PRENATAL DIAGNOSIS, VOL.
1 1,905-914 (1991)
COMPARATIVE AUDIT OF BOOKING AND MID-TRIMESTER ULTRASOUND SCANS IN THE PRENATAL DIAGNOSIS OF CONGENITAL ANOMALIES G . CONSTANTINE AND J. McCORMACK Birmingham Maternity Hospital. Queen Elizabeth Medical Centre, Edgbaston, Birmingham, B15 ZTB, U.K.
SUMMARY During I988 and 1989,3565 women booked under consultants who performed an ultrasound scan at booking, whilst 4984 booked under consultants who performed a formal midtrimester scan between 16 and 18 weeks. All significant anomalies diagnosed prenatally and in the neonatal period were recorded, the incidence in each group being 12.9/1000 and 9.83/1000, respectively (NS). The sensitivity of diagnosis before 20 weeks was 45 per cent in the ‘mid-trimester’ group (kappa 0.62) compared with 30 per cent in the ‘booking’ group (kappa 0.46), overall sensitivity of prenatal diagnosis, however, being similar in both groups (63 vs. 6 5 per cent, kappa 0.77 vs. 0.79). Cardiac anomalies were the single largest group which were not detected equally prenatally in both groups. This study shows that formal mid-trimester scanning leads to anomalies being detected significantly earlier in the antenatal period. Although not statistically significant, three lethal anomalies were missed prenatally in the ‘booking’ group which we would have expected to diagnose on a mid-trimester scan. These figures are discussed in the light of previous reports. KEY WORDS
Ultrasound scan Formal mid-trimester scan Congenital anomalies
INTRODUCTION It is now the standard practice of most obstetricians to perform at least one ultrasound examination on all pregnant women. The potential theoretical benefits of this approach were outlined in a report from the Royal College of Obstetricians and Gynaecologists (RCOG) in 1984 (RCOG, 1984), which concluded that a scan should be performed routinely on all women between 16 and 18 weeks’ gestation, the rationale being that at this gestation the pregnancy may be accurately dated and a wide range of fetal anomalies detected (Campbell and Pearce, 1983). Although such a policy appears to make sense, hard data to support such a claim are lacking, and other authorities favour a more selective approach (NIH, 1984; Thacker, 1985). Prompted by the RCOG report, many units have introduced formal midtrimester scanning usually supplementing, but occasionally supplanting, maternal serum alpha-fetoprotein (AFP) screening for neural tube defects. Other consultants continue to rely on scans at the time of booking together with a detailed or further Addresseefor correspondence:Dr G . Constantine,ConsultantObstetrician and Gynaecologist,Good Hope Hospital, Rectory Road, Sutton Coldfield, Birmingham B75 7RR, U.K.
0197-385 1/91/ 120905-10$05.00 0 1991 by John Wiley & Sons, Ltd.
Received 26 November 1990 Revised 6 May I991 Accepted 4 June I991
906
G . CONSTANTINE AND J. McCORMACK
scan if the maternal serum AFP is elevated or other indications are present. Neither of these approaches has been the subject of a randomized trial, and in the current climate is unlikely to be so. Given this fact, and the wide variation in quality and expertise of prenatal diagnostic ultrasound, effective auditing is vitally important in all obstetric ultrasound units. At the Birmingham Maternity Hospital during 1988 and 1989, the consultants were offered the choice of a routine booking scan or a formal mid-trimester scan, to have both being logistically impossible unless specific criteria such as bleeding in early pregnancy, unknown dates, or a previous anomaly, etc. were present. This presented an opportunity to audit and compare the effectiveness of the two philosophies. PATIENTS AND METHODS Of