Reprint from the Proceedings of the Equine Symposium and Annual Comference 2000, San Antonio, November 2000, pp 39-43.
FETAL MONITORING
Jon Palmer, VMD, DACVIM
New Bolton Center, University of Pennsylvania
Equine veterinarians have been searching for a reliable means of insuring fetal well-being for at least the last 40 years.1 Most techniques attempt to detect fetal physiologic responses which when abnormal could justify active intervention. Although it is now recognized that other antepartum insults are important in the pathogenesis of neonatal encephalopathy, 2,3 fetal monitoring is designed to detect a hypoxic ischemic asphyxial insults that are the most likely threat to fetal well being. Before discussing fetal monitoring techniques, a brief overview of the fetal response to hypoxemia will help with understanding the rationale of the procedures.
The fetal physiologic response to hypoxia is much different than that seen in the adult. In the adult, tissue hypoxia results in a dramatic increase in cardiac output and ventilation in an attempt to deliver more oxygen to tissues. The fetus, on the other hand, has no control over oxygen delivery to the uterus. As a result, the fetus must maximize the available limited resources. The first move by the fetus is to decrease cardiac output by a centrally mediated decrease in heart rate, decreasing cardiac work and thus limiting cardiac oxygen need. Simultaneously there is a centrally mediated vasoconstriction of nonessential circulations redirecting blood flow to the heart, brain and adrenals. The increased work caused by the increase in afterload is balanced by the decreased work associated with the decreased heart rate. The fetus is able to achieve dramatic circulatory changes without the penalty of increasing oxygen demand of the cardiac muscle. Another significant change is slowing of placental perfusion mediated by increased umbilical vein resistance. This helps maximize placental oxygen extraction by favoring increased fetal placental surface area thus improving maternofetal gas exchange. These adaptive responses require an intact CNS-adrenergic response.4-6
Further fetal adaptation to hypoxia occurs through shifts in oxygen utilization. Almost half of fetal/neonatal oxygen utilization is facultative (growth, anabolic processes, activity) and not essential for survival. It is not clear how the fetus turns off growth but it is an essential part of tolerance of acute hypoxia. Likewise, decreasing unnecessary activity such as fetal breathing and other fetal movement is also an important method of decreasing oxygen need.6
So the fetus adapting to hypoxic stress will have a slow heart rate and appear inactive with no fetal breathing, few movements and cardiac accelerations. If the insult becomes severe resulting in hypoxic ischemic conditions, the compensation will eventually fail as the insult results in central depression and disruption of the CNS-adrenergic response resulting in loss of vascular control. As this occurs, the fetal foal will develop a tachycardia in a desperate attempt to maintain essential perfusion. There will be no associated fetal activity with this tachycardia. As the myocardium becomes affected, a terminal bradycardia will occur just before death.
Fetal monitoring takes advantage of these physiologic responses using gross body movements as predicative of fetal health.7 When activity that normally arises from a given brain center is observed (e.g. fetal breathing) then that regulatory center is assumed to have adequate oxygenation and normal metabolism.7 Thus observing movement provides insight into fetal CNS integrity. The fetal respiratory center is particularly sensitive to hypoxemia resulting in early cessation of fetal breathing.8 Once normal activity such as sustained fetal breathing is seen further observations are not necessary. Failure to observe a response within a prescribed period of time (30 minutes) necessitates including hypoxia as a possible cause, but in clinical practice, fetal hypoxia is the least likely cause of the absence of a given acute fetal biophysical value. In most cases, the failure to observe the activity is a result of normal periodicity of the physiologic response or the effect of the normal rest/activity cycles of the fetus. The observation time is based on the average duration of sleep-wake cycles in the normal fetus. Maturation of CNS centers that regulate acute fetal biophysical actions occur independently so gross body movements first appear early in fetal life, breathing movements later and fetal movements causing heart rate accelerations even later.7
Taking advantage of these fetal physiologic responses, Manning, a pioneer in the area, developed a Fetal Biophysical Profile scoring system for the human fetus in 1980 to predict the presence or absence of fetal asphyxia.7 His profile consists of 4 ultrasound derived observations (presence of fetal breathing movements, gross body movements, general body tone, and amniotic fluid volume) and fetal heart rate responses to fetal movements (Nonstress Test - not ultrasound derived). Normal fetal breathing movements are defined as one or more episodes of > 20 seconds duration in 30 minutes of observation. Gross body movements are considered normal when 2 or more discreet body/limb movements are seen in 30 minutes of observation. Normal fetal tone is defined as one or more episodes of active extension with return to flexion of fetal limb or trunk occur during a 30 minutes observation period. The reactive heart rate (Nonstress Test) is measured using a Doppler method, which simultaneously records fetal heart rate and fetal movements during a 20 minute observation period. In the 20 years since it was developed, evaluation of over 155,000 tests has validated this scoring system. A normal score is a reliable indicator that the fetus is unlikely to die during the 7 days after the test.7
