The effect of maternal fasting during Ramadan on preterm delivery: a prospective cohort study


A Nassar, American University of Beirut Medical Centre, Department of Obstetrics and Gynaecology, PO Box 113-6044/B36, Beirut, Lebanon. Email


Please cite this paper as: Awwad J, Usta I, Succar J, Musallam K, Ghazeeri G, Nassar A. The effect of maternal fasting during Ramadan on preterm delivery: a prospective cohort study. BJOG 2012;119:1379–1386.

Objective  To determine the effect of fasting during the month of Ramadan on the rate of preterm delivery (PTD).

Design  A prospective cohort study of women with singleton pregnancies who elected to fast and matched controls.

Setting  Four medical centres in Beirut, Lebanon.

Population  Women presenting for prenatal care (20–34 weeks of gestation) during the month of Ramadan, September 2008.

Methods  Data were collected prospectively. The frequency of PTD was evaluated in relation to the duration of fasting and the stage of gestation at the time of fasting.

Main outcome measures  The primary endpoint was the percentage of pregnant women who had PTD, defined as delivery before 37 completed weeks of gestation.

Results  A total of 468 women were approached, of whom 402 were included in the study. There were no differences in smoking history and employment. There was no difference in the proportion of women who had PTD at <37 weeks (10.4% versus 10.4%) or PTD at <32 weeks (1.5% versus 0.5%) in the Ramadan-fasted group and the controls, respectively. The PTD rate was also similar in those who fasted before or during the third trimester. The mean birthweight was lower (3094 ± 467 g versus 3202 ± 473 g, = 0.024) and the rate of ketosis and ketonuria was higher in the Ramadan-fasted women. On multivariate stepwise logistic regression analysis, fasting was not associated with an increased risk of PTD (odds ratio 0.72; 95% confidence interval 0.34–1.54; = 0.397). The only factor that had a significant effect on the PTD rate was body mass index (odds ratio 0.43; 95% confidence interval 0.20–0.93; = 0.033).

Conclusions  Fasting during the month of Ramadan does not seem to increase the baseline risk of preterm delivery in pregnant women regardless of the gestational age during which this practice is observed.


Pregnancy is a physiological state with particular nutritional requirements critical for the achievement of optimal maternal and neonatal outcomes. Dietary recommendations for mothers have been established to ensure a balanced nutritional intake for a healthy pregnancy course.1–3 Circumstances that impose serious strains on the quality, quantity and distribution of food intake are believed to adversely affect maternal and neonatal outcomes.

Ramadan is the ninth lunar month of the Islamic calendar and symbolises the Revelation of the holy Qur’an. During this month, devout Muslims abstain from eating and drinking from dawn to dusk for a period of 29–30 days. Ramadan fasting represents a serious alteration to routine eating and drinking habits that deserves special consideration, namely that over one billion adherents globally are obligated to observe this essential tenet of the sacred law. Although the Ramadan fast is compulsory in Islam, pregnancy may represent a relative exemption if reasons for maternal/fetal hardship are suspected.4 Many Muslim women, nonetheless, still choose to fast during their pregnancy, creating a challenge to healthcare providers who feel compelled to provide them with credible advice. Unfortunately, a very limited literature database is available to offer a platform for sound counselling.

Few studies have examined the untoward effects of maternal fasting on the wellbeing of the mother and fetus.4–9 Significantly lower levels of glucose, insulin, lactate and carnitine were found in the serum of fasting mothers,6,7 with higher levels of triglycerides, non-esterified fatty acids and 3-hydroxybutyrate. A reduction in fetal breathing movements8 and fetal biophysical profile9 were also reported in association with maternal fasting. An increased risk of hyperemesis gravidarum5 was further associated with prolonged fasting. Other studies however failed to demonstrate any undesired effects on fetuses born to fasting mothers.10–16 No changes in flow velocimetry parameters were found in a uterine artery Doppler study of pregnant women during a Ramadan fast.10,17 Similarly, amniotic fluid index, birthweight and children’s IQ were all unaffected.10–16

