Cholecystectomy and risk of liver disease: a systematic review and meta-analysis of 27 million individuals

Background:

There is still a lack of knowledge on the association between cholecystectomy and liver disease. This study was conducted to summarize the available evidence on the association of cholecystectomy with liver disease and quantify the magnitude of the risk of liver disease after cholecystectomy.

Methods:

PubMed, Embase, Web of Science, and Cochrane Library were searched systematically from database inception to January 2023 to identify eligible studies that evaluated the association between cholecystectomy and the risk of liver disease. Meta-analysis was conducted to obtain a summary odds ratio (OR) and 95% confidence interval (CI) using a random-effects model.

Results:

We identified 20 studies with a total of 27 320 709 individuals and 282 670 liver disease cases. Cholecystectomy was associated with an increased risk of liver disease (OR: 1.63, 95% CI: 1.34–1.98). In particular, cholecystectomy was found to be significantly associated with a 54% increased risk of nonalcoholic fatty liver disease (OR: 1.54, 95% CI: 1.18–2.01), a 173% increased risk of cirrhosis (OR: 2.73, 95% CI: 1.81–4.12), and a 46% increased risk of primary liver cancer (OR: 1.46, 95% CI: 1.18–1.82).

Conclusions:

There is an association between cholecystectomy and the risk of liver disease. Our results suggest that strict surgical indications should be implemented to reduce unnecessary cholecystectomy. Additionally, the routine assessment of liver disease is necessary for patients with a history of cholecystectomy. More prospective large-sample studies are required for better estimates of the risk.

Introduction

Highlights

• This meta-analysis involving ~27 million individuals from various countries provided evidence that cholecystectomy was associated with a moderately increased risk of liver disease.

• Cholecystectomy was associated with an increased risk of nonalcoholic fatty liver disease.

• Cholecystectomy was associated with an increased risk of cirrhosis.

• Cholecystectomy was associated with an increased risk of primary liver cancer.

• Our findings suggest that strict surgical indications should be implemented to reduce unnecessary cholecystectomy. Additionally, clinicians should also inform patients with a history of cholecystectomy of liver surveillance to help with the early diagnosis of liver disease, enabling timely intervention.

Cholecystectomy is considered the standard treatment method for gallbladder disease1,2. Due to the increasing incidence of gallbladder disease and the introduction of laparoscopic cholecystectomy, the number of patients undergoing cholecystectomy has increased rapidly3. Although cholecystectomy is a common and simple procedure, concerns have been raised about the long-term health effects after cholecystectomy. Cholecystectomy is thought to affect surrounding tissues due to changes in bile flow, bile exposure, endocrine, and metabolic levels. Previous studies have identified an increased risk of metabolic syndrome and digestive system cancers after cholecystectomy4–7.

Primary liver cancer (PLC) is one of the most common cancers and the fifth leading cause of cancer-related deaths worldwide, and its incidence and mortality have been increasing8. Chronic liver disease (CLD), such as nonalcoholic fatty liver disease (NAFLD) and cirrhosis is the primary cause of PLC9. Additionally, CLD may result in death itself and is also a serious public health concern. Therefore, evaluating the association between cholecystectomy and the risk of PLC and CLD is clinically significant.

The association between cholecystectomy and liver disease remains poorly understood and there are conflicting conclusions among different studies. To date, no meta-analysis comprehensively has summarized the literature on the association between cholecystectomy and the risk of liver disease. Thus, we conducted the first systematic review and meta-analysis to determine if cholecystectomy would increase the risk of liver disease and quantify the magnitude of the association.

Methods

We conducted and reported this study following the Meta-analysis of Observational Studies in Epidemiology (MOOSE) guidelines10 and PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) Guidelines11, Supplemental Digital Content 1, http://links.lww.com/JS9/A219, Supplemental Digital Content 2, http://links.lww.com/JS9/A220. This study was registered with the International Prospective Register of Systematic Reviews (PROSPERO)12 with the registration number CRD42022352313.

