Accepted Manuscript Future perspectives in intrapartum fetal surveillance Lawrence D. Devoe, M.D., Professor Emeritus
PII:
S1521-6934(15)00110-8
DOI:
10.1016/j.bpobgyn.2015.06.006
Reference:
YBEOG 1518
To appear in:
Best Practice & Research Clinical Obstetrics & Gynaecology
Received Date: 15 June 2015 Accepted Date: 18 June 2015
Please cite this article as: Devoe LD, Future perspectives in intrapartum fetal surveillance, Best Practice & Research Clinical Obstetrics & Gynaecology (2015), doi: 10.1016/j.bpobgyn.2015.06.006. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Author:
Lawrence D. Devoe, M.D.
M AN U
Professor Emeritus
SC
RI PT
Title: Future perspectives in intrapartum fetal surveillance
Department of Obstetrics and Gynecology Medical College of Georgia Georgia Regents University
Correspondence:
TE D
Augusta, Georgia USA
EP
Lawrence D. Devoe, M.D.
AC C
Department of Obstetrics and Gynecolocy 1120 15th Street
Augusta, Georgia 30907 USA Phone: (706) 721-3556 Email: [email protected]
ACCEPTED MANUSCRIPT
Abstract
RI PT
Electronic fetal monitoring (EFM) has aided intrapartum fetal surveillance for more than four decades. In spite of numerous trials that have compared EFM with standard fetal
heart rate (FHR) auscultation, it remains unclear that this modality has led to improved
SC
perinatal outcomes, specifically lower rates of perinatal morbidity and mortality. A
variety of ancillary methods have been developed to improve the accuracy of EFM for
M AN U
prediction of fetal compromise. At present, a limited number of studies have shown that the addition of fetal ECG analysis to visual interpretation of FHR patterns has resulted in better fetal outcomes. However, the shortcomings of visual interpretation of FHR patterns persist and, while automated systems for FHR analysis have been developed,
TE D
such systems have not been widely used or proven to enhance the value of intrapartum fetal surveillance. This chapter will discuss future directions for novel intrapartum fetal surveillance systems that leverage the long experience gained from EFM to provide a
EP
higher level of risk assessment and prognosis.
AC C
Key words: computer, intelligent systems, electronic fetal monitoring, risk assessment
ACCEPTED MANUSCRIPT
Introduction and Background Electronic fetal monitoring (EFM) is entering its fifth decade in intrapartum care. In spite of the best of intentions, standard EFM using unaided visual interpretation of fetal heart
RI PT
rate (FHR) patterns has yet to demonstrate improved perinatal outcomes such as lower
rates of perinatal mortality or morbidity. (1) Recognized shortcomings of EFM include but are not limited to the reliability and reproducibility of FHR pattern recognition as
M AN U
interpretation among experienced clinicians. (2)
SC
demonstrated by numerous studies that have compared agreement levels for FHR
To begin to address some of the known shortcomings of EFM, the National Institute of Child Health and Human Development (NICHD) convened an American expert panel to develop standardized terminology for visual interpretation of FHR. The deliberations of
TE D
this panel, as well as those of a subsequent similar expert panel, resulted in an American College of Obstetricians and Gynecologists (ACOG) Practice Bulletin that defined a three-tier system for classifying FHR patterns. (3) To date, no rigorous studies have been
EP
performed to demonstrate that this classification scheme has resulted in improved
AC C
implementation of EFM or better obstetric outcomes.
Recognizing that the best opportunity to improve the effectiveness of EFM might begin with taking the “human factor” out of the equation, efforts to develop automated systems for FHR pattern recognition have been ongoing since the 1970s. (4). With improved computer operating systems, software and firmware, the development of such systems has continued (5-7), leading to mature products that are now being vended and used in
ACCEPTED MANUSCRIPT
selected obstetric units around the world. Figure 1 shows typical block-diagram architecture of such an analytic system. Concomitant with the appearance of such sophisticated analytic systems has been the development of a new generation of
RI PT
noninvasive transducers for obtaining both fetal and maternal signals. [8] The potential
benefit of more sensitive and accurate surface probes is a reduction in the “noise” of the system, thereby improving the quality of signal processing and analysis. Does the use of
SC
automated FHR analysis lead to better perinatal outcomes? A large randomized controlled trial, INFANT, was launched in the United Kingdom and Ireland
M AN U
(http://www.ucl.ac.uk/cctu/researchareas/womenshealth/infant) several years ago. Anticipated enrollment was to be 46,000 patients randomized to either a decision-support arm (automated FHR alerts available) or to a control arm (alerts not available). To date,
TE D
the final results of this trial have not been published.
It may well have been short sighted to consider that automation of FHR interpretation alone would improve intrapartum care and provide healthier infants. To overcome the
EP
historic limitations of unaided use of EFM, there have been numerous attempts to implement adjunct technologies, including fetal scalp blood sampling, fetal pulse
AC C
oximetry, and, most recently fetal ECG ST-segment analysis (9). The latter technology has shown much promise in European studies (10) but a large United States randomized controlled trial failed to show similar positive results. (11)
Future Perspectives in Fetal Surveillance Systems
ACCEPTED MANUSCRIPT
To forecast the future of intrapartum fetal surveillance, one must build on the large collective experience of decades of clinical investigation. Figure 2 shows the developmental stages of fetal surveillance systems as their degree of sophistication is
RI PT
increased. In Stage 1, the fetal monitor presents the raw data of the FHR and uterine activity to clinicians who must proceed to interpret these data and to apply their
interpretations to subsequent intrapartum care. In Stage 2, the fetal monitor provides
SC
automated FHR interpretation and/or higher-level alerts (v. Figure 1) that will be
incorporated in fetal status assessment and prognosis. In Stage 3, the fetal monitor
M AN U
combines additional clinical information about the patient with the current status of the automated alerting system and proposes a potential care plan for the patient. Stage 1 was accomplished at the introduction of EFM. Stage 2 has been more recently accomplished and such monitoring systems are currently available and have been used for intrapartum
TE D
surveillance in selected obstetric units. Stage 3 is the next horizon for future generation electronic fetal monitors and possible features of such systems will occupy the remainder
EP
of this monograph.
