- Email: [email protected]
Pathophysiology of congenital diaphragmatic hernia IV: Renal hyperplasia is associated with pulmonary hypoplasia
Pathophysiology Renal Hyperplasia
of Congenital Is Associated
Diaphragmatic Hernia IV: With Pulmonary Hypoplasia
By Yataro Hosoda, Jon E. Rossman, and Philip L. Glick Buffalo, New York l The hypothesis of this article is that growth of the fetal lung is stimulated by a pulmonary growth factor (PGF) produced by the kidneys, which is modulated by a feedback signal from the lungs, a pulmonary-derived renotropin (PDR). In the fetus with pulmonary hypoplasia (PH), the lungs may maximally stimulate this feedback loop to release more PDR, resulting in continual stimulation of the kidneys and renal enlargement. If such a schema plays a role in the pathophysiology of PH. newborn infants with congenital diaphragmatic hernia (CDH) or chronic amniotic fluid leak (CAFL) should have enlarged kidneys. To investigate this hypothesis, we created models of CDH in fetal lambs and CAFL in fetal rabbits, and then analyzed lung (Lu) and kidney (K) growth. When compared to controls, newborn CDH lambs had significantly smaller lungs and larger kidneys. The lungs were hypoplastic as defined by either decreased lung weight/ body weight (LuW/BW), lung DNA/body weight (Lu DNA/ BW), or lung total protein/ body weight (LuTP/BW) (P c .Ol). Renal hyperplasia was confirmed by KW/BW, K DNA/BW (P < .Ol), and KTP/BW (P < .05). An inverse relationship between lung size and kidney size could be described by the equation KW/BW = 1.04 - 0.12 LuW/BW (r = -.75). The CAFL model in newborn rabbits produced severe oligohydramnios when compared with controls (P c .Ol). This resulted in fetuses with smaller lungs and larger kidneys as compared with those of controls. The lungs were significantly smaller and more hypoplastic than controls when compared by LuW (P < .Ol), LuW/BW (P c .Ol), Lu DNA/BW (P < .05), and Lu TP/BW (P < .Ol). The kidneys were significantly larger and more hyperplastic than controls as judged by the similar criteria (P c .Ol). In addition, the earlier in gestation the operation was performed, the greater was the effect on the kidneys and lungs. These data support our hypothesis and demonstrate that significant renal enlargement is associated with PH. The presumed mechanisms of renal enlargement associated with PH and the relation to amniotic fluid volume are discussed. If such a PGF and PDR can be isolated, the diagnostic and therapeutic implication for fetuses with PH are considerable. Copyright o 7993 by W.B. Saunders Company
INDEX WORDS: nary hypoplasia; sia.
T
Congenital diaphragmatic hernia; pulmopulmonary growth factors; renal hyperpla-
HE MORTALITY rate for congenital diaphragmatic hernia (CDH) is thought to depend on both the severity of pulmonary hypoplasia (PH) and the pulmonary hypertension present in the neonatal period, and remains at approximately 50%.lm4Ironically, prenatal diagnosis and optimally managed prenatal, perinatal, and postnatal cases of CDH have mortality rates as high as 8O%.5,6 Polyhydramnios associated with CDH may further increase the mortality.5,6The etiology of the polyhydramnios in CDH is 464
Fig 1. Theoretical feedback modulation of normal in utero lung and kidney growth. The black arrows indicate normal modulation of the afferent and efferent limbs of this loop.
unclear, but recently a unifying hypothesis explaining the association of polyhydramnios, PH, and renal enlargement in CDH has been proposed.’ In this hypothesis, Glick et al suggest that in utero growth of the lungs is stimulated by a pulmonary growth factor (PGF) produced by the kidneys. It is further hypothesized that the release of PGF is modulated by a
From The Buffalo Institute of Fetal Therapy, the Division of Pediatric Surgery, The Children’s Hospital of Buffalo, and the Departments of Surgery and Pediatrics, School of Medicine and Biomedical Sciences, The University at Buffalo, State University of New York, Buffalo, NY Presented at the 23rd Annual Meeting of the American Pediatric Surgical Association, Colorado Springs, Colorado, May 13-I 6, 1992. Supported in part by The Women S and Children’s Health Research Foundation of the Children S Hospital of Buffalo, The March of Dimes Birth Defects Foundation Basic Research Grant (#l-1244), and The United States Surgical Carp, Norwalk, CT. Address reprint requests to Philip L. Glick, MD, The Buffalo Institute of Fetal Therapy, The Children’s Hospital of Buffalo, 219 Bryant St, Buffalo, NY 14222. Copyright o 1993 by W.B. Saunders Company 0022-3468/9312803-0031$03.00/O JournalofPediafric Surgery, Vol28, No 3 (March), 1993: pp 464-470
RENAL HYPERPLASIA & PULMONARY
HYPOPLASIA IN CDH
feedback signal from the lungs (to the kidneys) by a pulmonary-derived renotropin (PDR) (Fig 1). If this hypothesis is true, CDH and other causes of PH, ie, chronic amniotic fluid leak (CAFL), should be associated with renal enlargement. This theory requires experimental verification. The purpose of the present study is to document the effects of PH on renal growth and to determine if the changes seen are due to a change in the cell number (hyperplasia v hypoplasia), cell size (hypertrophy v atrophy), or both. MATERIALS
AND METHODS
CDH Model (Fetal Lambs) At 80 days of gestation, diaphragmatic hernias were created surgically in 24 fetuses by methods previously described.* We used the GIA stapling device (US Surgical Corp, Norwalk, CT) to open and the TA-90 stapler (US Surgical Corp) to close the uterus.9 At 139 to 14.5 days of gestation, the pregnant ewes underwent cesarean section using the anesthetic technique previously described.x The lambs were then killed by intravenous potassium
465
chloride injection as the cord was divided. Next, the lambs were delivered without spilling the amniotic fluid, and amniotic fluid volume was directly measured. The newborn lamb body weight (BW), lung wet weight (LuW), and kidney wet weight (KW) of all specimens were measured. In an effort to standardize the size of the organs and to account for differences in BW, ratios were calculated (LuW/BW. KW/BW). The lung and kidney samples were homogenized using a tissue homogenizer and an ultrasonicator (Braun-Sonic 1510; B. Braun Biotech Inc, Allentown, PA), and DNA was isolated by phenolchloroform extraction.10 The extracted DNA content was determined by the Hoechst 33258 method with the TKO 100 Mini Fluorometer (Hoefer Scientific Instruments, San Francisco, CA).lt,l” The total lung and kidney DNA (Lu DNA, K DNA) were estimated from the following equation: total DNA = DNA of aliquot x organ weight/weight of aliquot. Ratios were then calculated (Lu DNA/BW, K DNA/BW). The total DNA and the DNA/BW were assumed to be directly proportional to total cell population of the organ sampled; differences between experimental groups were thought to reflect hyperplastic or hypoplastic changes in the organ’s development.‘3-15 The total protein (TP) content of lung and kidney tissues was assayed (Lu TP, K TP) using the coomassie blue spectrophotometric analysisI and again corrected for differences in body weight (Lu
4
140
3
ae 120
C’ 7
100
T E w
OL 9
cm
comro,
CDH
COnnd
CD”
conwe,
Total Proteln/BW
DNAIBW
Lung WVBW
60
”
A
CDH
Control
Lung Weight
CDH
Control
Kidney Weight 0.06
D
CDH
Control
Lung
CDH
Control
Kidney
Fig 2. (A) The effect of the procedure on lung and kidney growth in the CDH fetal lamb model (**P < .Ol v controls). (6) The effect of the procedure on lung weight/body weight, lung DNA/body weight, and lung total protein/body weight in the CDH fetal lamb model (**P < .Ol v controls), indicating significant pulmonary hypoplasia. (C)The effect of the procedure on kidney weight/body weight, kidney DNA/ body weight, and kidney total protein/body weight in the CDH fetal lamb model (‘P c .05 v controls, l*P < .Ol Y controls), indicating significant renal hyperplasia. (D) Differences were not reflected in lung DNA/TP and kidney DNA/TP ratios, indicating that the cell size in the lungs and the kidneys was unchanged, but that the cell population was decreased in lungs (hypoplasia but not atrophy) and was increased in kidneys (hyperplasia but not hypertrophy) in the CDH fetal lamb model.