the nine consultant obstetricians at the Birmingham Maternity Hospital, four opted for a booking scan at whatever gestation the patient booked, whilst four opted for a formal mid-trimester scan, between 16 and 18weeks, or if greater than 18 weeks a scan at booking. If dates were unknown, or early pregnancy bleeding was present, an earlier scan was allowed in the mid-trimester group. One consultant elected to have scans performed only on selective indications. The ultrasound records of all women who booked under the eight consultants performing routine ultrasound scans during the years 1988 and 1989 were studied. Any women who were specifically referred because of a suspected fetal anomaly and all in-utero transfers were excluded. The neonatal records were also examined and all babies found to have a significant fetal abnormality in the first week of life noted. Some abnormalities which were unlikely to be detected by the level of routine ultrasound examination performed (these included cleft lip/palate, imperforate anus, Hirschsprung’s disease, extra digits, talipes and hypospadias) were excluded, as were chromosomal abnormalities with no marker (such as an exomphalos, etc.) observable by prenatal ultrasound, e.g., many cases of trisomy 21. For the purpose of this study, fetuses with more than one anomaly were classified under that most easily detectable by prenatal ultrasound. During this time, most scans were performed by trained and qualified ultrasonographers, many with the Diploma in Medical Ultrasound (DMU). The remainder were performed either by consultants/senior registrars trained in ultrasonography or by ultrasonographers being trained for the DMU. In particular, there was no selection of routine ultrasound examinations (whether booking or mid-trimester) to operators with differing levels of experience. All scans were performed on one of four machines: Hitachi 410, Hitachi 340, GE RT3000, Acuson 128, using linear or curved array transducers of 3.5 or 5MHz frequency. A booking scan below 16 weeks’ gestation involved measurement of the crown-rump length (CRL) and/or biparietal diameter (BPD). A mid-trimester scan was performed between 16 and 18 weeks’ gestation. Measurements were made of the biparietal diameter, head circumference,femur length, and transverse cerebellar diameter, and an assessment was made of the fetal head, spine, stomach, renal tract, abdominal wall, limbs and four-chamber view of the heart.
AUDIT OF BOOKING AND MID-TRIMESTER ULTRASOUND SCANS
907
Table 1. Abnormalities in booking scan group
Abnormality Choroid plexus cyst Anencephaly Spina bifida Isolated hydrocephaly Other CNS Major renal Dilated renal pelvis Casdiac Gut atresia Abdominal wall defects Diaphragmatic hernia
MCA Hydrops Hydatidiform mole CCAM Amniotic band Cranios y nostosis Cystic h ygroma Cord cyst Totals
Booking AFP Detailed 6 1 0 0 0 1 1 0 0 0 0 1 1 2
2 1 1 0 0 0 1 0 0 0 0 0 0 0
0 1
0
0 0 1 0 1 0 0 1 0 0 0 0
0 1 0
1 0 0 0
0 0 0 0 0 0 0
15
6
3
Early AN
Late AN
3
0
0 2
0 0 0 0 2
0 0
1 2 1 0 1 0 1 0 0 1 0 0
1 2 1 0 0 0 2 0 0
0 0
0 1 0 1
12
10
Missed
Totals
0 0 0
11 2 4 1 2 6 5 14 1 1 1
1 1 2 0 10 0 0 1 1 0
3
0
3 2 1 2 1 1 1
16
62
0 0 0
0 0
Booking: detected on ultrasound scan at booking; A F P detected because of raised maternal serum alpha-feloprotein; detailed: detected on detailed ultrasound scan generated on previous history; early AN: detected on ultrasound scan generated for other reasons prior to 28 weeks; late AN: detected on ultrasound scan generated for other reasons after 28 weeks; missed: abnormalitiesdetected only in early neonatal period; MCA: multiple congenital abnormalities; CCAM: congenital cystic adenomatoid malformation of the lung.
RESULTS During 1988 and 1989,8549 women who were booked to deliver at the Birmingham Maternity Hospital fulfilled the above criteria. Of this total, 4984 booked under consultants who routinely performed a mid-trimester scan, whilst 3565 booked under consultants who requested only a scan at the booking appointment together with further scans if clinically indicated. In the mid-trimester group, the mean gestation of scanning was 18-5 weeks (SD 4, range 15-37, median 17), whilst in the booking scan group the corresponding figure was mean 15 weeks (SD 5.8, range 7-38, median 13). This difference is significant at p < 0.001. The total abnormalities detected in the two groups, both prenatally and postnatally, are shown in Tables 1 and 2. Within the booking group, 62 fetuses with anomalies were present, which included 16 with choroid plexus cysts or minor degrees of dilatation of the renal pelves which resolved. Excluding these, the overall anomaly rate was 12.9/1000 bookers. In this group, 16 fetuses with abnormalities were not detected prenatally,
908
G. CONSTANTINE AND J. McCORMACK
Table 2. Abnormalities in mid-trimester scan group
Abnormality
Early AN
Late AN
Missed
Totals
0 0 0 0 0 0 1 1 2 2 0 0 1 0 1 0 0
0 0 0 0 0 1 1 0 11 1 0 2 0 1 1 0 0
19
0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 1 1 0 0 0 0 0 1 0 0
2
3
8
18
MTS
AFP
Detailed
Choroid plexus cyst Anencephaly Spina bifida Encephalocele Isolated hydrocephaly Other CNS Major renal Dilated renal pelvis Cardiac Gut atresia Abdominal wall defects Diaphragmatic hernia Volvulus MCA H ydrops Jeune’s syndrome Cystic hygroma
19 8 4 1 2 0 1 11 0 0 2 0 0 0 0 1 1
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 1 0
Totals
50
0
0 0 1 0
8 5 1 2 1
4 13 14 3 2 2 1 1
3 1 1 81
For abbreviationssee Table 1. MTS: Abnormalitiesnoted at mid-trimester scan.
Table 3. Sensitivitiesand kappa values for detection of abnormalities
Booking scan group All major abnormalities
All major abnormalities excluding cardiac Mid-trimester scan group All major abnormalities
A11major abnormalities excluding cardiac
Sensitivity (%)
Kappa
(a)
1 1,905-914 (1991)
COMPARATIVE AUDIT OF BOOKING AND MID-TRIMESTER ULTRASOUND SCANS IN THE PRENATAL DIAGNOSIS OF CONGENITAL ANOMALIES G . CONSTANTINE AND J. McCORMACK Birmingham Maternity Hospital. Queen Elizabeth Medical Centre, Edgbaston, Birmingham, B15 ZTB, U.K.
SUMMARY During I988 and 1989,3565 women booked under consultants who performed an ultrasound scan at booking, whilst 4984 booked under consultants who performed a formal midtrimester scan between 16 and 18 weeks. All significant anomalies diagnosed prenatally and in the neonatal period were recorded, the incidence in each group being 12.9/1000 and 9.83/1000, respectively (NS). The sensitivity of diagnosis before 20 weeks was 45 per cent in the ‘mid-trimester’ group (kappa 0.62) compared with 30 per cent in the ‘booking’ group (kappa 0.46), overall sensitivity of prenatal diagnosis, however, being similar in both groups (63 vs. 6 5 per cent, kappa 0.77 vs. 0.79). Cardiac anomalies were the single largest group which were not detected equally prenatally in both groups. This study shows that formal mid-trimester scanning leads to anomalies being detected significantly earlier in the antenatal period. Although not statistically significant, three lethal anomalies were missed prenatally in the ‘booking’ group which we would have expected to diagnose on a mid-trimester scan. These figures are discussed in the light of previous reports. KEY WORDS
Ultrasound scan Formal mid-trimester scan Congenital anomalies
INTRODUCTION It is now the standard practice of most obstetricians to perform at least one ultrasound examination on all pregnant women. The potential theoretical benefits of this approach were outlined in a report from the Royal College of Obstetricians and Gynaecologists (RCOG) in 1984 (RCOG, 1984), which concluded that a scan should be performed routinely on all women between 16 and 18 weeks’ gestation, the rationale being that at this gestation the pregnancy may be accurately dated and a wide range of fetal anomalies detected (Campbell and Pearce, 1983). Although such a policy appears to make sense, hard data to support such a claim are lacking, and other authorities favour a more selective approach (NIH, 1984; Thacker, 1985). Prompted by the RCOG report, many units have introduced formal midtrimester scanning usually supplementing, but occasionally supplanting, maternal serum alpha-fetoprotein (AFP) screening for neural tube defects. Other consultants continue to rely on scans at the time of booking together with a detailed or further Addresseefor correspondence:Dr G . Constantine,ConsultantObstetrician and Gynaecologist,Good Hope Hospital, Rectory Road, Sutton Coldfield, Birmingham B75 7RR, U.K.
