Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2018 May 27;596(23):6009–6019. doi: 10.1113/JP275682

Afferent neural feedback overrides the modulating effects of arousal, hypercapnia and hypoxaemia on neonatal cardiorespiratory control

Kathleen J Lumb 1, Jennifer M Schneider 1, Thowfique Ibrahim 1, Anita Rigaux 1, Shabih U Hasan 1,✉
PMCID: PMC6265552  PMID: 29676798

Abstract

Key points

  • Evidence obtained at whole animal, organ‐system, and cellular and molecular levels suggests that afferent volume feedback is critical for the establishment of adequate ventilation at birth.

  • As a result of the irreversible nature of the vagal ablation studies performed to date, it was difficult to quantify the roles of afferent volume input, arousal and changes in blood gas tensions on neonatal respiratory control.

  • During reversible perineural vagal block, profound apnoeas and hypoxaemia and hypercarbia were observed, necessitating the termination of perineural blockade.

  • Respiratory depression and apnoeas were independent of sleep state. We demonstrate that profound apnoeas and life‐threatening respiratory failure in vagally denervated animals do not result from a lack of arousal or hypoxaemia.

  • A change in sleep state and concomitant respiratory depression result from a lack of afferent volume feedback, which appears to be critical for the maintenance of normal breathing patterns and adequate gas exchange during the early postnatal period.

Abstract

Afferent volume feedback plays a vital role in neonatal respiratory control. Mechanisms for the profound respiratory depression and life‐threatening apnoeas observed in vagally denervated neonatal animals remain unclear. We investigated the roles of sleep states, hypoxic‐hypercapnia and afferent volume feedback on respiratory depression using reversible perineural vagal block during the early postnatal period. Seven lambs were instrumented during the first 48 h of life to record/analyse sleep states, diaphragmatic electromyograph, arterial blood gas tensions, systemic arterial blood pressure and rectal temperature. Perineural cuffs were placed around the vagi to attain reversible blockade. Postoperatively, during the awake state, both vagi were blocked using 2% xylocaine for up to 30 min. Compared to baseline values, pHa, Pao2 and Sao2 decreased and P ac o2 increased during perineural blockade (P < 0.05). Four of seven animals exhibited apnoeas of ≥20 s requiring the immediate termination of perineural blockade. Breathing rates decreased from the baseline value of 53 ± 12 to 24 ± 20 breaths min–1 during blockade despite an increased P ac o2 (P < 0.001). Following blockade, breathing patterns returned to baseline values despite marked hypocapnia (P ac o2 33 ± 3 torr; P = 0.03). Respiratory depression and apnoeas were independent of sleep states. The present study provides the much needed physiological evidence indicating that profound apnoeas and life‐threatening respiratory failure in vagally denervated animals do not result from a lack of arousal or hypoxaemia. Rather, a change in sleep state and concomitant respiratory depression result from a lack of afferent volume feedback, which appears to be critical for the maintenance of normal breathing patterns and adequate gas exchange during the early postnatal period.

Keywords: apnea, fetal to neonatal transition, respiratory control, pulmonary gas exchange, sleep states, vagal Input

Key points

  • Evidence obtained at whole animal, organ‐system, and cellular and molecular levels suggests that afferent volume feedback is critical for the establishment of adequate ventilation at birth.

  • As a result of the irreversible nature of the vagal ablation studies performed to date, it was difficult to quantify the roles of afferent volume input, arousal and changes in blood gas tensions on neonatal respiratory control.

  • During reversible perineural vagal block, profound apnoeas and hypoxaemia and hypercarbia were observed, necessitating the termination of perineural blockade.

  • Respiratory depression and apnoeas were independent of sleep state. We demonstrate that profound apnoeas and life‐threatening respiratory failure in vagally denervated animals do not result from a lack of arousal or hypoxaemia.

  • A change in sleep state and concomitant respiratory depression result from a lack of afferent volume feedback, which appears to be critical for the maintenance of normal breathing patterns and adequate gas exchange during the early postnatal period.

Introduction

Establishment of pulmonary gas exchange is one of the most vital adaptations that must occur for the successful transition from fetal to neonatal life (Wong et al. 1998; Greer et al. 2006; Greer, 2012; Hooper et al. 2015a; Hooper et al. 2015b; Samson et al. 2018). Evidence suggests that vagal innervation of the lung plays a critical role in establishing and maintaining adequate alveolar ventilation and pulmonary gas exchange during the early postnatal period (Fedorko et al. 1988; Wong et al. 1998; Harris & Milsom, 2001; Lalani et al. 2001; Lalani et al. 2002). However, the validity of some of these studies may have been compromised as a result of anaesthesia, possible secondary laryngeal obstruction and tracheostomy (Hasan et al. 2000). Furthermore, few studies have evaluated the role of afferent feedback in the immediate postnatal period. A number of studies have attempted to circumvent the limitations associated with the previous studies (Wong et al. 1998; Lalani et al. 2001; Lalani et al. 2002). Wong et al. (1998) reported that intrathoracic vagal denervation performed during the prenatal period caused fetal lambs to develop life‐threatening respiratory failure at birth. In a subsequent study, vagal denervation performed during the early postnatal period led to persistent hypoxaemia and apnoeas followed by respiratory failure within 24h post‐denervation (Lalani et al. 2001). Further investigations demonstrated that the mechanisms of respiratory failure included decreased functional residual capacity, attenuated expiratory braking, early suppression of augmented breaths, poor respiratory system compliance and atelectasis (Lalani et al. 2001; Lalani et al. 2002).

