A comparison of different modes of noninvasive ventilatory support: effects on ventilation and inspiratory muscle effort

Transcription

1 .-!naesthesia, 1994, Volume 49, pages A comparison of different modes of noninvasive ventilatory support: effects on ventilation and inspiratory muscle effort M. W. ELLIOTT, R. AQUILINA, M. GREEN, J. MOXHAM AND A. K. SIMONDS Summary The aims of noninvasive ventilation include the correction of hypoventilation and unloading of inspiratory muscles. Volume cycled flow generators, bi-level positive airway pressure and continuous positive airway pressure techniques have all been used with face and nasal masks. We have compared these modes of ventilatory support, administered by a nasal mask in stable, awake outpatients with chronic obstructive pulmonary disease or neuromusculo-skeletal disease in respect of their effects on ventilation, inspiratory muscle effort and oxygen saturation. There were no clinically significant differences between the volume cycledflow generator and bi-level positive airway pressure methods; compared to spontaneous ventilation, oxygen saturation increased and inspiratory muscle effort decreased. Tidal volume increased and respiratory rate reduced, the largest changes occurring with bilevel positive airway pressure. Only the volume cycledflow generator increased minute ventilation significantly. Ventilation and inspiratory muscle effort were unaffected by continuous positive airway pressure but oxygen saturation was lower than during spontaneous ventilation. In awake, stable outpatients acclimatised to nasal ventilation there were no clinically significant differences between volume cycledflow generator and bi-level positive airway pressure techniques, but continuous positive airway pressure was less effective. Key words Mechanical ventilation; intermittent positive pressure, continuous positive pressure, obstructive lung disease. Chronic obstructive airways disease; mechanical ventilation, positive pressure ventilation. Nasal positive pressure ventilation (NPPV) is an effective treatment for chronic respiratory failure due to neuromuscular disease [1,2], skeletal deformity [3] and for some patients with intrinsic lung disease [4,5]. It has also been used as an alternative to [6], or for weaning from [7,8], tracheal intubation and mechanical ventilation for acute on chronic respiratory failure. Most studies have reported experience with volume cycled (constant) flow generators (VCFG) but some have used the technique of bi-level positive airway pressure (BiPAP) (Respironics Inc, Murrysville, Pennsylvania, USA) [9]. This is achieved using a (variable) flow generator which permits independent control of inspiratory and expiratory pressures and provides pressure support ventilation. The machine is small and easily portable. It can augment spontaneous breaths by virtue of a sensitive trigger and increases flow to compensate for leaks from around the mask and through the open mouth. It therefore has a number of potential advantages over available VCFGs. However, pressure limited devices may be inadequate for controlling hypoventilation, particulary when the impedance to inflation is high, and the current model of the BiPAP can only generate pressures up to 20 cmh,o. The application of an expiratory positive airway pressure (EPAP) may be beneficial in patients with atelectasis or restrictive lung disease. EPAP may recruit collapsed alveoli and increase functional residual capacity (FRC), and reduce the work of breathing by improving pulmonary compliance [lo]. These features may also be helpful in patients with neuromusculo~skeletal disorders in whom basal atelectasis may be prominent. The addition of positive pressure during expiration in patients with obstructive pulmonary disease is controversial, but theoretically it may reduce hyperinflation by holding the airways open longer during expiration and allowing greater gas emptying. In addition some patients M.W. Elliott, MD, MRCP, Senior Registrar, R. Aquilina, BSc, MB, BS, FRCA, Research Fellow, M. Green, MD, FRCP, Consultant Physician, J. Moxham*, MD, FRCP, Professor of Thoracic Medicine, A.K. Simonds, MD, MRCP, Consultant Physician, Departments of Thoracic Medicine, The Royal Brompton National Heart and Lung Hospitals (Chelsea), Sydney Street, London SW3 6NP and *King s College Hospital and Medical School, Bessemer Road, London SE5 9RS, UK. Correspondence should be addressed to Dr M.W. Elliott, St. James s University Hospital, Leeds LS9 7TF. Accepted 22 September /94/ The Association of Anaesthetists of Gt Britain and Ireland 279