There is a very limited number of studies investigating the effects of caloric restriction on the risks of preterm labour (PTL) and preterm delivery (PTD) in pregnant women.18,19 The Dutch famine in 1944/45 provides evidence on the differential timing effect of poor nutrition on low birthweight and PTD.19 First-trimester food deprivation increased the risk of PTD, whereas third-trimester nutritional stress was associated with low birthweights.19 Fluid restriction during the fast could theoretically provoke contractions. Women with PTL have been reported to have lower plasma volume than controls20 and based on uncontrolled observations, intravenous hydration might decrease contractions or delay the delivery in these women, as proposed by some authors.20,21 It has been hypothesised that volume expansion could increase uterine blood flow, stabilise decidual lysosomes and decrease prostaglandin production.22 In addition, it could decrease the secretion of antidiuretic hormone from the posterior pituitary,21,22 so decreasing oxytocin secretion simultaneously.21

The clinical significance of PTD cannot be underestimated because it remains the leading public health burden in obstetric practice today. It is a major cause of neonatal morbidity and mortality, contributing an estimated 36.5% of infant deaths in 2005.23 Infant survivors born prematurely are themselves at increased risk for long-term neurological and developmental complications.24,25

Sharing the scientific evidence regarding the effects of fasting on maternal and neonatal outcome is the most appropriate approach to assist Muslim pregnant women in making informed decisions about fasting during the month of Ramadan. Unfortunately, the views of the medical literature with respect to this subject remain inconclusive. Available studies suffer many shortcomings, including small sample size, poorly defined outcome measures, and lack of well-matched controls. PTD has never been investigated as an outcome measure in the context of Ramadan fasting. The present study was therefore designed to investigate whether fasting during Ramadan increased PTD in pregnant women. Our null hypothesis was that daytime fasting would be unlikely to have any measurable effect on PTD rate in Ramadan-fasted women compared with matched controls.



This prospective study was conducted between the 1 and 30 September 2008 during the holy month of Ramadan in a geographic time-zone GMT +3. Women were identified during the course of routine prenatal care visits at four participating medical centres representative of the socioeconomic constitution of the Lebanese community during the month of August 2008 preceding Ramadan. Eligible women included all healthy Muslim women with singleton uncomplicated pregnancies who would be at 20–34 weeks of gestation at the beginning of the month of Ramadan irrespective of whether they would fast or not during the month of Ramadan. Exclusion criteria consisted of the following: conceptions by assisted reproductive technology; pre-existing maternal medical diseases (such as chronic hypertension, diabetes mellitus, asthma, seizures, cardiac disease, renal disease); multiple gestations; major fetal congenital abnormalities; polyhydramnios; obstetric complications (such as gestational diabetes, pre-eclampsia, chorioamnionitis); and previous poor obstetric outcome (such as PTD, intrauterine fetal deaths, recurrent pregnancy losses). The study was approved by the Institutional Review Board of the American University of Beirut Medical Centre (AUBMC). Informed consents were provided according to the Declaration of Helsinki. Consenting mothers were enrolled in August and followed up throughout the pregnancy thereafter.

Women who intended to fast were assigned to the ‘Ramadan-fasted group’ and each fasting woman was matched to a control non-fasting women by maternal age (±1 year), gestational age (±2 weeks), parity (nulliparous versus multiparous) and body mass index (BMI) (±2 kg/m2). The matching process was done sequentially within the same clinic and on the same or the following day. Women who intended to fast were provided with a timetable to document days and hours of fasting. During the month of Ramadan, phone calls were made every fifth day to monitor individual compliance in completing the self-reported document.


