Patient 1: A 10-day-old, 24-week neonate was extubated on day 5 and transitioned to noninvasive ventilation. Positive pressure was delivered via biprong nasal interphase. On day 2 of this treatment, significant erythema was noted despite provision of a protective barrier; prongs were changed to a mask, which caused deformation, erythema, and eventual tissue necrosis at the nasal root area. Prongs were reintroduced, and despite great effort columellar injury continued to progress, necessitating endotracheal intubation to allow healing (see Figure 1).

Patient 2: A 7 month old recovering from abdominal surgery was found to have a PI at the area of the peripheral intravenous (PIV) catheter hub (see Figure 2).

Patient 3: A 2 year old recovering from respiratory failure secondary to pneumonia required a bilevel positive airway pressure mask, which was difficult to keep on and to obtain an adequate seal for appropriate oxygen saturation. This injury was discovered on day 3 (see Figure 3).

The definition of PIs is based on research on adults. Adult medical units have been concerned with these “never events” for a quite a long time, so assessment scales, prevention dogma, and (of course) treatments have stemmed from adult literature. We slowly are starting to realize that children are not small adults, and neonates are not small children; PI risk factors, sites, prevention, and treatment need to be reevaluated from a different angle.

Most PI risk assessment scales have been validated for adults; few have been adjusted to allow application to pediatric patients, and only 1 or 2 specifically for the neonatal population. The Neonatal Skin Risk Assessment Scale (NSRAS), the Braden Q, Braden, Starkid, and Glamorgan scales are a few of the better known risk assessment tools for potential pediatric/neonatal use.¹ Their sensitivity and specificity are not optimal; NSRAS, Starkid, and Braden Q have not been validated in neonates, children older than 8 years of age, or patients with cardiac disease and do not account for the most common risk factor — medical devices. The Glamorgan instrument has not been validated in the neonatal population and is not commonly used in the United States. In 2018, the Braden QD, a revised and simplified version of Braden Q, was introduced and included preterm neonates (109 babies, gestation age may not be stipulated) to persons up to 21 years of age. The assessment population included intensive care patients and persons with multiple affected systems (eg, cardiac patients); therapeutic and diagnostic devices are part of the consideration. This new scale had 7 categories, 2 of which are device-related, with scores for each item from 0 to 2 (up to 8 devices can be counted in). The higher the score, the higher the risk. At a cutoff score of 13, the Braden QD has sensitivity and specificity of 86% and 59%, respectively, similar to the Braden Q. Given this scale is newly developed, I am not sure how many hospitals have started utilizing this scale and what the results will be. 

Currently, my neonatal unit uses the NSRAS and our pediatric units use the Braden Q/Braden, but often scale scores do not translate into actionable risk-reduction activity. In conversation with other neonatal units, I found they do not use any of these scales, believing their use contributes little to effective prevention or identification of patients at risk. I have to agree: if no clear action items are linked to a specific subscale score (the majority of the neonatal/pediatric units in the country do not have such algorithms or guidelines), and based on questionnaires, the nurse’s understanding of clinical risk factors and the risk each device presents produces better prevention and care than a scale score. 

Medical device-related injury (MDRPI). The traditional cause of PI (pressure exerted over the bony prominence) touches only the tip of “etiology iceberg” in the pediatric population. The majority of pediatric PIs (between 50% and 90%) are MDRPIs, and as such may not be over bony prominence.² Common medical devices include peripheral vascular catheters, central lines, tracheostomy systems, endotracheal tubes, pulse oximetry devices, and orogastric and gastrostomy tubes. Respiratory devices cause the majority of MDRPI — specifically, noninvasive ventilation interphases. Nasal prongs/masks often are placed tightly to deliver adequate pressure. Children may attempt to remove the apparatus, while neonates often are placed prone; this increases friction between the skin and the device.  

Anatomy. The uniqueness of neonatal skin contributes to PI: lack of a robust stratum corneum, thin dermis, poor epidermal/dermal adhesion, lack of subcutaneous tissue, and often poor end-vascularization potentiates the damage. The unique anatomy of the neonatal columella area, need for precise fit to deliver distending pressure, and poor tolerability of repositioning contributes to ongoing pressure on the nares, septum, and sometimes forehead/nasal root area. The columella area houses the bony maxillary spine, surrounding sparse vessels, and very little subcutaneous and dermal tissue to offload transmitted pressure. Thus, transmitted pressure may lead to deformation of skin layers, contributing to poor blood flow, ischemia, inflammation, and cell death. 

