Plain language summary
Screening for gestational diabetes and subsequent management for improving maternal and infant health
Gestational diabetes mellitus (GDM) is a form of diabetes that can develop during pregnancy. Having GDM increases the risk of complications during the rest of the pregnancy for the mother and her baby. Women with GDM are more likely to develop pre-eclampsia (high blood pressure and protein in the urine) and require a caesarean section. For the baby, potential problems include the baby growing larger than it normally would, causing difficulties with birth. The baby can also have low blood sugar levels after birth. Although GDM usually resolves following birth, both mother and child are at risk of developing type II diabetes in the future. There is strong evidence that treating GDM is beneficial and improves health outcomes.
It may therefore help if pregnant women are screened to identify as many as possible of those who do have GDM before they have symptoms, such as excessive thirst or urination, or fatigue. The two main approaches to screening are 'universal' where all women undergo a screening test for GDM; and 'selective' where only those women at 'high risk' are screened. The main risk factors are maternal age, high body mass index, family history and cigarette smoking. The different screening strategies used around the world to identify women with GDM include identifying women based on their risk factors, a blood sugar test one hour after a 50 g glucose drink, and random blood sugar measurements. It is however unclear whether screening for GDM leads to better health outcomes and if so, which screening strategy is the most appropriate.
This review included four trials involving 3972 women and their babies, and found that there is little high-quality evidence on the effects of screening for GDM on health outcomes for mothers and their babies. One trial compared risk factor screening with universal screening, and three trials evaluated different methods of administering a 50 g glucose load (the glucose load is used during the screening test). In one trial, women who were in the universal screening group were more likely to be diagnosed with GDM compared with women in the high-risk screening group. However, this trial was not of high quality. Few other differences between groups were shown in any of the trials. Further research is required to see which recommendations for screening practices for GDM are most appropriate.
Le dépistage et la prise en charge du diabète gestationnel pour une meilleure santé de la mère et de l'enfant
Le diabète sucré gestationnel (DSG) est une forme de diabète qui peut survenir pendant la grossesse. Le DSG augmente le risque de complications pendant le reste de la grossesse pour la mère et son bébé. Les femmes atteintes de DSG sont plus susceptibles de développer une pré-éclampsie (hypertension et protéine dans les urines) et de nécessiter une césarienne. Les complications potentielles pour l'enfant sont notamment une croissance anormalement importante, ce qui peut entraîner des difficultés lors de l'accouchement. Le bébé peut également présenter une hypoglycémie après la naissance. Bien que le DSG disparaisse généralement après l'accouchement, la mère comme l'enfant risquent de développer un diabète de type II par la suite. Tout porte à croire que le traitement du DSG apporte des bénéfices et améliore les résultats cliniques.
Par conséquent, il peut être utile de dépister les femmes enceintes afin d'identifier autant que possible celles atteintes de DSG avant qu'elles n'en développent les symptômes, tels qu'une soif ou une miction excessive ou la fatigue. Les deux approches principales de dépistage sont l'approche « universelle », lorsque toutes les femmes subissent un test de dépistage du DSG, et l'approche « sélective », lorsque seules les femmes à « risque élevé » sont dépistées. Les principaux facteurs de risque sont l'âge de la mère, un indice de masse corporelle élevé, les antécédents familiaux et le tabagisme. Les différentes stratégies de dépistage utilisées dans le monde pour identifier les femmes atteintes de DSG incluent l'identification des femmes d'après leurs facteurs de risque, un test de glycémie une heure après l'absorption d'une boisson contenant 50 g de glucose, et des mesures de glycémie aléatoires. Toutefois, on ne sait pas avec certitude si le dépistage du DSG améliore les résultats cliniques et, le cas échéant, quelle stratégie de dépistage est la plus efficace.
Cette revue a inclus quatre essais portant sur 3 972 femmes et leurs bébés, et a révélé qu'il existe peu de données de qualité élevée sur les effets du dépistage du DSG sur les résultats cliniques pour la mère et l'enfant. Un essai comparait le dépistage selon les facteurs de risque au dépistage universel, et trois essais évaluaient différentes méthodes d'administration d'une charge de 50 g de glucose (utilisée pendant le test de dépistage). Dans un essai, les femmes qui étaient dans le groupe de dépistage universel étaient plus susceptibles de recevoir un diagnostic de DSG, en comparaison avec les femmes dans le groupe de dépistage à haut risque. Cependant, cet essai n'était pas de qualité élevée. Peu d'autres différences entre les groupes ont été observées dans l'ensemble des essais. D'autres recherches sont nécessaires pour établir des recommandations quant aux meilleures pratiques de dépistage du diabète sucré gestationnel.
Notes de traduction
Traduit par: French Cochrane Centre 22nd June, 2014
Traduction financée par: Financeurs pour le Canada : Instituts de Recherche en Santé du Canada, Ministère de la Santé et des Services Sociaux du Québec, Fonds de recherche du Québec-Santé et Institut National d'Excellence en Santé et en Services Sociaux; pour la France : Ministère en charge de la Santé
Streszczenie prostym językiem
Badania przesiewowe oraz późniejsze postępowanie u kobiet z cukrzycą ciężarnych w celu poprawy zdrowia matki i dziecka
Cukrzyca ciężarnych (ang. Gestational Diabetes Mellitus, GDM) jest typem cukrzycy, który może rozwinąć się podczas ciąży. Zachorowanie na GDM zwiększa ryzyko powikłań podczas ciąży u matki i jej dziecka. U kobiet z GDM stwierdza się większe prawdopodobieństwo rozwoju stanu przedrzucawkowego (wysokiego ciśnienia krwi i obecności białka w moczu) i częściej istnieje potrzeba przeprowadzenia u nich cesarskiego cięcia. W przypadku dziecka potencjalne problemy obejmują zwiększenie rozmiarów dziecka ponad normę, co powodować może trudności podczas porodu. Dziecko może mieć również niski poziom cukru po urodzeniu. Chociaż GDM zwykle ustępuje po urodzeniu, zarówno matka oraz dziecko obciążone są ryzykiem rozwoju cukrzycy typu 2 w przyszłości. Istnieją silne dowody na to, że leczenie GDM jest korzystne i poprawia wyniki zdrowotne.
Badania przesiewowe kobiet ciężarnych, mogłyby okazać się przydatne w celu zidentyfikowania jak największej liczby kobiet chorujących na GDM jeszcze przed wystąpieniem objawów takich jak: nadmierne pragnienie, częste oddawanie moczu lub zmęczenie. Dwa główne podejścia do badań przesiewowych to podejście "uniwersalne", w którym wszystkie kobiety poddane są badaniu przesiewowemu pod kątem GDM; oraz podejście "wybiórcze", w którym badane są wyłącznie kobiety obciążone "dużym ryzykiem" wystąpienia cukrzycy. Główne czynniki ryzyka stanowią: wiek matki, duży wskaźnik masy ciała (BMI), dodatni wywiad rodzinny oraz palenie papierosów. Różne strategie badań przesiewowych wykorzystuje się na całym świecie do identyfikacji kobiet z GDM, obejmują one metody identyfikacji na podstawie czynników ryzyka, test obciążenia glukozą po jednej godzinie od wypicia płynu z 50 g glukozy oraz pomiary cukru we krwi w dowolnym momencie. Nie jest jednak jasne czy badania przesiewowe w kierunku GDM prowadzą do lepszych wyników zdrowotnych, a jeśli prowadzą, to która strategia badań przesiewowych jest najbardziej odpowiednia.
Do niniejszego przeglądu włączono cztery badania obejmujące 3972 kobiet oraz ich dzieci, stwierdzono, że istnieje niewiele doniesień o wysokiej jakości dotyczących wpływu badań przesiewowych w kierunku GDM na zdrowie matek oraz ich dzieci. W jednym badaniu porównywano badania przesiewowe u kobiet obciążonych czynnikami ryzyka z badaniem przesiewowym prowadzonym u wszystkich kobiet, w trzech innych badaniach porównywano różne metody podawania 50 g glukozy w teście obciążenia glukozą (obciążenie glukozą jest wykorzystywane podczas badania przesiewowego). W innym badaniu, u kobiet objętych uniwersalnym badaniem przesiewowym było większe prawdopodobieństwo rozpoznania GDM w porównaniu z grupą obejmującą badania przesiewowe jedynie u kobiet obciążonych dużym ryzykiem. Badanie to nie było jednak wysokiej jakości. Kilka innych różnic między grupami stwierdzono w opisanych badaniach. Potrzeba dalszych badań w celu sprawdzenia, które zalecenia dotyczące badań przesiewowych w kierunku GDM są najbardziej odpowiednie.
Uwagi do tłumaczenia
Tłumaczenie: Bartłomiej Matulewicz, Redakcja Paulina Rolska
Description of the condition
Gestational diabetes mellitus
Gestational diabetes mellitus (GDM) is defined as "carbohydrate intolerance of varying degrees of severity with onset or first recognition during pregnancy" (Metzger 1998). GDM therefore includes type I or type II diabetes previously undetected or with first presentation during pregnancy. GDM typically resolves following birth. However, these women are at risk for type II diabetes in the future (Kim 2002).
GDM affects up to 14% of pregnant women every year and accounts for 90% of pregnancies affected by diabetes mellitus (Coustan 1995; Setji 2005). There is growing concern over the increasing prevalence of GDM and its effects for individual mothers and infants and its impact on public health (Ferrara 2007; Hunt 2007; Metzger 2007). GDM is associated with numerous risk factors. Maternal age and body mass index (BMI) are among the most common risk factors (Di Cianni 2003; O'Sullivan 1973). Specific ethnicities are also at higher risk of developing GDM, namely Hispanic, black, Native American, South or East Asian, Pacific Islander and Indigenous Australian (Kjos 2005). These ethnicities are similar to those at high risk of type II diabetes mellitus, with suggestions that parallels may be drawn between these two forms of diabetes (Ben-Haroush 2004; Kuhl 1998). Other risk factors include previous birth of a large baby, a family history of diabetes mellitus, weight gain and cigarette smoking (Davey 2001; Di Cianni 2003; O'Sullivan 1973; Solomon 1997).
Normally, insulin is released by pancreatic beta cells in response to increasing blood glucose levels to achieve euglycaemia (normal blood glucose levels). This system can be disrupted in two ways. A problem with the release of insulin from beta cells can occur, such as in type I or insulin dependent diabetes mellitus. Alternatively, insulin may not act as effectively in promoting glucose uptake. This is known as insulin resistance, and is seen in the development of type II or non insulin dependent diabetes mellitus and GDM.
Placental hormones such as progesterone, cortisol, prolactin and human placental lactogen released mid-pregnancy contribute to decreased insulin action in pregnancy (Kuhl 1998). Physiologically, this ensures sufficient nutrient transport to the fetus as it develops, and promotes growth (Setji 2005). In a normal pregnancy, the action of these placental hormones is adequately compensated by increasing insulin release, creating an equilibrium between insulin supply and insulin demand.
