SUpporting Mothers Milk Intervention Trial (SUMMIT) — a randomised controlled trial comparing different doses of domperidone for treating lactation insufficiency in mothers of preterm infants – study protocol

Abstract

Background

Mothers of preterm infants face many challenges in establishing and maintaining an adequate supply of breast milk during their infant’s prolonged hospitalisation. Domperidone has been shown to be superior to placebo for the treatment of lactation insufficiency, however there is significant uncertainty regarding the optimal treatment dose. This trial has been designed to resolve the issue of whether a higher dose of domperidone (20 mg three times daily; 60 mg/day) leads to greater improvements in maternal breast milk supply compared to a lower dose (10 mg three times daily; 30 mg/day), while also evaluating differences in adverse events. This knowledge will be critical for guiding clinical practice and facilitating benefit versus risk evaluations surrounding the use of domperidone in lactation.

Methods

SUMMIT is a multicentre, double-blinded randomised controlled trial that will assess whether a higher dose of domperidone (60 mg/day) will further increase maternal breast milk supply compared to a lower dose (30 mg/day). Target population are mothers of preterm infants born at less than 34 weeks’ gestation, who are between 7 and 28 days postpartum, with lactation insufficiency, and can give informed consent. Participants will be randomly allocated into two parallel groups in a 1:1 ratio (n = 50 per group) to receive either 60 mg/day or 30 mg/day of domperidone. The primary outcome will be daily expressed breast milk volume over a 24-hour period on Day 21 post randomisation. As part of the main trial, participants and their infants will be followed until the infant reaches term corrected age or is discharged home from the neonatal unit (whichever occurs first).

Discussion

This will be the largest clinical trial to evaluate the efficacy and safety of domperidone for the treatment of lactation insufficiency following preterm birth. This trial will provide critical evidence to guide clinical practice recommendations regarding what the optimal treatment dose of domperidone should be, while also enabling assessment of the impacts of different dosage regimens on longer-term breast milk feeding outcomes, as well as maternal and infant health outcomes.

Trial registration

This study was prospectively registered via the Australian and New Zealand Clinical Trials Registry (https://www.anzctr.org.au: ACTRN12621000508875 (registered 30 April 2021)).

Background

Breast milk is considered to be the optimal nutrition for preterm infants [1, 2], and studies show that the use of mothers’ own breastmilk while in neonatal intensive care reduces the incidence and severity of preventable morbidities [3, 4]. However, mothers of preterm infants face many challenges in establishing and maintaining an adequate supply of breast milk during their infant’s prolonged hospitalisation.

Some mothers may respond well to non-pharmacological lactation support strategies (i.e., correct use of breast milk pump, increasing expressing frequency) for increasing breast milk supply, but for a large number of mothers, their breast milk supply continues to be insufficient to meet their infant’s needs. Insufficient volume of mothers’ own breast milk has been partly addressed through the introduction of human milk banks, however mothers’ own milk has been demonstrated to be superior to donor human milk with respect to composition and bioactivity, highlighting that focusing on supporting mothers to provide their own breast milk to their infants is key to optimising neonatal outcomes [4].

When insufficient milk supply persists despite the provision of appropriate lactation support, pharmacological treatment with a galactagogue (a medication that increases mother’s milk supply) is often considered [5]. Pharmacological interventions for enhancing breast milk supply have primarily focused on regulating oxytocin or prolactin [6]. Domperidone is the best-studied and most commonly used galactagogue and is thought to increase breast milk production by increasing prolactin levels [6], through its role as a dopamine antagonist [7]. While initially formally indicated for the treatment of gastroparesis and nausea and vomiting it was noted that an untoward effect of domperidone was the increase in prolactin levels (8) An audit of domperidone use as a galactagogue at an Australian tertiary level maternity hospital identified that between 2000 and 2010, domperidone dispensing was associated with approximately 1 in 20 of all hospital births [8]. Further analysis revealed that giving birth to a preterm infant was associated with a seven-fold increase of being dispensed domperidone (OR 7.05; 95%CI 6.20, 8.03) [9]. In the Australian setting, use of domperidone is associated with greater than 1 in 3 births (> 35%) occurring prior to 30 weeks [9]. Internationally, a Canadian study identified that domperidone was dispensed to 1 in 5 women with a full-term birth and 1 in 3 women with a preterm birth [10]. Similarly, an analysis of 247 349 general practice records for women giving birth in the United Kingdom identified that domperidone prescribing was much more common among those with preterm birth (11.6 per 100 postpartum person-years) compared to all births (2.0 per 100 postpartum person-years) [11] Outside of these studies, no other observational studies have examined the prevalence of domperidone use following preterm birth. Our meta-analysis of five RCTs [12–16] showed that compared to placebo, administration of 30 mg/day of domperidone is associated with a modest increase in maternal breast milk volume of 88 mL/day (95% CI 57, 120) in mothers of preterm infants [17]. A more recent Bayesian network meta-analysis of randomised controlled trials [18] showed a similar mean difference of 88.3 mL/day (95% CI 59.48, 118.62), across seven included randomised controlled trials [12–16, 19, 20], however two of these trials [19, 20] included infants born up to 37 weeks of gestation, with no further information regarding exact preterm gestation. The Bayesian network meta-analysis also examined mean differences in milk volumes stratified by daily dose and cumulative dose of domperidone [18]. Both daily doses of domperidone of 30 mg/day (mean difference 100.52 mL/day; 95% CI 50.49, 150.31) and > 30 mg/day (120.35 mL/day; 3.01, 240.43) were more effective than placebo at increasing breast milk volume, however the wide confidence interval for the > 30 mg/day mean difference suggests that more data is required for accurate interpretation of the effect of the higher dose [18]. Length of treatment effect was observed by comparing the mean difference to placebo of the equivalent of 7 days of 30 mg/day domperidone to greater dose and/or length of treatment. Both cumulative doses significantly increased the mean difference compared to placebo, however there was no difference in the mean difference between the two cumulative domperidone doses [18].

