Normal breathing occurs via the nose (Fig. 4.1). Inspired and expired air is conditioned by the nasal mucosa, whose surface area and blood supply are such that particles are filtered and temperature and humidity are modified. The nasal vascular system can achieve a 25 °C gradient between the nasal ostia and the nasopharynx (Godfrey 1994), a property that enables air breathing at very different environmental temperatures while maintaining a remarkable constancy of temperature and humidity in the lower airway, thereby optimizing cellular function and ciliary action. Given its vascularity, it is not surprising that the nose is a source of nitric oxide, that trauma to the nasal passages can be induced by nasal continuous positive airway pressure (NCPAP) prongs and other cannulae, and that epistaxis can require hospital admission.

The nose can sniff out noxious substances and protect the airway from particles by sneezing. It can increase its resistance rapidly, switching on a watery secretion “like a tap” (Godfrey 1994), and can trigger laryngeal closure and reduced ventilatory drive (Editorial 1992). The extreme protective reflex response to nasal irritation is the dive reflex that is typified by apnea with glottic closure, bradycardia, and redistribution of blood flow to the brain, heart, and adrenals (see also Sect. 47.​3.​1) (de Burgh Daly 1997; Editorial 1992). The dive reflex can be triggered by nasal water, tobacco smoke, or ice applied to the trigeminal area (Angell James and de Burgh Daly 1969; de Burgh Daly 1997). Water entry into the nose is dramatic even for adults who “suffer a sense of impending suffocation and the almost impossibility of voluntarily making an inspiratory effort” (de Burgh Daly 1997). Exaggerated dive reflex responses can cause total upper airway closure and cardiac arrest and have been implicated in accidental and deliberately induced deaths (de Burgh Daly 1997).

Nasal breathing in the newborn and infant is the predominant but not obligatory form of breathing and is related to the high position of the epiglottis (Fig. 4.1) (Seikel et al. 2005). There is a nasal cycle whereby resistance alternates from one nostril to the other every 2–4 h in 80 % of humans (Widdicombe 1986). Pressure in the axilla or on the lateral chest increases ipsilateral nasal resistance (Widdicombe 1986). Resistance is highest in the regions adjacent to the external ostia, whose dimensions are altered by the alae nasi muscles that cause nasal flaring during exercise hyperpnea and respiratory distress, especially in newborns. Newborn nasal resistance exceeds that of the adult in absolute terms but is less than that of their lower airways. Nearly half of the adult’s total airway resistance is nasal. Thus, minor increases in nasal resistance can produce major overall effects on breathing (Editorial 1992). In adult humans, nasal occlusion results in a switch to mouth breathing in <5 s (Editorial 1992). Both this switch to oral breathing and the resumption of nasal breathing following release of the occlusion are delayed by nasal and pharyngeal anesthesia. Occlusion of the nasal passages in infants results in mouth breathing in ~8 s (range <1–32 s) (Editorial 1992; Rodenstein 2004). The response is quicker with advancing age but is decreased in rapid eye movement (REM) sleep, more so at 6 weeks than in the newborn period (Editorial 1992). These observations led to a proposal that the response may be related to the sudden infant death syndrome (SIDS) (Editorial 1992). In premature infants, the change to oral breathing results in a sixfold increase in pulmonary resistance, which over time might result in fatigue with hypoventilation (Miller et al. 1987). Oral breathing places an increased burden on the lower airway to modify the temperature and humidity of inspired gas, presumably via changes in lower airway vascularity. Preventing nasal breathing can decrease gas exchange and functional residual capacity (FRC) (Editorial 1992). Postoperative nasal packing in adults can lead to arterial hypoxemia, apnea, and sleep disturbance, with the latter also occurring in normal volunteers after local anesthesia of the nasal passages (Editorial 1992). In newborn lambs, nasal obstruction also affects gas exchange, with hypoxemia, hypercapnia, and acidosis being observed. These effects are aggravated by carotid body denervation. The impact of nasal obstruction diminishes with advancing age (Editorial 1992). Blockage of the nose (choanal stenosis being an extreme example) can cause obstructive sleep apnea (OSA) (Marcus 2000). Human neonates whose nasal resistance is increased by birth trauma or trauma secondary to repeated suctioning breathe orally. They can have cyanotic episodes, stridor, and hypercapnia; these problems disappear with resolution of the nasal injury (Miller et al. 1987).

In summary, the nose provides air conditioning, and its sensory receptors initiate reflexes that alter upper airway patency and mediate extended ventilatory changes in response to a challenge.

The muscles of the tongue, especially the genioglossi, maintain patency of the oral cavity. Pharyngeal patency is a function of the cavity dimensions (Rodenstein 2004) and the balance between the opening forces exerted by pharyngeal muscle activities and the collapsing ones exerted by the tissue–intrapharyngeal pressure gradient and by pharyngeal mucosal adhesion (Fig. 4.2). Patency is critically influenced by posture. Narrowing occurs with neck flexion and hyperextension and can alter cavity pressures required to achieve flow. Narrowing can also follow an inspiratory occlusion of the mouth/nose (a load) that augments diaphragmatic activity and thus negative pressure within the pharynx. Upper and lower airway afferents induce reflex compensatory change in glossopharyngeal muscles to offset the load effect (Fig. 4.2) (Bailey and Fregosi 2006).

Obstructive apnea occurs during sleep in ~1–4 % of children and adults (Marcus 2000). In children, a persistent partial form of upper airway obstruction, obstructive hypoventilation, is seen (Marcus 2000). In adults, a spectrum of decreased inspiratory airflow is seen in snoring, upper airway resistance syndrome, and OSA. Pharyngeal OSA occurs when a critical tissue pressure (P _crit) exceeds the intrapharyngeal opening pressure (Fig. 4.2) (Marcus 2000; Dempsey et al. 2010) and is influenced by factors that alter pharyngeal dimensions, namely, adenotonsillar hypertrophy, obesity (less common in children than adults), low subglottic airway volume, and craniofacial structural abnormalities. The major role played by the central motor output to the oral cavity and pharyngeal muscles that modify P _crit is emphasized by the fact that OSA occurs only in sleep, predominantly in REM sleep in children (Marcus 2000). The following factors also stress the importance of central control in OSA. Obstructive apnea occurring during mixed apnea in preterm newborns involves a direct central mechanism, as shown by the onset of pharyngeal and/or laryngeal closure before diaphragmatic activation (Idiong et al. 1998) (see Sect. 47.​3.​1). In adults pharyngeal closure can accompany central apnea (Marcus 2000). Central arousal patterns during OSA differ between children and adults (Marcus 2000). Other factors affecting the incidence of OSA are gender, a familial tendency, and racial/ethnic factors (Marcus 2000). Anti-inflammatory medications can benefit OSA, suggesting a role for inflammation in its etiology (Praud and Dorion 2008). Therapy with NCPAP improves OSA by increasing intrapharyngeal pressure and its transverse diameter (Rodenstein 2004), by augmenting lower airway volume and probably by stimulating breathing via pharyngeal pressure sensors (Angell James and de Burgh Daly 1969). In children and adults with OSA, CPAP therapy can improve their metabolism.

