Feeding pattern in preterm neonates admitted in neonatology unit: a descriptive cross-sectional study

Background: The components of breast feeding including sucking, swallowing, and breathing, develop at different gestational ages, and are incomplete in preterm. Other modes of feeding like nasogastric/orogastric and cup are used till matured breast feeding skills appear. Material and methods: This descriptive cross-sectional study was done over a period of 1 year with objectives to observe the different feeding patterns in the preterm neonates, assess the time required to start feeding from the day of admission, and observe the time taken for transition from one feeding pattern to another feeding pattern. A total of 116 admitted preterm neonates born less than 34 weeks were included. The ethical clearance and informed consent were obtained. The time of admission, the duration between the time of admission and start of feeding, type of feeding, mode of feeding, and progression of feeding from one mode to another was noted, and appropriate statistical analysis done. Results: Cup feeding (57.8%) was the most common mode of starting of feeding followed by orogastric feeding (42.2%). Mothers breast milk was the most common feed. The median duration of starting of feed from the day of admission was 2 days. The time gap between day of admission and starting of orogastric feed and cup feeding were 2 days and 5 days, respectively. The median time taken for transition from orogastric feed to cup feed and from cup feed to mothers breast feed was 5 days and 6 days, respectively. Babies with a higher age at starting of feeding took significantly lesser time to transit from cup feeding to mother’s breast feeding (P=0.01). Babies with Respiratory Distress Syndrome had a significantly higher transit time from orogastric feeding to cup feeding (P<0.001). Conclusion: The rate of advancement of feed depends on maturation of feeding skills, gestational age, birth weight, presence of comorbidities, and nursing care.


Introduction
Proper nutrition is important for the growth and development of a neonate [1] .The WHO recommends exclusive breast feeding for the first 6 months of life [2] .There should be a proper coordination between sucking, swallowing, and breathing for effective breast feeding.These components of breast feeding are incomplete in preterm [3] .Sucking bursts develop by 28-31 weeks but there is lack of coordination between suck, swallow, and breathing cycle.Between 32 and 34 weeks, sucking matures to some extent and the coordination between breathing and swallowing begins.Sucking pattern further matures after 34 weeks with more coordination between breathing and swallowing [1,3,4] .Also, several comorbidities that arise out of prematurity hamper the feeding initiation and progression [5] .Controversies exist regarding the initial mode of feeding and rate of advancement of feeding.There have been some novel studies comparing the rate of advancement of feeding in preterm neonates [6] .Therefore, establishment of oral feeding is always challenging in preterms [7] .

HIGHLIGHTS
• Orogastric feeding and cup feeding are the initial modes of feeding in a preterm infant.• Orogastric feed can be started soon after the preterm baby gets admitted to the nursery.• It takes around a week to transition from orogastric feeding to cup feeding and cup feeding to breast feeding.This transition period is lesser in preterms born with higher gestational age.• Respiratory distress syndrome significantly impairs advancement of feeding.
Various modes of feeding like nasogastric/orogastric, cup, paladai, etc. are used to feed the preterm baby [8] .This study was done with an objective to observe the different feeding patterns in the preterm neonates, assess the time required to start feeding from the day of admission, and the time taken from transition from one feeding pattern to another feeding pattern.The study also points about the problems encountered during feeding.This will in turn help in formulating a feeding timeline protocol, so that babies could be provided with adequate and timely nutrition.