What might we anticipate in future intrapartum surveillance systems? As alluded to
AC C
earlier, electronic fetal monitors were originally developed to act as decision support tools. However, in so doing, undue emphasis was placed on FHR pattern interpretation alone. Advanced decision support systems would need to leverage what is now known about the prognostic value of specific FHR patterns that would be imposed on large knowledge bases as will be described later. The systems of the future will need to be
ACCEPTED MANUSCRIPT
user-friendly and provide displays that may incorporate the traditional EFM elements but add prognostic information that goes beyond basic alerting.
RI PT
Silent Alert Systems
Obstetrical units are often noisy and chaotic places where the conditions of individual
patients are subject to sudden and unpredicted change. Current monitoring systems link
SC
numerous individual devices to central stations that provide a continuing array of FHR data. Not only can this be an overwhelming setting but a setting that bombards care
M AN U
providers with large quantities of information that are often not clinically important. One possible solution is taken from a model that has been used in the banking business for decades: the silent alarm. Figure 3 analogizes how such systems might function in a busy labor unit. Using this model would keep fetal monitoring systems operating in the
TE D
background until a situation developed where critical alerts would need to be triggered. Currently available telemetry devices, enabling the physical monitors to reside outside of patient rooms, could support these “silent” systems. A potentially important downstream
EP
benefit from silent systems would be measureable reduction in stress levels for both patients and their caregivers who would remain undistracted by the often needless
AC C
intrusion of technologies.
Intelligent intrapartum surveillance systems Given the constantly changing intrauterine environment that characterizes active labor, the ability to evaluate fetal condition and to predict its course is currently well beyond the capability of conventional monitoring systems. Intrapartum conditions are not unlike
ACCEPTED MANUSCRIPT
those that are addressed routinely by the United States’ National Oceanic and Atmospheric Administration (NOAA) (http://www.noaa.gov). Equipped with a series of environmental satellites, courtesy of the United States’ National Aeronautics and Space
RI PT
Administration, NOAA has banks of intelligent computers that can predict natural
disasters, and, in some instances, render alerts far enough in advance to enable mass evacuations that could save thousands of lives. Key features of these predictive and
SC
alerting systems include the ability to identify and analyze specific climate patterns that are associated with severe weather events. At the backbone of such systems are hybrid
M AN U
software programs that employ algorithms, expert systems, and neural networks.
In attempting to deal with all of the environmental variables that occur during labor, there have been various efforts to develop computerized solutions using any and all of the
TE D
above methods [12]. While the mathematical formulae of algorithmic programs can address individual elements present during the course of labor, they are readily overwhelmed when presented with the massive number of varying data points that are in
EP
constant flux. Expert systems are often rule-based and, unlike algorithm-based systems, contain rules capable of recognizing a number of events that are occurring in real time. In
AC C
fact, a number of the automated FHR interpretative programs have used expert systems to recognize FHR patterns and to classify them (4). However, expert systems come up short when they are presented with events or conditions for which the rules are lacking or insufficient.
ACCEPTED MANUSCRIPT
Neural networks are a form of “intelligent” software that, rather than using algorithms or rules, look for associations among variables via an application of matrix algebra and derive solutions that can determine the probability of events of interest. Unlike expert
RI PT
systems, neural network solutions have the continuous capacity to improve through
“learning” by seeing increasing numbers of case examples. The downside of neural
networks is that they do not readily identify the importance of individual variables by
SC
measuring the magnitude of their effects. While there have been relatively few examples of neural networks developed to solve obstetric problems, our group successfully piloted
M AN U
such an approach to prediction of perinatal outcome associated with nonstress testing (Figure 4) (13). Of the model systems that are described below, the computerized solutions will most likely incorporate hybrid programs that may utilize more than one of
TE D
the above approaches.
Rapid Response Systems
As in automobile travel, serious accidents can also happen in an obstetrical unit.
EP
Fortunately, such accidents are infrequent but when they do occur, they can prove life threatening to both mother and child. While not every intrapartum accident can be
AC C
predicted far in advance, there are circumstances in which the events that precede them could be detected, for example, progressively worsening umbilical cord constriction that might ultimately lead to respiratory acidosis and even intrauterine death, if not relieved. Monitoring systems capable of modeling such events could provide sufficient warning in the same manner as conditions that precede automobile collisions, so that preemptive action, such as patient repositioning or amnioinfusion, could be undertaken (Figure 5).