One of the most consistent but deleterious effects of vagal denervation performed during the postnatal period included postoperative respiratory depression and apnoeas requiring prolonged assisted ventilation and supplemental oxygen (Hasan et al. 2000). The mechanisms of the profound life‐threatening respiratory depression in the immediate postoperative period remain unclear. The slow postoperative recovery cannot be explained on the basis of anaesthesia or surgical procedures because both sham and denervated animals underwent identical duration of anaesthesia and similar surgical procedures except vagal denervation (replaced with sham procedure). It is well established that arousal serves a potent stimulus for respiration and plays a critical role in the termination of apnoeas (Phillipson & Sullivan, 1978; Gauda et al. 2009; Dempsey et al. 2010; Guyenet & Abbott, 2013). In addition, an interaction exists between apnoea threshold and P ac o2 (Xie et al. 2001). Subsequent to the ground‐breaking discovery of mechanically activated Piezo1 and Piezo2 isoforms (Coste et al. 2010; Delmas et al. 2011), Piezo2 (FAM38B and B2) comprising more than 30 transmembrane domains has been demonstrated as the major and distinct mechanically‐activated cation channel in dorsal root ganglion neurons. In a series of elegant studies, (Nonomura et al. 2017) provided evidence that deletion of Piezo2 (involved in somatosensory mechanotransduction) in the jugular and dorsal root ganglion neurons detected in the subpopulation of vagal sensory neurons lead to death at birth, whereas Piezo2 deletion in the nodose ganglion was associated with abnormal breathing pattern and blunted Hering–Breuer reflex during adulthood (Goridis, 2017). Thus, recent evidence at the organ‐system (Wong et al. 1998; Harris & Milsom, 2001; Lalani et al. 2002) and cellular and molecular levels (Nonomura et al. 2017) suggests that pulmonary mechanical forces and volume signalling play an important role in pulmonary blood flow (Lang et al. 2017) and the establishment of breathing at birth. Because the knockout mice models and vagal denervation in our previous studies were not reversible, it remained unclear whether the uniformly observed severe postoperative respiratory depression in vagally denervated newborns was a result of the interruption of volume feedback, hypoxaemia, a change in P ac o2 threshold or a lack of arousal. One way to clarify the roles of sleep states, blood gas tensions and volume feedback on respiratory depression is to utilize reversible vagal perineural blockade.

The present study aimed to investigate the roles of sleep state, hypoxaemia and hypercarbia in respiratory depression using reversible perineural vagal blockade during the early neonatal period. We hypothesized that, during the early neonatal period, an absence of afferent volume feedback would override arousal state and hypercarbia with respect to maintaining adequate alveolar ventilation.

Methods

Surgical preparation

All procedures were performed in accordance with the Canadian Council on Animal Care and the study protocol was approved by the Animal Care Committee of the University of Calgary. To facilitate acclimatization, reduce animal stress and ensure time of birth within a 12 h window, time‐dated pregnant ewes of mixed breed were brought to the University of Calgary Animal Care Centre 7–10 days prior to the expected term (mean ± SD: 147 ± 3 days) delivery date. The pregnant ewes were observed for delivery every 2–4 h during the weekdays and at 12 h intervals during the weeknights and over the weekends by one of the investigators. After spontaneous onset of labour, all lambs delivered unassisted vaginally. None of the newborn lambs were delivered by Caesarean section or received resuscitation. Seven time‐dated neonatal lambs were instrumented and all studies were completed within 40 h of birth (six of seven lambs by 24 h and one of seven lambs by 40 h; 19.1 ± 12.4 h). The lambs weighed 4.8 ± 1.1 kg on the day of the studies.

Surgical instrumentation was performed under general anaesthesia using 5% isoflurane in oxygen for induction and 2–2.5% for maintenance. The duration of surgery was 147 ± 16 min. The animals were ventilated using a large animal volume controlled ventilator (Model 613; Harvard Apparatus, Holliston, MA, USA). Observing sterile conditions, the jugular vein and carotid artery were exposed via a 2 cm neck incision lateral to the trachea and directly below the thyroid cartilage. Once exposed, polyvinyl catheters (inner diameter 1 mm, outer diameter 2 mm; Portex, Hythe, Kent, UK) were inserted 7 cm into the jugular vein and carotid artery and secured in place. The arterial catheter was used to record arterial blood pressure and heart rate in addition to drawing blood samples for arterial pH and blood‐gas tensions. Intra‐operatively, after the insertion of the arterial catheter, arterial blood was drawn for pH and blood gas tensions (P ac o2 and Pao2) measurements. The pHa (7.35–7.45), P ac o2 (35–45 Torr) and Pao2 (80–100 Torr) were maintained by adjusting the ventilator rate and tidal volume during surgery. The rectal temperature was maintained (39–39.5°C) intra‐operatively and during the experiments using a heating pad (Microtemp water heating pump and Maxitherm vinyl blanket; Models 3783M and 3784C, respectively; Jorgensen Laboratories, Loveland, CO, USA).

Following the insertion of the vascular catheters, lambs were placed in prone position for the implantation of electrodes to record electrocorticogram (ECoG), electrooculogram (EOG), and nuchal electromyogram (EMGNK). The ECoG, EOG and EMGNK were used to define sleep states (Prechtl, 1967, 1974; Grigg‐Damberger, 2016). Subsequently, three diaphragmatic electrodes were implanted into the right costal diaphragm via a 2 cm incision, parallel to the ribs, at the tenth intercostal space. Only two of the three electrodes (bipolar) were used to obtain an optimal diaphragmatic signal. All of the incisions were sewn in layers using size 0 silk. An 8‐Fr chest tube (Argyle; CardinalHealth, Dublin, OH, USA) was inserted in the sixth intercostals space and secured with a purse‐string suture, to minimize the accumulation of intrapleural air, and placed under a water seal at –12 cmH2O.

The vagi were exposed through the midline incision previously used for the insertion of vascular catheters (as described above) and perineural vagal cuffs were placed to achieve reversible vagal blockade (Fig. 1 A). To reduce the risk of intra‐operative perineural cuff dislodgement and damage to the vagi, the perineural cuffs were placed following the implantation of vascular catheters and electrodes. The perineural vagal cuffs, placed perineurally around the vagi, were prepared from a polyvinyl cuff (inner diameter 2 mm, outer diameter 3 mm), 40 mm in length and slit longitudinally. Two fine catheters were attached to the caudal and rostral ends of the perineural cuff to infuse 2% xylocaine or saline. The open ends of the perineural cuff were sealed using inert and sterile modelling clay (Fig. 1 A). The combined dead space of the catheters and the vagal cuffs was established prior to the cuff placement around the vagi. Before closing the incisions, the vagal cuffs were checked for leaks using normal saline (0.9% NaCl) flushes.

