Impact of Ramadan fasting on glycemic control in type 2 diabetic patients treated with insulin.
SEBAI I, MAHJOUB F, EL FELAH E, GAMMOUDI A, LAHMER I, BEN OTHMAN R, BERRICHE O, JAMOUSSI H.
Service A, National Institute of Nutrition, Tunis, TUNISIA

Background:
Ramadan fasting, the fourth pillar of Islam, is a duty for all healthy Muslim adults. Quran exempts the sick from fasting. Nevertheless, some diabetic adults insist on observing Ramadan despite the potential risks incurred, the prohibition of doctors and the tolerance of religion. 
Aim: 
The aim of the study was to evaluate the effect of Ramadan fasting on glycemic control in type 2 diabetic patients treated with insulin.
Methods:
We performed a prospective study among type 2 diabetic patients at the National Institute of Nutrition in Tunisia. Patients were recruited one month before Ramadan 2015 which took place from 18 June to 16 July. Type 2 diabetic patients treated with insulin for more than three months, who had decided to fast Ramadan and who gave oral consent to participate in the study were included. Recruited patients were provided with glucometers, educated about diet and taught how to perform self-monitoring of blood glucose at home and when to break their fast. Insulin doses were altered according to the recommendations of the American Diabetes Association ADA 2010. All patients were instructed to call us for dose adjustment whenever necessary or if there was any episode of hypoglycemia. Glycosylated hemoglobin HbA1c was determined before and after Ramadan.
Results:
A total of 45 patients were included in the study (sex ratio (H/F) = 1,8; median age =  57,4 years(range 38-73 years), mean body mass index (BMI)= 28.56 kg/m2 ± 3,93 and mean duration of diabetes = 11,7 ± 6,6 years ). Type 2 diabetic patients were treated with insulin since a median of 3,9 years (range 4 monthes-23 years). Among them, 23 patients (51%) were categorized such as at high risk of hypoglycemia and 22 were stratified into the very high risk category. About 42% had chosen to fast without telling their doctors and 29% had insisted on fasting against the advice of their doctors. Half of our population (51%) had fasted Ramadan last year. This year, the average number of fasting days was 28,3 (range 20-29). We did not observe significant effect of Ramadan fasting on glycemic control. The mean hemoglobin A1c was 8,6% ± 1,46 before Ramadan and 8,4% ± 1,46 after Ramadan (p=0,142). During Ramadan, eight patients (18%) developed mild symptoms of hypoglycemia before Iftar. One patient hadn’t broken his fast. Four of them skipped fasting for subsequent 2-4 days. The majority (7/8) were treated with insulin analogs. No patient needed acute medical care or emergency hospitalization for diabetic ketoacidosis.  
Discussion:
To the best of our knowledge, this is the first study to assess the merits of the ADA recommendations in the management of type 2 diabetic patients on insulin during Ramadan. In this study, we have observed that a significant number of patients insist on fasting. In addition, we noted that there was no significant effect of Ramadan on glycemic control. However fasting was associated with an increased risk of hypoglycemia despite the treatment adjustments and the use of insulin analogs. These results suggested that assisting individuals with type 2 diabetes who choose to fast is necessary before and during Ramadan to minimize the risk of afternoon hypoglycemia. The need of specific recommendations for Tunisian diabetic people cannot be overemphasized.