Main

CDH is an anomaly occurring 1 in 3000 live births (1). It is characterized by a diaphragmatic defect, severe lung hypoplasia, and pulmonary hypertension, and in 40% of the patients other severe birth defects such as cardiac abnormalities are present (2, 3). Despite years of extensive research, the etiology of CDH remains unknown (4). Clinically, pulmonary hypoplasia and pulmonary hypertension form the major problems in CDH (5).

Many CDH studies have focused on treatment modalities such as conventional ventilation with gentle handling of the fragile lung, high-frequency ventilation, ECMO, in utero tracheal ligation with or without betamethasone injection, inhaled nitric oxide, and prenatal injections of betamethasone, TSH-releasing hormone, or a combination of both hormones. Although selected centers have reported improved survival (6, 7), the overall mortality rate, however, is still variably high, so that CDH continues to be a serious problem in the neonatal and pediatric surgical intensive care unit, with optimal treatment for CDH still the subject of ongoing debate (5, 8–10).

It is unknown why CDH infants can be so refractory to treatment, which makes it even more difficult to predict the outcome in the individual patient. Lungs from premature infants suffering from RDS are surfactant-deficient and major breakthroughs in the treatment of these infants were prenatal corticosteroid injections and the postnatal administration of exogenous surfactant (11–14). Surfactant is essential for proper lung function. It decreases the surface tension in the lung, thereby protecting alveoli against collapse at end-expiration (15). We realize that it is not justified to directly compare CDH lungs to RDS lungs, but the possibility that CDH lungs are surfactant-deficient is still not ruled out or proven. In human autopsy material of pulmonary hypoplasia including CDH, normal levels of SP-A, -B, and -C protein were reported (16). Also, normal levels of phospholipids and normal lecithin/sphingomyelin (L/S) ratios (used as indicator of fetal lung maturity) in amniotic and bronchoalveolar lavage (BAL) fluid have been reported (17, 18). Nevertheless, other studies have found decreased amounts of SP-A in human CDH (19, 20). It has to be kept in mind that a secondary surfactant deficiency may develop due to the negative effects of artificial ventilation on surfactant function after birth (21). So far, data from human and different animal CDH models (22) failed to report consistent results concerning the surfactant status in CDH (summarized in Table 1) (17–20, 23–39).

Table 1 Surfactant in congenital diaphragmatic hernia
Full size table

The most common method to determine the amount of RNA in tissue is Northern blot analysis. This method, however, requires tissue homogenization, with the result that all cellular relations are lost. Jonker et al.(40) have developed a method to quantify the signal obtained with the in situ hybridization procedure. This quantitative in situ hybridization technique allows the detection of the concentration of mRNA at the cellular level within the architecture of the tissue.

In this study, we have used this technique to accurately assess the cellular concentration of SP-A, -B, and -C mRNA in the nitrofen-induced CDH rat model (41). We observed no differences in the expression patterns or cellular concentration of the SP mRNAs between control and CDH embryonic rat lungs, except for the level of SP-C mRNA, which was higher in CDH than in control lungs at term [gestational day (E)22]. No differences between control and CDH lungs were found upon immunohistochemical visualization of the respective surfactant proteins. At term, the volume fraction of SP mRNA-expressing cells was higher in CDH lungs than in controls. Furthermore, no differences in the expression pattern, concentration of SP mRNA per cell or the volume fraction of SP mRNA-expressing cells was found between the ipsilateral (hernia) and contralateral (no hernia) side of a CDH lung. These results demonstrate that the embryonic rat with nitrofen-induced CDH is not deficient for pulmonary SP-A, -B, or -C.

METHODS

Animals.