The “off-label” use of domperidone for increasing maternal breast milk supply means that there are no approved dosing recommendations. Currently, there is no strong evidence that a higher dose of domperidone is more effective than a lower dose. Placebo controlled trials are limited to studies administering domperidone at a dose of 30 mg/day. Only one head-to-head trial has compared doses of 30 mg/day and 60 mg/day over 4-weeks [21]. This underpowered study (including just 13 women in total) identified a difference of 170 mL/day between groups (p = 0.41) [21]. This provides promise that a higher dose could further increase breast milk volume but requires confirmation in a sufficiently powered study.

Despite this limited evidence, uncertainty in clinical practice regarding optimal domperidone dosing is evident. Based on a 2015 clinical practice survey of Level 3 neonatal units across Australia (response fraction 73%; 16/22), of the 12 sites with a clinical guideline 11 (92%) recommended a standard dose of 30 mg/day, with 1 (8%) recommending a standard dose of 60 mg/day [22]. Further, 3 of 11 (27%) recommend increasing the dose to 60 mg/day if response to 30 mg dose was suboptimal [22]. A Canadian study of 45,000 postnatal women dispensed domperidone, identified that from 2002 to 2011 there was a 20% increase (from 50% to 70%) in the number of women prescribed a dose of 60 mg/day or greater [23]. An Australian cross-sectional survey showed that 355 of 1876 respondents (19%) reported using domperidone, and of those half (n = 179, 50%) reported using a dose of 30 mg/day or less, 44% (n = 155) reported using a dose between 31 and 60 mg/day, a further 6% (n = 22) reported using greater than 61 mg/day [24]. Therefore, there is significant clinical uncertainty regarding the optimal dose of domperidone, with frequent high dose usage in clinical practice despite absence of evidence for improved effectiveness.

Adverse events associated with the use of domperidone in clinical trials include headache, dry mouth, and gastrointestinal disturbances, however, results from our meta-analysis add to the evidence base that domperidone is not associated with increased adverse events compared with placebo (RR 1.05; 95% CI 0.65, 1.71) [17]. The more recent Bayesian network meta-analysis showed similar findings (OR 1.13; 95% CI 0.34, 3.87) [18]. An additional important finding is that no studies have reported any adverse neonatal events, in the preterm or term population, associated with the maternal use of domperidone [12, 14, 25]. Further, a number of studies have evaluated the concentrations of domperidone in breast milk, finding that the absolute infant dose is only 0.04 (95% CI 0.03, 0.07) micrograms/kg/day for a maternal dose of 30 mg/day, and 0.07 (95% CI 0.05, 0.11) micrograms/kg/day for a maternal dose of 60 mg/day [26]. However, domperidone may have indirect effects on infant growth through changes in breast milk composition. One small trial of 46 mothers demonstrated that compared to placebo, administration of domperidone at a dose of 30 mg/day for 14-days resulted in 10% lower protein concentration in expressed breast milk [14]. The clinical relevance of such changes are uncertain.

One small randomised cross-over dose-effect study demonstrated a higher prevalence of adverse events with a higher dose of domperidone [26]. Among the 6 study participants, there were a total of 5 adverse events reported at 30 mg/day (abdominal cramping [n = 1], dry mouth [n = 3], headache [n = 1]), and 12 adverse events reported at 60 mg/day (abdominal cramping [n = 2], constipation [n = 1], dry mouth [n = 5], depressed mood [n = 1], headache [n = 3]) [26]. Also, one woman withdrew from the study as a result of experiencing severe abdominal cramps while receiving 60 mg/day [26]. In contrast, a more recent double blind randomised controlled trial comparing 30 mg/day to 60 mg/day identified no adverse events in either group [21]. In an Australian cross-sectional survey, almost half of all women using domperidone (n = 172, 48%) experienced one or more adverse events, with 9% (n = 31) ceasing use due to adverse events [24]. Of those experiencing side effects, the most common side effect was weight gain (25%, n = 88), followed by headaches (17%, n = 59) and dry mouth (13%, n = 47). Thirteen (4%) women reported experiencing heart palpitations/racing heart with domperidone use. Higher doses were associated with higher rates of adverse events (≤ 30 mg/day, 38.2%; 31-≤ 60 mg/day, 48.4%, ≥ 60 mg/day 72.7%; p < 0.004 [24].

Domperidone has been associated with an increased risk of ventricular arrhythmia (VA) and sudden cardiac death of approximately 4 per 1000 person/years in non-lactating adults compared with no use of medication [27, 28]. This is due to the potential for domperidone to prolong the QT-interval [29]. The increased risk was largely observed among males over 60 years of age [27], and the relevance to younger, healthier lactating women has been questioned [30–32]. A randomised double-blind, placebo- and positive-controlled safety study identified that among healthy female volunteers, domperidone at doses up to 80 mg/day did not cause clinically relevant QTc interval prolongation [33]. Regardless of these facts, because of these cardiac concerns the European Medicines Agency in 2014 recommended that domperidone should only be used up to a maximum dose of 30 mg/day [34].

Evidence regarding the safety of higher doses of domperidone in lactation is provided by the 2016 publication of a Canadian population-based cohort study of 320,351 women in the 6-months postpartum [23]. In this cohort, 90% of women received a dose of domperidone greater than 30 mg/day [23]. No cases of VA or sudden cardiac death were identified among 45,163 women receiving domperidone AND had no previous history of VA [23]. All cases of VA were observed among women with a previous history of VA [23], which represents a contraindication to the use of domperidone and exclusion criteria for this RCT. Lastly, the most recent RCT investigating the use of domperidone in lactating women required that all women have an electrocardiogram (ECG) at study entry and at the end of the 4-week study period [12]. No women had evidence of QTc interval prolongation (n = 90) [12].

