Probiotic treatment for women with gestational diabetes to improve maternal and infant health and well‐being

Karaponi AM Okesene-Gafa; Abigail E Moore; Vanessa Jordan; Lesley McCowan; Caroline A Crowther

doi:10.1002/14651858.CD012970.pub2

Tratamiento con probióticos para mujeres con diabetes gestacional para mejorar la salud y el bienestar materno‐infantiles

Declaraciones de intereses de los autores

Versión publicada: 24 junio 2020 Historial de versiones

https://doi.org/10.1002/14651858.CD012970.pub2

Contraer todo Desplegar todo

Resumen

disponible en

Antecedentes

La diabetes mellitus gestacional (DMG) es una intolerancia a los carbohidratos que se detecta por primera vez durante el embarazo y se asocia con complicaciones para las madres y los neonatos. Los probióticos son microorganismos vivos naturales que, cuando se ingieren en la cantidad suficiente, pueden aportar beneficios para la salud. La evidencia del papel de los probióticos como tratamiento para la DMG es limitada.

Objetivos

Evaluar la seguridad y la efectividad de los probióticos en el tratamiento de las mujeres con DMG sobre los resultados materno‐infantiles.

Métodos de búsqueda

Se hicieron búsquedas en el registro de ensayos del Grupo Cochrane de Embarazo y Parto (Cochrane Pregnancy and Childbirth Group) ClinicalTrials.gov en la Plataforma de registros internacionales de ensayos clínicos de la OMS (ICTRP) (24 de julio de 2019) y en las listas de referencias de los estudios recuperados.

Criterios de selección

Ensayos controlados aleatorizados (ECA) que comparan el uso de probióticos versus placebo/cuidado estándar para el tratamiento de la DMG.

Obtención y análisis de los datos

Dos autores de la revisión evaluaron de forma independiente la elegibilidad del estudio, extrajeron los datos, comprobaron la exactitud de los mismos y evaluaron el riesgo de sesgo de los ensayos incluidos. La certeza de la evidencia para determinados resultados materno‐infantiles se evaluó mediante el uso de GRADE.

Resultados principales

Se identificaron nueve ECA (695 embarazadas con DMG) que compararon los probióticos con el placebo. El riesgo general de sesgo en los nueve ECA fue de bajo a incierto y la evidencia se rebajó por imprecisión, debido al bajo número de mujeres que participaron en los ensayos. Los ensayos se llevaron a cabo en hospitales y universidades de Irán (siete ensayos), Tailandia (un ensayo) e Irlanda (un ensayo). En todos se compararon los probióticos con placebo.

Resultados maternos

No se sabe con certeza si los probióticos tienen algún efecto en comparación con el placebo en los trastornos hipertensivos del embarazo, (riesgo relativo (RR) 1,50; intervalo de confianza (IC) del 95%: 0,64 a 3,53; participantes = 256; estudios = 3; evidencia de certeza baja) y en la modalidad de nacimiento por cesárea (RR promedio 0,64; IC del 95%: 0,30 a 1,35; participantes = 267; estudios = 3; evidencia de certeza baja) porque la certeza de la evidencia es baja y los IC del 95% abarcan el posible beneficio y el posible daño.

Ningún ensayo informó resultados primarios de: modo de nacimiento como vaginal/asistido y desarrollo posterior de diabetes tipo 2.

No se sabe si los probióticos tienen algún efecto en comparación con el placebo en la inducción del trabajo de parto (RR 1,33; IC del 95%: 0,74 a 2,37; participantes = 127; estudios = 1; evidencia de certeza muy baja).

En cuanto a otros resultados maternos secundarios, no se sabe con certeza si hay diferencias entre los probióticos y el placebo para la hemorragia posparto; el aumento de peso durante la intervención del embarazo y el aumento total de peso durante la gestación; glucosa plasmática en ayunas y necesidad de farmacoterapia adicional (insulina). Los probióticos podrían asociarse con una ligera reducción de los triglicéridos y del colesterol total.

En la comparación de probióticos y placebo, hubo evidencia de reducción de los marcadores de resistencia a la insulina (HOMA‐IR) y HOMA‐B; y de la secreción de insulina. También hubo un aumento del índice de comprobación de la sensibilidad cuantitativa a la insulina (del inglés QUICKI).

Los probióticos se asociaron con beneficios menores en los marcadores biológicos pertinentes, con evidencia de una reducción en los marcadores inflamatorios de la proteína C reactiva de alta sensibilidad (hs‐CRP), de la interleucina 6 (IL‐6) y del marcador del malondialdehído del estrés oxidativo; y un aumento del glutatión total antioxidante, pero no se sabe si hay alguna diferencia en la capacidad antioxidante total.

Ningún ensayo informó resultados secundarios de traumatismo perineal, retención de peso postnatal o retorno al peso anterior al embarazo y depresión postnatal.

Resultados del lactante/niño/adulto

No se sabe si los probióticos tienen algún efecto en comparación con el placebo en el riesgo de recién nacidos grandes para su edad gestacional (RR 0,73; intervalo de confianza (IC) del 95%: 0,35 a 1,52; participantes = 174; estudios = 2; evidencia de certeza baja) y en la modalidad de nacimiento por cesárea (RR promedio 0,85; IC del 95%: 0,39 a 1,84; participantes = 177; estudios = 3; evidencia de certeza baja) porque la certeza de la evidencia es baja y los IC del 95% abarcan el posible beneficio y el posible efecto perjudicial.

Ningún ensayo informó de resultados primarios de: mortalidad perinatal (fetal/neonatal); o discapacidad neurosensorial.

En cuanto a otros resultados secundarios, no se sabe con certeza si hay alguna diferencia entre los probióticos y el placebo en cuanto a la edad gestacional al nacer, el nacimiento prematuro, la macrosomía, el peso al nacer, el perímetro cefálico, la talla, la hipoglucemia infantil y los ingresos en la unidad de cuidados intensivos neonatales (UCIN).

Hubo evidencia de una reducción de la hiperbilirrubinemia infantil con probióticos en comparación con el placebo.

Ningún ensayo informó de resultados secundarios: adiposidad del lactante y adiposidad infantil posterior.

Ningún ensayo informó eventos adversos.

Conclusiones de los autores

La evidencia de baja certeza significa que no se sabe con seguridad si hay alguna diferencia entre los grupos de probióticos y de placebo en los trastornos hipertensivos maternos del embarazo, las cesáreas y los recién nacidos grandes para la edad gestacional

Ningún ensayo informó eventos adversos.

Debido a la variabilidad de los probióticos utilizados y al pequeño tamaño de las muestras de los ensayos, la evidencia de esta revisión tiene una capacidad limitada para informar la práctica. Se necesitan ensayos bien diseñados y con la potencia estadística suficiente para identificar si los probióticos pueden mejorar los niveles de glucosa en sangre de la madre o los resultados del recién nacido/niño/adulto; y si pueden utilizarse para tratar la DMG.

PICO

Population

Intervention

Comparison

Outcome

El uso y la enseñanza del modelo PICO están muy extendidos en el ámbito de la atención sanitaria basada en la evidencia para formular preguntas y estrategias de búsqueda y para caracterizar estudios o metanálisis clínicos. PICO son las siglas en inglés de cuatro posibles componentes de una pregunta de investigación: paciente, población o problema; intervención; comparación; desenlace (outcome).

Para saber más sobre el uso del modelo PICO, puede consultar el Manual Cochrane.

