Introduction
Diabetes mellitus (DM) is a syndrome characterized by hyperglycaemia, which leads to damage to the body’s tissues. Diabetes is also one of the world’s most common diseases, and it exerts severe healthcare burdens due to various associated complications such as nephropathy, neuropathy, retinopathy, peripheral vascular disease, and ischaemic heart disease. Estimates show that around 382 million individuals suffered from DM globally in 2013, and the numbers are expected to increase every year, reaching 592 million patients in 2035 [1]. The incidence of the 2 main types of diabetes, type 1 (T1DM) and type 2 (T2DM), is 15% and 85%, respectively [1]. T1DM results from insulin deficiency, while T2DM results from insulin resistance. In addition, current evidence suggests that DM is associated with the development of gallstones, which was found attributable to impaired gallbladder emptying and decreased bile salt secretion from the gallbladder [2–6].
Gallstones are crystalline deposits that are observed within the cavity of the gallbladder in some patients. Cholelithiasis, the disease of gallstones, affects up to 25% of adults globally [7, 8]. It has been reported that cholelithiasis is the most common gastrointestinal disorder, frequently observed within ambulatory clinics. Diagnosis can be quickly established using abdominal ultrasonography, reported with a sensitivity rate of 90% [9]. Patients with cholelithiasis can develop severe and even life-threatening complications from the disease, including acute pancreatitis, acute cholecystitis, and obstructive jaundice with varying incidence rates [10–13].
Acute cholangitis or acute pancreatitis is the most deadly, with an estimated mortality rate of 20% for patients suffering their first episode [14]. In addition, surgery is required in up to 2% of the cases per year, representing a heavy burden on healthcare resources associated with managing this disease [15]. Investigators have identified multiple risk factors associated with the development of cholelithiasis, such as increased body mass index, reduced physical activity, high triglycerides, hyperinsulinaemia, and even insulin resistance, which may be associated with an increased risk of cholelithiasis [16–21]. In this context, evidence regarding the association between DM and cholelithiasis is still inconsistent among studies, probably indicating that the risk of developing cholelithiasis is multifactorial [22–26]. Therefore, risk factors for developing gallstones must continue to be identified and eliminated to reduce healthcare burdens on patients and the healthcare system.
Aim
Our study aims to evaluate the association between diabetes mellitus and gallstones and common risk factors in a nationwide population-based cohort investigation.
Material and methods
Data sources
The data reported from the National Health and Nutrition Examination Survey (NHANES) database from 2017 to 2018 were extracted for this study. Administrators of the database collected the data, and each patient’s data was reported for usage for research purposes without identifying variables.
Study variables
Demographic variables were extracted, including sex, age, marital status, educational level, and race. Moreover, the outcomes of interest were collected, including diabetes diagnosis, insulin use, age at diabetes mellitus diagnosis, cholelithiasis diagnosis, age at cholelithiasis diagnosis, and need for surgery due to cholelithiasis. Comorbidities such as heart failure, stroke, coronary artery disease, angina pectoris, heart attack, and cancer were also analysed.
All patients who had a known diagnosis of diabetes mellitus and cholelithiasis were included. Those with missing data, borderline diabetes, and individuals who gave a “do not know” answer to the outcomes of interest were excluded.
Statistical analysis
R software version 4.1.0 was used for the analysis of the data. All the categorical data were analysed using the χ2 test, while all the continuous data were analysed using the Mann-Whitney H test in accordance with the normality of the data distribution. We further performed a logistic regression analysis interpreted as odds ratio (OR) and 95% confidence interval (CI) to test the association between our outcomes of interest. The logistical regression was done using 2 models, one was unadjusted, and the other was adjusted for potential covariates, including age, gender, race, marital status, educational level, body mass index (BMI), and all reported comorbidities.
Results
Sociodemographic characteristics
A total of 5376 individuals were included in the final analysis, with a mean age of 51.3 ±17.8 years. Females constituted 51.5% of the included individuals, with an overall mean BMI of 29.8 ±7.4 kg/m2. Non-Hispanic White was the most common race with 34.9% of the included patients, followed by Non-Hispanic Black (23.1%), other races – including multi-racial (19.2%), and Mexican Americans (131.3%), respectively. Of the included patients, 49.1% were married and 32.1% had some college courses or associate degrees. There were significant differences between patients with reported cholelithiasis and those who were not in all reported sociodemographic characteristics (Table I). Similarly, there were significant differences between patients with diabetes and those who were not in all reported sociodemographic characteristics, except for race (Table II).
Table I
Table II
Patient characteristics
The prevalence of diabetes was 16.2% among the included individuals, with a mean age of 50.6 ±13.6 years at diagnosis of diabetes, and only 4.5% were taking insulin. The prevalence of cholelithiasis was 11.2%, with a mean age of 44.4 ±16.1 years at the time of diagnosis, and 11.3% had a previous cholecystectomy (gallbladder surgery.) For cardiovascular disease, heart failure was present in 3.6%, coronary artery disease was present in 4.8%, angina pectoris was present in 3.0%, and myocardial infarction was present in 4.9% of patients. Moreover, 4.9% of the included patients had a history of stroke, and 10.3% of patients reported a history of cancer/malignancy (Table III).
