MANUSCRIPT CITATION

Kochan M, Leonardi B, Firestine A, McPadden J, Cobb D, Shah TA, et al. Elevated midline head positioning of extremely low birth weight infants: effects on cardiopulmonary function and the incidence of periventricular-intraventricular hemorrhage. J Perinatol 2019, 39(1): 54-62. PMID: 30655595

REVIEWED BY

David N. Matlock, Jr., MD
University of Arkansas for Medical Sciences

TYPE OF INVESTIGATION

Prevention

QUESTION

In infants with birth weight < 1000 grams (P), does the use of a supine position with the head of bed elevated 30 degrees (I) compared to a supine position with no elevation (C) decrease the incidence of periventricular and intraventricular hemorrhage seen on ultrasound (O) when used during the first four days of life (T)?

METHODS

• Design: Prospective, randomized, single center. Approved by institutional review board. There is no mention of prospective registration with ClinicalTrials.gov.
• Allocation: There is no mention of concealed allocation.
• Blinding: Partially blinded. Pediatric radiologists interpreting ultrasound results were blinded to the randomization. There is no mention of blinding other providers or research staff, and it would be difficult or impossible to blind caregivers due to the nature of the treatment of interest.
• Follow-up period: Infants were followed until death or hospital discharge.
• Setting: Single neonatal intensive care unit in Norfolk, VA, USA.
• Patients: Infants with birth weight < 1000 grams.
  • Inclusion Criteria: Birth weight < 1000 grams and able to be randomized, positioned according to randomization, and obtain initial ultrasound all within four hours of birth.
  • Exclusion Criteria: Congenital anomalies.
• Intervention: For neonates randomized to the treatment group, infants were maintained supine with head midline and elevated 30 degrees above horizontal. Infants in the control group were maintained supine and flat with head turned 90 degrees to the right or left every four hours. All infants were maintained in the control group position after the fourth day of life.
• Outcomes:
  • Primary outcome: Incidence of peri/intraventricular hemorrhage on ultrasound done daily during the first four days of life and on day of life seven.
  • Secondary outcomes: Indices of cardiopulmonary function (including daily maximal FiO₂, lowest pH, highest PCO₂, level of respiratory support, lowest mean blood pressure, and use of inotropic medications), number of chest radiographs, time to extubation, time to room air ventilation, and incidence of pneumothorax, pulmonary hemorrhage, and bronchopulmonary dysplasia.
• Analysis and sample size: From the previous four years of data at the institution, the rate of the primary outcome was predicted to be 40% in the control group. Assuming a 20% relative reduction, a sample size of 180 infants was calculated to achieve 80% power (α = 0.05). Infants were randomized using block randomization and analysis was based on the intention-to-treat.
• Patient follow-up: Of the enrolled infants, 180 (100%) were analyzed for the primary outcome and followed until death or discharge for the primary and secondary outcomes.

MAIN RESULTS

The reported characteristics of the mothers and infants were similar between treatment and control groups with the following exceptions: Pre-eclampsia was more common in the treatment group (p = 0.026) and prolonged rupture of membranes was more common in the control group (p = 0.018).

PRIMARY OUTCOME

The primary outcome, any peri/intraventricular hemorrhage documented on ultrasound, occurred in 34 of 90 infants in the treatment group and 31 of 90 infants in the control group (p value not reported, no significant difference). Rates of Grade I, II, and III hemorrhage detected on ultrasound also were not different between groups (p = 0.34, 0.23, and 0.72 respectively). However, Grade IV hemorrhage was more common in the control group affecting 14 infants, but only 6 infants in the treatment group (p = 0.036).

SECONDARY OUTCOMES

There were a few prominent differences in the secondary outcomes between groups. Infants in the treatment group had a lower maximal FiO₂ requirement during the first 24 hours (only), but higher PCO₂ levels on days 2 and 3. Level of respiratory support differed among groups with a higher odds ratio of increased support by day four (p = 0.01), lower mean airway pressure for infants on high frequency on day 1 (p = 0.023), and higher mean airway pressure for infants on conventional on day four (p = 0.027) in the treatment group than the control group. Lowest mean blood pressures were higher in the treatment than the control group on days one and three (p = 0.003, p = 0.03). Infants in the treatment group were more likely to survive to discharge (p = 0.033). Other cardiorespiratory physiologic parameters and rates of complications of prematurity did not differ between groups.