Resumen en términos sencillos

disponible en

Probióticos como tratamiento adicional para la diabetes gestacional para mejorar los resultados de la madre y su hijo

¿Cuál es el problema?

La diabetes mellitus gestacional (DMG) es la intolerancia a los carbohidratos que provoca concentraciones altas de glucosa en sangre y que se detecta por primera vez durante el embarazo. Las mujeres embarazadas con DMG corren el riesgo de sufrir hipertensión, inducción del parto y cesáreas. Sus hijos corren el riesgo de nacer grandes, tener dificultades en el nacimiento, sufrir problemas respiratorios, tener un nivel bajo de glucosa en sangre al nacer y de padecer ictericia que puede causar lesiones cerebrales. Hay un mayor riesgo de que la madre tenga diabetes a largo plazo y de que el recién nacido tenga sobrepeso. Los probióticos son microorganismos que se encuentran de forma natural en los alimentos y en la leche fermentada, el yogur o en cápsulas. Existen muchos probióticos diferentes; los dos más utilizados son Lactobacillus y Bifidobacterium, y si se consumen en cantidades adecuadas, pueden aportar beneficios para la salud.

¿Por qué es esto importante?

Los probióticos deben ser seguros y los niveles de glucosa en sangre de la madre deben controlarse cuidadosamente durante el embarazo.

Como tratamiento inicial, las mujeres con DMG pueden recibir formación sobre hábitos alimentarios y actividad física, junto con el control de los niveles de glucosa en sangre. Cuando los niveles de glucosa en sangre están por encima de un umbral concreto, a las mujeres con DMG se les prescriben fármacos para reducir la glucosa, incluyendo la metformina o la insulina. El objetivo de esta revisión fue determinar la seguridad y la efectividad de los probióticos en el tratamiento de las mujeres con diabetes gestacional.

¿Qué evidencia se encontró?

Se buscó evidencia de ensayos controlados aleatorizados (última búsqueda en julio de 2019). Se identificaron nueve estudios con 695 mujeres con DMG. En todos se compararon los probióticos con placebo. Se evaluó la certeza de la evidencia como baja o muy baja. El riesgo general de sesgo fue de bajo a incierto.

Siete ensayos se realizaron en Irán, uno en Tailandia y uno en Irlanda. Los ensayos tuvieron lugar en hospitales y universidades.

No se sabe si hay alguna diferencia entre el probiótico y el placebo en las tasas de trastornos por presión arterial alta (tres estudios, 256 participantes, evidencia de certeza baja), cesárea (tres estudios, 267 mujeres, evidencia de certeza baja), y recién nacidos grandes para la edad gestacional (dos estudios, 174 participantes, evidencia de certeza baja).

No se sabe si existe alguna diferencia entre el probiótico y el placebo en la inducción del trabajo de parto (un estudio, 127 participantes, evidencia de certeza muy baja) y en los niveles bajos de glucosa en la sangre del recién nacido (tres estudios, 177 participantes, evidencia de certeza baja). Tampoco se sabe si hay diferencias entre los probióticos y el placebo para el sangrado abundante inmediatamente después del nacimiento, el aumento de peso durante el embarazo o el aumento total de peso durante la gestación.

No se sabe si existe alguna diferencia en la glucosa en sangre en ayunas entre los probióticos y el placebo (siete estudios, 554 participantes). Los probióticos podrían asociarse con una ligera reducción de los triglicéridos y del colesterol total (cuatro estudios, 320 participantes). Hubo una reducción en la secreción de insulina con los probióticos (siete estudios, 505 participantes). Un ensayo (60 participantes) no mostró diferencias entre los grupos que necesitaban insulina.

Los biomarcadores mostraron una reducción en la resistencia a la insulina (HOMA‐IR), (siete estudios, 505 participantes) y en la resistencia a la insulina y en la función de células β (HOMA‐B) (dos estudios, 130 participantes) con los probióticos. El índice de comprobación de la sensibilidad cuantitativa a la insulina (del inglés QUICKI) aumentó (cuatro estudios, 276 participantes) con los probióticos.

Los marcadores inflamatorios, hs‐CRP (cuatro estudios, 248 participantes) y la interleucina 6 (dos estudios, 128 participantes) se redujeron con los probióticos. Aumentó el glutatión total antioxidante (dos estudios, 120 participantes) y se redujo el biomarcador de estrés oxidativo malondialdehído con probióticos (tres estudios, 176 participantes). No se sabe si hay alguna diferencia en la capacidad antioxidante total (cuatro estudios con 266 participantes).

En el caso del recién nacido, no se sabe con certeza si hay alguna diferencia entre los grupos en cuanto a: peso al nacer, edad gestacional al nacer, nacimientos prematuros, recién nacidos grandes, puntuaciones de perímetro cefálico y talla o necesidad de admisión en la unidad de cuidados intensivos neonatales. El número de recién nacidos con niveles altos de bilirrubina se redujo con los probióticos.

No se informaron eventos adversos en los ensayos.

¿Qué significa esto?

En base a los ensayos clínicos disponibles, existe poca evidencia para apoyar el uso de probióticos como tratamiento para las mujeres con DMG para mejorar los resultados del embarazo para las madres y los neonatos. Se necesitan ensayos controlados aleatorizados más grandes y bien diseñados para evaluar los efectos de los probióticos sobre el control de la concentración de glucosa y, cuando estén disponibles, podrán incluirse en la actualización de esta revisión.

Authors' conclusions

Implications for practice

Low‐certainty evidence showed that we were uncertain of any difference between probiotic and placebo groups in maternal hypertensive disorders of pregnancy and mode of birth by caesarean section; and neonatal/infant outcome of large for gestational age.

There were no reported adverse events reported by the trials.

Due to the variability of probiotics used and small sample sizes of the trials, the evidence from this review has limited ability to inform practice. Future studies need to consider the use of a specific probiotic type and strength, to be able to identify the impact on glucose metabolism and core clinical outcomes and to facilitate standardised reporting.

Implications for research

Further adequately powered, well‐designed randomised controlled trials are needed to clarify the effects of probiotics on glucose metabolism in pregnant women with gestational diabetes mellitus (GDM). There are currently six ongoing trials, with over 500 participants to be recruited. Once completed, these trials will add to the evidence base. These trials have been designed to examine some of our clinical outcomes but it would appear that they are not designed to measure long‐term infant outcomes. There is still a need for trials to be designed to address this need.

There is also a need for trials to be in agreement as to what probiotics should be used for research to avoid uncertainty of benefit with variability in probiotic type and strength. Clear evidence of benefit of these specific probiotics in improving fasting, pre‐prandial and/or postprandial glucose should also theoretically translate into improvement in important maternal and neonatal/infant clinical outcomes. In future trials, it is important to consider collecting important immediate and long‐term maternal and offspring outcomes as outlined in this review, but were not available. Important bio‐markers of insulin resistance (HOMA‐IR and HOMA‐B), insulin sensitivity (QUICKI), as well as inflammatory markers, antioxidants and markers of oxidative stress can be explored further to determine their role in modifying the effect of insulin resistance during pregnancy and whether they also play a role in improving glucose metabolism and consequently improve clinical outcomes. Another important factor to be considered in future trials is agreed criteria for diagnosing GDM.

We look forward to the results of current ongoing studies. The results of these studies will be included in future updates of this review.

Summary of findings

Open in table viewer

Summary of findings 1. Probiotic compared to placebo for treating women with gestational diabetes for improving maternal and infant health and well‐being ‐ maternal outcomes

Outcomes	№ of participants (studies) Follow up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	*Anticipated absolute effects^ (95% CI)**
Probiotic compared to placebo for treating women with gestational diabetes for improving maternal and fetal health and well‐being ‐ maternal outcomes
Patient or population: pregnant women diagnosed with gestational diabetes Setting: Iran (8), Ireland (1), Thailand (1) Intervention: probiotics (any type) administered by any route given during pregnancy to treat women with gestational diabetes Comparison: placebo (similar appearance and taste to the probiotics) or standard care
Outcomes	№ of participants (studies) Follow up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with placebo	Risk difference with probiotic
Hypertensive disorders (including pre‐eclampsia, pregnancy‐induced hypertension, eclampsia)	256 (3 RCTs)	⊕⊕⊝⊝ LOW ¹	RR 1.50 (0.64 to 3.53)	Study population
	256 (3 RCTs)	⊕⊕⊝⊝ LOW ¹	RR 1.50 (0.64 to 3.53)	63 per 1000	26 more per 1000 (26 fewer to 151 more)
Subsequent development of type 2 diabetes	(0 studies)		not estimable	No outcome data reported in the included studies.
Mode of birth (caesarean)	267 (3 RCTs)	⊕⊕⊝⊝ LOW ^{2 3}	RR 0.64 (0.30 to 1.35)	Study population
Mode of birth (caesarean)	267 (3 RCTs)	⊕⊕⊝⊝ LOW ^{2 3}	RR 0.64 (0.30 to 1.35)	351 per 1000	224 fewer per 1000 (105 fewer to 474 more)
Induction of labour	127 (1 RCT)	⊕⊝⊝⊝ VERY LOW ⁴	RR 1.33 (0.74 to 2.37)	Study population
	127 (1 RCT)	⊕⊝⊝⊝ VERY LOW ⁴	RR 1.33 (0.74 to 2.37)	231 per 1000	76 more per 1000 (60 fewer to 316 more)
Perineal trauma	(0 studies)		not estimable	No outcome data reported in the included studies.
Postnatal weight retention or return to pre‐pregnancy weight	(0 studies)		not estimable	No outcome data reported in the included studies.
Postnatal depression	(0 studies)		not estimable	No outcome data reported in the included studies.
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RR: Risk ratio.
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.
¹ Downgraded two levels due to serious concerns related to imprecision as only has 3 small studies with wide confidence intervals. ² Downgraded one level due to serious concerns related to imprecision as only has 3 small studies with wide confidence intervals. ³ Downgraded one level due to serious concerns related to inconsistency as I² of 69%, studies showed different findings. ⁴ Downgraded two levels due to serious concerns related to imprecision as only one small study with wide confidence intervals. We downgraded for indirectness as the population of one study will not reflect population of all women with GDM.

Open in table viewer

Summary of findings 2. Probiotic compared to placebo for treating women with gestational diabetes for improving maternal and infant health and well‐being‐ infant/child/adult outcome

Outcomes	№ of participants (studies) Follow up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	*Anticipated absolute effects^ (95% CI)**
Probiotic compared to placebo for treating women with gestational diabetes for improving maternal and infant health and well‐being ‐ infant/child/adult outcomes
Patient or population: pregnant women diagnosed with gestational diabetes Setting: Iran (1), Ireland (1) Intervention: probiotic Comparison: placebo
Outcomes	№ of participants (studies) Follow up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with placebo	Risk difference with probiotic
Perinatal (fetal and neonatal) mortality	(0 studies)		not estimable	No outcome data reported in the included studies.
Large‐for‐gestational age > 90 centile	174 (2 RCTs)	⊕⊕⊝⊝ LOW ¹	RR 0.73 (0.35 to 1.52)	Study population
Large‐for‐gestational age > 90 centile	174 (2 RCTs)	⊕⊕⊝⊝ LOW ¹	RR 0.73 (0.35 to 1.52)	159 per 1000	43 fewer per 1000 (103 fewer to 83 more)
Composite serious neonatal outcomes (variously defined by trials, e.g. infant death, shoulder dystocia, bone fracture, or nerve palsy	(0 studies)		not estimable	No data reported for composite serious neonatal outcomes (variously defined by trials, e.g. infant death, shoulder dystocia, bone fracture, or nerve palsy in any of the included studies.
Neurosensory disability	(0 studies)		not estimable	No outcome data reported in the included studies.
Neonatal hypoglycaemia requiring treatment (variously defined)	177 (3 RCTs)	⊕⊕⊝⊝ LOW ¹	RR 0.85 (0.39 to 1.84)	Study population
	177 (3 RCTs)	⊕⊕⊝⊝ LOW ¹	RR 0.85 (0.39 to 1.84)	135 per 1000	20 fewer per 1000 (82 fewer to 113 more)
Adiposity (neonatal/child/child as an adult)	(0 studies)		not estimable	No outcome data reported in the included studies.
Diabetes(type1 or type2) or impaired glucose tolerance (child/adult)	(0 studies)		not estimable	No outcome data reported in the included studies.
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RR: Risk ratio.
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect
¹ Downgraded two levels due to serious concerns related to imprecision as only has 2 small studies with wide confidence intervals.

Background

Description of the condition

Gestational diabetes (GDM) is defined as "carbohydrate intolerance resulting in hyperglycaemia of variable severity with onset or first recognition during pregnancy" (Alberti 1998). The prevalence of GDM is thought to vary from 1.5% to 14% worldwide and varies between ethnic groups (ACOG 2001; Dabelea 2005; Ekeroma 2015; Ferrara 2007; Poston 2013), and countries or institutions depending on the diagnostic criteria for GDM being used (ADA 2010; Diabetes Care 2010; Ekeroma 2015; NICE 2015). The global epidemic of obesity (a risk factor for GDM) is continuing to rise in developed and developing countries (Swinburn 2011), with the concomitant increase in rates of pregnancy complications (WHO 2016), including GDM. Health risks for women with GDM include pre‐eclampsia, induction of labour (Crowther 2005), caesarean section, and over half of women with GDM will develop type 2 diabetes within 10 years of the birth (Kim 2002). The risks for their infants include macrosomia (baby born much larger than average), respiratory distress syndrome, birth injuries such as nerve palsy, bone fracture and shoulder dystocia, jaundice, and hypoglycaemia, which if prolonged or severe can cause brain injury (Crowther 2005; Landon 2009). In addition, there is increasing recognition of the association between intrauterine fetal programming effects with adverse long‐term health consequences for the infant, creating a vicious intergenerational cycle of obesity, diabetes, and metabolic syndrome (Boney 2005; Dabelea 2005).

Description of the intervention

Probiotics are micro‐organisms that naturally occur in foods and when consumed in adequate amounts may confer health benefits for the host (FAO 2001). Probiotics are usually found in fermented milk products, yogurt or dietary supplements as well as in capsules. There are many different types of probiotics and the two most widely used genera are Lactobacillus and Bifidobacterium (Laitinen 2009).

The gut microbiota (micro‐organisms that colonise the gut) have the potential to influence obesity and type 2 diabetes through modification of energy extraction, inflammation, hunger and satiety, as well as lipid and glucose metabolism (Flint 2012; Nieuwdorp 2014; Turnbaugh 2006). Type 2 diabetes has been associated with changes in the gut microbiome (Larsen 2010). Obese women have also been identified to have a different gut microbiome compared to lean women (Nieuwdorp 2014; Turnbaugh 2006). Gut microbiota differences also exist between pregnant overweight and normal weight women (Collado 2008), as well as in the third trimester of pregnancy compared to the first trimester, with the third trimester microbiome being similar to non pregnant individuals with metabolic syndrome (Koren 2012). Supplementation with probiotics has been shown to improve glycaemic control in men and women with type 2 diabetes (Andreasen 2010; Ejtahed 2012). Probiotics have been shown to prevent GDM in a sample of pregnant women in a general population (Luoto 2010), and probiotics with dietary counselling reduced mean plasma glucose concentrations and improved insulin sensitivity in another study of healthy pregnant women both antenatally and postpartum (Laitinen 2009). Probiotic milk products reduced pre‐eclampsia in a large Norwegian cohort study (Braentsaeter 2011) and are considered safe to use in pregnancy (Allen 2010; Elias 2011). Probiotic capsules (Lactobacillus rhamnosus) in a double‐blind randomised controlled trial showed significant and sustainable weight loss in obese non pregnant women (Sanchez 2014). A larger randomised controlled trial of probiotic versus placebo in pregnant women in Australia (Nitert 2013), to determine whether probiotics can prevent GDM in overweight and obese women has recently been published (Callaway 2019). A systematic review and meta‐analysis looking at the effect of treatment of GDM on pregnancy outcomes showed that treatment significantly reduced the risks of fetal macrosomia, large‐for‐gestational‐age births, shoulder dystocia and gestational hypertension, as well as a tendency to reduction of perinatal/neonatal mortality and birth trauma (Poolsup 2014). A Cochrane Review of probiotics for prevention of GDM included one study that reported lower rates of women diagnosed with GDM and lower birthweight with probiotics (Barrett 2014). GDM treatment to date has mostly comprised of dietary and glucose‐lowering agents either insulin and or tablets (biguanides or second‐generation sulphonylureas) (Coustan 2013). The role of probiotics in treating pregnant women with GDM has yet to be clearly established.

How the intervention might work

Probiotics in the 1960s were hypothesised to have the beneficial effects of producing substances that may promote the growth of other micro‐organisms and was further defined in the 1980s as a microbial feed supplement that improves the intestinal balance of the host (FAO 2001). The discovery of the gut microbiome and its relationship to health and disease, together with DNA sequencing technology meant easier identification of the host genome and host micro‐organisms or microbiome (Solt 2015). Microbiome changes influence gut content by allowing the predominance of some organisms over others, which in turn can cause a generalised increase in inflammatory markers in the host and increasing risks of diseases (Solt 2015). Modification of the gut microbiome (Flint 2012) by probiotics may be used as an intervention to prevent or treat metabolic diseases through various complex intracellular metabolic pathways within the gut (Nieuwdorp 2014; Turpin 2010). The mechanisms are complex from probiotics actively competing with pathological bacteria to dampening their inflammatory effect possibly by producing more butyrate; to improving the bile acid pool to reduce insulin resistance; or binding to mucosal receptors in the gut altering metabolic pathways responsible for the metabolic syndrome and satiety (Nieuwdorp 2014). Furthermore, probiotics have an anti‐obesity action by influencing energy extraction in humans through increased lipolysis and reduction in lipoprotein lipase, which may reduce excess energy storage (Turpin 2010). The microbiomes of obese people have been found to have the ability to convert non digestible carbohydrates to digestible short‐chain fatty acids, with increased uptake in the gut increasing energy harvest, storage and consequently increasing adiposity (Flint 2012). High adiposity in human and animal studies has been associated with increased systemic inflammation, which impacts adversely on pregnancy outcomes especially increasing risks of pre‐eclampsia (Braentsaeter 2011), and increased insulin resistance. Probiotics have been shown to reduce the rates of severe pre‐eclampsia (Braentsaeter 2011), reduce insulin resistance (Asemi 2013) and improve insulin sensitivity (Laitinen 2009). Other beneficial effects of probiotics include reduction of psychological distress in healthy volunteers (Messaoudi 2011), and consumption of probiotic yoghurt improved mood (Benton 2007) possibly by reducing systemic inflammatory markers (Dinan 2011). Futhermore, individuals with depression have been shown to have a different microbiome to healthy individuals (Jiang 2015), as well as high levels of inflammatory cytokines (Dinan 2011), with probiotics predicted to dampen the negative effects of inflammation causing depression. Trials of probiotics in preterm neonates have demonstrated a reduction in necrotising enterocolitis and mortality (AlFaleh 2014).

Why it is important to do this review

The prevalence of GDM is increasing and the implementation of the International Association of Diabetes and Pregnancy Study Group diagnostic (IADPSG) criteria could also be contributing to the increase in women diagnosed with GDM. (Cundy 2014; Ekeroma 2015). All women with GDM may receive lifestyle advice (Metzger 2007), and for some women, this may be an effective treatment to maintain glycaemic control without the addition of pharmacotherapy (Brown 2017). The use of probiotics may prove a useful adjunct to lifestyle interventions and reduce the need for pharmacotherapy possibly by influencing metabolic pathways that lead to development of GDM (Nieuwdorp 2014). This review will establish the effectiveness of such an intervention in particular for women with GDM.

Objectives

To evaluate the safety and effectiveness of probiotics in treating pregnant women with gestational diabetes mellitus (GDM) on maternal and infant outcomes.

Methods

Criteria for considering studies for this review

Types of studies

Randomised controlled trials (RCTs). Cluster‐randomised trials were eligible for inclusion but none were identified. Quasi‐randomised and cross‐over trials were not eligible for inclusion. There were no restrictions to language or year of publication.

Types of participants

Pregnant women diagnosed with gestational diabetes (diagnosis as defined by the individual trial). Trials of women with type 1 or type 2 diabetes diagnosed prior to pregnancy were excluded.

Types of interventions

Probiotics (any type) administered by any route given during pregnancy to treat women with gestational diabetes and where the control group received placebo or standard care (as defined by the trialist).

Types of outcome measures

Primary outcomes

Maternal

Hypertensive disorders of pregnancy (including pre‐eclampsia, pregnancy‐induced hypertension, eclampsia)
Subsequent development of type 2 diabetes (as defined by trialist)
Mode of birth

Infant

Perinatal (fetal and neonatal) mortality
Large‐for‐gestational age (birthweight greater than the 90^th centile; or as defined by individual trial)
Composite of serious neonatal outcomes (variously defined by trials, e.g. infant death, shoulder dystocia, bone fracture or nerve palsy)
Neurosensory disability (defined by trialists)

Secondary outcomes

Maternal

Induction of labour
Perineal trauma
Placental abruption
Postpartum haemorrhage
Postpartum infection
Weight gain during pregnancy
Adherence to the intervention
Behaviour changes associated with the intervention
Relevant biomarker changes associated with the intervention (e.g. adiponectin, free‐fatty acids, triglycerides, high‐density lipoproteins (HDL), low‐density lipoproteins (LDL), insulin)
Sense of well‐being and quality of life (any validated Well‐being and Quality of life scores)
Views of the intervention
Breastfeeding (e.g. at discharge, six weeks postpartum)
Use of additional pharmacotherapy
Glycaemic control during/end of treatment (as defined by trialists)
Maternal hypoglycaemia
Maternal mortality

Long‐term maternal outcomes

Postnatal depression (any validated postnatal depression scores e.g. Edinburgh Postnatal Depression Scale (EPDS))
Postnatal weight retention or return to pre‐pregnancy weight
Body mass index (BMI)
GDM in a subsequent pregnancy
Type 1 diabetes
Type 2 diabetes
Impaired glucose tolerance
Cardiovascular health (as defined by trialists, including blood pressure (BP), hypertension, cardiovascular disease, metabolic syndrome)

Infant

Stillbirth
Neonatal mortality
Gestational age at birth
Preterm birth (less than 37 weeks' gestation and less than 32 weeks' gestation)
Apgar score (less than seven at five minutes)
Macrosomia
Small‐for‐gestational age
Birthweight and z‐score
Head circumference and z‐score
Length and z‐score
Ponderal index
Adiposity
Shoulder dystocia
Bone fracture
Nerve palsy
Respiratory distress syndrome
Hypoglycaemia requiring treatment (variously defined)
Hyperbilirubinaemia
Neonatal hypocalcaemia
Polycythaemia
Relevant biomarker changes associated with the intervention (e.g. cord c peptide, cord insulin)

Later childhood

Weight and z score
Height and z score
Head circumference and z score
Adiposity (including BMI, skinfold thickness)
BP
Type 1 diabetes mellitus
Type 2 diabetes mellitus
Impaired glucose tolerance
Dyslipidaemia or metabolic syndrome
Educational achievement

Adulthood outcomes

Weight
Height
Adiposity (including skin folds, fat mass)
Cardiovascular health (as defined by trialists, including BP, hypertension, cardiovascular disease, metabolic syndrome)
Type 1 diabetes mellitus
Type 2 diabetes mellitus
Impaired glucose tolerance
Dyslipidaemia or metabolic syndrome
Employment, education and social status/achievement

Health services

Number of antenatal visits or admissions
Number of hospital or health professional visits (including midwife, obstetrician, physician, dietician, diabetic nurse)
Admission to neonatal intensive care unit/nursery
Length of antenatal stay
Length of postnatal stay (maternal)
Length of postnatal stay (baby)
Cost of maternal care
Cost of offspring care (including neonatal intensive care unit admission)
Costs associated with the intervention
Costs to families associated with the management provide

Search methods for identification of studies

The following methods section of this protocol is based on a standard template used by Cochrane Pregnancy and Childbirth.

Electronic searches

We searched Cochrane Pregnancy and Childbirth’s Trials Register by contacting their Information Specialist (24 July 2019).

The Register is a database containing over 25,000 reports of controlled trials in the field of pregnancy and childbirth. It represents over 30 years of searching. For full current search methods used to populate Pregnancy and Childbirth’s Trials Register including the detailed search strategies for CENTRAL, MEDLINE, Embase and CINAHL; the list of handsearched journals and conference proceedings, and the list of journals reviewed via the current awareness service, please follow this link.

Briefly, Cochrane Pregnancy and Childbirth’s Trials Register is maintained by their Information Specialist and contains trials identified from:

monthly searches of the Cochrane Central Register of Controlled Trials (CENTRAL);
weekly searches of MEDLINE (Ovid);
weekly searches FSTA;
weekly searches Biosis;
weekly searches of Embase (Ovid);
monthly searches of CINAHL (EBSCO);
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.

Search results were screened by two people and the full text of all relevant trial reports identified through the searching activities described above is reviewed. Based on the intervention described, each trial report is assigned a number that corresponds to a specific Pregnancy and Childbirth review topic (or topics), and is then added to the Register. The Information Specialist searches the Register for each review using this topic number rather than keywords. This results in a more specific search set that has been fully accounted for in the relevant review sections (Included, Excluded, Awaiting Classification or Ongoing).

In addition, we searched ClinicalTrials.gov and the WHO International Clinical Trials Registry Platform (ICTRP) (24 July 2019) for unpublished, planned and ongoing trial reports using search methods detailed in Appendix 1.

Searching other resources

We searched the reference lists of all retrieved studies.

We did not apply any language or date restrictions.

Data collection and analysis

The methods was based on the standard template used by Cochrane Pregnancy and Childbirth.

Selection of studies

Two review authors Karaponi OKesene‐Gafa (KOG) and Abigail Moore (AM) independently assessed for inclusion all potential studies identified as a result of the search strategy. Any disagreement was resolved through discussion with senior author Professor Caroline A Crowther (CAC).

Data extraction and management

We extracted relevant data using the Cochrane Pregnancy and Childbirth Group's data extraction form. We collected information on type of intervention, frequency and route of administration; trialists' declarations of interest and trial dates. For eligible studies, two review authors extracted the data using the agreed form. Discrepancies were resolved through discussion. Data were entered into Review Manager software (RevMan 2014) and checked for accuracy. We contacted trial authors for the original reports to provide further details if required.

Assessment of risk of bias in included studies

Two review authors (KOG and AM) 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 our senior author (CAC).

(1) Random sequence generation (checking for possible selection bias)

We described for each included study the method used to generate the allocation sequence in sufficient detail to allow an assessment of whether it should produce comparable groups.

We assessed the method 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); we planned to exclude studies judged to be of high risk of bias.
unclear risk of bias.

(2) Allocation concealment (checking for possible selection bias)

We described for each included study the method used to conceal allocation to interventions prior to assignment and will assess 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);
unclear risk of bias.

(3.1) Blinding of participants and personnel (checking for possible performance bias)

We described for each included study the methods used, if any, to blind study participants and personnel from knowledge of which intervention a participant received. We considered that studies were at 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 and 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:

low, high or unclear risk of bias.

(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 and 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 supplied by the trial authors, we re‐included missing data in the analyses.

We assessed methods as:

low risk of bias (e.g. no missing outcome data; missing outcome data 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 (checking for reporting bias)

We described for each included study how we investigated the possibility of selective outcome reporting bias and what we found.

We assessed the methods as:

low risk of bias (where it is clear that all of the study’s pre‐specified outcomes and all expected outcomes of interest to the review have been reported);
high risk of bias (where not all the study’s pre‐specified outcomes have been reported; one or more reported primary outcomes were not pre‐specified; outcomes of interest were reported incompletely and so cannot be used; study fails to include results of a key outcome that would have been expected to have been reported);
unclear risk of bias.

(6) Other 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:

low risk of other bias;
high risk of other bias;
unclear whether there is risk of other bias.

(7) Overall risk of bias

We made explicit judgements about whether studies were at high risk of bias, according to the criteria given in the Handbook (Higgins 2011). With reference to (1) to (6) above, we assessed the likely magnitude and direction of the bias and whether we considered it was likely to impact on the findings. We planned to explore the impact of the level of bias through undertaking sensitivity analyses.

Assessment of the quality of the evidence using the GRADE approach

For the main comparison or probiotic versus placebo, the quality of the evidence will be assessed using the GRADE approach, outlined in the GRADE handbook and Chapters 11 and 12 of the Cochrane Handbook of Systematic Reviews of Interventions (Higgins 2011), for the outcomes listed below.

Maternal

Hypertensive disorders of pregnancy (including pre‐eclampsia, pregnancy‐induced hypertension, eclampsia)
Subsequent development of type 2 diabetes (as defined by trialist)
Mode of birth
Induction of labour
Perineal trauma
Postnatal weight retention or return to pre‐pregnancy weight
Postnatal depression

Infant/child/adult

Perinatal (fetal and neonatal) mortality
Large‐for‐gestational age (birthweight greater than the 90th centile; or as defined by individual trial)
Composite of serious neonatal outcomes (variously defined by trials, e.g. infant death, shoulder dystocia, bone fracture or nerve palsy)
Neurosensory disability (defined by trialists)
Neonatal hypoglycaemia
Adiposity (neonatal/child/adult)
Diabetes (type 1 or type 2) or impaired glucose tolerance (child/adult)

We used the GRADEpro Guideline Development tool to import data from Review Manager 5.3 (RevMan 2014) in order to create 'Summary of findings’ tables. A summary of the intervention effect and a measure of quality for each of the above outcomes was produced using the GRADE approach. The GRADE approach uses five considerations (study limitations, consistency of effect, imprecision, indirectness and publication bias) to assess the certainty of the body of evidence for each outcome. The evidence can be downgraded from 'high certainty' by one level for serious (or by two levels for very serious) limitations, depending on assessments for risk of bias, indirectness of evidence, serious inconsistency, imprecision of effect estimates or potential publication bias.

Measures of treatment effect

Dichotomous data

For dichotomous data, we presented the results as summary risk ratio (RR) with 95% confidence intervals (CIs).

Continuous data

For continuous data, we used the mean difference (MD) with 95% CIs as outcomes were measured in the same way between trials. We planned to use the standardised mean difference (SMD) with 95% CIs to combine trials that measured the same outcome, but used different methods.

Unit of analysis issues

Cluster‐randomised trials

We did not identify any cluster‐randomised trials in this review. If cluster‐randomised trials are identified in future updates of this review, we will include them in the analyses along with individually‐randomised trials. We will make adjustments using the methods described in the Handbook [Section 16.3.4 or 16.3.6] (Higgins 2011) using an estimate of the intra‐cluster correlation co‐efficient (ICC) derived from the trial (if possible), from a similar trial or from a study of a similar population. If we use ICCs from other sources, we will report this and conduct sensitivity analyses to investigate the effect of variation in the ICC. We will consider it reasonable to combine the results from both cluster‐randomised trials and individually‐randomised trials if there is little heterogeneity between the study designs and the interaction between the effect of intervention and the choice of randomisation unit is considered to be unlikely.

Multiple pregnancies

There were no multiple pregnancies identified in this review. In future updates of this review, if studies involving multiple pregnancies are identified, we will present maternal data as per woman randomised and neonatal data per infant.

Multiple‐arm studies

There were no studies with multiple arms identified in this review. If in future updates of this review, if studies with multiple intervention arms are identified, we will avoid 'double counting' of participants by combining groups to create a single pair‐wise comparison if possible. Where this is not possible, we will split the 'shared' group into two or more groups with smaller sample size and include two or more (reasonably independent) comparisons.

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 if the I² was greater than 30% and either a 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

Had we included more than 10 studies in the meta‐analysis, we planned to investigate reporting biases (such as publication bias) using funnel plots. We planned to assess funnel plot asymmetry visually. If asymmetry was suggested by a visual assessment, we planned to perform exploratory analyses to investigate it.

Data synthesis

We carried out statistical analysis using the Review Manager software (RevMan 2014). 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. If there was clinical heterogeneity sufficient to expect that the underlying treatment effects differed between trials, or if 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 of the range of possible treatment effects and we planned to discuss the clinical implications of treatment effects differing between trials. If the average treatment effect was not clinically meaningful, we did not combine trials.

Where we used random‐effects analyses, the results were presented as the average treatment effect with 95% CIs, and the estimates of Tau² and I².

Subgroup analysis and investigation of heterogeneity

Had we identified substantial heterogeneity, we planned to investigate it using subgroup analyses and sensitivity analyses. We will consider whether an overall summary is meaningful, and if it is, use random‐effects analysis to produce it.

We planned to carry out the following subgroup analyses.

Different types of probiotic (probiotic A versus probiotic B)
Mode of administration of probiotic (capsule versus yoghurt versus nutritional supplement)
Dosage (high versus low dose)
Diagnostic criteria used for GDM (IADPSG, American College of Obstetrics and Gynaecology, World Health Organization, Carpenter and Coustan, Australian Diabetes in Pregnancy Society, other criteria not specified above, diagnostic criteria not specified)

Subgroup analysis will be restricted to the review's primary outcomes.

We will assess subgroup differences by interaction tests available within RevMan (RevMan 2014). We will report the results of subgroup analyses quoting the Chi² statistic and P value, and the interaction test I² value.

Sensitivity analysis

In order to examine robustness of individual decisions being made to this systematic review, we planned to carry out sensitivity analysis restricting our analyses to:

studies at a low risk of bias (for allocation concealment);
full‐text papers;
number of participants > 300;
RCTs (excluding cluster‐randomised trials in order to investigate the effect of the randomisation unit);
studies without high levels of missing data.

Results

Description of studies

Results of the search

See: Figure 1

Figure 1

Study flow diagram.

We retrieved 31 trial reports from the Cochrane Pregnancy and Childbirth database searches plus an additional 17 from other sources. We included nine trials (22 reports) and excluded 17. Two trials (three reports) are awaiting further classification (Characteristics of studies awaiting classification), and six trials are ongoing (Characteristics of ongoing studies).

Included studies

Nine trials were selected and analysed (Ahmadi 2016; Badehnoosh 2018; Hajifaraji 2017; Jafarnejad 2016; Karamali 2016; Karamali 2018; Kijmanawat 2019; Lindsay 2015; Nabhani 2018). All trials randomised women with gestational diabetes mellitus (GDM) to probiotics or placebo.

Information regarding the included trials is reported in the Characteristics of included studies tables.

Design

All studies were randomised controlled clinical trials comparing probiotics with placebo. Probiotics used in most studies were different in strengths and combinations (refer to interventions and comparisons).

Sample sizes

From the nine included trials, sample sizes ranged from 60 to 149 participants. The total number of participants were 695 randomised and 674 analysed. Total number of participants per study randomised (final analysis) were: Hajifaraji 2017; randomised 64 (analysed 56); Kijmanawat 2019 randomised 60 (analysed 57); Badehnoosh 2018, Karamali 2016, Karamali 2018 randomised 60 (analysed 60) in each trial. Ahmadi 2016 randomised 70 (analysed 70); Jafarnejad 2016 randomised 82 (analysed 72) and the trial with largest number of participants was Lindsay 2015 which randomised and analysed 149; and Nabhani 2018 randomised and analysed 90.

Setting

Seven studies were carried out in hospital or university settings in Iran (Ahmadi 2016; Badehnoosh 2018; Hajifaraji 2017; Jafarnejad 2016; Karamali 2016; Karamali 2018; Nabhani 2018. One of the studies was carried out in Bangkok (Thailand) (Kijmanawat 2019), and one in Dublin (Ireland) (Lindsay 2015).

Dates of studies

Ahmadi 2016 took place between February and May 2016; Badehnoosh 2018 between April to September 2016, Hajifaraji 2017 during spring and summer 2014 (April to August); Jafarnejad 2016 between May 2014 to October 2015; Karamali 2016 between November 2015 to January 2016; Karamali 2018 between April and December 2016; Kijmanawat 2019 between July 2016 and February 2017; Lindsay 2015 between March 2012 and May 2014; Nabhani 2018 between January 2015 and September 2016.

