Therapeutic Hypothermia: Treatment for Hypoxic-Ischemic Encephalopathy in the NICU

Denise M. Casey, RN, MS, CCRN, CPNP Nancy Tella, RN, BSN, CCRN Rachel Turesky, RN, BSN Michelle Labrecque, RN, MSN, CCRN

Case Study admission, arterial access was established for hemodynamic Baby M was born limp, blue, and without respiratory effort at monitoring and ease of lab drawing purposes. Additional anti- 38 weeks gestation to a 38-year-old, gravida 5, para 1, woman. convulsants were given to control her seizures. Baby M’s father Delivery was vaginal after a rapid progression of labor leaving had spent the night going back and forth between our NICU no opportunity for a cesarean section. No other complications and the birth hospital because both his wife and newborn were noted during labor but a large surge at delivery, later diag- infant were quite sick. When he came in later that morning nosed as uterine rupture, initially raised concerns about placen- with Baby M’s adult stepsiblings, explanations were given for tal abruption. Apgar scores were 1, 2, and 4 at one, five, and ten each of the machines, her course of treatment up to that point, minutes, respectively. She was resuscitated in the delivery room, and our concerns about her neurologic status. Baby M’s father intubated, and transferred in critical condition to the neonatal tried to console his children despite his own fears and sadness. intensive care unit (NICU) at the birth hospital. Her initial During the next several days, Baby M remained intubated cord pH was 6.7 and was slightly improved at 7.17 on arterial for airway protection, was fluid restricted to prevent addi- blood gas after resuscitation. Our NICU team was consulted tional injury to her brain, and continued on anticonvulsants. because of her severe neurologic depression. The birth hospital Despite her traumatic delivery, Baby M remained stable from was within walking distance of our tertiary care center and our a cardiovascular standpoint without evidence of persistent neurologists went to evaluate her for the hypothermia protocol. pulmonary hypertension of the newborn (PPHN) or need Her neurologic exam was notable for dilated and unresponsive for pressors. Her kidneys were affected by the hypoxic events pupils, no spontaneous movements, and diminished reflexes at birth as evidenced by poor urine output and electrolyte and tone, consistent with moderate-to-severe encephalopathy. disturbances requiring further fluid restriction and electro- Seizure activity began at one hour of age and consisted of lip lyte boluses. Her liver enzymes were mildly elevated, but she smacking, which was later confirmed by electroencephalogram showed no evidence of coagulopathy. She was hemodynami- (EEG). Enrollment criteria were met based on respiratory cally stable after an airway was established. depression at birth requiring intubation and continued need She completed the hypothermia protocol after 72 hours for ventilation, concern for placental abruption, cord pH less of cooling and began the rewarming process. EEG tracings than 7, and encephalopathy on exam and EEG. After stabiliz- obtained on day of life (DOL) 5 showed no evidence of seizure ing her airway and achieving central access to treat acidosis and activity. Fosphenytoin was discontinued and she remained on seizures, the team prepared her for transfer to our NICU. At phenobarbital alone. A magnetic resonance imaging (MRI) this point, the primary concern became her neurologic status. of the brain done on the same day showed minor changes in Baby M was admitted to our NICU for whole-body hypo- the occipital cortex, but no significant abnormalities. thermia at four hours of age with a diagnosis of hypoxic-ischemic encephalopathy (HIE). Therapeutic hypothermia using whole- Disclosure body cooling was initiated within six hours of birth per pro- The author discloses no relevant financial interest or affiliations with any tocol with the Cincinnati Subzero Blanketrol II. Shortly after commercial interests.

Accepted for publication February 2011.

N EONATAL NETWORK 370 © 2011 Springer Publishing Company NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6 http://dx.doi.org/10.1891/0730–0832.30.6.370 Copyright © Springer Publishing Company, LLC As she recovered and her sedation was lightened, she a biphasic process. The initial phase of hypoxic-ischemia results began to breathe and was extubated to room air on DOL 6. in a primary brain energy failure in which there are reductions Intravenous fluids were liberalized and nasogastric feedings in cerebral blood flow, oxygen, high-energy phosphorylated were initiated on DOL 7. Her neurologic exam continued to metabolites, and brain acidosis.5,6 This phase is associated with stabilize. Her pupils became reactive, suck and gag reflexes intracellular derangements such as loss of membrane homeosta- were present, and she began to move spontaneously. sis, defective osmoregulation, and inhibition of protein synthesis. She was transferred back to her birth hospital on DOL 8 to Loss of membrane homeostasis can lead to increases in intracel- continue to recuperate with her mother. At that point, she was lular calcium and osmotic dysregulation.7 Elevated calcium levels tolerating full nasogastric feedings and was consistently inter- then trigger many destructive pathways.8 Approximately 6–12 acting with her family. Additional MRI was obtained on DOL hours after the initial insult, a secondary energy failure occurs 12 and the subtle abnormalities seen earlier were no longer leading to sustained brain injury. This phase typically is without apparent. An EEG done the same day showed no evidence of brain acidosis. Secondary energy failure is a marker of the begin- seizure activity. She was discharged to home with her parents ning of multiple pathways that lead to brain injury. The processes on DOL 15 on full oral feedings and off phenobarbital. She that take place within this phase are inflammation, apoptosis, was followed in the neurology clinic every three to six months oxidative injury, decrease growth factors, and protein synthesis. for the first 24 months of life. At the last scheduled visit, Baby Because the cerebral energy state can be restored after the primary M appeared to be growing well and meeting her developmen- energy failure, it suggests there is a therapeutic window to imple- tal milestones with no apparent neurologic deficits. ment interventions to avoid or diminish the secondary energy failure leading to brain injury.6 During this biphasic process, the INTRODUCTION therapeutic window to implement interventions to reduce brain Hypoxic-ischemic encephalopathy (HIE) is defined as an injury is during the 6–12 hour window prior to the secondary interruption in the supply of oxygen (hypoxia) and/or blood energy failure. A temperature reduction of 2°–4°C decreases the flow (ischemia) going to the brain and body. This kind of rate of cell death and delays the cascade of metabolic changes interruption occurring either hours before birth or during that occur with hypoxia. Cerebral metabolism is reduced and labor and delivery can happen for several reasons such as com- hypothermia can delay secondary brain injury in HIE infants.9 pression of placenta, tearing of placenta from uterine wall, or compression of the cord. REVIEW OF EVIDENCE HIE occurs in 1/1,000 term live births and remains an In 2005, two large clinical trials of 473 infants demon- important cause of mortality and neurodevelopmental deficits strated that therapeutic hypothermia reduced the risk of in infants.1 Moderate encephalopathy carries a 10 percent risk death/disability in neonates with HIE.1,2 The major differ- of death and 30 percent risk of disability, whereas 60 percent ence between these two trials was whole-body cooling versus of those patients with severe encephalopathy die and many, if head cooling only. The Shankaran and associates trial looked not all who survive, have neurologic deficits.2 The Sarnat and at whole-body cooling with core temperature of 33.5°C by Sarnat system for neonatal encephalopathy guides the classifi- esophageal probe for 72 hours followed by a rewarming phase. cation based on Stages 1 (mild), 2 (moderate), and 3 (severe).3 There were 239 infants enrolled in this study and the inclusion The stages are classified based on the distinguishing features criteria were infants $36 weeks, admitted less than six hours shown in Table 1. Until 2005, there was no other treatment after birth, severe acidosis or perinatal complications, resuscita- for HIE other than conventional intensive care. tion at birth, and moderate-to-severe encephalopathy (clinical HIE results from a lack of oxygen and blood supply that exam only). The neurodevelopmental outcomes were assessed leads to metabolic acidosis, ischemia, and subsequently neuro- at 18–22 months of age and showed a reduced risk of death/ logic dysfunction.4 Brain injury that ensues is characterized by disability in infants with moderate-to-severe encephalopathy.2 The Gluckman and associates trial was a head cooling TABLE 1 n Sarnat Staging of Encephalopathy study and core temperature was 34°–35°C by rectal probe for Sarnat Stage 1 Sarnat Stage 2 Sarnat Stage 3 72 hours followed by a rewarming phase. The sample size was (Mild) (Moderate) (Severe) 235 term infants. The inclusion criteria for this study were term Hyperalert Lethargic Stuporous infants with moderate-to-severe encephalopathy determined Normal tone Mild hypotonia Flaccid by clinical exam and an abnormal amplitude-integrated electroencephalogram (aEEG). The outcomes revealed that head Overactive stretch Overactive stretch Decreased or absent reflexes reflexes stretch reflexes cooling was not protective in a mixed population of infants Weak suck Weak or absent suck Absent suck with neonatal encephalopathy. The study did show there was improved survival without severe neurodevelopmental dis- No seizures Common; focal or Uncommon 1 multifocal ability in infants with less severe aEEG changes. An article published in 2009 by Azzopardi looked at neu- Less than 24 hours 2–14 days Hours to weeks rodevelopmental outcomes in infants at least 36 weeks of age

N EONATAL NETWORK VOL. 30, NO. 6, NOVEMBER/DECEMBER 2011 371

Copyright © Springer Publishing Company, LLC with perinatal asphyxia and found that infants in the cooled TABLE 2 n Whole-Body Cooling Inclusion Criteria group had an increased rate of survival without neurologic Evidence of fetal distress as evidenced by one of the following: abnormality. Among survivors, cooling resulted in reduced • History of acute perinatal event (abruptio placentae, cord prolapse, 10 risks of cerebral palsy and improved developmental scores. variable or late decelerations) Based on the 2005 studies by Gluckman and associates and • Biophysical profile ,6/10 within 6 hours of birth Shankara and colleagues, the National Institute of Child Health • Cord pH #7.0 or base deficit $16 mEq/L and Human Development (NICHD) held a workshop on hypo- and 1,2,11 thermia as a treatment for HIE. Experts summarized the Evidence of neonatal distress as evidenced by at least one of the evidence on hypothermia and concluded that mild therapeutic following: hypothermia offered a potential for short-term benefits when • Apgar score #5 at 10 minutes used under strict protocols. In addition, they stated, “therapeutic • Postnatal blood gas pH at ,1 hour #7.0 or base deficit $16 mEq/L hypothermia is an evolving therapy and the long-term safety and • Continued need for ventilation initiated at birth and continued for at efficacy are yet to be established.”11 They offered a framework least 10 minutes for patient care with standardized protocols to continue to deter- and mine the safety and efficacy of this therapy while also suggest- Evidence of neonatal encephalopathy by physical exam (neurologist) ing hypothermia registries as a way to monitor, develop, refine, and 11 Abnormal cerebral function monitor (CFM) with minimum 20 minutes and optimize this new therapy. One registry that has been of recording developed in response to this suggestion is the Vermont Oxford Network Registry for Neonatal Encephalopathy. The Vermont Adapted from: Shankaran, S., Laptook, A. R., Ehrenkranz, R. A., Tyson, J. E., McDonald, S. A., Donovan, E. F., . . . Jobe, A. H. Oxford Network was established in 2006. Hospitals enrolled (2005). Whole-body hypothermia for neonates with hypoxic- in the Vermont Oxford Network are able to participate as long ischemic encephalopathy. The New England Journal of Medicine, as they have internal review board (IRB) approval. The 4,464 353(15), 1574–1584. records of neonatal encephalopathy have been received since the beginning of the registry. The 1,797 infants have received either criteria: gestational age $36 weeks and birth weight .2,000 g; selective head cooling or whole-body therapeutic hypothermia. evidence of fetal distress; evidence of neonatal distress; evidence After an evidence-based review, our NICU developed a of neonatal encephalopathy; and abnormal cerebral function therapeutic hypothermia protocol in 2007. The two large ran- monitoring (CFM), which is a compressed three-lead EEG that domized clinical trials as well as the NICHD workshop rec- provides information on global cerebral activity and is evaluated ommendations formed the basis for our protocol. A team from by the consulting neurology team (Table 2). Once the patient is the NICU including neonatology, neurology, and clinical nurse determined to meet all of the inclusion criteria, the therapeutic specialists all met to discuss the current evidence and, using the hypothermia protocol is initiated. Prior to admission, the nec- Gluckman and associates and Shankaran and colleagues trials essary equipment is set up including the cooling unit, cooling as references, developed a total body hypothermia protocol blanket, esophageal temperature probe, and infant warmer, with for treatment of HIE.1,2 The protocol included guidelines for the heat off, but allowing skin temperature reading. Other nec- cooling and maintenance as well as rewarming phases. As we essary supplies are organized to ensure accuracy and readiness continue to gain experience with this new treatment, the proto- upon the infant’s arrival. col will be updated to use evidence-based research as it becomes available and to maintain consistency with current practice. Protocol Initiation Our NICU has evaluated 59 patients for the therapeu- On admission, the infant is placed on a radiant warmer in tic hypothermia protocol since initiation in 2007. Forty-two manual mode with the heat o f f . The attending neonatologist, patients have started the therapy and 32 patients have completed nurse practitioner, and the admitting nurse assess the infant, the 72-hour cooling period. Ten patients have been taken off and a neurology consult is obtained at the bedside. In collabo- the protocol for the following reasons: six withdrawal of care ration with neurology, the attending neonatologist confirms (removing life support), three abnormal coagulation levels, and that the infant is a candidate for therapeutic hypothermia based one for encephalopathy not caused by HIE. Twelve of these on clinical exam, inclusion criteria, and a 20-minute aEEG.1 patients had EEG-confirmed seizures. We had a total of eight The therapeutic hypothermia cooling procedure checklist and deaths. All surviving patients are followed by our neurology clinical practice guidelines are available at the bedside for ref- clinic on a regular basis to monitor for long-term outcomes. erence and are used throughout the care of the patient (see Figure 1). An esophageal temperature probe is placed by the NURSING CARE HIGHLIGHTS nurse to monitor core temperature, and the infant is placed OF A PATIENT UNDERGOING directly on the cooling blanket. In the cooling and mainte- THERAPEUTIC HYPOTHERMIA nance phase of therapy, the goal esophageal temperature is The timing for therapeutic hypothermia is crucial, optimally it 33.5°C with a range of 32.5°–34.5°C.2 Careful monitoring of is initiated within six hours of birth.2 To be considered for cooling the infant undergoing therapeutic hypothermia is a primary infants with presumed HIE, must meet the following inclusion focus of nursing care once hypothermia is initiated. Each

N EONATAL NETWORK 372 NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6

N EONATAL NETWORK VOL. 30, NO. 6, NOVEMBER/DECEMBER 2011 373

Copyright © Springer Publishing Company, LLC infant is monitored closely and for the purposes of this article, to maintain adequate blood pressure. Overall, tone is usually we have broken each of the monitoring categories out into low with little to no spontaneous movement and, frequently, systems to best describe the parameters individually. seizure activity is noted. Seizures are typically of early onset and occur in approximately 43–56 percent of patients.2,10 Systemic Effects Patients with moderate-to-severe HIE often appear sedated, Neurologic pale, and hypotonic because of the brain injury they have sus- Because of the potential for evolving injury, a neurologic tained. They usually require ventilator support to maintain ade- exam is, at a minimum, performed hourly by the infant’s nurse. quate oxygenation and ventilation. Peripheral perfusion is often Together with the medical team, those caring for the infant are poor, and the infant may require volume and inotropic support constantly monitoring for any changes in neurologic status. In

N EONATAL NETWORK 374 NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6

Copyright © Springer Publishing Company, LLC these infants, neurologic assessment should include pupil checks; Respiratory evaluation of level of consciousness and respiratory insufficiency; Infants receiving cooling are typically intubated, but some and evaluating for any evidence of seizures, signs of apnea/brady- patients are able to be extubated during the cooling phase. It is cardia, as well as signs of increased intracranial pressure.12 CFM important to note esophageal temperature on the requisition monitoring is continued throughout the 72-hour cooling period when arterial blood gases are sent to allow for temperature- with readings evaluated by the neurology service because it can corrected results.2 Hypothermia shifts the oxyhemoglobin provide information on duration, intensity, and frequency of curve and can result in a decreased oxygen delivery, but the neonatal seizures, and a full 20-lead EEG and MRI are arranged metabolic rate is also lowered, decreasing oxygen consumption at the discretion of the neurology service. An early MRI may be and carbon dioxide production. As a result of these reactions necessary for those patients with severe HIE to obtain additional to cooling, ventilator settings may require frequent adjust- information that may guide end-of-life discussions. The cooling ment using the temperature-corrected arterial blood gas.16 blanket is not compatible with MRI. To prevent rewarming, the infant should remain on the cooling blanket for as long as pos- Cardiovascular sible, and the nurse caring for the patient should coordinate with The nurse closely monitors the infant’s color and perfusion, radiology to minimize any wait time off the blanket. and watches for bradycardia and arrhythmias. The arrhythmia most likely seen with therapeutic hypothermia is a sinus bra- Sedation dycardia.10 This is a very frequent finding during therapeutic Infants undergoing hypothermia treatment in our NICU hypothermia and a normal physiologic response occurring with are typically sedated with a lower dose of fentanyl from the time most hypothermia patients; the infant’s perfusion is closely of initiation of cooling until the rewarming process is complete. monitored and the infant is treated individually as needed. The goal of sedation during the cooling period is to optimize Because of the initial hypoxic insult, these infants typically comfort and the efficacy of therapeutic hypothermia. Inadequate require volume resuscitation and initiation of inotropic support. sedation may result in an increased metabolic rate as the infant attempts to warm himself or herself, therefore decreasing the FIGURE 2 n Subcutaneous fat necrosis 1 and 2. effectiveness of the therapy. In our NICU, we use the State Behavioral Scale (SBS) to guide our sedation management with a goal of 21 to 22 during treatment. The fentanyl dosage is adjusted accordingly to maintain our target. Patients within this range are responsive to noxious stimuli and to a gentle touch or voice.13 In reviewing the literature on HIE, there was only one article that mentioned using sedation with morphine or chloral hydrate if the infants appeared distressed.10

Fluid and Electrolytes These patients are at risk for multiple electrolyte imbal- ances and do require frequent monitoring and correction based on laboratory values. Fluid restriction is anticipated with these patients to avoid fluid overload and thus cerebral edema. Maintaining sodium levels in the upper limits of normal also is important because these patients are at risk for cerebral edema.14 Serum magnesium levels are also maintained in the upper limits of normal because of its potential neuroprotective effect.15 This “osmotherapy” aids in improv- ing intracranial elastance and compliance by causing water to flow from the brain’s extracellular compartment into the vasculature, which decreases the intracranial volume.14

Laboratory Monitoring Ideally, arterial and venous access is obtained prior to initiation of cooling to allow ease of lab draws, continuous blood pressure monitoring, and optimal medication administration. Baseline labs are obtained with follow-up monitoring based on clinical indication and protocol. Our protocol requires that labs be performed upon admission, and at 4, 8, 12, 24, 48, and 72 hours.

N EONATAL NETWORK VOL. 30, NO. 6, NOVEMBER/DECEMBER 2011 375

N EONATAL NETWORK 376 NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6

N EONATAL NETWORK VOL. 30, NO. 6, NOVEMBER/DECEMBER 2011 377

(continued)

N EONATAL NETWORK 378 NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6

Copyright © Springer Publishing Company, LLC Thoresen recommends using the same treatment principles for (p. 257).18 Subcutaneous fat necrosis is rare but patients with all asphyxiated infants: first, correct the hypovolemia, then use HIE are at risk for developing this condition because of hypoxia inotropes (dopamine or dobutamine), depending on echocar- and induced hypothermia. Of the 42 infants we have treated in diogram and whether the cardiac contractility is normal or our NICU with whole-body cooling, two developed some form abnormal.17 of subcutaneous fat necrosis. These infants are presented with areas of redness, ill-defined erythema, bruise-like appearance, Hematology and the skin looked taut or shiny (Figure 2). Coagulopathy may be induced by hypothermia, and a platelet count of greater than 100,000 should be maintained Rewarming to compensate for decreased platelet function. Patients on At the completion of 72-hour cooling period, rewarming this protocol may require transfusions of fresh frozen plasma, is initiated and takes place over a period of 10 hours.1 The cryoprecipitate, and platelets. esophageal probe remains in place until the patient returns to standard care to allow for continuous monitoring through the Infection rewarming process (Figure 3). During rewarming, the nurse Blood cultures are obtained. Antibiotics are continued for should monitor for seizures and hypotension.17 Following the a minimum of 72 hours during the cooling period to provide cooling period, blood work is done every 24 hours during the prophylaxis in the setting of relative immune dysfunction. At rewarming process; particular attention is paid to serum potas- the completion of 72 hours of cooling, antibiotics are discon- sium levels because hyperkalemia may occur during rewarm- tinued if there are no obvious signs of infection. ing and as needed based on the infant’s clinical status.

Skin Family Support Infants are repositioned every two hours, and a skin assessment Support for the infant’s family begins in the delivery room is performed every four hours. Skin is monitored for color, per- and continues throughout the hospitalization. It is a very fusion, skin breakdown, or signs of subcutaneous fat necrosis.2–4 difficult time for families who are appropriately shocked by Subcutaneous fat necrosis is “characterized by induration, ery- the events surrounding the delivery of their infant and are thematous nodules and plaques over bony prominences such as worried about their child’s outcome. Once an infant is iden- back, arms, buttocks, thighs and cheeks of full term newborns” tified as a potential candidate for the hypothermia protocol,

N EONATAL NETWORK VOL. 30, NO. 6, NOVEMBER/DECEMBER 2011 379

Copyright © Springer Publishing Company, LLC the attending neonatologist talks to the infant’s parents to for neonates with hypoxic-ischemic encephalopathy. The New England explain the treatment and potential risks, and obtains consent Journal of Medicine, 353(15), 1574–1584. for treatment. A parent information guide to hypothermia 3. Sarnat, H. B., & Sarnat, M. S. (1976). Neonatal encephalopathy following treatment is also given to parents (Figure 4). In our NICU, fetal distress. A clinical and electroencephalographic study. Archives of parents have unlimited visitation 24 hours a day and are kept Neurology, 33(10), 696–705. updated on their infant’s condition and plan. An open envi- 4. Verklan, M. T., & Walden, M. (Eds.). (2004). Neurologic disorders. In Core curriculum for neonatal intensive care nursing (3rd ed., ronment of communication is maintained with parents and pp. 850–852). St. Louis, MO: Elsevier Saunders. they are encouraged to ask questions as they arise. Family 5. Lorek, A., Takei, Y., Cady, E. B., Wyatt, J. S., Penrice, J., Edwards, meetings are scheduled as needed and include parents, A. D., . . . Reynolds, E. O. R. (1994). Delayed (“secondary”) cerebral nursing, neonatology, neurology, and social work staff. Less energy failure after acute hypoxia-ischemia in the newborn piglet: formal updates also take place continually at the bedside. Continuous 48-hour studies by phosphorus magnetic resonance It is important for the team to maintain an understanding spectroscopy. Pediatric Research, 36(6), 699–706. 6. Laptook, A. R. (2009). Use of therapeutic hypothermia for term infants of the difficulties facing parents of a baby with HIE. Most with hypoxic-ischemia encephalopathy. Pediatric Clinics of North of these parents have encountered unexpected circumstances America, 56(3), 601–616. around the delivery of their child. Many of their expectations 7. Johnston, M. V., Trescher, W. H., Ishida, A., & Nakajima, W. (2001). about the birth of their baby have been shattered and they are Neurobiology of hypoxic-ischemic injury in the developing brain. understandably in crisis. The parents are likely worried that their Pediatric Research, 49(6), 735–741. infant may not survive, and as in Baby M’s case, the mother of 8. Siesjö, B. K., & Bengtsson, F. (1989). Calcium fluxes, calcium antagonists, the baby may also be critically ill. Parents are often concerned and calcium-related pathology in brain ischemia, hypoglycemia, and spreading depression: A unifying hypothesis. Journal of Cerebral Blood about their baby’s long-term neurologic outcome, and this can Flow and Metabolism, 9(2), 127–140. be an overwhelming concern for them. These feelings are com- 9. Perlman, J. M. (2006). Intervention strategies for neonatal hypoxic- pounded by the fact that they are unable to hold their baby, and ischemic cerebral injury. Clinical Therapeutics, 28(9), 1353–1365. they may be further frightened by the infant’s pale and cold 10. Azzopardi, D. V., Strohm, B., Edwards, A. D., Dyet, L., Halliday, H. appearance. Providing explanations and discussing concerns L., Juszczak, E., . . . Brocklehurst, P. (2009). Moderate hypothermia to can be helpful in reducing parental anxiety and in helping them treat perinatal asphyxial encephalopathy. The New England Journal of Medicine, 361(14), 1349–1358. to maintain some sense of control over a very stressful situation. 11. Higgins, R. D., Raju, T. N., Perlman, J., Azzopardi, D. V., Blackmon, L. Encouraging involvement in their infant’s care when feasible R., Clark, R. H., . . . Wyatt, J. (2006). Hypothermia and perinatal asphyxia: will reinforce their role as parents and as members of the team. Executive summary of the National Institute of Child Health and Human It is our goal and expectation to support both the baby and Development workshop. The Journal of Pediatrics, 148(2), 170–175. parents during this difficult and challenging time. 12. Volpe, J. J. (2001). Hypoxic-ischemic encephalopathy: Clinical aspects. In Neurology of the Newborn (4th ed., pp. 331–385). Philadelphia, PA: W. B. Saunders. CONCLUSION 13. Curley, M. A., Harris, S. K., Fraser, K. A., Johnson, R. A., & Arnold, Hypothermia is an evolving new therapy that requires strict J. H. (2006). State Behavioral Scale: A sedation assessment instrument protocols to safely and effectively care for infants with HIE. for infants and young children supported on mechanical ventilation. The nursing care is rigorous for these infants and the moni- Pediatric Critical Care Medicine, 7(2), 107–114. toring is quite specific. The collaborative effort among all dis- 14. Raslan, A., & Bhardwaj, A. (2007). Medical management of cerebral ciplines is essential for seamless delivery of care. Because this edema. Neurosurgical Focus, 22(5), E12. is a new therapy, there is a need for further research regarding 15. Crowther, C. A., Hiller, J. E., Doyle, L. W., & Haslam, R. R. (2003). Effect of magnesium sulfate given for neuroprotection before preterm long-term neurodevelopmental outcomes for infants having birth: A randomized controlled trial. The Journal of the American Medical undergone this therapy. Further research needs to be under- Association, 290(20), 2669–2676. taken on the sedation management of these infants because 16. Polderman, K. H. (2004). Application of therapeutic hypothermia in the this has not been studied thoroughly in this infant popula- intensive care unit. Opportunities and pitfalls of a promising treatment tion. We must also be cautious in our explanations to families modality—Part 2: Practical aspects and side effects. Intensive Care regarding what this treatment may be capable of for their Medicine, 30(5), 757–769. 17. Thoresen, M. (2008). Supportive care during neuroprotective infant, balancing honesty and sensitivity. hypothermia in the term newborn: Adverse effects and their prevention. Clinics in Perinatology, 35(4), 749–763. REFERENCES 18. Tran, J. T., & Sheth, A. P. (2003). Complications of subcutaneous fat 1. Gluckman, P. D., Wyatt, J. S., Azzopardi, D., Ballard, R., Edwards, A. necrosis of the newborn: A case report and review of the literature. D., Ferriero, D. M., . . . Gunn, A. J. (2005). Selective head cooling with Pediatric Dermatology, 20(3), 257–261. mild systemic hypothermia after neonatal encephalopathy: Multicentre randomised trial. Lancet, 365(9460), 663–670. For further information, please contact: 2. Shankaran, S., Laptook, A. R., Ehrenkranz, R. A., Tyson, J. E., McDonald, Denise M. Casey, RN, MS, CCRN, CPNP S. A., Donovan, E. F., . . . Jobe, A. H. (2005). Whole-body hypothermia E-mail: [email protected]

N EONATAL NETWORK 380 NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6

Sheila Malusky, DNP, RN, NNP-BC Ann Donze, MSN, RN, NNP-BC

n the United States each year, approximately 57,000 injury in the individual and the financial burden to care for Iinfants are born prematurely. With the advancement of these babies. The impact can also be devastating to the family neonatal medicine during the who has the responsibility of past several decades, including caring for a disabled child. The Ab s t r a c t improved methods of mechani- lifelong commitments to care for cal ventilation and the devel- With the advancement of neonatal medicine during the these individuals tax the family past several decades, premature and critically ill infants opment of total parenteral structure, family resiliency, and are living past the neonatal period and surviving. The nutrition (TPN) for neonates, survival of these infants at smaller birth weights and bring about the additional need even extremely low birth weight younger gestational ages puts them at an increased risk for for community support. (ELBW) infants are now living intraventricular hemorrhages (IVHs). Although shifts in There has been a multitude longer and surviving.1 Of these cerebral perfusion have been linked to the development of of research into the prevention infants, 20–25 percent of them these brain bleeds, many seemingly benign care activities of IVH in premature infants.6 will develop an intraventricular have been linked to changes in cerebral blood flow patterns, Some of these studies examine hemorrhage (IVH),2 with the possibly contributing to IVHs. The purpose of this article is prenatal factors such as antenatal incidence being inversely pro- to evaluate the current evidence to determine if the practice steroid use.7 Other studies have portional to gestational age.3 of midline positioning for infants born less than 32 weeks focused on antenatal factors such The total financial cost that gestation for possible IVH prevention is supported by the as delivery room resuscitation literature. Many of the researchers involved in these studies is estimated for these premature methods.8 Still other studies attributed the consequential venule leakage of blood to births is $26 billion or $51,600 occlusion of the jugular venous drainage system following have focused on neonatal preven- for each individual prema- a turn in the position of the head. Additionally, the articles tion methods such as pharmaco- ture birth. These costs include that examined the connection between the effects of head logic interventions and neonatal medical care, delivery costs, early tilting on brain hemodynamics attributed changes on the care management methods.9 intervention services, educational infants’ potential inability to autoregulate cerebral blood One neonatal care manage- services, and lost family income.4 flow adequately. Both of these findings were linked to ment activity that has been Additionally, the average cost of the development of IVHs. Based on physiologic data and examined in association with an IVH adds another $53,600 expert opinion, the authors found support in the literature IVH prevention is infant head to the cost of the initial hospi- and recommend implementing a plan of care that includes positioning. First studied in talization.5 But the costs of IVH midline head positioning for premature infants. adult patients, cerebral blood go far beyond the impact of the flow changes in response to head position were also examined in 10–12 Disclosure neonates beginning in the 1980s. The author discloses no relevant financial interest or affiliations with any The purpose of this article is to review current evidence on commercial interests. midline head positioning in the prevention of IVH. The goal

Accepted for publication May 2011.

N EONATAL NETWORK VOL. 30, NO. 6, NOVEMBER/DECEMBER 2011 © 2011 Springer Publishing Company 381 http://dx.doi.org/10.1891/0730–0832.30.6.381 Copyright © Springer Publishing Company, LLC of this review was to answer the clinical practice question: In TABLE 1 n Associated Risk Factors for IVH infants born at ,32 weeks gestation, does midline head Antenatal Risk Factors positioning along with head of bed tilted upward for the first 72 hours of life, when compared with standard posi- Prematurity High continuous airway pressure tioning, result in a lower incidence of IVH? Maternal infection Rapid fluid administration Maternal inflammatory Rapid alteration in blood pressure responses ETIOLOGY AND PATHOPHYSIOLOGY OF Hypotension Maternal hypertention Hypocarbia or hypercarbia INTRAVENTRICULAR HEMORRHAGE Maternal bleeding disorders Although this potentially devastating medical condition Pneumothorax Absent maternal steroid Asphyxia can occur at any age, premature infants are at an increased administration risk because of their immature brain vasculature and also their maternal diabetes Hypernatremia inability to autoregulate shifts in cerebral perfusion, described Placental insertion disorders Hypoglycemia as a pressure-passive circulatory state.13,14 Of the bleeds that Oligohydramnios Thrombocytopenia occur, 90 percent will develop during the first 72 hours of life, Maternal alcohol use Patent ductus arteriosis a time when these infants are in their most critically ill state.3 Maternal smoking Seizure activity Poor prenatal care Routine NICU care: Tracheal Although extreme prematurity and illness have been associ- suctioning, excessive handling, ated with shifts in brain perfusion, many seemingly benign Infertility treatments noxious stimulation, painful care activities and environmental factors have also been shown Out-born delivery and neonatal procedures, stress to cause changes in cerebral blood flow patterns.2,13–16 transport Any disease process or care activity Although there are many risk factors associated with Initial resuscitation efforts associated with alteration in Asynchronous ventilatory cerebral perfusion. the development of an IVH, some of which are noted in support Table 1, one of the most prominent risks is prematurity.3,17 This increased risk is caused by the presence of the germinal matrix, a network of delicate blood vessels within the premature brain that usually involutes between 32 weeks and A factor unique to this population is the inability of premature term gestation.3 Within this region, the capillary–venule infants to autoregulate cerebral perfusion in response to physi- juncture is the originating site of these hemorrhages.3 The ologic and positional changes.13 Autoregulation is the ability of fragile blood vessels within the germinal matrix are easily the body to maintain a constant blood flow to the brain despite ruptured with any rapid changes in the levels of cerebral cerebral perfusion.19 Inconsistencies in cerebral blood flow pat- perfusion, which may lead to bleeding into the brain tissue or terns have been observed during routine critical care of the ventricles.18 The structure of the venous system in this area premature infant.16 Impaired autoregulation can be markedly of the brain can also lead to IVH because the system has a pronounced in infants who are sick or extremely premature.14,20 U-shaped vessel pattern prone to venous congestion near the The mechanism of action that has been postulated is that germinal matrix, again causing vessel damage and bleeding. during head rotation to the side, an occlusion or obstruction A cranial ultrasound view of a neonate without IVH can be of the jugular venous–venule drainage system could occur on seen in Figure 1, whereas Figure 2 shows a drawing of the the ipsilateral side of the head. This is followed by increased U-shaped vascular anatomy. venous congestion in this area leading to vessel rupture.

Figure 1 n Cranial ultrasound of a neonate without IVH.

N EONATAL NETWORK 382 NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6

Copyright © Springer Publishing Company, LLC Figure 2 n U-shaped vascular anatomy. an origination site, although this is more common in term infants.25,26 Regardless of site of origination or description Medullary veins of pathology, many bleeds are now being associated with Choroidal vein the level of neurodevelopmental outcomes risk.25 Low-grade risk bleeds are associated with Grade I and II hemorrhages. Thalamostriate High-grade risk bleeds are associated with Grades III and vein IV hemorrhages. See Table 2 for a description of IVHs and statistics for very low birth weight (VLBW [,1,500 g]) infants. Because an IVH can be such a devastating event, there is a Terminal critical need to identify strategies to reduce IVH in this pop- vein ulation. One proposed strategy is the use of midline position- Internal ing during the first 72 hours of life, a time when 90 percent cerebral vein of all IVH occur.3 Vein of Golen DEVELOPING THE CLINICAL PRACTICE QUESTION USING PICO FORMAT When completing an evidence-based review of the literature, developing a question that helps focus the literature By maintaining a neutral head position, it is theorized that search is the first step. PICO is a mnemonic term used to venous obstruction could possibly be avoided, potentially focus and describe each part of the clinical practice question: preventing IVH caused by head position. “P” is the population, “I” is the intervention, “C” is the comparison group, and “O” is the outcome.33 INTRAVENTRICULAR HEMORRHAGE The PICO question focuses on an intervention that is SEVERITY AND OUTCOMES compared to the current standard of care. If there is evi- Diagnosed by cranial ultrasound, an IVH can occur fol- dence that an intervention may provide benefit without lowing serious illness of the infant or after no apparent insult harm, an intervention may be implemented into practice. at all. During the first four to five days of life, a time when The PICO or clinical practice question that prompted this premature infants are in their most critical state, 95 percent search was: of all cases of IVH will develop.21 Depending on the severity, “Do infants born at 32 weeks gestation who are posi- some of these bleeds may be accompanied by an acute deteri- tioned with head in midline position and head of bed tilted oration in clinical status, whereas some infants may show few upward for the first 72 hours of life have a lower incidence of symptoms until they reach school age.22 Overall, Paige and IVH than infants who receive standard positioning?” Carney found the associated sequelae of IVH to range from P 5 In infants born at 32 weeks gestation minimally distinguishable effects (50 percent), to abnormal I 5 does midline head positioning along with the head of neurologic outcomes (20–30 percent), to an increase in inci- bed tilted upward for the first 72 hours of life dence of mortality (10–30 percent).22 C 5 compared with standard positioning Although the IVH has been broken down into classifica- O 5 result in a lower incidence of IVH tions by grade, Volpe later developed an IVH labeling system based on a description of the neurologic pathology.3,23,24 LITERATURE SEARCH STRATEGIES This was caused by some abnormalities that can occur that A literature search using the keywords intracranial hem- do not fit into the original classifications. Such cases could orrhages, cerebral ventricles, infant, and newborn was per- include isolated ventriculomegaly or instances of white formed using Medline, Cumulative Index to Nursing and matter injury that are not associated with IVH.25 Volpe also Allied Health Literature (CINAHL), and Google Scholar. advocated for the cessation of labeling white matter brain The search years were limited from 1980 to 2010. On the tissue hemorrhages, or parenchymal hemorrhages, Grade IV first search, 935 articles were found. The search was then IVH.3 This was caused by the origination of these bleeds, limited to articles in English, human subjects, and infants at times, occurring secondary to parenchymal infarction from birth to 23 months of age. This search yielded 800 and not always being associated with bleeding within the articles. The search was then altered to infant, premature, ventricles.25 and cerebral ventricles and hemorrhage with the same limi- The Papile IVH grading system also does not describe tations of English, human subjects, and infants from birth the site of origin for the IVH.25 Most often, intracranial to 23 months of age. The reason that older research was hemorrhages in premature infants originate in the germinal admitted into this search is that this early time represents matrix, a highly vascular and fragile region in the premature the initial study into IVH and premature infants. Many of infant’s brain. Alternately, the choroid plexus can also be the positioning studies were conducted during the 1980s

N EONATAL NETWORK VOL. 30, NO. 6, NOVEMBER/DECEMBER 2011 383

Copyright © Springer Publishing Company, LLC and have not been repeated. The search then yielded 189 systematic review of nursing care activities or positioning articles. When the term prevention was added to the search, was found.34 77 articles were found that most appropriately fit the clinical Of the 77 articles reviewed, organization of the rest of the question. review consists of an evaluation of 11 articles. These articles Following this, a search of the Cochrane Systematic most appropriately answered the review question regarding Review Database was completed. Although many reviews positioning the premature infant and IVH prevention. The on IVH prevention were present, the reviews focused on other 66 articles that were discarded did not address research the medical management of IVH, such as medication relating to positioning and IVH occurrence in the premature administration. There was one review regarding devel- neonate. The following discussion synthesizes the evidence opmental care and the prevention of morbidities, but no collected.

Table 2 n Intraventricular Hemorrhages Papile’s Classification Volpe’s Classification: Occurrence Morbidity and Neurologic Mortality Progression by Severity Description by Pathology Rate Outcomes of Ventricular Dilatation Grade I: least severe Subependymal hemorrhage 25%–30% for Ten percent motor disability. This 5%3 4%28 mild IVH (SEH)—also called germinal Grades I–II27 rate is comparable to premature See Figure 3 matrix hemorrhage (GMH) infants without documented hemorrhage.24 Grade II: considered Intraventricular hemorrhage Most infants with Grade II IVH face 10%3 12%28 mild-to-moderate (IVH)—alternately called a SEH the same neurologic outcomes IVH with progression into the lateral associated with Grade I IVH, See Figure 4 ventricles by ,50% without although the extent of bleed can dilatation. lead to ventricular dilatation.18 Patra et al. found significantly poorer neurodevelopmental outcomes in extremely low birth weight (ELBW) infants (,1,000 g) with Grades I–II IVH at 20 months corrected age. This includes up to 15% of these ELBW infants with Grades I–II IVH who develop cerebral palsy (CP) and 9% who develop deafness.27 Grade III: considered IVH with ventricular dilatation— 10%–12% for Ventricular dilatation can result when 20%3 74%28 moderate-to-severe SEH with progression into the Grades III–IV29 blood blocks the cerebrospinal fluid IVH lateral ventricles by .50% and/ pathway, leading to progressive or with dilatation of ventricles. ventricular dilatation and increased intracranial pressure. The morbidity related to posthemorrhagic hydrocephalus is significant, with up to 90% of these infants having some degree of neuromotor deficits and 25% with visual and auditory impairments.30 In total, 76% of these infants will have pronounced disability and 56% have multiple impairments.28 Fifty percent of these children will require some special education and enrichment programs.24 Grade IV: considered Intraparenchymal hemorrhage Generally unilateral with the 50%3 71%28 severe IVH (IPH)—hemorrhagic infarct prognosis most often associated See Figure 5 into the white brain matter with poor motor deficits as well as significant cognitive impairments.24 Classified as the most severe IVH, many infants do not survive. Periventricular leukomalacia Ten percent very low birth weight (PVL)— parenchymal, or white (VLBW) infants with PVL will brain matter, necrosis often develop CP with spastic diplegia occurring following Grade IV and 50% will develop cognitive IVH or a parenchymal infarct. and behavioral deficits.31,32

N EONATAL NETWORK 384 NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6

Figure 4 n Ultrasound of Grade II IVH.

