The "88% Saturation Test": A Simple Lung Function Test for Young Children

(1)

The

“88%

Saturation

Test”:

A Simple

Lung

Function

Test

for

Young

Children

Carol L. Wagner, MD*; John C. Brooks, MDX; Susan E. Richter, CRRT;

Kimberly Pratt, RRT; and Dale L. Phelps, MD1J

ABSTRACT. Objective. The 88% saturation test

(88%-SAT) was developed as an alternative to standard

spirom-etry for those young children unable to perform standard

forced expiratory maneuvers. In adults, this test revealed

rapid desaturation in those persons with a history of

asthma when compared with healthy control subjects.

Similar findings in children were tested.

Setting. Tertiary care hospital.

Patients. Thirty-three former premature infants (28.3 ± 2.3 weeks gestation), aged 5 to 7 years, who were

participating in a follow-up study, were enrolled in this

study.

Design. The study compared the 88%-SAT with

stan-dard spirometry and respiratory health characteristics

as-certained through a parental questionnaire. The 88%-SAT

consists of continuous measurement of hemoglobin

satu-ration by pulse oximetry (Sao2) while the subject breathes

a nonhumidified 12% oxygen and nitrogen mixture for 10

minutes or until Sao2 decreases to 88%, whichever occurs

first. Abnormal 88%-SAT was defined as a decrease of

Sao2 to 88% within the 10-minute period, and abnormal

spirometry was defined using standardized values.

Results. Of the 20 children who successfully

corn-pleted both spirometry and the 88%-SAT, 10 had normal

spirornetry results and did not desaturate to 88%, and 5

had abnormal spirometry and 88%-SAT results. Four

chil-dren did not desaturate during the 88%-SAT, but had

ab-normal spirornetry results, and one child had abnormal

88%-SAT results, but normal spirometry.

Ten additional children completed the 88%-SAT, but

not standard spirometry. Three children were unable to

complete either test. Of those 30 children tested, 7 (23%)

had a history of reactive airways disease, and all 7 had

abnormal 88%-SAT results. The 88%-SAT had greater

sen-sitivity (100% vs 75%) and specificity (87% vs 63%) than

spirometry in identifying children with known reactive

airways disease.

The mean McCarthy general cognitive index (GCI) of

the group performing both spirometry and the 88%-SAT

(n = 20) achieved a mean (± SD) GCI of 96.2 ± 16.7, and the

group (n = 30) that completed the 88%-SAT had a mean (±

SD) GCI of 75.2 ± 26.3

(

P < .012). The 10 children able

to perform only the 88%-SAT had a mean CCI (± SD) of

From the *Depa.tnent of Pediatrics, Division of Neonatology, Children’s Hospital, Medical University of South Carolina, Charleston, SC; Depart-ment of Pediatrics, Divisions of #{182}Neonatology and jfulmonology,

Univer-sity of Rochester School of Medicine, Strong Children’s Research Center, Rochester, NY; and the §Department of Pediatrics, Division of Neonatology, Winthrop University Hospital, Mineola, NY.

Received for publication Mar 25, 1993; accepted Jun 18, 1993.

Reprint requests to (C.L.W.) Medical University of South Carolina,

Depart-ment of Pediatrics, Division of Neonatology, 171 Ashley Avenue, Charles-ton, SC 29425.

72.8 ± 26.9, and the 3 children unable to perform either test

had a mean GCI (± SD) of 63 ± 11.

Conclusions. Our data suggest that the 88%-SAT may be more effective than spirometry for identifying reactive

airways disease in young, uncooperative, or developmen-tally delayed children. The dry air of the hypoxic inspired gas may function as an airway challenge, leading to

de-creased oxygenation in patients with reactive airways.

Pediatrics 1994;93:63-67; pulmonaryfunction testing,

bron-chial challenge.

ABBREVIATIONS. 88%-SAT, 88% saturation test; Sao2,

hemoglo-bin oxygen saturation; FEFmax, maximum forced expiratory flow;

RAD, reactive airways disease; GCI, general cognitive index; Fio2,

fraction of inspired oxygen.

