| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
© 2004 European Association of Cardio-Thoracic Surgery
Management of the persistent ductus arteriosus in infants of very low birth weight: early and long-term results
a Clinic of Thoracic and Cardiovascular Surgery, Heinrich-Heine University Medical Center, Duesseldorf, Germany
* Corresponding author. Tel.: +49-211-811-9939; fax: +49-211-811-6996 Received December 9, 2003; received in revised form March 11, 2004; accepted March 31, 2004
The hemodynamically relevant persistent ductus arteriosus (PDA) impairs pulmonary and cardiac function. Frequently, PDA can be closed only via surgery. In this retrospective study, early and long-term results in very low birth weight newborns are evaluated. Eighty-seven of 634 very low weight newborns presented with PDA All patients (pts; age: ±14 days; weight: ±1064 g) were ventilator-dependent. Surgical closure (after 29±5 days) was indicated if echocardiography and prolonged ventilation (>20±2 days) evidenced a hemodynamically relevant PDA. Sixteen pts, in which indomethacin therapy failed preoperatively are included in the 36 surgically treated pts; no pt died intra- or early postoperatively (<3 day). Overall mortality 30 days after delivery was  Early plus late mortality was  Long-term follow-up (3–12 years) in 46 (68%) pts: 15 were solely physically, 11 were mentally and neurologically, and 4 were physically, mentally and neurologically retarded. From these 30 pts, 15 were severely (e.g. tetraspasm; severe cerebral paresis) and 15 were slightly (e.g. psychosomatic and language development prolongation) retarded. Sixteen pts exhibited no disability; no long-term complications owing to surgery. The relatively large number of neurological injuries was not owing to chromosomal syndromes or pre-existing abnormalities but can be explained by severe and frequent prematurity, hypoxia, and intracerebral bleeding. Indomethacin was successful only in a few patients. Early surgery (after frustran early indomethacin therapy) of a hemodynamically relevant PDA is recommended. In the long-term, severe disabilities develop.
Key Words: Persistent ductus arteriosus; Very low weight birth infants; Surgical closure
In healthy newborns closure of the duct happens almost in 100% of the cases during the first 96 h after birth. This is not the case in many very low birth weight newborns and inversely related to birth weight. The failure of duct closure is caused by a number of pathophysiologic maladaptations: hypoxia, increased production or decreased pulmonary meta-bolism of PGI2 and PGE2, and an increased sensitivity to the locally produced vasodilating prostaglandins and an ineffective contractile response to the increasing oxygen tension [1]. The ‘diastolic steal’ as a consequence of large left-to-right ductal shunting from the descending aorta into the pulmonary circulation may cause intestinal ischemia and lead to necrotizing enterocolitis, renal hypoperfusion, or cerebral ischemia. The volume overload of the pulmonary circulation may aggravate the common hyaline membrane disease (HMS) of these children [2]. The hemodynamically relevant persistent ductus arteriosus (PDA) causes an impairment of pulmonary and cardiac function by pulmonary hypertension in addition to a disturbance of lung maturation. In consequence, the duct should be closed as soon as possible, even in premature infants. This goal can be achieved using several therapeutic regimens, but in many cases closure is only warranted via surgery. It was the aim of the present retrospective and follow-up study to evaluate the early and long-term results in a group of very low birth weight newborns (545–1490 g). The following questions were to be answered: (1) Which therapeutic regimen was the most successful? (2) How was the occurrence of general and therapy-combined complications in the three treatment groups? (3) How was the development of the children in the long-term, and (4) Are there predictors in the newborn period for the long-term development of the children.
The records of 87 out of 634 consecutive very low weight newborns with PDA (January 1985–December 1994) were reviewed. The epidemiological data are shown in Table 1. Besides application modalities and success of various therapy regimens the following parameters were included: hemodynamic relevance of the PDA, intracranial hemorrhage (ICH—classification according to Levene), the rate and severity of the hyaline membrane syndrome (HMS—criteria from Giedion et al. [3]), necrotizing enterocolitis, oliguria or anuria, and sepsis.
Two types of ventilators were employed (Babylog 1 HF or Babylog 8000; Draeger, Luebeck, DE). The ventilation was volume-constant without pressure limitation. If necessary surfactant was applied (Survanta, Abbott: 1985–1990; since 1990 Alveofact, Thomae/Boehringer; Ingelheim, DE).