Maternal weight and blood pressure were recorded at the beginning and at the end of Ramadan. Maternal effects of fasting were assessed by measuring blood concentrations of glucose and 3β-hydroxybutyrate, and urine ketone levels. Samples for the above measurements were collected approximately 1 hour before breaking the fast (around 5:30–6:00 p.m.), both in the beginning (first 4 days) and at the end (last 4 days) of Ramadan. Collection was made at home upon scheduled appointments to facilitate individual compliance. Collected samples were then transported to the central laboratory of AUBMC for analysis.

Glucose levels were measured using a glucometer (Accu-Check Active; Roche Diagnostics, Basel, Switzerland). Hypoglycaemia was defined as serum levels <3.33 mmol/l. The 3β-hydroxybutyrate concentrations were measured by capillary test (Optimum Betaketone test strips; Medisense, Abingdon, Oxon, UK) to detect ketosis (≥1 mmol/l). Ketonuria was measured with urine dipsticks (Multistix®) read by Clinitek 50 (Bayer Leverkusen, North Rhine-Westphalia, Germany) where +, ++, or +++ urine ketones correspond to 1.5, 5, or 8 mmol/l of acetoacetate, respectively. On the last visit, women were investigated for medication intake and the frequency of episodic vomiting, diarrhoea, headache or dizziness during the month of fasting.

Definition of endpoints

Gestational age was calculated based on the last menstrual period or first-trimester ultrasonography whenever appropriate. The primary endpoint was the percentage of pregnant women who had PTD, defined as delivery before 37 completed weeks of gestation. Secondary endpoints included the percentage of women with confirmed hypoglycaemia, ketosis, ketonuria, those with deliveries before 32 weeks of gestation, and birthweight <2500 g and <1500 g. Small-for-gestational-age was defined as birthweight <10th centile for gestational age for singletons.

The frequency of PTD was evaluated in relation to the duration of fasting (≤15 days versus >15 days) and the stage of gestation at the time of fasting (first/second versus third trimester). Matching for maternal age, parity and BMI was carried out to control for previously established risk factors for PTD. Additional variables such as smoking and employment, considered confounders for PTD, were also collected.

Statistical analysis

Based on the yearly statistics from the Department of Obstetrics and Gynaecology at AUBMC and worldwide estimates of PTD rate of approximately 10%,20 a total sample size of 398 pregnant women (199 in the fasting group and 199 in the control group) was needed for the detection of a two-fold increase in the rate of PTD (from 10% to 20%), under the assumptions of a type I error (two-sided) of 5% and a power of at least 80%. Considering a dropout rate of 10%, the total number of women required was estimated to be 440 (220 in each arm).

Statistical analysis was performed using spss statistical package version 13 (SPSS Inc., Chicago, IL, USA). Categorical data were compared using a chi-square test. Otherwise, the two-tailed Fisher exact test was used if the expected cell frequencies were small. Continuous variables were compared using two-tailed Student’s t-test if assumptions of normality and homogeneity of variances appeared to be reasonable or the Wilcoxon rank-sum test. Multivariate stepwise logistic regression analysis was performed to examine the influence of confounding variables on the PTD: fasting, maternal age (<35 and ≥35 years); nulliparity, BMI (<25 kg/m2 and ≥25 kg/m2); employment, and smoking. A P-value of <0.05 was considered statistically significant. Data that were considered completely missing at random were deleted using the listwise approach.


Of 468 women interviewed, 427 (91.2%) agreed to participate in the study. After excluding 12 women for early withdrawal and 13 for missing data, the analysis was run on 402 women (201 Ramadan-fasted and 201 controls). The mean number of fasting days was 22 ± 9 (median 26; range 1–30). No significant differences were found in the baseline demographic characteristics between the two groups (Table 1).

Table 1.   Maternal characteristics of the Ramadan-fasted and control (non-fasting) groups
 Ramadan-fasted group
(n = 201)
Control group
(n = 201)
P value
  1. *Data presented as mean ± standard deviation.

  2. **Data presented as n (%).

Age (years)* 29.7 ± 5.230.0 ± 5.40.529
Gestational age at recruitment (weeks)* 27.0 ± 4.726.9 ± 4.90.672
Body mass index (kg/m2)* 24.8 ± 4.324.3 ± 3.60.252
Primigravida54/200 (27)69/201 (34.3)0.171
Gravida ≥2146/200 (73)132/201 (65.7)
Nulliparous91/200 (45.5)86/201 (42.8)0.282
Primiparous46/200 (23)60/201 (29.8)
Para ≥263/200 (31.5)55/201 (27.4)
Zero115/200 (57.5)136/201 (67.6)0.115
1–277/200 (38.5)58/201 (28.9)
≥38/200 (4)7/201 (3.5)
Smoker** 19/183 (10.4)9/175 (5.1)0.099
Employed** 47/191 (24.6)51/190 (26.8)0.648

The proportion of women who had PTD was comparable between the fasting and control groups (Table 2). However, the caesarean delivery rate was significantly lower in Ramadan-fasted women (28.4% versus 39.3%; P = 0.027) and their neonates had a significantly lower mean birthweight compared with those born to controls (3094 ± 467 versus 3202 ± 473 g; P = 0.024). When evaluating the obstetric effects of fasting in relation to the stage of gestation, no differences were found in the occurrence of PTD whether women fasted during the first/second trimester or third trimester (12/106 versus 9/95, = 0.844). Moreover, the duration of fasting did not appear to influence the endpoints measured. Fasting for ≤15 days or for >15 days did not affect significantly the percentage of women who had PTD (7/51 versus 14/150; P = 0.535).

Table 2.   Primary and secondary end-points in the Ramadan-fasted and control (non-fasting) groups
 Ramadan-fasted group
(n = 201)
Controls group
(n = 201)
P value
  1. *Data presented as mean ± standard deviation.

  2. **Data presented as n (%).

Gestational age at delivery (weeks)* 38.2 ± 1.738.4 ± 1.50.499
Preterm delivery**
<37 weeks21 (10.4)21 (10.4)1.000
<32 weeks3 (1.5)1 (0.5)0.623
Small for gestational age** 11/195 (5.6)12/194 (6.2)0.820
Birthweight (g)* 3094 ± 4673202 ± 4730.024
Birthweight <2500 g** 12/195 (6.2)7/194 (3.6)0.353
Birthweight <1500 g** 1/195 (0.5)2/194 (1.0)0.623
Caesarean delivery** 57/201 (28.4)79/201 (39.3)0.027

The likelihood of hypoglycaemia was similar between the fasting and control groups both at the beginning and at the end of Ramadan (Table 3). Ketosis and ketonuria were more frequent among Ramadan-fasted women compared with controls (Table 3). Ramadan-fasted women had a significantly lower mean increase in maternal weight compared with controls at the end of the fasting month (1.6 ± 2.2 kg versus 2.3 ± 2.0 kg; = 0.001). They were also more likely to report episodes of vomiting (= 0.011), diarrhoea (= 0.004) and dizziness (= 0.001) throughout the same month (Table 4).

Table 3.   Maternal laboratory data of Ramadan-fasted and control (non-fasting) groups
 Ramadan-fasted groupControl group P value
  1. All data presented as n (%).

  2. *Data collected during the first 4 days of Ramadan.

  3. **Data collected during the last 4 days of Ramadan.

Hypoglycaemia*48/150 (32)32/141 (22.7)0.076
Hypoglycaemia**38/99 (38.4)21/82 (25.6)0.068
Ketosis*17/150 (11.3)7/141 (5.0)0.048
Ketosis**15/99 (15.1)5/82 (6.1)0.053
Ketonuria*18/141 (12.8)6/137 (4.4)0.013
Ketonuria**12/82 (14.6)2/79 (2.5)0.006
Table 4.   Signs and symptoms reported by Ramadan-fasted and control (non-fasting) groups during the month of Ramadan
 Ramadan-fasted group
(n = 201)
Control group
(n = 201)
P value
  1. *Data presented as mean ± standard deviation.

