Pediatr Radiol (2012) 42 (Suppl 1):S92–S100 DOI 10.1007/s00247-011-2255-4

ADVANCES IN FETAL AND NEONATAL IMAGING

Update on the diagnosis and management of bronchopulmonary dysplasia/chronic lung disease of infancy: what the radiologist should know Richard B. Parad

Received: 29 August 2011 / Revised: 6 September 2011 / Accepted: 9 September 2011 # Springer-Verlag 2011

Abstract Pediatric radiologists are frequently called upon to render interpretations of chest radiographs performed on premature infants with chronic respiratory problems. After the acute phase of surfactant deficiency (respiratory distress syndrome), infants with persistent respiratory problems are loosely categorized by clinicians as evolving toward a broad, rather vague entity called bronchopulmonary dysplasia (BPD) or chronic lung disease (CLD). This review aims to update the radiologist on how the characteristics of the disease have shifted and how management, diagnosis and pathology have changed since the disorder was first described more than 40 years ago. The radiologist armed with this information might be better prepared to provide insightful reporting and address the needs of the neonatologist. Keywords Bronchopulmonary dysplasia (BPD) . Chronic lung disease (CLD) . Premature infants

Introduction Bronchopulmonary dysplasia (BPD) is a diagnosis commonly applied to newborns (most commonly those with immature lungs) who develop chronic respiratory abnormalities because of the toxicities of exposure to a combination of positive-pressure ventilation and supplemental oxygen. But there are many questions attached to this diagnosis. What medical literature should the radiologist be familiar with before interpreting the chest radiograph of a newborn that might have this disorder? What does “BPD” really mean? Does it have the same meaning as it did 20 or 40 years ago? Is the radiologist reporting an accurate and clinically useful result when dictating “consistent with evolving BPD”? What does the neonatologist want to know from the radiologists’ report? This update seeks to address these questions.

Epidemiology

R. B. Parad (*) Department of Newborn Medicine, Brigham and Women’s Hospital, Center for Women and Newborns, Room CWN418, 75 Francis St., Boston, MA 02115, USA e-mail: [email protected] R. B. Parad Division of Newborn Medicine, Children’s Hospital Boston, Boston, MA, USA R. B. Parad Harvard Program in Neonatology, Harvard Medical School, Boston, MA, USA

Of the approximately 4 million births per year in the United States, about 13% are premature (gestational age <37 weeks). Among premature infants, 15% require assisted mechanical ventilation, and of those ventilated, approximately 15% (12,000/year) are diagnosed with BPD. To arrive at this number another way, 20% of the 60,000 infants born annually at a birth weight <1,500 g (very low birth weight) are diagnosed with BPD. Although term infants who require prolonged mechanical ventilation as a result of primary disorders such as meconium aspiration syndrome, pneumonia or congenital anomalies of the heart, lung or GI tract might also be labeled with a BPD

Pediatr Radiol (2012) 42 (Suppl 1):S92–S100

S93

diagnosis, this review will focus on a discussion of BPD in very low birth-weight preterm infants.

Improvements in perinatal and neonatal care and a shift in the affected population: the Northway era (1967) vs. the post-surfactant and antenatal steroids era (circa post-1990) Northway’s original description of BPD correlated four radiographic patterns (stages I–IV) with postmortem histological findings from preterm infants who died from respiratory failure after initial surfactant deficiency in the pre-surfactant therapy era [1]. In the small study cohort, the approximate mean gestational age (GA) at birth was 33 weeks and mean birth weight was 1,600 g. With advances in both perinatal care and neonatal management since the time of that publication in 1967, the population currently described as having BPD differs significantly from Northway’s cohort, with mean GA closer to 27 weeks and mean birth weight <1,000 g. This review focuses on why the radiologist needs to replace Northway’s BPD label with Jobe’s [2] “new BPD”, and what current evidence evaluating clinical practices has shown regarding the reduction of risk for this newer form of disease. Pinpointing the incidence of BPD is like trying to hit a moving target. Randomized clinical trials (RCT) that have evaluated therapeutic interventions for protection against the development of BPD use the combined outcome of “Death + BPD”. The linkage of these two categories of premature infants is based on the concept that infants who died early because of severe respiratory failure (before a diagnosis of BPD could be officially designated) had a strong chance of ultimately receiving a BPD diagnosis if they had lived, so the combined incidence gives a worst case scenario of who would have developed BPD. The flip side of improving survival in increasingly premature infants is that the younger the gestational age, the more immature the lungs, the more mechanical ventilation and supplemental oxygen support required, the higher the risk of developing BPD. Thus, infants who previously died before they could be diagnosed with BPD now survive and move from the death to the BPD categories as shown (Fig. 1). In addition to increasing BPD risk with decreasing gestational age or birth weight, the more recent the year examined, the lower the BPD rates that might be seen at each gestational age or birth weight because of improvements in neonatal care. Thus a complex balance of competing factors is involved in determining the birth-weight-specific BPD rate and its changing incidence over time. Major advances in clinical care that have contributed to improved survival at lower gestational ages and birth

