As neonatal intensive care has evolved over the past decades, the clinical decision to treat preterm infants with persistent patent ductus arteriosus (PDA) with surgical ligation has become increasingly fraught with uncertainty. The high efficacy of early pharmacotherapy and advances in non-invasive ventilation have contributed to a progressive and sustained reduction in referrals for surgical ligation among very preterm infants. In addition, observational associations of ligation with adverse neonatal and neurodevelopmental outcomes have prompted concern for harm and rendered the ligation decision more challenging for clinicians, though these findings may be due to residual bias rather than a true causal detrimental effect of surgery. Ligation may still be indicated for some infants, though there is a need for comprehensive methods of evaluation to reliably identify high volume shunts whose surgical interruption may result in measureable benefit. A randomized clinical trial of surgical ligation vs. conservative management among preterm infants with persistent PDA after unsuccessful pharmacotherapy is necessary to appropriately assess the efficacy of surgical intervention. A trans-catheter approach to ductal closure in the very immature infant is an exciting therapeutic alternative that is still in its infancy. For infants who do undergo surgery, postoperative hemodynamic instability may be related to increased left ventricular afterload and potentially ameliorated by treatment with milrinone. This chapter will review the evidence regarding the benefits and risks of PDA ligation in preterm infants and propose a pathophysiology-based management paradigm to guide perioperative care.
Keywordsbronchopulmonary dysplasia, catheter, conservative management, death, extremely low birthweight, milrinone, mortality, neurodevelopmental impairment, patent ductus arteriosus (PDA), postligation cardiac syndrome, retinopathy of prematurity
Surgical ligation is a therapeutic option for persistent large patent ductus arteriosus (PDA) after unsuccessful medical therapy, although additional research is needed to determine the clinical and echocardiography characteristics associated with improved outcomes compared with conservative management.
Prophylactic surgical ligation in the first postnatal days is generally unnecessary due to the availability of pharmacologic therapy, the possibility of spontaneous ductal closure and the potential for coexistent pulmonary arterial hypertension.
Epidemiologic associations between PDA ligation and increased neonatal morbidity and neurodevelopmental impairment may be due to residual bias from confounding by indication, rather than a true detrimental causal effect of PDA surgery.
Transcatheter closure of PDA is an emerging therapeutic approach that requires further study as an alternate to open surgical ligation.
Postligation cardiac syndrome, characterized by oxygenation and ventilation failure and systemic hypotension with onset 6 to 12 hours postoperatively, is common in preterm infants and likely related to increased left ventricular afterload.
The early targeted administration of intravenous milrinone to preterm infants with low cardiac output in the immediate postoperative period may ameliorate the hemodynamic significance of postligation cardiac syndrome.
The approach to the acutely unstable postoperative infant should include consideration of postligation cardiac syndrome, nonspecific inflammatory response, potential surgical complications, adrenal insufficiency, and the possibility of occult infection (specific inflammation).
Surgical ligation is an immediate and definitive method for closure of patent ductus arteriosus (PDA). However, the selection of preterm infants for surgical treatment remains one of the most enduring controversies in neonatal medicine. There is marked center-to-center variation in the incidence of ligation among extremely low birth weight (ELBW) infants. Overall rates of ligation reported by international neonatal networks are decreasing, although the reasons behind this secular trend are likely multifactorial.
Observational studies have associated PDA surgery with adverse neonatal outcomes and neurodevelopmental impairment (NDI) in early childhood, although bias due to residual confounding threatens the validity of these studies. Infants treated with surgical ligation undergo major rapid changes in systemic hemodynamics which commonly lead to postoperative cardiorespiratory instability. In addition, given that dependence on mechanical ventilation is the sine qua non of referring an infant for PDA surgery, the increasing availability and use of advanced noninvasive methods of ventilation may permit earlier endotracheal extubation of extremely preterm infants. Consequently, clinician perceptions of the merits of surgical intervention may differ, prompting them to avoid PDA surgery with the hope of spontaneous ductal closure, yet this management strategy has not been rigorously evaluated.
As a result, contemporary practice is dominated by considerable uncertainty regarding the role of surgical ligation. In this chapter, we review the evidence regarding the benefits and risks of PDA surgery in very preterm neonates and provide a pathophysiology-based management paradigm to guide perioperative care in these high-risk infants.
Patent Ductus Arteriosus Ligation: Evidence of Benefit Versus Harm in Randomized Clinical Trials
A limited number of randomized clinical trials, all conducted more than 3 decades ago, have evaluated the impact of surgical ligation in preterm infants on neonatal outcomes ( Table 24.1 ). Two trials were conducted prior to the routine use of indomethacin treatment to pharmacologically achieve earlier ductal closure. Cotton et al. randomized 25 very low birth weight preterm infants (mean gestational age [GA] 28 weeks) with a clinical diagnosis of symptomatic PDA and requiring invasive mechanical ventilation at 1 week of postnatal life to surgical ligation versus ongoing medical management, which consisted of diuretic and digoxin therapy. Infants treated with surgical ligation had a significantly shorter time to successful extubation (median 6 vs. 22 days after enrollment, P < .05). In contrast, Levitsky et al. randomized 31 moderately preterm infants (mean GA 31 weeks) with respiratory distress syndrome and clinically diagnosed PDA to ligation or medical management. There were no differences in neonatal outcomes, although nearly half of the medically treated group was ultimately treated with surgery.
|Author (Year)||Size||Population||Intervention||Comparison||Outcome||Ligation group||Comparator group||P -value|
|Levitsky et al. (1976)||n = 31||Ligation ( n = 8)||No additional treatment, with ligation as backup ( n = 23) |
Rescue ligation: 11/23 (48%)
|Survival (in-hospital)||4 (50%)||16 (70%)||.32 a|
|Cotton et al. (1978)||n = 25||Ligation ( n = 10)||No additional treatment with ligation as backup ( n = 15) |
Rescue ligation: 2/15 (13%)
|Survival (in-hospital)||9 (90%)||12 (80%)||.50 a|
|Duration of endo-tracheal intub-ation (median)||6 days||22 days||<.05|
|Gersony et al. (1983)||n = 154||Ligation ( n = 79)||Indomethacin ( n = 75) |
Rescue ligation: 25/75 (33%)
|Survival (in-hospital)||67 (85%)||58 (77%)||.3|
|BPD||31 (39%)||23 (31%)||.3|
|ROP (III or IV)||12 (15%)||3(4%)||.02|
|NEC||5 (6%)||5 (7%)||.9|
|≥ 14 days invasive ventilation||22 (30%)||22 (32%)||.84 a|
|Cassady et al. (1989) and Clyman et al. (2009) [post hoc analysis]||n = 84||Prophylactic ligation within 24 h of birth ( n = 40)||Standard treatment ( n = 44)||Survival to 1 year||25 (63%)||26 (59%)||.75 a|
|NEC||3 (8%)||13 (30%)||.002|
|IVH III/IV b||16/35 (46%)||23/41 (56%)||.37 a|
|BPD b (Bancalari definition)||14 (35%)||16 (36%)||.90 a|
|Oxygen at 36 weeks c||11(48%)||4(21%)||<.05|
|Mechanical ventilation at 36 weeks c||6 (26%)||0 (0%)||<.05|
Only one trial, involving 154 preterm infants (median GA 28 to 29 weeks) with symptomatic PDA despite conservative management, compared surgical ligation versus a first course of indomethacin therapy. Infants randomized to ligation had a higher incidence of severe retinopathy of prematurity (ROP), and there were no other differences in neonatal outcomes. Cassady et al. reported reduced rates of necrotizing enterocolitis (NEC; 30% vs. 8%, P = .002) in ELBW preterm infants undergoing prophylactic ligation on the first postnatal day, although a recent post hoc secondary analysis reported an increased risk of moderate-severe bronchopulmonary dysplasia (BPD) in surgically treated infants.
Taken together, the common salient feature of trials of PDA ligation is a lack of external validity to permit their interpretation within contemporary practice. Although dependence on mechanical ventilation has remained the most common indication for referring infants for surgical treatment, advances in respiratory care (e.g., exogenous surfactant therapy, inline flow sensors, and noninvasive positive pressure ventilation) have dramatically altered the trajectory of mechanical ventilation for respiratory distress syndrome, including earlier endotracheal extubation. The advent of prophylactic indomethacin and the concept of exposing all infants (potentially including some with pathologically elevated pulmonary vascular resistance) to the incipient risks of early surgical ductal closure render prophylactic ligation untenable in modern practice. In addition, echocardiography is currently the definitive and integral method of assessing the magnitude of the ductal shunt and guiding management. This modality was conspicuously absent in these older trials alongside a limited, and predominantly clinical, definition of hemodynamic significance. Finally, over the past decade, surgical treatment has become uncommon in infants born at GA ≥27 weeks and is typically only considered after pharmacotherapeutic closure has failed or was contraindicated, representing a population and therapeutic approach that has not been evaluated.