Mothers of preterm infants often struggle in producing enough breast milk to meet the daily feed requirements of their infants, especially in the longer-term. This randomised controlled trial will resolve the issue of whether a higher dose of domperidone (60 mg/day) leads to greater improvements in maternal breast milk supply compared to a lower dose (30 mg/day), while also evaluating differences in adverse events, impacts on breast milk composition, and identifying predictors of treatment response to domperidone.

Methods

Study design/aim

The aim of this clinical trial is to investigate whether a higher dose (60 mg/day) of domperidone is more effective than a lower dose (30 mg/day) in increasing daily expressed breast milk volume in mothers of preterm infants < 34 weeks gestation (23^+ 0 to 33^+ 6 weeks), who are expressing breast milk for their infant(s) and are identified as having lactation insufficiency.

The SUMMIT trial is a multicentre, double-blinded randomised controlled trial with two parallel groups and a 1:1 allocation ratio. Enrolment to the trial and associated study visits will take place over a 5-year period between 2022 and 2026. The study protocol was prepared following the SPIRIT guideline and checklist [35].

Study settings

The sponsoring institution and trial coordinating centre is the South Australian Health and Medical Research Institute (SAHMRI) based at the Women’s and Children’s Hospital, North Adelaide, South Australia, Australia. The multicentre study will take place across six locations within Australia, including: Women’s and Children’s Hospital, North Adelaide, South Australia; Flinders Medical Centre, Bedford Park, South Australia; The Royal Women’s Hospital, Parkville, Victoria; Mercy Hospital for Women, Heidelberg, Victoria; Monash Children’s Hospital, Clayton, Victoria; and Royal Brisbane and Women’s Hospital, Herston, Queensland.

Participants

Women intending to supply breast milk to their premature infant(s), who have lactation insufficiency identified by the lactation consultant (LC) within the neonatal unit, will be given information about SUMMIT. Trial staff will contact consenting women to further discuss SUMMIT and screen the potential participants as each participant must meet all of the following eligibility criteria to be enrolled:

• Infant born < 34 weeks’ gestation (23^+ 0 to 33^+ 6 weeks)

• Between 7 and 28 days postpartum (inclusive)

• Lactation insufficiency: defined as breast milk volume less than 300 mL/day in previous 24 h if 7–13 days postpartum, or less than 500 mL/day in previous 24 h if 14–28 days postpartum

• Expressing with an electric pump an average of 6 times a day or more in previous 24–48 h

• Age ≥ 18 years

• Willing and able to comply with all study requirements, including repeat ECG, treatment, timing and/or nature of required assessment and follow-up of mother and baby until infant discharge

Women who agree to participate will provide informed written consent before participation in agreement with The Declaration of Helsinki.

Exclusion criteria

• Evidence of current Q-Tc prolongation (> 470 ms, Bazett’s formula [36], according to screening ECG)

• Previous involvement in SUMMIT study

• Current use of domperidone

• Allergy to domperidone

• History of known or suspected cardiac dysrhythmias or currently on anti-arrhythmic medication

• Currently experiencing mastitis

• Previous breast reduction surgery, or current nipple piercing

• Known chronic renal or hepatic impairment

• Contraindication to breastfeeding (e.g., HIV)

• Known to have a prolactin-releasing pituitary tumour

• Currently taking medications known to alter the metabolism and pharmacokinetics of domperidone and cause Q-Tc prolongation (e.g., oral azole antifungals, erythromycin antibiotics)

• Higher order pregnancies (triplet, or more).

Intervention

After screening, all women who consent to participate will be randomised in a 1:1 ratio to receive either 60 mg/day or 30 mg/day of domperidone (Fig. 1). Domperidone will be administered orally, in either 20 mg or 10 mg capsules three times daily. Participants will be provided with two bottles of study medication to be taken sequentially as outlined in Table 1. Bottle 1 contains 10 mg domperidone capsules to be taken three times daily for two days, subsequently. Bottle 2 contains either 10 mg or 20 mg domperidone capsules to be taken three times daily for the remaining 19 days. Participants will also be provided sufficient capsules for a recommended optional tapering dose of two capsules daily for four days, then one capsule daily for three days before stopping. The study medications will be manufactured in accordance with GMP by ACT Compounding Pty Ltd, T/A LifeCare Compounding Pharmacies (Cranbourne, West Australia). To preserve treatment blinding, different doses of study medications will be identical in appearance, weight and packaging.

Fig. 1.

SUMMIT study participant flow throughout the study

Table 1.

SUMMIT study treatment dosing regimen

	Domperidone Treatment Medication
	Bottle 1	Bottle 2
Intervention Group (60 mg/day)	1 × 10 mg capsule 3 times a day, for 2 days Total mg/day = 30	1 × 20 mg capsule 3 times a day, for 19 days Total mg/day = 60
Intervention Group (30 mg/day)	1 × 10 mg capsule 3 times a day, for 2 days Total mg/day = 30	1 × 10 mg capsule 3 times a day, for 19 days Total mg/day = 30
Total number of capsules per bottle	6 capsules	68 capsulesa

^a Participants are provided with additional domperidone capsules in bottle 2 to allow for a recommended tapering dose of two capsules daily for four days, then one capsule daily for three days before stopping

Adherence

Weekly monitoring of medication adherence will be performed by each study site, with daily tracking of doses taken (or missed) recorded in the milk diary [37] (Fig. 2). Study treatment will be discontinued if any of the of the following situations occur: unacceptable toxicity; treatment continuation is determined not to be in the participants’ best interest; occurrence of an exclusion criterion affecting participant safety (e.g. unavoidable contraindications to domperidone use), or participant declines taking study medication or withdraws their consent.

Fig. 2.

Excerpt from SUMMIT breast milk diary

All study medication will be dispensed and monitored by the site pharmacist. At the end of the study, there will be final reconciliation of study medications received, dispensed, consumed, and returned. Any discrepancies will be investigated, resolved, and documented by the study team. Unused study medications will be destroyed in compliance with applicable regulations.

Concomitant care

Participants in the trial may receive any medications required for the treatment of pre-existing medical conditions or new conditions requiring treatment in the postpartum period.

Women can take herbal galactagogues such as fenugreek, but data on use will be recorded in the electronic Case Report Form (eCRF). Women will also be asked whether they are taking other substances (e.g. lactation cookies) to increase breast milk supply. Off protocol use of domperidone or metoclopramide should not be used during this study. Women will be advised not to use any other medications that prolong the QT interval or inhibit CYP3A4 (e.g. fluconazole). Women will be asked to check with study research staff before starting any new medications to check for potential interactions and ensure all medications are documented in the eCRF.

SUMMIT outcomes

Primary

The primary outcome is daily expressed breast milk volume over a 24-hour period on Day 21 post study enrolment. Following randomisation all participants will be asked to maintain a daily breast milk diary [37] throughout each day of the study (Fig. 2). The diary will act as a record (self-reported) of each time breast milk was expressed, time spent expressing, whether they expressed one breast at a time or both breasts simultaneously, and the volume of expressed milk. Each individual volume of expressed milk will be summed to generate a total for expressed breast milk over a 24-hour period (midnight to midnight). No participant reminders will be utilised, however, breast milk diaries will be reviewed weekly during study visits.

Secondary

Safety outcomes

Key safety outcomes include maternal or infant mortality during the intervention period or within seven days of completing the intervention. Data on maternal adverse events will be reported overall and stratified according to adverse event type and timing. Maternal QT-c prolongation (defined as > 470 ms using Bazett’s formula [36]) will be assessed between day 6 to 15 post study medication starting date.

Maternal breastfeeding outcomes

Continued breast expression or breastfeeding on day 21 and infant discharge from hospital or infant term corrected age (if this occurs before infant discharge); exclusive breast milk feeding on day 21 and within 24 h prior to infant discharge from hospital or infant term corrected age (if this occurs before infant discharge); average breast milk volume over 21 days.

Infant feeding outcomes

Use and volume of supplementation to mother’s own breast milk, such as infant formula or donor breast milk on day 21 and within 24 h prior to infant discharge from hospital or infant term corrected age (if this occurs before infant discharge).

Blood and breast milk biomarkers

These will include exploratory studies of maternal blood and breast milk samples to identify biomarkers that are prognostic and/or predictive of treatment response to domperidone or are influenced by domperidone treatment. Since the identification of new biomarkers correlating with disease activity and the safety or efficacy of treatment is evolving, the definitive list of biomarkers remains to be determined. Studies may include but are not limited to: maternal blood for serum prolactin; maternal DNA for genetic and epigenetic tests to identify differences in gene variants, DNA methylation and telomere length associated with breast milk supply and domperidone treatment; breast milk for inflammatory cytokines and macronutrient composition.

Participant timeline

Table 2 shows the participant timeline for SUMMIT.

Table 2.

Participant timeline for SUMMIT

TIMEPOINT	STUDY PERIOD
	Baseline			Post-randomisation					Follow-up						Close-out
	Day 0			Day 7	Day 14	Day 21			Day 28			Infant discharge home/term CAa			24–36 months infant CA
ENROLMENT
Eligibility assessment/Informed consent	x
Randomisation	x
INTERVENTION
30 or 60 mg/day of domperidone	x			x	x	x
PRIMARY OUTCOME
Expressed breast milk volume on day 21						x
SECONDARY OUTCOMES
Participant Assessments
Maternal Electrocardiogram (ECG)	x			x
Daily Breast Milk Diary	x			x	x	x			x
Demographic/Lifestyle/Health	x
Postnatal Health				x	x	x			x			x
Breast Milk Expression/Breastfeeding	x			x	x	x			x			x
Breast Health	x			x	x	x			x
Breast Milk Expression Experience (BMEE)	x					x
Edinburgh Postnatal Depression Scale (EPDS)	x					x						x
Spielberger State-Trait Anxiety Inventory (STAI-6)	x					x						x
Parental Stressor Scale: NICU (PSS: NICU)	x											x
Maternal Postnatal Attachment Scale (MPAS)												x
Medical Record Data Extraction
Pregnancy/Birth History	x
Infant Feeding	x			x	x	x			x			x
Infant Growth and Anthropometric Measuresb	x			x	x	x			x			x
ANZNN Registry Data Extraction
Infant Morbidity												x
Bayley Scales of Infant Development (BSID)b;															x
Hearing/Vision Evaluationb															x
Maternal Biospecimens c
Blood Samples	x					x
Breast Milk Samples	x			x	x	x

Abbreviations: ANZNN, Australian and New Zealand Neonatal Network; CA, Corrected Age

^a Whichever time point occurs first

^b Infants born at less than 28 weeks’ gestation or less than 1,000 g at birth are routinely followed up at 2–3 years and undergo a neurological and developmental assessment

^c All biological samples are collected for research purposes and only at selected sites

Randomisation and blinding

Participants will be randomised using the Research Electronic Data Capture (REDCap; Vanderbuilt University, TN, USA [38, 39]) Randomisation Module. Participants will be randomised to the high dose (20 mg domperidone three times daily; 60 mg/day) or low dose group (10 mg domperidone three times daily; 30 mg/day) in a 1:1 ratio. The randomisation schedule will be prepared using ralloc.ado in Stata (v18, StataCorp, TX, USA) by an independent statistician who is not involved with trial participants or data. The randomisation schedule will be stratified according to study centre. Participants, care providers, outcome assessors, trial investigators, and data analysts will be blinded to randomisation group.

Data collection methods

Participants will be asked to partake in a total of up to six study related appointments for SUMMIT, these will be conducted within the hospital, within a designated research clinic space or by telephone (where appropriate). Participants will be encouraged to continue to partake in study related appointments irrespective of whether they discontinue or deviate from intervention protocols, under an intention-to-treat approach.

Questionnaires and assessments will be completed at study visits as summarised in Table 2.

Maternal blood and breast milk samples will be collected at various timepoints as indicated in Table 2. Approximately 30 mL of whole blood will be collected at baseline and again at day 21 post commencement of treatment (see Table 2). Samples will be used to evaluate changes in prolactin, as well as cardiometabolic and inflammatory biomarkers. Blood samples will also be used to obtain maternal DNA. This will be used to perform genetic and epigenetic tests to identify differences in gene variants, DNA methylation and telomere length associated with breast milk supply and domperidone treatment. Following enrolment to the study, breast milk samples (3–8 mL; Table 3) will be collected at 4 timepoints (see Table 2) to assess changes in macronutrient composition.

Table 3.

Biospecimen sample collection for the SUMMIT study^a

Sample type	Description
Maternal blood	• 30 mL blood (Serum tube 10 mL, potassium EDTA 10 mL and 5 mL, and Trace Elements 5 mL) • Cryopreservation of plasma trace element, serum, plasma, buffy coat, red blood cells aliquots • Sample collected at day 0 and day 21.
Breast milk	• Expressed milk (3–8 mL) from a single breast expression episode • Cryopreserved aliquots of whole milk • Sample collected at days 0, 7, 14 and 21.

^a All biological samples are collected for research purposes only at selected sites only

A repeat ECG (first ECG performed during screening to exclude women with current QT-c prolongation, defined as > 470 ms using Bazett’s formula [36]) will be conducted between day 6 and 15 post study medication starting date. Each ECG will be reviewed by a Cardiologist (NM). If QT prolongation, or any other incidental cardiac abnormalities, is observed the participant and their General Practitioner will be notified. The ECG will be conducted, interpreted and actioned in accordance with our standard operating procedure (SOP).

Summarised assessments and questionnaires at the various study visits are outlined in Table 4 and correlate to the corresponding schedule of assessments in Table 2.

Table 4.

Overview of data collected as part of the SUMMIT study

Study Assessments	Description
Participant Assessments
Maternal Electrocardiogram	Electrocardiogram (ECG) which records the electrical activity of the heart muscle and will display data on a trace, will be used as part of the Screening and Eligibility process to rule out clinically relevant pre-existing QT prolongation. A repeat ECG will also be conducted from day 6 to 15 post study medication start date. Each ECG will be reviewed by an adult Cardiologist. If QT prolongation (> 470 ms, Bazett’s formula [36]) or any other incidental cardiac abnormalities, is observed the participant and their General Practitioner will be notified.
Breast Milk Diary	Daily breast milk diary [37] recording the number of times each breast was expressed, the method used to express, the volume of expressed milk, duration and if applicable information pertaining to direct breastfeeding until at least seven days following the intervention phase (i.e. until at least 28 days after study enrolment). The breast milk diary will be commenced 24–48 h prior to randomisation to allow for baseline data to be collected. In addition, women will be asked to record medication doses taken (or missed) throughout the intervention period. Any adverse events can also be recorded in the diary.
General Information	Contact information, including name, address, phone number, email address, hospital ID number, secondary contacts and general practitioner details.
Demographic/Lifestyle/Health	Maternal age, height, pre-pregnancy weight, smoking status, alcohol intake, pre-existing medical conditions, ethnicity and Indigenous status, country of birth, languages spoken at home, combined household income status, workforce status, education level, marital status, previous pregnancy and breastfeeding history, pregnancy complications, medication use, postnatal contraception use, breast type/shape [40], breast development since puberty and during pregnancy [40]. Paternal ethnicity and Indigenous status, country of birth, workforce status, education level.
Postnatal Health	Postnatal health issues during the study (e.g. infections, cold/flu), as well as any medications (prescription or non-prescription), herbal supplements or multivitamins taken.
Breast Milk Expression/Breastfeeding	Breastfeeding values and intended duration of breastfeeding, breast development previous to and during pregnancy, time of first expressing following birth, perceived onset of lactogenesis, methods and type of breast milk expressing (including type of breast milk pump), duration of breast milk expressing or breastfeeding, and provision of lactation support.
Breast Health	Questionnaire adapted from Fetherston 2006 [41] to assess general breast health and mastitis risk. Includes data on breast fullness, breast pain and discomfort, duration of pain/discomfort, presence of cracks or grazes on the nipples, fever, as well as any areas of heat, swelling or redness on breasts.
Breast Milk Expression Experience (BMEE) Questionnaire	BMEE provides a measure of milk expression experience across three dimensions: (1) social support for milk expression; (2) ease of learning how to express milk; and (3) personal experiences of milk expression [42].
Mental health and wellbeing questionnaires • EPDS • STAI-6 • PSS: NICU • MPAS	Assessment of depression and anxiety symptoms will be collected using standardises validated questionnaires (Edinburgh Postnatal Depression Scale [EPDS] [43], Spielberger State-Trait Anxiety Inventory [STAI-6]) [44]. Stress related to infant hospitalisation will be assessed using the Parental Stressor Scale: NICU (PSS: NICU) [45]. Maternal-infant attachment will be assessed using the Maternal Postnatal Attachment Scale (MPAS) [46].
Medical Record Data Extraction
Pregnancy/Birth History	Previous pregnancy outcomes, method of conception for this pregnancy, antenatal anthropometrics, antenatal EPDS and Antenatal Risk Questionnaire (ANRQ) completion and scores, booking haemoglobin, ferritin, vitamin D, HbA1C and repeat haemoglobin, ferritin, vitamin D and oral glucose tolerance test (OGTT) at 28 weeks. Pregnancy complications and associated treatments (e.g., preterm premature rupture of membranes (PPROM), pre-term labour tocolytic use, antenatal corticosteroid use, pregnancy-induced hypertension, preeclampsia, HELLP, eclampsia, placental abruption, gestational diabetes mellitus, antepartum haemorrhage or intrauterine growth restriction. Labour and birth details including predelivery admission, onset of labour, mode of delivery, birthdate, gestation, analgesia for labour and/or caesarean section, active third stage management drugs, infant weight(s), retained placenta, placenta weight, estimated blood loss, perineal status, postpartum haemoglobin and ferritin, treatments for blood loss/anaemia, and Intensive Care Unit or High Dependency Unit admission. Antibiotic use prior to and during labour/delivery, antibiotic use following delivery and at discharge. Length of hospital admission.
Infant Feeding	Daily total enteral intake (expressed breast milk, donor breast milk, infant formula), parenteral intake, as well as proportion of enteral feeds consisting of mothers’ own breast milk.
Infant Growth and Anthropometric Measures	Daily record of anthropometrics (weight, length, head circumference) during admission.
ANZNN Registry Data Extraction [47][46]
Infant Morbidity	Infant morbidity data will be obtained from the Australian and New Zealand Neonatal Network (ANZNN) registry (maternal, perinatal and birth data, neonatal interventions and outcomes).
Bayley Scales of Infant Development (BSID)/ Hearing and Vision Evaluation	All infants born at less than 28 weeks’ gestation or less than 1,000 g at birth are routinely followed up at 2–3 years and undergo a neurological and development assessment. A standardised set of data collected and reported to the ANZNN will be obtained and includes BSID (Bayley-III or Bayley-4 (A&NZ)) developmental test results; growth, hearing and/or vision impairment.

