Ramadan Fasting in Diabetes Patients on Insulin Pump Therapy Augmented by Continuous Glucose Monitoring: An Observational Real-Life Study

Abstract

Background and Aims:

Hypoglycemia during the daytime of Ramadan fasting is the most feared complication of diabetes. Insulin pump therapy has been proposed as the ideal “theoretical” method for insulin delivery. We report a prospective observational, single-center study of insulin-treated patients using insulin pump therapy during Ramadan 2011.

Patients and Methods:

Twenty-one patients (10 males and 11 females) were selected; median age was 26 years. They adjusted their insulin as per their usual practices. Outcome measures obtained before and during Ramadan included body weight, glycosylated hemoglobin, blood glucose, total insulin dose differences, overriding tendency, suspension time during fasting, and number of hypoglycemic episodes.

Results:

The patients fasted for a median of 29 days. The observed changes during Ramadan were overall not significant quantitatively, but some trends were noted. The total insulin administered during Ramadan was not different from that in the pre-Ramadan period, but there was a redistribution of insulin over a 24-h period in relation to the changes in the daily lifestyle and eating patterns. Basal insulin was decreased during the daytime by 5–20% from before Ramadan and increased during the nighttime. The mean change in the overall amount of basal insulin was not significant. A larger than usual amount of insulin bolus was given at the meals Iftar, Fowala, and Suhur; the change in the total amount of bolus insulin as a percentage change from total insulin was also not significant. No major hypoglycemic episodes were reported. Minor hypoglcemic episodes were equally distributed between daytime and nighttime and were managed by either basal insulin adjustment or suspension from the pump.

Conclusions:

This study confirms the advantages provided by insulin pump use in patients with diabetes were enhanced by the use of continuous glucose monitoring. We provided more evidence-based advice on how best to adjust the insulin pump during fasting.

Introduction

R amadan, a holy month of Islam, is a lunar-based month, and its duration varies between 29 and 30 days. Fasting during Ramadan is an obligatory duty for all healthy adult Muslims.¹ Muslims during this month must abstain from eating, drinking, use of oral medications, and smoking from predawn to after sunset; however, there are no restrictions on food or fluid intake between sunset and dawn.^1,2 During this month, most people consume two meals a day with varying calorie content depending on local customs. The meals are generally referred to in their original Arabic names as Iftar and Suhur. Iftar (breaking the fast) is taken after sunset, and Suhur (predawn) usually is taken just before dawn.² Patients with chronic medical conditions are exempted from fasting if they risk worsening of their disease or delaying recovery or cure.¹ Diabetes has attracted much interest from physicians and religious scholars, with fairly clear and specific guidance on who should fast or refrain from fasting.³ Nevertheless, many patients with diabetes insist on fasting during Ramadan, thereby creating a challenge for and possibly a burden on themselves and their physicians.^4,5 The role of the physician is primarily to ascertain the safety of the therapeutic regimens prescribed to their patients.^6,7 Several authors have reiterated the various potential risks of fasting during Ramadan, including hyperglycemia, hypoglycemia, ketoacidosis, and dehydration.^6

–9 However, the risk of hypoglycemia during the daytime is by far the most disliked complications as its treatment means the intake of carbohydrate with resulting premature breaking of the fast, which induces a sense of guilt and failure by the faithful patients. Predictably, there is a general perceived fear by patients and their healthcare providers about the use of insulin therapy during Ramadan.

The professional view regarding the specific situation of patients on insulin pump therapy has radically changed over the past decade. Earlier authors deemed the use of insulin pumps as an absolute contraindication for fasting.^8,9 This could be partly due to the lack of experience and proper objective data on its use during Ramadan. Later published opinion statements argued favorably for insulin pump therapy to be an ideal “theoretical” solution.¹⁰ Previously our group reported in this journal an exploratory study on the feasibility of fasting in our adult patients on insulin pump therapy.¹¹ Our conclusions concurred with the opinions of others based on their adolescent population.^12,13 However, in both studies, monitoring was based on symptoms and self-monitoring of capillary blood glucose.¹⁴ It was suggested that utilization of more advanced methodology such as continuous glucose monitoring (CGM) should have given more insight into the degree of glycemic control. Such information would better inform decisions on how best to adjust the insulin infusion rates. We have therefore conducted this observational, single-center, prospective study of insulin-treated patients using insulin pump therapy during the month of Ramadan (2011).

Patients and Methods

Objectives

The objectives of the study were (1) to observe the behavioral attitude of the diabetes patients, making best use of most of the available software features inherent to the pump therapy, and (2) to provide a closer look at the typical eating patterns during Ramadan and their implications on the amount of insulin provided and the corresponding blood sugar excursions. The study protocol was approved by the Institutional Review Board at Sheikh Khalifa Medical City, Abu Dhabi, United Arab Emirates. All patients gave informed consents.