466
HOSODA, ROSSMAN, AND GLICK
TPIBW, K TP/BW). Ratios between DNA and TP were calculated (Lu DNA/TP, K DNA/TP, Lu TP/DNA, and K TP/DNA) as a measure of cell size, and differences between experimental groups were thought to reflect hypertrophic or atrophic changes in organ development.i7 Serving as controls, 15 nonoperated litter mates underwent cesarean section at 139 to 145 days of gestation and the lungs and the kidneys were analyzed in a similar manner.
CAFL Model (Fetal Rabbits) New Zealand white rabbits of known gestational age underwent operations to simulate CAFL to produce oligohydramnios and PH.‘s The operations were performed on day 17, 20, or 23 of gestation in the pseudoglandular period of lung development. Hysterotomy and amniotomy were carried out on 3 or 4 sacs of one uterine horn. An S-mm segment of Silastic tubing was placed into the amniotic sac and secured by a purse-string suture. This tubing assured continued drainage of amniotic fluid into the maternal peritoneal cavity. Six fetuses on day 17 of pregnancy, 24 fetuses on day 20, and 20 fetuses on day 23 underwent this procedure. Gestation was then allowed to continue until day 30 of pregnancy, when all fetuses underwent cesarean delivery. The newborn rabbits were checked for survival and then killed. By methods similar to the iamb study, BW, LuW, KW, LuW/ BW, KWIBW, Lu DNA, K DNA, Lu DNA/BW, K DNA/BW, Lu TP, K TP, Lu TPIBW, K TPIBW, Lu DNAITP, K DNA/TP, Lu TP/DNA and K TP/DNA were determined. The liver wet weight (LiW), LiW/BW, liver DNA (LiDNA), Li DNA/BW, liver TP (Li TP), Li TP/BW, Li DNAITP, and Li TPiDNA was used as a control organ to determine the effect of the procedure.la These data were compared to that of corresponding nonoperated litter mates for each group. In another study, 8 operated fetuses of each group underwent reexploration at 27 to 28 days of pregnancy, then each amniotic sac was aspirated of all fluid using a 20-gauge needle and syringe, and the volume was recorded.18.19 These were compared with control fetuses to assess the changes in amniotic fluid volume due to the procedure.
phy) in the CDH model. Comparison of LuW/BW and KW/BW showed that there was an inverse relationship between lung size and kidney size in the fetal lamb that could be described by the equation KW/BW = 1.04 - 0.12 LuW/BW (r = -.75) (Fig 3). The mean amniotic fluid volume of the CDH model was larger than that of controls (848 + 382 v 704 5 227 mL) but difference between groups was not statistically significant. CAFL Model
All six fetuses that were operated on day 17 of pregnancy died in utero for presumed technical causes, and this arm of protocol was stopped. Eighteen of 24 fetuses that were operated on at 20 days and all of 20 that were operated on at 23 days survived until term. This procedure produced severe oligohydramnios when compared to unoperated controls (P < .Ol, Fig 4A). The oligohydramnios from the CAFL resulted in fetuses with smaller lungs and larger kidneys as compared with those of controls (Fig 4B). The lungs were significantly smaller and more hypoplastic when compared by LuW (P < .Ol), LuW/BW (P < .Ol), Lu DNA/BW (P < .05), and Lu TP/BW (P < .Ol, Fig 4C). The kidneys were significantly larger and more hyperplastic than controls as judged by the similar criteria (P < .Ol, Fig 4D). In addition, the
2.0 -
Statistical Analysis The data were analyzed on a PC-286LE computer (EPSON Corp, Tokyo, Japan) using l-2-3 (Lotus Development Corp, Cambridge, MA) and Statview II (Abacus Concepts Inc, Berkeley, CA). Statistical determinations were performed using the MannWhitney U test, with P < .05 considered significant. RESULTS
2 E 9
n
1.5 -
5
CDH Model
Fifteen of 24 lambs that underwent fetal surgery survived until term. The operated lambs had significantly smaller lungs, and larger kidneys when compared with controls (Fig 2A). The lungs were significantly hypoplastic as defined by either decreased LuW/BW, Lu DNA/BW or Lu TP/BW (P < .Ol, Fig 2B). Renal hyperplasia was confirmed by KW/BW, K DNA/BW (P < .Ol) and K TP/BW (P < .05, Fig 2C). However, these differences were not reflected in Lu DNA/TP and K DNA/TP ratios (Fig 2D), indicating that the cell size in the lungs and the kidneys was unchanged, but that the cell population was decreased in lungs (hypoplasia but not atrophy) and was increased in kidneys (hyperplasia but not hypertro-
0.0
’ 0
I
,
!
I
,
1
2
3
4
5
Lung Wt I Body Wt (gm%) Fig 3. The relationship between LuW/BW and KW/BW can be described by the regression equation KW/BW = 1.04 - 0.12 LuW/ BW,r = -.75 (P c .OOl). Squares = controls. Triangles = CDH.
RENAL HYPERPLASIA
& PULMONARY
HYPOPLASIA
IN CDH
467
Rabbits : n=8
1
DNlVBW
Kidney WVBW
Total ProteinlBW
0.8
0.6
0.4
0.2 CAFL C 20 days
A
CAFL C 23 days
E
n “CAFL
C
CAFL 23 d
20 d
Lung
C
“CAFL
C
CAFL
20 d
C
23 d
Kidney
0.4
0.2
B
” “CAFL
C
20d
CAFL 23 d
Lung Weight
C
CAFL 20 d
C
CAFL 23 d
Kidney Weight
C
Fig 4. (A) The fetal rabbit CAFL model produced severe oligohydramnios by decreasing amniotic fluid volume to less than that of controls (**f c .Ol). (B) The effect of the procedure on lung and kidney growth in the rabbf chronic amniotic fluid leak model (**P e .Ol Y controls). (C) The effect of the procedure on lung weight/body weight, lung DNA/ body weight, and lung total protein/ body weight in the rabbit chronic amniotic fluid leak model (*p < .lXi v controls, l*p c .Ol v controls), indicating significant pulmonary hypoplasia. (D) The effect of the procedure on kidney weight/body weight, kidney DNA/body weight, and kidney total protein/body weight in the chronic amniotic fluid leak model (**P < .Ol v controls), indicating significant renal hyperplasia. (E) Differences were not reflected in lung DNA/TP and kidney DNA/TP ratios, indicating that the cell size in the lungs and the kidneys was unchanged, but that the cell population was decreased in lungs (hypoplasia but not atrophy) and was increased in kidneys (hyperplasia but not hypertrophy) in the rabbit CAFL model.
HOSODA, ROSSMAN,
468
earIier in gestation the operation was performed, the greater was the effect on the kidneys and lungs. Similar to the lamb study, DNA/TP and TP/DNA ratios of lungs and kidneys did not differ significantly between groups and thus cell atrophy or hypertrophy are not involved in the changes in organ size which were observed (Fig 4E). The liver did not vary significantly between groups, and its growth appeared to be unaffected by the procedure. DISCUSSION
These results in the fetal lamb support our hypothesis that PH in CDH is associated with significant renal enlargement. The renal enlargement is hyperplastic and not hypertrophic growth. This experimental verification confirms our previous clinical observation.’ According to our hypothesis, in the fetus with CDH, the space-occupying nature of the herniated viscera hinders normal pulmonary growth and development and results in PH. The lungs may “sense” this inadequate pulmonary growth and development and secrete excess PDR. The kidneys would respond to this afferent signal, enlarge by hyperplastic growth, and secrete increased amounts of PGF. We had previously speculated that, if our theory were true and broadly applicable to other instances of PH, then PH associated with oligohydramnios, ie, CAFL, should also be associated with renal enlargement. Our data now confirm this and the effect in rabbits seems to be proportional to the length of gestation in which the stimulus exists. Presumably in
AND GLICK
CAFL, as in CDH when the lungs “sense” inadequate growth, they also produce increased amounts of PDR. The kidneys become hyperplastic and produce more PGF, but the CAFL and oligohydramnios persists and PH results. The precise etiologic role of amniotic fluid in the process of lung growth is not known. Several possible etiologies for PH with oligohydramnios have been described.i8-20 Oligohydramnios may cause fetal thoracic compression by the amniotic membranes and the uterine wall and limit intrathoracic space and fetal breathing movement.2’-23 However, attempts at intraamniotic saline infusions to reconstitute amniotic fluid volume have not been successful in preventing PH. 24Therefore, we believe the quality of amniotic fluid rather than just the quantity may be important. PGF may be an important component of the amniotic fluid to promote proper lung growth and development. The present data encourage continued investigation to identify, characterize, and synthesize PGF and PDR. These may be unique growth factors specific to the lungs and kidneys, or ubiquitous growth factors being modulated in a unique fashion under the physiologic circumstances. In the future, PGF might be used therapeutically for fetuses and newborns with PH. Accelerating pulmonary growth and development in newborns with CDH or PH due to other causes may hasten weaning from mechanical ventilation or extracorporeal membrane oxygenation and decrease their mortality and morbidity.
REFERENCES 1. Anderson KD: Congenital diaphragmatic hernia, in Welch KJ, Randolph JG, Ravitch MM, et al (eds): Pediatric Surgery (ed 1). Chicago, IL, Year Book, 1986, pp 589-601 2. Reynolds M, Luck SR, Lappen R: The “critical” neonate with diaphragmatic hernia: A 21-year perspective. J Pediatr Surg 19:364-369,1984 3. Touloukian RJ, Markowitz RI: A preoperative x-ray scoring system for risk assessment of newborns with congenital diaphragmatic hernia. J Pediatr Surg 19:252-257, 1984 4. Simson JNL, Eckstein HB: Congenital diaphragmatic hernia: A 20 year experience. Br J Surg 72:733-736, 1985 5. 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 6. Steinhorn RH, Kriesmer PJ, Green PP, et al: Natural history of congenital diaphragmatic hernia in Minnesota: Impact of in utero diagnosis on survival. Pediatr Res 29:236A, 1991 (abstr) 7. Glick PL, Siebert JR, Benjamin DR: Pathophysiology of congenital diaphragmatic hernia I: Renal enlargement suggests feedback modulation by pulmonary derived renotropins-A unifying hypothesis to explain pulmonary hypoplasia, polyhydramnios, and renal enlargement in the fetus/newborn with congenital diaphragmatic hernia. J Pediatr Surg 25:492-495,199O
8. 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 9. Adzick NS, Harrison MR, Flake AW, et al: Automatic uterine stapling device in fetal surgery. Surgical Forum 36:479-481, 1985 10. Strauss WM: Preparation of genomic DNA from mammalian tissue, in Ausubel FM, Brent R, Kingston RE, et al (eds): Current Protocols in Molecular Biology 1. New York, NY, Wiley, 1988, Sections 2.2.1-2.2.3 11. Brunk CF, Jones KC, James TW: Assay for nanogram quantities of DNA in cellular homogenates. Anal Biochem 92:497500,1979 12. Labarca C, Paigen K: A simple, rapid, and sensitive DNA assay procedure. Anal Biochem 102:344-352,198O 13. Wigglesworth JS, Desai R: Use of DNA estimation for growth assessment in normal and hypoplastic fetal lungs. Arch Dis Child 56:601-605,198l 14. Wigglesworth JS, Desai R, Guerrini P: Fetal lung hypoplasia: biochemical and structural variations and their possible significance. Arch Dis Child 56:606-615, 1981 15. Wigglesworth JS, Desai R: Is fetal respiratory functional a major determinant of perinatal survival? Lancet 1:264-267, 1982
RENAL HYPERPLASIA & PULMONARY HYPOPLASIA IN CDH
16. Bradford MM: A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248-254, 1976
17. Winick M, Noble A: Quantitative change in DNA, RNA and protein during prenatal and postnatal growth in the rat. Dev Biol 12:451-466, 1965 18. Nakayama DK, Glick PL, Harrison MR, et al: Experimental pulmonary hypoplasia due to oligohydramnios and its reversal by relieving thoracic compression. J Pediatr Surg 18:347-353,1983 1Y. Adzick NS. Harrison MR, Glick PL, et al: Experimental pulmonary hypoplasia and oligohydramnios: Relative contributions of lung fluid and fetal breathing movements. J Pediatr Surg 19:658-665,1984 20. Perlman M. Williams J, Hirsch M: Neonatal pulmonary
469
hypoplasia after prolonged leakage of amniotic Huid. Arch Dis
Child 51:349-353,1976 21. Blott M, Greenough A, Nicolaides KH, et al: Fetal breathing movements as predictor of favorable pregnancy outcome after oligohydramnios due to membrane rupture in second trimester. Lancet 2:129-131, 1987 22. Collins MH, Moessinger AC, Kleinerman J. et al: Morphometry of hypoplastic fetal guinea pig lungs following amniotic fluid leak. Pediatr Res 20:955-960, 1986 23. Harrison MR, Jester JA, Ross NA: Correction of congenital diaphragmatic hernia in utero. I. The model: lntrathoracic balloon produced fetal pulmonary hypoplasia. Surgery 88:174-182,198O 24. Glick PL, Harrison MR, Adzick NS, et al: Management of fetus with congenital hydronephrosis II: Prognostic criteria and selection for treatment. J Pediatr Surg 20:376-387. 1985
Discussion T.F. Tracy (St Louis, MO): This excellent presentation by Dr Hosado and mentored by Dr Glick has really given us two distinct experiments that give us some insight into the long sought after pulmonaryrenal axis in fetal development. Throughout the 1970s Dr William Blanc of Columbia University, struggled with this problem and the problem of pulmonary hypoplasia that resulted from oligohydramnios. In a rat model with amniotic fluid leak in late gestation, he not only succeeded in inducing pulmonary hypoplasia but also succeeded in inducing global intrauterine growth retardation. Hence he had the wrong model. In the 1980s many studies by Dr Harrison and his colleagues at the University of San Francisco took up the banner with a successful larger animal model and were able to demonstrate that oligohydramnias indeed did induce pulmonary hypoplasia. However, with the removal of the thoracic compression, the hypoplasia was also reversed. Unfortunately, no observations throughout these several studies noted any renal hypertrophy or hyperplasia in any of these experiments. In the 1990s this search for this Holy Grail of lung development continues in Buffalo. In these elegant studies their hypothesis is that fetal lung growth is directed by fetal renal growth. Also there is a second arm to their hypothesis that there is feedback loop from the lung to the kidney. In this paper today, it is clear that renal hyperplasia follows two sequential modifications that induce pulmonary hypoplasia. This confirms the second arm of their hypothesis in that a small lung equals a large kidney. One of the most interesting figures was the regression analysis for those animals with congenital diaphragmatic hernia. Significant regression for small lungs leading to large kidneys as achieved by combining
control animals as well as experimental animals. My first question is, does this same regression coefficient apply in normal controls as well as those diaphragmatic controls? In other words, is there a graded hypoplasia of the lung that results in a graded hyperplasia within the kidney? If your hypothesis is correct, the same regression coefficient should be found in controls and experimental animals, yet both groups appear flat. The second question pertains to the problems with amniotic fluid leak. Your attempts at early induction of amniotic fluid leak led to increased fetal loss. Was Dr Blanc right in that amniotic fluid leak is capable of inducing global growth retardation and only late gestational changes in amniotic fluid volume will result in the changes that you found in your kidneys? One minor point with your DNA body weight data in this chronic amniotic fluid model was that your kidney weight loss was significant with increases in your margin of .07 to .04. You had some interesting liver data that showed a similar increase in DNA synthesis at the same time. Could we not be seeing a global regenerative response as a stimulus for proliferation? J. Lunger (St Louis, MO): If your hypothesis is correct, I assume there is a single cell type in the kidney that is producing this pulmonary growth factor. The hypertrophy in the kidney would be caused by hypertrophy of that single cell type. Have you done any histological studies to determine what cell type that is? .I. Wilson (Boston, MA): We looked at your data very carefully and even compared your kidney weights with those of our own control animals in our fetal studies and there is no question in my mind that your kidneys are hyperplastic. I think this is fascinating but I am not sure I agree with your hypothesis. In our own
470
studies of lung growth in fetal lambs, we have been able to correlate pulmonary hypoplasia with bilateral nephrectomy which is certainly in keeping with your model. However, by performing a tracheal ligation on these anephric animals, we have obtained lungs 3 times the size of normal lungs and 5 times the size of the lungs with nephrectomy alone. By volume, weight, alveolar count, and protein/DNA ratios, these are hyperplastic lungs despite the fact that no kidneys were present from either 70 or 90 days’ on. We have repeated this model utilizing experimental diaphragmatic hernia rather than nephrectomy as the promoter of pulmonary hypoplasia with similar results. I believe your description of renal hyperplasia is real but can you reconcile your model of feedback employing a renally previous pulmonary growth factor with our data in anephric fetal sheep? RL. Glick (response): Thank you very much for your comments. I would like to begin with Dr Wilson’s comments. We had originally speculated that PDR and PGF were unique growth factors produced by the lungs and the kidneys, respectively. In work we have not yet published from our laboratory, we have shown that fetal serum, amniotic fluid, and homoge-
HOSODA, ROSSMAN, AND GLICK
nates of lungs from the diaphragmatic hernia model stimulate kidney organ cell cultures. And serum, amniotic fluid and kidney homogenates in organ lung cultures stimulate the kidneys to grow. However, in light of Dr Wilson’s data, we have to rethink our hypothesis. We think that PDR and PGF may be unique growth factors specific to the lungs and kidneys. However, they may be ubiquitous growth factors being modulated in an unique fashion under these physiologic circumstances. We have not actually tried to find these growth factors as much as to prove that they are there. It may turn out that they are existing growth factors being modulated uniquely in this situation. So I don’t think that Dr Wilson’s data are incompatible with our original hypothesis. The PGF may just be coming from another organ. As far as Dr Tracy’s comments go, if you look at the regression curve, the control animals also fit right along that regression curve in that the control animals that have smaller lungs have larger kidneys. The liver data are interesting and that may go along with Dr Wilson’s thought that this is a ubiquitous growth factor.
of Congenital Is Associated
Diaphragmatic Hernia IV: With Pulmonary Hypoplasia
By Yataro Hosoda, Jon E. Rossman, and Philip L. Glick Buffalo, New York l The hypothesis of this article is that growth of the fetal lung is stimulated by a pulmonary growth factor (PGF) produced by the kidneys, which is modulated by a feedback signal from the lungs, a pulmonary-derived renotropin (PDR). In the fetus with pulmonary hypoplasia (PH), the lungs may maximally stimulate this feedback loop to release more PDR, resulting in continual stimulation of the kidneys and renal enlargement. If such a schema plays a role in the pathophysiology of PH. newborn infants with congenital diaphragmatic hernia (CDH) or chronic amniotic fluid leak (CAFL) should have enlarged kidneys. To investigate this hypothesis, we created models of CDH in fetal lambs and CAFL in fetal rabbits, and then analyzed lung (Lu) and kidney (K) growth. When compared to controls, newborn CDH lambs had significantly smaller lungs and larger kidneys. The lungs were hypoplastic as defined by either decreased lung weight/ body weight (LuW/BW), lung DNA/body weight (Lu DNA/ BW), or lung total protein/ body weight (LuTP/BW) (P c .Ol). Renal hyperplasia was confirmed by KW/BW, K DNA/BW (P < .Ol), and KTP/BW (P < .05). An inverse relationship between lung size and kidney size could be described by the equation KW/BW = 1.04 - 0.12 LuW/BW (r = -.75). The CAFL model in newborn rabbits produced severe oligohydramnios when compared with controls (P c .Ol). This resulted in fetuses with smaller lungs and larger kidneys as compared with those of controls. The lungs were significantly smaller and more hypoplastic than controls when compared by LuW (P < .Ol), LuW/BW (P c .Ol), Lu DNA/BW (P < .05), and Lu TP/BW (P < .Ol). The kidneys were significantly larger and more hyperplastic than controls as judged by the similar criteria (P c .Ol). In addition, the earlier in gestation the operation was performed, the greater was the effect on the kidneys and lungs. These data support our hypothesis and demonstrate that significant renal enlargement is associated with PH. The presumed mechanisms of renal enlargement associated with PH and the relation to amniotic fluid volume are discussed. If such a PGF and PDR can be isolated, the diagnostic and therapeutic implication for fetuses with PH are considerable. Copyright o 7993 by W.B. Saunders Company
INDEX WORDS: nary hypoplasia; sia.
T
Congenital diaphragmatic hernia; pulmopulmonary growth factors; renal hyperpla-
HE MORTALITY rate for congenital diaphragmatic hernia (CDH) is thought to depend on both the severity of pulmonary hypoplasia (PH) and the pulmonary hypertension present in the neonatal period, and remains at approximately 50%.lm4Ironically, prenatal diagnosis and optimally managed prenatal, perinatal, and postnatal cases of CDH have mortality rates as high as 8O%.5,6 Polyhydramnios associated with CDH may further increase the mortality.5,6The etiology of the polyhydramnios in CDH is 464
Fig 1. Theoretical feedback modulation of normal in utero lung and kidney growth. The black arrows indicate normal modulation of the afferent and efferent limbs of this loop.
unclear, but recently a unifying hypothesis explaining the association of polyhydramnios, PH, and renal enlargement in CDH has been proposed.’ In this hypothesis, Glick et al suggest that in utero growth of the lungs is stimulated by a pulmonary growth factor (PGF) produced by the kidneys. It is further hypothesized that the release of PGF is modulated by a
From The Buffalo Institute of Fetal Therapy, the Division of Pediatric Surgery, The Children’s Hospital of Buffalo, and the Departments of Surgery and Pediatrics, School of Medicine and Biomedical Sciences, The University at Buffalo, State University of New York, Buffalo, NY Presented at the 23rd Annual Meeting of the American Pediatric Surgical Association, Colorado Springs, Colorado, May 13-I 6, 1992. Supported in part by The Women S and Children’s Health Research Foundation of the Children S Hospital of Buffalo, The March of Dimes Birth Defects Foundation Basic Research Grant (#l-1244), and The United States Surgical Carp, Norwalk, CT. Address reprint requests to Philip L. Glick, MD, The Buffalo Institute of Fetal Therapy, The Children’s Hospital of Buffalo, 219 Bryant St, Buffalo, NY 14222. Copyright o 1993 by W.B. Saunders Company 0022-3468/9312803-0031$03.00/O JournalofPediafric Surgery, Vol28, No 3 (March), 1993: pp 464-470
RENAL HYPERPLASIA & PULMONARY
HYPOPLASIA IN CDH
feedback signal from the lungs (to the kidneys) by a pulmonary-derived renotropin (PDR) (Fig 1). If this hypothesis is true, CDH and other causes of PH, ie, chronic amniotic fluid leak (CAFL), should be associated with renal enlargement. This theory requires experimental verification. The purpose of the present study is to document the effects of PH on renal growth and to determine if the changes seen are due to a change in the cell number (hyperplasia v hypoplasia), cell size (hypertrophy v atrophy), or both. MATERIALS
AND METHODS
CDH Model (Fetal Lambs) At 80 days of gestation, diaphragmatic hernias were created surgically in 24 fetuses by methods previously described.* We used the GIA stapling device (US Surgical Corp, Norwalk, CT) to open and the TA-90 stapler (US Surgical Corp) to close the uterus.9 At 139 to 14.5 days of gestation, the pregnant ewes underwent cesarean section using the anesthetic technique previously described.x The lambs were then killed by intravenous potassium
465
chloride injection as the cord was divided. Next, the lambs were delivered without spilling the amniotic fluid, and amniotic fluid volume was directly measured. The newborn lamb body weight (BW), lung wet weight (LuW), and kidney wet weight (KW) of all specimens were measured. In an effort to standardize the size of the organs and to account for differences in BW, ratios were calculated (LuW/BW. KW/BW). The lung and kidney samples were homogenized using a tissue homogenizer and an ultrasonicator (Braun-Sonic 1510; B. Braun Biotech Inc, Allentown, PA), and DNA was isolated by phenolchloroform extraction.10 The extracted DNA content was determined by the Hoechst 33258 method with the TKO 100 Mini Fluorometer (Hoefer Scientific Instruments, San Francisco, CA).lt,l” The total lung and kidney DNA (Lu DNA, K DNA) were estimated from the following equation: total DNA = DNA of aliquot x organ weight/weight of aliquot. Ratios were then calculated (Lu DNA/BW, K DNA/BW). The total DNA and the DNA/BW were assumed to be directly proportional to total cell population of the organ sampled; differences between experimental groups were thought to reflect hyperplastic or hypoplastic changes in the organ’s development.‘3-15 The total protein (TP) content of lung and kidney tissues was assayed (Lu TP, K TP) using the coomassie blue spectrophotometric analysisI and again corrected for differences in body weight (Lu
4
140
3
ae 120
C’ 7
100
T E w
OL 9
cm
comro,
CDH
COnnd
CD”
conwe,
Total Proteln/BW
DNAIBW
Lung WVBW
60
”
A
CDH
Control
Lung Weight
CDH
Control
Kidney Weight 0.06
D
CDH
Control
Lung
CDH
Control
Kidney
Fig 2. (A) The effect of the procedure on lung and kidney growth in the CDH fetal lamb model (**P < .Ol v controls). (6) The effect of the procedure on lung weight/body weight, lung DNA/body weight, and lung total protein/body weight in the CDH fetal lamb model (**P < .Ol v controls), indicating significant pulmonary hypoplasia. (C)The effect of the procedure on kidney weight/body weight, kidney DNA/ body weight, and kidney total protein/body weight in the CDH fetal lamb model (‘P c .05 v controls, l*P < .Ol Y controls), indicating significant renal hyperplasia. (D) Differences were not reflected in lung DNA/TP and kidney DNA/TP ratios, indicating that the cell size in the lungs and the kidneys was unchanged, but that the cell population was decreased in lungs (hypoplasia but not atrophy) and was increased in kidneys (hyperplasia but not hypertrophy) in the CDH fetal lamb model.
466
HOSODA, ROSSMAN, AND GLICK
TPIBW, K TP/BW). Ratios between DNA and TP were calculated (Lu DNA/TP, K DNA/TP, Lu TP/DNA, and K TP/DNA) as a measure of cell size, and differences between experimental groups were thought to reflect hypertrophic or atrophic changes in organ development.i7 Serving as controls, 15 nonoperated litter mates underwent cesarean section at 139 to 145 days of gestation and the lungs and the kidneys were analyzed in a similar manner.
CAFL Model (Fetal Rabbits) New Zealand white rabbits of known gestational age underwent operations to simulate CAFL to produce oligohydramnios and PH.‘s The operations were performed on day 17, 20, or 23 of gestation in the pseudoglandular period of lung development. Hysterotomy and amniotomy were carried out on 3 or 4 sacs of one uterine horn. An S-mm segment of Silastic tubing was placed into the amniotic sac and secured by a purse-string suture. This tubing assured continued drainage of amniotic fluid into the maternal peritoneal cavity. Six fetuses on day 17 of pregnancy, 24 fetuses on day 20, and 20 fetuses on day 23 underwent this procedure. Gestation was then allowed to continue until day 30 of pregnancy, when all fetuses underwent cesarean delivery. The newborn rabbits were checked for survival and then killed. By methods similar to the iamb study, BW, LuW, KW, LuW/ BW, KWIBW, Lu DNA, K DNA, Lu DNA/BW, K DNA/BW, Lu TP, K TP, Lu TPIBW, K TPIBW, Lu DNAITP, K DNA/TP, Lu TP/DNA and K TP/DNA were determined. The liver wet weight (LiW), LiW/BW, liver DNA (LiDNA), Li DNA/BW, liver TP (Li TP), Li TP/BW, Li DNAITP, and Li TPiDNA was used as a control organ to determine the effect of the procedure.la These data were compared to that of corresponding nonoperated litter mates for each group. In another study, 8 operated fetuses of each group underwent reexploration at 27 to 28 days of pregnancy, then each amniotic sac was aspirated of all fluid using a 20-gauge needle and syringe, and the volume was recorded.18.19 These were compared with control fetuses to assess the changes in amniotic fluid volume due to the procedure.
phy) in the CDH model. Comparison of LuW/BW and KW/BW showed that there was an inverse relationship between lung size and kidney size in the fetal lamb that could be described by the equation KW/BW = 1.04 - 0.12 LuW/BW (r = -.75) (Fig 3). The mean amniotic fluid volume of the CDH model was larger than that of controls (848 + 382 v 704 5 227 mL) but difference between groups was not statistically significant. CAFL Model
All six fetuses that were operated on day 17 of pregnancy died in utero for presumed technical causes, and this arm of protocol was stopped. Eighteen of 24 fetuses that were operated on at 20 days and all of 20 that were operated on at 23 days survived until term. This procedure produced severe oligohydramnios when compared to unoperated controls (P < .Ol, Fig 4A). The oligohydramnios from the CAFL resulted in fetuses with smaller lungs and larger kidneys as compared with those of controls (Fig 4B). The lungs were significantly smaller and more hypoplastic when compared by LuW (P < .Ol), LuW/BW (P < .Ol), Lu DNA/BW (P < .05), and Lu TP/BW (P < .Ol, Fig 4C). The kidneys were significantly larger and more hyperplastic than controls as judged by the similar criteria (P < .Ol, Fig 4D). In addition, the
2.0 -
Statistical Analysis The data were analyzed on a PC-286LE computer (EPSON Corp, Tokyo, Japan) using l-2-3 (Lotus Development Corp, Cambridge, MA) and Statview II (Abacus Concepts Inc, Berkeley, CA). Statistical determinations were performed using the MannWhitney U test, with P < .05 considered significant. RESULTS
2 E 9
n
1.5 -
5
CDH Model
Fifteen of 24 lambs that underwent fetal surgery survived until term. The operated lambs had significantly smaller lungs, and larger kidneys when compared with controls (Fig 2A). The lungs were significantly hypoplastic as defined by either decreased LuW/BW, Lu DNA/BW or Lu TP/BW (P < .Ol, Fig 2B). Renal hyperplasia was confirmed by KW/BW, K DNA/BW (P < .Ol) and K TP/BW (P < .05, Fig 2C). However, these differences were not reflected in Lu DNA/TP and K DNA/TP ratios (Fig 2D), indicating that the cell size in the lungs and the kidneys was unchanged, but that the cell population was decreased in lungs (hypoplasia but not atrophy) and was increased in kidneys (hyperplasia but not hypertro-
0.0
’ 0
I
,
!
I
,
1
2
3
4
5
Lung Wt I Body Wt (gm%) Fig 3. The relationship between LuW/BW and KW/BW can be described by the regression equation KW/BW = 1.04 - 0.12 LuW/ BW,r = -.75 (P c .OOl). Squares = controls. Triangles = CDH.
RENAL HYPERPLASIA
& PULMONARY
HYPOPLASIA
IN CDH
467
Rabbits : n=8
1
DNlVBW
Kidney WVBW
Total ProteinlBW
0.8
0.6
0.4
0.2 CAFL C 20 days
A
CAFL C 23 days
E
n “CAFL
C
CAFL 23 d
20 d
Lung
C
“CAFL
C
CAFL
20 d
C
23 d
Kidney
0.4
0.2
B
” “CAFL
C
20d
CAFL 23 d
Lung Weight
C
CAFL 20 d
C
CAFL 23 d
Kidney Weight
C
Fig 4. (A) The fetal rabbit CAFL model produced severe oligohydramnios by decreasing amniotic fluid volume to less than that of controls (**f c .Ol). (B) The effect of the procedure on lung and kidney growth in the rabbf chronic amniotic fluid leak model (**P e .Ol Y controls). (C) The effect of the procedure on lung weight/body weight, lung DNA/ body weight, and lung total protein/ body weight in the rabbit chronic amniotic fluid leak model (*p < .lXi v controls, l*p c .Ol v controls), indicating significant pulmonary hypoplasia. (D) The effect of the procedure on kidney weight/body weight, kidney DNA/body weight, and kidney total protein/body weight in the chronic amniotic fluid leak model (**P < .Ol v controls), indicating significant renal hyperplasia. (E) Differences were not reflected in lung DNA/TP and kidney DNA/TP ratios, indicating that the cell size in the lungs and the kidneys was unchanged, but that the cell population was decreased in lungs (hypoplasia but not atrophy) and was increased in kidneys (hyperplasia but not hypertrophy) in the rabbit CAFL model.
HOSODA, ROSSMAN,
468
earIier in gestation the operation was performed, the greater was the effect on the kidneys and lungs. Similar to the lamb study, DNA/TP and TP/DNA ratios of lungs and kidneys did not differ significantly between groups and thus cell atrophy or hypertrophy are not involved in the changes in organ size which were observed (Fig 4E). The liver did not vary significantly between groups, and its growth appeared to be unaffected by the procedure. DISCUSSION
These results in the fetal lamb support our hypothesis that PH in CDH is associated with significant renal enlargement. The renal enlargement is hyperplastic and not hypertrophic growth. This experimental verification confirms our previous clinical observation.’ According to our hypothesis, in the fetus with CDH, the space-occupying nature of the herniated viscera hinders normal pulmonary growth and development and results in PH. The lungs may “sense” this inadequate pulmonary growth and development and secrete excess PDR. The kidneys would respond to this afferent signal, enlarge by hyperplastic growth, and secrete increased amounts of PGF. We had previously speculated that, if our theory were true and broadly applicable to other instances of PH, then PH associated with oligohydramnios, ie, CAFL, should also be associated with renal enlargement. Our data now confirm this and the effect in rabbits seems to be proportional to the length of gestation in which the stimulus exists. Presumably in
AND GLICK
CAFL, as in CDH when the lungs “sense” inadequate growth, they also produce increased amounts of PDR. The kidneys become hyperplastic and produce more PGF, but the CAFL and oligohydramnios persists and PH results. The precise etiologic role of amniotic fluid in the process of lung growth is not known. Several possible etiologies for PH with oligohydramnios have been described.i8-20 Oligohydramnios may cause fetal thoracic compression by the amniotic membranes and the uterine wall and limit intrathoracic space and fetal breathing movement.2’-23 However, attempts at intraamniotic saline infusions to reconstitute amniotic fluid volume have not been successful in preventing PH. 24Therefore, we believe the quality of amniotic fluid rather than just the quantity may be important. PGF may be an important component of the amniotic fluid to promote proper lung growth and development. The present data encourage continued investigation to identify, characterize, and synthesize PGF and PDR. These may be unique growth factors specific to the lungs and kidneys, or ubiquitous growth factors being modulated in a unique fashion under the physiologic circumstances. In the future, PGF might be used therapeutically for fetuses and newborns with PH. Accelerating pulmonary growth and development in newborns with CDH or PH due to other causes may hasten weaning from mechanical ventilation or extracorporeal membrane oxygenation and decrease their mortality and morbidity.
REFERENCES 1. Anderson KD: Congenital diaphragmatic hernia, in Welch KJ, Randolph JG, Ravitch MM, et al (eds): Pediatric Surgery (ed 1). Chicago, IL, Year Book, 1986, pp 589-601 2. Reynolds M, Luck SR, Lappen R: The “critical” neonate with diaphragmatic hernia: A 21-year perspective. J Pediatr Surg 19:364-369,1984 3. Touloukian RJ, Markowitz RI: A preoperative x-ray scoring system for risk assessment of newborns with congenital diaphragmatic hernia. J Pediatr Surg 19:252-257, 1984 4. Simson JNL, Eckstein HB: Congenital diaphragmatic hernia: A 20 year experience. Br J Surg 72:733-736, 1985 5. 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 6. Steinhorn RH, Kriesmer PJ, Green PP, et al: Natural history of congenital diaphragmatic hernia in Minnesota: Impact of in utero diagnosis on survival. Pediatr Res 29:236A, 1991 (abstr) 7. Glick PL, Siebert JR, Benjamin DR: Pathophysiology of congenital diaphragmatic hernia I: Renal enlargement suggests feedback modulation by pulmonary derived renotropins-A unifying hypothesis to explain pulmonary hypoplasia, polyhydramnios, and renal enlargement in the fetus/newborn with congenital diaphragmatic hernia. J Pediatr Surg 25:492-495,199O
8. 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 9. Adzick NS, Harrison MR, Flake AW, et al: Automatic uterine stapling device in fetal surgery. Surgical Forum 36:479-481, 1985 10. Strauss WM: Preparation of genomic DNA from mammalian tissue, in Ausubel FM, Brent R, Kingston RE, et al (eds): Current Protocols in Molecular Biology 1. New York, NY, Wiley, 1988, Sections 2.2.1-2.2.3 11. Brunk CF, Jones KC, James TW: Assay for nanogram quantities of DNA in cellular homogenates. Anal Biochem 92:497500,1979 12. Labarca C, Paigen K: A simple, rapid, and sensitive DNA assay procedure. Anal Biochem 102:344-352,198O 13. Wigglesworth JS, Desai R: Use of DNA estimation for growth assessment in normal and hypoplastic fetal lungs. Arch Dis Child 56:601-605,198l 14. Wigglesworth JS, Desai R, Guerrini P: Fetal lung hypoplasia: biochemical and structural variations and their possible significance. Arch Dis Child 56:606-615, 1981 15. Wigglesworth JS, Desai R: Is fetal respiratory functional a major determinant of perinatal survival? Lancet 1:264-267, 1982
RENAL HYPERPLASIA & PULMONARY HYPOPLASIA IN CDH
16. Bradford MM: A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248-254, 1976
17. Winick M, Noble A: Quantitative change in DNA, RNA and protein during prenatal and postnatal growth in the rat. Dev Biol 12:451-466, 1965 18. Nakayama DK, Glick PL, Harrison MR, et al: Experimental pulmonary hypoplasia due to oligohydramnios and its reversal by relieving thoracic compression. J Pediatr Surg 18:347-353,1983 1Y. Adzick NS. Harrison MR, Glick PL, et al: Experimental pulmonary hypoplasia and oligohydramnios: Relative contributions of lung fluid and fetal breathing movements. J Pediatr Surg 19:658-665,1984 20. Perlman M. Williams J, Hirsch M: Neonatal pulmonary
469
hypoplasia after prolonged leakage of amniotic Huid. Arch Dis
Child 51:349-353,1976 21. Blott M, Greenough A, Nicolaides KH, et al: Fetal breathing movements as predictor of favorable pregnancy outcome after oligohydramnios due to membrane rupture in second trimester. Lancet 2:129-131, 1987 22. Collins MH, Moessinger AC, Kleinerman J. et al: Morphometry of hypoplastic fetal guinea pig lungs following amniotic fluid leak. Pediatr Res 20:955-960, 1986 23. Harrison MR, Jester JA, Ross NA: Correction of congenital diaphragmatic hernia in utero. I. The model: lntrathoracic balloon produced fetal pulmonary hypoplasia. Surgery 88:174-182,198O 24. Glick PL, Harrison MR, Adzick NS, et al: Management of fetus with congenital hydronephrosis II: Prognostic criteria and selection for treatment. J Pediatr Surg 20:376-387. 1985
Discussion T.F. Tracy (St Louis, MO): This excellent presentation by Dr Hosado and mentored by Dr Glick has really given us two distinct experiments that give us some insight into the long sought after pulmonaryrenal axis in fetal development. Throughout the 1970s Dr William Blanc of Columbia University, struggled with this problem and the problem of pulmonary hypoplasia that resulted from oligohydramnios. In a rat model with amniotic fluid leak in late gestation, he not only succeeded in inducing pulmonary hypoplasia but also succeeded in inducing global intrauterine growth retardation. Hence he had the wrong model. In the 1980s many studies by Dr Harrison and his colleagues at the University of San Francisco took up the banner with a successful larger animal model and were able to demonstrate that oligohydramnias indeed did induce pulmonary hypoplasia. However, with the removal of the thoracic compression, the hypoplasia was also reversed. Unfortunately, no observations throughout these several studies noted any renal hypertrophy or hyperplasia in any of these experiments. In the 1990s this search for this Holy Grail of lung development continues in Buffalo. In these elegant studies their hypothesis is that fetal lung growth is directed by fetal renal growth. Also there is a second arm to their hypothesis that there is feedback loop from the lung to the kidney. In this paper today, it is clear that renal hyperplasia follows two sequential modifications that induce pulmonary hypoplasia. This confirms the second arm of their hypothesis in that a small lung equals a large kidney. One of the most interesting figures was the regression analysis for those animals with congenital diaphragmatic hernia. Significant regression for small lungs leading to large kidneys as achieved by combining
control animals as well as experimental animals. My first question is, does this same regression coefficient apply in normal controls as well as those diaphragmatic controls? In other words, is there a graded hypoplasia of the lung that results in a graded hyperplasia within the kidney? If your hypothesis is correct, the same regression coefficient should be found in controls and experimental animals, yet both groups appear flat. The second question pertains to the problems with amniotic fluid leak. Your attempts at early induction of amniotic fluid leak led to increased fetal loss. Was Dr Blanc right in that amniotic fluid leak is capable of inducing global growth retardation and only late gestational changes in amniotic fluid volume will result in the changes that you found in your kidneys? One minor point with your DNA body weight data in this chronic amniotic fluid model was that your kidney weight loss was significant with increases in your margin of .07 to .04. You had some interesting liver data that showed a similar increase in DNA synthesis at the same time. Could we not be seeing a global regenerative response as a stimulus for proliferation? J. Lunger (St Louis, MO): If your hypothesis is correct, I assume there is a single cell type in the kidney that is producing this pulmonary growth factor. The hypertrophy in the kidney would be caused by hypertrophy of that single cell type. Have you done any histological studies to determine what cell type that is? .I. Wilson (Boston, MA): We looked at your data very carefully and even compared your kidney weights with those of our own control animals in our fetal studies and there is no question in my mind that your kidneys are hyperplastic. I think this is fascinating but I am not sure I agree with your hypothesis. In our own
470
studies of lung growth in fetal lambs, we have been able to correlate pulmonary hypoplasia with bilateral nephrectomy which is certainly in keeping with your model. However, by performing a tracheal ligation on these anephric animals, we have obtained lungs 3 times the size of normal lungs and 5 times the size of the lungs with nephrectomy alone. By volume, weight, alveolar count, and protein/DNA ratios, these are hyperplastic lungs despite the fact that no kidneys were present from either 70 or 90 days’ on. We have repeated this model utilizing experimental diaphragmatic hernia rather than nephrectomy as the promoter of pulmonary hypoplasia with similar results. I believe your description of renal hyperplasia is real but can you reconcile your model of feedback employing a renally previous pulmonary growth factor with our data in anephric fetal sheep? RL. Glick (response): Thank you very much for your comments. I would like to begin with Dr Wilson’s comments. We had originally speculated that PDR and PGF were unique growth factors produced by the lungs and the kidneys, respectively. In work we have not yet published from our laboratory, we have shown that fetal serum, amniotic fluid, and homoge-
HOSODA, ROSSMAN, AND GLICK
nates of lungs from the diaphragmatic hernia model stimulate kidney organ cell cultures. And serum, amniotic fluid and kidney homogenates in organ lung cultures stimulate the kidneys to grow. However, in light of Dr Wilson’s data, we have to rethink our hypothesis. We think that PDR and PGF may be unique growth factors specific to the lungs and kidneys. However, they may be ubiquitous growth factors being modulated in an unique fashion under these physiologic circumstances. We have not actually tried to find these growth factors as much as to prove that they are there. It may turn out that they are existing growth factors being modulated uniquely in this situation. So I don’t think that Dr Wilson’s data are incompatible with our original hypothesis. The PGF may just be coming from another organ. As far as Dr Tracy’s comments go, if you look at the regression curve, the control animals also fit right along that regression curve in that the control animals that have smaller lungs have larger kidneys. The liver data are interesting and that may go along with Dr Wilson’s thought that this is a ubiquitous growth factor.