ACCEPTED MANUSCRIPT
Risk Assessment Systems
RI PT
Intrapartum surveillance is intended to provide continuous assessment of fetal condition and to alert those providing intrapartum care when the risk for fetal compromise
increases. Unfortunately, given the lack of specificity of FHR patterns, assessing such
SC
changes in fetal risk has proved to be an elusive target. Contributing to this problem is
the inability of current monitoring systems to integrate the extensive clinical database for
M AN U
any given pregnancy that establishes an a priori level of risk present on entry into the labor unit and how it may be modified by changes in FHR patterns. Levels of risk for fetal compromise may change frequently during the course of labor. Both immediate and progressive alterations in such risk (v. Figures 6 and 7) should be made available to those providing care. Fortunately, high-level solutions for risk assessment and prediction are
TE D
now more readily available in clinical medicine than they have been in the past. Systems like IBM’s Watson (http://www.ibm.com/smarterplanet/us/en/ibmwatson) confer the ability to handle vast amounts of data in real-time and is being used in various areas of
EP
diagnosis and treatment, such as medical oncology. Looking ahead, it would seem
AC C
logical that intrapartum surveillance would be another clinical area for such “intelligent” systems to be developed and implemented.
Barriers to the development of new intrapartum surveillance systems
New technologies in clinical obstetrics have had a checkered past. Migrating great ideas to real-life clinical applications litter the highways of progress and, in this regard,
ACCEPTED MANUSCRIPT
intrapartum surveillance is no exception. Systems that provided continuous scalp pH or fetal oximetry [14] are examples of good ideas that foundered either due to the failure of early adoption or lack of support from clinical trials. The rate-limiting step for the
RI PT
research, development, and acceptance of new technologies underscores the obvious
reluctance of industry to invest in novel methods of clinical assessment. Nowhere is this more apparent than in the field of intrapartum fetal surveillance. Regardless of the new
SC
cosmetics for the monitoring machines and web-based communication of intrapartum
data, the basic functionality of electronic fetal monitors has changed little, if at all, in the
M AN U
past four decades.
The hard realities faced by obstetricians and midwives who desire better monitoring systems must deal with how industry looks at the development of innovative medical
TE D
systems. Validation of new systems, software and firmware is extremely time-, labor-, and cost-intensive. The early days of EFM often witnessed shoestring business operations with no expectation that such medical devices would ever recover research
EP
and development costs let alone turn a profit. Those days are over and, in today’s world, the introduction of a novel monitoring system would have to pass through a number of
AC C
daunting hurdles. The first of these hurdles would involve a pilot study showing that the system worked as described and satisfied the requirements of patient safety to enable approval by the Federal Drug Administration (in the United States). Most likely, such a study would be followed by a randomized trial comparing the new system to the so-called “gold standard.” Should this hurdle be successfully negotiated, the company that developed this system would then need to incur a lengthy and costly program of
ACCEPTED MANUSCRIPT
introduction, marketing, and education of a new generation of end-users, a process that can often take years. Assuming that a group of early adopters emerges, acceptance by the
wider adoption of novel devices into practice is years.
RI PT
general obstetric community would need to follow. Again, the measurement standard for
New products engender potential medico-legal liability implications that involve both the
SC
users and the companies that provide the new systems. [15] Companies willing to take on such a risky business must have deep enough pockets to escrow monies for legal
M AN U
issues, should they arise. A final and most important question is whether a company willing to invest in new technologies would be dissuaded by their ability to recover their investment and show a profit.
TE D
Summary
In conclusion, the nature of future advances in intrapartum surveillance, if any, remains uncertain. It is obvious that standard EFM is a limited technology at its very best and
EP
that every effort to improve the reliability of visual interpretation through education and automation, laudable as it may be, will not take this modality to the next level. Potential
AC C
avenues for improving the value of intrapartum fetal surveillance are clear and this monograph has highlighted a few of them. The pathway toward better and smarter intrapartum surveillance systems will almost certainly require the mating of conventional fetal monitors with smart computer systems that can acquire, analyze, and prioritize the vast data array that a single labor can generate. The next step forward will require a very bold move and one for which, hopefully, we will not have to wait another four decades.
ACCEPTED MANUSCRIPT
Practice points •
Current EFM systems that rely on unaided visual interpretation for intrapartum surveillance have limited ability to detect fetal compromise. With the exception of fetal monitors that add fetal ECG analysis to standard FHR
RI PT
•
interpretation, ancillary methods to improve the accuracy of EFM have failed to improve perinatal outcomes.
Automated systems for FHR analysis and interpretation are now available but
SC
•
have yet to be linked to better perinatal outcomes
The development of advanced fetal monitoring systems is essential if significant
M AN U
•
progress in improving the accuracy of intrapartum surveillance and timeliness of obstetric interventions when fetal compromise is detected.
•
TE D
Research agenda
Development of intelligent fetal monitoring systems that can interpret FHR patterns accurately and determine current and projected risk for fetal compromise
•
EP
by including other relevant clinical data. Comparative trials of such systems with standard fetal monitoring systems that
AC C
rely on visual interpretation of FHR patterns.
REFERENCES
[1] Alfirevic Z, Devane D, Gyte GM. Continuous cardiotocography (CTG) as a form of electronic fetal monitoring (EFM) for fetal assessment during labour. Cochrane Database
ACCEPTED MANUSCRIPT
Syst Rev 2006;3:CD006066.* [2] Devoe L, Golde S, Kilman Y, Morton D, Shea K, Waller J. A comparison of visual analyses of intrapartum fetal heart rate tracings according to the new National Institute of
RI PT
Child Health and Human Development guidelines with computer analyses by an
automated fetal heart rate monitoring system. Am J Obstet Gynecol 2001;183:361-6.* [3] American College of Obstetricians and Gynecologists. Intrapartum fetal heart rate
SC
monitoring: nomenclature, interpretation, and general management principles. ACOG Practice Bulletin no. 106. Washington DC: ACOG; 2009*.