Both transabdominal and transrectal ultrasound have an important place in evaluating the health and well-being of the late term equine fetus. A number of transabdominal ultrasound observations have been related to fetal health.9-12 Fetal heart rate and rhythm, fetal aortic diameter, fetal breathing movements, fetal activity, fetal tone, fetal fluid depths, uteroplacental thickness and integrity and fetal positioning have all been utilize in evaluating the fetal foal. The skilled ultrasonographer can obtained useful morphometric variables, survey fetal morphology, note fetal positioning, survey for abnormalities in the viewable placenta and search for twins quite efficiently. Unfortunately, there is a significant learning curve for the neophyte in obtaining accurate and repeatable observations. Some of the problems are outlined as follows. Measuring fetal heart rate may be simple when the fetus is inactive, but when trying to measure an active fetus’ heart rate, it can take considerable patients and skill. Problems arise measuring fetal aortic diameter when the aorta is not measured close to the heart, the aortic cross-section is biased or when the caudal vena cava is confused for the aorta. Fetal breathing movements are best visualized by watching for fetal diaphragm movement in relationship to fetal ribs. This can be difficult since the fetus must be inactive to clearly see these movements and maternal breathing motions often confuse the picture. A continuous period of 20 seconds of fetal breathing should be observed before it can be considered normal with confidence. Even for the experienced ultrasonographer this can prove impossible. Assessment of the uteroplacental unit is a very important part of transabdominal ultrasound assessment. Although it is relatively easy to measure the uteroplacental unit since the borders are distinct, the uteroplacental unit is quite thick in the nonfetal horn and the areas adjacent to it. This can cause considerable confusion in uteroplacental unit thickness measurement when an inexperienced operator is not clear as to how close to the nonfetal horn the measurements are made. With all the opportunity for error in morphometric measurements and subjective evaluations necessary, transabdominal ultrasound might better be considered an art most efficiently learned through apprenticeship and experience rather than a science that can easily be reproduced by rote.
Attempts have been made to produce an equine biophysical profile.9,12 Unfortunately, the predictive value of a normal test is not as uniformly accurate as it is in human medicine 7,9,12and clinical experience indicates that the predictive value of a abnormal test falls short of the ideal. The most recent version of the equine biophysical profile12 includes transabdominal ultrasound measurements of the fetal aortic diameter in relationship to the mare's weight, uteroplacental thickness, maximum amniotic and allantoic fetal fluid depth, uteroplacental contact, fetal activity level subjectively graded after the observation period and heart rate as estimated from spot ultrasound observations taken sporadically during the examination period. The lack of sensitivity and specificity of the profile may have to do with the selection of parameters measured. In development of the profile for the human fetus, Manning selected parameters that would reflect an acute hypoxia (fetal movement/tone, fetal heart rate and amniotic fluid volume). Amniotic fluid volume may reflect either acute or chronic hypoxia. The equine biophysical profile has a mix of parameters resulting from acute or chronic hypoxia. Changes in heart rate may reflect an acute hypoxia but, despite the observation period being more than 30 minutes, basing fetal heart rate observations on a dozen or so spot measurements does not give the quality of information continuous heart rate observations give over 30 minutes as in the Nonstress Test used by Manning. Observation of fetal movement and tone are important parameters of acute hypoxia, but in the equine profile these observations are subjective and lack the definition of discrete movement over a standard time interval. The difficulty in imaging large areas of the fetus in order to observe coordinated movements is a great disadvantage in equine medicine. Loss of fetal breathing movements, perhaps the most sensitive indicator of fetal hypoxia, is currently not included in the equine biophysical profile. Its inclusion is problematic considering the skill required to confidently detect 20 seconds of continuous fetal breathing. Observation of placental health (probably the most important ultrasound based observations) may detect a cause of hypoxia but does not indicate whether the fetal foal is successfully coping.13 Likewise, fetal aortic diameter reflects fetal size. Any change is a chronic result of in utero growth retardation or other problems. This is important information and, like placental changes, increases the risk of problems but does not address the current state of the fetus and how well it is coping. The equine biophysical profile shows promise but still requires refinement.
Invaluable information about placental health is obtained from transabdominal ultrasound examination. Although only the ventral and parts of the lateral aspect of the placenta can be imaged, careful examination of this area can add important information about placental dysfunction, which is the most significant threat to fetal foal well-being.
