The study demonstrated a survival rate of 67.5% among CDH infants managed without ECMO at our institution. We have included all fetuses with in-utero and postnatal CDH diagnosis born at our institution over the 7-years study period with no exclusions or selective terminations during the study period. Previous single centre studies from the Kingdom of Saudi Arabia showed variable survival rates in CDH infants ranging from 71.4% (5) to 57% (6). Our survival rate is comparable to that reported from the CDH Study Group (67%) (4). Recently, multicenter European study involving 975 infants born between 2004–2013, reported an overall CDH survival of 71.9%, a 68.7% survival for the group of CDH infants who underwent ECMO and a lower survival at 52.4% for a subgroup of CDH infants who had (FETO) (20).
In our series, 32 infants (80%) were antenatally diagnosed, and the rates of antenatal diagnosis did not differ between survivors and non-survivors. Antenatal diagnosis offers the advantages of a planned delivery at a tertiary centre with availability of expertise dealing with CDH cases. Nevertheless, antenatal diagnosis might imply a longer duraration of abdominal visceral herniation into the thoracic cage, resulting in an increased degree of lung hypoplasia (21). Therefore, some studies have reported a significantly higher mortality in antenatally diagnosed patients, compared with those diagnosed postnatally (21, 22).
We have shown that the Apgar scores at 1 and 5 minutes after birth were significantly lower among CDH non-survivors. This is in agreement with other previous studies that have demonstrated a lower 5 min Apgar score among CDH non-survivors (12, 23). Indeed, the Apgar score is indicator of early cardiopulmonary adaptation and response to resuscitation. Nevertheless, others have found no differences in the Apgar scores between CDH survivors and non-survivors (24). Both the need for iNO therapy and HFOV were significantly higher among CDH non-survivors in our cohort. This is similar to the findings from the largest observational study on the resuscitation of CDH infants that showed a higher use of HFOV and iNO therapy among CDH non-survivors (24). Furthermore, we have found that the echocardiographic findings of PPHN were significantly higher among CDH non-survivors. Indeed, pulmonary hypertension is a major determinant of postnatal survival in infants with a CDH. The Pulmonary hypertension treatment in our patients varied considerably and included a combination of different vasopressors, milrinone, sildenafil and iNO therapy. Indeed, optimal treatment for pulmonary hypertension in infants with CDH is still an area with a lot of unanswered questions. We await the results the currently ongoing Randomized Control Trials (RCTs) looking at the best initial therapy for pulmonary hypertension in infants with CDH (25) (26).
This study showed that the best (lowest), mean and highest oxygenation indices in day 1 were all excellent predictors of survival with the mean day 1 OI marginally outperforming the other two indices with higher sensitivity and specificity. The mean day 1 OI is likely to be more reflective of the overall first day cardiopulmonary stability compared to the lowest and highest day OI as these indices may vary in response to alternations in the clinical condition and the ventilatory support changes. Indeed, the excellent survival predictive ability of the best day 1 OI has been described in the literature (15–17, 27). In addition, we have also demonstrated that WHSRpf (highest PaO2 – highest PaCO2) (19) is predictive of CDH survival in our study. Nevertheless, the predictive ability of the formula is lower than that of day 1 oxygenation indices as evidenced by the AUC analysis. We have also shown that the first arterial PCO2, the highest arterial PCO2 and PaO2 in day 1 as well as the first arterial blood gas pH were all predictive of survival, albeit with lower predictive ability in comparison to the OIs and the WHSRpf. Others have also shown that the maximum PCO2 in the first 24 hours after birth is a useful predictor of survival in CDH infants (14).
We have shown that the pre-operative OI had a significant correlation with the time to surgical repair of the diaphragmatic hernia as well as the length of hospital stay. Similarly, Tan YW et al (27), found that pre-operative OI did show significant, correlations with time to surgery, length of stay and the duration of mechanical ventilation (27). Furthermore, in addition to the predictive ability of the first day oxygenation indices for CDH survival, we have also found that they have a significant correlation with the time to surgery, duration of ventilation and the length of hospital stay. To our knowledge, the correlation of the first day oxygenation indices with DOV, LOS and TTS have not been described before. This finding therefore adds to the usefulness of first day OIs as a predictor of both CDH survival and adverse outcomes among survivors.
Our study has some Strengths and limitations of our study. We report data on consecutive infants with CDH with no exclusions. Additionally, we were able to obtain a full set of oxygenation indices from our electronic patients records that combine the ventilatory settings and the corresponding blood gases results used to calculate the oxygenation indices. With regard to the limitations of study, we did not have the data for the LHR in the majority of the infants in our study. This attributable to the lack of LHR data for infants with postnatal diagnosis and variation in the method of recording the LHR (absolute value vs. Observed/expected percentage) among antenatally diagnosed CDH cases.