The pesky PDA. A puzzle after all these years
Like many Neonatologists, I experience a complete body sigh when I discover that an ELBW infant has a PDA. Once I find out that once again another duct has reared it’s ugly head so to speak a number of thoughts run through my head. Should I treat it? With what? For how long? What if the first treatment doesn’t work? Should I have given prophylactic indomethacin? If I do that how any ELBWs will experience significant renal impairment or worse NEC based on that decision? Then we move into a completely different territory that occurs after the duct has been unsuccessfully treated. To ligate or not to ligate? With so many questions and so many conflicting papers in the literature some of which say the ductus is associated but not causative for this or that outcome it is no wonder I am still largely in the dark as to what is truly the best approach.
What about prophylactic indomethacin?
With respect to the use of prophylactic indomethacin the TIPP study clearly showed that while units could reduce the incidence of PDA and with in severe intraventricular hemorrhage the impact on neurodevelopment was unchanged. This was definitely unfortunate news as the trade off then to exposing all of the infants < 1000g in a unit to an increased risk of renal impairment and NEC would not be seemingly worth it. As PDAs are found more commonly in those ELBWs who are not exposed to antenatal steroids, a 2011 paper questioned whether provision of indomethacin prophylaxis only to those babies without steroids was published. Unfortunately the results were not as hoped as no benefit to this subgroup was noted.
Ibuprofen has been associated with less renal impairment and lower risk of NEC but unfortunately we do not have any long term outcome results from ELBWs treated with such therapy. It is promising though that the Cochrane review on the subject found a number of positive short term outcomes from the use of such treatment. In 7 studies included the rates of development of a PDA, repeat courses of NSAIDs and surgical ligations were all reduced with this therapy.
More recently several papers studying paracetamol (tylenol) as a medicine to promote ductal closure have been published but results have been mixed and the lack of large RCTs make it difficult to advocate for it’s regular use instead of ibuprofen or indomethacin. A cochrane review comparing oral ibuprofen to oral paracetamol has been published in 2015 and does show based on the two studies included that the drugs are likely equally efficacious in closing a ductus with an added benefit of paracetamol being less oxygen usage and hyperbilirubinemia compared to ibuprofen. Recent data in mice however linking autism to fetal exposure to paracetamol has necessitated longer term outcome data before this treatment can be recommended instead of the current ibuprofen or indomethacin.
What about the option of benign neglect?
Another option exists however which is to manage any symptoms resulting from the PDA and allow it to close on it’s own. This was the subject of a recent paper published in Archives of Diseases and Childhood Fetal and Neonatal Ed by Rolland et al. The authors of this study describe their experience retrospectively during a time in which there was intentional avoidance of treatment of any kind including ligation for the ductus. What this allows for is a comprehensive assessment of the natural history of the ductus in their cohort of 103 infants between 24-27 weeks gestational age.
Although the study began with 103 infants there were 12 infants (12% of the cohort) that died prior to 72 hours which was the time when the identification of the PDA would have been done. We do not know if these infants died from the PDA or not but it is fair to agree with the authors that unless prophylactic indomethacin was being used this outcome would not have been avoidable. Looking at the remainder of the group, 8 were found to have no PDA at 72 hours while an additional 13 had either no ECHO (10) or died (3) so were not followed. In the ten cases that did not have an ECHO the reason was the lack of respiratory support so presumably the ECHO result would have been irrrelevant to care. The remaining 70 are described in the following table where HNSPDA is a hemodynamically non significant PDA, HSPDA refers to a significant PDA and IDHS means insufficient data on the hemodynamic status although the PDA is still there.
|HNSPDA (n=30)||HSPDA (N=23)||IDHS (N=17)|
|Spontaneous closure||22 (73)||18 (78)||11 (65)|
|Death before discharge||4 (13)||2 (9)||5 (29)|
|Survival at discharge with a PDA||4 (13)||2 (9)||1 (6)|
|Ligation of PDA||0||1 (4)||0|
The paper is intriguing for a number of reasons. The first is that by an average age of two months 73% of the PDAs closed from the 70 patients that had documented persistence after 72 hours. We know that from the group of 91, 8 had closure at 72 hours which leaves 83 with a PDA. An additional 10 were not studied as mentioned above as they had no symptoms so we can assume they did not have a HSPDA. Of the three that died however we do not know if the PDA contributed. Looking at the 70 patients left 11 more patients died and 7 were discharged with a PDA. The authors do not disclose the fate of those additional 7. Did they have a ligation after discharge or not and by what method? If they were ligated this is still a positive outcome inn my my mind as the larger infant at that point would likely tolerate a surgical or catheter based closure much better with less morbidity.
It can also not be ignored that in this study the incidence of pulmonary hemorrhage was high at 25% and severe IVH in 21%. Both of these outcomes may be affected by the presence of a PDA so one important question raised here is an ethical one. Clearly many of the children studied who were not treated avoided complications related to treatment which is a good thing. The concern however is that by not treating there may have been excessive morbidity in those who developed the above complications and lacking from the study is follow-up of the survivors to see if their outcome. We also clearly don’t know what the PDA may have contributed to death in this study which is also clearly an important outcome to consider.
I wish I could give you the answer of what to do with these kids. Watch and wait or treat? I suspect in the end we will likely settle on a hybrid approach guided by information gleaned from Targeted Neonatal Echocardiography as was discussed in a post on the topic which can be found here. I think the future state will likely see us using a strategy of selecting some infants by risk categorization to receive prophylactic indomethacin, some to have TNE done between 48 – 72 hours with or without the use of biomarkers to identify those who are likely to get a HSPDA and then treat those and then a final group that we may watch.
What the above study adds to the literature though is that in this final category who we may watch and wait there is about a 75% chance that they will close on their own. If we can pick out the ones that are not HSPDA I suspect the spontaneous closure rate would be even higher. While I am grateful for the publication of this article for now I will continue to pick and choose as best I can which ducts we need to deal with and those we don’t. Hopefully with time and more knowledge my body sighs will be replaced by a look of confidence as I explain to families what is needed for their child.