Max, maximum. SD, standard deviation. HFV, high frequency ventilation. CV, conventional ventilation. Min, minimum.

CONCLUSIONS

The authors conclude that the use of an elevated midline head position for the first four days of life results in fewer grade IV intracranial hemorrhages in infants with birth weight < 1000 grams.

COMMENTARY

Germinal matrix-intraventricular hemorrhage (IVH) of the premature infant is the most common form of neonatal intracranial hemorrhage. A portion of infants with IVH suffer a related parenchymal injury called periventricular hemorrhagic infarction. Fluctuations in cerebral blood flow and increases in cerebral venous pressure have been implicated in the pathogenesis of these injuries (1). Head position can affect cerebral hemodynamics and has been evaluated for preventing IVH (2). A recent systematic review evaluated trials involving head positioning strategies to prevent IVH, reviewed two trials (n = 110), and found insufficient evidence to determine the effectiveness of a particular strategy (3). Both trials reviewed examined the effect of head rotation. No trials were identified which investigated head elevation (3, 4). The current study aims to close this knowledge gap.

This single-center, prospective, randomized trial compared the use of an elevated midline head position to a flat head position for the prevention of intracranial hemorrhage in neonates < 1000 grams at birth. Baseline characteristics differed among the groups with more pre-eclampsia in the treatment group and more prolonged rupture in the control group. There was no difference in the primary outcome. However, there was improved survival and decreased grade IV (but not grade III) IVH in the treatment group. There was a trend toward escalating respiratory support in the treatment group. Of interest, none of the infants with grade IV IVH developed cystic PVL.

A strength of the study included the early enrollment by four hours of life. Although enrollment at birth (with prenatal consent) would be preferred, this may be impractical as flat positioning may be required during initial stabilization procedures.

There is risk for bias with this study as it is unclear if allocation was concealed. Although the interpreting radiologists were blinded, blinding of all caretakers is not possible when evaluating this intervention. It is unclear if the effect of the intervention was due to the absence of periodic turning or the elevation, as infants in the two groups were treated differently in both regards. In fact, the control group may have been affected by compromised venous return due to turning the head only rather than head and body.

A weakness of the study is the lack of long-term developmental follow up. As demonstrated in the largest trial of prophylactic indomethacin, a decrease is severe IVH does not always produce an improvement in long-term developmental outcome (5). Another weakness is the lack of magnetic resonance imaging (MRI) to assess for periventricular leukomalacia (PVL). Mean blood pressures were slightly higher in the elevated group, but cerebral perfusion may have been affected by the elevation placing the infants at risk for PVL.

These results suggest that an elevated head position may be beneficial over flat supine head positioning with periodic turning for improving survival and decreasing grade IV IVH. Further, the results suggest that the control and intervention conditions are both safe with no harm shown in either group. With the differences in baseline characteristics, lack of allocation concealment, lack of MRI to assess for PVL, and most importantly, lack of developmental follow up, it is not possible to strongly suggest a practice change from the results of this trial. However, these results should inform the design of a large, multicenter trial of this simple intervention.

REFERENCES

1. Volpe JJ. Volpe’s neurology of the newborn, Sixth edition. edn. Elsevier: Philadelphia, PA, 2018.
2. Pellicer A, Gaya F, Madero R, Quero J, Cabanas F. Noninvasive continuous monitoring of the effects of head position on brain hemodynamics in ventilated infants. Pediatrics 2002; 109: 434-440.
3. Romantsik O, Calevo MG, Bruschettini M. Head midline position for preventing the occurrence or extension of germinal matrix-intraventricular hemorrhage in preterm infants. Cochrane Database Syst Rev 2017, 7: CD012362.
4. Al-Abdi SY, Nojoom MS, Alshaalan HM, Al-Aamri MA. Pilot-randomized study on intraventricular hemorrhage with midline versus lateral head positions. Saudi Med J 2011; 32: 420-421.
5. Schmidt B, Davis P, Moddemann D, Ohlsson A, Roberts RS, Saigal S, et al. Long-term effects of indomethacin prophylaxis in extremely-low-birth-weight infants. N Engl J Med 2001; 344: 1966-1972.