Search strategy

The literature was searched in PubMed, Embase, Web of Science, and Cochrane Library using predefined search terms for eligible studies evaluating the association between cholecystectomy and the risk of liver disease. Searches were originally conducted in August 2022 and updated in January 2023. The main search terms included ‘cholecystectomy’, ‘nonalcoholic fatty liver disease’, ‘nonalcoholic steatohepatitis’, ‘metabolic associated fatty liver disease’, ‘dysfunction-associated fatty liver disease’, ‘cirrhosis’, ‘liver cancer’, ‘hepatocellular carcinoma’, and ‘intrahepatic cholangiocarcinoma’. To identify additional studies, references from the included studies were manually searched. An independent search was carried out by two investigators (Y.D. and M.-J.K.), Supplmental Digital Content 3, http://links.lww.com/JS9/A221.

Study selection

An eligible study must evaluate the association between cholecystectomy and liver disease. Effect estimates [odds ratio (OR), relative risk (RR), or hazard ratio (HR)] with a 95% confidence interval (CI) must be provided. We included the recently published study with the most comprehensive data analyses when several studies used the same database or cohort. Inclusion was not restricted by sample size. Two investigators (Y.D. and M.-J.K.) independently determined whether the studies were eligible. Disagreements were resolved by discussion.

Data extraction

The data were extracted and recorded in the predefined form independently by two investigators (Y.D. and M.-J.K.) to ensure accuracy. The following data were extracted: the first author’s name, publication year, study design, geographic location, sample size, population characteristics, the method used for liver disease diagnosis, length of follow-up for cohort studies, adjusted effect estimate with 95% CI, and covariate in the multivariable analysis.

Quality assessment

The risk of bias of included case–control and cohort studies was assessed independently by two investigators (Y.D. and F.K.) using the Newcastle–Ottawa scale13. The scale awards 4 points for participant selection, 3 points for exposure or outcome, and 2 points for comparability, giving a maximum total of 9 points. The quality of the study is classified as high (7–9 points), medium (4–6 points), or low (0–3 points). The risk of bias in cross-sectional studies was assessed based on 14 questions using the National Institutes of Health Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies14. There is no points system used to calculate an overall quality score. Instead, the overall judgment is based on the answers to each of the questions. There are no more than three questions answered ‘no’ in studies rated ‘good’, and there are less than three questions answered ‘yes’ in studies rated ‘poor’. Articles other than those are rated ‘fair’. The grading of recommendations assessment, development, and evaluation (GRADE) approach was used to determine the quality of the evidence15. Each result is assigned a certainty level (high, moderate, low, or very low) based on the risk of bias, publication bias, indirectness, imprecision, incoherence, inconsistency, intransitivity, large effect, plausible confounding, and dose–response gradient.

Statistical analysis

The effect estimate with 95% CI were extracted from each included study. We extracted the effect size reflecting the greatest degree of adjustment for potentially confounding factors when multiple effect sizes with different degrees of covariate adjustment were reported in a study. We then calculated an overall estimate of effect size using a random-effects meta-analysis based on the adjusted OR of all eligible studies. It was determined that heterogeneity among studies was low (25%), moderate (25–55%), and high (>75%) according to the I ² statistic16. For the sensitivity analysis, we removed one study at a time to observe the effects on the results. Stata version 16.0 was used to conduct the meta-analysis. The asymmetry was visualized using funnel plots in the case of more than 10 studies available, and Egger’s test was performed to assess publication bias.

Results

Literature search and characteristics of studies

Based on the title and abstract of 3220 selected citations (after excluding duplicates), we initially identified 33 potentially eligible studies from PubMed, Embase, Web of Science, and Cochrane Library. After examining the full text of these 33 potentially eligible studies, 13 studies were excluded because of the following reasons: no outcome of interest was reported (n=3), the risk of cancer was shown with a standardized incidence ratio (SIR) instead of OR, HR, or RR (n=4), or study populations overlapped (n=6). Finally, 20 studies17–36 were included in the meta-analysis consisting of 13 cohort studies17–20,22,24–26,28,29,32,33,36, 3 matched case–control studies30,31,34, and 4 cross-section studies21,23,27,35. The results of the literature research and study selection are detailed in Figure 1. The meta-analysis involved a total of 27 320 709 individuals, 882 675 of whom had a history of cholecystectomy. In the meta-analysis, 282 670 liver disease cases (NAFLD: 233 537, cirrhosis: 397, and PLC: 48 736) were included. Most included studies reported that individuals with liver disease or other causes of CLD at baseline were excluded. Half of the studies included in this meta-analysis were conducted in Asia (n = 10), whereas 8 studies were carried out in the United States, and two studies were carried out in Europe. Of the 20 studies, 7 studies18,21–24,27,33 evaluated the association between cholecystectomy and NAFLD, 3 studies evaluated cirrhosis26,29,35, and 11 studies17,19,20,25,28–32,34,36 evaluated PLC [3 studies17,29,30 evaluated the hepatocellular carcinoma (HCC), 3 studies17,31,34 evaluated the intrahepatic cholangiocarcinoma (ICC), and 6 studies19,20,25,28,32,36 did not distinguish HCC and ICC]. The majority of studies17–20,22,24–26,28–34,36 included in this meta-analysis were of high quality. These individuals had a median follow-up period of 11 years (4.9–26 years). The characteristics and quality assessment of included studies are described in Table 1.