The abilities to suck, swallow, and breathe are developed during fetal life (Harding 1986). At all ages, swallowing is accompanied by laryngeal closure. Term newborns and infants up to 12–18 months of age can lock the larynx into the nasopharynx (Figs. 4.1 and 4.3). This separation of the nasopharyngeal–laryngeal air passage from the oropharyngeal nutritive passage enables the term infant to breathe and feed simultaneously but with decreased ventilatory drive. This ability is not developed in the preterm neonate whose sucking pattern is also immature (Lau 2006). Coordination of sucking–swallowing–breathing is defined by the ability to feed “by mouth with no overt signs of aspiration, oxygen desaturation, apnea, or bradycardia” and when “a ratio of 1:1:1 or 2:2:1 suck: swallow: breathe” is attained (Lau 2006). Safe swallowing occurs at end-expiration and end-inspiration (Lau 2006). In term newborns, 55 % of swallows occur at these points of the breathing cycle. In preterm infants, 55 % of swallows occur during deglutition apnea (Lau 2006).

Nonnutritive swallowing (NNS) is “essential for survival,” since >2 L of oral and nasal secretions are produced daily. Without NNS, “the lungs rapidly fill with these secretions, producing death within a few days” (Thach 2005). When pharyngeal secretions reach a critical volume, neural receptors, in the interarytenoid notch, stimulate the laryngochemoreflex which triggers NNS (Thach 2005). The critical pharyngeal volume for NNS may be increased by applied positive pharyngeal pressure, providing an explanation for the decrease in NNS frequency with CPAP (Thach 2005). In adult humans NNS occurs mainly during expiration. In infants NNS occurs at any time in the respiratory cycle (Thach 2005). Volitional swallowing in adults clears the pharynx and reduces the protective laryngeal adductor reflex. During sleep, the protection provided by NNS and glottic closure is imperfect, and some aspiration is common in normal individuals (Thach 2005). Swallow breaths are short inspirations that precede a swallow, occurring when the upper pharyngeal sphincters are closed (Thach 2005). These breaths are thought to result in pharyngeal air being inhaled as opposed to being swallowed (Thach 2005). The increased gastric air found when positive pharyngeal pressure is applied suggests that CPAP can alter normal pharyngeal clearance mechanisms and infers the need to titrate the applied pressure carefully (Thach 2005). Finally, swallowing may be related to sleep apnea and be part of recovery from sleep apnea (see Sect. 47.​3.​1) (Thach 2005).

In summary, patency of the oral cavity and the pharynx and the coordination with swallowing and sucking are critical for breathing. The term, but not preterm, newborn can suck and breathe simultaneously, although ventilatory drive is decreased. Physiological control of breathing and swallowing involves important sensory input that limits aspiration and protects the lower airway. Intensive care therapies can alter normal protective, breathing and swallowing functions.

Understanding the actions of the laryngeal intrinsic muscles is a necessary first step in gaining insights into the nature of the central control of breathing. The posterior cricoarytenoids (PCA) are the sole abductors of the vocal cords. There are several adductors, with the thyroarytenoids (TA) being the major ones (Fig. 4.4). Some adductors, the cricothyroid and the vocalis part of TA, enhance abduction by tensing the vocal cords. This restricts vertical movement, improving flow control in the more horizontal plane. The magnitude of abductor activity exceeds that of the adductors when glottic size is increased and vice versa, consistent with relationships between glottic size and flow resistance (England et al. 1982a; Insalaco et al. 1990; Kuna et al. 1988). Animal studies indicate that, in general, glottic opening and closing are achieved by reciprocal laryngeal abductor and adductor activities (Fig. 4.5) (Bartlett 1989; Harding 1986; Hutchison et al. 1993). Concurrent actions of laryngeal abductor and adductor muscles during inspiration and expiration may occur in adult humans, but their nature is unknown (Insalaco et al. 1990; Kuna et al. 1988, 1990).

Behavioral state is a major controlling factor of breathing pattern. In fetal lambs, during the high-voltage electrocorticogram (ECoG) state, apnea (no phasic laryngeal abductor or diaphragmatic activities) and tonic laryngeal adductor activity occur. Laryngeal adductor activity can be enhanced by laryngeal contact with cooled lung liquid, by fetal movements, by “arousals,” by swallowing, by uterine contractures, and by fetal hypoxemia. In the low-voltage ECoG state, laryngeal and diaphragmatic activities are similar to those seen postnatally (Harding 1986). Hypercapnia mainly enhances fetal breathing in the low-voltage ECoG state (Harding 1986). In general, active laryngeal retardation of lung liquid egress is reported in the absence of fetal breathing movements and is related to lung growth (Harding 1986). Stimulation of the superior laryngeal nerve (SLN) produces laryngeal closure (Harding 1986) that will prevent aspiration of noxious material, e.g., meconium. The coordination of the acts of fetal swallowing (Harding 1986) and “inspiratory” activities of the respiratory muscles stresses the developmental importance of the controller’s ability to switch patterns rapidly between upper airway closure (laryngeal adduction) with lack of diaphragmatic activity (apnea) and laryngeal abductor and pump activities (see Sect. 47.​3.​1).

Airway volume is high in the fetus compared to the newborn. It has been speculated that during different fetal behavioral states, the brain is setting the homeostatic limits for high and low airway volume to prepare the fetus for the major needs of subglottic volume control at birth (Hutchison 2007) (see also Sect. 47.​3.​1). Life’s greatest airway volume challenge, the establishment and maintenance of an air-filled airway at birth, is usually overcome without a hitch by breathing patterns whose complexities exceed those used during the majority of adult life. At birth, an incremental breathing pattern retains more inspired volume than is expired (Fig. 4.6) (Hutchison et al. 1994). Incremental breathing shares features with grunting with the addition of timing changes in laryngeal and pump muscle activities at end-expiration when the onset of pump muscle activity occurs before laryngeal muscles open the glottis fully. The result is a decrease in expired volume and an increase in EEV (Hutchison et al. 1994). This incremental mechanical process is aided by the airway stability accrued from the increased end-expiratory positive subglottic pressure that limits the development of a negative airway pressure during inspiration. Given their tendency to chest wall distortion, it is not surprising that incremental breathing can occur in preterm neonates (Eichenwald et al. 1992). Incremental breathing is also seen in gasping during acute cerebral hypoxia–ischemia (Hutchison et al. 2002). Gasping is typified by brief laryngeal abductor and pump muscle activities followed by long periods of laryngeal adductor activity with apnea (Hutchison et al. 2002). Subglottic pressure is increased, maintaining airway volume and likely promoting autoresuscitation (Hutchison et al. 2002).

Nasal and laryngeal afferent inputs alter laryngeal abductor and adductor activities and thus glottic size. The laryngochemoreflex (LCR), whose afferent pathway is the SLN, results in protective responses, including swallowing, apnea, obstructed respiratory efforts, cough, hypertension, and arousal from sleep (Thach 2008). An obstructive apnea with bradycardia results if the LCR stimulus is persistent, especially in the immature or depressed brain. This LCR response alters with age – coughing becomes the major response. However, acquisition of a viral infection in infancy can rekindle its potency (Thach 2008). It is augmented by hypoxemia and anemia and has been implicated in apnea of prematurity and SIDS (Thach 2008) (see Sect. 47.​3.​1).

Other laryngeal SLN afferents, including those from mechanoreceptors, drive receptors, and temperature receptors, affect laryngeal intrinsic muscle activities. Although SLN section does not alter the eupneic breathing pattern, these afferents are important. Upper airway bypass can alter the inspiration-inhibition Hering–Breuer reflex. Application of acid to the larynx in animals, mimicking chronic aspiration, alters the subsequent response to an applied airway load (see Sect. 47.​3.​1). During noninvasive ventilation, nonsynchronous delivery of airflow to the upper airway in neural expiration results in laryngeal closure (Rodenstein 2004; Scharf et al. 1978). Thus, noninvasive ventilation and pacing of a paralyzed diaphragm should be applied synchronously with neural inspiration (Rodenstein 2004; Scharf et al. 1978).

In intact animals, compensatory glottic abduction follows single-breath total occlusion of inspiration or expiration at the mouth/nose. During the unimpeded expiration following a single inspiratory occlusive load, unopposed TA activity can produce laryngeal closure. Thus when no air enters the lung, a compensatory reflex mechanism can prevent expiratory volume loss, maintaining absolute subglottic volume. Lower airway afferent inputs affect glottic size (Bailey and Fregosi 2006). Stretch receptor discharges, induced by PEEP, decrease expiratory TA activity (Harding 1986). By contrast increased expiratory glottic adduction follows rapidly adapting receptor stimulation, e.g., with deflation or irritant stimulation secondary to a pneumothorax (Bartlett 1989), or follows C-fiber stimulation, e.g., with experimental pulmonary edema (Bartlett 1989). Direct and indirect stimulation of chest wall muscle afferents can invoke several flow patterns involving glottic closure (Bartlett 1989; Stecenko and Hutchison 1991).

While pseudoasthma can be fabricated by partial glottic closure (Rodenstein 2004), matching laryngeal control of flow pattern to maintain optimal subglottic airway volume may explain changes reported with true increased airway resistance. Subglottic inspiratory flow limitation in croup is associated with expiratory flow resistance, probably glottic in origin (Argent et al. 2008). Increased lower airway resistance in adult asthmatic patients induces a breathing pattern characterized by laryngeal expiratory flow retardation (Collett et al. 1983; Sekizawa et al. 1987). In adults, resistive loads applied at the mouth also decrease expiratory glottic size (Brancatisano et al. 1985). This change may occur immediately, suggesting that resistive loading, unlike total occlusive (elastic) loading, produces sudden flow changes that stimulate airway receptors to cause laryngeal adduction (Brancatisano et al. 1985). Therapy with CPAP reverses the expiratory laryngeal adduction in some asthmatic subjects, suggesting that airway pressure changes may alter the dynamics between stretch receptor and irritant receptor stimulations, thus changing the breathing pattern (Collett et al. 1983). In normal adults a voluntary deep breath can decrease laryngeal resistance and lower airway resistance (Sekizawa et al. 1987). By contrast, bronchoconstrictive stimulation in normal adults can result in inspiratory and expiratory glottic narrowing (Higenbottam 1980). An increased resistance to inspiratory flow may be advantageous in that more transpulmonary pressure is applied to opening a constricted peripheral airway. If lower airway volume and resistance changes affect laryngeal function, can absence of laryngeal control affect lower airway resistance? Laryngeal bypass during invasive ventilation is associated with atelectasis and with the development of increased lower airway resistance (Hutchison and Bignall 2008). The latter effect may stiffen the conducting airways and thus stabilize total airway volume but demand increased effort. In summary, the data reflect the importance of laryngeal subglottic volume control as a determinant of lower airway resistance and point to interactions between lower airway afferents and coordinated laryngeal and pump muscle activities in this dynamic process (see also Sect. 47.​3.​1).

Central control of coordinated laryngeal and diaphragmatic muscle activities is determined by developmental stage, behavioral states, metabolism (temperature), central inputs, and centrally acting chemicals. Postnatally in lambs, increased expiratory TA activity with decreased PCA activity occurs during the NREM state. By contrast, expiratory TA activity in REM is mainly absent, while PCA activity is variable (Harding 1986). In normal adult humans, expiratory TA activity is absent during stable NREM, while expiratory PCA activity decreases in NREM and is variable in REM (Kuna et al. 1988, 1990). At all ages, expiratory TA activity occurs at arousal. Thus behavioral state is a key factor in subglottic airway volume control that, in turn, is important in sleep apnea.

Increased central drive, with hyperventilation and/or hypercapnia, promotes laryngeal abduction in inspiration and expiration in adults and term newborns, in whom increased PCA, diaphragmatic and intercostal muscle activities occur (Bartlett 1989; Insalaco et al. 1990; Kuna et al. 1994; Wozniak et al. 1993). Laryngeal adduction can occur during hypercapnic hyperventilation in preterm neonates, probably due to chest wall distortion (but see also Sect. 47.​3.​1) (Eichenwald et al. 1993). This adduction also occurs in adults at the mechanical limits of airway volume (Brancatisano et al. 1983). In general, however, increased central drive decreases laryngeal resistance unless mechanical limitations are present.

Decreased central drive with hypocapnia diminishes expiratory glottic size and is associated with periodic breathing in adults (Kuna et al. 1993; Rodenstein 2004). Laryngeal closure has been noted in human newborns and infants with apnea or suspected apparent life-threatening events (Ruggins and Milner 1991, 1993). During central apnea and periodic breathing in lambs, expiratory TA activity is noted (Praud 1999). In depressed human infants at birth, laryngeal closure can block intubation, and, in lambs, acute cerebral hypoxia–ischemia results in expiratory TA activity with laryngeal closure (Hutchison et al. 2002). Centrally acting depressant drugs diminish central drive and, in lambs, produce laryngeal closure with apnea (Praud 1999). In former preterm infants, up to ~55–60 postmenstrual weeks, exposure to anesthesia and/or operative stress can result in apnea postoperatively (see Sect. 47.​3.​1). Thus, the intensivist should avoid hypocapnia during noninvasive ventilation and be aware that central depression can result in apnea with glottic closure.

Hypoxia stimulates expiratory laryngeal abduction and increases ventilation in human adults and newborn lambs (England et al. 1982b; Insalaco et al. 1990; Praud et al. 1992). In lambs, the increase in ventilation is dependent upon carotid body input – a sudden decrease in that input induces expiratory TA activity (Praud et al. 1992). However, animal and human data indicate that the effects of carotid body input vary depending upon central state and the presence/absence of other inputs (de Burgh Daly et al. 1979). If animals are vagotomized and paralyzed, direct carotid body stimulation can induce expiratory TA activity, and exposure to hypoxia after airway vagal blockade or intrathoracic vagal section results in laryngeal adduction (Bailey and Fregosi 2006). In adult humans, the degree of expiratory glottic adduction with hypoxia exceeds that during hypercapnia (England et al. 1982b). Thus, it is argued that the “pure” carotid body reflex response is laryngeal expiratory adduction that will retain airway volume and promote oxygenation. This pure response can be countered by the presence of ventilation which increases airway stretch receptor feedback that results in an overall expiratory laryngeal abduction response. In the author’s view, a unifying explanation for the diverse findings with hypoxia is that the effect of carotid body input is to augment the central coordinated output to laryngeal and pump muscles that is selected under different conditions (see Sect. 47.​3.​1). When central depression exists, protective inputs and/or absence of volume-related inputs produce an apneic response with glottic closure that is augmented by hypoxia. This can be seen in the “vagal” preterm infant who is sedated for a surgical procedure. At intubation laryngeal closure and apnea/bradycardia can be triggered. If hypoxemia ensues, the problems are aggravated. When tidal ventilation and airway volume feedback are restored, the impact of any ongoing hypoxemia is to augment breathing. Hering and Breuer found that their volume-related reflexes were active even with hydrogen breathing. This fits the modern focus on ventilation for resuscitation from asphyxia.

