Nebulized 3% Hypertonic Saline

Solution Treatment in Hospitalized

Infants With Viral Bronchiolitis*
Avigdor Mandelberg, MD; Guy Tal, MD; Michaela Witzling, MD;
Eli Someck, MD; Sion Houri, MD; Ami Balin, MD; and
Israel E. Priel, MD, FCCP

Objective: To determine the utility of inhaled hypertonic saline solution to treat infants
hospitalized with viral bronchiolitis.
Design: Randomized, double-blind, controlled trial. Fifty-two hospitalized infants (mean ⴞ SD
age, 2.9 ⴞ 2.1 months) with viral bronchiolitis received either inhalation of epinephrine, 1.5 mg,
in 4 mL of 0.9% saline solution (group 1; n ⴝ 25) or inhalation of epinephrine, 1.5 mg, in 4 mL
of 3% saline solution (group 2; n ⴝ 27). This therapy was repeated three times every hospital-
ization day until discharge.
Results: The percentage improvement in the clinical severity scores after inhalation therapy was
not significant in group 1 on the first, second, and third days after hospital admission (3.5%, 2%,
and 4%, respectively). In group 2, significant improvement was observed on these days (7.3%,
8.9%, and 10%, respectively; p < 0.001). Also, the improvement in clinical severity scores differed
significantly on each of these days between the two groups. Using 3% saline solution decreased
the hospitalization stay by 25%: from 4 ⴞ 1.9 days in group 1 to 3 ⴞ 1.2 days in group 2 (p < 0.05).
Conclusions: We conclude that in nonasthmatic, nonseverely ill infants hospitalized with viral
bronchiolitis, aerosolized 3% saline solution/1.5 mg epinephrine decreases symptoms and length
of hospitalization as compared to 0.9% saline solution/1.5 mg epinephrine.
(CHEST 2003; 123:481– 487)

Key words: ␤2-agonist; epinephrine; hypertonic saline solution; respiratory syncytial virus; viral bronchiolitis

Abbreviations: CF ⫽ cystic fibrosis; NS ⫽ not significant; RSV ⫽ respiratory syncytial virus

R ofespiratory syncytial virus (RSV) is the chief cause

hospital admission for respiratory tract illness
hospital days could potentially greatly reduce health-
care expenditures. Despite 4 decades of efforts,
in young children. In the 1980s, an estimated there are no effective means to control RSV,1 and
100,000 children were hospitalized with RSV infec- still the mainstay of treatment for RSV infection is
tion in the United States annually, at a cost of $300 supplemental oxygen and hydration.5 Controversies
million.1–3 The magnitude of the cost is understand- exist related to the available treatments for acute
able, since virtually all children become infected bronchiolitis.1,5 Antiviral agents are available, but
with RSV within 2 years after birth,1 with 1% their use in most patients is controversial and there-
requiring hospitalization. Nearly two thirds of the fore not routinely indicated. The efficacy of ribavirin,
cost related to annual RSV epidemics is attributable the only specific drug available for the treatment of
to hospitalization.1,4 Therefore, therapies that reduce RSV infection,1 has not been demonstrated conclu-
*From the Pediatric Pulmonary Unit (Drs. Mandelberg and Tal),
sively.1,5–10 Most of the studies using glucocorticoids
the Department of Radiology (Dr. Witzling), the Unit of Pediat- in the treatment of bronchiolitis denied a positive
ric Infectious Diseases (Dr. Someck), the Pediatric Critical Care therapeutic effect in previously normal children with
(Dr. Houri), the Pediatric Department (Dr. Balin), and the
Department of Pulmonary Medicine (Dr. Priel), The Edith bronchiolitis.5,11–12 The use of ␤2-agonists occasion-
Wolfson Medical Center, Holon, and The Sackler School of ally resulted in a short-term improvement in patients
Medicine, Tel Aviv University, Tel Aviv, Israel. with bronchiolitis, especially when using epineph-
Manuscript received January 9, 2002; revision accepted July 23,
2002. rine,13–16 while others failed to show a significant
Correspondence to: Avigdor Mandelberg, MD, Director, Pediat- effect.5,17
ric Pulmonary Unit, The Department of Pulmonary Medicine and
Pediatrics, The Edith Wolfson Medical Center, 62 Halochamim Pathophysiologically, bronchiolitis is an infection
St, Holon 58100, Israel; e-mail: [email protected] of the bronchiolar epithelium, with subsequent pro-

