Secondary Logo

Journal Logo


Use of a Midliner Positioning System for Prevention of Dolichocephaly in Preterm Infants

McCarty, Dana B. PT, DPT; O'Donnell, Shannon PT, DPT; Goldstein, Ricki F. MD; Smith, P. Brian MD; Fisher, Kimberley PhD, FNP-BC; Malcolm, William F. MD

Author Information
doi: 10.1097/PEP.0000000000000487


Dolichocephaly (positional scaphocephaly) is defined as an elongated anterior-posterior axis of the head as a result of head flattening during side-to-side head positioning.1,2 Bilateral flattening of the lateral skull develops due to the weight of the head and pressure of gravity.3 This deformity can occur during the hospitalization of preterm infants 32 weeks' or less gestational age (GA) due to caregiver preference for the side-lying or prone position for ease of containment, decreased gastroesophageal reflux (GER) episodes, and improved respiratory function.3–5 Use of positioning aids intended to maintain the head midline when in the supine position and caregiver education about the importance of positional variety are common interventions used to reduce dolichocephaly.1,6–9 Despite these interventions, up to 73% of very preterm infants continue to have dolichocephaly at term-corrected age.10

The long-term consequences of dolichocephaly are not fully known. Association with delayed reaching skills, tightness in the spinal extensors and scapular retractors, and asymmetrical motor development have been noted.1,3–5 Other longer-term effects of dolichocephaly have not been well studied. Elliman et al11 reported comparable developmental quotients at 3 years of age when comparing a preterm group with significant head narrowing with controls with typical head shapes. Mewes and colleagues,12 however, suggested that the shift in cortical structures that occurs in infants with dolichocephaly may affect the preterm brain, which continues to develop rapidly after birth.

A recent retrospective study in our neonatal intensive care unit (NICU) demonstrated that 54% of premature infants born less than 32 weeks' gestation developed dolichocephaly during the hospital stay.13 Dolichocephaly has been quantified using Ballert Orthopedic calipers and calculation of the cranial index (CI) or the ratio of the widest transverse diameter of the head over the length of the head.2,14 While various reported ranges exist for cranial molding norms,7,1516 the more widely adopted definition of dolichocephaly, defined as a CI of <76%,23,910,17–19 was used.

The highest percentage of dolichocephaly was noted between 32 and 34 weeks' postmenstrual age (PMA). Mean chronologic age of the infants at the time of diagnosis was 5.5 weeks. Infants with dolichocephaly at 32 to 34 weeks' PMA and at discharge were also more likely to have the deformity at outpatient follow-up, suggesting that cranial molding deformity may be difficult to reverse in the first 6 months postdischarge. Furthermore, our study found no association between infant GA at birth, birth weight (BW), bronchopulmonary dysplasia (BPD), diagnosis of GER, or severe intraventricular hemorrhage (IVH) and the development of dolichocephaly. These results suggested that infant positioning had a greater effect on cranial molding than neonatal comorbidities.

The most noteworthy finding in this study was the correlation between dolichocephaly at 32 to 34 weeks' PMA and referral for outpatient physical therapy (PT) services in early infancy.13 These results support the importance of early prevention and treatment of dolichocephaly in the hospital from birth to 34 weeks' PMA—a critical time for cervical muscle development.13

Currently, a variety of cranial molding devices are commercially available,20–26 but the best available method for prevention of cranial molding deformity in high-risk premature infants has not been determined. The purpose of this study was to determine the effectiveness of using a midliner positioning system20 (MPS, Tortle Midliner, Tortle Products, LLC) to prevent dolichocephaly in premature infants during their NICU stay. The study aims to determine whether the use of this type of MPS is a more effective prevention strategy for dolichocephaly than current standard-of-care interventions in the NICU.


Study Design

This study was a nonrandomized, prospective study of 31 premature infants using the MPS in Duke University's Intensive Care Nursery and Transitional Care Nursery (level IV and level II/III, respectively, combined as a 67-bed NICU) and was approved by the Duke University Institutional Review Board. Parents of all infants provided written informed consent. Infants were enrolled at the time of initial PT evaluation (0-3 weeks' chronological age).

The 31 infants in our study cohort (SC) were compared with a group of 65 infants from our previous retrospective study,13 who were positioned at registered nurse discretion with or without the use of moldable pillows.24 Infants in both the SC and the retrospective study cohort (RSC) received PT intervention 1 to 2 times per week, and their caregivers (nursing team and parent) received education about cranial molding from a PT per standard-of-care intervention. Cranial index measures (see procedures later) between the 2 cohorts were compared. Cranial index measures for the RSC were recorded between September 2013 and January 2015, and CI measures for the SC were collected between May 2015 and March 2016.

