|
|
||||||||
a Department of Pediatric and Congenital Cardiac Surgery, University of Padua, Padua, Italy
b Department of Pediatrics, University of Padua, Padua, Italy
c Department of Statistics, University of Padua, Padua, Italy
Accepted for publication December 17, 2008.
* Address correspondence to Dr Vida, Pediatric and Congenital Cardiac Surgery Unit, Via Giustiniani 2, Padua, 35100, Italy (Email: vladimirovida{at}yahoo.it).
Presented at the Forty-fifth Annual Meeting of The Society of Thoracic Surgeons, San Francisco, CA, Jan 26–28, 2009.
| This article has been selected for the open discussion forum on the CTSNet Web Site: http://www.ctsnet.org/sections/newsandviews/discussions/index.html
|
| Abstract |
|---|
|
|
|---|
Methods: Included were 201 premature babies (
32 weeks gestational age), from January 2001 to June 2007, with PDA who received primary medical treatment with ibuprofen. Number of ibuprofen cycles, gestational age, body weight, and presence of symptomatic hypotension requiring vasoactive/inotropic drugs were related to hospital outcome, including hospital mortality, presence of necrotizing enterocolitis, acute renal failure, intraventricular hemorrhage, retinopathy and bronchopulmonary dysplasia at week 36. Data were analyzed with a logistic regression model.
Results: Medical treatment was effective in 149 patients (75%), but 52 (25%) required surgical ligation after medical treatment failed. They had younger gestational age (25 weeks [IQR, 24 to 27 weeks] vs 27 weeks [IQR, 25 to 28 weeks], p < 0.0001), lower body weight at birth (730 g [IQR, 595 to 915 g] vs 840 g [IQR, 670 to 1016], p = 0.05), and a higher incidence of symptomatic hypotension (38 of 52 [73%] vs 56 of 149 [38%], p < 0.0001) than patients who responded to ibuprofen. More than two cycles of ibuprofen was significantly associated with an increased risk for bronchopulmonary dysplasia (odds ratio [OR], 2.81; p = 0.03) and acute renal failure (OR, 3.81; p = 0.09).
Conclusions: The prolonged patency of the ductus arteriosus in preterm infants is related to an increased morbidity. Surgical ligation of PDA is a safe and effective treatment and should be done soon after two complete cycles of ibuprofen, especially in selected patients, to improve clinical outcome.
Patent ductus arteriosus (PDA) with significant left to right shunting is an important problem in premature infants [1]. When pulmonary resistance reduces, systemic effects result from the diastolic steal and retrograde diastolic blood flow, often producing hemodynamic and respiratory derangement, necessitating the use of inotropic drugs and an escalating ventilatory support [1–3].
The ideal initial treatment of symptomatic preterm infants with PDA is not well established [4–6]. Currently, one or more cycles of medical therapy, when indicated, are usually preferred before planning a surgical ligation [2, 7]. Intravenous ibuprofen is now widely used in premature infants because it has been proved to be equally as effective as intravenous indomethacin and results in fewer side effects [8–12].
Surgical PDA ligation, which is usually considered by the neonatologists when medical treatment has failed or is contraindicated, has low surgical morbidity and mortality [1, 13–16]. Goals of surgical ligation include a prompt improvement in cardiorespiratory failure, a lower risk for a prolonged mechanical ventilation and subsequent bronchopulmonary dysplasia (BPD), and better survival to discharge [3, 16, 17].
We have analyzed our recent experience with the concept in mind that prolonged medical treatment, especially in selected premature babies in whom medical therapy frequently fails, often delays the closure of a PDA and, therefore, exposes these patients to an increased hospital mortality and morbidity. We sought to define the variables associated with increased hospital mortality and morbidity in preterm infants with PDA and also to identify the optimal timing for PDA closure to improve the hospital outcome.
| Patients and Methods |
|---|
|
|
|---|
The study included all premature infants between January 2001 and June 2007 with an isolated PDA after 48 hours of life who were 32 weeks of gestational age (GA) or younger and who received primary medical treatment with ibuprofen. Exclusion criteria were association with either congenital intracardiac anomalies or chromosomal abnormalities. Also excluded were patients who had contraindications to medical treatment and underwent primary surgical closure.
Ductal patency was established by a routine 2-dimensional echocardiography screening performed at 48 hours of life by a consultant pediatric cardiologist. A PDA was defined by the echocardiographic presence of a continuous left–right shunting into the main pulmonary artery from the PDA [13]. All patients with a PDA graded as moderate or more according to Van Overmeire's classification were included in our study [18]. Patients with small or minimal shunting were excluded.
The intravenous (IV) ibuprofen therapeutic cycle included a 10 mg/kg dose, followed by two 5-mg/kg doses at 24-hour intervals. The PDA was classified as closed when clinical and echocardiographic assessments 24 hours after the end of the ibuprofen cycle were negative for patency. In case of persistent patency after the first cycle, all patients with a PDA graded as moderate or more were given a second ibuprofen cycle. After the second ibuprofen cycle, all PDAs were treated according to the consultant neonatologist's preference (Fig 1).
|
Patient demographics, gestational age in weeks, body weight (BW) in grams, age in days at PDA closure, and systemic hypotension were recorded. Systemic hypotension was defined as mean systemic arterial pressure of less than 30 mm Hg according to gestational age together with signs of hypoperfusion [20]. Gestational age was determined by the date of the mother's last menstrual period and early ultrasound (before 24 weeks gestation).