www.chestjournal.org CHEST / 123 / 2 / FEBRUARY, 2003 481

found submucosal and adventitial edema, increased sign the informed consent: three patients intended for the 3%
secretion of mucus, peribronchiolar mononuclear saline solution group, and five patients intended for 0.9% saline
solution group. All eligible patients were randomly assigned to
infiltration, and epithelial cell necrosis. These one of two groups: group 1 received inhalation of epinephrine,
changes obstruct flow in the small airways, leading to 1.5 mg, in 4 mL of 0.9% saline solution; group 2 received
hyperinflation, atelectasis, and wheezing.1,5,18 A sin- inhalation of epinephrine, 1.5 mg, in 4 mL of 3% saline solution.
gle inhalation of recombinant human deoxyribonu- Patients in each group received three treatments for every day of
clease has been recently used as a mucolytic agent in hospital stay delivered at intervals of 8 h, until the patient was
ready for discharge. Additional inhalations as needed of epineph-
RSV bronchiolitis with some success.19 However, rine in 0.9% saline solution were recorded and calculated as
this expensive drug was administered only once to add-on therapy.
each baby tested by Nasr et al,19 and had no effect on Patients were examined at the study entry and every day. All
the length of hospital stay, nor did it improve patients were enrolled within 24 h of admission to the hospital. At
postinhalation clinical severity scores significantly. treatment time and 30 min after the beginning of each inhalation
session, the following parameters were measured and recorded
Hypertonic saline solution, by absorbing water using a clinical severity score described by Wang et al.29 This
from the submucosa, can theoretically reverse some scoring system assigns a number from 0 to 3 to each variable, with
of the submucosal and adventitial edema and de- increased severity receiving a higher score, as follows: respiratory
crease the thickness and dryness of the mucous rate, ⬍ 30 breaths/min, 31 to 45 breaths/min, 46 to 60 breaths/min,
plaques inside the bronchiolar lumen. It has been or ⬎ 60 breaths/min; wheezing, none, terminal expiratory or only
with stethoscopy, or inspiration and expiration without stetho-
shown to increase mucociliary transit time in various scope; retraction, none, intercostal only, tracheosternal, or severe
situations: in vitro, in normal subjects, in patients with nasal flaring; and general condition, 0 point for normal, and
with cystic fibrosis (CF), and in patients with sinona- 3 points for irritable, lethargic, or poor feeding. After random-
sal diseases.20 –28 ization, the intended therapy was begun. Anteroposterior and
The common practice in our institution is to treat lateral chest radiographs were obtained at the time of hospital
admission and 3 days afterwards. The radiograph assessment
hospitalized babies with acute bronchiolitis with score described by Nasr et al19,30 was utilized.
inhalation of epinephrine diluted in normal saline The combination of the therapeutic package (0.9% [normal]
solution. We hypothesized that simply substituting saline solution vs 3% saline solution) was not available to the
normal saline solution with hypertonic saline solution investigator or medical personnel. No detectable difference in
in the inhalation mixture for delivering epinephrine color, smell, or other physical properties existed between 0.9%
saline solution and 3% saline solution. The code was deposited
to these babies may improve clinical severity scores with the statistician.
after inhalations and decrease hospitalization stay. The decisions to discharge babies were made at morning
rounds by the attending physician, based on clinical grounds
alone. The attending physician was blinded to the combination of
the therapeutic package (0.9% saline solution vs 3% saline
Materials and Methods solution).
Antigen detection was performed using a commercial immu-
Devices nochromatographic assay (ImmunoCard STAT! RSV; Meridian
Diagnostics Europe; Bad Homburg, Germany). The sensitivity of
We utilized a nebulizer (Aeromist Nebulizer Set 61400; B&F the test is 80 to 90%.1
Medical by Allied; Toledo, OH) routinely available in our ward
connected to a source of pressurized oxygen, from the wall, set to
a flow rate of 5 L/min. This device has an output of 3 mL in Statistical Methods
6 min, an aerodynamic diameter mass medium of 0.5- to 4-␮m Two major outcomes of interest were considered: duration of
range, and a geometric SD of 96% of all liquid nebulized. The hospitalization, and the change in clinical severity score after the
nebulizers were administered until empty. 3% saline solution/0.9% saline solution (with epinephrine) inha-
lations each day. Other minor outcomes were pulse rate, satura-
Study Design tion on room air, radiograph assessment score, and number of
add-on treatments. Each variable was visually scanned for nor-
This was a randomized, double-blind, controlled trial. Signed malcy of distribution. Only age and hospitalization days were
informed consent was obtained from the parents of each child, highly skewed. These variables therefore were log transformed
and the human ethics committee (Helsinki) of our hospital prior to analysis. All continuous variables were examined using
approved the study. Fifty-three infants who were hospitalized in the paired or unpaired t test as appropriate. Noncontinuous
the Edith Wolfson Medical Center for acute viral bronchiolitis variables (sex and atopy) were examined using the ␹2 test. After
during the winter of 2000 –2001 were recruited. Inclusion criteria log transformation, days of illness at hospital admission were
were clinical presentation of viral bronchiolitis with temperatures regressed against days of hospitalization to examine the linear
⬎ 38°C that lead to hospitalization. Exclusion criteria were as correlation between them using the least-squares method. The
follows: cardiac disease, chronic respiratory disease, previous mean ⫾ SD expresses the central tendency of the data. The
wheezing episode, age ⬎ 12 months, saturation ⬍ 85% on room mean ⫾ SE was used in the graphs. To examine the change in
air, obtunded consciousness, and/or progressive respiratory fail- clinical severity score after the inhalations, paired t tests were
ure requiring mechanical ventilation. carried out for each day in each treatment group separately. For
The patients were recruited sequentially and were randomized this analysis, a p value ⬍ 0.006 for two-tailed t test was consid-
in a double-blind fashion. However, eight potentially eligible ered significant due to multiple comparisons. Otherwise,
patients were excluded because their parents did not agree to p ⬍ 0.05 was considered statistically significant.

