Pathophysiology of congenital diaphragmatic hernia XI: Anatomic and biochemical characterization of the heart in the fetal lamb CDH model

Pathophysiology of Congenital Diaphragmatic Anatomic and Biochemical Characterization in the Fetal Lamb CDH Model By Hratch L. Karamanoukian, Philip L. Glick, Duncan T. Wilcox, Jon E. Rossman, and Richard G. Azizkhan Buffalo. New York e Aim: The purpose of this study was to determine whether the presence of bowel in the chest during development in the fetal iamb model of congenital diaphragmatic hernia (CDH) results in structural and/or biochemical hypoplasia of the left venticle. Methods: The model was created at 80 days’ gestation and delivered at term. The hearts were fixed in 4% formaldehyde solution, components weighed, and right ventricular (RV) and left ventricular (LV) wall thicknesses and both sot-tic (Ao) and pulmonary artery (PA) root diameters were measured. Fresh specimens were analyzed for protein, DNA, hydroxyproline, and elastin content. All CDH measurements are compared with littermate control tissues. Results: There were no differences in body weight (kg) between CDH and control littermates (4.25 * 0.26 versus 3.71 f 0.24, P = NS). CDH lambs have significantly decreased total heart (4.88 + .25* versus 6.75 f .49, P < .05), left ventricular (1.65 r .ll* versus 2.15 r .19, P < .05), septal (1.29 r .ll* versus 1.99 ? .21, P < .05), and combined atrial (0.68 -C .06* versus 1.14 f .15, P < .05) weights (g/kg lamb) without differences in RV weights (1.26 * .07 versus 1.57 Z!I .17, P = NS). LV and RV wall thickness, and Ao root diameters (cm) were found to be identical in both CDH and control lambs. However, PA root diameters (0.47 f .Ol* versus 0.38 k .Ol, P < ,005) and ductus arteriosus diameters were increased in CDH (0.35 f .Ol* versus 0.22 +- .02, P < .005). Total protein, DNA collagen, and elastin content and DNA/ total protein ratios were identical in RV and LV in both CDH and control lambs. Conc/usion: Newborn lambs with leftsided CDH have a significantly lower total heart, LV, septal, and atrial weights without differences of RV weight or ventricular wall thicknesses. Given these findings, the unchanged DNA/protein ratio implies that the left ventricle is hypoplastic in CDH. Ao/PA root ratios suggest that LV hypoplasia in utero may result in increased left atrial pressures, decreased right-to-left shunting through the foramen ovale, and increased PA pressures and flow, resulting in increased PA root and ductus arteriosus diameters. This model simulates the clinical data from human fetuses/ neonates with CDH. Further investigations are necessary to determine the functional significance of these findings. Copyright o 7995 by W.B. Saunders Company INDEX WORDS: Congenital tricular hypoplasia, cardiac anomalies.

diaphragmatic maldevelopment,

hernia,

left venassociated

ESPITE OPTIMAL perinatal care, the “hidden mortality” on congenital diaphragmatic hernia (CDH) diagnosed before 24 weeks’ gestation exceeds 6O%.l.’ The pathophysiology results from a combination of pulmonary hypoplasia, pulmonary hypertension, and both a qualitative and quantitative surfactant deficiency.3-” Autopsy reports from patients with

D

Journal

of Ped/atr/c

Surgery,

Vol30,

No 7 (July),

1995:

pp 925-929

Hernia XI: of the Heart Stuart J. O’Toole,

CDH have shown that in human newborns with CDH, the left ventricles, interventricular septum, and atria are smaller than those in age-matched controls.1° Fetal and newborn echocardiography and human autopsy studies now suggest that underdevelopment of the left ventricle in CDH may be caused by compression by herniated viscera during development.‘OJl However, it is not clear to what extent the cardiovascular system contributes to the natural history of the untreated fetus or the pathophysiology of the newborn with CDH.‘* The purpose of this study was to determine whether the presence of bowel in the chest during critical stages of in utero development in the fetal lamb model of CDH results in structural and/or biochemical underdevelopment of the heart. MATERIALS

AND METHODS

The Model Thirty-five time-dated pregnant ewes underwent operation. Diaphragmatic hernias were created surgically in fetal lambs at 80 days’ gestation (term. 145 days) as previously described by our laboratories.13 At 141 days’ gestation, the pregnant ewes were fasted for 24 hours and then underwent cesarean section using the anesthetic and hysterotomy techniques described above. The lambs were killed with an intravenous overdose of sodium pentobarbital.

