THE RELATIONSHIP BETWEEN PROLONGED MECHANICAL VENTILATION AND DELAYED FEEDING MILESTONES IN NEWBORNS WITH HYPOXIC-ISCHEMIC ENCEPHALOPATHY

Summary

Methods: This retrospective cohort study included 102 newborns with HIE who received therapeutic hypothermia in a neonatal intensive care unit. Outcomes included postnatal day of first oral feeding (POfirst), achievement of oral feeding for at least half of daily feeds (POhalf), achievement of full oral feeding (POFull), transition to breastfeeding, length of hospital stay, and Early Feeding Skills (EFS) scores at POfirst and discharge. Associations with ventilation duration were examined, infants were compared by ventilation duration (<7 vs. ≥7 days), and multiple linear regression analyses were performed after adjustment for HIE stage, gestational age, birth weight, convulsion status, and MRI findings.

Results: Longer mechanical ventilation duration was associated with delayed feeding milestones, later transition to breastfeeding, prolonged hospitalization, and lower EFS scores at POfirst and discharge. In adjusted regression models, mechanical ventilation duration remained significantly associated with later POfirst, POhalf, POFull, transition to breastfeeding, longer hospital stay, and lower EFS scores after adjustment for HIE stage, gestational age, birth weight, convulsion status, and MRI findings. Infants ventilated for ≥7 days showed significantly later feeding milestones and lower EFS scores than those ventilated for <7 days, while weight outcomes did not differ between groups.

Conclusions: In newborns with HIE treated with therapeutic hypothermia, prolonged mechanical ventilation was associated with delayed oral feeding and breastfeeding transitions and lower early feeding skill scores. These findings should not be interpreted as evidence of a causal effect of ventilation alone, because ventilation duration may also reflect underlying encephalopathy severity and overall illness burden. Nevertheless, ventilation duration may serve as a practical clinical marker for identifying infants who require closer feeding surveillance and early multidisciplinary support.

Introduction

In infants with HIE, feeding involves more than nutrition; it requires coordinated sucking, swallowing, breathing, and stable physiological and behavioral regulation.[3,4] Hypoxic ischemic injury may interfere with these processes, resulting in delayed readiness for oral feeding, reduced endurance, poor suck–swallow–breath coordination, and impaired oral motor control, even when tube feeding is tolerated.[3-5] Beyond feeding performance, these difficulties also affect early bonding, regulation, and neurodevelopment.

Successful breastfeeding depends on intact sensory processing, oral motor coordination, and neurobehavioral stability. Among infants with HIE receiving therapeutic hypothermia, signs of oropharyngeal dysphagia highlight early vulnerability in feeding skills.[5] In some infants, these difficulties persist beyond the neonatal period and may influence long-term growth and health.[6]

Many infants with HIE require respiratory support during NICU care because of the complex interaction between perinatal asphyxia, respiratory failure, impaired respiratory drive, therapeutic hypothermia-related physiological changes, and mechanical ventilation management.[7] Although life-saving, prolonged ventilation may limit early oral experience and delay development of coordinated sucking and swallowing.[8,9]

Across neonatal and pediatric populations, prolonged mechanical ventilation has been associated with feeding and swallowing dysfunction, possibly related to reduced use of oral motor structures, altered aerodigestive coordination, and decreased sensory input.[8-11] In preterm infants, longer ventilation is linked to delayed oral feeding and longer hospital stays.[12] However, these findings may not directly apply to HIE, where feeding difficulties primarily reflect neurological injury. Evidence examining ventilation duration and feeding outcomes in HIE remains limited.

Only a limited number of studies have examined feeding transitions in infants with HIE. Gunes et al.[13] reported delayed enteral feeding with longer ventilation, while Asgarshirazi et al.[14] described swallowing impairment after ventilation lasting seven days or longer. However, evidence on oral feeding milestones, breastfeeding, and early feeding skills remains scarce.

Early feeding skills support not only nutrition but also neonatal communication, self-regulation, neurobehavioral development, and caregiver infant bonding.[3-5] Clarifying factors associated with delayed feeding in HIE is therefore essential for early identification and timely multidisciplinary intervention. Accordingly, this study examined the relationship between ventilation duration, feeding milestones, breastfeeding, and early feeding skills in infants with HIE.

