As an advocate for patient safety and for the introduction of appropriate monitoring technology into perioperative practice, I read the recent article by Taenzer et al.  1with anticipation. Unfortunately, the conclusions reached by the authors do not appear to be supported by the evidence provided in the article. Specifically, the authors conclude “our results demonstrate that continuous patient surveillance can improve outcomes in a postoperative orthopedic ward setting.” There was no meaningful difference in death, intensive care unit transfer, or hospital stay; the only reported difference was in the number of “rescue” events. The rescue events consisted of several levels of intervention ranging from conventional code blue teams to a bedside visit of an intensive care nurse and a respiratory care technician within 10 min of call. Surely, the “rescues” at the latter end of the range cannot be considered significant clinical or resource utilization outcomes as described within. Although the authors note that the types of rescues activated were collected, the actual distribution by type of event before and after surveillance was not provided in the article.

This article does break with tradition in a positive way in that it studies the impact of Spo2surveillance in a clinical area where, by routine practice, patients are only assessed intermittently and where hypoxic events are not rare. In the past, the value of pulse oximetry was assessed in areas where intensive monitoring was already the rule.2 

One cannot help but be a bit confounded by the results of this study because important data are absent. At its heart, the authors claim that surveillance reduces interventions, but how is this possible? A priori , more monitoring should detect more true hypoxia, which in turn should lead to more interventions (at an earlier stage, perhaps), not fewer, in order to improve true clinical outcomes. Are there important patient care interventions that are excluded from reporting in this article, such as direct nursing care and calls to and action by responsible physicians, among others? No mention is made of what process a floor nurse was to follow, protocol-driven or ad hoc , once notified by the central paging system. Did the frequency of nurse intervention in adjusting a patient's posture or supplemental oxygen delivery, among other actions, change with surveillance?

Finally, one is left curious about the impact of more intense respiratory monitoring on postoperative management and patient satisfaction because these are not addressed in the article. We look forward to further research using this model once adequately powered to discern clinically relevant outcomes.

*Health Science Center, Stony Brook University, Stony Brook, New York.

Taenzer AH, Pyke JB, McGrath SP, Blike GT: Impact of pulse oximetry surveillance on rescue events and intensive care unit transfers: A before-and-after concurrence study. Anesthesiology 2010; 112:282–7
Pedersen T, Hovhannisyan K, Møller AM: Pulse oximetry for perioperative monitoring. Cochrane Database of Systematic Reviews  2009, Issue 4. Art. No.: CD002013