Tracheomalacia with esophageal atresia and tracheoesophageal fistula in fetal rats

Tracheomalacia

With

By Bao Quan Qi, Jamal

Merei,

Esophageal Atresia and Tracheoesophageal Fistula in Fetal Rats Pam Farmer, Suzanne Hasthorpe, and Spencer W. Beasley Melbourne,

Background: Many patients who have esophageal atresia and tracheoesophageal fistula (EA-TEF) have associated tracheomalacia, which is thought to be one of the reasons for respiratory complications after surgical correction of the abnormality. Methods: In this study, tracheas from Adriamycin-induced EA-TEF fetal rats were examined histologically and relevant cross-sectional parameters of the tracheas were measured. Resu/ts:The tracheal lumen in tracheomalacia was small and irregular, losing its normal “D” shape. In most rats, the cartilaginous ring was broken into two to four segments, making the trachea lose its rigid support. The submucosa was thickened with prominent bulging of its membranous part into the tracheal lumen. The ratio of the inner luminal cross-sectional area to the outer tracheal cross-sectional area in EA-TEF rats was 15.7%, compared with a control ratio of 47.2%. In EA-TEF rats, the length of the cartilaginous ring was

T

HE POSTOPERATIVE COURSE of many patients who have esophageal atresia and tracheoesophageal fistula (EA-TEF) is complicated by the effects of tracheomalacia,‘-4 and in some infants may be life threatening. In a review of the records of 76 patients who had EA-TEF, Benjamin found that 20 of 21 patients in whom respiratory symptoms developed had tracheomalacia confirmed by bronchoscopy as well as by contrast radiography or cineradiography.5 Because the trachea and esophagus have a common origin, namely the primitive foregut,6.7 one would expect that the same factors that induce the development of EA and TEF might also exert a detrimental influence on the development of the respiratory system. Alternatively, the presence of EA-TEF itself may be responsible for tracheomalacia. To see whether this occurs, EA-TEF was induced in fetal rats by Adriamycin as previously described.8 The tracheas were studied histologically and the relevant parameters were measured to determine the degree to which tracheomalacia was an accompaniment of EA-TEF and to elucidate the structural characteristics of the abnormality. MATERIALS The method of inducing previously.* Timed-pregnant Animal Research Laboratory Journal

ofPediatric

AND

METHODS

EA-TEF in fetal rats has been described Sprague-Dawley rats maintained in the at the Royal Children’s Hospital were

Surgery,Vol32,

No 11 (November),

1997: pp 1575-1579

John

M. Hutson,

Nate A. Myers,

Australia

significantly shortened (P < .OOl), but not the length of membranous trachea, thus resulting in a cartilaginous/ membranous (C/M) ratio of 1.55:1, markedly lower than that of normal rats (4.34:1, P< .OOl). The reduction of anteriorposterior diameter of the tracheal lumen was more marked than that of the transverse diameter. Conclusions: These observations suggest that the trachea in EA-TEF rats has a smaller lumen and is more flaccid than normal, making it prone to airway obstruction. The fact that tracheomalacia developed only in fetuses who had EA-TEF indicates that the factors that result in EA-TEF also cause tracheomalacia. J Pediatr Surg 32: 1575- 1579. Copyright o 1997 by W. B. Saunders Company. INDEX WORDS: Tracheomalacia, sia, tracheoesophageal fistula,

cartilage, esophageal atrerat model, Adriamycin.

