Objective. To assess the therapeutic effects of breathing a low-density gas mixture (heliox: 70% helium and 30% oxygen) in infants with bronchiolitis.
Design. Prospective, interventional, comparative study.
Setting. A pediatric intensive care unit (PICU) in a tertiary care, teaching hospital.
Patients. Thirty-eight infants, 1 month to 2 years old, consecutively admitted to the PICU for treatment of moderate-to-severe acute respiratory syncytial virus bronchiolitis.
Interventions. The first 19 patients were enrolled as the control group and received supportive care and nebulized epinephrine. In the next 19 patients, heliox therapy was added through a nonrebreather reservoir face mask.
Measurements and Outcomes. Respiratory distress score, respiratory rate, heart rate, end-tidal CO2 (etCO2), and pulse oximetry oxygen saturation (satO2) values were recorded at baseline and at regular intervals. Data obtained during the first 4 hours were analyzed for comparison purposes. Demographic data, age, time elapsed from the start of the symptoms to the admission to PICU, length of stay in PICU (PICU-LOS), and duration of heliox therapy were also collected for each patient. Reductions in clinical scores and PICU-LOS were considered primary outcomes.
Main Results. At baseline, the heliox and control groups had similar age (5.5 ± 3.1 vs 5.9 ± 3 months), previous length of course (47.3 ± 19.3 vs 45.4 ± 18.6 hours), clinical score (6.7 ± 1.1 vs 6.6 ± 1), heart rate (160 ± 24 vs 165 ± 20 beats per minute), respiratory rate (64 ± 7 vs 61 ± 7 respirations per minute), satO2 (91 ± 2.3 vs 91 ± 2.5%), and etCO2 (34 ± 7 vs 33 ± 6 mm Hg). Clinical score, heart rate, respiratory rate, and satO2 improved during the study in both groups. After 1 hour, the improvement in clinical score was significantly higher in the heliox group than in the control group (3.6 ± 1.16 vs 5.5 ± 0.89), and these differences continued to be significant at the end of the observation period (2.39 ± 0.69 and 4.07 ± 0.96, respectively), with a total average decrease in the score of 4.2 points in the heliox group versus 2.5 points in the control group. Heart and respiratory rates were also significantly lower in the heliox group compared with the control group after 1 hour and stayed lower throughout the rest of the study period. No changes were noted either in satO2 between groups or in etCO2 within or between groups throughout the study. Mean duration of heliox administration was 53 ± 24 hours (range: 24–112 hours) and no adverse effects were detected. PICU-LOS was significantly shorter in the heliox group (3.5 ± 1.1 days) than in the control group (5.4 ± 1.6 days).
Conclusions. In infants with moderate-to-severe respiratory syncytial virus bronchiolitis, heliox therapy enhanced their clinical respiratory status, according to the marked improvement in their clinical scores and the reduction of the accompanying tachycardia and tachypnea. This beneficial response occurred within the first hour of its administration and was maintained as long as heliox therapy continued. In addition, PICU-LOS was reduced in heliox-treated patients. Long-term prospective studies are required to corroborate these findings and to establish the proper place of heliox in the therapeutic schedule of bronchiolitis.
- helium-oxygen mixture
- respiratory syncytial virus infections
- airway obstruction
- respiratory therapy
- work of breathing
Since use of helium-oxygen mixtures (heliox) was proposed by Barach1,2 as a medical therapy for severe asthma and upper airway obstruction, several studies on its role in respiratory therapy have been performed, with controversial results in many cases. Because of this sparse and conflicting evidence of utility, heliox has not achieved wide application, although several potential indications have been suggested.3–5
The beneficial effects of heliox have been attributed to its substantially lower density compared with an air-oxygen mixture, reducing the driving pressure required under turbulent flow conditions and preserving laminar flow at higher flow rates.6 Therefore, heliox decreases the work of breathing by means of a marked reduction in resistance to gas flow. These properties have been reported to be of benefit in different pathologic settings such us upper respiratory obstruction of varied causes, asthma, acute exacerbations of chronic obstructive pulmonary disease, and others.7–14
Acute viral bronchiolitis attributable to respiratory syncytial virus (RSV) is the most common lower respiratory tract infection in the first year of life.15,16 An estimated 3% to 8% of hospitalized infants with bronchiolitis develop acute respiratory failure that requires mechanical ventilation.15,16 RSV bronchiolitis is characterized by increased airway resistance; thus, heliox could have beneficial physiologic effects in these patients. Despite the high incidence of RSV bronchiolitis and the absence of clearly effective treatments, heliox therapy has been scarcely tested in this setting.17,18
The aim of the present study was to assess the clinical efficacy of heliox in infants admitted to the pediatric intensive care unit (PICU) with moderate-to-severe acute RSV bronchiolitis.