Patent Ductus Arteriosus Ligation and Outcomes: Associations from Observational Studies
Studies have reported increased neonatal morbidity and NDI in early childhood among infants treated with PDA ligation compared with medical management alone. In these large retrospective cohort studies, infants were categorized by treatment assignment (conservative management, cyclooxygenase inhibitor [COXI] only, COXI followed by surgical ligation, or primary ligation) and outcomes compared between treatment groups.
In a large retrospective cohort study of preterm infants born less than 32 weeks’ GA with a symptomatic PDA, Mirea et al. compared neonatal outcomes according to PDA treatment assignment. After adjustment for antenatal and perinatal confounders, infants treated with surgical ligation had lower mortality but higher odds of BPD and ROP, compared with infants treated with medical management alone. Similarly, in a retrospective review of 426 ELBW infants, Kabra et al. detected higher BPD and ROP in 110 infants who underwent surgical ligation compared with 316 infants who received medical management only. Higher rates of BPD among infants with PDA ligation was also detected by Madan et al. in a review of 2838 ELBW treated for symptomatic PDA. Several studies also reported an association of PDA ligation with increased rates of NDI in early childhood. A recent systematic review and metaanalysis of randomized trials and controlled observational studies demonstrated that compared with medically treated infants, ligated infants were more likely to develop BPD, severe ROP, and NDI, although with improved survival ( Table 24.2 ).
|Odds Ratio (95% Confidence Interval) a|
|Study||Characteristics||Antenatal/Perinatal Confounders Adjusted||Postnatal Confounders Adjusted||Death or NDI||Death||NDI||Severe ROP||CLD|
|Kabra et al. 2007||ELBW infants with symptomatic PDA |
PDA ligation ( n = 110)
Medical only ( n = 316)
|GA, sex, ACS, multiples, mother’s education, total dose indomethacin||None||1.55 (0.97, 2.50)||0.56 (0.29, 1.10)||1.98 (1.18, 3.30)||2.20 (1.19, 4.07)||1.81 (1.09, 3.03)|
|Madan et al. 2009 b||ELBW infants with PDA |
Primary ligation ( n = 135), Indo only ( n = 1525), Indo and ligation ( n = 775)
No treatment ( n = 403)
|GA, BW, gender, ACS, center, prophylactic indomethacin, Apgar score, RDS, IUGR, antenatal infection, maternal marital status and age||Postnatal sepsis||1.03 (0.80, 1.33)||0.46 (0.35, 0.62)||1.53 (1.16, 2.03)||n/a||3.10 (2.26, 4.26)|
|Mirea et al. 2012 c||Infants with GA ≤ 32 weeks with PDA. Conservative ( n = 577), Indo only ( n = 2026), Indo + ligation ( n = 626), Primary ligation ( n = 327)||GA, ACS, multiples, gender, IUGR, SNAP II||None||n/a||0.41 (0.31, 0.54)||n/a||1.91 (1.51, 2.41)||2.30 (1.91, 2.77)|
|Bourgoin et al. 2016 d||Infants with GA ≤ 28 weeks with PDA. Conservative ( n = 505), Ibuprofen only ( n = 248), Ligation ( n = 104)||gender, GA, BW Z-score, ACS, gestational hypertension, clinical chorioamnionitis, Apgar score, place of hospitalization, place of birth, year of birth, delivery characteristics||None||n/a||n/a||1.9 (1.1, 3.1)||n/a||n/a|
|Weisz et al. 2017 e||Infants with GA < 28 weeks with clinical and echo diagnosis of PDA |
Ligation ( n = 184)
Medical treatment only ( n = 570)
|GA, IUGR, ACS, gender, multiples, SNAP II score ≥ 20, center||sepsis, severe IVH, inotropes, NEC >2, average daily mean airway pressure, No. days invasive ventilation, total dose of indomethacin, systemic corticosteroids||0.83 (0.52, 1.32)||0.09 (0.04, 0.21)||1.27 (0.78, 2.06)||1.61 (0.85, 3.06)||1.36 (0.78, 2.39)|
Retrospective epoch studies have reported a similar association between PDA ligation and NDI; specifically, an improvement in neurodevelopmental outcomes coincided with a reduced proportion of infants referred for surgical ligation. Wickremasinghe et al. evaluated 18- to 36-month neurodevelopmental outcomes after moving from an early, aggressive surgical approach (infants with echocardiographic evidence of PDA after medical therapy were immediately referred for surgery) to a delayed selection ligation approach (infants with a persistent PDA were referred for ligation if the PDA was clinically and echocardiographically significant). Infants in the selective ligation epoch were less likely to be treated with surgery (66% vs. 100%) and had less NDI (adjusted odds ratio [OR] 0.07, 95% confidence interval [CI] 0.00 to 0.96), potentially reflecting a benefit of avoiding ligation in infants with smaller ductal shunts. These studies highlight the importance of the ascertainment of hemodynamic significance, particularly in appraising the merits of surgical intervention.