Patient and public involvement

Members of the public were involved at several stages of the trial, including the design, management and conduct of the trial. Input was sought from members of the Robinson Research Institute consumer group comprising women who had previously given birth to a preterm infant. Main trial results will be disseminated to trial participants, and we will seek patient and public involvement in the development of dissemination materials.

Sample size calculation

A sample of 100 women (50 per arm) yields 90% power to show a difference in the mean daily breast milk volume of 100 mL/day between the high and low dose domperidone groups, allowing for 10% loss to follow-up. This is based on a type I error rate of 5%, an estimated standard deviation of 220 mL, and an estimated correlation between the outcome at baseline (to be adjusted for in the analysis) and day 21 of 0.75.

Data management

A purpose-built REDCap database will be used for all data entry and study management. Trained researchers at each study site will collect and enter data directly into REDCap, no confidential data are stored on data entry machines. Paper-based CRFs will also be available where needed and securely stored in a locked office at the study site. Confidentiality will be maintained at all times, and only research staff directly involved in the study will have access to the information. Regular blinded reviews of data quality and data checks will be performed during the duration of the study. Study documents will be retained at the study site or in archives for at least 33 years after study completion in line with the data retention schedules for research involving minors. The study electronic data will be stored indefinitely on SAHMRI’s secure servers with access only granted to authorised study personnel.

De-identified sequencing datafiles will be archived in the Gene Expression Omnibus (GEO or equivalent database) and will be free of all identifiers that would permit linkages to individual research participants and variables that could lead to deductive disclosure of the identity of individual subjects, in accordance with the NIH Statement on Sharing Research Data [48].

Statistical analysis methods

The primary analysis will be performed according to the treatment group to which participants were randomised (intention-to-treat principle). A secondary per-protocol analysis will also be performed for each of the primary and secondary outcomes.

The primary outcome of daily expressed breast milk volume on Day 21 post intervention will be compared between treatment groups using a linear regression model. The results will be expressed as a difference in means with a 95% confidence interval and two-sided p-value. Adjustment will be made for baseline breast milk volume and study site. A p-value of less than 0.05 will be considered to indicate statistical significance. Analysis of secondary outcomes will use log- binomial regression models for binary outcomes and linear regression models for continuous outcomes with adjustment for stratification variables and other pre-specified prognostic baseline variables. Results will be presented as relative risks and differences in means respectively, along with 95% confidence intervals. Missing data will be addressed using multiple imputation. Planned sub-group analyses of the primary outcome will be undertaken according to plurality (singleton vs. twins), parity (primiparous vs. multiparous), infant gestation at birth (23–26 vs. 27–29 vs. 30–31 vs. 32–33 weeks’ gestation), postpartum day at enrolment (7–13 days vs. 14–28 days) and maternal body mass index (underweight/normal weight vs. overweight/obese). Effect modification by each of these factors will be assessed separately by including interaction effects with treatment group in the statistical models. Sensitivity analyses will also be performed using the original unimputed data. All analyses will follow a pre-specified statistical analysis plan. No interim analysis is planned.

Data monitoring and safety

The trial coordinating centre will perform ongoing monitoring via weekly reports of participants screened/enrolled/randomised, visits due/overdue/missed, adverse and serious adverse events, product inventory and dispense records, sample collection logs, and participant communication logs. Reports will be assessed at monthly trial management meetings of the SUMMIT Executive Management Team.

In addition, an independent Data and Safety Monitoring Committee (DSMC) will be set up to review the yearly progress of the trial and provide feedback to the Trial Steering Committee. The DSMC will review general study progress (recruitment, compliance, loss to follow-up), breast milk supply and key secondary/safety outcomes, and will have access to unblinded data. The DSMC will consist of a neonatologist/obstetrician, clinical pharmacologist and a biostatistician.

Any unfavorable and unintended sign, symptom or illness that develops or worsens during the period of the study will be classified as an adverse event (AE), whether or not it is considered to be related to the study treatment. Adverse events will include unwanted side effects, injury or inter-current illnesses, and may be expected or unexpected. These will be recorded on the eCRF, completed during phone or in-person interviews. Participants also have the ability to document AEs in a specific section of the provided breast milk diary. Any adverse events or complications arising from trial participation will be promptly addressed, with participants referred to the neonatal unit lactation consultant or their usual treating medical practitioner as required.

Weekly safety evaluations will be conducted during the intervention period. The study or site Principal Investigator can be directly contacted by participants if there are any concerns regarding their medication. The period for adverse event reporting will be from the time of first dose until seven days post administration of final study medication intervention (Day 28), and participants will be followed up face-to-face or by telephone interview at this time.