Methods
This was a descriptive cross-sectional study done over a period of The time of admission, and the duration between the time of admission and start of feeding in the neonate was noted.Initial feeding (any one of: expressed breast milk or commercially available formula feed) was given to the baby via either orogastric feeding (defined as enteral feeding done via orogastric tube), cup feeding (defined as enteral feeding done via cup) or breast feeding (defined as feeding done directly through mother's breast), depending on the baby's feeding skills, gestational age, and comorbidities.Initial feed also included the trophic feed that was started in the newborn for gut priming.The type of feeding (formula vs expressed breast milk), amount (ml/kg/feed), mode of feed (orogastric or cup feeding or directly breast feeding), and progression of feeding from one mode to another (e.g.: from orogastric feed to cup feed/ cup feed to breast feed) was noted.The time period needed for transition from one mode of feeding to another, as mentioned earlier, was noted.The presence of comorbidities in the babies, feeding intolerance (defined as inability to tolerate enteral feedings characterized by vomiting (altered milk/bile or blood-stained), systemic features (lethargy, apnea, cyanosis, bradycardia, abdominal distension (with or without visible bowel loops), increased gastric residuals: > 2 ml/ kg or any change from previous pattern, abdominal tenderness, reduced, or absent bowel sounds), was also taken into account.The baby was weighed daily and recorded.
The instructions for feeding was solely given by the neonatologist and the treating team, as done routinely on a standard basis, and no deviation was made from the existing feeding protocol.The initial volume of feeding was started from 10 to 20 ml/kg/day and then gradually increased in volume of 20 ml/kg/ day, till 180 ml/kg/day.The usual frequency of feeding was every two hourly.The feeding volume and frequency were altered depending on the tolerance of the feed by the neonate.No prokinetic drugs were given to any neonates to accelerate the bowel motility.All the babies enrolled, were followed till the time of discharge from the hospital.At discharge, the weight of the baby and mode of feeding was noted.Data was analyzed using Statistical Package for Social Sciences (SPSS 11.5) for statistical analysis, and appropriate statistical tests used as needed.The work has been reported in line with the strengthening the reporting of cohort, cross-sectional, and case-control studies in surgery (STROCSS) criteria [9] .This research work was registered in the research registry and the registry Unique ID number is 9323.The registration details can be reached at the following link https://www.researchregistry.com/browse-the-registry#home/registrationdetails/64c24d3cf693a2002735c538/.

Results
Out of 116 mothers who gave birth to the preterm neonates, more than half of them were primipara [n = 63 (54.3%)], and about 97 (84%) had a singleton pregnancy.The mean age of the mothers was 24.8 3.5 years and mean period of gestation was about 31.5 1.6 weeks.There were 81 (69.8%) male babies and 35 (30.2%) female babies.The baseline characteristics of the neonates are shown in Table 1.The most common method used for initiation of feeding was cup feeding (n = 67, 57.8%,) followed by orogastric feeding (n = 49, 42.2%).None of the babies were started directly on breast feeding.Mother's breast milk was the most common type of first feeding used (n = 88, 76%) followed by formula feeding (n = 28, 24%).Formula feeding was used in those babies whose mother's had decreased milk secretion, or had taken any form of drugs like antiepileptics, or were comatose.
All the babies were admitted on first day of life.The feeding related timeline of the neonates is given in Table 1.
The various factors, which could influence the choice of initial mode of feeding, be it orogastric or cup feed, were studied.Orogastric feeding, as an initial mode of feeding, was significantly higher in those babies who were born on or before 30 weeks period of gestation and those who had respiratory distress syndrome (RDS) (P < 0.001).Similarly, cup feeding as an initial mode of feeding was significantly higher in those babies who had a birth weight above 1.5 kg and those with outborn delivery (P < 0.001, P = 0.001, respectively).Other factors like age at starting of first feed, sex, appropriate for gestational age/small for gestational age status and use of supplemental oxygen did not significantly determine the initial mode of feeding.The details are given in Table 2.There were different comorbidities seen in the babies during the hospital stay.Early Onset Neonatal Sepsis (EONS) was the most common comorbidity seen followed by Neonatal Hyperbilirubinemia (NNH).The frequency distribution of the different comorbidities seen are given in Table 3.
The effect of various factors on the median time taken for transition from orogastric feed to cup feed was studied.It was found that those babies who were more than 48 h old, when first feed was started, took significantly lesser time for transition from orogastric feed to cup feed (P = 0.01).Similarly, babies who developed respiratory distress syndrome took significantly longer time for transition from orogastricfeed to cup feed (P < 0.001).However, various factors like sex, gestational age, AGA/SGA status, place of delivery, use of supplemental oxygen, and mechanical ventilation had a comparable transition time from orogastric feeding to cup feeding.Also, presence of various comorbidities like Jaundice, Hypoxic Ischemic Encephalopathy (HIE), Late Onset Neonatal Sepsis (LONS), Early Onset Neonatal Sepsis (EONS), shock, polycythemia, hypoglycemia, and hypocalcemia had no significant difference in the time taken for transition from orogastric feeding to cup feeding.The details are given in Table 4.
The effect of various factors like age, sex, supplemental oxygen, use and presence of comorbidities on the time taken for transition from cup feeding to mother's breast feeding was analyzed.It was observed that different factors like age at starting of feeding, sex, gestational age, AGA/SGA status, birth weight, use of supplemental oxygen, and mechanical ventilation (MV) and presence of comorbidities like RDS, Jaundice, HIE, LONS, EONS, shock, polycythemia, hypoglycemia, and hypocalcemia had no significant difference in the time taken for transition from cup feeding to mother's breast feeding (Table 5).
Feeding intolerance was seen in 36 babies (31%).Feeding intolerance was significantly higher in those babies who had orogastric feed as initial mode of feed, and who were born with gestational age less than or equal to 30 weeks.None of the babies developed necrotizing enterocolitis (NEC).Similarly babies with birth weight more than 1.5 kg had significantly lesser incidence of feeding tolerance.Factors like age, sex, and AGA/SGA status had a comparable rate of feeding intolerance.

Discussion
This descriptive cross-sectional study was conducted among 116 preterm babies born <34 weeks of gestation over a period of 1 year.There was a male preponderace in this study, a finding  [8,10,11] .Also, the mean birth weight (1.5 0.3 kg) and mean gestational age (31.5 1.6 weeks) of the patients in this study were similar to studies done by Kwok TC et al. and Paudel L et al. [8,10] .The most common comorbidity seen in this study were Early Onset Neonatal Sepsis followed by Neonatal hyperbilirubinemia.Contrasting to this, were the findings of Kwok TC et al. [8] , who observed Late Onset Neonatal Sepsis as the main comorbidity.Also, Shrestha and Shrestha [12] , in their study assessing mortality and morbidity of preterm neonates, observed a lower incidence of hyperbilirubinemia, sepsis, and hypoglycemia as compared to this study.The higher rate of sepsis, hyperbilirubinemia, and hypoglycemia in our set up could be somewhat attributed to hygiene in the delivery room, gestational age, and place of delivery (outborn babies having higher rate of these comorbidities).
In this study, the most common initial mode of starting of feeding was cup feeding, a finding similar to that of Jadcherla SR et al., Flint A et al., and Park J et al. [13][14][15] .We also observed that, among newborns born with gestational age less than 30 weeks, the initial mode of feeding was through orogastric route, while as age advanced, this was changed to cup feeding.Similar was the finding made by Lang S et al. and Dowling DA et al., who observed that cup feeding was the initial mode of starting of feed on babies born with higher postconceptional age (32-36 weeks) as compared to those with lower postconceptional age [16,17] .This finding is obvious as complexity of mode of feeding changes with gestational age.The more mature the baby, advanced is the mode of feeding.
Orogastric feed was the initial mode of feeding in babies who were born less than 30 weeks gestational age and those who had respiratory distress syndrome.The presence of respiratory distress syndrome and immaturity impairs the cup feeding skills.None of the babies were started directly on mothers breast feed because of the obvious reason of immaturity of sucking and swallowing skills in the study population.Mothers milk was the choice of feed in 76% babies in our study.The remaining were kept on formula feed due to inadequacy or unavailability of mothers milk.Contrary to this finding, Kwok TC et al. observed that only 53% babies were fed with mothers milk, while 43% received formula milk.As per the authors, the reason behind the use of formula milk in nearly half of the babies was unavailability of mothers milk in the initial few days [8] .
In this study, the median time between day of admission and starting of orogastric feed and cup feed was 2 days and 5 days, respectively.Similarly, the time taken for transition from orogastric feed to cup feeding, and cup feeding to breast feeding was 5 days and 6 days, respectively.These transition times were lesser in our study than those of the study done by Simpson S et al., Gianni ML et al., and Nyec [18][19][20] .Lesser transition time suggests early advancement of mode of feeding.This difference in the findings between our study and other studies could be attributed to a combination of factors like race, presence of comorbidities, and quality of medical, and nursing care.This transition time was significantly higher in presence of respiratory distress syndrome, and lesser in those neonates with a greater age at the time of starting of first feeding.In the same context, Kwok TC et al. [8] observed that a higher gestational age was associated with a lesser transition time.Similar to our study, Amoris and Nascimento [11] observed that various factors like use of mechanical ventilation, continuous positive airway pressure, and increased hospitalization duration had a longer transition time from one mode of feed to the next.We observed that the median duration between day of admission and starting of direct mothers breast feed was 10 days in our study.This finding was similar to that of Pickler et al. [21] in which the median duration to breast feeding from the day of starting of feeding was 14.5 days.The reason for the longer time taken for starting mothers breast feed is due to the time taken for maturity of feeding skills.However, Kwok et al. [8] had a time gap of 8 days between day of admission and starting of breast feeding.The reason for this early initiation of mothers breast feed could be lesser comorbidities and higher mean gestational age of neonates in their study, as compared to ours.Likewise, Amoris and Nascimento [11] observed a shorter time (4 days) for achievement of full feed.This might be due to inclusion of older babies in their study as compared to our study.The rate of feeding intolerance in this study was 31%, similar to the findings of Quing Chen et al. [22] However, Huang et al. [23] observed a higher rate of feeding intolerance in their study.In our study, factors like orogastric feed as the initial mode of feeding and lower gestational age had significantly higher rate of feeding intolerance.Similar to our study Quing Chen et al. also observed a higher rate of feeding intolerance among neonates with lesser gestational age [22] .Huang et al. [23] in their study observed that a lower birth weight (< 1 kg), use of formula feed and caffeine citrate were associated with feeding intolerance.
The median hospital stay in this study was nearly 2 weeks (13.5 days).Since the mean gestational age is nearly 32 weeks, so a mean duration after 2 weeks of this gestational age could have enabled the babies to feed on full mothers breast milk.However, compared to this study, Kwok TC et al. and Pickler RH et al. observed a higher duration of hospital stay (26 days and 21 days) and Amoris EVN et al. had a lower duration of hospital stay [8,11,21] .The shorter duration of hospital stay in the study done by Amoris and Nascimento [11] might be the inclusion of babies above 1.5 kg in their study.
There were certain limitations in the study.The presence of comorbidities with a potential to cause cardiorespiratory disturbances (like RDS and shock) could have hampered the innate feeding ability and progression of feeding of the babies.Also, there was no follow-up for the babies after hospital discharge.However, similar studies with apparently healthy preterm neonates (without underlying serious comorbidities) and with longer follow-up time could be done in the future.

