PEDIATRICS Vol. 107 No. 6 June 2001, pp. 1323-1328

Feasibility of Tidal Volume-Guided Ventilation in Newborn Infants: A Randomized, Crossover Trial Using the Volume Guarantee Modality

Irfan Ulhaq Cheema, MRCP^* and Jagjit Singh Ahluwalia, FRCPCH

From the ^* Department of Paediatrics, University of Cambridge; and the  Neonatal Intensive Care Unit, The Rosie Hospital, Cambridge, United Kingdom.

	ABSTRACT

Top Abstract Methods Results Discussion Conclusion References

Background and Aim.  Volume guarantee (VG) is a new composite mode of pressure-limited ventilation, available on the Dräger Babylog 8000 ventilator, which allows the clinician to set a target mean tidal volume to be delivered while still maintaining control over peak airway pressures. This study aimed to investigate the feasibility and efficacy of this mode of ventilation in premature newborn infants with respiratory distress syndrome (RDS).

Methods.  Two groups of infants were studied: those receiving synchronized intermittent positive pressure ventilation (SIPPV) in early phase of RDS (group 1) and those in recovery phase of RDS being weaned from artificial ventilation through synchronized intermittent mandatory ventilation (SIMV; group 2). Both groups of infants were studied over a 4-hour period. Before the start of the study, the infants were either receiving SIPPV (group 1) or SIMV (group 2). Infants in group 1 were randomized to either continue on SIPPV for the first hour of the study or to receive SIPPV plus VG for the first hour. Subsequently, the 2 modes were used alternately for the remaining three 1-hour periods. Similarly, infants in group 2 were randomized to either continue on SIMV for the first hour of the study or to receive SIMV plus VG for the first hour. Data on ventilation parameters and transcutaneous carbon dioxide and oxygen were collected continuously.

Results.  Forty infants were studied, 20 in each group. The mean (standard error) gestational age was 27.9 (0.3) weeks; birth weight was 1064 (60) g. No adverse events were observed during the study. Fractional inspired oxygen during SIMV plus VG was 0.31 (0.3); during SIMV, 0.31 (0.3); during SIPPV plus VG, 0.41 (0.4); and during SIPPV, 0.40 (0.4). Transcutaneous carbon dioxide pressure during SIMV plus VG was 6.0 (2.2) kPa; during SIMV, 5.9 (2.2) kPa; during SIPPV plus VG, 6.4 (2.9) kPa; and during SIPPV, 6.4 (2.8) kPa. Transcutaneous partial pressure of oxygen during SIMV plus VG was 8.4 (8.7) kPa; during SIMV, 8.6 (8.8) kPa; during SIPPV plus VG, 7.6 (4.0) kPa; and during SIPPV, 7.7 (4.2) kPa. None of these differences was statistically significant. The mean (standard error) peak inspiratory pressure used during SIMV was 17.1 (3.4) cm of water; during SIMV plus VG, 15.0 (7.5) cm of water; during SIPPV plus VG, 17.1 (9.3) cm of water; and during SIPPV, 18.7 (8.3) cm of water. The mean airway pressure during SIMV plus VG was 6.5 (3.1) cm of water; during SIMV, 6.9 (2.8) cm of water; during SIPPV plus VG, 9.6 (4.5) cm of water; and during SIPPV, 9.8 (4.6) cm of water.

Conclusion.  VG seems to be a stable and feasible ventilation mode for neonatal patients and can achieve equivalent gas exchange using statistically significant lower peak airway pressures both during early and recovery stages of RDS.ventilation, airway pressure, volume guarantee, tidal volume.