Figure 5 n Ultrasound of Grade IV IVH.

N EONATAL NETWORK VOL. 30, NO. 6, NOVEMBER/DECEMBER 2011 385

Copyright © Springer Publishing Company, LLC SYNTHESIS AND SUMMARY for these studies centered on positioning or altering position Details of all appraised studies can be found in the Appendix. of the newborn infant. Some of these articles focused on the effects of midline head positioning, whereas others looked at The Five Strengths of Evidence35 the effects of changes in the tilting position. Regardless of Type 1: strong evidence from at least one systematic review of particular position change that each individual study exam- multiple, well-designed randomized, controlled trials. ined, they all evaluated changes in cerebral hemodynamics or Type 2: strong evidence from at least one properly designed ICP in response to position change. randomized, controlled trial of appropriate size. Following the positioning interventions, an evaluation of Type 3: evidence from well-designed trials without random- cerebral blood flow or ICP was assessed. There were various ization, single group pre–post, cohort, time series, or instruments used to assess these measures. In the 10 articles matched case-control studies. with patient enrollment, 5 used near infrared spectroscopy Type 4: evidence from well-designed, nonexperimental (NIRS), 4 used ultrasound, and 2 used a transfontanel pres- studies from more than one center or research group. sure transducer. The transfontanel pressure transducers were Type 5: opinions of respected authorities (based on clinical evi- used in the first studies, followed by the use of ultrasound, dence), descriptive studies, or reports of expert committees. and then NIRS methodology because technology has pro- There was no meta-analysis or randomized controlled gressed. A brief description of the instrumentation used for trials. The studies included nine predesigns and postdesigns, the evaluations is listed in Table 3. two repeated measures design, and one expert review panel The articles were then examined to determine the homoge- report. One of the studies used both a premethodology and neity of the “O,” or outcomes. When evaluating the outcomes postmethodology as well as a repeated measure design, total- reported in these studies, the groups could be separated into ing 11 reviewed articles.36 two divisions: those that evaluated the effects of head or When synthesizing the evidence gathered regarding body position changes and those that evaluated the effects of the positioning of premature infants and the potential tilting. In those that evaluated head and body changes, several effects of these positions on cerebral hemodynamics, the outcome measures were used and not all assessed the incidence information gathered was analyzed following the PICO of IVH, making the evaluation of the intervention difficult. format. The purpose of this analysis was to determine In general, these articles demonstrated alterations in cere- whether the studies reviewed were homogenous. This is bral blood flow following position changes. One study found an important step in evaluating whether the study simi- a significant decrease in tissue hemoglobin index and tissue larities lend to the pooling of evidence that may support a oxygen index during head rotation in infants ,26 weeks ges- practice change. tation.37 A second study found a significant increase in cerebral First, did all of the studies ask the same question? Not all of blood volume (CBV) during 90-degree head rotation, which them. Although the articles were included in the review because was pronounced in infants ,1,200 g.38 The third study found of similar subject matter, the focus of some of these articles cerebral blood flow velocities were significantly higher in the was not exactly homogenous. Eight of the articles studied changes in cerebral hemodynamic and two studied changes in intracranial pressure (ICP) in response to position changes. TABLE 3 n Definition of Study Instruments The final article gave expert opinion about infant positioning after benchmarking hospitals with low IVH rates. Ultrasonography A non-invasive radiological exam that uses a transducer to pass Next, the articles were examined to determine the homoge- sound waves through soft tissue neity of the “P,” or population. Of the articles reviewed, nine and fluid. A picture is produced studies examined preterm infants. A tenth article that exam- when the returning echo bounces off internal structures and returns ined full-term infants was also included because this study is to the transducer. The resulting often cited as a seminal article that examines changes in neo- picture is formed when the data natal cerebral hemodynamic in response to position change.11 entered into the transducer is read by the ultrasound computer and Of the articles examined, there was a wide variation in ges- is analyzed to produce real-time tational age, weight, postnatal age, and level of illness. These images.44 infants ranged from the most premature and critically ill infants Trans-fontanel Pressure A non-invasive device that measures to infants who were described by the authors as healthy pre- Transducer intracranial pressure through a mature infants. Because of the profoundly wide variation in probe secured over the anterior fontanel.10 levels of health and gestational age of the subjects, comparing outcomes for infants in this review was difficult. Additionally, Near Infrared Spectroscopy A non-invasive neuro-imaging device (NIRS) that uses near-infrared light to none of these studies strictly examined the neonates during evaluate real-time tissue oxygen their first 72 hours of life, a time when most IVHs occur.32 and blood volume to interpret Next, t he a r t icles were exa m i ned to determ i ne t he homoge- blood hemodynamics of the brain.43 neity of the “I,” or interventions. The interventions reported

N EONATAL NETWORK 386 NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6

Copyright © Springer Publishing Company, LLC supine position at one month of age during evaluation, and and technology and medical practices may have changed. vertebral arterial flows in prone position were decreased.39 A Finally, all nine studies that examined preterm infants fourth study found jugular blood flow was decreased during would be considered “stable” neonates and may not reflect 90-degree head rotation.11 The three studies that evaluated the group of patients who are at greatest risk for IVH, those ICP all found a significant decrease in ICP, with the head in who are severely premature during their first 72 hours of midline position and the head of the bed elevated.10–12 life. The final article, based on expert opinion, benchmark- The second group of studies examined the effects of tilting. ing, and review of the literature, reviewed multiple practices These studies evaluated CBV and found a significant increase and their recommendations varied between level, depending with the head lowered in a dependent position. These studies on the strength of the evidence.6 The level of evidence for also found significant alterations in CBV in response to the recommendations on head positioning by VON was IV tilting, especially in preterm or brain-injured infants. Another and VI.6 According to the Muir Gray schema of evidence study evaluated for biphasic responses in cerebral blood flow that was used to rate the evidence, IV is defined as a well- velocities, demonstrating autoregulatory responses in preterm designed, nonexperimental study, and VI is defined as an infants, and found significantly more reliable responses were evidence supported by casual theory of disease.35 Although elicited as gestational age increased but IVH outcome was there are limitations to these 11 studies, they may still, not evaluated.36 A complete description of the findings of however, provide some benefit to our patient population. these articles can be found in the Appendix. Many investigators included discussion about how their Although the final article was not a research study, infor- study results could be interpreted. The expert opinion of mation from such an article can still be valuable. This article many of the researchers attribute changes in cerebral oxy- by Carteaux and colleagues detailed the work of a multidis- genation, increased CBV, and/or increased ICP to occlu- ciplinary focus group that was formed to complete an evi- sion of the jugular venous drainage system following a turn dence-based literature review, benchmarking activities, and in the position of the head. The proposed consequential expert committee review. The purpose of this evaluation was backup of cerebral blood is an ongoing theme presented by to identify potentially better practices (PBPs) that could lead many of the authors, although jugular blood flow was only to the reduction of IVH and periventricular leukomalacia analyzed in one study.6,11,12,37–39 These authors then specu- (PVL) in VLBW and premature infants.6 late that the risk of IVH was increased because of ruptures The group identified benchmark NICUs with the lowest in the cerebral venous–venule drainage system following incidence of IVH reported to the Vermont Oxford Network blood accumulation secondary to the occlusion. Although (VON). The VON, a consortium of more than 700 NICUs jugular obstruction studies during head rotation have been associated with the improvement of safety and quality of documented in the adult population, results in the neonatal newborn care, formed a focus group to evaluate methods for population are limited and similar results in this population reducing the incidence of IVHs in premature infants.40 It then are theorized.11,38 The researchers concluded that venous developed an NICU practice questionnaire for VON sites. Four obstruction caused by head position could be detrimental to sites with low incidence of intracranial hemorrhage were iden- these infants who are already at increased risk for IVH. tified and used as a benchmark for clinical practice. Specific A second observation that was discussed by the investiga- clinical practices at these sites were described through struc- tors was the connection between the effects of head tilting on tured questionnaires and site visits. The information obtained brain hemodynamics. Of the four studies that examined tilting, from these sites was analyzed and used to help identify NICU the experts attributed the results, an increase in CBV and/or practices that might be related to IVH prevention. A complete increased ICP, to the infants’ potential inability to autoregulate literature review was then completed on each of these NICU cerebral blood flow adequately. Because these findings were sig- practices. Following the benchmarking and literature review, nificantly increased in infants with PVL, brain injury, and those the group identified a final list of ten recommended practices who were premature, the authors further speculated these find- for NICUs that were PBPs, which could help with the reduc- ings may put these infants at a greater risk for developing IVH. tion of incidence of IVH. Use of midline positioning and bed All of the methods of measurement used throughout the elevation of 30 degrees was identified as PBPs. studies showed a difference in cerebral hemodynamics when positioning was changed. These findings were present in DISCUSSION the head rotation studies as well as the tilting ones. These The decision to make a practice change should be based differences were most marked in the ELBW infant. Present on the grade, quality, and strength of the evidence after studies, unfortunately, only include outcome measures that synthesizing all data. Of the 11 articles reviewed, 10 of were short-term and any changes in practice should include them involved clinical trials. Of these, all were Type 3 evi- long-term outcome measurement. dence with a quasi-experimental, nonrandomized conve- The investigators who completed the expert review6 did nience sample design with five studies including a control discuss the lack of high-level evidence when evaluating infant group. All of the studies were small and no power analy- positioning.6 The decision to recommend the use of neutral sis was commented upon. Many of the studies were older, head position was based on the potential benefits of this

N EONATAL NETWORK VOL. 30, NO. 6, NOVEMBER/DECEMBER 2011 387

Copyright © Springer Publishing Company, LLC practice and the lack of harm. Based on these recommenda- areas to assess are organizational culture, organizational tions, many NICUs currently use these positioning practices. infrastructure, and organizational resources.41 In assessing organizational culture, one would assess the values of the IMPLEMENTATION OF NEUTRAL unit. Do the caregivers understand the importance of imple- HEAD POSITIONING menting evidence-based recommendations? Do the nurses The decision to implement midline/neutral head posi- and other caregivers understand the potential benefits and tioning and a 30-degree elevation in the HB in infants less how to achieve neutral head positioning? Does the practice than 32 weeks gestation for the first 72 hours of life can be change support family-centered care, an important value to recommended, at this point, based on physiologic data and NICU caregivers? Does this practice change support devel- the views of experts in the field. Furthermore, there have not opmental care practices, another important care value in been any adverse consequences identified when implement- NICUs? Have the team given input in identifying potential ing these positioning changes. barriers to this practice change, including nursing barriers to For units considering a change in neonatal positioning care and equipment needs? practices for potential IVH prevention such as the implemen- In assessing organizational infrastructure, one would assess tation of neutral head positioning, there are several impor- the organization’s willingness to support evidence-based care tant steps to facilitate change. These steps should only occur practice changes. Does the unit have goals that state support following critical appraisal of the evidence. of practices based on the most current evidence?41 Has the organization made efforts to hire or train employees in the Gather the Stakeholders evidence-based process? The stakeholders incorporated in this practice change The organizational resource assessment evaluates whether should include registered nurses who care for the infants and an organization is willing to support the man-hours needed understand the fine nuances of caring for the infants. The to evaluate the evidence and implement these changes. Can physicians, advance practice nurses (APNs), and bedside staff the organization financially support the EBP process, which nurses would be essential in identifying infants appropriate includes the evaluation, implementation, and assessment for this practice change and for ordering this care practice. phase of this process? Is the organization willing to supply Physical therapists would be needed to assist with positioning equipment and training time? If the organization cannot and obtaining positioning devices needed on an individual give full financial support for the practice change, are they bases. Respiratory therapists would be important to help posi- willing to support further planning to identify creative alter- tion infants in neutral head positioning while still receiving nates that support the evidence? the necessary respiratory support. Specialized equipment may be needed to positioning these infants midline, especially if Use Multiple Implementation Strategies the infant is on an oscillating ventilator. Pharmacists input to When implementing a practice change, multiple imple- maintain patient comfort may also be a necessity. Our unit has mentation strategies can help ensure success. Does the developed a multidisciplinary IVH prevention taskforce com- education plan include multiple methods to reach the care- prised of representation from all the mentioned stakeholders giving team, such as demonstrations of midline positioning to evaluate the evidence and provide recommendations. and posters with pictures? Does the education presentations included multiple levels of medical knowledge, with the dif- Create a Detailed Action Plan ference being parental education being easier to understand The first steps to creating an action plan would be to for the layperson and the medical education, including a more use the stakeholders to discuss potential obstructions, plan pathophysiologic approach to address the caregivers under- nursing and medical team education, and plan parent educa- standing of the rationale for change? When implementing tion. An audit of current positioning practices can help the these changes, are there multiple approaches to support the team assess the degree of change that is being proposed for bedside nurses such as the team members available to address the unit. This can help the team determine the amount of technical questions and nursing champions to encourage time this change may require and the amount of support practice change use through exemplifying the practice? needed to be successful with this change. Planning to assess outcome measures (IVH rates) prior to EVALUATING OUTCOMES the practice change is important in evaluating if the practice Process Outcomes change of midline positioning has made a difference in this Once the practice change has occurred, a method to assess patient population. Our unit has done this step in the IVH the practice change is essential. Is the practice implemented prevention taskforce. as designed? Can the caregiving team describe midline positioning and demonstrate this practice change correctly? Is Assessing Environmental Readiness midline positioning being performed routinely? Have the The stakeholders must evaluate and address the organi- caregiving team identified further barriers to midline posi- zation, environment, and whether it is supportive to this tioning as the practice occurs on a routine basis? A quality evidence-based practice (EBP) change at this time. The three assurance plan to assess that the practice change is being

N EONATAL NETWORK 388 NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6

Copyright © Springer Publishing Company, LLC executed regularly and correctly is essential. This is impor- 5. Russell, R. B., Green, N. S., Steiner, C. A., Meikle, S., Howse, J. L., tant because it assesses whether a practice change is truly Poschman, K., . . . Petrini, J. R. (2007). Cost of hospitalization for preterm and low birth weight infants in the United States. Pediatrics, being carried out as planned. A method to evaluate this 120(1), e1–e9. http://dx.doi.org/10.1542/peds.2006-2386 could be the formation of “positioning super users” to audit 6. Carteaux, P., Cohen, H., Check, J., George, J., McKinley, P., Lewis, W., . . . patient positioning and evaluate the need for further educa- McConnell, C. (2003). Evaluation and development of potentially better tion based on their findings. The identification of barriers practices for the prevention of brain hemorrhage and ischemic brain injury and development of solutions to barriers is essential in the in very low birth weight infants. Pediatrics, 111(4 Pt. 2), e489–e496. success of the implementation of this change. 7. Crowley, P. (2000). Prophylactic corticosteroids for preterm births. Cochrane Database of Systematic Reviews, (2), CD000065. Clinical Outcomes 8. Kattwinkel, J., Niermeyer, S., Nadkarni, V., Tibballs, J., Phillips, B., Zideman, The final step in this process is an evaluation of the clini- D., . . . Osmond, M. (1999). Resuscitation of the newly born infant: An advisory statement from the Pediatric Working Group of the International cal outcomes. Has this change made an impact? How do the Liaison Committee on Resuscitation. Resuscitation, 40(2), 71–88. current IVH rates compare with the rates prior to the prac- 9. Vohr, B., & Ment, L. (1996). Intraventricular hemorrhage in the preterm tice change, as well as the rates of others such as NICUs? Can infant. Early Human Development, 44(1), 1–16. we follow long-term outcomes such as incidence of cognitive, 10. Emery, J. R., & Peabody, J. L. (1983). Head position affects intracranial behavioral, and physical disabilities? pressure in newborn infants. The Journal of Pediatrics, 103(6), 950–953. 11. Cowan, F., & Thoresen, M. (1985). Changes in superior sagittal sinus Future Direction blood velocities due to postural alterations and pressure on the head of Although the cause of IVH in premature infants may be the newborn infant. Pediatrics, 75(6), 1038–1047. multifactorial and complicated, many investigators are cur- 12. Goldberg, R. N., Joshi, A., Moscoso, P., & Castillo, T. (1983). The effect rently searching for prevention methods. Some neonatal of head position on intracranial pressure in the neonate. Critical Care Medicine, 11(6), 428–430. units have adopted IVH prevention bundles or multiple care 13. Owens, R. (2005). Intraventricular hemorrhage in the premature neonate. practice changes that can potentially reduce the incidence Neonatal Network, 24(3), 55–71. 42 of IVH. Regardless of whether a unit is looking to make 14. Perlman, J. M. (2009). The relationship between systemic hemodynamic several practice changes or just one change at a time, there is perturbations and periventricular-intraventricular hemorrhage—A historical little dispute that work toward a decrease in the incidence of perspective. Seminars in Pediatric Neurology, 16(4), 191–199. http:// IVH in premature infants should continue. dx.doi.org/10.1016/j.spen.2009.09.006 Additionally, further research into neonatal positioning 15. Ballabh, P. (2010). Intraventricular hemorrhage in premature infants: for IVH prevention should continue. Although the strength Mechanism of disease. Pediatric Research, 67(1), 1–8. of evidence to support this practice change could be stronger, 16. Limperopoulos, C., Gauvreau, K. K., O’Leary, H., Moore, M., Bassan, H., Eichenwald, E. C., . . . du Plessis, A. J. (2008). Cerebral hemodynamic the decision to adopt this practice change should be evalu- changes during intensive care of preterm infants. Pediatrics, 122(5), ated and discussed in individual neonatal units. The adoption e1006–e1013. http://dx.doi.org/10.1542/peds.2008-0768 of the practice of midline positioning could still be recom- 17. Vergani, P., Locatelli, A., Doria, V., Assi, F., Paterlini, G., Pezzullo, J. C., mended based on its potential benefits. & Ghidini, A. (2004). Intraventricular hemorrhage and periventricular With the increased survival of extremely premature infants, leukomalacia in preterm infants. Obstetrics and Gynecology, 104(2), 225–231. the incidence of IVH, a common neonatal morbidity, can 18. Bloch, J. R. (2005). Antenatal events causing neonatal brain injury in be expected to rise proportionately. Although research into premature infants. Journal of Obstetric, Gynecologic, and Neonatal Nursing, 34(3), 358–366. the prevention of this potentially devastating illness should 19. Kaiser, J. R., Gauss, C. H., & Williams, D. K. (2005). The effects of continue, caregivers must continue to evaluate the litera- hypercapnia on cerebral autoregulation in ventilated very low birth weight ture in evaluation of their current practices. Because IVH infants. Pediatric Research, 58(5), 931–935. can be so devastating to the infant, caregivers must strive 20. Wong, F. Y., Leung, T. S., Austin, T., Wilkinson, M., Meek, J. H., Wyatt, to provide evidence-based care that can potentially prevent J. S., & Walker, A. M. (2008). Impaired autoregulation in preterm infants these occurrences. identified by using spatially resolved spectroscopy. Pediatrics, 121(3), e604–e611. http://dx.doi.org/10.1542/peds.2007-1487 21. Linder, N., Haskin, O., Levit, O., Klinger, G., Prince, T., Naor, N., . . . REFERENCES Sirota, L. (2003). Risk factors for intraventricular hemorrhage in very low 1. Committee on Hospital Care, American Academy of Pediatrics. (2003). birth weight premature infants: A retrospective case-control study. Pediatrics, Family-centered care and the pediatrician’s role. Pediatrics, 112(3 Pt. 1), 111(5 Pt. 1), e590–e595. 691–697. 22. Paige, P. L., & Carney, P. R. (2002). Neurological disorders. In G. B. 2. McCrea, H. J., & Ment, L. R. (2008). The diagnosis, management, and Merenstein & S. L. Gardner (Eds.), Handbook of neonatal intensive care postnatal prevention of intraventricular hemorrhage in the preterm neonate. (5th ed., pp. 644–678). St. Louis, MO: Mosby. Clinics in Perinatology, 35(4), 777–792. http://dx.doi.org/10.1016/ 23. Papile, L. A., Burstein, J., Burstein, R., & Koffler, H. (1978). Incidence j.clp.2008.07.014 and evolution of subependymal and intraventricular hemorrhage: A 3. Volpe, J. J. (2008). Intracranial hemorrhage: Germinal matrix-intraventricular study of infants with birth weights less than 1,500 gm. The Journal of hemorrhage of the premature infant. In Neurology of the newborn (5th ed., Pediatrics, 92(4), 529–534. pp. 517–588). Philadelphia, PA: Saunders Elsevier. 24. Papile, L. A. (2002). Intracranial hemorrhage. In A. A. Fanaroff & R. 4. March of Dimes. (2009). About prematurity: The economic costs. Retrieved J. Martin (Eds.), Neonatal-perinatal medicine: Diseases of the fetus and from http://www.marchofdimes.com/prematurity/21198_10734.asp infant (7th ed., pp. 1001–1011). St. Louis, MO: Mosby.

