You are here

Article Text

Article menu

Download PDFPDF

Original articles
Outcomes following the surgical ligation of the patent ductus arteriosus in premature infants in Scotland
  1. Anne Marie Heuchan1,
  2. Lindsey Hunter2,
  3. David Young3
  1. 1Department of Neonatal Medicine, The Royal Hospital for Sick Children, Glasgow, UK
  2. 2Department of Paediatric Cardiology, The Royal Hospital for Sick Children, Glasgow, UK
  3. 3Department of Mathematics and Statistics, University of Strathclyde, Glasgow, UK
  1. Correspondence to Dr Anne Marie Heuchan, Department of Neonatal Medicine, The Royal Hospital for Sick Children, Yorkhill, Glasgow G3 8SJ, UK; annemarie.heuchan{at}ggc.scot.nhs.uk

Abstract

Objective To determine morbidity, mortality and associated risk factors following patent ductus arteriosus (PDA) ligation in premature infants.

Methods Retrospective case note audit of premature infants referred to a national paediatric cardiothoracic surgical service (2001–2007) with univariate and multivariate analysis of potential risk factors for mortality and morbidity.

Results 125 infants were enrolled (median gestational age 26 weeks (IQR 25–27 weeks), median birth weight 840 g (IQR 730–1035 g)). Referral characteristics were median LA:Ao 1.8 (IQR 1.5–2.0), 80% ventilated, 18.4% continuous positive airway pressure, 70% diuretics and 58% prior treatment with cyclooxygenase inhibitors (COIs). Median age at PDA ligation was 31 days (IQR 25–41 days). Postoperative characteristics were median time to extubation 5 days (IQR 3–10 days), 36.0% corticosteroids, 46.8% domiciliary oxygen and 4.8% vocal cord palsy. The 30-day and 1-year mortality rates were 4.8% and 12.8%, respectively, with neurodisability in 32% of survivors. All deaths occurred in the ventilated group and were mainly attributable to bronchopulmonary dysplasia (BPD). Gestation and fractional inspired oxygen (FiO2)>60% were significantly associated with 30-day mortality. FiO2, ventilation, lack of prior COIs and postoperative corticosteroids were significantly associated with 1-year mortality. Preoperative FiO2>40% and lack of prior COIs retained independent significance for death at 1 year.

Conclusions PDA ligation is well tolerated, with evidence of early benefit. The incidence of neurodisability or death from BPD at 1 year remains high. Increasing preoperative FiO2 and lack of prior treatment with COIs are associated with increased mortality at 1 year.

https://doi.org/10.1136/adc.2010.206052

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Background

    The safety of and benefits of management of the patent ductus arteriosus (PDA), and in particular surgical ligation, remain uncertain.1 The Royal Hospital for Sick Children (RHSC), Glasgow became the designated National Paediatric Cardiothoracic Surgery centre for Scotland in 2000. The first publication of national Congenital Cardiac Audit Database (CCAD) audit data in 2007 suggested that postoperative mortality following PDA ligation in Scotland was higher than the UK average, prompting a detailed audit of the service. Additionally, despite a relatively small national population of 5 million people and a birth rate of approximately 11/1000 per annum, there were a total of 12 neonatal intensive care units (NICUs) offering full intensive care during this audit period. Management of PDA between Scottish NICUs is known to be diverse,2 so this study also provided the opportunity to explore the relationship of treatment prior to PDA ligation and outcome. The primary aims of the study were to describe the population characteristics of premature infants referred for surgical ligation of the PDA. The secondary aims were to report early (≤30 days) and late (≤1 year) mortality and morbidity following PDA ligation and to identify significant risk factors for these outcomes.

    What is already known on this topic

    • ▶ The population of infants referred for surgical patent ductus arteriosus (PDA) ligation is extremely premature.

    • ▶ The incidence of bronchopulmonary dysplasia and neurodevelopmental disability is high in infants who have undergone surgical PDA ligation.

    What this study adds

    • ▶ Surgical ligation of the PDA is safe and well tolerated, with subsequent death almost exclusively due to chronic lung disease.