M AN U
[4] Alonso-Betanzos A, Moret-Bonillo V, Devoe L, Searle J, Banias B, Ramos E. Computerized antenatal assessment: The NST-Expert Project. Automedica 1992 14:3-22.*
[5] Bernardes J, Ayres-de-Campos D, Costa-Pereira A, Pereira-Leite L, Garrido A.
TE D
Objective computerized fetal heart rate analysis. Int J Gynecol Obstet 1998;62:141-7 [6] Hamilton EF, Warrick PA. New perspectives in electronic fetal surveillance. J Perinat Med. 2012;0:1-10.*
EP
[7] Warrick PA, Hamilton EF, Precup D, Kearney RE. Classification of normal and hypoxic fetuses from systems modeling of intrapartum of intrapartum cardiotocography.
AC C
IEEE Trans Biomed Eng. 2010;57:771-779. [8] Wolfberg A. The future of fetal monitoring. Rev Obstet Gynecol. 2012;5(3/4):e132e136 doi: 10.3909/riog0197* [9] Amer-Wåhlin I, Yli B, Arulkumaran S. Foetal ECG and STAN technology – a review. Eur Clinics Obstet Gynaecol (2005) 1: 61–73.*
ACCEPTED MANUSCRIPT
[10] Norén H, Carlsson A. Reduced prevalence of metabolic acidosis at birth: an established STAN usage in the total population of deliveries in a Swedish district hospital. Am J Obstet Gynecol 2010;202:546.e1-7.
RI PT
[11]. Saade, G. Fetal ECG analysis of the ST segment as an adjunct to intrapartum fetal
heart rate monitoring: a randomized clinical trial. Am J Obstet Gynecol 2015; 212, Issue 1, Supplement, Page S52
SC
[12] Devoe, L.D. Searle N, Castillo RA, Searle J. Prognostic components of
M AN U
computerized fetal biophysical testing. Am J Obstet Gynecol 1988; 158:1144-1148.*
[13] Devoe LD: Computerized FHR Analysis and Neural Networks in Antepartum Fetal Surveillance. Current Opinion in Ob/Gyn 1996: Vol. 8, 119-122*
[14] Bloom SL, Spong CY, Thom E, Varner MW, et al. Fetal pulse oximetry and
TE D
cesarean delivery. N Eng J Med. 2006; 355: 2195-2002*
[15] Richards EP, The Supreme Court Rules on Medical Device Liability - Or Does It?
AC C
EP
IEEE Engineering In Medicine And Biology Magazine 1997; 16: 87
Conflict of Interest Statement:
Dr. Devoe is a paid consultant for Neoventa Medical, AB, Molndal, Sweden.
RI PT
ACCEPTED MANUSCRIPT
SC
Figures
M AN U
1. Schematic block-diagram of an automated fetal monitor alerting system
2. Stages of development of intrapartum surveillance systems
3. Silent alarm systems
TE D
4. Model for neural network
5. Characteristics of rapid response systems
EP
6. Features of a novel intrapartum risk assessment and alerting system
AC C
7. Graphic display of risk trends displayed during labor
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Stage 1
Stage 2
Conventional
MACHINE Local Diagnostics
Global Diagnostics
Stan S31
Dx fetal condition
FHR Monitor
SC
PeriCalm
RI PT
MACHINE
Stage 3 MACHINE
Propose Care Plan
M AN U
TraceVue
Alerting/Rules Data Flow
Data Flow
DONE Local Dx External Factors Prognosis
AC C
Global Dx
EP
FHR Tracing
TE D
DONE
Data Flow
FUTURE
Global Dx
Inform machine
External
of external
Factors
factors, control
Prognosis
thresholds, etc
Tasks left for Clinicians
Invisible unless a problem is detected: Silent Bank Alarms “Silent” Alarms will not distract or interfere with other essential care elements 2. Patient should not be held “hostage” by system: Wireless transmission (Monica) Machines do not need to be in same room to provide alerts
AC C
EP
TE D
M AN U
SC
1.
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Gestation Risk Age Factor
Mat Age
Testing Interval
React
RI PT
Layer
M AN U
SC
INPUT
AC C
OUTPUT
EP
TE D
HIDDEN
Perinatal Outcomes
Race
Parity
HISTORIC EXAMPLE: THE CARDIFF PUMP
SC
M AN U
Intelligent systems that sense and respond to sudden changes in maternal or fetal status
RI PT
ACCEPTED MANUSCRIPT
EP
TE D
Feed-back loop regulation of oxytocin infusion: automatic shut-off for abnormal CTG
AC C
CURRENT EXAMPLE: VOLVO CARS
Systems of sensors are already used in automobile technology to monitor fuel mixture operating temperature, tire pressure, and ……
Collision Risk
RI PT
ACCEPTED MANUSCRIPT
M AN U TE D
EP
Prediction of risk based on post-test likelihood : (How much does monitoring data alter risk of hypoxia or acidosis).
SC
Baseline assessment of fetal risk Linking to probabilistic database of known fetal risk factors to current physiologic status (EFM)
AC C
Alerts trigger clinical prompts: Present best practice alternatives for patient care (evidence-based)
ACCEPTED MANUSCRIPT
RI PT
Estimated Risk Over Time
M AN U TE D EP AC C
Fetal Risk Status
SC
Time Course of Labor
ACCEPTED MANUSCRIPT
Highlights for Future Perspectives in Intrapartum Surveillance
• •
AC C
EP
TE D
M AN U
SC
•
Standard EFM is a limited prognostic tool Adjunctive assessment tools have not consistently improved EFM performance Future surveillance systems are possible with today’s technologies Such systems will need to blend current EFM systems with intelligent computers Industry must be willing to take on the risk of developing such novel systems
RI PT
• •
PII:
S1521-6934(15)00110-8
DOI:
10.1016/j.bpobgyn.2015.06.006
Reference:
YBEOG 1518
To appear in:
Best Practice & Research Clinical Obstetrics & Gynaecology
Received Date: 15 June 2015 Accepted Date: 18 June 2015
Please cite this article as: Devoe LD, Future perspectives in intrapartum fetal surveillance, Best Practice & Research Clinical Obstetrics & Gynaecology (2015), doi: 10.1016/j.bpobgyn.2015.06.006. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Author:
Lawrence D. Devoe, M.D.
M AN U
Professor Emeritus
SC
RI PT
Title: Future perspectives in intrapartum fetal surveillance
Department of Obstetrics and Gynecology Medical College of Georgia Georgia Regents University
Correspondence:
TE D
Augusta, Georgia USA
EP
Lawrence D. Devoe, M.D.
AC C
Department of Obstetrics and Gynecolocy 1120 15th Street
Augusta, Georgia 30907 USA Phone: (706) 721-3556 Email: [email protected]
ACCEPTED MANUSCRIPT
Abstract
RI PT
Electronic fetal monitoring (EFM) has aided intrapartum fetal surveillance for more than four decades. In spite of numerous trials that have compared EFM with standard fetal
heart rate (FHR) auscultation, it remains unclear that this modality has led to improved
SC
perinatal outcomes, specifically lower rates of perinatal morbidity and mortality. A
variety of ancillary methods have been developed to improve the accuracy of EFM for
M AN U
prediction of fetal compromise. At present, a limited number of studies have shown that the addition of fetal ECG analysis to visual interpretation of FHR patterns has resulted in better fetal outcomes. However, the shortcomings of visual interpretation of FHR patterns persist and, while automated systems for FHR analysis have been developed,
TE D
such systems have not been widely used or proven to enhance the value of intrapartum fetal surveillance. This chapter will discuss future directions for novel intrapartum fetal surveillance systems that leverage the long experience gained from EFM to provide a
EP
higher level of risk assessment and prognosis.
AC C
Key words: computer, intelligent systems, electronic fetal monitoring, risk assessment
ACCEPTED MANUSCRIPT
Introduction and Background Electronic fetal monitoring (EFM) is entering its fifth decade in intrapartum care. In spite of the best of intentions, standard EFM using unaided visual interpretation of fetal heart
RI PT
rate (FHR) patterns has yet to demonstrate improved perinatal outcomes such as lower
rates of perinatal mortality or morbidity. (1) Recognized shortcomings of EFM include but are not limited to the reliability and reproducibility of FHR pattern recognition as
M AN U
interpretation among experienced clinicians. (2)
SC
demonstrated by numerous studies that have compared agreement levels for FHR
To begin to address some of the known shortcomings of EFM, the National Institute of Child Health and Human Development (NICHD) convened an American expert panel to develop standardized terminology for visual interpretation of FHR. The deliberations of
TE D
this panel, as well as those of a subsequent similar expert panel, resulted in an American College of Obstetricians and Gynecologists (ACOG) Practice Bulletin that defined a three-tier system for classifying FHR patterns. (3) To date, no rigorous studies have been
EP
performed to demonstrate that this classification scheme has resulted in improved
AC C
implementation of EFM or better obstetric outcomes.
Recognizing that the best opportunity to improve the effectiveness of EFM might begin with taking the “human factor” out of the equation, efforts to develop automated systems for FHR pattern recognition have been ongoing since the 1970s. (4). With improved computer operating systems, software and firmware, the development of such systems has continued (5-7), leading to mature products that are now being vended and used in
ACCEPTED MANUSCRIPT
selected obstetric units around the world. Figure 1 shows typical block-diagram architecture of such an analytic system. Concomitant with the appearance of such sophisticated analytic systems has been the development of a new generation of
RI PT
noninvasive transducers for obtaining both fetal and maternal signals. [8] The potential
benefit of more sensitive and accurate surface probes is a reduction in the “noise” of the system, thereby improving the quality of signal processing and analysis. Does the use of
SC
automated FHR analysis lead to better perinatal outcomes? A large randomized controlled trial, INFANT, was launched in the United Kingdom and Ireland
M AN U
(http://www.ucl.ac.uk/cctu/researchareas/womenshealth/infant) several years ago. Anticipated enrollment was to be 46,000 patients randomized to either a decision-support arm (automated FHR alerts available) or to a control arm (alerts not available). To date,
TE D
the final results of this trial have not been published.
It may well have been short sighted to consider that automation of FHR interpretation alone would improve intrapartum care and provide healthier infants. To overcome the
EP
historic limitations of unaided use of EFM, there have been numerous attempts to implement adjunct technologies, including fetal scalp blood sampling, fetal pulse
AC C
oximetry, and, most recently fetal ECG ST-segment analysis (9). The latter technology has shown much promise in European studies (10) but a large United States randomized controlled trial failed to show similar positive results. (11)
Future Perspectives in Fetal Surveillance Systems
ACCEPTED MANUSCRIPT
To forecast the future of intrapartum fetal surveillance, one must build on the large collective experience of decades of clinical investigation. Figure 2 shows the developmental stages of fetal surveillance systems as their degree of sophistication is
RI PT
increased. In Stage 1, the fetal monitor presents the raw data of the FHR and uterine activity to clinicians who must proceed to interpret these data and to apply their
interpretations to subsequent intrapartum care. In Stage 2, the fetal monitor provides
SC
automated FHR interpretation and/or higher-level alerts (v. Figure 1) that will be
incorporated in fetal status assessment and prognosis. In Stage 3, the fetal monitor
M AN U
combines additional clinical information about the patient with the current status of the automated alerting system and proposes a potential care plan for the patient. Stage 1 was accomplished at the introduction of EFM. Stage 2 has been more recently accomplished and such monitoring systems are currently available and have been used for intrapartum
TE D
surveillance in selected obstetric units. Stage 3 is the next horizon for future generation electronic fetal monitors and possible features of such systems will occupy the remainder
EP
of this monograph.
What might we anticipate in future intrapartum surveillance systems? As alluded to
AC C
earlier, electronic fetal monitors were originally developed to act as decision support tools. However, in so doing, undue emphasis was placed on FHR pattern interpretation alone. Advanced decision support systems would need to leverage what is now known about the prognostic value of specific FHR patterns that would be imposed on large knowledge bases as will be described later. The systems of the future will need to be
ACCEPTED MANUSCRIPT
user-friendly and provide displays that may incorporate the traditional EFM elements but add prognostic information that goes beyond basic alerting.
RI PT
Silent Alert Systems
Obstetrical units are often noisy and chaotic places where the conditions of individual
patients are subject to sudden and unpredicted change. Current monitoring systems link
SC
numerous individual devices to central stations that provide a continuing array of FHR data. Not only can this be an overwhelming setting but a setting that bombards care
M AN U
providers with large quantities of information that are often not clinically important. One possible solution is taken from a model that has been used in the banking business for decades: the silent alarm. Figure 3 analogizes how such systems might function in a busy labor unit. Using this model would keep fetal monitoring systems operating in the
TE D
background until a situation developed where critical alerts would need to be triggered. Currently available telemetry devices, enabling the physical monitors to reside outside of patient rooms, could support these “silent” systems. A potentially important downstream
EP
benefit from silent systems would be measureable reduction in stress levels for both patients and their caregivers who would remain undistracted by the often needless
AC C
intrusion of technologies.
Intelligent intrapartum surveillance systems Given the constantly changing intrauterine environment that characterizes active labor, the ability to evaluate fetal condition and to predict its course is currently well beyond the capability of conventional monitoring systems. Intrapartum conditions are not unlike
ACCEPTED MANUSCRIPT
those that are addressed routinely by the United States’ National Oceanic and Atmospheric Administration (NOAA) (http://www.noaa.gov). Equipped with a series of environmental satellites, courtesy of the United States’ National Aeronautics and Space
RI PT
Administration, NOAA has banks of intelligent computers that can predict natural
disasters, and, in some instances, render alerts far enough in advance to enable mass evacuations that could save thousands of lives. Key features of these predictive and
SC
alerting systems include the ability to identify and analyze specific climate patterns that are associated with severe weather events. At the backbone of such systems are hybrid
M AN U
software programs that employ algorithms, expert systems, and neural networks.
In attempting to deal with all of the environmental variables that occur during labor, there have been various efforts to develop computerized solutions using any and all of the
TE D
above methods [12]. While the mathematical formulae of algorithmic programs can address individual elements present during the course of labor, they are readily overwhelmed when presented with the massive number of varying data points that are in
EP
constant flux. Expert systems are often rule-based and, unlike algorithm-based systems, contain rules capable of recognizing a number of events that are occurring in real time. In
AC C
fact, a number of the automated FHR interpretative programs have used expert systems to recognize FHR patterns and to classify them (4). However, expert systems come up short when they are presented with events or conditions for which the rules are lacking or insufficient.
ACCEPTED MANUSCRIPT
Neural networks are a form of “intelligent” software that, rather than using algorithms or rules, look for associations among variables via an application of matrix algebra and derive solutions that can determine the probability of events of interest. Unlike expert
RI PT
systems, neural network solutions have the continuous capacity to improve through
“learning” by seeing increasing numbers of case examples. The downside of neural
networks is that they do not readily identify the importance of individual variables by
SC
measuring the magnitude of their effects. While there have been relatively few examples of neural networks developed to solve obstetric problems, our group successfully piloted
M AN U
such an approach to prediction of perinatal outcome associated with nonstress testing (Figure 4) (13). Of the model systems that are described below, the computerized solutions will most likely incorporate hybrid programs that may utilize more than one of
TE D
the above approaches.
Rapid Response Systems
As in automobile travel, serious accidents can also happen in an obstetrical unit.
EP
Fortunately, such accidents are infrequent but when they do occur, they can prove life threatening to both mother and child. While not every intrapartum accident can be
AC C
predicted far in advance, there are circumstances in which the events that precede them could be detected, for example, progressively worsening umbilical cord constriction that might ultimately lead to respiratory acidosis and even intrauterine death, if not relieved. Monitoring systems capable of modeling such events could provide sufficient warning in the same manner as conditions that precede automobile collisions, so that preemptive action, such as patient repositioning or amnioinfusion, could be undertaken (Figure 5).