Options for quantitative pulmonary function

as-sessment of young children are limited by the level of

cognitive function and developmental maturity

nec-essary to adequately perform standard spirometry.1’2

The 88% saturation test (88%-SAT) was developed as

an alternative to standard spirometry in young

pa-tients with developmental disabilities and lung

ab-normalities who are unable to complete standard

forced expiratory maneuvers. Preliminary testing of

adult volunteers breathing Fio2 0.12 revealed rapid

hemoglobin desaturation in those with a history of

asthma when compared with healthy control subjects

(unpublished data).

The purpose of this study was to compare the

88%-SAT test with spirometry in young children with

re-gard to completion success rate, abnormality, and

questionnaire responses regarding respiratory health.

Subjects

METHODS

After approval from the University of Rochester’s Research

Review Board, parental signed consent, and the testing of 14

healthy adult volunteers, 33 children with a history of prematurity

and/or respiratory distress syndrome as neonates, who were

con-secutive participants in a neonatal follow-up study, were recruited

between November 1989 and July 1991 to participate in this test

evaluation. The children were 5 to 7 years old at recruitment. The

mean birth weight (± SD) of the group was 1055 ± 317 g with a

mean gestational age of 28.3 ± 2.3 weeks (range = 25 to 34 weeks).

Children with a history of anemia (hematocrit <30) or congenital heart disease (shunting and admixing lesions, exclusive of patent

ductus arteriosus in premature infants, and valvular and great

vessel disease] were excluded from the study.

Measures

Spirometry was performed using the Medical Graphics model

1070 and standard procedures with the children seated.F5 The

tests included forced vital capacity, forced expiratory volume in I

at Viet Nam:AAP Sponsored on September 1, 2020

www.aappublications.org/news

(2)

RESULTS

Both spirometry and the 88%-SAT were attempted

in 33 children, aged 5 to 7 years (mean ± SD, 5.9 ±

0.7

years), with a history of prematurity and

respi-ratory distress syndrome. Other neonatal complica-tions included patent ductus arteriosus (medical do-sure in 5/33 [15%] or surgical closure in 3/33 [9%]),

pulmonary interstitial emphysema in 3/33 (9%),

pneumothorax in 4/33 (12%), and intraventricular

hemorrhage in 4/33 (12%). Eighteen (55%) infants

had been diagnosed with bronchopulmonary

dyspla-sia at 28 days of age, and three (9%) had a persistent

diagnosis of chronic lung disease at 40 weeks

post-conceptional age. Two infants (6%) had been

dis-charged to home receiving supplemental oxygen and

diuretic therapy.

Respiratory illnesses were frequent during the first year of life, with 9/33 (27%) requiring rehospitaliza-tion for bronchiolitis or pneumonia. Other neonatal

and general health characteristics are summarized in

Table I.No children were excluded because of anemia or congenital heart disease.

Seven of the 33 children had a history of RAD as

defined by history of asthma or chronic wheeze/

cough, and 4 of these 7 (57%) had a prior diagnosis of

bronchopulmonary dysplasia. Two of those 7

chil-dren had been hospitalized for an acute exacerbation of their RAD 2 and 3 years before testing, but had not required intensive care. At the time of testing, one

child was taking oral theophylline and nebulized

al-buterol treatments three times a day (Table 2, subject 3). The remaining six children had taken no

medica-tions for at least I year. There were no children in

whom other forms of chronic lung disease had been

diagnosed. No child had undergone anesthesia for a

surgical procedure during the 3 years before testing.

As shown in Table 2, 20 of the 33 children (61%,

subjects 1 through 20) were evaluated successfully

with both spirometry and the 88%-SAT. Only subject

9 had the pattern of spirometry abnormality expected with generalized airways obstruction. Eight subjects

9 (27)

7 (21) 4(12)

second, maximum forced expiratory flow (FEFmax), and the

av-erage forced expiratory flow rate over the middle half of

expira-tion. Normal values were defined using the data of Weng and

Levison3’4 and Zapletal et al. Pulmonary function results were

expressed as percent predicted based on height and gender of

child.

In a single session, the children underwent spirometry, and

then the 88%-SAT testing. Pulse oximetry was measured

continu-ously throughout the 88%-SAT test period (Nelcor model 200).

After a 1- to 2-minute equilibration period with the child seated and breathing quietly to achieve a stable baseline Sao2 and heart

rate, a noseclip was applied and the child began breathing

non-humidified 12% oxygen through a mouthpiece. The mouthpiece

was attached to a “T” piece with one-way valves (Airlife 001504)

at each of the other two outlets. The inspiratory limb of the “T”

piece was attached by tubing to a tank of 12% oxygen in nitrogen,

which was run at a flow rate of 8 to 16 L/min throughout the

hypoxic challenge. Oxygen concentration of inspired air was

monitored just upstream from the “T” piece by an oxygen

ana-lyzer (MiniOXI, Catalyst Research Corporation, Pittsburgh, PA).

The timing for the hypoxic exposure (maximum duration of 10

minutes) began when the child started to breathe through the

mouthpiece. Oximetry was validated by continuous heart rate

recording for comparison with the pulse rate detected by the

oximeter throughout the 88%-SAT testing procedure. Earlier

stud-ies with adult subjects had indicated that the Fio2 had to be

low-ered to 0.12 for the 88%-SAT to distinguish between normal

con-trol subjects and those with a history of asthma (unpublished

data). In the adult studies, if desaturation were to occur, it

oc-curred within 5 minutes of the initiation of hypoxic breathing. We

selected the maximum duration of hypoxia of 10 minutes for this

study to allow for greater subject variability in reaching a steady

state. Any hemoglobin desaturation to 88% by pulse oximetry

during the 10-minute period was considered abnormal. The limit

of 88% Say, was chosen empirically as the lowest “safe” level of

oxygenation.

Precautions were taken to insure that no subject suffered any respiratory, physical, or mental discomfort during the testing. The

hypoxic exposure was to be discontinued immediately if there was

shortness of breath, dizziness, tachypnea, wheezing, or signs of

respiratory decompensation, but this never occurred. Termination

of testing occurred when pulse oximetry reached 88% saturation

or after a maximum 10 minutes of breathing hypoxic gas,

which-ever came first. Subjects were monitored continuously for heart

rate and respiratory status via oximetry and physical assessment

throughout the 88%-SAT test. Monitoring of oximetry continued

for 2 minutes after Fio2 was returned to 0.21 to insure normaliza-tion of oximetry to >95% or previous baseline.

After pulmonary assessment, each child’s development was

assessed using the McCarthy Scales of Children’s Abilities.b A

health questionnaire, including specific questions regarding

res-piratory illnesses and hospitalizations, was administered to the

parent(s) in a standardized fashion via personal interview to

es-tablish their child’s past medical history. Reactive airways disease

(RAD) was defined by either a previous history of asthma,

includ-ing exercise-induced asthma, diagnosed and/or treated by a

phy-sician, consisting of acute, recurrent episodes of wheezing with

labored breathing, or chronic wheezing/coughing with a

pro-longed expiratory phase. An isolated episode of wheezing

associ-ated with bronchiolitis during infancy was not considered a

reli-able manifestation of RAD.

All medical history from the questionnaire was validated by

review of medical records from hospital and physicians’ offices.

The interviewer was unaware of pulmonary and cognitive

assess-ment results at the time of the interview. In addition, all examiners

were unaware of other test scores or the child’s past medical

history during pulmonary and cognitive testing.

Statistical Analysis

Data were expressed as the mean ± standard deviation where

appropriate. The Pearson test and Fisher’s Exact Test were

applied for analysis of discrete variables, and the Pearson corre-lation coefficient was applied to ascertain the association among

the 88%-SAT and subtests of spirometry. The primary outcome

measures of the 88%-SAT were time to 88% Sao2 or Sao2 at 10

minutes. All P values were based on two-tailed tests. Significance was set at a P value <.05.

TABLE 1. General Health Characteristics* of Study Group

(n = 33) (mean ± SD)

Health Characteristics n

Neonatal

Birth weight, g 1055 ± 317

Gestation, wk 28.3 ± 2.3

Gender, M/F 16/17

1-mm Apgar score 4.2 ± 2.0

5-mm Apgar score 7.3 ± 1.4

Respiratory distress syndrome, n (%) 33 (100)

Pneumothorax, n (%) 4 (12)

Pulmonary interstitial emphysema, n (%) 3 (9)

Duration of 02 therapy, d 34 ± 32

Duration of intermittent mandatory 30 ± 25

ventilation, d

Bronchopulmonary dysplasia, n (%), at 28 d 18 (55)

Other

Respiratory illness requiring hospitalization during first year, n (%)

Reactive airways disease, n (%)f Cerebral palsy, n (%)

* Obtained from medical records and physician(s)’ documentation

t Reactive airways disease: History of asthma or chronic wheezing and coughing with verification by a physician.

(3)

Subjects Age, y FEVI, % PlC, % FEFn%75%, #{176}‘ FEF, % 88%SAT RAD by Historyt

(ni 804 (ni 80) (nl 70) (ni 80)

I 6 141 88 107 107 Completed

2 6 122 104 123 101 0.5 min

3 6 86 76 77 66 3min

4 7 92 86 88 81 Completed

5 5 140 83 118 76 3 min

6 7 81 71 100 69 5 min

7 5 207 100 192 94 Completed

8 6 167 99 174 123 Completed

9 7 76 89 49 58 I .5 min

10 5 104 87 85 51 Completed

11 6 135 95 151 93 Completed

12 5 200 81 162 100 Completed

13 6 121 88 101 121 Completed

14 5 81 68 72 60 Completed

15 5 119 91 99 78 3 min

16 6 135 83 189 96 Completed

17 6 132 87 157 91 Completed

18 5 102 86 98 77 Completed

19 7 91 93 67 73 Completed

20 7 92 .. . . . . Completed

+ +

+

* 88% saturation test (88%-SAT): The time of hemoglobin desaturation to 88% is given in minutes. Those who completed the test had an

Sao2 >90% for the 10-minute test period. FEV1 = forced expiratory volume at I second; FVC, forced vital capacity; FEFn%The, forced

expiratory flow rate over the middle haff of expiration; FEF, maximum forced expiratory flow. Pulmonary function results are

expressed as percent predicted based on height and gender.

t History of reactive airways disease.

:1:

ni = normal.

Value in abnormal range.

TABLE 2. Comparison of Spirometry and 88%SAT* in Study Subjects

(3, 5, 6, 10, 14, 15, 18, 19) had the greatest abnormality

in FEFmax suggesting either a narrowing of the large

airway (eg, tracheal narrowing), or improper perfor-mance of the forced expiratory maneuvers. Four (sub-jects 2, 3, 6, 9) of the 20 who completed spirometry had a previous diagnosis of RAD.

Ten additional children completed the 88%-SAT

test who had been unable to adequately perform

spi-rometry maneuvers (Table 3, subjects 21 through 30).

Three children were unable to perform either test due

to difficulty in breathing through the mouthpiece with the noseclip in place. Thus, a total of 30/33 (91%) cooperated for the 88%-SAT. All pretest oximetries were 96%. In 10 of these children, Sao2 decreased to

88% after breathing Fio2 0.12 for 5 minutes or less

(range, 45 seconds to 5 minutes). The seven children

with a history of RAD desaturated to 88% at a mean

time (±SD) of 2.9 ± 1.1 minutes. With the exception

TABLE 3. 88%-SATe Results in Subjects

Spirometry Satisfactorily

Unable to Perform

Subjects Age, y 88%-SAT RAD by Historyt

21 7 Completed

22 5 3min +

23 6 3.5 mini +

24 6 1.3min +

25 6 Completed

26 6 4.5 mini

27 5 Completed

28 6 Completed

29 5 Completed

30 6 Completed

* 88% saturation test (88%-SAT): The time of hemoglobin

desatu-ration to 88% is given in minutes. Those who completed the test

had an Sao2 >90% for the 10-minute test period. I History of reactive airways disease.

I

Value in abnormal range.

of a history of RAD, there were no statistically sig-nificant differences between those who desaturated to

88% and those who did not, with respect to

gesta-tional age, birth weight, history of bronchopulmonary dysplasia, postnatal age, resting Sao2, or heart rate. The 20 children who completed 10 minutes of testing,

maintaining a Sao2 >88%, had a mean (± SD)

maxi-mum decrease in their Sao2 of 4.6 ± I .6% (range, 2 to 9% decrease). There were two patients (subjects 4 and

10) who had transient decreases in Sao2 from 98%

to 90% and then stabilized at Sao2 of 93% and 95%,

respectively.

Although resting heart rates of the 30 children

tested increased a mean (±SD) of 7.9 ± 4.3 beats per

minute (range, 4 to 14) during the test period, no

symptoms were reported by any of the children, so

testing was never discontinued because of patient

dis-tress. At no time during any testing session did the

Sao2 decrease to less than 88%. Room air oximetry and heart rates of all 30 children returned to baseline within 30 seconds after completion of testing.

Of the 20 children who underwent both spirometry

and the 88%-SAT, 10 (50%) had normal spirometry

results and did not desaturate to 88%. The mean

(±

SD) time for completion of the spirometry

maneu-vers by those 20 children was 42 ± 12 minutes, and

all the same children completed the 88%-SAT within

the designated 2-minute equilibration/b-minute

test/2-minute posttest recovery periods (<15

mm-utes). Five children (25%) had abnormal results of

both spirometry and 88%-SAT. Another child with a

history of partial upper airway obstruction secondary

to enlarged adenoids and tonsils (subject 10) had an

abnormal peak flow rate (FEFmax), but other

spirom-etry and 88%-SAT results were within the normal

(4)

range. Subject 14 did not desaturate during the 88%-SAT, but had abnormal spirometry results suggestive of restrictive airways disease. Subjects 18 and 19 had

abnormal FEF values, but otherwise normal

spirom-etry results and completed the 88%-SAT without a

decline in oximetry. One child, subject 20, who could

not perform a full forced expiratory maneuver had a

normal FEV1 and a normal 88%-SAT.

The positive predictive values of spirometry and

the 88%-SAT in correctly identifying children with a

history of RAD in this sample were 33% and 70%,

respectively (see Table 4); the negative predictive

val-ues of spirometry and the 88%-SAT were 9b % and

100%, respectively. Although the sensitivity of spi-rometry in accurately detecting a history of RAD was

75%, the sensitivity of the 88%-SAT was 100%. The

specificity was 63% for spirometry and 87% for the

88%-SAT. Correlation among the results of the

88%-SAT and spirometry tests within individual subjects was weak (r < .4).

The mean McCarthy general cognitive index (CCI)

was higher in the group of children who performed

both spirometry and the 88%-SAT than in the group

of children who could perform only the 88%-SAT.

Specifically, the group performing both tests (n = 20)

achieved a mean (±SD) CCI of 96.2 ± 16.7, whereas

the group (n = 30) that completed the 88%-SAT had

a mean (± SD) CCI of 75.2 ± 26.3 (j P < .012). The

bO children able to perform only the 88%-SAT had a

mean CCI (± SD) of 72.8 ± 26.9, whereas the 3

chil-dren who were unable to successfully complete either

spirometry or the 88%-SAT all had McCarthy GCIs of

<70 (mean ± SD: 63 ± 11). Subject 22 achieved a CCI

of 1 10, but because of motor impairments secondary to spastic quadriplegia, was unable to adequately

per-form spirometry maneuvers. History of RAD was not

associated with a lower CCI (P > .05).

DISCUSSION

This study was designed to assess the efficacy and safety of the 88%-SAT as an alternative to standard

spirometry in the objective evaluation of pulmonary

function of young children with developmental

dis-abilities. Based on preliminary evaluation of the

88%-SAT test in adults, we hypothesized that children

with a known history of RAD would have abnormal

88%-SAT results, and that these results would

corre-late with abnormal spirometry results. Although the

88%-SAT results did not correlate very well with

spi-rometry results, the 88%-SAT was superior to

spirom-etry in identifying those children with a history of

RAD.

TABLE 4. Test Comparison in the Identification of Children

With Reactive Airways Disease (RAD)

History_of RAD Total

(+) (-) Spirometry Abnormal Normal 3 I 6 10 9 11

Total 4 16 20

88% saturation test Abnormal Normal 7 0 3 20 10 20

Total 7 23 30

Historically, hypoxia challenge tests had been

de-veloped to better define the physiologic responses that occur during both acute and chronic hypoxia.714 Controversy exists in the literature about whether hy-poxia induces nonspecific bronchial reactivity in

asth-matic patients.74 Denjean et al showed nonspecific

bronchial hyperreactivity in subjects with a history of

asthma who breathed nonhumidified 13% oxygen as

part of an hypoxia challenge test.7 In a later study

using a sheep animal model, Denjean et a!8 could not

demonstrate altered pulmonary mechanics during

the hypoxia-induced state, but did show that

bron-chial responsiveness to methacholine was

signifi-cantly increased by hypoxia (nonhumidified F1o2

15%).

These findings have been replicated by others

us-ing sheep and dog animal models, demonstrating

that hypoxia (F1o2 < 15%) enhances a nonspecific

bronchial responsiveness.92 Various mechanisms of

action have been implicated. The data support the

concept that the hypoxia-induced increase in

bron-chial responsiveness is the result of a reflex

trig-gered by stimulation of carotid chemoreceptors,

im-plicating a centrally mediated mechanism as the

cause of the bronchial hyperresponsiveness.8 Others

have demonstrated mast cell degranulation during

hypoxia and a concomitant increase in the release

of leukotrienes.12

In contrast to human and animal data of Denjean8

and others712 Alberts et a113 did not demonstrate a

change in bronchial responsiveness when asthmatic

subjects breathed a humidified 15.5% oxygen

mix-ture. In another study, 117 infants at 9 months cor-rected age with a history of neonatal pulmonary

func-tion abnormality underwent a series of pulmonary

function tests that included a 17% humidified oxygen

exposure test. No group differences could be

dem-onstrated, although infants’ mean Sao2 decreased to

93% during the hypoxia challenge test.14 In summary,

among published studies on hypoxia and bronchial

reactivity, increases in bronchial reactivity were dem-onstrated only in those studies in which a nonhumi-difed oxygen concentration of <15.5% had been used. Potential differences between subjects may have gone

undetected with humidified F1o2 .b5.

During the 88%-SAT, each subject underwent a

hypoxic challenge in an attempt to elicit airways re-activity, thereby potentiating any differences in

pul-monary function between subjects. Subjects with a

history of RAD or bronchospasm rapidly

desatu-rated to 88%. These findings are consistent with the

previous findings of Denjean et al7’8 and Ahmed

and Marchette.11 We found that the nonhumidified

oxygen-nitrogen gas mixture created a hypoxic

en-vironment that may have induced subclinical

bron-choconstriction in susceptible subjects, resulting in

more marked hemoglobin desaturation than was

demonstrated in subjects with no history of RAD.

For reasons of safety, testing was terminated when

a subject’s Sao2 reached 88%. Had we prolonged the

88%-SAT test, some subjects may have developed

more obvious clinical signs of bronchospasm such

as wheezing or tachypnea. The 88%-SAT may serve,

then, as a bronchial challenge test that is sensitive in

(5)

detecting the specific subclass of pulmonary

abnor-mality referred to as RAD. It is not clear whether

the dry inspired air or the hypoxic inspired air was

the most important factor in unmasking the

pulmo-nary abnormality in the patient.

The 88%-SAT was administered to the 30 children

tested without complications. There were no patients

in whom testing was discontinued because of adverse signs or symptoms. The safety of this test in children with a history of congenital heart disease or anemia

cannot be assured because such children were

ex-cluded from this study.

The usefulness of the 88%-SAT in children with

cognitive disability was demonstrated by this study.

Although successful completion of spirometry was

associated with a higher level of cognitive function as

demonstrated by a higher mean McCarthy CCI, the

88%-SAT was less dependent on cognition, with

suc-cessful completion of the test by those with a lower

mean CCI. Thus, when compared with standard

spi-rometry, the 88%-SAT safely assessed lung function

in a greater number of children of varying cognitive abilities enrolled in a follow-up program of

prema-ture infants, and identified four children with

pulmonary abnormality who, because of cognitive

limitations, could not be identified via standard

spirometry.

Although the sensitivity and specificity of the

88%-SAT were higher than was found for spirometry, the

positive predictive value of both tests in correctly

identifying children with a known history of RAD in

this sample was low. There are two possible

expla-nations for the poor positive predictive value of the tests: (1) Neither is a good test of pulmonary abnor-mality (although the 88%-SAT is definitely better); and (2) the poor predictive value of both tests may be due to the lack of a sensitive and accurate “gold stan-dard” test of lung function in these children to which

the 88%-SAT and spirometry can be compared.

His-tory of RAD as a gold standard has significant limi-tations because it may not reflect the “true” disease state at the time of testing. In addition, Josephs et al point out that the relationship between nonspecific bronchial hyperreactivity (as measured by

methacho-line challenge tests) and asthma is complex. At any

given time, bronchial hyperreactivity is only one

mechanism contributing to the clinical expression of

the disease and, therefore, conclusions regarding the presence or absence of the disease based on bronchial challenge results must be made with caution.’5

In summary, these data from a small group of

pa-tients indicate that the 88%-SAT may be a safe and

simple alternative to spirometry as a bronchial

chal-lenge test for identifying RAD in the cognitively

im-mature or developmentally disabled patient

popula-tion. Further studies to determine the value of the

88%-SAT in younger children and in those with

spe-cific lung diseases will be necessary. The 88%-SAT

must be compared with other bronchial challenge

tests (eg, methacholine) before our hypothesis about

the mechanisms which it evaluates can be accepted.

ACKNOWLEDGEMENTS

This research was supported by a Specialized Center of

Re-search (SCOR) Grant HL-36543, and in part, by a General Clinical

Research Centers Grant RR000-44, both from the National

Insti-tutes of Health (D. L. P.). This research was also supported in part

by Maternal and Child Health Pediatric Pulmonary Center Grant

MCJ-369071 (J.G. B.).

The authors wish to thank Dr. William B. Pittard for his

thoughtful editorial assistance during the preparation of this

manuscript.

REFERENCES

I.Sly PD, Robertson CF. A review of pulmonary function testing in

chil-dren. JAsthma. 1990;27:137-147

2. Taussig LM. Maximal expiratory flows at functional residual capacity: a test of lung function for young children. Am Rev Respir Dis. 1977;116: 1031-1038

3. Weng T, Levison A. Pulmonary function in children with asthma at acute attack and symptom-free status. Am Rev Respir Dis. 1969;99: 719-728

4. Weng T, Levison A. Standards of pulmonary function in children. Am Rev Respir Dis. 1969;99:879-894

5. Zapletal A, Motoyama EK, Van De Woestijne KP, et al. Maximum

expiratory flow volume curves and airway conductance in children and

adolescents. IApp! Physio!. 1969;26:308

6. McCarthy D. McCarthy Scales of Children’s Abilities. New York:

Psycho-logical Corporation; 1972

7. Denjean A, Roux C, Herve P. et al. Mild isocapnic hypoxia enhances the bronchial response to methacholine in asthmatic subjects. Am Rev Respir Dis. 1988;138:789-793

8. Denjean A, Canet E, Praud JP, et al. Hypoxia-induced bronchial

respon-siveness in awake sheep: role of carotid chemoreceptors. Respir Physiol.

1991 ;83:201-210

9. Ahmed T, Oliver W, Frank BL. Hypoxic pulmonary vasoconstriction: role of mast cell degranulation. Am Rev Respir Dis. 1982;126:291-297

10. Ahmed T, Oliver W. Does slow-reacting substance of anaphylaxis me-diate hypoxic pulmonary vasoconstriction? Am Rev Respir Dis. 1983;127: 566-571

1 1. Ahmed T, Marchette B. Hypoxia enhances nonspecific bronchial reac-tivity. Am Rev Respir Dis. 1985;132:839-844

12. D’Brot J, Ahmed T. Hypoxia-induced enhancement of non-specific bronchial reactivity: role of leukotrienes. JApp! Physio!. 1988;651 :94-199 13. Alberts WM, Colice GC, Hammond MD, et al. Effect of mild hypoxemia

on bronchial responsiveness. Ann Allergy. 1990;65:189-193

14. HiFi Study Group. High-frequency oscillatory ventilation compared with conventional mechanical ventilation in the treatment of respiratory failure in preterm infants: assessment of pulmonary function at 9 months corrected age. IPediatr. 1990;116:933-941

15. Josephs LK, Gregg I, Mullee MA, Holgate ST. Nonspecific bronchial

reactivity and its relationship to the clinical expression of asthma: a longitudinal study. Am Rev Respir Dis. 1989;140:350-357

(6)

1994;93;63

Pediatrics

Carol L. Wagner, John G. Brooks, Kimberly Pratt, Susan E. Richter and Dale L. Phelps