  2. **Data presented as n (%).

Weight gain (kg)*1.6 ± 2.22.3 ± 2.00.001
Vomiting**10/200 (5.0)1/200 (0.5)0.011
Diarrhoea**9/200 (4.5)00.004
Headache**5/201 (2.5)3/201 (1.5)0.724
Dizziness**17/201 (8.5)2/201 (1.0)0.001
Systolic pressure*110 ± 10110 ± 110.678
Diastolic pressure*68 ± 867 ± 100.158

On multivariate stepwise logistic regression analysis, none of the factors studied had a significant effect on the PTD rate except for BMI (odds ratio 0.43; 95% confidence interval 0.20–0.93; = 0.033). Fasting was not associated with an increased risk of PTD (odds ratio 0.72; 95% confidence interval 0.34–1.54; = 0.397).


The findings of the present study showed that Ramadan fasting during the month of September in a geographic time zone GMT +3 did not increase the risk of PTD in a cohort of healthy Lebanese pregnant women with singletons between 20 and 34 weeks of gestation. A PubMed search of the English literature from 1980 to August 2011 using the key words ‘fasting’, ‘Ramadan’, ‘preterm labour’ and ‘preterm delivery’, reveals this study to be the first to determine the effect of Ramadan fasting on PTD during the course of pregnancy. Data generated consequently add valuable information to the existing body of knowledge that will be useful to the counselling process of pregnant women interested in fasting during this holy month.

Our findings are in contrast with several human and animal studies that evaluated various patterns of experimental caloric deprivation and the risks of PTD. Meal pattern variations for instance were shown to influence the frequency of PTD in pregnant women. Siega-Riz et al.26 reported that women who ate fewer than three meals and two snacks per day had a 30% higher risk for PTD compared with pregnant women who met this frequency. Pregnant women who abstained from eating for >13 hours per day had a three-fold greater risk of delivering preterm at <34 weeks of gestation (TS Herrmann, unpublished data from 1999). Improving the nutritional status of pregnant women was therefore proposed by some as a means of reducing the risk of PTD.27,28 Few animal studies also indicate that food deprivation of 12–48 hours during late gestation upregulates corticotrophin-releasing hormone (CRH) messenger RNA (mRNA) in various regions of the rodent brain29,30 and triggers preterm birth.31,32 Similarly, work in sheep confirmed that acute undernutrition in early gestation triggered PTD.33 Herrmann et al.34 found that a 13-hour fast in 237 pregnant women increased significantly maternal CRH concentrations compared with fasts of shorter duration. An inverse linear relationship between maternal CRH concentrations and gestational age at delivery was also established.34 It is plausible then to assume that CRH pathways may be involved in the pathogenesis of starvation-induced PTD, possibly stimulating placental–fetal signalling during late gestation to hasten fetal delivery from a potentially adverse environment. These experimental starvation models nevertheless may not always be representative of a 15-hour Ramadan fast. It is likely that the nature of stressors associated with energy deprivation during religious fasting expresses a different pattern of manifestations on the obstetric outcome.

The metabolic adaptation to energy restriction in humans involves the release of ketoacids, β-hydroxybutyric acid, and acetoacetic acid to replace glucose as the primary substrate in tissues. Transient ketonuria and ketonaemia may occur during pregnancy as a result of caloric restriction and also secondary to hypohydration. Ketones were described to affect amniotic fluid volume and composition in sheep,35 and potentially to cause damage to the neurological system of rodents.36 Felig and Lynch37 suggested that pregnant women were more vulnerable to heightened ketonaemia and hypoglycaemia after a brief period of fasting (12 hours) compared with non-pregnant ones. The term ‘accelerated starvation’ was coined by Metzger et al.38 who confirmed these findings following 16 hours of fasting. Hizli et al.17 in a case–control design of 56 Ramadan-fasting pregnant women concurred with the above. On the other hand, Dikensoy et al.39 could not detect significant ketonaemia or ketonuria in pregnant women after a 13- to 14-hour daily fast. The findings of the present study revealed that a 15-hour fast resulted in an increased likelihood of ketonuria and ketosis in pregnant women, but failed to correlate these findings with a higher rate of PTD.

