Hypothalamic-Pituitary-Adrenal Axis Function in Very Low Birth Weight Infants Treated With Dexamethasone

(1)

Hypothalamic-Pituitary-Adrenal

Axis

Function

in Very

Low Birth Weight

Infants

Treated

With

Dexamethasone

Arie L Alkalay,

MD; Jeffrey

J. Pomerance,

MD, MPH;

Asha

R. Pun,

MD; Berwyn

J.C. Lin, MD; Arnold

L. Vinstein,

MD;

Naomi

D. Neufeld,

MD; and Alan

H. Klein,

MD

From the Divisions of Neonatology and Endocrinology, Ahmanson Pediatric Center, and

Department of Radiology, Cedars-Sinai Medical Center, University of California,

School of Medicine, Los Angeles

ABSTRACT.

The effect of dexamethasone therapy on

hypothalamic-pituitary-adrenal axis function was

pro-spectively investigated in very low birth weight infants

with bronchopulmonary dysplasia. Ten infants (mean ±

SD birth weight 825 ± 265 g, gestation 25.8 ± 1.9 weeks, postnatal age 33.1 ± 17.7 days) initially received

intra-venous dexamethasone, 0.5 mg/kg per day for 3 days, and then were weaned over a period of 45 ± 19.0 days to a replacement dose, followed by a metyrapone test.

Morn-ing plasma cortisol and 11-deoxycortisol levels were

measured before and after an oral metyrapone dose given at midnight. Five infants (group A: birth weight 876 ± 313 g, gestation 26.2 ± 1.3 weeks, age of entry 31.8 ± 22.8 days) had normal metyrapone test results, and five

in-fants (group B: 778 ± 234 g, 25.4 ± 2.5 weeks, 34.4 ± 13.4 days) had suppressed test results. Group A infants, in

comparison with group B infants, had higher basal

cor-tisol plasma levels (14.52 ± 12.53 and 3.00 ± 1.38 g/dL,

P = .047), higher postmetyrapone 11-deoxycortisol

plasma levels (3.11 ± 3.93 and 0.55 ± 0.51 sg/dL, P =

.028), larger differences between basal and postmetyra-pone cortisol levels (7.10 ± 4.67 and 2.12 ± 1.31 sg/dL,

P = .047), and larger differences between basal and

postmetyrapone 11-deoxycortisol levels (2.99 ± 3.93 and 0.29 ± 0.25 ;Lg/dL, P = .009). The

hypothalamic-pitui-tary-adrenal axis function in group B infants eventually

returned to normal when they continued to receive low-dose dexamethasone therapy after a period of 36.8 ± 16.6

days. The results suggest that dexamethasone therapy given for several weeks or more may be associated with prolonged suppression of hypothalamic-pituitary-adrenal

axis function in a substantial number of very low birth

weight infants with bronchopulmonary dysplasia.

Eval-uation of hypothalamic-pituitary-adrenal axis function

before discontinuation of dexamethasone therapy is nec-essary to ensure proper adrenal secretory response.

Pe-Received for publication Feb 9, 1989; accepted Aug 8, 1989. Reprint requests to (A.L.A.) Cedars-Sinai Medical Center, 8700

Beverly Blvd, Room 4310, Los Angeles, CA 90048.

American Academy of Pediatrics.

diatrics 1990;86:204-210; bronchopulmonary dysplasia. dexamethasone therapy, hypothalamic-pituitary-adrenal

axis, metyrapone test.

ABBREVIATIONS. HPAA, hypothalamic-pituitary-athenal axis; Fio2, fraction of inspired oxygen.

Bronchopulmonary dysplasia occurs most

com-monly in mechanically ventilated, very low birth weight infants.”2 The overall incidence of

broncho-pulmonary dysplasia in very low birth weight

in-fants is 15% and increases to 38% in those with respiratory distress syndrome. The mortality rate of infants with bronchopulmonary dysplasia during the first 2 years of life is 25%.’ Hydrocortisone therapy for the acute phase of respiratory distress syndrome has not been found to be helpful.4 How-ever, several studies have reported the usefulness, as well as the complications, of dexamethasone therapy in infants with bronchopulmonary dyspla-sia.9 In three controlled studies using

dexameth-asone therapy in respirator-dependent, very low

birth weight infants with bronchopulmonary dys-plasia,’#{176}’2 a significant decrease in the need for respiratory support and rapid weaning from the

respiratory were found. The potential suppression

(2)

bronchopulmo-nary dysplasia and the side effects and outcome of

dexamethasone therapy are reported.

PATIENTS

AND METHODS

This study assessed HPAA function in very low birth weight infants treated with dexamethasone

for respirator dependency in the neonatal intensive

care unit at Cedars-Sinai Medical Center in Los

Angeles between December 1985 and November

1988. The study group included the first 10 patients who met eligibility criteria. Institutional human subject committee approval was not obtained as the study protocol involved only the routine evaluation of adrenal function after a course of dexametha-sone. To be included, patients had to fulfill the following criteria: (1) birth weight of <1500 g; (2) clinical and roentgenologic findings consistent with

respiratory distress syndrome; (3) mechanically

ventilated for at least 2 weeks without respiratory improvement in the last 3 days; (4) no clinical or laboratory evidence of patent ductus arteriosus,

congestive heart failure, or infection; (5) fluid

re-striction to 150 mL/kg per day; and (6) receiving dexamethasone therapy according to the protocol of Kramer and Huylzen.5

Each infant initially was treated with

dexameth-asone at a dose that was many-fold higher than the

physiologic (14.0 ± 2.9 mg/rn2 per day)’3 secretory rate of hydrocortisone in newborns. Treatment was then tapered over a period of several weeks. The starting dose, 0.5 mg/kg per day, was administered

intravenously in twice-a-day aliquots for 3 days,

reduced to 0.3 mg/kg per day for 3 days, and there-after reduced by 10% of the current dose every 3 days until a dose of 0.1 mg/kg per day was reached. At this point dexamethasone was given on alternate days at 0.05 mg/kg every 12 hours. If an unex-plained respiratory setback occurred during the weaning process, a continuation of the same

dexa-methasone dose was given, or the dose was

in-creased to the previous dose and then tapered as

before. To evaluate the effect of this regimen on the HPAA, a metyrapone test was conducted after

the infant had been receiving alternate-day

dexa-methasone therapy for approximately 1 week.

Blood for measurement of basal plasma cortisol and li-deoxycortisol levels was drawn at 9:00 AM, fol-lowed by administration of 35 mg/kg of metyrapone at midnight by gavage.’4 At 9:00 AM the following

morning, blood for measurement of

postmetyra-pone plasma cortisol and 11-deoxycortisol levels

was drawn. A normal response was defined as a

fourfold or more increase in plasma

11-deoxycorti-sol levels accompanied by approximately a 50% or

more postmetyrapone decrease in plasma cortisol levels.’5 Infants who had an abnormal test response

were considered to have suppression of their HPAA function and therefore continued to receive low-dose dexamethasone (s0.1 mg/kg per day) on a!-ternate days for 2 or more weeks (depending on

their respiratory status), after which the

metyra-pone test was repeated. When the metyrapone test

response was normal, dexamethasone, 0.1 mg/kg

every alternate day, was discontinued. The

follow-ing variables were compared in infants with normal

and abnormal metyrapone test results: basal and postmetyrapone plasma levels of cortisol and 11-deoxycortisol, time interval (hours) between the last dexamethasone dose and measurement of plasma levels of basal and postmetyrapone cortisol and 11-deoxycortiso!, duration (days) of dexameth-asone therapy (from the day of initiation of therapy until the day of the first metyrapone test), weaning period (days) of dexamethasone therapy after the initial 3 days of 0.5 mg/kg per day of

dexametha-sone, duration (days) of alternate-day

dexametha-sone therapy, birth weight (grams), gestational age (weeks), and postnatal age (days) when dexameth-asone therapy was begun. Plasma cortisol and 11-deoxycortisol levels were analyzed by radio-immunoassay.

The short-term effect of dexamethasone therapy on respiratory status was evaluated in each infant by comparing his or her own predexamethasone therapy respiratory status with respiratory status 48 hours later. Blood gases were monitored approx-imately every 4 hours and respirator settings were adjusted to maintain the Pa02 and Paco2 between 50 and 70 mm Hg and 45 and 55 mm Hg, respec-tively. The average of the respiratory settings based on four blood gases obtained approximately 12 hours before the initiation of dexamethasone ther-apy and 12 hours after the infants had been receiv-ing dexamethasone therapy for 48 hours were corn-pared with regard to fraction of inspired oxygen (Fio2), intermittent mandatory ventilation rate, peak inspiratory pressure, and positive

end-expir-atory pressure.

The methods used for statistical analysis were as follows: Mann-Whitney U test for comparison be-tween infants, Wilcoxon signed-rank test for corn-parison within the infants, and Fisher’s exact test for categorical data.’6 All reported P values are

two-tailed.

RESULTS

(3)

il-deoxycortisol, in comparison with group B in-fants (Table 2). There were no statistically signifi-cant differences between the groups regarding the time interval between the dexamethasone dose be-fore measurement of plasma levels of basal and postmetyrapone cortisol and il-deoxycortisol. However, duration of dexamethasone therapy was longer in group A than group B infants. Despite normal metyrapone test results, because of respi-ratory status, two infants in group A continued to

E -J

0

I-0 C)

U

< -4

‘I)

le.

4/6 825 ± 265

25.8 ± 1.9 33.1 ± 17.7

48 ± 19.0

45 ± 19.0

6.3 ± 2.7

10 (100)

2 (20)

6 (60)

2 (20)

8 (80) 9 AM._DAY ₁ 9_DAYAM.₂ 9 AM._DAY ₁ 9 AM._DAY ₂

eEF0 , AFTER BEFORE , AFTER

METYRAPONE DOSE METYRAPONE DOSE

test results are depicted in Table 2 and Figs 1 and

2. Five infants had normal test results (group A)

and five had abnormal test results (group B). Group A and B infants were not statistically different with

regard to birth weight, gestational age, and age

when entered into the study. Group A infants had

statistically significantly higher basal plasma

cor-tisol and postmetyrapone li-deoxycortisol levels,

and larger differences between the respective basal and postmetyrapone plasma levels of cortisol and

TABLE 1. Characteristics of the Study Patients (N =

10)

Sex, M/F

Birth wt, g (mean ± SD)

Gestational age, wk (mean ± SD)

Age when started DEXM therapy, d

(mean ± SD)

Duration of DEXM therapy, d

(mean ± SD)

Weaning period of DEXM therapy

(mean ± SD)

Alternate-day DEXM therapy, d

(mean ± SD)

History of respiratory distress syndrome, no. (%)

Stage II BPD, no. (%)

Stage III BPD, no. (%)

Stage IV BPD, no. (%)

Diuretic and bronchodilator therapy, no.(%)

4 DEXM, dexamethasone; BPD, bronchopulmonary

dys-plasia stages according to Northway et al’7 (before Fig 1. Plasma cortisol and 11-deoxycortisol levels

be-DEXM therapy). fore and after 35 mg/kg of metyrapone.

TABLE 2.

Metyrapone Test Results and Characteristics of Groups A and B4

Group A

(n=5)

Group B (n=5)

P Value

Basal CORT, g/dL 14.52 ± 12.53 3.00 ± 1.38 .047

Post-METY CORT, ig/dL 7.42 ± 7.99 0.88 ± 0.52 .066

Pre-post-METY CORT difference, 7.10 ± 4.67 2.12 ± 1.31 .047

ig/dL

Basal 11-DC, g/dL 0.12 ± 0.07 0.26 ± 0.28 .347

Post-METY 11-DC, g/dL 3.11 ± 3.93 0.55 ± 0.51 .028

Pre-post-METY 11-DC difference, 2.99 ± 3.93 0.29 ± 0.25 .009

iig/dL

Duration of DEXM therapy, d 59.2 ± 15.2 36.8 ± 16.4 .046

Weaning period of DEXM therapy, 56.2 ± 15.2 33.8 ± 16.4 .055

d

Alternate-day DEXM therapy, d 5.6 ± 1.6 7.0 ± 3.5 .690 Time interval between DEXM and 31.6 ± 25.2 17.8 ± 12.0 .528

basal CORT and 11-DC, h

Time interval between DEXM and 34.8 ± 27.3 29.8 ± 14.3 .834 post-METY CORT and 11-DC, h

Sex, M/F 1/4 3/2 .524

Birth wt, g 876 ± 313 778 ± 234 .530

Gestational age, wk 26.2 ± 1.3 25.4 ± 2.5 .591

Age when started DEXM therapy, 31.8 ± 22.8 34.4 ± 13.4 .528

d

4 Infants in group A had no suppression of hypothalamic-pituitary-adrenal axis function;

infants in group B had suppressed function. Values are mean ± SD. CORT, cortisol;

(4)

.-. Nonstepreseed i’ifants o-o Suppressed mfants

30 V 25 C) E

:

20 0 Cl) 15 0 C) 10 C,) -J C

9 AM. 9 AM.

DAY1 DAY2

BEFORE , AFTER

METYRAPONE DOSE

9 AM. 9 AM.

DAY1 DAY2

BEFORE , AFTER METYRAPONE DOSE

Fig 2. Mean ± SD plasma cortisol and 11-deoxycortisol

levels before and after 35 mg/kg of metyrapone.

receive 0.05 mg/kg of dexamethasone every 12

hours every day and then on alternate days for a

total period of 4 and 5 weeks, respectively. Their respiratory requirements after the metyrapone tests were as follows: Fio2 0.42 and +5 cm H2O nasal

continuous positive airway pressure and Fio2 0.42,

respectively; continuation of therapy was

consid-ered prudent. Group B infants continued to receive

low-dose dexamethasone therapy for 36.8 ± 16.6

days. Results of a second metyrapone test were normal in four infants and dexamethasone therapy

was discontinued. In one infant, low-dose

dexa-methasone therapy was continued for 14 days after

the second metyrapone test showed abnormal

re-sults and discontinued after the third metyrapone

test, the results of which were normal. Two group B infants had postmetyrapone plasma cortisol

1ev-els similar to their basal levels. However, they had

large increases in their postmetyrapone

11-deoxy-cortisol plasma levels, approximately 35- and

60-fold, respectively. Therefore, the metyrapone test results in these two infants were considered normal.

The short-term effect on respiratory status of dexamethasone-treated infants 48 hours after

mi-tiation of therapy was manifested by a significant

improvement in ventilator settings (Table 3). Of the 10 infants, 9 were extubated 105 ± 83 (SD)

TABLE 3.

48 Hours Af

Average of Respiratory Settings ter Dexamethasone Therapy4

Before and

Before Therapy After Therapy

(n=10) (n=7)t P Value Fio2 IMV rate PIP PEEP

65.4 ± 21.1 49.4 ± 18.4

37.6 ± 13.8 17.1 ± 16.7

22.0 ± 6.4 12.4 ± 10.7

3.1 ± 0.7 1.9 ± 1.4

.0001 .0001

.0001

.0264

4 Values are mean ± SD. Fio2, fraction of inspired oxygen; IMV, intermittent mandatory ventilation; PIP, peak in-spiratory pressure; PEEP, positive end-expiratory pres-sure.

t Forty-eight hours after dexamethasone therapy, 3 of

the 10 study infants were extubated.

hours after initiation of dexamethasone therapy. Of

these infants, 7 received Fio2 0.23 to 0.42, and 2

received +5 cm H2O nasal and continuous positive

airway pressure and Fio2 0.25 and 0.27. In 1 infant,

extubation was not achieved until 35 days after initiation of dexamethasone therapy and therefore was not attributed directly to this therapy. After discontinuation of dexamethasone therapy, all

in-fants remained extubated and required Fio2

be-tween 0.21 and 0.34. Eight infants had worsening of the roentgenographic findings of bronchopul-monary dysplasia from grade II and III before dex-amethasone therapy to grade IV after therapy, and in 2 infants bronchopulmonary dysplasia remained at grade IV.

Side effects of dexamethasone therapy, depicted in Table 4, were as follows: (1) Four infants had steroid-related gastrointestinal bleeding manifested

as coffee-ground gastric aspirate. These infants

were treated successfully by discontinuation of

feedings and/or antacid therapy and gastric

irriga-tion with saline. (2) Three infants had hyperten-sion,’8 which was transient and normalized during weaning from dexamethasone in two; one infant required short-term antihypertensive therapy. (3) Three infants had hyperglycemia (two or more con-secutive chemstrip glucose levels >180 mg/dL,

5 :c without apparent cause). The hyperglycemia was

transient and responded to a decrease in glucose

4 intake. (4) Three infants acquired sepsis, two with

Staphylococcus epidermidis and one with Kiebsiella

3 pneumoniae. These infants responded to

appropri-ate antimicrobial therapy and recovered. (5) In two

2 infants cushingoid facies developed but resolved

several weeks after discontinuation of

dexametha-1 sone therapy.

No patients had air leak. One infant died at a

3 postnatal age of 8 months of cardiorespiratory

fail-ure after slow and progressive deterioration caused

by the chronic lung disease. Of the nine surviving

infants, eight were discharged at postnatal ages of

6.1 ± 2.8 months (range, 2 to 10 months). One

infant was discharged 3 weeks after completion of the study at a postnatal age of 9 months. This

infant is currently receiving supplemental oxygen

(5)

4/10 (40) 3/10 (30) 3/10 (30) 3/10 (30) 2/10 (20) 9/10 (90)

1/10 (10)

8/9 (89)

5.7 ± 2.7 23.1 ± 7.4

27 ± 5.8

6/7 (86)

2/7 (29)

3/7 (43)

6/7 (86) 4/7 (57)

4 Of the 10 study infants, 1 died, 1 is still hospitalized, and 1

fraction of inspired oxygen. t Pulmonary-related hospitalizations. :1:Corrected to gestational age.

is lost to follow-up. Fio2,

TABLE 4.

Dexamethasone Therapy Side Effects and Infant Outcome (N = 10)

Side effects and outcome, no. (%) Gastrointestinal tract bleeding

Hypertension

Hyperglycemia Sepsis

Cushingoid facies

Survival

Death

Discharge

Discharged infants: follow-up data (n = 7)

Hospitalization period, mo (mean ± SD)

Follow-up period, mo (mean ± SD)

Postnatal age, mo (mean ± SD)

Rehospitalization, no. (%)t Fio2 supplementation, no. (%)

Diuretic and/or bronchodilator therapy, no. (%) Weight (<10th percentile), no.

Length (<10th percentile), no. (%4

DISCUSSION

Increasing numbers of neonatal intensive care units are currently using dexamethasone therapy for respirator-dependent, premature infants who have failed fluid restriction and diuretic and bron-chodilator therapy and who have no contributing factors for respiratory dependency other than

bron-chopulmonary dysplasia. It is postulated that the

antiinflammatory properties of dexamethasone, which has far more potent glucocorticoid activity than hydrocortisone, play a role in its short- and long-term effects on chronic lung disease. Dexa-methasone therapy in these infants has the benefit of reducing exposure to chronic intubation, baro-trauma, and oxygen toxicity, but carries the risks of complications due to steroid therapy.5’1 The

starting dose of dexamethasone, which is currently

given by many nurseries and found effective, is

significantly higher than the physiologic secretory

rate of hydrocortisone, and the tapering period extends over several weeks.5”#{176}12 This therapeutic

regimen has the potential of causing suppression of

HPAA function. Evaluating HPAA function in treated infants is particularly important because of the likelihood of their encountering stressful

con-ditions such as infection, surgery, and/or

respira-tory distress. In infants and children, a single oral metyrapone test is a reliable means of evaluating HPAA function.’4’15”9’2#{176} In premature infants, ref-erence metyrapone test results are not available. However, there are well-documented reports of plasma cortisol and 11-deoxycortisol levels in healthy newborns,21 plasma cortisol levels in healthy and sick premature infants,’23 and the

capacity of healthy and sick premature infants to increase their plasma cortisol and

il-hydroxypro-gesterone levels twofold to threefold in response to

adrenocorticotropic hormone.24 Although prema-tore infants may well not have established diurnal variation of cortisol secretion, the metyrapone test was administered as if the subject were an older child.

Fifty percent of dexamethasone-treated patients in the study had initially suppressed HPAA

func-tion, which subsequently became normal after

pro-longed dexamethasone therapy. The infants whose HPAA function was not suppressed received

dexa-methasone therapy for longer periods than the

in-fants with suppressed HPAA function; the groups were statistically comparable. The longer weaning period of the infants with nonsuppressed HPAA

function may have accounted for the lack of

suppression of their HPAA function. The infants with nonsuppressed HPAA function and basal plasma cortisol levels significantly higher than those of the infants with suppressed function. The infants with suppressed function also had lower basal plasma cortisol levels when compared with those reported in very low birth weight infants.22’23 Recently, a transient reduction in basal serum cor-tisol level with no decrease in cortisol response to adrenocorticotropic hormone was reported in pre-mature infants with bronchopulmonary dysplasia who were treated with high doses of dexamethasone therapy for 1 week.24

(6)

may be a factor in the test results. This factor probably did not play a role in the metyrapone test

results of the present study for the following

rea-sons: (1) Of 20 blood samples used for the

metyra-pone test, 15 were obtained at least 12 hours after the last dose of dexamethasone, and 2 blood sam-ples from group A infants and 3 from group B were obtained 3 to 11 hours after the last dose of dexa-methasone. (2) There was not a statistical signifi-cant difference between groups A and B with regard to the time interval between the metyrapone test and the last dose of dexamethasone.

The results of the present study confirm those of

previous investigations512 with regard to the

short-term respiratory benefits of dexamethasone ther-apy. Even though significant morbidity is associ-ated with dexamethasone therapy, the side effects were transient and easily manageable. The sepsis rate in the study group during the period of dexa-methasone therapy was 30%, which is higher than the sepsis rate in nonsteroid-treated very low birth weight infants in our unit during the years 1986 and 1987 (25% and 21%, respectively). However, the study infants were among the most ill very low birth weight infants and, therefore, were more likely to acquire infection. The mortality rate in our small

study group was 1 infant of 10. This mortality rate

is lower than that reported in comparable studies

of dexamethasone-treated very low birth weight infants (50% [3/6 infants]: 2 died of infection and 1 of sudden infant death syndrome’#{176}; and 37% [3/

8 infants]: all 3 died of pulmonary-related

diseases11). In contrast with previous studies, the

present study investigated HPAA function, and discontinuation of dexamethasone therapy was contingent on normalization ofthe metyrapone test results. Despite the fact that stage IV bronchopul-monary dysplasia, visible on chest roentgenogram, developed in all study infants, the overall clinical respiratory outcome was favorable. The explana-tion for this discrepancy is probably the lack of good correlation between bronchopulmonary dys-plasia findings by chest roentgenography and din-ical respiratory status.25

The following conclusions are derived from the

present study: (1) Dexamethasone therapy is

asso-ciated with suppression of HPAA function in a substantial number of very low birth weight infants with bronchopulmonary dysplasia. This suppres-sion is reversible with continuation of prolonged, low-dose dexamethasone therapy for several weeks. (2) Improvement of respiratory status by means of dexamethasone therapy may result in a rapid wean-ing from mechanical ventilation. (3) Side effects of dexamethasone therapy were easily manageable and transient in nature. (4) Controlled studies are

needed to establish the long-term outcome and the risks of steroid therapy.

We recommend that decisions to use dexameth-asone therapy in respirator-dependent, very low birth weight infants with bronchopulmonary dys-plasia should be individualized. Benefits and risks of the therapy have to be carefully evaluated.

As-sessment of the HPAA function before

discontin-uation of therapy is necessary to ensure proper adrenal secretory capacity. This approach may take on added importance as these infants are at high risk for infection, subsequent surgery, and/or res-piratory distress.

ACKNOWLEDGMENT

We thank Paula J. Chou, MD, for her helpful com-ments in preparation of this manuscript.

REFERENCES

1. Tooley WH. Epidemiology of bronchopulmonary dysplasia.

J Pediatr. 1979;95:851-55

2. Hodson WA, Truog WE, Mayock DE, et al. Bronchopul-monary dysplasia: the need for epidemiologic studies. J Pediatr. 1979;95:848-851

3. Wung J-T, Koons AH, Driscoll JM, et al. Changing mci-dence of bronchopulmonary dysplasia. J Pediatr. 1979;

95:845-847

4. Taeusch HW Jr, Wang NS, Baden M, et al. A controlled trial of hydrocortisone therapy in infants with respiratory distress syndrome, II: pathology. Pediatrics. 1973;52:850-854

5. Kramer LI, Huylzen C. The role of steroid in early broncho-pulmonary dysplasia (BPD). Pediatr Res. 1978;12:564. Ab-stract

6. Schick JB, Goetzman BW, Smith LE, et al. Corticosteroid response in chronic lung disease of prematurity. Pediatr Res.

1981;15:680. Abstract

7. Pomerance JJ, Pun AR. Treatment of severe neonatal

bron-chopulmonary dysplasia with dexamethasone. Pediatr Res.

1982;16:395. Abstract

8. Donn SM, Faix RG, Banagale RC. Dexamethasone for bron-chopulmonary dysplasia. Laricet. 1983;2:460. Letter

9. Schick JB, Goetzman BW. Corticosteroid response in chronic lung disease of prematurity. Am J Perinatol. 1983;1:23-27

10. Mammel MC, Green TP, Johnson DE, et al. Controlled trial of dexamethasone therapy in infants with bronchopulmo-nary dysplasia. Lancet. 1983;1:1356-1358

11. Avery GB, Fletcher AB, Kaplan M, Brudno DS. Controlled trial of dexamethasone in respirator-dependent infants with bronchopulmonary dysplasia. Pediatrics. 1985;75:106-111

12. Cummings JJ, D’Eugenio OB, Gross SJ. A controlled trial

of dexamethasone in preterm infants at high risk for bron-chopulmonary dysplasia. N EnglJMed. 1989;320:1505-1510

13. Kenny FM, Preeyasombat C, Mingeon CJ. Cortisol produc-tion rate, II: normal infants, children, and adults. Pediatrics.

1966;37:34-42

14. Collu R, Ducharme RJ, Guyda H. Pediatric endocrinology.

In: Migeon CJ, ad. Physiology and Pathology of Adrenocor-tical Function in Infancy and Childhood. New York, NY: Raven Press; 1981:465-522

15. Kelley VC. Metabolic endocrine and genetic disorders of

(7)

16. Conover WJ, ed. Practical Nonparametric Statistics. New York, NY: John Wiley & Sons; 1980:166-288

17. Northway WH Jr. Rosan RC, Porter DY. Pulmonary disease

following respirator therapy in hyaline membrane disease: bronchopulmonary dysplasia. N Engl J Med. 1967;276:357-368

18. Tan RL. Blood pressure in very low birth weight infants in first 70 days of life. J Pediatr. 1988;112:266-270

19. Limal JM, Basmaciogullari A, Rappaport R. Evaluation of single oral dose metyrapone tests in children with hypopi-tuitarism. Acta Paediatr Scand. 1976;65:177-183

20. Kohlberg IJ, Doret AM, Paunier L, et al. Assessment of pituitary adrenal cortex axis by single dose metyrapone test.

Helu Paediatr Acta. 1972;27:437-447

21. Wiener D, Smith J, Dahlem 5, et al. Serum adrenal steroid

levels in healthy full-term 3-day-old infants. J Pediatr.

1987;110:122-124

22. Kanarek KS, Villareces C, Duckett G, et al. Serum concen-trations of growth hormone, insulin, free thyroxine, thyro-tropin, and cortisol in very low birth weight infants receiving total parenteral nutrition. AJDC. 1988;142:993-995

23. Wilson DM, Baldwin RB, Ariagno RL. A randomized, pla-cebo-controlled trial of effects of dexamethasone on hy-pothalamic-pituitary-adrenal axis in preterm infants. J Pe-diatr. 1988;113:764-768

24. Thomas 5, Murphy JF, Dyas J, et al. Response to ACTH in

the newborn. Arch Dis Child. 1986;61:57-60

25. Ballard RA. Pediatric care of the ICN graduate. In: Platzker

ACG, ad. Chronic Lung Disease of Infancy. Philadelphia,

PA: WB Saunders Co; 1988:129-156

COMPETITION

AND

COST

-

HOSPITAL

STYLE

Hospital construction is booming despite anguish over rising health-care costs and despite the fact that far more hospitals have empty beds than waiting lists. Although some hospitals in urban areas, especially New York City, have extremely high occupancy rates, more than one-third of the nation’s 947,000 community-hospital beds are empty.

Instead of building new facilities, many hospitals are adding services and

modernizing as a marketing strategy to attract patients from competing

insti-tutions. But in the health business, unlike most others, competition drives prices up, not down. . . .prices [are] generally higher in competitive,

multihos-pita! markets than in single-hospital areas.

“Hospitals don’t compete on price; they compete on services,” explains. . .the

president of the Health Insurance Association of America. With occupancy low, “the pure economic response is that you develop things to fill your hospital.” Aside from offering new, expensive technology, hospitals have set up centers for diet control, sports medicine, drug and alcohol treatment and psychiatry to attract patients to empty beds.

But the cost of excess capacity goes beyond higher prices.”More capacity to provide more services. . .simply raises volume and expenditures,” contends Rep.

Fortney Stark who heads a House Ways and Means health subcommittee. “Therefore, the high rate of excess capacity in beds and services leads directly to unnecessary admissions and procedures.”

He adds: “There’s nobody in the American Hospital Association whose mother has ever told him that enough is enough.”

Bacon KH. Hospital construction booms, driving cost of health care up. The Wall Street Journal.

(8)

1990;86;204

Pediatrics

Naomi D. Neufeld and Alan H. Klein

Arie L. Alkalay, Jeffrey J. Pomerance, Asha R. Puri, Berwyn J.C. Lin, Arnold L. Vinstein,

Treated With Dexamethasone