All study personnel received human subjects research training from the Collaborative Institutional Training Initiative Program. The Tortle Champions were 10 nurses covering both daytime and evening shifts and 2 daytime shift PTs who were available to other nurses and caregivers in the NICU to provide on-site training for use of the MPS in real time when caring for consented subjects. All Tortle Champions received training from a Tortle Products LLC representative on the use of the MPS immediately prior to the study period and were also involved in the development of a nursing questionnaire to evaluate the feasibility of the MPS in daily care. Additional information about this questionnaire is described later in this article and in Supplemental Digital Content 2 (available at: The contact information for the primary investigator (PI) and Tortle Champions were posted at the bedside of each infant enrolled in the study. The PI was available by phone during all shifts and called the evening and weekend shift nurse daily for the first week of an infant's enrollment in the study. In addition, a study and MPS informational sheet was added to the bedside chart of each enrolled infant for nurse review.

Selection of Subjects

Inclusion Criteria. Inclusion criteria were a birth weight of less than 1500 g, a GA less than 31 weeks (≤30 weeks 6 days) at birth and at the time of enrollment, less than 3 weeks' chronological age at the time of enrollment, and stability on continuous positive airway pressure (CPAP), nasal cannula, or room air. These inclusion criteria were selected to best match the RSC.

Exclusion Criteria. Exclusion criteria were genetic/chromosomal abnormality, congenital neuromuscular disorder, craniofacial abnormalities, congenital or posthemorrhagic hydrocephalus, and other diagnoses determined by the PI that could impact generalizability of results.

Determination of Cranial Index

The Ballert Orthopedic Cranial Caliper was used to measure CI defined as the ratio of head width to head length expressed as a percentage. The CI measured with spreading calipers has good sensitivity (93%) and high accuracy (96%) and is a better measure of head shape than visual assessment.19 Two therapists were trained using the Ballert Orthopedic caliper user guide instructions.14 Interrater reliability was established between the therapists until agreement to the lead therapist was within 0.25 cm for head width and length measurements for both the retrospective and prospective studies. The lead therapist was the same on both the retrospective and prospective studies.

Cranial index was determined by calculating the ratio of the biparietal diameter (BiPD) over the occipitofrontal diameter (OFD). The BiPD is defined as the widest transverse diameter of the head. The BiPD was measured from the most prominent lateral point on each side of the skull in the area of parietal and temporal bones.2 The OFD is defined as the diameter of the head from the glabella—the most prominent midline point of the frontal bone—to the opisthocranion—the most prominent point on the occiput.2,14 Dolichocephaly was defined as a CI of less than76%23,910,17–19 in alignment with the CI values used for the RSC. The normative CI range is 76% to 85% for prone and supine sleeping infants.23,10

Midliner Positioning System

The MPS20 is a breathable, knit beanie with 2 support rolls to position the infant's head in midline while in the supine position (Figure 1). It can also be worn in the side-lying or prone position. The original design used in this study includes Velcro adjustments and tabs for nasal cannula and feeding tubes. It is compatible with some ventilation devices, nasal CPAP, x-ray, and bilirubin shades. The MPS is designed to reduce dolichocephaly and provide passive stretch to cervical rotators if head preference has developed. It comes in 3 sizes and is designed to fit premature infants weighing 500 to 2500 g.20

Fig. 1.
Fig. 1.:
Tortle Midliner positioning system.

Positioning Protocol

Nursing orders to follow a bedside positioning protocol and provide MPS use 24 hours a day were implemented for all enrolled infants at the beginning of the study period. The positioning protocol (Figures 2–5) was to be maintained at all times, but the nurse could use discretion to reposition the infant if he/she demonstrated signs of discomfort or autonomic instability related to positioning. In addition, the MPS was to be worn at all times and in every position unless the infant developed skin irritation, was receiving kangaroo care, or if the infant demonstrated signs of discomfort or instability by nurse observation. To ensure consistent wear and use of the MPS, the primary investigator (PI) or coinvestigator visited the study infant and assigned nurse at bedside 5 days a week, and the PI called each nurse assigned to a study infant over the weekend throughout the study period. In addition, the PI called each nurse assigned to a study infant every evening shift during the first week of study enrollment. If during any of these encounters, the PI or coinvestigator found the MPS to be removed, they provided the nurse with instructions or strategies for positioning the MPS on the infant by the next care time. These measures reduced sustained periods (>12-24 hours) when the MPS was not in place. Two Tortle Champions developed a positioning log (supine, right side-lying, left side-lying, prone, and kangaroo care) via chart review at the end of the study period by for each infant in the SC.

