INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Impaired diaphragm function can cause respiratory distress in the neonatal period (1). Accurate assessment of diaphragm function in the neonate could aid diagnosis of respiratory distress, evaluation of therapeutic interventions, and identification of infants ready to wean from mechanical ventilation. Real-time ultrasonography (2), respiratory inductive plethysmography (3), and electromyographic (EMG) analysis (4) have been used as indicators of diaphragm function, but provide only a qualitative measure and generate no information about diaphragm force production. Maximal inspiratory pressures during crying have also been used (5), but the test is effort dependent and, as such, the results can be submaximal and variable. By stimulating the phrenic nerves in the neck and recording the transdiaphragmatic pressure (Pdi) generated, diaphragm function can be directly assessed. Transcutaneous electrical stimulation has been used in neonates to measure phrenic nerve latency (6), but the technique is uncomfortable and technically difficult to perform. Many of the problems associated with electrical stimulation can be overcome by magnetic stimulation (MS) which is both easily applied and relatively painless. MS is performed in adults using a coil placed over the cervical spine to supramaximally stimulate the phrenic nerve roots bilaterally (CMS) (7) or on the anterolateral aspect of the neck, allowing both unilateral and bilateral stimulation to be performed (AMS) (8). The objective of this study was to assess the feasibility of CMS and AMS of the phrenic nerves in neonates.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Twenty-five infants without respiratory distress or requirement for additional oxygen supplementation were recruited. The study was approved by King's College Hospital research ethics committee. Infants were studied once informed written consent had been obtained from the parents.

Equipment

Force production was assessed as Pdi recorded from two latex balloon catheters placed in the lower third of the esophagus (esophageal pressure [Pes]) and stomach (gastric pressure [Pgas]) (P.K. Morgan, Gillingham, Kent, UK). The balloon catheters were passed by trained staff without difficulty or complication. The length of the esophageal balloon was 2.5 cm and the gastric balloon 4.0 cm. Correct positioning of the esophageal balloon catheter was checked by comparing Pes to mouth pressure (Pm) during an occluded breath. Pm was measured from the side-port of a pneumotachograph attached to a face mask (total dead space 4.5 ml) placed over the infant's nose and mouth. Agreement of Pes and Pm within 90 to 110% was deemed acceptable (9). Pressures were measured by differential pressure transducers (MP45-1; Validyne, Northridge, CA) and carrier amplifiers (CD 280; Validyne). The pressure signals were analyzed and displayed in real time by computer (Quadra 650; Apple Computer Inc., Cupertino, CA) running Labview software (Ver. 2.2; National Instruments, Austin, TX) with analogue-to-digital sampling at 100 Hz (12 bit NB-MIO-16; National Instruments, Austin, TX). Transdiaphragmatic pressure was obtained on line by subtraction of Pes from Pgas. A Magstim 200 (high-power) magnetic stimulator (Magstim Co., Whitland, Dyfed, UK) was used to power the magnetic coils. Two Magstim 90-mm (SPC 8632) double circular coils were used, each having a peak field of 2.5 Tesla at the face of the coil and 1.4 Tesla 25 mm below the coil. A diagram of the experimental apparatus is given in Figure 1.

View larger version (19K): [in this window] [in a new window]   Figure 1.   Diagrammatic representation of the apparatus used with the magnetic coil positioned for unilateral AMS of the phrenic nerve.

Functional residual capacity (FRC) was measured using a helium (He) gas dilution system (Equilibrated Bio Systems, Melville, NY) specifically designed for infants (total volume 95 ml). The system consisted of a He analyzer, a rebreathing bag, and a three-way valve to which the infant was connected via a face mask. The initial and equilibration He concentrations were used to calculate FRC, which was corrected for oxygen consumption, assumed to be 7 ml/kg/min (10), and to BTPS conditions. FRC was estimated twice in each infant and expressed as the mean of the paired measurements and related to body weight. The median FRC of term infants is 30 ml/kg (range 24 to 36) (11). Oxygen saturation was monitored continuously during each study using a pulse oximeter (Biox 3740; Ohmeda, Louisville, CO).