injected intraperitoneally with Adriamycin at a dose of 2 mg/kg body weight from day 6 to 9 of gestation. The pregnant dams were kept in an air-conditioned, 12-hour light/dark cycle and fed regular rat chow and tap water ad libitum. The fetuses were retrieved by cesarean section on day 21 of gestation. Two fetuses were picked randomly from each litter until a total of 12 treated fetuses were collected. Six control fetuses were obtained from dams that received only normal saline injections during the corresponding period of gestation. The fetuses were fixed in Bouin’s solution for 48 hours and cut transversely. from neck to upper abdomen, into three to four blocks. Each block was embedded in paraffin and sectioned at 5 pm thickness. Every tenth section was mounted and stained with Hematoxylin and Eosin. All histological slides were examined carefully. with special attention to the presence or absence of EA-TEF and to the histological structure of the trachea. Cross sections of the trachea with complete or near-complete cartilaginous rings were chosen and projected through the microscope with a photographic enlarger (Leitz Wetzlar, Germany) at a constant magnification of 100 X. The images were traced onto paper by a calibrated map measure, and then measured using Sigma Scan computer software (Jandel Scientific. Corte, Madera, California). The following parameters were observed and analyzed: (I) the area of

From the F: Douglas Stephens Surgical Research Laborator), Royal Childveil S Hospital, Melbourne, Australia. Presented at the 30th Annual Meeting qf the Pac$ic Association of Pediatric Surgeons, Phoeni.x, Arizona. May 9-13. 1997. Address reprint requests to Spencer W Beasley, Department of Surgeq Christchurch Hospital, Private Bag 4710, Christchurch, New Zealand. Copyright o 1997 by WB. Sawlders Company 0022.3468/97/3211-0013$03.00/O 1575

1576

QI ET AL

the tracheal lumen (mm2); (2) the cross-sectional area of the trachea; (3) ratio of inner luminal area to tracheal cross-sectional area; (4) the length of (inner border of) the cartilaginous ring (C, mm); (5) the length of the membranous part of the trachea (M); (6) ratio of the length of the cartilaginous ring to the length of the membranous trachea (C/M ratio); (7) the anterior-posterior diameter of the tracheal lumen (APD); (8) the transverse diameter of the tracheal lumen (TD), and (9) ratio of APDITD. The quantitative data between the different groups were analyzed statistically by unpaired Student’s t test. When P < .05, the difference was considered significant.

RESULTS

Esophageal atresia with a distal tracheoesophageal fistula (EA-TEF) was confirmed histologically in 8 of 12 Adriamycin-treated fetuses (67%), identical to the abnormality described previously.* No abnormality was found in any of the six control fetuses. Transverse sections of the trachea and esophagus from the control animals showed normal configuration of both structures. The trachea lies in front of the esophagus with a characteristic “D” shape and a “C”-shaped cartilaginous ring in its wall. The thin ciliated epithelium and submucosa layer attaches to the inner surface of the cartilage and its membranous part bulges slightly into the trachea lumen. The membranous muscle bundle seals the mouth of the C-shaped cartilaginous ring (Fig 1). The horizontal sections of the trachea from the four Adriamycin-treated animals in whom EA-TEF did not develop showed the same histological appearance as their control counterparts, although the dimensions of their tracheas were reduced in proportion (Table 1). However, in those Adriamycin-treated fetuses in whom EA-TEF developed, the configuration of the trachea became irregular, losing its D shape. In addition, it showed a significant reduction in size: In severe cases, the shape of the tracheal lumen was highly irregular or slitlike with malformation or distortion of the cartilaginous ring. In many cases, the cartilage was broken into two to four segments with the median or the main piece of cartilage “compressing” the anterior wall of the trachea (Fig 2). Increase in the thickness of submucosa resulted in prominent bulging of its membranous part into the tracheal lumen, reducing its anterior-posterior diameter. When the cartilaginous ring was complete, its horizontal length was greatly reduced and the length of the membranous trachea in the coronal plan was increased. There was no abnormality in the lining epithelium. Squamous epithelium was not found in any sections of the tracheas studied (Fig 3). Parameters measured from cross sections of the trachea are shown in Table 1. The length of both the cartilaginous ring and the membranous trachea of fetuses without EA-TEF (NO-EA group) were reduced proportionally, compared with the control group. The C/M ratio

Fig from with layer

1. The transverse section of the trachea (T) and esophagus (E) a normal control fetus. The trachea is a characteristic D shape a C-shaped cartilaginous ring in its wall and a thin submucosa attached to its inner surface. (Original magnification x 100.)

