- Email: [email protected]
Airway muscle in preterm infants: Changes during development
Airway muscle in preterm infants: Changes during development Susan L. Sward-Comunelli, MD, Sherry M. Mabry, MS, William E. Truog, MD, a n d Donald W. Thibeault, MD From the Department of Pediatrics,Children's Mercy Hospital, Universityof Missouri-Kansas City School of Medicine, Kansas City, Missouri
Objective: To quantitate airway muscle changes in infants born at 23 to 41 weeks' gestation (control subjects) and to compare the changes with those in infants with chronic lung disease. Methods: Fifty-five human lungs (from infants born at 23 to 41 weeks' gestation) were studied: 46 from infants who died of various diseases within 72 hours of birth, and 9 from infants with CLD (infants born at 26.9 ± 0.5 weeks' gestation, who lived 17 ± 8 days). All the lungs were perfused via the trachea and pulmonary artery in a standardized protocol. Formalin-fixed tissues in paraffin blocks were cut 5 pm thick. Sections were immunohistochemically stained for =-smooth muscle actin, By using computerized image analysis to quantitate images digitized into the computer, we measured the area of muscle, epithelium, airway lumen, and length of basement membrane in 18 airways, from the smallest bronchioles to bronchi, in each infant. Results: Muscle was present at 23 weeks' gestation at all levels of the bronchial tree, and from 25 weeks to term the control lungs had a similar quantity of muscle at any given airway circumference. Relative to airway size, there was more muscle in small airways, less than 1000 IJm in circumference, than in larger airways. In airways greater than 1500 IJm in circumference, infants with CLD had significantly more muscle than did control lungs. Conclusions: Airway muscle is present at 23 weeks' gestation at all levels of the conducting airways. The 25-week gestation infants had a quantity of airway muscle relative to airway circumference similar to that of term infants. Preterm infants with CLD who were aged 9 to 29 days have increased airway muscle in airways greater than 1500 pm in circumference. Bronchospasm in very low birth weight infants is possible within the first days of life. (J Pediatr 1997;130:570-6) Lack of information on the development of airway muscle confounds the interpretation of postnatal muscle changes in infants at risk of having bronchopulmonary dysplasia or chronic lung disease. In infants with BPD the rationale for bronchodilator therapy is supported by a decrease in pulmoSupported by the Katharine B. Richardson Foundation and a Physician Scientist Award (Dr. Truog) from The Children's Mercy Hospital. Submitted for publication April 25, 1996; accepted Sept. 27, 1996. Reprint requests: Donald W. Thibeault, MD, Children's Mercy Hospital, 2401 Gillham Rd., Kansas City, MO 64108. Copyright © 1997 by Mosby-Year Book, Inc. 0022-3476/97/$5.00 + 0 9121178375
570
nary resistance after its use, as well as by clear anatomic evidence of increased airway muscle mass. ]4 In contrast, in very low birth weight infants receiving assisted ventilation during the first 2 weeks of life, there is limited evidence BPD CLD PBSS
Bronchopulmonary dysplasia Chronic lung disease Phosphate-buffered saline solution
documenting the efficacy of bronchodilator therapy.I' 5 Anatomic studies demonstrate a lack of adequate airway muscle distal to the small bronchi) A detailed, systematic study of the ontogeny of bronchial tree smooth muscle in relatively normal preterm infants could not be found. 3' 6 The purpose
The Journal of Pediatrics Volume 130, Number 4
Sward-Comunelli et al.
571
] ' a b l e . Clinical characteristics of infants living less than 72 hours compared with those with chronic lung disease Groups (wk or CLD)
Gestational ages (wk) Birth weight) (gm) Days lived* Primary cause of death'~ IRDS PDA Pulmonary hypertension PIE MAS Respiratory failure Asphyxia Septic shock NEC Intracranial hemorrhage Other hemorrhage Arrhythmia Drug treatment Surfactant Prenatal steroids Thyroid-releasing hormone
23-24 (n = 7)
25-26 (n = 10)
27-30 (n = 10)
31-36 (n = 8)
37-41 (n = 1 I)
CLD (n = 9)
23.6 -+ 0.5 578 -+ 76 0.9 -+ 0.7
25.4 -+ 0.5 759 -+ 115 0.9 +- 0.9
28 -+ 0.6 1020 -+ 153 1.5 + 1.0
33.1 -+ 2.1 2057 -+ 775 0.6 -+ 0.5
39.4 -+ 1.4 3199 -+ 928 1.6 -+ 1.1
26.9 -+ 1.5 841 -+ 276 17 +- 8
3 0 0 1 0 0 1 0 1 0 0 1
4 0 0
6 1 0
1 1 0
1 0 2
0 0 0
1
0
0
0
0
0 0 2 0 0 2
0 0 0 0 0 2
0 0 3 2 0 0
4 0 2 0 0 1
0 7 0 1 0 0
1
1
0
1
0
0
0
1
0
1
3 3 0
6 4 0
7 4 0
5 3 0
0 0 0
6 4 0
Values for first three rows are expressed as mean _+SD. IRDS, Idiopathic RDS; PDA,patent ductus arteriosus; PIE, pulmonary interstitial edema; NEC, necrotizing enterocolifis;pPHN, persistent pulmonary hypertension of the newborn. *Not significantly different among groups without chronic lung disease. -~Acomparison of muscle area versus airway circumference in infants with PPHN or MAS was not different from those with other causes of death.