2 280 M. W. Elliott et al. have intrinsic positive end-expiratory pressure (PEEP,). This decreases the effective trigger sensitivity because it must be overcome before pressure change and flow occur and can be sensed at the nose [Ill. Unlike most conventional ventilators used in hospital, those for domiciliary use do not have the facility for setting the trigger above atmospheric pressure. Because the BiPAP triggers on changes in flow, matching EPAP to PEEPi should increase effective trigger sensitivity. Continuous positive airway pressure (CPAP) has been shown to reduce the work of breathing [12], may obviate the need for tracheal intubation and mechanical ventilation, and has been used to facilitate the weaning of patients from mechanical ventilation [13]. A major advantage is that CPAP flow generators are substantially cheaper than both the BiPAP and currently available VCFGs. The aims of assisted ventilation include improved oxygenation, the control of hypoventilation and unloading of respiratory muscles. We have, therefore, compared oxygenation, ventilation and inspiratory muscle effort during spontaneous ventilation (SV) with that during ventilatory assistance with a VCFG, BiPAP and CPAP in awake patients. Patients and methods Eleven patients with chronic ventilatory failure were studied. The study was approved by the Ethics Committee of the Royal Brompton National Heart and Lung Hospitals and all patients gave informed consent. All were receiving noninvasive ventilation using VCFGs (nine Bromptonpac) (Pneupac Ltd, Crescent Road, Luton, UK), one Lifecare PLV (Medic-Aid Ltd, Pagham, West Sussex, UK) and one Monnal D (Deva Medical Electronics, Runcorn, Cheshire, UK). The patients were all acclimatised to the BiPAP before the study. Respironics nasal masks were used and patients were encourged to keep their mouths firmly shut throughout the study period. Medication could not be standardised but each patient study took place over a 2 to 3 h period in one day. Oesophageal (Poes) and gastric (Pgas) pressures were measured using balloon tipped catheters 100 cm in length (PK Morgan, Rainham, Kent, UK) positioned in the standard manner [14]. All recordings were made with the patient semisupine on a bed with the same position maintained throughout the study period. In addition mask pressure (Pmask) was measured with a separate catheter connected to an integral port. The catheters were connected to Validyne MP45-1 differential pressure transducers (range f 250 cmh,o; Validyne Corp, Northridge, CA, USA), calibrated before each study and referenced to atmospheric pressure. PEEP, was measured during a period of spontaneous breathing on a mouthpiece with airflow recorded using a Fleisch number 4 pneumotachograph head (Fleisch, Lausanne, Switzerland), connected to a Mercury CS6 electrospirometer (Mercury Electronics, Glasgow, UK). The level of PEEPi was measured as the difference between the end-expiratory oesophageal pressure (EEPoes) and the Poes at the onset of inspiratory airflow [ 151. Ribcage and abdominal movements were monitored using inductance plethysmography (Respitrace Systems, Ardsley, New York 10502, USA) calibrated using an Ohio 840 dry spirometer (Ohio Medical Products, Wisconsin, USA). Tidal volume (V,) and minute ventilation (MV) were computed from the sum of rib cage and abdominal wall motion. Peripheral oxygen saturation (Spo,) was measured with an Ohmeda Biox 111 pulse oximeter (Ohmeda Ltd, Louisville, USA). All signals were written on to paper by a Mingograf 800 inkjet recorder (Siemens- Elema AB, Stockholm). Once the balloon catheters were positioned