A Serious Adverse Event (SAE) Committee blinded to the treatment allocation will be established and chaired by LEG and will include medical practitioners with expertise in obstetrics and neonatology. The primary role of the SAE Committee is to review all maternal and infant SAEs to determine whether there is any likelihood that involvement in the trial could have contributed. All maternal and infant SAEs will be reported to the Chair of the Steering Committee (LEG) within 24-hours of the site becoming aware. The SAE Committee will assess expectedness of all SAEs by considering the adverse effects identified within the study Investigational Brochure and Australian Product Information for domperidone. As it is not expected that the Investigational Medicinal Product or study protocol will cause death to the infant, all infant SAEs will be classified as unexpected. The Australian Therapeutic Goods Administration and HRECs will also be notified for any SAEs which are unexpected and deemed to be related to the study treatments.

An individual participant may only be unblinded in emergency situations, where the Investigator decides a participant cannot be adequately treated without knowing the identity of their treatment allocation. To break the randomisation code the Principal Investigator must contact the randomisation facility/personnel. The time, date, participant study ID and reason for unblinding will be documented. Events leading to the emergency breaking will be recorded in an SAE report form. Any protocol deviation will be recorded in the patient medical record and on the eCRF and reported to the Principal Investigator, who will determine the potential impact and action(s) required.

Ethical issues and dissemination

This study will be conducted according to the Note for Guidance on Good Clinical Practice (CPMP/ICH/135/95) annotated with TGA comments (Therapeutic Goods Administration DSEB July 2000) and in compliance with applicable laws and regulations. The study will be performed in accordance with the NHMRC Statement on Ethical Conduct in Research Involving Humans, and the NHMRC Australian Code for the Responsible Conduct of Research, and the principles laid down by the World Medical Assembly in the Declaration of Helsinki 2008. All study materials and study protocols (SUMMIT, current version 7.2, dated 27 July 2025) have been reviewed and approved by the Women’s and Children’s Health Network Human Research Ethics Committee (HREC), South Australia (HREC/19/WCHN/149) and Mercy Health HREC, Victoria (2021-065), as well as the research governance officers at each of the study sites. Any change to the protocol or Informed Consent Form that affects the scientific intent, study design, patient safety or may affect a participant’s willingness to continue participation in the study is considered an amendment, which require approval from the HRECs before becoming effective.

Beyond the core objectives of the main trial, participants will be asked to provide extended consent for the use of their data and biological specimens to be used in future research related to breast milk production or how breast milk feeding influences maternal or infant health and development.

Results from this study will be published in a peer-reviewed journal and shared at domestic and international conferences. A brief lay summary of the study findings will be shared with all study participants. No identifiable information will be included in any reports or publications, all data collected will be kept confidential.

Discussion

Domperidone is utilised “off-label” for the treatment of lactation insufficiency, therefore there are no approved dosing recommendations. Only one, notably underpowered study, compared doses of 30 mg/day with 60 mg/day, and identified a mean difference of 170 mL/day between groups (p = 0.41) [21]. The SUMMIT study is sufficiently powered to show a difference in the mean daily breast milk volume of 100 mL/day between the 60 mg/day and 30 mg/day domperidone doses.

The study also seeks to monitor the safety of the increased dosage of the domperidone. Domperidone has the potential to prolong the QT interval and has been associated with ventricular arrhythmia and torsades des pointes in studies involving non-lactating adults. However, several post-marketing surveillance studies involving women of childbearing age or lactating women have failed to demonstrate any differences in either QT-interval prolongation [12, 33, 49] or cases of ventricular arrhythmia [23, 50]. Safety evaluations will be conducted each week during the intervention period, and the period of adverse event reporting will be from the time of the first dose until seven days post final study medication administration.

Further explorations within this study include the impact of domperidone on breast milk composition and infant growth. Preterm infants have higher protein, energy, mineral, and electrolyte requirements than term infants and are at risk of cumulative nutritional deficits and postnatal growth restriction during early life [51]. Multi-component breast milk fortifiers (including protein) are routinely added to expressed breast milk to assist preterm infants in meeting their nutrition and growth requirements. Whether domperidone alters macronutrient composition of human milk has been examined in only one small previous study [14]. Compared to placebo, administration of domperidone at a dose of 30 mg/day for 14-days resulted in 10% lower protein concentration [14]. Given the clinically significant reduction in protein, confirming such findings in a larger clinical trial is critically important. This is particularly important given lower protein intake can lead to growth faltering in human milk-fed preterm infants, which is a strong predictor of poor neurodevelopment [52]. Despite the expected beneficial effects of improvements in maternal breast milk supply on infant growth and neurodevelopment, no studies have specifically evaluated these outcomes following maternal use of domperidone in lactation. Further, no studies have examined the impact of higher doses of domperidone (i.e. 60 mg/day) on breast milk composition. If domperidone does reduce protein concentration, it would be important for clinicians to adjust for the lowered protein intake in order for preterm infants to grow appropriately.

Determinants of treatment response to domperidone will also be examined. Increased prolactin levels have been observed in most [14, 15, 21, 26], but not all [16] domperidone studies. When restricted to studies comparing a higher (60 mg/day) versus lower (30 mg/day) dose of domperidone, one demonstrated a greater increase in serum prolactin concentration associated with the higher dose [21], while the other found no difference [26]. Overall, the magnitude of increase in prolactin is variable and its relationship to domperidone dose and milk volume is uncertain. Individual variation in prolactin response could be one explanation for the observed wide variability in treatment response to domperidone.