Methods

The study protocol was approved by the İstanbul Medipol University Non-Interventional Clinical Research Ethics Committee (approval no: 447, date: 17.04.2025). Owing to the retrospective nature of the study, families of eligible infants were contacted using information available in medical records, and written informed consent was obtained from parents or legal guardians prior to data inclusion. All data were anonymized before analysis and managed in accordance with the principles of the Declaration of Helsinki.

Participants and Diagnosis of Hypoxic–Ischemic Encephalopathy
Infants were eligible for inclusion if they were diagnosed with HIE by a consultant neonatologist, received therapeutic hypothermia, and had complete data on mechanical ventilation duration, feeding milestones, breastfeeding transition, and EFS assessments. Because the primary outcomes focused on the timing of oral feeding and breastfeeding transitions, infants without complete feeding milestone data were excluded. This inclusion strategy was acknowledged as a potential source of selection bias.

The diagnosis of HIE was established according to institutional protocols and Turkish Neonatal Society guidelines.[15] Eligibility for therapeutic hypothermia was determined using national criteria, including admission within the first 6 hours of life, evidence of perinatal hypoxic ischemic insult (such as metabolic acidosis, low Apgar scores at 10 minutes, or ongoing need for resuscitation), and clinical signs of moderate to severe encephalopathy.[4,15]

In selected cases, infants with lower gestational age who met clinical criteria for HIE also received therapeutic hypothermia at the discretion of the attending neonatologist. Neurological severity was evaluated daily during the first 72 hours of life using the Sarnat and Sarnat staging system, and the most severe stage observed during this period was recorded for analysis.[16]

Clinical Care and Feeding Management
All infants received standardized neonatal intensive care unit management, including therapeutic hypothermia and respiratory support. During hypothermia treatment, enteral feeding was provided via orogastric tube, and oral feeding was withheld in accordance with standard clinical practice.[5]

Both invasive and non-invasive mechanical ventilation were administered as clinically indicated. In infants with HIE, respiratory support was most commonly required because of neurological instability, impaired respiratory drive, seizures, or the need for sedation during hypothermia, rather than primary pulmonary disease.[3,8,17]

Initiation of oral feeding was determined by the NICU team based on cardiorespiratory stability, regular spontaneous breathing, and absence of clinically significant apnea or oxygen desaturation.[3] Feeding readiness was additionally evaluated by a speech-language therapist through assessment of oral motor function, non-nutritive sucking, coordination, and behavioral regulation.[3]

Syringe feeding trials were initiated and advanced per infant tolerance and clinical status. Breastfeeding was supported in infants with adequate oral skills and no aspiration signs.[5]

Data Collection
Demographic and clinical data extracted from medical records included gestational age, sex, birth weight, maternal age, mode of delivery, Apgar scores at 1 and 5 minutes, HIE stage, and duration of mechanical ventilation.

Feeding-related outcomes included the postnatal day and weight at first oral feeding (POfirst), achievement of oral feeding for at least half of daily feeds (POhalf), achievement of full oral feeding (POFull), transition to breastfeeding, length of hospital stay, and corresponding weight outcomes. POhalf was defined as the postnatal day on which the infant received at least half of daily feeds orally, corresponding to at least 4 oral feeds in infants fed 8 times per day or at least 6 oral feeds in infants fed 12 times per day.[12,13]

Early Feeding Skills were assessed using the Early Feeding Skills (EFS) Assessment, which evaluates respiratory regulation, oral-motor function, swallowing coordination, engagement, and physiological stability during feeding.[19] EFS assessments were performed by a speech-language therapist experienced in neonatal feeding assessment at POfirst and again before discharge. The validated Turkish version of the EFS Assessment was used. The assessor was not blinded to the clinical course because of the retrospective clinical nature of the study, and this was considered a methodological limitation. Infant weight was measured daily with calibrated digital scales before morning feeding.

Statistical Analysis
Statistical analyses were performed using SPSS version 26.0. Categorical variables were summarized as frequencies and percentages, whereas continuous variables were presented as means ± standard deviations or medians with minimum–maximum values, as appropriate. Distribution normality was evaluated by visual inspection of histograms and the Kolmogorov–Smirnov test. As most variables were not normally distributed, non-parametric statistical methods were applied.