and calibrations completed, patients rested for at least 5 min after which recordings were made over 5 min with the patient breathing air spontaneously. The levels of EPAP and CPAP were matched to the mean EEPoes during SV with a minimum pressure of 5 cmh,o. Patients then received, in random order, inspiratory positive airway pressure (IPAP) alone, IPAP and EPAP, CPAP (all delivered using the BiPAP ventilator) and NPPV using their usual volume cycled flow generator. Ventilator settings for the VCFG limb were those used normally at home and had been shown to improve blood gas tensions during both wakefulness and sleep. For the BiPAP studies the maximum IPAP that the patient could tolerate was used and the respiratory rate was set at 10 breath.min-i. Patients received each mode of respiratory support for a run in period of at least 5 min and then the following measurements were made for all breaths over the following 5 min: respiratory frequency, sum of rib cage and abdominal wall motion, EEPoes, the negative deflection in Poes from the end-expiratory level of the previous breath (APoes) and the number of breaths associated with no negative deflection in Poes. APoes was used as an index of inspiratory muscle effort. Tidal volume and minute ventilation were computed for each mode of ventilatory support. Results are expressed as mean (standard deviation). Paired Student s t-tests were used for comparisons between spontaneous ventilation and the different modes of assisted ventilation, and unpaired tests for comparison between patients with chronic obstructive pulmonary diseasc: (COPD) and neuromusculo-skeletal disorders. Analysis of variance (ANOVA) was used to compare IPAP, IPAP/EPAP, VCFG and CPAP and the level of significance was set at p < Results The functional characteristics and level of PEEPi of the patients are shown in Table 1. The arterial blood gas tensions shown were those at the time of the study and in all cases were better than before starting domiciliary nasal ventilation. Comparison of each mode of ventilatory support with spontaneous ventilation (Figs 1 and 2) Minute ventilation increased during all modes of ventilatory support but only reached statistical significance with VCFG (+ 1.6 (2.2) l.min-, p = 0.04). V, was increased by VCFG (+0.17 (0.09)1, p = O.OOOl), IPAP (+0.31 (0.28)1 p = 0.005), and IPAP/EPAP ( (0.19)l p = 0.01) but not CPAP (-0.01 (0.09)1 p = 0.77). This was associated with a reduction in respiratory rate (breath.min-i) during VCFG (-3.0 (3.8), p = 0.02), IPAP (-6.6 (4.8), p = 0.001) and IPAP/EPAP (-4.5 (5.3), p = 0.02) but not during CPAP (-0.9 (3.6), p = 0.46).

3 Noninvasive ventilatory support Y s-" * Fig. 1. Mean (SD) tidal volume (VT), respiratory rate and minute volume (MV) for spontaneous ventilation (S), nasal positive pressure ventilation with a volume cycled flow generator (V), inspiratory positive airway pressure with (I/E) and without (I) expiratory positive airway pressure, and continuous positive airway pressure (C). *p < 0.05 (compared to S). The APoes for each breath was reduced during VCFG ( -9.5 (3.2)cmH20, p < 0.001), IPAP (-8.8 (3.0)cmHz0, p < 0.001) and IPAP/EPAP (-9.3 (2.9)cmH20, p < 0.001) but not during CPAP (-2.5 (4.5)cmH20, p = 0.11). All breaths during SV and CPAP occurred with a negative deflection in Poes, but the mean (SD) percentage of breaths associated with no negative deflection in Poes was markedly reduced with VCFG - 64% (44) p < 0.001; IPAP - 39% (39) p < 0.001; IPAPiEPAP - 48% (45) p < Gastric pressure during expiration was unchanged by any of the modes of ventilatory support; mean difference compared to SV: IPAP +0.7 (3.5) cmh?o, IPAP/EPAP +0.3 (3.3) cmh,o, NPPV +0.8 (3.1) cmhzo and CPAP +0.6 (3.5) cmhzo, (p > 0.05). Oxygen saturation was better during VCFG (+4.0 (2.6)%, p = 0.003), IPAP (+ 3.0 (2.6)%, p = 0.003) and IPAPiEPAP (+ 2.7 (2.1)%, p = 0.002) and worse during CPAP (-2.0 (1.8)%, p = Comparison between volume cjdedjoir generator and inspiratory positive airway pressure There were no significant differences in any of the measured variables between VCFG and IPAP except that