Fig. 2.
Fig. 2.:
Tortle Midliner Positioning System positioning protocol for infants less than 32 weeks' postmenstrual age without dolichocephaly in prevention phase.
Fig. 3.
Fig. 3.:
Tortle Midliner Positioning System positioning protocol for infants less than 32 weeks' postmenstrual age with dolichocephaly in treatment phase.
Fig. 4.
Fig. 4.:
Tortle Midliner Positioning System positioning protocol for infants 32 weeks' or more postmenstrual age without dolichocephaly in prevention phase.
Fig. 5.
Fig. 5.:
Tortle Midliner Positioning System positioning protocol for infants 32 weeks' or more postmenstrual age with dolichocephaly in treatment phase.

The PI and coinvestigator updated the positioning protocol weekly based on the PMA of the infant and the weekly CI measurement. This positioning protocol is recommended by Tortle Products LLC and is provided in the packaging with each Tortle Midliner.20 The protocol recommends increased time in the supine position at 32 weeks' PMA in accordance with the American Academy of Pediatrics guidelines for safe sleep practices for high-risk infants and preparation for discharge home.27 Infants less than 32 weeks' PMA received supine positioning 25% of the time (1 in 4 positions). Because all infants were enrolled in the study at less than 31 weeks' PMA, all infants were positioned in the supine position at this rate initially. When infants were more than 32 weeks' PMA, the positioning protocol was updated, and the infants received supine positioning 33% of the time (1 in 3 positions).

In addition, if the CI was 76% or more (no dolichocephaly, prevention phase), then the infant experienced true side-lying positioning with pressure through the lateral skull. For infants with a CI less than 76% (dolichocephaly, treatment phase), the partial side-lying position was used to provide weight bearing on the posterolateral aspect of the head for cranial reshaping. The partial side-lying position was achieved using a small blanket roll to maintain the head in line with the body.

The following positioning protocols were implemented by the PI or coinvestigator weekly for each infant based on PMA and CI as follows:

  • For infants less than 32 weeks' PMA with a CI 76% or more (no dolichocephaly, prevention phase), positions were supine (25%), right side-lying (25%), prone (25%), and left side-lying (25%) (Figure 2).
  • For infants less than 32 weeks' PMA with a CI less than 76% (dolichocephaly, treatment phase), positions were supine (25%), right partial side-lying (25%), prone (25%), and left partial side-lying (25%) (Figure 3).
  • For infants 32 weeks' or more PMA with a CI 76% or more (no dolichocephaly, prevention phase), positions were supine (33%), right or left side-lying (33%), and prone (33%) (Figure 4).
  • For infants 32 weeks' or more PMA with a CI less than 76% (dolichocephaly, treatment phase), positions were supine (33%), right or left partial side-lying (33%), and prone (33%) (Figure 5).

Questionnaire for Nursing Staff

Each nurse who served as a “primary” team member of a study infant was asked to answer 10 questions about the feasibility of using the MPS in daily care (Supplemental Digital Content 2, available at: The PI and the developer of the MPS created this questionnaire with input from the Tortle Champion nurses after they had been trained to use the device. Nursing feedback was considered crucial to the success of the MPS device since the nurse manipulates and adjusts the device during care times and ensures compatibility with other medical devices such as feeding tubes and respiratory equipment. Furthermore, nurses have the opportunity to educate the parents about how to use the MPS. The PI emailed the questionnaire link to the primary nurse within 1 week of the end of the infant's study period.

Data Collection

Based on the retrospective nature of the RSC study, CI measures for the RSC were measured only once at variable time points between 32 and 34 weeks' PMA depending on infant availability and stability at the time of PT assessment. Of the 65 infants in the RSC, 38 were measured at 32 weeks' PMA, 18 were measured at 33 weeks' PMA, and 4 were measured at 34 weeks' PMA. Demographic and medical comorbidity information collected for RSC infants included infant GA, BW, diagnosis of BPD at discharge, diagnosis of GER at discharge, and presence of severe IVH (grade III or IV).14,28 Bronchopulmonary dysplasia was defined as requiring supplemental oxygen or other forms of respiratory support at 36 weeks' PMA.29 Gastroesophageal reflux was defined as receiving medical treatment for reflux, including antacid therapy and/or a promotility agent at discharge.30

Cranial index measures for the SC were collected at the beginning of the study period and then weekly until 34 weeks' PMA (end of the study period). Demographic and comorbidity information collected for the SC group included infant GA, BW, number of days on CPAP during the study period, diagnosis of BPD at discharge, and diagnosis of GER at discharge. Intraventricular hemorrhage and BPD were not evaluated for association with dolichocephaly as part of the prospective analysis because only 2 infants in the SC had diagnosis of severe IVH and only 1 infant had a diagnosis of BPD at the time of data collection. Number of days on CPAP during the study period was determined to establish any association with CPAP use and cranial molding deformity.7,1516 Continuous positive airway pressure use and history of GER were recorded due to nursing reports of repositioning the infant in the side-lying or prone position from the supine position due to respiratory instability and reflux episodes. Therefore, if infants had a history of either greater CPAP use or GER episodes, they may have been positioned at a lower frequency in supine, which is an important position to maintain round head shape.3,67 Frequency of time spent in the supine position for the SC was recorded in a positioning log obtained through chart review at the end of the study period by Tortle Champion nurses to evaluate adherence to the positioning protocol (ie, the supine position 25% of time for infants <32 weeks' PMA and the supine position 33% of time for infants ≥32 weeks' PMA).

The primary nurse(s) for each infant completed a questionnaire within 1 week of the end of the study period to evaluate ease of use, feasibility, and compatibility of the MPS in daily care. A total of 43 nurse responses were recorded. The nurses rated frequency of having a favorable experience in several areas related to the feasibility of use and infant adaptability to the MPS (frequently: 75%-100% of time, sometimes: 50% of time, rarely or never: 0%-25% of time). For the purposes of this study, a favorable response was defined as a rating of “frequently: 75%-100% of the time.” If the nurse chose “rarely or never” as a rating for a questionnaire item, they were prompted to provide a comment to specify reasons for the response.

Statistical Analysis

Descriptive statistics were used to characterize the demographics and comorbidities of the SC, hospital course, and clinical outcomes including prevalence of dolichocephaly. The Wilcoxon rank sum and Fisher exact tests were used to compare characteristics and outcomes of infants in the SC and the RSC. Paired t tests were developed to evaluate changes between baseline and final CI measures in the RSC and the SC. STATA (version 12, College Station, Texas) was used for the analysis. A P value of <.05 was considered statistically significant.


A total of 31 infants were enrolled in the study. Of these, 1 infant developed posthemorrhagic hydrocephalus after enrollment, 4 infants were transferred to outside facilities prior to 34 weeks' PMA, and 2 infants were withdrawn from the study per parent request. Of these infants who did not complete the study, all but 1 had CI measures up to at least 31 weeks' PMA and were included in analysis (30 infants). Regarding demographics and comorbidities and associations with development of dolichocephaly, the RSC and SC groups were similar. For the RSC, no significant associations were found between infant GA (P = .9), BW (P = .3), BPD (P = .3), GER (P = .7), and the development of dolichocephaly.13 In the SC, infant GA (P = .7), BW (P = .9), GER (P = .9), PMA at the time of enrollment (P = .2), days on CPAP during the study period (P = .2), and percentage of time spent in the supine position (<32 weeks' PMA, P = .4; >32 weeks' PMA, P = .1) were not significantly associated with the development of dolichocephaly (Table 1).

TABLE 1 - Demographic Variables and Neonatal Comorbidities Associated With Dolichocephalya
SC RSC P Value(Group Comparison)
N = 30 P Value N = 65 P Value
Gestational age,b wk 27 (25-30) .7 27 (23-32) .9 .3
Birth weight,b g 975 (590-1430) .9 900 (475-1430) .3 .2
BPD, % 4 - 31 .3 .003
GER, % 35 .9 60 .7 .04
PMA at study initiation,b wk 29 .2 - - -
Days on CPAPb 7 .2 - - -
Supine frequency at <32-wk PMA (goal 25%) 23% .4 - - -
Supine frequency at >32-wk PMA (goal 33.33%) 26% .1 - - -
Abbreviations: BPD, bronchopulmonary dysplasia; CPAP, continuous positive airway pressure; GER, gastroesophageal reflux; PMA, postmenstrual age; RSC, retrospective study cohort; SC, study cohort.
aNo significant associations between demographic variables in the SC or RSC. Significant differences noted between RSC and SC in number of infants with BPD and GER.
bMean (range).

Differences in the Cranial Index by PMA

Weekly CI measures from the beginning of the study period until 34 weeks' PMA were obtained for all infants in the SC, with the exception of 1 infant who was not available at 33 weeks' PMA. Incidence of dolichocephaly between 32 and 34 weeks' PMA in RSC infants versus SC infants were as follows: 32 weeks' PMA: 15/38 (40%) versus 2/25 (8%); 33 weeks' PMA: 6/18 (33%) versus 2/23 (9%); 34 weeks' PMA: 2/4 (50%) versus 3/24 (12.5%). The RSC 32 to 34 weeks' mean CI values and the SC final CI values between 32 and 34 weeks' PMA (based on last CI measure during study period) were significantly different (P = .03), indicating a larger average CI in the SC. In addition, CI values at 32 weeks' PMA and 34 weeks' PMA were significantly greater in the SC group (P = .04 and P = .03, respectively; Table 2).