In pregnant women with abnormal glucose intolerance, the insulin resistance of pregnancy is not adequately compensated for, resulting in carbohydrate or glucose intolerance. It is suggested that women who develop GDM may also have an underlying insulin resistance, such as high maternal adiposity, or beta cell dysfunction that potentiates the insulin resistance of pregnancy (Buchanan 2005; Kuhl 1998; Richardson 2007). Recent theories relating to the pathogenesis of GDM include inflammation (Richardson 2007).
These effects culminate in a disruption of the action of insulin in maintaining glucose levels, resulting in maternal hyperglycaemia (high blood glucose). Glucose is transferred, via the placenta, to the fetus. Maternal hyperglycaemia therefore stimulates a fetal hyperinsulinaemia to counter the excess placental glucose transfer.
There is strong evidence confirming the continuum of risk associated with increasing carbohydrate intolerance (Dodd 2007; HAPO 2008; Sermer 1995). The point at which this increasing carbohydrate intolerance becomes pathological remains uncertain.
Excess insulin due to maternal hyperglycaemia acts in two ways on the fetus. Firstly, insulin promotes fat deposition due to the state of nutrient excess (Pedersen 1954; Whitelaw 1977). Secondly, insulin acts as a growth factor, stimulating further growth of the infant in utero (Hunt 2007). Thus, fetal hyperinsulinaemia results in excessive growth of the fetus, leading to one of the major perinatal concerns in GDM, macrosomia (birthweight greater than 4000 g). Macrosomia may lead to birth trauma including shoulder dystocia, nerve palsies and fractures (Dodd 2007; Metzger 1998). GDM is associated with respiratory distress syndrome, neonatal hypoglycaemia (low blood glucose), hyperbilirubinaemia (high bilirubin levels), polycythaemia (excess red blood cells), and hypocalcaemia (low calcium) (ADA 2003; Metzger 1998). In utero exposure to hyperglycaemia has long lasting effects on the infant, increasing their risk of future obesity and type II diabetes mellitus (Pettitt 1985; Silverman 1998).
With the implementation of screening protocols, GDM is usually diagnosed before it becomes symptomatic during pregnancy. However, where GDM is undetected, the pregnant woman may experience polyuria (increased urinary frequency), polydipsia (excessive thirst) or fatigue. Macrosomia in utero or polyhydramnios (excess amniotic fluid volume) may also indicate GDM.
In the mother, evidence supports an association between GDM and increased rates of caesarean delivery and pre-eclampsia (ACOG 2001). As with their infants, the consequences of GDM for the mother extend beyond the perinatal period. There are strong links between GDM and future development of type II diabetes mellitus. Within 10 years of developing GDM, half the women develop type II diabetes mellitus (Kim 2002).
Diagnosis of GDM
Although diagnostic criteria vary (ACOG 2001; ADA 2003; Berger 2002; IADPSG 2010; NICE 2008; Oats 2004; RANZCOG 2008; WHO 1999), the oral glucose tolerance test (OGTT) is considered the ‘gold standard’ for diagnosis of GDM (Scott 2002). Minor degrees of abnormal carbohydrate tolerance, such as one abnormal value on OGTT or positive oral glucose challenge test (OGCT) with normal OGTT are also associated with similar outcomes to GDM. This is in line with the increasing awareness of the continuum of risk associated with increasing carbohydrate intolerance (Dodd 2007; HAPO 2008; Sermer 1995).
Management of GDM
The importance of management for women with GDM has been widely accepted and is evaluated by several Cochrane reviews (Alwan 2009; Boulvain 2001; Ceysens 2006) and the treatment of GDM is widely supported (ADA 2003; Crowther 2005; Hoffman 1998; Metzger 1998; O'Sullivan 1966). Treatment focuses on reducing the hyperglycaemia driving the complications of GDM (Metzger 1998). In general, management includes any or all of: nutritional therapy, exercise, blood glucose monitoring and insulin therapy. The results from two large, multi-centred randomised controlled trials provide strong support for the treatment of women with mild GDM (Crowther 2005; Landon 2009). Crowther and colleagues reported reduced infant morbidity in those treated for GDM in addition to suggesting that maternal quality of life was improved by treatment. Landon and colleagues showed treatment for GDM reduced the risks of fetal overgrowth, shoulder dystocia and pre-eclampsia.
Medical nutrition therapy
The American Diabetes Association and Australasian Diabetes in Pregnancy Society, in line with other governing bodies, recommend nutrition therapy in the treatment of GDM (ADA 2003; ADA 2007; Hoffman 1998). Both guidelines focus on managing carbohydrate intake for blood glucose maintenance.
Exercise is often used in conjunction with dietary therapy to maintain normal glucose levels. The Cochrane review ‘Exercise in diabetic pregnancy’ found that there was insufficient evidence to make a recommendation (Ceysens 2006). However, there is growing consensus on the safety of moderate exercise in pregnancy and its benefits in the treatment of GDM.
Blood glucose monitoring
Blood glucose monitoring is often recommended (ACOG 2001; Hoffman 1998). Postprandial hyperglycaemia monitoring demonstrates a closer association with rates of fetal macrosomia and obviously correlates with peaks of blood glucose (Hoffman 1998). Blood glucose monitoring provides the health professional with a representation of glycaemic control while providing the woman with feedback on her management progress.