Outcomes included hospital mortality, presence of necrotizing enterocolitis (NEC) [21–22], acute renal failure (ARF) [23–24], intraventricular hemorrhage (IVH) [25], retinopathy of prematurity (ROP) [26], and BPD [3–6]. NEC was defined as Bell classification II or greater, this included NEC that was treated medically or surgically and so-called spontaneous perforations. ARF was defined when a sustained decrease in urine output (< 1 mL/kg/h after the first 48 hours) was present or the serum creatinine concentration was 1.5 mg/dL or higher, or both. IVH was defined as a bleeding of the ventricles. ROP was defined as stage 2 or greater. BPD was defined as the supplemental oxygen requirement at 36 weeks gestational age to maintain adequate oxygen saturation (> 90%).
Quantitative variables were summarized as medians and interquartile ranges (IQR). Because quantitative variables are not normally distributed, comparison among groups was by Kruskal-Wallis test [27]; when two groups were to be compared, the Wilcoxon signed rank test was used. Difference between proportions was assessed by means of the Fisher exact test or its extension proposed by Freeman and Halton [28] when there are more than two groups. The following predictors were related to hospital outcomes with a stepwise logistic regression model: birthweight (
750 g or not), gestational age (
26 weeks or not), systemic hypotension (yes or no), Apgar score (
6 or > 6), ibuprofen (> 6 or
6 cycles). Predictors were considered significant at value of p
0.09. Data were analyzed using SAS 9.1.3 software (SAS Institute, Cary, NC). Values of p < 0.05 were considered significant.
| Results |
|---|
|
|
|---|
|
|
|
|
|
The patency of the ductus arteriosus was significantly longer in patients who required surgical ligation, at 16 days (IQR, 12 to 22 days) vs 7 days (IQR, 6 to 9 days; p < 0.0001). The prolonged patency of the ductus arteriosus in patients who received more than two cycles of ibuprofen was significantly associated with and increased the risk for BPD (OR, 3.13; p = 0.01), IVH (OR, 4.39; p = 0.01), and ARF (OR, 4.07; p = 0.05), and had a higher incidence of surgical ligation (OR, 3.42; p = 0.005; Table 4). Patients who received fewer than two cycles of ibuprofen had a shorter median time on mechanical ventilation of 9 days (IQR, 4 to 25 days) vs 27 days (IQR, 17 to 35 days) and a shorter median stay in the NICU of 36 days (IQR, 19 to 60 days) vs 57 days (IQR, 48 to 75 days).
|
|
| Comment |
|---|
|
|
|---|
Our institution is tertiary care hospital and a referring center for all premature infants in the northeast region of Italy. Premature babies are admitted in our NICU with a body weight as low as 400 g and a gestational age as young as 23 weeks. Since 2001, we have routinely adopted the use of IV ibuprofen to treat premature babies with a hemodynamically significant PDA diagnosed within 48 hours of life. In our preterm babies, ibuprofen therapy led to a 74% PDA closure rate; and in 137 of 149 (92%), the PDA closed within the second cycle of ibuprofen (6 doses). In only 12 of 201 patients (6%) was prolonged medical treatment (more than two cycles of ibuprofen) required before the PDA closed. The remaining 52 patients (26%) who did not respond to medical treatment had a safe surgical PDA ligation.
Data from published reports suggest that PDA ligation should be preferred in very low birth weight patients, especially when less than 1000 g of body weight [2, 4, 30]. These patients seem to have a lower closure rate after medical therapy compared with heavier premature infants. Our results show that we have identified a selected population of patients in whom ibuprofen therapy frequently fails. All babies had a younger gestational age, a lower body weight at birth, and a higher frequency of symptomatic hypotension requiring vasoactive/inotropic drugs than patients who responded to ibuprofen.
Chorne and colleagues [28] and Kabra and colleagues [17], have recently reported that surgical PDA ligation in premature babies can be associated with an increased risk for hospital morbidity and a worse long-term outcome. However, to the best of our knowledge, only few previous studies have reported a possible correlation between the prolonged PDA and the hospital outcome [16]. This is in consideration of the fact that babies who had undergone PDA ligation had a prolonged PDA that was delayed by unsuccessful medical therapy.
Our study demonstrated that surgical ligation is a safe and effective treatment; it can be done at the bedside in the NICU in critically ill neonates [31–34], with no surgically related death and a low rate of complications. We have also demonstrated that a prolonged PDA, defined as a ductus arteriosus still patent after two cycles of medical treatment, is associated with a significantly higher short-term morbidity due to an increased risk for BPD, IVH, and ARF. A prolonged PDA was also associated with a longer mechanical ventilation time and a longer stay in the NICU.