TABLE 2 - Cranial Index Measures Between 32 and 34 Weeks' PMA
N Mean (Range) N Mean (Range) P Value
Final CI (mean 32-34 wk) 30 79 (71-83) 65 77 (69-88) .03
32 wk 25 79 (71-86) 38 77 (69-88) .04
33 wk 24 79 (70-85) 18 77 (70-83) .28
34 wk 24 79 (73-83) 4 75 (72-78) .03
Abbreviations: CI, cranial index; PMA, postmenstrual age; RSC, retrospective study cohort; SC, study cohort.

Use of Tortle MPS

To determine the effectiveness of the MPS in prevention of dolichocephaly, baseline and final CI measures between the RSC and the SC were compared. In the RSC, 43 infants had both baseline measures (taken at <3 weeks' chronological age) and reassessment measures at 32 to 34 weeks' PMA. The mean baseline CI for the RSC was 80% and the 32 to 34 weeks' PMA CI was 77%, indicating a significant decrease in the CI over an average of 5.5 weeks (P < .0001). In the SC, the mean baseline CI of 30 infants and the mean final CI (32-34 weeks' PMA depending on when last CI measure was taken) of 24 infants were both 79% over an average study period of 5.7 weeks, indicating no significant difference between CI measures (P = .6) (Table 3).

TABLE 3 - Baseline and Final Cranial Index Measures
Baseline Final
N Mean (Range) N Mean (Range) P Value
SC 30 79 (71-87) 24 79 (71-83) .6
RSC 43 80 (72-88) 43 77 (69-88) <.0001
Abbreviations: RSC, retrospective study cohort; SC, study cohort.

Nursing Feedback

Questionnaire results from 43 primary nurses (Figure 6) were evaluated to determine percentages of favorable (frequently: 75%-100% of time) and less favorable responses (Rarely: 0%-25% of time). Nearly two-thirds or greater favorable responses regarding the use of the MPS were recorded in the following categories: “compatible with nasal cannula” (78%), “easy to adjust” (62%), “infant autonomically stable” (62%), and “infant skin integrity maintained” (76%).

Fig. 6.
Fig. 6.:
Nurse questionnaire responses.

Smaller percentages of nurses rated the MPS as favorable in the following categories: “easy to apply” (24%), “compatible with CPAP” (14%), “compatible during procedures” (13%), and “infant head position maintained” (33%). The comments section of the questionnaire provided valuable feedback to the investigators, citing specific concerns with the MPS. Ease of application concerns with regard to waking/disturbing the infant and lifting the infant's head to apply the MPS were noted in 8 nurse comments. One nurse commented that using the MPS and CPAP stocking cap together caused infant temperature to rise, and 7 nurses reported that it was difficult to secure the MPS over CPAP stocking cap. Twelve nurses noted that the MPS was usually removed for IV placement in the scalp or cranial ultrasound and was therefore less compatible during procedures. Regarding maintenance of head position, 8 nurses reported infant cervical hyperflexion and head turning within the MPS.

During daily rounds of the PI, coinvestigator, or Tortle Champions, occasional removal of the MPS was noted based on nursing evaluation of infant comfort or appropriate fit of MPS. However, no infants were removed from the study because of autonomic instability related to the use of the MPS. In every instance of MPS removal, each infant was able to tolerate the use of the MPS under that nurse's supervision after PI, coinvestigator, or Tortle champion provided education and positioning strategies to the nurse, minimizing the total time of MPS removal to no more than 24 hours at once.

Only 3 incidences of skin irritation were reported to or noted by the PI during the study period. Three infants developed forehead erythema (2 blanchable, 1 nonblanchable) directly under the Velcro closure tab during the study period that all resolved within 1 week of initial report. During this period, the infants wore the MPS as a “nest” in supine only without connecting the Velcro closures.


Overall, infants who used the MPS had better cranial molding outcomes than infants who received standard-of-care intervention for cranial molding prevention. Of the 3 infants in the SC who had dolichocephaly at 34 weeks' PMA (final CI <76% for infants completing the study period), the CI measures ranged from 73% to 75%, which are considered within normative range in a few studies.7,1516

No significant associations were found between demographic variables and comorbidities in either RSC or SC. These findings further support that GA, BW, and infant acuity are not the only variables that influence the development of cranial molding as previously thought,31 but, instead, lack of variable positioning and environment may have a greater effect.

