Abstract:

Background: Diabetes mellitus is a chronic metabolic disease which is characterized by hyperglycemia (high blood glucose), which in turn causes serious damage to the heart, blood vessels, eyes, kidneys, liver, and nerves over time and exists in more than two forms in which type one and two being the most common forms. Non-alcoholic fatty liver disease (NAFLD) is the presence of fat infiltration in the liver in the absence of excessive alcohol consumption. Type 2 diabetes mellitus (T2DM) is among the risk factors for developing non-alcoholic liver disease because insulin resistance (IR) induces hepatic de novo lipogenesis, influx of free fatty acids (FFA) into hepatocytes. Despite the significant high prevalence of NAFLD around the world, in Sub-Saharan Africa and Tanzania clinical management is not included in the guidelines. Furthermore, there are nodocumented data about the prevalence and predictors of NAFLD using non-invasive tests among diabetic patients attending a Zonal Hospital in Northwestern Tanzania.

Aim: To determine the prevalence and factors associated with NAFLD by non-invasive tests among adults with T2DM at Bugando Medical Center (BMC).

Methodology: Hospital-based cross-sectional study that was conducted at BMC's outpatient department from November 2023 to February 2024. During the study period 254 T2DM patients aged 18 and more were enrolled using a convenient sampling method at the BMC Outpatient Clinic. Structured questionnaires were used to collect demographic and clinical information from patients. Blood samples were taken to assess the fasting blood triglyceride, fasting blood glucose, hepatitis B surface antigen, hepatitis C antibodies, aspartate aminotransferase, alanine aminotransferase, and transient liver elastography was performed for each patient. Data were cleaned, entered, and analyzed by STATA Version 16. Data were summarized using measures of central tendency and dispersion for numeric variables and frequency and percentages for categorical variables. Multivariable logistic regression was used to determine independent factors associated with NAFLD and a p-value of 0.05 was considered statistically significant. A nonparametrical receiver operating characteristics (ROC) curve analysis without covariates was used to determine the optimal cut-off points for Tyg-I, Tyg-BMI and Tyg-WC for diagnosis of NAFLD. A two-by-two tables were formulated so as to calculate the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV).

Results: The prevalence of NAFLD was 62.6% (159/254). NAFLD was best predicted by obesity (BMI >29.9kg/m2 p-value <0.003), high waist circumference [ (>96cm); p-value 0.028] and uncontrolled blood pressure [(SBP>130 or DBP >80mmhg);p-value <0.001] in the multivariate logistic regression. The best cut offs for triglyceride index (Tyg-I) were 4.8 [AUC 0.525;sensitivity 66.66%, specificity 37.89%, Tyg-BMI 139.85[AUC 0.6192; sensitivity 62.89%, specificity 63.16%] and for Tyg-WC was 442.2 with AUC 0.676; sensitivity 67.30% specificity 52.63% .

Conclusion: Our study showed a higher prevalence of NAFLD among T2DM patients. Obesity, high waist circumference and uncontrolled blood pressure were associated with NAFLD. Other non-invasive test for detection of fatty liver showed a poor performance in terms of sensitivity, specificity, positive and negative predictive values as compared to transient elastography (TLE). We recommend all T2DM patient should be screened for NAFLD using TLE.