482 Clinical Investigations

 Results The clinical severity scores at baseline were
8.08 ⫾ 1.3 in group 1 and 8.29 ⫾ 1.3 in group 2
Fifty-three previously healthy infants with viral (p ⫽ NS). The percentage fall of the clinical severity
bronchiolitis were enrolled in the study between score after inhalation therapy was NS in group 1 on
December 2000 and March 2001. Their mean age the first, second, and third days after hospital admis-
was 2.9 ⫾ 2.1 months (range, 0.5 to 12 months). One sion (3.5%, 2%, and 4%, respectively). In group 2,
patient from group 1 was excluded from the analysis significant differences were observed on each of the
because of deterioration immediately after the first first 3 days (7.3%, 8.9%, and 10%, respectively;
treatment inhalation, which required mechanical p ⬍ 0.001) [Fig 2]. Also, these falls in clinical scores
ventilation; this deterioration was attributed to inha- differed significantly between the two groups on
lation of epinephrine. Of the 52 infants who took each of these days.
part in the study analysis, 25 infants were enrolled to The radiograph assessment scores at baseline were
receive epinephrine, 1.5 mL (1.5 mg), in 4 mL of 0.9% 3.9 ⫾ 2.32 in group 1 and 3.52 ⫾ 2.49 in group 2
saline solution as a wet nebulized aerosol (group 1), and (p ⫽ NS). The second radiograph assessment score
27 infants received epinephrine, 1.5 mL (1.5 mg), in improved in both groups (3.43 ⫾ 2.35 in group 1 and
4 mL of 3% saline solution administered in the same 3.38 ⫾ 2.28 in group 2), but it did not reach signif-
manner as above (group 2). The two groups had icance. There was no significant difference between
similar clinical characteristics and variables at base- the groups.
line. Using the immunochromatographic assays, 47 No adverse effects were observed. Pulse rate and
of 52 infants (87%) were RSV positive. The positive the room air saturation of oxyhemoglobin did not
rate for RSV in group 2 was 23 of 27 infants (85%) differ at any day between the two groups. In the first
and in group 1 was 22 of 25 infants (88%), p ⫽ not 3 days, there was a trend for more add-on inhalation
significant (NS). therapy needed per day for group 1 (1.2 ⫾ 0.9 as
The mean duration of hospitalization was compared to 0.9 ⫾ 0.7 for group 2). However, this
3.5 ⫾ 1.6 days for the whole population, and it did not reach significance according to the two-tailed
differed significantly between the two groups: test (p ⫽ 0.1; p ⫽ 0.05 for one-tailed test). Analysis
4 ⫾ 1.9 days in group 1 and 3 ⫾ 1.2 days in group 2 of intention to treat showed that two patients in
(Table 1) [p ⬍ 0.05]. Days of illness prior to hospital group 1 and no patients in group 2 required mechan-
admission (as obtained by history taking) were ical ventilation. However, as mentioned above, one
3 ⫾ 1.6 days in group 1 vs 3.9 ⫾ 2.9 days in group 2. of these patients was excluded from the analysis.
This could theoretically affect the hospital stay.
Therefore, after log transformation, days of illness at
hospital admission were regressed against days of
hospitalization to examine the linear association be- Discussion
tween them using the least-squares method. This Our study indicates that by substituting normal
association was not significant (r ⫽ 0.14, p ⫽ 0.3) saline solution with hypertonic saline in the inhala-
and indicated that variation in days of illness at tion mixture for delivering epinephrine to hospital-
hospital admission accounted for only 2% (r2) of the ized infants with viral bronchiolitis, the in-hospital
variation in days of hospitalization. The percentage stay was reduced by 25%, from 4 days in the 0.9%
of infants remaining in the hospital each day for each saline solution group (group 1) to 3 days in the 3%
group is shown in Figure 1. saline solution group (group 2). This possible effect
could bear an important economic and clinical im-
Table 1—Baseline Clinical Characteristics pact worldwide; in the United States, ⬎ 100,000
0.9% Saline 3% Saline children are hospitalized annually at a cost of $300
Solution Solution million.1–3 Decreasing this burden by almost 25%
(Group 1; (Group 2; could theoretically save nearly $75 million annually
Characteristics n ⫽ 25) n ⫽ 27) p Value in the United States alone.
Age, mo 2.6 ⫾ 1.9 3 ⫾ 1.2 NS This study demonstrated a significantly better
Female/male gender 9/15 12/15 NS improvement in clinical severity score after epineph-
Baseline clinical severity 8.08 ⫾ 1.3 8.29 ⫾ 1.35 NS rine inhalation in hypertonic 3% saline solution as
scores
Baseline radiograph 3.9 ⫾ 2.32 3.52 ⫾ 2.49 NS
compared to epinephrine in 0.9% saline solution
scores (Fig 2). The clinical severity scores before inhalation
Days of illness at 3 ⫾ 1.6 3.9 ⫾ 2.9 NS decreased progressively in a significant way each day
hospital admission, until discharge. No significant differences were dem-
No. onstrated for the preinhalation clinical severity
Baseline saturation, % 94.7 ⫾ 3.3 93.8 ⫾ 3.2 NS
scores between the two groups (Fig 2). One possible

www.chestjournal.org CHEST / 123 / 2 / FEBRUARY, 2003 483

 Figure 1. The percentage of infants remaining in the hospital each day for each group.