N EONATAL NETWORK VOL. 30, NO. 6, NOVEMBER/DECEMBER 2011 389

Copyright © Springer Publishing Company, LLC 25. Vasileiadis, G. T. (2004). Grading intraventricular hemorrhage with no 41. Smith, J. R., & Donze, A. (2010). Assessing environmental readiness: grades. Pediatrics, 113(4), 930–931. First steps in developing an evidence-based practice implementation 26. Annibale, D. J. (2010). Periventricular hemorrhage-intraventricular culture. The Journal of Perinatal & Neonatal Nursing, 24(1), 61–71. hemorrhage. Medscape. Retrieved from http://emedicine.medscape.com/ 42. Bedwell, S. M., Sekar, K. C., & Bright, B. C. (2010, May). Decrease in article/976654-overview the incidence of intraventricular hemorrhages after the introduction of an 27. Patra, K., Wilson-Costello, D., Taylor, H. G., Mercuri-Minich, N., & IVH prevention bundle in the NICU. Presented at the Pediatric Academic Hack, M. (2006). Grade I-II intraventricular hemorrhage in extremely Society Conference, Neonatal Neurology Platform, Vancouver, British low birth weight infants: Effects on neurodevelopment. The Journal of Columbia, Canada. Pediatrics, 149(2), 169–173. 43. Bozkurt, A., Rosen, A., Rosen, H., & Onaral, B. (2005). A portable 28. Murphy, B. P., Inder, T. E., Rooks, V., Taylor, G. A., Anderson, N. J., near infrared spectroscopy system for bedside monitoring of newborn Mogridge, N., . . . Volpe, J. J. (2002). Posthaemorrhagic ventricular dilatation brain. Biomedical Engineering OnLine, 4(1), 29. http://dx.doi. in the premature infant: Natural history and predictors of outcome. Archives org/10.1186/1475-925X-4-29 of Disease in Childhood. Fetal and Neonatal Edition, 87(1), F37–F41. 44. U. S. Food and Drug Administration. (2008). Taking a close look 29. Ment, L. R., Allen, W. C., Makuch, R. W., & Vohr, B. (2005). Grade 3 to 4 at ultrasound. FDA Consumer Health Information. Retrieved from intraventricular hemorrhage and Bayley scores predict outcome. Pediatrics, http://www.fda.gov/downloads/ForConsumers/ConsumerUpdates/ 116(6), 1597–1598. http://dx.doi.org/10.1542/peds.2005-2020 UCM095487.pdf 30. Chumas, P., Tyagi, A., & Livingston, J. (2001). Hydrocephalus—what’s 45. Pichler, G., Urlesberger, B., Schmölzer, G., & Müller W. (2004). new? Archives of Disease in Childhood. Fetal and Neonatal Edition, 85(3), Effect of tilting on cerebral haemodynamics in preterm infants with F149–F154. periventricular leucencephalomalacia. Acta Paediatrica, 93(1), 70–75. 31. Perlman, J. M. (1998). White matter injury in the preterm infant: An http://dx.doi.org/10.1111/j.1651-2227.2004.tb00677.x important determination of abnormal neurodevelopment outcome. Early 46. Pichler, G., van Boetzelar, M. C., Müller, W., & Urlesberger, B. (2001). Human Development, 53(2), 99–120. Effect of tilting on cerebral hemodynamics in preterm and term infants. 32. Volpe, J. J. (2003). Cerebral white matter injury of the premature infant- Biology of the Neonate, 80(3), 179–185. more common than you think. Pediatrics, 112(1 Pt. 1), 176–180. 47. Schrod, L., & Walter, J. (2002). Effect of head-up body tilt position on 33. Melnyk, B. M., & Fineout-Overholt, E. (2005). Evidence-based practice autonomic function and cerebral oxygenation in preterm infants. Biology in nursing and healthcare: A guide to best practice. Philadelphia, PA: of the Neonate, 81(4), 255–259. Lippincott, Williams & Wilkins. 34. Symington, A., & Pinelli, J. (2003). Developmental care for promoting About the Authors development and preventing morbidity in preterm infants. The Cochrane Sheila Malusky is a neonatal nurse practitioner with over 19 years Library, (4). Retrieved from http://www.nichd.nih.gov/COCHRANE/ of neonatal nursing care experience, currently working in the level III symington/symington.htm NICU at St. Louis Children’s Hospital. Her interests include neo- 35. Muir Gray, J. A. (1997). Evidence-based healthcare: How to make health natal neurology and family-centered care. She would like to thank policy and management decisions. London, United Kingdom: Churchill Livingstone. Dr. Lyla Lindholm at UMKC and St. Louis Children’s Hospital NICU IVH Prevention Taskforce. Ms Malusky would especially like 36. Anthony, M. Y., Evans, D. H., & Levene, M. I. (1993). Neonatal cerebral blood flow velocity responses to changes in posture.Archives of Disease in to thank Ms. Donze for her continuing mentorship and caring guid- Childhood, 69(3 Spec. No.), 304–308. ance. Ms. Malusky received her undergraduate degree from Maryville 37. Ancora, G., Maranella, E., Aceti, A., Pierantoni, L., Grandi, S., Corvaglia, University in St. Louis, her graduate degree from Barnes-Jewish L., & Faldella, G. (2010). Effect of posture on brain hemodynamics in College of Nursing and Allied Health, and her doctoral degree from preterm newborns not mechanically ventilated. Neonatology, 97(3), the University of Missouri, Kansas City. 212–217. http://dx.doi.org/10.1159/000253149 Ann Donze has over 34 years of experience in the NICU, with the 38. Pellicer, A., Gayá, F., Madero, R., Quero, J., & Cabañas, F. (2002). past 15 years as a neonatal nurse practitioner. She has coordinated Noninvasive continuous monitoring of the effects of head position on the neonatal nurse practitioner program at Barnes-Jewish College of brain hemodynamics in ventilated infants. Pediatrics, 109(3), 434–440. Nursing and Allied Health. Ms. Donze currently cochairs the St. Louis http://dx.doi.org/10.1542/peds.109.3.434 Children’s NICU research committee. Ms. Donze received her nursing 39. Eichler, F., Ipsiroglu, O., Arif, T., Popow, C., Heinzl, H., Urschitz, M., diploma from Barnes School of Nursing, her undergraduate degree & Pollak, A. (2001). Position dependent changes of cerebral blood flow from Maryville University, and her graduate degree from Southern velocities in premature infants. European Journal of Pediatrics, 160(10), Illinois University-Edwardsville. 633–639. http://dx.doi.org/10.1007/s004310100806 40. Vermont Oxford Network. (2008). What is the Vermont Oxford For further information, please contact: Network? In About Us. Retrieved from http://www.vtoxford.org/home. Sheila Malusky, DNP, RN, NNP-BC aspx?p5about/index.htm E-mail: [email protected]

N EONATAL NETWORK 390 NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6

Copyright © Springer Publishing Company, LLC ) continued ( 26 and 26 weeks 27 size with only patientsin 8 the , , GA groups. minutes10 in each which position, might not be long full for enough evaluation. performed to numberdetermine of subjects needed to reach statistical significance. Investigators it that noted sometimes was difficultout rule to artifacts. significantly had gestational lower birthweight, age, They weight. and significantly had higher PCA and chronologicalage. and position each not all infants were able to complete the sequence. Study Limitations Study Small sample only spent Infants No power analysis Small sample size. group PVL in Infants in minutes 30 Only No power analysis. 26 weeks 26 26 weeks, 26 during 0.05). TOIs remained 0.01). Post tilting Post 0.01). ,

p p  to evaluate tissue hemoglobinindex (nTHI) tissueand oxygenation index (TOI) in all positions.Nosignificant TOI or changesnTHIs in for infants . gestation. was nTHI significantly reduced in infant head rotation. nTHI supinewas in positions (both30 flatat and degreeselevated) were significantlyhigher than supineposition with head rotated to the side ( stable in all positions. postnataland CPAP age significantly werenot associatied with changes andin nTHI TOI. increased significantly volume blood cerebral tilting a following downward maneuver, volume blood cerebral hemoblobin cerebral and oxygen index was increased significantly in infants with PVL infants to compared tilting PVLpost without ( up, infants with PVL had PVLhad with infants up, decrease pronounced a tilting post and CBV in pronounced a had down increase in CBV. Outcomes ANOVA was performed had groups both Although : : NIRS : : Near- : Right lateral. degrees.

: Infantswere : : Infantswere : minutes. minutes. minutes.

nstrument nstrument : midline head position or position head midline : : prone or supine. episodes of head tilted up tilted head episodes of up tilted head episodes of : flat or elevated. : From HOB flat to elevated.

measured after placement in afterplacement measured 6 different positions. degrees to 90 rotated head the side. analyzed were postnatalage variables. independent as measured before and after and before measured tilting by changes position 20 up bed 24 minutes, 30 degrees for 20 episodeshorizontal 19 and 30 for 24 minutes, 30 degrees for 20 episodes horizontal 23 and 30 for infrared spectroscopy (NIRS). spectroscopy infrared tissue in Changes (hgb) hemoglobin index (nTHI) and tissue oxygenation index (TOI) after posture variations. Method Head Body HOB and CPAP Gestational age, Study I Method Head and Body HOB PVLinfantswith had 10 The PVLinfantswith had 25 The Study I Intervention Biological MeasuresBiological

: : : : : : : : : A A ge ge : : : 35 stable: 35 27.5 weeks 27.5 grams925 days 10.3 30 weeks 30 grams1225 days 14 24 stable24 preterm infants. normalwith All studies.brain nasal on Eleven CPAP. preterm infants: infants: preterm Control group— infants 25 with normal studies. brain Experimental group—10 infants with PVL. nfants nfants Mean Weight: Mean A Mean Weight Mean A Population I Mean G I Mean G alterations in the brain homodynamics of preterm newborns following head and bodyposition changes? gestational age, postnatal age, and nasal also was CPAP evaluated. tilting bed up degrees, 20 any there are effects in the cerebral hemodynamics of preterm orinfants with PVLwithout identified? PICO Question PICO The influence of Following Rating Strength/ Quality Level 2 Are there Level 2 6,10–12,36–39,45–47 non-randomized, non-randomized, convenience sample, within- subject before- design and-after with participants their servingas own controls non-randomized non-randomized convenience with sample, control group design. Evidence Type Evidence Quasi-experimental, Quasi-experimental, Quasi-experimental, - - - Summary of Evidence

n posture on brain hemodynamics preterm in newborns not mechanically ventilated cerebral hemo dynamics in pre infantswithterm periventricular leucencepha lomalacia Schmolzer, & Muller, 2004 Article Citation The effects of Ancora et al., 2010 Effects of tilting on Pichler, Urlesberger, Urlesberger, Pichler, appendix

N EONATAL NETWORK VOL. 30, NO. 6, NOVEMBER/DECEMBER 2011 391

Copyright © Springer Publishing Company, LLC ) continued ( CBF measurements CBF were unsuccessful infants.Tried 9 in to minimize bias randomly by theassigning startingposition. Also all HUS and read were NIR all samethe by investigator. were based on lower level when evidence no were RCTs available. Study Limitations Study Small sample size. Some of the PBPs -test for -testfor p 5 0.026). -test using -testusing U p 5 0.05). This change volume significantly increased with head turned 90 degrees. ( was most pronounced in infants , 1200 grams. There was also a significant change in cerebral blood flow relative to time spent in supine with head turned to side compared to head in midline ( There was no significant change in cerebral blood flow or any other physiologic variable: BP, oxygen saturation, PCO2. using Student t paired analysis and Mann- Whitney Statview software. practicesidentified, were neutralhead including positioning and the use developmentalcare of strategies. Outcomes Change in cerebral blood The analysis was completed completed was analysis The Ten potentially better potentially Ten : : : NIRS HUS : CBV) and cerebral  : Infants measured : midline or head rotated rotated head or midline : : prone or supine. : Flat or elevated elevated or Flat : changes in cerebral bloodcerebral in changes volume ( before and afterplaced being and before in multiple positions. 90 degrees to the side. 30 degrees. in minutes 30 for minutes 10 each position. obtained afterto study was detect changes. blood flow (CBF). Measured Cerebral bloodCerebral Measured (CBV) cerebralvolume and hemoglobin oxygen index (cHbD). capnography, oximetry, pulse and respiratory effort. members participated in a QI project to evaluate practices in benchmarked hospital’s IVH prevention methods. They utilized benchmarking of practices in institutions with low incidence of IVH and PVL, systematic review of the literature, and expert consutation, the group then made recommendations. Intervention Method Head Body HOB everyInfants measured Study Instrument MeasuresBiological Biological MeasuresBiological Investigators also recorded EKG, Five NICUs who were VON : : : : : : :   30.9 weeks 4.9 1575 grams803 5.8 days NICUs. preterm21 infants. on 13 conventional on 8 ventilators, oscillators Population Mean Weight Mean Age Five Benchmark Infants Mean GA evaluation and evaluation development of potentially better practices for the prevention brain of hemorrhage and ischemic injurybrain in very low birth weight infants identified? be position head brain on hemodynamics or changes cerebral in venous blood flow/ volume ventilated in infants be identified? PICO Question PICO Can an Can Expert Committee Report. they Do a have for level clinical practice guideline? This really fits that definition. Rating Strength/ Quality Level 4 Level Level 2 Can the effects of (continued) literature review, benchmarking, and expert committee review design non-randomized, convenience sample, within- subject before- design and-after with participants their servingas own controls. Evidence Type Evidence Evidence-based Evidence-based Quasi-experimental, Summary of Evidence

n x i development of of development better potentially thepractices for of prevention hemorrhagebrain and ischemic injurybrain in very low birth infants. weight continuous the of monitoring effectshead of brain on position hemodynamicsin infants. ventilated Quero, Madero, & Cabanas, 2002 ppend Article Citation Evaluation and Evaluation Carteaux et al., 2003 Noninvasive Gaya, Pellicer, a

N EONATAL NETWORK 392 NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6

Copyright © Springer Publishing Company, LLC ) continued ( did not comment comment not did direction on of head (midline or rotated). completed. follow up at 10 10 at up follow an and days infant additonal lost to follow up at 1 month corrected age. theincluded bias use of a second investigator infant to blinded positions. Study Limitations Study Small sample size. investigatorsThe not analysis Power Small sample size. to infants lost 6 eliminate Attemptsto 1,500 grams showed revealed initial maximal initial revealed total fluctuations of hemoglobin cerebral content (tHb) up to 42% After HETP. following stabilization within several prolongedtilting minutes, any in result not did furtherchangessignificant heart rate, tHb, of arterial mean pressure saturation. oxygen and Only preterm infants , or 5 of decreasesignificant a oxygen cerebral regional (rSO(2))saturation from 2–5% about of day 2 to 8, measured by pulseoxymetry. completed using SPSS statistic program. Non- parametric tests were applied. velocitieswere the in significantly higher one the at position supine evaluation. age of month researchers found The a decrease in vertebral pronearterial in flow position, likely due to unilateral vessel compression. signfificantly not did age bloodcerebral influence flow. Outcomes Continuous recordings recordings Continuous The study analysis was Cerebral blood flow Birthweightgestational and : : Total : Total : NIRs : : supine : Preterm infants were : Infants were nstrument nstrument : Centered when supine, supine, when Centered : : prone or supine. : horizontal then elevated then horizontal : : Not discussed. measured after being placed in multiple head/body tilt positions. degrees each with 30 least lastingat position 20 minutes each. positionsvarious the timesin during study. cerebral hemoglobin content. content. hemoglobin cerebral arterial mean oximetry, Pulse respiratory and pressure, curve. impedance measured before and after 4 position changes on 3 separate occasions: postnatal day 3-5, at one week, and at one month. when side either turnedto supine. Cerebral blood flow velocitiesbloodflow Cerebral artery, carotid internal of vertebralbasilar and artery, artery. Ultrasound Method Head and Body HOB Infants were each measured 4 Study I Intervention Investigators also recorded EKG, Study I Biological MeasureBiological Method Head Body HOB Biological MeasuresBiological : : : 2 to to 2 : : : 25-36 : A A ge ge : : 23 stable 23 : weeks

36 preterm 36 infants weeks.Median GA 32.5 weeks. 880-2980 Mediangrams. weight grams.1460 days of life.12 preterm infants preterm normalwith all studiesbrain none and mechanically ventilated. infantsAll were studies postnatal 3-5 days. 26.7 nfants nfants I Mean G Mean Weight Mean A Population I Mean G Mean Weight grams 1027 Mean A infants, are any there negative effectshead of body elevated position tilt (HETP)on systemic and cerebral oxygenation, circulation, and sympathetic- balance? vagal dependent changes in cerebral blood velocitiesflow identified be premature in infants? In preterm In PICO Question PICO Can position Level 2 Rating Strength/ Quality Level 2 Level (continued) non-randomized non-randomized convenience sample, non- control equivocal group before- design, and-after and Repeated- measuresdesign. Quasi-experimental Evidence Type Evidence Quasi-experimental, Quasi-experimental, Summary of Evidence

n body tilt position autonomic on function and cerebral oxygenation in preterm infants. 2002 changes of cerebral blood flow velocities in infants. premature Effecthead-up of Schrod & Walter, Article Citation Position dependent Position Eichler et al., 2001 appendix

N EONATAL NETWORK VOL. 30, NO. 6, NOVEMBER/DECEMBER 2011 393

Copyright © Springer Publishing Company, LLC ) continued ( each position. were supine or whether and prone in was head midline. tilting table. autoregulation. on Study Limitations Study Small sample size. in minutes 30 Only No power analysis. Unclear if ICU infantsICU if Unclear on only Concentrated wasstudy of Focus Small study.