Figure 1.

Selection of studies for inclusion.

Table 1.

Characteristics of the included studies.

Study	Region	Study design	Sample size	Liver disease/causes of liver disease at baseline	Follow-up (year)	Diagnosis of liver disease	No. of cases	Results (95% CI)	Confounder adjustment	Quality assessment
Ahn et al.17	Asia	Cohort study	10 545 447	No	15	ICD code	HCC: 27 283 ICC: 8265	HCC: HR 0.93 (0.87–0.99); ICC: HR 1.69 (1.49–1.92)	Age, gender, HBV, HCV, alcoholic liver disease, chronic hepatitis, liver fibrosis, cirrhosis, diabetes, acute cholangitis, and acute cholecystitis	8
Chang et al.18	Asia	Cohort study	219 641	No	6	Ultrasound	NAFLD: 49 301	HR 1.17 (1.03–1.33)	Age, gender, BMI, year of examination, education level, smoking, alcohol intake, physical activity, energy intake, hypertension, diabetes, and medication for dyslipidemia	8
Chen et al.19	Asia	Cohort study	77 725	No	11	ICD code	PLC: 541	HR 1.17 (0.90–1.52)	Age, gender, and comorbidities	8
Goldacre et al.20	America	Cohort study	374 067	No	NA	NA	PLC: 344	RR 1.45 (1.09–1.90)	NA	7
Kichloo et al.21	America	Cross-section study	14 294 784	No	NA	ICD code	NAFLD: 159 259	OR 1.97 (1.93–2.01)	Age, gender, race, alcohol intake, diabetes, dyslipidemia, hypertension, metabolic syndrome, smoking, and obesity	Fair
Konyn et al.22	America	Cohort study	11 153	No	23	Ultrasound	NAFLD: 3835	HR 2.77 (2.01–3.83)	Age, gender, race, BMI, smoking, diabetes, hypertension, TC, HDL-C, coffee consumption, alcohol intake, sedentary lifestyle, hormone replacement therapy, and energy intake	8
Kwak et al.23	Asia	Cross-section study	17 612	No	NA	Ultrasound	NAFLD: 5337	OR 1.35 (1.03–1.77)	Age, gender, hypertension, diabetes, BMI, smoking, TC, TG, and HDL-C	Fair
Latenstein et al.24	Europe	Cohort study	4307	No	10	Ultrasound	NAFLD: 1269	OR 1.04 (0.62–1.77)	Age, gender, education level, physical activity, energy intake, hypertension, diabetes, metabolic syndrome, and BMI	8
Liu et al.25	Asia	Cohort study	95 021	No	9.05	ICD code	PLC: 306	HR 5.25 (1.95–14.17)	Age, gender, BMI, ALT, fasting blood glucose, cirrhosis, HBV, NAFLD, metabolic syndrome, alcoholic liver disease, smoking, alcohol intake, hypertension, and physical activity	9
Ioannou26	America	Cohort study	9072	No	13.3	ICD code	Cirrhosis: 121	HR 2.1 (1.1–4.0)	Age, gender, race, BMI, subscapular-to-triceps skinfold ratio, alcohol intake, education level, the U.S. geographical region, diabetes, and coffee or tea consumption	7
Lu et al.27	Asia	Cross-section study	4325	No	NA	Ultrasound+pathology	NAFLD: 2084	OR 1.95 (1.55–2.45)	Age, gender, BMI, waist circumference, diabetes, SDP, DBP, glycosylated hemoglobin, TC, TG, and LDL-C	Fair
Luo et al.28	America	Cohort study	173 229	No	26	Medical or pathologic reports	PLC: 204	HR 1.33 (0.90–1.95)	Age, study period, cohort, race, aspirin use, smoking, energy intake, alcohol intake, coffee consumption, physical activity, BMI, hypercholesterolemia, and diabetes	9
Martin et al.29	America	Cohort study	395	Hepatitis C virus infection	4.9	Pathology	Cirrhosis: 142 HCC: 34	Cirrhosis: OR 2.85 (1.11–7.36) HCC: OR 1.34 (1.17–1.52)	BMI, alcohol intake, tobacco use, metabolic syndrome, and HIV	8
Nogueira et al.30	America	Matched case–control study	1 238 390	No (in control populations)	NA	ICD code	HCC: 8104 PLC: 10 219	HCC: OR 1.34 (1.17–1.52) PLC: OR 1.26 (1.12–1.41)	Age, gender, year of selection, and diabetes	9
Tao et al.31	Asia	Matched case–control study	441	No (in control populations)	NA	Pathology	ICC: 61	OR 3.6 (0.9–15.1)	Age, gender, and HBV	9
Vogtmann et al.32	Asia	Cohort study	134 546	No	NA	ICD code	PLC: 412	HR 1.38 (0.92–2.07)	Age, education level, smoking, alcohol intake, menopausal status (for women only), family history of liver cancer, BMI, physical activity, total energy intake, diabetes, hepatitis, and chronic liver disease	8
Wang et al.33	Asia	Cohort study	32 428	No	NA	Ultrasound	NAFLD: 12 452	OR 1.10 (0.94–1.28)	Age, gender, BMI, SBP, DBP, fasting plasma glucose, TC, TG, HDL-C, LDL-C, ALT, AST, GGT, albumin, and SUA	7
Welzel et al.34	Europe	Matched case–control study	3820	No (in control populations)	NA	Pathology+ICD code	ICC: 764	OR 1.56 (0.65–3.73)	Age at ICC diagnosis, gender, and year of birth	9
Xie et al.35	America	Cross-section study	4497	NA	NA	Transient elastography	Cirrhosis: 134	OR 3.29 (1.50–7.22)	Age, gender, race, education level, alcohol intake, diabetes, HBV, HCV, physical activity, serum cotinine levels, BMI, and income	Fair
Zhao et al.36	Asia	Cohort study	79 809	No	11	ICD code	PLC: 303	HR 2.81 (0.68–11.51)	Cirrhosis and HBsAg status	7