explanation to this observation points to lower per- saline solution on the respiratory epithelium and the
centage of infants remaining hospitalized in the 3% mucociliary transport. Hypertonic saline solution has
saline solution group as compared to the 0.9% saline been shown to enhance mucociliary clearance in
solution group. As more infants with lower (better) vivo.27 Moreover, hypertonic saline solution had a
clinical severity scores were discharged from group 2 greater effect on mucus clearability in vitro than
than from group 1, the average clinical severity score deoxyribonuclease.27 Tomooka et al28 suggested four
of group 2, now including relatively more sick pa- mechanisms for the favorable effect of hypertonic
tients, would resemble the clinical severity score of saline solution in a study of patients suffering from
group 1 (Fig 1). Another possible explanation is that sinonasal diseases: (1) decreasing mucosal edema,
the acute effect of the hypertonic saline solution on (2) decreasing inflammatory mediators concentra-
symptoms (clinical severity) is shorter than the inter- tion, (3) mechanically clearing inspissated mucus,
vals between inhalations, so that the favorable de- and (4) improvement in overall mucociliary function
cline in clinical severity score after each inhalation and transport. A current review and meta-analysis of
does not persist overnight and is not apparent on the the literature, including seven high-quality selected
next morning before the next inhalation. The exact recent randomized controlled studies, concluded
duration of the effect of one hypertonic saline that in CF patients nebulized hypertonic saline
solution inhalation and therefore its continuing im- solution of ⱖ 3% concentration improves mucocili-
pact on clinical severity score is not known and ary clearance immediately after administration with
should be investigated further. Conceivably, more possible long-term beneficial effect.20 –27 The postu-
inhalations should be tried before assuming a maxi- lated molecular mechanism of the favorable effect of
mal dose/effect of hypertonic 3% saline solution hypertonic saline solution on the mucus membrane
inhalations in infants hospitalized with bronchiolitis. and mucus transport in patients with CF, according
In our patient population, the three-times-daily dose to these articles, was as follows: (1) hypertonic saline
and subsequent duration of effect proved to be suffi- solution induces an osmotic flow of water into mucus
cient to shorten hospital stay significantly (Fig 1). layer, rehydrating secretions, and thereby improving
The precise pathophysiologic mechanism of hy- mucus rheology24; (2) hypertonic saline solution
pertonic saline solution action specifically in bron- breaks the ionic bonds within the mucus gel, which
chiolitis has not been investigated in this study. could reduce the degree of cross-linking and entan-
However, some mechanisms have been studied and glement and lower viscosity and elasticity20; and
proposed for the favorable action of hypertonic (3) hypertonic saline solution increases the ionic

484 Clinical Investigations

 Figure 2. The clinical severity scores in group 1 and group 2. The fall of the clinical score after the
inhalation therapy was NS in group 1 on the first, second, and third days after hospital admission. In
group 2, significant differences were observed on each of the first 3 days (p ⬍ 0.001). Also, these falls
in clinical scores differed significantly between the two groups on each of these days. IN,
INH ⫽ inhalation.

concentration of the mucus and causes a conforma- be documented in viral bronchiolitis in animal or in
tional change by shielding the negative charges and vitro studies. It is possible that in our study, an
thereby reducing repulsion. This would result in a improvement in mucociliary transport and a better
more compact mucus macromolecule that would elimination of intracellular debris may have reduced
allow more effective clearance.24 However, as rheo- viral load and decreased ongoing inflammation
logic properties of CF sputum differ from that of within the airways. This might have reduced oppor-
RSV bronchiolitis, any direct extrapolation of these tunity for secondary bacterial overgrowth and
mechanisms to infants with bronchiolitis should be thereby may contribute to the favorable effect of
taken with caution, until examined specifically in decreasing postinhalation therapy clinical severity
RSV bronchiolitis. Interestingly, it has been shown in score and shortening hospital stay as noted (Figs 1, 2).
different models that using much higher concentra- Finally, hypertonic saline solution inhalation can
tion than we used, 7.5% hypertonic saline solution cause sputum induction and cough, which can help
can potentially reduce lung damage by suppressing to clear the sputum outside of the bronchi and thus
neutrophil activation.31–33 improve airway function in infants with bronchiolitis.
Pathophysiologically, viral bronchiolitis is an infec- Our patient population included only hospitalized
tious inflammation of the whole respiratory mucosal infants ⬍ 12 months old (mean age, 2.9 ⫾ 2.1
epithelium, although more pronounced in small months). However, the extent of RSV bronchiolitis is
bronchioles.34 This leads to tissue edema and mucus much wider. Virtually all children become infected
production resulting in thick mucus plaques within with RSV within 2 years after birth, and the histo-
the airway lumen and increase in intraluminal DNA logic evidence of recovery reveals that complete
concentration due to lysis of inflammatory and restoration of ciliated epithelial cells requires 4 to 8
sloughed respiratory epithelial cells.19,34 Taking this weeks in correlation with the common clinical find-
into consideration, the exact contribution and impor- ings of prolonged cough, wheezing, and altered
tance of each of these possible mechanisms waits to pulmonary function.1,35

www.chestjournal.org CHEST / 123 / 2 / FEBRUARY, 2003 485

Safety Issues 8 Moler FW, Steinhart CM, Ohmit SE, et al. Effectiveness of
ribavirin in otherwise well infants with respiratory syncytial
We used a relatively low concentration of 3% virus-associated respiratory failure. J Pediatr 1996; 128:442–
hypertonic saline solution in order to decrease the 448
possible negative effect of a higher concentrations 9 Guerguerian A-M, Gauthier M, Lebel MH, et al. Ribavirin in
(⬎ 7% saline solution) on the ciliary beat frequency ventilated respiratory syncytial virus bronchiolitis: a random-
ized, placebo-controlled trial. Am J Respir Crit Care Med
and to decrease risk of bronchospasm.36 We always 1999; 160:829 – 834
administered hypertonic saline solution in conjunc- 10 Wheeler JG, Wofford J, Turner RB. Historical cohort evalu-
tion with epinephrine solution in order to avoid any ation of ribavirin efficacy in respiratory syncytial virus infec-
possible bronchoconstriction effect. In our study, we tion. Pediatr Infect Dis J 1993; 12:209 –213
found no such detrimental effect. This is in concor- 11 Van Woensel JB, Wolfs TF, van Aalderen WM, et al.
dance with excellent safety profile reported by Wark Randomized double blind placebo controlled trial of pred-
and McDonald20 and others,21–26 who found no nisolone in children admitted to hospital with respiratory
syncytial virus bronchiolitis. Thorax 1997; 52:634 – 637
reports of bronchospasm in 143 reviewed patients 12 Cade A, Brownlee KG, Conway SP, et al. Randomised
with relatively severe CF treated with hypertonic placebo controlled trial of nebulised corticosteroids in acute
saline solution inhalations. They attributed this reas- respiratory syncytial viral bronchiolitis. Arch Dis Child 2000;
suring observation to the cotreatment of the hyper- 82:126 –130
tonic saline solution inhalations with ␤-agonists. 13 Menon K, Sutcliffe T, Klassen TP. A randomized trial
comparing the efficacy of epinephrine with salbutamol in the
treatment of acute bronchiolitis. J Pediatr 1995; 126:1004 –
Conclusion 1007
14 Dobson JV, Stephens-Groff SM, McMahon SR, et al. The use
On the basis of a faster rate of discharge from the of albutarol in hospitalized infants with bronchiolitis. Pediat-
hospital and a significant posttreatment improve- rics 1998; 101:361–368
15 Barr FE, Patel NR, Newth, CJL. The pharmacologic mech-
ment in clinical severity score, we conclude that 3%
anism by which inhaled epinephrine reduces airway obstruc-
saline solution/1.5 mg epinephrine is more effective tion in respiratory syncytial virus associated bronchiolitis.
than normal saline solution/epinephrine in a nonse- J Pediatr 2000; 136:699 –700
verely ill infant population hospitalized with acute 16 Bertrand P, Aranibar H, Castro E, et al. Efficacy of nebulized
bronchiolitis. Additionally, there was no difference in epinephrine vs salbutamol in hospitalized infants with bron-
safety profile. chiolitis. Pediatr Pulmonol 2001; 31:284 –288
17 Flores G, Horwitz RI. Efficacy of ␤2-agonists in bronchiolitis:
ACKNOWLEDGMENT: Mona Boaz, MSc, biostatistician of the a reappraisal and meta-analysis. Pediatrics 1997; 100:233–239
Edith Wolfson Medical Center, Holon, advised on statistics. 18 Welliver JR, Welliver RC. Bronchiolitis. Pediatr Rev 1993;
14:134 –139
19 Nasr SZ, Strouse PJ, Soskoline E, et al. Efficacy of recombi-
References nant human deoxyribonuclease I in the hospital management
1 Hall CB. Respiratory syncytial virus and parainfluenza virus. of respiratory syncytial virus bronchiolitis. Chest 2001; 120:
N Engl J Med 2001; 344:1917–1928 203–208
2 Glezen WT, Greenberg SB, Atmar RL, et al. Impact of 20 Wark PAB, McDonald V. Nebulised hypertonic saline for
respiratory virus infections on persons with chronic underly- cystic fibrosis (Cochrane Review). In: The Cochrane Library,
ing conditions. JAMA 2000; 283:499 –505 2, 2001. Oxford, UK: Update Software
3 The prospects for immunizing against respiratory syncytial 21 Eng PA, Morton J, Douglass JA, et al. Short-term efficacy of
virus. In: Institute of Medicine Committee on issues and ultrasonically nebulized hypertonic saline in cystic fibrosis.
priorities for new vaccine development. New vaccine devel- Pediatr Pulmonol 1996; 21:77– 83
opment: establishing priorities. Vol 2. Diseases of importance 22 Riedler J, Reade T, Button B, et al. Inhaled hypertonic saline
in developing countries. Washington, DC: National Academy increases sputum expectoration in cystic fibrosis. J Paediatr
Press, 1986; 299 –308 Child Health 1996; 32:48 –50
4 Langley JM, Wang EEL, Law BJ, et al. Economic evaluation 23 Robinson M, Regnis JA, Bailey DL, et al. Effect of hypertonic
of respiratory syncytial virus infection in Canadian children: a saline, amiloride, and cough on mucociliary clearance in
Pediatric Investigators Collaborative Network on Infections patients with cystic fibrosis. Am J Respir Crit Care Med 1996;
in Canada (PICNIC) study. J Pediatr 1997; 131:113–117 153:1503–1509
5 Darville T, Yamauchi T. Respiratory syncytial virus. Pediatr 24 Robinson M, Hemming AL, Regnis JA, et al. Effect of
Rev 1998; 19:55– 61 increasing doses of hypertonic saline on mucociliary clearance
6 Smith DW, Frankel LR, Mathers LH, et al. A controlled trial in patients with cystic fibrosis. Thorax 1997; 52:900 –903
of aerosolized ribavirin in infants receiving mechanical ven- 25 Robinson M, Daviskas E, Eberl S, et al. The effect of inhaled
tilation for severe respiratory syncytial virus infection. N Engl mannitol on bronchial mucus clearance in cystic fibrosis
J Med 1991; 325:24 –29 patients: a pilot study. Eur Respir J 1999; 14:678 – 685
7 Meert KL, Sarniak AP, Gelmini MJ, et al. Aerosolized 26 King M, Dasgupta B, Tomkiewicz RP, et al. Rheology of
ribavirin in mechanically ventilated children with respiratory cystic fibrosis sputum after in vitro treatment with hypertonic
syncytial virus lower respiratory tract disease: a prospective saline alone and in combination with recombinant human
double-blind, randomized trial. Crit Care Med 1994; 22:566 – deoxyribonuclease I. Am J Respir Crit Care Med 1997;
572 156:173–177