ACCEPTED MANUSCRIPT
Risk Assessment Systems
RI PT
Intrapartum surveillance is intended to provide continuous assessment of fetal condition and to alert those providing intrapartum care when the risk for fetal compromise
increases. Unfortunately, given the lack of specificity of FHR patterns, assessing such
SC
changes in fetal risk has proved to be an elusive target. Contributing to this problem is
the inability of current monitoring systems to integrate the extensive clinical database for
M AN U
any given pregnancy that establishes an a priori level of risk present on entry into the labor unit and how it may be modified by changes in FHR patterns. Levels of risk for fetal compromise may change frequently during the course of labor. Both immediate and progressive alterations in such risk (v. Figures 6 and 7) should be made available to those providing care. Fortunately, high-level solutions for risk assessment and prediction are
TE D
now more readily available in clinical medicine than they have been in the past. Systems like IBM’s Watson (http://www.ibm.com/smarterplanet/us/en/ibmwatson) confer the ability to handle vast amounts of data in real-time and is being used in various areas of
EP
diagnosis and treatment, such as medical oncology. Looking ahead, it would seem
AC C
logical that intrapartum surveillance would be another clinical area for such “intelligent” systems to be developed and implemented.
Barriers to the development of new intrapartum surveillance systems
New technologies in clinical obstetrics have had a checkered past. Migrating great ideas to real-life clinical applications litter the highways of progress and, in this regard,
ACCEPTED MANUSCRIPT
intrapartum surveillance is no exception. Systems that provided continuous scalp pH or fetal oximetry [14] are examples of good ideas that foundered either due to the failure of early adoption or lack of support from clinical trials. The rate-limiting step for the
RI PT
research, development, and acceptance of new technologies underscores the obvious
reluctance of industry to invest in novel methods of clinical assessment. Nowhere is this more apparent than in the field of intrapartum fetal surveillance. Regardless of the new
SC
cosmetics for the monitoring machines and web-based communication of intrapartum
data, the basic functionality of electronic fetal monitors has changed little, if at all, in the
M AN U
past four decades.
The hard realities faced by obstetricians and midwives who desire better monitoring systems must deal with how industry looks at the development of innovative medical
TE D
systems. Validation of new systems, software and firmware is extremely time-, labor-, and cost-intensive. The early days of EFM often witnessed shoestring business operations with no expectation that such medical devices would ever recover research
EP
and development costs let alone turn a profit. Those days are over and, in today’s world, the introduction of a novel monitoring system would have to pass through a number of
AC C
daunting hurdles. The first of these hurdles would involve a pilot study showing that the system worked as described and satisfied the requirements of patient safety to enable approval by the Federal Drug Administration (in the United States). Most likely, such a study would be followed by a randomized trial comparing the new system to the so-called “gold standard.” Should this hurdle be successfully negotiated, the company that developed this system would then need to incur a lengthy and costly program of
ACCEPTED MANUSCRIPT
introduction, marketing, and education of a new generation of end-users, a process that can often take years. Assuming that a group of early adopters emerges, acceptance by the
wider adoption of novel devices into practice is years.
RI PT
general obstetric community would need to follow. Again, the measurement standard for
New products engender potential medico-legal liability implications that involve both the
SC
users and the companies that provide the new systems. [15] Companies willing to take on such a risky business must have deep enough pockets to escrow monies for legal
M AN U
issues, should they arise. A final and most important question is whether a company willing to invest in new technologies would be dissuaded by their ability to recover their investment and show a profit.
TE D
Summary
In conclusion, the nature of future advances in intrapartum surveillance, if any, remains uncertain. It is obvious that standard EFM is a limited technology at its very best and
EP
that every effort to improve the reliability of visual interpretation through education and automation, laudable as it may be, will not take this modality to the next level. Potential
AC C
avenues for improving the value of intrapartum fetal surveillance are clear and this monograph has highlighted a few of them. The pathway toward better and smarter intrapartum surveillance systems will almost certainly require the mating of conventional fetal monitors with smart computer systems that can acquire, analyze, and prioritize the vast data array that a single labor can generate. The next step forward will require a very bold move and one for which, hopefully, we will not have to wait another four decades.
ACCEPTED MANUSCRIPT
Practice points •
Current EFM systems that rely on unaided visual interpretation for intrapartum surveillance have limited ability to detect fetal compromise. With the exception of fetal monitors that add fetal ECG analysis to standard FHR
RI PT
•
interpretation, ancillary methods to improve the accuracy of EFM have failed to improve perinatal outcomes.
Automated systems for FHR analysis and interpretation are now available but
SC
•
have yet to be linked to better perinatal outcomes
The development of advanced fetal monitoring systems is essential if significant
M AN U
•
progress in improving the accuracy of intrapartum surveillance and timeliness of obstetric interventions when fetal compromise is detected.
•
TE D
Research agenda
Development of intelligent fetal monitoring systems that can interpret FHR patterns accurately and determine current and projected risk for fetal compromise
•
EP
by including other relevant clinical data. Comparative trials of such systems with standard fetal monitoring systems that
AC C
rely on visual interpretation of FHR patterns.
REFERENCES
[1] Alfirevic Z, Devane D, Gyte GM. Continuous cardiotocography (CTG) as a form of electronic fetal monitoring (EFM) for fetal assessment during labour. Cochrane Database
ACCEPTED MANUSCRIPT
Syst Rev 2006;3:CD006066.* [2] Devoe L, Golde S, Kilman Y, Morton D, Shea K, Waller J. A comparison of visual analyses of intrapartum fetal heart rate tracings according to the new National Institute of
RI PT
Child Health and Human Development guidelines with computer analyses by an
automated fetal heart rate monitoring system. Am J Obstet Gynecol 2001;183:361-6.* [3] American College of Obstetricians and Gynecologists. Intrapartum fetal heart rate
SC
monitoring: nomenclature, interpretation, and general management principles. ACOG Practice Bulletin no. 106. Washington DC: ACOG; 2009*.
M AN U
[4] Alonso-Betanzos A, Moret-Bonillo V, Devoe L, Searle J, Banias B, Ramos E. Computerized antenatal assessment: The NST-Expert Project. Automedica 1992 14:3-22.*
[5] Bernardes J, Ayres-de-Campos D, Costa-Pereira A, Pereira-Leite L, Garrido A.
TE D
Objective computerized fetal heart rate analysis. Int J Gynecol Obstet 1998;62:141-7 [6] Hamilton EF, Warrick PA. New perspectives in electronic fetal surveillance. J Perinat Med. 2012;0:1-10.*
EP
[7] Warrick PA, Hamilton EF, Precup D, Kearney RE. Classification of normal and hypoxic fetuses from systems modeling of intrapartum of intrapartum cardiotocography.
AC C
IEEE Trans Biomed Eng. 2010;57:771-779. [8] Wolfberg A. The future of fetal monitoring. Rev Obstet Gynecol. 2012;5(3/4):e132e136 doi: 10.3909/riog0197* [9] Amer-Wåhlin I, Yli B, Arulkumaran S. Foetal ECG and STAN technology – a review. Eur Clinics Obstet Gynaecol (2005) 1: 61–73.*
ACCEPTED MANUSCRIPT
[10] Norén H, Carlsson A. Reduced prevalence of metabolic acidosis at birth: an established STAN usage in the total population of deliveries in a Swedish district hospital. Am J Obstet Gynecol 2010;202:546.e1-7.
RI PT
[11]. Saade, G. Fetal ECG analysis of the ST segment as an adjunct to intrapartum fetal
heart rate monitoring: a randomized clinical trial. Am J Obstet Gynecol 2015; 212, Issue 1, Supplement, Page S52
SC
[12] Devoe, L.D. Searle N, Castillo RA, Searle J. Prognostic components of
M AN U
computerized fetal biophysical testing. Am J Obstet Gynecol 1988; 158:1144-1148.*
[13] Devoe LD: Computerized FHR Analysis and Neural Networks in Antepartum Fetal Surveillance. Current Opinion in Ob/Gyn 1996: Vol. 8, 119-122*
[14] Bloom SL, Spong CY, Thom E, Varner MW, et al. Fetal pulse oximetry and
TE D
cesarean delivery. N Eng J Med. 2006; 355: 2195-2002*
[15] Richards EP, The Supreme Court Rules on Medical Device Liability - Or Does It?
AC C
EP
IEEE Engineering In Medicine And Biology Magazine 1997; 16: 87
Conflict of Interest Statement:
Dr. Devoe is a paid consultant for Neoventa Medical, AB, Molndal, Sweden.
RI PT
ACCEPTED MANUSCRIPT
SC
Figures
M AN U
1. Schematic block-diagram of an automated fetal monitor alerting system
2. Stages of development of intrapartum surveillance systems
3. Silent alarm systems
TE D
4. Model for neural network
5. Characteristics of rapid response systems
EP
6. Features of a novel intrapartum risk assessment and alerting system
AC C
7. Graphic display of risk trends displayed during labor
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Stage 1
Stage 2
Conventional
MACHINE Local Diagnostics
Global Diagnostics
Stan S31
Dx fetal condition
FHR Monitor
SC
PeriCalm
RI PT
MACHINE
Stage 3 MACHINE
Propose Care Plan
M AN U
TraceVue
Alerting/Rules Data Flow
Data Flow
DONE Local Dx External Factors Prognosis
AC C
Global Dx
EP
FHR Tracing
TE D
DONE
Data Flow
FUTURE
Global Dx
Inform machine
External
of external
Factors
factors, control
Prognosis
thresholds, etc
Tasks left for Clinicians
Invisible unless a problem is detected: Silent Bank Alarms “Silent” Alarms will not distract or interfere with other essential care elements 2. Patient should not be held “hostage” by system: Wireless transmission (Monica) Machines do not need to be in same room to provide alerts
AC C
EP
TE D
M AN U
SC
1.
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Gestation Risk Age Factor
Mat Age
Testing Interval
React
RI PT
Layer
M AN U
SC
INPUT
AC C
OUTPUT
EP
TE D
HIDDEN
Perinatal Outcomes
Race
Parity
HISTORIC EXAMPLE: THE CARDIFF PUMP
SC
M AN U
Intelligent systems that sense and respond to sudden changes in maternal or fetal status
RI PT
ACCEPTED MANUSCRIPT
EP
TE D
Feed-back loop regulation of oxytocin infusion: automatic shut-off for abnormal CTG
AC C
CURRENT EXAMPLE: VOLVO CARS
Systems of sensors are already used in automobile technology to monitor fuel mixture operating temperature, tire pressure, and ……
Collision Risk
RI PT
ACCEPTED MANUSCRIPT
M AN U TE D
EP
Prediction of risk based on post-test likelihood : (How much does monitoring data alter risk of hypoxia or acidosis).
SC
Baseline assessment of fetal risk Linking to probabilistic database of known fetal risk factors to current physiologic status (EFM)
AC C
Alerts trigger clinical prompts: Present best practice alternatives for patient care (evidence-based)
ACCEPTED MANUSCRIPT
RI PT
Estimated Risk Over Time
M AN U TE D EP AC C
Fetal Risk Status
SC
Time Course of Labor
ACCEPTED MANUSCRIPT
Highlights for Future Perspectives in Intrapartum Surveillance
• •
AC C
EP
TE D
M AN U
SC
•
Standard EFM is a limited prognostic tool Adjunctive assessment tools have not consistently improved EFM performance Future surveillance systems are possible with today’s technologies Such systems will need to blend current EFM systems with intelligent computers Industry must be willing to take on the risk of developing such novel systems
RI PT
• •