was 4.26:1, similar to that of control fetuses (4.34:1). However, in the EA-TEF group, the length of the cartilaginous ring was significantly shortened and the length of the membranous trachea relatively increased compared with the other two groups. This resulted in a C/M ratio of 1.55:1 in the EA-TEF group, which was significantly lower than that of the other two groups (P < .OOl). In the Adriamycin-treated NO-EA group, there was a proportionate decrease in both the anterior-posterior diameter (APD) and transverse diameter (TD) resulting in a APD/TD ratio of 0.718, which was similar to that of the control rats (0.715). Although the TD of the trachea in EA-TEF animals was substantially less than that in NO-EA group, the reduction of APD was even more marked, resulting in a APD/TD ratio significantly lower than that of control and NO-EA animals (P < .OOl), implying that the anterior and posterior tracheal walls of EA-TEF rats closely approached each other, leaving little space for air passage. The ratio of inner luminal area to outer area of the cross sections of the trachea was 47.2%

TRACHEOMALACIA

AND

ESOPHAGEAL

1577

ATRESIA

Table

Controlh = 110) NO-EA (n = 60) EA-TEF (n = 177)

1. Parameters

from

Tracheal

Cross

Sections

1n.A (mm21

0ut.A (mm21

In/Out

C-L (mm)

M-L (mm)

C/M

APD (mm)

TD (mm)

APD/TD

0.249 + 0.043 0.127 ? 0.021 0.026 i; 0.019

0.527 t 0.041 0.327 + 0.045 0.158 2 0.053

0.472 L 0.059 0.389 2 0.025 0.157 + 0.069

1.644 i 0.087 1.233 i 0.087 0.625 + 0.205

0.387 + 0.059 0.296 + 0.074 0.413 2 0.074

4.342 r 0.661* 4.263 ? 0.781* 1.553 2 0.542

0.448 2 0.057 0.312 -t 0.067 0.067 2 0.033

0.631 2 0.057 0.435 i; 0.021 0.291 2 0.089

0.715 2 0.104" 0.718 i 0.153" 0.238 2 0.099

NOTE. n refers to the number of sections examined histologicaily (Number of fetuses, Control. 6; NO-EA, 4; EA-TEF, 8). Abbreviations: IRA, inner luminal area of trachea; Out.A, outer area of cross-section of trachea; C-L, the length of cartilaginous ring; M-L, the length of membranous trachea. *P> .05. when comparisons were made between the two groups. In the remaining groups, P< ,001, when comparisonswere made among them.

and 38.9% for control and NO-EA animals, respectively. But in EA-TEF fetuses this ratio was only 15.7%, reflecting a significant reduction of inner luminal area and the relative increase of thickness of the tracheal wall (Table 1). DISCUSSION

Tracheomalacia is a structural and functional weakness of the trachea that results in partial respiratory obstruction in infants who have EA-TEE9 Respiratory complications develop in about 20% to 30% of patients who have EA-TEF after surgical correction of their esophageal

Fig 2. Transverse section from a EA-TEF fetus. The esophagus is absent at this level. The shape of the trachea looks bizarre. The ring of cartilage has lost its continuity and is broken into three pieces. T, trachea. (Original magnification x100.)

anomaly.‘~z~5Tracheomalacia is thought to be a significant factor contributing to these respiratory complications. The diagnosis of tracheomalacia initially is a clinical one, supported by endoscopic observations and cinetracheobronchography with contrast and CT scan.5Jo-12The structural abnormality underlying tracheomalacia is largely unknown, mainly because it is difficult to obtain adequate pathological material from affected patients. This EATEF rat model may provide a useful tool to investigate

Fig 3. Transverse section from a EA-TEF fetus. The tracheal submucosa is thickened resulting in prominent bulging of its membranous part into the tracheal lumen and reducing the area of the lumen. The length of cartilaginous ring on cross section is decreased and the length of membranous trachea relatively increased, leading to a reduced C/M ratio. T, trachea: V, vertebra. (Original magnification x 100.)