ALT, alanine aminotransferase; AST, aspartate aminotransferase; DBP, diastolic blood pressure; GGT, gamma-glutamyl transpeptidase; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; HDL, high-density lipoprotein; HDL-C, high-density lipoprotein cholesterol; ICC, intrahepatic cholangiocarcinoma; ICD, international classification of disease; LDL-C, low-density lipoprotein cholesterol; NA, not available; NAFLD, nonalcoholic fatty liver disease; PLC, primary liver cancer; SBP, systolic blood pressure; SUA, serum uric acid; TC, total cholesterol; TG, triglycerides.

Association between cholecystectomy and the risk of liver disease

Twenty studies17–36 provided data suitable for the pooled primary analysis, involving a total of 27 320 709 individuals with 282 670 liver disease cases. The cholecystectomy was associated with a moderately increased risk of liver disease (OR: 1.63, 95% CI: 1.34–1.98) with considerable heterogeneity (I ²=96.7%, P<0.0001) (Fig. 2). The sensitivity analysis showed that the removal of any individual study had little effect on the pooled OR, suggesting the result was stable (Supplement Figure 1, Supplemental Digital Content 4, http://links.lww.com/JS9/A222). The funnel plot demonstrated the distribution of studies on both sides was relatively symmetrical (Supplement Figure 2, Supplemental Digital Content 5, http://links.lww.com/JS9/A223). Meanwhile, Egger’s test for asymmetry (P=0.20) showed no evidence of publication bias. Subgroup analyses were also conducted by geographic location, study design, study quality, and follow-up duration. Table 2 shows the results of the subgroup meta-analyses. In general, the positive association between cholecystectomy and the risk of liver disease was consistent across all subgroups except for the subgroup meta-analysis conducted by two European studies (OR: 1.16, 95% CI: 0.74–1.82)24,34.

Figure 2.

Forest plot and pooled estimates of the effect of cholecystectomy on the risk of liver disease, stratified by types of disease. HCC: hepatocellular carcinoma; ICC: intrahepatic cholangiocarcinoma; NAFLD, nonalcoholic fatty liver disease; PLC: primary liver cancer.

Table 2.

Subgroup analysis of the association between cholecystectomy and risk of liver disease.

Subgroup	Pooled OR (95% CI)	No. of studies	No. of individuals included	I 2 (%)
Liver disease
Geographic location
Asia	1.41 (1.14–1.73)	10	11 206 995	91.1
America	1.90 (1.50–2.39)	8	16 105 587	89.1
Europe	1.16 (0.74–1.82)	2	8127	0.0
Study design
Cohort study	1.52 (1.25–1.84)	13	12 756 840	89.6
Case–control study	1.35 (1.00–1.82)	3	1 242 651	14.0
Cross-section study	1.85 (1.52–2.25)	4	14 321 218	67.0
Study quality
Low risk of bias	1.52 (1.25–1.86)	12	12 504 115	90.2
Medium risk of bias	1.68 (1.31–2.16)	8	14 816 594	89.6
Follow-up duration
<10 years	2.53 (1.14–5.64)	3	315 057	83.9
≥10 years	1.52 (1.06–2.17)	6	10 727 513	94.4
NAFLD
Geographic location
Asia	1.34 (1.06–1.68)	4	274 006	83.8
America	2.25 (1.62–3.11)	2	14 305 937	76.6
Europe	1.04 (0.62–1.76)	1	4307	NA
Study design
Cohort study	1.38 (0.99–1.93)	4	267 529	89.0
Case–control study	None	–	–	–
Cross-section study	1.79 (1.47–2.18)	3	14 316 721	73.2
Study quality
Low risk of bias	1.51 (0.82–2.78)	3	235 101	91.8
Medium risk of bias	1.55 (1.12–2.15)	4	14 349 149	95.1
Follow-up duration
<10 years	1.17 (1.03–1.33)	1	219 641	NA
≥10 years	1.74 (0.67–4.53)	2	15 460	89.7
Cirrhosis
Geographic location
Asia	None	–	–	–
America	2.73 (1.81–4.12)	3	13 964	0.0
Europe	None	–	–	–
Study design
Cohort study	2.54 (1.57–4.12)	2	9467	0.0
Case–control study	None	–	–	–
Cross-section study	3.29 (1.50–7.22)	1	4497	NA
Study quality
Low risk of bias	3.24 (1.57–6.68)	1	395	NA
Medium risk of bias	2.52 (1.53–4.14)	2	13 569	0.0
Follow-up duration
<10 years	3.24 (1.57–6.68)	1	9072	NA
≥10 years	2.1 (1.1–4.0)	1	395	NA
PLC
Pathological type
HCC	1.23 (0.86–1.77)	3	11 784 232	93.0
ICC	1.70 (1.50–1.92)	3	10 549 708	0.0
HCC and ICC combined	1.36 (1.15–1.60)	7	2 172 787	50.0
Geographic location
Asia	1.55 (1.08–2.23)	6	10 932 989	92.9
America	1.33 (1.16–1.52)	4	1 786 081	41.8
Europe	1.56 (0.65–3.74)	1	3820	NA
Study design
Cohort study	1.50 (1.13–2.00)	8	11 480 239	91.2
Case–control study	1.35 (1.00–1.82)	3	1 242 651	14.0
Cross-section study	None	–	–	–
Study quality
Low risk of bias	1.45 (1.14–1.84)	19	12 269 014	90.5
Medium risk of bias	1.49 (1.13–1.95)	2	453 876	0.0
Follow-up duration
<10 years	3.81 (1.92–7.56)	2	95 416	0.0
≥10 years	1.30 (0.85–1.98)	3	10 702 981	95.7

HCC, hepatocellular carcinoma; ICC, intrahepatic cholangiocarcinoma; NA, not available; NAFLD, nonalcoholic fatty liver disease; PLC, primary liver cancer.

