Hepatic Cysts in Korea: Prevalence, Incidence, and Risk Factors Based on a 5.8-Year Follow-up Health Screening Cohort Authors: Kang Seo, Yujeong Kim, Sang Hyuk Park, Nam-Joon Yi, Jeong Hee Yoon, Goh Eun Chung, Jeong-Moo Lee, Seoung Yoon Rho, Min Su Park, Eun Kyung Choe, Aesun Shin Published: 2026 ABSTRACT AbstractBackgroundHepatic cysts are fluid-filled liver lesions that are mostly asymptomatic, but fewer than 15% of patients present clinically significant symptoms. Reported prevalence ranges from 2.5% to 71%, yet prior studies have been limited by small populations and incomplete assessment of risk factors. This study aimed to determine the prevalence, incidence, and risk factors using data from a large health screening cohort in Korea.MethodsWe performed a retrospective cohort analysis of 78,676 participants in the Health and Prevention Enhancement cohort who underwent abdominal ultrasound at the Seoul National University Hospital Healthcare System Gangnam Center from 2003 and 2014. Hepatic cysts were identified using a validated rule-based text-mining algorithm applied to ultrasound reports. Prevalence was estimated from baseline data. Incidence was assessed in 41,179 participants initially negative for hepatic cysts. Risk factors were determined through Cox proportional hazards modeling.ResultsThe prevalence of hepatic cysts at baseline was 8.54% (95% confidence interval [95% CI], 8.34–8.73). The incidence was 23.4 per 1,000 person-years at risk (95% CI, 22.8–24.0). Cox regression identified hepatitis B virus surface antigen (HBs-Ag) positivity (hazard ratio [HR], 1.20; 95% CI, 1.00–1.45) and renal cyst presence (HR, 1.46; 95% CI, 1.31–1.64) as independent risk factors, whereas diabetes mellitus (HR, 0.64; 95% CI, 0.54–0.77) and high physical activity (HR, 0.89; 95% CI, 0.81–0.97) appeared protective.ConclusionThis study provides the prevalence of 8.54%, incidence, and risk factors of hepatic cysts. Renal cyst presence and HBs-Ag positivity are significant risk factors, whereas high physical activity and diabetic status are protective.Keywords: Cysts, Liver Diseases, Prevalence, Incidence, Risk Factors, Korea INTRODUCTION Hepatic cysts are fluid-filled lesions in the liver lined by a single layer of cuboidal or columnar biliary epithelium.1 Generally, hepatic cysts are asymptomatic, with fewer than 10–15% of patients experiencing clinically significant symptoms.2 Most simple hepatic cysts are found incidentally on abdominal imaging, such as ultrasound (US), computed tomography, or magnetic resonance imaging (MRI).3,4 Advancements in imaging technology and wide accessibility of imaging have resulted in a greater number of incidental diagnoses of hepatic cysts than in the past.5 Previous studies have demonstrated wide variability in the prevalence of hepatic cysts, ranging from 2.5% to 71%, depending on study design, population characteristics, and image modality.4,6,7,8,9 The incidence data are very limited, with only one prior study reporting a 5-year cumulative incidence of 24.6% in 431 individuals.6 Reported risk factors exhibit significant inconsistencies across studies. While older age4,6,10,11 and renal cyst4,10 or pancreatic cyst presence4 are consistently associated with a higher prevalence, sex associations remain controversial. Most studies indicate a female predominance,4,7,8 yet others report a male predominance6 or no difference.10 Additionally, conflicting associations exist regarding metabolic factors. A large German population study found that hepatic cysts are associated with favorable metabolic profiles, such as a lower body mass index, triglycerides, serum protein and alanine amino-transferase levels. Conversely, a Japanese study revealed associations with different metabolic parameters, including low albumin and gamma-glutamyl transpeptidase (GGT) but high low-density lipoprotein and high systolic blood pressure. These conflicting findings suggest intricate relationships between hepatic cysts and metabolic factors that require further investigation. Given the wide variability in reported prevalence and incidence, limited understanding of the risk factors, and significant inconsistencies among existing studies, large-scale epidemiological research using standardized methodologies is critically needed. The increasing prevalence of benign hepatic lesions in Korea highlights the clinical importance of characterizing incidental hepatic findings.12 Therefore, the aim of our study was to evaluate the prevalence and incidence of hepatic cysts in a large population cohort over 10 years and identify consistent risk factors for hepatic cyst development. METHODS Study design and population This study was based on the Health and Prevention Enhancement (H-PEACE) cohort, which specifically analyzed data from the Seoul National University Hospital Healthcare System Gangnam Center, an institution providing comprehensive health checkups for the Korean population.13 Participants in the H-PEACE cohort were enrolled from October 2003 to December 2014, encompassing a total of 91,152 participants. The inclusion criterion for the current analysis was abdominal ultrasound (US) examination at the time of enrollment. Patients without an abdominal US or basic clinical data were excluded. For the incidence and risk factor analysis, we excluded participants who i) had hepatic cysts on baseline US or ii) did not have any follow-up US. US examination and reports All abdominal US examinations were performed and interpreted by board-certified radiologists at the Seoul National University Hospital Healthcare System Gangnam Center. To maximize the acoustic window and minimize interference from bowel gas, all patients underwent a strict 12-hour overnight fasting protocol prior to the examination. Although hepatic steatosis can increase sound attenuation, the diagnosis of hepatic cysts and renal cysts were diagnosed based on their characteristic findings on US: anechoic, thin-walled, well-circumscribed lesions with or without posterior acoustic enhancement.14,15 These features typically remain detectable by radiologists even in the presence of moderate hepatic steatosis. Abdominal US reports were written by the radiologist in bilingual (Korean and English) format. Definition of hepatic cysts and renal cysts The presence of hepatic cysts and renal cysts was derived from abdominal US reports using a rule-based algorithm employing word distance calculations. Distance computations were restricted to words within the same sentence and excluded cross-sentence word pairs. For algorithm development and validation, training and test sets were obtained through 5% random sampling from baseline US reports containing both liver-meaning and cyst-meaning words. Both datasets were manually annotated by two medical doctors (interrater agreement: κ = 0.978 for hepatic cysts, κ = 0.975 for renal cysts). A custom R function was designed to handle Korean and English US reports appropriately. The detailed algorithm specifications are provided in the Supplementary Data 1. Covariates Variables that are known or hypothesized to be associated with hepatic cysts were selected as covariates. The chosen covariates were sociodemographic information (sex, age, education level), lifestyle factors (alcohol consumption, smoking and physical activity), body measure (waist circumference), medical history (diabetes mellitus, hypertension, renal cysts), and laboratory results (hepatitis B virus surface antigen [HBs-Ag], hepatitis C virus antibody [anti-HCV]). Baseline age was stratified into groups of < 30, 30–39, 40–49, 50–59, 60–69, and ≥ 70. Education was dichotomized as ‘less than high school’ and ‘high school graduate.’ Alcohol consumption was dichotomized as ‘drinking more than 140 g/week’ and ‘other.’ Smoking status was dichotomized as ‘current smokers’ and ‘nonsmokers.’ Physical activity was categorized as ‘health-enhancing physically active (HEPA),’ ‘minimally active,’ or ‘inactive.’ Waist circumference was dichotomized by a cutoff of 85 cm for women and 90 cm for men. Diabetes, hypertension, and renal cysts were categorized as ‘yes’ or ‘no.’ HBs-Ag and anti-HCV were categorized as positive or negative. Outcome assessment This study had two primary outcomes: the prevalence and incidence of hepatic cysts. Prevalence was calculated by dividing the number of participants with hepatic cysts identified in the baseline US by the total number of participants. The incidence of hepatic cysts was evaluated among participants who did not have hepatic cysts at baseline. For each participant, the baseline date was defined as the date of their first US examination. The endpoint for this study was characterized as either 1) the date of the US examination in which a hepatic cyst was first identified or 2) the date of the last US examination if all examinations were negative for hepatic cysts. Hepatic cyst incidence was calculated by counting the number of new cases of hepatic cysts and then dividing this figure by the person-years at risk (PYR), with results expressed per 1,000 PYR. All variables used in the Cox proportional hazards model were categorized, with a cutoff known to be related to mortality or certain chronic conditions.16,17,18,19,20,21 Statistical analysis The performance of the rule-based algorithm for identifying hepatic and renal cysts was assessed using a test dataset independently labeled by two medical doctors. The metrics for comparison were accuracy, sensitivity, specificity, precision and the F1 score. Demographic variables are expressed as the means and standard deviations for continuous variables, whereas categorical variables are presented as absolute and relative frequencies. Comparisons between two groups were conducted with Student’s t-test for continuous variables and the χ2 test for categorical variables. Cox proportional hazards regression models were used to identify covariates associated with the incidence of hepatic cysts by using age as a time scale. To identify independent risk factors for hepatic cyst incidence, Cox proportional hazards models were used with age as a time scale. The purposeful selection approach recommended by Hosmer et al.22 was employed to identify independent risk factors. Variables with P < 0.10 in univariable analysis were selected for inclusion in the multivariable Cox proportional hazards model. Hazard ratios (HRs) with 95% confidence intervals (95% CIs) are reported for each covariate, calculated by the exact method. The proportional hazard assumption was tested by examining Schoenfeld residuals and creating time-dependent covariates for each predictor, with P < 0.05 indicating a violation of the assumption. All analyses were performed using R version 4.4.1 (R Foundation for Statistical Computing, Vienna, Austria). Ethics statement This study was approved by the Institutional Review Board (IRB) of Seoul National University Hospital, Seoul National University College of Medicine (IRB No. 2407-013-1549). The requirement for informed consent was waived by the IRB due to the retrospective nature of the study. RESULTS The study flowchart (Fig. 1) illustrates the participant selection process, resulting in 78,676 participants whose baseline US data and clinical information were included in the prevalence analysis. For the incidence and risk factor analysis, 41,179 participants were selected. Study flowchart of the study population at the Seoul National University Hospital Gangnam Center, Seoul, Korea. US = ultrasound. Cyst identification performance The performance of the rule-based algorithm for identifying hepatic and renal cysts was assessed using a manually labeled test dataset (n = 601; Table 1). The rule-based approach resulted in high scores for every metric (F1 score: 95.1 for hepatic cysts, 97.2 for renal cysts). Table 1Cyst identification performance for test data (N = 601)MetricRule-based approachHepatic cystRenal cystAccuracy94.3 (92.2–96.1)97.2 (94.9–98.0)Sensitivity94.3 (91.3–96.5)97.0 (94.4–98.6)Specificity94.4 (90.8–96.9)97.3 (94.8–98.8)Precision95.9 (93.3–97.8)97.3 (94.8–98.8)F1 score95.1 (87.7–100.0)97.2 (90.7–100.0)Data given as percentages with 95% confidence intervals in parentheses.Precision: positive predictive value.F1 score: 2 × (precision × sensitivity)/(precision + sensitivity). Cyst identification performance for test data (N = 601) Data given as percentages with 95% confidence intervals in parentheses. Precision: positive predictive value. F1 score: 2 × (precision × sensitivity)/(precision + sensitivity). Prevalence of hepatic cysts The baseline prevalence of hepatic cysts was 85.4 per 1,000 persons (8.54%; 95% CI, 8.34–8.73), indicating that a substantial portion of the study population harbored hepatic cysts. Table 2 shows the demographic and clinical characteristics of the participants. Age and the presence of renal cysts were associated with the presence of hepatic cysts. Interestingly, patients with higher physical activity had a higher prevalence and patients with diabetes had a lower prevalence. Table 2Demographic and clinical characteristics, and the prevalence of hepatic cysts in the study population at the Seoul National