of this study was to quantitate airway muscle mass in 23- to 41-week gestation infants who died within 72 hours of birth and to compare the findings with those in infants with CLD. METHODS
Fifty-five human infants, born at 23 to 41 weeks of gestation and appropriate in size for gestational age, were studied (Table I). Forty-six infants died at between 1 and 72 hours (28 _+ 23 hours). The 9 infants with CLD were born at 26.9 _+ 0.5 weeks' gestation and lived 17 _+ 8 days (range, 9 to 29 days). All infants with CLD received mechanical ventilation with a pressure-limited ventilator and required increased inspired oxygen concentrations throughout life; 7 died of respiratory failure. BPD was diagnosed in all these infants after death, on the basis of altered alveolar inflation patterns, airway changes, and fibroproliferative acinar changes. W e used the clinical term chronic lung disease rather than BPD because of the •relatively short life span. The time from death to autopsy was 13 _+ 11 hours. Lungs with widespread bronchopneumonia or leaks that could not maintain the perfusion pressure were excluded from the study. The lungs were wanned to 38 ° C, and the pulmonary artery was injected with a 60 ° C radiopaque barium-gelatin mixture (4 minutes, 100 cm H 2 0 pressure). 7' 8 The lungs, via
the trachea, were then perfused 72 hours at 24 cm H20 with 10% buffered formalin. The following sections were cut: an entire cross section of the left upper lobe; the base of the lingula; a 1.5 x 1.0 cm piece of the left lower lobe (one side containing pleura); and a longitudinal hilar cut (1.5 x 1.0 cm), including the central hilar structures surrounded by parenchyma. After immersion in 10% buffered formalin, the area of the cut surface of each section was measured by computerized image analysis (Optimas Corp.~ Edmonds, Wash.). The tissue samples were processed and embedded in paraffin. Sections, 5 pm thick, were stained with hematoxylin and eosin and with M i l l e r - v a n Gieson. Additional 5 pm thick sections were mounted on poly-Llysine-coated glass slides and immunohistochemically stained for actin. A kit (Sigma Immunochemicals, St. Louis, Mo.) using the avidin-biotin complex method 9 was used immunohistochemically to demonstrate s - s m o o t h muscle actin. The mouse monoclonal primary antibody used specifically identifies the smooth muscle actin ~ isotype, reacting with normal and neoplastic smooth muscle and myoepithelial cells. 1° Sections were deparaffinized, rehydrated, and treated with 3% hydrogen peroxide for 5 minutes to remove endogenous peroxidase; they were then washed in phosphate-buffered saline solution (pH 7.4) and incubated first
572
Sward-Comunelli et al.
The Journal of Pediatrics April 1997
Fig. 1. A bronchiole with a diameter of 263 pm from an infant who was born at 27 weeks' gestation and died of respiratory distress syndrome at age 23 hours. Arrow, Dark band of immunohistochemically stained airway muscle actin (x200).
25 WEEKS GA
CHRONIC LUNG DISEASE BRONCHUS
BRONCHUS
v
PRETERMINAL BRONCHIOLE
PRETERMINAL BRONCHIOLE
.13
MUSCLE AREA
.22
173
DIAMETER (~m}
179
Fig. 2. Airways from two representative 25-week-gestation infants. Airways on the left represent an infant who died of RDS at age 18 hours. Those onthe right represent an infant with CLD who died at 14 days. The ratios were computed by dividing the three areas (muscle, epithelial, and airway luminal) by the area bounded by the perimeter of the epithelial basement membrane (BM). The muscle and epithelial areas are larger, relative to airway size, in smaller, preterminal bronchioles than in bronchi.
with normal goat serum for 10 minutes and then with primary antibody for 60 minutes, After being rinsed with PBSS, sections were incubated for 20 minutes with biotinylated secondary antibody, washed with PBSS again, and incubated for
20 minutes with avidin-conjugated peroxidase reagent. Antigenic sites were visualized by addition of the chromogen 3,3'-diaminobenzidine, which formed a brown precipitate. Slides were counterstained with methyl green (Fig. 1). Neg-
~,0
100000
80000
/k • • O [] •
CLD 37-41 WKS 31-36 WKS 27-30 WKS 25-26 WKS 23-24 WKS
4~
CLD 0.8
o
v
60000
°
o
•
•
#e
Oo
O
04
<
2
< ©
23-41 WKS
06
E
-~
•
Z~x o
0
£
57 3
Sward-Comunelli et al.
The Journal of Pediatrics Volume 130, Number 4
:s
0.2
40000 0.0
20000 Airway Area (um2X 106
i
i
1000 2000
i
30'00 4000
i
i
5000 6000
"7000
Airway Circumference (um) Fig. 3. Airway muscle area increases in all groups nonlinearly as airway circumference increases. Data overlap in all control groups. Infants with CLD have increased muscle in airways greater than 1500 gm in circumference. Data from all infants, born at 25 to 41 weeks' gestation, were best fit to the polynomial equation: Muscle area = 1040 + Airway circumference + 10 x 1.28.3 x Airway circumference2 (r = 0.945; p <0.0001). In the 23- to 24-week group, there is a trend toward less muscle area in airways larger than 4000 Wn in circumference.
ative control slides were stained by using the same procedure, with the primary antibody omitted or a nonimmune serum added. In two lungs, comparisons were made of eight bronchial muscle areas measured on sequential 5 gm sections by using the actin immunohistochemical stain and the Gomori Irichrome stain. In large airways the two techniques produced similar muscle-area measurements. For small bronchioles, the actin stain was more discrete and far superior for image analysis. Therefore actin stains were used for all measurements reported in the Results section, below. Staining of the same bronchiolar area, in 5 p r n serial cuts (n = 6), revealed an airway muscle area of 14,879 -+ 2177 N n 2, indicating that the reproducibility of measuring the muscle area was -+ 14%. However, because the muscle spirals, each serial section is not identical to the previous one; thus some of the measurement variation is biologic. The area of sections immunohistochemically stained for actin was compared with the area of the cut surface before
Fig. 4. Nonlinear increase in airway muscle area as the airway area increases. Airway area is the area bounded by the epithelial basement membrane and includes the epithelial and luminal areas. The data overlap on infants born at 23 to 41 weeks' gestation (open circles). The best-fit polynomial equation is as follows: Muscle area = 0.1449 + (0.027 x Airway area) - (3.38 x 10.9 x Airway area)2 (r = 0.952; p <0.001). Infants with CLD had increased muscle area compared with control infants but only in airways greater than 1500 Nn in circumference.
processing to determine the shrinkage correction factor. In the 27- to 30-week group, muscle areas from infants who died within 24 hours of birth were compared with muscle areas from those who died after 25 to 72 hours, to determine whether the muscle,mass changed with time. Eighteen airways throughout the respiratory tree, from preterminal bronchioles to cartilaginous bronchi, were measured in each study infant. Measurements were made by investigators masked to clinical history. With the use of image analysis, the epithelial basement membrane perimeter was traced by mouse. The circumference, length, and breadth of each airway were calculated by computer. Airways whose length-to-breadth ratio was equal to or greater than 2 were considered to be cut obliquely and were not analyzed. Color thresholds were set for the brown-stained muscle actin and the white airway lumen. The computer calculated the following areas, schematically shown in Fig. 2: muscle, region within the perimeter of the epithelial basement membrane, and airway lumen. Epithelial area was obtained by subtraction. The epithelial and muscle areas were standardized for each airway size by dividing by airway circumference (length of the epithelial basement membrane). This yields the effective mean muscle and epithelium thickness. Basement membrane length was used as a standardized measure of airway size because it does not change with
574
Sward-Comunelli et al.