Insulin/oral hypoglycaemic agents
Where the maternal hyperglycaemia cannot be managed by dietary or exercise advice and blood glucose levels remain elevated, insulin is added for greater control (Metzger 1998). The methods for administering insulin are discussed in the Cochrane review 'Continuous subcutaneous insulin infusion versus multiple daily injections of insulin for pregnant women with diabetes' (Farrar 2007). Oral hypoglycaemics such as glyburide (Langer 2000) and metformin (Rowan 2008) have been suggested as alternatives to insulin therapy.
The Cochrane review ‘Elective delivery in diabetic pregnant women’ suggested that induction of labour at 38 to 39 weeks may be suitable for diabetic women treated with insulin (Boulvain 2001).
It is recommended that women whose pregnancies were affected by GDM receive an OGTT between six and 12 weeks postpartum to detect diabetes (ACOG 2009; Berger 2002; Hoffman 1998; Metzger 2007; Oats 2004; RANZCOG 2008). Because of the high risk of future diabetes, these women are often advised to be retested on a regular basis (Hoffman 1998; Metzger 2007; Oats 2004; RANZCOG 2008).
Description of the intervention
A screening tool establishes the risk of disease in an otherwise well person (NSC 2007). Ordinarily, the presentation of symptoms prompts testing for disease. However, screening aims to identify the illness earlier, before symptoms arise. Identification of an illness by screening allows for earlier management, which may result in better health outcomes. While screening can be beneficial, it can also cause unnecessary anxiety due to the testing process itself. This is further complicated by the occurrence of false positives, where screening has suggested an increased risk for the disease but the diagnostic test does not show evidence of the disease.
An accepted screening process must first meet certain criteria (NSC 2003). In addition to the illness being an important health problem, the screening process must benefit the individual. This includes the acceptability of the screening process clinically, socially and ethically and the availability of an effective treatment. These benefits must outweigh any possible harms such as discomfort from any testing and costs of administering the screening process. From an economic perspective, the screening process must also be cost effective.
Screening does not always involve a clinical test, and may include, for example, a series of history questions. Furthermore, it is important to distinguish screening from diagnostic procedures which provide a diagnosis in an already symptomatic or high-risk individual. It is important to distinguish between a screening and diagnostic test. While a screening test will identify those at risk for a disease, diagnostic tests are generally designed to give a definitive yes or no. Diagnostic tests are also often more complex and expensive than screening tests.
While screening tools will identify those at risk of an illness, it is the subsequent management of the result that ultimately affects health outcomes. 'Screening' can be used to refer to an individual screening tool or to a screening program, protocol or guideline, which includes the screening tool and subsequent management such as diagnostic testing and treatment of any illness identified. By identifying individuals at high and low risk of a particular illness, a screening tool therefore identifies those who require diagnostic testing and those who do not. It therefore follows that the implementation of a screening program, which includes a combination of screening tool, subsequent diagnostic testing and management, is able to affect health outcome.
Screening for GDM
Whether to screen for GDM, and which methods to use, remain controversial. This is compounded by the lack of clearly defined, universally accepted screening criteria, and the uncertainty as to the severity of glucose intolerance at which treatment is beneficial. Even with screening protocols in place, GDM is diagnosed at the end of the second trimester or early third trimester based on physiology. This leaves little time for management of GDM. Without screening, the diagnosis of GDM, and therefore treatment, is potentially delayed.
Screening for GDM is often implemented despite the uncertainty of its utility. A wide variety of strategies have been employed in screening for GDM that provide varying degrees of sensitivity and specificity. Universal or routine screening, usually where all women are offered a 50 g OGCT, and risk factor screening (by women's history) are the most commonly used methods and combinations of these and other methods have been used to form various screening protocols (ACOG 2001; Gabbe 2004; Metzger 2007; Mires 1999; Rumbold 2001).
The OGCT was originally proposed by O’Sullivan and colleagues to provide a more sensitive screening process than risk factor screening (O'Sullivan 1973). An OGCT involves a 50 g glucose drink and a blood glucose measurement after one hour. While the predefined risk factors used vary between centres and countries, they commonly include maternal age, BMI, ethnicity, previous GDM and family history of diabetes mellitus (ADA 2003; Berger 2002; Hoffman 1998; Metzger 2007; USPSTF 2008). Other methods used to screen for GDM include urine testing for glucosuria, fructosamine testing, random plasma glucose measurements, fasting plasma glucose measurements and HbA1c (a measure of how well blood glucose has been controlled over the previous two to three months) (Scott 2002).
The variation in screening protocols is reflected in surveys conducted around the world. In a UK survey of obstetric units in 1996, it was found that 89% screened for GDM, with 81% of those units using risk factor based screening (Mires 1999). There was a lack of consensus on the appropriate screening method (Mires 1999). In a similar Australian survey of obstetric practice conducted in 1999, it was found that 87% of the obstetric population was being screened for GDM. Again there was no strong consensus on how to screen (Rumbold 2001). An American survey in 2004 found that 95.2% of obstetricians screening for GDM adopted a universal one-hour 50 g OGCT (Gabbe 2004). This diversity in preferred screening protocols may reflect a number of factors, such as the cost of screening, the expected prevalence of GDM and test accuracy, in addition to the lack of definitive evidence in favour of a particular screening protocol.
A largely accepted time for screening is the end of the second trimester, ranging between 24 to 28 weeks' gestation (ACOG 2001; Hoffman 1998; Metzger 2007; Oats 2004; RANZCOG 2008). This value reflects a balance between having adequate time to manage GDM and the ability to detect the development of carbohydrate intolerance (ACOG 2001; Brody 2003). There is little evidence on the benefits and detriments of screening prior to 24 weeks' gestation (USPSTF 2008).
The negative impact of screening for GDM also needs to be considered. The importance of identification of GDM should be weighed against any discomfort experienced by the woman and anxiety from testing. While the 50 g OGCT is considered to be a quick and simple test, it is unpleasant to drink and is associated with side effects such as dizziness, headaches, nausea and vomiting, and requires a blood test. For many women, the inconvenience of testing can be significant. Screening by any method can create anxiety for the mother, including women identified as having risk factors for GDM, those identified through routine OGCT and those with elevated random blood glucose levels. In particular, a false positive result has been associated with a decline in women's perception of health (Kerbel 1997; Rumbold 2002). It also follows that with the introduction of screening, that more women are offered diagnostic testing, usually an OGTT, which requires them to fast overnight, drink a higher glucose load, requires more blood tests and can take up to three hours to complete.
An evaluation of cost is imperative with any screening procedure. While screening processes may affect detection, management and therefore improve maternal and infant health, this must also be weighed against the cost of screening all pregnant women, any subsequent diagnostic tests and treatment for additional women diagnosed with GDM.
Why it is important to do this review
Many believe that the incidence of GDM, the adverse outcomes arising from GDM and the benefits of treatment suggest a need for some screening process. However, high-quality evidence demonstrating the effectiveness of screening on detection of GDM and subsequent maternal and infant health is required for screening to be implemented (NSC 2003). Whether a screening protocol adequately identifies those at risk of GDM, and whether this knowledge improves the health outcomes for women with GDM and their babies through subsequent management of a screening result, are important factors to consider when recommending a screening process. It is also equally important that a recommended screening protocol does not harm women without GDM. Moreover, given the lack of consensus on a method for screening, an evaluation of the different screening protocols on the detection of GDM and subsequent maternal and infant health is required.
This review updates a previously published Cochrane review on screening and subsequent management for GDM for improving maternal and infant health (Tieu 2010). The previous review included four randomised trials and found that there was insufficient evidence to determine if screening for GDM, or what types of screening, can improve maternal and infant health outcomes; and concluded that high-quality, large trials are required in this area.
This review update therefore aims to evaluate the current evidence regarding the effects of screening for GDM as an intervention on maternal and infant health. The evaluation of test performance of individual screening methods is not included in the scope of this review. The strategies for diagnosis of GDM have been evaluated in a separate Cochrane review 'Alternative strategies for glucose tolerance testing to diagnose gestational diabetes and impaired glucose tolerance during pregnancy' (Farrar 2011). The Cochrane review 'Treatments for gestational diabetes' (Alwan 2009) assesses management after diagnosis of GDM or impaired glucose tolerance and therefore is looking at a later stage in the pathways of care and management than this review which addresses the process from screening onwards
To assess the effects of different methods of screening for gestational diabetes mellitus on maternal and infant outcomes.
Criteria for considering studies for this review
Types of studies
Randomised controlled trials, quasi-randomised controlled trials and cluster-randomised trials. We planned to exclude cross-over trials. We planned to include studies published as abstracts provided there was sufficient information to allow us to assess study eligibility and risk of bias. If sufficient information was not available, the study would await assessment pending the publication of the full trial report, or the provision of further information by trial authors.
Types of participants
Pregnant women, excluding women who have already been diagnosed with GDM in this pregnancy or who have pre-existing diabetes mellitus.
Types of interventions
Any individual screening tool or screening program, protocol or guideline for GDM compared with the absence of screening; or any individual screening tool or screening program, protocol or guideline for GDM with another.
Types of outcome measures
Diagnosis of GDM*;
positive screen for GDM*;
mode of birth (normal vaginal birth, operative vaginal birth, caesarean section).
Large-for-gestational age (birthweight greater than or equal to 90th percentile);
macrosomia (greater than 4000 g or greater than 4500 g).
* as defined by author(s)
induction of labour;
weight gain in pregnancy;
augmentation of labour;
insulin or oral hypoglycaemic agent required to treat GDM;
women who screen positive and are not subsequently diagnosed with GDM;
women's sense of well-being and quality of life*.
Development of type II diabetes mellitus;
GDM in subsequent pregnancy;
development of type I diabetes mellitus;
impaired glucose tolerance*;
body mass index (BMI);
BMI greater than 25;
BMI greater than 30;
women's sense of well-being and quality of life*.
death of liveborn infants prior to hospital discharge;
infant death (up to one year of life);
gestational age at birth;
preterm birth (less than 37 weeks' gestation);
respiratory distress syndrome;
hypoglycaemia requiring treatment;
hyperbilirubinaemia requiring treatment;
five minute Apgar score less than seven;
five minute Apgar score less than four.
BMI greater than 25;
BMI greater than 30;
fat mass/fat-free mass;
skinfold thickness measurements;
impaired glucose tolerance*;
type I diabetes;
type II diabetes;
BMI greater than 25;
BMI greater than 30;
fat mass/fat-free mass;
skinfold thickness measurements;
impaired glucose tolerance*;
type I diabetes;
type II diabetes;
Acceptability of testing
Adverse effects of testing (e.g. nausea, vomiting);
women's acceptance of screening protocol*.
Cost of screening each woman;
number of hospital visits/antenatal visits for mother;
medical physician visits;
length of postnatal stay (mother);
length of postnatal stay (baby);
cost of maternal care;
cost of offspring care.
* as defined by author(s)
Search methods for identification of studies
We contacted the Trials Search Co-ordinator to search the Cochrane Pregnancy and Childbirth Group’s Trials Register (1 December 2013).
The Cochrane Pregnancy and Childbirth Group’s Trials Register is maintained by the Trials Search Co-ordinator and contains trials identified from:
quarterly searches of the Cochrane Central Register of Controlled Trials (CENTRAL);
weekly searches of MEDLINE;
weekly searches of Embase;
handsearches of 30 journals and the proceedings of major conferences;
weekly current awareness alerts for a further 44 journals plus monthly BioMed Central email alerts.
Details of the search strategies for CENTRAL, MEDLINE and Embase, the list of handsearched journals and conference proceedings, and the list of journals reviewed via the current awareness service can be found in the ‘Specialized Register’ section within the editorial information about the Cochrane Pregnancy and Childbirth Group.
Trials identified through the searching activities described above are each assigned to a review topic (or topics). The Trials Search Co-ordinator searches the register for each review using the topic list rather than keywords.
We did not apply any language restrictions.
Data collection and analysis
For methods used in the previous version, please see Tieu 2010.
For this update, the following methods were used.
Selection of studies
Two review authors independently assessed for inclusion all the potential studies we identified as a result of the search strategy. We resolved any disagreement through discussion or, if required, we consulted a third review author.
Data extraction and management
We designed a form to extract data. For eligible studies, two review authors extracted the data using the agreed form. We resolved discrepancies through discussion or, if required, we consulted a third review author. We entered data into Review Manager software (RevMan 2012) and checked for accuracy.
When information regarding any of the above was unclear, we attempted to contact authors of the original reports to provide further details.
Assessment of risk of bias in included studies
Two review authors independently assessed risk of bias for each study using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We resolved any disagreement by discussion or by involving a third author.
(1) Random sequence generation (checking for possible selection bias)
We described for each included study the methods used to generate the allocation sequence in sufficient detail to allow an assessment of whether it should produce comparable groups.
We assessed the methods as:
low risk of bias (any truly random process, e.g. random number table; computer random number generator);
high risk of bias (any non-random process, e.g. odd or even date of birth; hospital or clinic record number);
unclear risk of bias.
(2) Allocation concealment (checking for possible selection bias)
We described for each included study the method used to conceal the allocation sequence and determined whether intervention allocation could have been foreseen in advance of, or during recruitment, or changed after assignment.
We assessed the methods as:
low risk of bias (e.g. telephone or central randomisation; consecutively numbered sealed opaque envelopes);
high risk of bias (open random allocation; unsealed or non-opaque envelopes, alternation; date of birth);
unclear risk of bias.
(3.1) Blinding of participants and personnel (checking for possible performance bias)
We described for each included study, the methods, if any, used to blind study participants and personnel from knowledge of which intervention a participant received. We considered studies to be at a low risk of bias if they were blinded, or if we judged that the lack of blinding would be unlikely to affect results. We assessed blinding separately for different outcomes or classes of outcomes.
We assessed the methods as:
low, high or unclear risk of bias for participants;
low, high or unclear risk of bias for personnel.
(3.2) Blinding of outcome assessment (checking for possible detection bias)
We described for each included study the methods used, if any, to blind outcome assessors from knowledge of which intervention a participant received. We assessed blinding separately for different outcomes or classes of outcomes.
We assessed methods used to blind outcome assessment as:
(4) Incomplete outcome data (checking for possible attrition bias due to the amount, nature and handling of incomplete outcome data)
We described for each included study and for each outcome or class of outcomes, the completeness of data including attrition and exclusions from the analysis. We stated whether attrition and exclusions were reported, the numbers included in the analysis at each stage (compared with the total randomised participants), reasons for attrition or exclusion where reported, and whether missing data were balanced across groups or were related to outcomes. Where sufficient information was reported or was supplied by the trial authors, we included missing data in the analyses which we undertook.
We assessed the methods as:
low risk of bias (e.g. where there were no missing data or where reasons for missing data were balanced across groups);
high risk of bias (e.g. numbers or reasons for missing data imbalanced across groups; 'as treated' analysis done with substantial departure of intervention received from that assigned at randomisation);
unclear risk of bias.
(5) Selective reporting bias (checking for reporting bias)
We described for each included study how the possibility of selective outcome reporting bias was examined by us and what we found.
We assessed the methods as:
low risk of bias (where it was clear that all of the study’s pre-specified outcomes and all expected outcomes of interest to the review had been reported);
high risk of bias (where not all the study’s pre-specified outcomes had been reported; one or more reported primary outcomes were not pre-specified; outcomes of interest were reported incompletely and so could not be used; study failed to include results of a key outcome that would have been expected to have been reported);
unclear risk of bias.
(6) Other sources of bias (checking for bias due to problems not covered by (1) to (5) above)
We described for each included study any important concerns we had about other possible sources of bias. We assessed whether each study was free of other problems that could put it at risk of bias:
(7) Overall risk of bias
We made explicit judgements about whether studies were at a high risk of bias, according to the criteria given in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). With reference to (1) to (6) above, we assessed the likely magnitude and direction of the bias and whether we considered it is likely to impact on the findings. We planned to explore the impact of the level of bias through undertaking sensitivity analyses - see Sensitivity analysis.
Measures of treatment effect
For dichotomous data, we presented results as risk ratio with 95% confidence intervals.
For continuous data, we used the mean difference when outcomes were measured in the same way between trials. If necessary, we would have used the standardised mean difference to combine trials that measured the same outcome, but used different methods.
Unit of analysis issues
We planned to include cluster-randomised trials in the analyses along with individually-randomised trials. We planned to adjust their sample sizes using the methods described in the Handbook using an estimate of the intracluster correlation co-efficient (ICC) derived from the trial (if possible), from a similar trial or from a study of a similar population. If we had used ICCs from other sources, we planned to report this and conduct sensitivity analyses to investigate the effect of variation in the ICC. If we had identified both cluster-randomised trials and individually-randomised trials, we planned to synthesise the relevant information. We would have considered it reasonable to combine the results from both if there was little heterogeneity between the study designs and the interaction between the effect of intervention and the choice of randomisation unit was considered to be unlikely.
We planned to also acknowledge heterogeneity in the randomisation unit and perform a sensitivity analysis to investigate the effects of the randomisation unit.
We considered cross-over trials inappropriate for this research question.
Dealing with missing data
For included studies, we noted levels of attrition. We planned to explore the impact of including studies with high levels of missing data in the overall assessment of treatment effect by using sensitivity analysis.
For all outcomes, we carried out analyses, as far as possible, on an intention-to-treat basis, i.e. we attempted to include all participants randomised to each group in the analyses, and all participants were analysed in the group to which they were allocated, regardless of whether or not they received the allocated intervention. The denominator for each outcome in each trial was the number randomised minus any participants whose outcomes were known to be missing.
Assessment of heterogeneity
We assessed statistical heterogeneity in each meta-analysis using the Tau², I² and Chi² statistics. We regarded heterogeneity as substantial where the I² was greater than 30% and either the Tau² was greater than zero, or there was a low P value (less than 0.10) in the Chi² test for heterogeneity.
Assessment of reporting biases
In future updates of this review, if there are 10 or more studies in the meta-analysis, we will investigate reporting biases (such as publication bias) using funnel plots. We will assess funnel plot asymmetry visually. If asymmetry is suggested by a visual assessment, we will perform exploratory analyses to investigate it.
We carried out statistical analysis using the Review Manager software (RevMan 2011). We used fixed-effect meta-analysis for combining data where it was reasonable to assume that studies were estimating the same underlying treatment effect: i.e. where trials were examining the same intervention, and the trials’ populations and methods were judged sufficiently similar. Where there was clinical heterogeneity sufficient to expect that the underlying treatment effects differed between trials, or where substantial statistical heterogeneity was detected, we used random-effects meta-analysis to produce an overall summary, if an average treatment effect across trials was considered clinically meaningful. The random-effects summary was treated as the average range of possible treatment effects and we have discussed the clinical implications of treatment effects differing between trials. If the average treatment effect was not clinically meaningful, we would not have combined trials.
Where we have used random-effects analyses, we have presented the results as the average treatment effect with its 95% confidence interval, and the estimates of Tau² and I²
Subgroup analysis and investigation of heterogeneity
If we had identified substantial heterogeneity, we planned to investigate it using subgroup analyses and sensitivity analyses. We planned to consider whether an overall summary was meaningful, and if it was, use random-effects analysis to produce it.
We planned an assessment of trials comparing any screening protocol with none, with data analysed separately for different methods of screening. We analysed trials comparing one method of screening with another, with data from different comparisons analysed separately.
We planned to carry out the following subgroup analyses for primary outcomes:
high risk for GDM (variously defined) (we will explore risk factors individually if sufficient data become available);
gestational age at screening (less than 24 weeks, 24 to 30 weeks, 30 weeks or more);
number of stages in the screening protocol;
type of management protocol.
There were insufficient data to conduct subgroup analyses.
We planned to assess subgroup differences by interaction tests available in within RevMan (RevMan 2012). We planned to report the results of the subgroup analysis quoting the Chi² statistic and P value, and the interaction test I² value.
We planned to carry out sensitivity analysis to explore the effect of trial quality assessed by concealment of allocation, by excluding studies with clearly inadequate allocation of concealment (rated high risk of bias).
We planned to then carry out sensitivity analysis to explore the effect of trial quality on primary outcomes. This would have involved analysis based on an assessment of selection bias and attrition bias. We planned to exclude studies of poor quality in the analysis (those rating unclear or high risk) in order to assess for any substantive difference in the overall result.
There were insufficient data to conduct sensitivity analyses.
Gestational diabetes mellitus (GDM) is widely accepted as a serious health issue, associated with serious maternal and infant morbidity. Recent, high-quality evidence exists to suggest that treatment of GDM is beneficial (Crowther 2005; Landon 2009). The variation in screening practices and guidelines, both in national surveys and worldwide, further demonstrates the need for large, high-quality trials evaluating the effect of screening for GDM (ACOG 2001; ADA 2003; CDA 2008; Gabbe 2004; Hanna 2008; Mires 1999; NICE 2008; Oats 2004; RANZCOG 2008; Rumbold 2001). Despite its use, this review found little evidence on the effect of screening on maternal and infant health outcomes. This is consistent with other reviews on screening for GDM (Hillier 2008; Hollander 2007; NICE 2008; Scott 2002).
We included four trials evaluating various methods of screening for GDM, which can be considered in two categories. Griffin 2000 evaluated different screening protocols. Bergus 1992, Martinez Collado 2003 and Murphy 1994 compared different types of glucose for a 50 g OGCT.
Griffin 2000 was a large quasi-randomised trial comparing universal screening by 50 g OGCT with risk factor screening and reported on diagnosis of GDM, positive screening for GDM and gestational age at birth. Of the 1299 women in the universal screening group who completed an OGCT, 366 were referred for an oral glucose tolerance test (OGTT). By comparison, fewer women (249 of the 1853 women in the risk factor group) were referred for an OGTT. Unsurprisingly, with more women offered diagnostic testing for GDM in the universal screening group, more women were diagnosed with GDM. It is, however, difficult to interpret how these increases in diagnoses and subsequent management translate to clinical maternal and infant health outcomes in the review. The marginal reduction seen in gestational age at birth is likely a result of the large sample size rather than a clinically relevant difference. The trial reported on the remaining health outcome data with women reorganised into three groups, those who were not diagnosed with GDM, women who were diagnosed with GDM from the universal screening group and those diagnosed with GDM in the risk factor screening group.
While more women received glucose testing in the universal screening group, compared to those in the risk factor group, Griffin 2000 did not report on women's views on the two forms of screening. In addition to side effects of receiving a glucose load, practical issues for testing and the anxiety from screening and false positive results, the anxiety created by receiving an abnormal result on screening compared with receiving information on background risk of GDM is also important in selecting an appropriate screening protocol. Women's views on their health status and the screening protocol would be further affected by whether or not they are subsequently diagnosed and managed for GDM. Given that some screening protocols will result in more women being diagnosed with GDM than others, this may influence women's views on their health status.
Furthermore, although management of diagnosed GDM improves maternal and infant health outcomes, the diagnosis of GDM may also be associated with increased intervention or monitoring such as induction of labour and neonatal intensive care unit admission (Alwan 2009). Therefore, with various protocols for screening, the subsequent management of women with a positive or negative screening result as part of these protocols may impact on maternal and infant health. Large studies are required to address these issues and given that only a proportion of women are subsequently diagnosed with GDM, these trials require sufficient power for subgroup analyses by diagnosis of GDM are meaningful.
Interestingly, women in the risk factor group with risk factors for GDM received an oral glucose tolerance test at 32 weeks. Women in the universal screening group underwent a glucose challenge test at 26 to 28 weeks' gestation and were referred for subsequent OGTT if this was positive. Further to this, if glucose testing was negative, repeat screening was performed where the woman had risk factors for GDM. The risk factor group is therefore more likely to be diagnosed at a later stage in pregnancy than those in the universal screening group. It is unclear, however, if the timing of glucose testing has affected health outcomes.
The trials (Bergus 1992; Martinez Collado 2003; Murphy 1994) evaluating different types of glucose for the challenge test were primarily interested in the side effects and number of women diagnosed by different forms of glucose. Therefore, unlike Griffin 2000, the women in these trials were all offered the same diagnostic testing pathway regardless of the screening result. Although no health outcome data were included in the review, data on health outcome would be unlikely to be affected by the screening test because the women received the same subsequent management.
Bergus 1992 and Murphy 1994 included a comparison of glucose monomer (d-glucose) and glucose polymer. Overall, fewer side effects were reported by women receiving a glucose polymer drink than glucose monomer. There was, however, significant heterogeneity in this result, with a stronger effect seen in Murphy 1994 than Bergus 1992. This may be explained by the different approaches taken by the trials, with Murphy 1994 only reporting side effects where they were rated moderate to severe. Bergus 1992 however used a binary approach. The differences in data presentation therefore limit interpretation of these outcomes. Maternal and infant health outcomes were not recorded by the trials. Although not statistically significant, the glucose polymer group generally reported fewer side effects in the moderate to severe range than the candy bar group.
Interpretation of the results of this arm of the review are, however, limited by the number of participants in each trial. Although women were screened at 24 to 28 weeks' gestation in the trials, Murphy 1994 also included women in their first trimester with risk factors for GDM and Martinez Collado 2003 included women with a 'high-risk' pregnancy. Both trials used a 100 g OGTT for diagnosis of GDM.
The results of this review were limited by the number of participants and methodological quality of the trials, which, overall, was assessed to be moderate to low. Therefore, the results of this review are to be interpreted with caution. Bergus 1992, Martinez Collado 2003 and Murphy 1994 included a total of 230 women and were both primarily interested in the side effects and numbers of women diagnosed by different forms of glucose. As a consequence, most of the maternal and infant health outcomes included in this review were not reported by these trials. While Griffin 2000 was a large quasi-randomised trial enrolling 3792 women, the format of outcome reporting precluded the outcome data of interest for this review from being included.
The trials included in this review were conducted in the United States, Ireland and Mexico. Geographical location, socio-economic circumstances and ethnicity are important factors that can alter the most appropriate method of screening with regard to feasibility and practicality. Compliance is probably not only related to side effects, as investigated by Bergus 1992, Martinez Collado 2003 and Murphy 1994, but also the requirements for the screening protocol. For example, the inconvenience posed by a one-hour 50 g OGCT should be considered against random capillary blood glucose testing or risk factor screening. This emphasises the need for future research to report not only on subsequent management and maternal and infant health outcomes, but also on the acceptability of and adherence to particular screening protocols.
One trial (Bebbington 1999) published as an abstract, and awaiting assessment, also compared universal and selected screening in 2401 women with no reported risk factors for GDM. In abstract form, there were insufficient data to include results in the review. The abstract reports no significant difference in birthweight and the incidence of macrosomia between the two groups.
We thank Emily Bain, Australian Research Centre for Health of Women and Babies, The University of Adelaide, for her support for updating this review.
We thank Dr George Bergus for contributing extra information and clarification on Bergus 1992. We also acknowledge the efforts of Dr Richard Firth and Dr Michael Bebbington in locating extra data.
The National Institute for Health Research (NIHR) is the largest single funder of the Cochrane Pregnancy and Childbirth Group. The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the NIHR, NHS or the Department of Health.
Contributions of authors
For this update, Joanna Tieu and Philippa Middleton assessed the new studies for inclusion/exclusion. Joanna Tieu and Philippa Middleton drafted changes to the text, and Caroline Crowther and Andy McPhee commented on drafts and contributed to the final version.