Fig. 1 U.S. Birth-weight-specific mortality and CLD incidence in 2007 (data from Vermont Oxford Network). Note that the mean birth weight of the Northway cohort was 1,600 g

weights include improved neonatal mechanical ventilation equipment and strategies, better capabilities for nourishing premature infants and optimizing growth, use of antenatal glucocorticoids in mothers (to mature the synthesis of fetal pulmonary surfactant), and the development of exogenous surfactants administered by endotracheal tube. Since the mid-1990s, steroids for the mother and endotracheally administered surfactant for the preterm newborn have become the standard of care, and now that their utilization is near-universal, drops in death rates and BPD rates have leveled off (mortality and BPD rates between 1994 and now are flat). Data shown in Fig. 1 reveal the breakdown by birth weight of survival with and without BPD. In 2007, 80% of infants born at 500–750 g died or had BPD. This cohort differs dramatically from that described by Northway, who reported a mean of 1,600 g in his cohort who died with BPD. Currently, most BPD risk resides in the group of infants born at less than 1,250 g, mainly below 1,000 g. The fact that birth weight and gestational age are proxies for lung maturity (the smaller and younger the infant, the more immature the structural and biochemical developmental status of the lung) has led to a change in the neonatologist’s understanding of the pathophysiology of BPD. Overall, for preterm infants born weighing less than 1,000 g, a risk of 20–30% of developing BPD can be anticipated, with the tiniest babies carrying risk as high as 70%. Please note there is a wide variability of this number depending on the institution where care is provided (differences in care practices that are not well understood).

S94

Update on definitions: problems and confusions “Old” vs. “new” BPD In response to Northway’s findings, the original clinically practical BPD definition applied to living preterm infants was related to the initial duration of exposure to oxygen (>28 days) in conjunction with the presence of abnormalities on chest radiograph. However, in response to the increasing neonatal survival at decreasing gestational ages that resulted from advances in medical care of mothers and preterm newborns, the clinical definition had to evolve. By the mid-1980s, the 28-day supplemental oxygen requirement definition became nondiscriminate in predicting outcome for infants born at a GA ≤30 weeks, so it was supplanted by a definition that had an improved predictive value: the persistent need for supplemental oxygen therapy at or beyond 36 weeks’ corrected gestational age (CGA) [3]. Pathology/histology: structural injury or small airway obstruction vs. alveolar arrest In the histology findings that Northway correlated with radiographic patterns of small airway obstruction, there was airway epithelial metaplasia, debris obstructing small airways, emphysema, atelectasis, interstitial fibrosis and smooth muscle hyperplasia of both airways and the vascular tree. Pulmonary hypertension eventually led to right ventricular hypertrophy and cor pulmonale. Those 33week, 1,600-g preterm infants are no longer afflicted with the BPD described by Northway. We currently see a different group of BPD babies, who are much lower in birth weight and gestational age, with lungs that are much more immature at the time injury takes place from barotrauma, oxygen toxicity and inflammation and who develop a much different lung disease. In these babies with new BPD, histology shows fewer airway abnormalities and more parenchymal changes, particularly the arrest of alveolarization, which is demonstrated by fewer, larger alveolar spaces, decreased septation, decreased vascularization and decreased surface area [2]. In the developing lung, alveolarization does not begin until 23–24 weeks’ gestation and normally progresses over the 3rd trimester and ex-utero during the early years of life. It appears that the injury response of the preterm lung differs depending on when in the developmental spectrum the insults occur (Fig. 2). Radiographic findings: “old” vs. “new” BPD Figure 2 compares the typical chest radiographic patterns seen in what was formerly called BPD to what is now