University Hospital Gangnam CenterCharacteristicsTotalHepatic cyst, positiveP valueTotal78,6766,717 (8.5)Sex< 0.001Female34,8383,123 (9.0)Male43,8383,594 (8.2)Age group, yr< 0.001< 305,93453 (0.9)30–3919,278666 (3.5)40–4924,9621,961 (7.9)50–5919,1662,385 (12.4)60–697,7341,304 (16.9)≥ 701,602348 (21.7)Education< 0.001Less than high school10,7551,135 (10.6)High school graduate67,9215,582 (8.2)Alcohol consumption< 0.001≤ 140 g/wk60,7225,423 (8.9)> 140 g/wk17,9541,294 (7.2)Smoking0.002Non-smokers41,3533,652 (8.8)Current smokers37,3233,065 (8.2)Physical activity< 0.001Inactive31,3722,243 (7.1)Minimally active22,5042,077 (9.2)HEPA24,8002,397 (9.7)Waist circumference, cm< 0.001Female < 85, male < 9054,9024,313 (7.9)Female ≥ 85, male ≥ 9023,7742,404 (10.1)Diabetes mellitus0.005No72,9216,283 (8.6)Yes5,755434 (7.5)Hypertension< 0.001No67,3395,276 (7.8)Yes11,3371,441 (12.7)HBs-Ag positive< 0.001No75,4296,384 (8.5)Yes3,247333 (10.3)Anti-HCV positive< 0.001No77,9226,616 (8.5)Yes754101 (13.4)Renal cyst< 0.001No70,7725,230 (7.4)Yes7,9041,487 (18.8)Values are presented as number (%).P value is calculated by χ2 test for each covariate.HEPA = health-enhancing physically active, HBs-Ag = hepatitis B virus surface antigen, anti-HCV = hepatitis C virus antibody. Demographic and clinical characteristics, and the prevalence of hepatic cysts in the study population at the Seoul National University Hospital Gangnam Center Values are presented as number (%). P value is calculated by χ2 test for each covariate. HEPA = health-enhancing physically active, HBs-Ag = hepatitis B virus surface antigen, anti-HCV = hepatitis C virus antibody. Incidence of hepatic cysts The incidence of hepatic cysts during the study period was calculated utilizing PYR, yielding a rate of 23.4 per 1,000 PYR (95% CI, 22.8–24.0). Table 3 shows that most of the factors related to prevalence failed to show an association with incidence. However, higher physical activity levels, hypertension, and the presence of renal cysts were significantly associated with increased incidence. Conversely, patients with diabetes mellitus exhibited a lower incidence. Fig. 2 visually depicts the age-specific prevalence and incidence by sex. There was a sex difference in prevalence, with a higher prevalence among women; however, no significant difference by sex was observed for incidence. A linear increase in prevalence was observed with age, but the incidence decreased after age 50 for women and after age 60 for men. Table 3Incidence density of hepatic cysts from 2003–2024 in the study population at the Seoul National University Hospital Gangnam CenterCharacteristicsTotalPYRNo. of new casesHepatic cyst incidenceP valueTotal41,179240,7015,62823.4 (22.8–24.0)Sex0.173Female17,602 (42.7)104,2102,38622.9 (22.0–23.8)Male23,577 (57.3)136,4923,24223.8 (23.0–24.6)Education0.244Less than high school3,777 (9.2)20,77546122.2 (20.2–24.3)High school graduate37,402 (90.8)219,9265,16723.5 (22.9–24.1)Alcohol consumption0.579≤ 140 g/wk31,163 (75.7)183,0094,26123.3 (22.6–24.0)> 140 g/wk10,016 (24.3)57,6921,36723.7 (22.5–25.0)Smoking0.776Non-smokers21,431 (52.0)127,2392,96423.3 (22.5–24.1)Current smokers19,748 (48.0)113,4632,66423.5 (22.6–24.4)Physical activity< 0.001Inactive15,145 (36.8)87,7921,85021.1 (20.1–22.0)Minimally active12,681 (30.8)74,6741,85924.9 (23.8–26.0)HEPA13,353 (32.4)78,2351,91924.5 (23.5–25.6)Waist circumference, cm0.606Female < 85, male < 9029,231 (71.0)172,4124,04923.5 (22.8–24.2)Female ≥ 85, male ≥ 9011,948 (29.0)68,2901,57923.1 (22.0–24.3)Diabetes mellitus0.004No38,438 (93.3)225,3855,32223.6 (23.0–24.2)Yes2,741 (6.7)15,31730620.0 (17.8–22.3)Hypertension< 0.001No35,473 (86.1)208,1334,72722.7 (22.1–23.4)Yes5,706 (13.9)32,56890127.7 (25.9–29.5)HBs-Ag positive0.130No39,508 (95.9)231,6705,39523.3 (22.7–23.9)Yes1,671 (4.1)9,03223325.8 (22.6–29.3)Anti-HCV positive0.460No40,829 (99.2)238,7925,57823.4 (22.8–24.0)Yes350 (0.8)1,9095026.2 (19.5–34.4)Renal cyst< 0.001No37,449 (90.9)220,4134,81121.8 (21.2–22.4)Yes3,730 (9.1)20,28981740.3 (37.6–43.1)Values are presented as number (%), or number (95% confidence interval range).Median value for PYR: 5.64 for total; 5.98 for incidence (−); 4.06 for incidence (+).Incidence is calculated per 1,000 PYRs.P value is calculated by χ2 test for each covariate.PYR = person-years at risk, HEPA = health-enhancing physically active, HBs-Ag = hepatitis B virus surface antigen, anti-HCV = hepatitis C virus antibody. Incidence density of hepatic cysts from 2003–2024 in the study population at the Seoul National University Hospital Gangnam Center Values are presented as number (%), or number (95% confidence interval range). Median value for PYR: 5.64 for total; 5.98 for incidence (−); 4.06 for incidence (+). Incidence is calculated per 1,000 PYRs. P value is calculated by χ2 test for each covariate. PYR = person-years at risk, HEPA = health-enhancing physically active, HBs-Ag = hepatitis B virus surface antigen, anti-HCV = hepatitis C virus antibody. Comparison of the prevalence and incidence of hepatic cysts by age and sex. Prevalence is calculated per 1,000 people and grouped by baseline age. Incidence is calculated per 1,000 PYR and grouped by age at diagnosis. Error bars represent 95% confidence intervals. PYR = person-years at risk. Risk factors for hepatic cysts Results of the Cox proportional hazards regression are shown in Table 4. The final multivariable model revealed that HBs-Ag positivity (HR, 1.20; 95% CI, 1.00–1.45) and the presence of renal cysts (HR, 1.46; 95% CI, 1.31–1.64) were risk factors for hepatic cyst formation. Diabetes mellitus (HR, 0.64; 95% CI, 0.54–0.77), GGT levels greater than 50 U/L (HR, 0.77; 95% CI, 0.68–0.87), and blood urea nitrogen (BUN) levels greater than 20 mg/dL (HR, 0.79; 95% CI, 0.63–0.99) and triglyceride greater than 150 mg/dL (HR, 0.89; 95% CI, 0.79–0.99) were shown to be protective against hepatic cyst development. The incidence of hepatic cysts was nearly twofold greater in individuals with renal cysts than in those without renal cysts according to the multivariable analysis (HR, 1.46; 95% CI, 1.31–1.64). Greater physical activity and the presence of diabetes were also inversely associated with hepatic cyst incidence in the multivariable model. Table 4Cox regression model of risk factors for hepatic cystsVariablesUnivariate analysisMultivariate analysisaHazard ratio (95% CI)P valueHazard ratio (95% CI)P valueSex, male 0.92 (0.86–1.00) 0.043 1.02 (0.92–1.14)0.673High school graduate1.12 (0.99–1.28)0.0731.10 (0.97–1.25)0.140Alcohol > 140 g/wk0.93 (0.85–1.02)0.131--Current smokers 0.91 (0.85–0.99) 0.020 0.98 (0.89–1.09)0.702Physical activity, HEPA 0.90 (0.82–0.99) 0.034 0.89 (0.81–0.97) 0.013 Waist circumference, cm (female ≥ 85, male ≥ 90) 0.86 (0.79–0.94) 0.001 0.94 (0.86–1.03)0.207Diabetes mellitus, yes 0.60 (0.50–0.71) < 0.001 0.64 (0.54–0.77) < 0.001 Hypertension, yes 0.88 (0.79–0.99) 0.031 0.95 (0.84–1.06)0.345HBs-Ag positive, yes 1.23 (1.03–1.48) 0.025 1.20 (1.00–1.45) 0.048 Anti-HCV positive, yes0.70 (0.43–1.15)0.159--Renal cyst, positive 1.42 (1.27–1.59) < 0.001 1.46 (1.31–1.64) < 0.001 Albumin, < 4 g/dL1.10 (0.90–1.33)0.342--Platelet, ≥ 2.46 E90.94 (0.87–1.01)0.106--ALT, ≥ 120 U/L1.00 (0.58–1.73)0.986--GGT, > 50 U/L 0.71 (0.64–0.79) < 0.001 0.77 (0.68–0.87) < 0.001 BUN, > 20 mg/dL 0.78 (0.62–0.97) 0.025 0.79 (0.63–0.99) 0.037 Uric acid, mg/dL (female ≥ 4.9, male ≥ 6.3) 0.92 (0.85–0.99) 0.036 0.97 (0.90–1.05)0.490Triglyceride, ≥ 150 mg/dL 0.79 (0.71–0.87) < 0.001 0.89 (0.79–0.99) 0.032 LDL, ≥ 160 mg/dL0.90 (0.80–1.02)0.0930.94 (0.83–1.06)0.328TSH, > 4.0 mIU/L0.88 (0.74–1.04)0.142--CI = confidence interval, HEPA = health-enhancing physically active, HBs-Ag = hepatitis B virus surface antigen, anti-HCV = hepatitis C virus antibody, ALT = alanine amino-transferase, GGT = gamma-glutamyl transferase, BUN = blood urea nitrogen, LDL = low density lipoprotein, TSH = thyroid-stimulating hormone.Bold value indicates statistical significance (P < 0.05).aAdjusted for variables with a P value < 0.10 in univariate analyses. Cox regression model of risk factors for hepatic cysts CI = confidence interval, HEPA = health-enhancing physically active, HBs-Ag = hepatitis B virus surface antigen, anti-HCV = hepatitis C virus antibody, ALT = alanine amino-transferase, GGT = gamma-glutamyl transferase, BUN = blood urea nitrogen, LDL = low density lipoprotein, TSH = thyroid-stimulating hormone. Bold value indicates statistical significance (P < 0.05). aAdjusted for variables with a P value < 0.10 in univariate analyses. DISCUSSION In our study, the prevalence of hepatic cysts was 8.54%, and the incidence was 23.4 per 1,000 person-years in the study population. HBs-Ag positivity and the presence of renal cysts were risk factors for hepatic cyst development. These findings are clinically significant given that hepatic cysts are now diagnosed in more than 8,000 individuals annually in South Korea.23 The reported prevalence of hepatic cysts in existing studies ranges widely from 2.5% to 71%, complicating direct comparisons. Notably, studies utilizing abdominal US have reported prevalences of 2.5%,7 3.64%,9 3.9%,24 5.8%,8 and 21.9%.4 In contrast, our study observed a prevalence of 8.54%. These discrepancies could be due to differences in the timing of the studies, diagnostic performance, and study population characteristics. Recent large-scale Korean epidemiological surveys have documented a rising prevalence of benign hepatic lesions,25 highlighting the need for systematic surveillance of incidental findings in health screening programs. While many studies have been conducted on the prevalence of hepatic cysts, research into their incidence remains limited. To date, only one study has specifically investigated the incidence of hepatic cysts by MRI. That study reported a five-year cumulative incidence of 24.6% (36 out of 143) during a median follow-up of 4.8 years (standard deviation 0.8) among individuals without hepatic cysts on initial US.6 In our study, the 5-year cumulative incidence was approximately 11.7%, which was lower than that reported in previous studies. The previous study also reported a considerably higher hepatic cyst prevalence of 71%. The relatively higher prevalence and incidence may be attributed to the use of MRI, which offers greater sensitivity. Additionally, the higher mean age of 54.4 ± 12.1 years, compared with 45.56 ± 11.4 years in our study, may also be contributing to this difference. Our findings revealed a greater prevalence of hepatic cysts in women, which aligns with the findings of previous studies.4,7,8 However, some studies have reported a higher prevalence in men6 or found no significant difference between the sexes.10 In terms of incidence, our study revealed no significant difference in incidence between men and women. Despite no difference in incidence, the higher prevalence observed in women suggests the possibility of a factor influencing the persistence or growth of cysts rather than their formation. One notable factor is estrogen. A previous study showed that the proliferative and secretory activities of the cystic epithelium can be regulated by estrogens either directly or by synergizing growth factors, such as insulin-like growth factor 1, follicle-stimulating hormone, and vascular endothelial growth factor (VEGF).26 These findings indicate a possible association between estrogen and cyst maintenance. Our findings revealed that the prevalence of hepatic cysts increased significantly with increasing age, which is consistent with the findings of previous studies.4,10,11 However, our study also revealed a distinct pattern in the incidence, with a rapid increase from the 20s to the 40s, followed by a plateau and eventual decline in older individuals. This contrasts with a previous study that reported no significant age-related differences in incidence.6 It is well established that the overall metabolism rate decreases with age.27 Although no specific studies have examined the association between the metabolic rate and hepatic cysts, a decrease in the metabolic rate could lead to a decrease in hepatic cyst formation, lowering the incidence in elderly individuals. Additionally, our study revealed a greater prevalence of hepatic cysts among individuals with renal cysts, which is consistent with previous research.4,10,28 Furthermore, we found that renal cysts were also associated with an increased incidence of hepatic cysts. Prior studies10,28 have suggested that this association may be age dependent, pointing to shared pathophysiological mechanisms. In our study, HBs-Ag positivity was an independent risk factor for hepatic cyst incidence. This association may be partially explained by potential surveillance bias, as hepatocellular carcinoma screening guidelines recommend 6-monthly US surveillance for HBs-Ag-positive individuals.29,30 Although HBs-Ag-positive participants had a shorter time to the first follow-up (1.15 vs. 1.34 years), HBs-Ag-negative participants had longer total follow-up duration (6.59 vs. 5.74 years) with the same total and 5-year numbers of US examinations (median 4 and 2, respectively). Taken together, these findings suggest that while some degree of surveillance bias is possible, it is unlikely to fully explain the observed association. Notably, several biological mechanisms have been identified. Chronic hepatitis B virus (HBV) infection induces hepatic fibrogenesis via transforming growth factor-β1 signaling, which disrupts the biliary architecture and may facilitate cyst development.26 Additionally, HBV-induced VEGF upregulation can alter peribiliary vascular permeability, potentially promoting cystic dilation.31 These pathophysiological processes provide a mechanistic basis for the observed associations. An unexpected finding in our study was the significantly greater prevalence and incidence of hepatic cysts among individuals categorized as minimally active or active (classified as HEPA) than among those with no physical activity. This contradicts the general expectation that physical activity plays a protective role against the development of liver disease.32 However, the Cox proportional hazards model indicated a gradual increase in protective effects against hepatic cyst incidence, which aligns with our initial hypothesis. Physical activity stimulates and activates the VEGF signaling pathways.33 Another study conducted in mice demonstrated that liver cells with VEGF induced a proliferative response in liver cysts.34 These findings suggest a possible association between increased physical activity and hepatic cyst development through mechanisms similar to estrogen. Our study also revealed a greater prevalence and incidence of hepatic cysts in nondiabetic individuals than in diabetic individuals, which is contrary to typical expectations. Metformin, an oral diabetic medication, may influence hepatic cyst formation. An experimental study on polycystic liver disease revealed that metformin slowed liver cyst formation and fibrosis in a rat model.35 This raises the possibility that diabetic patients taking metformin may exhibit lower rates of hepatic cyst formation, contributing to the negative association between hepatic cysts and diabetes. Another notable finding was the higher prevalence of hepatic cysts among individuals who do not consume alcohol, despite no significant difference in incidence on the basis of alcohol consumption. Alcohol consumption is known to contribute to liver fibrosis and steatosis.36,37,38 Although there is no established association between fibrosis or steatosis and hepatic cysts, we can hypothesize that these changes might interrupt hepatic cyst development. While hypertension was not a significant risk factor of hepatic cysts in our study, prior studies have suggested a potential association between hypertension and hepatic cysts. Multiple case reports have indicated that systemic and portal hypertension may be induced by enlarged hepatic cysts,39,40,41 highlighting their possible impact on the cardiovascular system. The protective effect of high BUN is surprising, given that high BUN is mostly associated with adverse health effects, including mortality.42 The liver converts ammonia into urea for safe excretion. In chronic liver disease, impaired hepatocyte function reduces urea production, leading to lower BUN levels.43 This phenomenon is especially pronounced in cirrhosis, and this pathophysiology may be related to the observed protective effects of high BUN against hepatic cyst formation. In our study, hepatic cysts were evaluated using abdominal US imaging reports. This method has demonstrated reliability and diagnostic accuracy. The application of standardized evaluation criteria established by experts ensured a high level of performance, leading to clinically significant findings. Additionally, this approach is both straightforward and flexible to use, making it applicable in other research settings and clinical evaluations. Current clinical guidelines recommend against routine follow-up of asymptomatic simple hepatic cysts.23 Therefore, understanding the prevalence, incidence, and risk factors of hepatic cysts could reassure patients and healthcare professionals that these findings, though common (8.54% of our population), typically follow a benign, non-progressive course without the need for routine follow-up. Thus, our epidemiological findings on risk factors do not support more frequent imaging surveillance, regardless of baseline risk factors, consistent with current guidelines. Our study has certain limitations. As a single-center study conducted within a specific health care system, the generalizability of our findings to the broader population may be limited. Selection bias is also a potential concern, as individuals undergoing routine health checkups may differ from the general population in terms of age, sex, socioeconomic status, obesity, smoking habits, alcohol consumption, comorbidities, and medication use.44 However, comparison with the 2009 Korean National Health and Nutrition Examination Survey showed similar distributions of diabetes and physical activity in our cohort (Supplementary Table 1). This suggests that selection bias for these key covariates is unlikely to substantially distort our findings. Additionally, our study did not account for the size, number, or location of hepatic cysts. Because cyst identification was based on text diagnosis from US imaging reports, detailed information on cyst characteristics is limited. Furthermore, the use of abdominal US as the detection modality introduces inherent limitations: US may have blind spots, and small hepatic cysts—particularly those less than 1.5 cm—can easily be overlooked, potentially leading to underestimation of the true prevalence and incidence.45 Lastly, we used a uniform alcohol consumption threshold (over 140 g/week) for both sexes due to data limitations, whereas sex-specific World Health Organization thresholds (over 140 g/week for women; over 210 g/week for men) would be more appropriate.46 Despite these limitations, our study is the first to investigate the incidence of hepatic cysts in a large-scale, retrospective cohort over a 10-year period. Our methodology of using person-years for estimating incidence allowed the identification of key risk factors associated with hepatic cyst development. This study provides important epidemiological insights into hepatic cysts. In conclusion, the focus of our study was on the prevalence of, incidence of, and risk factors for hepatic cysts. Since there are only a few studies on the incidence of and risk factors for hepatic cysts, we hope that our findings will provide basic scientific evidence and a better understanding of the epidemiology, characteristics, and distribution of hepatic cysts. ACKNOWLEDGMENTS The authors would like to thank Hyeryun Park, from the interdisciplinary program for bioengineering, Seoul National University, for her assistance with natural language processing, and Tae-hoon Kim and Kun-hee Oh, both of whom are preventive medicine residents from Seoul National University, School of Medicine, for their assistance with manual labeling of the training and test sets of ultrasound reports. Footnotes Funding: This study was supported by the Korean Association of Hepato-Biliary-Pancreatic Surgery (2024). Disclosure: The authors have no potential conflicts of interest to disclose. Author Contributions: Conceptualization: Shin A, Choe EK, Yi NJ.Data curation: Yoon JH, Chung GE, Lee JM, Rho SY, Park MS.Formal analysis: Seo K, Park SH, Shin A.Methodology: Shin A, Choe EK.Writing - original draft: Seo K, Kim Y.Writing - review & editing: Seo K, Kim Y, Park SH, Yi NJ, Yoon JH, Chung GE, Lee JM, Rho SY, Park MS, Choe EK, Shin A. Conceptualization: Shin A, Choe EK, Yi NJ. Conceptualization: Shin A, Choe EK, Yi NJ. Data curation: Yoon JH, Chung GE, Lee JM, Rho SY, Park MS. Data curation: Yoon JH, Chung GE, Lee JM, Rho SY, Park MS. Formal analysis: Seo K, Park SH, Shin A. Formal analysis: Seo K, Park SH, Shin A. Methodology: Shin A, Choe EK. Methodology: Shin A, Choe EK. Writing - original draft: Seo K, Kim Y. Writing - original draft: Seo K, Kim Y. Writing - review & editing: Seo K, Kim Y, Park SH, Yi NJ, Yoon JH, Chung GE, Lee JM, Rho SY, Park MS, Choe EK, Shin A. Writing - review & editing: Seo K, Kim Y, Park SH, Yi NJ, Yoon JH, Chung GE, Lee JM, Rho SY, Park MS, Choe EK, Shin A. SUPPLEMENTARY MATERIALS Supplementary Data 1 Rule-based algorithm for defining hepatic and renal cysts from abdominal ultrasound reports Supplementary Table 1 Comparison of H-PEACE cohort to 2009 Korea National Health and Nutrition Examination Survey References Associated Data Supplementary Materials Supplementary Data 1 Rule-based algorithm for defining hepatic and renal cysts from abdominal ultrasound reports Supplementary Table 1 Comparison of H-PEACE cohort to 2009 Korea National Health and Nutrition Examination Survey