PEDIATRICS Vol. 110 No. 4 October 2002, pp. 847-848
Heliox Therapy
To the Editor.—I read with interest the article by Martinón-Torres et al1 that evaluated the benefits of heliox therapy in infants with bronchiolitis. However, the methods and results raise several questions.
The duration of the study spanned 2 respiratory syncytial virus seasons, and the sequential randomization scheme would have caused the placebo and treatment group to potentially be from 2 separate seasons where the severity of the viral infection and the clinical approach may have differed. The lack of blinding also allows the introduction of bias. All patients could have been given either heliox or oxygen via a facemask with the tank covered or placed behind a screen. The cry pitch of infants on heliox is often indistinguishable to children not receiving heliox and may have not influenced the blinding of the study as the authors claim.
Most interesting is the approach to those infants receiving heliox who required supplemental oxygen therapy to maintain their oxygen saturations. The use of nasal cannula oxygen in addition to the 70/30 heliox mixture via facemask would significantly decrease the amount of helium inspired. The amount of decrease would depend on the flow through the cannula. This would introduce error against the use of heliox assuming its use is beneficial, thus potentially making the clinical improvement noted in the heliox group more significant. However, the use of nasal cannula oxygen may have administered inadvertent positive end expiratory pressure. If a large number of infants in the heliox group required nasal cannula oxygen (data not given in the article), then the benefits noted with heliox may be entirely from inadvertent positive end expiratory pressure distending the small airways.
Finally, the transfer criteria were not set before the study, although the authors do state that the normal discharge policies for the intensive care unit were followed. The heliox patients were discharged 45 hours sooner than the conventional therapy group. How do the authors explain this change when heliox has no direct therapeutic effect and is indeed a temporizing or "cosmetic" therapy? Is the difference found attributable to changing discharge criteria in this unblinded study, the study spanning 2 respiratory syncytial virus seasons, or is there an as of yet undefined therapeutic effect of heliox?
Despite the potential problems with the study, the clinically relevant finding is that the apparent effects of heliox therapy have a rapid onset within the first hour of treatment and may allow time for other therapies to take effect without the risk of respiratory failure.
John P. Straumanis, MD, FAAP
Department of Pediatrics
Division of Pediatric Critical Care
University of Maryland Hospital for Children
Baltimore, MD 21201
REFERENCE
- Martinón-Torres F, Rodríguez-Núñez A, Martinón-Sánchez JM. Heliox therapy in infants with acute bronchiolitis.
Pediatrics.2002; 109
:68
–73
[Abstract/Free Full Text]
In Reply.—
We thank Dr Straumanis for his interest in our article.1 Most of the limitations of our study that he noted had already been stated in our discussion, and we partially agree with some of his comments.
His first comment concerns the duration of the study. Two respiratory syncytial virus seasons were required to include an adequate number of patients, attributable to the limited number of patients admitted to our hospital per year. Bigger centers would probably achieve it in one season or even in less time. Even so, comparison of patients’ data grouped by season of inclusion did not show any significant difference.
Another comment concerns the lack of blinding and proposes some suggestions that might possibly solve it. Although it has already been stated in our original paper, it seems essential to insist that in spontaneously breathing patients heliox should be delivered by a nonrebreather face mask with reservoir to reduce the amount of external air entrained on inspiration and the subsequent dilution of helium concentration. Conventional oxygen masks produce dilution of supplied gas with air according to patient factors such as peak inspiratory flow rates, duration of expiratory pause, and variation in tidal volume, and also according to "mask factors" such as type of mask, fresh gas flow rate, tightness of fit, and deadspace of the device.2 In any case, if we had used nonrebreathing facemasks in the control group, the FiO2 delivered would be extremely high in comparison to the FiO2 delivered in the heliox group. To avoid it and make groups comparable, we would have had to mix air and oxygen, which would complicate the logistics and introduce the risk of administering a hypoxic mixture. Regarding the change in cry pitch of infants on heliox, although we agree with Dr Straumanis that it is often indistinguishable to that of children not receiving heliox, this is not always the case, particularly in bigger patients. In this context, the disguise of the tanks seems unimportant to us, although easily feasible; currently, wall-supply for both oxygen and heliox is available at our unit.
Dr Straumanis wonders, on the one hand, whether the use of nasal cannula oxygen in addition to heliox would decrease the amount of helium inspired and, on the other hand, whether its use would lead to inadvertent positive end expiratory pressure (PEEP), which could explain the positive effects found in the heliox group. Nasal cannula beneath the facemask was used to supply oxygen when needed, to maintain pulse oximetry oxygen saturations above 90%, and, simultaneously, to measure end-tidal carbon dioxide. The possibility of some background PEEP had been considered; however, the maximal flow used through the nasal cannula was 3 L/min, and thus the inadvertent PEEP would be nearly trivial. In any event, we should take into account that all patients in the control group—in contrast to only 6 patients in the heliox group—received some degree of supplemental oxygen through the nasal cannula. Therefore, even accepting that inadvertent PEEP occurred, this would benefit control group patients, making the clinical improvement noted in the heliox group less significant than it actually was. Regarding the dilutional issue, we agree with Dr Straumanis that the more supplemental oxygen required, the less proportion of helium in the airway, and the fewer potential benefits; although significantly lessened, heliox properties exist even with proportions down to 40%. Moreover, we should remember that heliox properties are not restricted to its effect in airway resistance. Comparison of the data from the 6 patients in the heliox group receiving additional oxygen supply to those from the remaining 13 patients of this group showed no differences.
Regarding the transfer criteria applied in our study, we have followed the discharge policies normally used in our pediatric intensive care unit for bronchiolitis patients (adapted from other standard policies), which we set before the study (although not included in the text) and equally applied to both groups. However, we cannot indeed explain the reduction in the mean length of stay of heliox-treated patients, and possibly the factors pointed out by Dr Straumanis and other uncontrolled factors different from heliox properties and/or discharge policies could have influenced it.
Finally, we believe that our article’s discussion explored the strengths and weaknesses of our investigation in a balanced fashion. Readers are not expected to conclude, on the basis of our study, that a definitive treatment effect of heliox therapy in bronchiolitis patients has been demonstrated, but the results are encouraging and additional studies are warranted.
Federico Martinón-Torres, MD, PhD
Antonio Rodríguez-Núñez, MD, PhD
Jose María Martinón-Sánchez, MD, PhD
Pediatric Emergency and Critical Care Division, Department of Pediatrics, Hospital
Clinico Universitario de Santiago
University of Santiago
Santiago de Compostela, Spain
REFERENCES
- Martinón-Torres F, Rodríguez-Núñez A, Martinón-Sánchez JM. Heliox therapy in infants with acute bronchiolitis. Pediatrics.2002; 109 :68 –73
- Stillwell PC, Quick JD, Munro PR, Mallory GB. Effectiveness of open-circuit and oxyhood delivery of helium-oxygen.
Chest.1989; 95
:1222
–1224
[Abstract/Free Full Text]
PEDIATRICS (ISSN 1098-4275). ©2002 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