Previous studies attempted to evaluate the effectiveness of various methods to prevent or treat cranial molding deformity by use of orthotic devices,1516 and various pillows32–34 and mattresses.3536 Pillows made of gel32 and water,3334 as well as pressure-relief mattresses,36 have not consistently demonstrated significant differences in the head shapes32,36 in very low-birth-weight infants or have had small samples sizes with inconclusive results.3334 These poor outcomes may be explained by differences in positioning of the infants' bodies, such as amount of time spent in the supine, side-lying, or prone position, which varied between studies, or simply because the devices were ineffective. Schwirian and colleagues33 reported that using a combination of water pillows and torso support resulted in improved cranial molding for healthy preterm infants less than 36 weeks' PMA who required no respiratory or feeding support, but the water pillow has not been determined to be effective in sick or very young preterm infants. Although studies have documented the phenomenon of preterm cranial molding for decades, studies testing interventions with positive outcomes used small sample sizes78,33–35 and current evidence supports the continued need for effective interventions in prevention and treatment.3,7,10,11,13,32–36

Recently, Knorr et al15 determined that a foam sleep surface called the “cranial cup” was a safe and effective device to prevent and treat cranial molding deformity in preterm infants despite early concerns regarding its safety.16 These researchers, however, called for more studies to address cranial molding treatment in extremely low-birth-weight infants and long-term outcomes associated with use of the device.1516

Based on previous studies,3–5 we hypothesized that infants with poorer respiratory function (ie, requiring CPAP) would not tolerate supine positioning with the head in midline per the study protocol based on frequent need for repositioning and on the potential for cranial reshaping by the CPAP device itself. However, infants who were on room air or nasal cannula during most of the study period had similar CI measures as infants who spent increased time on CPAP (P = .2). These findings are comparable to the findings of Knorr et al15 and DeGrazia et al,16 who did not find an association between ventilation and CPAP or ventilation and head shape in preterm infants using the cranial cup.

Infants who did not meet the recommended supine frequency (either 25% or 33% of time depending on PMA during study period) had similar CI measures as those who did meet supine frequency as recommended in the MPS positioning protocol throughout the study period (<32 weeks' PMA, P = .4; >32 weeks' PMA, P = .1). Because decreased time in the supine position would indicate a greater period in side-lying or prone positions with increased pressure on the lateral head, these results may indicate that the MPS has some head positioning benefit in side-lying and prone positions.

Reasons for infant intolerance of the supine position per nurse questionnaire comments were GER symptoms (1 comment) and apnea or bradycardia episodes (9 comments). These issues have commonly been cited as a result of supine positioning of the premature infant.3–5 When comparing cohorts, the SC had a lower incidence of BPD and GER than the RSC (Table 1). Therefore, the SC may have had a better tolerance for supine positioning as a result. Our analysis of risk factors, however, revealed a lack of association between GER and the development of dolichocephaly in both SC (P = .9) and RSC (P = .7), and a lack of association between BPD and dolichocephaly in the RSC (P = .3).13 Association between BPD and dolichocephaly could not be determined in the SC due to small incidence of BPD in this cohort.

Because the RSC did not have a uniform positioning protocol that required a specific supine frequency, our results may also indicate that the MPS positioning protocol may have increased the total amount of time that the SC spent in the supine position as compared with the RSC, even if recommended frequency of the supine position was not met by all infants in the SC. Furthermore, infants in this study may have benefited from the study positioning protocol, which provided variable head and body positioning guidelines.37 Because every infant in the SC did not meet recommended supine positioning frequency and the cohorts could not be compared, optimal frequency of time spent in supine to ensure appropriate cranial molding could not be concluded from the results of this study.

Nursing evaluation feedback was an important outcome measure to evaluate the feasibility of use in the NICU. While the MPS received favorable responses in “compatible with nasal cannula,” “easy to adjust,” “infant autonomically stable,” and “infant skin integrity maintained” categories, it is important to note potential barriers for use in the NICU. Less favorable responses were reported in the areas of “easy to apply,” “compatible during procedures,” “infant head position maintained,” and “compatible with CPAP.” Compatibility of the MPS with CPAP was poorly rated, with nurses citing difficulty securing the MPS in place over CPAP stockinette cap in the questionnaire. In addition, the PI or coinvestigator recorded MPS removal in 6 study infants due to nursing report of CPAP incompatibility. In each of these instances, the nurse was educated on a strategy to reapply the MPS at the next infant care time (usually within 3-4 hours) while addressing nursing concerns. Each infant was able to tolerate the use of the MPS at the next care time after the PI, coinvestigator, or Tortle Champion provided education and positioning strategies to the nurse. The most commonly used strategy to address this concern was to position the MPS open as a “nest” in the supine position to prevent difficulty with Velcro closure and to prevent infant overheating. Despite the reported concerns with fit of the MPS over the CPAP, infants who required CPAP at a higher frequency than others in the MPS study did not demonstrate worse cranial molding. An updated design of the MPS (currently available) that includes a strapping system compatible with a variety of CPAP devices20 would eliminate the need for wearing the MPS over the CPAP stockinette, but this new design will need to be evaluated for feasibility of use from a nursing and respiratory therapy perspective.

“Easy to apply,” “compatibility with procedures,” and “infant head position maintained” categories may demonstrate improved ratings with additional nurse training and support, as many nurses were unfamiliar with the MPS at the beginning of the study period, and this may have affected their ability to correctly apply and discretely adjust the MPS while keeping the head in alignment. Furthermore, these reported concerns including skin integrity issues (noted in 3 study infants) may demonstrate improvement based on an updated design of the MPS that includes seamless Velcro adjustment to prevent pressure points, nonslip foam lining to reduce head movement, and an open top design to decrease infant overheating and improve access to scalp for procedures and IV placement.20

Limitations of this study include the small number of infants enrolled and the nonrandomized design. There were significant differences in the number of infants with GER and BPD in the RSC compared with the SC. Based on previous study findings13 and prospective analysis of demographic information and comorbidities, however, none of these factors was associated with the development of dolichocephaly in either group. Another limitation of this study is the lack of outcome data regarding cranial asymmetry. Because no consistent objective data for this measure existed for cranial asymmetry in our retrospective study, we did not include this measure in the prospective study. Because data in the RSC were collected at variable time points between 32 and 34 weeks' PMA instead of at weekly time points as in the SC, the data points were collected differently between groups. Larger randomized cohorts should be evaluated to compare the MPS to conventional practices to treat and prevent cranial molding and asymmetry from birth to hospital discharge. Future studies should evaluate the safety and effectiveness of using this type of positioning device immediately after birth, regardless of GA, to determine the optimal time to begin intervention as well as to determine optimal supine positioning frequency. In addition, infants using the MPS should be followed up postdischarge to determine the effect on motor outcomes in early infancy.


In a small cohort of premature infants, the use of the MPS resulted in less cranial molding compared with a larger group of infants who had previously received standard of care in the same NICU. The SC using the MPS had equal mean baseline and final measures over a period of 5.7 weeks, demonstrating stable CI measures throughout the study period. Infant GA, BW, GER, time on CPAP, and time spent in the supine position did not contribute to the development of dolichocephaly. These findings indicate that the MPS may be a good option for cranial molding prevention for premature infants with a variety of demographics and comorbidities. Further studies are needed in larger cohorts of infants to confirm these findings and evaluate recommended timing for intervention with this type of MPS, as well as to determine optimal frequency of supine positioning and the potential effect on motor outcomes postdischarge.


The authors thank Jennifer Peat, MSPT, NTMC, for her valuable consultation and conception of this project. This project would not have been possible without the sustained efforts and excellent care by the NICU nursing staff, particularly those who provided valuable support and education to the NICU staff during the study period and collected outcomes data: Dana Robinson, RN, BSN, CCRN, Lindsey McMahan, MSN, CCRN, Samantha Malestenic, RN, BSN, Theresa Roach, RN, BSN, and Mary Thundathil, RN, BSN, CCRN.

The authors also thank Jennifer Edelschick, PT, DPT, PCS, Jan Fitch, PT, and Jennifer Richardson, MSPT, for their commitment to this project and valuable contributions through administrative oversight and staff education.