-test U -test for -testfor 0.01) and cerebral 0.01) 0.05). 0.001) were both were 0.001)   25 mmHG) influenced

p p  p  significantlyincreased blood volume cerebral tilting a following maneuver, downward blood volume cerebral ( index oxygen hemoblobin ( in significantly altered preterm infants in both downtilting and up tilting wasmaneuvers.There between correlation a also age postconceptional and degree of change in blood volume cerebral ( using Student t paired analysis and Mann-Whitney using Statviewsoftware. using babies were monitored. percentage Increasing responses biphasic of in cerebral blood flow velocities as infant got responses Biphasic older. were considered seconday autoregulatory to preterm the In responses. high a neither infants, norPco2 a low MAP ( the percentage of biphasic biphasic of percentage the responses. Outcomes Although both groups had had groups Although both completed was analysis The Total of 501 episodes of 501 Total in 60 : : : NIRS : : : Right lateral. Measures

: Infantsmeasured : degreesfrom down

: Infantswere : Instrument Instrument episodes of head tilted up tilted head episodes of : From HOB flat to elevated.

measured before and after and before measured tilting by changes position degrees. 20 up bed 24 minutes, 30 degrees for 20 episodes horizontal 23 and minutes. 30 for episodes12 of head tilted up 20 degrees for 30 minutes, episodesand 10 horizontal for 30 minutes. Measured Cerebral blood blood Cerebral Measured cerebral and (CBV) volume index oxygen hemoglobin (cHbD). Pulse oximetry, capnography, and respiratory effort. before and afterposition and before changes by tilting. In NICUinfants,mattress In each for altered was position testing period 20 degrees up 20 and infants Healthy horizontal. a with chair a in placed were alternated mechanism tiliting lyingbetween and upright back position. UltrasoundDopplerprobe skin each attached to was over temporal bone. Method Head and Body HOB The preterm infants had The term infants had Study Intervention Investigators also recorded EKG, Biological Methods Head, Body, and HOB Study : : 50, 50, : : : : GA Weight Age

Infants

1 week. infants:25 mean preterm, weeks 33 GA and mean 1899 weight grams. Term: infantswith 13 mean GA 39 weeksand mean weight of grams. 2969 60 infants all with normal normal with all brain studies. weeks28 grams 1150  Thirty eight Population Infants NICU Mean Mean Mean degrees,

tilting bed up 20 any there are effects in the cerebral hemodynamics and term of infants preterm identified? can cerebral blood flow as changes, demonstrated uniphasic by biphasic or responses,be elicited after changes in posture? PICO QuestionPICO In neonates, - Cross- sectional study using healthy term full infants as compari son group. Level 2 Level Following Rating Strength/ Quality Level 2 Level (continued) non-randomized convenience control sample, group before- and-after design. non randomized randomized non convenience using sample repeated measuresdesign. Quasi-experimental, Quasi-experimental, Evidence Type Evidence Quasi-experimental Summary of Evidence

n x i d n on cerebral on hemodynamicsin term and preterm infants Muller, & 2001 Urlesberger, blood flow velocity responses in changes to posture Levene, 1993 ppe Effects of tilting Boetzelar, Pichler, Article Citation Neonatalcerebral Evans, Anthony, a

N EONATAL NETWORK 394 NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6

Copyright © Springer Publishing Company, LLC ) continued ( investigators reading the CBVV. reliabilityof fontanometer accurately to used measure intracranial pressure. difficult was it to hand the hold Doppler held transducer in place study. the during Investigatorswere not blinded to position infant interpretingwhen results. other many by studies. Study Limitations Study Unclear if Small sample size. regarding Unclear Investigators stated cited but Olderstudy,

degrees

90 —Occlusionof

turned

left

ipsilateral jugular resultedjugular ipsilateral in no change in flow in 25/26 subjects. jugular resulted in 100% occlusion in 20/26 patients decreasedwith correlated bloodflowcerebral Correlation velocity. coefficient of 0.91. turn occludes the jugular jugular the occludes turn side ipsilateral the on vein venous of occlusion and contralateral the on flow blood force not does side drainage through the subsequent with occlusion pressure cranial increased head External changes. increases also pressure pressure and impedes drainage. venous reduction: Bilateral occlusion of right left and jugular—100% all in reductionflow in subjects to Outcomes Occlusion of contralateral contralateral of Occlusion pressure fontanel in Increase Results revealed that 90˚ head head 90˚ that revealed Results Cerebral flood flow Supine Head : : : Measures Measures

: Infantswere : Instrument : midline head position, head midline : : prone, supine, or lateral. : flat throughout the study. Cerbralperfusion pressure was measure at each be to noted and change equivocal, no response, (reflecting uniphasic passive alteration in cerebral blood velocity),flowbiphasic or (reflectingpassive initial alteration, then active in change responseto cerebral blood flow velocity). measured after placement measured different 8 positions. in measured also Infantswere while applying external head venous jugular and pressure occlusion. head rotated 90 degrees to flexedand neck side, the extended. Ultrasound velocitiesbloodflow Cerebral (CBFV)superiorthe of sagittal sinus. Intervention Biological Study Methods Head Body HOB Biological : :

: : : : 10 : : Term

: 18 18 : GA Weight Age GA Weight Age

1 week Infants weeks39 grams3070  term healthy newborns weeks39 grams3683 6 days Healthy Mean Population Mean Mean Infants Mean Mean Mean changes superior in sagittal blood sinus velocities due postural to changes, jugular venous occlusion, increased and head manual the in pressure newborn? PICO QuestionPICO Are there Rating Strength/ Quality Level 2 Level (continued) non-randomized non-randomized convenience control sample, group before- and-afterdesign. Evidence Type Evidence Quasi-experimental, Quasi-experimental, Summary of Evidence

n x i sagittal sinus velocities blood postural to due and alterations onpressure the of head the newborninfant. 1985 ppend Article Citation Changes in superior Thoresen, & Cowen a

N EONATAL NETWORK VOL. 30, NO. 6, NOVEMBER/DECEMBER 2011 395

Copyright © Springer Publishing Company, LLC time periodtime for study. position order randomly was assigned. Study Limitations Study Small sample size Older study Small sample size. Older study. Power not calculated. brief a is minutes 10 avoid bias,To - - - 0.01) and 0.01) 0.001) for 0.001) 0.001) for both for 0.001) p  p  p  0.02). p  terior, lateral and occipital and lateral terior, variable to lead areas results. Pressure on poste rior fontanel always result decreasedcerebral in ed blood flow velocity and dependant positiondependant and when head turned to the right ( groups. and midline head with dependant in while turned position ( both groups. and midline in head when elevated ( decreasedin more even infants asphyxiated ( intracranial pressureintracranial was significantly lower positions midline in positions to compared turned. head the with pressureintracranial The significantly also was lower in patients whose to elevated was head degrees. 30 intracranial measured midline head was pressure elevatedbed of head with degrees. 30 Head pressure anterior pos Outcomes Significant ICP in horizontal in ICP Significant ICP in increase Significant ICP in decrease Significant The results suggest The position with the lowest : cerebral :

: Student Student : : : Ladd : Analysis Measures

test using Bonferroni using test t : Infants measured : ICP measured every Instrument : midline or 90 degrees 90 or midline : : midline or turned to the turnedto or midline : : supine : : supine : Flat, elevated 30 degrees 30 elevated Flat, : : flat or elevated elevated or flat : Ultrasound and transfontanel transducer. pressure Measured at least 30 seconds in each position. paired correction. or dependant 30 degrees. (CBf)bloodflow and intracranial pressure (ICP). before and after 6 different position changes. head turn to right. 1 minute for 10 minutes in minutes 10 for minute 1 each of 4 positions. A 4-minute stabilizationA period was given after every position change. right. 30 degrees. monitor to study intracranialstudy to monitor pressure. Intervention Biological Study Instrument Statistical Method Head Body HOB Method Head Body HOB Study

: : 8 : : 6 : : : : 14 14 : : days. days. GA GA Age GA Weight Age Age

weeks weeks preterm hours of life. life. of hours weeks

- preterm infants—with 6 a history of asphyxia and 8 without. Infants Asphyxiated Infants: 54 35 36 26 infants. hours43 of life. 33 grams 1497 3 days, with a range of 1-10 Infants Asphyxiated Mean Population Non Mean Mean Mean Infants Mean Mean Mean infants, doeshead affect position intracranial pressure? identified effectshead of on position intracranial in pressure the neonate, found is as in the adult literature? In newborn In PICO QuestionPICO Level 2 Rating Strength/ Quality Level 2 Level any there Are (continued) non-randomized non-randomized convenience control sample, group design. non-randomized non-randomized convenience sample design. Quasi-experimental, Quasi-experimental, Evidence Type Evidence Quasi-experimental, Quasi-experimental, Summary of Evidence

n x i d n affectsintracranial in pressure newborn infants. 1983 head position head intracranial on the in pressure was neonate studied. Moscoso, & Castillo, 1983 ppe Head position Head Emery & Peabody, Article Citation The effectThe of Goldberg, Joshi, a

N EONATAL NETWORK 396 NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6

Tara L. Marion, RN, MSN, NNP-BC Wanda T. Bradshaw, BSN, MSN, PMC, DNP

ongenital central hypoventilation syndrome PATHOPHYSIOLOGY C(CCHS), a unique disorder of respiratory control, CCHS is a disorder of the central nervous system in which results from polyalanine repeat expansion mutations in the the automatic control of breathing is absent or impaired. paired-like homeobox 2B (PHOX2B) gene in more than Normally, the desire to breathe occurs with increasing 90 percent of cases, and alternative PHOX2B mutations in carbon dioxide (CO2) levels in the brain. In CCHS, a child’s remaining cases. It is a disorder respiratory response to low in which affected individuals fail blood oxygen (hypoxemia) or to to breathe during sleep despite Ab s t r a c t CO2 retention (hypercapnia) is progressive hypercapnia and Congenital central hypoventilation syndrome (CCHS) typically delayed while they are hypoxia. Diagnosis of CCHS is is a rare syndrome of dysfunction of the autonomic awake, and absent, to varying established by clinical findings nervous system characterized by a decreased response to degrees, during sleep, serious and confirmatory molecular hypercarbia. It is a disorder in which affected individuals illness, and stress. Classic CCHS genetic testing. The diagnosis of fail to breathe during sleep despite progressive hypercapnia is characterized by adequate CC h s de p end s on t he do c u men- and hypoxia. Infants simply fall asleep and quit breathing. ventilation while the affected tation of hypoventilation during They are found by their parents or caregivers blue and individual is awake, and by sleep in the absence of primary lifeless. CCHS is an autosomal dominant disease. It has hypoventilation with normal neuromuscular, lung, cardiac, or been linked with tumors of neural crest origin, segmental respiratory rates and shallow aganglionosis of the colon, and diffuse autonomic metabolic disease, or an identi- breathing during sleep. More dysregulation but can occur alone. Discovery of the genetic fiable brainstem lesion. Because link between the paired-like homeobox 2B (PHOX2B) severely affected individuals many carriers of the disease genetic mutations and CCHS represents a breakthrough hypoventilate when both awake CCHS are asymptomatic, it is in the diagnosis of CCHS, association of mutated alleles and asleep. Both phenotypes difficult to determine the pres- with disease severity, and clues to the pathophysiology present in the newborn period.1 ence of the PHOX2B genetic responsible for the disorder. Early genetic screening and The autonomic nervous system mutation prior to birth. Most intervention can provide the families of these infants with controls breathing during quiet infants are diagnosed within hope for achieving a normal life. sleep. For this reason, ventilation the first 48 hours of life, after a is severely affected. Ventilation is dusky period during sleep. This better in active or rapid eye move- is a critical time in the care of an infant with CCHS. Many ment (REM) sleep when cortical input is at its greatest although times, a diagnosis is made after asphyxia has occurred and still not normal. In the newborn period, many affected infants irreversible brain damage has been done. will not have the classically described sleep–wakefulness dif- The purpose of this article is to provide an overview ferences; thus, they may appear to have chronic intermittent of CCHS and to discuss the implementation and value of genetic screening to assist in the diagnosis. All full-text Disclosure English language articles from 1999 to 2010 were reviewed The author discloses no relevant financial interest or affiliations with any using CINAHL and PubMed. commercial interests.

Accepted for publication March 2011.

N EONATAL NETWORK VOL. 30, NO. 6, NOVEMBER/DECEMBER 2011 © 2011 Springer Publishing Company 397 http://dx.doi.org/10.1891/0730–0832.30.6.397 Copyright © Springer Publishing Company, LLC duskiness, cyanosis, and measurable hypercapnia.2 As oxygen radiograph, fluoroscopy of the diaphragm, bronchoscopy, saturations fall and carbon dioxide levels rise, affected infants electrocardiogram, Holter recording, echocardiogram, and demonstrate no increase in respiratory rate or effort and usually magnetic resonance imaging (MRI) of the brain and brain- do not arouse or appear distressed. These children, if unde- stem. Serum and urinary organic acids, amino acids, and tected or misdiagnosed, will present again at a later age with carnitine levels should be obtained to rule out inborn errors signs of right-sided heart failure and pulmonary hypertension of metabolism. Genetic testing is recommended for confir- from prolonged periods of hypoxia and hypercapnia.2 Infants matory diagnosis. with CCHS typically present with apnea and respiratory arrest shortly after falling asleep. The primitive responses to hypoxia CHARACTERISTIC FACIAL PHENOTYPES and hypercapnia that ordinarily stimulate respiratory drive in Todd and colleagues describe a characteristic facial phe- normal breathing are altered, and the infant fails to breathe notype in individuals with CCHS, which includes facies that effectively, or even struggle to do so. For this reason, patients are generally shorter and flatter, with significantly decreased with CCHS often do not display signs of respiratory distress, upper-face height, excessive nasal tip protrusion, decreased such as tachypnea, nasal flaring, or retractions. They do not nasolabial angle, short upper-lip height, and an inferior inflec- struggle or gasp, they simply quit breathing. Without the aid tion of the lateral segment of vermillion border on the upper of objective monitoring, hypoxia and hypercapnia are detected lip.5 Usi ng fi ve va r iable s to cha r ac ter i z e f ac ie s (upp er-l ip heig ht , only at a late stage, after the onset of severe cyanosis and central biocular width, upper facial height, nasal tip protrusion, and nervous system depression. By this time, it can be too late, and the lip trait), Todd and colleagues found that 85.7 percent irreversible neurologic damage may have been done. Children of individuals with CCHS and 82.2 percent of controls were with CCHS often have physiologic and anatomic manifesta- correctly predicted.5 These authors also found 86 percent tions of a generalized autonomic nervous system dysfunction/ of patients with CCHS could be identified by facial features dysregulation (ANSD). Some may have altered development (box shaped, generally shorter and flatter) and correctly dis- of neural crest-derived structures (i.e., Hirschsprung disease tinguish them from 82 percent of controls. These authors also [HSCR]) and tumors of neural crest origin, including neuro- suggest that facial features may be distinctive in individuals blastoma, ganglioneuroma, and ganglioneuroblastoma. with CCHS because the dorsal rhombencephalon and caudal midbrain give rise to neural crest tissue that develops into the DIAGNOSING CONGENITAL CENTRAL facial structures, and the PHOX2B gene is expressed when HYPOVENTILATION SYNDROME these structures are developing in the embryo.5 AND IMPLICATIONS FOR CARE Clinical signs and symptoms and later performed molecular COMORBIDITIES AND CONGENITAL genetic testing confirm the diagnosis of CCHS. The diagno- CENTRAL HYPOVENTILATION SYNDROME sis of CCHS depends on the documentation of hypoventila- A case-control study found 89 percent of CCHS patients tion during sleep in the absence of primary neuromuscular, had cardiovascular symptoms (e.g., decreased heart rate vari- lung, cardiac or metabolic disease, or an identifiable brain- ability, vasovagal syncope, cardiac dysrhythmias), 84 percent stem lesion.3 Other reasons for hypoventilation during sleep, exhibited gastrointestinal autonomic symptoms (constipa- including infection, asphyxia, and trauma, must be delin- tion, dysphagia, or gastroesophageal reflux), 82 percent eated before the diagnosis of CCHS is made. The American demonstrated altered temperature regulation (lack of fever Thoracic Society has issued a statement on the diagnosis of with infection), and 52 percent had altered pain perception CCHS, and preparation of a revised statement is in process.4 as compared respectively with 5 percent, 5 percent, 0 percent, CCHS is diagnosed in individuals with the following: and 4 percent of the control subjects. Eighty-six percent had • Hypoventilation with absent or negligible ventilatory sen- ophthalmologic abnormalities (sluggish or unreactive pupils, sitivity to hypercarbia and absent or variable ventilatory abnormal tearing, strabismus, anisocoria, miosis) versus sensitivity to hypoxemia 2 percent of the control group.6 • Generally adequate ventilation while awake, but hypoven- HSCR is characterized by the absence of intramural tilation with normal respiratory rate and shallow breathing ganglion cells in the distal gut, which can result in bowel (diminished tidal volume) during sleep, or obstruction shortly after birth. A combination of CCHS • Hypoventilation both while awake and asleep and HSCR is a rare condition with variable severity. Both • Absent perception of asphyxia (i.e., absent behavioral CCHS and HSCR are uncommon, and their co-occurrence awareness of hypercarbia and hypoxemia) and absent may suggest a common etiology, probably involving a fault of arousal neural crest development.7 These two disorders are thought • No evidence of primary neuromuscular, lung, or cardiac to be related within a class of diseases known as neurocris- disease or identifiable brainstem lesion that might account topathy. Neurocristopathies result from the malfunction of for the constellation of symptoms neural crest cells.8 Examples of neurocristopathies include The initial evaluation may include a detailed neuro- pheochromocytoma, neuroblastoma, medullary carcinoma of logic evaluation that may require a muscle biopsy, chest the thyroid, and carcinoid tumors.8

N EONATAL NETWORK 398 NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6

Copyright © Springer Publishing Company, LLC GENETIC TESTING AND CONGENITAL of children with identified polyalanine expansion mutations CENTRAL HYPOVENTILATION SYNDROME should be screened by the PCR assay PHOX2B Screening Prior to genetic testing and screening and discovery of the Test to determine mosaicism.3 PHOX2B gene, CCHS was a diagnosis of exclusion, meaning all other causes of hypoventilation were ruled out, includ- Sequence Analysis ing X-linked myotubular myopathy, multiminicore disease, The process of sequence analysis evaluates the order of congenital myasthenic syndrome, altered airway or intratho- nucleotide bases, the structural units of both RNA and racic anatomy, diaphragmatic dysfunction, congenital cardiac DNA. Three base pairs form a codon. The polyalanine repeat disease, brainstem abnormality, and various metabolic dis- responsible for CCHS can be made up of any one of four eases. This process was timely and costly and could have codon combinations—GCA, GCT, GCC, or GCG—as each wasted valuable time in the management and treatment of one encodes the amino acid alanine. The automated process the infant. of sequence analysis is used in cases of CCHS when the PCR PHOX2B mutations can now be detected by two methods: is negative for the PHOX2B mutation. Sequence analysis polymerase chain reaction (PCR) assay and sequencing of the entire coding region and intron-exon boundaries of microarray analysis. These tests are clinically available with PHOX2B can detect mutations in the 8 percent of individu- the most cases being detected by PCR. Introduction of clini- als with the CCHS phenotype who do not have an expansion cally available PHOX2B genetic testing allows for distinction mutation.3 between CCHS and other disorders in the differential diagnosis, such as severe prematurity, identifiable brainstem find- LIMITATIONS OF THE SCIENCE ings, asphyxia, infection, trauma, tumor, and infarction.9,10 Because many carriers of the disease CCHS are asymptom- Because more than 90 percent of individuals with CCHS atic, it is difficult to determine the presence of the PHOX2B have a PHOX2B polyalanine expansion mutation and because genetic mutation prior to birth. Most infants are diagnosed PHOX2B polyalanine expansion testing (PCR) is a more sen- within the first 48 hours of life, after a dusky period following sitive test for detection of mosaicism, such testing should be falling to sleep, and a possible anoxic event. Genetic testing performed first. Only if a PHOX2B polyalanine expansion for CCHS can be costly (U.S. $399). It is also important to mutation is not found in an individual with the CCHS phe- note that results for a full gene analysis can take 10–21 days, notype should sequencing of the entire coding region and and specific mutation analysis 10–14 days. This is a critical intron-exon boundaries of the PHOX2B gene be performed.3 time in the care of an infant with CCHS. Prenatal diagnosis for pregnancies at increased risk is possible by analysis of Polymerase Chain Reaction DNA extracted from fetal cells obtained by amniocentesis A PCR assay produces thousands to millions of copies of usually performed at about 15–18 weeks gestation or chori- a small, specific DNA segment in a matter of hours. It is the onic villus sampling (CVS) at about 10–12 weeks gestation. first genetic test that should be carried out. A PCR should be The disease-causing allele of an affected family member must performed on both the parents and the infant. It permits rapid be identified before prenatal testing can be performed. detection of genetic aberrations. Each PCR assay is specific for one particular abnormality. In CCHS, the PCR assay is used CARE RECOMMENDATIONS FOR MEDICAL to directly amplify and size the second polyalanine-coding AND NURSING MANAGEMENT triplet repeat sequence in exon 3 of the PHOX2B gene. This The treatment of CCHS is to ensure adequate ventila- triplet repeat is expanded in most (about 90 percent) indi- tion for the patients who are unable to achieve adequate gas viduals with CCHS. The remaining individuals with CCHS exchange during spontaneous breathing, or simply put, to will have mutations not detected by PCR. Those individu- breathe for them. als with suspected CCHS and a negative PCR should have It is important to note that simply providing supplemental follow-up sequencing of the coding regions of the PHOX2B oxygen to individuals with CCHS is not warranted. CCHS is a gene. It should be noted that 5–10 percent of CCHS patients problem with ventilation. Thus, children with CCHS require inherited their PHOX2B mutation from a parent who has home mechanical ventilation—possibly for life. As children mosaicism or a “lesser dose” of the mutation, which explains with CCHS do not usually have severe lung disease, they have why the parents are not affected with the CCHS phenotype. many options for different techniques to provide mechanical Because parents with PHOX2B mosaicism can pass the same assisted ventilatory support at home.2 Infants are best venti- PHOX2B mutation onto other children, it is necessary to test lated with positive pressure ventilation (PPV) via tracheostomy all parents of CCHS probands for mosaicism. The PCR assay and should have the tracheostomy placed soon after diagnosis. PHOX2B Screening Test (and not sequencing) is the best It is usually preferable to ventilate infants for 24 hours per available assay for identifying and quantifying mosaicism. day for some time, assuring that oxygenation and ventilation Therefore, children suspected to have CCHS should ideally remain adequate to minimize possible damage to a developing be tested by the PCR assay PHOX2B Screening Test, with brain. Passy-Muir (Irvine, California) one-way speaking valves follow-up sequencing if no mutation is found. All parents and tracheostomy capping can be done while awake patients

N EONATAL NETWORK VOL. 30, NO. 6, NOVEMBER/DECEMBER 2011 399

Copyright © Springer Publishing Company, LLC are older to allow for vocalization and use of the upper airway. CONCLUSION Individuals with CCHS are not like other children on home Since 2003 and the discovery of the PHOX2B gene, the mechanical ventilation. They must be managed with extreme treatment and genetic basis for CCHS has evolved new under- vigilance because of their lack of objective or subjective standing and implications. Infants and children with CCHS responsivity to hypoxemia and hypercapnia. Transition to a are living healthier and less complicated lives because of the home portable ventilator should be made in the hospital, and advances in care. With more genetic screening options being discharge planning for home is extensive. Arrangements for offered to families, and hopes of more to come, earlier detec- the home ventilator, a backup ventilator, 16–24 hours per day tion of this life-altering disease is becoming a reality. Because of in-home nursing, supplemental oxygen, a pulse oximeter, CCHS is rare, it has not received research funding from private and an end-tidal carbon dioxide monitor are required in the sources and must rely heavily on grants and donations from home prior to discharge. families. Although it is true that CCHS is a rare disease, the Responses to respiratory infections in children with CCHS families that it affects are severely disrupted and rearranged. differ from non-CCHS ventilator-dependent children. Children A national support group exists to assist families in dealing with CCHS do not typically develop a fever, increase their with this diagnosis and its implications. The CCHS Family respiratory rate or have dyspnea in response to pneumonia, or Network works to arm families and individuals living with exhibit symptoms of severe hypoxia and/or hypercapnia. These CCHS with information to make living with this diagnosis limitations emphasize the importance of both objective pulse possible. Early detection of affected infants with CCHS is the oximetry and end-tidal carbon dioxide monitoring, as well as single most important aspect of their care and management. highly skilled and consistent caretakers in the home.2 The earlier intervention occurs, the better the outcome and All individuals with CCHS require assisted ventilation during prognosis of the child. Educating families about CCHS and sleep for life; thus, weaning these individuals off mechanical providing them with resources in the community will arm ventilation totally is not a realistic goal. For these children, them with a lifetime of knowledge and resourcefulness. ventilators are adjusted to provide end-tidal carbon dioxide values consistently between 30–35 torr and oxygen saturation values greater than 95 percent.2 Whereas it should be empha- REFERENCES 1. Weese-Mayer, D. E., Berry-Kravis, E. M., & Marazita, M. L. (2005). sized that children with CCHS are not candidates for weaning In pursuit (and discovery) of a genetic basis for congenital central off mechanical-assisted ventilation while asleep, mobility and hypoventilation syndrome. Respiratory Physiology & Neurobiology, quality of life are maximized if the child can breathe unassisted 149(1–3), 73–82. http://dx.doi.org/10.1164/rccm.200807-1069ST for time while awake. Some CCHS children gradually develop 2. Chen, M. L., & Keens, T. G. (2004). Congenital central hypoventilation the ability to breathe adequately during wakefulness.9 syndrome: Not just another rare disorder. Paediatric Respiratory Reviews, Increased mobility and improved quality of life can be 5(3), 182–189. achieved by using diaphragm pacing. Diaphragm pacing 3. Berry-Kravis, E. M., Zhou, L., Rand, C. M., & Weese-Mayer, D. E. (2006). Congenital central hypoventilation syndrome: PHOX2B mutations and entails surgical implantation of an electrode on the phrenic phenotype. American Journal of Respiratory and Critical Care Medicine, nerve, which is connected to a subcutaneous receiver. There is 174(10), 1139–1144. http://dx.doi.org/10.1164/rccm.200602-305OC an external battery-operated transmitter and antenna placed 4. Weese-Mayer, D. E., Shannon, D. C., Keens, T. G., & Silvestri, J. M. on the skin over the receiver. The transmitter emits energy, (1999). American Thoracic Society Statement. Idiopathic congenital similar to a radio transmission, which is converted into an central hypoventilation syndrome. Diagnosis and management. American electrical current by the receiver. This stimulates the phrenic Journal of Respiratory and Critical Care Medicine, 160(1), 368–373. pmid:10390427 nerve resulting in a diaphragmatic contraction. Settings on the 5. Todd, E. S., Weinberg, S. M., Berry-Kravis, E. M., Silvestri, J. M., transmitter include respiratory rate and electrical voltage and Kenny, A. S., Rand, C. M., . . . Weese-Mayer, D. E. (2006). Facial are adjusted to give enough tidal volume to allow for adequate phenotype in children and young adults with PHOX2B-determined oxygenation and ventilation. Therefore, diaphragm pacing is congenital central hypoventilation syndrome: Quantitative pattern of an attractive alternative mode of mechanically assisted venti- dysmorphology. Pediatric Research, 59(1), 39–45. http://dx.doi.org/ lation for many patients with CCHS.11 Diaphragm pacing is 10.1203/01.pdr.0000191814.73340.1d not typically recommended for the young child who requires 6. Grigg-Damberger, M., & Wells, A. (2009). Central congenital hypoventilation syndrome: Changing face of a less mysterious but more only nighttime ventilatory support because the benefits do complex genetic disorder. Seminars in Respiratory and Critical Care not outweigh the risks; however, for older adolescents and Medicine, 30(3), 262–274. http://dx.doi.org/10.1055/s-0029-1222440 young adults, this could be an appropriate consideration. 7. Ou-Yang, M. C., Yang, S. N., Hsu, Y. M., Ou-Yang, M. H., Haung, Education for parents and caretakers should include avoid- H. C., Lee, S. Y., . . .Liu, C. A. (2007). Concomitant existence of total ing the use of alcohol and other impairing substances while bowel aganglionosis and congenital central hypoventilation syndrome in caring for the child. Small toys and other items should be a neonate with PHOX2B gene mutation. Journal of Pediatric Surgery, 42(2), e9–e11. http://dx.doi.org/10.1016/j.jpedsurg.2006.10.022 removed from within the child’s reach because of their ability 8. Lai, D., & Schroer, B. (2008). Haddad syndrome: A case of an infant with to occlude the tracheostomy. Education about their equip- central congenital hypoventilation syndrome and Hirschsprung disease. ment and disease should begin as the child attains the ability Journal of Child Neurology, 23(3), 341–343. http://dx.doi.org/10.1177/ to understand. 0883073807309242

N EONATAL NETWORK 400 NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6

Copyright © Springer Publishing Company, LLC 9. Bachetti, T., Robbiano, A., Parodi, S., Matera, I., Merello, E., Capra, V., . . . Ottonello, G. (2006). Brainstem anomalies in two patients affected by congenital central hypoventilation syndrome. American Journal of Respiratory and Critical Care Medicine, 174(6), 706–709. http://dx. doi.org/10.1164/rccm.200602-266CR 10. Bajaj, R., Smith, J., Trochet, D., Pitkin, J., Ouvrier, R., Graf, N., . . . Kluckow, M. (2005). Congenital central hypoventilation syndrome and Hirschsprung’s disease in an extremely preterm infant. Pediatrics. 115(6), e737–e738. http://dx.doi.org/10.1542/peds.2004-1910 11. Chen, M. L., Turkel, S. B., Jacobson, J. R., & Keens, T. G. (2006). Alcohol use in congenital central hypoventilation syndrome. Pediatric Pulmonology, 41(3), 283–285. http://dx.doi.org/10.1002/ppul.20366

About the Author Tara L. Marion is a graduate of Duke University School of Nursing. She currently practices as a neonatal nurse practitioner with Brenner Children’s Hospital, Wake Forest Baptist Health. She is the mother of Max, who is two and a half, and Georgia, who is four months.

For further information, please contact: Tara L. Marion, RN, BSN, NNP-BC E-mail: [email protected]

N EONATAL NETWORK VOL. 30, NO. 6, NOVEMBER/DECEMBER 2011 401

eonatal alloimmune percent of NA IT cases,14 followed Nt h r o m b o c y t o p e n i a Neonatal Alloimmune by HPA-5a, 15-a, and 15-b.2 In (NAIT) is a life-threatening disor- A s i a n s , H PA - 4 i s t h e p r e d o m i n a nt der caused by fetomaternal plate- cause of NAIT.3 HPAs are num- let incompatibility analogous to Thrombocytopenia: bered from 1 to 24 in the order in that seen in rhesus (Rh) disease. which they were discovered. The In NAIT, maternal immunoglob- A Case Study letters “a” and “b” describe the ulin G (IgG) antiplatelet antibod- frequency of occurrence with “a” Jodi Beachy, RNC, MSN, NNP ies cross the placenta, resulting in being high and “b” being low.9 rapid destruction and removal of HPA-1a, HPA-3a, and HPA-4a fetal platelets by the reticuloen- Ed i t o r cause severe thrombocytopenia dothelial system.1,2 Studies have Patricia Nash, MSN, NNP-BC and HPA-5a and -5b are milder shown that NAIT has an incidence and rarely have intracranial hem- of 1 of 1,000 live births,3,4 with a orrhage (ICH).3 mortality rate of 10–15 percent1 and the risk of long-term morbidities up to 20–60 percent if CLINICAL PRESENTATION intracranial hemorrhage (ICH) occurs.3,5 A newborn with NAIT will usually appear healthy. The This column will discuss the pathophysiology, differential mother will also be healthy and have a normal platelet count. diagnosis, morbidities, and treatment of NAIT and conclude The infant will often present with petechiae, bruising, exces- with a relevant case study. sive bleeding, and mucocutaneous purpura within the first few hours of life. When the newborn presents with these signs PATHOPHYSIOLOGY and symptoms, the platelet count, when checked, is typically Platelets are small cell fragments of very large bone less than 20,000.2 The platelet level may begin to drop early marrow cells called megakaryocytes.6 Platelets normally in pregnancy and may or may not self-correct.3 Generally, live for 10 days and then are removed by the spleen and though, a rapid postnatal drop in platelet levels is seen and liver. Neonatal thrombocytopenia historically is defined as caused in part by the increased exposure of other platelets a platelet count of less than 150,000.7 However, in recent from reticuloendothelial cells in the blood flow to the lungs.2 studies, neonatal thrombocytopenia has been defined as a Fourteen to twenty percent of newborns with NAIT will platelet count less than or equal to 123,000,7 with severe develop an ICH.2,5 Fifty percent of ICH occur in utero. The neonatal thrombocytopenia as a platelet count less than presentation of ICH is variable ranging from no clinical signs 50,000.7, 8 to obvious signs, such as seizures, retinal hemorrhage, leth- NAIT subsequently occurs when the mother produces argy, tense fontanel, stupor, apnea, and bradycardia.15 antiplatelet antibodies to the fetal platelets similar to what There is currently no routine prenatal screen for NAIT; is seen with hemolytic disease of the newborn (HDN). therefore, the diagnosis is usually not made until after birth.9 Unlike HDN, though, NAIT can occur in the first preg- The mother’s pregnancy is at high risk for NAIT if there is nancy; in fact, 50 percent of all NAIT cases are seen in a history of a previous infant with NAIT or an infant with the first pregnancy.1 This is because immunoglobulin G thrombocytopenia of unknown etiology. (IgG) is the only maternal antibody that crosses the normal placenta as early as 14–16 weeks of gestation. At the same DIFFERENTIAL DIAGNOSIS time, platelet antigens are seen in the fetus by 18 weeks of The causes of newborn thrombocytopenia in the oth- gestation, setting the stage for NAIT to develop. On the erwise healthy newborn differ from thrombocytopenia other hand, Rh sensitization occurs when there is a breech seen in the sick newborn. The mother’s pregnancy history in the uterine wall, which most often occurs after delivery and physical assessment can help determine a diagnosis. of the placenta. This is why with HDN, the first pregnancy With NAIT, the mother usually experiences an unevent- is not affected.2,9–11 ful pregnancy with normal platelet levels.3 Alternately, For NAIT to develop, human platelet antigens (HPA, thrombocytopenia may be seen in the neonate of women formally called PLA1, human platelet antigen 1a) must be with a history of pregnancy-induced hypertension (PIH), absent in the mother but present in the father and subse- drug use, or infection.16 The most common cause of severe quently present in the fetus.12 Human platelet antigens (HPA-1a) are found in 98 percent of the Caucasian pop- Disclosure 13 ulation and caused 75 percent of all NAIT cases. The The author discloses no relevant financial interest or affiliations with any HPA-5b is second in frequency and is responsible for 16 commercial interests.

Accepted for publication May 2011.

N EONATAL NETWORK 402 © 2011 Springer Publishing Company NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6 http://dx.doi.org/10.1891/0730–0832.30.6.402 Copyright © Springer Publishing Company, LLC thrombocytopenia in the well newborn is NAIT.11 NAIT Intravenous Immune Globulin accounts for 20 percent of cases of thrombocytopenia in Intravenous immune globulin (IVIG) is IgG in con- the healthy newborn.5 The second most common cause, centrated form.27 IVIG is administered to decrease the resulting in 10 percent of the cases of neonatal thrombocy- production of and neutralize circulating antiplatelet antibod- topenia, is maternal idiopathic thrombocytopenia purpura ies. IVIG also works by blocking platelet Fc receptors and (ITP).17 Although only 10 percent of ITP mothers have therefore decreasing destruction of platelets.9,28 IVIG has infants with thrombocytopenia, it can be severe.6,18 The been shown to reduce the frequency of ICH.15 The effect of differential diagnosis for healthy newborns with thrombo- IVIG takes place in one to three days.23,28 Typical dosage is cytopenia are presented in Table 1.19 500–750 mg/kg, but for infants with NAIT, dosages range from 400 to 1,000 mg/kg. IVIG is given over 2–6 hours LABORATORY TESTING and is compatible with dextrose 5% water (D5W), D15W, and The infant’s complete blood count (CBC) is generally total parenteral nutrition (TPN).27 IVIG can be given up to normal other than the low platelet count; although if bleed- two days in a row.20 ing is severe, anemia may be present. The parents, not the The infant’s temperature, blood pressure, and respira- infant, are screened for antigens.20 NAIT is confirmed by tions must be monitored closely with all transfusions.29 identifying antiplatelet antibodies in the mother’s blood When giving IVIG, in addition to monitoring the vital as well as antigen incompatibility between the mother and signs and intravenous site, also monitor the infant for the the father.12,20 The mother should be screened for HPA-1, rare side effects of bronchospasm, renal failure, and laryngeal HPA-3, and HPA-5, as well as for HPA-4, if the mother is edema.2,27 Asian.20 The father should also be screened. It is important to find the specific antigen to protect future pregnancies.21 Fresh Frozen Plasma This is best done through a laboratory that has DNA testing Generally, fresh frozen plasma (FFP) comes from whole and the ability to find rare antigens if needed.21,22 blood and is frozen within 6–8 hours after collection.25 FFP In the early stages of neonatal thrombocytopenia, infec- contains one international unit of clotting factors for every tion and bleeding disorders must be ruled out. However, 10–15 mL/kg24; FFP also contains albumin and many other most healthy newborns with severe thrombocytopenia have plasma proteins. FFP is most often used to prevent bleed- NAIT.21 ing in the infant with severe thrombocytopenia of unknown origin. The recommended dose of FFP is 10–20 mL/kg.25 TREATMENT Platelet Transfusions Cryoprecipitate Treatment of NAIT should begin without delay.15 A plate- Cryoprecipitate is thawed FFP, refrozen with plasma. The let transfusion should be given if the infant’s platelet count is plasma contains high concentrations of clotting factors, espe- less than 20,000–30,000 or less than 50,000 if the infant is cially Factors VII and VIII and fibrinogen.29 Cryoprecipitate bleeding or in critical condition.9,23 In an emergent situation, is used to treat infants with low fibrinogen clotting disorders the first course of action is to transfuse irradiated random and is not used once NAIT has been confirmed. donor platelets. HPA-1a negative donor platelets or washed and irradiated maternal platelets are ideal because the recovery SCREENING FOR INTRACRANIAL is quicker, the effect is faster, and it lasts longer. These HPA-1a HEMORRHAGE negative transfusions do not have the offending antigen, Intracranial hemorrhage is the most devastating compli- which is responsible for platelet destruction. However, the cation of NAIT, most of which occur in utero and can result negative platelets are not always available when the infant is in mortality or major morbidity. Consequently, all infants bleeding and needs immediate assistance.2,3,9 Therefore, if with a confirmed diagnosis or suspected NAIT should be random donor platelets are used that are HPA-1a positive, screened by cranial ultrasound, computed tomography (CT) the infant will have a temporary improvement, but the plate- scan, or magnetic resonance imaging (MRI).22 Cranial let destruction will continue until the infant has cleared all ultrasounds are usually the least expensive and fastest antigens from the circulation. screen available for visualizing a hemorrhage. However, CT The target platelet level is 100,000 or greater.22 A platelet imaging and MRI can be more sensitive in finding some infusion of 5–10 mL/kg should raise the platelet count by smaller intraventricular hemorrhages (IVHs) and subarach- 50,000–100,000,24 although most references recommend noid bleeds.30 platelets of 10 mL/kg per transfusion.25 Platelet transfu- Intracranial hemorrhage is found in 1 of 1,500 term new- sions are repeated until the level is greater than 100,000.22 borns. Twenty-five percent are caused by NAIT, making it It is especially important to maintain the platelet levels the most common cause of severe ICH. Parenchymal hemor- during the first 3–4 days of life to minimize the risk of rhage and IVH are the most common ICH seen in NAIT.31 ICH.26 Once the platelet level has reached a normal level, it The connection between thrombocytopenia and ICH is will remain there. most likely caused by the lack of protection of the vessel walls

N EONATAL NETWORK VOL. 30, NO. 6, NOVEMBER/DECEMBER 2011 403

Copyright © Springer Publishing Company, LLC TABLE 1 n Differential Diagnosis for Well Newborn Presenting with Thrombocytopenia1,2,6,18,19,20,21,22,26 Diagnosis Distinguishing Features Typical Platelet Count and Other Notes (Maternal History Symptoms) Laboratory Exams Maternal idiopathic Maternal platelets low (though Infant platelets 50,000–100,000 Autoimmune thrombocytopenia purpura may be recovering) Platelets drop during first days of life ICH 3% NAIT Maternal platelet levels normal Infant platelets under 20,000 Seen within first several hours of life ICH 14% Neonatal drug exposure Positive history Heparin Quinine Rare TARS (thrombocytopenia-absent Severe thrombocytopenia from Absent radius radius syndrome) decreased production ICH CAMT (Congenital amegakaryocytic Severe thrombocytopenia Few megakaryocytes in bone marrow thrombocytopenia) Rare Mimics NAIT Maternal drugs Quinidine Penicillin Dioxin Indomethacin Phenytoin Heparin Chromosomal abnormalities: Clinical features of syndrome Less than 100,000 Low-percentage platelets Trisomy 18 87% Trisomy 13 54% Trisomy 21 28% Turner syndrome 31% Placental insufficiency—IUGR and Maternal history, small infant Borderline low Recover by Day 10 PIH Wiskott-Aldrich syndrome Moderate thrombocytopenia Eczema and immunodeficiency Fanconi’s anemia Congenital anomalies (muscular, microcephaly, and GU) Rare Cardiac anomalies Thrombosis Indwelling catheters or protein C deficiency Kasabach-Merritt syndrome Severe thrombocytopenia Lesions on trunk Prolonged PT and PTT Giant hemangiomas

Key: GU 5 genitourinary; ICH 5 intracranial hemorrhage; IUGR 5 intrauterine growth restriction; NAIT 5 Neonatal alloimmune thrombocytopenia; PIH 5 pregnancy-induced hypertension; PT 5 prothrombin time; PTT 5 partial thromboplastin time. Note. Neonatal alloimmune thrombocytopenia; Maternal idiopathic thrombocytopenia purpura; Systemic lupus erythematosus; Giant hemangioma; Thrombosis (large renal vein thrombus); Neonatal drug exposure (heparin and quinine); Congenital thrombocytopenia—bone marrow failure (thrombocytopenia-absent radius syndrome and congenital amegakaryocytic thrombocytopenia); Maternal pregnancy-induced hypertension; IUGR from placental insufficiency; Preeclampsia or chronic hypertension; Maternal drug exposure (heparin, penicillin, dioxin, antiseizure medications, and quinine); Congenital heart disease; Chromosome abnormalities (trisomies 21, 18, and 13 and Turner syndrome); Wiskott-Aldrich syndrome; Fanconi’s anemia.

N EONATAL NETWORK 404 NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6

Copyright © Springer Publishing Company, LLC normally provided by platelets. It is unknown whether the Group B beta Streptococcus [GBBS] negative). Vacuum was antiplatelet antibodies actually cause the platelets to be less used for nonreassuring fetal heart tones. The infant delivered effective in protecting the vessel walls.31 in the vertex position with a loose nuchal cord. At delivery, Maternal antibodies begin to leave the infant’s circulation she cried loudly and became pink quickly. Routine care was at 48 hours of age.32 Most courses of NAIT resolve by two given, and the Apgar scores were 8 at one minute and 9 at five weeks of age, with platelet levels usually normalizing com- minutes. The infant stayed with the mother in the labor and pletely by four weeks.20,24 delivery unit for the first hour of life. The one-hour assessment was normal except for bruis- COMPLICATIONS AND LONG- ing on the scalp. As the mother’s blood type was O1, the TERM OUTCOMES cord blood was sent and noted to be A1 with a negative Long-term outcomes are dependent on the severity of Coombs. By six hours of age, the infant was feeding poorly NAIT and the swiftness of treatment.3 If there is no ICH, and was hypothermic. On assessment, she was pale pink, with morbidity is low, although these infants are at a small risk decreased tone, irritability, and hyperresponsiveness to stim- for vision problems.3,33 One-third of infants with NAIT ulation; her anterior and posterior fontanels were full. and ICH will die. The rest are at risk for hydrocephaly, Bruising was noted over the entire scalp, with petechiae especially if the ICH occurs in utero; developmental delays, covering the entire chest, abdomen, arms, and legs. The mental retardation, cerebral palsy, and seizures may also be infant had clear breath sounds, mild subcostal retractions seen.2,33 with periodic breathing, and one episode of apnea; the rest of the exam was normal. The workup included a CBC with dif- MATERNAL FOLLOW-UP AND PREVENTION ferential, prothrombin time (PT), international normalized IN FUTURE PREGNANCIES ratio (INR), activated partial thromboplastin time (aPTT), Women who have had a newborn with NAIT must be taught blood gas, fibrinogen, glucose level, blood culture, and CT the importance of early and regular prenatal care;2 especially scan of the brain. The CBC showed a left shift with a plate- because all future pregnancies are at 90–100 percent risk for let count of 16,000; all other laboratory results were within severe fetal and neonatal thrombocytopenia, as the symptoms an acceptable range. The CT scan showed a very large acute worsen with each subsequent pregnancy.4,12 Close maternal subdural hematoma on the right side of the cerebellum. The follow-up with high-risk obstetrics is especially crucial if the subdural hematoma extended along the falx (located between first infant had an ICH.34 Antenatal management focuses on the cerebellar hemispheres) and the cerebellum was shifted minimizing the risk of ICH and currently includes treatment to the left. The compressed cerebellum caused obstruction of with IVIG and prednisone throughout the pregnancy to the cisterns and increased the size of the temporal horns of block maternal production of antibodies.4,9,20 the lateral ventricles, resulting in increased intracranial pressure (ICP) (Figure 1). The platelet count was repeated and NEONATAL FOLLOW-UP dropped from 16,000 to 6,000 in one hour. Platelets were Although NAIT is resolved by two weeks, the outpatient ordered but not available before the transport team arrived. follow-up should include platelet levels for the rare but potential risk of the platelet levels dropping. Developmental and neurologic follow-up is also warranted if the infant experi- FIGURE 1 n CAT scan shows intracranial hemorrhage. Blood from enced an ICH. an acute bleed is seen as white once a clot is formed.

CONCLUSION NAIT causes ICH and death in the otherwise healthy newborn. It is vital that immediate action is taken when NAIT occurs in the first days of life. With quick and proper treatment, the risks of death and long-term disabilities are diminished.

NEONATAL ALLOIMMUNE THROMBOCYTOPENIA: A Case Study Baby girl A delivered vaginally at 41 weeks gestation, with vacuum assistance, to a 36-year-old, O1, Caucasian, primi- gravidous mother. The pregnancy was complicated by hypo- thyroidism, treated with Synthroid. All maternal serologies were negative (rubella immune, Venereal Disease Research From Manco-Johnson, M., Rodden, D., & Collins, S. (2007). Newborn Laboratory [VDRL] test negative, human immunodeficiency hematology. In G. Merenstein & S. Gardner (Eds.), Handbook of virus [HIV] negative, hepatitis B virus [HBV] negative, neonatal intensive care (6th ed., pp. 521–547). St. Louis, MO: Mosby.

N EONATAL NETWORK VOL. 30, NO. 6, NOVEMBER/DECEMBER 2011 405

Copyright © Springer Publishing Company, LLC The infant was then transferred to a Level III NICU. With 12. Cota, F., Zuppa, A. A., Luciano, R., Gallini, F., Savarese, I., Alighieri, G., the results of the CT scan in hand, neurosurgery made an . . . Romagnoli, C. (2008). A severe case of intracranial hemorrhage unsuccessful attempt to drain the blood at the bedside to due to alloimmune thrombocytopenia. The Journal of Maternal-Fetal & Neonatal Medicine, 21(11), 852–854. decrease the ICP. The infant was therefore taken to the oper- 13. Sola, M. C., Del Vecchio, A., & Rimsza, L. M. ( 2000). Evaluation ating room for emergency surgery to remove the subdural and treatment of thrombocytopenia in the neonatal intensive care unit. clot. After surgery, the ICP dropped, and the ventricular size Clinics in Perinatology, 27(3), 655–679. was normal on repeat CT scan. In the meantime, the mother 14. van den Akker, E. S., & Oepkes, D. (2007). Fetal and neonatal alloimmune was tested for antibodies. Her platelet count was normal, but thrombocytopenia. Best Practice & Research Clinical Obstetrics and her screen was positive for anti-HPA-1 antibodies, confirm- Gynaecology, 22(1), 3–14. ing the diagnosis of NAIT. 15. Blickstein, I., & Friedman, S. (2006). Fetal effects of autoimmune disease. During the first 36 hours of life, the infant received four In A. Fanaroff, R. Martin, & M. Walsh (Eds.) Fanaroff and Martin’s neonatal-perinatal medicine, diseases of the fetus and infant (8th ed., transfusions of random donor platelets, one transfusion of pp. 368–370). Philadelphia, PA: Elsevier Mosby. packed red blood cells, one unit of FFP, and one dose of 16. Cloherty, J. P., & Goorin, A. M. (2004). Thrombocytopenia. In J. P. cryoprecipitate. The platelet count was followed every eight Cloherty, E. C. Eichenwald, & A. R. Stark (Eds.), Manual of neonatal care hours and showed improvement. By Day 5, the platelet count (6th ed., pp. 455–462). Philadelphia, PA: Lippincott Williams & Wilkins. was 228,000, increasing to 346,000 by discharge on Day 7 17. Manco-Johnson, M., Rodden, D., & Collins, S. (2006). Newborn of life. The infant was eating well and gaining weight by hematology. In G. Merenstein & S. Gardner (Eds.), Handbook of neonatal discharge. The platelet count remained stable at two weeks intensive care (6th ed., pp. 521–547). St. Louis, MO: Mosby. and one month of age. On the one-year follow-up, the 18. Sola-Visner, M., Saxonhouse, M. A., & Brown, R. E. (2008). Neonatal thrombocytopenia: What we do and don’t know. Early Human infant is thriving and, most importantly, is developmentally Development, 84(8), 499–506. appropriate. 19. Luchtman-Jones, L., Schwarts, A., & Wilson, D. (2006). The blood and The infant had outpatient laboratory studies to check her hematopoietic system. In R. J. Martin, A. A. Fanaroff, & M. C. Walsh platelet levels, which remained normal; she was also followed (Eds.), Neonatal perinatal medicine, diseases of fetus and infant (8th ed., up in the developmental clinic. pp. 1338–1339). Philadelphia, PA: Elsevier Mosby. 20. Bussel, J. B. (2009). Diagnosis and management of the fetus and neonate REFERENCES with alloimmune thrombocytopenia. International Society of Thrombosis 1. Manco-Johnson, M., Rodden, D., & Collins, S. (2007). Newborn and Haemostasis, 7(1) 253–257. hematology. In G. Merenstein & S. Gardner (Eds.), Handbook of neonatal 21. Cloherty, J. P., & Goorin, A. M. (2008). Thrombocytopenia. In J. P. intensive care (6th ed., pp. 521–547). St. Louis, MO: Mosby. Cloherty, E. C. Eichenwald, & A. R. Stark (Eds.), Manual of neonatal 2. Arnold, D. M., Smith, J. W., & Kelton, J. G. (2008). Diagnosis and care (6th ed., pp. 455–467). Philadelphia, PA: Wolters Kluwer, Lippincott management of neonatal alloimmune thrombocytopenia. Transfusion Williams & Wilkins. Medicine Reviews, 22(4), 255–267. 22. Bussel, J. B., & Sola-Visner, M. (2009). Current approaches to the 3. Kaplan, C. (2006). Alloimmune thrombocytopenia of the fetus and evaluation and management of the fetus and neonate with immune newborn. Blood Reviews, 16(1), 69–72. thrombocytopenia. Seminars in Perinatology, 33(1), 35–42. 4. Bussel, J. B., Berkowitz, R. L., Hung, C., Kolb, E. A., Wissert, M., 23. Bassler, D., Greinacher, A., Okascharoen C., Klenner, A., Ditomasso, J., Primiani, A., . . . Macfarland, J. G. (2010). Intracranial hemorrhage Kiefel V., . . . Paes, B. (2008). A systematic review and survey of the in alloimmune thrombocytopenia: Stratified management to prevent management of unexpected neonatal alloimmune thrombocytopenia. recurrence in the subsequent affected fetus. American Journal of Obstetrics Transfusion Practice, 48(1), 92–98. and Gynecology, 203(2), 135.e1–135.e14. 24. Bagwell, G. A. (2007). Hematologic system. In C. Kenner & J. W. 5. Ghevaert, C., Campbell, K., Walton, J., Smith, G. A., Allen, D., Lott (Eds.), Comprehensive neonatal care, an interdisciplinary approach Williamson, L. M., . . . Ranasinghe, E. (2007). Management and outcome (pp. 247–251). Philadelphia, PA: Saunders Elsevier. of 200 cases of fetomaternal alloimmune thrombocytopenia. Transfusion, 25. Poterjoy, B. S., & Josephson, C. D. (2009). Platelet, frozen plasma, 47(5), 901–910. cryoprecipitate: What is the clinical evidence for their use in the neonatal 6. Jones, C. W. (2004). Platelet disorders. Newborn and Infant Nursing intensive care unit? Seminars in Perinatology. 33(1), 66–74. Reviews, 4(4), 181–190. 26. Fernandes, C. J., Garcia-Prats, J. A., Mahoney, D. H., Jr., & Kim, M. S. 7. Wiedmeier, S. E., Henry, E., Sola-Visner, M. C., & Christensen, R. D. (2009). Neonatal thrombocytopenia. UptoDate. Retrieved from http:// (2008). Platelet reference ranges for neonates, defined using data from www.uptodate.com/contents/neonatal-thrombocytopenia over 47,000 patients in a multihospital healthcare system. Journal of 27. Young, T., & Mangum, B (2009). Neofax (22nd ed.). Montvale, NJ: Perinatology, 29(2), 130–136. Thomson Reuters. 8. Roberts, I., & Murray, N. A. (2008). Neonatal thrombocytopenia. 28. Leong, H., Stachnik, J., Bonk, M. E., & Matuszewski, K. A. (2008). Seminars in Fetal & Neonatal Medicine, 13(4), 256–264. Unlabeled uses of intravenous immune globulin. American Journal of 9. Symington, A., & Paes B. (2010). Fetal and neonatal alloimmune Health-System Pharmacologists, 65(19), 1815–1824. thrombocytopenia: Harvesting the evidence to develop a clinical approach 29. Watson, D., & Hearnshaw, K. (2010). Understanding blood groups and to management. American Journal of Perinatology, 28(2), 137–144. transfusion in nursing practice. Nursing Standard, 24(30), 41–48. 10. Kaplan, C. (2003). Fetal and neonatal alloimmune thrombocytopenia. 30. Gupta, S. N., Kechli, A. M., & Kanamalla, U. S. (2009). Intracranial Orphanet Encyclopedia. Retrieved http://www.orpha.net/data/patho/ hemorrhage in term newborns: Management and outcomes. Pediatric GB/uk-NAIT.pdf Neurology, 40(1), 1–12. 11. Michelson, A. (2007). Platelets (2nd ed., pp. 23–55). Burlington, MA: 31. Bussel, J. B., & Primiani, A. (2007). Fetal and neonatal alloimmune throm Elsevier. bocytopenia: Progress and ongoing debates. Blood Reviews, 22(1), 33–52.

N EONATAL NETWORK 406 NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6

Copyright © Springer Publishing Company, LLC 32. Harkness, M. (2002). Neonatal alloimmune thrombocytopenia. British Journal of Midwifery, 10(2), 99–103. 33. Ward, M. J., Pauliny, J., Lipper, E. G., & Bussel, J. B. (2006). Long- term effects of fetal and neonatal alloimmune thrombocytopenia and its antenatal treatment on the medical and developmental outcomes of affected children. American Journal of Perinatology, 23(8), 487–492. 34. van den Akker, E. S., & Oepkes, D. (2008). Fetal and neonatal alloimmune thrombocytopenia. Best Practice & Research. Clinical Obstetrics & Gynaecology, 22(1), 3–14.

About the Author Jodi Beachy is a neonatal nurse practitioner in the Nationwide Children’s Hospital Neonatal Special Care unit at Dublin Methodist Hospital Ohio Health. The author would like to thank Patricia Nash, RNC, MSN, NNP, for her editing expertise. For further information, please contact: Jodi Beachy, RNC, MSN, NNP E-mail: [email protected]

N EONATAL NETWORK VOL. 30, NO. 6, NOVEMBER/DECEMBER 2011 407

affeine is a drug Throughout the next two to Ccommonly used in the Caffeine Citrate Therapy three decades, additional studies— neonatal intensive care unit. given summarized in a 2001 Cochrane its frequent use, providers need to Review—confirmed the benefits of be well informed of the drug, its for Apnea of Prematurity methylxanthines, including theo- indications, mechanism of action, Patricia J. Johnson, DNP, MPH, RN, NNP phylline and caffeine for the treat- and pharmacokinetics. Caffeine ment of apnea in the premature.14 has an important history of skeptics Of note, early preparations of caf- and variable practice preference, ultimately leading to fairly new feine included the use of caffeine benzoate, given intramuscu- evidence supporting its pharmacologic attributes for treatment larly. This compound is now contraindicated in the newborn of apnea of prematurity (AOP). Caffeine is also used to stim- because it displaces bilirubin from albumin binding sites, ulate preextubation respiratory drive in the nursery and fol- potentiating hyperbilirubinemia.15 Alternative or adjunct to lowing anesthesia.1,2 The purpose of this column is to review pharmacologic management of AOP, supportive therapies the history of methylxanthine therapy as a treatment for AOP, have been recommended to alleviate AOP including prone examine the benefits of caffeine citrate (Cafcit®) as the meth- positioning, head elevation, and continuous positive airway ylxanthine of choice, including the pharmacology and pharma- pressure, with and without synchronized nasal ventilation.4 cokinetics of the drug, and review the current evidence-based Although caffeine has been used in the preterm with practice for the use of Cafcit® in the treatment of AOP. apnea for three decades, commercially available preparations AOP is a self-resolving disorder of developmental of both injectable and oral Cafcit® received U.S. Food and immaturity, but it has potential for serious clinical conse- Drug Administration (FDA) approval in 1999 and 2000, quences including an increased risk of neurodevelopmental respectively. The FDA approved both preparations of Cafcit® impairment.3,4 AOP, occurring in infants less than 37 weeks for the short-term treatment of AOP. Once approved, Cafcit® gestation, is defined as the cessation of breathing for greater became the methylxanthine of choice for AOP given its than 20 seconds or cessation of breathing for 15 seconds, long half-life allowing once daily dosing.16 Prior to Cafcit® with associated bradycardia or cyanosis.5–8 The reported inci- regulatory approval by FDA, AOP was widely treated with dence of AOP varies, but it is clearly inversely related to gesta- theophylline or its parenteral form, aminophylline—both tional age. The incidence is 25 percent in newborns less than “off label” and without regulatory approval for AOP. Despite 2,500 g or those 34 weeks gestation at birth; but in new- their off-label use, both preparations of theophylline and borns less than 34 weeks gestation, the incidence increases aminophylline had available commercial preparations. Prior from 35 percent to as frequent as 85 percent depending on to Cafcit® release, the lack of commercial preparations meant gestation. Ninety percent of the extremely low birth weight each dose of caffeine had to be prepared by the hospital phar- newborn population, less than 1,000 g, are reported to have macy, which was time consuming and carried an increased AOP. 6,9–11 AOP is a diagnosis of exclusion with no definitive risk of compounding errors and dose stability issues. diagnostic test beyond its observed occurrence. Alternative Studies before and after the commercial production and causes of AOP such as central nervous system (CNS) disor- approval of caffeine, reported in a 2010 Cochrane Review, ders, primary lung disease, anemia, sepsis, metabolic distur- ultimately confirmed its clinical benefits compared with theo- bances, cardiovascular abnormalities, or airway obstruction phylline in the treatment of AOP.8 Caffeine’s comparative should be ruled out.5 clinical and pharmacologic benefits include its longer half- life allowing daily dosing, wide therapeutic margin reducing HISTORICAL REVIEW the likelihood of toxicity, smaller plasma concentration fluc- Historically, AOP was first recognized in the late 1960s and tuations, and greater CNS penetration without producing early 1970s. At that time, the only remedy for AOP was the fluctuations in CNS blood flow.11 use of mechanical ventilation and supplemental oxygen to treat the lack of respirations and resulting hypoxia. However, the CAFFEINE CITRATE DESCRIPTION problem recurred in many of the infants when ventilation was Caffeine (1,3,7-trimethylxanthine [Figure 1]) is a naturally discontinued. At the same time that AOP was recognized, the occurring fruit of the Coffea arabica plant. The other natu- use of aminophylline suppositories for toddlers with asthma rally occurring methylated xanthines are theophylline (1,3- was found to be beneficial. During a regional meeting in dimethylxanthine [Figure 2]) and theobromine (not used 1970, a group of pediatricians considered the potential use of 12 aminophylline for apnea management in the premature. The Disclosure first trial using aminophylline in the premature was completed The author discloses no relevant financial interest or affiliations with any in the United Kingdom and published in 1973.13 commercial interests.

Accepted for publication June 2011.

N EONATAL NETWORK 408 © 2011 Springer Publishing Company NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6 http://dx.doi.org/10.1891/0730–0832.30.6.408 Copyright © Springer Publishing Company, LLC FIGURE 1 n Caffeine base chemical structure. FIGURE 2 n Theophylline chemical structure.

