© 2004 American Thoracic Society
Acknowledging Previous Work Is Part of Scientific ProcessTo the Editor:We have read with great interest the recently published article on hypoxic response under hypocapnic conditions by Corne and colleagues (1). In this study performed in spontaneously breathing normal subjects who were noninvasively mechanically ventilated, the authors have studied the muscle pressure output (Pmus) amplitude during the course of hypoxia at three different levels of steady-state PETCO2. They found that when hypoxia was induced in moderate hypocapnia (mean PETCO2 ± SD of 30.7 ± 2.6 mm Hg), there were no changes in respiratory rate and Pmus, whereas hypoxia induced during mild hypocapnia (mean PETCO2 ± SD of 36.8 ± 3.8 mm Hg) resulted in an increase in Pmus and respiratory rate that was significant, but less than those observed in eucapnia. They concluded that the ventilatory response to hypoxia is maintained, although attenuated, under mild hypocapnia, whereas it is totally lost during moderate hypocapnia. These conclusions are exactly the same as those we reached with a different experimental setting, and that have been published in April 2002 (2). We used noninvasive ventilation in the controlled mode to obtain different steady-state levels of hypocapnia, and then induced hypoxia by adding N2 to the admission filter of the volumetric respirator used in the experiments, without changing the delivered ventilation or the PETCO2 levels. We found that some kind of ventilatory response (i.e., increase in diaphragmatic surface EMG, deformation of the mask pressure curve, triggered breaths, increases in actual tidal volume) could be elicited at moderate levels of hypocapnia. On the contrary, below a threshold level of 29.3 mm Hg PETCO2, there was no response whatsoever to hypoxia. This figure of 29.3 mm Hg is almost the same as that found by Corne and colleagues (1). We acknowledged this fact, referring to their then published abstract (3). It should be mentioned that our own findings were presented in an abstract form in 1994 (4). In our study, we assessed the vigilance state of the subjects, but could not get sufficient data to determine whether the hypocapnic threshold is different in wakefulness and sleep. It is unfortunate that Corne and colleagues (1) have not addressed this issue in their experiments. The question remains to be studied. The confirmation by Corne and colleagues (1) of our own findings that there is a CO2 threshold around the figure of 30 mm Hg, below which there is a complete loss of the ventilatory response to hypoxia, is welcome and appears to settle this crucial concept in ventilatory control.
a Centre Hospitalier Universitaire Sud Amiens, France FOOTNOTES Conflict of Interest Statement: V.J. and D.R. have no declared conflict of interest. REFERENCES
From the Authors: We cannot agree more that acknowledgment of previous work is important and is the fair thing to do. Yet, failure to cite earlier work is common, at least as judged by how often relevant work from our laboratory is not acknowledged. We believe omissions are mostly benign; even with best intentions some previous work may be missed. Some omissions, however, are deliberate and intended to enhance the importance of authors' own work. The preaching nature of this letter's title implies that (1) our failure to cite the work of Drs. Jounieaux and Rodenstein was deliberate, and (2) Drs. Jounieaux and Rodenstein are exemplary in acknowledging the work of others. We reject both implications. The article by Jounieaux and colleagues (1) appeared after we had submitted our manuscript (2). We did not redo the literature search between submission and revision. This was unfortunate but not a deliberate omission. We had nothing to gain by ignoring their work. We extensively reviewed previous work on the effect of hypocapnia on hypoxic responses, including a 1995 study by Roberts and colleagues (3; also Reference 20 in our article [2]) that used the same approach used by Jounieaux and colleagues: controlled mechanical ventilation (CMV) to induce central apnea followed by determination of the O2 saturation required to reinitiate respiratory efforts. We indicated that results of this approach are confounded by neuromechanical inhibition. With CMV, central apnea develops regardless of whether PCO2 decreases (4), and the chemical drive required to reestablish efforts is considerably greater than what existed before apnea ("control system inertia" [4, 5]). The mechanism of CMV-induced apnea is unknown but may be related to inhibitory laryngeal influences as air is forced through adducted vocal cords during neural expiration (6). Regardless of mechanism, failure of efforts to reappear at a given O2 saturation need not reflect lack of hypoxic response at the same PCO2 during spontaneous breathing; the hypoxic stimulus may simply not be enough to overcome neuromechanical inhibition. Citing the work of Jounieaux and colleagues (1) would have simply involved adding one more reference to the discussion dealing with the CMV approach. There would have been no threat whatsoever to the significance of our findings. The article by Jounieaux and colleagues (1) adds nothing (i.e., relevant to our study) to the findings of Roberts and colleagues (3). Preachers should practice what they preach. In the article by Jounieaux and colleagues (1), extensive previous work on effect of hypocapnia on hypoxic responses (see DISCUSSION in Reference 2) was ignored. Particularly notable is failure to cite Roberts and colleagues (3) who used similar methodology and obtained similar results.
a University of Toronto Toronto, Ontario, Canada FOOTNOTES Conflict of Interest Statement: M.Y. and S.C. have no declared conflict of interest. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
