Keywords
premature - chronic lung disease - definition
Since its first description by Northway et al 50 years ago,[1] bronchopulmonary dysplasia (BPD) continues to be the most significant respiratory
morbidity affecting premature infants. Significant advances in obstetrics and neonatal
medicine over the last half century have resulted in an evolution in the pathophysiology
and clinical picture of BPD. In contrast to Northway et al's description of BPD as
the end result of severe lung injury due to high ventilator pressures and oxygen concentrations
in relatively mature preterm infants, the most common form of BPD today is the end
result of a complex interplay between abnormal lung development and injury/repair
in extremely preterm infants. This has presented a unique challenge for neonatologists
to find a definition that describes the severity of the lung damage and at the same
time can predict with some accuracy long-term respiratory outcome for a disease that
is continuously evolving.
Evolution of the Definition of Bronchopulmonary Dysplasia
To overcome the problem of changes in pathogenesis over time, most diagnostic criteria
have used treatment variables such as oxygen concentration and positive pressure at
different time points during the clinical course. With improving survival of extremely
immature infants and changes in care practices, these diagnostic criteria have been
modified multiple times.[2]
[3] In a major review in 2001, Jobe and Bancalari proposed severity-based diagnostic
criteria for BPD that included oxygen need for 28 days and an assessment of respiratory
support at 36 weeks' postmenstrual age (PMA; [Table 1]).[4] This definition has been used inconsistently with various modified versions either
to more accurately assess the need for oxygen by room air challenge test at the time
of assessment (physiologic definition) or to simplify the use in large populations.
This has resulted in the use of different definitions of BPD not only across trials
but also in clinical practice, thereby making meaningful comparisons of outcomes very
difficult.[5] [Fig. 1] shows the incidence of BPD in a group of extremely low birth weight infants using
some of the definitions that have been suggested in different publications.
Fig. 1 Incidence of bronchopulmonary dysplasia (BPD) by diagnostic criteria across gestational
ages (GAs).The figure shows the incidence of BPD according to different diagnostic
criteria of BPD across different GAs. A large proportion of infants in lower GA groups
need oxygen at day 28. Nearly two-thirds of those infants remain oxygen dependent
at 36 weeks'postmenstrual age (PMA) and approximately one-third of them need ≥30%
oxygen. Data are shown from 1,539 infants of 23 to 30 weeks' GA admitted to the neonatal
intensive care unit of Holtz Children's Hospital of the Jackson Health System/University
of Miami during the years 2005 to 2015 who were alive at 36 weeks' PMA.
Table 1
NIH workshop definition of BPD
|
Gestational age
|
<32 wk
|
≥32 wk
|
|
Time point of assessment
|
36 wk' PMA or discharge to home, whichever comes first
|
>28 d but <56 d postnatal age or discharge to home, whichever comes first
|
|
Treatment with oxygen >21% for at least 28 d plus
|
|
Mild BPD
|
Breathing room air at 36 wk' PMA or discharge, whichever comes first
|
Breathing room air by 56 d postnatal age or discharge, whichever comes first
|
|
Moderate BPD
|
Need for <30% at 36 wk' PMA or discharge, whichever comes first
|
Need for <30% at 56 d postnatal age or discharge, whichever comes first
|
|
Severe BPD
|
Need for ≥30% oxygen and/or positive pressure (PPV or NCPAP) at 36 wk' PMA or discharge,
whichever comes first
|
Need for ≥30% oxygen and/or positive pressure (PPV or NCPAP) at 56 d postnatal age
or discharge, whichever comes first
|
Abbreviations: BPD, bronchopulmonary dysplasia; NCPAP, nasal continuous positive airway
pressure; NIH, National Institutes of Health; PMA, postmenstrual age; PPV, positive
pressure ventilation.
Source: Adapted from Jobe and Bancalari.[4]
Pitfalls of the Current Definition of Bronchopulmonary Dysplasia
BPD is one of the few diseases arbitrarily defined by its treatment rather than by
its pathophysiology or clinical picture. This potentially has resulted in lumping
together of infants with diverse respiratory pathologies and subsequent long-term
outcomes. This limitation is reflected in an inconsistent association of diagnosis
of BPD with long-term respiratory outcomes across different studies. This has brought
to question the value of BPD diagnosis as an outcome measure for various antenatal
and postnatal treatment strategies.
Another important element in the definition of BPD is to find an optimum timing of
assessment. Though a continuous assessment of respiratory support over a period of
time, such as area under the curve, is likely to more accurately reflect the severity
of lung disease, it would be time demanding and more difficult to achieve in research
or clinical settings. As reported by Isayama et al, the predictive accuracy of the
definition of BPD for later respiratory outcomes progressively improves by delaying
the time of assessment from 34 to 44weeks'PMA.[6] However, this improvement in accuracy comes at the cost of decreasing the number
of infants who can be classified prior to transfer or discharge from the hospital.
One of the significant limitations of current definitions of BPD is the inability
to classify infants dying with severe respiratory failure before 36 weeks' PMA. These
infants reflect a significant subset of extremely preterm infants who are likely to
be the most at risk for developing severe BPD if they survived.[7] One of the problems of including these infants in any classification of BPD is that
it is very difficult to consistently differentiate between mortality from severe respiratory
failure and other causes of death.
Some of the changes in clinical practice over the last decade have posed additional
challenges to the definition of BPD. The increasing use of high-flow nasal cannula
oxygen therapy has resulted in a new respiratory support strategy that combines oxygen
supplementation with a variable degree of positive end-expiratory pressure. The classification
of these infants on the basis of oxygen concentration alone, not including the positive
pressure that they receive, may result in inaccurate estimation of lung disease severity.
On the other extreme are the infants who receive 100% oxygen at very low nasal cannula
flows. The classification of these infants on the basis of nasal cannula oxygen is
currently a challenge as there are very limited data on the effect of different nasal
cannula flows on effective fraction of inspired oxygen and airway pressure.[8]
Another group of infants who are difficult to classify as BPD are those requiring
positive-pressure ventilation for abnormal respiratory control or severe airway abnormalities
with no or minimal parenchymal lung disease. These infants should not be classified
as BPD, but their proper classification would require some degree of clinical, radiographic,
and/or functional assessment of parenchymal lung disease, all of which are subject
to interpretation, have low specificity, and are difficult to apply in a large clinical
population.
The Way Forward
The definition of BPD has different meanings for different stakeholders. While for
an epidemiologist, a definition should accurately reflect the disease process and
should be easy to use across neonatal units over time, for clinicians and researchers
involved in developing preventive or treatment strategies, a definition with some
correlation with long-term outcome is important. On the other hand, for a parent of
an infant admitted to a neonatal intensive care unit, survival and the effect on long-term
respiratory health is paramount. A key challenge for an ideal definition of BPD is
to devise objective diagnostic criteria to appropriately categorize infants with different
severity of lung disease while excluding infants with nonpulmonary causes of respiratory
insufficiency. In addition, these criteria will need to be simple to use and require
no or minimal extra effort by caregivers to ensure consistent use in large populations
in different clinical settings.
One of the key issues with the diagnostic criteria of BPD has been the prediction
of later respiratory course. This good intention is fraught with a risk of coming
short on expectations since unlike a dichotomous definition of presence or absence
of BPD, lung development and injury is a continuum from normal to severe dysfunction.[9] In addition, lung development, additional injury, and the repair process continue
well beyond the time of diagnosis of BPD and can be influenced by various factors
after discharge from the hospital. With increasing survival of more immature infants,
many of them meet the criteria of oxygen use at 36 weeks and are diagnosed as BPD,
but a large proportion of them have only mild or minimal problems with long-term respiratory
health. As expected, the diagnosis of severe BPD is the one more consistently associated
with worse long-term respiratory outcomes.[10] It is these infants with severe lung injury who must be properly classified so that
they can be the primary target for prevention or treatment strategies that should
improve their long-term outcome.
Recently, there has been increasing realization of the inadequacies of current definitions
of BPD, and a concerted effort is being made to refine the diagnostic criteria and
reflect current clinical practices more accurately.[11] These new criteria will need to include infants dying of severe respiratory failure
prior to 36weeks' PMA, categorize infants on different flows of nasal cannula oxygen,
and provide specific guidance for including only infants with significant parenchymal
lung disease. The National Institute of Child Health and Development recently held
a workshop focused on BPD and its definition, and a report will be published in the
near future. Long-term biomarkers for evaluating lung injury,[12] imaging studies such as functional magnetic resonance imaging (MRI) to evaluate
lung structure,[13] and innovative ways to reliably assess pulmonary function will be needed to move
toward a new classification of BPD that is more based on pathophysiology and objective
lung function evaluation.
In conclusion, it is unlikely that a new definition of BPD will fulfill the requirements
of all the stakeholders, but a consensus definition, which is relatively simple to
use, encompasses current clinical practices, and provides a reasonable prediction
of respiratory outcomes, is attainable and will improve the care of these infants.