Although the mechanism for ARF seems to be related to the prolonged hypotension due to bloods steal from the left–right shunt at the PDA level, the exact cause of the increased rate of BPD in patients with prolonged PDA is still unknown. We can hypothesize that the prolonged and significant left–right shunting at the PDA level may also interfere with the adequate development of the premature lungs of these babies, or over-circulation to the lung could lead to microcirculatory lung damage. All these theories remain unproven, however.
In conclusion, a PDA must be ligated anytime medical treatment has failed or is contraindicated. Nonetheless, the ideal timing for surgical PDA ligation after initial medical treatment is yet to be defined. According to our results, we strongly believe that prolonged medical therapy with the aim of avoiding a surgical procedure can be detrimental and can lead to increased hospital morbidity that will consequently prolong the hospital stay. Early surgical PDA ligation should always be recommended soon after the second unsuccessful cycle of ibuprofen, with the aim of improving patient's clinical outcome. In accordance with our results, we have recently modified our institutional policy, which will recommend early surgical PDA ligation in all babies with a body weight of 750 g or less.
Teamwork and cooperation between neonatologists and cardiac surgeons is essential. Optimal planning of the ideal timing for surgical ligation is essential for optimizing the clinical course of these patients. Nevertheless, further prospective evaluation with a larger population is necessary for firmly recommending a modification of the current management for this group of patients.
| Discussion |
|---|
|
|
|---|
DR VIDA: Yes, it did. We didn't have any events of renal failure or NEC after surgical ligation.
DR CALDARONE: And did you evaluate the interval between the completion of the medical therapy and the surgery? The reason I am asking that is it is a little bit unclear how safe it is to wait and how long one can wait in order to get one of these patients on the operating room schedule. They frequently get stuffed in gaps in the operating room schedule rather than given OR [operating room] time with the same priority as other cases.
DR VIDA: Can I show back a slide, please? First, ducts were not ligated in the OR but in the NICU [neonatal intensive care unit] at the bed space of the patient. According to our experience, we think that this is an easy and safe treatment for these patients, avoiding in this way cumbersome transportation in the OR, squashing them in the OR schedule. As surgical complications, we had 3 pneumothorax that resolved spontaneously and a small bleeding during PDA [patent ductus arterious] during clipping that did not require any blood transfusion.
To understand which is the best time to ligate the duct, we have analyzed the time intervals between the first, second, and third cycle of medical treatment. It is clear that the bigger are the kids, the higher is the success of medical treatment. The smaller and the younger are the kids, who frequently also presented with systemic hypotension, the less they respond to medical therapy.
Looking at the outcomes, as shown in column 1, there is no statistical difference between kids who underwent one or two cycles of ibuprofen. The only difference is that if you go ahead with a second cycle of ibuprofen, you can have a 22% of patients who can have medical closure.
The ideal day for PDA ligation seems to be the 11th or 12th day after birth, which usually corresponds to the end of the second cycle of ibuprofen. We believe that planning PDA closure soon after the second unsuccessful cycles of ibuprofen will avoid any deleterious delay preventing in this way any further co-morbidity.
DR CALDARONE: It's a very thoughtful analysis.
DR JOHN A. HAWKINS (Salt Lake City, UT) : I have one question. You found an association between two cycles of ibuprofen and the increased incidence of renal failure, intraventricular hemorrhage, and BPD [bronchopulmonary dysplasia]. I can understand acute renal failure and I can understand the intraventricular hemorrhage, but is there any explanation for why there is an association between more ibuprofen and the development of BPD?
DR VIDA: Actually, there is no explanation. We have no histological findings, and currently there are no supporting data in the literature. Perhaps we can speculate that a longer period of pulmonary over-circulation and hypotension can induce some intravascular changes on these premature lungs. But this is just a hypothesis.
DR HAWKINS: A second question. You also found in your study that low gestational age, low birth rate, and hypotension predicted that medical therapy wouldn't work. Has this changed your practice to where there are certain weight or gestational age cutoffs that you would recommend surgical therapy earlier than two rounds of ibuprofen?
DR VIDA: As it is shown in the logistic regression analysis, the cut-point which is the best predictor for the majority of outcomes is represented by the body weight less than 750 grams. On the other had the gestational age of the patients falls out of the analysis. This made a change in our institutional policy in treating these patients, promoting early surgical ligation immediately after the second unsuccessful cycle of ibuprofen, especially in patients less than 750 grams of body weight.
DR HAWKINS: So you don't necessarily not give medical therapy, it's just that you wait less time after the second course if their weight is under 750 grams?
DR VIDA: Yes. That is what we do.
DR JEFF L. MYERS (Boston, MA): The biggest argument we get from our neonatologists and cardiologists is the risk of vocal cord paralysis. Do you have an incidence of zero, or did you just not look that as you did your chart review for these patients? I know it's hard in the 700-gram kids, by the time they get extubated and you can figure this out, to really track that, but do you have any sense of that complication?
DR VIDA: Fortunately, we did not have this complication.
| References |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |