|Year : 2017 | Volume
| Issue : 1 | Page : 12-17
Prevalence of glucose intolerance in cirrhotics and risk factors predicting its progression to diabetes mellitus
Pavan Hanchanale, Jayanthi Venkataraman, Mayank Jain, Vijaya Srinivasan, Joy Varghese
Institute of Gastro Sciences and Institute for Liver Disease and Transplant, Gleneagles Global Health City, Chennai, Tamil Nadu, India
|Date of Web Publication||9-May-2017|
Institute of Gastro Sciences and Institute for Liver Disease and Transplant, Gleneagles Global Health City, Chennai - 600 100, Tamil Nadu
Source of Support: None, Conflict of Interest: None
Background: Liver cirrhosis is a unique clinical entity wherein long-standing diabetes mellitus (DM) can predispose an individual to cirrhosis, and on the other hand, a long-standing cirrhosis liver can predispose a cirrhotic to DM, secondary to deranged glucose and insulin metabolism. Aim of the Study: The aim of this study is to determine the prevalence of impaired glucose tolerance (IGT) in patients with liver cirrhosis and risk factors predicting these patients to progression to DM. Materials and Methods: Adult patients above 18 years of age, with a diagnosis of cirrhosis of liver, irrespective of the diabetic status, the aetiology and severity of liver disease, were enrolled for the prospective cross-sectional study. The American Diabetes Association guidelines were followed for classification of DM. Fasting serum insulin assay was done for calculation of homeostatic model assessment insulin resistance (HOMA-IR) with a cut-off of >2.5. Patients with normal baseline glucose tolerance test (GTT) or IGT were followed up prospectively for 6 months at 3 monthly interval. Appropriate statistical tests were applied. Results: Of the 505 cirrhotic patients screened, 158 patients fulfilled the selection criteria. The overall mean age of these 158 patients was 53.3 ± 11.5 years; majority were men (93.7%). As a significant proportion of patients belonged to Child-Turcotte-Pugh (CTP)-C (51.3%) followed by CTP-B and CTP-A. Seventy-four patients had self-reported diabetes (46.8%) at registration. Post-GTT, the prevalence of impaired IGT was 27.4%, and new-onset diabetes mellitus (NODM) was 11.9%. Patients with CTP-C were at greatest risk for either IGT or NODM (P < 0.05). IGT+NODM patients had high median values of 2-h GTT, fasting serum insulin and HOMA-IR compared to non-diabetic participants at admission with a significant trend increase at 3 and 6 months. Conclusion: The prevalence of glucose intolerance is significantly high in individuals with cirrhosis. A 2-h GGT unmasks a significant number of cirrhosis with normal fasting blood sugar and glycated haemoglobin. A trend towards an increase in glucose intolerance and insulin resistance is noted with increasing duration of illness.
Keywords: Cirrhotics, diabetes, glucose intolerance
|How to cite this article:|
Hanchanale P, Venkataraman J, Jain M, Srinivasan V, Varghese J. Prevalence of glucose intolerance in cirrhotics and risk factors predicting its progression to diabetes mellitus. J Diabetol 2017;8:12-7
|How to cite this URL:|
Hanchanale P, Venkataraman J, Jain M, Srinivasan V, Varghese J. Prevalence of glucose intolerance in cirrhotics and risk factors predicting its progression to diabetes mellitus. J Diabetol [serial online] 2017 [cited 2019 Jul 22];8:12-7. Available from: http://www.journalofdiabetology.org/text.asp?2017/8/1/12/205984
| Introduction|| |
Diabetes mellitus (DM) is a complex metabolic clinical entity that is associated with high risk for micro- and macro-vascular complications, a poor quality of life and reduced life expectancy. Worldwide, there are at least over 170 million people suffering from DM, and this is likely to touch almost 266 million by 2030. In India, the prevalence is approximately 10% i.e., 31 million people are likely to be diabetic and these figures are likely to cross 80 million by the year 2030. Studies have shown that Indians are predisposed to DM at a much younger age with a greater risk for DM-related complications on a background of less degree of obesity.
