Thoracoabdominal Asynchrony in Acute Upper Airway Obstruction in Small Children 1 , 2

YAKOV SIVAN, TIMOTHY W. DEAKERS, and CHRISTOPHER J. L. NEWTH

Introduction

The assessment of the severity and response to therapy of acute upper airway obstruction (UAO) in small children relies on vague and subjective parameters. Although several clinical and radiologic scoring systems havebeen suggested, they have been found to be unreliable and none have become standard (1, 2). Even such objective measurements as arterial blood gases do not correlate reliably with clinical findings because arterial hypoxemia may occur very early and hypercarbia occurs only late in the course of acute UAO when urgent intervention is already needed (3,4). Obtaining such measurements may agitate the child who is already in respiratory distress. Hence, pediatric departments, emergencyrooms, and pediatric intensive care units are all in need of techniques that will enable objective noninvasive serial assessment of the severity of UAO in small children and their response to therapy. Because of the highly compliant chest wall in infants and young children, much of the ventilatory effort during UAO is wasted by "sucking in ribs" rather than "sucking in air" during forced inspiration. This results in the asynchronous motion between the rib cage (RC) and abdominal (AB) compartments commonly observed in small children during respiratory stress and may be clinically manifested as chest retractions. However, this phenomenon has generally been descriptive and there is no simple way to quantitate it. There is some evidence that as airway obstruction in small children becomes greater, the phase lag between the RC and AB motions (asynchrony) increases (5). This study was conducted to test the hypotheses that RC-AB asynchrony accompanies acute UAO in small children and that monitoring the RC-AB asynchrony can serve as a noninvasive, objective scoring method for the response of UAO to therapy in small children. 540

SUMMARY The assessment of the severity and response to therapy of acute upper airway obstruction (UAO) In small children relies on subjective parameters. Using a noncallbrated respiratory Inductence plethysmograph (RIP), we quantlteted the rib cage (RC) to abdominal (AB) asynchrony and the lag phase In chest well expansion by the phase angle from the RC versus AB signal curve. Phase angles were obtelned In 17children aged 1 to 50 months with acute UAOand 30 normal control subjects. The phase angle In UAO(16to 165°; mean = 83.6°) wassignificantly higher than In control subjects (3 to 25°; mean = 11.5°), P < 0.001. following 29 episodes of Inhalation treatment with 0.03 ml/kg of racemic epinephrine, the phase angle In the UAOgroup decreased to 7 to 1600(mean = 38.3; P = 0.001) as the shape of the RC versus AB loop became narrower. In response to the treatment, the clinical severity of UAOdacreased and the tidal breathing flow-volume loop Improved. A high association was observed between the phase angle and the degree of stridor (p < 0.005 Fisher's exact test), and In 90% (26 of 29) the changes In the phase angle and In the degree of stridor were In agreement. Weconclude that the RC-AB asynchrony In acute UAOcan be objectively quantlteted by phase-angle measurement from a noncallbrated RIP and Is thus sulteble for use In Infants and small children. The phase angle may be used to assess objectively the response of UAO to therapy. AM REV RESPIR DIS 1990; 142:540-544

Methods Study Population RC-AB asynchrony was assessed in 17 children during 29 episodes of acute VAG that required treatment with racemic epinephrine in the pediatric intensive care unit (PICV) and in 30 normal control subjects. The age ranged from 1 to 50 months (mean = 25), and body weight from 3.6 to 20 kg (mean = 12.1). The cause of acute VAG was acute infectious laryngotracheitis (croup) in 5 patients (8 episodes) and postextubation acute VAG in the other 12 (21 episodes). The normal subjects were age-matched hospitalized children without respiratory problems (agel to 50 months, mean = 25; body weight 3.1 to 20 kg, mean = 10.7).

In each patient of the VAG group, we measured the heart rate and respiratory rate and also transcutaneous Pco, and pulse oximeter saturation whenever these instruments were applied for clinical management reasons (17 of 29 and 22 of 29 episodes, respectively).The degree of stridor was assessed using a fourpoint scoring system (table 1). In 14episodes of stridor (8 patients) tidal breathing flow volume (TBFV) loops were also obtained using the Sensormedics 2600 pulmonary function unit (Anaheim, CA) and analyzed for tidal volume, midtidal and peak tidal inspiratory flows (MTIF and PTIF, respectively), and midtidal expiratory flow/midtidal inspiratory flow (MTEF/MTIF) (6). All these variables were measured before and after inhalation

treatment with 0.03 ml/kg of racemic epinephrine (RE) diluted in 2 ml normal saline. The diagnosis of VAG secondary to subglottic edema and the decision to give racemic epinephrine was made by the attending physician in the PICV, and not by the investigators. RC-AB asynchrony was measured during all these episodes before and after RE as well as in normal subjects. The study was approved by the committee of clinical research of our institution, and informed consent was obtained in each case.

Technique RC and AB motions were depicted from the output of a noncalibrated respiratory inductance plethysmograph (RIP; Respitraces, Ambulatory Monitoring Inc, Ardsley, NY) whose bands were placed at the levels of the nipples and upper abdomen. Runs of 12to 30 breaths

(Received in originalform November 14,1989and in revised form March 21, 1990) 1 From the Division of Pediatric Intensive Care, Children's Hospital of Los Angeles, University of Southern California School of Medicine, Los Angeles, California. • Correspondence and requests for reprints should be addressed to Christopher J. L. Newth, M.B., FRCP(C), Pediatric ICU Administration, Children's Hospital of Los Angeles, 4650 Sunset Boulevard, Los Angeles, CA 90027.

541

BREATHING ASYNCHRONY IN UPPER AlfNlAY OBSTRUCTION

TABLE 1

TABLE 2

ACUTE UAO SCORING SYSTEM

Stridor occurring only during agitation or crying Stridor at rest but no respiratory distress Stridor with distress, chest retractions, or nasal flaring Severe stridor as in score 3 but with cyanosis

1 2 3 4

Phase Angle (0)

Mean Median SEM Range

using a sine-wave signal generator at fivepreset phase shifts (0, 45, 90, 135, and 180°). The reproducibility ofthe technique in patients was evaluated by the coefficient of variation of the 12 to 30 breaths in each of 38 different runs from the first 10patients (6 patients with UAO and 4 control subjects) enrolled in the study (28 runs in the UAO group and 10 in normal children). In patients with UAO, two to three runs and phase-angle calculations were obtained before as well as after treatment with RE. In normal children, runs and phase-angle measurements were obtained every 15 to 30 min over 2 to 6 h (total of 5 to 10runs per normal subject). In all patients, only breaths that were obtained when the patient was still (either asleep or awake) were included in the calculations.

of RC and AB signals versus time were displayed in real time on the screen of a personal computer. Using specifically designed software, RC-AB asynchrony was automatically calculated and displayed for each of these breaths, and the mean RC-ABasynchrony for the run was calculated. RC-AB asynchrony was obtained from the RC to AB display by the phase angle e using the principles employed by Agostoni and Mognoni (7) and the analysis of phase shifts of sinusoid electrical waves of equal frequency (Lissajous figures (8). Phase angle was thus calculated according to the equation sin e = m/s, where e is the phase angle; m is the distance between the intercepts of the RC-AB loop on a line drawn parallel to the X axis, which is placed at half the distance between the maximal and minimal RC excursions; and s is the maximal AB excursion (figure 1). Normally, the RC and AB expand and decrease in synchrony, producing a closed or very narrow loop on the XY plot, phase angle = 0° (figure lA). However, when these compartments move in opposite directions, as when the RC decreases while the AB expands, such as in severeUAO,a closed or very narrow loop is also created but with a negative slope (figure IB). This is defined as paradoxical breathing, phase angle = 180°. Between these two extremes,there are many possible degrees of asynchrony in which the RC lags behind the AB (figure Ie). The accuracy ofthe hardware and software was verified before the beginning ofthe study

A

Results

The reproducibility of phase-angle measurements in patients from consecutive breaths during a run was 1.2 to 14070 (mean = 9.7%), allowing mean phaseangle definition as the average of phaseangle measurements for the same run. In patients without respiratory illness, e ranged from 3 to 25° with a mean of 11.5° and standard error of the mean (SEM) = 0.9. In the UAO group, e was significantlyhigher (P < 0.001) and ranged from 16 to 165° with a mean of 83.6°

C

B RC

RC

RC

50%

RC-AB ASYNCHRONY CALCULATED AS THE PHASE ANGLE IN PATIENTS WITH UAO AND NORMAL CONTROLS (p < 0.001)

Description

Score

~

50%

I

s

&=0

I

ABO

synchronous 0

8

50%

s

I

ABO

asynchronous &= goO sl"&=

i~ I

s

ABO

perllCloxlcel

&= 180

0

m

s

Fig. 1. Phase-angle measurement from RC-AB curve. (A). RC and AB expand and decrease in synchrony. (B) RC and AB move in opposite directions throughout the respiratory cycle. (C) RC lags behind the AS during half the inspiratory and half the expiratory cycle.

Normal SUbjects

UAO

11.5 11.5

83.6 65.0 8.2 16-165

0.9 3-25

(table 2). An example of phase-angle measurement from a patient with UAO versus a normal subject is presented in figure 2. In response to inhalation treatment with racemic epinephrine, e decreased significantly from a mean ± SEMof83.6 ± 8.2° to 38.3 ± 7.5° (range, 7 to 160°; p = 0.001, paired t test, figures 3A and 4). The phase-angle posttreatment, however, remained above that of normal subjects (P < 0.05, unpaired t test). The degree of stridor improved after treatment with racemic epinephrine from a mean score of 2.8 to 1.7 (p = 0.001; Wilcoxon signed rank test). On four occasions neither the phase angle nor the degree of stridor changed. In one patient the phase angle increased from 115 to 130° but both the degree of stridor (3 points) and the TBFV measurements remained unchanged. This wasthe fourth consecutive episode that required racemic epinephrine treatment (in 6 h) for this patient, who had demonstrated good response to the previous three treatments. In another patient the phase angle (45°) and the degree of stridor (3 points) remained unchanged despite racemic epinephrine (TBFV curves were not obtained). A third patient did not respond to two treatments. The first was given when the degree of stridor was 3 and phase angle 105°, and the second when the degree of stridor was 3 points and the phase angle 160°. In both episodes neither the phase angles nor the degree of stridor changed. TBFV curves obtained during the second of these two episodes did not show any significant change. The patient was intubated after the second treatment failure (clinical decision). Of the 29 patients, 22 (76%) showed improvement in both the degreeof stridor and the phase angle (responders); in 4 patients neither of these improved (nonresponders); in 1 the phase angle improved but the degree of stridor stayed unchanged; and in 2 the phase angle did

542

SIVAN, DEAKEAS, AND NEWTH

Brnt hs I 4 to 4 I

1,21-

r1.11~

( .....__

:LHL ~I .rn-

-J l - .ti 11.I4 1.14

II"

I

1,2~

Fig. 2. RC versus AS display from a patient with UAO (left) and a normal control (right). Note the difference in the shape of the curve and the direction of the slope.

-.IlII-

-.17

-,n '. 21 ' .11 1:12 1:1l 1.2~ Udutn

1.3~

Phase : I .i

PhMoe : 141.1

UAO

..:

I1S

~

:!!.

t 2S

....'" CI

" Xl

. .if'" Z

III

NOR MAL CONTROL

t"

~

"

.

I~=~~~~~~~~~i;

2S

L - - - , , -PRE·RE

. ~

~ :!!.

...'" CI

.

-

-

-

-

+2 so _ I - m.ln

- - - - r --

-

-

-2 so

POST·R E

m

lO.

Fig. 3. (top) Phase angle in patients (clinical responders and non-responders) with UAO before and after racemic epinephrine. The shadowed area is the normal range (mean ± 2 SO). (bottom) Results in 24 patients who responded clinically,

t2S

tOO

Z

.'" ".. III

~

:I: ll.

2S

tI:::~~~~~~~~i=

L-_,PRE·RE

+2 so normaI _ml.n

~---- .2

so

POST-RE

not change but the degree of stridor improved (clinical responders). Thus a total of 24 of 29 children showed a positive clinical response, and in 26 of 29 (90010) the change in the phase angle agreed with the change in the degree of stridor (p < 0.005, two-tailed Fisher's exact test). Assuming that the random chance of agreement is generally 0.5 (50%), then the critical ratio test showed that our degree of agreement differed from chance at p < 0.003. Excluding the 5 patients who did not respond clinically, the phase angle decreased from a mean ± SEM of 78.2 ± 8.9 to 27.8 ± 5.5° (figure 3B). The same analysis for phase angle and change in tidal volume showed 71.4% agreement (10 of 14). The following variables also improved significantly (paired t test) after treatment (mean ± SEM): tidal volume from 4.4 ± 0.4 to 5.7 ± 0.5 mIlkg (p = 0.0001), MTIF from 127.9 ± 17.1 to 158.2 ± 20.1

mIls(p