Over the past decade I have spent considerable time observing fetal heart rate patterns in our high-risk pregnancy population. Although our understanding of changing patterns is certainly rudimentary, observations can be helpful in gauging fetal health. Fetal heart rate patterns in relationship to uterine contractions continue to be the primary method of intrapartum fetal assessment in man despite questions of its usefulness.14,15 Continuous fetal heart rate measurement over a 20 min. interval in association with fetal activity is a common antepartum measure of fetal well being. Fetal heart rate was first heard and described in 17th-century France and first proposed as a measure fetal distress in the mid-19th century. In 1960 the first technique for recording equine fetal ECGs was published.1 Since then there have been numerous publications with occasional reference to relating fetal heart rate patterns to fetal foal health.16,17
When recording fetal heart rate (FHR) patterns both beat-to-beat variations and changes in baseline heart rate levels with periodic accelerations are important. Periodic accelerations are generally associated with fetal activity and suggest fetal health. The presence of accelerations associated with fetal activity increase with gestational age. Persistent tachycardia in the absence fetal activity suggests fetal distress. Fetal bradycardia may be present as a normal adaptive heart rate pattern suggesting either cardiovascular efficiency or early adaptation to hypoxia. Extreme fetal bradycardia may occur during terminal stages of fetal distress. So the usual pattern of fetal heart rate changes seen with fetal distress are an initial bradycardia dropping below the baseline heart rate without periodic accelerations during early compensation followed by persistent fetal tachycardia as the fetus decompensates followed finally by terminal bradycardia.
The presence of significant beat-to-beat variation suggests intact sympathetic/parasympathetic tone and central control indicating normal CNS responsiveness and normal local CNS metabolic environment reflecting fetal health. The disappearance of beat-to-beat variation is an ominous sign. Beat-to-beat FHR variability increases with increasing gestational age in association with maturation of the parasympathetic-sympathetic interaction and function.18 Strong correlations have been found between reduced heart rate variability and increase disease severity in clinical illness. Common causes of decreased beat-to-beat FHR variability include fetal compromise from hypoxia ischemia, asphyxia, sedation, brain death or other clinical illness. The fetal state and breathing movements affect heart rate variability, with periods of quiet sleep showing reduced variability and the presence of breathing movements increasing heart rate variability.19 Stimulation of the fetus increases the heart rate variability due to the change in the fetal state.18 Maternal medications such as detomidine or butorphenol reduce fetal heart rate variability transiently. Neonatal heart rate variability increases postnatally associated with gestational age and increases until adulthood.18
Other changes that may be noted on FHR ECG tracings include 2 distinct fetal patterns associated with twins, changes in complex orientation associated with fetal movements, changes in complex height associated with changes in fetal fluid volumes and fetal arrhythmias. Extrasystoles are an occasional finding which generally have little consequence if they are rare. The most common fetal arrhythmia (other than tachycardia/bradycardia) is an irregularly irregular pattern suggestive of atrial fibrillation. This pattern is usually transient and seen in fetal foals that subsequently have some degree of physical signs of fetal compromise. Very rarely paroxysmal tachycardia or a wide complex tachycardia may be identified which suggests extreme fetal distress.
Cardiotocographic recordings are possible on the fetal foal, which adds detection of fetal movement in relation to fetal heart rate accelerations and also records fetal heart rate variability. However expensive custom-made equipment is needed to perform this efficiently on the pregnant mare because of the depths of penetration needed. I believe that fetal monitoring techniques should be developed that can be used universally and not just by a select few.
Any ECG recorder that can print a tracing can be used to measure FHR. The electrical signal from the fetal heart is low amplitude necessitating placing the electrodes as close to the fetus as possible. Electrodes are basically placed in a pattern that produces the best amplitude in fetal complexes. The ideal pattern varies from mare to mare. In general, one electrode is placed in the lumbosacral area of the mare and the other 2 in the mid flank region. Initially the 2 flank electrodes are placed just below the lumbosacral electrode. However if the tracing is not adequate, one of the flank electrodes may be moved cranially or ventrally. If a single daily observation is made, it should be over a minimum of 10 minutes with periodic recordings. Currently, I routinely use telemetry ECG on pregnant mares from 6 PM to 6 AM, which allows periodic inspection of the heart rate throughout the 12-hour period and a 20 second recording of FHR every 2 hours or more often. It should be noted that fetal foal activity levels often appear to be circadian, with more activity during the night, so if baseline fetal heart rates without periodic accelerations are recorded during one part of the day, recordings should be made at other times.