Conclusion
The present study highlighted the various timeline of feeding methods in preterm neonates.It demonstrated the median duration between each feeding modes.This study also showcased the effect of various factors like age at starting of feed, sex, birth weight, and presence of comorbidities on these feeding timelines.Hence, it can be concluded that feeding in preterms is a challenging task which depends on various factors.The rate of advancement of feed depends on maturation of feeding skills, gestational age, birth weight, presence of comorbidities, and nursing care.
1 year (March 2020 to February 2021) among 116 neonates at B.P. Koirala Institute of Health Sciences (BPKIHS), a tertiary care referral centre of Eastern Nepal.Babies born less than 34 weeks gestational age, either inborn or outborn, and who were admitted to Neonatology Unit (neonatal ICU, nursery, and neonatal ward) of BPKIHS were included in the study, while babies with major congenital malformation of the orofacial region (e. g.: cleft lip/ cleft palate), down syndrome and genetic disorders, which directly hamper the feeding, and babies whose parents refused to give consent, were excluded.Since babies born above 34 weeks might theoretically demonstrate a more mature sucking and swallowing coordination as compared to the younger ones, hence these babies were not included.The ethical clearance was obtained from the Institutional Review Committee, BPKIHS (Ref.no.303/076/077 IRC).Written informed consent in local language was obtained from the parents.Nonprobability purposive sampling technique was used.The sociodemographic, anthropometry, and clinical parameters (including focused maternal and birth history, and physical examination findings) of the babies were recorded in a predesigned proforma.The weight of the baby was classified as extremely low birth weight (ELBW; birth weight less than 1 kg), very low birth weight (VLBW; birth weight between 1 and 1.5 kg), and low birth weight (LBW; birth weight above 1.5 kg and less than 2.5 kg).

Table 1
Baseline characteristics of the neonates.

Table 2
Feeding timeline of the neonates.

Table 3
Comparison of various clinical characteristics with initial mode of starting of feed. to studies done by Kwok TC et al., Paudel L et al., and Amoris EVN et al., respectively *Chi-square test.**Fisherexact test.AGA, appropriate for gestational age; SGA, small for gestational age.similar

Table 4
Comorbidities seen in the neonates.

Table 5
Comparison of various clinical characteristics with time taken for transition from orogastric feed to cup feed.