Anatomic Dissectionof the Hearts Both CDH and littermate control hearts were fixed in 4% paraformaldehyde and were dissected systematically by the same mdividual (HLK). The left ventricles (LV) and right ventricles (RV), interventricular septum, and atria were dissected and weighed in a standard manner as described by Rowlan et a1.r”J4

From the Buffalo Instztute of Fetal Therapy (BIFT); the ChzldrenS Hospital of Buffalo, Departments of Surgery and Pediatrics; and the State Unzverszty of New York at Buffalo, School of Medzcine and Biomedical Sciences, Bzzffalo, NY Presented at the 1994 Annual Meeting of the Sectzon on Surgery of the American Academy of Pedzatrics, Dallas, Texas, October 21-23, 1994. Suppovted in part by grants from the American Lung Assoczation, the Women and Children’s Health Research Foundation, NIH Grant No. 1504816A, and the US Surgzcal Corporation, Norwalk, CT. Address reprint requests to Philip L. Glick, MD, Buffalo Instztute of Fetal Therapy (BIFT}, Chzldren’s Hospztal of Buffalo, State Vnzversit) of New York at Buffalo, 219 Bryant St, Buffalo, NY14222. Copyright o 1995 by W.B. Saunders Company 0022-3468/95/3007-0004$03.00/O 925

926

KARAMANOUKIAN

Ventricular wall thicknesses were measured at two predetermined distances from the atrioventricular groove (see Results).10%14In addition, both aortic and pulmonary root diameters and the diameter of the ductus arteriosus at its juncture with the aorta were measured in both CDH and control hearts.‘r

Biochemical Analyses Assays were performed on precisely weighed, homogenized, and sonicated tissue aliquots. Approximately lOO-mg aliquots were used to determine collagen (hydroxyproline) content by a standard technique.15 The determination of elastin was accomplished by weak base hydrolysis of tissue fragments followed by the gravimetric determination of remaining elastin residue.16 The total protein concentration was determined by the method of Bradford using commercial Coomassie dye reagent” (BioRad, Richmond, CA). Cellular DNA was quantitated fluorimetrically, using the fluorochrome Hoechst 33258 compound (bzs-benzimidazole)18 (Sigma Chemical, St Louis, MO).

Statistical Analysis All data is expressed as mean rt standard error (SEM). Unpaired t tests were used for comparisons within each group (CDH versus CDH, control versus control) and between groups (CDH versus controls). RESULTS

There were no differences in body weight (expressed in kg) between CDH and control lambs (4.25 + .26 versus 3.71 + .24, P = NS). Total heart weight, left ventricular, septal, and atria1 weights (g/kg lamb) were significantly decreased when compared with those of littermate controls (Table 1). Equivalent decrease in weight after fixation of both CDH and control tissues in 4% paraformaldehyde solution ensured accurate comparisons and statistical analyses. Wet/dry weight ratios of heart tissue from both CDH and littermate controls were identical, implying that tissue edema was not responsible for differences in total weights. DNA/total protein concentration ratios were not found to be different between RV tissues of both CDH and control hearts (.Oll ?z .003 versus .013 -t .005, P = NS), and LV tissues between CDH and control hearts (.015 -t- .004 versus .015 f .006, P = NS). LV wall thicknesses measured at 0.5 cm from the atrioventricular groove (0.59 f .03 versus 0.63 4 .03, P = NS), and halfway between the atrioventricular groove and the apex (0.53 + .04 versus 0.59 ? .03, P = NS) were identical between CDH and control lambs. The RV wall thicknesses measured at 0.5 cm Table Group CDH (n = 7) CON (n = 7)

Right and Left Atria 0.68

1. Total Right Ventricle

L 0.06*

1.26

k 0.07

1.14 k 0.15

1.57

+- 0.17

NOTE. Data expressed as g/kg *P < .05for CDH versus control

lamb, mean httermates.

+- standard

error.

ET AL

from the atrioventricular groove (0.50 5 .05 versus 0.51 ~fr.04, P = NS), and halfway between the atrioventricular groove and the apex (0.53 +- .04 versus 0.50 +- .03, P = NS) were identical between CDH and control lambs. Aortic (Ao) root diameters (measured in cm) were identical between CDH and control hearts. However, pulmonary artery (PA) root diameters were increased in CDH, resulting in a decreased Ao/PA root ratio in CDH (Table 2). When compared with controls, ductus arteriosus diameters were also found to be increased in CDH (Table 2). Hydroxyproline (.519 r .03 versus .586 + .05, P = NS) and elastin (.45 2 .36 versus .61 ? .12, P = NS) content were not found to be different between RV and LV in CDH hearts. Hydroxyproline (.538 f .04 versus .514 + .09, p = NS) and elastin (.69 + .34 versus .61 + .18,p = NS) content were also found to be different between RV and LV in control hearts. Similarly, there were no differences in hydroxyproline or elastin content in the LV when CDH hearts were compared with controls. This relationship was also maintained for both RV hydroxyproline and elastin content. DISCUSSION

Over the last 2 decades, the contribution of the cardiovascular system in the pathophysiology of CDH has been uncommonly alluded to in the surgical literature.1°J9 Despite this, there is now a growing body of evidence to suggest that the heart is affected in CDH, that cardiac parameters may be a useful predictor of outcome,10J2Jg-22 and that focusing new therapeutic strategies at these cardiovascular alterations might improve outcome in patients with CDH. Results from our study show that total heart weight and left ventricular, septal, and atria1 weights are significantly decreased when CDH is compared with littermate controls. These results have duplicated the findings in human neonates with CDH, which have been shown to have hypoplastic left ventricles.1° Furthermore, identical DNA to total protein concentration ratios of the left ventricles of both CDH and littermate controls imply that the decreased weight of the left ventricle in CDH is caused by hypoplasia, and not atrophy, of the left ventricles. This is consistent with our previous biochemical analyses of the lungs in CDH, which were similarly shown to be hypoplastic.23

and Component

Heart

Weights

Left Ventricle

Interventricular

1.65 k 0.11” 2.15 k 0.19

I.25 1.99

Septum

f O.ll* k 0.21

Total Heart 4.88 6.75

f 0.25* i- 0.49

PATHOPHYSIOLOGY

OF CDH

927

Table Group

Aortlc

2. Aortic,

(Ao) Root

Pulmonary

Artery,

Pulmonary

and

Ductus

(PA) Artery Root

Arteriosus

Diameters Ductus

Artenosus

AolPA

Ratio

CDH

(n = 8)

0.37

2 .Ol

0.47

2 .01*

0.35

‘. .01*

0.81

CON

(n = 5)

0.37

z!I .Ol

0.38

2 .Ol

0.22

z!z .02

1.00 + .03

NOTE. Data expressed in centimeters, mean *P < .005 for CDH versus cantrol littermates.

+ standard

‘- .Ol*

error.