MATERIALS AND METHODS
Infants 1 month to 2 years old admitted to the PICU with moderate-to-severe respiratory distress attributable to RSV-positive bronchiolitis were eligible for the study. Diagnostic criteria of bronchiolitis included tachypnea, cough, prolonged expiratory time, wheezing, rales, chest retractions, and hyperinflation of the lungs on chest radiographs. Before including a patient in the study, RSV infection was confirmed by enzyme-linked immunoadsorbent assay of nasal secretions. Moderate-to-severe respiratory distress was defined as a modified Wood’s Clinical Asthma Score (M-WCAS) of 5 or greater (Table 1). This score, previously used in other studies, includes “mild” categories of 0.5 points to better define the clinical response to therapy.19–21
Excluded from the study were patients who had underlying cardiopulmonary disease, who suffered from bronchiolitis and/or persistent airway hyperreactivity in the 3 months before the study, or who had received corticosteroids and/or bronchodilators within 2 hours of the initiation of the study.
Because of the difficulties of blinding the study, 19 consecutive patients who fulfilled the inclusion criteria during early winter 1999 were enrolled as a control group. These patients received the usual care used in our unit: oxygen, fluid hydration, careful monitoring, and nebulized epinephrine. After having completed the control group, the next 19 consecutive patients who fulfilled the inclusion criteria constituted the study group. In this group, heliox therapy was added.
The study was approved by the local ethical committee, and written informed parental consent was obtained in all cases before enrollment.
Tanks of dry heliox with a prefixed concentration (70% He and 30% O2) were used (Air Liquide Medicinal, Madrid, Spain). The tank was connected by plastic tubing to a conventional warmer and humidifier system, and delivered to the patient by a nonrebreather reservoir face mask. A starting gas flow rate of 10 liters per minute (lpm) was established, which was increased to a maximum of 15 lpm, provided the reservoir was adequately filled and pulse oximetry oxygen saturations (satO2) were ≥90%. If supplemental oxygen was required to maintain satO2 above 90%, it was supplied by a nasal cannula beneath the mask and titrated to this endpoint. When nebulized medication was required, it was administered using heliox, and gas flow rate was increased 5 lpm above the previous rate.
Heliox therapy was slowly discontinued when M-WCAS was maintained for at least 6 hours below 2. Heliox therapy was considered a failure when either the clinical score worsened or improvement was <2 points after the first hour of the study, when satO2 was below 90% despite supplementary oxygen through the nasal cannula, or if patients’ clinical condition deteriorated at any time during the study. In these circumstances, patients were switched to conventional therapy alone.
Measurements and Outcomes
M-WCAS, respiratory rate, heart rate, end-tidal CO2 (etCO2), and satO2 values were recorded at baseline, at 1 hour intervals for 4 hours, and afterward, each 8 hours, until either heliox therapy was discontinued (study group) or the patient was discharged to the floor (control group). Data obtained during the first 4 hours were analyzed for comparison purposes. Demographic data, including age, time elapsed from the start of the symptoms to the admission to PICU, length of stay in PICU (PICU-LOS), and duration of heliox therapy, were collected for each patient. Administration of bronchodilators or corticosteroids before the study was also recorded.
The principal outcome measures were the change in M-WCAS and in PICU-LOS, and the secondary outcome measures included changes in satO2, respiratory rate, heart rate, and etCO2. Our sample size of 19 patients in each group yielded and >95% power to detect a difference of change in M-WCASs of 1.5 between groups and >80% power to detect a change in PICU-LOS of 36 hours between groups.