486 Clinical Investigations

27 Dasgupta B, Tomkiewicz RP, Boyd WA, et al. Combined effect phil activation after hemorrhagic shock. Shock 1998; 9:
of hypertonic saline and rhDNase on cystic fibrosis sputum in 164 –170
vitro. Pediatr Pulmonol 1995; 20(Suppl 12):A201–A236 32 Angle N, Hoyt DB, Cabello-Passini R, et al. Hypertonic saline
28 Tomooka LT, Murphy C, Davidson TM. Clinical study and resuscitation reduces neutrophil margination by suppressing
literature review of nasal irrigation. Laryngoscope 2000; neutrophil L selectin expression. J Trauma 1998; 45:7–12;
110:1189 –1193 discussion 12–13
29 Wang EEL, Milner RA, Navas L, et al. Observer agreement 33 Murao Y, Hoyt DB, Loomis W, et al. Does the timing of
for respiratory signs and oximetry in infants hospitalized with hypertonic saline resuscitation affect its potential to prevent
lower respiratory infections. Am Rev Respir Dis 1992; 145: lung damage? Shock 2000; 14:18 –23
106 –109 34 Spencer H. Pathology of the lung. Oxford, UK: Pergamon
30 Nasr SZ, Kuhns LR, Brown RW, et al. Use of computerized Press, 1985; 131–165
tomography and chest X-rays in evaluating efficacy of aero- 35 Swingler GH, Hussy GD, Zwarenstein M. Duration of illness
solized recombinant human DNase in cystic fibrosis patients in ambulatory children diagnosed with bronchiolitis. Arch
younger than age 5 years: a preliminary study. Pediatr Pediatr Adolesc Med 2000; 154:997–1000
Pulmonol 2001; 31:377–382 36 Boek WM, Keles N, Graamans K, et al. Physiologic and
31 Angle N, Hoyt DB, Coimbra R, et al. Hypertonic saline hypertonic saline solutions impair ciliary activity in vitro.
resuscitation diminishes lung injury by suppressing neutro- Laryngoscope 1999; 109:396 –399

www.chestjournal.org CHEST / 123 / 2 / FEBRUARY, 2003 487