A Comparison of On-Site and Off-Site Patent Ductus Arteriosus Ligation in Premature Infants
Introduction. Persistent patent ductus arteriosus (PDA) often produces hemodynamic and respiratory derangement necessitating use of inotropic drugs and escalating ventilatory support in premature infants. When medical therapy fails, surgical ligation is indicated. Because of the risks of transferring unstable neonates to the operating room, ductal ligation is routinely performed at the neonatal intensive care unit (NICU) bedside. Some patients, however, require transfer from hospitals without pediatric cardiac surgical teams. In an attempt to eliminate the risks associated with transfer, a surgical team from our institution offered to perform duct ligation in the NICUs of referring institutions. This experienced team consisted of a pediatric cardiac attending anesthesiologist and certified registered nurse anesthetist, cardiac operating room nurses, an attending cardiothoracic surgeon, and a cardiothoracic surgery fellow. We retrospectively reviewed our experience.
Methods. After approval from the Committee for the Protection of Human Subjects, the charts of premature neonates who underwent PDA ligation in the NICU at the Children’s Hospital of Philadelphia NICU or in a network NICU between January 1996 and April 2002 were reviewed. Data abstracted included institution, gender, gestational age, birth weight, weight at surgery, and number of courses of indomethacin. Mean arterial blood pressure and use of inotropic drugs and ventilatory parameters (fraction of inspired oxygen, peak inspiratory pressure) were recorded at the time of surgery and 96 hours postoperatively. Perioperative complications were recorded.
Results. Seventy-two patients met the criteria for inclusion. PDA ligation was performed in the Children’s Hospital of Philadelphia NICU in 38 of 72 patients, 53% (group 1). The remainder, 34 of 72 (47%) underwent PDA ligation in the NICU at 1 of 6 referring institutions (group 2). There were no significant differences between groups with respect to demographics, number of courses of indomethacin, or use of inotropic drugs or ventilatory support. The incidence of perioperative complications did not differ between groups: 3 in group 1 (bleeding, chylothorax, and pleural effusion) and 3 in group 2 (pneumothorax ). There were no anesthetic-related complications. Seven patients died (4 in group 1 and 3 in group 2), none within 96 hours of surgery and none secondary to the procedure.
Discussion. The data demonstrate that an experienced team can perform PDA ligation safely in NICUs of hospitals without on-site pediatric cardiac surgical capabilities in critically ill neonates without incurring the risks inherent in patient transport. Most importantly, patient care is continued by the neonatology team most familiar with the infant’s medical and social history, and the patient’s family is minimally inconvenienced.
Although the ductus arteriosus usually closes spontaneously shortly after birth in the term infant, a higher incidence of persistent patency exists in the premature infant, resulting in left-to-right shunting with increased pulmonary blood flow and decreased systemic perfusion. The left ventricular volume overload and increased pulmonary blood flow may result in myocardial dysfunction and respiratory insufficiency requiring escalating ventilatory support. Complications of patent ductus arteriosus (PDA) include necrotizing enterocolitis, renal failure, bronchopulmonary dysplasia, and intraventricular hemorrhage.1,2 When medical therapy with indomethacin, fluid restriction, and diuretics fails or is deferred to minimize adverse side effects, surgical ligation is indicated.
Transfer of these infants from the neonatal intensive care unit (NICU) to the operating room has been shown to be associated with hemodynamic instability, inability to maintain optimal ventilation and oxygenation, hypothermia, endotracheal tube dislodgment, and loss of indwelling lines with interruption of continuous infusions. Ductal ligation is routinely performed at the NICU bedside at the Children’s Hospital of Philadelphia and others around the country to minimize the risk of patient transport.3–6
Some patients, however, are transferred from referring hospitals in which a pediatric cardiac surgical team is not available to other facilities for surgical intervention. Frequently, duct ligation is the only indication for transfer, and most patients return to the NICU of origin 24 to 48 hours after the procedure. In addition to exposing the infant to the greater distance and time inherent with interhospital transport, this process disrupts the continuity of care provided by the primary neonatology team and places additional demands on already stressed families.7–9
Seven years ago, in an attempt to eliminate the risks inherent in transfer of these neonates, a team consisting of cardiac surgeons, cardiac anesthetists, and dedicated cardiac operating room nurses offered to perform ductal ligation in the NICUs of our referring institutions. The local neonatologists, alone or in conjunction with the traveling team, provided postoperative care. The following is an observational study of our experiences in performing PDA ligation in both on- and off-site NICU settings.
All premature infants (≤37 weeks gestation) who underwent PDA ligation by our team in a network NICU between January 1996 and April 2002 were identified by using the cardiothoracic surgery database. After obtaining approval from the Committee for the Protection of Human Subjects, all of these patient’s charts were sought from medical records, and those obtained were reviewed. Patients with additional cardiac lesions were excluded.
Documents reviewed included anesthetic record, operative report, nursing and respiratory flow sheets, and physician progress notes. Abstracted data included the institution where the procedure was performed, gender, gestational age, birth weight, weight at time of surgery, and number of courses of indomethacin. Use of inotropic drugs, mean arterial pressure, and ventilatory parameters (fraction of inspired oxygen, peak inspiratory pressure) were recorded both at time of surgery and 96 hours postoperatively as surrogate indicators of severity of illness. The number and type of surgical and anesthetic complications were noted. In those patients who expired, the cause of death and relationship to surgery were recorded. In both groups, records for the entire hospital stay were reviewed.
The patients were separated into 2 groups based on the site at which the procedure was performed; group 1 represents those done in the NICU at Children’s Hospital of Philadelphia, and group 2 is comprised of those done in the NICUs of 6 referring institutions.
Within 24 hours of planned surgery, each patient was seen by a pediatric cardiologist, and the presence of a PDA was confirmed by transthoracic echocardiography. Patients in the Children’s Hospital of Philadelphia NICU were evaluated by surgical and anesthesia personnel before the procedure. Preoperative information on those patients at off-site NICUs was obtained by telephone the day before the procedure from nursing and neonatology staff and updated immediately before surgery. All surgical instruments, resuscitative equipment, medications, and other supplies that might be needed in the conduct of the procedure were brought to the NICU bedside. Two 1/4 units of packed red blood cells, washed or <7 days postdonation, were cross-matched and available at the bedside. Before positioning the patient in the right lateral decubitus position, proper endotracheal tube placement was confirmed with a recent chest radiograph and auscultation. Patients received 5 to 10 μg/kg of fentanyl citrate, 0.1 mg/kg of pancuronium bromide, and 25 mg/kg of cefazolin. All medication infusions were continued. A left thoracotomy was performed in all cases. The duct was mobilized superiorly and inferiorly and occluded with 1 or 2 titanium hemaclips. To maintain adequate oxygenation and ventilation in the face of surgical retraction on the left lung, ventilation during the procedure was accomplished by increasing the fraction of inspired oxygen to 1.0 and either increasing the ventilator minute ventilation or manually ventilating with a Mapleson circuit. A 5 French pleural catheter was inserted, and a postoperative chest radiograph was obtained. The pleural catheter was removed by the NICU staff 24 to 72 hours postoperatively.
Sample characteristics for the 2 groups were calculated and are presented in Table 1. Statistical comparisons between the 2 groups were conducted by using the Mann-Whitney U test of location for 3 sets of variables (demographic, preoperative, and postoperative). All preoperative data were obtained on the day of surgery, whereas all postoperative data reported were obtained 96 hours after the procedure. The Mann-Whitney test was used as the test of choice for 2 reasons: the rather skewed nature of the distributions and the rather small sample sizes for some of the variables (eg, amplitude [AMP]DOS and AMP96). Hence, all descriptive statistics provided in Table 1 are presented as medians and ranges. The hypothesis-wise error rate was adjusted for the 13 comparisons by using Tukey, Ciminera, and Heyse’s adjustment for multiple moderately correlated comparisons.10
A total of 80 patients who met criteria were identified from the cardiothoracic surgery database. Of these, 72 charts were available for review. Of the 8 unavailable charts, 3 were from group 1 and 5 were from group 2. Duct ligation in the Children’s Hospital of Philadelphia NICU accounted for 38 of 72 (53%) (group 1), and 34 of 74 (47%) were at 6 referring NICUs (group 2). The referring NICUs are located in Pennsylvania and New Jersey and range in distance from 1 block to 75 miles from our facility. They are affiliated with a variety of institutions varying in size from large academic medical centers to small community hospitals. All the referring facilities have traditionally transferred patients to our center for PDA ligation. Of the patients in group 1, 21 of 38 (55%) were transferred to our institution specifically for the purpose of undergoing PDA ligation. The majority of these patients were referred from out-of-network NICUs.
There were no differences with respect to patient characteristics or indomethacin therapy between the 2 groups (Table 1). At the time of surgery, 3 of 38 (8%) patients in group 1 and 6 of 34 (18%) in group 2 were being treated with high-frequency jet ventilation, 33 of 38 in group 1 (87%) and 26 of 34 (76%) in group 2 were on conventional ventilatory support, and the remainder, 2 of 38 (5%) in group 1 and 2 of 34 (6%) in group 2 had natural airways. Three surgical complications occurred in group 1: bleeding from the ductus after clip application (1), chylothorax (1), and pneumothorax (1). Three surgical complications occurred in group 2: pneumothorax (3). There were no anesthetic-related complications, defined as extubation or intraoperative ventilatory or hemodynamic instability.
The majority of patients in both groups (84% and 76%, respectively) had their chest tubes removed by postoperative day 1, and only 1 patient had an indwelling chest tube after postoperative day 3.
There were 7 deaths, 4 in group 1 and 3 in group 2. Causes of death were sepsis, intraventricular hemorrhage, and respiratory failure. None of the deaths occurred within 96 hours of duct ligation.
There were no statistically significant differences between the 2 groups of patients for any of the variables presented here using the adjusted α (ά = 0.005).
The data demonstrate that an experienced team can perform PDA ligation safely in both on- and off-site NICUs in critically ill infants without incurring the risks inherent in patient transport. Additionally, continuity of care is maintained by the neonatology team most familiar with the infant’s medical and social history with minimal inconvenience to the patient’s family.
In all cases, the same experienced team of surgeons, anesthesia providers, and operating room nurses, all of whom work together on a routine basis caring for patients with congenital heart disease, performed the procedure. There were no differences in technique, personnel, or equipment—only in the hospital where the procedure was performed. The safety and efficacy of this technique depends on the utilization of a team to perform off-site duct ligation. Travel by an isolated surgeon, working without familiar equipment and nursing/anesthesia personnel, could introduce additional risks.
Meticulous preoperative assessment, planning, and intraoperative attention to detail are crucial. The needs of the visiting team must be made clear to the host NICU well in advance of arrival for surgery. Appropriate planning and cooperation between surgeons, neonatologists, nurses, anesthesia providers, and cardiologists along with support services such as blood bank, laboratory, and radiology is essential. Recently, nursing staff from non-network NICUs have observed duct ligation in our NICU in anticipation of hosting our visiting team.
In addition to safety and efficacy, performing PDA ligation in the primary NICU setting offers multiple advantages. The neonatology team caring for a premature neonate invests a great deal of time in devising a plan of care and establishing rapport and trust-based relationships with family members. Transfer to another NICU for PDA ligation, even for a relatively brief period, disrupts the continuity of care and exposes family and patient to a potentially different and confusing environment. Allowing the primary neonatology group, with the cooperation and assistance of the surgical team, to manage the postoperative care of the patient offers the benefit of having those most familiar with the patient and family immediately present throughout the perioperative period. Furthermore, the risk of delaying surgical therapy because of concerns over transfer is avoided. A surgical time slot (within 36 hours of initial request for elective procedures and 4 hours for acute procedures) has routinely been offered to all NICUs.
Having 1 (or more) infant hospitalized in the NICU is a tremendously stressful time for parents or guardians. This only intensifies when the infant must undergo a surgical procedure. Often the decision to proceed with surgery is presented at the same time that plans are made for patient transfer. The prospect of following one’s infant to a new environment with unfamiliar personnel is associated with significant stress. Managing to arrange for transportation, lodging, and the care of other children can be problematic and overwhelming.
Performing PDA ligation in the NICU, on- or off-site, is safe and feasible and can work to the advantage of both patients and families. There has been an overwhelmingly positive response from the NICUs that we serve, and a number of additional sites have been added. In view of the potential benefits of this approach, we believe that additional experience is both needed and warranted.
The purpose of this study was to demonstrate the safety and feasibility of this approach to PDA ligation in the NICU. We recognize that additional work with respect to cost-benefit analysis would be of value.
We thank William J. Greeley, MD, and David R. Jobes, MD, for their thoughtful reviews of the manuscript.
- Copyright © 2003 by the American Academy of Pediatrics