M-WCAS scores were assessed in each patient by 1 of 3 pediatricians unaware of the purpose of the study and previously trained in this scoring system by the principal investigator; interobserver agreement was checked before the study in 10 infants with moderate-to-severe bronchiolitis, and the agreement was almost perfect (k = 0.83).
Data are presented as mean ± standard deviation. The differences between the heliox and control groups for each parameter were assessed by independent sample t tests. Repeated-measures analysis of variance was used to compare changes in measured parameters over the course of the study. Statistical significance was indicated by P < .01. All statistical analysis were performed using SPSS software 10.0 version (SPSS Inc, Chicago, IL).
Forty-one eligible patients were admitted to our PICU during the study period (November 1, 1999–February 1, 2001). One patient with underlying heart disease and 2 patients who had had bronchiolitis in the 2 preceding months were excluded from the study. Finally, a total of 19 patients receiving heliox on admission to the PICU were compared with 19 controls who did not; all of them completed the study. All patients received 1 dose of nebulized epinephrine at the start of the study; afterward, it was administered at 4-hour intervals as needed at the discretion of the pediatric intensivist responsible for the infant. Two patients in the control group and 1 patient in the heliox group had received nebulized adrenaline 2 hours before the initiation of the study. The rest of the patients received neither nebulized medications other than epinephrine nor systemic corticosteroids before enrollment or during the study. No patients required intubation. At baseline, the control and heliox groups had similar age, previous length of course, clinical score, heart rate, respiratory rate, satO2, and etCO2(Table 2).
M-WCAS, heart rate, respiratory rate, and satO2 improved over study time in both groups (P < .01 in each case). After 1 hour, the improvement in M-WCAS was significantly higher in the heliox group compared with the control group (3.6 ± 1.16 vs 5.5 ± 0.89; P < .01), with an average decrease in M-WCAS of 3 versus 1.1 points, respectively (P < .01). The score continued to improve in both groups (Fig 1), although differences in M-WCAS between the heliox and control groups remained statistically significant at the end of the observation period (2.39 ± 0.69 and 4.07 ± 0.96, respectively; P < .01). The total average decrease in the score since the beginning of the study was 4.2 points in the heliox group and 2.5 points in the control group (P < .01). Heart rate (Fig 2) and respiratory rate (Fig 3) were also significantly lower in the heliox group compared with the control group after 1 hour (P < .01); these rates stayed lower throughout the study period, with P < .01 at all time intervals. No differences in satO2 between groups were detected. No statistically significant differences were noted in etCO2 within or between groups throughout the study.
In the study group, duration of heliox administration ranged from 24 to 112 hours (mean: 53 hours, standard deviation: 24 hours). No adverse effects attributable to heliox therapy were detected. PICU-LOS was significantly shorter in the heliox group (3.5 ± 1.1 days) compared with the control group (5.4 ± 1.6 days; P < .01). No patient required PICU readmission after being discharged to floor.
Our results suggest heliox therapy as a safe and effective option for the management of acute RSV bronchiolitis. Heliox enhanced the clinical respiratory status of these patients, according to the marked improvement in their clinical scores and the reduction of the accompanying tachycardia and tachypnea. Interestingly, the onset of this beneficial response to heliox compared with conventional therapy occurred within the first hour of its administration and was maintained as long as heliox therapy was continued. Moreover, length of PICU stay was shorter in infants treated with heliox.
Helium is a biologically inert gas of low molecular weight that is one-eighth the density of nitrogen.3–5 When blended with oxygen, the resulting gas mixture has a marked reduction in density compared with air (specifically, it has a threefold reduction when blended with 21% oxygen). This reduction of density will have 2 consequences, depending on the flow conditions within the airways. First, the lower the density, the lower the Reynolds number and the more likely it is that laminar conditions will prevail. Consequently, the flow for a given driving pressure will be higher. Second, in turbulent conditions, the flow rate is inversely related to density, meaning that for a given driving pressure, flow will again be higher if density is lower. In summary, the use of heliox in the setting of obstructive airway disease is equivalent to decreasing airway resistance to flow and ultimately the work of breathing. Previous studies have shown that breathing heliox has beneficial physiologic effects on intubated and nonintubated patients with asthma, upper respiratory obstruction of varied cause, and chronic obstructive pulmonary disease.3–5,7–14 On the other hand, heliox has no inherent therapeutic effect, and thus, it can be used only as a temporizing agent: it provides time until definitive therapies act or the subjacent pathologic circumstance spontaneously resolves.
