- Email: [email protected]
Fetal pulse oximetry: a critical appraisal
Best Practice & Research Clinical Obstetrics and Gynaecology Vol. 18, No. 3, pp. 477–484, 2004 doi:10.1016/j.bpobgyn.2004.02.010 available online at http://www.sciencedirect.com
9 Fetal pulse oximetry: a critical appraisal Gary A. Dildy III*
MD
Professor of Obstetrics and Gynecology School of Medicine, Louisiana State University Health Sciences Center, 1542 Tulane Avenue, New Orleans, LA 70112-2822, USA
Despite 40 years of cumulative clinical experience, and a number of randomized clinical trials, electronic fetal heart rate monitoring is perceived by many obstetrical caregivers as a suboptimal method of intrapartum fetal assessment. Fetal pulse oximetry emerged 15 years ago as a promising new technology intended to improve assessment of fetal condition during labor. A large amount of physiologic data and one large randomized clinical trial have brought this technology into clinical practice. We know that fetal acidemia is rare when the arterial oxygen saturation is . 30% but fetal pulse oximetry as currently understood and applied does not reduce the overall cesarean rate. Thus, many clinicians remain unconvinced of the benefit of this technology and its utilization has stalled in the US and Europe. We need to further understand if there is a way to use fetal pulse oximetry in the setting of labor dystocia and a non-reassuring fetal heart rate pattern. Although hypoxemia is an accepted mechanism of fetal brain injury, other potential mechanisms should be explored. Current controversies and fertile areas of research are presented. Key words: acidemia; fetus; intrapartum assessment; pulse oximetry.
CURRENT PROBLEMS WITH INTRAPARTUM FETAL ASSESSMENT Electronic fetal heart rate monitoring (EFM), introduced in the 1960s1 and now the predominant method of intrapartum fetal assessment in most developed countries, has turned out to be disappointing from the standpoint of its subjective nature, frequency of falsely non-reassuring patterns, and persistent questions regarding efficacy.2 A Cochrane review of 13 randomized controlled trials (RCTs) concluded that the only clinically significant benefit from the routine use of continuous EFM compared with intermittent auscultation was the reduction of neonatal seizures, with an increase in cesarean and operative vaginal deliveries.3 The widespread implementation of EFM does not appear to have brought about a reduction in cerebral palsy.4 Despite a multitude of approaches, efforts to develop the ideal method of intrapartum fetal acid – base assessment have remained unfulfilled. Intermittent fetal * Corresponding author. Tel.: þ1-504-568-4931; Fax: þ1-504-568-5140. E-mail address: [email protected]; [email protected] (G.A. Dildy III). 1521-6934/$ - see front matter Q 2004 Elsevier Ltd. All rights reserved.
478 G. A. Dildy III
scalp blood sampling, also developed in the 1960s5, is now rarely utilized in the US.6,7 Scalp and acoustic stimulation, while simple to perform, are limited by falsely nonreassuring results.8,9 Attempts to develop devices that continually assess fetal pO210, pCO211, and pH12 have all but been abandoned. More recently, several new technologies with practical potential have emerged, namely computerized fetal heart rate (FHR) analysis13, ST segment analysis14, near-infrared spectroscopy15, and fetal pulse oximetry (FPO).16
DEVELOPMENT OF FETAL PULSE OXIMETRY Since its introduction in the early 1980s, pulse oximetry—the measurement of oxyhemoglobin and deoxyhemoglobin in arterial blood using alternating pulses of red and near-infrared light—has had a significant positive impact on patient care in a variety of settings.17 To obstetrical investigators and industry, it seemed logical that the natural extension of pulse oximetry would be the obstetrical patient, as intrapartum fetal asphyxia has generally been viewed to occur via hypoxemia. FPO made its formal debut in the late 1980s, when investigators in the UK independently reported their initial experiences in measuring fetal oxygen saturation (FSpO2) using cannibalized components of adult oximeters.18 – 20 During the 1990s, investigators in the UK, Europe, Australia, Asia, and the US published a large volume of data regarding feasibility, physiology and clinical application of this technology.16,21 – 26 One of the most important questions addressed was the ‘critical threshold’ of FSpO2— the level of fetal arterial oxygen saturation above which acidemia does not occur. Clinical observations of thousands of cases, and published studies, suggest that a critical threshold of 30% would be appropriate for human clinical use, a threshold used in the American RCT and subsequently approved by the US Food and Drug Administration (FDA).27 – 30
FETAL PULSE OXIMETRY CLINICAL TRIALS The American multicenter RCTwas the first large-scale study to assess clinical utility of this new technology and remains the only RCT published to date.31 Eligible patients included singleton, vertex fetuses at least 36 weeks of gestation, with specifically defined mild to moderately non-reassuring EFM tracings. The primary hypothesis was that the cesarean delivery rate for non-reassuring EFM patterns would be halved by using FPO as an adjunct to EFM. However, although the cesarean rate for nonreassuring EFM was reduced by more than 50%, the overall cesarean rate was unchanged between the control and the test group, secondary to an increased incidence of cesareans for dystocia in the FPO group (Table 1). The failure to reduce the overall cesarean rate led many to question whether there was any clear benefit to FPO monitoring, despite improved prediction of fetal condition using FPO. It should be recognized that the study measured short-term neonatal outcomes (to discharge from the hospital) and was not designed to measure long-term outcomes. Subanalysis of these data was performed to explain the unanticipated increase in cesarean rate for dystocia. The first hypothesis was that patients were mislabeled for the indication of cesarean delivery. However, this appeared not to be the case because similar numbers of test and control patients sectioned for dystocia had documented
Fetal pulse oximetry: a critical appraisal 479
Table 1. Summary of results from the American RCT. Outcomes
EFM alone ðn ¼ 502Þ
EFM þ FPO ðn ¼ 508Þ
P
Cesarean NRFS Dystocia 5-minute Apgar ,7 Umbilical artery pH , 7.0 Neonatal death
130 (26%) 51 (10%) 43 (9%) 19 4 2
147 (29%) 23 (5%) 94 (19%) 8 3 3
NS ,0.0001 ,0.0001 0.05 NS NS
EFM, electronic fetal monitoring; FPO, fetal pulse oximetry; NRFS, non-reassuring fetal status; RCT, randomized clinical trial.