Aspects of PDA ligation that have been postulated to contribute to the risk of NDI include surgical and anesthesia effects and postoperative hemodynamic compromise. Vocal cord paresis is a common surgical complication and is associated with an increased risk of death, extubation failure and chronic lung disease, need for gastrostomy tube, and gastroesophageal reflux disease. Recent studies have associated use of halothane gases for anesthesia in young children with NDI. Preterm infants are at risk of postoperative hypotension and cardiogenic shock due to postligation cardiac syndrome (PLCS), which may result in cerebral hypoperfusion and injury. Although it is important to acknowledge the clinical importance of these complications, causality is not implied.
In light of concerns regarding NDI and neonatal morbidities, the safety of PDA ligation has been questioned. These concerns have been associated with a secular trend toward a reduction in infants being treated with surgical ligation in North American centers. On the contrary, the reduction in ligation has been associated with increased BPD in one large American neonatal network, suggesting caution before imposing radical changes in practice ( Fig. 24.1 ).
Neonatal Morbidity, Neurodevelopmental Impairment, and Patent Ductus Arteriosus Ligation: Residual Bias in Observational Studies
Residual selection bias and bias due to confounding by indication threaten the validity of observational studies that have associated PDA ligation with increased neonatal morbidity and NDI but lower mortality compared with medical management alone. The divergence of these competing outcomes (lower mortality but increased BPD, ROP, and NDI) may be explained by several possible situations: “First, surgical ligation may improve the survival of infants with PDA but may be simultaneously neurologically detrimental, either directly through the surgery-associated inflammatory response or indirectly via worsening lung disease and ventilator dependency.” Second, ligation may improve the survival of infants with PDA, but the infants referred for ligation are at higher preligation risk of NDI (confounding by indication and increased preligation illness severity). Finally, the decreased mortality may be a spurious finding influenced by survival bias (where moribund nonligated infants with a PDA die before becoming eligible for ligation), and the increase in NDI may be due to either a true detrimental effect of ligation, the effect of confounding by indication, or a combination of these two possibilities.
Studies to date have inadequately addressed confounding by indication—that infants referred for ligation may be more “ill” and/or have larger ductal shunts at the time of the decision to treat with surgery, compared with infants who are treated with medical management alone. Illness severity, characterized by postnatal morbidities such as periventricular/intraventricular hemorrhage (P/IVH), NEC, and sepsis, and parameters of physiologic instability such as hypotension, predict both neonatal morbidities and NDI (see Table 24.2 ).
The intensity and duration of invasive mechanical ventilation is a particularly important confounder of the association between ligation and adverse outcomes. Prolonged ventilator dependence is a commonly used clinical criterion in the decision to treat with ligation (vs. medical management) and also strongly predicts neonatal morbidities such as BPD, ROP, and NDI.
Survival bias may have influenced the reported lower mortality among ligated infants. Ligation is generally undertaken later in life after failure of medical therapy, meaning that ligated infants are more likely to have already survived the critical period of high early neonatal mortality. This treatment paradigm implies that some of the sickest infants, treated initially with conservative management and/or COXIs, may have died prior to becoming “eligible” for ligation, resulting in selection bias in assembling the cohort of ligated infants. The possibility of survival bias is supported by studies which found no difference in mortality between medically and surgically treated groups of infants who both received treatments at a similar postnatal age. Taken together, survival bias may be present and any beneficial effect of ligation on mortality requires additional study prior to clinicians being confident of a survival benefit of PDA surgery.
A recent large cohort study ( n = 754) of extremely preterm infants with PDA compared neonatal and neurodevelopmental outcomes of infants treated with ligation versus medical treatment, while minimizing the effect of residual bias due to confounding by indication (see Table 24.2 ). Multivariable logistic regression analysis was used to adjust for antenatal, perinatal, and postnatal, preductal closure confounders, including the duration and intensity of mechanical ventilation and morbidities such as NEC and sepsis. Infants treated with ligation had higher ventilation requirements than did infants who were not treated with ligation, a confounder not controlled for in studies to date ( Fig. 24.2 ). As with previous observational studies, when adjusting only for antenatal and perinatal confounders, ligation was associated with increased BPD, ROP, and NDI. However, after further adjustment for postnatal, preductal closure confounders (e.g., burden of mechanical ventilation, sepsis), ligation was no longer associated with any adverse outcome. This study suggests that the associations of ligation with adverse outcomes from earlier studies were likely due to confounding by indication rather than a true detrimental causal effect of ligation. These findings have direct clinical relevance; specifically, neonatologists and pediatric cardiac surgeons may now consider to no longer prioritize the risk of adverse neonatal and neurodevelopmental outcomes as a reason to avoid surgical PDA ligation.
Patent Ductus Arteriosus Ligation Decision: Timing, Patient Selection, and Staging
In contemporary practice the “PDA ligation decision” remains an enduring controversy for clinicians. There is currently a paucity of knowledge regarding the clinical and echocardiography characteristics of infants with persistent PDA who may benefit from ligation. The relative risks and benefits of rescue surgical ligation compared with conservative management are unknown. Although dependence on mechanical ventilation remains the primary criterion for surgical referral, the relative independent contribution of the PDA to ongoing respiratory insufficiency is, at present, difficult to quantify. Infants with similar echocardiography indices of PDA hemodynamic significance may have varying degrees of respiratory failure (severe to none) owing to differences in nascent lung disease of prematurity, pulmonary arterial pressure, and tolerance of the increased pulmonary blood flow from the ductal shunt.
Limited evidence from clinical trials and careful reflection on contemporary treatment practices provides some boundaries for surgical treatment, mostly by limiting the use of ligation in the first postnatal week. Although the trial by Cassady et al. reported reduced NEC with prophylactic ligation on the first postnatal day in ELBW infants, several factors render this practice now untenable. First, prophylactic indomethacin has both relatively high efficacy for early closure and reduces all grades of P/IVH, providing a therapeutic alternative with additional benefit. Second, PDA ligation may result in increased right ventricular (RV) afterload, so early ligation may be harmful in preterm infants with increased pulmonary artery pressure, a common finding in severe respiratory distress syndrome. Finally, the natural history of ductal closure has been described, with many infants experiencing early spontaneous closure. Exposing all infants to the risk of prophylactic surgical ligation is therefore inappropriate.
Epoch studies have reported that preterm infants treated with delayed selective ligation (surgical PDA closure after failure of medical therapy only if infants had concomitant respiratory failure, systemic hypotension, or signs of end-organ hypoperfusion) had improved rates of BPD and NDI compared with infants treated with early, routine ligation. Ligation prior to the 10th postnatal day was independently associated with adverse outcomes.
Beyond the first 2 weeks after delivery, there is a paucity of data specifically addressing the impact of ligation versus conservative management in infants with persistent PDA after unsuccessful pharmacologic therapy. In most cases the persistent ductal shunt may be considered a chronic, consistent contributor to impaired pulmonary compliance. Infants with a persistent PDA may be considered for surgical ligation when the clinical and echocardiography evaluations identify a shunt of sufficient volume that may be actively contributing to respiratory insufficiency. Persistent dependence on invasive or noninvasive ventilation in combination with moderate-severe echocardiography indicators of hemodynamic significance suggests an impaired ability to compensate for the ductal shunt.
However, it is important to consider preterm infants who have clinical features of chronic end-organ hypoperfusion (systemic hypotension, renal failure, feeding intolerance) in the absence of an acute etiology (e.g., sepsis, NEC) accompanied by echocardiographic indices of a large ductal shunt. In these infants, echocardiographic indices are often in the “severe” range across all parameters, although there are little data regarding the direct clinical relevance of these deviations from normal ( Table 24.3 ). Early surgical ligation after failure of medical therapy may be indicated for this subgroup of infants.