Bronchiolitis occurs mainly in children under 2 years of age.15,16 RSV is estimated to be responsible for over half of all bronchiolitis cases.15,16 A heterogeneous disease, RSV bronchiolitis involves mainly small airways but also lung interstitium.22 The obstruction of airways results in the increase of inspiratory and expiratory resistances. In these circumstances, the patient tries to counteract for the marked increase in airway resistance and to maintain adequate gas exchange, by increasing both respiratory effort and rate, namely, work of breathing. This response can lead to muscle fatigue and ultimately respiratory failure. In fact, RSV bronchiolitis can lead to respiratory insufficiency in >50% of bronchiolitis patients admitted to the PICU despite aggressive supportive therapy.16,23 Although no treatment has been shown to be consistently effective in bronchiolitis, its supportive therapy usually includes oxygen, intravenous hydration, nutrition, bronchodilators, and corticosteroids.24,25 Considering the physiopathologic chain of bronchiolitis, heliox properties could be of benefit in these patients. Some evidence suggests that small airway disease, where laminar flow prevails, might predominate in bronchiolitis22; if so, the expected benefits of heliox in this setting would be smaller.26 However, the exact proportion of peripheral versus large airways involved and its consequent influence on the clinical picture of bronchiolitis remain controversial and can vary from case to case. Furthermore, even when laminar flow conditions predominate, heliox has shown beneficial effects.6
When heliox was added to conventional therapy, our patients markedly improved not only in their clinical scores, but in their respiratory and heart rates as well. This reduction of a patient’s tachypnea and tachycardia could be mainly a consequence of the relief of the patient’s dyspnea and subsequent subjective recovery.3–5 Although heliox has been shown to enhance alveolar ventilation because of its high diffusion coefficient,26,27 no differences in etCO2 were noted within or between the control and heliox groups throughout our study. Apart from the technical limitations of capnography, this fact could be partially explained by the fact that basal etCO2 was within normal range in all studied patients.
In all the cases treated with heliox, its beneficial effects began within the first hour of therapy and were maintained as long as it was continued, which is reasonable if we consider its mechanism of action. Therefore, 1 hour seems to be long enough to detect nonresponders to heliox. Once the patients achieved the endpoint, they were slowly weaned from the heliox mixture and continued to do well. Hence, heliox seems able to maintain the patient in better respiratory condition while bronchiolitis follows its natural course, avoiding further potentially injuring and/or nonproven efficacious therapies. In addition, our study showed a mean decrease of 45 hours in the PICU-LOS of those patients treated with heliox, which could represent a significant savings in the cost of hospitalization. Although we have followed the discharge policies normally used in our PICU for bronchiolitis patients, larger specific studies are required to confirm this finding.
Heliox was first suggested as a therapy for bronchiolitis by Paret et al. The only prospective study of heliox therapy in spontaneously breathing children with acute bronchiolitis is by Hollman et al.18 They found that heliox improved the overall respiratory status of children with mild-to-moderate bronchiolitis, most markedly in those patients more severely affected. Similarly, our study showed an improvement in the clinical score of those infants treated with heliox compared with those treated with conventional therapy alone. In addition, in our series, the response to heliox was more marked according to the clinical score, it had an associated significant reduction in the accompanying tachypnea and tachycardia, and the positive effects were maintained as long as heliox was continued. The greater degree of respiratory compromise of the patients included in our study could explain the more pronounced response, as previously suggested by Hollman et al18 and other authors3–5; however, the differences in the study designs make comparing the results obtained difficult.