Figure 1. Perineural vagal cuff and experimental design.

Figure 1

Open in a new tab

Schematic representation of perineural vagal cuff (A): the sterile polyvinyl perineural vagal cuffs, 40 mm in length and slit longitudinally, had two catheters attached to the caudal and rostral ends to infuse saline or 2% xylocaine (perineural blockade). The longitudinal slit and the open ends of the perineural cuff were sealed using silk ties and sterile inert modelling clay, respectively. The dead space of the ingress, perineural cuff and egress catheters was measured preoperatively and just prior to the experiments. Experimental design (B): once spontaneous breathing and normal arterial blood gas tensions were established, both vagi were bathed in saline for 15 min (baseline period). Thereafter, the vagi were bathed in 2% xylocaine. Subsequently, the vagi were flushed and immersed in normal saline for 15 min (washout period). The vagal blockade was performed during the awake state. EMGdia, sleep states, arterial blood pressure (mmHg), heart rate (beats min–1) and rectal temperature (°C) were monitored and recorded continuously throughout the study. [Color figure can be viewed at http://wileyonlinelibrary.com]

Experimental design and data analysis

The experimental design is illustrated in Fig. 1 B. Postoperatively, the ventilator rate and isoflurane were decreased to facilitate the onset of spontaneous breathing. We ensured that, prior to initiating the studies, the animals were breathing spontaneously, and arterial pH and blood gas tensions were within normal range as documented under the baseline period (Fig. 2). The positive end expiratory pressure was kept at 5 cmH2O to prevent pulmonary atelectasis. Both vagi were flushed with saline for 15 min (baseline period). Following the baseline measurements, the vagi were bathed in 2% xylocaine. The mean ± SD (median) length of time from the initiation of xylocaine perfusion to the onset of vagal blockade was variable 145 ± 91 (105) s. The experimental design included maintenance of vagal blockade for 30 min or less if respiratory failure ensued. Thereafter, the vagi were flushed and immersed in normal saline for 15 min (washout period). The amount of saline or xylocaine infused onto the vagi was three times the dead space of the perineural cuff. The vagal blockade was performed during the awake state. Diaphragm electromyogram (EMGdia), sleep states, arterial blood pressure (mmHg), heart rate (beats min–1) and rectal temperature (°C) were monitored and recorded continuously throughout the study.

Figure 2. Arterial pH and blood gas tensions, and breathing patterns.

Figure 2

Open in a new tab

Arterial pH (A), blood gas tensions (B and C), Hgb oxygen saturation (Sao2) (D), breathing rate (E) and inspiratory time (F) during baseline, perineural vagal blockade and washout periods. The arterial pH was lower during the vagal blockade compared to the washout and baseline periods. The P ac o2 (torr) increased during the perineural blockade compared to the baseline and washout periods. Furthermore, P ac o2 was lower during the washout period compared to the baseline values. The Pao2 (torr) and the Sao2 (%) were both lower during the vagal blockade compared to the baseline and washout periods. *P < 0.05: perineural blockade vs. baseline and washout. ¶P < 0.05: baseline vs. washout. The breathing rate (breaths min–1) was lower during the vagal blockade compared to the baseline (P = 0.001) and washout (P = 0.001) periods. However, no statistically significant difference in the T i (s) was observed during vagal blockade compared to the baseline or washout periods. *P < 0.05: perineural blockade vs. baseline and washout.

Breathing rate was determined by the number of diaphragmatic contractions per minute. A breathing pause lasting ≥20 s was defined as an apnoea (Eichenwald, 2016). The sleep states were determined using bioelectric (ECoG, EOG and nuchal EMG) signals and behavioural criteria. The lambs were considered awake by the simultaneous occurrence of fast wave, low amplitude ECoG, presence of eye movements and nuchal tone. The behavioural criteria included open eyes and the ability to lift their heads. A pressure transducer (Statham P23 ID; Gould Instruments Division, Cleveland, OH, USA) was used to measure arterial blood pressure and heart rate.

Arterial blood was drawn to analyse pH and blood gas tensions every 2–3 min. Samples used for measurement of blood‐gas tension were corrected for body temperature (IL 1312 blood‐gas manager; Instrumentation Laboratories, Lexington, MA, USA). All bioelectric signals were displayed using an eight‐channel chart recorder (Gould Brush 2800s; Gould Instruments Division) and then digitized and stored on a videocassette using an eight‐channel Neurocorder (DR‐886; Neurodata Instruments, New York, NY, USA). To prevent hypoglycaemia and dehydration, dextrose (10%, w/v) in normal saline was continuously infused i.v. at 90–120 ml kg−1 day−1. To prevent infection, the animals received 2.5 mg kg–1 of gentamicin sulphate (Garamycin injectable; Shering Canada, Pointe‐Claire, PQ, Canada) and 25 mg kg–1 of cefazolin sodium in saline (Ancef; Smith Kline Beecham Pharma. Oakville, ON, USA) intra‐operatively. After the studies were completed, all animals were killed using a lethal dose of pentobarbitone.

Statistical analysis

To test the hypothesis, each animal served as its own control. The effects of saline (baseline), xylocaine blockade and washout (recovery) on arterial pH, blood gas tensions, breathing patterns, heart rate and arterial blood pressure were analysed using ANOVA for repeated measures. The F test values and their respective P values are given for each variable. The repeated measures included the a priori post hoc contrast tests (SPSS, version 15; SPSS Inc., Chicago, IL, USA). All values are reported as the mean ± SD. P < 0.05 was considered statistically significant.

Results

Arterial pH and blood‐gas tensions

During the vagal perineural blockade, significant effects were observed with respect to arterial pH (F 2,12 = 6.03; P = 0.015), P ac o2 (F 2,12 = 17.97; P < 0.001), Pao2 (F 2,12 = 19.0; P < 0.001) and arterial oxygen saturation (Sao2) (F 2,10 = 25.59; P < 0.001) (Fig. 2 A–D).

The arterial pH was significantly lower during vagal blockade (7.39 ± 0.04) compared to baseline (7.44 ± 0.03; P = 0.008) and washout (7.43 ± 0.05; P = 0.02) values (Fig. 2 A). No difference was observed between the baseline and washout periods (P = 0.8). The P ac o2 during the xylocaine blockade was higher (45.6 ± 5 torr) compared to baseline (38.6 ± 5 torr) and washout (33.3 ± 3 torr) periods (P = 0.02 and P = 0.001, respectively). Furthermore, the P ac o2 was lower during the washout period compared to baseline values (P = 0.03) (Fig. 2 B). The Pao2 was lower during the vagal blockade (26.1 ± 8 torr) compared to baseline (56 ± 6 torr; P = < 0.001) and washout (53 ± 14 torr; P < 0.01) periods. The Pao2 was similar between the baseline and washout periods (P = 0.6) (Fig. 2 C). The change in Sao2 mirrored the changes in Pao2 . The Sao2 was lower (47 ± 15%) during the perineural blockade vs. baseline (86 ± 7%; P = 0.001) and washout values (83 ± 11; P = 0.003). No difference was observed in Sao2 between baseline and washout periods (P = 0.6; Fig. 2 D). Arterial base excess remained unchanged; however, bicarbonate (mmol L–1) was lower (P < 0.02) during the washout (23 ± 4) period compared to the perineural vagal blockade (28 ± 1) values. No significant differences were observed between the baseline vs. blockade period (P = 0.19) or baseline vs. washout period (P = 0.08).

Breathing patterns

The effects of perineural vagal blockade on respiratory variables are shown in Fig. 2 E–F and Table 1. The duration of the perineural vagal block was 11.3 ± 8.6 min (mean ± SD). The vagal blockade had a significant effect on breathing rate (F 2,12 = 26.85; P < 0.001) (Fig. 2 E). Breathing rate was lower during the perineural blockade (24 ± 20 beats min–1) compared to the baseline (53 ± 12 beats min–1; P = 0.001) and washout (57 ± 24 beats min–1; P = 0.001) periods. However, no difference in breathing rate was observed between the baseline and washout periods (P = 0.5). A non‐significant increase was observed in the inspiratory time (T i) during vagal blockade compared to the baseline period (P = 0.06); however, no differences were observed between the vagal blockade and washout periods (P = 0.49) (Fig. 2 F).

Table 1.

Effect of perineural vagal blockade on respiratory variables

Variable Mean ± SD Median (range)
Duration of perineural blockade (min) 11.3 ± 8.6 9.0 (3–23)
Lowest breathing rate during block (breaths min–1) 20.4 ± 14.0 22.0 (5–43)
Apnea length (s) 23 ± 4.7 21.0 (20–30)
Time to achieve pre‐block breathing rate after saline flush (min) 16.6 ± 8.9 17.0 (6–34)
Open in a new tab

Apnoeas

Of the seven lambs studied, four exhibited apnoeas during the perineural blockade requiring immediate termination of the blockade using normal saline flushes. Three apnoeas coincided with the onset of sleep, whereas one lamb exhibited apnoea during the awake state. The mean ± SD duration of the longest apnoea was 23 ± 4.7 s. The time to achieve baseline breathing rates following the discontinuation of vagal blockade was 16.6 ± 8.9 min (Table 1).

Sleep and behavioural states

All vagal blockades were initiated during the electroencephalographic and behavioural awake states (Fig. 3). Four of the seven lambs stayed behaviourally awake during the entire duration of vagal blockade, whereas two animals transitioned to non‐rapid eye movement (REM) and one towards REM sleep state. During the washout period, all seven lambs were electroencephalographically and behaviourally awake, including the three who transitioned to REM or non‐REM sleep during the perineural vagal blockade. Figure 3 highlights the changes in breathing patterns and sleep states during baseline, perineural blockade and washout periods.

Figure 3. Cardiorespiratory variables and sleep states.

Figure 3

Open in a new tab

Representative recording of bioelectric signals during baseline (A), perineural vagal nerve blockade (B) and washout (C) periods. The perineural blockade was performed in spontaneously breathing neonatal lambs during the awake state. During vagal blockade, respiratory rate markedly decreased and the sleep state transitioned from indeterminate to non‐REM (B). During the washout period, breathing rate increased to near‐baseline values followed by an intense behavioural and electroencephalographic arousal response (C). [Color figure can be viewed at http://wileyonlinelibrary.com]

Cardiovascular variables

The results for systolic, diastolic, mean blood pressure (mmHg) and heart rate (beats min–1) are given in Fig. 4. The vagal blockade had no overall effect on the systolic blood pressure (F 2,12 = 3.3; P = 0.07). However, an overall significant effect was observed on diastolic (F 2,12 = 7.2; P = 0.009) and mean blood pressures (F 2,12 = 5.8 P = 0.02). The multiple comparisons, however, did not show any significant differences between the various experimental conditions (P > 0.05). The perineural vagal blockade had a significant overall effect on heart rate (F 2,12 = 4.9; P = 0.03); however, a non‐significant decrease in heart rate was observed during the vagal blockade compared to the baseline values (P = 0.056).

Figure 4. Cardiovascular variables.

Figure 4

Open in a new tab

Systolic blood pressure (A), diastolic blood pressure (B), mean blood pressure (C) and heart rate (D) during baseline, vagal blockade and washout periods. No statistically significant differences were observed.

Discussion

The design of the present study clarifies the roles of afferent volume feedback, sleep states, carbon dioxide tension and hypoxaemia in the neonatal control of breathing. We provide evidence that, in previous studies, an inability to establish adequate alveolar ventilation and profound postoperative apnoeas and life‐threatening respiratory failure in vagally denervated animals did not result from a lack of arousal, hypoxaemia or hypocapnia. Rather, a change in sleep state and concomitant respiratory acidosis were the result of absent afferent volume feedback, which appears to be critical for the maintenance of normal breathing patterns and adequate gas exchange during the early newborn period. Furthermore, the data demonstrate that afferent volume feedback takes precedence over arousal and changes in blood gas tensions with respect to maintaining normal breathing rhythmicity during the early neonatal period.

The physiological mechanisms underlying respiratory failure and inadequate pulmonary gas exchange at the organ‐system level in vagally denervated animals have been attributed to an absence of critical volume feedback, pulmonary atelectasis, attenuated expiratory braking and augmented breaths (Lalani et al. 2001; Lalani et al. 2002), as well as a decrease in pulmonary blood flow in partially aerated lungs (Lang et al. 2017). However, the molecular mechanisms underlying the effects of absence of vagal feedback have remained unknown until recently. Nonumura et al. (2017) demonstrated that ablation of mechanically activated Piezo2 ion channels led to respiratory distress at birth and all newborn mice died within 24 h of life. Furthermore, Piezo2−/− pups had a lower respiratory rate and oxygen saturation, and also exhibited gasping and pulmonary atelectasis, compared to wild‐type animals. The respiratory distress and death were not attributable to prenatal lung hypoplasia as a result of Piezo2 ablation (Nonomura et al. 2017). It is important to note that the breathing patterns, lethality and lung histological changes in Piezo2−/– pups were identical to the vagal denervation studies previously reported by our group indicating that volume‐mediated mechanotransduction via vagal afferents is required at birth to establish regular breathing and pulmonary gas exchange (Wong et al. 1998).

Arousal from sleep serves as a vital protective mechanism against hypoxia (Horne et al. 2005) and respiratory failure and plays an important role in the termination of apnoeas (Phillipson & Sullivan, 1978; Dempsey et al. 2010; Guyenet & Abbott, 2013). Furthermore, a number of compensatory or back‐up ventilatory mechanisms may be compromised during sleep (Dempsey et al. 1996) (Thach, 2008). It has been demonstrated that apnoeas are longer and occur more consistently during sleep versus the arousal state during hypocapnoea (Datta et al. 1991). In our previous studies, animals did not establish an eupnoeic breathing pattern and exhibited life‐threatening apnoeas requiring prolonged assisted ventilation following surgery. It is possible that the animals were not fully awake in the immediate postoperative period and/or the anaesthesia‐mediated peripheral chemoreflex loop was suppressed (Stuth et al. 2008). To avoid the confounding effect of sleep states on the establishment of normal breathing patterns and pulmonary gas exchange, we did not initiate perineural vagal blockades in the present study until we ensured that the animals were behaviourally and electrographically awake. During the washout period, resumption of breathing towards baseline (pre‐blockade) patterns occurred regardless of the sleep state. Arousal did not appear to be a pre‐requisite for the resumption of a normal breathing pattern. These observations provide evidence that afferent vagal feedback takes precedence over arousal in reversing respiratory depression during the early neonatal period. The mechanisms for the behavioural change from awake to active or quiet sleep in some of our animals during perineural blockade remain unclear. However, two possibilities exist. First, the nucleus tractus solitarii (NTS) receive projections from vagi in addition to the afferent fibres from the facial and glossopharyngeal cranial nerves. There are extensive anatomical and functional direct connections between NTS and locus coeruleus (LC), (Counts & Mufson, 2011; Lopes et al. 2016; Benarroch, 2017) and indirectly via nucleus paragigantocellularis (Mello‐Carpes & Izquierdo, 2013). The LC is a primary synthetic site for the synthesis of norepinephrine and is one of the profound modulators of arousal. Thus, the change in behavioural state towards active or quiet sleep could result from the absence of vagal input to LC via NTS projections (Counts & Mufson, 2011; Espana & Scammell, 2011; Lopes et al. 2016). Second, hypoxia could also lead to increased time in quiet sleep (Beuchee et al. 2012).

Afferent input from the cardiovascular and respiratory systems travel via the vagus nerve and terminate in the NTS. The vagal afferent input integrated via the NTS contributes to the inspiratory off switch that regulates the respiratory pattern, especially in the younger age group (Dutschmann et al. 2009). Ventilatory control is mediated via the Pao2 and P ac o2 sensitive peripheral and/or central chemoreceptors and the pulmonary‐pontine feedback loops (Molkov et al. 2013). CO2 plays a central role in maintaining breathing rhythmicity and may increase ventilation several‐fold over the baseline (Nattie, 2001; Nattie & Li, 2002; Nattie & Li, 2009) and has been demonstrated to be mediated mainly by neurons in the raphe and retrotrapezoid nucleus and other neurons in the brainstem (Erlichman et al. 2009; Fu et al. 2017). The stimulatory effects of hypercapnia are mediated via the peripheral carotid and central chemoreceptors with the latter comprising the NTS, the retrotrapezoid nucleus (RTN) and the ventral respiratory group, amongst others (Guyenet et al. 2010; Guyenet & Bayliss, 2015). Studies have demonstrated that focal acidosis of the NTS increases ventilation during sleep and wakefulness (Nattie & Li, 2002). Moreover, peripheral chemoreceptor stimulation of the NTS provides excitatory input to the rhythm generating preBotC via the RTN/parafacial respiratory group (Guyenet, 2008). It is well documented that CO2 chemical drive is of critical importance for maintaining breathing rhythmicity in the very early postnatal period (Praud et al., 1997). Vagal denervation blunts the breathing frequency response to hypercapnia (Kashani & Haigh, 1975; Marsland et al. 1975). In the present study, during the vagal blockade, breathing rate decreased despite an increase in P ac o2. Conversely, during the washout periods, P ac o2 was lower than baseline and blockade values. However, breathing rate and breathing patterns returned to baseline values despite markedly low P ac o2 values during the washout period suggesting that, in the neonatal lamb, afferent input probably at the NTS level is a critical determinant of breathing rhythmicity.

Newborn term mammals exhibit a biphasic ventilatory response to hypoxia: 15% FIO2 (Cross & Warner, 1951; Rigatto et al. 1975; Praud et al. 1995; Waters & Gozal, 2003). The initial increase in minute ventilation results from an increase in both respiratory rate and tidal volume. However, this increase is not maintained as a result of a decrease in tidal volume despite a sustained increase in respiratory rate (Cross & Warner, 1951). Praud et al. (1995) demonstrated an increase in both breathing frequency and tidal volume leading to an increase in minute ventilation in neonatal lambs over a 15 min hypoxic exposure during the first 72 h of life. A downward (biphasic) trend was observed compared to the initial robust increase in minute ventilation during sustained hypoxia; however, breathing frequency, tidal volume and minute ventilation remained significantly higher compared to the baseline normoxic values. In another study, after a very large peak increase in minute ventilation, these values decreased to ∼90% and 70% above the baseline values at 9 and 15 min of hypoxia, respectively (Delacourt et al. 1995). In the present study, during the perineural block, no initial increase in respiratory rate was observed despite hypoxaemia, suggesting that the vagal afferent feedback loop is necessary for respiratory stimulation during hypoxaemia in very young animals. Praud et al. (1992) and Delacourt et al. (1995) also demonstrated a blunted breathing frequency increase in response to hypoxia in vagally denervated 2‐week‐old lambs. Under normoxic conditions, ventilation increases as PCO2 rises, further increases in ventilation are seen if PO2 is decreased (West, 2012). Such an interaction of hypercapnic and hypoxic ventilatory responses probably results from components of central and peripheral chemical drive (West, 2012). In the present study, despite markedly low Pao2 levels and elevated P ac o2 levels, we did not observe either a biphasic response or any increase in breathing rate during the neural blockade, highlighting the overriding role of vagal feedback on neonatal respiratory control.

The present study has some limitations. As a result of the profound life‐threatening respiratory depression observed in previous studies (Lalani et al. 2002), the animals needed to remain endotracheally intubated during the perineural blockade. Reversible blockade was warranted to test the proposed hypothesis and the use of each animal as its own control decreased the number of animals used and avoided the confounding effects of inter‐subject variability on a number of cardiorespiratory and behavioural responses. Reversible perineural blockade is not feasible at the intrathoracic level as a result of the anatomical limitations of the left vagal nerve. Similarly, rapid onset of hypoxaemia, apneic episodes and slow recovery from respiratory depression precluded determination of Pco2 and Pao2 thresholds for the onset or termination of apnoeas. Furthermore, vagal blockade might impact CO2 chemoreception integration in the solitary complex comprising the nucleus tractus solatarii and dorsal motor nucleus (Dean & Putnam, 2010). Every effort was made to ensure that the anaesthetic was off and ventilatory rates were decreased to zero at the same time as maintaining a positive end expiratory pressure of 5 cmH2O. The studies were not started until the lambs were spontaneously breathing. However, it is possible that there were mild remaining effects of anaesthesia.

In summary, over two decades, studies have systematically investigated the role of afferent vagal feedback on cardiorespiratory control during transition from fetal to neonatal and during the early neonatal period. The present study advances the field in terms of several aspects: we demonstrate that a lack of afferent feedback overrides the stimulatory effects of carbon dioxide on neonatal breathing and also that the profound apnoeas and life‐threatening respiratory failure observed in vagally denervated animals were not a result of hypoxia or a lack of arousal. Rather, a change in sleep state and attendant respiratory depression result from a lack of afferent volume feedback, which takes precedence over other respiratory stimuli and appears to be vital for the maintenance of normal breathing patterns and adequate gas exchange during the early postnatal period.

Additional information

Competing interests

The authors declare that they have no competing interests.

Author contributions

The studies were performed in Dr Shabih U. Hasan's laboratory at the University of Calgary, Alberta, Canada. KJL, TI, AR and SUH were responsible for the conception and design of the work. KJL, JMS, TI, AR and SUH were responsible for acquisition, analysis and interpretation of the data. KJL, JMS, TI, AR and SUH were responsible for drafting, review and critical revision of the work. All authors approved the manuscript submitted for publication, agree to be accountable for all aspects of the work, including accuracy and integrity, and qualify for authorship in accordance with the requirements of The Journal of Physiology.

Funding

This work was supported by the Canadian Institutes of Health Research.

Acknowledgements

We thank Dr Magdy Younes for his input in the experimental design and Ms Shylin Miljan for providing editorial assistance.

Biography

Kathleen J Lumb received her undergraduate degree in Zoology from the University of Calgary, Alberta, Canada, sparking her interest in physiology and leading her to pursue a Master's degree in Medical Science. The research reported in the present study was conducted as part of her graduate degree. She is very interested in the basic physiological mechanisms of cardiorespiratory control during the perinatal period. Using a reversible perineural blockade model, her team has demonstrated the significance of vagal innervation on cardiorespiratory control during early neonatal life. There is a great deal of further research needed to explore the complexities of the initiation and establishment of adequate ventilation and pulmonary gas exchange in neonates and this will hopefully further clarify our understanding of tragic illness such as sudden unexpected death in infancy.

graphic file with name TJP-596-6009-g001.gif

Edited by: Laura Bennet & Dino Giussani

References

  1. Benarroch EE (2017). Locus coeruleus. Cell Tissue Res, 10.1007/s00441-017-2649-1. [Epub ahead of print] [DOI] [PubMed] [Google Scholar]
  2. Beuchee A, Hernandez AI, Duvareille C, Daniel D, Samson N, Pladys P & Praud JP (2012). Influence of hypoxia and hypercapnia on sleep state‐dependent heart rate variability behavior in newborn lambs. Sleep 35, 1541–1549. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Coste B, Mathur J, Schmidt M, Earley TJ, Ranade S, Petrus MJ, Dubin AE & Patapoutian A (2010). Piezo1 and Piezo2 are essential components of distinct mechanically activated cation channels. Science 330, 55–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Counts SE & Mufson EJ (2011). Locus voeruleus In The Human Nervous System, 3rd edn, ed. Mai JK. & Paxinos G, pp. 425–438. Academic Press, Elsevier, Amsterdam, The Netherlands. [Google Scholar]
  5. Cross KW & Warner P (1951). The effect of inhalation of high and low oxygen concentrations on the respiration of the newborn infant. J Physiol 114, 283–295. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Datta AK, Shea SA, Horner RL & Guz A (1991). The influence of induced hypocapnia and sleep on the endogenous respiratory rhythm in humans. J Physiol 440, 17–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Dean JB & Putnam RW (2010). The caudal solitary complex is a site of central CO(2) chemoreception and integration of multiple systems that regulate expired CO(2). Respir Physiol Neurobiol 173, 274–287. [DOI] [PubMed] [Google Scholar]
  8. Delacourt C, Canet E, Praud JP & Bureau MA (1995). Influence of vagal afferents on diphasic ventilatory response to hypoxia in newborn lambs. Respir Physiol 99, 29–39. [DOI] [PubMed] [Google Scholar]
  9. Delmas P, Hao J & Rodat‐Despoix L (2011). Molecular mechanisms of mechanotransduction in mammalian sensory neurons. Nat Rev Neurosci 12, 139–153. [DOI] [PubMed] [Google Scholar]
  10. Dempsey JA, Smith CA, Harms CA, Chow C & Saupe KW (1996). Sleep‐induced breathing instability. University of Wisconsin‐Madison Sleep and respiration Research Group. Sleep 19, 236–247. [PubMed] [Google Scholar]
  11. Dempsey JA, Veasey SC, Morgan BJ & O'Donnell CP (2010). Pathophysiology of sleep apnea. Physiol Rev 90, 47–112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Dutschmann M, Morschel M, Rybak IA & Dick TE (2009). Learning to breathe: control of the inspiratory‐expiratory phase transition shifts from sensory‐ to central‐dominated during postnatal development in rats. J Physiol 587, 4931–4948. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Eichenwald EC, Committee on F & Newborn AAoP (2016). Apnea of prematurity. Pediatrics 137, e20153757. 10.1542/peds.2015-3757. [DOI] [Google Scholar]
  14. Erlichman JS, Boyer AC, Reagan P, Putnam RW, Ritucci NA & Leiter JC (2009). Chemosensory responses to CO2 in multiple brain stem nuclei determined using a voltage‐sensitive dye in brain slices from rats. J Neurophysiol 102, 1577–1590. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Espana RA & Scammell TE (2011). Sleep neurobiology from a clinical perspective. Sleep 34, 845–858. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Fedorko L, Kelly EN & England SJ (1988). Importance of vagal afferents in determining ventilation in newborn rats. J Appl Physiol (1985) 65, 1033–1039. [DOI] [PubMed] [Google Scholar]
  17. Fu C, Xue J, Wang R, Chen J, Ma L, Liu Y, Wang X, Guo F, Zhang Y, Zhang X & Wang S (2017). Chemosensitive Phox2b‐expressing neurons are crucial for hypercapnic ventilatory response in the nucleus tractus solitarius. J Physiol 595, 4973–4989. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Gauda EB, Carroll JL & Donnelly DF (2009). Developmental maturation of chemosensitivity to hypoxia of peripheral arterial chemoreceptors – invited article. Adv Exp Med Biol 648, 243–255. [DOI] [PubMed] [Google Scholar]
  19. Goridis C (2017). Physiology: mechanosensor of lung inflation identified. Nature 541, 165–166. [DOI] [PubMed] [Google Scholar]
  20. Greer JJ (2012). Control of breathing activity in the fetus and newborn. Compr Physiol 2, 1873–1888. [DOI] [PubMed] [Google Scholar]
  21. Greer JJ, Funk GD & Ballanyi K (2006). Preparing for the first breath: prenatal maturation of respiratory neural control. J Physiol 570, 437–444. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Grigg‐Damberger MM (2016). The visual scoring of sleep in infants 0 to 2 months of age. J Clin Sleep Med 12, 429–445. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Guyenet PG (2008). The 2008 Carl Ludwig Lecture: retrotrapezoid nucleus, CO2 homeostasis, and breathing automaticity. J Appl Physiol 105, 404–416. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Guyenet PG & Abbott SB (2013). Chemoreception and asphyxia‐induced arousal. Respir Physiol Neurobiol 188, 333–343. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Guyenet PG & Bayliss DA (2015). Neural Control of Breathing and CO2 Homeostasis. Neuron 87, 946–961. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Guyenet PG, Stornetta RL, Abbott SB, Depuy SD, Fortuna MG & Kanbar R (2010). Central CO2 chemoreception and integrated neural mechanisms of cardiovascular and respiratory control. J Appl Physiol (1985) 108, 995–1002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Harris MB & Milsom WK (2001). Vagal feedback is essential for breathing in unanesthetized ground squirrels. Respir Physiol 125, 199–212. [DOI] [PubMed] [Google Scholar]
  28. Hasan SU, Lalani S & Remmers JE (2000). Significance of vagal innervation in perinatal breathing and gas exchange. Respir Physiol 119, 133–141. [DOI] [PubMed] [Google Scholar]
  29. Hooper SB, Polglase GR & Roehr CC (2015a). Cardiopulmonary changes with aeration of the newborn lung. Paediatr Respir Rev 16, 147–150. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Hooper SB, Te Pas AB, Lang J, van Vonderen JJ, Roehr CC, Kluckow M, Gill AW, Wallace EM & Polglase GR (2015b). Cardiovascular transition at birth: a physiological sequence. Pediatr Res 77, 608–614. [DOI] [PubMed] [Google Scholar]
  31. Horne RS, Parslow PM & Harding R (2005). Postnatal development of ventilatory and arousal responses to hypoxia in human infants. Respir Physiol Neurobiol 149, 257–271. [DOI] [PubMed] [Google Scholar]
  32. Kashani M & Haigh AL (1975). The effects of vagotomy on ventilation and blood gas composition in dog, sheep and rabbit. Q J Exp Physiol Cogn Med Sci 60, 285–298. [DOI] [PubMed] [Google Scholar]
  33. Lalani S, Remmers JE, Green FH, Bukhari A, Ford GT & Hasan SU (2001). Effects of vagal denervation on cardiorespiratory and behavioral responses in the newborn lamb. J Appl Physiol (1985) 91, 2301–2313. [DOI] [PubMed] [Google Scholar]
  34. Lalani S, Remmers JE, MacKinnon Y, Ford GT & Hasan SU (2002). Hypoxemia and low Crs in vagally denervated lambs result from reduced lung volume and not pulmonary edema. J Appl Physiol (1985) 93, 601–610. [DOI] [PubMed] [Google Scholar]
  35. Lang JA, Pearson JT, Binder‐Heschl C, Wallace MJ, Siew ML, Kitchen MJ, Te Pas AB, Lewis RA, Polglase GR, Shirai M & Hooper SB (2017). Vagal denervation inhibits the increase in pulmonary blood flow during partial lung aeration at birth. J Physiol 595, 1593–1606. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Lopes LT, Patrone LG, Li KY, Imber AN, Graham CD, Gargaglioni LH & Putnam RW (2016). Anatomical and functional connections between the locus coeruleus and the nucleus tractus solitarius in neonatal rats. Neuroscience 324, 446–468. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Marsland DW, Callahan BJ & Shannon DC (1975). The afferent vagus and regulation of breathing in response to inhaled CO2 in awake newborn lambs. Biol Neonate 27, 102–107. [DOI] [PubMed] [Google Scholar]
  38. Mello‐Carpes PB & Izquierdo I (2013). The nucleus of the solitary tract → nucleus paragigantocellularis → locus coeruleus → CA1 region of dorsal hippocampus pathway is important for consolidation of object recognition memory. Neurobiol Learn Mem 100, 56–63. [DOI] [PubMed] [Google Scholar]
  39. Molkov YI, Bacak BJ, Dick TE & Rybak IA (2013). Control of breathing by interacting pontine and pulmonary feedback loops. Front Neural Circuits 7, 16. eCollection 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Nattie EE (2001). Central chemosensitivity, sleep, and wakefulness. Respir Physiol 129, 257–268. [DOI] [PubMed] [Google Scholar]
  41. Nattie EE & Li A (2002). CO2 dialysis in nucleus tractus solitarius region of rat increases ventilation in sleep and wakefulness. J Appl Physiol 92, 2119–2130. [DOI] [PubMed] [Google Scholar]
  42. Nattie EE & Li A (2009). Central chemoreception is a complex system function that involves multiple brain stem sites. J Appl Physiol 106, 1464–1466. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Nonomura K, Woo SH, Chang RB, Gillich A, Qiu Z, Francisco AG, Ranade SS, Liberles SD & Patapoutian A (2017). Piezo2 senses airway stretch and mediates lung inflation‐induced apnoea. Nature 541, 176–181. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Phillipson EA & Sullivan CE (1978). Arousal: the forgotten response to respiratory stimuli. Am Rev Respir Dis 118, 807–809. [DOI] [PubMed] [Google Scholar]
  45. Praud JP, Canet E & Bureau MA (1992). Chemoreceptor and vagal influences on thyroarytenoid muscle activity in awake lambs during hypoxia. J Appl Physiol (1985) 72, 962–969. [DOI] [PubMed] [Google Scholar]
  46. Praud JP, Diaz V, Kianicka I, Chevalier JY, Canet E & Thisdale Y (1997). Abolition of breathing rhythmicity in lambs by CO2 unloading in the first hours of life. Respir Physiol 110, 1–8. [DOI] [PubMed] [Google Scholar]
  47. Praud JP, Kianicka I, Leroux JF & Dalle D (1995). Laryngeal response to hypoxia in awake lambs during the first postnatal days. Pediatr Res 37, 482–488. [DOI] [PubMed] [Google Scholar]
  48. Prechtl HF (1967). Polygraphic studies in newborn infants. Electroencephalogr Clin Neurophysiol 23, 494. [PubMed] [Google Scholar]
  49. Prechtl HF (1974). The behavioural states of the newborn infant (a review). Brain Res 76, 185–212. [DOI] [PubMed] [Google Scholar]
  50. Rigatto H, Brady JP & de la Torre Verduzco R (1975). Chemoreceptor reflexes in preterm infants: I. The effect of gestational and postnatal age on the ventilatory response to inhalation of 100% and 15% oxygen. Pediatrics 55, 604–613. [PubMed] [Google Scholar]
  51. Samson N, Fortin‐Pellerin E & Praud JP (2018). The contribution of ovine models to perinatal respiratory physiology. Front Biosci (Landmark Ed) 23, 1195–1219. [DOI] [PubMed] [Google Scholar]
  52. Stuth EA, Stucke AG, Brandes IF & Zuperku EJ (2008). Anesthetic effects on synaptic transmission and gain control in respiratory control. Respir Physiol Neurobiol 164, 151–159. [DOI] [PubMed] [Google Scholar]
  53. Thach B (2008). Tragic and sudden death – potential and proven mechanisms causing sudden infant death syndrome. Embo Rep 9, 114–118. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Waters KA & Gozal D (2003). Responses to hypoxia during early development. Respir Physiol Neurobiol 136, 115–129. [DOI] [PubMed] [Google Scholar]
  55. West JB (2012). Respiratory Physiology: The Essentials. Wolters Kluwer Health/Lippincott Williams & Wilkins, Philadelphia, PA. [Google Scholar]
  56. Wong KA, Bano A, Rigaux A, Wang B, Bharadwaj B, Schurch S, Green F, Remmers JE & Hasan SU (1998). Pulmonary vagal innervation is required to establish adequate alveolar ventilation in the newborn lamb. J Appl Physiol 85, 849–859. [DOI] [PubMed] [Google Scholar]
  57. Xie A, Skatrud JB & Dempsey JA (2001). Effect of hypoxia on the hypopnoeic and apnoeic threshold for CO(2) in sleeping humans. J Physiol 535, 269–278. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES