian subpopulations. Further-more, we did not perform comprehensive serological testing to exclude all secondary causes of steatotic liver disease, such as Wilson’s disease, autoimmune hepatitis, or monogenic lipid disorders. Finally, while we identified significant genetic-dietary interactions, other potential confounders, including environmental toxins, detailed socioeconomic factors, and the gut microbiome, were not assessed and warrant investigation in future studies.

Recommendations

The findings of this research underscore the urgent need for nationwide MASLD screening programs, tailored lifestyle interventions addressing Egypt’s unique dietary patterns, and genetic risk stratification, particularly for individuals carrying the PNPLA3 variant. Physician education is crucial for enhancing disease recognition and management. Future longitudinal studies should investigate progression rates and evaluate targeted prevention strategies in this high-risk population. Furthermore, future research should investigate AI-driven precision medicine approaches for managing MASLD in resource-constrained settings.

CONCLUSION

This extensive multicenter study reveals an alarmingly high prevalence of MASLD (67.7%) among Egyptian adults, far exceeding global averages and establishing Egypt as a critical hotspot for this condition. The epidemic is driven by a confluence of powerful factors, including a high prevalence of obesity (conferring a 27.5-fold increased risk), insulin resistance, a sedentary lifestyle, and a diet high in refined carbohydrates, and fats. Furthermore, the high population frequency of the PNPLA3 rs738409 variant, which interacts with dietary intake, creates a potent genetic-dietary nexus that modulates disease severity. These findings underscore the urgent public health imperative to implement national screening programs and culturally tailored lifestyle interventions, aligning with recent global calls to action for a coordinated response to the MASLD pandemic.

ACKNOWLEDGEMENTS

The authors would like to acknowledge Ashraf Gayel Ahmed (consultant Clinical Nutritionin Road Elfarag Hospital, Egypt), Manal Farouk Ahmed Abdelhafeez, and Hala Wassef Abdallah (consultants of clinical pathology in El-Helal Hospital, Egypt); for their participation in the data collection in this study.