The inert nature of helium explains its scarce secondary effects. The most common risks of heliox administration, especially in infants, are both the incidence of hypoxemia because of high Fio2 requirements or the delivery of hypoxic mixtures, and the development of hypothermia attributable to the high thermal conductivity of helium.3–5 Both adverse effects can be at least partially prevented in 2 ways: 1) by using prefixed concentrations of helium-oxygen, which avoids the need of continuous Fio2 monitoring and ensures the constant delivery of a nonhypoxic mixture that contains enough helium to preserve its properties; and 2) with the adequate warming and humidification of the heliox mixture using standard devices. Neither hypoxemia nor hypothermia attributable to heliox occurred in our patients throughout the study. On the other hand, the best way to deliver heliox noninvasively is through a nonrebreather reservoir face mask3–5; this way the amount of external air entrained on inspiration and its resulting dilution of helium concentration are reduced. Although it has been suggested that this system may overwhelm the patient,5 in our experience the mask was well-tolerated.
There are several limitations to our study: sample size, lack of adequate blinding, and the use of changes in the clinical score as a primary outcome. Despite the reduced number of participants and the low power of the study to detect small differences between groups, all patients completed the study, the control and study groups were homogeneous, and the sample size was large enough to detect differential improvements in the measured primary outcomes. On the other hand, clinical scores, beyond their known limitations, are often used as a research tool to provide an objective measure of clinical improvement, and the measures already mentioned in the methodology section could have reasonably counteracted observer bias. The need for a nonrebreather reservoir face mask to deliver heliox adequately, together with the changes in voice/cry pitch secondary to heliox breathing, hindered adequate blinding.
Heliox seems to be effective in improving the respiratory condition of infants with acute bronchiolitis in a safe, noninvasive, simple manner. Taking into account the lack of proven efficacious treatments in this setting and considering its theoretical advantages, we believe that heliox might be considered a first-step therapy for patients with moderate-to-severe RSV bronchiolitis. Nevertheless, long-term prospective studies are needed to assess this belief and to establish the proper place of heliox in the therapeutic schedule of bronchiolitis.
- ↵Martinón-Torres F, Nuñez AR, Martinón JM. Heliox: perspectivas de aplicación en pediatría. An Esp Ped.1999;128(suppl) :42S– 45S
- ↵Martinón-Torres F. Otros modos de terapia respiratoria: heliox. In: Ruza Tarrío F, ed. Tratado de Cuidados Intensivos Pediatricos. Madrid, Spain: Norma-Capitel; In press
- ↵Manthous CA, Morgan S, Pohlman A, Hall JB. Heliox in the treatment of airflow obstruction: a critical review of the literature. Respir Care.1997;42 :1034– 1042
- ↵Wang EE, Law BJ, Stephens D. Pediatric Investigators Collaborative Network on Infections in Canada (PICNIC) prospective study of risk factors and outcomes in patients hospitalized with respiratory syncytial viral lower respiratory tract infection. J Pediatr.1995;126 :212– 219
- Tal A, Bavilski C, Yohai D, Bearman JE, Gorodischer R, Moses SW. Dexamethasone and salbutamol in the treatment of acute wheezing in infants. Pediatrics.1983;71 :13– 18
- ↵O’Brodovich HM, Haddad GC. The functional basis of respiratory pathology and disease. In: Chernick V, Boat TF, eds. Kendig’s Disorders of the Respiratory Tract in Children. Philadelphia, PA: WB Saunders; 1998:27–74
- ↵Shay DK, Holman RC, Roosevelt GE, Clarke MJ, Anderson LJ. Bronchiolitis-associated mortality and estimates of respiratory syncytial virus-associated deaths among US children, 1979–1997. J Infect Dis.2001;183 :16– 22
- ↵Deschildre A, Thumerelle C, Bruno B, Dubos F, Santos C, Dumonceaux A. Acute bronchiolitis in infants. Arch Pediatr.2000;7(suppl) :21S– 26S
- ↵Wood LDH, Engel LA, Griffin P. Effect of gas physical properties and flow on lower pulmonary resistance. J Appl Physiol.1976;41 :234– 244
- Copyright © 2002 by the American Academy of Pediatrics