The respiratory system is in a passive state when there is no inspiratory muscle activity (P _mus = 0). A passive condition is present in the paralyzed patient but can also occur without paralysis following hyperventilation (Mortola 2004). Under passive conditions, resting lung volume is determined by two opposing forces. The outward recoil of the chest wall results from the elastic characteristics of the diaphragm/abdomen and of the rib cage. The outward recoil of the chest wall directly opposes the inward recoil of the lung. Inward recoil is determined by the viscoelastic properties of the lung, including the mechanical properties of the lung tissues, and the collapsing pressure generated at the alveolar air–liquid interface.

During spontaneous breathing, the passive respiratory mechanics can be measured using occlusion techniques. Occlusion of the airways at lung volumes above the resting lung volume relaxes the respiratory muscles by stimulating the slowly adapting stretch receptors and the vagally mediated Hering–Breuer inflation reflex. Passive tests assess the mechanical properties of the entire respiratory system. Simultaneous measurement of transpulmonary pressure allows measurement of respiratory system compliance (C _rs) to be partitioned into lung (C _L) and chest wall (C _w) components. Transpulmonary pressure (P _tp) is derived from the difference in pressure at the airway opening (P _ao) and pleural pressure (or elastic recoil pressure, P _el), measured using an esophageal balloon, liquid-filled catheter, or miniature pressure transducer.

Other passive measurements of respiratory mechanics include forced expiratory maneuvers to provide measures of maximal flow at FRC (V′_max,FRC) (Lum et al. 2006a). Partial forced expiratory flow–volume curves in infants are obtained by rapid compression of the chest through end-inspiratory inflation of a jacket placed around the chest and abdomen. During tidal breathing, this rapid thoracoabdominal compression (RTC) technique provides a V′_max,FRC that is similar to forced flows at low lung volumes (e.g., maximal expiratory flow MEF_{25 %,FRC}). Modification of the RTC technique by raising lung volume of the infant towards total lung capacity (so-called raised volume RTC or RVRTC) allows assessment of forced expiratory flows over a more extended range of volumes. Reported measures include forced vital capacity (FVC) and forced expiratory volumes after defined time periods (e.g., 0.4, 0.5, or 1 s, termed FEV_0.4, FEV_0.5, and FEV₁, respectively) and the ratio of these values e.g., (FEV_t/FVC, where t is the time period chosen).

In contrast, dynamic tests of respiratory mechanics allow the relative contributions of the airways and lung tissues to be partitioned, in addition to separately identifying the chest wall contribution to airway and tissue mechanical properties. Dynamic mechanics derive mechanical variables from analysis of dynamic waveforms obtained either during normal (tidal) breathing or via imposed oscillatory waveforms. During tidal breathing, multilinear regression is used to determine dynamic respiratory system compliance (C _dyn,rs) and resistance (R _rs). Kano and colleagues showed that consideration of volume dependence is important when using this approach to assess dynamic mechanics during mechanical ventilation. Consideration of volume dependence is important because ventilatory settings may “push” the pressure–volume curve onto the flattened upper portion of the curve (Kano et al. 1994). In the overdistended lung, fits obtained by multilinear regression are improved by the inclusion of a volume-dependent elastance term in the regression model:
where P _ao is the airway opening pressure, V is the tidal volume, E ₁ + E ₂ V is the total elastance over the cycle (E ₁ and E ₂ V representing the volume-independent and volume-dependent portions of elastance, respectively), and the final term P _A,EE represents the alveolar pressure at end-expiration.

An important consideration during the interpretation of dynamic respiratory mechanics is the concept of frequency dependence. Kano showed that increasing the ventilation rate from 30 to 80 breaths/min in a group of near-term newborn infants increased E _rs and R _rs by a mean of 8.3 and 17.5 %, respectively (Kano et al. 1994). Frequency dependence and the mechanical properties of the respiratory system under normal tidal breathing conditions are also determined using forced oscillatory techniques (FOT). These techniques are essentially noninvasive and are especially useful in infants and children as no active subject cooperation is required. The technique involves application of a pressure waveform (forcing function) to the airway opening and measurement of the resultant flow. The measured impedance reflects the complex viscoelastic resistance of the respiratory system (Z _rs). Simultaneous measurement of the transpulmonary pressure and hence chest wall impedance (Z _w) allows determination of the lung impedance (Z _L) by subtraction. The impedance (Z) can be separated into the resistance (R _rs), the impedance component in phase with flow, and reactance (X _rs), representing the impedance components in phase with volume (elastance) and acceleration (inertance).

The low-frequency forced oscillation technique (LFOT) measures impedance over a range of frequencies (usually ~0.5–20 Hz). LFOT is obtained in infants by invoking the Hering–Breuer inflation reflex to inhibit temporarily any respiratory muscle activity, during which brief time (6–10 s) the LFOT signal is applied. Fitting the constant-phase model to the resultant pressure and flow data permits partitioning of respiratory mechanics into frequency-independent airway resistance (R _aw) and airway inertance (I _aw) and a constant-phase tissue component. This constant-phase tissue component is described by [(G − jH)/ω^α, where G and H are coefficients for tissue damping and elastance, respectively, ω is angular frequency, and α determines the frequency dependence of the real (pressure change in phase with flow) and imaginary (pressure change in phase with volume) parts of the impedance] (Hantos et al. 1992a).

The interrupter technique provides a further option to assess partitioned mechanics. The interrupter technique involves measurement of changes in airway opening pressure after a mid-expiratory occlusion, which is used to calculate airway resistance (R _aw). After airway occlusion at mid-expiration, there is a biphasic change in P _ao: the immediate rapid rise in P _ao represents the resistive pressure drop across the conducting airways and is followed by a secondary slower increase in P _ao (often referred to as P _dif) generally attributed to stress recovery in the respiratory tissues (lung and chest wall) and gas redistribution associated with ventilation inhomogeneity (Bates et al. 1988). A major limitation of this technique is that it requires mechanical ventilation and paralysis to obtain reliable data.

Lung development represents a period of considerable change in both lung architecture and volume and is affected by several factors including genetic, in utero and postnatal environmental exposures, somatic growth, puberty, changes in muscularity, and ongoing alveolarization. Our understanding of the interactions between these factors is based on a mix of pathological and physiological data, which are predominantly cross-sectional in nature. Consensus between these studies is lacking. While some of the lack of agreement is attributable to technical approaches used by different authors, a comprehensive understanding of factors influencing respiratory mechanics over time can only be answered by strong longitudinal data, which are urgently required. These tests of lung mechanics provide us with essential insight into how tissue properties change with development, and a full understanding of the changes that occur during healthy development is an important step to detect the subsequent effects of disease and response of the individual to therapeutic interventions. Measurements need to be accompanied by accurate records of height (length) and weight to facilitate interpretation of longitudinal data.

