WE read with great interest the article by Severgnini et al.1
in which the authors evaluated a protective ventilation strategy with low tidal volume (7.7 ± 0.8 ml/kg of predicted body weight), positive end-expiratory pressure (10 cm H2
O), and recruitment maneuvers (RMs) as compared with ventilation with higher tidal volumes (9.5 ± 1.1 ml/kg of predicted body weight), no positive end-expiratory pressure (zero end-expiratory pressure), and no RMs in patients undergoing open abdominal surgery and mechanically ventilated for at least 2 h. Their results showed better pulmonary functional tests, fewer alterations on chest radiograph, and higher arterial oxygenation on different postoperative days. One month later, Futier et al.
one of Severgnini’s coauthors, published a similar study on the efficacy of the protective ventilation strategy in intermediate- to high-risk patients undergoing major abdominal surgery. The authors similarly found that a protocol based on low tidal volume (6.4 ± 0.8 ml/kg of predicted body weight), positive end-expiratory pressure 6 cm H2
O (interquartile range, 6–8), and RMs was beneficial for major pulmonary and extrapulmonary complications evaluated within the first 7 days after surgery when compared with a more “standard” treatment based on a higher tidal volume (11.1 ± 1.1 ml/kg of predicted body weight), zero end-expiratory pressure, and no RMs. Futier, Severgnini, and their coauthors contributed important findings to the controversy regarding the best tidal volume and the usefulness of positive end-expiratory pressure and RMs during general anesthesia, thus improving our knowledge on this issue. Nonetheless, in both studies, no clear indications are given about the optimal oxygen inspiratory fraction (FIO2
) to be delivered during mechanical ventilation to limit the toxic effects of oxygen. In the study by Severginini et al.
, the authors state that “All patients were preoxygenated with F
IO2 0.8 before tracheal intubation, and maintained at 0.4 during the entire anesthesia procedure, irrespective of study group
,” whereas in the study by Futier et al.
, the FIO2
was 47.2 ± 7.6% and 46.4 ± 7.3% in the traditional versus
lung-protective groups, respectively (P
= 0.27). Not targeting the FIO2
for arterial partial oxygen pressure and saturation (PaO2
) could lead to an oversupply of oxygen and excessive reactive oxygen species production, which has been clearly identified as causing alveolar and organ inflammatory damage.3–8
Moreover, cardiovascular negative effects (such as an increase in vascular resistance, reduction in cardiac output, carotid and downstream cerebral arteries vasoconstriction, or a decrease in coronary blood flow) have been demonstrated in healthy people and during medical emergencies during routine use of supplemental oxygen.7
Depending on the concentration and duration of oxygen exposure, excessive production of reactive oxygen species may lead to the development of “oxidative stress” and consequently damage the lungs and other tissues.3–8
Effectively, to quote Winslow7
: “Oxygen: the poison is the dose.” Indeed, it has been demonstrated that even oxygen administration delivered at medium concentration (6 l/min) may have negative effects on outcome even in patients with cardiac ischemia.9
In this regard, we agree with Marino10
who wrote: “… why an organism that requires oxygen for survival is designed to carry on metabolism in an oxygen-limited environment? The answer may be related to the toxic potential of oxygen. Oxygen is well known for its ability to produce lethal cell injury via the production of toxic metabolites … so, limiting the oxygen concentration in the vicinity of cells may be the mechanism for protecting cells from oxygen-induced cell injury.
” That is like saying that there is surely a reason why Mother Nature provided 21% oxygen in the air we breathe. In conclusion, we believe that protective ventilation strategies should definitely be administered to patients undergoing surgery, but because oxygen clearly has a double-edged nature, we hope that forthcoming trials will include an FIO2
titration for optimal oxygenation (with the use of point-of-care blood gas analyzer and lactate concentration), keeping it as low as possible to deliver appropriate oxygen for any given patient, yet avoiding hyperoxia.
The authors declare no competing interests.
Stefano Romagnoli, M.D., Sergio Bevilacqua, M.D., Zaccaria Ricci, M.D., Angelo Raffaele De Gaudio, M.D.
Azienda Ospedaliero-Universitaria Careggi, Florence, Italy (S.R.). firstname.lastname@example.org