Clinical practice guidelines for neonatal therapeutic hypothermia across European Society for Paediatric and Neonatal Intensive Care (ESPNIC) member countries: a multidisciplinary blind comparison

Background Therapeutic hypothermia (TH) decreases mortality and neuro-developmental disability in neonates greater or equal than 36 weeks’ gestation with moderate-to-severe hypoxic-ischaemic encephalopathy (HIE) due to perinatal asphyxia. TH is often applied in different ways, as national guidelines and target populations may significantly differ. We aimed to compare TH guidelines from various European countries as we hypothesised that guidelines are significantly different and often of insufficient quality. Methods Multidisciplinary, blind, critical appraisal of clinical practice guidelines using AGREE-II and RIGHT tools to assess their quality and completeness of the reporting, respectively. The appraisal was performed by neonatologists, obstetricians, paediatric and adult intensivists. The raters were kept unaware of the identity of the other panellists and each of them performed the evaluation of the guidelines independently. Findings Total AGREE-II score was different amongst the guidelines (p = 0.008): the German guidelines scored highest (5.2 (95% CI: 4.7–5.7), while the others had comparable intermediate scores (approximately between 2.2 and 4.4). A good score in the rigour of development was not achieved by any guideline (German guidelines scored “moderate”). The overall completeness of reporting was different amongst the guidelines, with total completeness ranging between 20% and 49% (p < 0.0001). Completeness of reporting was incomplete for most rater-item responses. German and French guidelines had the best domain-based reporting completeness for AGREE-II and RIGHT, respectively. Inter-rater agreement was satisfactory (intraclass correlation coefficient varying between 0.83 and 0.91; p < 0.001). Interpretation European guidelines for neonatal TH have a general insufficient quality and completeness of reporting.