Pediatr Radiol (2012) 42 (Suppl 1):S92–S100

found in babies born with more immature lungs. The latter tend to have a more homogeneous diffuse lung disease than the classic inhomogeneous pattern with hyperinflation and cystic disease. Although the former pattern is certainly still seen, the latter is generally more common (Fig. 3). Evolution of BPD definition: ‘oxygen at 28 days of life’ to ‘oxygen at 36 weeks’ corrected gestational age’ with severity stratification In 2005, a U.S. National Institutes of Health consensus panel attempted to improve the predictive value of the diagnosis of BPD at 36 weeks’ CGA on severity of pulmonary outcome by stratifying the degree of supplemental respiratory support [4]. Mild, moderate and severe categories were defined. Mild BPD required a prior period of supplemental oxygen requirement for 28 days, moderate requires a continued oxygen requirement <30% and infants with severe BPD have an oxygen requirement ≥30% or any form of positive-pressure ventilation or CPAP. Oxygen challenge test Clinicians vary in what oxygen saturation (SpO2) cutoff mandates the use of supplemental oxygen. Thus, a 36-week CGA infant with SpO2 of 91% undergoing care at a NICU that only provides supplemental oxygen when SpO2 registers below 90% would not be on supplemental oxygen at 36 weeks and might fit the above NIH criteria differently than if that infant were at a NICU that required the SpO2 be maintained at >94% (and thus would require supplemental O2 to maintain the target SpO2). Walsh [5, 6] proposed the “oxygen challenge test” as standardization at 36 weeks’ CGA [5, 6]. An infant requiring supplemental oxygen would be placed in room air, and if SpO2 dropped below a threshold, supplemental oxygen would be appropriate. This has mainly been used for outcome measures in recent clinical trials for BPD but could become a clinical standard of care. This is important given that two neonatologists might grade the same infant differently as to whether the label of BPD should be applied. On which day of life one is allowed to diagnose or label a newborn as having BPD remains a point of confusion and debate. The radiologist might be providing a reading believed to be consistent with a diagnosis of BPD based on either the pattern of old or new BPD while that infant has not reached 36 weeks’ CGA. This might be at a time when no neonatologist would officially be able to make that diagnosis. It should also be noted that the term chronic lung disease is also used by some as a wastebasket diagnosis, in order to be less specific about applying the formal BPD label but implying that all indicators from imaging and clinical course are heading toward a chronic lung disorder.

Pediatr Radiol (2012) 42 (Suppl 1):S92–S100

S95

Fig. 2 Chest radiographic examples of old BPD (left) and new BPD (right)

Poor predictive value of current definitions Unfortunately, even if all clinicians were in agreement on the attachment of a BPD label, even the NIH consensus panel definition and the diagnosis by oxygen challenge test are problematic in that they have poor predictive value for which infants will go on to have chronic respiratory morbidity beyond 36 weeks’ CGA. A BPD diagnosis is associated with a 1.5-fold increased risk (any oxygen at 36 weeks’ CGA) and 1.7-fold increased risk (severe BPD by NIH consensus criteria) of being on asthma medications at 18 months of age, as compared to preterm infants who are not labeled BPD. However these controls, just by being premature, have a 2.8-fold increased risk of respiratory morbidity compared to term infants. We really need better definitions and better predictors, and radiographs are also not so predictive of which babies will go on to have later problems. Table 1 summarizes some studies that have looked at the sensitivity and specificity of supplemental oxygen requirement at 36 weeks’ CGA in predicting long-term respiratory outcomes; this definition is neither sensitive nor specific. While oxygen requirement at 36 weeks is the most common Fig. 3 Pathophysiology of old and new BPD (modified from Baraldi [26])

primary outcome variable used in clinical trials, it is known to have high false-positive and false-negative values for predicting respiratory morbidity in the first years of life. Thus, better markers would be valuable. Future definitions: chronic respiratory morbidity and measures of symptoms and function Currently, bronchopulmonary dysplasia is defined by the surrogate O2 requirement at 36 weeks’ post-menstrual age (PMA) or corrected gestational age (CGA). Although the predictive value of this definition is somewhat improved by both the severity stratification proposed by the NIH consensus statement on BPD and oxygen challenge testing, the general term BPD and its inconsistent usage leads to unreliability in predicting long-term pulmonary outcome. In 2003, the publication of negative results for a randomized trial of an intratracheally administered antioxidant (recombinant human copper/zinc superoxide dismutase [rhCuZnSOD]) every 48 h for the first month of life for the prevention of BPD (as defined by an old BPD definition adhered to by the U.S. Food and Drug Administration from use in surfactant clinical trials: O2 requirement at 28 days of

S96

Pediatr Radiol (2012) 42 (Suppl 1):S92–S100

Table 1 Accuracy of supplemental oxygen requirement at 36 weeks’ CGA (BPD diagnosis) in the prediction of chronic respiratory morbidity (1– 5 years) in preterm infants. Modified from Lefkowitz and Rosenberg [27] Outcome

Sensitivity

Specificity

PPV

NPV

Johnson et al., Marlow et al., Thomas et al. [8–10]

-Wheeze -Cough

0.27–0.32

0.86–0.89

0.35–0.47

0.80–0.81

Palta et al. [28, 29]

-Use of respiratory meds -Asthma

0.33–0.43

0.80–0.95

0.38–0.48

0.82

0.53–0.67

0.59–0.62

0.11–0.39

0.74–0.96

0.46

0.82

0.75

0.57

0.52–0.54

0.60–0.62

0.36–0.44

0.71–0.74

Grégoire et al. [30]

-Hospitalizations -Hospitalizations -Symptoms

Davis et al. [7]

-Use of respiratory meds -ER visits

Ehrenkranz et al. [4]

-Use of respiratory meds -Hospitalizations

-Hospitalizations

PPV positive predictive value; NPV negative predictive value

life) reported that while the short-term BPD definition showed no drug effect, evaluation of longer-term (12-month) respiratory outcomes revealed that the drug might indeed demonstrate clinical benefit. While the drugtreated infants showed no benefit in terms of development of BPD at 28 days, when followed at 1 year CGA there was a nearly 50% reduction in the use of asthma medications and hospitalization in superoxide-dismutase-treated infants born at 27 weeks’ or less gestational age [7]. Thus, the observation that the potential benefit of a treatment provided in the newborn period might not be detectable under our current diagnostic terms for BPD or CLD is of concern in that we might miss a therapy that could impact the long-term outcomes that are of primary importance, i.e. chronic respiratory morbidity. Rather than considering the current definitions of BPD as diagnostic of a specific disease or marker of a specific underlying pathology, we should really consider its importance as a biomarker of later respiratory morbidity. If there is no correlation between BPD and ultimate clinical outcome, then it is not a good biomarker. Figure 4 demonstrates this concept of BPD as a proxy for 6 to12-month outcomes such as asthma, wheezing, coughing, respiratory hospitalizations and others. The term “chronic respiratory morbidity” is being increasingly used as a more long-term non-specific wastebasket descriptor of late chronic lung disease, much as “chronic lung disease” is used in the NICU. Data are accumulating on the predictors of this outcome, and it might become the trend to refer to a risk of chronic respiratory morbidity rather than BPD. We see in Table 1 that recent studies suggested poor correlation between short-term (BPD) definitions and long-term outcome (chronic respiratory morbidity). Clinicians should be most concerned about the long-term respiratory outcome of these infants beyond their care in the NICU. Given that

improving long-term pulmonary outcomes is our clinical goal it is critical to optimize the accuracy of surrogate markers for evaluating preventive therapeutic strategies. Researchers [8–10] have developed a chronic respiratory morbidity definition based on the quantitative data from self-reported 4-week parental diaries collected at 12 months’ CGA regarding the presence of coughing, wheezing and use of respiratory medications (Yes/No). These diaries have been shown to correlate closely with physiological measurements of pulmonary function performed at 12 months’ CGA and later respiratory questionnaires at 24 months’ CGA [11]. About one-third had CRM in this cohort of babies who were born before 29 weeks’ gestational age and required mechanical ventilation. Use of oxygen at 36 weeks’ CGA did not correlate with PFT abnormalities or chronic respiratory morbidity as defined by clinical data at 12 and 24 months’ CGA. Injured/maladapted premature lung

BPD: O2 @ 36wks • NIH severity strata • O2 challenge test

Cough Wheeze Tachypnea Increased WOB Exercise intolerance Respiratory medication use Respiratory infections MD visits ER visits Hospitalizations

Chronic Respiratory Morbidity 12 months corrected age and beyond

Fig. 4 Proxies for respiratory clinical outcome in preterm infants: BPD and CRM – what are their relationships? A Relationship between histopathology and biomarker. B Relationship betweem biomarker and outcome. Modified from Lefkowitz and Rosenberg [27]

