Elevated Midline Positioning During Diaper Change in Preterm Infants

Background

Germinal matrix-intraventricular hemorrhage (GM-IVH) is one of the most common and serious neurological complications of preterm birth. Fluctuations in cerebral blood flow and increases in cerebral venous pressure play a central role in its pathogenesis. Turning the head to one side can functionally occlude the ipsilateral jugular vein, impairing cerebral venous drainage and increasing intracranial pressure and cerebral blood volume (Cowan \& Thoresen, 1985; Pellicer et al., 2002). Intracranial pressure in newborns is lowest when the head is elevated by 30 degrees and kept in the midline (Emery \& Peabody, 1983), and midline head positioning with head-of-bed elevation has become a core component of IVH-prevention nursing care bundles (de Bijl-Marcus et al., 2017). In a randomized controlled trial, Kochan et al. (2019) reported less grade 4 hemorrhage and higher survival to discharge in extremely low birth weight infants kept in an elevated midline position during the first 96 hours of life compared with flat supine positioning; however, the most recent Cochrane review (Romantsik et al., 2020) concluded that the evidence remains uncertain and that high-quality randomized trials are needed.

Routine care procedures, particularly diaper changing, increase heart rate, decrease oxygen saturation and elicit behavioral stress responses in preterm infants (Holsti et al., 2005). Keeping the infant in a side-lying position with the head turned together with the trunk during diaper change may further impair cerebral venous return while the infant is already physiologically stressed. Although the effect of body position on physiological stability has been shown (Ancora et al., 2010; Ghorbani et al., 2013), no randomized controlled trial has directly evaluated 30-degree elevated midline positioning during an acute, stressful routine procedure such as diaper change.

Objective

To determine the effect of 30-degree elevated midline positioning applied during routine diaper change, compared with standard care, on physiological stability (desaturation and bradycardia episodes; recovery time of heart rate and oxygen saturation) and comfort/stress behaviors in stable preterm infants born at 28-34 weeks of gestation.

Hypotheses

• H1: The number of desaturation and bradycardia episodes during and after diaper change will be lower in infants receiving 30-degree elevated midline positioning than in the standard care (control) group.
• H2: NIPS pain/stress scores will be lower in the intervention group than in the control group.
• H3: The time required for heart rate and oxygen saturation to return to baseline values (recovery time) will be shorter in the intervention group than in the control group.

Design and Setting

Single-center, prospective, two-arm, parallel-group randomized controlled trial with 1:1 allocation, conducted in the level III NICU of Antalya City Hospital, Antalya, Türkiye. The trial will be reported according to the CONSORT statement.

Participants and Timing

Hemodynamically stable preterm infants born at 28-34 weeks of gestation who meet the eligibility criteria will be enrolled after written informed consent is obtained from a parent. To avoid the period of highest IVH risk, no study procedure will be performed before 96 hours of postnatal age. All procedures will be carried out while the infant is clinically stable, with the approval of the neonatal nurse and the principal investigator. Each infant will be evaluated during a single diaper change.

Randomization and Allocation Concealment

Allocation will follow a list generated with a computer-based random number generator. Allocation concealment will be ensured with sequentially numbered, opaque, sealed envelopes that are opened by the intervention nurse only after eligibility has been confirmed and consent has been obtained.

Interventions

Common preparation (both groups): the infant is fed at least 30 minutes before the procedure; continuous SpO2 and heart rate monitoring is in place; ambient light and noise are minimized; thermoregulation is maintained; baseline (T0) values are recorded.

Intervention group (30-degree elevated midline position): the head of the bed is elevated to 30 degrees (verified with a protractor or the bed indicator), tilting the entire back-spine plane without flexion at the hips. The head is kept in the midline in line with the body, with rolled supports on both sides and a neutral neck position (chin not touching the chest) to maintain airway patency. The diaper is changed with a log-roll technique without lifting the legs by the ankles: head, shoulders and pelvis are turned as one unit by the smallest angle sufficient for cleaning (approximately 15-30 degrees) to one side and then to the other, while a second nurse continuously keeps the head in the midline and aligned with the trunk. The clean diaper is slid into place without lifting the pelvis. After the procedure the infant is returned to the supine midline position and the 30-degree elevation is maintained throughout the 5-minute post-procedure observation period.

Control group (standard care): the diaper is changed according to the unit's current standard practice with the infant in a side-lying (lateral) position on a flat bed (0 degrees). The head is turned to one side together with the trunk as a natural consequence of the lateral position; no active midline positioning, rolled supports, head-of-bed elevation or log-roll technique is applied.