Table III
Association between diabetes and gallstones
There was a significant increase in gallstone rates among diabetic patients as compared to non-diabetics in the unadjusted (OR = 2.30; 95% CI: 1.89–2.79; p < 0.001) and adjusted (OR = 1.46; 95% CI: 1.12–1.89; p = 0.004) models. Moreover, this association was not time-dependent where the duration of diabetes did not show a significant influence on the gallstone rate, whether in unadjusted (OR = 0.99; 95% CI: 0.98–1.01; p = 0.194) or adjusted (OR = 1.01; 95% CI: 0.99–1.02; p = 0.285) models. Furthermore, age, BMI, female gender, black race, higher education, angina, stroke, and cancer were all significant independent predictors persisting in the multivariate model (Table IV).
Table IV
Discussion
Diabetes mellitus is one of the world’s most costly diseases and is responsible for significant comorbidity and mortality. Despite the rising incidence of DM globally, a noticeable reduction in its complications has been reported in the past decade [27]. Contemporary research has focused on the association between diabetes and the risk of complicated cholelithiasis. However, there is a heterogeneity of data in these studies, and the prevalence of cholelithiasis varies across populations, from 5% to 33% [28–31]. We provide the first and largest study for the US population, investigating all other factors that may affect the studies association.
Our study demonstrated a 46% increase in the odds of developing gallstones in patients with diabetes after the adjustment of all available covariates (OR = 1.52; 95% CI: 1.05–2.19). A similar observation was reported by the retrospective cohort of Chen et al. that reported a 55% increase in the incidence of gallstone formation in patients with type 2 diabetes mellitus. However, this study reported an inverse relationship between type 1 diabetes mellitus, where the odds of cholelithiasis were reduced by 52% compared to nondiabetic controls. The results of Chen et al. regarding type 1 diabetes mellitus were attributed to the younger age of patients with type 1 diabetes mellitus and the subsequent low prevalence of comorbidities compared to patients with type 2 diabetes mellitus, which were known risk factors for gallstone formation [29]. Furthermore, in an Italian population-based study, a higher prevalence of cholelithiasis was found in obese patients with type 2 diabetes mellitus compared to lean subjects with normal glucose tolerance, lean subjects with type 2 diabetes mellitus, and obese normal glucose tolerance groups, at 23%, 9%, 17%, and 14%, respectively [30].
In a case-control study in Iran, a significantly higher prevalence of gallstones was found in diabetic patients than in nondiabetics (33% vs. 17%). The risk was further elevated with the longer duration of diabetes mellitus (> 10 years) or with poor control with haemoglobin A1c (HbA1c) levels (8 and higher) [28]. This was supported by the fact that diabetes control can affect the incidence of cholelithiasis. A retrospective cohort study demonstrated a significant reduction in gallstone formation among diabetic patients with long-term use of metformin [32]. In addition, Liu et al. observed that around 10% of diabetic patients had symptomatic cholelithiasis requiring surgical intervention compared to 8% of the nondiabetic patients. This risk was further increased in diabetic females than in diabetic males [33].
Insulin is the primary therapeutic modality for uncontrolled type 2 diabetes and type 1 diabetes [34]. In animal models of insulin resistance, hepatic insulin resistance was implicated in gallstone formation. In this model, insulin resistance increases the heterodimeric cholesterol transporters responsible for regulating cholesterol secretion. Furthermore, an enzyme involved in bile acid synthesis, Cyp7b1, which has a protective effect against gallstone formation, was significantly downregulated [35]. Our analysis demonstrates that patients currently on insulin therapy had a higher odds ratio of developing cholelithiasis. A similar observation was reported in an Italian case-control study, in which participants with high insulin levels were associated with a higher prevalence of gallstones [36].
Diabetes is associated with comorbidities, particularly obesity and hypercholesterolaemia, which are known risk factors for cholelithiasis in previously predisposed individuals [37, 38]. Diabetes affects the biliary tract via microvascular and neurologic complications, potentiating a reduction in gallbladder (GB) blood supply, reducing GB emptying, raising the risk of infection, and increasing the concentration of bile, which may lead to infection and the development of cholelithiasis [39]. Diabetes has also been implicated in alterations of gastrointestinal hormones that affect gallbladder emptying. This may allow bile stasis, precipitating cholelithiasis [40]. Lastly, the dysfunction of the sphincter of Oddi in patients with DM may increase the risk of bile concentration and subsequent cholelithiasis [41].
This analysis suggests that insulin use may significantly predict cholelithiasis development in diabetic patients. One might conclude that these data support the initiation of non-insulin therapies as first-line, to reduce the risk of developing cholelithiasis. However, there is little evidence regarding the effect of these therapies on the development of cholelithiasis. Further research should be directed regarding the association of these drugs with cholelithiasis.
One of the study’s main strengths includes a large sample size of patients. Furthermore, our results were consistent in both the adjusted and unadjusted analyses. We also investigated whether time dependency and insulin intake are associated with gallstones, which are essential parameters that can affect the development of any diabetic complication and were not adequately reported by previous investigations. We also acknowledge several limitations that should be considered. Variables from the NHANES database are self-reported and subject to recall bias that cannot be avoided. The data regarding the type of diabetes mellitus (type 1 and type 2) and the duration of diabetes were not reported for all patients in our study. Finally, very few patients reported the current levels of HbA1c; therefore, the effect of diabetes control on cholelithiasis could not be examined.
Conclusions
There is an association between developing cholelithiasis and diabetes. Moreover, patients on insulin therapy showed greater odds of developing gallstones than patients who did not. Further investigation is needed to discover the role of insulin therapy, its duration, and dose in the development of gallstones.