Spearman's rank correlation was used to test associations between ventilation duration and feeding outcomes, including POfirst, POhalf, POFull, transition to breastfeeding, length of hospital stay, weight outcomes, and EFS scores. For subgroup analysis, infants were grouped according to total mechanical ventilation duration as <7 days and ≥7 days, and the Mann–Whitney U test was used for comparisons.

Multiple linear regression analyses were performed to examine whether mechanical ventilation duration remained associated with time-based feeding outcomes and EFS scores after adjustment for HIE stage, gestational age, birth weight, convulsion status, and MRI findings. Separate models were constructed for each outcome. Convulsion status was coded as absent or present, and MRI findings were coded as normal or abnormal. Both variables were entered into the regression models as binary variables. Standardized beta coefficients (β) and p-values were reported. Multicollinearity was assessed using variance inflation factor values. Statistical significance was defined as p < 0.05. A sensitivity analysis was performed after excluding infants born before 36 weeks' gestation to evaluate whether the findings were robust in the gestational-age group most commonly represented in therapeutic hypothermia protocols.

Results

Table 1: Demographic, Clinical, and Nutritional Characteristics of Participants (n = 102).

Associations Between Ventilation Duration and Feeding Outcomes
Total mechanical ventilation duration showed strong positive correlations with the timing of feeding milestones and length of hospital stay (Table 2). Longer total mechanical ventilation duration was associated with later POfirst (ρ = 0.747), POhalf (ρ = 0.779), POFull (ρ = 0.780), transition to breastfeeding (ρ = 0.747), and longer hospital stay (ρ = 0.721) (all p < 0.001). Total mechanical ventilation duration was also negatively correlated with EFS scores at POfirst (ρ = −0.584) and at discharge (ρ = −0.477) (both p < 0.001). Invasive and non-invasive mechanical ventilation durations demonstrated similar associations with delayed feeding milestones and lower EFS scores.

Table 2: Spearman Correlations Between Mechanical Ventilation Duration and Feeding Outcomes.

In contrast, mechanical ventilation duration was not significantly correlated with weight outcomes at any feeding milestone or at discharge (Table 3, all p > 0.05).

Table 3: Spearman Correlations Between Mechanical Ventilation Duration and Weight Outcomes.

Subgroup Comparison by Ventilation Duration
Infants were stratified according to total mechanical ventilation duration as <7 days (n = 57) and ≥7 days (n = 45) (Table 4). Infants ventilated for ≥7 days had significantly later POfirst, POhalf, POFull, transition to breastfeeding, and longer hospital stay compared with those ventilated for <7 days (all p < 0.001). In contrast, weight at feeding milestones and discharge did not differ significantly between groups (all p > 0.05). EFS scores at POfirst and discharge were significantly lower in infants ventilated for ≥7 days than in those ventilated for <7 days (both p < 0.001).

Table 4: Comparison of Feeding, Weight, and Early Feeding Skill Outcomes According to Mechanical Ventilation Duration.

Multiple linear regression analyses were performed to examine whether mechanical ventilation duration remained associated with time-based feeding outcomes after adjustment for HIE stage, gestational age, birth weight, convulsion status, and MRI findings (Table 5). Mechanical ventilation duration remained significantly associated with later POfirst (β = 0.701, p < 0.001), POhalf (β = 0.733, p < 0.001), POFull (β = 0.685, p < 0.001), transition to breastfeeding (β = 0.653, p < 0.001), and longer hospital stay (β = 0.657, p < 0.001).

Table 5: Adjusted Linear Regression Models for Time-Based Feeding Outcomes.

HIE stage was significantly associated with later POFull (β = 0.187, p = 0.015), transition to breastfeeding (β = 0.256, p = 0.001), and longer hospital stay (β = 0.254, p < 0.001). Abnormal MRI findings were significantly associated with later POfirst (β = 0.156, p = 0.008), POhalf (β = 0.132, p = 0.029), and POFull (β = 0.132, p = 0.027). Convulsion status was not significantly associated with time-based feeding outcomes in the adjusted models. No evidence of problematic multicollinearity was observed, with all variance inflation factor values below 2.