Seibert et allo have shown that in human neonates dying of CDH, total heart, atrial, left ventricular, and septal weights were decreased when compared with those of age-matched controls.1° They concluded that the left ventricles in CDH were “hypoplastic,” but failed to measure DNA/protein ratios. Perhaps they should have concluded that the left ventricles were smaller than those of controls because without biochemical data, the decreased weights could be attributable to hypoplasia, atrophy, or both. More recently, it has been shown that infants with cardiac malformations associated with CDH (excluding patent ductus arteriosus and patent foramen ovale) had a significantly higher mortality than neonates with isolated CDH; hypoplastic left heart syndrome (HLHS) being the most common associated cardiac defect.24 However, it should be noted that isolated hypoplasia of the left ventricle is different than HLHS, which includes aortic and mitral valve atresia, and severe left ventricular and proximal aortic hypoplasia and may or may not show common developmental mechanisms.25 Nevertheless, the conclusions from this and other studies have reinvigorated interest in the contribution of the cardiovascular system in the pathophysiology of CDH.12 These clinical reports have been confirmed by previous laboratory studies. In surgically created CDH at 110 days’ gestation, it has been determined that cardiac output is significantly lower than the cardiac output of control lambs.ig In this study, it was further noted that CDH lambs had a ratio of pulmonary vascular resistance to systemic vascular resistance that was twice as high as in the control group. It was hypothesized that the observed differences in cardiovascular hemodynamics in CDH were attributable to the physiological sequelae of the pulmonary hypoplasia and pulmonary hypertension in CDH.lg A structural abnormality of the heart resulting in these hemodynamic alterations was not considered by the authors to be a likely explanation for the differences in hemodynamics between CDH and controls.lg However, with our results in a similar fetal lamb CDH model, it is retrospectively plausible to invoke developmental cardiac anomalies to partially explain their results. When compared with controls, human neonates

with left-sided CDH have both left ventricular hypoplasia and decreased mean aortic to pulmonary artery (Ao/PA) root ratios. l1 We wanted to show whether this finding was also present in the surgically created CDH lamb model. Our data has confirmed identical Ao root diameters between CDH and control lambs. Therefore, a decreased Ao/PA root ratio in CDH is caused by an increased PA diameter. Presumably, left ventricular hypoplasia may result in utero in increased left atria1 pressure, decreased right-to-left shunting through the foramen ovale, and increased PA pressures and flow, thus increasing both PA and ductus arteriosus diameters.” We have shown previously that the decreased compliance of the lungs in CDH is partly caused by an increased collagen content.26 Because the heart is intimately related to the lung during development, we hypothesized that like the lungs in CDH, the heart may have an increased collagen content. However, our results show no differences in both collagen and elastin content between right and left ventricles in both CDH and control lambs. Similarly, no differences in collagen or elastin content were found between left (or right) ventricles when CDH iambs were compared with control lambs. Although CDH is a simple anatomic lesion amenable to surgical correction, current medical treatment is directed toward improving oxygenation and ventilation and toward preventing fatal episodes of pulmonary hypertension. However, at best, the results have been disappointing. A significant group of neonates with CDH either do not survive to operation, die after surgical correction, or die after a successful ECMO run.27 It is conceivable that unrecognized and thus inadequately treated left ventricular hypoplasia might have been a significant factor in many of these cases. I2 Presumably, cardiac dysfunction secondary to left ventricular hypoplasia may significantly contribute to decreased cardiac output, decreased systemic blood pressures, increased rightto-left shunting, hypoxia, and acidosis. Traditionally, this cascade of events has been ascribed to result solely from pulmonary hypertension in CDH.28 Analysis of data from echocardiographic studies performed routinely before placement of neonates on ECMO have enlightened our thinking about the

928

KARAMANOUKIAN

“missing link” in the pathophysiology of CDH.12 Until recently, it has been near impossible to predict which fetuses/newborns with CDH will or will not survive.29~30The major challenge to investigators in the field has been to identify prenatal risk factors in the CDH fetus that are predictors of survival.12 To this effect, fetal echocardiography may have identified a crucial element contributing to the “hidden mortality” in CDH.12 Fetal cardiac ventricular disproportion, as shown by a reduced left/right ventricular size before 24 weeks’ gestation, has been shown to

ET AL

have an associated mortality rate reaching 100%.11,21 Other investigators have duplicated these studies with measurements of left ventricular mass index. Careful analysis has shown that left ventricular mass index in CDH is a good predictor of survival.22 Prospective studies will determine whether the left ventricular hypoplasia is a prognostic indicator in CDH.r2 ACKNOWLEDGMENT The authors thank Terri Jones and Cynthia Williams for technical assistance.

REFERENCES 16. Soskel NT, Wolt TB, Sandberg LB: Isolation and characterization of insoluble and soluble elastins, in Cunningham LW (ed): Methods in Enzymology, vol 144: Structural and Contractile Proteins, Part D. New York, NY, Academic, 1987 17. Bradford MM: A raptd and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72248-254.1976 18. Brunk CF, Jones KC, James TW: Assay for nanogram quantities of DNA in cellular homogenates. Anal Biochem 92:497-

1. Harrison MR, Bjordal RI. Langmark F, et al: Congenital diaphragmatic hernia: The hidden mortality. J Pediatr Surg 13:227230,1978 2. Adzick NS, Harrison MR, Glick PL, et al: Diaphragmatic hernia in the fetus: Prenatal diagnosis and outcome in 94 cases. J Pediatr Surg 20:357-361, 1985 3. Geggel RL, Murphy JD, Landleben D, et al: Congenital diaphragmatic hernia: Arterial structural changes and persistent pulmonary hypertension after surgical repair. J Pediatr 107:457464,1985 4. Bohn

D, Tamura M, Perrin D, et al: Ventilatory predictors of pulmonary hypoplasia in congenital diaphragmatic hernia, confirmed by morphologic assessment. J Pediatr 111:423-431, 1987 5. Levin DL: Morphologic analysis of the pulmonary vascular bed in congenital left-sided diaphragmatic hernia. J Pediatr 92:805809,1979 6. Glick

PL, Stannard VA, Leach CL, et al: Pathophysiology of congenital diaphragmatic hernia II: The fetal lamb CDH model is surfactant deficient. J Pediatr Surg 27:382-388, 1992 7. Glick PL, Leach CL, Besner GE, et al: Pathophysiology of congenital diaphragmatic hernia III: Exogenous surfactant therapy for the high risk neonate with CDH. J Pediatr Surg 27:866-869, 1992 8. Wilcox DT, Glick PL, Karamanoukian HL, et al: Pathophysiology of congenital diaphragmatic hernia V: Exogenous surfactant therapy improves gas exchange in the lamb congenital diaphragmatic hernia model. J Pediatrics 124:289-293, 1994 9. Willcox DT, Glick PL! Karamanoukian HL, et al: Pathophysiology of congenital ‘diaphragmatic hernia VI: Type II cell function 1 in the lamb congenital diaphragmatic hernia model. J Pediatr Surg ’ (in press) 10. Seibert JR, Haas JE, Beckwith JB: Left ventricular hypoplasia in congenital diaphragmatic hernia. J Pediatr Surg 19:567-571, 1984

11. Sharland GK, Lockhart SM, Heward AJ, et al: Prognosis in fetal diaphragmatic hernia. Am J Obstet Gynecol 166:9-13, 1992 12. Karamanoukian HL, Wilcox DT, Glick PL: The “missing link” in congenital diaphragmatic hernia. J Pediatr Surg 29:954955, 1994 (letter) 13. Adzick NS, Harrison MR. Outwater KM, et al: Correction of congenital diaphragmatic hernia in utero IV. An early gestational fetal lamb model for pulmonary vascular morphometric analysis. J Pediatr Surg 20:673-680, 1985 14. Rowlatt UF, Rimoldi HJA, Lev M: The quantitative anatomy of the normal child’s heart. Pediatr Clin North Am 10:499-591, 1963 15. Stegemann H, Stalder K: Determination of hydroxyproline. Clin Chem Acta 18:267-273, 1967

500,1979 19. Olivet

RT, Rupp WM, Telander RL, et al: Hemodynamics of congenital diaphragmatic hernia in lambs. J Pediatr Surg 13:231-235, 1978 20. Crawford DC, Drake DP, Kwaithkowski D, et al: Prenatal diagnosis of reversible cardiac hypoplasia associated with congenital diaphragmatic hernia: Implications for postnatal management. J Clin Ultrasound 14:718-721,1986 21. Crawford DC, Wright VM, Drake DP, et al: Fetal diaphragmatic hernia: The value of fetal echocardiography in the prediction of postnatal outcome. Br J Obstet Gynecol96:705-710. 1989 22. Schwartz SM, Vermillion RP, Hirsch1 RB: Evaluation of left ventricular mass in children with left-sided congenital diaphragmatic hernia. J Pediatr 125:447-451, 1994 23. Hosoda Y, Rossman JE, Glick PL: Pathophysiology of congenital diaphragmatic hernia IV: Renal hyperplasia is associated with pulmonary hypoplasia. J Pediatr Surg 28:464-470,1993 24. Fouza DO, Wilson JM: Congenital diaphragmatic hernia and associated anomalies: Their incidence, identification, and impact on prognosis. J Pediatr Surg 29:1113-1117,1994 25. Zahka KG, Spector M, Hanisch D: Hypoplastic left-heart syndrome N&wood operation, transplantation, or compassionate care. Clin Perinatol20:145-154, 1993 26. Hassett M, Glick PL, Karamanoukian HL, et al: Elevated pulmonary collagen in the lamb congenital diaphragmatic hernia model. J Pediatr Surg (in press) 27. Lally KP, Paranka MS, Roden J, et al: Congenital diaphragmatic hernia. Stabilization and repair on ECMO. Ann Surg 216~569.573,1992 28. Wilcox

DT, Karamanoukian HL, Glick PL: Antenatal diagnosis of pediatric surgical anomalies. Pediatr Clin North Am 40:1273-1287, 1993 29. Harrison MR, Adzick NS, Flake AW, et al: Correction of congenital diaphragmatic hernia in utero VI. Hard-earned lessons. J Pediatr Surg 28:1411-1418,1993 30. Harrison MR, Langer JC, Adzuck NS, et al. Correction of congenital diaphragmatic hernia in utero V. Initial clinical experience. J Pediatr Surg 25:47-57, 1990

PATHOPHYSIOLOGY

OF CDH

929

Discussion A. Coran (Ann Arbor, MI): I’d like to congratulate the authors on an important and well-presented study and thank them for allowing me to look at the manuscript ahead of time. This study shows a decrease in total heart, left ventricular, septal, and atria1 weights, but no differences in right ventricular weight or ventricular wall thickness. We have carried out a study that was done by Dr Hirsch1 and our group at Ann Arbor looking at echocardiographic evaluation of the left ventricular mass and congenital diaphragmatic hernia. We looked at the left ventricular mass in controlled newborns with pulmonary hypertension not related to a CDH and in newborns with a left-sided CDH. The left ventricular mass is decreased in this CDH group. We then looked at that relationship in terms of ECMO need, and the need for ECMO was much greater in those with a lower left ventricular (LV) mass and survival also was correlated with the LV mass. I have two questions for the authors. In this model were all the CDHs left-sided? If yes, do you think you would have observed something different if you had created right-sided CDHs in these lambs? Secondly, what do you think is the pathophysiology going on here? Does the hypoplasia come first and lead to increased left atria1 pressure and decreased pulmonary flow? Or, on the other hand, does the decreased pulmonary flow secondary to pulmonary

hypoplasia come first and lead to decreased pulmonary venous return to the left side of the heart, which then results in left ventricular hypoplasia? H.L. Karamanoukian (response): The observations that Ron Hirsch1 has made have certainly substantiated the similar observations that Sharland et al made in fetuses with CDH in terms of indicating that a diagnosis of ventricular disproportion before 24 weeks’ gestation is certainly associated with a 100% mortality. Sharland’s group has shown that in right-sided diaphragmatic defects, this ventricular disproportion is reversed. Although we don’t have data in our lamb model, in the human condition it has been observed that there is right ventricular hypoplasia in rightsided CDH. In terms of your second question, its’s been thought that this is due to a compressive phenomenon on the left ventricle. Because left ventricular development is related directly to blood flow through the left ventricle, it should be noted that the amount of blood going through the left ventricle is only about 8% of the total blood flow. We are in the process of doing microsphere studies in our CDH lamb model to determine whether there is an alteration of flow through the foramen ovale. I think that will give us a better understanding of the pathophysiology of this selective left ventricular hypoplasia in left-sided CDH.