Patient population

In total, 72 patients were screened. Patients were considered for inclusion in the study if they met the following criteria: adults (16–69 years old), with C-peptide <0.1 ng/mL, already on insulin pump therapy as defined by the standard pump qualifications, who were able and willing to perform the required reliable data collection for the study. The following groups of patients were excluded: patients who were pregnant or planning to get pregnant during the projected course of the study, those using systemic steroids at the time of screening or enrollment, patients who had end-stage renal disease, those who had recurrent episodes of skin infections, and those who had participated in any investigational clinical trial during the 30 days prior to screening. Fifty-one patients were excluded for the following reasons: 20 out of the 72 patients were excluded because they had no intention of fasting and/or refused to adhere to the regular visits and note-taking during Ramadan either because of distance issue or for personal reasons, one became pregnant during recruitment phase, two had missing crucial monitoring data, and 28 were below the predefined age limit. Twenty-one patients (10 males and 11 females) completed the study. The baseline characteristics of the study population are given in Table 1.

Table 1.

Baseline Characteristics of the Study Population

				Interquartile range
Characteristic	Mean	SD	Median	25^th percentile	75^th percentile
Age (years)	29.3	12.5	26.0	20.0	38.0
Duration (years) of
Diabetes	12.7	8.2	11.0	8.0	16.0
Pump therapy	3.4	1.2	3.0	3.0	4.0
Body mass index (kg/m²)	27.3	4.6	26.3	24.1	31.1
HbA1c (%)	7.7	1.0	7.4	7.1	8.1
Average blood glucose (mg/dL)	166	38	157	139	189
Daily insulin dose (IU)
Total	59	21	54	42	61
Basal	26	11	25	20	32
Bolus	33	14	29	25	36
Carbohydrate intake (g/day)	194	71	191	153	231

HbA1c, glycosylated hemoglobin; IU, international units.

Dietary changes and patient education

Most patients took Iftar (after the sunset) and Suhur (just before dawn). In the United Arab Emirates, there is traditionally a snack between Iftar and Suhur (Fowala). Patients were advised to adjust their insulin as per their usual carbohydrate counting practices. Patients were instructed to adjust their basal insulin should minor hypoglycemia (defined as blood sugar less than 70 mg/dL) occur or to break their fast should severe hypoglycemia (blood sugar less than 40 mg/dL, requiring the assistance of a third party or hospitalization) develop. Examples of meal samples consumed during Ramadan along with their corresponding amount of carbohydrates are given in Table 2.

Table 2.

Samples of Iftar, Fowala, and Suhur Meal Compositions

Sample meal, food type	Portion size	Carbohydrate (g)
Iftar
Dates	3–5 pieces	15–25
Sugary drink concentrates (Vimto or Tang with sugar)	1 cup	30–45
Somoza (vegetable, meat, or cheese)	2–3	15–45
Harees	½ cup	15
Thareed	½ cup	25
Soup with oats	1 cup	15–20
Rice	2 cups	90
Meat, chicken, or fish	1 piece	0
Yogurt or laban	1 cup	12
Total carbohydrates		177–200
Fowala
Fruit	1–2	15–30
Sweets (Jello, cream caramel, legeemat, custard, mohalabia)		30–45
Tea with sugar		10
Total carbohydrates		55–85
Suhur
Rice with salona and chicken or with yogurt or		50–65
Bread with cheese or eggs or foul with tea		50–65
Total carbohydrates		50–65

See text for definition of the three main meals.

foul, cooked and mashed fava beans served with olive oil; harees, pureed boiled cracked wheat and meat or chicken; laban, drinking yogurt; legeemat, fried sweet dumplings in sugar syrup; mohalabia, milk pudding; salona, meat boiled with a variety of vegetables and tomato paste; somoza, fried pastries stuffed with meat or chicken or vegetables; Tang, a proprietary powdered orange juice; thareed, pieces of bread in cooked vegetable broth with meat or chicken; Vimto, a proprietary concentrated mixed berry juice.

Outcome measures

Outcome measures were taken in two periods—namely, “Pre-Ramadan” (defined as 30 days preceding the first day of Ramadan) and “during Ramadan.” Measurements included the differences in body weight and body mass index, changes in glycosylated hemoglobin, average blood sugar differences, total amount of insulin differences, basal insulin differences, percentage change in bolus insulin from the total amount, number of fasting days, overriding tendency, suspension time from pump during Ramadan, and the number of minor and major hypoglycemic episodes.