Ventilation-related respiratory morbidity remains high in small premature infants despite the technical advances made in neonatal ventilation over the last 3 decades.¹ Intermittent positive pressure ventilation (IPPV) using time-cycled, pressure-limited, continuous-flow ventilators has been the standard for conventional ventilation of premature infants. The design characteristics of these ventilators make them suitable to address 2 common problems in ventilation of premature infants: that of leak around the uncuffed endotracheal tube² and the low compliance ratio between the patient's lungs and the ventilator circuit. IPPV also offers control over peak inspiratory pressure (PIP), potentially reducing barotrauma and the risk of airleaks and bronchopulmonary dysplasia.³ However additional research in ventilation-induced lung injury suggests that lung over expansion induced by large lung volumes or volutrauma also play a crucial role in its causation.^4-7 During IPPV, the tidal volume delivered to the patient is dependant on lung compliance and resistance, the patient's inspiratory effort, as well as the applied ventilator pressure. Improving lung mechanics, either because of surfactant administration^8-10 or during recovery phase of respiratory distress syndrome (RDS), can lead to delivery of inadvertently large tidal volumes. Large variations in tidal volume and alveolar ventilation not only raise the possibility of volutrauma, but if the problem is not recognized in time, could also lead to hypocarbia with its recognized deleterious effects on cerebral blood flow.^11-13 Furthermore, the anticipated benefits expected of triggered ventilation over conventional IPPV have yet to be realized.^14,15 These limitations of IPPV and the persisting respiratory morbidity in premature infants demand the development of newer ventilatory modalities to address these problems.

Volume guarantee (VG) is one such modality, offered on the Dräger Babylog 8000 ventilator (Dräger Medical, Lübeck, Germany). It can only be used in conjunction with the patient triggered modalities available on that ventilator, ie, synchronized intermittent positive pressure ventilation (SIPPV), synchronized intermittent mandatory ventilation, and pressure support ventilation. The addition of VG to one of the triggered modes allows the clinician to set a mean tidal volume to be delivered as well as the standard ventilator settings of PIP, positive end expiratory pressure (PEEP), inspiratory time, and respiratory rate.

This study aimed to assess the short-term efficacy of this new ventilation modality in premature newborn infants with RDS. We were particularly interested to see the effects of VG on airway pressures and on stability of gas exchange. We hypothesized that if the addition of VG to the conventional trigger modes worked as intended, then for periods in which the infant was making good inspiratory effort, the ventilator would be able to use a lower PIP to deliver set tidal volume.

	METHODS

Top Abstract Methods Results Discussion Conclusion References

The study was performed in the tertiary level neonatal intensive care unit at The Rosie Hospital in Cambridge, United Kingdom, after approval by the local research ethics committee. All infants <34 weeks' gestation receiving artificial ventilation for RDS on the Dräger Babylog 8000 were considered eligible, subject to written informed parental consent. The diagnosis of RDS was based on clinical and radiologic criteria. Infants requiring muscle relaxants and those with lethal congenital anomalies were excluded.

Infants in the study were divided into 2 groups; those in early phase of RDS ventilated on SIPPV (group 1), and those in the recovery stage of RDS, receiving SIMV (group 2).

For both groups of patients, the study was designed as a 4-period, 2-treatment, double-crossover trial. Before the study, all infants in group 1 were receiving SIPPV, on the ventilatory settings selected by the clinical team. The infants were randomized to receive 1 of the 2 modalities, either SIPPV alone or SIPPV plus VG for the first 1-hour period. Subsequently, the 2 modes were used alternately for the remaining three 1-hour periods, ie, SIPPV+VG/SIPPV/SIPPV+VG/SIPPV or, alternately, SIPPV/SIPPV+VG/SIPPV/SIPPV+VG. By the end of the study, the patients had received 2 hours each of SIPPV plus VG and SIPPV. Infants in group 2, those receiving SIMV, were also randomized in a similar manner.

The set tidal volume, to be delivered during SIMV plus VG and SIPPV plus VG was the mean of the tidal volumes delivered during the 30 minutes on SIMV or SIPPV before the start of the study, and which had achieved good gas exchange. Once selected, the set tidal volume was kept the same for the duration of the study in individual patients. Other ventilation settings, including the PIP limit were left unchanged. All nonurgent interventions were kept to a minimum during the study. Any deterioration in the infant's clinical status during the study was managed according to the unit's clinical guidelines. The device used for recording transcutaneous carbon dioxide and oxygen values (Radiometer Limited, Crawley, United Kingdom) was calibrated with a blood gas at the onset. The ventilator's flow sensor was calibrated at the start of the study. After every 10 seconds, the ventilator transmitted the last stored value of PIP, mean airway pressure, tidal volume, minute volume, and fractional inspired oxygen concentration to a multichannel analyzer (Grove Medical, Hampton, Middlesex, United Kingdom). The data were recorded continuously and were logged every minute by the analyzer. Incidents of air leaks during the study were also recorded.

