Oral Feeding in the Preterm Infant

Abstract

After completing this article, readers should be able to: Oral feeding issues in preterm infants are a growing concern for neonatologists because attainment of independent oral feeding is one of the prerequisites for hospital discharge. With the increase in survival of infants born continuously more preterm, understanding such issues has a certain urgency. Concerns do not pertain only to difficulties encountered by neonatologists during the birthstay hospitalization, but also by pediatricians and pediatric gastroenterologists who attend to long-term feeding difficulties/disorders, such as oral feeding aversion. Indeed, greater than 40% of patients followed in feeding disorder clinics are former preterm infants. Research over the last decade has begun to shed light on the development of oral feeding skills in these infants as they mature, which has increased understanding of their limited skills at varying postmenstrual ages. Such knowledge is crucial in clinical practice insofar as expectations of these infants’ oral feeding performance must take into account the ever-changing level of maturity of their skills.This article focuses on the development of sucking, swallowing, and respiration (as it pertains to oral feeding) and the coordination of these three functions. Additional factors, separate from infant feeding skills, also are discussed because they can affect oral feeding performance. This review examines information gathered from bottle feeding because more extensive research has been conducted on bottle feeding than on breastfeeding. However, in no way should bottle feeding be interpreted as favored over breastfeeding.Clinicians look for signs of readiness to feed orally before weaning preterm infants from tube feeding. However, such signs are ill-defined because of limited information on the oral feeding skills necessary for infants to feed safely and successfully by mouth. At present, it is customary for infants who have stable cardiopulmonary status to be introduced to oral feeding around 33 to 34 weeks postmenstrual age. At this age, their sucking pattern resembles that of term infants, ie, rhythmic alternation of suction and expression, the two components of sucking. However, studies have shown that the mature sucking pattern is not necessary for safe and successful oral feeding. Indeed, infants can feed orally using only the expression component, with no suction.It is important to recognize that adequate oral feeding in infants does not rest solely on adequate sucking, but also on appropriate swallowing, respiration, and the coordination of sucking, swallowing, and respiration. Furthermore, independent of such skills, it is now acknowledged that an infant’s behavioral state and organization during feeding, the nursery environment (eg, light, sound), and a caretaker’s approach to oral feeding can affect the infant’s performance significantly. Such awareness has grown over the last few years as a result of the introduction of the developmental care program in growing numbers of nurseries.Sucking includes nutritive sucking, when milk is ingested from a bottle or breast, and non-nutritive sucking, when no liquid is involved (eg, pacifier). Mature nutritive sucking is defined by the rhythmic alternation of suction (the negative intraoral pressure that draws milk into the mouth) and expression (the positive pressure generated by the compression/stripping of the nipple [breast or bottle] that ejects milk into the mouth). In a longitudinal study conducted during bottle feeding of “healthy” infants, feeders and growers, born at less than 30 weeks’ gestation (ie, who had no major medical issues), we observed that mature sucking is attained sequentially and characterized this progression into five primary stages (Fig. 1). In brief, sucking begins with the emergence of an arrhythmic expression with no suction (Stage 1). As expression becomes rhythmic, arrhythmic suction appears (Stage 2), in time also acquiring rhythmicity (Stage 3). During these initial three stages, infants can switch from using the expression component only to a developing alternation of suction/expression within a feeding session. As the alternation of rhythmic suction/expression matures, suction amplitude increases along with the duration of sucking bursts (Stages 4 and 5). Such characterization of the progression of nutritive sucking was positively correlated not only with postmenstrual age, as expected, but also with oral feeding performance, as defined by the rate of milk transfer (mL/min) and the ability of the infants to complete their feedings within an allocated time (20 min).Non-nutritive sucking, although not directly implicated in feeding, has its benefits. Studies showed that a pacifier offered during gavage feeding accelerated the transition from tube to oral feeding and