Participants

All participants were women diagnosed with GDM according to the criteria chosen by each research team at between 24 to 28 weeks' gestation. Ahmadi 2016, Badehnoosh 2018, Hajifaraji 2017, Karamali 2016, Karamali 2018, Kijmanawat 2019, Nabhani 2018 used the American Diabetes Association (ADA) criteria after taking a 75 g oral glucose tolerance test (OGTT), and having a fasting blood glucose of ≥ 92 mg/dL, one‐hour OGTT ≥ 180 mg/dL and two‐hour OGTT ≥ 153 mg/dL. Kijmanawat 2019, as well as using the ADA diagnostic criteria also used a fasting plasma glucose ≥ 92 mg/dL at the first prenatal visit as a diagnosis for GDM. Jafarnejad 2016 used the Australasian Diabetes in Pregnancy Society (ADIPS) criteria with 75 g OGTT with results of fasting venous plasma glucose level, ≥ 5.5 mmol/L⁻¹ or two‐hour venous plasma glucose level, ≥ 8.0 mmol/L⁻¹. Lindsay 2015 used the Carpenter Coustan criteria results of a three‐hour 100 g OGTT with fasting ≥ 95 mg/dL, one‐hour ≥ 180 mg/dL, two‐hours ≥ 155 mg/dL, three‐hour 140 mg/dL for newly diagnosed impaired glucose tolerance (IGT) (1 raised value) or GDM (≥ 2 raised values).

Participants were between 18 to 45 years of age.

Three trials specifically reported participants as nulliparous (Badehnoosh 2018; Hajifaraji 2017; Karamali 2016). The other six trials did not clearly report baseline information related to parity (Ahmadi 2016; Jafarnejad 2016; Karamali 2018; Kijmanawat 2019; Lindsay 2015; Nabhani 2018).

Interventions and comparisons

Probiotics used in studies were of different strengths and combinations and given to participants in capsule form daily for either four, six or eight weeks. Participants in Ahmadi 2016 were given Lactobacillus casei and, Bifidobacterium bifidum (2 × 10⁹ colony‐forming units (CFU)/g each) plus 0·8 g inulin for six weeks. Badehnoosh 2018 gave participants Lactobacillus acidophilus, L casei and B bifidum (2 x 10⁹ CFU/g each) for six weeks. Hajifaraji 2017 used (4Biocap capsules) containing 180 mg (4 x 10⁹ CFU) standard power including freeze‐dried cultures of Lactobacillus acidophilus LA‐5, Bifibacterium BB12, Streptococcus thermophilus STY‐31, and Lactobacillus delbrueckii bulgaricus LBY‐27 + dextrose anhydrate filler and magnesium stearate lubricant for eight weeks. Jafarnejad 2016 used VSL#3, a freeze‐dried probiotic preparation containing eight strains of lactic acid bacteria (S thermophilus, Bifidobacterium breve, Bifidobacterium longum, Bifidobacterium infantis, L acidophilus, Lactobacillus plantarum, Lactobacillus paracasei, and L delbrueckii subsp. Bulgaricus (112.5 × 10⁹ CFU/capsule), plus microcrystalline cellulose, stearic acid, magnesium stearate, and vegetable capsule (hydroxypropyl methylcellulose), silicon dioxide for eight weeks. Karamali 2016 and Karamali 2018 used three viable freeze‐dried strains: L acidophilus (2 × 10⁹ CFU/g), L. casei (2 × 10⁹ CFU/g), L casei (2 × 10⁹ CFU/g) and B bifidum (2 × 10⁹ CFU/g) for six weeks. Kijmanawat 2019 gave participants L acidophilus and B bifidum (1 x 10⁹) CFU for four weeks. Lindsay 2015 used 100 mg Lactobacillus salivarus UCC118 (10⁹ CFU/capsule) for four to six weeks. Nabhani 2018 used Lacidophilus, Lactobacillusplantarum, Lactobacillus fermentum, Lactobacillus gasseri (1.5–7.0 x 10⁹–10 CFU/g) – with fructo‐oligosaccharide (38.5 mg) with lactose (300 mg), magnesium stearate, talc, colloidal silicon dioxide (each of them 5.5 mg), flavourings and sweeteners that have neutral effects for six weeks.

All studies had placebo capsules as a comparison. Placebo in studies are explained in more detail in the Blinding (performance bias and detection bias) section of the review.

To support the interventions, four trials sent daily reminder text messages to participants (Ahmadi 2016; Badehnoosh 2018; Karamali 2016; Karamali 2018). Four trials carried out weekly phone interviews ( Hajifaraji 2017; Jafarnejad 2016; Kijmanawat 2019; Nabhani 2018). One study did not use phone interviews in their processes (Lindsay 2015).

Outcomes

Three trials (Badehnoosh 2018; Karamali 2018; Lindsay 2015) reported hypertensive disorders of pregnancy including pre‐eclampsia and pregnancy‐induced hypertension. No trials reported eclampsia. Three trials (Badehnoosh 2018; Karamali 2018; Lindsay 2015) reported caesarean section rates. Two trials (Badehnoosh 2018; Lindsay 2015) reported large‐for‐gestational age > 90 centile. One trial (Lindsay 2015) reported induction of labour and postpartum haemorrhage. Six trials (Ahmadi 2016; Badehnoosh 2018; Jafarnejad 2016; Karamali 2016; Karamali 2018; Kijmanawat 2019) reported weight gain during pregnancy (during the intervention) . Three trials (Badehnoosh 2018; Kijmanawat 2019; Lindsay 2015) reported total gestational weight gain.

For relevant biomarker for oxidative stress, three trials (Badehnoosh 2018; Hajifaraji 2017; Karamali 2018) reported malondialdehyde (MDA); two trials (Badehnoosh 2018; Karamali 2018) reported total glutathione (GSH), and one trial (Hajifaraji 2017) reported uric acid.

For inflammatory bio markers: four trials (Badehnoosh 2018; Hajifaraji 2017; Jafarnejad 2016; Karamali 2018) reported high‐sensitivity C‐reactive protein (hs‐CRP); one trial (Jafarnejad 2016) reported interleukin 10 (IL‐10) and interferon c (IFN‐c); two trials (Hajifaraji 2017; Jafarnejad 2016) reported interferon 6 (IL‐6) and tumour necrosis factor alpha (TNF‐α).

For antioxidants: one trial (Karamali 2018) reported nitrous oxide; four trials (Badehnoosh 2018; Hajifaraji 2017; Karamali 2018; Nabhani 2018) reported total antioxidant capacity (TAC), and one trial (Hajifaraji 2017) reported serum GSH reductase (GSHR), erythrocyte superoxide dismutase (SOD) and erythrocyte glutathione peroxidase (GPx).

In biomarkers for insulin resistance: seven trials (Ahmadi 2016; Hajifaraji 2017; Jafarnejad 2016; Karamali 2016; Kijmanawat 2019; Lindsay 2015; Nabhani 2018) reported Homeostatic Model Assessment of Insulin Resistance (HOMA‐IR); and two trials (Ahmadi 2016; Karamali 2016) reported HOMA‐B (β‐cell function ).

The biomarker for insulin sensitivity QUICKI ( quantitative insulin‐sensitivity check index) was reported by four trials (Ahmadi 2016; Hajifaraji 2017; Karamali 2016; Nabhani 2018).

Insulin secretion was reported by seven trials (Ahmadi 2016; Hajifaraji 2017; Jafarnejad 2016; Karamali 2016; Kijmanawat 2019; Lindsay 2015; Nabhani 2018).

For lipids: four trials (Ahmadi 2016; Karamali 2016; Lindsay 2015; Nabhani 2018) reported triglycerides (TAG); two trials (Ahmadi 2016; Karamali 2016) reported very low‐density lipoprotein (VLDL); and four trials (Ahmadi 2016; Karamali 2016; Lindsay 2015; Nabhani 2018) reported low‐density lipoprotein (LDL) cholesterol, high‐density lipoprotein (HDL) cholesterol, and total cholesterol.

Use of additional pharmacotherapy was reported by one trial (Badehnoosh 2018).

For glycaemic control: seven trials (Ahmadi 2016; Hajifaraji 2017; Jafarnejad 2016; Karamali 2016; Kijmanawat 2019; Lindsay 2015; Nabhani 2018) reported fasting plasma glucose.