Liver in diabetics is affected by either simple fat infiltration i.e., fatty liver or is associated with inflammation (steatohepatitis), cirrhosis liver and even hepatocellular carcinoma (HCC).
There is ample epidemiological evidence to suggest that diabetics are at a greater risk for chronic liver disease, cirrhosis and HCC and vice versa i.e. chronic liver disease such as cirrhosis can predispose an individual to a diabetic state , which is associated with high mortality.
Patients with liver cirrhosis have an impaired glucose tolerance (IGT) in more than 80% of patients. Overt diabetes exists in approximately 20% of cirrhotics. This may be related to insulin resistance that exists in the periphery, in the liver and decreased β-cell function.
The aim of the present study was to determine the prevalence of IGT with cirrhosis of the liver and to ascertain the risk factors predisposing these patients to overt diabetes.
| Materials and Methods|| |
Adult patients above 18 years of age, with a diagnosis of cirrhosis of liver (for at least 6 months), registered in the liver clinic at the Global Health City, Chennai, a tertiary care referral centre between 1st June 2014 and 30 May 2015 were included in the prospective cross-sectional study.
For the study, patients with liver cirrhosis with or without diabetes, irrespective of aetiology, Child-Turcotte-Pugh (CTP) or model for end-stage liver disease (MELD) scores, were included in the study. Patients who consented for the study alone were included in the study. The American Diabetes Association (ADA) guidelines were followed for assessing the diabetic state in these patients.
Overseas patients, individuals below the age of 18 years, post-liver transplant, non-diabetic cirrhotics in Grade III or more Hepatic Encephalopathy (HE), cirrhotics on steroids and hepatocellular carcinoma were excluded from the study.
Diagnosis of cirrhosis was based on standard clinical, laboratory and/or radiological findings (ultrasound or computed tomography). Demographic information included age, gender, co-morbidity including diabetes (and its duration), aetiology, CTP and MELD score. Investigations at registration included complete blood count, fasting plasma glucose, glycated haemoglobin (HbA1C), liver function tests, coagulation profile, thyroid function tests and lipid profile and workup for aetiology [Figure 1]. Liver biopsy was considered as optional.
Cirrhotic patients with fasting plasma glucose <100 mg/dL were considered as non-DM, those >126 mg/dL with HbA1C >6.5% as diabetic and the rest with plasma glucose between 100 and 125 mg/dL as pre-diabetes (impaired fasting glucose). Two-hour glucose tolerance test (GTT) was done for non- and pre-diabetic patients. A patient with a 2-h GTT blood glucose level <139 mg/dL was defined as non-diabetic. When the value exceeded 200 mg/dL, the patient was categorised as a diabetic, and when 2-h blood sugar value ranged between 140 and 199 mg/dL, the individual was classified as pre-DM (IGT).
Fasting serum insulin assay was done for calculation of homeostatic model assessment insulin resistance (HOMA-IR) with a cut-off of >2.5. Patients with normal baseline GTT or IGT were followed up prospectively for 6 months at 3 monthly interval with blood sugar, HbA1C, 2-h GTT (as and when appropriate) and serum insulin assay.
Ethics Committee of the institution approved the study. Ethical principles as dictated by the Declaration of Helsinki, which provides guidance to physicians and other participants in medical research involving human participants, were strictly followed. Informed consent was obtained from all the patients pertaining to the details of the study and confidentiality of the data.
Calculation of sample size
The study required 250 cases with the assumption that DM was likely to be prevalent in 15% of cirrhotic patients with a confidence interval (CI) of 95%, a power of 80% and an acceptable risk ratio of 1.5.
Data were analysed using the IBM-SPSS version 20.0 application. Chi-square test or proportion tests were used for comparison of percentages, ANOVA test for comparison of means and Kruskal–Wallis/Mann–Whitney test for medians. Risk computation was done for factors that emerged as significant by one-way or two-way analysis. Relative risk (RR) with 95% CI was used to compute risk rates and ratios. Linear regression analysis and logistic regression analysis were done. P < 0.05 was considered significant.
Prevalence rates and their 95% CI were computed. Univariate and multivariate analysis was done. Risk computation was done for factors that emerged as significant on one-way or two-way analysis.