    • ▶ There is a significant association between fractional inspired oxygen at the time of referral for PDA ligation and subsequent death.

    • ▶ Prior treatment with cyclooxygenase inhibitors is associated with reduced 1-year mortality following PDA ligation.

    Methods

    A retrospective case note audit of selected preoperative and postoperative variables and 1-year outcomes for preterm infants referred for PDA ligation from all Scottish NICUs was undertaken during the 7-year period 2001–2007. Approval was obtained from the Scottish Neonatal Consultants Group and local Caldicott Guardians.

    Prior to PDA ligation, all infants had a full echocardiographic and clinical assessment by consultant neonatologists and paediatric cardiologists at RHSC. Surgical ligation was undertaken where there was agreement that there was evidence of significant left heart volume loading and that this was contributing to the ongoing need for respiratory support. The surgical approach was via left posteriolateral thoracotomy with clipping of the PDA and placement of a left sided chest drain.

    Cases were identified from the cardiac database at RHSC and from NICU admission records. Infants were excluded if they had other significant congenital abnormalities or were not born prematurely. We did not include closure by interventional catheterisation.

    Major preoperative, postoperative and perioperative clinical variables were obtained from case notes at RHSC and the referring NICUs. Preoperative respiratory data were defined by the need for ventilation, continuous positive airway pressure (CPAP) or low flow oxygen. A median preoperative oxygen requirement was calculated for each infant for the 24 h period prior to surgical ligation. Pre-referral treatment patterns with cyclooxygenase inhibitors (COIs) and diuretics were obtained by searching case note entries and prescription charts. Abnormal neurodevelopmental outcome at 1 year was defined as a diagnosis of cerebral palsy, hearing impairment requiring aids or registered visual impairment recorded by 1 year of age corrected for prematurity.

    Echocardiographic data were obtained from formal cardiology reports or by review of stored images by a paediatric cardiologist (second author Lindsey Hunter). All images were obtained by fully trained paediatric cardiology technicians utilising a standardised technique for recording images and obtaining measurements.

    The survival status of all infants at 1 year of age and at the time of completing the audit was checked against the referring hospital case notes, medical records appointment systems and the cardiology database at RHSC. The latter are notified of all Scottish deaths at <25 years of age by the Registrar General, Scotland. Where there was uncertainty, direct enquiry was made to the Registrar General. Deaths were defined as early (≤30 days postsurgery) or late (death by 1 year following surgery) based on CCAD UK definitions (http://www.ccad.org.uk/congenital). The cause of death was as recorded on the patient's death certificate.

    Statistical analysis was performed with Minitab (v 15) and SPSS (v 15) using a significance level of 5%. Possible relationships between clinical variables and the outcomes survival at 30 days, survival at 1 year and the combined outcome of death or significant neurodisability at 1 year were explored. Univariate analyses were initially performed to identify variables associated with the outcomes of interest. Quantitative variables were compared using t tests or Mann–Whitney tests as appropriate for the distribution of the data. Qualitative variables were compared using χ2 tests or Fisher's exact tests. Fractional inspired oxygen (FiO2) was analysed both as a continuous variable and as blocks stratified by 5% increments. Receiver operator characteristic (ROC) curves were generated to determine the optimal cut-off in terms of maximising the sensitivity and specificity for FiO2 and prediction of death. For the 1-year follow-up data, a binary logistic regression model was constructed which included all biologically plausible variables with p values <0.1 on univariate analysis as possible predictors. Stepwise selection was used to determine independent predictors of death.

    Results

    A total of 145 infants underwent surgical PDA ligation from 2001 to 2007. Inclusion criteria were met by 129 patients. Missing case notes resulted in four exclusions, leaving a final study cohort of 125. The indication for PDA ligation was predominantly (98.4%) failure to wean from ventilation or CPAP, with evidence of a significant clinical PDA. Data acquisition was complete for all explored variables except preoperative FiO2 and left atrial (LA):aortic root (Ao) ratio, for which data were available from 93.6% and 83% of infants, respectively.

    All data items collected are presented in the tables (tables 1–4), apart from time between referral and admission for ligation (median 3 days, IQR 1–5). Population characteristics and relationships with the outcomes early death and death by 1 year are summarised in table 1. For early death, the only significant variables identified were gestational age and preoperative FiO2. For deaths by 1 year, there were significant associations with lack of prior treatment with COIs, increasing preoperative FiO2, ventilation at referral and postligation corticosteroid treatment. The optimal cut-off for preoperative FiO2 values as predictors of death were preoperative FiO2>40% (ROC 0.723, 95% CI 0.603 to 0.842) for death by 1 year and preoperative FiO2>60% (ROC 0.634, 95% CI 0.365 to 0.887) and death by 30 days. FiO2>40% was selected for inclusion in logistic regression analysis for deaths by 1 year. There were insufficient deaths by 30 days to undertake logistic regression analysis for this group.

    View this table:
    Table 1

    Preoperative variables and death by 30 days and 1 year after PDA ligation

    The relationship between pre-referral COI treatment patterns, number of courses of COIs, timing of PDA ligation and 1-year mortality are presented in table 2. Significant differences in both age at ligation and mortality at 1 year were found between the groups referred for primary PDA ligation and those previously treated with COI, but there was no significant difference in gestational age or weight at birth. The numbers of patients receiving multiple courses of COI are presented. The median age at ligation increased with each increasing course of COI, but the differences between subgroups were not significant.

    View this table:
    Table 2

    COI treatment subgroup analysis by number of courses, age of commencing treatment, age at PDA ligation and mortality at 1 year

    Postoperative variables following PDA ligation are summarised in table 3. The median time to treatment with postoperative corticosteroids following surgery was 17 days (IQR 9.5–23.5), with corticosteroid treatment being initiated in 33% of infants prior to extubation and 66% of infants after extubation. Postoperative complications of clinical importance occurred in 6.4% of patients (table 4).

    View this table:
    Table 3

    Postoperative outcomes

    View this table:
    Table 4

    Postoperative complications (first postoperative week)

    There were 6/125 deaths (4.8%) at ≤30 days and 16/125 deaths (12.8%) by 1 year after ligation. Early deaths occurred on days 3, 9, 10, 13, 15 and 16 postoperatively. The primary cause of death for nearly all cases (early and late) was bronchopulmonary dysplasia (BPD). The two exceptions were deaths secondary to necrotising enterocolitis (NEC) (day 3) and sepsis (day 16). Pneumothoraces, both left and right, complicated the postoperative course in two early deaths, but no deaths were attributed to direct surgical complications. Neurodevelopmental impairments were hearing loss (32%), cerebral palsy (25%), visual impairment (6.2%) and combined impairment (32%). There were no significant predictive associations with neurodevelopmental impairment, but ventilation preoperatively was significantly associated with the combined outcome of death and neurodisability (p=0.02).

    Logistic regression analysis for the outcome death by 1 year was undertaken on the population who were ventilated at referral. Only FiO2>40% (OR 8.7, 95% CI 1.8 to 42.4) and prior COI (OR 0.24, 95% CI 0.1 to 0.8) retained independent significance.

    Discussion

    The optimal timing and even the benefits of treating PDA remain controversial.1 The decision to refer patients for surgical PDA ligation can be particularly difficult because of a lack of clear evidence of long term benefits, the risk of surgical complications and concern about adverse long term outcomes.3 4 There is an increasing body of physiological evidence based on ultrasound studies5 and, more recently, near infrared spectroscopy6 that PDA adversely affects end-organ blood flow. However, relatively little has been published about the indications and benefits of surgical PDA ligation since the large trial by Gersony and colleagues in the 1980s7 reported that COIs had marginal advantages over surgery. There is conflicting evidence8 9 about the effect of very early surgical ligation on the incidence of NEC in very low birth weight babies. Surgical ligation has been proposed as the treatment of choice for infants <800 g with PDA.10 11 Lack of access to surgical PDA ligation has also been associated with a higher mortality in infants with persistent symptomatic PDA following treatment with COIs.12 Conversely, there are reports of an increased incidence of BPD and neurodisability in infants who have undergone duct ligation.3 4

    Population characteristics

    In our population, the infants referred were very premature and of extremely low birth weight. They were predominantly referred because of a continuing need for positive pressure ventilation, with a minority unable to wean from CPAP. All were defined as having a clinically significant PDA and the data on LA:Ao ratio are consistent with the presence of large shunts producing left ventricular overload, despite diuretic treatment in the majority of patients.

    Timing of surgery

    The timing of PDA ligation in this population is late compared with other series, where late ligation is defined as ligation beyond days 10–14 or day 21 from birth.9 13 14 Although the infants referred for primary PDA ligation were referred significantly earlier than those with prior COI treatment (26 vs 36 days), these cases would still be considered to be late ligations. Reasons for this late referral pattern in the present study are unclear. Our audit did not identify significant difficulties with access to the service, so it may reflect anxieties about the safety of PDA ligation or uncertainty about optimal management of PDA. The timing of treatment with COIs was also relatively late, with the median time to commencing treatment with even the first course of COIs at the margin of the suggested optimal therapeutic window.15 16 Multiple courses of treatment with COIs were also given in many cases, factors which may have contributed to substantially later PDA ligation.

    Surgical complications

    Overall, surgical PDA ligation appears to have been well tolerated, with postoperative times to extubation providing evidence of some short term benefit. The only direct iatrogenic complications were related to femoral arterial lines, an uncommon practice which has been discontinued, and two left sided pneumothoraces requiring treatment. In contrast to other studies,14 we did not find a frequent requirement for postoperative inotrope support, although such cardiorespiratory instability was associated with much earlier PDA ligation. The frequency of clinical stridor and confirmed left vocal cord paralysis in this study was low compared with published frequencies of 40–67%.17 18 This condition is likely to be underdiagnosed without routine nasal microlaryngoscopy. A high index of suspicion should be maintained in infants with severe lung disease and feeding difficulties following PDA ligation.18

    Neurodevelopmental outcomes

    The incidence of serious neurodisability and, in particular, hearing impairment in survivors at 1 year of age in this study was high and is almost certainly an underestimate. This has been noted in other large studies3 4 and may be due to the relative immaturity of infants typically referred for PDA ligation. Additionally, prolonged exposure to reductions in cerebral perfusion and ototoxic drugs while awaiting PDA ligation may have adverse neurological effects. Cerebral oxygenation may also be compromised during surgery19 and this effect and its impact on subsequent development require further study.

    Mortality, BPD and antecedent risk factors

    Mortality rates at 30 days postsurgery were low (4.8%) and consistent with average CCAD outcome for the UK, but increased substantially by 1 year (12.8%). Other reported UK mortality rates are 2%, 8% and 20% at 7 days, at 30 days and at hospital discharge, respectively.20 Death in our population was almost exclusively due to BPD and the incidence of severe BPD, as indicated by domiciliary oxygen requirement (46.8%), was high. Early deaths were associated with decreasing gestational age and FiO2>60% preoperatively, suggesting that infants referred with the most severe lung disease were least able to tolerate thoracotomy. Postoperative corticosteroid use was associated with an increased rate of death by 1 year, but only FiO2>40% at the time of surgery and lack of prior COI treatment (primary PDA ligation) retained independent significance in a multivariate model for death during the first postoperative year. The association between higher oxygen requirements preoperatively and death by 1 year suggests that, at the time of PDA ligation, these infants already had established lung disease. Lee and colleagues,19 in a similar UK series, report an association between multiple courses of COIs prior to PDA ligation and an increased incidence of chronic lung disease. A similar trend in our data, associated with later ligation, raises the possibility that earlier treatment could be advantageous.

    This protective association of prior COI treatment may be due to selection bias with the sickest babies referred for surgical ligation. We did not have illness severity scores to compare the two groups, but there was no significant difference in birth weights and gestational age between the two groups. The finding is supported by Madan and colleagues,4 who, in an evaluation of therapies for PDA using the American National Institute for Child Health database, noted that infants referred for ligation following COI treatment had significantly lower mortality than those referred for primary PDA ligation. A mechanism for this association might be partial constriction of the PDA with COI, resulting in reduced blood flow to the lungs and modulation of BPD development. Animal models using ventilated preterm baboons have demonstrated arrested alveolar development and abnormal lung mechanics after exposure to a ductal shunt for 14 days and that early ibuprofen-induced PDA closure prevented this alveolar arrest.21 Conversely, surgical PDA ligation may have detrimental effects. Infant and animal research studies suggest that there may be adverse effects on lung function and a decrease in alveolar fluid clearance following surgical PDA ligation.22,–,24 Despite these findings, our patients appeared to have evidence of improvement in respiratory outcomes.

    Conclusions

    The conclusions from this study are that surgical PDA ligation is well tolerated and associated with short term respiratory improvements. Our findings that a lack of prior COI treatment, and increasing preoperative FiO2 in ventilated infants are associated with an increasing risk of death, are biologically plausible. There is a need for ongoing research to improve perioperative management and to refine the optimal patient selection and timing for PDA ligation. In the interim, efforts should be focused on earlier identification of infants at risk of significant PDA, rationalising treatment with COIs and considering earlier staged referral for surgical PDA ligation6 where there are contraindications to using COI or the PDA fails to respond to COI therapy.

    Acknowledgments

    The authors wish to acknowledge the support of the Scottish Neonatal Consultants Group and the staff of the Department of Paediatric Cardiology and Cardio-thoracic Surgery at the Royal Hospital for Sick Children, Glasgow. They would particularly like to thank Sheena Mitchell for assistance with data collection.

    References

    1. ↵
      1. Bose CL,
      2. Laughon MM
      . Patent ductus arteriosus: lack of evidence for common treatments. Arch Dis Child Fetal Neonatal Ed 2007;92:F498–502.
      OpenUrlAbstract/FREE Full Text
    2. ↵
      1. Weffers B,
      2. Heuchan AM
      . A national survey of treatment of PDA in infants <30 weeks gestation. Proc BAPM Ann Meet 2006; page 2.
    3. ↵
      1. Kabra NS,
      2. Schmidt B,
      3. Roberts RS,
      4. et al
      .; Trial of Indomethacin Prophylaxis in Preterms Investigators. Neurosensory impairment after surgical closure of patent ductus arteriosus in extremely low birth weight infants: results from the Trial of Indomethacin Prophylaxis in Preterms. J Pediatr 2007;150:229–34, 234.e1.
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. Madan JC,
      2. Kendrick D,
      3. Hagadorn JI,
      4. et al
      .; National Institute of Child Health and Human Development Neonatal Research Network. Patent ductus arteriosus therapy: impact on neonatal and 18-month outcome. Pediatrics 2009;123:674–81.
      OpenUrlAbstract/FREE Full Text
    5. ↵
      1. Shimada S,
      2. Kasai T,
      3. Hoshi A,
      4. et al
      . Cardiocirculatory effects of patent ductus arteriosus in extremely low-birth-weight infants with respiratory distress syndrome. Pediatr Int 2003;45:255–62.
      OpenUrlCrossRefPubMedWeb of Science
    6. ↵
      1. Lemmers PM,
      2. Toet MC,
      3. van Bel F
      . Impact of patent ductus arteriosus and subsequent therapy with indomethacin on cerebral oxygenation in preterm infants. Pediatrics 2008;121:142–7.
      OpenUrlAbstract/FREE Full Text
    7. ↵
      1. Gersony WM,
      2. Peckham GJ,
      3. Ellison RC,
      4. et al
      . Effects of indomethacin in premature infants with patent ductus arteriosus: results of a national collaborative study. J Pediatr 1983;102:895–906.
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. Cassady G,
      2. Crouse DT,
      3. Kirklin JW,
      4. et al
      . A randomized, controlled trial of very early prophylactic ligation of the ductus arteriosus in babies who weighed 1000 g or less at birth. N Engl J Med 1989;320:1511–16.
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Jhaveri N,
      2. Moon-Grady A,
      3. Clyman RI
      . Early surgical ligation versus a conservative approach for management of patent ductus arteriosus that fails to close after indomethacin treatment. J Pediatr 2010;157:381–7, 387.e1.
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Palder SB,
      2. Schwartz MZ,
      3. Tyson KR,
      4. et al
      . Management of patent ductus arteriosus: a comparison of operative v pharmacologic treatment. J Pediatr Surg 1987;22:1171–4.
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Trus T,
      2. Winthrop AL,
      3. Pipe S,
      4. et al
      . Optimal management of patent ductus arteriosus in the neonate weighing less than 800 g. J Pediatr Surg 1993;28:1137–9.
      OpenUrlCrossRefPubMedWeb of Science
    12. ↵
      1. Brooks JM,
      2. Travadi JN,
      3. Patole SK,
      4. et al
      . Is surgical ligation of patent ductus arteriosus necessary? The Western Australian experience of conservative management. Arch Dis Child Fetal Neonatal Ed 2005;90:F235–9.
      OpenUrlAbstract/FREE Full Text
    13. ↵
      1. Hsiao CC,
      2. Wung JT,
      3. Tsao LY,
      4. et al
      . Early or late surgical ligation of medical refractory patent ductus arteriosus in premature infants. J Formos Med Assoc 2009;108:72–7.
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      1. Teixeira LS,
      2. Shivananda SP,
      3. Stephens D,
      4. et al
      . Postoperative cardiorespiratory instability following ligation of the preterm ductus arteriosus is related to early need for intervention. J Perinatol 2008;28:803–10.
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      1. Clyman RI
      . Mechanisms regulating the ductus arteriosus. Biol Neonate 2006;89:330–5.
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Clyman RI
      . Recommendations for the postnatal use of indomethacin: an analysis of four separate treatment strategies. J Pediatr 1996;128:601–7.
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Clement WA,
      2. El-Hakim H,
      3. Phillipos EZ,
      4. et al
      . Unilateral vocal cord paralysis following patent ductus arteriosus ligation in extremely low-birth-weight infants. Arch Otolaryngol Head Neck Surg 2008;134:28–33.
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Benjamin JR,
      2. Smith PB,
      3. Cotten CM,
      4. et al
      . Long-term morbidities associated with vocal cord paralysis after surgical closure of a patent ductus arteriosus in extremely low birth weight infants. J Perinatol 2010;30:408–13.
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      1. Lemmers PM,
      2. Molenschot MC,
      3. Evens J,
      4. et al
      . Is cerebral oxygen supply compromised in preterm infants undergoing surgical closure for patent ductus arteriosus? Arch Dis Child Fetal Neonatal Ed 2010;95:F429–34.
      OpenUrlAbstract/FREE Full Text
    20. ↵
      1. Lee LC,
      2. Tillett A,
      3. Tulloh R,
      4. et al
      . Outcome following patent ductus arteriosus ligation in premature infants: a retrospective cohort analysis. BMC Pediatr 2006;6:15.
      OpenUrlCrossRefPubMed
    21. ↵
      1. McCurnin D,
      2. Seidner S,
      3. Chang LY,
      4. et al
      . Ibuprofen-induced patent ductus arteriosus closure: physiologic, histologic, and biochemical effects on the premature lung. Pediatrics 2008;121:945–56.
      OpenUrlAbstract/FREE Full Text
    22. ↵
      1. Chang LY,
      2. McCurnin D,
      3. Yoder B,
      4. et al
      . Ductus arteriosus ligation and alveolar growth in preterm baboons with a patent ductus arteriosus. Pediatr Res 2008;63:299–302.
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      1. Waleh N,
      2. McCurnin DC,
      3. Yoder BA,
      4. et al
      . Patent ductus arteriosus ligation alters pulmonary gene expression in preterm baboons. Pediatr Res 2011;69:212–16.
      OpenUrlCrossRefPubMedWeb of Science
    24. ↵
      1. Dzukou T,
      2. Rakza T,
      3. Bouissou A,
      4. et al
      . Impaired lung function after intubation and mechanical ventilation for surgical ligation of the ductus arteriosus in preterm infants. J Cardiothorac Vasc Anesth 2011;25:1000–4.
      OpenUrlCrossRefPubMed

    Footnotes

    • Competing interests None.

    • Provenance and peer review Not commissioned; externally peer reviewed.