respiratory rate was lower during IPAP (Table 2 and Figs 1 and 2). The number of breaths associated with no negative deflec- tion in Poes was less during IPAP, but this did not reach statistical significance. Comparison between inspiratory positive airway pressure and IPAPIEPAP Tidal volume was less and respiratory rate higher when EPAP was added but otherwise there were no statistically significant differences (Table 2 and Figs 1 and 2). Comparison of continuous positive airway pressure with the other niodes of ventilatory support CPAP was inferior to the other modes of ventilatory support in terms of APoes and oxygen saturation (Table 2 and Figs 1 and 2). Comparison between patients with chronic obstructive pulnionarj, disease and neuromusculo-skeletal disorders Comparisons of the changes, compared with baseline values during SV, seen during the different modes of ventilatory support between the COPD and the neuromusculoskeletal patients did not show any statistically significant differences other than in oxygen saturation during VCFG (COPD + 6.4(2.4)%; neuromusculo-skeletal + 1.2( 1.7)%, Table 1. Physical characteristics, diagnosis, respiratory values and positive pressures applied in the individual patients shown. FEV,:FVC Pao, Paco2 PEEP, IPAP EPAP Agejsex Diagnosis 1 mmhg mmhg cmh,o cmh,o cmh,o 1 61 M COPD 2 64M COPD 3 64F COPD 4 66M COPD 5 54M COPD 6 71 M COPD 7 46 M KS 8 58 M KS/asthma 9 58 M KS M KS M KS : : ; y / !l ! ! ! f * FEV,, forced expiratory volume in 1 s; FVC, forced vital capacity; COPD, chronic obstructive pulmonary disease; KS, kyphoscoliosis; PEEP,, intrinsic positive end-expiratory pressure; IPAP, inspiratory positive airway pressure; EPAP, expiratory positive airway pressure; * = not measured.

4 282 M. W. Elliott et al. Fig. 2. Mean (SD) APoes and inspiratory muscle effort for spontaneous ventilation (S), nasal positive pressure ventilation with VCFG (V), inspiratory positive airway pressure with (I/E) and without (I) expiratory positive airway pressure and continuous positive airway pressure (C). *p < 0.05 (compared to S). p = 0.003) and IPAP (COPD +4.5(1.9)% and neuromusculo-skeletal + 1.3(2.3)%, p = 0.04). Discussion The volume cycled flow generator increased minute ventilation above the level seen during spontaneous ventilation, but VCFG, IPAP and IPAP/EPAP all increased tidal volume and decreased the respiratory rate. This change in the pattern of breathing improves alveolar ventilation. In addition patients adopt a rapid shallow pattern of breathing in response to loading of the respiratory muscles, for instance during a failed weaning trial [16] and therefore the change to slower deeper breaths suggests that the respiratory muscles are under less load. Oesophageal pressure was reduced by VCFG, IPAP and IPAP/EPAP, but not by CPAP. This, together with the large number of breaths associated with no negative deflection in oesophageal pressure, suggests that all three modes of ventilatory support substantially reduce inspiratory muscle work. Compared with spontaneous ventilation, oxygen saturation was improved by VCFG, IPAP and IPAP/EPAP but was worse with CPAP. The change in oxygen saturation during VCFG and IPAP was significantly better in the COPD patients than in the neuromusculo-skeletal patients. This can be explained by the fact that oxygen saturation during spontaneous ventilation was usually less than 90% in the COPD patients, and there was therefore room for improvement, whereas it was near normal in the neuromusculo-skeletal patients. Respiratory rate during IPAP was slower than that during VCFG but otherwise there were no significant differences between these two modes of ventilatory support. This difference in rate may simply reflect different ventilator settings: during ventilation with a VCFG functioning in the assist/control mode, most breaths were controlled, i.e. initiated by the machine, whereas many breaths were initiated by the patient, i.e. assisted, during IPAP. Analysis of the individual traces, particularly in the patients with neuromuscular/skeletal disorders, showed that the BiPAP often triggers