enhanced weight gain, leading to an overall shortened hospital stay. Although non-nutritive sucking enhanced gastrointestinal motility, it did not increase the levels of gastrointestinal hormones, namely, gastrin, motilin, insulin, or insulin-like growth factor-1. Such sucking also improved the infant’s behavioral state and organization during oral feeding and decreased distress and pain.A rhythmic non-nutritive sucking pattern resembling that of mature nutritive sucking has been used as an indicator of readiness for oral feeding. However, this approach is now debatable. We observed that preterm infants who demonstrated a mature sucking pattern when sucking on a pacifier did not necessarily do so during bottle feeding. Figure 2 shows the non-nutritive and nutritive sucking patterns of a preterm infant (30 weeks’ gestation) at 43 days after birth (36.1 weeks postmenstrual age) when taking four oral feedings per day. He was offered the pacifier for the first 3 minutes of the monitored session and given a bottle immediately after. During non-nutritive sucking, he demonstrated a mature rhythmic alternation of suction/expression (Fig. 2a), which was not maintained when offered the bottle (Fig. 2b). Figure 3 shows the results obtained in a similar manner from an infant (29 weeks’ gestation) at 54 days after birth (36.4 weeks postmenstrual age) when taking eight oral feedings per day. For this infant, the sucking pattern was similar during non-nutritive and nutritive sucking.We speculated that the discrepancy observed in the first infant reflected uncoordinated suck-swallow-respiration during oral feeding. During non-nutritive sucking, with swallowing being at a minimum (but for the infant’s own secretion), sucking and respiration can function independently from each other (Fig. 4a). However, when frequent swallowing occurs, as in the case of nutritive sucking, the sucking, swallowing, and respiration need to be closely linked to avoid aspiration (Fig. 4b). As such, immature nutritive sucking does not necessarily reflect sucking ability, but also the coordination of suck, swallow, and respiration. Thus, non-nutritive sucking is a good index of sucking skills, but not necessarily of an infant’s readiness to feed by mouth.With maturation, the swallowing process becomes more adaptable, handling both larger and more varied bolus sizes. It also is swifter, as evidenced by the increase in swallowing rates observed as infants mature (Table 1). This may be partly explained by the increase in intrabolus pressure, the lingual force generated that propels the bolus to the pharynx to trigger the swallowing reflex. This reflex can be initiated with only 0.04 mL of fluid if it is delivered to the correct region of the swallowing receptor nerve ending located in the posterior region of the pharynx. Thus, it would be expected that the better the bolus formation and the stronger the intrabolus pressure, the more rapidly a swallow would be initiated. Such observations emphasize the close link between sucking and swallowing.Oral feeding is safe if swallowing occurs with the proper timing of tracheal closure to prevent tracheal penetration/aspiration into the lungs. This necessitates not only timely closure of the epiglottitis, but also of the aryepiglottics and vocal folds. However, penetration and aspiration may occur prior to swallowing because of poor bolus formation, during swallowing due to improper laryngeal closure, or after swallowing when residual pools around the valleculae and pyriform sinuses as a result of poor pharyngeal clearance.Preterm infants often have respiratory issues, ranging from severe respiratory distress syndrome, including bronchopulmonary dysplasia, to varying degrees of oxygen supplementation early in life. However, as they mature, “healthy” preterm infants normally experience decreasing episodes of oxygen desaturation or apnea during oral feeding. Baseline respiratory rates in these infants range from 40 to 60 breaths/min or 1.0 to 1.5 seconds per respiratory cycle (inspiration/expiration). The duration of the swallowing event when airflow is interrupted can range between 0.35 and 0.7 seconds, as measured by intrapharyngeal pressure in term and preterm infants. With such respiratory rates and swallow duration, little time may be left for respiration. In addition, during oral feeding, a number of studies noted decreased minute ventilation, prolonged expiration, and shortened inspiration, albeit no significant change in tidal volume. Thus, with all the respiratory alterations taking place, it is not surprising that some “healthy” preterm infants incur episodes of desaturation, apnea, or bradycardia when feeding by mouth.A number of studies have examined the coordination of sucking, swallowing, and breathing. However, they have been primarily descriptive, focusing on the structural elements (eg, tongue