Influence of non-alcoholic fatty liver disease on the development of diabetes mellitus.

BACKGROUND AND AIM: Non-alcoholic fatty liver disease (NAFLD) is linked to metabolic syndrome, and is known to be associated with impaired fasting glycemia and diabetes mellitus. This prospective community-based study was conducted to determine the association between NAFLD and incidence of diabetes mellitus in an urban adult population in Sri Lanka. METHODS: Participants of the Ragama Health Study cohort were assessed for NAFLD using established ultrasound criteria in 2007. Those who were free of diabetes at baseline were followed up for 3 years. Incidence rates of diabetes mellitus were compared between subjects with and without NAFLD at baseline. RESULTS: Out of 2984 subjects, 926 had NAFLD and 676 had diabetes in 2007. Of the 2276 subjects who were free of diabetes in 2007, 1914 were re-assessed in 2010. After 3 years, 104 out of 528 subjects with NAFLD and 138 out of 1314 subjects without NAFLD had developed diabetes mellitus de novo. Incidence rates of diabetes were respectively 64.2 and 34 per 1000 person-years of follow up for those with and without NAFLD. NAFLD was an independent predictor of developing diabetes mellitus. Other independent predictors were impaired fasting glycemia and dyslipidemia. CONCLUSIONS: Subjects with ultrasonically diagnosed NAFLD have an increased risk of developing diabetes mellitus. Intervention for NAFLD through lifestyle modification may prevent progression of the current diabetes epidemic.

Clinically helpful rickettsial disease diagnostic IgG titers in relation to duration of illness in an endemic setting in Sri Lanka.

BACKGROUND: Although an initial IFA-IgG titer greater or equal to 1/64 or 1/128 is considered positive in presumptive diagnosis, in clinical practice in an endemic setting for rickettsioses in Sri Lanka, some patients with IFA-IgG titer of 1/128 for either spotted fever group (SFG) or scrub typhus (ST) did not respond to treatment. FINDINGS: To determine a clinically helpful diagnostic algorithm, IFA-IgG results of serologically confirmed treatment responders were analyzed in relation to duration of illness at sampling. Of 146 suspected SFG, 3 responders of 25 patients had titers ≤1/128 with < 7 days of illness while all 9 with titers ≥1/256 responded (false negative with 1/256 cutoff was 12%, false positive was 0%). For illness > 7 days, the false negative and positive rates were 4.3% (3/59) and 11.3% (6/53). Of 115 suspected ST, false negative and positive rates with ≥1/256 cutoff at <7 days of illness were 14.2% (2/14) and 0% (0/8) respectively while > 7 days, false negative and positive rates were 2% (1/51) and 0% (0/42). CONCLUSIONS: For clinical decision making, duration of illness at sampling is important in interpreting serology results in an endemic setting. If sample is obtained ≤7 day of illness, an IgG titer of ≤1/128 requires a follow up sample in the diagnosis and > 7 days of illness, a single ≥1/256 titer is diagnostic for all ST and 90% of SFG.

Survival of patients with alcoholic and cryptogenic cirrhosis without liver transplantation: a single center retrospective study.

BACKGROUND: There is no recent data addressing the long term survival of cirrhosis patients without transplantation, but with the availability of optimal pharmacological and endoscopic therapies. We compared the long term transplant free survival of alcoholic (AC) and cryptogenic (CC) cirrhosis patients in a setting where liver transplantation was, until very recently, not available. AC and CC patient details were extracted from our database, maintained since 1995. For those who had not attended clinics within the past 4 weeks, the patient or families were contacted to obtain survival status. If deceased, cause of death was ascertained from death certificates and patient records. Survival was compared using Kaplan-Meier curves. RESULTS: Complete details were available in 549/651 (84.3%) patients (AC 306, CC 243). Mean follow up duration (SD) (months) was 29.9 (32.6). 82/96 deaths (85.4%) among AC and 80/94 deaths (85.1%) among CC were liver related. Multivariate analysis showed age at diagnosis and Child's class predicted overall survival among all groups. The median survival in Child's class B and C were 53.5 and 25.3 months respectively. Survival was similar among AC and CC. Among AC survival was improved by abstinence [HR = 0.63 (95% CI: 0.40-1.00)] and was worse with diabetes [HR=1.59 (95% CI: 1.02- 2.48)] irrespective of alcohol status. CONCLUSIONS: The overall survival of AC was similar to CC. Death in both groups were predominantly liver related, and was predicated by age at diagnosis and Child class. Among AC, presence of diabetes and non-abstinence from alcohol were independent predictors for poor survival.