Pediatr Radiol (2012) 42 (Suppl 1):S92–S100

Update on management Many very large, expensive, complex clinical trials on clinical management issues that might prevent BPD (defined as O2 requirement at 36 weeks’ SGA) have been published in reputable journals during the last several years. As suggested above, while it is not clear that this is the best primary outcome on which to base trials, the conclusions are still worth reviewing for radiologists so that they can have insight into current management strategies for respiratory care in the NICU. Therapies evaluated Minimization of lung injury through reduced barotrauma and oxygen toxicity Currently, the mantra of the neonatologist is to strive for kinder, gentler ventilation. This translates into devising a ventilation strategy that involves the choice of a type of mechanical ventilator and combination of settings that results in the minimum possible barotrauma and volutrauma (two of the most significant toxic exposures that lead to lung injury). Positive-pressure ventilator support is to be minimized, in part, reflected in the lowest obtainable mean airway pressure tolerated to provide adequate gas exchange. It has been proposed since the 1980s [12] that managing surfactant-deficient newborns with aggressive, early use of continuous positive airway pressure (CPAP) and avoidance of intubation and positive-pressure ventilation (with or without surfactant replacement therapy) leads to a lower risk for developing BPD—no intubation, no surfactant, just immediate institution of CPAP in the delivery room. Below is an overview of some major recent clinical trials involving optimal ventilator management strategies. High-frequency oscillatory ventilation High-frequency oscillatory ventilation (HFOV) with small oscillations of pressure around a constant distending pressure has been shown to be particularly effective in lowering pCO2 in preterm infants with severe lung disease at relatively lower peak airway pressures than required with conventional mechanical ventilators. Hypothetically there is less barotrauma when gas is not moved by tidal breathing, as with conventional mechanical ventilation. Recently Cools et al. [13, 14] utilized a new form of meta-analysis (individual patient meta-analysis) to combine the results of 17 randomized clinical trials of HFOV vs. conventional mechanical ventilation in preterm infants with surfactant deficiency (more than 3,500 patients) and found no clear benefit of HFOV over conventional mechanical ventilation in lowering BPD risk (as defined by an oxygen requirement

S97

at 36 weeks’ CGA). While theoretically the more gentle form of ventilation should have reduced lung injury and thus BPD risk, this has not been clearly demonstrated by trials to date. Volume-targeted and patient-triggered ventilation For infants who are intubated and treated with conventional ventilators, management strategies have evolved during the last several years that allow the patient to have more control over how and when each breath is provided by the ventilator. Synchronized intermittent mandatory ventilation allows the baby to initiate ventilator-supported breaths rather than forcing metered breaths at regular intervals out of step with the baby’s spontaneous efforts. In addition, with technical improvements in flow sensors that can accurately measure air movement in the breathing circuits used on the smallest babies, use of volume-targeted ventilation strategies can potentially minimize exposure of the lung to excess inspiratory pressure, thus minimizing barotrauma/volutrauma. Rather than setting breaths to be provided at a fixed inspiratory pressure, new neonatal ventilators can be safely programmed to generate a set tidal volume (e.g., 4–5 ml/kg per breath) and the baby determines how much inspiratory pressure is required. If that pressure is less than what would have been given by the ventilator through conventional settings, then each such breath results in less trauma. Less pressure and less injury theoretically lead to a lower risk of developing chronic lung disease. Several randomized clinical trials have compared volume-limited, patient-triggered ventilator strategies to conventional ones. In a meta-analysis of five studies involving 439 babies, volume-limited, patient-triggered strategies resulted in a lower-risk of bad outcome: death or requiring O2 at 36 weeks’ CGA (RR=0.73 [95% CI 0.57–0.93]) [15]. In response, clinical practice is changing in the NICU so that the neonatologist is thinking more about lung volumes and is increasingly cautious about avoiding over-distension. The radiologist should be aware that a comment on lung volume status is being sought in the report, even though this might have been influenced by the inspiratory status at the time the image was obtained. Although hyperinflation might be intrinsic to small airway obstruction, in the infant receiving CPAP or PPV, it might be a signal that the clinician is providing too much pressure, and thus unnecessary injury. Continuous positive airway pressure (CPAP) vs. conventional mechanical ventilation With the kinder, gentler mindset of minimizing the barotrauma and volutrauma associated with positive-

