Is there a determining factor that predicts mortality in patients with congenital diaphragmatic hernia?

Introduction

Congenital diaphragmatic hernia (CDH) is a rare congenital anomaly causing severe respiratory distress. In this abnormality, the abdominal organs enter the thorax through a diaphragmatic defect. Although it varies between 1/2000 and 1/5000, its incidence is believed to be higher due to stillbirths and terminations of pregnancy due to accompanying lethal anomalies. Factors such as pulmonary hypoplasia and pulmonary hypertension lead to high mortality and morbidity rates in babies born with this condition [1–3]. Hence, many factors have been proposed to explain the prognosis of the high mortality, including lung-heart ratio (LHR), presence of stomach or liver in the thorax, blood gases in the first 24 h and the modified ventilation index (MVI). Because of the limited number of cases, no single factor alone has been proven to determine the prognosis.

Aim

In this study, we aimed to evaluate whether factors such as prenatal LHR, presence of stomach or liver in the thorax, blood gases in the first 24 h and MVI in infants who were treated in our clinic contributed to the prognosis in prenatally diagnosed CDH.

Material and methods

This study was carried out retrospectively in the Istanbul Faculty of Medicine, Department of Pediatric Surgery with 30 isolated CDH patients seen between January 2007 and December 2013. All the neonates were diagnosed and followed up in the Division of Perinatology of the University prenatally and transferred to the Department of Pediatric Surgery. All of the lesions were left sided. The associated anomalies were bilateral grade II hydronephrosis in 1 patient, cleft palate in 1 patient, patent ductus arteriosus (PDA) in 1 patient, atrial septal defect (ASD) in 2 patients, patent foramen ovale (PFO) in 1 patient, and PDA and ventricular septal defect (VSD) in 1 patient.
All of the patients underwent prenatal two-dimensional ultrasonography and had a prenatal diagnosis of CDH. The LHR was measured prenatally by two-dimensional ultrasonography; the two longest perpendicular diameters of the contralateral lung are multiplied and this is divided by head circumference (in mm) [4]. All diagnoses were made prenatally at 24–26 weeks of gestational age and LHR was measured at the same time. The deliveries were all sectio. The prematurity rate was 5/30 (16.7%).
The time of the surgery was 2.5 ±1.1 days. We did not use a patch in any of the cases.
Twenty-one newborns had normal chromosomal analysis, 8 patients refused analysis, and 1 patient had t(1,5)(p21;p14) on analysis. The mean time of newborns stay at the hospital was 22.2 ±18.8 days. We do not have extracorporeal membrane oxygenation (ECMO). The presence of the stomach or liver in the thoracic region was also noted during ultrasonography. Upon delivery, the newborns were transferred to our neonatal intensive care unit (NICU). Conventional ventilation was used in all cases. Following hemodynamic and respiratory stabilization (pH > 7.45, PaCO2 < 60 mm Hg, and PaO2 > 45 mm Hg), the patients underwent the operation. Each prenatally diagnosed neonate’s sex, gestational weight, gestational age, LHR, presence of stomach or liver in thorax, blood gases in the first 24 h and MVI were recorded. The MVI is a parameter defined in determining the prognosis of neonates with CDH under mechanical ventilation [5]. It is calculated from the factors of PIP, ventilatory frequency, and PCO2 (PIP  PCO2  ventilation frequency/1000).

Statistical analysis

Descriptive statistical methods were used to analyze the results for the groups. Kolmogorov-Smirnov and Shapiro-Wilk tests were used for the normality analysis. ANOVA and t-test methods were used to compare the groups and to ensure that a normal distribution was found. The Kruskal-Wallis test was used for the variance analysis, and the Mann-Whitney U test was used to compare the groups that did not have a normal distribution. Ninety-five percent was accepted as the confidence interval, and a p-value ≤ 0.05 was considered significant for the analysis.

Results

Of the 30 neonates enrolled in the study, 20 (66.6%) were male and 10 (33.3%) were female, mean gestational weight was 2900 ±471 g, and mean gestational age was 37.8 ±1.6 weeks.
Data were grouped and analyzed according to mortality. Overall mortality was 14 (46.6%). Fourteen of the 30 neonates died. Eleven (55%) of them were male and 3 (33%) were female. Seven neonates died before surgery and 23 neonates underwent operations. The mean defect area in our patients was 3 × 4 cm and no patch was used in any of our patients. Of these 23 patients, 7 died post-operatively.
Among 13 patients with LHR < 1.4, 7 (53%) neonates were lost. Among 17 neonates whose LHR was > 1.4, again 7 (29%) neonates were lost. The median LHR in the overall group was 1.6. (For non-survivors it was 1.49 and for survivors 1.51.) No statistical significance was detected among the groups according to LHR.
The stomachs were present in the thoraxes of 19 (63.3%) patients, 9 (47%) of whom died. The livers were present in the thoraxes of 15 (50%) neonates, and 9 (60%) of them died. A statistically significant difference was not found between survivors and non-survivors in the intrathoracic liver or intrathoracic stomach groups either (Tab. I).
For the evaluation of the prognostic values of LHR, pH, PaCO2, PaO2, base excess in the extracellular fluid compartment (BEecf), and cHCO3, the neonates were grouped as total non-survivors and survivors. From perinatal delivery until discharge, 14 neonates were lost and 16 neonates survived. A comparison between the non-survivors and survivors showed a median pH value of 7.10 in non-survivors and 7.24 in survivors (p = 0.002). The median PaCO2 value was 69.4 mm Hg in non-survivors and 51.9 mm Hg in survivors (p = 0.01). There were statistically significant differences in pH and PaCO2 values (Tab. II).
For investigating the prognostic value of MVI, the data were compared between overall non-survivors and survivors. The median MVI value in non-survivors was 100 (32–167), whereas in survivors it was 33 (19–63). Only 1 neonate was lost with MVI < 37; however, all the neonates were lost with MVI > 80. There was a statistically significant difference between overall non-survivors and survivors in the MVI value (p < 0.05).
To determine a threshold value for MVI, an receiver operating characteristic (ROC) curve was generated. The MVI threshold value was 37 with a sensitivity of 92% and specificity of 82% (Fig. 1).

Discussion

The search for prognostic factors affecting the CDH prognosis has been a long process. It may be that a sole factor will not be singled out to determine the survival for this congenital anomaly, but rather, a combination of several factors will have to be taken into account [6].
The LHR calculated by prenatal ultrasound was first described by Metkus et al. in 1996. Fifty-five neonates were reviewed in this retrospective study, and it was reported that all neonates were lost when LHR was < 0.6, and 61% of neonates survived when LHR was 0.6–1.30. Survival reached 100% when LHR was > 1.4 with the aid of ECMO and conventional therapies [7]. Following this report and many others, mortality and morbidity were shown to decline when LHR was < 1.0 [8]. Based on this information, experimental studies with fetal tracheal occlusion are continuing in North America and Europe [9]. However, because there are a limited number of cases and every clinic has unique criteria for selecting patients, standardized protocols for LHR have not been established. In our study, no statistically significant difference in LHR was observed between survivors and non-survivors. However, 4 out of 6 of our patients with LHR below 1.0 were lost. This provides a clue about this parameter, which should be investigated within a larger study population.
The presence of the stomach in the thorax is another prognostic factor that has been proposed in survival. Burge et al. reported high mortality in their study investigating the prognostic effect of stomach position; that is, when the stomach was in the thorax, mortality was greater, but when the stomach was in the abdomen, survival was 100% [10]. Hatch et al. reported survival of 93% when the stomach was below the diaphragma and 29% when it was in the thorax [11]. In a multicenter study from Japan, Usui et al. defined intrathoracic stomach as when more than half of the stomach was herniated into the thorax. Of the 117 neonates enrolled in their study, the stomach was found in the thoraxes of 70% of the neonates in the high-risk group, 25.7% in the medium-risk group, and 4.2% of neonates in the low-risk group. Survival was reported at 20% in the high-risk group, 74.3% in the medium-risk group, and 100% in the low-risk group [12]. In our study, the stomach was present in the thoraxes of 63.3% of the neonates, and survival was 53%, which is lower than in the previous series. Nine neonates were lost in the intrathoracic stomach group and five neonates were lost in the intra-abdominal stomach group. The presence of the stomach in the thorax did not relate to survival in our study. This result is not concordant with the literature, which was attributed to pulmonary hypertension.
The presence of the liver in the thorax is a factor for determining prognosis and planning fetal interventions. In the antenatal CDH registry group, Jani et al. reported 86 intrathoracic livers out of 184 neonates [13]. In the same study, previous reports were also evaluated. The study of Metkus et al. included 38 neonates of which 80% had intrathoracic livers, while Heling et al. reported a rate of 64% for intrathoracic livers in 22 neonates. Victoria et al. conducted a study with 85 neonates, wherein 51 (60%) had intrathoracic livers and 23 were lost with a 45% survival rate [14]. In their meta-analysis, Mullasery et al. evaluated 21 studies that were published in the Medline and Embase databases. These studies focused on the presence of the liver in thorax. There were 407 livers detected in thoraxes. Of the 407 neonates, 222 were lost and 185 survived; overall survival was 45.4% [15]. In our study, 9 of 15 neonates with intrathoracic livers were lost and 5 of 15 neonates with intra-abdominal livers were lost; survival was 40% for these patients. A statistically significant difference between the groups was not revealed; therefore, the presence of the liver in the thorax could not be shown as a prognostic factor, contradicting the current literature. Pulmonary hypertension was the main reason for this discrepancy.
Blood gases in the first 24 h have been proposed as a prognostic factor in many studies since Boix-Ochoa presented their report in 1974 [16]. Early era studies showed an association with pH, PaCO2, and mortality between survivors and non-survivors. Low PaCO2 and high PaO2 levels that were either previously normal or corrected by mechanical ventilation have been found to be related to a good prognosis. On the other hand, despite mechanical ventilation, persistently high levels of PaCO2 were related to a poor prognosis. Blood gases have also been investigated for timing of surgery when pre-operative preparations were deemed necessary. Gentili et al. reported that pH > 7.35 with PaCO2 < 55 mm Hg is a valid indicator for pre-surgical stabilization and surgical intervention [17]. Haricharan et al. stated that hypercapnia in initial blood gases is related to high mortality [18]. Similarly, Hoffman et al. reported a survival rate of 27% when PaCO2 < 60 mm Hg, but all the patients were lost when PaCO2 > 70 mm Hg [19]. Neonates who were lost in our study had median pH at 7.1 and median PaCO2 at 69.4 mm Hg, whereas survivors had median pH at 7.24 and median PaCO2 at 51.9 mm Hg. Low pH and high PaCO2 levels between survivors and non-survivors had statistical significance, consistent with previous reports.
The MVI is a parameter that relates PaCO2 to peak inspiratory pressure in mechanically ventilated patients. Norden et al. reported their results of a study associating pre-operative PaCO2 values and ventilatory indices obtained in the first 24 h in 90 neonates. They showed that the MVI threshold value must be 40 for survival [20]. When the MVI is < 40, they predicted the survival rate at 91% with a sensitivity of 94% and specificity of 85%. Dimitriou et al. studied the blood gases and MVI values by calculating the lung area to determine the prognosis and MVI threshold. They observed an average of 49 in the good prognosis group [21]. Ilce and Celayir revealed similar values in their study with 30 CDH patients. Of the 30 patients, the mean MVI value of the survivors was 38.8, but the non-survivors’ mean MVI value was 114.3 [5]. In our study, median MVI value for survivors was 33 and for the non-survivors was 100. We found a threshold value of 37 for MVI for predicting mortality. The MVI values were < 40 in 92% of our neonates, which were all lost. All neonates with MVI > 80, as suggested by previous reports, were lost.
This study has limitations due to its retrospective nature and limited number of cases. The limited number of patients did not permit two analyses missing for a prediction study, i.e. a multivariate analysis and analysis of AUC/predictive value. Another limitation of the study is that the protocol of these patients is unique for our study; this prediction model pertains to this center and neonatal protocol, so it may not be applicable elsewhere. However, one advantage of our study is that there are no missing values in any of the patients. The mentioned parameters are evaluated in all of the patients.

Conclusions

In light of these findings, pH, PaCO2 levels and MVI have a positive predictive value in prognosis. LHR, liver, and/or stomach presence in the thorax could not be related to prognosis. Future research is necessary to reveal the outcome of the prognostic factors in isolated CDH as well as long-term morbidity.

Disclosure

Authors report no conflict of interest.

References

1. Arensman RM, Bambini DA. Congenital diaphragmatic hernia and eventration. In: Pediatric Surgery. Ashcraft KW et al. (eds). WB Saunders Company, Philadelphia 2000; 300-317.
2. Ba’ath ME, Jesudason EC, Losty PD. How useful is the lung-to-head ratio in predicting outcome in the fetus with congenital diaphragmatic hernia? A systematic review and meta-analysis. Ultrasound Obstet Gynecol 2007; 30: 897-906.
3. Bahlmann F, Merz E, Hallermann C, Stopfkuchen H, Kramer W, Hofmann M. Congenital diaphragmatic hernia: ultrasonic measurement of fetal lungs to predict pulmonary hypoplasia. Ultrasound Obstet Gynecol 1999; 14: 162-168.
4. Ruano R, Lazar DA, Cass DL, Zamora IJ, Lee TC, Cassady CI, Mehollin-Ray A, Welty S, Fernandes CJ, Haeri S, Belfort MA, Olutoye OO. Fetal lung volume and quantification of liver herniation by magnetic resonance imaging in isolated congenital diaphragmatic hernia. Ultrasound Obstet Gynecol 2014; 43: 662-669.
5. Ilçe Z, Celayir S. Modified ventilation index (MVI) in newborns with congenital diaphragmatic hernia. Indian Pediatr 2003; 40: 920.
6. Pediatric Surgery. 6th edn. Grosfeld JL, Fonkalsrud EW, Coran AG (eds.). Mosby-Year Book Inc., St. Louis, 2006; 931-946.
7. Metkus AP, Filly RA, Stringer MD, Harrison MR, Adzick NS. Sonographic predictors of survival in fetal diaphragmatic hernia. J Pediatr Surg 1996; 31: 148-151.
8. Harrison MR, Keller RL, Hawgood SB, Kitterman JA, Sandberg PL, Farmer DL, Lee H, Filly RA, Farrell JA, Albanese CT. A randomized trial of fetal endoscopic tracheal occlusion for severe fetal congenital diaphragmatic hernia. N Engl J Med 2003; 349: 1916-1924.
9. Jani JC, Nicolaides KH. Fetal surgery for severe congenital diaphragmatic hernia? Ultrasound Obstet Gynecol 2012; 39: 7-9.
10. Burge DM, Atwell JD, Freeman NV. Could the stomach site help predict outcome in babies with left sided congenital diaphragmatic hernia diagnosed antenatally? J Pediatr Surg 1989; 24: 567-569.
11. Hatch EL Jr, Kendall J, Blumhagen J. Stomach position as an in utero predictor of neonatal outcome in left-sided diaphragmatic hernia. J Pediatr Surg 1992; 27: 778-779.
12. Usui N, Kitano Y, Okuyama H, Saito M, Masumoto K, Morikawa N, Takayasu H, Nakamura T, Hayashi S, Kawataki M, Ishikawa H, Nose K, Inamura N, Sago H. Prenatal risk stratification for isolated congenital diaphragmatic hernia: results of a Japanese multicenter study. J Pediatr Surg 2011; 46: 1873-1880.
13. Jani J, Keller RL, Benachi A, Nicolaides KH, Favre R, Gratacos E, Laudy J, Eisenberg V, Eggink A, Vaast P, Deprest J; Antenatal-CDH-Registry Group. Prenatal prediction of survival in isolated left-sided diaphragmatic hernia. Ultrasound Obstet Gynecol 2006; 27: 18-22.
14. Victoria T, Bebbington MW, Danzer E, Flake AW, Johnson MP, Dinan D, Scott Adzick N, Hendrick HL. Use of magnetic resonance imaging in prenatal prognosis of the fetus with isolated left congenital diaphragmatic hernia. Prenatal Diag 2012; 32: 715-723.
15. Mullasery D, Ba’ath ME, Jesudason EC, Losty PD. Value of liver herniation in prediction of outcome
in fetal congenital diaphragmatic hernia: a systematic review and meta-analysis. Ultrasound Obstet Gynecol 2010; 35: 609-614.
16. Boix-Ochoa J, Peguero G, Seijo G, Natal A, Canals J. Acid-base balance and blood gases in prognosis and therapy of congenital diaphragmatic hernia. J Pediatr Surg 1974; 9: 49-57.
17. Gentili A, Giuntoli L, Bacchi Reggiani ML, Masciopinto F, Lima M, Baroncini S. Neonatal congenital diaphragmatic hernia: respiratory and blood-gas derived indices in choosing surgical timing. Minerva Anestesiol 2012; 78: 1117-1125.
18. Haricharan R, Barnhart DC, Cheng H, Delzell E. Identifying neonates at a very high risk for mortality among children with congenital diaphragmatic hernia managed with extracorporeal membrane oxygenation. J Pediatr Surg 2009; 44: 87-93.
19. Hoffman SB, Massaro AN, Gingalewski C, Short BL. Predictors of survival in congenital diaphragmatic hernia patients requiring extracorporeal membrane oxygenation: CNMC 15-year experience. J Perinatol 2010; 30: 546-552.
20. Norden MA, Butt W, McDougall P. Predictors of survival for infants with congenital diaphragmatic hernia. J Pediatr Surg 1994; 29: 1442-1446.
21. Dimitriou G, Greenough A, Davenport, Nicolaides K. Prediction of outcome by computer-assisted analysis of lung area on the chest radiograph of infants with congenital diaphragmatic hernia. J Ped Surg 2000; 35: 489-493.