The Journal of Pediatrics April 1997
30
90 • CLD - -I- . 37-41 WKS
25
:;:
.
,m • ••
o o
-
80
- 25-26 WKS - -~, - 23-24 WKS -~
oo°°
-
CLD
o
oO 70 X
X
20
[
••
<
0
:
<.-~ 15
0
0
°8°
°o
50
•
40
oj
<
A!~B%oo o 0 o 9 b
{.l::" 4~
-.
O~q6:Yo
"
,3~:"~ ~_.--_.._i ". .-
o-,o o r~
30
"
• 20 . . . .
:.: 0
1000
2000 3000 40'00 5~0
.
~oOo
~ o
10
6000 7000
Airway Circumference (um)
o o o
i
0
1000
2000
3000
4000
5000
6000
7000
Airway Circumference (um)
Fig. 5. The muscle area/airway area ratio is exponentially and inversely related to airway circumference. As circumference decreases to less than 1000 gm, muscle area increases relative to airway area. There are no differences among the control groups. However, the ratio is greater in infants with CLD in airways greater than 1500 #in in circumference.
bronchocons~iction, n The effective mean thickness represents the muscle thickness that would be seen in completely relaxed airways or airways distended at 24 cm H20. Muscle and epithelial areas also were standardized for airway size by dividing by the' total area internal to the epithelial basement membrane (i.e., airway area, which is the epithelial plus luminal area). Statistical analysis, Values are given as mean + SD in the Table. Statistical analysis in the Table and text was performed with t tests or analysis of variance when there were more than two groups. A value of p <0.05 was considered significant. Regression analysis and best-fit equations were determined by computer analysis (SigmaStat; Jandel Scientific, San Rafael, Calif.).
RESULTS Absolute airway muscle area in infants of 23 to 41 weeks' gestation increases nonlinearly from the smallest bronchiole to the bronchial level (Fig. 3). In infants born at 23 to 24 weeks of gestation, airways greater than 4000 grn in circumference trend toward less muscle than do those from
Fig. 6. Epithelial area/airway area ratio is nonlinearly and inversely related to airway circumference. The control data points (open circles) show wide scatter. The best-fit equation for the control data is as follows: Epithelial area/Airway area x 100= 384 x Airway circumference qz3 (r = 0.644; p <0.001). Infants with CLD were not different from control subjects. later-gestation infants. In infants with CLD, airway muscle was greater than control airway muscle at an airway circumferences greater than 1500 pm. In smaller airways the 95% confidence limits of control and CLD data overlap. Muscle area as a function of airway area also was nonlinear (Fig. 4). Infants with CLD had increased muscle compared with control lungs in airways greater than 1500 pm in diameter. The steeper slope in the smaller airways shown in Fig. 4 indicates that there is more muscle relative to airway size in small airways. This can be better visualized in Fig. 5, in which the muscle area/airway area ratio is plotted against airway circumference. In control lungs, muscle area relative to the airway area increases sharply in airways smaller than 1000 pm in circumference. In larger airways the muscle area/airway area ratio remains steady at 0.02 to 0.04. Control data overlapped at all airway sizes and best fit the exponential equation: Muscle area/airway areax 100=406e225/airway circumference (r=0.720; p <0.0001). In infants with CLD the muscle area/airway area ratio is increased compared with control airways geater than 1500 wn in circumference. CLD lung ratios tend to be greater in smaller airways
The Journal of Pediatrics Volume 130, Number 4
but overlap the 95% confidence limits of the control data points. The epithelial thickness linearly increases at all ages as the airway circumference increases: Epithelial thickness= 12.3 + 0.015 x Airway circumference (r = 0.824; p <0.0001). The data, similar in all groups, including infants with CLD, should be interpreted cautiously because 28% of airway measurements were excluded because of lost or severely damaged epithelium. Similar to muscle, the ratio of epithelial area to airway area was nonlinear, with the epithelium constituting a relatively larger area in small airways (Fig. 6). Bronchial muscle area measurements on sequential cuts with the Gomori trichrome stain were not significantly different from measurements from immunohistochemically actin-stained slides. Muscle measured in infants who died within 24 hours of birth did not significantly differ from that measured in infants 25 to 72 hours of age. Prenatal steroid treatment did not change the relation of muscle area to airway circumference in any group. DISCUSSION The results demonstrate that newborn infants from 23 weeks' gestation to term have airway muscle from the smallest preterminal bronchiole to the bronchi. Furthermore, this muscle mass, relative to airway size, is the same at 25 weeks' gestation as it is at term. As in adults, it increases as airway size increases, and relative to airway size there is more muscle in smaller airways. 12 Preterm infants with early or established histol0gic BPD changes, who died after prolonged ventilator assistance, had more muscle in all conducting airways, particularly where the circumference was greater than 1500 gin. Increased airway muscle in BPD has been shown repeatedly,3,4, i3, 14 but we show that it can occur within 10 days of birth, before a clinical BPD diagnosis is established. However, it is not entirely clear whether ventilator-dependent preterm infants without CLD radiographic changes, who require low inspiratory pressures and minimal oxygen, have increased muscle in conducting airways. 3, 13, ~5 Nevertheless, the muscle mass found in 23-week to term control infants is compatible with that reported in studies demonstrating increased airway resistance and response to bronchodilators before BPD development. 1,2,5 Increased airway resistance in small preterm infants may be a function of both muscle mass and epithelial thickness, both of which are increased relative to airway size in small conducting airways. This is schematically shown in Fig. 2. In both control infants and the infants with CLD, 28% of all airways, particularly the small bronchioles, have damaged or denuded areas of epithelium. These damaged areas may also be sites of inflammation and secretions that could further increase air-
Sward-Comunelli et aL
575
way reactivity, narrowing, and resistance. The demonstration of muscle in the most-immature infants allows the possibility of bronchoconstriction and may further justify the study and use of bronchodilators despite their potential side effects. The presence of airway muscle so early in gestation suggests that muscle may serve an important function in lung development. The finding of relatively more muscle in small airways, compared with larger, upper airways (Fig. 5) throughout development suggests that the role of fetal muscle in lung growth may differ in various parts of the tracheobronclfial tree. It is known that fetal lung development is, in part, regulated by lung liquid pressure acting against a resistance to outflow in the upper airways. 16' 17Muscle in large and small fetal airways can undergo relaxation and contraction, changing overall intralung airway pressure and thereby influencing growth. 18,19 The relatively increased muscle mass at the small-bronchiolar level could differentially regulate intraacinar fluid pressure in the lung, ensuring symmetric growth. 19-22 Smooth-muscle tone also may provide stability in th e conducting airways surrounded by the fetal fluid-filled air sacs. This would be particularly important in smaller, more compliant airways. 23 Use of the c~-smooth muscle actin antibody to determine muscle area gave results similar to those of traditional muscle stains. However, in small airways the actin stain is superior for image-analysis quantitation. With lrichrome staining of small airways, it is often difficult or impossible to separate the muscle from the surrounding tissue. Tissue stained for muscle, or antibody-identified muscle actin, is not guaranteed to be functional muscle. Nevertheless, our data are compatible with, and extend in important ways, previous anatomic and embryologic studies, as well as clinical studies. REFERENCES
1. Motoyama EK, Fort MD, Klesh KE, Mutich RL, Guthrie RD. Early onset of airway reactivity in premature infants with hronchopulmonary dysplasia. Am Rev Respir Dis 1987;136: 50-7. 2. Wilkie RA, Bryan MH. Effect of bronchodilators on airway resistance in ventilator-dependent neonates with chronic lung disease. J Pediatr 1987;111:278-82. 3. Hislop AA, Haworth SG. Airway size and smacture in the normal fetal and infant lung and the effect of premature delivery and artificial ventilation. Am Rev Respir Dis 1989;140:171726. 4. Bonikos DS, Bensch KG, Northway WH, Edwards DR. Bronchopulmonary dysplasia: the pulmonary pathologic sequel of necrotizing bronchiolitis and pulmonary fibrosis: Hum Pathol 1976;7:643-66. 5. Gomez-Del Rio M, Gerhardt T, Hehre D, Feiler R, Bancalari E. Effect of a beta-agonist nebulization on lung function in neonates with increased pulmonary resistance. Pediatr Pulmonol 1986;2:287-91. 6. O'Brodovich HM, Mellins RB. Bronchopulmonary dysplasia;
576
7.
8.
9.
10.
11.
12.
13. • 14.
Sward-Comunelli et al.
The Journal of Pediatrics April 1997
unresolved neonatal acute lung injury. Am Rev Respir Dis 1985;132:694-709. Meyrick B, Reid L. Pulmonary arterial and alveolar development in normal postnatal rat lung. Am Rev Respir Dis 1982; 125:468-73. Downing GJ, Thibeault DW. Pulmonary vasculatnre changes associated with idiopathic closure of the ductus arteriosus and hydrops fetalis. Pediatr Cardiol 1994;15:71-5. Hsu SM, Raine L, Fanger H. Use of avidin-biotin peroxidase complex (ABC) in irrlmunoperoxidase techniques: a comparison between ABC and unlabeled antibody (PAP) procedures. J Histochem Cytochem 1981;29:577-80. Bussolati G, Gugliotta P, Fulcheri E. Immunohistochemistryof actin I normal and neoplastic tissues. In: DeLellis RA, editor. Advances in immunochemistry.New York: Masson, 1984:325. Ebina M, Yaegashi H, Takahashi T, Motomiya M, Tanemura M. Distribution of smooth muscles along the bronchial tree: a morphometric study on ordinary autopsy lungs. Am Rev Respir Dis 1990;141:1322-6. Matsuba K, Thurlbeck WM. A morphometric ~study of bronchial and bronchiolar walls in children. Am Rev Respir Dis 1972;105:908-13. Taghizadeh A, Reynolds EOR. Pathogenesis of bronchopulmonary dysplasia following hyaline membrane disease. Am J Pathol 1976;82:241-58. Hislop AA, Haworth SG. Pulmonary vascular damage and the development of cor pulmonale following hyaline membrane disease. Pediatr Pulmonol 1990;9:152-61.
15. Sward-Comunelli SL, Mabry SM, Truog WE, Thibeault DW. Airway muscle in preterm infants: changes during development [abstract]. Pediatr Res 1995;37:353A. 16. Moessinger AC, Harding R, Adamson TM, Singh M, Kiu GT. Role of lung fluid volume in growth and maturation of the fetal sheep lung. J Clin Invest 199i3;86:1270-7. 17. Harding R. The upper respiratory tract in perinatal life. In: Johnston BM, Gluckman PD, editors. Respiratory control and lung development in the fetus and newborn. Ithaca (NY): Perir~atology Press, 1986:331-76. 18. Isabel JB, Towers B, Adams FIt, Gyepes MT. The effects of ganglionic blockade on tracheobronchial muscle in fetal and newborn lambs. Respir Physiol 1972;15:255-67. 19. Sparrow MP, Warwick SP, Mitchell HW. Foetal airway motor tone in prenatal lung development of the pig. Eur Respir J 1994;7:1416-24. 20. Hooper SB, Hart VKM, Harding R. Changes in lung expansion alter pulmonary DNA synthesis and IGF-II gene expression in fetal sheep. Am J Physiol 1993;265 Lung Cell Mol Physiol 9:L403-9. 21. McCray PB. Spontaneous contractility of human fetal airway smooth muscle. Am J Respir Dis 1993;8:573-80. 22. Sparrow MP, Warwick SP, Everett AW. Innervation and function of the distal airways in the developing bronchial tree of fetal pig lung. Am J Respir Dis 1995;13:518-25. 23. Bhutani VK, Koslo ILl, Shaffer TH. The effect of tracheal smooth muscle tone on neonatal airway collapsibility. Pediatr Res 1986;20:492-5.
FELLOWSHIPS Fellowships available in pediatric subspecialties and those for general academic pediatric training are listed once a year, in J a n u a r y , in The Journal of Pediatrics. Each June, forms for listing fellowships available for the academic year beginning 18 months after publication are sent to the Chairman of the Department of Pediatrics at major hospitals in the United States and Canada. In addition, a copy of the application form appears in the July, August, and September issues of The Journal (please use the current form). Should you desire to list fellowships, a separate application must be made each year for each position. All applications must be returned to Mosby-Year Book, I n c , by October 15 preceding the listing year to ensure publication. Additional forms will be supplied on request from the Journal Editing Department, Mosby-Year Book, Inc., 11830 Westline Industrial Drive, St. Louis, MO 63146-3318/800-325-4177, ext. 4317, or 314-453-4317.
Objective: To quantitate airway muscle changes in infants born at 23 to 41 weeks' gestation (control subjects) and to compare the changes with those in infants with chronic lung disease. Methods: Fifty-five human lungs (from infants born at 23 to 41 weeks' gestation) were studied: 46 from infants who died of various diseases within 72 hours of birth, and 9 from infants with CLD (infants born at 26.9 ± 0.5 weeks' gestation, who lived 17 ± 8 days). All the lungs were perfused via the trachea and pulmonary artery in a standardized protocol. Formalin-fixed tissues in paraffin blocks were cut 5 pm thick. Sections were immunohistochemically stained for =-smooth muscle actin, By using computerized image analysis to quantitate images digitized into the computer, we measured the area of muscle, epithelium, airway lumen, and length of basement membrane in 18 airways, from the smallest bronchioles to bronchi, in each infant. Results: Muscle was present at 23 weeks' gestation at all levels of the bronchial tree, and from 25 weeks to term the control lungs had a similar quantity of muscle at any given airway circumference. Relative to airway size, there was more muscle in small airways, less than 1000 IJm in circumference, than in larger airways. In airways greater than 1500 IJm in circumference, infants with CLD had significantly more muscle than did control lungs. Conclusions: Airway muscle is present at 23 weeks' gestation at all levels of the conducting airways. The 25-week gestation infants had a quantity of airway muscle relative to airway circumference similar to that of term infants. Preterm infants with CLD who were aged 9 to 29 days have increased airway muscle in airways greater than 1500 pm in circumference. Bronchospasm in very low birth weight infants is possible within the first days of life. (J Pediatr 1997;130:570-6) Lack of information on the development of airway muscle confounds the interpretation of postnatal muscle changes in infants at risk of having bronchopulmonary dysplasia or chronic lung disease. In infants with BPD the rationale for bronchodilator therapy is supported by a decrease in pulmoSupported by the Katharine B. Richardson Foundation and a Physician Scientist Award (Dr. Truog) from The Children's Mercy Hospital. Submitted for publication April 25, 1996; accepted Sept. 27, 1996. Reprint requests: Donald W. Thibeault, MD, Children's Mercy Hospital, 2401 Gillham Rd., Kansas City, MO 64108. Copyright © 1997 by Mosby-Year Book, Inc. 0022-3476/97/$5.00 + 0 9121178375
570
nary resistance after its use, as well as by clear anatomic evidence of increased airway muscle mass. ]4 In contrast, in very low birth weight infants receiving assisted ventilation during the first 2 weeks of life, there is limited evidence BPD CLD PBSS
Bronchopulmonary dysplasia Chronic lung disease Phosphate-buffered saline solution
documenting the efficacy of bronchodilator therapy.I' 5 Anatomic studies demonstrate a lack of adequate airway muscle distal to the small bronchi) A detailed, systematic study of the ontogeny of bronchial tree smooth muscle in relatively normal preterm infants could not be found. 3' 6 The purpose
The Journal of Pediatrics Volume 130, Number 4
Sward-Comunelli et al.
571
] ' a b l e . Clinical characteristics of infants living less than 72 hours compared with those with chronic lung disease Groups (wk or CLD)
Gestational ages (wk) Birth weight) (gm) Days lived* Primary cause of death'~ IRDS PDA Pulmonary hypertension PIE MAS Respiratory failure Asphyxia Septic shock NEC Intracranial hemorrhage Other hemorrhage Arrhythmia Drug treatment Surfactant Prenatal steroids Thyroid-releasing hormone
23-24 (n = 7)
25-26 (n = 10)
27-30 (n = 10)
31-36 (n = 8)
37-41 (n = 1 I)
CLD (n = 9)
23.6 -+ 0.5 578 -+ 76 0.9 -+ 0.7
25.4 -+ 0.5 759 -+ 115 0.9 +- 0.9
28 -+ 0.6 1020 -+ 153 1.5 + 1.0
33.1 -+ 2.1 2057 -+ 775 0.6 -+ 0.5
39.4 -+ 1.4 3199 -+ 928 1.6 -+ 1.1
26.9 -+ 1.5 841 -+ 276 17 +- 8
3 0 0 1 0 0 1 0 1 0 0 1
4 0 0
6 1 0
1 1 0
1 0 2
0 0 0
1
0
0
0
0
0 0 2 0 0 2
0 0 0 0 0 2
0 0 3 2 0 0
4 0 2 0 0 1
0 7 0 1 0 0
1
1
0
1
0
0
0
1
0
1
3 3 0
6 4 0
7 4 0
5 3 0
0 0 0
6 4 0
Values for first three rows are expressed as mean _+SD. IRDS, Idiopathic RDS; PDA,patent ductus arteriosus; PIE, pulmonary interstitial edema; NEC, necrotizing enterocolifis;pPHN, persistent pulmonary hypertension of the newborn. *Not significantly different among groups without chronic lung disease. -~Acomparison of muscle area versus airway circumference in infants with PPHN or MAS was not different from those with other causes of death.
of this study was to quantitate airway muscle mass in 23- to 41-week gestation infants who died within 72 hours of birth and to compare the findings with those in infants with CLD. METHODS
Fifty-five human infants, born at 23 to 41 weeks of gestation and appropriate in size for gestational age, were studied (Table I). Forty-six infants died at between 1 and 72 hours (28 _+ 23 hours). The 9 infants with CLD were born at 26.9 _+ 0.5 weeks' gestation and lived 17 _+ 8 days (range, 9 to 29 days). All infants with CLD received mechanical ventilation with a pressure-limited ventilator and required increased inspired oxygen concentrations throughout life; 7 died of respiratory failure. BPD was diagnosed in all these infants after death, on the basis of altered alveolar inflation patterns, airway changes, and fibroproliferative acinar changes. W e used the clinical term chronic lung disease rather than BPD because of the •relatively short life span. The time from death to autopsy was 13 _+ 11 hours. Lungs with widespread bronchopneumonia or leaks that could not maintain the perfusion pressure were excluded from the study. The lungs were wanned to 38 ° C, and the pulmonary artery was injected with a 60 ° C radiopaque barium-gelatin mixture (4 minutes, 100 cm H 2 0 pressure). 7' 8 The lungs, via
the trachea, were then perfused 72 hours at 24 cm H20 with 10% buffered formalin. The following sections were cut: an entire cross section of the left upper lobe; the base of the lingula; a 1.5 x 1.0 cm piece of the left lower lobe (one side containing pleura); and a longitudinal hilar cut (1.5 x 1.0 cm), including the central hilar structures surrounded by parenchyma. After immersion in 10% buffered formalin, the area of the cut surface of each section was measured by computerized image analysis (Optimas Corp.~ Edmonds, Wash.). The tissue samples were processed and embedded in paraffin. Sections, 5 pm thick, were stained with hematoxylin and eosin and with M i l l e r - v a n Gieson. Additional 5 pm thick sections were mounted on poly-Llysine-coated glass slides and immunohistochemically stained for actin. A kit (Sigma Immunochemicals, St. Louis, Mo.) using the avidin-biotin complex method 9 was used immunohistochemically to demonstrate s - s m o o t h muscle actin. The mouse monoclonal primary antibody used specifically identifies the smooth muscle actin ~ isotype, reacting with normal and neoplastic smooth muscle and myoepithelial cells. 1° Sections were deparaffinized, rehydrated, and treated with 3% hydrogen peroxide for 5 minutes to remove endogenous peroxidase; they were then washed in phosphate-buffered saline solution (pH 7.4) and incubated first
572
Sward-Comunelli et al.
The Journal of Pediatrics April 1997
Fig. 1. A bronchiole with a diameter of 263 pm from an infant who was born at 27 weeks' gestation and died of respiratory distress syndrome at age 23 hours. Arrow, Dark band of immunohistochemically stained airway muscle actin (x200).
25 WEEKS GA
CHRONIC LUNG DISEASE BRONCHUS
BRONCHUS
v
PRETERMINAL BRONCHIOLE
PRETERMINAL BRONCHIOLE
.13
MUSCLE AREA
.22
173
DIAMETER (~m}
179
Fig. 2. Airways from two representative 25-week-gestation infants. Airways on the left represent an infant who died of RDS at age 18 hours. Those onthe right represent an infant with CLD who died at 14 days. The ratios were computed by dividing the three areas (muscle, epithelial, and airway luminal) by the area bounded by the perimeter of the epithelial basement membrane (BM). The muscle and epithelial areas are larger, relative to airway size, in smaller, preterminal bronchioles than in bronchi.
with normal goat serum for 10 minutes and then with primary antibody for 60 minutes, After being rinsed with PBSS, sections were incubated for 20 minutes with biotinylated secondary antibody, washed with PBSS again, and incubated for
20 minutes with avidin-conjugated peroxidase reagent. Antigenic sites were visualized by addition of the chromogen 3,3'-diaminobenzidine, which formed a brown precipitate. Slides were counterstained with methyl green (Fig. 1). Neg-
~,0
100000
80000
/k • • O [] •
CLD 37-41 WKS 31-36 WKS 27-30 WKS 25-26 WKS 23-24 WKS
4~
CLD 0.8
o
v
60000
°
o
•
•
#e
Oo
O
04
<
2
< ©
23-41 WKS
06
E
-~
•
Z~x o
0
£
57 3
Sward-Comunelli et al.
The Journal of Pediatrics Volume 130, Number 4
:s
0.2
40000 0.0
20000 Airway Area (um2X 106
i
i
1000 2000
i
30'00 4000
i
i
5000 6000
"7000
Airway Circumference (um) Fig. 3. Airway muscle area increases in all groups nonlinearly as airway circumference increases. Data overlap in all control groups. Infants with CLD have increased muscle in airways greater than 1500 gm in circumference. Data from all infants, born at 25 to 41 weeks' gestation, were best fit to the polynomial equation: Muscle area = 1040 + Airway circumference + 10 x 1.28.3 x Airway circumference2 (r = 0.945; p <0.0001). In the 23- to 24-week group, there is a trend toward less muscle area in airways larger than 4000 Wn in circumference.
ative control slides were stained by using the same procedure, with the primary antibody omitted or a nonimmune serum added. In two lungs, comparisons were made of eight bronchial muscle areas measured on sequential 5 gm sections by using the actin immunohistochemical stain and the Gomori Irichrome stain. In large airways the two techniques produced similar muscle-area measurements. For small bronchioles, the actin stain was more discrete and far superior for image analysis. Therefore actin stains were used for all measurements reported in the Results section, below. Staining of the same bronchiolar area, in 5 p r n serial cuts (n = 6), revealed an airway muscle area of 14,879 -+ 2177 N n 2, indicating that the reproducibility of measuring the muscle area was -+ 14%. However, because the muscle spirals, each serial section is not identical to the previous one; thus some of the measurement variation is biologic. The area of sections immunohistochemically stained for actin was compared with the area of the cut surface before
Fig. 4. Nonlinear increase in airway muscle area as the airway area increases. Airway area is the area bounded by the epithelial basement membrane and includes the epithelial and luminal areas. The data overlap on infants born at 23 to 41 weeks' gestation (open circles). The best-fit polynomial equation is as follows: Muscle area = 0.1449 + (0.027 x Airway area) - (3.38 x 10.9 x Airway area)2 (r = 0.952; p <0.001). Infants with CLD had increased muscle area compared with control infants but only in airways greater than 1500 Nn in circumference.
processing to determine the shrinkage correction factor. In the 27- to 30-week group, muscle areas from infants who died within 24 hours of birth were compared with muscle areas from those who died after 25 to 72 hours, to determine whether the muscle,mass changed with time. Eighteen airways throughout the respiratory tree, from preterminal bronchioles to cartilaginous bronchi, were measured in each study infant. Measurements were made by investigators masked to clinical history. With the use of image analysis, the epithelial basement membrane perimeter was traced by mouse. The circumference, length, and breadth of each airway were calculated by computer. Airways whose length-to-breadth ratio was equal to or greater than 2 were considered to be cut obliquely and were not analyzed. Color thresholds were set for the brown-stained muscle actin and the white airway lumen. The computer calculated the following areas, schematically shown in Fig. 2: muscle, region within the perimeter of the epithelial basement membrane, and airway lumen. Epithelial area was obtained by subtraction. The epithelial and muscle areas were standardized for each airway size by dividing by airway circumference (length of the epithelial basement membrane). This yields the effective mean muscle and epithelium thickness. Basement membrane length was used as a standardized measure of airway size because it does not change with
574
Sward-Comunelli et al.
The Journal of Pediatrics April 1997
30
90 • CLD - -I- . 37-41 WKS
25
:;:
.
,m • ••
o o
-
80
- 25-26 WKS - -~, - 23-24 WKS -~
oo°°
-
CLD
o
oO 70 X
X
20
[
••
<
0
:
<.-~ 15
0
0
°8°
°o
50
•
40
oj
<
A!~B%oo o 0 o 9 b
{.l::" 4~
-.
O~q6:Yo
"
,3~:"~ ~_.--_.._i ". .-
o-,o o r~
30
"
• 20 . . . .
:.: 0
1000
2000 3000 40'00 5~0
.
~oOo
~ o
10
6000 7000
Airway Circumference (um)
o o o
i
0
1000
2000
3000
4000
5000
6000
7000
Airway Circumference (um)
Fig. 5. The muscle area/airway area ratio is exponentially and inversely related to airway circumference. As circumference decreases to less than 1000 gm, muscle area increases relative to airway area. There are no differences among the control groups. However, the ratio is greater in infants with CLD in airways greater than 1500 #in in circumference.
bronchocons~iction, n The effective mean thickness represents the muscle thickness that would be seen in completely relaxed airways or airways distended at 24 cm H20. Muscle and epithelial areas also were standardized for airway size by dividing by the' total area internal to the epithelial basement membrane (i.e., airway area, which is the epithelial plus luminal area). Statistical analysis, Values are given as mean + SD in the Table. Statistical analysis in the Table and text was performed with t tests or analysis of variance when there were more than two groups. A value of p <0.05 was considered significant. Regression analysis and best-fit equations were determined by computer analysis (SigmaStat; Jandel Scientific, San Rafael, Calif.).
RESULTS Absolute airway muscle area in infants of 23 to 41 weeks' gestation increases nonlinearly from the smallest bronchiole to the bronchial level (Fig. 3). In infants born at 23 to 24 weeks of gestation, airways greater than 4000 grn in circumference trend toward less muscle than do those from
Fig. 6. Epithelial area/airway area ratio is nonlinearly and inversely related to airway circumference. The control data points (open circles) show wide scatter. The best-fit equation for the control data is as follows: Epithelial area/Airway area x 100= 384 x Airway circumference qz3 (r = 0.644; p <0.001). Infants with CLD were not different from control subjects. later-gestation infants. In infants with CLD, airway muscle was greater than control airway muscle at an airway circumferences greater than 1500 pm. In smaller airways the 95% confidence limits of control and CLD data overlap. Muscle area as a function of airway area also was nonlinear (Fig. 4). Infants with CLD had increased muscle compared with control lungs in airways greater than 1500 pm in diameter. The steeper slope in the smaller airways shown in Fig. 4 indicates that there is more muscle relative to airway size in small airways. This can be better visualized in Fig. 5, in which the muscle area/airway area ratio is plotted against airway circumference. In control lungs, muscle area relative to the airway area increases sharply in airways smaller than 1000 pm in circumference. In larger airways the muscle area/airway area ratio remains steady at 0.02 to 0.04. Control data overlapped at all airway sizes and best fit the exponential equation: Muscle area/airway areax 100=406e225/airway circumference (r=0.720; p <0.0001). In infants with CLD the muscle area/airway area ratio is increased compared with control airways geater than 1500 wn in circumference. CLD lung ratios tend to be greater in smaller airways
The Journal of Pediatrics Volume 130, Number 4
but overlap the 95% confidence limits of the control data points. The epithelial thickness linearly increases at all ages as the airway circumference increases: Epithelial thickness= 12.3 + 0.015 x Airway circumference (r = 0.824; p <0.0001). The data, similar in all groups, including infants with CLD, should be interpreted cautiously because 28% of airway measurements were excluded because of lost or severely damaged epithelium. Similar to muscle, the ratio of epithelial area to airway area was nonlinear, with the epithelium constituting a relatively larger area in small airways (Fig. 6). Bronchial muscle area measurements on sequential cuts with the Gomori trichrome stain were not significantly different from measurements from immunohistochemically actin-stained slides. Muscle measured in infants who died within 24 hours of birth did not significantly differ from that measured in infants 25 to 72 hours of age. Prenatal steroid treatment did not change the relation of muscle area to airway circumference in any group. DISCUSSION The results demonstrate that newborn infants from 23 weeks' gestation to term have airway muscle from the smallest preterminal bronchiole to the bronchi. Furthermore, this muscle mass, relative to airway size, is the same at 25 weeks' gestation as it is at term. As in adults, it increases as airway size increases, and relative to airway size there is more muscle in smaller airways. 12 Preterm infants with early or established histol0gic BPD changes, who died after prolonged ventilator assistance, had more muscle in all conducting airways, particularly where the circumference was greater than 1500 gin. Increased airway muscle in BPD has been shown repeatedly,3,4, i3, 14 but we show that it can occur within 10 days of birth, before a clinical BPD diagnosis is established. However, it is not entirely clear whether ventilator-dependent preterm infants without CLD radiographic changes, who require low inspiratory pressures and minimal oxygen, have increased muscle in conducting airways. 3, 13, ~5 Nevertheless, the muscle mass found in 23-week to term control infants is compatible with that reported in studies demonstrating increased airway resistance and response to bronchodilators before BPD development. 1,2,5 Increased airway resistance in small preterm infants may be a function of both muscle mass and epithelial thickness, both of which are increased relative to airway size in small conducting airways. This is schematically shown in Fig. 2. In both control infants and the infants with CLD, 28% of all airways, particularly the small bronchioles, have damaged or denuded areas of epithelium. These damaged areas may also be sites of inflammation and secretions that could further increase air-
Sward-Comunelli et aL
575
way reactivity, narrowing, and resistance. The demonstration of muscle in the most-immature infants allows the possibility of bronchoconstriction and may further justify the study and use of bronchodilators despite their potential side effects. The presence of airway muscle so early in gestation suggests that muscle may serve an important function in lung development. The finding of relatively more muscle in small airways, compared with larger, upper airways (Fig. 5) throughout development suggests that the role of fetal muscle in lung growth may differ in various parts of the tracheobronclfial tree. It is known that fetal lung development is, in part, regulated by lung liquid pressure acting against a resistance to outflow in the upper airways. 16' 17Muscle in large and small fetal airways can undergo relaxation and contraction, changing overall intralung airway pressure and thereby influencing growth. 18,19 The relatively increased muscle mass at the small-bronchiolar level could differentially regulate intraacinar fluid pressure in the lung, ensuring symmetric growth. 19-22 Smooth-muscle tone also may provide stability in th e conducting airways surrounded by the fetal fluid-filled air sacs. This would be particularly important in smaller, more compliant airways. 23 Use of the c~-smooth muscle actin antibody to determine muscle area gave results similar to those of traditional muscle stains. However, in small airways the actin stain is superior for image-analysis quantitation. With lrichrome staining of small airways, it is often difficult or impossible to separate the muscle from the surrounding tissue. Tissue stained for muscle, or antibody-identified muscle actin, is not guaranteed to be functional muscle. Nevertheless, our data are compatible with, and extend in important ways, previous anatomic and embryologic studies, as well as clinical studies. REFERENCES
1. Motoyama EK, Fort MD, Klesh KE, Mutich RL, Guthrie RD. Early onset of airway reactivity in premature infants with hronchopulmonary dysplasia. Am Rev Respir Dis 1987;136: 50-7. 2. Wilkie RA, Bryan MH. Effect of bronchodilators on airway resistance in ventilator-dependent neonates with chronic lung disease. J Pediatr 1987;111:278-82. 3. Hislop AA, Haworth SG. Airway size and smacture in the normal fetal and infant lung and the effect of premature delivery and artificial ventilation. Am Rev Respir Dis 1989;140:171726. 4. Bonikos DS, Bensch KG, Northway WH, Edwards DR. Bronchopulmonary dysplasia: the pulmonary pathologic sequel of necrotizing bronchiolitis and pulmonary fibrosis: Hum Pathol 1976;7:643-66. 5. Gomez-Del Rio M, Gerhardt T, Hehre D, Feiler R, Bancalari E. Effect of a beta-agonist nebulization on lung function in neonates with increased pulmonary resistance. Pediatr Pulmonol 1986;2:287-91. 6. O'Brodovich HM, Mellins RB. Bronchopulmonary dysplasia;
576
7.
8.
9.
10.
11.
12.
13. • 14.
Sward-Comunelli et al.
The Journal of Pediatrics April 1997
unresolved neonatal acute lung injury. Am Rev Respir Dis 1985;132:694-709. Meyrick B, Reid L. Pulmonary arterial and alveolar development in normal postnatal rat lung. Am Rev Respir Dis 1982; 125:468-73. Downing GJ, Thibeault DW. Pulmonary vasculatnre changes associated with idiopathic closure of the ductus arteriosus and hydrops fetalis. Pediatr Cardiol 1994;15:71-5. Hsu SM, Raine L, Fanger H. Use of avidin-biotin peroxidase complex (ABC) in irrlmunoperoxidase techniques: a comparison between ABC and unlabeled antibody (PAP) procedures. J Histochem Cytochem 1981;29:577-80. Bussolati G, Gugliotta P, Fulcheri E. Immunohistochemistryof actin I normal and neoplastic tissues. In: DeLellis RA, editor. Advances in immunochemistry.New York: Masson, 1984:325. Ebina M, Yaegashi H, Takahashi T, Motomiya M, Tanemura M. Distribution of smooth muscles along the bronchial tree: a morphometric study on ordinary autopsy lungs. Am Rev Respir Dis 1990;141:1322-6. Matsuba K, Thurlbeck WM. A morphometric ~study of bronchial and bronchiolar walls in children. Am Rev Respir Dis 1972;105:908-13. Taghizadeh A, Reynolds EOR. Pathogenesis of bronchopulmonary dysplasia following hyaline membrane disease. Am J Pathol 1976;82:241-58. Hislop AA, Haworth SG. Pulmonary vascular damage and the development of cor pulmonale following hyaline membrane disease. Pediatr Pulmonol 1990;9:152-61.
15. Sward-Comunelli SL, Mabry SM, Truog WE, Thibeault DW. Airway muscle in preterm infants: changes during development [abstract]. Pediatr Res 1995;37:353A. 16. Moessinger AC, Harding R, Adamson TM, Singh M, Kiu GT. Role of lung fluid volume in growth and maturation of the fetal sheep lung. J Clin Invest 199i3;86:1270-7. 17. Harding R. The upper respiratory tract in perinatal life. In: Johnston BM, Gluckman PD, editors. Respiratory control and lung development in the fetus and newborn. Ithaca (NY): Perir~atology Press, 1986:331-76. 18. Isabel JB, Towers B, Adams FIt, Gyepes MT. The effects of ganglionic blockade on tracheobronchial muscle in fetal and newborn lambs. Respir Physiol 1972;15:255-67. 19. Sparrow MP, Warwick SP, Mitchell HW. Foetal airway motor tone in prenatal lung development of the pig. Eur Respir J 1994;7:1416-24. 20. Hooper SB, Hart VKM, Harding R. Changes in lung expansion alter pulmonary DNA synthesis and IGF-II gene expression in fetal sheep. Am J Physiol 1993;265 Lung Cell Mol Physiol 9:L403-9. 21. McCray PB. Spontaneous contractility of human fetal airway smooth muscle. Am J Respir Dis 1993;8:573-80. 22. Sparrow MP, Warwick SP, Everett AW. Innervation and function of the distal airways in the developing bronchial tree of fetal pig lung. Am J Respir Dis 1995;13:518-25. 23. Bhutani VK, Koslo ILl, Shaffer TH. The effect of tracheal smooth muscle tone on neonatal airway collapsibility. Pediatr Res 1986;20:492-5.
FELLOWSHIPS Fellowships available in pediatric subspecialties and those for general academic pediatric training are listed once a year, in J a n u a r y , in The Journal of Pediatrics. Each June, forms for listing fellowships available for the academic year beginning 18 months after publication are sent to the Chairman of the Department of Pediatrics at major hospitals in the United States and Canada. In addition, a copy of the application form appears in the July, August, and September issues of The Journal (please use the current form). Should you desire to list fellowships, a separate application must be made each year for each position. All applications must be returned to Mosby-Year Book, I n c , by October 15 preceding the listing year to ensure publication. Additional forms will be supplied on request from the Journal Editing Department, Mosby-Year Book, Inc., 11830 Westline Industrial Drive, St. Louis, MO 63146-3318/800-325-4177, ext. 4317, or 314-453-4317.