O CH3 O H H C H3C 3 N N N N

N O N N O N

CH CH3 3

in the newborn). Cafcit® (commercial caffeine citrate), is a monophosphate (cAMP) and cyclic guanosine monophosphate preservative-free citrate salt in solution with citric acid. Each (cGMP) and leading to bronchodilatation.21–25 20 mg of Cafcit ® contains the equivalent of 10 mg of the active caffeine base. These methylxanthines are structurally related PHARMACOKINETICS to important endogenous metabolites, including purine, xan- Absorption thine, and uric acid.11 Purine nucleotides are metabolized into Based on a study of 85 premature newborns, 28–33 weeks uric acid for excretion in the urine, and xanthine is an interme- gestation with AOP, oral administration of 10 mg/kg of caf- diate metabolite in the breakdown of purine into uric acid. feine base (20 mg/kg of Cafcit®) resulted in peak plasma level (Cmax) for caffeine of 6–10 mg/liter and a mean time to reach MECHANISM OF ACTION pea k concent rat ion (Tma x) of 30 –120 m i nutes. 7 Ca fcit ® admin- As a methylxanthine, caffeine produces smooth muscle relax- istered orally is absorbed rapidly with minimal-to-no first pass ation, stimulates the CNS and cardiac muscle, is a diuretic, and metabolism, and 90 percent of the orally administered dose reduces apneic events.17 In the treatment of AOP, caffeine is reaches systemic circulation in less than two hours. The route thought to (a) stimulate the respiratory center of the brain; (b) of administration does not alter the pharmacokinetics of caf- increase minute ventilation; (c) decrease the CNS’s threshold to feine and there is almost complete bioavailability following oral carbon dioxide, hypercapnia; (d) increase the CNS response to administration as with intravenous (IV) administration.11,26,27 hypercapnia; (e) increase skeletal muscle tone; (f) decrease diaphragmatic fatigue; (g) increase metabolic rate; and (h) increase Distribution oxygen consumption.11,18 Caffeine, as with other methylxan- Being hydrophobic, caffeine distributes rapidly, passing thines, may also provide an anti-inflammatory action in the through most membranes without tissue accumulation. immature lung, similar to the anti-inflammatory benefits seen It is rapidly distributed into the brain, and levels in the with asthmatics of other age groups.19,20 The specific mecha- cerebrospinal fluid of preterm neonates approximate the nism in reducing AOP is not known, but clinically it acts as a plasma levels. The mean volume of distribution in infants was respiratory stimulant overcoming the developmental immatu- 0.8–0.9 liter/kg or slightly higher than volume of distribu- rity leading to interrupted breathing or periodic breathing.20 tion in adults at 0.6 liter/kg.28 At therapeutic levels, caffeine readily crosses the blood–brain barrier. It acts as a competitive antagonist of adenosine at cell Metabolism surface receptors, competitively inhibiting adenosine binding, Caffeine is metabolized by the hepatocytes partially via thereby preventing the neuronal suppressing activity of adenosine. the soluble enzyme xanthine oxidase and primarily by the Caffeine allows glutamate disinhibition with a resulting increase hepatic enzyme, cytochrome P450 1A2 (CYP1A2), leading in dopamine, reward or pleasure, and wakefulness. Adenosine to biotransformation of caffeine through the process of receptors A1 and A 2a are thought to be important in controlling demethylation. The demethylation process is postnatal age respiratory drive and potentially decreasing neuronal respiratory dependent, regardless of birth weight or gestational age with motor tract outflow, ultimately impeding respiration. Therefore, maturation by four months of age.29 Therefore, metabolism adenosine is presumed the source of central respiratory depres- of caffeine in the preterm newborn is limited by the imma- sion. The caffeine molecule, as a selective adenosine antagonist ture hepatic enzyme system. Interconversion between caf- at the A2a receptors and a nonselective adenosine antagonist at feine and theophylline has been reported with a greater rate A1 receptor, is thought to block this adenosine effect. Caffeine of theophylline converting to caffeine than caffeine convert- increases catecholamine release by modulating the cellular ing to theophylline. Approximately 3–10 percent of caffeine influx of calcium. It further stimulates the CNS by inhibiting converts to theophylline, whereas up to 50 percent of theo- phosphodiesterase to increased levels of cyclic adenosine 39,59 phylline converts to caffeine.18,28,30

N EONATAL NETWORK VOL. 30, NO. 6, NOVEMBER/DECEMBER 2011 409

Copyright © Springer Publishing Company, LLC Elimination DOSING Caffeine elimination is slower in the premature and I n t reat i ng AOP, a load i ng dose of 20 – 4 0 mg/ kg of C a fcit ®, term neonate, compared with older children and adults, equivalent to 10–20 mg/kg of caffeine base administered IV because of developmental immaturity of the liver and kidney or orally, is recommended. The maintenance-recommended impacting hepatic metabolism and renal excretion. In the dose range for caffeine is 5–10 mg/kg/dose, equivalent first postnatal weeks, caffeine is eliminated primarily by to 2.5–5 mg/kg of caffeine base, commencing 24 hours renal excretion.31 The mean half-life (T½) and the fraction after loading dose administration and administered every 35 excreted unchanged in urine (Ae) in neonates are inversely 24 hours. In an observational study of 101 premature neo- related to gestational and postconceptual age. In neonates, nates, 23–32 weeks gestation and with median age of six days, the T½ is approximately three to four days with a range this dosing schedule has been shown to yield a predictable of 40–230 hours, and the Ae is approximately 86 percent (94.8 percent) desired therapeutic trough serum concentra- within six days. The metabolism and elimination of caffeine tion of 5–20 mcg/mL based on this observational study, and approximates adult levels with T½ 5 five hours and eA 5 the therapeutic caffeine level was attained despite gestational 1 percent by nine months of age 11,16,24,28,31–35 or at least age, renal, or hepatic dysfunction.40 These authors infer that by 60 weeks postmenstrual age.31 Postnatal age influences although many neonatal intensive care units routinely monitor clearance and some researchers have suggested parenteral caffeine levels in the absence of toxic symptoms, drug level nutrition may alter predictable clearance and half-life.31 monitoring may be unnecessary unless higher target levels Half-life of caffeine may be prolonged in premature neo- are preferred for neonates with suboptimal responses. They nates with cholestatic hepatitis.31 further recommend eliminating routine caffeine level monitoring, concluding a benefit of less blood loss from testing as PRECAUTIONS, ADVERSE EFFECTS, well as cost savings of at least $50 per assay.40 AND INTERACTIONS Although caffeine is well tolerated by most neonates, early CURRENT EVIDENCE studies suggested potential transient alterations in blood Although caffeine use for AOP has gradually become a flow to vital organs with caffeine administration to prema- practice standard, some concerns have been raised about caf- ture neonates. Specifically, the studies identified reduced flow feine’s potential to alter adenosine metabolism or regional to brain and gut for the first two hours after administration cerebral and intestinal blood flow, perhaps resulting in an of caffeine with potential for cerebral and gastrointestinal increased risk of NEC and long-term neurologic seque- impairment.36,37 One additional study found an increased lae.36 Schmidt and colleagues have reported positive results rate of necrotizing enterocolitis (NEC) in the caffeine study from analysis of the short-term and longer term outcomes group using the open-label caffeine preparation.7 These con- as well as post hoc subgroup analysis of their large, mul- cerns were addressed by the large, well-controlled study by ticenter, randomized, placebo-controlled study of caffeine Schmidt and colleagues indicating no difference in NEC therapy for AOP in premature infants with birth weights among study and placebo groups, and reduced incidence of of 500–1,250 g.17, 38,41 Their results not only confirmed cerebral palsy, reduced incidence of bronchopulmonary dys- reduction of apnea episodes, but also demonstrated a sig- plasia, and improved survival without neurodevelopmental nificant reduction in bronchopulmonary dysplasia, as well disability.38 as improved survival without neurodevelopmental disabil- Overdose leading to toxicity has been reported, resulting ity at the corrected age of 18–21 months. The incidences of in tachyarrhythmias, irritability, jitteriness, feeding intoler- cerebral palsy and cognitive delay were also reduced in the ance, and increased urine output.39 Additionally with levels caffeine-treated group. Furthermore, there was a reduct ion in in excess of 50 mcg/mL, fever, tachypnea, hypertonia, vomit- respiratory support, supplemental oxygen, postnatal steroids, ing, hyperglycemia, elevated blood urea nitrogen, and leuko- and surgical ligation of patent ductus arteriosus without a cytosis as well as seizures have been reported in the product difference in the rates of death, deafness, or blindness literature for Cafcit®.28 between the control and caffeine groups. Despite a dimin- Caffeine has the potential to interact with drugs that are ished weight gain during the first three weeks of caffeine metabolized by the CYP1A2 enzyme. Cimetidine and keto- treatment, growth was not affected and long- and short-term conazole can inhibit caffeine metabolism, requiring a lower weight loss may be overshadowed by the long-term benefits caffeine dose. Phenytoin and phenobarbital can increase in survival without neurodevelopmental delay.42 The results caffeine elimination, necessitating higher doses of caffeine. of later neurodevelopmental outcome and school perfor- Given that caffeine is an adenosine receptor antagonist, treat- mance are expected in future reports of continued follow-up ment of tachyarrhythmias with adenosine may require altered of this important study population.38,41 In an ad hoc sub- dosing for desired therapeutic conversion.24 It may be impor- group report of the 2006 Caffeine for Apnea of Prematurity tant to consider caffeine dosing or levels if maternal caffeine (CAP) trial, the authors further found that early initiation intake was excessive during pregnancy or breast milk feeding; of caffeine was associated with a greater reduction in time however, this is only a theoretical consideration. on ventilation.17

N EONATAL NETWORK 410 NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6

Copyright © Springer Publishing Company, LLC SUMMARY 15. Marchal, F., Bairam, A., & Vert, P. (1987). Neonatal apnea and apneic Cafcit® is the methylxanthine of choice for the treatment of syndromes. Clinics in Perinatology, 14(3), 509–529. AOP, with demonstrated benefits beyond reduction of apnea 16. U.S. Food and Drug Administration. (2005). Drug approval package. episodes. Caffeine is also an effective treatment for stimu- Retrieved from http://www.accessdata.fda.gov/Scripts/cder/Drugsat FDA/index.cfm?fuseaction5Search.Label_ApprovalHistory#apphist; lating the respiratory drive prior to extubation in newborns http://www.accessdata.fda.gov/drugsatfda_docs/nda/99/020793 ventilated for respiratory failure, respiratory distress, and fol- _000_Cafcit®TOC.cfm 1,8,14 lowing operative procedure. Compared with theophyl- 17. Davis, P. G., Schmidt, B., Roberts, R. S., Doyle, L. W., Asztalos, E., Haslam, line/aminophylline, caffeine is preferred because of its longer R., . . . Tin, W. (2010). Caffeine for Apnea of Prematurity trial: Benefits half-life, allowing daily maintenance dosing; wider therapeu- may vary in subgroups. The Journal of Pediatrics, 156(3), 382–387. tic range with lower risk of toxicity; low rate of adverse effects; 18. Buck, M. L., Hofer, K. N., & McCarthy, M. W. (2008). Caffeine citrate for and ease in transitioning between IV and oral administration the treatment of apnea of prematurity: Dosing recommendations. Pediatric Pharmacology, 14(6), 1–4. Retrieved from http://www.medscape.com because both routes have equivalent bioavailability. 19. Page, C. P. (1999). Recent advances in our understanding of the use of As discussed, the recently published results of the CAP theophylline in the treatment of asthma. Journal of Clinical Pharmacology, ® trial provide support for the safe and effective use of Cafcit in 39(3), 237–240. the treatment of AOP. The results of their ongoing follow-up 20. Bancalari, E. (2006). Caffeine for apnea of prematurity. New England conclude that the benefits of treatment clearly outweigh any Journal of Medicine, 345(20), 2179–2181. risks. Caffeine has established an accepted place in the routine 21. Martin, R. J., Miller, M. J., & Carlo, W. A. (1986). Pathogenesis of apnea clinical management of premature infants.17,38,41,42 in preterm infants. The Journal of Pediatrics, 109(5), 733–741. 22. Herlenius, E., & Lagercrantz, H. (1999). Adenosinergic modulation of respiratory neurones in the neonatal rat brainstem in vitro. The Journal of REFERENCES Physiology, 518(Pt. 1), 159–172. 1. Henderson-Smart, D. J., & Davis, P. G. (2003). Prophylactic 23. Wilson, C. G., Martin, R. J., Jaber, M., Abu-Shaweesh, J., Jafri, A., methylxanthines for extubation in preterm infants. Cochrane Database of Haxhiu, M. A., & Zaidi, S. (2004). Adenosine 2A2 receptors interact Systemic Reviews, (1), CD000139. with GABAergic pathways to modulate respiration in neonatal piglets. 2. Steer, P., Flenady, V., Shearman, A., Charles, B., Gray, P. H., Henderson- Respiratory Physiology & Neurobiology, 141(2), 201–211. Smart, D., . . . Walsh, A. (2004). High dose caffeine citrate for extubation 24. Natarajan, G., Lulic-Botica, M., & Aranda, J. V. (2007). Pharmacology of preterm infants: A randomised controlled trial. Archives of Diseases in review: Clinical pharmacology of caffeine in the newborn. NeoReviews, Childhood. Fetal and Neonatal Edition, 89(6), F499–F503. 8(5), e214–e221. Retrieved from http://neoreviews.aappublications. 3. Poets, C. F. (2008). Apnea of prematurity, sudden infant death syndrome, org/cgi/content/abstract/neoreviews;8/5/e214 and apparent life-threatening events. In L. M. Taussig & L. I. Landau (Eds.), 25. Sui, A., & James, A. (2010). Apnea of prematurity pharmacotherapy. US Pediatric respiratory medicine (pp. 413–434). Philadelphia, PA: Mosby. Pharmacist, 35(3), HS2–HS6. Retrieved from http://www.medscape.com 4. Poets, C. F. (2010). Interventions for apnoea of prematurity: A personal 26. Aranda, J. V., Cook, C. E., Gorman, W., Collinge, J. M., Loughnan, view. Acta Paediatrica, 99(2), 172–177. P. M., Outerbridge, E. W., . . . Neims, A. H. (1979). Pharmacokinetic 5. Aranda, J. V., & Turmen, T. (1979). Methylxanthines in apnea of profile of caffeine in the premature newborn infant with apnea. The prematurity. Clinics in Perinatology, 6(1), 87–108. Journal of Pediatrics, 94(4), 663–668. 6. Barrington, K., & Finer, N. (1991). The natural history of the appearance 27. Giacoia, G. P., Jungbluth, G. L., & Jusko, W. J. (1989). Effect of formula of apnea of prematurity. Pediatric Research, 29(4 Pt. 1), 372–375. feeding on oral absorption of caffeine in premature infants. Developmental 7. Erenberg, A., Leff, R. D., Haack, D. G., Mosdell, K. W., Hicks, G. M., Pharmacology and Therapeutics, 12(4), 205–210. & Wynne, B. A. (2000). Caffeine citrate for the treatment of apnea of 28. Mead Johnson and Co. (2003). Cafcit® (caffeine citrate) injection prematurity: A double-blind, placebo-controlled study. Pharmacotherapy, and oral solution prescribing information. Retrieved from http://www 20(6), 644–652. .bedfordlabs.com 8. Henderson-Smart, D. J., & Steer, P. A. (2010). Caffeine versus 29. al-Alaiyan, S., al-Rawithi, S., Raines, D., Yusuf, A., Legayada, E., Shoukri, theophylline for apnea in preterm infants. Cochrane Database of Systematic M. M., & el-Yazigi, A. (2001). Caffeine metabolism in premature infants. Reviews, 20(1), CD000273. Journal of Clinical Pharmacology, 41(6), 620–627. 9. Henderson-Smart, D. J. (1981). The effect of gestational age on the 30. Davis, J. M., Spitzer, A. R., Stefano, J. L., Bhutani, V., & Fox, W. W. incidence and duration of recurrent apnoea in newborn babies. Australian (1987). Use of caffeine in infants unresponsive to theophylline in apnea Paediatric Journal, 17(4), 273–276. of prematurity. Pediatric Pulmonology, 3(2), 90–93. 10. Scanlon, J. E., Chin, K. C., Morgan M. E., Durbin, G. M., Hale, K. A., 31. Le Guennec, J. C., Billon, B., & Paré C. (1985). Maturational changes & Brown, S. S. (1992). Caffeine or theophylline for neonatal apnoea? of caffeine concentrations and disposition in infancy during maintenance Archives of Disease in Childhood, 67(4 Spec. No.), 425–428. therapy for apnea of prematurity: Influence of gestational age, hepatic 11. Comer, A. M., Perry, C. M., & Figgitt, D. P. (2001). Caffeine citrate: A disease, and breast-feeding. Pediatrics, 76(5), 834–840. review of its use in apnoea of prematurity. Paediatric Drugs, 3(1), 61–79. 32. Arnaud, M. J. (1987). The pharmacology of caffeine. Progress in Drug 12. Kuzemko, J. A. (2003). Historical perspectives: Use of methylxanthines Research, 31, 273–313. in the management of apneic attacks in the newborn. NeoReviews, 4(3), 33. Lee, T. C., Charles, B., Steer, P., Flenady, V., & Shearman, A. (1997). e62–e64. Retrieved from http://neoreviews.aappublications.org Population pharmacokinetics of intravenous caffeine in neonates with apnea 13. Kuzemko, J. A., & Paala, J. (1973). Apnoeic attacks in the newborn treated of prematurity. Clinical Pharmacology & Therapeutics, 61(6), 628–640. with aminophylline. Archives of Disease in Childhood, 48(5), 404–406. 34. Falcão, A. C., Fernández de Gatta, M. M., Delgado Iribarnegaray, M. 14. Henderson-Smart, D. J., & Steer, P. (2001). Methylxanthine treatment F., Santos Buelga, D., García, M. J., Dominguez-Gil, A., & Lanao, J. M. for apnea in preterm infants. Cochrane Database Systematic Reviews, (3), (1997). Population pharmacokinetics of caffeine in premature neonates. CD000140. European Journal of Clinical Pharmacology, 52(3), 211–217.

N EONATAL NETWORK VOL. 30, NO. 6, NOVEMBER/DECEMBER 2011 411

Copyright © Springer Publishing Company, LLC 35. Young, T. E., & Magnum, B. (2010). Caffeine citrate. In Neofax: A 41. Schmidt, B., Roberts, R. S., Davis, P., Doyle, L. W., Barrington, K. J., manual of drugs used in neonatal care (23rd ed., pp. 272–273). Montvale, Ohlsson, A., . . . Tin, W. (2006). Caffeine therapy for apnea of prematurity. NJ: Thomson Reuters. The New England Journal of Medicine, 354(20), 2112–2121. 36. Hoecker, C., Nelle, M., Poeschl, J., Beedgen, B., & Linderkamp, O. 42. Stevenson, D. K. (2007). On the caffeination of prematurity. The New (2002). Caffeine impairs cerebral and intestinal blood flow velocity in England Journal of Medicine, 357(19), 1967–1968. preterm infants. Pediatrics, 109(5), 784–787. 37. Hoecker, C., Nelle, M., Beedgen, B., Rengelshausen, J., & Linderkamp, O. (2006). Effects of a divided high loading dose of caffeine on circulatory About the Author variables in preterm infants. Archives of Disease in Childhood. Fetal and Patricia J. Johnson is a practicing neonatal nurse practitioner in the Neonatal Edition, 91(1), F61–F64. Phoenix metropolitan area. She piloted the NNP role during her master’s 38. Schmidt, B., Roberts, R. S., Davis, P., Doyle, L. W., Barrington, K. J., program in nursing in the early 1970s and subsequently established NNP Ohlsson, A., . . . Tin, W. (2007). Long-term effects of caffeine therapy teams in the Midwest and Southwest. She also taught and coordinated one for apnea of prematurity. The New England Journal of Medicine, 357(19), of the early NNP certificate programs in Arizona. She has been an active 1893–1902. volunteer with neonatal nursing organizations and is a recognized expert 39. Anderson, B. J., Gunn, T. R., Holford, N. H., & Johnson, R. in the delivery of neonatal care by the advanced practice nurse. She received (1999). Caffeine overdose in a premature infant: Clinical course and a doctorate in nursing practice from Arizona State University in 2008. pharmacokinetics. Anaesthesia and Intensive Care, 27(3), 307–311. 40. Natarajan, G., Botica, M. L., Thomas, R., & Aranda, J. V. (2007). For further information, please contact: Therapeutic drug monitoring for caffeine in preterm neonates: An Patricia J. Johnson, DNP, MPH, RN, NNP unnecessary exercise? Pediatrics, 119(5), 936–940. E-mail: [email protected]

N EONATAL NETWORK 412 NOVEMBER/DECEMBER 2011, VOL. 30, NO. 6