For neonatal outcomes: three trials (Badehnoosh 2018; Karamali 2018; Lindsay 2015) reported gestational age at birth; two trials (Badehnoosh 2018; Karamali 2018) reported preterm birth; three trials (Badehnoosh 2018; Karamali 2018; Lindsay 2015) reported macrosomia; one trial (Lindsay 2015) reported small‐for‐gestational age (SGA); four trials (Badehnoosh 2018; Karamali 2018; Kijmanawat 2019; Lindsay 2015) reported birthweight; three trials (Badehnoosh 2018; Karamali 2018; Lindsay 2015) reported head circumference, length and infant hypoglycaemia (requiring treatment, variously defined); two trials (Badehnoosh 2018; Karamali 2018) reported hyperbilirubinaemia; one trial (Lindsay 2015) reported Cord C peptide; and two trials (Badehnoosh 2018; Lindsay 2015) reported on neonatal intensive care unit (NICU) or nursery admissions.

All trials reported no significant issues or important clinical adverse effects with probiotics.

Sources of funding

Grants from: the Vice Chancellor for Research AUMS, Iran funded Ahmadi 2016; Vice Chancellor for Research, IUMS, Tehran, Iran funded Badehnoosh 2018; Tehran, Shahid Beheshti, University Medical Sciences funded Hajifaraji 2017; Vice Chancellor for Research, IUMS, Tehran, Iran funded Karamali 2016 and Karamali 2018; Thailand Research Fund (TRF) funded Kijmanawat 2019; National Maternity Hospital Medical Fund with support from the Ivo Drury Award and the European Union’s Seventh Framework Program (FP7/2007‐2013), project Early Nutrition under grant agreement number 289346 funded Lindsay 2015; and Tabriz University of Medical Sciences, Iran, and Nutrition Research Center funded Nabhani 2018.

No details of funding for one trial (Jafarnejad 2016),

Declarations of conflict of interest

A total of nine trials declared no conflict of interest except two trials that declared conflict of interest of at least one of its members (Kijmanawat 2019) (SR received grant support from Merck Sharp and Dohme, research equipment support from ResMed, and speaker honoraria from Sanofi, Novo Nordisk and Medtronic), (Lindsay 2015) (F.S. was a shareholder in Alimentary Health Ltd and has received grants from GlaxoSmithKline and the Procter and Gamble Company in the past).

Excluded studies

There were 17 articles that were excluded including one conference abstract.

In seven of the trials (Asemi 2013; Asemi 2013a; Barthow 2016; Luoto 2010; Nitert 2013; Okesene‐Gafa 2018; Wickens 2017), the participants in the randomised controlled trial (RCTs) were not women with GDM. Four articles (Al‐Dughaishi 2016; Gomez 2015; Lindsay 2013; Lindsay 2014) were not RCTs. Two of the papers (Barrett 2012; Barrett 2014) were systematic reviews; the latter a Cochrane Review. One of the papers (Muktabhant 2015) was a Cochrane Review on diet and exercise. Two of the trials (Fei 2014; Zhang 2018) used prebiotics and not probiotics.

Studies awaiting classification

Two trials (Gonai 2014; Jamilian 2019) are awaiting classification.

Risk of bias in included studies

Risk of bias is summarised in Figure 2 and Figure 3.

Figure 2

'Risk of bias' graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Figure 3

'Risk of bias' summary: review authors' judgements about each risk of bias item for each included study.

Random sequence generation

We assessed all nine studies as low risk of bias for random sequence generation because they appeared to have adequate randomisation processes.

Eight trials reported that their random sequence was computer‐generated (Ahmadi 2016; Badehnoosh 2018; Hajifaraji 2017; Jafarnejad 2016; Karamali 2016; Karamali 2018; Lindsay 2015; Nabhani 2018).

One of the trials reported that a statistician generated the randomisation sequence using blocks (Kijmanawat 2019).

Three trials used computer block randomisation (Hajifaraji 2017, Kijmanawat 2019, Nabhani 2018).

Two trials specified that a separate researcher assistant (counsellor/therapist/trained personnel) carried out the randomisation and allocated the capsule packages according to the random sequence generated by the computer program (Hajifaraji 2017; Jafarnejad 2016).

Allocation concealment

Of the nine trials, one was rated as low risk of bias for allocation concealment ( Lindsay 2015). The remaining eight studies were rated as unclear as we were unable to determine if researchers were aware of the allocation sequence when recruiting participants (Ahmadi 2016; Badehnoosh 2018; Jafarnejad 2016; Karamali 2016; Nabhani 2018; Hajifaraji 2017; Karamali 2018; Kijmanawat 2019).

Blinding of participants and personnel (performance bias)

All nine trials were specifically reported as double‐blind, placebo‐controlled randomised trials and were all graded as low risk of performance bias.

Six trials specified that probiotics and placebo were indistinguishable from each other (Ahmadi 2016; Hajifaraji 2017; Karamali 2016; Kijmanawat 2019; Lindsay 2015; Nabhani 2018). One trial stated their placebo capsules were identical to probiotics and contained 40 mg microcrystalline cellulose (Jafarnejad 2016).

Two trials did not offer adequate details of their placebo capsules: one trial stated that the placebo contained starch (Badehnoosh 2018); the other trial stated that their placebo contained gelatin (Kijmanawat 2019. One trial reported use of placebo with no specifics (Karamali 2018).

Four trials reported that a coder or supplier of capsules anonymously labelled the packages as A or B, whereas the contents of the packages were unknown to the researcher allocating the treatment in four trials (Hajifaraji 2017; Jafarnejad 2016; Lindsay 2015; Nabhani 2018). In one of these studies the packages (A or B) were placed in sequentially‐numbered, sealed opaque envelopes (Lindsay 2015).

Blinding of outcome assessment (detection bias)

We assessed all nine trials as low risk of detection bias since they all reported adequate blinding of outcome assessors.

Incomplete outcome data

All nine studies were graded as low risk of bias, as there were minimal dropouts and no differential attrition. Two trials also stated that they used intention‐to‐treat analysis (Ahmadi 2016; Hajifaraji 2017.

Selective reporting

We assessed all nine trials as unclear risk of reporting bias because none of them had published study protocols, nor were any of them registered prospectively in any clinical trials registry, therefore we had insufficient information to judge which outcomes were pre‐specified outcomes and if they were reported in full.

Other potential sources of bias

One trial had a significant difference in baseline cholesterol level between the probiotics and placebo groups. After adjusting for biochemical values, maternal age and body mass index (BMI) at baseline, there was no significant differences in these results (Ahmadi 2016).

One trial had significant differences in baseline levels of fasting plasma glucose (FPG) and HDL cholesterol between the two groups, but after further adjusting these variables as well as for baseline maternal age and BMI, the results were similar in both groups except for HOMA‐B (P = 0.08) (Karamali 2016).

One trial had a slightly lower rate of Caucasian ethnicity and obesity and a higher rate of primiparity in the probiotic compared to placebo group, although these differences were not significant (Lindsay 2015).

One trial showed that there was a difference in energy, protein and total fat intakes (P < 0.05); thus, final analyses were adjusted for the measures of energy intake, BMI and baseline values (Nabhani 2018).

One trial stated that the women in their trial were taking 400 µg early pregnancy and 60 mg/day of ferrous sulphate from the second trimester (Badehnoosh 2018).

We assessed all nine studies as low risk of other sources of bias.