arrest of labor or failed induction on blinded review of partograms. The second hypothesis was that the sensor itself slowed labor. However, Kaplan– Meyer analysis showed no slowing effect of group assignment for patients delivered by any mode or indication of delivery. We then considered the possibility that there was an imbalance in patient risk factors, such as induction of labor. However, logistic regression analysis indicated an independent effect of group assignment and unbalanced factors did not explain the difference. Finally, we explored the possibility that an abnormal EFM pattern, itself, is a marker for dystocia. However, the data—acquired prospectively—were not sufficient to adequately test this hypothesis. It is of interest to note that previous studies showed that malpositions of the fetal occiput are strongly associated with dystocia and that occipitoposterior positions are associated with a marked increase in variable decelerations.32 Thus, the Nellcor-sponsored investigators conducted a follow-up study to further elucidate the increased cesarean rate for dystocia observed in the American RCT.33 The entry criteria for EFM patterns in this prospective cohort study were similar to those used in the original trial.31 The investigators found that women with more clinically worrisome EFM patterns had a significantly higher incidence of cesarean delivery than those who had less worrisome patterns while managed with a standardized labor protocol, despite the fact that both groups had similar FSpO2 trends and similar neonatal outcomes. The investigators concluded that non-reassuring EFM patterns predict cesarean delivery for dystocia among nulliparous patients with normally oxygenated fetuses.
CLINICAL IMPLEMENTATION OF FETAL PULSE OXIMETRY The Nellcor (Pleasanton, California, USA) device was approved by the FDA in May 2000 and soon thereafter became available for clinical use in the US. Currently, the Nellcor system is the only device approved by the FDA and commercially available in the US. It is important to note which technologies (transmission versus reflectance oximeters, light-emitting diode wavelengths, etc.) are reported in the methodology sections of peer-reviewed publications; the vast majority (. 99%) of published FPO cases have been performed with a Nellcor device (N400 monitor with FS10 or FS14 reflectance sensor). In Europe, FPO technology is available from Nellcor and from OB Scientific (Germantown, Wisconsin, USA).
480 G. A. Dildy III
In September 2001, the American College of Obstetricians and Gynecologists (ACOG) released a Committee Opinion (Number 258) on FPO. This stated that the: …Committee on Obstetric Practice cannot endorse the adoption of this device in clinical practice at this time because of concerns that its introduction could further escalate the cost of medical care without necessarily improving clinical outcome. The committee recommends that prospective randomized clinical trials be conducted to evaluate the clinical use of this new technology in conjunction with fetal well-being assessment’.34 It appeared that the ACOG had three concerns regarding FPO, specifically: † signal registration time, † possible false negative readings, † lack of proven cost benefit. The effect of this publication on American obstetricians appears to be a significant reversal in enthusiasm for the technology. In March 2002, the Society of Obstetricians and Gynaecologists of Canada published guidelines for the surveillance of fetal health in labor. It recommended that FPO, as an adjunct to EFM in patients with non-reassuring fetal status, should not be considered a standard of care.35,36 None of these organizations, however, has recommended that FPO should not be used by clinicians. In early 2003, FDA labeling for the Nellcor monitor was revised. This was based upon post-approval surveillance of over 12 000 cases. Thirteen unexpected adverse fetal outcomes were the focus of investigation; in all but one case there appeared to be a deviation from the clinical management protocol. Nevertheless, a re-categorization of ‘ominous’ EFM patterns added two new definitions (‘markedly reduced or absent variability with late decelerations’ and ‘markedly reduced or absent variability with severe variable decelerations’) to the pre-existing definition (‘prolonged deceleration below 70 beats per minute for at least 7 minutes’) of ‘ominous’ EFM requiring delivery. The new FDA recommendations re-emphasize that FPO is meant as an adjunct to EFM, not as a replacement, and that no technology is 100% predictive of the fetal acid-based condition.
CLINICAL CONTROVERSIES Obstetrical clinicians and academicians continue to debate for37 and against38 FPO. Objections to FPO, as heard by this author, include: † ‘FPO is unnecessary if one knows how to interpret EFM properly’. It is widely agreed that falsely non-reassuring EFM patterns are common and are the most substantial problem with this form of intrapartum fetal assessment; we have all seen EFM patterns interpreted as reassuring, only to later prove erroneous, leading to severe fetal compromise. Even ‘experts’ have shown inconsistencies in interpretation of the same EFM patterns at later readings. There appears to be a problem with the teaching or learning process of EFM pattern interpretation, despite four decades of experience. This is probably due to the inherently subjective nature of EFM patterns. Whereas EFM is satisfactory most of the time, there is a clear need for an adjunct method of intrapartum fetal surveillance when concerns arise about the EFM pattern.
Fetal pulse oximetry: a critical appraisal 481
† ‘What benefit is FPO if the cesarean rate cannot be reduced?’. It was originally hoped that a dual benefit of good neonatal outcome with a lower cesarean rate could be accomplished with FPO. However, although the largest experience to date—the American RCT—improved on prediction of fetal outcome, it did not reduce the overall cesarean rate and thus fell short of expectations. One might argue that the cesarean section rate is not the real problem facing obstetrical caregivers today, but rather the need for improved fetal assessment. Future studies should aim to determine whether FPO improves prediction of fetal condition, as seen in the American RCT. If a reduction in cesarean delivery is mandated, further research to better manage dysfunctional labor in the setting of non-reassuring EFM will be needed. † ‘FPO is not accurate enough for safe clinical use.39 – 43’ The accuracy and precision of FPO readings have improved steadily over the last decade as a result of dedicated research and development. Although it now appears to many—including the FDA after rigorous review—that the existing Nellcor device is satisfactory for clinical use, there is no question that further improvements are needed to provide more continuous, easy-to-acquire, accurate, and precise information. To a great extent, this will duty will lie with those industries that provide the commercial devices. † ‘Fetal brain injury is due to ischemia, not hypoxemia.38’ If true, this theory (that normally oxygenated blood could perfuse the skin of the presenting part while the brain suffers ischemic damage) would be the Achilles’ heel of FPO. The argument is based on anecdotal experience and published data are lacking. Nonetheless, in my opinion this theory merits study in an animal model experiment.
FUTURE RESEARCH One ongoing study is likely to impact the long-term future of FPO. The NICHHD MFM Units Network is currently conducting an RCT involving 10 000 nulliparous women in labor. Subjects will be randomized to FPO with information available to the clinician (open device arm) or FPO with information masked to the clinician (blinded device arm). The primary outcome is the impact of FPO as an adjunct to EFM on the cesarean delivery rate. The study was initiated in 2002 will run for a period of several years; thus the results are not anticipated in the near future.
SUMMARY Since the publication of ACOG Committee Opinion #258, clinical utilization, research, and development have all substantially declined, leaving the future of FPO in question. Any new insights into the utility of FPO will probably be provided by large clinical trials. We know that fetal acidemia is rare when the arterial oxygen saturation is . 30% but FPO, as currently understood and applied, does not reduce the overall cesarean rate. We need to further establish if there is a way to use fetal pulse oximetry in the setting of labor dystocia and a non-reassuring FHR pattern. Whereas hypoxemia is an accepted pathophysiologic pathway to fetal brain injury, other potential mechanisms should be explored. For practicing clinicians, the decision of whether to utilize this technology should depend on a critical review of existing information.
482 G. A. Dildy III
Practice points † as an adjunct to EFM, fetal pulse oximetry improves the prediction of fetal acidbase condition † fetal pulse oximetry will reduce the cesarean rate for non-reassuring EFM patterns but, as currently used, will not decrease the overall cesarean rate † new FDA guidelines expand EFM patterns to be considered ominous, requiring prompt delivery despite reassuring fetal oximetry readings
Research agenda † to establish whether there is a way to employ fetal pulse oximetry effectively in the setting of labor dystocia and a non-reassuring fetal heart rate pattern to allow safe vaginal delivery † to determine whether other mechanisms of intrapartum fetal brain injury besides hypoxemia (e.g. infection, ischemia) could escape surveillance by fetal pulse oximetry † to establish the long-term outcome of children monitored with fetal pulse oximetry, compared to standard methods
ACKNOWLEDGEMENTS Thanks to Suzanne Harper for manuscript preparation.
REFERENCES 1. Hon EH. Instrumentation of fetal heart rate and fetal electrocardiography. III. Fetal ECG electrodes: further observations. Obstetrics and Gynecology 1967; 30: 281–286. 2. Freeman RK. Problems with intrapartum fetal heart rate monitoring interpretation and patient management. Obstetrics and Gynecology 2002; 100: 813 –826. 3. Thacker SB, Stroup D & Chang M. Continuous electronic heart rate monitoring for fetal assessment during labor. Cochrane Database of Systematic Reviews 2001; : CD000063. 4. Clark SL & Hankins GD. Temporal and demographic trends in cerebral palsy—fact and fiction. American Journal of Obstetrics and Gynecology 2003; 188: 628– 633. 5. Saling E & Schneider D. Biochemical supervision of the foetus during labour. Journal of Obstetrics and Gynaecology of British Commonwealth 1967; 74: 799–811. 6. Clark SL & Paul RH. Intrapartum fetal surveillance: the role of fetal scalp blood sampling. American Journal of Obstetrics and Gynecology 1985; 153: 717 –720. 7. Goodwin TM, Milner-Masterson L & Paul RH. Elimination of fetal scalp blood sampling on a large clinical service. Obstetrics and Gynecology 1994; 83: 971 –974. 8. Tan KH & Smyth R. Fetal vibroacoustic stimulation for facilitation of tests of fetal wellbeing (Cochrane Review). Cochrane Database of Systematic Reviews 2001; : 1. 9. Porter TF & Clark SL. Vibroacoustic and scalp stimulation. Obstetrics and Gynecology Clinics of North America 1999; 26: 657 –669. 10. Aarnoudse JG, Huisjes HJ, Gordon H, et al. Fetal subcutaneous scalp PO2 and abnormal heart rate during labor. American Journal of Obstetrics and Gynecology 1985; 153: 565–566. 11. Bergmans MG, van Geijn HP, Weber T, et al. Fetal transcutaneous PCO2 measurements during labour. European Journal of Obstetrics, Gynecology, and Reproductive Biology 1993; 51: 1– 7.
Fetal pulse oximetry: a critical appraisal 483 12. Young BK, Katz M, Wilson SJ & Klein SA. Continuous fetal tissue pH monitoring in labor. Progress in Clinical and Biological Research 1980; 44: 123– 132. 13. Devoe L, Golde S, Kilman Y, et al. A comparison of visual analyses of intrapartum fetal heart rate tracings according to the new national institute of child health and human development guidelines with computer analyses by an automated fetal heart rate monitoring system. American Journal of Obstetrics and Gynecology 2000; 183: 361–366. 14. Westgate J, Harris M, Curnow JS & Greene KR. Plymouth randomized trial of cardiotocogram only versus STwaveform plus cardiotocogram for intrapartum monitoring in 2400 cases. American Journal of Obstetrics and Gynecology 1993; 169: 1151–1160. 15. Aldrich CJ, D’Antona D, Wyatt JS, et al. Fetal cerebral oxygenation measured by near-infrared spectroscopy shortly before birth and acid –base status at birth. Obstetrics and Gynecology 1994; 84: 861–866. 16. Dildy GA, Clark SL & Loucks CA. Intrapartum fetal pulse oximetry: past, present, and future. American Journal of Obstetrics and Gynecology 1996; 175: 1– 9. 17. Yelderman M & New Jr W. Evaluation of pulse oximetry. Anesthesiology 1983; 59: 349–352. 18. Peat S, Booker M, Lanigan C & Ponte J. Continuous intrapartum measurement of fetal oxygen saturation [letter]. Lancet 1988; 2: 213. 19. Johnson N, Johnson VA, Bannister J & Lilford RJ. Measurement of fetal peripheral perfusion with a pulse oximeter [letter]. Lancet 1989; 1: 898. 20. Gardosi J, Carter M & Becket T. Continuous intrapartum monitoring of fetal oxygen saturation [letter] [see comments]. Lancet 1989; 2: 692 –693. 21. Dildy GA, Clark SL & Loucks CA. Preliminary experience with intrapartum fetal pulse oximetry in humans. Obstetrics and Gynecology 1993; 81: 630 –635. 22. Carbonne B, Audibert F, Segard L, et al. Fetal pulse oximetry: correlation between changes in oxygen saturation and neonatal outcome. Preliminary report on 39 cases. European Journal of Obstetrics, Gynecology, and Reproductive Biology 1994; 57: 73– 77. 23. Luttkus A, Fengler TW, Friedmann W & Dudenhausen JW. Continuous monitoring of fetal oxygen saturation by pulse oximetry. Obstetrics and Gynecology 1995; 85: 183– 186. 24. East CE & Colditz PB. Women’s evaluations of their experience with fetal intrapartum oxygen saturation monitoring and participation in a research project. Midwifery 1996; 12: 93–97. 25. Yam J, Chua S & Arulkumaran S. Intrapartum fetal pulse oximetry. Part I. Principles and technical issues. Obstetrical and Gynecological Survey 2000; 55: 163–172. 26. Yam J, Chua S & Arulkumaran S. Intrapartum fetal pulse oximetry. Part 2. Clinical application. Obstetrical and Gynecological Survey 2000; 55: 173 –183. 27. Richardson BS, Carmichael L, Homan J & Patrick JE. Electrocortical activity, electroocular activity, and breathing movements in fetal sheep with prolonged and graded hypoxemia. American Journal of Obstetrics and Gynecology 1992; 167: 553–558. 28. Seelbach-Gobel B, Heupel M, Kuhnert M & Butterwegge M. The prediction of fetal acidosis by means of intrapartum fetal pulse oximetry. American Journal of Obstetrics and Gynecology 1999; 180: 73–81. 29. Kuhnert M, Seelbach-Goebel B & Butterwegge M. Predictive agreement between the fetal arterial oxygen saturation and fetal scalp pH: results of the German multicenter study [see comments]. American Journal of Obstetrics and Gynecology 1998; 178: 330–335. 30. Dildy GA, Thorp JA, Yeast JD & Clark SL. The relationship between oxygen saturation and pH in umbilical blood: implications for intrapartum fetal oxygen saturation monitoring. American Journal of Obstetrics and Gynecology 1996; 175: 682 –687. 31. Garite TJ, Dildy GA, McNamara H, et al. A multicenter controlled trial of fetal pulse oximetry in the intrapartum management of nonreassuring fetal heart rate patterns. American Journal of Obstetrics and Gynecology 2000; 183: 1049–1058. 32. Ingemarsson E, Ingemarsson I, Solum T & Westgren M. Influence of occiput posterior position on the fetal heart rate pattern. Obstetrics and Gynecology 1980; 55: 301–304. 33. Porreco RP, Boehm FH, Dildy GA, et al. Dystocia in nulliparous patients monitored with fetal pulse oximetry. American Journal of Obstetrics and Gynecology 2004; 190: 113 –117. 34. ACOG Committee Opinion, Number 258, September 2001. Fetal pulse oximetry. Obstetrics and Gynecology 2001; 98: 523–524. 35. Liston R, Crane J, Hamilton E, et al. Fetal health surveillance in labour. Journal of Obstetrics and Gynaecology Canada 2002; 24: 250 –276. quiz 277 –280. 36. Liston R, Crane J, Hughes O, et al. Fetal health surveillance in labour. Journal of Obstetrics and Gynaecology Canada 2002; 24: 342 –355. 37. Thorp JA. Is fetal pulse oximetry ready for clinical practice? Writing for the PRO position. MCN. The American Journal of Maternal Child Nursing 2003; 28: 64.
484 G. A. Dildy III 38. Schifrin BS. Is fetal pulse oximetry ready for clinical practice? Writing for the CON position. MCN. The American Journal of Maternal Child Nursing 2003; 28: 65. 39. Schram CM & Gardosi JO. Artifacts in fetal pulse oximetry: nonarterial pulsatile signals. American Journal of Obstetrics and Gynecology 1994; 170: 1174–1177. 40. Gardosi JO, Damianou D & Schram CM. Artifacts in fetal pulse oximetry: incomplete sensor-to-skin contact. American Journal of Obstetrics and Gynecology 1994; 170: 1169–1173. 41. Nijland R, Jongsma HW, Nijhuis JG & Oeseburg B. Accuracy of fetal pulse oximetry and pitfalls in measurements. European Journal of Obstetrics, Gynecology, Reproductive Biology 1997; 72: S21–S27. 42. Mannheimer PD. Design and validation of pulse oximetry for low saturation. Anesthesia and Analgesia 2002; 94: S21– S25. 43. Alshimmiri M, Bocking AD, Gagnon R, et al. Prediction of umbilical artery base excess by intrapartum fetal oxygen saturation monitoring. American Journal of Obstetrics and Gynecology 1997; 177: 775 –779.
9 Fetal pulse oximetry: a critical appraisal Gary A. Dildy III*
MD
Professor of Obstetrics and Gynecology School of Medicine, Louisiana State University Health Sciences Center, 1542 Tulane Avenue, New Orleans, LA 70112-2822, USA
Despite 40 years of cumulative clinical experience, and a number of randomized clinical trials, electronic fetal heart rate monitoring is perceived by many obstetrical caregivers as a suboptimal method of intrapartum fetal assessment. Fetal pulse oximetry emerged 15 years ago as a promising new technology intended to improve assessment of fetal condition during labor. A large amount of physiologic data and one large randomized clinical trial have brought this technology into clinical practice. We know that fetal acidemia is rare when the arterial oxygen saturation is . 30% but fetal pulse oximetry as currently understood and applied does not reduce the overall cesarean rate. Thus, many clinicians remain unconvinced of the benefit of this technology and its utilization has stalled in the US and Europe. We need to further understand if there is a way to use fetal pulse oximetry in the setting of labor dystocia and a non-reassuring fetal heart rate pattern. Although hypoxemia is an accepted mechanism of fetal brain injury, other potential mechanisms should be explored. Current controversies and fertile areas of research are presented. Key words: acidemia; fetus; intrapartum assessment; pulse oximetry.
CURRENT PROBLEMS WITH INTRAPARTUM FETAL ASSESSMENT Electronic fetal heart rate monitoring (EFM), introduced in the 1960s1 and now the predominant method of intrapartum fetal assessment in most developed countries, has turned out to be disappointing from the standpoint of its subjective nature, frequency of falsely non-reassuring patterns, and persistent questions regarding efficacy.2 A Cochrane review of 13 randomized controlled trials (RCTs) concluded that the only clinically significant benefit from the routine use of continuous EFM compared with intermittent auscultation was the reduction of neonatal seizures, with an increase in cesarean and operative vaginal deliveries.3 The widespread implementation of EFM does not appear to have brought about a reduction in cerebral palsy.4 Despite a multitude of approaches, efforts to develop the ideal method of intrapartum fetal acid – base assessment have remained unfulfilled. Intermittent fetal * Corresponding author. Tel.: þ1-504-568-4931; Fax: þ1-504-568-5140. E-mail address: [email protected]; [email protected] (G.A. Dildy III). 1521-6934/$ - see front matter Q 2004 Elsevier Ltd. All rights reserved.
478 G. A. Dildy III
scalp blood sampling, also developed in the 1960s5, is now rarely utilized in the US.6,7 Scalp and acoustic stimulation, while simple to perform, are limited by falsely nonreassuring results.8,9 Attempts to develop devices that continually assess fetal pO210, pCO211, and pH12 have all but been abandoned. More recently, several new technologies with practical potential have emerged, namely computerized fetal heart rate (FHR) analysis13, ST segment analysis14, near-infrared spectroscopy15, and fetal pulse oximetry (FPO).16
DEVELOPMENT OF FETAL PULSE OXIMETRY Since its introduction in the early 1980s, pulse oximetry—the measurement of oxyhemoglobin and deoxyhemoglobin in arterial blood using alternating pulses of red and near-infrared light—has had a significant positive impact on patient care in a variety of settings.17 To obstetrical investigators and industry, it seemed logical that the natural extension of pulse oximetry would be the obstetrical patient, as intrapartum fetal asphyxia has generally been viewed to occur via hypoxemia. FPO made its formal debut in the late 1980s, when investigators in the UK independently reported their initial experiences in measuring fetal oxygen saturation (FSpO2) using cannibalized components of adult oximeters.18 – 20 During the 1990s, investigators in the UK, Europe, Australia, Asia, and the US published a large volume of data regarding feasibility, physiology and clinical application of this technology.16,21 – 26 One of the most important questions addressed was the ‘critical threshold’ of FSpO2— the level of fetal arterial oxygen saturation above which acidemia does not occur. Clinical observations of thousands of cases, and published studies, suggest that a critical threshold of 30% would be appropriate for human clinical use, a threshold used in the American RCT and subsequently approved by the US Food and Drug Administration (FDA).27 – 30
FETAL PULSE OXIMETRY CLINICAL TRIALS The American multicenter RCTwas the first large-scale study to assess clinical utility of this new technology and remains the only RCT published to date.31 Eligible patients included singleton, vertex fetuses at least 36 weeks of gestation, with specifically defined mild to moderately non-reassuring EFM tracings. The primary hypothesis was that the cesarean delivery rate for non-reassuring EFM patterns would be halved by using FPO as an adjunct to EFM. However, although the cesarean rate for nonreassuring EFM was reduced by more than 50%, the overall cesarean rate was unchanged between the control and the test group, secondary to an increased incidence of cesareans for dystocia in the FPO group (Table 1). The failure to reduce the overall cesarean rate led many to question whether there was any clear benefit to FPO monitoring, despite improved prediction of fetal condition using FPO. It should be recognized that the study measured short-term neonatal outcomes (to discharge from the hospital) and was not designed to measure long-term outcomes. Subanalysis of these data was performed to explain the unanticipated increase in cesarean rate for dystocia. The first hypothesis was that patients were mislabeled for the indication of cesarean delivery. However, this appeared not to be the case because similar numbers of test and control patients sectioned for dystocia had documented
Fetal pulse oximetry: a critical appraisal 479
Table 1. Summary of results from the American RCT. Outcomes
EFM alone ðn ¼ 502Þ
EFM þ FPO ðn ¼ 508Þ
P
Cesarean NRFS Dystocia 5-minute Apgar ,7 Umbilical artery pH , 7.0 Neonatal death
130 (26%) 51 (10%) 43 (9%) 19 4 2
147 (29%) 23 (5%) 94 (19%) 8 3 3
NS ,0.0001 ,0.0001 0.05 NS NS
EFM, electronic fetal monitoring; FPO, fetal pulse oximetry; NRFS, non-reassuring fetal status; RCT, randomized clinical trial.
arrest of labor or failed induction on blinded review of partograms. The second hypothesis was that the sensor itself slowed labor. However, Kaplan– Meyer analysis showed no slowing effect of group assignment for patients delivered by any mode or indication of delivery. We then considered the possibility that there was an imbalance in patient risk factors, such as induction of labor. However, logistic regression analysis indicated an independent effect of group assignment and unbalanced factors did not explain the difference. Finally, we explored the possibility that an abnormal EFM pattern, itself, is a marker for dystocia. However, the data—acquired prospectively—were not sufficient to adequately test this hypothesis. It is of interest to note that previous studies showed that malpositions of the fetal occiput are strongly associated with dystocia and that occipitoposterior positions are associated with a marked increase in variable decelerations.32 Thus, the Nellcor-sponsored investigators conducted a follow-up study to further elucidate the increased cesarean rate for dystocia observed in the American RCT.33 The entry criteria for EFM patterns in this prospective cohort study were similar to those used in the original trial.31 The investigators found that women with more clinically worrisome EFM patterns had a significantly higher incidence of cesarean delivery than those who had less worrisome patterns while managed with a standardized labor protocol, despite the fact that both groups had similar FSpO2 trends and similar neonatal outcomes. The investigators concluded that non-reassuring EFM patterns predict cesarean delivery for dystocia among nulliparous patients with normally oxygenated fetuses.
CLINICAL IMPLEMENTATION OF FETAL PULSE OXIMETRY The Nellcor (Pleasanton, California, USA) device was approved by the FDA in May 2000 and soon thereafter became available for clinical use in the US. Currently, the Nellcor system is the only device approved by the FDA and commercially available in the US. It is important to note which technologies (transmission versus reflectance oximeters, light-emitting diode wavelengths, etc.) are reported in the methodology sections of peer-reviewed publications; the vast majority (. 99%) of published FPO cases have been performed with a Nellcor device (N400 monitor with FS10 or FS14 reflectance sensor). In Europe, FPO technology is available from Nellcor and from OB Scientific (Germantown, Wisconsin, USA).
480 G. A. Dildy III
In September 2001, the American College of Obstetricians and Gynecologists (ACOG) released a Committee Opinion (Number 258) on FPO. This stated that the: …Committee on Obstetric Practice cannot endorse the adoption of this device in clinical practice at this time because of concerns that its introduction could further escalate the cost of medical care without necessarily improving clinical outcome. The committee recommends that prospective randomized clinical trials be conducted to evaluate the clinical use of this new technology in conjunction with fetal well-being assessment’.34 It appeared that the ACOG had three concerns regarding FPO, specifically: † signal registration time, † possible false negative readings, † lack of proven cost benefit. The effect of this publication on American obstetricians appears to be a significant reversal in enthusiasm for the technology. In March 2002, the Society of Obstetricians and Gynaecologists of Canada published guidelines for the surveillance of fetal health in labor. It recommended that FPO, as an adjunct to EFM in patients with non-reassuring fetal status, should not be considered a standard of care.35,36 None of these organizations, however, has recommended that FPO should not be used by clinicians. In early 2003, FDA labeling for the Nellcor monitor was revised. This was based upon post-approval surveillance of over 12 000 cases. Thirteen unexpected adverse fetal outcomes were the focus of investigation; in all but one case there appeared to be a deviation from the clinical management protocol. Nevertheless, a re-categorization of ‘ominous’ EFM patterns added two new definitions (‘markedly reduced or absent variability with late decelerations’ and ‘markedly reduced or absent variability with severe variable decelerations’) to the pre-existing definition (‘prolonged deceleration below 70 beats per minute for at least 7 minutes’) of ‘ominous’ EFM requiring delivery. The new FDA recommendations re-emphasize that FPO is meant as an adjunct to EFM, not as a replacement, and that no technology is 100% predictive of the fetal acid-based condition.
CLINICAL CONTROVERSIES Obstetrical clinicians and academicians continue to debate for37 and against38 FPO. Objections to FPO, as heard by this author, include: † ‘FPO is unnecessary if one knows how to interpret EFM properly’. It is widely agreed that falsely non-reassuring EFM patterns are common and are the most substantial problem with this form of intrapartum fetal assessment; we have all seen EFM patterns interpreted as reassuring, only to later prove erroneous, leading to severe fetal compromise. Even ‘experts’ have shown inconsistencies in interpretation of the same EFM patterns at later readings. There appears to be a problem with the teaching or learning process of EFM pattern interpretation, despite four decades of experience. This is probably due to the inherently subjective nature of EFM patterns. Whereas EFM is satisfactory most of the time, there is a clear need for an adjunct method of intrapartum fetal surveillance when concerns arise about the EFM pattern.
Fetal pulse oximetry: a critical appraisal 481
† ‘What benefit is FPO if the cesarean rate cannot be reduced?’. It was originally hoped that a dual benefit of good neonatal outcome with a lower cesarean rate could be accomplished with FPO. However, although the largest experience to date—the American RCT—improved on prediction of fetal outcome, it did not reduce the overall cesarean rate and thus fell short of expectations. One might argue that the cesarean section rate is not the real problem facing obstetrical caregivers today, but rather the need for improved fetal assessment. Future studies should aim to determine whether FPO improves prediction of fetal condition, as seen in the American RCT. If a reduction in cesarean delivery is mandated, further research to better manage dysfunctional labor in the setting of non-reassuring EFM will be needed. † ‘FPO is not accurate enough for safe clinical use.39 – 43’ The accuracy and precision of FPO readings have improved steadily over the last decade as a result of dedicated research and development. Although it now appears to many—including the FDA after rigorous review—that the existing Nellcor device is satisfactory for clinical use, there is no question that further improvements are needed to provide more continuous, easy-to-acquire, accurate, and precise information. To a great extent, this will duty will lie with those industries that provide the commercial devices. † ‘Fetal brain injury is due to ischemia, not hypoxemia.38’ If true, this theory (that normally oxygenated blood could perfuse the skin of the presenting part while the brain suffers ischemic damage) would be the Achilles’ heel of FPO. The argument is based on anecdotal experience and published data are lacking. Nonetheless, in my opinion this theory merits study in an animal model experiment.
FUTURE RESEARCH One ongoing study is likely to impact the long-term future of FPO. The NICHHD MFM Units Network is currently conducting an RCT involving 10 000 nulliparous women in labor. Subjects will be randomized to FPO with information available to the clinician (open device arm) or FPO with information masked to the clinician (blinded device arm). The primary outcome is the impact of FPO as an adjunct to EFM on the cesarean delivery rate. The study was initiated in 2002 will run for a period of several years; thus the results are not anticipated in the near future.
SUMMARY Since the publication of ACOG Committee Opinion #258, clinical utilization, research, and development have all substantially declined, leaving the future of FPO in question. Any new insights into the utility of FPO will probably be provided by large clinical trials. We know that fetal acidemia is rare when the arterial oxygen saturation is . 30% but FPO, as currently understood and applied, does not reduce the overall cesarean rate. We need to further establish if there is a way to use fetal pulse oximetry in the setting of labor dystocia and a non-reassuring FHR pattern. Whereas hypoxemia is an accepted pathophysiologic pathway to fetal brain injury, other potential mechanisms should be explored. For practicing clinicians, the decision of whether to utilize this technology should depend on a critical review of existing information.
482 G. A. Dildy III
Practice points † as an adjunct to EFM, fetal pulse oximetry improves the prediction of fetal acidbase condition † fetal pulse oximetry will reduce the cesarean rate for non-reassuring EFM patterns but, as currently used, will not decrease the overall cesarean rate † new FDA guidelines expand EFM patterns to be considered ominous, requiring prompt delivery despite reassuring fetal oximetry readings
Research agenda † to establish whether there is a way to employ fetal pulse oximetry effectively in the setting of labor dystocia and a non-reassuring fetal heart rate pattern to allow safe vaginal delivery † to determine whether other mechanisms of intrapartum fetal brain injury besides hypoxemia (e.g. infection, ischemia) could escape surveillance by fetal pulse oximetry † to establish the long-term outcome of children monitored with fetal pulse oximetry, compared to standard methods
ACKNOWLEDGEMENTS Thanks to Suzanne Harper for manuscript preparation.
REFERENCES 1. Hon EH. Instrumentation of fetal heart rate and fetal electrocardiography. III. Fetal ECG electrodes: further observations. Obstetrics and Gynecology 1967; 30: 281–286. 2. Freeman RK. Problems with intrapartum fetal heart rate monitoring interpretation and patient management. Obstetrics and Gynecology 2002; 100: 813 –826. 3. Thacker SB, Stroup D & Chang M. Continuous electronic heart rate monitoring for fetal assessment during labor. Cochrane Database of Systematic Reviews 2001; : CD000063. 4. Clark SL & Hankins GD. Temporal and demographic trends in cerebral palsy—fact and fiction. American Journal of Obstetrics and Gynecology 2003; 188: 628– 633. 5. Saling E & Schneider D. Biochemical supervision of the foetus during labour. Journal of Obstetrics and Gynaecology of British Commonwealth 1967; 74: 799–811. 6. Clark SL & Paul RH. Intrapartum fetal surveillance: the role of fetal scalp blood sampling. American Journal of Obstetrics and Gynecology 1985; 153: 717 –720. 7. Goodwin TM, Milner-Masterson L & Paul RH. Elimination of fetal scalp blood sampling on a large clinical service. Obstetrics and Gynecology 1994; 83: 971 –974. 8. Tan KH & Smyth R. Fetal vibroacoustic stimulation for facilitation of tests of fetal wellbeing (Cochrane Review). Cochrane Database of Systematic Reviews 2001; : 1. 9. Porter TF & Clark SL. Vibroacoustic and scalp stimulation. Obstetrics and Gynecology Clinics of North America 1999; 26: 657 –669. 10. Aarnoudse JG, Huisjes HJ, Gordon H, et al. Fetal subcutaneous scalp PO2 and abnormal heart rate during labor. American Journal of Obstetrics and Gynecology 1985; 153: 565–566. 11. Bergmans MG, van Geijn HP, Weber T, et al. Fetal transcutaneous PCO2 measurements during labour. European Journal of Obstetrics, Gynecology, and Reproductive Biology 1993; 51: 1– 7.
Fetal pulse oximetry: a critical appraisal 483 12. Young BK, Katz M, Wilson SJ & Klein SA. Continuous fetal tissue pH monitoring in labor. Progress in Clinical and Biological Research 1980; 44: 123– 132. 13. Devoe L, Golde S, Kilman Y, et al. A comparison of visual analyses of intrapartum fetal heart rate tracings according to the new national institute of child health and human development guidelines with computer analyses by an automated fetal heart rate monitoring system. American Journal of Obstetrics and Gynecology 2000; 183: 361–366. 14. Westgate J, Harris M, Curnow JS & Greene KR. Plymouth randomized trial of cardiotocogram only versus STwaveform plus cardiotocogram for intrapartum monitoring in 2400 cases. American Journal of Obstetrics and Gynecology 1993; 169: 1151–1160. 15. Aldrich CJ, D’Antona D, Wyatt JS, et al. Fetal cerebral oxygenation measured by near-infrared spectroscopy shortly before birth and acid –base status at birth. Obstetrics and Gynecology 1994; 84: 861–866. 16. Dildy GA, Clark SL & Loucks CA. Intrapartum fetal pulse oximetry: past, present, and future. American Journal of Obstetrics and Gynecology 1996; 175: 1– 9. 17. Yelderman M & New Jr W. Evaluation of pulse oximetry. Anesthesiology 1983; 59: 349–352. 18. Peat S, Booker M, Lanigan C & Ponte J. Continuous intrapartum measurement of fetal oxygen saturation [letter]. Lancet 1988; 2: 213. 19. Johnson N, Johnson VA, Bannister J & Lilford RJ. Measurement of fetal peripheral perfusion with a pulse oximeter [letter]. Lancet 1989; 1: 898. 20. Gardosi J, Carter M & Becket T. Continuous intrapartum monitoring of fetal oxygen saturation [letter] [see comments]. Lancet 1989; 2: 692 –693. 21. Dildy GA, Clark SL & Loucks CA. Preliminary experience with intrapartum fetal pulse oximetry in humans. Obstetrics and Gynecology 1993; 81: 630 –635. 22. Carbonne B, Audibert F, Segard L, et al. Fetal pulse oximetry: correlation between changes in oxygen saturation and neonatal outcome. Preliminary report on 39 cases. European Journal of Obstetrics, Gynecology, and Reproductive Biology 1994; 57: 73– 77. 23. Luttkus A, Fengler TW, Friedmann W & Dudenhausen JW. Continuous monitoring of fetal oxygen saturation by pulse oximetry. Obstetrics and Gynecology 1995; 85: 183– 186. 24. East CE & Colditz PB. Women’s evaluations of their experience with fetal intrapartum oxygen saturation monitoring and participation in a research project. Midwifery 1996; 12: 93–97. 25. Yam J, Chua S & Arulkumaran S. Intrapartum fetal pulse oximetry. Part I. Principles and technical issues. Obstetrical and Gynecological Survey 2000; 55: 163–172. 26. Yam J, Chua S & Arulkumaran S. Intrapartum fetal pulse oximetry. Part 2. Clinical application. Obstetrical and Gynecological Survey 2000; 55: 173 –183. 27. Richardson BS, Carmichael L, Homan J & Patrick JE. Electrocortical activity, electroocular activity, and breathing movements in fetal sheep with prolonged and graded hypoxemia. American Journal of Obstetrics and Gynecology 1992; 167: 553–558. 28. Seelbach-Gobel B, Heupel M, Kuhnert M & Butterwegge M. The prediction of fetal acidosis by means of intrapartum fetal pulse oximetry. American Journal of Obstetrics and Gynecology 1999; 180: 73–81. 29. Kuhnert M, Seelbach-Goebel B & Butterwegge M. Predictive agreement between the fetal arterial oxygen saturation and fetal scalp pH: results of the German multicenter study [see comments]. American Journal of Obstetrics and Gynecology 1998; 178: 330–335. 30. Dildy GA, Thorp JA, Yeast JD & Clark SL. The relationship between oxygen saturation and pH in umbilical blood: implications for intrapartum fetal oxygen saturation monitoring. American Journal of Obstetrics and Gynecology 1996; 175: 682 –687. 31. Garite TJ, Dildy GA, McNamara H, et al. A multicenter controlled trial of fetal pulse oximetry in the intrapartum management of nonreassuring fetal heart rate patterns. American Journal of Obstetrics and Gynecology 2000; 183: 1049–1058. 32. Ingemarsson E, Ingemarsson I, Solum T & Westgren M. Influence of occiput posterior position on the fetal heart rate pattern. Obstetrics and Gynecology 1980; 55: 301–304. 33. Porreco RP, Boehm FH, Dildy GA, et al. Dystocia in nulliparous patients monitored with fetal pulse oximetry. American Journal of Obstetrics and Gynecology 2004; 190: 113 –117. 34. ACOG Committee Opinion, Number 258, September 2001. Fetal pulse oximetry. Obstetrics and Gynecology 2001; 98: 523–524. 35. Liston R, Crane J, Hamilton E, et al. Fetal health surveillance in labour. Journal of Obstetrics and Gynaecology Canada 2002; 24: 250 –276. quiz 277 –280. 36. Liston R, Crane J, Hughes O, et al. Fetal health surveillance in labour. Journal of Obstetrics and Gynaecology Canada 2002; 24: 342 –355. 37. Thorp JA. Is fetal pulse oximetry ready for clinical practice? Writing for the PRO position. MCN. The American Journal of Maternal Child Nursing 2003; 28: 64.
484 G. A. Dildy III 38. Schifrin BS. Is fetal pulse oximetry ready for clinical practice? Writing for the CON position. MCN. The American Journal of Maternal Child Nursing 2003; 28: 65. 39. Schram CM & Gardosi JO. Artifacts in fetal pulse oximetry: nonarterial pulsatile signals. American Journal of Obstetrics and Gynecology 1994; 170: 1174–1177. 40. Gardosi JO, Damianou D & Schram CM. Artifacts in fetal pulse oximetry: incomplete sensor-to-skin contact. American Journal of Obstetrics and Gynecology 1994; 170: 1169–1173. 41. Nijland R, Jongsma HW, Nijhuis JG & Oeseburg B. Accuracy of fetal pulse oximetry and pitfalls in measurements. European Journal of Obstetrics, Gynecology, Reproductive Biology 1997; 72: S21–S27. 42. Mannheimer PD. Design and validation of pulse oximetry for low saturation. Anesthesia and Analgesia 2002; 94: S21– S25. 43. Alshimmiri M, Bocking AD, Gagnon R, et al. Prediction of umbilical artery base excess by intrapartum fetal oxygen saturation monitoring. American Journal of Obstetrics and Gynecology 1997; 177: 775 –779.