Statistical methods

Standard statistical methods for paired data, notably the paired t test, were used for significance tests. A significance level of 0.05 was used throughout. Data were summarized as either median and (interquartile) ranges or means and SD values. Box plots were used to graphically present the distribution of variables.

Insulin pump protocol and CGM data capture

Intermittent CGM was applied to all the patients in the study. Data were extrapolated from the patients' pumps using the Medtronic (Northridge, CA) Minimed 722 model. Data were correlated with the data obtained from a sample questionnaire given to patients relating to their perceptions of the changes in their daily lifestyle and eating patterns during Ramadan; the sensed hypoglycemic episodes were also correlated with the ones recorded during intermittent CGM. Continuous glucose sensing during Ramadan was performed on average for 10 days per patient. The data obtained were then compared with those from the month preceding Ramadan, so that the same patients served as their own controls.

Results

Fasting days and body weight changes

The median number of days fasted was 29 days (Table 1). Intermittent abstinence from fasting was mainly observed in women and more precisely due to menses (as they are not supposed to be fasting during this period). The changes during Ramadan were overall not significant quantitatively, but some trends were noted qualitatively. There was no significant difference in body mass index from before Ramadan (Table 3 and Fig. 1).

FIG. 1.

Box plot of percentage increase during Ramadan of the main outcome measures. Basal, basal insulin; BMI, body mass index; Bolus, bolus insulin; BS, blood sugar; CHO, daily carbohydrates.

Table 3.

Summary of the Outcome Measures Before and During Ramadan and Their Mean Change

	Mean±SD value
Variable	Before Ramadan	During Ramadan	Change	Significance two-tailed ^a
Body mass index (kg/m²)	27.3±4.6	27.1±4.3	−0.1±1.0	0.45
HbA1c (%)	7.7±1.0	7.6±8.8	−0.13±0.8	0.5
Average blood glucose (mmol/L)	166±38	171±40	+6±26	0.274
Total daily insulin dose (units)	59±21	57±20	−0.84±9.71	0.718
Basal insulin (units/day)	26±11	25±11	−1.4±4.0	0.125
Bolus insulin/total daily dose (%)	33±14	34±11	+3.2±7.8	0.108
Carbohydrates (g/day)	194±71	216±104	+26±85	0.225

By paired-samples t test.

HbA1c, glycosylated hemoglobin.

Insulin infusion rates

The overall amount of total insulin per day was not significantly different (Table 3). However, there was an expected redistribution of insulin over a 24-h period to match the changes in the daily lifestyle and eating patterns (i.e., basal insulin was decreased during the daytime [by 5–20%] and increased during the nighttime period between Iftar and Suhur). However, the mean change in the overall amount of basal insulin was not significant (Table 3). Predictably, no insulin bolus was taken during the daytime because of abstinence from food and drink. However, larger amounts of insulin bolus were infused during the Iftar, Fowala, and Suhur meals as described above. The change in the total amount of bolus insulin given, estimated as a percentage change from total insulin, was also not significant (Table 3). In parallel, there was a trend toward a positive change in the average blood glucose that was not significant (Table 3).

Hypoglycemia

There were no major hypoglycemic episodes observed in the study period. Thirty-three minor hypoglycemic episodes were equally distributed between day and night. The reactions during the fasting period (16/33) were adequately managed by either basal insulin decrement or temporary suspension of the pump. The median time suspension from the pump was 1 h. There was no perceived need to interrupt the fast prematurely in any patient. Nearly half of the minor hypoglycemic reactions occurred between the Iftar and the Suhur period (17/33 episodes) because of overcorrection (70%) or reduced carbohydrate intake (30%). A Wilcoxon signed rank test to compare the periods of hypoglycemia before and during Ramadan was used. The P value was 0.097; a t test of the square root of the transformed number of episodes gave a P value of 0.119, indicating that there was no statistical significance between the episodes of hypoglycemia before or during Ramadan.

Discussion

Fasting during Ramadan involve alternating periods of absolute fasting and refeeding within a 24-h period.² This alternating pattern produces two contrasting metabolic environments within the same day. To address this perpetual pattern, several strategies have been suggested for the patients receiving multiple insulin dose injections.^15,16 More recently, insulin pump therapy has been suggested as one innovative method for use during Ramadan fasting.^10,17 However, most of the advantages relating to insulin therapy during Ramadan proposed in the literature are based on both anecdotal or “theoretical” arguments and not well-planned studies.^18
–20 To our knowledge there are only a few studies of this sort,^10

–14 and further research in this area is badly needed.^17,21

There are special aspects relating to the eating patterns observed during Ramadan that are worth highlighting. First, eating patterns change dramatically during Ramadan with a tendency for grouping most of the calories consumed during a day over 4–8 h.² Furthermore, as Ramadan follows the lunar calendar, the fasting month is brought forward by about 10 days each year; as such, the daylight hours can reach up to 20 h in nonequatorial areas, reducing the time between sunset to dawn to as little as 4–6 h and maximizing the postprandial hyperglycemia in a shorter time frame. Second, the amount of carbohydrates and fat consumed during Ramadan is on average increased in comparison with that of a nonfasting day. The estimated amount of carbohydrate consumed during Ramadan can vary from 300 up to 500 g daily. An average estimation is given in Table 2. Postprandial hyperglycemia is also sustained with an increase in fat consumption delaying food absorption. In this regard, 801 episodes of postprandial hyperglycemia were recorded in our patients post-Iftar (hyperglycemia defined as blood sugar exceeding 180 mg/dL), explaining the high percentage of corrective insulin boluses given with meals and postprandially representing 70% of the total insulin administered.

Insulin pump therapy addresses elegantly the above-mentioned metabolic issues mainly the potential daytime hypoglycemia due to deprivation of caloric intake and unopposed insulin action as well as the compensatory overeating after the breaking of the fast with its resulting hyperglycemia.

The results of the present study reflect on the advantages offered by usage of the insulin pump during Ramadan. Patients had the flexibility of adjusting their insulin pattern with no or minimal risk for complications. It was possible that hypoglycemia can be aborted, reduced, prevented, or reversed by timely downward adjustments or even totally stopping of insulin delivery from the pump. Such facility is not available to those on multiple daily insulin injection therapy. The episodes of hypoglycemia observed were minor and easily addressed at a time when the patients are quite awake and are free to treat it in the usual manner. The excessive amount of hyperglycemia after Iftar is not inherent to the pump and does not constitute a pump failure or a bad outcome but rather a mismatching between excessive carbohydrates consumed and the corresponding amount of insulin infused. The rate of hypoglycemia taking place during the day (fasting period), totaling only 16 minor episodes in 21 individuals with diabetes studied over an entire month, is remarkably low. In addition to the inherent capability adjustments of the pump, other reasons could have contributed such as the physical inactivity observed during the daytime of Ramadan. In the present study, the questionnaires only counted the hypoglycemia episodes and did not document possible causes.

There are, however, some limitations of this study. The small number of patients could explain the presence of trends rather than statistically significant results. A larger study, perhaps involving more than one center, would have been more relevant. The self-selective nature of our group of patients could also pose a challenge to our attempt to generalize. Patients involved in this study are well educated about their insulin pump. The importance of education has indeed been strongly associated with control.²² We have also excluded two important subgroups on s pragmatic basis—namely, young (less than 16 years of age) patients, who might have a more brittle diabetes, and pregnant women. We uphold the decree that recommends that these two subgroups among others should indeed be advised to refrain from fasting.³ In this regard, the results of the study of Bin Abbas and co-workers^12,13 in children and adolescents using an insulin pump concur with ours. They demonstrated a decreased frequency of hypoglycemic episodes that was “serious” enough to require breaking the fast during the holy month of Ramadan. It could be argued that comparison of insulin pump therapy with an alternative type therapy such as multiple daily insulin injections would have been appropriate to illustrate any advantages from the pump over the conventional method. However, we believed that using the patients as their own controls is also relevant for sound statistical analysis.

We conclude that the use of insulin pump therapy augmented with CGM capability during Ramadan is feasible physiologically and realistically. It provides flexibility in matching food patterns with the more appropriate insulin dosages. It is also particularly useful in reducing the severity and duration of hypoglycemia, facilitated by the newly added pump features of sensing and automatic suspension in the presence of low blood sugars.^23,24

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

The Holy Quran. Sura, 2:183–185.

Sulimani

. Ramadan fasting: medical aspects in health and in disease. Ann Saudi Med, 1991; 11:637–641.

Beshyah

. Fasting Ramadan for people with diabetes: medicine and Fiqh united at last. Ibnosina J Med Biomed Sci, 2009; 1:1–5.

Salti

, Bénard

, Detournay

, Bianchi-Biscay

, Le Brigand

, Voinet

, Jabbar

. EPIDIAR Study Group. A population based study of diabetes and its characteristics during the fasting month of Ramadan in 13 countries: results of the epidemiology of diabetes and Ramadan 1422/2001 (EPIDIAR) study. Diabetes Care, 2004; 27:2306–2311.