without any appreciable negative deflection in APoes. The mandatory 2 cmh,o of EPAP may be important in this respect by facilitating inspiratory aidlow and hence triggering the inspiratory Table 2. Comparison between volume cycled flow generator (VCFG) and inspiratory positive airway pressure (IPAP) (positive value represents VCFG > IPAP and vice versa), IPAP and IPAP/EPAP (expiratory positive airway pressure) (positive value represents IPAP > IPAP/EPAP and vice versa) and analysis of variance for VCFG, IPAP, IPAP/EPAP and continuous positive airway pressure (CPAP). Values are mean (SD). CPAP vs VCFG vs IPAP vs IPAP/ VCFG-IPAP IPAP-IPAP/EPAP EPAP Spo,; Yo -0.9 (2) -0.3 (2.0) F = 11.1, p < MV; I.min- (p = 0.15) (4.5) (p = 0.59) 0.24 (1.8) F = 0.21, p = 0.89 (p = 0.98) (p = 0.67) VT; (0.26) 0.13 (0.14) F = 1.65, p = 0.19 Respiratory rate; breathmin- (p = 0.11) +3.6 (3.2) (p = 0.002) -2.2 (1.2) F = 8.3, p = < (p = 0.004) (p = 0.OOOl) Breaths with no negative Poes; % 25 (40) -9 (40) F = 5.31, p = APoes; cmh,o -0.7 (2.6) (p = 0.07) 0.48 (1.9) (p = 0.49) F = 16.4, p = < (p = 0.38) (p = 0.42)

5 Noninvasive ventilatory support 283 phase. It is not surprising that an average of 6 cmh,o of CPAP did not provide the same reduction in inspiratory effort as the much higher pressures used with VCFG and BiPAP. However, higher levels of CPAP are often poorly tolerated by patients because of difficulty in breathing out. Pgas during expiration was unchanged by all the modes of ventilatory support suggesting that the expiratory work of breathing was unaffected. We had hypothesised that the addition of EPAP to IPAP might be beneficial. However, tidal volme was less and respiratory rate more when EPAP of at least 5 cmh,o was added to IPAP. Because the level of IPAP was the same during the IPAP and the IPAP/EPAP studies, increasing the level of EPAP reduces the differential pressure between inspiration and expiration (making it more like CPAP) and this might be expected to make ventilatory support less effective. In other words, any possible beneficial effects of EPAP are offset by the reduction in effective inflation pressure. A high level of EPAP may of itself be deleterious and in three patients (nos. 2, 4 and 8) it exceeded PEEP,. This difference may have been greater in the COPD patients because PEEPi may have been an overestimate as a consequence of activation of expiratory muscles, as evidenced by the high levels of Pgas at end-expiration [17]. Oxygen saturation with CPAP was worse than during spontaneous ventilation and APoes unchanged, suggesting that CPAP through a nasal mask has little role in this group of patients with established chronic respiratory failure. All the patients used VCFGs at home and therefore the changes seen may have been biased in favour of these devices. However, patients were acclimatised to the BiPAP before the study, which did not start until they were comfortable and happy with the technique. Furthermore, the results with the BiPAP in this study are very similar to those of Ambrosino and colleagues [ 181. Measurements during spontaneous ventilation were made first in all patients because positive pressure ventilation per se may affect lung mechanics, for example by recruiting atelectatic lung, improving chest wall compliance or worsening hyperinflation. Bias towards any mode of ventilation is unlikely in view of the fact that ventilator trials were performed in random order. This also makes it unlikely that the differences seen in the COPD patients were because of changes in airflow limitation as a consequence of differences in the time intervals since last receiving bronchodilator therapy. It is important to note that higher minute volumes could have been used with the VCFGs, whereas IPAP was at, or near to, the maximum level possible with the BiPAP. In conclusion, there was little difference between VCFG, IPAP and IPAP/EPAP used during wakefulness in these stable outpatients well acclimatised to assisted ventilation. It is important to note that these findings may not apply during sleep, particularly in patients who have episodes of