Procedure

The infants were studied in the supine position at least 1 h after a feed and during periods of sleep. Sleep state was not formally assessed, but infants were only studied during periods of quiet sleep when rapid eye movements and gross body movements were absent. No sedation was used during the study. CMS was performed with the stimulating coil placed in the midline over the cervical spine to bilaterally stimulate the phrenic nerve roots, C3-C5. Subjects breathed through a tightly fitting face mask which was occluded during the stimulation. AMS was performed with the coil placed over the phrenic nerve on the anterior aspect of the neck at the posterior border of the sternomastoid muscle at the level of the cricoid cartilage (Figure 1). Unilateral and bilateral stimulation of the phrenic nerves was performed. To produce bilateral stimulation two coils were discharged simultaneously. With both techniques, the phrenic nerves were stimulated at end expiration with 10 stimulations at maximal stimulator output being made on each side. During the study, end-expiratory Pes was used as an indicator of lung volume relative to FRC and MS was only performed when Pes was at its resting baseline FRC value. Only twitches occurring at end expiration, as indicated by the Pdi, Pes, and Pgas waveforms, were subsequently analyzed. To avoid twitch potentiation of the diaphragm there was a period of 10 min quiet breathing before stimulation was commenced and a 20-s interval between each twitch (12). Preliminary data were obtained to achieve optimal coil positioning. The supramaximality of stimuli was assessed in five infants by examining the mean diaphragm twitch Pdi (TwPdi) at varying stimulator outputs imposed in random order over the range 80, 85, 90, 95, and 100%. Five to 10 stimulations were obtained at each power setting.

Airway occlusion was not routinely performed during AMS. To assess the effect of airway occlusion during AMS, which allows a quasi-isometric contraction of the diaphragm, three mechanically ventilated patients were studied. Intubated patients were selected as the endotracheal tube maintains upper airway patency preventing glottic closure, thus allowing the influence of airway occlusion on the transmission of twitch pressure to be assessed. Airway occlusion during phrenic nerve stimulation (PNS) was achieved using a low dead space occlusion valve added to the ventilator circuit directly above the endotracheal tube. TwPdi for occluded and unoccluded breaths were compared in each patient.

Statistical Analysis

Data for age, weight, and FRC are given as median and range and TwPdi expressed as mean ± SD. The mean intrapatient coefficient of variation (CV) for left and right unilateral stimulation was calculated in five infants and based in each on at least five stimulations at 100% power output.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Diaphragm twitch responses were obtained in all subjects. MS was generally well tolerated using both techniques with the patients continuing to sleep throughout. The mean (range) ratio of Pm to Pes was 0.99 (0.91 to 1.06). Their median FRC was 25.2 ml/kg (range 22.5 to 35.1). Seven infants (1 female, 6 male) with a median gestational age (GA) of 35 wk (range 33 to 40) received CMS at a median postconceptional age (PCA) of 37 wk (range 35 to 41) and weight of 2,470 g (range 1,830 to 2,760). Their mean (SD) TwPdi was 2.5 (0.8) cm H₂O. The responses were not supramaximal as reducing stimulator output below 100% led to reductions in TwPdi. Pes was commonly positive, probably indicating activation of extradiaphragmatic muscles either with or without chest wall distortion (Figure 2). Eighteen infants (7 female, 11 male) with a median GA of 37 wk (range 25 to 41) received AMS at a median PCA of 39 wk (range 33 to 44) and weight of 3,151 g (range 2,260 to 4,670). Mean (SD) TwPdi was 4.5 (1.3) cm H₂O on the left (n = 12), 4.1 (0.9) cm H₂O on the right (n = 12), and 8.7 (3.9) cm H₂O for left and right together (n = 12) (Figures 3 and 4). The mean (range) intrapatient CV of TwPdi was assessed in five patients (GA 37 wk, range 28 to 40; PCA 37, range 35 to 44; weight 2,700, range 1,350 to 4,568) and was 17% (4 to 30%) for left and 21% (14 to 29%) for right unilateral stimulation. The supramaximality of the stimulus was tested in five infants, median GA 34 wk (range 25 to 37), PCA 37 wk (range 35 to 42), and weight 2,530 g (range 1,098 to 3,900). Supramaximal stimulation was achieved with both right and left unilateral AMS as indicated by plateauing of TwPdi with increasing stimulator output (Figure 5). In the three ventilated patients in which occluded and unoccluded twitches were compared, the difference in each subject for unilateral stimulation was 1, 0.3, and 0.6 cm H₂O, respectively, and the group mean TwPdi was 4.6 cm H₂O occluded and 4.5 cm H₂O unoccluded.

View larger version (14K): [in this window] [in a new window]   Figure 2.   Representative diaphragm pressure response after CMS. Pes and Pgas are shown in the upper trace, and the resulting Pdi has been transposed and shown in the lower trace. Note the positive Pes wave, probably indicating activation of extradiaphragmatic muscles.

View larger version (15K): [in this window] [in a new window]   Figure 3.   Representative diaphragm pressure response after unilateral AMS. Pes and Pgas shown in the upper trace, and the resulting Pdi has been transposed and shown in the lower trace.

View larger version (14K): [in this window] [in a new window]   Figure 4.   Representative diaphragm pressure response after bilateral AMS. Pes and Pgas are shown in the upper trace, and the resulting Pdi has been transposed and shown in the lower trace.

View larger version (14K): [in this window] [in a new window]   Figure 5.   Mean (± SE) diaphragm pressure responses (TwPdi) versus stimulator output for left and right unilateral AMS in five subjects. Supramaximal stimulation was achieved as indicated by the plateauing of TwPdi with increasing stimulator output.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present study demonstrates that supramaximal stimulation of the phrenic nerves can be achieved in neonates using an anterior approach. We used a 90-mm double circular coil to allow optimal positioning on the neck and a magnetic field strength, depth of penetration, and induced current in the tissues of sufficient magnitude to produce supramaximal stimulation of the phrenic nerves using an anterior approach.

CMS was not supramaximal, as reducing stimulator output led to marked reductions in TwPdi and may not have stimulated the diaphragm alone, as indicated by a positive esophageal pressure wave after stimulation (Figure 2). CMS has been shown to cause significant extradiaphragmatic muscle depolarization and rib cage expansion in adults (7). Unlike CMS, AMS was supramaximal as indicated by the plateau in TwPdi at 95% of stimulator output (Figure 5) and the shape and timing of the pressure waveforms were similar to the results obtained in adults. The specificity of the stimulus, however, requires further investigation in infants and children.

The pressure developed by the diaphragm is dependent on muscle length and geometry (13). As hyperinflation reduces diaphragm contractility (14), the lung volume was assessed by helium dilution in each infant and was demonstrated to be within the normal range (24 to 36 ml/kg) (11). To reduce variability, PNS was performed at end expiration with the infant supine to minimize alterations in body position and therefore rib cage configuration.

Because of the difficulty of performing airway occlusion during PNS in spontaneously breathing infants with a rapid respiratory rate, such maneuvers were not routinely performed during AMS. No difference was found between the occluded and unoccluded TwPdi in the three patients studied, suggesting that any pressure loss due to air flow and therefore additional volume change was negligible.

Bilateral TwPdi was only marginally greater than the sum of TwPdi for left and right unilateral AMS, unlike in the adult when it is usually 25 to 30% greater (15). During unilateral PNS, the stimulated hemidiaphragm contracts and descends while the contralateral hemidiaphragm ascends, the resulting TwPdi being dependent upon the compliance of the unstimulated hemidiaphragm and of the chest wall (15). The diaphragm of the newborn infant is relatively flat with a large angle of insertion on the rib cage and a small area of apposition (16). This, in association with a compliant chest wall, leads to chest wall distortion during respiratory efforts. Bilateral hemidiaphragm contraction may cause proportionately more chest wall distortion than unilateral hemidiaphragm contraction although this difference requires further investigation. As in adults, TwPdi was higher on the left than on the right (17), possibly owing to differences in the compliance of the abdominal compartment between left and right. The mean within-occasion variability of AMS in this study was 17% for right and 21% for left unilateral stimulation. In adults, the variability has been reported as approximately 8% for unilateral MS and 10% for unilateral electrical stimulation. There are no data available for comparison in neonates with electrical stimulation.

MS is relatively painless and as such is an ideal technique to assess diaphragm function in the neonate. Painful stimuli would arouse the patient and possibly prevent reproducible measurements of TwPdi because of twitch potentiation from crying. As the stimulating currents are produced in situ their intensities can be very low, possibly explaining the absence of pain when using MS (18). MS is able to stimulate underlying nervous tissue without contact with, or pressure on, the overlying skin and could therefore be useful when contact is painful or access is limited, for example, due to vascular catheters. Furthermore, the stimulating coils are easy to position and compared with electrical stimulation require less precise positioning. The stimulus does not irritate the skin and it is nonthermal, the heat generated in the nervous tissue being negligible [approximately 300 times less than international safety standards (18)]. Neural tissue can be easily damaged by any electrical stimulus, but the short stimuli at low frequencies used in MS are unlikely to cause lesions in nervous structures (19). To date no side effects of single, repeated MS, whether used transcortically or peripherally have been reported (18, 20). In conclusion, these data suggest that anterior MS of the phrenic nerves is a useful technique to assess diaphragm function in neonates.