Association between cholecystectomy and the risk of NAFLD

Seven studies18,21–24,27,33 provided data suitable for the pooled primary analysis, involving a total of 14 584 250 individuals with 233 537 NAFLD cases. The cholecystectomy was associated with a 54% increased risk of NAFLD (OR: 1.54, 95% CI: 1.18–2.01, I²=95.4%, P<0.0001) (Fig. 2). When stratifying by geographic location, the increased risk of NAFLD was prominent in the American population (OR: 2.25, 95% CI: 1.62–3.11, I²=76.6%, P<0.0001), but no significant association in the European population (OR: 1.04, 95% CI: 0.62–1.76). Results of subgroup meta-analyses also showed that cholecystectomy was significantly associated with an increased risk of NAFLD in most of the subgroups, while the association disappeared for studies that had a follow-up of over 10 years (two studies22,24, OR: 1.74, 95% CI: 0.67–4.53) (Table 2). The sensitivity analysis showed that the removal of any individual study had little effect on the pooled OR, suggesting the result was stable.

Association between cholecystectomy and the risk of cirrhosis

Three studies26,29,35 provided data suitable for the pooled primary analysis, involving a total of 13 964 individuals with 397 cirrhosis cases. The cholecystectomy was associated with a 173% increased risk of cirrhosis with no heterogeneity (OR: 2.73, 95% CI: 1.81–4.12, I²=0%, P=0.59) (Fig. 2). Furthermore, the results of all subgroup meta-analyses showed that cholecystectomy was associated with an increased risk of cirrhosis without significant heterogeneity (Table 2). The removal of a single study at a time also had little effect on the pooled OR, suggesting that the result remained stable.

Association between cholecystectomy and the risk of PLC

Eleven studies17,19,20,25,28–32,34,36 provided data suitable for the pooled primary analysis, involving a total of 12 722 890 individuals with 48 736 PLC cases. The cholecystectomy was associated with a 46% increased risk of PLC (OR: 1.46, 95% CI: 1.18–1.82, I²=88.9%, P<0.0001) (Fig. 2). There was a significantly increased risk of ICC (OR: 1.70, 95% CI: 1.50–1.92, I²=0%), but the association between cholecystectomy and HCC was not significant (OR: 1.23, 95% CI: 0.86–1.77, I²=93%). Results of subgroup meta-analyses also demonstrated that cholecystectomy was significantly associated with an increased risk of PLC, while the association disappeared for studies conducted in Europe (one study34, OR: 1.56, 95% CI: 0.65–3.74) and with a follow-up of over 10 years (three studies17,19,36, OR: 1.30, 95% CI: 0.85–1.98) (Table 2). The sensitivity analysis showed that the removal of any individual study had little effect on the pooled OR, suggesting the result was stable.

Grading the quality of evidence

GRADE determined that the overall quality of evidence was low since all the studies were observational (Table 3).

Table 3.

Grading of recommendations, assessment, development, and evaluation (GRADE) criteria for studies included in the meta-analysis evaluating the risk of liver disease after cholecystectomy.

			Quality assessment
Outcome	Number of participants	Study design	Risk of bias	Inconsistency	Indirectness	Imprecision	Publication bias	Other considerations (large magnitude of effect, plausible confounding, and dose–response gradient)	Overall quality of evidence
LD	27, 320, 709	Observational studies	Not serious	Serious	Serious	Not serious	Not serious	Very strong association	Low
NAFLD	14, 584, 250	Observational studies	Not serious	Serious	Serious	Not serious	Not serious	Very strong association	Low
Cirrhosis	13, 964	Observational studies	Not serious	Not serious	Serious	Not serious	Not serious	Very strong association	Low
PLC	12, 722, 890	Observational studies	Not serious	Serious	Serious	Not serious	Not serious	Very strong association	Low

LD, liver disease; NAFLD, nonalcoholic fatty liver disease; PLC, primary liver cancer.

Discussion

This meta-analysis of 20 observational studies (involving ∼27 million individuals from various countries and ∼280 000 cases of liver disease) provided evidence that cholecystectomy was associated with an increased risk of liver disease. The magnitude of this risk remained essentially unchanged when subgroup analyses stratified by geographic location, study design, study quality, and follow-up duration were conducted.

