322 - Retinopathy of Prematurity is Associated with Exposure to Hemodynamically Significant Ductal Shunt

Poster Board #: 322
Publication #: 3906-HP-QA.322

Background: Hemodynamically significant patent ductus arteriosus (hsPDA) shunt may predispose extremely preterm neonates to retinopathy of prematurity (ROP) because pulmonary recirculation may lead to higher pre-ductal cardiac output and blood oxygen content, which may augment ocular oxygen delivery.


Objective: We hypothesize that extremely preterm neonates with severe and/or prolonged exposure to hsPDA are at increased risk of the composite of death <32 weeks postmenstrual age (PMA) or severe ROP.


Design/Methods: A retrospective cohort study of preterms born ≤26⁺⁶ weeks gestation and admitted at <24h postnatal age to University of Iowa Stead Family Children’s Hospital from 09/2018-07/2020 was conducted. Patients were considered exposed to hsPDA if any Targeted Neonatal Echocardiogram (TnECHO) demonstrated Iowa PDA score ≥6 [Table 1] at <32 weeks PMA vs. controls with ‘no hsPDA’. The primary outcome was death<32 weeks or severe ROP [≥stage 2+] in either eye. Univariate analysis of patient characteristics and outcomes was performed. Logistic regression was used to adjust ROP risk for factors with p<0.05. Magnitude of PDA exposure was calculated [mean PDA score*days exposed] and a receiver operator characteristics (ROC) curve was generated for the outcome of ROP≥stage 2+.


Results: 91 patients were screened, of whom 86 [n=54 hsPDA, n=32 controls] were eligible. hsPDA patients were younger and lighter at birth and had a higher burden of hyperglycemia and respiratory illness. More hsPDA infants received hydrocortisone.[Table 2] The composite outcome (p<0.001) and any ROP (p=0.002) was more frequent in hsPDA group and there was 4x more ROP intervention following hsPDA.[Table 2] After adjustment, hsPDA remained independently associated with the composite outcome [OR 27.6] as did hyperglycemia and respiratory severity. hsPDA shunt exposure was independently associated with development of any ROP among survivors to assessment (p=0.004).[Table 3] Among preterms<24weeks GA, shunt exposure predicted ≥stage 2+ ROP with AUC=0.79. PDA exposure score of 78 [clinical equivalent=7 days high volume shunt] predicts stage 2+ ROP [sensitivity=80%, specificity=78%].[Fig 1]

Conclusion(s): Among infants born 22-26⁺⁶ weeks, hsPDA shunt is associated with a composite of death or severe ROP, as well as ROP of any stage. Shunt modulation as a strategy to reduce ROP represents a biologically plausible avenue for investigation.


Table 1: Iowa PDA score


Table 2: Demographic Data and Neonatal Outcomes


Table 3: Logistic Regression to adjust risk of death or severe ROP (model 1) and Any ROP (model 2) for PDA shunt, hyperglycemia and respiratory severity.


Figure 1: Receiver operator characteristics curve for shunt exposure to predict the outcome of ROP stage 2+ or greater among patients born at 22-23⁺⁶ weeks GA.

---

Authors/Institutions: Alison Cunningham, The University of Iowa, Iowa City, Iowa, United States; Madeline Beauchene, The University of Iowa, Iowa City, Iowa, United States; Danielle R. Rios, The University of Iowa, Iowa City, Iowa, United States; Jonathan M. Klein, The University of Iowa, Iowa City, Iowa, United States; Amy Stanford, The University of Iowa, Iowa City, Iowa, United States; Megan Fellows, The University of Iowa, Iowa City, Iowa, United States; John M. Dagle, The University of Iowa, Iowa City, Iowa, United States; Patrick J. McNamara, The University of Iowa, Iowa City, Iowa, United States; Regan E. Giesinger, The University of Iowa, Iowa City, Iowa, United States