Skip to main content
PMC home page
HRB Open Research logoLink to HRB Open Research
. 2023 Oct 20;6:63. [Version 1] doi: 10.12688/hrbopenres.13802.1

Association between intrapartum fetal pulse oximetry and adverse perinatal and long-term outcomes: a systematic review and meta-analysis protocol

Jill M Mitchell 1,2,a, Siobhan Walsh 2, Laura J O'Byrne 1,2, Virginia Conrick 3, Ray Burke 4, Ali S Khashan 1,5, John Higgins 2, Richard Greene 2,6, Gillian M Maher 1,5, Fergus P McCarthy 1,2
PMCID: PMC11019289  PMID: 38628596

Abstract

Background: Current methods of intrapartum fetal monitoring based on heart rate, increase the rates of operative delivery but do not prevent or accurately detect fetal hypoxic brain injury. There is a need for more accurate methods of intrapartum fetal surveillance that will decrease the incidence of adverse perinatal and long-term neurodevelopmental outcomes while maintaining the lowest possible rate of obstetric intervention. Fetal pulse oximetry (FPO) is a technology that may contribute to improved intrapartum fetal wellbeing evaluation by providing a non-invasive measurement of fetal oxygenation status.

Objective: This systematic review and meta-analysis aims to synthesise the evidence examining the association between intrapartum fetal oxygen saturation levels and adverse perinatal and long-term outcomes in the offspring.

Methods: We will include randomised control trials (RCTs), cohort, cross-sectional and case-control studies which examine the use of FPO during labour as a means of measuring intrapartum fetal oxygen saturation and assess its effectiveness at detecting adverse perinatal and long-term outcomes compared to existing intrapartum surveillance methods. A detailed systematic search of PubMed, EMBASE, CINAHL, The Cochrane Library and Web of Science will be conducted following a detailed search strategy until August 2023. Three authors will independently review titles, abstracts and full text of articles. Two reviewers will independently extract data using pre-defined data extraction and assess the quality using the Risk of Bias tool for RCTs and Newcastle-Ottawa Scale for observational studies. We will use random-effects meta-analysis for each exposure-outcome association to calculate pooled estimates using the generic variance method. This systematic review will follow the Preferred Reporting Items for Systematic reviews and Meta-analyses guidelines.

PROSPERO registration: CRD42023457368 (04/09/2023)

Keywords: Labour, intrapartum, fetal monitoring, oximetry, oxygen saturation, blood gas monitoring, Sp02

Introduction

Intrapartum fetal monitoring aims to improve perinatal outcome while avoiding unnecessary operative interventions 1 . The gold standard for assessment of fetal well-being continues to be the auscultation of the fetal heart, along with the interpretation of alterations in the fetal heart rate pattern, as demonstrated through cardiotocography (CTG) to try to predict babies at risk of hypoxic brain injury and adverse perinatal outcomes. Despite its status as the established benchmark in care, employed in approximately 90% of births in the United States (US) and on a global scale, there is a pervasive consensus that current fetal monitoring devices based on heart rate, do not prevent or accurately detect fetal hypoxic brain injury 2– 7 . The current use of cardiotocography (CTG) as a method of monitoring intrapartum fetal well‐being during labour is associated with an increased caesarean section rate, compared with intermittent auscultation of the fetal heart rate, resulting in a reduction in neonatal seizures, although no differences in other neonatal outcomes 8 . CTG is complicated by significant inter- and intra-observer variation 2, 9, 10 . Furthermore, CTG demonstrates a low positive predictive value for fetal hypoxia, meaning that among those fetuses that CTG indicates are at risk for hypoxia, a smaller proportion will truly be hypoxic 2, 9, 10 . This can lead to an increased frequency of false-positive results, thus potentially prompting unnecessary interventions based on an overestimated risk of fetal hypoxia. Consequently, such misinterpretations could contribute to the escalating rates of caesarean deliveries observed worldwide. For instance, in the US, the caesarean delivery rate has surged, increasing from 20.7% in 1996 to 32.1% in 2021 11 . Similarly in the United Kingdom, the caesarean delivery rate has increased from 14.7% from 1990 to 1999 to 35.21% in 2021–2022 12, 13 .

Fetal blood sampling (FBS) is widely used as a complementary tool to improve the specificity and sensitivity of CTG 14 . FBS has been shown to reduce operative vaginal delivery rates without affecting neonatal outcomes but it is a complex, invasive procedure 2 . There is a need for more accurate methods of intrapartum fetal surveillance that will decrease the incidence of adverse neonatal and long-term neurodevelopmental outcomes while maintaining the lowest possible rate of obstetric intervention.

Fetal pulse oximetry (FPO) is a technology that may contribute to improved intrapartum fetal wellbeing evaluation by providing a non-invasive measurement of fetal oxygenation status 15– 18 . FPO offers a potential dual advantage over traditional fetal heart rate monitoring. It quantifies the percentage of oxygenated haemoglobin directly, thereby assessing fetal oxygenation, which is pivotal in mediating the harmful consequences of hypoxia/ischaemia. Additionally, it utilises a well-established, non-invasive technology, regarded as safe and broadly implemented in all modern intensive care units and operating theatres 8 . Based on data from both human and animal studies, average intrapartum fetal oxygen saturation (FSp02) range from 35% to 65% 19– 21 . FSpO2 levels of 30% or higher are generally considered reassuring for the human fetus. However, if FSpO2 levels are less than 30% for a duration of 10 minutes or more, additional evaluation or intervention is warranted 17, 19, 22– 24 A study by McNamara et al. (n=100) demonstrated that babies born in poor condition had abnormal fetal oximetry values 25 . FPO is advantageous over FBS in that it is a non-invasive and continuous monitoring technique.

A Cochrane Review published in 2014 compared fetal intrapartum pulse oximetry with other fetal surveillance techniques 8 . This review included seven trials reporting on a total of 8013 pregnancies. The primary outcomes were caesarean section, hypoxic‐ischaemic encephalopathy, neonatal seizures and long‐term neurodevelopmental outcomes. The authors found no significant differences in the overall caesarean delivery rate between those monitored with FPO and those not monitored with FPO or for whom the FPO results were not displayed to the clinician or woman (four studies, n = 4008, risk ratio (RR) 0.99, 95% confidence intervals (CI) 0.86 to 1.13, I 2 = 45). The authors noted a reduction in the number of caesarean sections performed for cases of non-reassuring fetal status in the FPO plus CTG group compared to the CTG only group in two of the four analyses: firstly when considering pregnancies at or beyond 34 weeks where fetal FBS was not required prior to study entry (comprising four studies with 4008 participants, RR 0.65, 95% CI 0.46 to 0.90, I 2 = 63) Secondly, in situations where FBS was a prerequisite before study participation (one study involving 146 participants), the RR was notably low at 0.03, with a 95% confidence interval spanning from 0.00 to 0.44. Additionally, the review reported a reduction in operative births (comprising caesarean sections or operative vaginal births) for non-reassuring fetal status when FPO was combined with CTG monitoring, compared to CTG monitoring alone. This finding was consistent across two studies involving 1610 participants, with a RR of 0.74 (95% confidence interval: 0.62 to 0.89). However, no statistically significant differences were observed in several other outcome measures, including Apgar scores less than four at five minutes or less than seven at five minutes, umbilical arterial pH less than 7.10, neonatal intensive care unit (NICU) admissions, length of hospital stays, mortality, or fetal skin trauma, when comparing the use of FPO in conjunction with CTG to fetal electrocardiography combined with CTG. The results of this review may have been influenced by a large randomised control trial (RCT) by Bloom et al. 26 which had several limitations. Namely, that they did not describe the number of caesarean births indicated by the FPO results, nor did they describe counter-measures, including posture change according to the fetal oxygen saturation values, administration of tocolytic agents, and expedited delivery. Furthermore, observational studies were not included in the previous systematic review. In contrast, our planned review will take a more comprehensive approach, incorporating evidence not just from RCTs, but also from cohort studies, case-control studies, and observational studies.

Uchida et al. conducted a narrative review on FPO as a measure of intrapartum fetal condition. They concluded that FPO with fetal heart rate monitoring in selected cases of non-reassuring fetal status may reduce the caesarean section rate 27 . While Uchida and colleagues discussed 31 studies and seven RCTs in their review of previous literature, they did not conduct a systematic review of previous literature. Therefore, we aimed to synthesise the previous literature examining the association between intrapartum fetal oxygen saturation and perinatal and long-term outcomes in the offspring in the form of a systematic review and meta-analysis.

This review has been registered with PROSPERO (CRD42023457368) on 4 th September 2023 and follows the PRISMA-P guidelines 28 .

Review question

Does low intrapartum fetal oxygen saturation as measured by fetal pulse oximetry increase the risk of adverse perinatal and long-term neurodevelopmental outcomes?

Does the addition of fetal pulse oximetry to established forms of intrapartum fetal monitoring such as cardiotocographs, improve perinatal and long-term neurodevelopmental outcomes?

Methods

Eligibility criteria

The following PICO criteria will guide this systematic review.

Population

Women in labour with a cephalic, singleton live baby.

Intervention

The use of fetal pulse oximetry during labour as a means of measuring intrapartum fetal oxygen saturation or low intrapartum fetal oxygen saturation in labour

Comparison

We will compare the perinatal and long-term neurodevelopmental outcomes of offspring who had low oxygen saturations in labour versus those who had normal oxygen saturations in labour as measured by FPO or the use of FPO during labour as an adjunctive method of intrapartum surveillance compared to conventional fetal monitoring e.g. fetal heart rate monitoring by intermittent/continuous cardiotocography or the use of fetal scalp electrode, fetal blood sampling or electrocardiogram.

Outcomes

Any measure of compromise of neonatal or childhood wellbeing - such as mortality; neonatal morbidity, including cord blood acidaemia, low APGAR scores, hypoxic ischaemic encephalopathy, intraventricular haemorrhage, seizures, periventricular leukomalacia, admission to the NICU and long term neurodevelopmental outcomes including cerebral palsy and developmental delay. See Table 1 for primary and secondary outcome measures.

Table 1. Review Outcome Measures.

Primary Outcomes Perinatal Outcomes Cord Blood acidaemia
Abnormal Arterial or Venous Base Excess
Abnormal Umbilical arterial or venous lactate levels
Low 5 minute APGAR Score
Admission to NICU
Neonatal seizures
Neonatal or intrapartum death
Cardiopulmonary resuscitation or intubation
required within 24 hours of life
Hypoxic Ischaemic Encephalopathy
Secondary Outcomes Low umbilical vein oxygen saturation
Low umbilical artery oxygen saturation
Operative Delivery for Non-Reassuring Fetal Status
Operative Delivery for Dystocia
Abnormal intrapartum fetal blood sample: scalp
pH
Abnormal intrapartum fetal blood sample: scalp lactate
Long Term Outcomes Cerebral Palsy
Developmental Delay
Open in a new tab

Studies

Randomised control trials and observational studies including cohort, case-control and cross-sectional studies

We will not exclude studies based on time frame or language. We will include published peer-reviewed studies only.

Review exclusion criteria

  • 1.

    Studies only available in abstract form.

  • 2.

    Non-human studies.

  • 3.

    Review articles, case reports, case series.

  • 4.

    Conference proceedings, letters, commentaries, notes, editorials, dissertations.

Literature search

We will use a two-part search strategy to identify studies meeting the inclusion criteria: (1) we will search electronic bibliographic databases for published work, using a comprehensive search strategy for fetal intrapartum pulse oximetry and perinatal and long-term outcomes; (2) we will hand-search the reference lists of studies included in the review and the reference lists of relevant, previously published reviews. The following electronic bibliographic databases will be searched: PubMed, EMBASE, CINAHL, The Cochrane Library and Web of Science. The search strategies for all databases can be found as Extended data 29 .

Study screening and selection

Titles and abstracts of the studies retrieved from each database search will be stored and managed in the Endnote reference manager and de-duplicated. Three independent reviewers (JM, SW, LOB) will screen all titles and abstracts. Full texts will be obtained where necessary to screen for eligibility. Where consensus on eligibility cannot be achieved, a fourth review author (FMcC) will be involved in the discussion.

Data extraction and management

Two review authors (JM, SW) will independently extract data and discrepancies will be identified and resolved through discussion with a third author (FMcC), where necessary. A standardised, pre-piloted data extraction form will be used to extract data from the included studies. We will extract data including the author and year of publication, study design, country and setting of study, sample size, definition and or assessment of the exposures and outcome(s) of interest, comparison group, length of follow up, confounders adjusted for (if any), crude and adjusted estimates. If additional data is required from an eligible study, the corresponding author will be contacted via email. A reminder email will be sent two weeks later if the corresponding author does not reply.

Quality appraisal of included studies

Articles which meet the selection criteria will be assessed for methodological quality independently by two reviewers using the Risk of Bias tool 30 for randomised controlled trials (RCT) and the Newcastle Ottawa Scale 31 for observational studies. Disagreements between the review authors over the quality assessment of each study will be resolved by discussion, with involvement of a third review author where necessary.

Data synthesis, including assessment of heterogeneity

We will undertake separate meta-analyses for RCTs and observational studies using RevMan 5.4. We will also perform separate meta-analyses for each exposure-outcome associations. For example, low intrapartum fetal oxygen saturation and cord blood acidaemia, the addition of FPO as a monitoring method and cord blood acidaemia, low intrapartum fetal oxygen saturation and low 5 minute Apgar score and low intrapartum fetal oxygen saturation and admission to NICU. Heterogeneity will be assessed statistically using the I 2 statistic and also explored using subgroup analyses based on the different study designs included in this review. We will perform the subgroup/sensitivity analysis where the data allow, according to the study design (RCT, cohort, case-control and cross-sectional) and study quality/risk of bias (minimal/low versus moderate/high). We will perform a subgroup analysis using different cut-off values of cord blood acidaemia and FPO if different cut-offs are used in studies. Random effects meta-analyses will be performed to calculate overall pooled estimates where data allow. We will use the generic inverse variance method to display crude and adjusted results where possible. First, we will conduct a meta-analysis of all crude estimates for each exposure-outcome association. We will then conduct a meta-analysis of all adjusted estimates for each exposure-outcome association. We will consider any adjusted estimate as adjusted regardless of the variables adjusted for. When a meta-analysis cannot be conducted because of lack of data, a narrative synthesis of the results will be included.

The presence of publication bias will be evaluated using a funnel plot, provided a minimum inclusion of 10 studies or more in the meta-analysis. In instances where additional subgroup/sensitivity analyses are found within the meta-analysis, such as examinations to investigate potential high heterogeneity, these will be labelled as post-hoc analyses.

Presenting and reporting the results

A PRISMA flow diagram will be incorporated to detail the sequential process of study selection, along with explanations for any studies excluded during the full-text review phase. Study characteristics and quality assessment of included studies will be displayed in tables, while pooled estimates will be presented using forest plots. Where data which is unsuitable for meta-analysis, results will be narratively synthesised.

Conclusions

The systematic review and meta-analysis will summarise the existing literature investigating the association between intrapartum fetal oxygen saturation and adverse perinatal and long term outcomes in offspring. This review is of considerable importance as it explores the potential utility of fetal pulse oximetry as a method for intrapartum fetal monitoring. There is a pressing need for innovative and reliable approaches to monitor fetal well-being during labour, and this review could provide pivotal insights in this regard.

Potential strengths and limitations of this study

The robustness of this review is bolstered by the implementation of a thorough search strategy, a prospectively registered protocol, and strict compliance with PRISMA guidelines. Additionally, the engagement of three reviewers in the process of eligibility screening and two reviewers in the process of data extraction, and quality assessment of the included studies serves to substantially mitigate the potential for reviewer-based bias in the systematic review. Furthermore, this review will not have language restrictions, reducing the risk that relevant indexed studies be overlooked. In the scope of this review, we will only incorporate studies that have been formally published. this may render our review susceptible to publication bias, as studies with significant or positive results are often likely to be published 32 . If possible, a funnel plot will be used to assess the presence of publication bias. Moreover, the existence of confounding variables poses a significant challenge in observational research. Possible confounders might encompass the age of the mother, parity, maternal body mass index, heterogenous clinical approaches, different methods of monitoring FSp02 and pregnancy complications such as intrauterine growth restriction, pre-eclampsia and gestational diabetes mellitus. As noted previously, our meta-analyses will present both unadjusted and adjusted outcomes, whenever feasible, using the generic inverse variance approach. This adjustment will be based on the definitions provided in each of the studies we've reviewed.

Dissemination

It is anticipated that findings of this review will be disseminated through publication in a peer-reviewed journal and presented at scientific conferences.

Study status

Not commenced yet.

Funding Statement

The author(s) declared that no grants were involved in supporting this work.

[version 1; peer review: 1 approved, 1 approved with reservations]

Data availability

Underlying data

No data are associated with this article.

Extended data

Fighsare: Search Strategy - Association between Intrapartum Fetal Pulse Oximetry and Adverse Perinatal and Long-term Outcomes- a Systematic Review and Meta-analysis Protocol.docx. https://doi.org/10.6084/m9.figshare.24049890.v3 29 .

Reporting guidelines

Fighsare: PRISMA-P checklist for ‘Association between intrapartum fetal pulse oximetry and adverse perinatal and long-term outcomes: a systematic review and meta-analysis protocol. https://doi.org/10.6084/m9.figshare.24049899 28 .

Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).

References

  • 1. Chandraharan E, Wiberg N: Fetal scalp blood sampling during labor: an appraisal of the physiological basis and scientific evidence. Acta Obstet Gynecol Scand. 2014;93(6):544–547. 10.1111/aogs.12416 [DOI] [PubMed] [Google Scholar]
  • 2. Alfirevic Z, Devane D, Gyte G: Continuous cardiotocography (CTG) as a form of electronic fetal monitoring (EFM) for fetal assessment during labour. Cochrane Database Syst Rev. 2006; (3): CD006066. 10.1002/14651858.CD006066 [DOI] [PubMed] [Google Scholar]
  • 3. Vintzileos AM, Antsaklis A, Varvarigos I, et al. : A randomized trial of intrapartum electronic fetal heart rate monitoring versus intermittent auscultation. Obstet Gynecol. 1993;81(6):899–907. [PubMed] [Google Scholar]
  • 4. MacDonald D, Grant A, Sheridan-Pereira M, et al. : The Dublin randomized controlled trial of intrapartum fetal heart rate monitoring. Am J Obstet Gynecol. 1985;152(5):524–539. 10.1016/0002-9378(85)90619-2 [DOI] [PubMed] [Google Scholar]
  • 5. Kelso IM, Parsons RJ, Lawrence GF, et al. : An assessment of continuous fetal heart rate monitoring in labor. A randomized trial. Am J Obstet Gynecol. 1978;131(5):526–532. 10.1016/0002-9378(78)90114-x [DOI] [PubMed] [Google Scholar]
  • 6. Grant A, O'Brien N, Joy MT, et al. : Cerebral palsy among children born during the Dublin randomised trial of intrapartum monitoring. Lancet. 1989;2(8674):1233–1236. 10.1016/s0140-6736(89)91848-5 [DOI] [PubMed] [Google Scholar]
  • 7. Grivell RM, Alfirevic Z, Gyte GM, et al. : Antenatal cardiotocography for fetal assessment. Cochrane Database Syst Rev. 2015;2015(9): Cd007863. 10.1002/14651858.CD007863.pub4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. East CE, Begg L, Colditz PB, et al. : Fetal pulse oximetry for fetal assessment in labour. Cochrane Database Syst Rev. 2014;2014(10): Cd004075. 10.1002/14651858.CD004075.pub4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Nonnenmacher A, Hopp H, Dudenhausen J: Predictive value of pulse oximetry for the development of fetal acidosis. J Perinat Med. 2010;38(1):83–86. 10.1515/jpm.2010.006 [DOI] [PubMed] [Google Scholar]
  • 10. Rhöse S, Heinis A, Vandenbussche F, et al. : Inter- and intra-observer agreement of non-reassuring cardiotocography analysis and subsequent clinical management. Acta Obstet Gynecol Scand. 2014;93(6):596–602. 10.1111/aogs.12371 [DOI] [PubMed] [Google Scholar]
  • 11. Osterman MJK, Hamilton BE, Martin JA, et al. : Births: Final Data for 2021.In: Statistics DoV. (ed.). Natl Vital Stat Rep. 2023;72(1):1–53. [PubMed] [Google Scholar]
  • 12. Black C, Kaye JA, Jick H: Cesarean Delivery in the United Kingdom: Time Trends in the General Practice Research Database. Obstet Gynecol. 2005;106(1):151–155. 10.1097/01.AOG.0000160429.22836.c0 [DOI] [PubMed] [Google Scholar]
  • 13. NHS Maternity Statistics, England - 2021-22.2022. Reference Source
  • 14. Jørgensen J, Weber T: Fetal scalp blood sampling in labor - a review. Acta Obstetricia et Gynecologica Scandinavica. 2014;93(6):548–555. 10.1111/aogs.12421 [DOI] [PubMed] [Google Scholar]
  • 15. Yam J, Selina C, Arulkumaran S: Intrapartum Fetal Pulse Oximetry. Part I: Principles and Technical Issues. Obstet Gynecol Surv. 2000;55(3):163–172. 10.1097/00006254-200003000-00025 [DOI] [PubMed] [Google Scholar]
  • 16. Colditz P, Begg L, East C: Fetal pulse oximetry: instrumentation and recent clinical experience. Clin Perinatol. 1999;26(4):869–880. 10.1016/S0095-5108(18)30024-1 [DOI] [PubMed] [Google Scholar]
  • 17. Kuhnert M, Seelbach-Gobel B, Di Renzo G, et al. : Guidelines for the use of fetal pulse oximetry during labor and delivery. Prenat Neonatal Med. 1999;3(4):432–433. [Google Scholar]
  • 18. East CE, Colditz PB, Begg LM, et al. : Update on intrapartum fetal pulse oximetry. Aust N Z J Obstet Gynaecol. 2002;42(2):119–124. 10.1111/j.0004-8666.2002.00119.x [DOI] [PubMed] [Google Scholar]
  • 19. Kuhnert M, Seelbach-Gobel B, Butterwegge M: Predictive agreement between the fetal arterial oxygen saturation and fetal scalp pH: results of the German multicenter study. Am J Obstet Gynecol. 1998;178(2):330–335. 10.1016/s0002-9378(98)80021-5 [DOI] [PubMed] [Google Scholar]
  • 20. Seelbach-Gobel B, Heupel M, Kuhnert M, et al. : The prediction of fetal acidosis by means of intrapartum fetal pulse oximetry. Am J Obstet Gynecol. 1999;180(1 Pt 1):73–81. 10.1016/s0002-9378(99)70153-5 [DOI] [PubMed] [Google Scholar]
  • 21. Carbonne B, Langer B, Goffinet F, et al. : Clinical importance of fetal pulse oximetry. II. Comparative predictive values of oximetry and scalp pH. Multicenter study. J Gynecol Obstet Biol Reprod (Paris). 1999;28(2):137–144. [PubMed] [Google Scholar]
  • 22. Nijland R, Jongsma HW, Crevels J, et al. : The ductus arteriosus, pre- and post-ductal oxygen saturation measurements in fetal lambs. Eur J Obstet Gynecol Reprod Biol. 1994;55(2):135–140. 10.1016/0028-2243(94)90068-x [DOI] [PubMed] [Google Scholar]
  • 23. Nijland R, Jongsma HW, Nijhuis JG, et al. : Arterial oxygen saturation in relation to metabolic acidosis in fetal lambs. Am J Obstet Gynecol. 1995;172(3):810–819. 10.1016/0002-9378(95)90004-7 [DOI] [PubMed] [Google Scholar]
  • 24. Dildy GA, Thorp JA, Yeast JD, et al. : The relationship between oxygen saturation and pH in umbilical blood: implications for intrapartum fetal oxygen saturation monitoring. Am J Obstet Gynecol. 1996;175(3 Pt 1):682–687. 10.1053/ob.1996.v175.a74922 [DOI] [PubMed] [Google Scholar]
  • 25. Mc Namara H, Johnson N: Fetal monitoring by pulse oximetry and CTG. J Perinat Med. 1993;21(6):475–480. 10.1515/jpme.1993.21.6.475 [DOI] [PubMed] [Google Scholar]
  • 26. Bloom SL, Spong CY, Thom E, et al. : Fetal pulse oximetry and cesarean delivery. N Engl J Med. 2006;355(21):2195–2202. 10.1056/NEJMoa061170 [DOI] [PubMed] [Google Scholar]
  • 27. Uchida T, Kanayama N, Kawai K, et al. : Reevaluation of intrapartum fetal monitoring using fetal oximetry: A review. J Obstet Gynaecol Res. 2018;44(12):2127–2134. 10.1111/jog.13761 [DOI] [PubMed] [Google Scholar]
  • 28. Mitchell J, Walsh S, O'Byrne L, et al. : Preferred Reporting Items for Systematic review and Meta-analysis protocol (PRISMA-P) checklist and flow diagram.docx. figshare. Dataset.2023. 10.6084/m9.figshare.24049899.v2 [DOI]
  • 29. Mitchell J, Walsh S, O'Byrne L, et al. : Search Strategy - Association between Intrapartum Fetal Pulse Oximetry and Adverse Perinatal and Long-term Outcomes- a Systematic Review and Meta-analysis Protocol.docx. figshare. Dataset.2023. 10.6084/m9.figshare.24049890.v3 [DOI] [PMC free article] [PubMed]
  • 30. Higgins J, Altman D, Gøtzsche P, et al. : The Cochrane Collaboration's tool for assessing risk of bias in randomised trials. BMJ. 2011;343: d5928. 10.1136/bmj.d5928 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Wells GA, Shea B, O'Connell D, et al. : The Newcastle-Ottawa Scale (NOS) for Assessing the Quality of Nonrandomised Studies in Meta-Analyses.2014. Reference Source
  • 32. Nair AS: Publication bias - Importance of studies with negative results! Indian J Anaesth. 2019;63(6):505–507. 10.4103/ija.IJA_142_19 [DOI] [PMC free article] [PubMed] [Google Scholar]
HRB Open Res. 2023 Nov 27. doi: 10.21956/hrbopenres.15101.r37131

Reviewer response for version 1

James X Sotiropoulos 1, Sol Libesman 1

We thank you for the opportunity to review this study protocol. 

We commend the authors for their practice of open science and publishing the protocol at this stage with pre-registration following PRISMA-P guidelines. We will focus mainly on the methods in this review, as it is most aligned with my experience, and provide some (hopefully) constructive feedback. 

Introduction 

The introduction is clear and describes the need for this study – with a recent narrative review without systematic review and meta-analysis indicating potential findings.  

Methods 

Note Table 1 formatting appears odd on the online system – please check 

Are there any inclusion/exclusion criteria beyond cephalic, singleton live baby? 

The authors' inclusion/exclusion criteria is singleton pregnancies only. Is there a good justification for this? If possible, it may provide future utility in collecting data for multiples, even if only reporting it in stratified analysis separately. If FPO is not feasible/reliable in multiple pregnancies this should be clearly stated and cited. 

The authors could consider further specifying the outcomes. For example, the cut-off for “low”, and “abnormal” are not provided. Could you specify how developmental delay will be measured including a cut-off (if this has been established)? Could you also specify the time in which these outcomes are measured (e.g., the primary endpoints), the measurement units (if relevant), and the relative effects that will be used to compare the groups? 

In the section “comparator” you mention “We will compare …offspring who had low oxygen saturations in labour versus those who had normal oxygen saturations in labour as measured by…”. Could you specify the cut-off you plan to use for low?  For further guidance on the description of treatments and outcomes, the ‘estimand’ framework may help you clarify elements of treatment, population, outcome, population level summaries, and more. 

There are two review questions nominated, but the comparator section only appears to address the first review question.

The authors plan to use I 2 as a measure of heterogeneity. I 2 is a flawed measure of heterogeneity, although it permeates systematic review and meta-analysis across all disciplines. Please see this article (https://doi.org/10.1002/jrsm.1230, 1 ). Will the authors consider, instead or in addition to I 2, graphically representing heterogeneity with prediction intervals or another measure of between-study variance? The Cochrane Handbook provides some useful guidance. 

The authors have not included an assessment of the overall certainty of evidence (e.g., GRADE). This has also been marked as “NA” in the PRISMA-P document. I strongly recommend the use of GRADE to assess the overall certainty of evidence, or if it is to be omitted for a valid reason that this be explained in the methods and/or discussion. Alternatively, if it is omitted a clear plan as to how the evidence will be evaluated should be provided. 

The authors have noted the potential impact of publication bias and plan to use a funnel plot. They may consider a contour-enhanced funnel plot, PET-PEESE, or p-curve analysis to examine publication bias. They might also consider a search of clinical trial registers to look for other registered studies that may be unpublished but have useable data for inclusion in meta-analysis. Guidance for searching clinical trial registers is here https://doi.org/10.1136/bmj-2021-068791  2 .

The authors intend to search only until August 2023. By the time the results of this study are ready for publication the search may be outdated. I note the PROSPERO registration says the search will begin in 2024. This may be a typo. Otherwise, I suggest the search is updated beyond August 2023 prior to the results publication. 

The authors have prespecified their outcomes, however, have listed many primary outcomes and stated “Any measure of compromise of neonatal or childhood wellbeing…” will be used. I suggest pre-specifying a single (or at most two) key primary outcomes, and other outcomes be listed as secondary. Mortality and neurodevelopment may be appropriate given the introduction; however, this is beyond my expertise and knowledge of the literature. I also suggest revising the sentence so that the measures of compromised are more specific, and not “any” measure is included. 

There are many secondary outcomes. I would recommend a sentence acknowledging the increased family wise error rate when conducting multiple comparisons. E.g., “because family wise error rate increases when comparing many secondary outcomes, significant results will be interpreted with caution based on plausibility, theory and uncertainty.” 

Discussion 

Strengths and limitations are appropriately noted.

Is the study design appropriate for the research question?

Yes

Is the rationale for, and objectives of, the study clearly described?

Yes

Are sufficient details of the methods provided to allow replication by others?

Partly

Are the datasets clearly presented in a useable and accessible format?

Yes

Reviewer Expertise:

Systematic Reviews, Meta-analysis, Research Integrity, Neonatal and Perinatal Research

We confirm that we have read this submission and believe that we have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however we have significant reservations, as outlined above.

References

  • 1. : Basics of meta-analysis: I2 is not an absolute measure of heterogeneity. Res Synth Methods .2017;8(1) : 10.1002/jrsm.1230 5-18 10.1002/jrsm.1230 [DOI] [PubMed] [Google Scholar]
  • 2. : Searching clinical trials registers: guide for systematic reviewers. BMJ .2022;377: 10.1136/bmj-2021-068791 e068791 10.1136/bmj-2021-068791 [DOI] [PubMed] [Google Scholar]
HRB Open Res. 2024 Mar 13.
Jill Mitchell 1

Dear Dr Sotiropoulos and Dr Libesman,

We thank you both for your helpful review of our manuscript entitled “Association between Intrapartum Fetal Pulse Oximetry and Adverse Perinatal and Long-term Outcomes: a Systematic Review and Meta-analysis Protocol”.

Please find below an itemized reply addressing each comment in your Reviewer’s Report.

1.Introduction 

The introduction is clear and describes the need for this study – with a recent narrative review without systematic review and meta-analysis indicating potential findings. 

We thank you for your positive comment regarding our Introduction. 

2.Methods 

Table 1

Table 1. Review Outcomes

https://hrbopenresearch.s3.eu-west-1.amazonaws.com/linked/197134.13802-Author_response_to_Referee_James_X_Sotiropoulos_v1__2.pdf

3. Are there any inclusion/exclusion criteria beyond cephalic, singleton live baby? The authors' inclusion/exclusion criteria is singleton pregnancies only. Is there a good justification for this? If possible, it may provide future utility in collecting data for multiples, even if only reporting it in stratified analysis separately. If FPO is not feasible/reliable in multiple pregnancies this should be clearly stated and cited. 

We have removed the word “singleton” from the inclusion criteria so that our search will include multiple pregnancies although we don’t expect to find studies where fetal oxygen saturation is measured in both twins as it is therefore only technically possible to monitor fetal oxygen saturation using fetal pulse oximetry for the presenting twin using the available technology. This being said, a study is currently ongoing investigating the concurrence of internal and external fetal oxygen sensors using a novel transabdominal fetal pulse oximetry device (3). However, an inclusion criteria in this study is singleton pregnancy.

We have updated the manuscript to reflect this, see p9, lines 95 to 111.

“ Population

Women in labour with a cephalic baby.

Our search will not exclude multiple pregnancies although we don’t expect to find studies where FPO is measured in the non-presenting fetus as it is only technically possible to monitor fetal oxygen saturation using fetal pulse oximetry for the presenting twin using the available technology as the devices are positioned between the fetal cheek or scapula and uterine wall or placed on the fetal scalp (4-9) . This being said, a study is currently ongoing investigating the concurrence of internal and external fetal oxygen sensors using a novel transabdominal fetal pulse oximetry device (3) . However, an inclusion criteria in this study is singleton pregnancy.”

We decided not to exclude based on gestational age as we want to include as many relevant studies as possible. Where the data allow, we will perform subgroup analyses based on gestational age. See p14, lines 257-260.

“Where the data allow, we will perform subgroup analyses based on gestational age.”

4. The authors could consider further specifying the outcomes. For example, the cut-off for “low”, and “abnormal” are not provided.

Thank you for this suggestion. We have updated our protocol and Table 1 to further define the outcomes. See p10-11, lines 124-153;

“The primary outcomes of this review are; (i) umbilical artery pH less than 7.20, (ii) umbilical artery pH less than 7.15 and (iii) umbilical artery pH less than 7.0, (iv) 5 minute APGAR Score less than 7. Secondary outcomes include; (i) umbilical artery base excess less then -10mmol/L, (ii) umbilical artery lactate greater than 4.8 mmol/L, (iii) admission to the neonatal intensive care unit (NICU), (iv) neonatal or intrapartum death, (v) cardiopulmonary resuscitation or intubation required within 24 hours of life, (vi) hypoxic ischaemic encephalopathy, (vii) umbilical vein oxygen saturation less than 55%, (viii) low umbilical artery oxygen saturation less than 30%, (ix) operative delivery for non-reassuring fetal status (as defined by local protocols in each study) (x) operative delivery for dystocia, (xi) fetal scalp pH <7.20, (xii) fetal scalp lactate >4.8mmol/L, (xiii) cerebral palsy and (xiv) severe neurodevelopmental disability. There is no consensus on the specific definition of developmental delay (10) . Multiple different assessment tools exist and the recommended tools have changed over the past 30 years (11, 12) . Severe neurodevelopmental disability was defined by a previous Cochrane review (1) ; “any one or a combination of the following: non‐ambulant cerebral palsy, developmental delay (developmental quotient less than 70), auditory and visual impairment assessed at 12 months of age or more. Development should have been assessed by means of a previously validated tool, such as Bayley Scales of Infant Development (Psychomotor Developmental Index and Mental Developmental Index (2) .”

5. Could you specify how developmental delay will be measured including a cut-off (if this has been established)?

We have considered your suggestion and with this in mind have decided to update our objective from “developmental delay” to “severe neurodevelopmental disability” as defined by a previous Cochrane review(1). This is outlined in our on p11, lines 146 to 153, as follows:

“There is no consensus on the specific definition of developmental delay (10) . Multiple different assessment tools exist and the recommended tools have changed over the past 30 years (11, 12) . Severe neurodevelopmental disability was defined by a previous Cochrane review (1) ; “any one or a combination of the following: non‐ambulant cerebral palsy, developmental delay (developmental quotient less than 70), auditory and visual impairment assessed at 12 months of age or more. Development should have been assessed by means of a previously validated tool, such as Bayley Scales of Infant Development (Psychomotor Developmental Index and Mental Developmental Index (2) .”

6. Could you also specify the time in which these outcomes are measured (e.g., the primary endpoints),the measurement units (if relevant), and the relative effects that will be used to compare the groups? 

We have defined the term “perinatal outcomes” as outcomes occurring within the first 24 hours of life (see p19, Table 1, footnote 1) and severe neurodevelopmental disability was defined as any one or a combination of the following: non‐ambulant cerebral palsy, developmental delay (developmental quotient less than 70), auditory and visual impairment assessed at 12 months of age (see p19, Table 1, footnote 2).

“* Perinatal outcomes are defined as outcomes occurring within the first 24 hours of life.

** Severe neurodevelopmental disability was defined by a previous Cochrane review (1) ; “any one or a combination of the following: non‐ambulant cerebral palsy, developmental delay (developmental quotient less than 70), auditory and visual impairment assessed at 12 months of age or more. Development should have been assessed by means of a previously validated tool, such as Bayley Scales of Infant Development (Psychomotor Developmental Index and Mental Developmental Index” (2) .”

The summary measures will be reported as odds ratio with 95% confidence interval. We will present the correlation between FSp02 and umbilical artery and vein pH in narrative form.

This has been clarified on p14, lines 242 to 244.

“The summary measures will be reported as odds ratio (OR) with 95% confidence interval (CI). We will present the correlation between FSp02 and umbilical artery and vein pH in narrative form.”

7. In the section “comparator” you mention “We will compare …offspring who had low oxygen saturations in labour versus those who had normal oxygen saturations in labour as measured by…”. Could you specify the cut-off you plan to use for low?  For further guidance on the description of treatments and outcomes, the ‘estimand’ framework may help you clarify elements of treatment, population, outcome, population level summaries, and more. There are two review questions nominated, but the comparator section only appears to address the first review question.

We have now edited our “intervention/exposure” and “comparison” to incorporate your suggestion. . This is outlined in our manuscript as follows (p10, lines 113 to 122):

“Intervention/Exposure

  1. Low fetal oxygen saturation defined as less than 30% (exposure).

  2. The use of fetal pulse oximetry during labour as a means of measuring intrapartum fetal oxygen saturation (intervention).

Comparison

  1. Normal fetal oxygen saturation defined as greater than or equal to 30% (unexposed group).”

  2. Conventional fetal monitoring e.g. fetal heart rate monitoring by CTG or the use of fetal scalp electrode, fetal blood sampling or fetal electrocardiogram without the measurement of fetal oxygen saturation (control).

8. The authors plan to use I 2 as a measure of heterogeneity. I 2 is a flawed measure of heterogeneity, although it permeates systematic review and meta-analysis across all disciplines. Please see this article (https://doi.org/10.1002/jrsm.1230,  1 ). Will the authors consider, instead or in addition to I 2, graphically representing heterogeneity with prediction intervals or another measure of between-study variance? The Cochrane Handbook provides some useful guidance. 

Thank you for this comment. We have amended methods section to include prediction intervals where meta-analysis include three or more studies, see p14, lines 251 to 253.

“Furthermore, for meta-analyses which include at least three studies, 95% prediction intervals will be calculated in Microsoft Excel using formulae previously outlined by Borenstein et al (13, 14) ’

9. The authors have not included an assessment of the overall certainty of evidence (e.g., GRADE). This has also been marked as “NA” in the PRISMA-P document. I strongly recommend the use of GRADE to assess the overall certainty of evidence, or if it is to be omitted for a valid reason that this be explained in the methods and/or discussion. Alternatively, if it is omitted a clear plan as to how the evidence will be evaluated should be provided. 

Thank you for this recommendation. We have decided to use an assessment of the overall certainty of evidence using the GRADE approach. This has been added to the manuscript. See p13, lines 216 to 220.

“The grading of recommendations, assessment, development, and evaluation (GRADE) approach will be used to evaluate the certainty of the evidence for primary outcomes by two reviewers independently (J.M. and S.W). The certainty of the evidence was assessed using the GRADEpro software on the basis of the following domains: the study design, risk of bias, imprecision, inconsistency, indirectness, and publication bias (15) .”

10. The authors have noted the potential impact of publication bias and plan to use a funnel plot. They may consider a contour-enhanced funnel plot, PET-PEESE, or p-curve analysis to examine publication bias.

Thank you for this suggestion. We have decided to assess publication bias using Egger’s test in addition to funnel plots. See p14, lines 254 to 256.

“The presence of publication bias will be evaluated using a funnel plot, and by conducting an Egger’s test for asymmetry of the funnel plot , provided a minimum inclusion of 10 studies or more in the meta-analysis.”

11. They might also consider a search of clinical trial registers to look for other registered studies that may be unpublished but have useable data for inclusion in meta-analysis. Guidance for searching clinical trial registers is here  https://doi.org/10.1136/bmj-2021-068791   2 .

We have decided to include clinical trial registers, namely ClinicalTrials.Gov and WHO ICTRP in our search. We have amended our manuscript to read (p12, lines 185 to 187);

“The following electronic bibliographic databases and clinical trial registers will be searched: PubMed, EMBASE, CINAHL, The Cochrane Library, Web of Science, ClinicalTrials.Gov and WHO ICTRP until February 2024.”

12. The authors intend to search only until August 2023. By the time the results of this study are ready for publication the search may be outdated. I note the PROSPERO registration says the search will begin in 2024. This may be a typo. Otherwise, I suggest the search is updated beyond August 2023 prior to the results publication. 

Thank you for this suggestion. We have extended our search timeline to February 2024, see p12, lines 185 to 187.

“The following electronic bibliographic databases and clinical trial registers will be searched: PubMed, EMBASE, CINAHL, The Cochrane Library, Web of Science, ClinicalTrials.Gov and WHO ICTRP until February 2024.”

13. The authors have prespecified their outcomes, however, have listed many primary outcomes and stated “Any measure of compromise of neonatal or childhood wellbeing…” will be used. I suggest pre-specifying a single (or at most two) key primary outcomes, and other outcomes be listed as secondary. Mortality and neurodevelopment may be appropriate given the introduction; however, this is beyond my expertise and knowledge of the literature. I also suggest revising the sentence so that the measures of compromised are more specific, and not “any” measure is included. 

Thank you for this comment. We have taken this suggestion on board and have updated our outcomes accordingly. See p10, lines 124 to 146.

“The primary outcomes of this review are; (i) umbilical artery pH less than 7.20, (ii) umbilical artery pH less than 7.15 and (iii) umbilical artery pH less than 7.0, (iv) 5 minute APGAR Score less than 7. Secondary outcomes include; (i) umbilical artery base excess less then -10mmol/L, (ii) umbilical artery lactate greater than 4.8 mmol/L, (iii) admission to the neonatal intensive care unit (NICU), (iv) neonatal or intrapartum death, (v) cardiopulmonary resuscitation or intubation required within 24 hours of life, (vi) hypoxic ischaemic encephalopathy, (vii) umbilical vein oxygen saturation less than 55%, (viii) low umbilical artery oxygen saturation less than 30%, (ix) operative delivery for non-reassuring fetal status (as defined by local protocols in each study) (x) operative delivery for dystocia, (xi) fetal scalp pH <7.20, (xii) fetal scalp lactate >4.8mmol/L, (xiii) cerebral palsy and (xiv) severe neurodevelopmental disability.

14. There are many secondary outcomes. I would recommend a sentence acknowledging the increased family wise error rate when conducting multiple comparisons. E.g., “because family wise error rate increases when comparing many secondary outcomes, significant results will be interpreted with caution based on plausibility, theory and uncertainty.” 

We have included this sentence and will also include it in the limitations section of our review. See p16, page 309 to 311.

“Family wise error rate increases when comparing many secondary outcomes, therefore significant results will be interpreted with caution based on plausibility, theory and uncertainty.”

15.Discussion 

Strengths and limitations are appropriately noted.

We thank you for your positive comment regarding our Discussion. 

We hope these amendments address the suggestions raised. Should any further amendments be necessary, we would be happy to address them accordingly. 

Sincerely,

Dr Jill Mitchell

Corresponding author.  

Research Fellow

Department of Obstetrics and Gynaecology, University College Cork, Cork, Ireland.

References

1.         Devane D, Lalor JG, Daly S, McGuire W, Cuthbert A, Smith V. Cardiotocography versus intermittent auscultation of fetal heart on admission to labour ward for assessment of fetal wellbeing. Cochrane Database of Systematic Reviews. 2017(1).

2.         Bayley N. Manual for the Bayley Scales of Infant Development 2nd Edition: San Antonio: The Psychological Corporation; 1993.

3.         Rosen M. Trans-abdominal Fetal Pulse Oximetry. 2023.

4.         Siristatidis C, Salamalekis E, Kassanos D, Loghis C, Creatsas G. Evaluation of fetal intrapartum hypoxia by middle cerebral and umbilical artery Doppler velocimetry with simultaneous cardiotocography and pulse oximetry. Archives of gynecology and obstetrics. 2004;270(4):265-70.

5.         Siristatidis C, Salamalekis E, Vitoratos N, Loghis C, Salloum J, Kassanos D, et al. Intrapartum surveillance of IUGR fetuses with cardiotocography and fetal pulse oximetry. Biol Neonate. 2003;83(3):162-5.

6.         Vitoratos N, Salamalekis E, Saloum J, Makrakis E, Creatsas G. Abnormal fetal heart rate patterns during the active phase of labor: The value of fetal oxygen saturation. Journal of Maternal-Fetal Medicine. 2002;11(1):46-9.

7.         Carbonne B, Langer B, Goffinet F, Audibert F, Tardif D, Le Goueff F, et al. Multicenter study on the clinical value of fetal pulse oximetry: II. Compared predictive values of pulse oximetry and fetal blood analysis. American Journal of Obstetrics and Gynecology. 1997;177(3):593-8.

8.         Luttkus AK, Callsen TA, Stupin JH, Dudenhausen JW. Pulse oximetry during labour - Does it give rise to hope? Value of saturation monitoring in comparison to fetal blood gas status. European Journal of Obstetrics and Gynecology and Reproductive Biology. 2003;110(SUPPL.):S132-S8.

9.         Uchida T, Kanayama N, Mukai M, Furuta N, Itoh H, Suzuki H, et al. Examiner's finger-mounted fetal tissue oximetry: a preliminary report on 30 cases. Journal of Perinatal Medicine. 2016;44(7):745-9.

10.       Petersen MC, Kube DA, Palmer FB. Classification of developmental delays. Seminars in Pediatric Neurology. 1998;5(1):2-14.

11.       Rosenbaum P. Screening tests and standardized assessments used to identify and characterize developmental delays. Seminars in Pediatric Neurology. 1998;5(1):27-32.

12.       Çelik P, Ayranci Sucakli I, Yakut HI. Which Bayley-III cut-off values should be used in different developmental levels? Turk J Med Sci. 2020;50(4):764-70.

13.       Borenstein M, Higgins JP, Hedges LV, Rothstein HR. Basics of meta-analysis: I(2) is not an absolute measure of heterogeneity. Res Synth Methods. 2017;8(1):5-18.

14.       Borenstein M, Hedges LV, Higgins JPT, Rothstein HR. Prediction Intervals.  Introduction to Meta‐Analysis2009. p. 127-33.

15.       Balshem H, Helfand M, Schünemann HJ, Oxman AD, Kunz R, Brozek J, et al. GRADE guidelines: 3. Rating the quality of evidence. Journal of Clinical Epidemiology. 2011;64(4):401-6.

HRB Open Res. 2023 Nov 10. doi: 10.21956/hrbopenres.15101.r36909

Reviewer response for version 1

Maeve Eogan 1,2

This protocol for a systematic review & metanalysis synthesising evidence on this relevant topic for parents and practitioners is already registered with Prospero, and is well summarised and detailed in the attached documents. The background to the topic is eloquently described, the review questions are distinct and the methods are described in sufficient detail. Inclusion and exclusion criteria for identified studies are prescribed, and the methodology and proposed reporting of findings appear appropriate.

Is the study design appropriate for the research question?

Yes

Is the rationale for, and objectives of, the study clearly described?

Yes

Are sufficient details of the methods provided to allow replication by others?

Yes

Are the datasets clearly presented in a useable and accessible format?

Yes

Reviewer Expertise:

Clinical obstetrics

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Data Citations

    1. Mitchell J, Walsh S, O'Byrne L, et al. : Preferred Reporting Items for Systematic review and Meta-analysis protocol (PRISMA-P) checklist and flow diagram.docx. figshare. Dataset.2023. 10.6084/m9.figshare.24049899.v2 [DOI]
    2. Mitchell J, Walsh S, O'Byrne L, et al. : Search Strategy - Association between Intrapartum Fetal Pulse Oximetry and Adverse Perinatal and Long-term Outcomes- a Systematic Review and Meta-analysis Protocol.docx. figshare. Dataset.2023. 10.6084/m9.figshare.24049890.v3 [DOI] [PMC free article] [PubMed]

    Data Availability Statement

    Underlying data

    No data are associated with this article.

    Extended data

    Fighsare: Search Strategy - Association between Intrapartum Fetal Pulse Oximetry and Adverse Perinatal and Long-term Outcomes- a Systematic Review and Meta-analysis Protocol.docx. https://doi.org/10.6084/m9.figshare.24049890.v3 29 .

    Reporting guidelines

    Fighsare: PRISMA-P checklist for ‘Association between intrapartum fetal pulse oximetry and adverse perinatal and long-term outcomes: a systematic review and meta-analysis protocol. https://doi.org/10.6084/m9.figshare.24049899 28 .

    Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).

    Articles from HRB Open Research are provided here courtesy of Health Research Board Ireland

    RESOURCES