To date, several review articles have shown that cholecystectomy was associated with an increased risk of liver disease. But most of these studies only focused on the association between cholecystectomy and NAFLD. A recent article provided a detailed review of the positive association and pathophysiological relationship between cholecystectomy and NAFLD37. Another review article showed that cholecystectomy was a risk factor for metabolic syndrome and increased the prevalence of fatty liver5. In addition, the underlying mechanisms by which cholecystectomy might cause metabolic syndrome were also analyzed. In Qi et al.’s38 review article, they described the epidemiologic evidence that linked cholecystectomy to an increased risk of NAFLD and discussed the possible mechanisms behind these connections. In contrast to these review articles, we explored the effect of cholecystectomy on NAFLD, cirrhosis, and PLC. In addition, we used the method of meta-analysis to quantitatively analyze the included articles to quantify the magnitude of the risk of liver disease after cholecystectomy.

A few previous meta-analyses have been conducted to assess the association between cholecystectomy and the risk of NAFLD and liver cancer. For NAFLD, one previous meta-analysis reported that cholecystectomy was significantly associated with a 46% increased risk of NAFLD39. The previous meta-analysis included only four articles23,33,40,41. However, we found that one included study in the previous meta-analysis only reported the association between cholelithiasis and fatty liver40. So, this study should not be included in the meta-analysis. In the present study, we excluded this study40 and included seven eligible studies that evaluated the association between cholecystectomy and the risk of NAFLD. The result of the present study showed the association between cholecystectomy and an increased risk of NAFLD, which is consistent with previous clinical studies and reviews37,41,42. Two studies based on the National Survey of Health and Nutrition Examination database analysis demonstrated that cholecystectomy showed an increased risk for NAFLD42,43. The independent association between NAFLD and cholecystectomy was reported, but not between NAFLD and gallstones, indicating that cholecystectomy could be a risk factor for NAFLD. A cross-sectional study conducted at the Seoul University Hospital also confirmed the independent association between cholecystectomy and NAFLD, but not with gallstones23. This study supported the idea that cholecystectomy had some effect on NAFLD development. Another cross-sectional study demonstrated that NAFLD risk increased significantly in patients with a cholecystectomy compared to patients with gallstone disease who preserve the gallbladder, even after adjusting in the shared metabolic risk factors44. Therefore, this association was mainly attributable to a previous cholecystectomy, not to gallstones, establishing cholecystectomy as a risk factor for NAFLD. In contrast to these results from clinical studies, a recent study in the mouse model demonstrated that cholecystectomy increased NAFLD activity score but did not increase the incidence of NAFLD or aggravate preexisting NAFLD45. Several limitations were associated with this animal study. The number of mice in the sham group was only two at month 4 postsurgery, which may affect the statistical power. Furthermore, the observation period of 6 months after cholecystectomy may not be enough. More importantly, the effect of cholecystectomy on humans may be different from that on mice.

To date, no meta-analysis has been conducted to examine the association between cholecystectomy and cirrhosis. Three eligible studies that evaluated the association between cholecystectomy and the risk of cirrhosis were included in the present study. Consistent with the findings of the present study, a recent cross-sectional study based on data from the 2017 to 2018 National Health and Nutrition Examination Survey showed that cholecystectomy was positively associated with cirrhosis in U.S. adults35. In addition, the association remained statistically significant even after adjusting for possible confounders. For liver cancer, two meta-analyses were conducted to examine the association46,47. The meta-analysis by Liu et al.46 included 12 studies and demonstrated that individuals who had their gallbladder removed had a 62% greater risk of liver cancer. But the meta-analysis included 5 studies, among which 2 studies19,48 and 3 studies43,49,50 using the same database or cohort respectively. Furthermore, some studies included in the meta-analysis estimate the risk of liver cancer by using SIR43,49,50. The SIR is the ratio of observed to expected cases, which is inappropriately used to compare the risk of liver cancer in individuals with and without cholecystectomy. The meta-analysis by Wang et al.47 suggested that cholecystectomy was associated with a 47% increased risk of liver cancer. Similar to Liu et al.’s study, this meta-analysis also included two studies using SIR as effect size43,51. To date, there was no meta-analysis assessing the association between cholecystectomy and the risk of cirrhosis. So, we conducted this study to comprehensively assess the association between cholecystectomy and the risk of liver disease. In the present meta-analysis, we conducted a comprehensive literature search and included eligible studies strictly according to the predefined criteria. When multiple studies used the same database or cohort, we included only the most recently published study with the most comprehensive data analyses.