The planned difference between the groups is the 30-degree head-of-bed elevation and the midline head position. The log-roll technique is the means of maintaining this position during the diaper change and is not a separate comparison variable.

Study Personnel and Blinding

Eligibility (inclusion/exclusion criteria, clinical stability, postnatal age) is confirmed by the principal investigator. Envelope opening, positioning and the diaper change are performed by a neonatal nurse researcher with at least one year of NICU experience who has been trained on the study protocol; a second neonatal nurse supports the infant's head in the midline. Heart rate and SpO2 are recorded at T0-T3 by a researcher who is not involved in the procedure, and NIPS scoring is performed by direct observation by an independent neonatal nurse who is not involved in the procedure and is not informed of group allocation. Because the positioning difference is visible, full blinding of the nurses and bedside assessors is not possible; this will be reported as a limitation. Data analysis will be performed on a data set in which the groups are coded as A and B, and the codes will be revealed only after the analysis is completed. No video or photographic recordings will be made.

Data Collection and Outcome Assessment

Data will be collected with the Infant and Parent Information Form / Case Report Form and the Neonatal Infant Pain Scale (NIPS; Lawrence et al., 1993; Turkish validity and reliability study by Akdovan, 1999). Heart rate, SpO2 and NIPS scores will be recorded at four time points: T0, 1 minute before the procedure (baseline); T1, during the diaper change (lowest SpO2, the heart rate value with the greatest deviation from baseline, and the highest NIPS score); T2, 1 minute after the procedure; and T3, 5 minutes after the procedure. A desaturation episode is defined as SpO2 \< 85% lasting at least 10 seconds and a bradycardia episode as heart rate \< 100 beats/min lasting at least 10 seconds; two episodes are counted separately only if the value remains above the threshold for at least 10 seconds between them. Episodes are counted from the start of the diaper change until 5 minutes after its completion. Recovery time is the time in seconds from completion of the diaper change (clean diaper closed and infant in the final position) until SpO2 returns to no more than 2 percentage points below the T0 value, or heart rate returns to within ±10 beats/min of the T0 value, and remains there for at least 10 seconds. Infants already within the baseline range at completion are assigned 0 seconds; infants who have not recovered by T3 are assigned 300 seconds and flagged as not recovered.

Safety

All infants are continuously monitored. The procedure is stopped immediately if SpO2 falls below 80% for more than 20 seconds, heart rate falls below 80 beats/min, apnea lasting 20 seconds or longer occurs, or the nurse performing the procedure considers it clinically necessary; the infant is then placed in a neutral/stabilizing position and the standard NICU stabilization protocol is applied. Brief, self-resolving falls that do not meet these criteria (SpO2 \< 85% or heart rate \< 100 beats/min) are closely monitored, recorded as episodes, and the procedure is continued; the nurse may stop the procedure at any time for the infant's safety. All adverse events and protocol deviations are recorded on the case report form and on the adverse event and protocol deviation log of the follow-up form.

Sample Size

Sample size was calculated with G*Power 3.1 for an independent-samples t-test (two-tailed, α = 0.05, power = 0.80, effect size Cohen's d = 0.70), requiring at least 34 infants per group (68 in total). Allowing for approximately 15% data loss, 40 infants per group (80 in total) will be enrolled.

Statistical Analysis

Data will be analyzed with SPSS 25.0. Normality will be assessed with the Shapiro-Wilk test. Continuous variables will be presented as mean ± standard deviation or median (interquartile range). Groups will be compared with the independent-samples t-test or the Mann-Whitney U test, repeated measurements with mixed models or repeated-measures ANOVA, and categorical variables with the chi-square test. Because episode counts are count data, groups will also be compared with Poisson regression (negative binomial regression in case of overdispersion), reporting incidence rate ratios with 95% confidence intervals. Recovery times, which may be censored at 5 minutes, will also be analyzed with the Kaplan-Meier method and the log-rank test. Postnatal age, postmenstrual age on the day of the procedure and type of respiratory support will be recorded as potential confounders and included as covariates in regression models when appropriate. Analyses will be performed on a data set in which groups are coded as A and B. Effect sizes will be reported as Cohen's d. Statistical significance will be set at p \< 0.05.

Ethics

The study was approved by the Akdeniz University Medical Scientific Research Ethics Committee (approval no: TBAEK-828, date: 27.08.2026); an amendment clarifying outcome definitions, stopping criteria and the study timeline was submitted to the committee in October 2026. The trial is registered before enrollment of the first participant. Written informed consent will be obtained from a parent of each infant, and consent may be withdrawn at any time without affecting the infant's care.