and perioral muscles) implicated in the generation of sucking and swallowing. No study, to our knowledge, has yet defined the precise temporal relationship(s) for the proper coordination of suck, swallow, and respiration. Clinically, coordination is presumed attained when infants take their feedings by mouth with no overt signs of aspiration, oxygen desaturation, apnea, or bradycardia and demonstrate a ratio of 1:1:1 or 2:2:1 suck:swallow:breathe.Based on the close link between sucking and swallowing and between swallowing and respiration, we investigated a different aspect of coordination by examining the relationships between suck-swallow and swallow-respiration. This study was conducted during bottle feeding in preterm infants who had no major medical issues and were born at less than 30 weeks’ gestation (26.8±2.7 weeks) and term infants (39.1±1.1 weeks’ gestation) between 1 and 3 weeks after birth. We followed the gradual maturation of various oral feeding skills from 34 to 42 weeks postmenstrual age (Table 1). During this time period, there occurred a steady increase in bolus size, sucking and swallowing rate, strength of the suction component, and rate of milk transfer. It is of interest that suck-swallow coordination already was attained when infants were introduced to oral feeding at approximately 34 weeks postmenstrual age.We further explored the timing of swallows in relation to the phases of the respiratory cycle. The schematic in Figure 5 depicts the various phases of the respiratory cycle when swallows may occur (swallow-respiratory interfacings). We computed the frequency of occurrence of these potential interfacings when preterm infants were taking one to two and six to eight oral feedings per day and when term infants were at 1 week and 2 to 3 weeks after birth. Table 2 shows how, with maturation, a gradual trend occurs toward swallowing at a safer phase of respiration (ie, start of inspiration or end of expiration when airflow is minimal or stopped). Considering the risks involved if swallows were to occur at other phases of the respiratory cycle (eg, inspiration or deglutition apnea), we propose that coordination of suck-swallow-respiration is attained with a consistent suck-swallow ratio (eg, 1:1, 2:1) and a safe swallow-respiration interfacing (eg, start of inspiration or start of expiration).In a recent study investigating whether an optimal bottle nipple can be identified to enhance oral feeding performance, we observed that “healthy” preterm infants (<30 weeks’ gestation) can alter their sucking pattern depending on the flow rate of the bottle nipples used. From these results, we speculated that infants can modify their sucking skills to maintain a rate of milk transfer that they can handle safely or that is compatible with the level of suck-swallow-breathe coordination they have attained. Because the studied infants were “healthy” and medically stable, we hypothesize that their ability to modify specific sucking variables to maintain appropriate flow rates requires the integrity of afferent sensory feedback to the sucking, swallowing, and respiratory centers in the brainstem. This concept is supported by the work of Finan and Barlow, who speculated that the sucking motor pattern of infants is under the control of a central pattern generator that adapts to changing environmental conditions via afferent sensory feedback. Craig and Lee have proposed a similar theory, with the existence of an intrinsic tau-guide acting as a common process linking various forms of timing events within the motor function (eg, the control of sucking pressure in infants). Under such conditions, oral feeding can be successful if the infants are allowed to regulate their own milk flow.Oral feeding safety and success is not solely dependent on an infant’s oral feeding skills. Indeed, it may be disrupted by fluid penetration or aspiration into the larynx caused by unrelated factors, such as gastrointestinal immaturity or the infant’s behavioral state and organization at the time of the feedings. Fluid backflow into the larynx may occur not only as a result of pooling at the level of the valleculae and pyriform sinuses resulting from poor pharyngeal clearance, but also as a result of improper timing of the relaxation or constriction of the upper or lower esophageal sphincters, gastroesophageal reflux, improper esophageal motility, or delayed gastric emptying. Because these different aspects of gastrointestinal immaturity are addressed in other articles in this issue, only the infant’s behavioral state and organization and environmental factors are discussed here.The National Neonatal Individualized Developmental Care and Assessment Program (NIDCAP) distinguishes six levels of state organization: quiet sleep, active sleep, drowsiness, quiet alert, active alert, and crying/fussing. Based on this classification, the