The objective of this study is to evaluate the efficacy of an oral stimulation program on the length of the transition period in preterm infants (primary outcome), the length of hospital stay and the breastfeeding rates at discharge (secondary outcomes).

Oral feeding issues in preterm infants are a growing concern for neonatologists because the difficulties preterm infants show in transitioning from tube to independent oral feeding lead to delayed hospital discharge, maternal stress and rising financial burden.

Safety in infant oral feeding implies minimal risk of aspiration and requires the adequate coordination of sucking, swallowing and respiration.

Sucking and swallowing are present in early foetal life, but the coordination of sucking with swallowing and breathing is not thought to occur before 34 weeks post-menstrual age (PMA). Thus, preterm infants tend to suck, swallow and breathe in an alternate, rather than coordinated fashion.

Accelerated maturation of sucking reflex and earlier readiness for bottle-feeding is reported when preterm neonates were given non-nutritive sucking (NNS) opportunities during gavage feeding. Recent evidence suggests that the sensory consequences associated with the production of NNS have beneficial effects on oral feeding performance and the development of specific sucking skills.

At present, it is customary for infants who have stable cardiopulmonary status to be introduced to oral feeding around 33 to 34 weeks PMA. They usually take days or weeks in the transition period of combined gavage and oral feeding before reaching full oral feeding.

Recent studies suggest that an oral stimulation program associated to NNS applied to preterm infants for at least 10 days in the period of full gavage feeding can facilitate their oral feeding progress, improves breastfeeding rates among preterm infants and decreases the length of stay.

A meta-analysis found that NNS decreases significantly the length of hospital stay in preterm infants with an acceleration of the transition from tube to bottle feeds and better bottle feeding performance, without any negative outcomes reported.

Actually, in our neonatal intensive care unit, preterm infants don't receive any oral stimulation or a pacifier to develop NNS before oral feeding introduction and they show frequently, even for the infants born after 32 weeks' gestational age, difficulties in the transitioning from tube to independent oral feeding.

In a prospective randomized controlled clinical trial, prefeeding oral stimulation will be compared with the conventional approach for oral feeding initiation.

Considering a 15 days follow-up for each patient after the introduction of oral feeding, an expected rate of patients performing the full oral feeding after 15 days of 95% in the stimulated group and of 70% in the non stimulated group, the minimal sample size needed to observe a difference between the two groups would be 37 patients in each group with a power of 80% and an alpha level of 5%.

A total of 74 preterm infants are going to be included from the neonatal intensive care unit at the Children's Hospital of the Centre Hospitalier de Luxembourg, Luxembourg.

The investigators are going to include preterm infants from 26 to 33+6 weeks' gestational age, as determined by date of last menstruation and first-trimester ultrasound. Prior informed consent is going to be obtained from the parents/guardians of the newborns for their participation in the study.

Infants' randomization, into an experimental and a control group, is going to be done when they reach 32 weeks PMA for the infants born before 32 weeks, and at birth for the infants born after 32 weeks' gestational age. The process is done using sequential numbers, kept in sealed, opaque, non-translucid envelopes. Randomization is going to be stratified based on gestational age ranges (26-27+6, 28-29+6, 30-31+6, 32-33+6) to ensure a similar gestational age distribution in the two groups.

Infants in the interventional group are going to receive a prefeeding oral stimulation (group 1), the infants in the control group (group 2) are neither being stimulated nor will them be proposed a pacifier before or gavage feeding.

The intervention will start at 32 weeks PMA in a medically stable infant feed by gavage feeding for the infants born before 32 weeks' gestational age, and, as soon as clinical stability is acquired, for the infants born after 32 weeks.

The investigators defined as a complete oral feeding a minimal quantity of 5 millilitres that the preterm infant has to drink in less than 10 minutes by bottle or by breastfeeding. These 5 millilitres are the usual quantity that a preterm baby born at 34 weeks' gestational age is able to drink by himself at each meal.

The program is going to be interrupted if infants are medically unstable and/or have any episodes of oxygen desaturation, apnoea and/or bradycardia during the intervention.

Gavage feeding will begin when the preterm infant is clinically stable in terms of hemodynamics and has presented peristalsis according to criteria established by the caretaker staff. Oral diet progress depends on infant's tolerance and will be about 20 ml/kg/day.

The change from gavage to oral feeding is going to be initiated at 34 weeks PMA, after the beginning of oral stimulation program in group 1.

The gavage feeding will be stopped and the orogastric tube removed, when the infant attains an oral feeding volume of more than 120 ml/kg/day. A weight lost of 5 % is authorized after stopping gavage feeding.

Full oral diet is defined by an oral intake of milk > 150 ml/kg/day, by bottle or by breastfeeding, for 3 consecutive days.

The expected benefits of this study are a reduction in the length of stay in our neonatal unit in the stimulated group, due to the reduction in the length of the transition period in these preterm infants. A minimisation of the hospitalization cost and a reduction in the parental stress can by expected with an earlier hospital discharge.

An increase of the breastfeeding rates at discharge is expected. With the improved feeding performance in the stimulated group, the maternal stress concerning breastfeeding is reduced and the risk of failure is smaller. A better mother-child attachment can be expected with this study.

Inclusion Criteria:

• preterm infants from 26 to 33+6 weeks' gestational age, as determined by date of last menstruation and first-trimester ultrasound, hospitalized in our neonatal unit.

Exclusion Criteria:

• Congenital malformations (chromosomal disorders, malformations of head and face, neurological, cardiac, digestive or pulmonary malformations)
• Severe asphyxia (hypoxic-ischemic encephalopathy)
• Presence of third or fourth degree intracranial haemorrhage
• Severe periventricular leukomalacia
• Severe chronic lung disease
• Severe hospital infection during the study period
• Necrotising enterocolitis during the study period
• Feeding interruption for more than 10 days during the study period
• Death during the study period
• Transfer to another hospital before discharge.