The most widely used indicator of maternal nutritional status during pregnancy is birthweight. Our study showed a significant reduction in birthweight in the Ramadan-fasted group. Weight nonetheless provides a limited perspective on fetal growth and development during gestation. Data from the 1944/45 Dutch famine provide valuable information on the effects of severe energy rationing on birthweights. A daily energy restriction of 400–800 kcal/person/day from a baseline of 1800 kcal for 7 months resulted in a significant drop in birthweights only when it coincided with the third-trimester of pregnancy.40 A retrospective cohort study of neonates born to 284 mothers who fasted in Ramadan during their pregnancy did not show any significant effects on birthweights.13 Similarly, a case–control study revealed that fasting in 56 pregnant women did not affect birthweight compared with pregnant controls.17 Reviewing studies that evaluate the effects of Ramadan fasting on maternal and fetal health, we found only one that matches closely our study in relation to the characteristics of the fasting period.41 In that respect, both were performed during September 2008 and within the same geographic time zone GMT +3. Unlike the findings of the present study however, Ozturk et al.41 in their prospective controlled design, showed no significant differences in maternal weight gain and birthweights among 42 Ramadan-fasted pregnant women compared with non-fasting controls.41 The Turkish study nevertheless suffered a major limitation related to the small number of women studied. The lower birthweight in our study could have contributed to the lower caesarean delivery rate in the Ramadan-fasted group.

The majority of reported health-related findings in relation to fasting and energy deprivation are complex and often contradictory.42 Energy deprivation may be either: (i) ‘elective’, often reflective of a real-time social experience such as religious fasting; (ii) ‘experimental’, targeted to measure specific physiological outcome measures; or (iii) ‘compulsory’, best exemplified by catastrophic real-life events, such as famine. Fundamental differences exist among these three patterns of energy deprivation, in terms of the magnitude of caloric deprivation, temporal characteristics, individual background energy reserves, and consequently bodily physiological adjustments. ‘Elective’ caloric stress, being self-imposed and predictive, is normally associated with episodic replenishment of individual energy reserves. It should be differentiated from the non-repetitive highly standardised ‘experimental’ type occurring in the context of healthy individual energy reserves, and the cumulative coercive ‘compulsory’ type leading to the progressive erosion of individual energy reserves.

The non-obstetric medical literature provides a wealth of data evaluating the effects of religious fasting on health, and shows that changes in energy balance are not consistent between communities observing a Ramadan fast. As would be expected during this period, several studies reported either reduced daily energy intake43,44 or demonstrated an overall energy balance.45,46 Contrary to any prediction nonetheless, few studies reported on a paradoxical increase in the daily energy intake during Ramadan leading to an overall weight gain.47–49 In a survey of Saudi families, 59.5% of women reported weight gain and 79.4% increased their food expenditure during Ramadan.49 The behavioural response of Muslims to religious requirements therefore does not appear to be uniform across cultures. A process of overcompensation could be driven by binge eating or modification of caloric sources and food preferences. The current design did not account for the total daily energy intake of the pregnant Lebanese women throughout the study. Yet, the significantly lower weight gain found in the current study in association with Ramadan fasting points to a likely decrease in maternal daily caloric intake. It should be noted that body mass measurements are not the best index of energy intake, because they can be confounded by a negative water balance as a result of hypohydration43 and a significant fall in daily energy expenditure secondary to decreased habitual activity profile.