Linear regression analysis was done for regressing numerical variables such as fasting insulin with GTT. Logistic regression analysis was done for variables with RR of >1.5. A RR of >1.5 with the lower boundary of 95% CI that did not include 1 or below and was considered significant.
| Results|| |
Of the 505 cirrhotic patients screened during the study period, 158 patients fulfilled the selection criteria. The overall mean age of these 158 patients was 53.3 ± 11.5 years; majority were men (93.7%). Approximately 50% (81 patients) of patients belonged to CTP-C followed by CTP-B (58 patients; 36.7%) and CTP-A (19 patients; 12%). Median MELD at registration was 16 (5–41), and the median duration of illness was approximately 2 years (25.5 months; 6–120).
The flow chart [Figure 2] shows the diabetic status of the 158 patients at registration. Seventy-four patients were known diabetic patients (46.8%) and were on treatment for more than 5 years. Of the remaining 84 patients, post-GTT, 51 (60.7%) patients remained non-diabetic, 23 patients i.e., 27.4% (95% CI 23.1 to 31.7%) had IGT and 10 were new-onset diabetes mellitus (NODM) i.e., in 11.9% patients (95% CI 9.5 to 14.2%), and the difference between the two was statistically significant (P < 0.01) [Figure 2]. The 2-h median GTT and range for IGT and NODM were 165 (140–198) and 224 (211–274) mg/dL, respectively. As the differences in the median levels were large, the two subgroups of IGT and NODM were combined for further analysis (IGT+NODM) [Figure 2].
|Figure 2: Prevalence of impaired glucose tolerance and newly diagnosed diabetes in cirrhosis patients|
Click here to view
Comparing the demography of three groups of patients i.e., DM, (74) no DM (51) and IGT+NODM (33), majority of patients were men and above 45 years (P < 0.001). There was a strong family history of diabetes (P < 0.0001) in known diabetics. There was no significant difference in aetiology, MELD and duration of cirrhosis between the three groups. The CTP score was, however, significantly higher in diabetic and newly diagnosed IGT+NODM patients compared to non-diabetic participants (P < 0.05) [Table 1].
|Table 1: Comparative demography and clinical data in patients with diabetes, no diabetes and combined impaired glucose tolerance and new-onset diabetes mellitus|
Click here to view
Between non-diabetic and IGT+NODM, there was no difference in age, aetiology, duration of illness and MELD between the two groups. Patients with CTP-C were at greatest risk for either IGT or NODM i.e., two times (RR 2.0; 95% CI 1.1–3.6) compared to CTP-A and CTP-B and this was statistically significant (P < 0.05). This difference was largely contributed by raised serum bilirubin (RR 1.4; 95% CI 1–2) and low serum albumin (RR 1.4) in CTP-C patients [Table 2]. The HbA1C, fasting insulin and HOMA-IR were significantly high in IGT+NODM compared to non-diabetics (P < 0.0001 for all the three parameters).
|Table 2: Comparative demography between non-diabetics and combined impaired glucose tolerance and new-onset diabetes mellitus group of cirrhosis patients|
Click here to view
Patients with IGT+NODM had high median values of 2-h GTT, fasting serum insulin and HOMA-IR compared to non-diabetic participants at admission with a significant trend increase at 3 and 6 months [Table 3] and [Figure 3].
|Table 3: Comparison of changing trend in 2-h glucose tolerance test, fasting insulin and homeostasis model assessment of insulin resistance levels|
Click here to view
|Figure 3: Changing trend in glucose tolerance, fasting insulin and homeostatic model assessment|
Click here to view
| Discussion|| |
Bohan  for the first time described in 1947, an association between DM and liver cirrhosis. This association was labelled as hepatogenous diabetes by Megyesi et al., and 57% of the cirrhotic patients showed insulin resistance. The terminology of hepatogenous diabetes as a distinct entity has not yet been accepted by the WHO or the ADA. Our study which included 158 patients of liver cirrhosis, the overall prevalence of IGT and NODM was 39.2% (33 of 84) of whom 27.4% (23 patients) had an IGT and 11.9% (10 patients) had NODM. Thus, IGT was significant in cirrhotic patients (P < 0.01). Perme et al. found the prevalence of DM to be 12.9% among patients with chronic liver disease with higher rates of complications in these patients. Garcia-Compean et al. in a study of 100 cirrhotic patients reported a normal GTT in 20%, IGT in 29.3% and NODM in 17.3%.
The major risk factor for IGT and NODM in our series was CTP Grade C, and the risk for acquiring this in liver cirrhosis was two times in Child C compared to Child A and B. None of the aetiological factors in our series predisposed to this though Zein et al. reported alcohol and HCV to predispose cirrhotic patients to HD. The observations in our series may be referral biased as majority of our patients had alcohol-related cirrhosis and belonged to Child C. Despite this, the most common cause of cirrhosis at our centre is alcohol related, followed by hepatitis B virus, hepatitis C virus (HCV) and non-alcoholic fatty liver disease (NAFLD). Mohan et al. reported the prevalence of NAFLD in an urban South Indian population and correlated this with different grades of glucose intolerance and metabolic syndrome. The prevalence of cirrhosis in these patients is not known.
During follow-up of 23 patients with IGT and 51 patients with normal GTT, 7 patients in the former and one patient in the latter group progressed to NODM at the end of 6 months. Other studies , have reported similar observations i.e., with increasing duration of illness.
DM in patients with liver cirrhosis may be subclinical and therefore a GTT is essential to detect impaired glucose metabolism in these patients. Fasting blood sugar and HbA1C are likely to less informative as cirrhotic patients have a reduced red blood cell liver span and one may get false low HbA1C levels  and reduce the rates of detection of both IGT and NODM. In the present study, 47.8% of cirrhotic patients whose fasting glucose level and HbA1c were normal at GTT had had pre-diabetes. In a recent study by Nishida et al. involving cirrhotic patients with normal fasting blood glucose levels and without a family history of type 2 DM, 77% had DM or IGT on GTT.
HOMA-IR has been reported to parallel the degree of liver fibrosis and has been typically described in HCV-infected patients.
The coefficient correlation between HbA1c and HOMA-IR was poor in all cirrhotic patients. This observation is likely to be the result of insufficient sample size. A trend towards rising 2-h GTT blood glucose levels and serum insulin was seen across the patients with IGT and NODM compared to non-diabetic patients (P < 0.05) both at 3 and 6 months. The median fasting insulin, HOMA-IR and HbA1c in IGT and NODM patients in the present study were 14.4 μIU/ml, 3.3 and 5.7%, respectively, that was significantly higher compared to non-diabetic patients. HOMA-IR was conspicuously higher in the IGT and NDM patients. Goswami et al. made similar observations i.e. a significantly high HOMA-IR in recently diagnosed diabetes in cirrhosis patients compared to those without DM i.e. 4.9 ± 1.9 (P < 0.0001). Insulin resistance is common in euglycemic cirrhosis and with advancement of liver disease; there is a compensatory increase in pancreatic β-cell insulin secretion to overcome the insulin resistance. However, despite this over a period of time with a fall in β-cell function, hepatogenous diabetes is likely to set in.
Our study has several limitations. The calculated sample size of 250 could not be reached despite the large cohort at registration because of multiple reasons such as non-compliance for follow-up, likelihood of death during the study period or proceeding for liver transplant. Drugs such as propranolol are known to cause glucose intolerance. We believe that since all are patients were on propranolol for secondary prophylaxis, it is less likely to be a confounding factor to influence blood sugar levels. In patients with IGT and NODM, the use of insulin and oral drugs and cirrhosis-related complications could have altered the progression to DM. This aspect was not looked into in this study and needs to be addressed in future studies. Finally, as the centre is a referral for liver transplant, the high prevalence of CTP-C among our patients could have affected the outcome due to this referral bias. Future multicentre studies are required need to address some of the limitations of this study with larger sample size.
In summary, patients with liver cirrhosis are predisposed to IGT and NODM which could alter the natural history of cirrhosis and also could be a major detrimental factor that could influence post-liver transplant outcome.
| Conclusion|| |
Patients with liver cirrhosis have a high prevalence of glucose intolerance, which is best unmasked by a 2-hr GTT in cirrhotics with normal blood sugar and glycated hemoglobin values. With increasing duration of illness there is a trend towards an increase in glucose intolerance as well as insulin resistance.
We thank Mr. Tom Michael, PhD scholar, Madras University, for technical assistance and data analysis.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Wild S, Roglic G, Green A, Sicree R, King H. Global prevalence of diabetes: Estimates for the year 2000 and projections for 2030. Diabetes Care 2004;27:1047-53.
Mohan V. Why are Indians more prone to diabetes? J Assoc Physicians India 2004;52:468-74.
Matteoni CA, Younossi ZM, Gramlich T, Boparai N, Liu YC, McCullough AJ. Nonalcoholic fatty liver disease: A spectrum of clinical and pathological severity. Gastroenterology 1999;116:1413-9.
Zein NN, Abdulkarim AS, Wiesner RH, Egan KS, Persing DH. Prevalence of diabetes mellitus in patients with end-stage liver cirrhosis due to hepatitis C, alcohol, or cholestatic disease. J Hepatol 2000;32:209-17.
Mason AL, Lau JY, Hoang N, Qian K, Alexander GJ, Xu L, et al.
Association of diabetes mellitus and chronic hepatitis C virus infection. Hepatology 1999;29:328-33.
Bianchi G, Marchesini G, Zoli M, Bugianesi E, Fabbri A, Pisi E. Prognostic significance of diabetes in patients with cirrhosis. Hepatology 1994;20(1 Pt 1):119-25.
Knobler H, Schihmanter R, Zifroni A, Fenakel G, Schattner A. Increased risk of type 2 diabetes in noncirrhotic patients with chronic hepatitis C virus infection. Mayo Clin Proc 2000;75:355-9.
Goswami A, Bhargava N, Dadhich S, Kulamarva G. Insulin resistance in euglycemic cirrhosis. Ann Gastroenterol 2014;27:237-43.
American Diabetes Association. Classification and diagnosis of diabetes. Diabetes Care 2017;40 Suppl 1:S11-24.
Bohan EM. Diabetes mellitus and cirrhosis of the liver; a case report. Del Med J 1947;19:212-5.
Megyesi C, Samols E, Marks V. Glucose tolerance and diabetes in chronic liver disease. Lancet 1967;2:1051-6.
Perme O, Singh YI, Singh KR, Devi BS, Rao A, Singh SK. Prevalence of diabetes in chronic liver disease patients admitted in medicine ward in RIIMS Hospital, Imphal. J Med Soc 2016;30:84-8. [Full text]
Garcia-Compean D, Jaquez-Quintana JO, Gonzalez-Gonzalez JA, Maldonado-Garza H. Liver cirrhosis and diabetes: Risk factors, pathophysiology, clinical implications and management. World J Gastroenterol 2009;15:280-8.
Mohan V, Farooq S, Deepa M, Ravikumar R, Pitchumoni CS. Prevalence of non-alcoholic fatty liver disease in urban South Indians in relation to different grades of glucose intolerance and metabolic syndrome. Diabetes Res Clin Pract 2009;84:84-91.
García-Compean D, Jaquez-Quintana JO, Maldonado-Garza H. Hepatogenous diabetes. Current views of an ancient problem. Ann Hepatol 2009;8:13-20.
Nishida T, Tsuji S, Tsujii M, Arimitsu S, Haruna Y, Imano E, et al.
Oral glucose tolerance test predicts prognosis of patients with liver cirrhosis. Am J Gastroenterol 2006;101:70-5.
Kawaguchi T, Taniguchi E, Itou M, Sakata M, Sumie S, Sata M. Insulin resistance and chronic liver disease. World J Hepatol 2011;3:99-107.
Petta S, Cammà C, Di Marco V, Macaluso FS, Maida M, Pizzolanti G, et al.
Hepatic steatosis and insulin resistance are associated with severe fibrosis in patients with chronic hepatitis caused by HBV or HCV infection. Liver Int 2011;31:507-15.
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2], [Table 3]