Notably, one study demonstrated that 1 in 3 women who received domperidone experienced no improvement in breast milk volume [26]. Understanding factors that contribute towards response to domperidone is important in identifying groups of women most likely to benefit from treatment, while also enabling potential identification of other important targets for improving breast milk volume. One such factor may be maternal BMI, with obesity associated with reduced prolactin actions subsequent to alterations in the lactating mammary gland [53]. This may be due to hormonal or inflammatory differences associated with obesity and warrants further investigation. Of particular interest is the role of the toll-like receptor 4 (TLR4) immune signalling pathway, which plays an important role in obesity and the metabolic syndrome [54]. Further, the TLR4 pathway, regardless of whether it is activated by an infection or a sterile inflammatory pathway, is recognised as a key mediator of preterm labour [55]. This is important as TLR4 signalling may be a key factor involved in delayed secretory activation and development of low milk supply, commonly experienced by many mothers of preterm infants. When exposed to a bacterial by-product in a mouse model of mastitis [56], it is TLR4 signalling that is responsible for low milk supply, not the bacterium itself [57]. This surprising finding has led to a new paradigm for low milk supply, whereby TLR4 signalling leads to lactocyte cell death and consequently a reduced capacity to produce milk [58]. In a pilot study, 40% (6 of 15) of healthy Australian women exhibited elevated interleukin 8 (IL8), an inflammatory cytokine produced through TLR4 signalling, in their breast milk 4 weeks postpartum (Ingman & Amir: unpublished). The presence of asymptomatic breast inflammation could result in unresponsiveness to domperidone whereby the prolactin-mediated signal to stimulate lactocytes to produce more milk is negated by the inflammatory signal triggering lactocyte cell death.

To date, only one placebo-controlled study has reported longer-term breastfeeding outcomes following the use of domperidone for the treatment of low milk supply. Campbell-Yeo et al. observed that two weeks following the completion of the trial, the proportion of women reporting any breastfeeding was higher among those who had received domperidone compared to placebo (86% vs. 63%) [14]. However, by the time of infant discharge from the neonatal unit, there was no longer any observed difference between those who received domperidone or placebo (55% vs. 52%) [14]. Notably, type of breastfeeding was not reported, so it is unclear whether there were any differences in the proportion of women exclusively breastfeeding their infants at the later time points. It is therefore uncertain what impact early identification and pharmacological management of lactation insufficiency with domperidone has on longer-term breastfeeding outcomes.

This study has several limitations that warrant consideration. The trial population is restricted to mothers of preterm infants less than 34 weeks’ gestation, which may limit generalisability to later preterm or term populations with lactation insufficiency. Milk volume is recorded using maternal diaries, introducing potential for missed entries or measurement inaccuracies despite standardised training. Although powered to detect a clinically meaningful difference in daily breast milk volume, the study may be underpowered to identify rare adverse events, particularly those related to cardiac safety, or to evaluate longer-term infant growth and neurodevelopmental outcomes. Furthermore, variation in adherence to study medication, use of concomitant galactagogues, and differences in lactation support services across recruitment sites may introduce heterogeneity that could influence observed effects.

Despite these limitations, the SUMMIT trial addresses an important gap in evidence regarding the optimal domperidone dose for treating lactation insufficiency in mothers of preterm infants. The SUMMIT trial will provide high-quality data on the comparative efficacy of 30 mg/day and 60 mg/day dosing, while rigorously monitoring maternal safety. By also exploring the effects of domperidone on milk composition, biomarkers of treatment response, and breastfeeding outcomes, the trial will generate valuable insights beyond milk volume alone. Findings from this study will inform clinical practice, guide safe and effective prescribing, and ultimately support mothers in achieving sufficient breast milk supply to optimise outcomes for preterm infants.

Abbreviations

AE

Adverse Event

ANRQ

Antenatal Risk Questionnaire

ANZNN

Australian and New Zealand Neonatal Network

BSID

Bayley Scales of Infant and Toddler Development

BMEE

Breast Milk Expression Experience questionnaire

CA

Corrected Age

DSMC

Data Safety Monitoring Committee

ECG

Electrocardiogram

EPDS

Edinburgh Postnatal Depression Scale

HREC

Human Research Ethics Committee

IL

Interleukin

MPAS

Maternal Postnatal Attachment Scale

NEC

Necrotizing enterocolitis

NICU

Neonatal Intensive Care Unit

PSS:NICU

Parental Stressor Scale: Neonatal Intensive Care Unit

REDCap

Research Electronic Data Capture

SAE

Serious Adverse Event

SAHMRI

South Australian Health and Medical Research Institute

SOP

Standard Operating Procedure

SF36

36-item Short Form Health Survey

STAI-6

Six-item Spielberger State-Trait Anxiety Inventory

SUMMIT

Supporting Mothers Milk Intervention Trial

TLR-4

Toll-like receptor - 4

VA

Ventricular Arrhythmia

Author contributions

Drafted the manuscript: LEG. Revised the manuscript: LW, MS, KAM, AK, WVI, LS, SJ, EK, LY, NM, PP, ARR, LHA. Conceptualised and designed the study: LEG, LW, WVI, LS, SJ, EK, LY, NM, PP, ARR, and LHA. Obtained grant funding to conduct the study: LEG, MS, WVI, LS, SJ, EK, and LHA. All authors read and approved the final manuscript.

Funding

This study is funded by the Channel 7 Children’s Research Foundation, South Australia, Australia awarded to LEG (CRF-210323), Rebecca L Cooper Medical Research Foundation (PG2019314), The Hospital Research Foundation (CS0064096) and The National Health and Medical Research Council (GNT1165457). The funder/s have no role in the study design; collection, management, analysis and interpretation of data; writing of the report; and the decision to submit the report for publication and have no authority over any of these activities.

Data availability

The trial is ongoing. We do not have consent from the participants to publish the full dataset, but de-identified individual participant data will be available from the corresponding author on reasonable request when findings of the trial has been published. The results will be presented in international peer reviewed journals and at relevant international conferences. Copies of study related materials (e.g. trial protocol, statistical analysis plan, and data collection forms) will be made available on request to the corresponding author.

Declarations

Ethics approval and consent to participate

Ethics approval has been granted from the Women’s and Children’s Health Network Human Research Ethics Committee (HREC), Women’s and Children’s Hospital, 72 King William Road, North Adelaide, South Australia (HREC/19/WCHN/149) and Mercy Health HREC, Mercy Hospital, 163 Studley Road, Heidelberg, Victoria (2021-065). Informed consent is obtained from all individual participants in this study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.