Adult Wistar rats were purchased from the Broekman Institute B.V. in Someren (The Netherlands). Rats were mated at the end of the morning (E0). To induce CDH and lung hypoplasia, 100 mg of nitrofen dissolved in 1 mL olive oil was administered orally on E10 (term = E23) (38, 41, 42). To determine the surfactant status of lungs from embryonic rats with nitrofen-induced CDH, we examined the expression patterns of SP-A, -B, and -C mRNA and protein in control and CDH fetuses of E15, 18, 20, and 22. In rat, E15 corresponds to the mid-pseudoglandular stage of lung development, E18 to late pseudoglandular, E20 to late canalicular, and E22 to the saccular stage of lung development. Animal experiments were performed in accordance with the guidelines of the animal research committee of the Academic Medical Center of the University of Amsterdam.

Tissue preparation.

At E15–E22, embryos were delivered by cesarean section and their thoraxes were fixed in 4% phosphate-buffered formaldehyde (wt/vol) (4°C, 16–18 h), dehydrated in a graded series of ethanol solutions, cleared with 1-butanol, and embedded in Paraplast Plus (Monoject, Kildare, Ireland). Seven-micrometer frontal sections were cut and mounted onto RNAse-free 3-aminopropyltriethoxysilane-(Sigma Chemical Co., St. Louis, MO, U.S.A.) coated slides. From each embryo, the entire thorax was embedded and cut, and only embryos that had a visible diaphragmatic hernia and lung hypoplasia were used for the study.

In situ hybridization.

Plasmids containing mouse SP-A and SP-C (both in pGEM3Z) and SP-B (in pBS-SKII) cDNA were obtained from Dr. J.A. Whitsett (Division of Pulmonary Biology, Children's Hospital Medical Center, Cincinnati, OH, U.S.A.) (43–45). [α-35S]dCTP-labeled antisense probes for SP-A, -B, and -C were generated with T7 polymerase, after linearization of the plasmids with ApaLI, Eco RI, and Hin dIII, respectively. The hybridization conditions were as described elsewhere (46). Exposure time to nuclear autoradiographic emulsion (Ilford Nuclear Research Emulsion G-5; Ilford, Cheshire, UK) was 14 d for SP-A and 7 d for SP-B and -C. The development time was 4 min. After developing, the sections were dehydrated in a graded series of ethanol and xylol, and mounted in Malinol (Chroma-Gesellshaft, Schmidt Gmbh+Co, Köngen, Germany).

Image acquisition and analysis.

For image acquisition, a Photometrics cooled-CCD camera (Tucson, AZ, U.S.A.; 12-bit dynamic range; 1317 × 1035 pixels), attached to an Axioplan microscope (Carl Zeiss, Oberkochen, Germany) equipped with a 5× objective (N.A.: 0.15), a stabilized power supply, and an infrared-blocking filter was used. The low-power objective was used to assure the sampling of several respiratory acini in an image, representing 2.5 mm2 of the lung section. Digital images from the in situ hybridization procedure were recorded using white light (40). The digital transmission images were converted to OD images by calculating the negative logarithm of the transmission image divided by an image of the light source (OD = −10log(I/I0) for each pixel). This conversion implicitly corrects background shading.

The OD images were analyzed using the public domain image analysis program from the National Institutes of Health—Image (available at rsb.info.nih.gov/nih-image; version 1.61). The areas to be measured were marked by an interactive density slice, which identifies structures based on a selected lower and upper density value. Tissue background was defined as nonstaining tissue such as esophagus or cartilage tissue. Signal was defined as specific positive staining due to hybridization (signal in the bronchiolar and alveolar epithelium) (40, 46). The definitive signal value, expressed as mean OD, is obtained by subtracting the mean tissue background density from the mean signal density (40, 46). The positive signal is plotted as mean OD (± SEM) per group and age. The volume fraction of SP mRNA-expressing cells (% Area) was calculated from the area of positive staining cells in the section and the total lung area (excluding airspaces). All lung measurements were carried out three times with a randomized series of images. No differences were observed between the three measurements and, therefore, a mean value per lung was calculated.

Immunohistochemistry.

All antibodies were obtained from Dr. J.A. Whitsett (Division of Pulmonary Biology, Children's Hospital Medical Center, Cincinnati, OH, U.S.A.). A guinea pig polyclonal antibody against rat SP-A, a rabbit polyclonal antibody against mature bovine SP-B (R28031), and a rabbit polyclonal antibody against human proSP-C (R68514) were applied at a dilution of 1:300, 1:2000, and 1:5000, respectively, to 7 μm deparaffinized sections (47, 48). For SP-A, a Vectastain ABC Peroxidase kit and for SP-B and proSP-C a Vector Elite ABC-DAB kit (Vector Laboratories, Burlingame, CA, U.S.A.) were used to detect antigen-antibody complexes (48). The enzymatic reaction product was enhanced with nickel/cobalt to produce a black precipitate. Sections were counterstained with nuclear fast red. For each different protein, slides were handled in similar ways concerning the concentration of the antibody and the times for incubation and detection. Analysis of the pattern and intensity of protein staining was done blindly by two persons.

Statistical analysis.

Variations in staining intensity due to the different in situ hybridization sessions were removed by subtracting the session effects calculated from the log-transformed data with the general linear model ANOVA without interaction (SPSS version 10.0.7; SPSS Inc., Chicago, IL, U.S.A.). Differences between groups (control versus CDH) and age [two animals per group per age for E15, E18, E20, and E22 (control) and three animals per group for E22 (CDH)] and between ipsilateral (hernia) and contralateral (no hernia) lung and age were tested with a two-way ANOVA.

RESULTS

In situ hybridization.

SP-A and -B mRNA are normally expressed in both bronchiolar and alveolar epithelial cells, whereas SP-C mRNA is only expressed in alveolar epithelial cells. The patterns of mRNA expression for SP-A, -B, and -C did not differ between control (Figs. 1–3, panels A, C, E) and CDH (Figs. 1–3, panels B, D, F) lungs at the gestational ages analyzed. SP-A and -B mRNA expression started at a very low level at E15 (not shown). SP-A mRNA was expressed in alveolar epithelial cells and from E20 onwards also at low levels in bronchiolar epithelial cells (Figs. 1–3, panels A and B). SP-B mRNA was expressed in alveolar epithelial cells and from E20 onwards as well at high levels in bronchiolar epithelial cells (Figs. 1–3, panels C and D). At E15, SP-C mRNA was already expressed at a high level in pulmonary epithelial cells (not shown). At later stages, SP-C mRNA was expressed only in alveolar (distal) epithelial cells (Figs. 1–3, panels E and F). No differences in the expression patterns of the respective SP mRNAs were observed between the ipsilateral and the contralateral side of CDH lungs (not shown). Figure 3 also revealed that at E22 the morphology of the CDH lungs differed substantially from the control lungs. Although pictures were taken at the same magnification, CDH lungs (Fig. 3, B, D, F) were smaller than control lungs (Fig. 3, A, C, E) and contained only a few visible airspaces, in contrast to the control lungs where airspaces were abundant at this gestational age. Note also that the airspaces in CDH lungs (Fig. 3, B, D, F) are poorly expanded.

Figure 1
figure 1

Expression of SP-A (A, B), SP-B (C, D), and SP-C (E, F) mRNA in serial sections of lungs of E18 control (A, C, E) and CDH (B, D, F) rat fetuses. SP-A, -B, and -C mRNA were expressed in the alveolar epithelial cells of both control and CDH lungs. Bar = 200 μm.

Full size image
Figure 3
figure 2

Expression of SP-A (A, B), SP-B (C, D), and SP-C (E, F) mRNA in serial sections of lungs of E22 control (A, C, E) and CDH (B, D, F) rat fetuses. In both control and CDH lungs, SP-A, -B, and -C mRNA were detected in alveolar epithelial cells. SP-A and -B mRNA were also expressed in bronchiolar epithelial cells, whereas SP-C mRNA was never detected in these cells. v, blood vessel; b, bronchiole. Bar = 200 μm.

Full size image

Quantification of the in situ hybridization results.

The cellular concentration (mean OD) of SP-A, -B, and -C mRNA in control and CDH lungs increased with advancing gestational age (Fig. 4). All SP mRNAs increased in cellular concentration (Fig. 4, A, C, E), whereas the volume fraction of cells (% Area) expressing the respective mRNAs also increased (Fig. 4, B, D, F). Nevertheless, the behavior of the SP mRNAs differed slightly from each other. SP-C mRNA levels increased early (Fig. 4E) and before SP-A and SP-B mRNA levels (Fig. 4, A and C). The mean OD of SP-A and SP-B mRNA was similar between control and CDH lungs at all ages (Fig. 4, A and C). The mean OD of SP-C mRNA was similar between control and CDH lungs between E15 and E20, and higher in CDH lungs at E22 (Fig. 4E, group · age effect, p = 0.007). For SP-A, -B, and -C mRNA, the volume fraction of mRNA-expressing cells (% Area) was similar between CDH and control lungs until E20 and was higher in CDH lungs at E22 (Fig. 4, B, D, F; group · age effect, p = 0.009, p = 0.014, p = 0.002, respectively). These results show that the cellular concentration (mean OD) of SP mRNA in the CDH lungs is equal to the concentration found in a similar cell in control lungs for SP-A and -B. For SP-C mRNA, there is increased expression in CDH lungs at the end of gestation (E22). Furthermore, the volume fraction of cells (% Area) expressing SP mRNA as a percentage of total lung tissue (excluding airspaces) is higher near term in CDH lungs. In summary, these results indicate that there is no deficiency of SP-A, -B, or -C mRNA in CDH lungs.

Figure 4
figure 3

Quantification of the in situ hybridization signals for SP-A (A, B), SP-B (C, D), and SP-C (E, F) mRNA in control (solid circles) and CDH (open squares) lungs. Mean OD was measured as described in the “Methods” section and indicates the cellular concentration of SP mRNA (A, C, E). The volume fraction of cells expressing SP mRNA is given as a percentage (% Area) of the total lung tissue measured (both staining and nonstaining lung tissue, airspaces excluded) (B, D, F). Both parameters increased with age in control and CDH lungs. At E22, the mean OD of SP-C mRNA and the % Area of cells expressing SP-A, -B, and -C mRNA was higher in CDH than in control lungs (significant group · age effect, p = 0.007, 0.009, 0.014, and 0.002, respectively).

Full size image

Because the lung at the side of the diaphragmatic defect (ipsilateral lung) was always more hypoplastic than the contralateral lung, we investigated the cellular concentration (mean OD) of SP mRNA (Fig. 5, A, C, E) and the volume fraction (% Area) of SP-expressing cells (Fig. 5, B, D, F) between the ipsilateral and contralateral side of a CDH lung. In both the ipsilateral and contralateral CDH lung, the mean OD of SP-A, -B, and -C mRNA increased with gestational age and no differences were observed between the two lungs (Fig. 5, A, C, E). Similarly, the % Area of SP-expressing cells increased with gestational age and no differences were found between the ipsilateral and contralateral side of a CDH lung (Fig. 5, B, D, F).

Figure 5
figure 4

Quantification of the in situ hybridization signals for SP-A (A, B), SP-B (C, D) and SP-C (E, F) mRNA in the contralateral (no hernia, solid circles) and ipsilateral (hernia, open squares) side of CDH lungs. Mean OD was measured as described in the “Methods” section and indicates the cellular concentration of SP mRNA (A, C, E). The volume fraction of cells expressing SP mRNA is given as a percentage (% Area) of the total lung tissue measured (both staining and nonstaining lung tissue, airspaces excluded) (B, D, F). Both parameters increased with advancing gestational age and no differences were found between the ipsilateral and contralateral side of CDH lungs.

Full size image

Immunohistochemistry.

SP-A (Fig. 6) and SP-B (Fig. 7) proteins were detected in the bronchiolar and alveolar epithelial cells of both control and CDH lungs. Expression of both proteins increased in control and CDH fetuses with advancing gestational age and no differences in expression pattern or staining intensity were observed between the control (Figs. 6 and 7, A and C) and CDH (Figs. 6 and 7, B and D) lungs at E20 (Figs. 6 and 7, A and B) or E22 (Figs. 6 and 7, C and D). No differences in the pattern or intensity of SP-A or -B staining were observed between the ipsilateral (hernia) and contralateral (no hernia) side of E22 CDH lungs (not shown). Clear proSP-C staining was observed in control (Fig. 8, A and C) and CDH (Fig. 8, B and D) lungs at E20, where it appeared higher in CDH (Fig. 8B) lungs than in control (Fig. 8A) lungs. However, at E22 (Fig. 8, C and D) this difference could no longer be observed. In both control and CDH lungs, proSP-C was only detected in alveolar epithelial cells. At E22, there was no difference in intensity or pattern of proSP-C staining between the ipsilateral and the contralateral side of CDH lungs (not shown). These results show that the respective mRNAs are properly translated into protein and therefore indicate that lungs of nitrofen-induced CDH rat fetuses do not have a primary deficiency of surfactant proteins.

Figure 6
figure 5

Immunohistochemical staining for SP-A at E20 (A, B) and E22 (C, D) in control (A, C) and CDH (B, D) rat lungs. SP-A was expressed in alveolar and bronchiolar epithelial cells of both control and CDH lungs. Note the difference in morphology between control and CDH lungs at E22. v, blood vessel; b, bronchiole. Bar = 100 μm.

Full size image
Figure 7
figure 6

Immunohistochemical staining for SP-B at E20 (A, B) and E22 (C, D) in control (A, C) and CDH (B, D) rat lungs. SP-B was expressed in alveolar epithelial cells of both control and CDH lungs. Some staining is also detected in E20 bronchiolar epithelial cells. At E22, SP-B was expressed at high levels in alveolar and bronchiolar epithelial cells of both control and CDH lungs. v, blood vessel; b, bronchiole; br, bronchus. Bar = 100 μm.

Full size image
Figure 8
figure 7

Immunohistochemical staining for proSP-C at E20 (A, B) and E22 (C, D) in control (A, C) and CDH (B, D) rat lungs. ProSP-C was expressed in alveolar epithelial cells of both control and CDH lungs. ProSP-C was never observed in bronchiolar epithelial cells. v, blood vessel; b, bronchiole; br, bronchus. Bar = 100 μm.

Full size image

DISCUSSION

CDH is characterized by a diaphragmatic defect, pulmonary hypoplasia, and pulmonary hypertension. In the CDH neonate, pulmonary hypoplasia is often complicated by respiratory failure, low lung compliance, increased inspiratory resistance, and hyaline membrane formation (5). Based on inconsistent results from human and experimental animal studies with regard to the use of surfactant (prophylactic or rescue therapy) in CDH, we tested the hypothesis that CDH lungs are surfactant-deficient, which could explain in part the respiratory failure and difficulties in treating CDH infants. In this study, we showed that mRNA and protein expression of SP-A, -B and -C, in both control and CDH lungs, increased with advancing gestational age and no differences were present in the cellular concentration (mean OD) of SP-A, -B, and -C mRNA between control and CDH lungs except for a higher concentration of SP-C mRNA at term (E22) in CDH lungs. The volume fraction of SP mRNA-expressing cells (% Area) similarly increased with gestational age in control and CDH lungs and near term more SP mRNA-expressing cells were found in CDH lungs than in controls. This increased volume fraction of SP mRNA-expressing cells is in agreement with the increased number of surfactant-producing type II cells in CDH lungs. Using electron microscopy, it was shown that CDH lungs from fetal lambs (36, 49) and rats (50) had increased numbers of pulmonary type II cells, and more recently it was shown that those cells exhibit normal type II cell maturation (35). Furthermore, no differences were found in the cellular concentration of the SP mRNAs or in the volume fraction of SP mRNA-expressing cells between the ipsilateral (hernia) and contralateral (no hernia) side of CDH lungs. This study therefore demonstrates that there is no primary deficiency of surfactant proteins in the nitrofen-induced CDH rat model. It is, however, possible that the total amount of surfactant protein is less because CDH animals have smaller lungs.

Data on the surfactant status of CDH humans and experimental animals are not consistent. Both normal and decreased levels for DSPC and SP-A protein and mRNA were reported in CDH lungs. SP-C protein and mRNA levels were more often reported normal than decreased, whereas SP-B protein levels were often decreased, however, normal mRNA levels were also found in CDH compared with control lungs (Table 1). Despite animal data, neither a primary surfactant deficiency nor a beneficial effect of surfactant replacement therapy has been proven in human CDH infants (18, 51–53). These apparent discrepancies and the variably high mortality rate of CDH infants led us to investigate the surfactant protein status in the nitrofen-induced CDH rat model using a novel in situ hybridization technique.

The key finding in this study was that with quantitative in situ hybridization we showed that in the same sample, the expression pattern and the cellular concentration of SP-A and SP-B mRNA did not differ between CDH and control lungs, whereas the expression pattern of SP-C mRNA was similar, but the level of expression was higher in CDH than in control lungs at the end of gestation (E22). The quantitative in situ hybridization technique is an adequate approach to measure differences in mRNA level in the context of changing tissue architecture or a changing expression pattern of a gene (40, 46, 54). These simultaneous observations cannot be made with techniques that require tissue homogenization, such as Northern blot analysis. The advantage of quantitative in situ hybridization is that it is possible to locate a positive staining cell in the original tissue and that at the same time it is possible to measure mRNA expression levels per cell. Our study was carried out in the laboratory where the quantitative in situ hybridization technique was developed and extensively validated with Northern blot analysis (40, 46). Jonker et al.(40) showed that the integrated OD (mean OD) of silver grains produced in liver and intestinal sections by the in situ hybridization procedure using 35S-labeled riboprobes (i.e. positive signal) is directly proportional to the signal obtained by quantitative Northern blot analysis and, more recently, Moorman et al.(55) used the very same technique to distinguish cardiac specific mRNA expression levels in different parts of the developing heart. In further support, the line of increase of SP mRNA levels with advancing gestational age in control rat lungs in our study is in agreement with published data from Schellhase et al.(56). These authors used Northern blot analysis to show that in control fetal rat lungs, SP-A and SP-B mRNA levels are very low before E17 and markedly increase from E18 to E19 and further increase till E21 (term = E22). Similar to our study (term = E23), they also showed that SP-C mRNA levels are already high at E17, increase dramatically toward E19, and further increase toward adult levels by E21 (56).

We think that the measurement of mRNA concentration in the cell of interest (via in situ hybridization), as opposed to the measurement of mRNA levels per tissue volume (via e.g. Northern blot analysis) can account for the observed differences between our study and that of Thébaud et al.(38). Using Northern blot analysis, these latter authors found normal levels of SP-B and decreased levels of SP-A and SP-C mRNA in nitrofen-induced CDH rats at E21 (term = E22). Northern blot analysis includes all cells, including non-SP-producing cells, in the tissue base, whereas the in situ hybridization approach only includes SP-producing cells. In view of the altered morphology of CDH lungs, the contribution of non-SP-producing cells (e.g. fibroblasts) is larger in hypoplastic CDH lungs than in control lungs, especially near term as was recently shown by Guilbert et al.(35). In this case, the key advantage of quantitative in situ hybridization is the possibility to avoid the contribution of fibroblast and other nonsurfactant-producing cells and selectively measure SP mRNA levels in pulmonary epithelial cells.

From the study by Guilbert et al.(35) a similar conclusion can be drawn. These authors used RNA protection assay to measure the different levels of SP mRNA in nitrofen-induced CDH rats and showed normal SP-A, -B, and -C mRNA levels in contralateral CDH lungs compared with control lungs at any time and only decreased levels of SP-B and -C mRNA in ipsilateral CDH lungs at E21 (term = E22), which is the time when morphologic abnormalities (increased ratio of mesenchyme to epithelium) in CDH lungs are most apparent (35, 57). However, they also showed that in situ hybridization for SP-C mRNA did not show any difference between control and CDH lungs at E21, which supports the results presented in our study.

Using immunohistochemistry, we found no differences in the protein staining pattern or intensity of SP-A, -B, and proSP–C, indicating that the respective mRNAs are properly translated into proteins. The surfactant protein antibodies used in this study were previously shown to be highly specific (16, 48). Our immunohistochemistry results are in line with results from Zhou et al.(16), who found no differences in the expression of SP-A, -B, and proSP-C proteins in various cases of human lung hypoplasia, including CDH.

DSPC is as important for the function of surfactant as the surfactant proteins. Secreted surfactant is composed of 90% lipids, 9% proteins, and 1% carbohydrate (15). Experimental animal data support both decreased and normal levels of DSPC in CDH (Table 1). Interestingly, a recently published paper showed that DSPC synthesis in human CDH infants on ventilation was not different from control infants (no respiratory disease) on ventilation (52). Similarly, our group showed that CDH infants had a normal % DSPC in their BAL fluid (18) and more recently that CDH infants on ECMO are not surfactant-deficient when compared with infants on ECMO who suffer from meconium aspiration or with ventilated CDH infants (51). These results indicate that a primary surfactant deficiency is not likely in CDH infants.

That impaired lung development and subsequent respiratory distress can develop without a primary SP deficiency was also shown in an entirely different experimental model (58). Fetal rats with oligohydramnios suffer from a marked pulmonary hypoplasia; in humans, this often leads to respiratory distress at birth. However, despite the profound pulmonary hypoplasia, normal levels of saturated PC (phosphatidyl choline), SP-A, SP-B, and SP-C mRNA were found (58).

Although we did not find a primary deficiency in surfactant proteins, it is still possible that a secondary deficiency develops postnatally due to the nature of the disease with morphologically immature lungs and/or postnatal ventilation strategies. If (secondary) surfactant deficiency were a major component in the clinical course of CDH, one would expect beneficial results from prenatal corticosteroid injections or surfactant replacement therapy, treatments that have dramatically reduced the mortality rate in surfactant-deficient RDS infants. Although animal CDH models showed an improvement of lung morphology after prenatal corticosteroid treatment (59–62), unfortunately, no human randomized controlled trials for the use of surfactant or prenatal corticosteroids in CDH have been carried out and only anecdotal data are available (63–65), which, as recently reviewed (53), do not support a primary surfactant deficiency in CDH.

CONCLUSION

In conclusion, this study showed that there is no primary deficiency of surfactant proteins in the nitrofen-induced rat model of CDH. If extrapolation to the human situation is allowed, it is possible that the respiratory failure in CDH infants does not necessarily result from biochemically immature lungs but rather from morphologic immaturity and perhaps vascular abnormalities. Above all, a significant contribution to respiratory insufficiency in CDH is iatrogenic, as suggested in recent papers and a secondary surfactant deficiency cannot be excluded under these circumstances (21).

Figure 2
figure 8

Expression of SP-A (A, B), SP-B (C, D), and SP-C (E, F) mRNA in serial sections of lungs of E20 control (A, C, E) and CDH (B, D, F) rat fetuses. In both control and CDH lungs, the three mRNAs were detected in alveolar epithelial cells. At this stage of development, SP-B mRNA was also detected in bronchiolar epithelial cells of both control and CDH lungs. Weak SP-A and no SP-C mRNA was detected in bronchiolar epithelial cells. v, blood vessel; b, bronchiole. Bar = 200 μm.

Full size image