Outcome and Sample Size

The main outcome measure was the average peak airway pressure used during the 2 modes of ventilation. A sample size of 20 patients in each of the 2 groups (groups 1 and 2) of infants was required to show a 10% difference in the mean PIP between the conventional trigger modes and VG, with power of 0.95 and a significance level of 0.05. Secondary outcome measures included mean airway pressure, expired tidal volume, minute volume, fractional inspired oxygen concentration, and transcutaneous carbon dioxide and oxygen pressures.

Statistical Analyses

Data from groups 1 and 2 were analyzed separately. The minute-by-minute logged data on the ventilatory parameters and the transcutaneous carbon dioxide and oxygen concentrations were averaged for each 1-hour period. A linear regression model was then fitted to these data, which enabled us to look at the effect of the 2 modes of ventilation in each group and the statistical significance of this effect, while controlling for effects attributable to period and patient. To assess any possible variation in effects of VG modality based on body weight, infants with weight at study 1000 g were compared with those weighing >1000 g within the same group, ie, group 1 or 2, for any significant differences in ventilatory parameters and transcutaneous gas pressures. Data were analyzed using SPSS Software, Version 9.0 (SPSS Inc, Chicago, IL).

	RESULTS

Top Abstract Methods Results Discussion Conclusion References

Forty infants were studied. Thirty-five infants received at least 1 dose of antenatal steroids. The 5 infants who did not receive antenatal steroids were all in group 2. Demographic characteristics of the study infants are summarized in Table 1. The comparative values of mean and peak airway pressures, delivered tidal volume, minute volume, fractional inspired oxygen, and transcutaneous carbon dioxide and oxygen are given in Table 2. The regression model fitted to the data showed that the mode of ventilation was the only factor that significantly affected PIP and mean airway pressure. When infants with study weight 1000 g were compared with those weighing >1000 g within the same group, there were no significant differences in the effects of VG on any of the parameters.

The use of VG with either SIPPV or SIMV allowed us to achieve stable gas exchange with significantly lower mean delivered PIP and mean airway pressure than with either SIPPV or SIMV alone. There were no incidents of pneumothoraces or other air leaks during the study.

In only 8 of the (10-second) epochs of transmitted data, was the delivered PIP equal to the PEEP. This was treated as PIP delivered for that period and was taken into account when the delivered PIP was averaged. On no occasion during SIMV plus VG or SIPPV plus VG was the actual PIP delivered by the ventilator higher than the set PIP, thus maintaining similar control over peak airway pressure as in conventional SIMV and SIPPV. Figure 1, A and B compare the PIP used during SIMV plus VG versus SIMV and during SIPPV versus SIPPV plus VG, respectively. Figure 2, A and B compare the range of tidal volumes delivered during SIMV versus SIMV plus VG and SIPPV versus SIPPV plus VG, showing an equally wide variation in the delivered tidal volumes during the 2 modalities in the 2 groups of patients.

View larger version (19K): [in this window] [in a new window]   Fig. 1.   A, Box and whisker plot comparison of the PIP (cm of water) used during the 2 modes of ventilation in group 2 (SIMV vs SIMV + VG). The limits of the box represent interquartile range and the thick line through the box represents the median. The whiskers represent minimum and maximum values with extreme outliers removed for clarity. B, Box and whisker comparison of the PIP (cm of water) used during the 2 modes of ventilation in group 1 (SIPPV vs SIPPV + VG). The limits of the box represent interquartile range and the thick line through the box represents the median. The whiskers represent minimum and maximum values with extreme outliers removed for clarity.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   A, Box and whisker comparison of expired tidal volume (mL/kg) during the 2 modes of ventilation in group 2 (SIMV vs SIMV + VG). The limits of the box represent interquartile range and the thick line through the box represents the median. The whiskers represent minimum and maximum values with extreme outliers removed for clarity. B, Box and whisker comparison of expired tidal volume (mL/kg) during the 2 modes of ventilation in group 1 (SIPPV vs SIPPV + VG). The limits of the box represent interquartile range and the thick line through the box represents the median. The whiskers represent minimum and maximum values with extreme outliers removed for clarity.

	DISCUSSION

Top Abstract Methods Results Discussion Conclusion References

VG is another addition to the already overwhelming array of options available on various ventilators. The PIP set on the ventilator during the standard trigger modes acts as an upper limit for the PIP during VG, while the ventilator attempts to deliver the set tidal volume using the lowest airway pressure possible. A software algorithm for VG is shown in Fig 3. According to this algorithm, the ventilator constantly attempts to deliver the tidal volume set by the operator. If the expired tidal volume measured by the ventilator is smaller than the tidal volume set by the clinician, and the ventilator cannot deliver the set tidal volume within the PIP limit set by the clinician, an alarm is triggered, suggesting that the clinician increases the PIP limit, the inspiratory time, or flow, to deliver the set tidal volume. After breaths in which the measured expired tidal volume exceeds the set tidal volume, the ventilator uses a lower PIP during the next breath. This process of readjustment is repeated after every mechanical breath and the PIP delivered by the ventilator varies between the set PIP and PEEP.

View larger version (13K): [in this window] [in a new window]   Fig. 3.   Software algorithm of VG mode on Dräger Babylog 8000 (Dräger Medical).

Data on the efficacy and feasibility of VG are scant. The single published study¹⁶ to date on VG reported data in abstract form from 8 patients. The data from this study showed that VG allowed ventilation with significantly lower airway pressures. We report our study to provide additional observations on VG in clinical use and to determine whether the current system would behave in a stable manner and whether infants could be ventilated adequately using VG. It is not presented as a definitive study on VG.

We ventilated 40 infants with acceptable gas exchange on VG without any detectable adverse effects. VG achieved similar ventilation and oxygenation to SIMV and SIPPV, while using lower PIP and lower mean airway pressure. Our results support the previously published data. These data would support the undertaking of a much larger trial of VG to determine whether clinical outcomes can be improved using this mode.

The lower PIP and mean airway pressure used during VG compared with SIMV or SIPPV alone were not unexpected. During conventional trigger modes, the ventilator responds to the patient's inspiratory efforts by delivering a mechanical breath. The PIP used during each of these breaths is the PIP set by the operator. The tidal volume delivered to the patient is not taken into account by the ventilator and varies according to airway resistance, lung compliance, and the strength and duration of patient's inspiratory effort. Thus, inadvertently, large tidal volumes may be delivered during SIMV with rapidly improving lung mechanics unless PIP is altered frequently. In contrast, during VG, the ventilator attempts to continuously compare expired tidal volume with the set tidal volume and reduces the PIP in subsequent ventilatory cycles if the expired (delivered) tidal volume exceeds the set tidal volume.

The use of expired tidal volume as the basis for determining delivered tidal volume is an important distinction, compared with most volume-controlled ventilators. In the presence of a leak around the endotracheal tube, using inspired tidal volumes may falsely overestimate the actual tidal volume delivered to the lungs. Expired tidal volume measurements avoid this error because gas leaving the lungs must have been delivered to the lungs in the first place. Only in the presence of a substantial leak around the endotracheal tube would there be concerns that expired tidal volume would falsely underestimate the actual tidal volume delivered to the lungs, given that leak is proportional to airway pressures and these are highest during inspiration rather than expiration. Furthermore, even if a ventilator error in estimating delivered tidal volume occurred, the deployed PIP would still not exceed that level determined by the clinician. Theoretically, VG could allow for airway pressures lower than set PIP when the lung characteristics or infant effort allows but limit deployed PIP to the clinician set PIP when needed.

The study failed to show any reduction in the variability in the delivered tidal volume between the conventional trigger modes and VG in the 2 groups of infants. Because the ventilator monitors the expired tidal volume during VG and alters the PIP accordingly, we might have expected fewer variations in the delivered tidal volume compared with SIMV or SIPPV. This lack of reduction in variability during VG may reflect continually changing patient effort but may also be attributable to the fact that VG uses historical expired tidal volume data to determine the PIP to use rather than real time data. This may be an important limiting factor in its clinical applicability.

	CONCLUSION

Top Abstract Methods Results Discussion Conclusion References

This study has demonstrated that the VG option, when combined with existing triggered modes of ventilation, allows for stable gas exchange in this group of infants with RDS and potentially enables the use of lower airway pressures.

	FOOTNOTES

Received for publication May 9, 2000; accepted Sep 21, 2000.

Reprint requests to (J.S.A.) Neonatal Intensive Care Unit, The Rosie Hospital, Robinson Way, 226, Cambridge, United Kingdom, CB2 2QQ. E-mail: jag.ahluwalia{at}msexc.addenbrookes.anglox.nhs.uk

	ABBREVIATIONS

IPPV, intermittent positive pressure ventilation; PIP, peak inspiratory pressure; RDS, respiratory distress syndrome; VG, volume guarantee; SIPPV, synchronized intermittent positive pressure ventilation; PEEP, positive end expiratory pressure.

	REFERENCES

Top Abstract Methods Results Discussion Conclusion References

1. Hack M, Friedman H, Fanaroff AA Outcomes of extremely low birth weight infants. Pediatrics 1996; 98:931-937 [Abstract/Free Full Text]
2. Kondo T, Matsumoto I, Lanteri CJ, Sly PD Respiratory mechanics during mechanical ventilation: a model study on the effects of leak around a tracheal tube. Pediatr Pulmonol 1997; 24:423-428 [CrossRef][Medline]
3. Greenough A, Roberton NRC Neonatal ventilation. Early Hum Dev 1986; 13:127-136 [CrossRef][Medline]
4. Dreyfuss D, Solar P, Basset G, Saumon G High inflation pressure pulmonary edema: respective effects of high airway pressure, high tidal volume, and positive end-expiratory pressure. Am Rev Respir Dis 1988; 137:1159-1164 [Medline]
5. Hernandez LA, Peevey KJ, Moise AA, Parker JC Chest wall restriction limits high airway pressure-induced lung injury in young rabbits. J Appl Physiol 1989; 66:2364-2368 [Abstract/Free Full Text]
6. Dreyfuss D, Saumon G Barotruama is volutruama, but which volume is responsible? Intensive Care Med 1992; 18:139-141 [CrossRef][Medline]
7. Parker JC, Hernandez LA, Peevy KJ Mechanisms of ventilator-induced lung injury. Crit Care Med 1993; 21:131-142 [Medline]
8. Stenson BJ, Glover RM, Parry GJ, Wilkie RA, Laing IA, Tarnow-Mordi WO Static respiratory compliance in the newborn: early changes after exogenous surfactant treatment. Arch Dis Child 1994; 70:F19-F24
9. de Winter JP, Merth IT, van Bel F, Egberts J, Brand R, Quanjer PH Changes of respiratory system mechanics in ventilated lungs of preterm infants with two different schedules of surfactant treatment. Pediatr Res 1994; 35:541-549 [Medline]
10. Choukroun ML, Llanas B, Apere H, Pulmonary mechanics in ventilated preterm infants with respiratory distress syndrome after exogenous surfactant administration: a comparison between two surfactant preparations. Pediatr Pulmonol 1994; 18:273-278 [Medline]
11. Greisen G, Munck H, Lou H Severe hypocarbia in preterm infants and neurodevelopmental deficit. Acta Paedatr Scand 1987; 76:401-404 [Medline]
12. Wyatt JS, Edwards AD, Cope M, Response of cerebral blood volume to changes in arterial carbon dioxide tension in preterm and term infants. Pediatr Res 1991; 29:553-557 [Medline]
13. Fujimoto S, Togari H, Yamaguchi N, Mizutani F, Suzuki S, Sobajima H Hypocarbia and cystic periventricular leukomalacia in premature infants. Arch Dis Child 1994; 71:F107-F110
14. Beresford MW, Shaw NJ, Manning D Randomized controlled trial of patient triggered and conventional fast rate ventilation in neonatal respiratory distress syndrome. Arch Dis Child Fetal Neonatal Ed 2000; 82:F14-F18 [Abstract/Free Full Text]
15. Baumer JH International randomized controlled trial of patient triggered ventilation in neonatal respiratory distress syndrome. Arch Dis Child Fetal Neonatal Ed. 2000; 82:F5-F10 [Abstract/Free Full Text]
16. Herrera CM, Gerhardt T, Everett R, Claure N, Musante G, Bancalari E Randomized crossover study of VG versus SIMV in very low birth weight infants recovering from respiratory failure. Pediatr Res. 1999; 45:304A