S98

Pediatr Radiol (2012) 42 (Suppl 1):S92–S100

pressure ventilation, two derivatives are logical: (1) If it is possible to avoid intubating a newborn, that is preferable, and (2) If the newborn requires intubation, the goal should be to minimize mean airway pressure and extubate as soon as possible. Animal models have suggested that starting CPAP immediately after birth prevents enough microatelectasis that some preterm infants would not require intubation (or treatment with exogenous surfactant). To test this hypothesis in humans, several trials have attempted to examine pulmonary outcomes (combined outcome of BPD or death) in infants randomized to receiving either early CPAP or intubation/surfactant. Three recent, large trials all concluded that there was no clear benefit to providing early CPAP over early intubation. Although clinical practice has been evolving toward trying early CPAP and then intubating if CPAP fails, particularly for extremely low birth-weight infants, it is not clear whether delaying surfactant therapy (early use leads to better response) could negatively impact outcome (i.e. if the infant who failed CPAP had been given early surfactant, would pulmonary outcome have been better?). It is misleading to conclude that early CPAP precludes the need for intubation; in these trials, for those infants randomized to CPAP, 40–50% ultimately failed CPAP alone and required intubation. Three major trials from three continents are summarized in Table 2. The COIN [16] (Cpap Or INtubation) trial (Australian) found a three-fold increased risk of pneumo-

thorax in infants randomized to early CPAP; however, the starting pressures used (8 cm H2O) were higher than are typically used in the U.S. and Europe (5–6 cm H2O). The SUPPORT trial [17] (U.S.) used lower CPAP pressures and did not demonstrate an increased pneumothorax risk but didn’t show a significant difference in the outcome of BPD or death. While these trials suggest a possible trend toward less BPD in the CPAP groups, no significant differences were found. The CURPAP trial [18] (European) came to the same conclusions; however, interestingly the baseline BPD rates used were much lower than those typically found in the U.S. in the age group studied. The latter confirms the implication that variation in clinical practices that are not well characterized can influence the risk of developing lung injury. Oxygen as a drug (SaO2/SpO2) Supplemental oxygen therapy has long been available for treatment of lung diseases in the newborn. Unfortunately, as with all drugs, oxygen can be toxic, and this toxicity is a major contributor to the development of BPD. While too much oxygen can cause oxidant injury, hypoxemia can also be a source of end organ damage (e.g., development of pulmonary and retinopathy of prematurity). With the advent of non-invasive monitoring of blood oxygen saturation (oximetry), the clinical management of lung disease in premature infants underwent a major change. Rather than

Table 2 Comparison of RCTs evaluating efficacy of early CPAP vs. intubation for the treatment of surfactant deficiency COIN21

SUPPORT22

CURPAP23

Year published Number of babies randomized

2008 615

2010 1,316

2010 208

GA (weeks) Comparison

25–28 early CPAP vs. intubation +/− surfactant O2 @ 36 weeks’ CGA

24–28 early CPAP vs. intubation/surfactant O2 challenge test @ 36 weeks’ CGA

25–28 early CPAP vs. intubation/surfactant O2 @ 36 weeks’ CGA

34% 39% N.S. 0.8 [0.58–1.12]

49% 54% N.S. 0.91 [0.83–1.01]

21% 22% N.S. 0.91 [0.83–1.01]

9.1% 3.0% P=0.001 No clear benefit of early CPAP High CPAP pressure. Surfactant treatment not required in intubated infants (only received by 77%)

6.80% 7.40% N.S. No clear benefit of early CPAP Those intubated and treated with surfactant were to extubated by 24 h, if possible

1.0% 7.0% N.S. No clear benefit of early CPAP Those intubated and treated with surfactant were extubated aggressively by protocol (by 1 h if possible)

BPD definition % death or BPD CPAP Intubation Significance RR [95% CI]: death or BPD Pneumothorax CPAP Intubation Significance Conclusion Design issues

Pediatr Radiol (2012) 42 (Suppl 1):S92–S100

intermittent blood sampling for analysis of blood gases (direct measurements of PaO2 and SaO2), continuous indirect monitoring of oxygen saturation (SpO2) has allowed for the titration of supplemental oxygen to avoid levels too low and too high. However, it has become apparent that limits of the SpO2 range that were associated with the best outcomes (minimizing pulmonary hypertension, retinopathy and chronic lung disease) are not at all clear. Recently, a number of trials have attempted to answer this question: at a particular gestational age, or in the setting of a particular disorder (such as retinopathy), what is the optimal range to maintain SpO2? Some studies suggest that mortality increases if the lower limit of the range is too low [19]. Excess hyperoxic exposure has been associated not only with increased short-term lung morbidity but with increased long-term risk of developing malignancies [20]. A recent meta-analysis of studies that evaluated SpO 2 ranges suggests that a trade-off is required: a higher dose with higher toxicity in return for a lower mortality [21]. Pharmacotherapy Inhaled nitric oxide A controversy surrounds the use of inhaled nitric oxide (iNO) in the NICU. While its value as a selective pulmonary vasodilator for the treatment of pulmonary hypertension in term neonates is undisputed, its use as an anti-BPD prophylactic, perhaps by acting as an anti-inflammatory agent, remains debatable. Many trials have attempted to assess the use of iNO as a preventor of BPD. Because it is such an expensive therapy, NIH recently assembled a consensus panel to review the available clinical trial data and make a clinical recommendation. This review concluded that the evidence does not support the routine use of iNO. Although some nurseries might be using iNO in premature infants for this purpose, such use is not supported by the evidence [22, 23]. Caffeine Caffeine has been used in the NICU for the treatment of apnea of prematurity. A recent large randomized controlled trial was performed to assess the safety of this drug regarding neurodevelopment, a result of theoretical concerns about impact on neuronal migration. To the surprise of all, not only was it found that neurodevelopmental outcome was improved in the infants randomized to caffeine, but it was noted that the risk of BPD was reduced. It is not clear whether these findings are a direct effect of the drug or whether they are because episodes of hypoxemia are avoided and thus exposure to hyperoxia (during desaturation episodes) is decreased. As a result, caffeine is now evolving as a therapy that improves neurodevelopmental outcomes and is also a potential therapeutic agent for the prevention of BPD [24, 25].

S99

What does the neonatologist want to know? What should the radiologist describe? With this update on BPD background, nomenclature, understanding of pathophysiology, and current management practices, the radiologist can better provide readings that meet the needs of the neonatologist. Classic descriptors of pattern, symmetry, degree of hyperinflation, interstitial density, focal or diffuse densities, heart size and comparison to prior status are most useful. Simply stating “evolving BPD or chronic lung disease” is not clear. In the clinical setting of an acute decompensation, the neonatologist wants to know about inferred mechanics (ETT placement, atelectasis), inflation status and whether there might be increased interstitial fluid. These findings might respond to mechanical adjustments or therapeutic agents (e.g., diuretics, bronchodilators, antibiotics) that directly diminish pulmonary fluid or infiltrates or dilate small airways. Comments that allow for minimizing barotrauma are particularly appreciated as they might lead to lowering the risk of BPD development. Remember that in the year 2011, one can’t make a clinical BPD diagnosis until 36 weeks’ CGA, and that the pattern of new BPD differs from old BPD.

Conclusion The pathology of new BPD or chronic lung disease is different from old BPD. The definition remains problematic, and the management has changed significantly: kinder, gentler ventilation to minimize barotrauma and minimized oxygen exposure to lessen oxygen toxicity. We have a better understanding of caffeine’s impact. However, since the institution of exogenous surfactant therapy, no new pharmacological therapies have made any clear impact on either the risk or natural history of the disorder. It is important for the radiologist to be aware of these issues in order to be able to intelligently communicate radiographic interpretations to the neonatologist.

Disclaimer The supplement this article is part of is not sponsored by the industry. Dr. Parad has no financial interest, investigational or offlabel uses to disclose.

References 1. Northway WH, Rosan RC, Porter DY et al (1967) Pulmonary disease following respirator therapy of hyaline-membrane disease. Bronchopulmonary dysplasia. N Engl J Med 276:357–368 2. Jobe AJ (1999) The new BPD: an arrest of lung development. Pediatr Res 46:641–643