When I measure heart rates and beat-to-beat variability, I generally measure the distance between each complex noting rate, accelerations, deaccelerations, complex polarity changes, rhythm and beat-to-beat variation. During the last weeks of pregnancy fetal foals usually have a baseline heart rate between 75- 60 with a low heart rate in the range of 75-40 bpm (80% will have a low fetal heart rate < 70, 55% low FHR < 60, 14% low FHR < 50) and the high FHR in the range of 83-250 bpm (86% will have a high fetal heart rate > 100, 50% high FHR > 120, 20% high FHR > 200). As indicated, transient low heart rates <60 bpm are very common and should not be considered ominous unless they are consistent with no accelerations. Also, FHR transiently may be > 200 bpm. Transient FHR > 120 bpm are not ominous unless they are consistently increased and do not dropped to baseline levels. In either case, when FHR are <60 or >120 throughout an observation period, repeat assessment within 24 hours or less is indicated. Beat-to-beat variability generally ranges from 0.5-4 mm with most in the range of 1 mm. When measuring the variation, periods when the heart rate is not accelerating or decelerating should be used for an accurate observation. The finding of no beat-to-beat variation in the absence of maternal drugs that may sedate the fetus is an ominous sign and repeat observations are indicated.
All fetal assessments using fetal monitoring techniques must be tempered by the whole clinical picture and not taken as a defining assessment. The goal is to treat the fetus and mare and not the test results. When non-reassuring findings are consistent, fetal resuscitation interventions are indicated. If the neonatal death rate is likely to exceed the fetal death rate the fetal foal should be left where it is. Even an abnormal uterine environment is often more successful at maintaining the fetal foals life than neonatal intensive care. Delivery is only indicated if extrauterine survival is more likely than continued survival in utero.
References
1. Holmes JR, Darke PGG: Foetal Electrocardiograpy in the Mare. Vet Rec 82:651-655, 1968
2. Edwards AD and Nelson KB: Neonatal encephalopathies: Time to reconsider the cause of encephalopathies. British medical journal 317: 1537-1538, 1998.
3. Nelson K. B. and Willoughby RE: Infection, inflammation and risk of cerebral palsy. Current Opinion in Neurology 13:133-139, 2000.
4. Adamson SL, Myatt L, Byrne BMP: Chapter 90: Regulation of Umbilical Blood Flow. In ed. Polin RA and Fox WW Fetal and Neonatal Physiology WB Saunders, Philadelphia 1998, pp 977-989.
5. Long WA, Henry W, Llanos AJ: Autonomic and Central Neuroregulation of Fetal Cardiovascular Function. In ed. Polin RA and Fox WW Fetal and Neonatal Physiology WB Saunders, Philadelphia 1998, pp 943-961.
6. Sukumar M and Morin FC: Chapter 93: Response of the Fetal Circulation to Stress. In ed. Polin RA and Fox WW
7. Manning FA, Platt LD: Fetal movements in human pregnancies in the third trimester. Obstetrics and Gynecology 54:699, 1979
8. Dawes S, Fox HE, Leduc BM, et al.: Respiratory movements and paradoxical sleep in the fetal lamb. Journal Physiology (London) 210:47, 1970
9. Adams-Brendemuehl C and Pipers FS: Antepartum evaluations of the equine fetus. Journal of Reproduction Fertility, Supplement 35:565-5 73, 1987.
10. Pipers FS and Adams-Brendemuehl C: Techniques and applications of transabdominal ultrasonography in the pregnant mare. JVM may 185: 766-771, 1984.
11. Reef VB, Vaala WE, Worth LT, et. al.: Ultrasonographic evaluation of the fetus and intrauterine environment in healthy mares during late gestation. Veterinary Radiology & Ultrasonography 36:533-5 41, 1995.
12. Reef VB, Vaala WE, Worth LT, et. al.: Ultrasonographic assessment of fetal well-being during late gestation: development of an equine biophysical profile. Equine Veterinary Journal 28:200-208, 1996.
13. Vintzileos AM, Campbell WA, Ingardia CT: The fetal biophysical profile and its predictive value. Obstetrics and gynecology 62: 271, 1983.
14. Whittle, M: Is it time to abandon cardiotocographic ECG and analysis? The Lancet 355 (9202) :422, 2000.
15. Parer JT and King T: Fetal heart rate monitoring: Is it salvageable? American journal of obstetrics and gynecology 182:928-987, 2000.
16. Colles CM, Parks RD and May CJ: Foetal Echocardiography in the Mare. Equine Veterinary journal 10:32-37, 1978.
17. Yamamoto K., Yasuda Jane and Kimehiko T: Electrocardiography findings during parturition and blood gas tensions immediately after birth in thoroughbred foals. Japanese journal vet research 39:143-157, 1991.
18. Dunster KR: Physiologic variability in a perinatal period: Origins, measurement, and applications. Clinics in perinatology 26 (4):801-89, 1999
19. Divon MY, Zimmer EZ, Platt LD, et al: Human fetal breathing: Associated changes in heart rate and beat-to-beat variability. American Journal Obstetrics and Gynecology 151: 403, 1985