Sex-specific differences in respiratory mechanics are well recognized in infants and young children and are a particularly important consideration in forced expiratory maneuvers in infants. Airways are larger in females before puberty, and this difference is detectable from infancy (Tepper et al. 1986a; Hoo et al. 2002a): V′_max,FRC is approximately 20 % higher in girls over the first 9 months of life (Hoo et al. 2002a). Similar sex-dependent increases in V′_max,FRC are evident in preterm girls compared to preterm boys, suggesting such differences in lung function may be developmental rather than environmental in origin (Stocks et al. 1997). Larger lungs have been reported in males between the ages of 6 weeks and 14 years (Thurlbeck 1982). Presence and timing of the growth spurt is an important confounding factor, with lung growth occurring out of phase to somatic growth, especially in boys (Degroodt et al. 1986; Quanjer et al. 1989; Schrader et al. 1984). Lung development in females is almost complete following menarche but continues throughout puberty in males (Neve et al. 2002).

There is increasing evidence that the periconceptional and intrauterine exposures impact on development of the respiratory system and consequently impact on respiratory mechanics in infancy and childhood. Factors known to influence lung mechanics after birth include maternal smoking, nutritional deprivation and intrauterine growth restriction, infection and/or inflammation, and maternal glucocorticoids. Prenatal exposure to nicotine may cause abnormal airway branching and dimensions as well as increase airway smooth muscle and collagen deposition resulting in reduced lung function at birth that persists into early adulthood regardless of postnatal exposure (Hayatbakhsh et al. 2009; Svanes et al. 2004). Lung function irregularities include airflow limitation, reduced FEV₁, and airway hyperreactivity (Sandberg et al. 2011; Wongtrakool et al. 2012). Chronic restriction of nutrients and/or oxygen in late pregnancy impairs development of the fetal small airways and lungs, including reduced numbers of alveoli that are also enlarged with increased septal wall thickness and basement membranes compared to nonexposed infants (Pike et al. 2012). Such differences in alveolar number can influence the subsequent rate of disease progression, including an accelerated rate of decline in FEV₁ with advancing age (Stocks and Sonnappa 2013).

Several cross-sectional studies have looked at the changes in airway resistance (reflecting airway size) during childhood. Hall and colleagues using LFOT described a quasilinear decrease in R _aw with increasing length during infancy in 37 infants, including some with repeat measurements (Hall et al. 2000). Lanteri and Sly examined changes across the pediatric age range and also reported linear decreases in both R _rs and R _aw with increasing height (when plotted on a log–log plot) in 51 children (range 3 weeks–15 years) with healthy lungs (Fig. 4.7) (Lanteri and Sly 1993). Pooled data across the preschool age range reinforces the consistent relationship seen with increasing height (Beydon et al. 2007a). In a study of 40 subjects, across the first 5 years of life, Gerhardt et al. demonstrated an increase in lung conductance (the reciprocal of resistance), of approximately fivefold over the weight range examined (Gerhardt et al. 1987a). This increase occurred at a slower rate than the increase in FRC, which led to a rapid drop in specific conductance (Fig. 4.8b). This rapid decrease in specific airway resistance or conductance through early life has been described previously (Doershuk et al. 1974; Stocks and Godfrey 1977). Tepper et al. measured maximal FEF at FRC, from partial expiratory flow–volume curves and corrected for changes in lung volume: they reported higher flows in the neonatal range which decreased to a steady value through the 2nd year of life, similar to those reported elsewhere in older children (Tepper et al. 1986a). Differential increases in anatomic dead space volume (twofold), compared to FRC (threefold), have also been shown in the pediatric population (Wood et al. 1971).

Rapid postnatal increases in lung volume are reported widely. FRC increases by a magnitude of 40 times (from 80 mL in infants to 3,000 mL in adults) (Dunnill 1982) and over ten times in total lung weight (Polgar and Weng 1979). The rapid growth of the distal lung, in comparison to relatively steady changes in airway dimensions, is a major feature of respiratory system development beyond birth. After the appearance of primitive saccules at approximately 28 weeks gestation, mature alveoli with secondary septa appear from 34 weeks gestation onwards. Alveolar multiplication was initially thought to cease at around 2 years of age (Thurlbeck 1982). More recent estimates indicate that increase in alveolar number continues until at least 8 years (Dunnill 1982). Recent advanced imaging studies using hyperpolarized helium magnetic resonance imaging (MRI) provide evidence of ongoing alveolar multiplication in adolescence (Narayanan et al. 2012). Increase in alveolar size (expansion) is also evident during infancy and childhood, with recent epidemiology studies suggesting lung size increases into early adulthood, when chest wall development is complete (Quanjer et al. 1989; Reid 1977).

Based on cross-sectional studies, specific lung compliance falls during early infancy before remaining relatively constant from the early preschool years. Tepper et al., based on measurements of compliance from the linear portion of the static pressure–volume (PV) curve in almost 50 infants, described an increase in lung compliance over the age range studied, but decreasing specific lung compliance with increasing body length in the first 2 years of life (Tepper et al. 2001). Gerhardt et al. examined similar changes in 40 infants and children over the first 5 years of life. Lung compliance increased markedly (by ~20–25-fold) over the weight range examined (Fig. 4.8a) but in proportion to FRC, resulting in a constant-specific compliance over the first 5 years of life. The same pattern of findings was reported in 63 children aged 2–7 years (Greenough et al. 1986). This initial rapid period of alveolar growth is also supported by evidence from measurements using the interrupter technique and FOT. Lanteri and Sly showed a biphasic change in P _dif with age: P _dif was highest in young infants, falling rapidly over the first 12 months, before increasing again after approximately 5 years of age (Lanteri and Sly 1993). Hall and colleagues, using LFOT measures of tissue mechanics derived by fitting data to the constant-phase model, showed that both tissue parameters G and H decreased in a quasihyperbolic manner with increasing length between 7 weeks and 2 years of age (Hall et al. 2000). The difference in the pattern of change in G and H compared to the change in R _aw with increasing length is further evidence of dysanapsis: tissue mechanical properties change more rapidly than airway mechanics over the first 2 years of life (Fig. 4.9).

The chest wall of the neonate and infant, and to a lesser extent the growing child, is especially compliant compared to the compliance of the chest wall in the adult subject. A highly compliant chest wall in infancy results in less opposition to the tendency of the lung to collapse, promoting a low residual lung volume and also distortion of the chest wall, resulting in a loss of volume during inspiration. The disproportionately high compliance of the chest wall compared to the compliances of the lung means that the compliance of the respiratory system (C _rs) is approximately equal to the compliance of the lung (C _L) during childhood, especially in neonates and infancy (Davis et al. 1988; Gerhardt et al. 1987a). Several active (dynamic) mechanisms of respiratory mechanics serve to partially compensate for the instability in FRC associated with a compliant chest wall and tendency for small airway closure during tidal breathing. Infants acutely increase their end-expiratory lung volume (EELV) and maintain it above resting lung volume by modulating (abbreviating) their expiratory time (Mortola and Saetta 1987) (Fig. 4.10a). In addition, they reduce expiratory flow through post-inspiratory activity of the diaphragm and inspiratory chest wall muscles (Kosch and Stark 1984) (Fig. 4.10b) (Lopes et al. 1981), stiffening the chest wall and by increasing laryngeal resistance (± glottic closure) briefly (Kosch et al. 1988). Such dynamic modifications of respiratory mechanics occur via neural (vagal) receptor-mediated reflexes. Restriction of lung function measurement to periods of quiet sleep is necessary to reduce the variability in measurements resulting from such dynamic processes (Henschen and Stocks 1999).

The changes in chest wall, lung, and respiratory system mechanics over the first 4 years of life were examined by Papastamelos et al., in 40 subjects (Papastamelos et al. 1995). The authors used a modified Mead–Whittenberger technique (Cook et al. 1957): manual ventilation overrode respiratory drive and relaxed the respiratory muscles, avoiding the need for intubation or induction of a Hering–Breuer reflex. In this cohort, the ratio of chest wall to lung compliance fell from a mean (SD) of 2.9 (1.1) to 1.3 (0.4) (p < 0.005). Chest wall to lung compliance ratios are even higher in preterm infants (Gerhardt and Bancalari 1980). The stiffening of the chest wall, due to rapid rib ossification, increasing musculature, and altered chest wall configuration (Bryan and Wohl 1986; Mead 1979; Openshaw et al. 1984; Leiter et al. 1986), is such that near-adult values, where lung and chest wall compliance are equal (Mittman et al. 1965), are reached after the first year of life.

Stiffening of the chest wall over the first year of life allows the infant to shift its maintenance of FRC from dynamic elevation of EELV to a more passive mechanism achieved by the balance of outward recoil of the chest wall and inward recoil of the lung (Colin et al. 1989). Stiffening of the chest wall likely also explains the increase in P _dif in children after age of 5 years observed by Lanteri and Sly (Lanteri and Sly 1993). FRC as a proportion of total lung capacity (TLC) increases with age through childhood, in the majority (Engstrom et al. 1956; Mansell et al. 1977; Weng and Levison 1969), but not all (Schrader et al. 1988), studies to date. Residual volume (RV), as a proportion of TLC, also increases with age through childhood (Merkus et al. 1993), while closing capacity (measured using single-breath nitrogen washout) decreases significantly with age, as a proportion of TLC, converging towards RV (Mansell et al. 1977). This increased RV/TLC ratio suggests that increased stiffness of the chest wall is not entirely compensated for by increase in expiratory muscle force over time (Schrader et al. 1988). The study by Merkus et al. also highlighted the need to use TLC as a measure of lung growth and not FVC, due to the fact that an increasing RV/TLC with age led to errors in maximal expired flows due to measurement at progressively higher lung volumes (Merkus et al. 1993).

The postnatal period is characterized by ongoing rapid lung development, with differing rates of growth seen in the airway, lung parenchyma, and chest wall. The dysanapsis that occurs between the components of the respiratory system has important consequences for measurements of lung mechanics performed during infancy and childhood. Accounting for these developmental changes and dysanapsis is necessary to optimize disease detection and subsequent assessment of an individual’s response to interventions. Insights into these changes are largely derived from cross-sectional studies, which have identified important sources of potential error in existing methodology. Strong longitudinal data are now urgently required to confirm these apparent patterns and answer the important questions that remain.

Transient tachypnea of the newborn (TTN) is characterized by delayed resorption of fetal lung fluid within the distal airspaces. The mechanical consequences of TTN include reduced lung volume (functional residual capacity, FRC) and consequently also decreased lung compliance (increased elastance), tidal volume, and respiratory rate (Benito Zaballos et al. 1989). Although increased interstitial fluid might be expected to increase lung tissue resistance, there are no reported studies of tissue resistance in infants with TTN.

Like TTN, RDS is predominantly a disease of the distal lung parenchyma. Surfactant deficiency promotes atelectasis, typically characterized by low lung volumes and reduced lung compliance. Lung function is also reflective of the degree of structural maturation, as surfactant deficiency is predominantly a disease of the premature infant.

In the intubated infant, lung resistance measures are highly variable and not reproducible (22–32 %, ICC <0.75), in contrast with less variable and more reproducible measurements of lung compliance (obtained from esophageal manometry). Hence, the clinical relevance of dynamic mechanics measurements of resistance in intubated newborn infants with respiratory distress is questionable (Gappa et al. 2006a).

Measurements of impedance, obtained using forced oscillatory mechanics (a quasistatic lung mechanics measurement), may provide a more reliable assessment of lung mechanics. Dorkin measured respiratory impedance in six paralyzed and intubated infants, three of whom also had pulmonary interstitial emphysema (Dorkin et al. 1983). The tracheal tube contributed to almost all the inertance and approximately 50 % of the respiratory system resistance in the intubated infant. After subtraction of the impedance of the endotracheal tube, resistance ranged from 22 to 34 cm H₂O/L/s, compliance from 0.22 to 0.68 mL/cm H₂O, and inertance from 0.0056 to 0.047 cm H₂O/L/s (Gappa et al. 2006a; Dorkin et al. 1983).

The low-frequency forced oscillation technique (LFOT) evaluates impedance simultaneously across a range of frequencies (usually ~0.5–14 Hz). Fitting of the resultant impedance data to the constant-phase model permits estimation of partitioned mechanical variables of the airways and the parenchymal tissues (Hantos et al. 1992b). Using the LFOT, impedance measurements in naïve (surfactant-deficient) newborn preterm lambs showed that respiratory system resistance (R _rs) and reactance (X _rs) are markedly frequency dependent (Pillow et al. 2001a). Respiratory system resistance in the preterm infant is also dominated by the resistive properties of the tissues, contrasting sharply with the predominant airway contribution to respiratory system resistance in later life (Pillow et al. 2005). Although compliance is also low (increased elastance), the tissue resistance is disproportionately greater, resulting in mechanical uncoupling of the parenchyma and increased hysteresivity (ratio of tissue resistance to tissue elastance) (Pillow et al. 2001a, 2005). Increased surfactant pool size (Pillow et al. 2004a) and lung volume recruitment (Pillow et al. 2004b) both enhance mechanical coupling of the tissues, evident as a lowering of the hysteresivity ratio in the lung. Increased contribution of the tissues to respiratory system mechanics, and the associated increased frequency dependence, results in elevation of the resonance frequency of the lung (Pillow et al. 2001a). Frequency dependence also becomes less marked with increasing postnatal age (Fig. 4.11) (Pillow et al. 2005).

Variability in measurements of lung mechanics in newborn infants in part reflects lung maturation associated with advancing gestation: lung compliance increases 0.17 mL/cm H₂O/week, to reach mean (SD) values of 2.50 (0.07) mL/cm H₂O at term equivalent (Bhutani et al. 2005). Other factors include in utero fetal exposures such as placental nutrition, inflammation, antenatal steroids, and postnatal treatments including caffeine, glucocorticoids, surfactant, and ventilation modality. Early treatment with caffeine in surfactant-treated immature baboons increased C _rs over the first day of life and increased ventilatory efficiency index (Yoder et al. 2005).

The initial benefit of antenatal steroids on improved lung compliance at birth fades beyond 1 week after initial exposure (McEvoy et al. 2008). There is no evidence of that antenatal steroids have a persistent effect on lung mechanics: infants born at term after exposure to up to three courses of betamethasone at gestations between 25 and 33 weeks had no difference in gas mixing or lung volume/mechanics when compared to nonexposed term controls (Hjalmarson and Sandberg 2011). Like antenatal steroids, postnatal steroids also increase respiratory system compliance (Durand et al. 1995). Low-dose and high-dose postnatal dexamethasone for CLD achieve a similar increase in C _rs suggesting equal effectiveness for improvement of lung volume (McEvoy et al. 2004). Early postnatal dexamethasone may be more effective than late dexamethasone in achieving lower R _aw/R _rs and higher specific conductance (sG _aw) (Merz et al. 1999; Vento et al. 2004), which may indicate a degree of fixed change in the airway walls in infants receiving delayed glucocorticoid treatment.