1578

the tracheal abnormalities that accompany esophageal atresia. Our data show that the C/M ratio is reduced from 4.3: 1 in normal control rats to 1.5: 1 in EA-TEF rats. This results from a shortened ring of cartilage and a widened posterior membranous part in EA-TEF animals, with a consequent loss of the normal rigid support of the tracheal wall. This explains why the trachea becomes flaccid and can collapse easily in the presence of positive intrathoracic pressure, leading to obstruction of the upper airway. This is believed to be the mechanism of the wheezing, strider, and pathognomonic seal bark cough characteristically seen in tracheomalacia patients.5,9J1 Usui et aL5 using bronchoscopy, examined the trachea in 32 patients who had EA-TEF, and found that ratio of circumferential length of cartilaginous trachea to circumferential length of membranous trachea (C/M ratio) was 1.7: 1 in those patients who had respiratory complications, compared with 3.0: 1 in those who had no complications.5 Wailoo and Emery’sI series of postmortum histopathologic examinations of the trachea showed a C/M ratio in the normal trachea of 4.5: 1, which is in accordance with our findings in normal rats. In addition, they found in EA-TEF patients a significant reduction in the C/M ratio, mainly caused by a reduction in the length of the cartilage and a concomitant increase in the length of the transverse muscle of the tracheal ring.t3 This is similar to our findings in rats, The present study also demonstrated abnormal histopathologic changes in the trachea of EA-TEF rats. In most cases, the cartilaginous rings of affected rats displayed a loss of continuity with between two and four broken segments (Figs 2 and 3). Moreover, the cartilaginous segment on the anterior aspect of the trachea was collapsed backward causing the tracheal lumen to lose its characteristic D shape. In addition, there was an increase in the thickness of the tracheal submucosa. This was more noticeable in the membranous trachea where the mucosa and submucosa layers strikingly bulged into the lumen, further reducing the cross-sectional area of the trachea and producing some bizarre luminal shapes. It is not difficult to see how the narrowed tracheal lumen would tend to obstruct in the presence of edema from infection or repeated tracheal intubation, both of which may occur after surgical correction of EA-TEF in humans. Benjamin14 classified tracheomalacia in infants into two groups: the primary group included premature infants who had dyschondroplasia and the secondary group included infants who had EA-TEF or those with external compression from an anomalous innominate artery, vascular ring, or congenital cysts. It has been postulated previously that in the case of EA-TEF the extrinsic

01 ET AL

compressive force from the dilated and hypertrophied proximal esophageal pouch in early fetal life might adversely affect the development and maturation of the tracheal wall, and that the distal tracheoesophageal fistula would encourage tracheal collapse by its decompression effect, ie, by allowing the abnormal leakage of lung fluid into the esophagus with a resultant loss of its tracheal splinting effect. I1 This postulate does not fit with the observations of the present study for two reasons. First, the upper esophageal pouch in this EA-TEF model always ended near the level of the cricoid with no sign of dilatation or hypertrophy of the upper pouch. The concept of an extrinsic compression effect on the trachea from the upper pouch has not been confirmed from this model. Second, the abnormal decompression effect through the fistula did not seem to be a factor because most rat fetuses with EA-TEF also had duodenal atresia and a tiny collapsed stomach. As a result, the fluid had no outlet and decompression of the respiratory tract could not occur. It might be reasonable to consider that the occurrence of tracheomalacia in this model was more likely to be caused by a primary intrinsic abnormality rather than a secondary compressive effect from the upper esophageal pouch, or abnormal decompression through the fistula. It has been suggested previously that the tracheal abnormality in patients who have EA-TEF is an intrinsic one as evidenced by abnormal tracheal innervationI and squamous epithelium in the respiratory tract.r6 Because both the esophagus and respiratory tract develop from the primitive foregut, one may expect that the same detrimental factors that cause the occurrence of EA-TEF would also interfere with the normal development and maturation of the trachea. Interestingly, the fact that the abnormal reduction of C/M ratio and the abnormal histological findings of the trachea existed only in EA-TEF rats but not in NO-EA group in the present study suggests that this speculation may be true. Tracheomalacia is a common accompaniment of EA-TEF induced by Adriamycin in fetal rats. The main abnormalities of the trachea are a significant reduction of C/M ratio caused by decrease in the length of the cartilaginous ring and a relative increase in the length of the membranous trachea, loss of continuity of the cartilaginous ring, and prominent bulging of the mucosa and submucosa into the lumen, with a resultant reduction of the tracheal lumen. The fact that tracheomalacia occurred only in the fetuses affected by EA-TEF suggests that the factor causing EA-TEF also causes tracheomalacia. ACKNOWLEDGMENTS The authors thank the Esophageal Atresia Research Auxiliary and the Education and Research Fund of the Department of General Surgery, Royal Children’s Hospital, Melbourne, for supporting this project.

TRACHEOMALACIA

AND

ESOPHAGEAL

Al-RESIA

1579

REFERENCES 1. Leendertse-Verloop K, Tibboel D, Hazebroek FWJ, et al: Postoperative morbidity in patients with esophageal atresia. Pediatr Surg Int 2:2-s, 1987 2. Spitz L, Kiely EM, Brereton RJ: Esophageal atresia: Five year experience with 148 cases. J Pediatr Surg 22:103-108,1987 3. Blair GK, Cohen R, Filler RM: Treatment of tracheomalacia: Eight years’ experience. J Pediatr Surg 21:781-785, 1986 4. Kiely EM, Spitz L, Brereton R: Management of tracheomalacia by aortopexy. Pediatr Surg Int 2:13-1.5,1987 5. Benjamin B, Cohen D, Glasson M: Tracheomalacia in association with congenital tracheoesophageal fistula. Surgery 79:504-508,1976 6. O’Rahilly R, Miiller F: Respiratory and alimentary relations in staged human embryos-New embryological data and congenital anomalies. Ann Otol Rhino1 Laryngol93:421-429, 1984 7. Beasley SW: “Embryology,” in Beasley SW, Myers NA, A&list AW (eds): Oesophageal Atresia, London, England, Chapman & Hall Medical, 1991, pp 31-42 8. Diez-Pardo JA, Qi BQ, Tovar JA: A new rodent experimental model of esophageal atresia and tracheoesophageal fistula: Preliminary report. J Pediatr Surg 31:498-502, 1996 9. Spitz L, Phelan PD: “Tracheomalacia,” in Beasley SW, Myers

NA, Auldist AW (eds): Oesophageal Atresia, London, England Chapman&Hall Medical, 1991, pp 331-340 10. Usui N, Kamata S, Ishikawa S, et al: Anomalies of the tracheobronchial tree in patients with esophageal atresia. J Pediatr Surg 31:258-262, 1996 11. Davies MRQ, Cywes S: The flaccid trachea and tracheoesophageal congenital anomalies. J Pediatr Surg 13:363-367, 1978 12. Griscom NT, Martin TR: The trachea and esophagus after repair of esophageal atresia and distal fistula: Computed tomographic observations. Pediatr Radio1 20:447-450, 1990 13. Wailoo MP, Emery JL: The trachea in children with tracheooesophageal fistula. Histopathology 3:329-338, 1979 14. Benjamin B: Tracheomalacia in infants and children. Ann Otol Rhino1 Laryngol93:438-442, 1984 1.5. Nakazato Y, Wells TR, Landing BH: Abnormal tracheal innervation in patients with esophageal atresia and tracheoesophageal fistula: Study of the intrinsic tracheal nerve plexuses by a microdissection technique. J Pediatr Surg 21:838-844, 1986 16. Emery JL, Haddadin AJ: Squamous epithelium in respiratory tract of children with trachea-oesophageal fistula. Arch Dis Child 46~236-242.1971