PI progression. Injured cells undergo reperfusion injury within a few hours of the original insult, which may worsen the damage in the next few days even as circulation returns. Damage can be caused by ischemia in just 2 hours. In addition, the microclimate comprising heated and humidified air delivered by continuous positive airway pressure (CPAP) apparatus places nasal skin at increased risk for injury. Every 1˚ C increases oxygen consumption by 10% and increases metabolic demands; it also weakens the intercellular cytoskeleton’s connections, reducing the tolerance for deformation. 

Pain can be difficult to assess in a neonate or young, sick child. 

No robust guidance exists with regard to the management/prevention of neonatal and pediatric skin injury. 

According to the National Pressure Ulcer Advisory Panel in conjunction with European and Pan-Pacific peer organizations,³ anyone provided medical equipment should be deemed a high-risk patient. Strategies to reduce risk include correct fit and positioning, frequent assessments, keeping skin dry and clean, applying a nonirritant barrier (dressings that do not gel are recommended to avoid skin stripping or maceration), and clinician awareness of the tubing and the position of the supporting structure. Researchers have supported frequent assessments, repositioning, nursing education, and barriers as preventive methods. Many researchers have concluded nursing care is paramount to injury prevention. Yet despite the importance of nursing vigilance and expertise, studies do not support its efficacy in fully eliminating PI.⁴ 

The literature supports the success of pressure-relieving barriers in eliminating PI from tracheostomy, gastrostomy, PIV, miscellaneous tubing, and noninvasive respiratory devices. Hydrocolloid and foam barriers represent the most commonly used products. Soft, thin silicone or polyurethane foam is a product of choice for many (over bony prominence, to offload extracorporeal membrane oxygenation [ECMO] cannulas, under gastrostomy tubes, and under tracheostomy ties and flanges),⁵ and it is a commonly used barrier under CPAP masks. Others have used hydrocolloid dressings to decrease friction and sheer. Both options have drawbacks when used under noninvasive ventilatory devices, especially in humidified isolettes (hydrocolloids increase the risk of epidermal stripping, they are difficult to reposition, and not recommended to remove frequently; foam may not stay in place, become moist, or prevent appropriate fit).

Device-specific options. Weng at al⁶ reported decreased PI when hydrocolloid and film dressings were used (versus none). Hsu⁷ suggested caution when using a hydrocolloid because epidermal stripping was prevalent during dressing changes, and a sticky residue remained, especially in humidified environment. Hsu⁷ noted the lowest injury rate when soft silicone was used, followed by hydrocolloid and no barrier. Gunlemez et al⁸ had success using a silicone gel sheet at the nasal surface to alleviate direct pressure. My unit found the use of Neoseal (Neotech, Valencia, CA) foam a reasonable barrier for various biprong respiratory interphases, as well occasional use of a hydrocolloid for lip and cheek areas, and Mepilex Lite foam (Mölnlycke Health Care, Norcross, GA) worked well with mask devices. We use Mepilex Lite foam as a prophylactic barrier with tracheostomies, casts, and gastrostomies (if erythema is noted) and to offload catheters. A hydrocolloid sheet also can be used if very thin material is required. We offload all PIV hubs with sterile stretchable foam tape strips (Veni-Gard kits; Con-Med, Utica, NY). We use nonalcohol skin polymers if a hydrocolloid is used, and adhesive dressings are always removed using a silicone-based dressing remover.

Care bundles. A collection of processes can be combined to effectively and safely address a particular condition or prevent one from occurring. Consistent delivery of combined interventions has been shown to improve reliability of a process; a PI prevention bundle — a collection of actionable items required for every patient at risk — should be an integral part of PI program. Since 2012, many pediatric hospitals participate in The Children’s Hospitals Solutions for Patients Safety National Children’s Network, a group of 135+ hospitals that focus on sharing their efforts in preventing hospital-acquired harm, including PI, by implementing a PI bundle. 

In my hospital, we have a general PI bundle (Actionable items: Assessment scale/Device Assessment/Reposition/Patient Turning/Repositioning/Bed Surface/Moisture/Nutrition), an ECMO-specific PI bundle, and a Respiratory Devices PI Prevention bundle. It is imperative that all components are followed. The literature supports that if nurses do not know about PI prevalence, do not understand importance of PI prevention, believe it is time-consuming or that many PIs are not avoidable, or in general are less interested in PI prevention than other aspects of nursing care, they will not follow bundle actions. Extensive nursing education highlighting prevention and recognition of PI is paramount for bundle compliance. A culture of “zero tolerance of PI” should be promoted, highlighting preoccupation with prevention and nontolerance of complacency. This culture change remains the most critical yet hardest component of any successful PI prevention program.