During administration, surfactant represents a fluid bolus in the airway that increases inspiratory and expiratory time constant, associated with an acute increase in respiratory system resistance. Effective delivery of surfactant to the lung thus benefits from a transient increase in the inspiratory time and potentially brief increase in the delta P (peak inspiratory pressure–positive end-expiratory pressure) to overcome the increase resistance resulting from the fluid bolus. Treatment with exogenous surfactant (da Silva et al. 1994) or increases in positive end-expiratory pressure (PEEP) during recruitment from atelectasis increases lung volume: (Dimitriou et al. 1999) lung volume increases by 61 ± 39 % within a median of 4 min after surfactant administration, with altered distribution of lung volume towards the dorsal rather than ventral compartment. The increase in lung volume after surfactant or volume recruitment maneuvers increases maximal compliance of the respiratory system, with higher tidal volumes achieved at lower pressures than required prior to treatment (Miedema et al. 2011a; McEvoy et al. 2010). Importantly, the increase in compliance also impacts on the time constant (τ = R _rs × C _rs) of the respiratory system. Changes in the inspiratory and expiratory time constants need to be monitored as they necessitate adjustment of the inspiratory and expiratory times and influence the maximal frequency that can be used during ventilation with passive expiration.

Compared to treatment of RDS with synchronized intermittent mandatory ventilation (SIMV), a small clinical trial showed that infants treated with high-frequency oscillatory ventilation (HFOV) have early and sustained improvement in pulmonary mechanics and higher dynamic respiratory compliance (Vento et al. 2005). The improvement in pulmonary mechanics in patients treated with HFOV is likely associated with enhanced lung volume recruitment. The neonatal lung with RDS exhibits hysteresis (Miedema et al. 2011b). Hence lung recruitment maneuvers that aim to recruit the lung and subsequently ventilate it on the most compliant part of the deflation pressure volume curve can achieve effective ventilation with minimum pressure and volume cost of ventilation. Importantly, recruitment of the lung by increasing mean distending pressure during HFOV is also associated with changes in compliance and the time constant for volume delivery during HFOV (Pillow 2012). Consequently, volume delivery can vary substantially over the time course of an HFOV volume recruitment maneuver (Miedema et al. 2012) and potentially result in rapid fluxes of the partial pressure of arterial carbon dioxide. The effect of compliance on volume delivery during HFOV is more evident at lower frequencies (Pillow 2012; Pillow et al. 2001b). The development of HFOV ventilators that incorporate volume guarantee during HFOV will provide protection from such rapid changes in ventilation during HFOV.

Infants with RDS often have their clinical course complicated by the persistence of a patent ductus arteriosus (PDA), which presents as a left–right shunt. Ligation of a PDA (left–right shunt) increased dynamic compliance by 77 % in 16 newborn infants measured before and after ligation, but did not influence dynamic R _rs or mean airway pressure (Szymankiewicz et al. 2004a).

Although meconium aspiration syndrome (MAS) is often considered as a high airway resistance disease due to the presence of meconium in the airways, studies in rabbits show that the acute aspiration of meconium also causes a significant reduction in the lung–thorax compliance (Sun et al. 1993). Impaired compliance after MAS is responsive to standard surfactant instillation (Sun et al. 1993). Larger volume (15 mL/kg) surfactant lung lavage is an emerging treatment for MAS: following surfactant lung lavage, dynamic compliance approximately doubled, and airway resistance nearly halved in a group of MAS infants on mechanical ventilation (Szymankiewicz et al. 2004b). These changes in dynamic respiratory mechanics were associated with a decrease in mean (SD) airway pressure from 12.4 ± 3.6 to 5.4 ± 2.1 cm H₂O within 48 h after surfactant lung lavage.

Meconium appears to have a differential effect on hyperreactivity of the airways and the lung tissue. Whereas tracheal smooth muscle reactivity increases with increasing meconium concentration in response to histamine and acetylcholine, a negative correlation was observed in the lung tissue (Mokry et al. 2007). Airway hyperresponsiveness 2 weeks after birth (approximately 5 days after cessation of ECMO) is responsive to bronchodilator treatment (Koumbourlis et al. 1995) as evidenced by percent change in MEF₂₅ and the MEF₂₅/FVC ratio. Airway hyperreactivity after MAS is also reduced by budesonide (Mokry et al. 2007).

Limited information is available for the long-term lung function outcomes after MAS. Neonates treated with ECMO for meconium aspiration have better long-term lung function outcomes than infants with diaphragmatic hernia treated with ECMO in the neonatal period (Spoel et al. 2012a), most likely reflecting differences in lung capacity and structure. At least 50 % of children who had MAS in the neonatal period have evidence of trapped air on lung function during mid-late childhood. There are no published reports evaluating responsiveness to bronchodilators beyond infancy after neonatal MAS.

Persistent pulmonary hypertension of the newborn (PPHN) presents with profound hypoxia either in the absence of significant lung disease (primary PPHN) or hypoxia out of proportion with the degree of respiratory disease (secondary PPHN – complicating RDS, meconium aspiration, pneumonia, etc.). Lung function in 1-year-old infants who had PPHN and who were treated with inhaled nitric oxide (iNO) but not extracorporeal membrane oxygenation (ECMO) was compared with lung function outcomes of infants who were randomized to receive either ECMO or conventional management as part of the UK ECMO trial. V′_max,FRC was lower than predicted in all three groups (p < 0.001). There were no statistical differences between the three groups in the Z-scores for V′_max,FRC (Hoskote et al. 2008). Nonetheless, only 26 % of iNO-treated infants had V′_max,FRC Z-scores below normal compared to 37 and 56 %, respectively, for ECMO and CM groups. A prospective evaluation of lung function in infants with PPHN treated with/without iNO was reported by Dobyns and colleagues (1999): compared to healthy control infants of the same age, there were no differences in lung volume or passive respiratory mechanics for infants with PPHN, nor was there any effect of iNO on later pulmonary function in the PPHN group. Together, these results suggest that iNO treatment does not worsen outcome compared to ECMO or conventional management.

Infants with congenital diaphragmatic hernia have impaired airway function in the first year of life: maximal expiratory flows at FRC were a mean Z-score of −1.5 lower than healthy term infants (i.e., 1.5 standard deviations below the mean FRC for healthy term infants) with no evidence of significant change between 6 and 12 months of age (Spoel et al. 2012b). Conversely, and perhaps surprisingly, given the usual association of CDH with pulmonary hypoplasia in at least one lung, Spoel noted that measured values of functional residual capacity were relatively high (47 % fell above the normal range) (Spoel et al. 2012b). As expected, mean Z-score for FEV₁ and FVC was negatively influenced by the presence of chronic lung disease, the duration of ventilation, and ECMO support (Spoel et al. 2012a). Patients with CDH develop hyperinflation, with elevated plethysmographic FRC at 1 year (Hofhuis et al. 2011) that is still evident at 8–12 years (Spoel et al. 2012a; Hamutcu et al. 2004; Majaesic et al. 2007).

Spoel showed that greater impairment was evident in infants with CDH who required extracorporeal membrane oxygenation (Spoel et al. 2012b). However, poorer outcome in ECMO-treated CDH may reflect the initial severity of disease rather than ECMO: Beardsmore showed that ECMO per se did not worsen respiratory function at 1 year of age compared to conventionally ventilated controls (Beardsmore et al. 2000).

Spoel and colleagues also noted deterioration in lung function over time (5–12 years) in patients with CDH who were treated with ECMO in the neonatal period (see Table 4.1) (Spoel et al. 2012a): mean (SE) SDS score for FEV₁ after bronchodilation was higher at 5 years (−0.71 (0.40)) than at 8 years (−2.27 (0.36)) and 12 years (−2.73 (0.61)). Deterioration over time may be related to maldevelopment of the alveoli and pulmonary vessels with disturbed lung growth. It is also possible that increased susceptibility to recurrent respiratory tract infections may contribute to deterioration in lung function over time. An active lifestyle and healthy eating pattern may be especially important for children with CDH to counteract the deterioration in lung function over time, given the known positive effect of participation in sports and the negative interaction of BMI with exercise capacity (van der Cammen-van Zijp et al. 2011). Similarly, it may be especially important to treat recurrent infections of the respiratory tract and actively manage asthma to avoid limitation of exercise capacity and deterioration of lung function. Prolonged follow-up of children with CDH is warranted.

BPD is most commonly considered as the chronic respiratory condition complicating extreme prematurity and/or prolonged neonatal mechanical ventilation. Factors that independently influence the tidal flow–volume-derived indices include variations in disease severity, degree of alteration to airway and lung mechanics, and also the requirement for oxygen supplementation (Baldwin et al. 2006). There are, nonetheless, quantitative differences in tidal breathing function measured in BPD infants when compared with measurements obtained from healthy term and preterm controls. BPD infants are consistently more tachypneic than their healthy counterparts (Baldwin et al. 2006), but differences in other lung function variables are less consistent. Tidal volume (V _T) is either less than (Paetow et al. 1999) or similar to (Patzak et al. 1999; Schmalisch et al. 2003) values obtained in healthy infants: V _T relative to healthy infants may also decrease with advancing postnatal age (Tepper et al. 1986b), likely reflecting the altered maturational progression in lung structure and function. BPD infants have increased minute ventilation (MV). Increased MV is primarily due to increased respiratory rate rather than increases in V, in part reflecting increased dead space ventilation.

Infants with BPD have an increased work of breathing. Within the tidal flow–volume loop, increased work of breathing is evident from linear or concave expiratory limb morphology. BPD infants also have a less variable shape to the tidal flow–volume waveform, indicating that they are functioning near their maximum respiratory capability with minimal reserve capacity to adapt to changes in intrinsic (e.g., airway obstruction, infection, or aspiration) or extrinsic (e.g., cold stress) environment (Baldwin et al. 2006). Tidal flow indices are difficult to interpret in part due to opposing effects of neurorespiratory control and lung mechanics on the morphology of the tidal flow waveform. Whereas BPD infants have higher respiratory drive (quantified from the mean inspiratory flow: V _T/t _I where t _I is inspiratory time), the ratio of the time to peak expiratory flow relative to expiratory time (t _PTEF:t _E) after methacholine-induced airway obstruction can either increase or decrease (Clarke et al. 1994). Normal fluctuations in V _T, t _I, and MV seen in healthy infants in response to alternate hypoxic–normoxic breath testing are not evident in BPD infants, indicating reduced chemoreceptor sensitivity to hypoxic stimulus (Calder et al. 1994).

Lung and chest wall mechanics of infants with bronchopulmonary dysplasia (BPD) were extensively reviewed by Gappa and colleagues (2006b). Changes are reflective of maldevelopment of the lungs and airways and remodelling of the lung parenchyma and airway walls associated with increased collagen content. In preterm infants, respiratory system compliance (C _rs) increases relative to body weight over the first 2 years of life, while an initially high respiratory system resistance (R _rs) decreases over the same period (Baraldi et al. 1997a). Airway walls of preterm infants are more compliant than term infants as evidenced through measurements obtained using the high-speed interrupter technique (HIT) (Henschen et al. 2006).

Following the NICHD consensus conference in 2001 (Jobe and Bancalari 2001), some investigators evaluated differences in lung function according to whether infants had mild, moderate, or severe chronic lung disease. Using the single occlusion technique, Hjalmarson and Sandberg showed that infants with severe bronchopulmonary dysplasia (born at mean 25 weeks GA) had increased specific conductance and decreased specific compliance (sC _rs) compared to healthy preterm infants (born at mean 29 weeks gestation) (Hjalmarson and Sandberg 2005). There were no differences in lung mechanics between the healthy preterm infants and those with mild or moderate bronchopulmonary dysplasia. The data published by Tortorolo et al. indicate a similar trend towards lower C _rs at day 7 in infants who later went on to develop mild BPD and to a greater extent in severe BPD (Tortorolo et al. 2002). Their results contrast with the findings by Shao and colleagues in infants of similar gestations that infants with BPD had decreased C _rs (10.8 vs 16.0 mL/kPa/Kg), but no difference in specific sC _rs (0.69 vs 0.75 mL/kPa/kg) or R _rs (6.6 vs 7.3 cm H₂O/mL/s) compared to non-ventilated preterm infants. Infants with BPD do not show an improvement in airway resistance (R _aw, measured by plethysmography) over a 4-month period (Shao et al. 1998).

Several authors have attempted to use lung function as a predictor of BPD. The value of initial R _rs for prediction of lung disease is unclear: some have found increased initial R _rs is associated with respiratory outcome at 1 year (Choukroun et al. 2003; Lui et al. 2000; Snepvangers et al. 2004), whereas others observed that R _rs is not predictive of BPD (Tortorolo et al. 2002). Both Lui (Lui et al. 2000) and Merth (Merth et al. 1997) showed that early RDS is not an important determinant of later lung function. Serial measurements of compliance, however, show that lower C _rs values after day 5 were evident in infants who later developed severe BPD. Merth observed that BPD infants have reduced C _rs at 1 year of age irrespective of RDS (Merth et al. 1997).

V′_max,FRC obtained using the tidal rapid thoracic compression technique is reported consistently as being lower than healthy controls throughout the first 3 years of life, regardless of the BPD era or treatment strategies (Lum et al. 2006b). Reduced expiratory flows are indicative of abnormal structural and functional development of the airways (Fig. 4.12). Serial lung function measurements in infants with both BPD and “healthy” unsedated preterm infants indicate a decline over the first year of life, suggesting that factors other than BPD may contribute to abnormal airway function in preterm infants (Hoo et al. 2002b). Raised volume rapid thoracic compression (RVRTC) measurements show similar mild–moderately severe airflow obstruction. However the predominance of airflow obstruction primarily at low lung volumes indicates the pathology is more likely to involve the small peripheral rather than larger central airways (Lum et al. 2006b). Concurrent increased residual volume (RV), FRC, and RV/TLC (total lung capacity) ratios were more marked in infants with recurrent wheeze, indicative of a degree of hyperinflation and gas trapping. Intermediate values (between healthy controls and wheezy BPD infants) were observed for RV, FRC, and RV/TLC in non-wheezy preterm infants (Robin et al. 2004), likely reflective of abnormal airway growth or the effects of preterm delivery. Forced deflation from FRC showed that preterm infants with developing BPD have severe lower airway obstruction as early as 3–4 weeks postnatal age. Increased flows and FVC following bronchodilation indicate that reopening of obstructed airways is achieved with bronchodilators (Motoyama et al. 1987).