Near infrared spectroscopy in paediatric cardiac surgery - a review of local experience and national survey

Abstract

Introduction: Many centres have adopted Near infrared spectroscopy (NIRS) as a standard of care for adult cardiac surgery, despite limited evidence [1]. A search of the literature on the use of NIRS in paediatric cardiac surgery reveals mostly case reports and no randomised controlled trials. NIRS has been used for selected cases at Bristol Children's Hospital (BCH) for 2 years. We sought to establish firstly, whether there was evidence that NIRS had guided perioperative management at BCH by auditing current practice and, secondly, whether NIRS was used routinely in the other UK paediatric cardiac centres. Methods: For the first phase of the project, downloaded NIRS data, anaesthetic and bypass records covering a 10-week period at BCH were collated retrospectively. Periods of significant cerebral desaturation were mapped to contemporaneous notes and recordings to determine whether they related to specific perioperative events. In the second phase, a 10-question survey was e-mailed to a paediatric cardiac anaesthetist at all UK paediatric cardiac centres. A reminder email was sent 10 days later, if required and a second anaesthetist was contacted after a further week if there was still no response. Anaesthetists were asked questions concerning their peri- and-postoperative use of NIRS, how reliant they were on it, what indicators they used to signify cerebral hypoxia and whether they have used NIRS to guide therapy. Results: NIRS at Bristol Children's Hospital Twenty eight sets of records were analysed and, of these, two provided clear evidence of NIRS guiding perioperative management. In the first, concerning a 19-month-old, a Glenn shunt was performed on bypass, following a non-bypass attempt, aborted due to poor cerebral oxygen saturations. The second concerned low cerebral tissue saturation due to temporary obstruction of the superior vena cava by the cannula after commencing bypass in a 3-year-old. National Survey: We received 11 completed surveys from 13 centres. NIRS is used in 6/11 (55%) centres and, of these, three use it for all cases. Of the four that don't use NIRS, two plan to in the near future. All of those using NIRS routinely stated they would now be uncomfortable undertaking a case without it. Concerning those with experience of NIRS (7), cerebral hypoxia is determined by either a percentage drop (3), absolute values (1) or a combination (3). The cerebral saturation % for concern given ranged from 30% to 55%. All centres with NIRS also used it postoperatively (although half did so only occasionally) and 6/7 of those with experience of NIRS were aware of situations where NIRS had significantly altered patient management. Discussion: There is a large disparity in the use of NIRS in paediatric cardiac surgery in the UK with around half of the centres surveyed using it. Furthermore, of those that do use NIRS, there is a lack of consensus between centres on the most accurate mechanisms and thresholds used to identify cerebral hypoxia. This is probably related to the lack of clarity in the literature of the benefits of NIRS in this patient group. Our local review illustrated two instances, over 10 weeks, where NIRS was valuable in directing patient management. Furthermore, the vast majority of those surveyed with experience of NIRS had also used it to guide management at some point. There are some limitations to this study. Our local note review was retrospective and relied on accurate reporting by the anaesthetist at the time. If anything, however, this would result in the under-reporting of events. We acknowledge that we only surveyed one anaesthetist in each centre and their answers may not reflect those of all of their colleagues. There were also two centres that we were unable to get a reply from. Conclusion: We have found two local examples of perioperative management guided by NIRS. We have further identified, through the survey, evidence of the inconsistency and deficiency of standard in the use of NIRS in the UK. Conflict of interest: The authors declare no conflict of interest in this study.