central apnoea and in whom the impedance to inflation is high. In acutely ill patients, with erratic breathing and a changing impedance to inflation, small differences between the different modes may assume greater significance. Further comparative studies are required in these areas. The greater increase in minute ventilation with VCFG and the ability of current machines to develop higher inflation pressures than those possible with BiPAP suggest that a VCFG is preferable when hypoventilation is the dominant problem. However, conventional intensive care ventilators are capable of delivering higher levels of pressure support and may be used with a nasal mask in hospital. CPAP with a nasal mask is less effective than the other modes of ventilatory support. References ELLIS ER, BYE PTB, BRUDERER JW, SULLIVAN CE. Treatment of respiratory failure during sleep in patients with neuromuscular disease. American Review of Respiratory Disease 1987; HECKMATT JZ, LOH L, DUBOWITZ V. Night-time nasal ventilation in neuromuscular disease. Lancet 1990; 335: ELLIS ER, GRUNSTEIN RR, CHAN S, BYE PTB, SULLIVAN CE. Noninvasive ventilatory support during sleep improves respiratory failure in kyphoscoliosis. Chest 1988; 94: CARROLL N, BRANTHWAITE MA. Control of nocturnal hypoventilation by nasal intermittent positive pressure ventilation. Thorax 1988; 43: ELLIOTT MW, SIMONDS AK, CARROLL MP, WEDZICHA JA, BRANTHWAITE MA. Domiciliary nocturnal nasal intermittent positive pressure ventilation in hypercapnic respiratory failure due to chronic obstructive lung disease: effects on sleep and quality of life. Thorax 1992; 47: ELLIOTT MW, STEVEN MH, PHILLIPS GD, BRANTHWAITE MA. Non-invasive mechanical ventilation for acute respiratory failure. British Medical Journal 1990; 300: UDWADIA ZF, SANTIS GK, STEVEN MH, SIMONDS AK. Nasal ventilation to facilitate weaning in patients with chronic respiratory insufficiency. Thorax 1992; 47: RESTRICK LJ, SCOTT AD, WARD EM, FENECK RO, CORNWELL WE, WEDZICHA JA. Nasal intermittent positive-pressure ventilation in weaning intubated patients with chronic respiratory disease from assisted intermittent positivepressure ventilation. Respiratory Medicine 1993; 87: STRUMPF DA, MILLMAN RP, CARLISLE CC, GRATTAN LM, RYAN SM, ERICKSON AD, HILL NS. Nocturnal positivepressure ventilation via nasal mask in patients with severe chronic obstructive pulmonary disease. American Review of Respiratory Disease 1991; 144: KATZ JA, OZANNE GM, ZINN SE, FAIRLEY HB. Time course and mechanisms of lung-volume increase with PEEP in acute pulmonary failure. Anesthesiology 198 I; 54: SMITH TC, MARINI JJ. Impact of PEEP on lung mechanics and work of breathing in severe airflow obstruction. Journal of Applied Physiology 1988; PETROF BJ, KIMOFF RJ, LEVY RD, COSIO MG, GOTTFRIED SB. Nasal continuous positive airway pressure facilitates respiratory muscle function during sleep in severe chronic obstructive pulmonary disease. American Review of Respiratory Disease 1991; DEHAVEN CB, HURST JM, BRANSON RD. Postextubation hypoxemia treated with a continuous positive airway pressure mask. Critical Care Medicine 1985; 13: 468. MILIC-EMILI J, MEAD J, TURNER JM, GLAUSER EM. Improved technique for estimating pleural pressures from esophageal balloons. Journal of Applied Physiology 1964; DAL VECCHIO L, POLESE G, Poca R, Ross1 A. Intrinsic positive end-expiratory pressure in stable patients with chronic obstructive pulmonary disease. European Respiratory Journal 1990; 3: TOBIN MJ, PEREZ W, GUENTHER SM, SEMMES BJ, MADOR MJ, ALLEN SJ, LODATO RF, DANTZKER DR. The pattern of breathing during successful and unsuccessful trials of weaning from mechanical ventilation. American Review of Respiratory Disease 1986; 134: NINANE V, YERNAULT JC, DE TROYER A. Abdominal muscles, active expiration and intrinsic PEEP (PEEP,) in stable patients with chronic airflow limitation (CAO). European Respiratory Journal 1992; 5 (Suppl.): 42 1 s. [ 181 AMBROSINO N, NAVA S, BERTONE P, FRACCHIA C, RAMPULLA C. Physiologic evaluation of pressure support ventilation by nasal mask in patients with stable COPD. Chest 1992; 101: