Strategy Matters

In this issue of the American Journal of Perinatology, Simbruner et al[1] report their findings from an elegant study of ventilator-induced lung injury conducted in a group of rabbits with a relatively mild degree of lung injury induced by saline lung lavage. Their results are consistent with the well-documented adverse effects of excessive tidal volume, but contradict the widely held belief that faster respiratory rates are safe, or indeed protective, with conventional ventilation and even more so with high-frequency ventilation.

There are many undeniable strengths of this study, especially the concept of examining lung injury at multiple levels, including biophysical, biochemical, histological, and transcriptional levels. The findings with these multiple measures of lung injury are internally consistent for the different experimental conditions, greatly strengthening confidence in the fundamental validity of the conclusions. However, the authors had a great deal of difficulty in publishing this fine work, encountering rejection after rejection, most likely because the damaging effects of higher respiratory rates simply flew in the face of everything we “know” about lung injury.

In reality, although the damaging effects of the higher respiratory rate may seem counterintuitive and at odds with accepted dogma, the findings are actually quite plausible. When the finer points of the study design, especially the ventilator strategy, are examined more closely, it becomes evident that the results are quite consistent with current knowledge in the field. Moreover, they are quite relevant because similar ventilator strategies are still used in certain clinical conditions, such as persistent pulmonary hypertension (PPHN).

The authors chose a rather injurious ventilator strategy as the baseline state. The initial normal tidal volume was 10 mL/kg, which resulted in mild hypocapnia despite interposition of a flow sensor and end-tidal CO₂ monitor that added 4 mL of dead-space volume. The 10-mL tidal volume was based on reported values from older literature, which probably relied on tidal volume measurement that did not take into account loss of tidal volume in the ventilator circuit. Moreover, the positive end-expiratory pressure (PEEP) was kept at a very low value of 2 cm H₂O throughout the study. Saline lavage depletes surfactant and leads to atelectasis. Ventilating atelectasis-prone lungs with high tidal volume and low PEEP has been documented to be the most injurious mode of ventilation.[2] Thus, it is not surprising that even a relatively short period of 6 hours resulted in considerable degree of lung injury even at baseline conditions.

Doubling the already generous tidal volume resulted in the most severe degree of lung injury, with rather high mortality in the fragile rabbit model that was used. Doubling the ventilator rate was also quite damaging, though less so than doubling the volume. This finding appears to contradict several published studies that documented better outcomes with rapid rate ventilation.[3] [4] [5] There is, of course a key difference: in these studies and in ordinary clinical practice, the faster rates are typically combined with lower tidal volumes, maintaining essentially the same minute ventilation and trading off the potential damage from the larger number of mechanical breaths against the lesser damage from smaller tidal volumes, with the balance apparently favoring the more rapid rates. High-frequency ventilation uses very rapid respiratory rates and very small tidal volumes. Though not all clinical trials showed a reduction in chronic lung disease, none showed increased lung damage despite these very fast rates.

More importantly, it is critical to recognize that the many laboratory and clinical studies that did show decreased lung injury with high-frequency ventilation all used the optimal lung volume strategy.[6] Animal studies have demonstrated clearly that optimizing lung volume decreases surfactant turnover and protects the lung from injury by preventing the repeated collapse and re-expansion of alveoli, avoiding shear stress where atelectatic and aerated alveoli coexist, and by distributing delivered tidal volume evenly into a homogeneously aerated lung. When a substantial portion of alveoli remains collapsed, even a normal tidal volume may overexpand the limited portion of aerated lung. With large tidal volumes and low PEEP, the effect of increased rate would therefore be expected to result in increased lung injury, given that the number of cycles of collapses and re-expansions is doubled. When smaller tidal volumes are used and sufficient PEEP is used to avoid alveolar collapse during exhalation, higher rates are not detrimental; indeed, they are the cornerstone of lung-protective ventilation strategies.[7] Thus, it is critical that the findings of this study are interpreted in the context of this specific ventilator strategy and not generalized to the usual clinical situation.

However, strategies very similar to that used in the study by Simbruner et al[1] are still used by many clinicians who continue to use hyperventilation as a treatment for PPHN. In that setting, doubling or tripling of minute ventilation with rapid rates and rather large tidal volumes is often combined with low PEEP in an effort to avoid air-trapping. It has been increasingly recognized that hyperventilation, though effective in improving oxygenation in the short term, commonly leads to worsening of the patient's status over time, and probably results in increased need for rescue therapies such as extracorporeal membrane oxygenation. The study by Simbruner et al[1] was meticulously done; for the first time, lung injury was examined at all levels, thus providing very strong evidence that high ventilator rates are not necessarily safe. Clearly, under certain conditions that are still relevant today, high ventilator rates may result in a substantial degree of lung injury. It is my hope that this important publication will hasten the disappearance of this most lung-injurious approach to the treatment of PPHN.

REFERENCES

• 1 Simbruner G, Mittal R A, Smith J et al.. Effects of duration and amount of lung stretch at biophysical, biochemical, histological, and transcriptional levels in an in vivo rabbit model of mild lung injury.  Am J Perinatol. 2007;  24 149-160
• 2 Dreyfuss D, Saumon G. Ventilator-induced lung injury: lessons from experimental studies.  Am J Respir Crit Care Med. 1998;  157 294-323
• 3 Bland R D, Kim M H, Light M J et al.. High frequency mechanical ventilation in severe hyaline membrane disease: an alternative treatment?.  Crit Care Med. 1980;  8 275-280
• 4 Oxford Region Controlled Trial of Artificial Ventilation (OCTAVE) . Multicentre randomized controlled trial of high against low frequency positive pressure ventilation.  Arch Dis Child. 1991;  66 770-775
• 5 Pohlandt F, Saule H, Schroder H et al.. Decreased incidence of extraalveolar air leakage or death prior to air leakage in high versus low rate positive pressure ventilation: results of a randomised seven-centre trial in preterm infants.  Eur J Pediatr. 1992;  151 904-909
• 6 Keszler M, Durand D J. Neonatal high-frequency ventilation: past, present, and future.  Clin Perinatol. 2001;  28 579-607
• 7 Clark R H, Slutsky A S, Gertsmann D R. Lung protective strategies of ventilation in the neonate: what are they?.  Pediatrics. 2000;  105 112-114