In accordance with the three different treatments, the total cohort was divided into three groups: (1) fluid restriction to 100 ml/kg/die ( 2.1. Statistical analysis The statistical analysis was performed using SPSS software (SPSS Inc, Chicago, Ill). The significance of differences between the groups for all epidemiological data and the occurrence of general and specific complications were assessed using Fisher's exact test, t-test-analyses and the Mann–Withney-U-test, when appropriate. For testing of hypotheses, an adjusted procedure has been used since several hypotheses were tested simultaneously. Thus, we adjusted the P-values according to the Sture–Holm procedure within the variable groups, such that a global of 0.05 was guaranteed. Several candidates were tested for their power to predict the children's developmental outcome using Fisher's exact test alone or in combination with a discriminant analysis. Still, the results have a descriptive character.
All 87 patients (pts) were ventilator-dependent (mean age: 14 day; mean weight 1064 g, range 545–1490 g). Duration of ventilation was 27±20 days (mean ± standard deviation) in group I, 42±21 days in group II, and 63±36 days in group III; duration of ventilation from the onset of the indomethacin therapy (group II) until extubation was essentially the same as duration of ventilation between the day of operation and extubation (group III); Table 2. Indomethacin therapy was started on day 15±2 and operation was performed on day 29±5 after delivery.
In 41 pts intracranial bleeding occurred (class 1: 14; class 2: 13, class 3: 14 pts, according to Levene). No intracranial bleeding occurred after indomethacin or after operation. In 16 pts indomethacin therapy failed preoperatively; a total of 36 pts was treated surgically (20 pts primary surgery plus 16 pts after failed indomethacin therapy); no patient died intra- or early postperatively (<3 days). The average postoperative body core temperature was 34.3±1.4 °C on arrival to the NICU.
Mortality until day 30 after delivery was
Length of hospital stay was 88±58 days in the total cohort (74±36 in the fluid group, 92±70 in the indomethacin group, and 111±72 days in the surgical group). Long-term follow-up (3–12 years) data was available in 46 (68%) out of 68 long-term survivors: 15 were solely physically, 11 were mentally and neurologically, and 4 were physically, mentally and neurologically retarded. From these 30pts, 15 were severely retarded (e.g. tetraspasm; severe cerebral paresis), and 15 were slightly retarded (e.g. psychosomatic and language development prolongation). 16pts did not exhibit any disability; no complications were owing to surgery; 1pt suffered from voice disturbances following tracheal stenosis. To test for differences between the surviving and the deceased patients, weight at birth, duration of pregnancy, maturity score according to Farr, the Apgar-scores, duration of ventilation and hospitalisation of the three different groups were compared. On the basis of the above measures, no significant differences were detected. In the long-term, no indomethacin- or operation-related complications were observed.
In this restrospective study on 87 of 634 consecutive, very low birth weight newborns, data are analyzed that were collected over 10 years. Three different therapies were employed: fluid restriction, pharmacological therapy, and surgical therapy. 4.1. Diagnosis The accuracy to correctly diagnose a PDA is low [4]. Therefore, the ratio of the diameter of the left atrium to the aortic root together with ventilation parameters was used to evaluate the significance of the PDA. However, this method can be falsely positive for different reasons, e.g. in case of a ventricular septal defect, any cause of left ventricular dysfunction, and mitral valve abnormalities. Hence, appropriate definition of a hemodynamic significant PDA presents a fundamental problem.4.2. Therapy In most cases with a hemodynamically non-relevant PDA, fluid restriction alone or in combination with diuretics was a successful therapy. Pharmacological PDA closure (indomethacin) was successful in 28%. This result is different from those in many recent studies, in which success rates vary between 50% [5] and 92% [6]. Both the application form and the time of treatment differ between the present study and the other studies. We applied indomethacin orally, whereas it was given intravenously in both other studies. Moreover, we applied indomethacin only from day 16 postpartum, whereas it was applied already from day 3 [6] and day 4 [7] in the other studies. The different success rates strongly suggest an early begin of indomethacin therapy after diagnosis of a PDA, independent from whether it is hemodynamically relevant or not. The late start of indomethacin therapy during the reported period was chosen to avoid side effects.The known side effects of indomethacin in small neonates vary within a wide range: one group reported no episode of decreasing urinary output and necrotizing enterocolitis after very early application (6–12 postnatal hours) [8]. Another group describes that early i.v. indomethacin treatment (day 3) in contrast to late treatment (day 7) improved PDA closure but was associated with increased renal side effects and even more severe complications [9]. Recently, ibuprofen has been proven to be as effective as indomethacin in promoting ductal closure in premature infants, while inducing significantly lower rates of oliguria [10]. Thus, ibuprofen might prove a promising alternative in future to treat PDA. Surgical closure of the duct was, as expected, the most successful form of the applied therapy regimens. Closure of the duct was achieved in 100% of our cases without a recurrent opening, which corresponds with the current literature. However, a residual flow might persist in up to 23% of the cases. Thus, earlier surgery may lead to shortening of the entire duration of ventilation and in consequence, to a decrease in ventilation-induced injuries. Therefore, we propose a concise diagnostic and therapeutic regimen soon after birth as presented in the summary and conclusion. 4.3. Mortality and morbidity 4.3.1. Perioperative mortalityNo patient was lost intraoperatively, but about 25% of these patients died early or late postoperatively. These results nicely agree with the literature [11]. Because of the good intraoperative results, surgical therapy was suggested as the best option for babies with a body weight below 800 g [5].
4.3.2. Perioperative morbidity The length of hospital stay was possibly too long in the present study. Likely, this was owing in part to the delayed completion of the routine diagnosis and the duration of an as safe as possible distinction between a hemodynamically relevant and irrelevant PDA. The great importance of this distinction is underlined by the finding that 80% of preterm infants with HMS will present a physically patent duct during the first four postnatal days. However, only about one third of these infants will develop shunts large enough to cause symptoms [4], because the clinical impact of ductal shunting on the individual infant does vary to a large extent [14]. Thus, functional criteria decide the hemodynamic significance of a PDA. In accordance with our findings it seems unlikely to close the duct pharmacologically, if this therapy is initiated only 14 days postpartum. Therefore, it is suggested to abbreviate the delay until the decision, whether a significant PDA demands a specific therapy, and we mention again that the question ‘What is a hemodynamically relevant PDA’ cannot easily be answered. Already before 20 years and more, any gold standard for the detection of a PDA was questioned since ductal tissue is dynamic, and since the direction and magnitude of shunting can change dramatically during the newborn period [15]. This statement seems to be valid till this day. 4.4. Summary and conclusion Indomethacin was successful only in a few cases in the present study. In the long-term, frequently severe disabilities develop. Prognostic markers for survival after any intervention seem not to exist among our measures. Surgical closure of the duct can be achieved without additional morbidity and mortality. On the basis of previous and our findings, the following suggestions for PDA diagnosis and PDA closure in very low weight newborns are made: in all newborns weighing less than 1500 g at birth, echocardiography should be performed on day 3 of life. A hemodynamically relevant duct should be closed pharmacologically. Otherwise, if the ventilator therapy has to be intensified even after day 7 of life, echocardiography should be repeated. If the echocardiographic findings and the clinical signs speak for a hemodynamically relevant duct, the PDA should be surgically closed in the NICU.The findings strongly suggest an early surgical procedure in case the duct is of hemodynamically relevance after failed indomethacin therapy has been initiated timely.
This original article was presented in part at the 49th International Congress of the European Society for Cardiovascular Surgery 2000, Dresden, Germany. doi:10.1016/j.icvts.2004.03.007
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
Top
Abstract
1. Introduction
pts); (2) pharmacological therapy with indomethacin (amuno, Merck, Sharp and Dohme, UK): initially 0.2 mg/kg, and if needed a second dose of 0.2 mg/kg after 12 h and a third dose of 0.2 mg/kg after 36 h (
pts; (3) surgical closure (primary 20 pts plus 16 pts after failed indomethacin therapy) via a lateral thoracotomy of the third intercostal space in the operating room. Indication for surgical closure of the duct was a hemodynamically relevant PDA evidenced via echocardiography (significantly increased ratio between left atrium and aortic root) together with a prolonged ventilation (on average 20 days preoperatively). 
(10%): 7 after conservative therapy in group I, and 1 pt died on day 6 and one pt on day 7 post surgery in group III. Postoperative ventilation was on average 31 days (2–101 days). Duration of intubation was on average 27 days (1–104 days) in group I, 42 days (16–90 days) in group II, and 58 days (16–139 days) in group III. Early and late mortality in the three groups was 19 (22%). Even the 10 late deaths were in-hospital deaths caused by: sepsis syndrome 4pts, massive intracerebral hemorrhage 2pts, irreversible cardiac decompensation 3pts, and fatal aspiration in 1pt. There were no significant differences between the three groups with respect to general complications (