British Journal of Obstetrics and Gynaecology January 1992, Vol. 99, pp. 26-3 1
FETAL A N D NEONATAL M E D I C I N E
Fetal acidaemia, the cardiotocograph and the T/QRS ratio of the fetal ECG in labour N E I L A. MACLACHLAN J O H N A . D. S P E N C E R KATE HARDING SABARATNAM ARULKUMARAN Institute of Obstetrics and Gynaecology, Royal Postgratuate Medical School, University of London, Queen Charlotte’s and Chelsea Hospital, Goldhawk Road, Chiswick, London W6 OXG
ABSTRACT Objective To relate the T/QRS ratio of the fetal electrocardiogram (ECG) to the cardiotocogram (CTG) and fetal pH during labour.
Design Prospective data collection from selected monitored labours. Setting A postgraduate teaching hospital delivery suite. Subjects 113 women in labour at term. Main outcome measures Correlation of fetal T/QRS ratio values with pH values at the time of fetal blood sampling and at birth (umbilical artery blood). Comparison of the predictive values of raised T/QRS ratio and a pathological CTG for fetal acidaemia. Results Complete data sufficient for analysis was available for 51 fetal scalp blood samples and 93 umbilical artery pH samples. The median (range) of T/QRS ratio values before birth of 88 babies not requiring admission to the neonatal unit was 0.13 (0.00-0.32) with a 97.5th centile value of 0.28. T/QRS ratios did not correlate with fetal scalp pH values. Fetal scalp acidaemia (pHc7.20) was detected with rates of 50 and 13% respectively by a pathological CTG and by a T/QRS ratio above 0.28, the positive predictive values being 40% and 50%, respectively. There was a significant correlation between increasing T/QRS ratio and falling pH. Detection rates (sensitivities) for umbilical artery acidaemia (pH 0.28) had a considerably lower detection rate for fetal acidaemia during labour than a pathological CTG.
The ability to make an accurate assessment of fetal well-being during labour from the cardiotocogram (CTG) has recently been questioned (Murphy et al. 1990). Fetal blood sampling, whilst necessary to minimize the increased rate of operative intervention associated with fetal heart rate (FHR) abnormalities seen on the CTG (Thacker 1991), is practised in less than 50% of consultant units in the UK (Wheble et al. 1989). Reflex FHR decelerations indicate adaptive cardiovascular adjustments to hypoxia in fetal sheep (Itskovitz et al. 1982) which include maintaining or increasing blood supply to the brain, heart and adrenals (Cohn et al. 1974). Artery blood pH may remain normal (Parer et al. 1980). Severe hypoxaemia stimulates catecholamine release (Gu et al. 1985) and associated myocardial anaerobic glycogenolysis changes the potassium gradient across the myocardial cell membrane producing ST elevation and a raised T/QRS ratio (Rosen et al. 1984). Analysis of the fetal electrocardiogram (ECG) together with the FHR on the CTG has been suggested as a possible means by which the prediction of fetal acidaemia may be improved (Jenkins et Correspondence: John A. D. Spencer, Senior Clinical Lecturer and Consultant, Department of Obstetrics and Gynaecology, University College and Middlesex School of Medicine, University College London, 86-96 Chenies Mews, London, WClE 6HX, UK.
26
al. 1986; Arulkumaran et al. 1990) and a new monitor has been designed to provide on-line calculation of the T/QRS ratio from the fetal ECG during labour for display on the CTG (Rosen & Lindecrantz 1989). This study was designed to compare the predictive values of a raised T/QRS ratio with a pathological CTG for fetal acidaemia during labour and at delivery and to relate the T/QRS ratio to pH and to neonatal condition.
Subjects and methods A total of 113 women in labour with a singleton fetus presenting by the head at term (37-42 weeks) was included in the study. Pregnancies with an abnormal CTG were selected preferentially even though the abnormality did not always persist. The FHR was monitored using a standard scalp electrode (Copeland, Surgicraft Ltd, Redditch, UK) and an FM6 (Oxford Sonicaid, Chichester, Sussex, UK) or HP 803 1 (Hewlett Packard, Uxbridge, Middlesex, UK) fetal monitor. The fetal ECG was taken from the same scalp electrode and maternal leg plate for on-line calculation and recording of the T/QRS ratio by a STAN monitor (Cinventa, Gothenberg, Sweden). The T/QRS ratio data were not used for clinical management. The study
TIQRSR A T ~ O 27
FETAL A C I D A E M I A , THE CARDIOTOCOGRAPH A N D T H E
was approved by the hospital ethics committee and all women gave informed consent. The method of signal processing by the STAN monitor has been described (Rosen & Lindecrantz 1989). Ten values of the fetal T/QRS ratio were averaged and printed onto paper at approximately 2-min intervals. The mean of three such average ratios (at least 6-min of record) was taken as representative of the signal before each fetal blood sample (FBS) and before delivery. Waveforms included for analysis were within 10 min of the FBS or within 30 min of delivery. Signals with excessive 50 Hz noise, a wandering baseline or those without a clear P wave were not considered for analysis. An FBS was collected when clinically indicated, usually to assess FHR changes. All scalp samples were collected with women in the left lateral position and only first samples were used in the analysis. Umbilical artery blood was collected into a preheparinized syringe from a double-clamped section of cord for immediate acid-base analysis. The pH and base deficit was obtained using a standard blood gas analyser (ABL 30, Radiometer, Copenhagen, Denmark), and extracellular base deficit was derived from a nomogram (Siggaard-Andersen 1971). The CTG during the 30 min before each FBS and before delivery was classified as normal, suspicious or pathological (FIGO News 1987). Patient data and information regarding outcome were collected, including mode of delivery, birthweight, Apgar scores and admission to the neonatal unit. A normal range of values for the T/QRS ratio was determined from those cases where the baby did not require admission to the neonatal unit.
Predictive values (Wald & Cuckle 1989) of the CTG and T/QRS ratio for fetal acidaemia during the first stage of labour (FBS pH
FETAL A N D NEONATAL M E D I C I N E
Fetal acidaemia, the cardiotocograph and the T/QRS ratio of the fetal ECG in labour N E I L A. MACLACHLAN J O H N A . D. S P E N C E R KATE HARDING SABARATNAM ARULKUMARAN Institute of Obstetrics and Gynaecology, Royal Postgratuate Medical School, University of London, Queen Charlotte’s and Chelsea Hospital, Goldhawk Road, Chiswick, London W6 OXG
ABSTRACT Objective To relate the T/QRS ratio of the fetal electrocardiogram (ECG) to the cardiotocogram (CTG) and fetal pH during labour.
Design Prospective data collection from selected monitored labours. Setting A postgraduate teaching hospital delivery suite. Subjects 113 women in labour at term. Main outcome measures Correlation of fetal T/QRS ratio values with pH values at the time of fetal blood sampling and at birth (umbilical artery blood). Comparison of the predictive values of raised T/QRS ratio and a pathological CTG for fetal acidaemia. Results Complete data sufficient for analysis was available for 51 fetal scalp blood samples and 93 umbilical artery pH samples. The median (range) of T/QRS ratio values before birth of 88 babies not requiring admission to the neonatal unit was 0.13 (0.00-0.32) with a 97.5th centile value of 0.28. T/QRS ratios did not correlate with fetal scalp pH values. Fetal scalp acidaemia (pHc7.20) was detected with rates of 50 and 13% respectively by a pathological CTG and by a T/QRS ratio above 0.28, the positive predictive values being 40% and 50%, respectively. There was a significant correlation between increasing T/QRS ratio and falling pH. Detection rates (sensitivities) for umbilical artery acidaemia (pH 0.28) had a considerably lower detection rate for fetal acidaemia during labour than a pathological CTG.
The ability to make an accurate assessment of fetal well-being during labour from the cardiotocogram (CTG) has recently been questioned (Murphy et al. 1990). Fetal blood sampling, whilst necessary to minimize the increased rate of operative intervention associated with fetal heart rate (FHR) abnormalities seen on the CTG (Thacker 1991), is practised in less than 50% of consultant units in the UK (Wheble et al. 1989). Reflex FHR decelerations indicate adaptive cardiovascular adjustments to hypoxia in fetal sheep (Itskovitz et al. 1982) which include maintaining or increasing blood supply to the brain, heart and adrenals (Cohn et al. 1974). Artery blood pH may remain normal (Parer et al. 1980). Severe hypoxaemia stimulates catecholamine release (Gu et al. 1985) and associated myocardial anaerobic glycogenolysis changes the potassium gradient across the myocardial cell membrane producing ST elevation and a raised T/QRS ratio (Rosen et al. 1984). Analysis of the fetal electrocardiogram (ECG) together with the FHR on the CTG has been suggested as a possible means by which the prediction of fetal acidaemia may be improved (Jenkins et Correspondence: John A. D. Spencer, Senior Clinical Lecturer and Consultant, Department of Obstetrics and Gynaecology, University College and Middlesex School of Medicine, University College London, 86-96 Chenies Mews, London, WClE 6HX, UK.
26
al. 1986; Arulkumaran et al. 1990) and a new monitor has been designed to provide on-line calculation of the T/QRS ratio from the fetal ECG during labour for display on the CTG (Rosen & Lindecrantz 1989). This study was designed to compare the predictive values of a raised T/QRS ratio with a pathological CTG for fetal acidaemia during labour and at delivery and to relate the T/QRS ratio to pH and to neonatal condition.
Subjects and methods A total of 113 women in labour with a singleton fetus presenting by the head at term (37-42 weeks) was included in the study. Pregnancies with an abnormal CTG were selected preferentially even though the abnormality did not always persist. The FHR was monitored using a standard scalp electrode (Copeland, Surgicraft Ltd, Redditch, UK) and an FM6 (Oxford Sonicaid, Chichester, Sussex, UK) or HP 803 1 (Hewlett Packard, Uxbridge, Middlesex, UK) fetal monitor. The fetal ECG was taken from the same scalp electrode and maternal leg plate for on-line calculation and recording of the T/QRS ratio by a STAN monitor (Cinventa, Gothenberg, Sweden). The T/QRS ratio data were not used for clinical management. The study
TIQRSR A T ~ O 27
FETAL A C I D A E M I A , THE CARDIOTOCOGRAPH A N D T H E
was approved by the hospital ethics committee and all women gave informed consent. The method of signal processing by the STAN monitor has been described (Rosen & Lindecrantz 1989). Ten values of the fetal T/QRS ratio were averaged and printed onto paper at approximately 2-min intervals. The mean of three such average ratios (at least 6-min of record) was taken as representative of the signal before each fetal blood sample (FBS) and before delivery. Waveforms included for analysis were within 10 min of the FBS or within 30 min of delivery. Signals with excessive 50 Hz noise, a wandering baseline or those without a clear P wave were not considered for analysis. An FBS was collected when clinically indicated, usually to assess FHR changes. All scalp samples were collected with women in the left lateral position and only first samples were used in the analysis. Umbilical artery blood was collected into a preheparinized syringe from a double-clamped section of cord for immediate acid-base analysis. The pH and base deficit was obtained using a standard blood gas analyser (ABL 30, Radiometer, Copenhagen, Denmark), and extracellular base deficit was derived from a nomogram (Siggaard-Andersen 1971). The CTG during the 30 min before each FBS and before delivery was classified as normal, suspicious or pathological (FIGO News 1987). Patient data and information regarding outcome were collected, including mode of delivery, birthweight, Apgar scores and admission to the neonatal unit. A normal range of values for the T/QRS ratio was determined from those cases where the baby did not require admission to the neonatal unit.
Predictive values (Wald & Cuckle 1989) of the CTG and T/QRS ratio for fetal acidaemia during the first stage of labour (FBS pH