In the present study, we acknowledge some limitations. Because Ramadan is a lunar month, it may occur in any season of the year. The daily duration of the fast may therefore vary from 10 to 19 hours depending on the season and the geographic location, being much shorter the nearer one is to the equator. For this reason, the usefulness of our findings is more relevant to pregnant women who observe Ramadan around September in a time zone GMT +3. Furthermore, the study design did not account for the daily caloric intake of the women studied, which renders the comparison with other studies more difficult. Also, it did not address alterations in food quality which could profoundly affect gut metabolism potentially influencing the risk of PTD via modulation of local and systemic inflammatory markers. Although the sample size of pregnant women enrolled in this study is much larger than the majority of comparable studies, it is only capable of detecting a two-fold or greater increase in PTD rates, i.e. 10 to 20% PTD rate. Considering that preterm birth is a complex entity with multiple predisposing factors, no single alteration in behaviour during pregnancy may be held accountable alone for any significant increase in PTD rate. Consequently, failing to reject the null hypothesis does not necessarily imply equivalence. At the other end, we also recognise several strengths to this study. (i) The prospective controlled nature of the design involved a rigid process of patient matching by recruiting controls who matched study women at every level of known confounding variables, i.e. maternal age, gestational age, parity and BMI. In addition, the control population was restricted to Muslim women so the unexposed group could have counterfactually been part of the exposed group. (ii) The endpoint measured had not been previously investigated in the context of a Ramadan fast, namely PTD. (iii) A rigorous follow-up system of study women was adopted to reduce recall bias. It consisted of contacting women by phone every fifth day to reinforce personal compliance in completing timetable sheets. (iv) A home collection service was designed to ensure that blood and urine samples were promptly obtained as previewed by the study design. This approach reduced noncompliance and dropouts as a result of fatigue and dehydration from food and fluid deprivation.


In conclusion, the findings of the present study showed that Ramadan fasting during the month of September in a geographic time zone GMT +3, did not increase maternal risk for preterm delivery in healthy singleton pregnancies >20 weeks of gestation, despite a smaller gain in maternal weight, increased incidence of ketosis and ketonuria, and significant reduction in birthweight.

Since behavioural responses of Muslims to religious requirements are not always uniform, further research is greatly needed to investigate underlying cultural backgrounds and dietary habits, and evaluate their effects on maternal and fetal wellbeing. It is also important not to overlook the fact that the effects of fasting on pregnancy may not always be immediate. Concerns that carryover metabolic consequences could continue late into adulthood are real. In that respect, the present data alone may not be sufficient to ascertain that Ramadan fasting is safe during pregnancy. Whether the significant effect on birthweight is clinically relevant can only be confirmed by longitudinal studies to assess the effects of poor fetal growth on the origin of adult diseases according to the ‘Barker Hypothesis’.50 The build-up of similar research is highly valuable in offering healthcare providers new insights on addressing this issue, and providing directions to drive best practices in the future.

Disclosure of interest

None of the authors of this manuscript have conflicts of interest with this research.

Contribution to authorship

JA provided guidance regarding the study question, and helped with the writing up and editing of the article. IMU conducted the statistical analyses and helped with editing of the article. JS helped in data collection, entry and called women to arrange for collection of their blood and urine samples. KMM helped with the analysis, as well as the editing of the article. GG provided expert guidance on the article and helped in editing. AHN conceived the idea, developed the study objectives, and led the preparation of the article, including writing and editing.

Details of ethics approval

This study was approved by the Institutional Review Board at the American University of Beirut on 15 January 2008.


This study was funded by a grant from the Medical Practice Plan at the American University of Beirut, Beirut, Lebanon. (Principal investigator: Anwar H. Nassar, MD). The funding agency did not play any role in any aspect of the study.


We are grateful to Dr Dima Dandashi who helped in the recruitment of the women and in data entry, and to Miss Suad Katerji, the Laboratory Reception Supervisor, and her team who visited the women participating in this study at their homes to secure blood and urine samples.