Multiple linear regression analyses were also performed for EFS scores at POfirst and discharge (Table 6). Mechanical ventilation duration remained significantly associated with lower EFS scores at POfirst (β = −0.500, p < 0.001) and at discharge (β = −0.480, p < 0.001) after adjustment for HIE stage, gestational age, birth weight, convulsion status, and MRI findings. Abnormal MRI findings were also significantly associated with lower EFS scores at POfirst (β = −0.209, p = 0.013) and at discharge (β = −0.213, p = 0.016). Birth weight was positively associated with EFS at discharge (β = 0.228, p = 0.030). Convulsion status was not significantly associated with EFS scores in the adjusted models. No evidence of problematic multicollinearity was observed, with all variance inflation factor values below 2.

Table 6: Adjusted Linear Regression Models for Early Feeding Skills.

Sensitivity Analysis
A sensitivity analysis was performed after excluding infants born before 36 weeks' gestation (n = 91). In this analysis, total mechanical ventilation duration remained significantly correlated with later POfirst (ρ = 0.773), POhalf (ρ = 0.800), POFull (ρ = 0.800), transition to breastfeeding (ρ = 0.761), and longer hospital stay (ρ = 0.750), and remained negatively correlated with EFS scores at POfirst (ρ = −0.582) and discharge (ρ = −0.463) (all p < 0.001). In adjusted regression models restricted to infants born at ≥36 weeks' gestation, mechanical ventilation duration also remained significantly associated with all time-based feeding outcomes and EFS scores, with standardized β values ranging from 0.599 to 0.699 for time-based outcomes and from −0.455 to −0.456 for EFS outcomes (all p < 0.001) (Table 7).

Table 7: Sensitivity Analysis Restricted to Infants Born at ≥36 Weeks' Gestation (n = 91).

Discussion

In infants with HIE, feeding represents more than nutritional intake and instead reflects a complex neurodevelopmental process that depends on coordinated sucking, swallowing, and breathing, along with physiological stability and behavioral state regulation.[3,4] Hypoxic ischemic injury may disrupt these interrelated systems, resulting in oral feeding difficulties even in infants who tolerate enteral tube feeding.[3-5] Over time, feeding problems in neurologically impaired infants have been associated with adverse effects on growth and long-term health.[6]

The observed associations between ventilation duration and feeding milestones should be interpreted within the broader clinical context of HIE. In this population, prolonged ventilation is unlikely to represent an isolated exposure; rather, it may reflect the combined influence of encephalopathy severity, impaired respiratory drive, seizure burden, sedative or anticonvulsant exposure, abnormal neuroimaging findings, and overall clinical instability.[17] Prolonged respiratory support may also reduce early oral experiences and contribute to altered aerodigestive coordination, sensory input, and suck–swallow–breath organization, mechanisms that have been described in high-risk neonatal and pediatric populations.[3,8-10] Although the association between ventilation duration and feeding outcomes persisted after adjustment for several measured clinical covariates, the retrospective design does not allow mechanical ventilation to be identified as the sole cause of delayed feeding. Instead, ventilation duration appears to function as a practical bedside indicator of infants who may require closer feeding surveillance and earlier multidisciplinary support.

Consistent with this interpretation, the lower EFS scores observed in infants with prolonged ventilation suggest that delayed feeding in HIE reflects not only later initiation of oral intake but also reduced functional feeding competence. The EFS Assessment evaluates respiratory regulation, oral-motor function, swallowing coordination, engagement, and physiological stability, thereby capturing clinically relevant aspects of feeding readiness and safety beyond chronological feeding milestones.[18] These findings are consistent with previous reports describing oropharyngeal dysphagia signs in infants with HIE during therapeutic hypothermia and support the need to evaluate feeding skill quality rather than relying solely on the timing of oral feeding initiation.[5]

The ≥7-day ventilation threshold used in subgroup analyses was selected as a clinically interpretable and exploratory cut-off, supported by previous studies reporting increased feeding or swallowing risk after prolonged respiratory support.[8,10,11,14] In our cohort, infants ventilated for seven days or longer exhibited delayed POfirst, POhalf, POFull, transition to breastfeeding, and poorer EFS scores, although no significant differences in weight outcomes were observed. This finding suggests that adequate weight gain may be achieved through enteral nutritional support while functional oral feeding readiness, coordination, and endurance remain delayed. Therefore, weight gain alone may not fully reflect feeding competence in infants with HIE.

The lack of association between ventilation duration and weight outcomes contrasts with previous studies reporting links between prolonged respiratory support, delayed feeding progression, and growth-related outcomes in high-risk neonatal populations.[12,13] In the present cohort, this finding may be explained by the structured nutritional support provided in the NICU, as infants with delayed oral feeding can maintain caloric intake and weight gain through orogastric or other enteral feeding routes even when oral feeding skills are immature. This may explain why ventilation duration was strongly associated with delayed feeding milestones and lower EFS scores, but not with weight at feeding milestones or discharge. In infants with HIE, feeding progression may be influenced by neurological injury, clinical instability, and therapeutic hypothermia-related care processes, underscoring the importance of population-specific evaluation.[4,13]

From a clinical perspective, these findings support the use of early, risk-based feeding assessments in infants with HIE who require prolonged mechanical ventilation. Particular attention should be paid to functional feeding skills, including oral-motor organization, suck–swallow–breath coordination, physiological stability, and feeding endurance, rather than relying only on the timing of oral feeding initiation or weight gain. Early involvement of speech-language therapists may aid in readiness evaluation, individualized progression planning, and family support.[3,9,18] Although evidence for HIE-specific feeding interventions remains limited, studies in other high-risk neonatal populations suggest that structured oral-motor and sensory-based interventions may support feeding development.[2,20,21] Stepwise, individualized feeding pathways may therefore be more appropriate than rigid time-based transitions in this clinically heterogeneous population.

Several limitations should be acknowledged. First, the retrospective design prevents causal inference, and mechanical ventilation duration may partly reflect underlying encephalopathy severity and overall illness burden rather than an isolated exposure. Second, although the regression models were adjusted for HIE stage, gestational age, birth weight, convulsion status, and MRI findings, other potentially relevant confounders, including sedative exposure, anticonvulsant treatment details, sepsis, inotropic support, EEG findings, and detailed respiratory disease severity, could not be fully controlled because of the retrospective nature of the available records. Third, the inclusion of infants with complete oral feeding and breastfeeding transition data may limit generalizability to more severely affected infants who died, were transferred, or did not achieve these feeding milestones during hospitalization. Fourth, although the Turkish version of the EFS Assessment has been validated, it has not been specifically validated in infants with HIE. Finally, although lower-gestational-age infants were included because therapeutic hypothermia was applied at the discretion of the attending neonatologist in selected cases, a sensitivity analysis restricted to infants born at ≥36 weeks' gestation showed findings consistent with the main analysis. Nevertheless, future prospective studies should confirm these results in more homogeneous gestational-age subgroups.

Conclusion

Reference

1) Glass HC. Neonatal seizures: advances in mechanisms and management. Clin Perinatol. 2014;41(1):177-190. doi:10.1016/j.clp.2013.10.004. [ Özet ]

2) Arora I, Bhandekar H, Lakra A, Lakra MS, Khadse SS. Filling the gaps for feeding difficulties in neonates with hypoxic-ischemic encephalopathy. Cureus. 2022;14(8):e28564. doi:10.7759/cureus.28564. [ Özet ]

3) Jadcherla SR. Dysphagia in the high-risk infant: Potential factors and mechanisms. Am J Clin Nutr 2016; 103: 622S-628S. doi:10.3945/ajcn.115.110106. [ Özet ]

4) Martinez-Biarge M, Diez-Sebastian J, Wusthoff CJ, Mercuri E, Cowan FM. Antepartum and intrapartum factors preceding neonatal hypoxic-ischemic encephalopathy. Pediatrics. 2013;132(4):e952-e959. doi:10.1542/peds.2013-0511. [ Özet ]

5) Krüger E, Kritzinger A, Pottas L. Oropharyngeal dysphagia in breastfeeding neonates with hypoxic-ischemic encephalopathy on therapeutic hypothermia. Breastfeed Med. 2019;14(10):718-723. doi:10.1089/bfm.2019.0048. [ Özet ]

6) Fung EB, Samson-Fang L, Stallings VA, et al. Feeding dysfunction is associated with poor growth and health status in children with cerebral palsy. J Am Diet Assoc 2002; 102: 361-373. doi:10.1016/s0002-8223(02)90084-2. [ Özet ]

7) Salvo V, Gazzolo D, Zimmermann LJ. The complex interrelationship between mechanical ventilation and therapeutic hypothermia in asphyxiated newborns: a review. Ther Hypothermia Temp Manag. 2024;14(2):80-88. doi:10.1089/ther.2023.0032. [ Özet ]

8) Bier JA, Ferguson AE, Cho C, Oh W, Vohr BR. The oral motor development of low-birth-weight infants who underwent orotracheal intubation during the neonatal period. Am J Dis Child. 1993;147(8):858-862. [ Özet ]

9) da Silva PSL, Lobrigate NL, Fonseca MCM. Postextubation dysphagia in children: the role of speech-language pathologists. Pediatr Crit Care Med. 2018;19(10):e538-e546. doi:10.1097/PCC.0000000000001688. [ Özet ]

10) Raol N, Schrepfer T, Hartnick C. Aspiration and dysphagia in the neonatal patient. Clin Perinatol. 2018;45(4):645-660. doi:10.1016/j.clp.2018.07.005. [ Özet ]

11) Hoffmeister J, Zaborek N, Thibeault SL. Postextubation dysphagia in pediatric populations: incidence, risk factors, and outcomes. J Pediatr. 2019;211:126-133.e1. doi:10.1016/j.jpeds.2019.02.019. [ Özet ]

12) McCain GC, Gartside PS, Greenberg JM, Lott JW. A feeding protocol for healthy preterm infants that shortens time to oral feeding. J Pediatr. 2001;139(3):374-379. doi:10.1067/mpd.2001.117077. [ Özet ]

13) Gunes AO, Karadag N, Topcuoglu S, Ozalkaya E, Toptan HH, Dincer E, et al. Factors associated with the transition time to full enteral feeding in newborns with hypoxic ischemic encephalopathy. Arch Iran Med. 2022;25(8):547-551. doi:10.34172/aim.2022.87.

14) Asgarshirazi M, Shariat M, Moradi R, Farahani Z, Ziaei F. Evaluation of Swallowing Coordination in Infants with a History of Mechanical Ventilation Longer Than 7 Days in the Neonatal Period and Assessment of Response to Oral Stimulation Maneuver. Iran J Pediatr. 2023;33(4):e135552. doi:10.5812/ijp-135552.

15) Akisu M, Kumral A, Canpolat FE. Turkish Neonatal Society Guideline on neonatal encephalopathy. Turk Pediatri Ars. 2018;53(Suppl 1):S32-S44. doi:10.5152/TurkPediatriArs.2018.01805. [ Özet ]

16) Sarnat HB, Sarnat MS. Neonatal encephalopathy following fetal distress: a clinical and electroencephalographic study. Arch Neurol. 1976;33(10):696-705. doi:10.1001/archneur.1976.00500100030012. [ Özet ]

17) Szakmar E, Jermendy A, El-Dib M. Respiratory management during therapeutic hypothermia for hypoxic-ischemic encephalopathy. J Perinatol. 2019;39(6):763-773. doi:10.1038/s41372-019-0349-2. [ Özet ]

18) Thoyre SM, Shaker CS, Pridham KF. The Early Feeding Skills Assessment for preterm infants. Neonatal Netw. 2005;24(3):7-16. doi:10.1891/0730-0832.24.3.7. [ Özet ]

19) Girgin BA, Gözen D, Uslubaş R, Bilgin L. The evaluation of oral feeding in preterm infants: Turkish validation of the Early Feeding Skills Assessment Tool. Turk Arch Pediatr. 2021;56(5):440-446. doi:10.5152/TurkArchPediatr.2021.21008. [ Özet ]

20) Thakkar PA, Rohit HR, Ranjan Das R, Thakkar UP, Singh A. Effect of oral stimulation on feeding performance and weight gain in preterm neonates: a randomised controlled trial. Paediatr Int Child Health. 2018;38(3):181-186. doi:10.1080/20469047.2018.1435172. [ Özet ]

21) Arora K, Goel S, Manerkar S, Konde N, Panchal H, Hegde D, Mondkar J. Prefeeding Oromotor Stimulation Program for Improving Oromotor Function in Preterm Infants — A Randomized Controlled Trial. Indian Pediatr. 2018;55(8):675-678. doi:10.1007/s13312-018-1357-6. [ Özet ]