Several potential mechanisms of cholecystectomy may increase the risk of liver disease. As a link between the liver and the gut, the gallbladder maintains metabolic homeostasis of bile acid, cholesterol, and triglyceride52. Enterohepatic circulation of bile acids is altered after the removal of the gallbladder, which would affect the metabolism and transport of bile acids as well as other key aspects of metabolic homeostasis38,53,54. An animal study demonstrated that cholecystectomy increased liver fat accumulation, which led to NALFD and liver inflammation55. Ioannou’s26 study showed that individuals who have undergone cholecystectomy have higher levels of serum alanine transaminase and gamma-glutamyl transferase than those without cholecystectomy, suggesting the liver injury induced by cholecystectomy. Hepatic steatosis is considered the underlying mechanism linking cholecystectomy with hepatocyte damage, elevated serum liver enzymes, cirrhosis, and chronic inflammation. Chronic inflammation plays an important role in the development of many cancers, including liver cancer56–58. Persistent chronic inflammation can lead to liver fibrosis, cirrhosis, and eventually liver cancer.

The present study has some strengths. It is the first meta-analysis to assess the association between cholecystectomy and the risk of liver disease. In addition, a relatively large number of studies allowed us to conduct subgroup analyses by type of liver disease, geographic region, study design, study quality, and follow-up duration. The present meta-analysis included the most recent cohort, case–control, and cross-section studies and the largest number of study participants. In addition, most of the studies included in the analysis were adjusted for confounding factors such as age, gender, energy intake, alcohol intake, smoking, metabolic syndrome, and body mass index.

Despite these strengths, several potential limitations of our meta-analysis should be considered. First, the present study included observational studies, and therefore, potential methodological biases, including selection bias or recall bias, might be considered. Second, heterogeneity was found among the studies. To address this limitation, subgroup analyses were conducted. Third, as we only included studies published in English, we may have missed relevant non-English studies.

Conclusion

The results of the present study indicate that cholecystectomy is associated with an increased risk of liver disease. Our results suggest that strict surgical indications should be implemented to reduce unnecessary cholecystectomy. Additionally, the routine assessment of liver disease is necessary for patients with a history of cholecystectomy to help with the early diagnosis of liver disease, enabling timely intervention. However, more prospective large-sample studies are still needed to further verify the effect of cholecystectomy on liver disease. Moreover, mechanistic studies are also needed to better understand the link between cholecystectomy and liver disease.

Ethical approval

Ethical approval was not given because we conducted systematic reviews that did not collect personal information from patients.

Sources of funding

The study is supported by the National Natural Science Foundation of China (U22A2039).

Author contribution

D.L.: designed the study; Y.D. and M.-J.K.: performed the searches and data extraction; Y.D. and F.K.: conducted the risk of bias assessment; D.L. and F.K.: performed the statistical analysis; D.L. and X.-P.C.: drafted and edited the manuscript; B.L. and S.S.: provided oversight, critical evaluation, and verification of the manuscript. All authors approved the final version of the manuscript.

Conflicts of interest disclosure

There are no conflicts of interest.

Research registration unique identifying number (UIN)

1. Name of the registry: PROSPERO.

2. Unique identifying number or registration ID: CRD42022352313.

3. Hyperlink to your specific registration (must be publicly accessible and will be checked): https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID=352313

Guarantor

Bo Li and Song Su.

Data availability statement

No new data were generated during the current study. All data used for meta-analyses in the current study are from published articles. Data are available from these published articles or the original study authors.

Supplementary Material

js9-109-1420-s001.docx ^{(31.3KB, docx)}

js9-109-1420-s002.docx ^{(43.4KB, docx)}

js9-109-1420-s003.pdf ^{(392.6KB, pdf)}

js9-109-1420-s004.pdf ^{(2.2MB, pdf)}

js9-109-1420-s005.pdf ^{(60.5KB, pdf)}