S100 3. Shennan AT, Dunn MS, Ohlsson A et al (1988) Abnormal pulmonary outcomes in premature infants: prediction from oxygen requirement in the neonatal period. Pediatrics 82:527–532 4. Ehrenkranz RA, Walsh MC, Vohr BR et al (2005) Validation of the National Institutes of Health consensus definition of bronchopulmonary dysplasia. Pediatrics 116:1353–1360 5. Walsh MC, Wilson-Costello D, Zadell A et al (2003) Safety, reliability, and validity of a physiologic definition of bronchopulmonary dysplasia. J Perinatol 23:451–456 6. Walsh MC, Yao Q, Gettner P et al (2004) Impact of a physiologic definition on bronchopulmonary dysplasia rates. Pediatrics 114:1305–1311 7. Davis JM, Parad RB, Michele T et al (2003) Pulmonary outcome at 1 year corrected age in premature infants treated at birth with recombinant human CuZn superoxide dismutase. Pediatrics 111:469–476 8. Johnson AH, Peacock JL, Greenough A et al (2002) Highfrequency oscillatory ventilation for the prevention of chronic lung disease of prematurity. N Engl J Med 347:633–642 9. Marlow N, Greenough A, Peacock JL et al (2006) Randomised trial of high frequency oscillatory ventilation or conventional ventilation in babies of gestational age 28 weeks or less: respiratory and neurological outcomes at 2 years. Arch Dis Child Fetal Neonatal Ed 91:F320–F326 10. Thomas M, Greenough A, Johnson A et al (2003) Frequent wheeze at follow up of very preterm infants: which factors are predictive? Arch Dis Child Fetal Neonatal Ed 88:F329–F332 11. Parad RB, Davis JM, Marlow N et al (2009) A chronic respiratory morbidity (CRM) definition derived from 12 month respiratory symptom diaries predicts 12 m pulmonary function (PFT) and 24 m respiratory health in extremely low gestational age newborns (ELGANs). E-PAS 3450.1 12. Avery ME, Tooley WH, Keller JB et al (1987) Is chronic lung disease in low birth weight infants preventable? A survey of eight centers. Pediatrics 79:26–30 13. Cools F, Askie LM, Offringa M et al (2010) Elective highfrequency oscillatory versus conventional ventilation in preterm infants: a systematic review and meta-analysis of individual patients’ data. Lancet 375:2082–2091 14. Parad RB (2010) HFOV in preterms: an individual patients’ data meta-analysis. Lancet 375:2054–2055 15. Wheeler KI, Klingenberg C, Morley CJ et al (2011) Volumetargeted versus pressure-limited ventilation for preterm infants:

Pediatr Radiol (2012) 42 (Suppl 1):S92–S100

16.

17. 18.

19. 20. 21.

22. 23.

24. 25.

26. 27. 28.

29.

30.

a systematic review and meta-analysis. Neonatology 100:219– 227 Morley CJ, Davis PG, Doyle LW et al (2008) Nasal CPAP or intubation at birth for very preterm infants. N Engl J Med 358:700–708 Finer NN, Carlo WA, Walsh MC et al (2010) Early CPAP versus surfactant in extremely preterm infants. N Engl J Med 362:1970–1979 Sandri F, Plavka R, Ancora G et al (2010) Prophylactic or early selective surfactant combined with nCPAP in very preterm infants. Pediatrics 125:e1402–e1409 Saugstad OD (2007) Optimal oxygenation at birth and in the neonatal period. Neonatology 91:319–322 Spector LG, Klebanoff MA (2005) Childhood cancer following neonatal oxygen supplementation. J Pediatr 147:27–31 Carlo WA, Finer NN, Walsh MC et al (2010) Target ranges of oxygen saturation in extremely preterm infants. N Engl J Med 362:1959–1969 Van Marter LJ (2009) Epidemiology of bronchopulmonary dysplasia. Semin Fetal Neonatal Med 14:358–366 U.S. Dept. of Health and Human Services (2010) NIH consensus development conference: inhaled nitric oxide therapy for premature infants. Available via http://consensus.nih.gov/2010/ inofinalstatement.htm. Accessed 7 Sept 2011 Schmidt B, Roberts RS, Davis P et al (2006) Caffeine therapy for apnea of prematurity. N Engl J Med 354:2112–2121 Schmidt B, Roberts RS, Davis P et al (2007) Long-term effects of caffeine therapy for apnea of prematurity. N Engl J Med 357:1893–1902 Baraldi E, Filippone M (2007) Chronic lung disease after premature birth. N Engl J Med 357:1946–1955 Lefkowitz W, Rosenberg SH (2008) Bronchopulmonary dysplasia: pathway from disease to long-term outcome. J Perinatol 28:837–840 Palta M, Gabbert D, Weinstein MR et al (1991) Multivariate assessment of traditional risk factors for chronic lung disease in very low birth weight neonates. The Newborn Lung Project. J Pediatr 119:285–292 Palta M, Gabbert D, Fryback D et al (1990) Development and validation of an index for scoring baseline respiratory disease in the very low birth weight neonate. Severity Index Development and Validation Panels and Newborn Lung Project. Pediatrics 86:714–721 Grégoire MC, Lefebvre F, Glorieux J et al (1998) Health and developmental outcomes at 18 months in very preterm infants with bronchopulmonary dysplasia. Pediatrics 101:856–860