1. Sweeney JK, Gutierrez T. Musculoskeletal implications of preterm infant positioning in the NICU. J Perinat Neonatal Nurs. 2002;16(1):58–70.
2. Allanson JE, Cunniff C, Hoyme HE, McGaughran J, Muenke M, Neri G. Elements of morphology: standard terminology for the head and face. Am J Med Genet A. 2009;149A(1):6–28.
3. Hummel P, Fortado D. Impacting infant head shapes. Adv Neonatal Care. 2005;5(6):329–340.
4. Balaguer A, Escribano J, Roque M. Infant position in neonates receiving mechanical ventilation. Cochrane Database Syst Rev. 2006;18(4):CD003668.
5. Orenstein SR, Whitington PF. Positioning for prevention of infant gastroesophageal reflux. J Pediatr. 1983;103(4):534–537.
6. McMullen SL. Transitioning premature infants supine: state of the science. MCN Am J Matern Child Nurs. 2013;38(1):8–12.
7. McManus BM, Capistran PS. A case presentation of early intervention with dolichocephaly in the NICU: collaboration between the primary nursing team and the developmental care specialist. Neonatal Netw. 2008;27(5):307–315.
8. Cartlidge PH, Rutter N. Reduction of head flattening in preterm infants. Arch Dis Child. 1988;63(7):755–757.
9. Schreiber J, Orlin M, Palisano R. Campbell's Physical Therapy for Children Expert Consult, 5th Edition. Elsevier (HS-US); 2016:672–696.
10. Ifflaender S, Rudiger M, Konstantelos D, Wahls K, Burkhardt W. Prevalence of head deformities in preterm infants at term equivalent age. Early Hum Dev. 2013;89(12):1041–1047.
11. Elliman AM, Bryan EM, Elliman AD, Starte D. Narrow heads of preterm infants—do they matter? Dev Med Child Neurol. 1986;28(6):745–748.
12. Mewes AU, Zollei L, Huppi PS, et al. Displacement of brain regions in preterm infants with non-synostotic dolichocephaly investigated by MRI. Neuroimage. 2007;36(4):1074–1085.
13. McCarty DB, Peat J, Malcolm WF, Smith B, Fisher K, Goldstein R. Dolichocephaly in preterm infants: a retrospective investigation of the prevalence and risk factors for cranial molding deformities in the NICU. Am J Perinatol. 2017;34:372–378.
14. Ballert Orthopedic. Cranial Anthropometry: A User's Guide to Cranial Measurements Using the Ballert Cranial Caliper. Chicago, IL:Ballert Orthopedic; 2012:2–4.
15. Knorr A, Gauvreau K, Porter CL, Serino E, DeGrazia M. Use of the cranial cup to correct positional head shape deformities in hospitalized premature infants. J Obstet Gynecol Neonatal Nurs. 2016;45(4):542–552.
16. DeGrazia M, Giambanco D, Hamn G, Ditzel A, Tucker L, Gauvrea K. Prevention of deformational plagiocephaly in hospitalized infants using a new orthotic device. J Obstet Gynecol Neonatal Nurs. 2015;44(1):28–41.
17. Likus W, Bajor G, Gruszczynska K, et al. Cephalic index in the first three years of life: study of children with normal brain development based on computed tomography. Sci World J. 2014;2014:1–6.
18. Beckett JS, Pfaff MJ, Diluna M, Steinbacher DM. Dolichocephaly without sagittal craniosynostosis. J Craniofac Surg. 2013;24(5):1713–1715.
19. Ruiz-Correa S, Sze RW, Starr JR, Lin H, Speltz M. New scaphocephaly severity indices of sagittal craniosynostosis: a comparative study with cranial index quantifications. Craniofacial Journal. 2006;43:211–221.
20. “The Tortle System.” Tortle&Tortle Air for Optimal Head Positioning. Accessed March 1, 2016.
21. GELShield. GELShield: Cradle the Cranium. Invictus Medical: Improving the Lives of Newborns. Accessed March 1, 2016.
22. AliMed: Innovative Solutions. AliGel Head Positioner Donut, Neonatal. Accessed January 4, 2017.
23. Medline. Squishon Pillows by Respironics, Inc. Accessed January 4, 2017.
24. Sundance Solutions: Therapeutic Positioning in the NICU. Tortoise: Neo—Small Utility Pillow. Accessed January 4, 2017.
25. Dandle-LION Medical. Gel-filled Pillows. Accessed January 4, 2017.
26. The Shape Right System home page. Accessed January 4, 2017.
27. American Academy of Pediatrics, Committee on Fetus and Newborn. Hospital discharge of the high-risk neonate. Pediatrics. 2008;122(5):1119–1126.
28. Iyer KK, Roberts JA, Hellström-Westas L, et al. Early detection of preterm intraventricular hemorrhage from clinical electroencephalography. Crit Care Med. 2015;43(10):2219–2227.
29. Natarajan G, Pappas A, Shankaran S, et al. Outcomes of extremely low birth weight infants with bronchopulmonary dysplasia: impact of the physiologic definition. Early Human Development. 2012;88:509–515.
30. Malcolm WF, Gantz M, Martin RJ, Goldstein RF, Goldberg RN, Cotton CM. Use of medications for gastroesophageal reflux at discharge among extremely low birth weight infants. Pediatrics. 2008;121:22.
31. Baum JD, Searls D. Head shape and size of newborn infants. Develop Med Child Neurol. 1971;13:572–575.
32. Schultz AA, Goodwin PA, Jesseman C, Toews HG, Lane M, Smith C. Evaluating the effectiveness of gel pillows for reducing bilateral head flattening in preterm infants: a randomized controlled pilot study. Appl Nurs Res. 2008;21(4):191–198.
33. Schwirian PM, Eesley T, Cuellar L. Use of water pillows in reducing head shape distortion in preterm infants. Res Nurs Health. 1986;9(3):203–207.
34. Marsden DJ. Reduction of head flattening in preterm infants. Dev Med Child Neurol. 1980;22(4):507–509.
35. Cartilidge PT, Rutter N. Reduction of head flattening in preterm infants. Arch Dis Child. 1988;63(7):755–777.
36. Chan JS, Kelley ML, Khan J. The effects of a pressure relief mattress on postnatal head molding in very low birth weight infants. Neonatal Network. 1993;12(5):19–22.
37. Hemingway M, Oliver S. Bilateral head flattening in hospitalized premature infants. Online J Knowl Synth Nurs. 2000;7:3.

cranial index; cranial molding; Dolichocephaly; premature infants; preterm infants; Tortle Midliner

© 2018 Academy of Pediatric Physical Therapy of the American Physical Therapy Association