Beshyah

, Benbarka

, Sherif

. Practical management of diabetes during Ramadan fast. Libyan J Med, 2008; 2:185–189.

Beshyah

, Sherif

. Diabetes during Ramadan and Haj. Lakhdar

, Gill

. Diabetes in the Arab World. Cambridge, UK: FSG Communications Ltd., 2005; 225–240.

Benaji

, Mounib

, Roky

, Aadil

, Houti

, Moussamih

, Maliki

, Gressier

, El Ghomari

. Diabetes and Ramadan: review of the literature. Diabetes Res Clin Pract, 2006; 73:117–125.

Diabetes and Ramadan Advisory Board. International Medical Recommendations for Muslim subjects with diabetes mellitus who fast during the month of Ramadan. Clin Diabetes (Middle East Ed), 2004; 3:143–145.

Al-Arouj

, Bouguerra

, Buse

, Hafez

, Hassanein

, Ibrahim

, Ismail-Beigi

, El-Kebbi

, Khatib

, Kishawi

, Al-Madani

, Mishal

, Al-Maskari

, Nakhi

, Al-Rubean

. Recommendations for management of diabetes during Ramadan. Diabetes Care, 2005; 28:2305–2311.

10.

Al-Arouj

, Assaad-Khalil

, Buse

, Fahdil

, Fahmy

, Hafez

, Hassanein

, Ibrahim

, Kendall

, Kishawi

, Al-Madani

, Nakhi

, Tayeb

, Thomas

. Recommendations for management of diabetes during Ramadan: update 2010. Diabetes Care, 2010; 33:1895–1902.

11.

Benbarka

, Khalil

, Beshyah

, Marjei

, Awad

. Insulin pump therapy in Moslem patients with type 1 diabetes during Ramadan fasting: an observational report. Diabetes Technol Ther, 2010; 12:287–290.

12.

Bin-Abbas

, Sakati

, Raef

, Al-Ashwal

. Continuous subcutaneous insulin infusion in type 1 diabetic Saudi children. A comparison with conventional insulin therapy. Saudi Med J, 2005; 26:918–922.

13.

Bin-Abbas

. Insulin pump therapy during Ramadan fasting in type 1 diabetic adolescents. Ann Saudi Med, 2008; 28:305–306.

14.

Lessan

, Abdullah

, Al Mousavi

, Nasir

, Athamneh

. Utility of continuous glucose monitoring (CGMS) to explore glycemic patterns in patients with diabetes mellitus in the UAE [abstract] Endocr Abstr, 2006; 11:P324.

15.

Azar

, Khairallah

, Merheb

, Zantout

, Fliti

. Insulin therapy during Ramadan fast for patients with type 1 diabetes mellitus. J Med Liban, 2008; 56:46.

16.

Kassem

, Zantout

, Azar

. Insulin therapy during Ramadan fast for Type 1 diabetes patients. J Endocrinol Invest, 2005; 28:802–805.

17.

Ahmed

, Abdu

. Diabetes and Ramadan: an update on use of glycemic therapies during fasting. Ann Saudi Med, 2011; 31:402–406.

18.

Pickup

. Insulin pumps. Int J Clin Pract Suppl, 2011; 170:16–19.

19.

Merheb

, Chaiban

, Salti

, Sammak

, Choucair

, Nasrallah

. Insulin pump therapy in type 1 diabetic patients: experience at the American University of Beirut Medical Center. Int J Diabetes Metab, 2008; 16:57–61.

20.

Ooi

, Wu

. Insulin pump therapy in children and adolescents with Type 1 diabetes: improvements in glycemic control and patients' satisfaction—Hospital UKM experience. Med J Malaysia, 2011; 66:308–312.

21.

Sherif

, Lakhdar

. Ramadan fasting and the medical patient: consensus is welcome but more evidence is needed. Ibnosina J Med Biomed Sci, 2010; 2:237–239.

22.

Jayasekara

, Munn

, Lockwood

. Effect of educational components and strategies associated with insulin pump therapy: a systematic review. Int J Evid Based Healthc, 2011; 9:346–361.

23.

Garg

, Brazg

, Bailey

, Buckingham

, Slover

, Klonoff

, Shin

, Welsh

, Kaufman

. Reduction in duration of hypoglycemia by automatic suspension of insulin delivery: the in-clinic ASPIRE study. Diabetes Technol Ther, 2012; 14:205–209.

24.

Agrawal

, Welsh

, Kannard

, Askari

, Yang

, Kaufman

. Usage and effectiveness of the low glucose suspend feature of the Medtronic Paradigm Veo insulin pump. J Diabetes Sci Technol, 2011; 5:1137–1141.