optimal states for oral feeding appear to be drowsiness and quiet and active alert. However, due to their immaturity, one of the characteristics of preterm infants is their inability to regulate their states. Because they spend most of their time transitioning from one state to another, some infants cannot feed by mouth for extended periods of time. Neurobehavioral studies have proposed that an organized pattern of state regulation is a good indicator of an infant’s ability to cope with his or her environment, particularly that of a neonatal intensive care unit. As a result, improvement in state regulation is interpreted as evidence of central nervous system maturation and can be key to an infant’s ability to orally feed successfully for an extended period of time.Behavioral organization is identified by the calmness, relaxation, and regular breathing that an infant exhibits at a particular time. Not necessarily linked to a particular behavioral state, this condition needs to be taken into account when feeding infants. Feeding therapists commonly identify infants as organized when their bodies are gently flexed, their arms are folded toward midline, hands are placed under the chin, and there are no facial adverse expressions (eg, grimace). It is under these circumstances that oral feeding is best achieved, likely because infants, in the absence of any additional stimulation, can focus on the task at hand. Thus, behavioral state and organization can affect significantly the ability of a preterm infant to feed by mouth.Given the type of stimulations that preterm infants experience in nurseries, debate has arisen regarding the appropriate stimulations that such infants should receive. Caretakers have long recognized different types of sensory stimulations. Some provided during routine nursing care are aversive (eg, venipuncture, heelsticks, suctioning) and “impersonal” (eg, regular diaper changes, temperature monitoring, positioning). Others, often provided by parents, are calming and soothing (eg, holding, stroking, or rubbing). To address these issues, two different approaches are taken. The first reduces or limits the aversive stimulations occurring in nurseries and the second provides additional stimulations that benefit the infants.The establishment of developmental care programs primarily led by nursing staff focuses on the first approach. For example, effort is made to cluster care or to carry out painful procedures within a specific time of the day rather than throughout the day. The light in nurseries is dimmed, isolette covers are used to reduce background light and noise, and containment methods (eg, swaddling) are used as calming interventions. The second approach focuses on providing supplemental stimulations to counterbalance the negative effect(s) of the immediate adversities and the developmental deficits associated with prematurity, such as skin-to-skin holding.Currently, when infants are deemed ready to feed orally, an important sign of success is completion of their feedings (ie, the ability to take the volume prescribed within an allotted period of time), which is key to advancement of oral feeding and earlier attainment of independent oral feeding. Under such pressure, caretakers often inappropriately “encourage” infants to finish their feeding no matter what the consequences (eg, choking, fatigue, emesis). We speculate that the high occurrence of oral feeding aversion observed in nurseries and following discharge may result from such technique. Consequently, we have proposed a developmental oral feeding approach that, most importantly, takes into consideration the limitations of these infants. In brief, infants are advanced in their daily oral feedings only if they do not demonstrate any adverse events or aversive behavior. Completion of feedings is not a requirement for advancement. Under such a program, caretakers must become cognizant of all aspects of their patient’s limitations and “adjust” their feeding approach accordingly. Table 3 shows some of the facts and recommendations we propose.From our studies on nutritive sucking, we observed that the mature sucking pattern is not necessary for safe and successful oral feeding. Readiness to feed orally should not be based only on sucking skills, but rather on the coordination of sucking, swallowing, and respiration. Because the various skills implicated in oral feeding mature at different times, it is speculated that neurologically intact “healthy” infants can coordinate these three functions if allowed to regulate and control their own feeding. In addition, oral feeding performance does not depend only on oral motor skills but also on the infant behavioral state and organization and his or her environment. It is essential, therefore, that all these factors be taken into consideration when weaning an infant from tube feeding.