0197-385 1/91/ 120905-10$05.00 0 1991 by John Wiley & Sons, Ltd.
Received 26 November 1990 Revised 6 May I991 Accepted 4 June I991
906
G . CONSTANTINE AND J. McCORMACK
scan if the maternal serum AFP is elevated or other indications are present. Neither of these approaches has been the subject of a randomized trial, and in the current climate is unlikely to be so. Given this fact, and the wide variation in quality and expertise of prenatal diagnostic ultrasound, effective auditing is vitally important in all obstetric ultrasound units. At the Birmingham Maternity Hospital during 1988 and 1989, the consultants were offered the choice of a routine booking scan or a formal mid-trimester scan, to have both being logistically impossible unless specific criteria such as bleeding in early pregnancy, unknown dates, or a previous anomaly, etc. were present. This presented an opportunity to audit and compare the effectiveness of the two philosophies. PATIENTS AND METHODS Of the nine consultant obstetricians at the Birmingham Maternity Hospital, four opted for a booking scan at whatever gestation the patient booked, whilst four opted for a formal mid-trimester scan, between 16 and 18weeks, or if greater than 18 weeks a scan at booking. If dates were unknown, or early pregnancy bleeding was present, an earlier scan was allowed in the mid-trimester group. One consultant elected to have scans performed only on selective indications. The ultrasound records of all women who booked under the eight consultants performing routine ultrasound scans during the years 1988 and 1989 were studied. Any women who were specifically referred because of a suspected fetal anomaly and all in-utero transfers were excluded. The neonatal records were also examined and all babies found to have a significant fetal abnormality in the first week of life noted. Some abnormalities which were unlikely to be detected by the level of routine ultrasound examination performed (these included cleft lip/palate, imperforate anus, Hirschsprung’s disease, extra digits, talipes and hypospadias) were excluded, as were chromosomal abnormalities with no marker (such as an exomphalos, etc.) observable by prenatal ultrasound, e.g., many cases of trisomy 21. For the purpose of this study, fetuses with more than one anomaly were classified under that most easily detectable by prenatal ultrasound. During this time, most scans were performed by trained and qualified ultrasonographers, many with the Diploma in Medical Ultrasound (DMU). The remainder were performed either by consultants/senior registrars trained in ultrasonography or by ultrasonographers being trained for the DMU. In particular, there was no selection of routine ultrasound examinations (whether booking or mid-trimester) to operators with differing levels of experience. All scans were performed on one of four machines: Hitachi 410, Hitachi 340, GE RT3000, Acuson 128, using linear or curved array transducers of 3.5 or 5MHz frequency. A booking scan below 16 weeks’ gestation involved measurement of the crown-rump length (CRL) and/or biparietal diameter (BPD). A mid-trimester scan was performed between 16 and 18 weeks’ gestation. Measurements were made of the biparietal diameter, head circumference,femur length, and transverse cerebellar diameter, and an assessment was made of the fetal head, spine, stomach, renal tract, abdominal wall, limbs and four-chamber view of the heart.
AUDIT OF BOOKING AND MID-TRIMESTER ULTRASOUND SCANS
907
Table 1. Abnormalities in booking scan group
Abnormality Choroid plexus cyst Anencephaly Spina bifida Isolated hydrocephaly Other CNS Major renal Dilated renal pelvis Casdiac Gut atresia Abdominal wall defects Diaphragmatic hernia
MCA Hydrops Hydatidiform mole CCAM Amniotic band Cranios y nostosis Cystic h ygroma Cord cyst Totals
Booking AFP Detailed 6 1 0 0 0 1 1 0 0 0 0 1 1 2
2 1 1 0 0 0 1 0 0 0 0 0 0 0
0 1
0
0 0 1 0 1 0 0 1 0 0 0 0
0 1 0
1 0 0 0
0 0 0 0 0 0 0
15
6
3
Early AN
Late AN
3
0
0 2
0 0 0 0 2
0 0
1 2 1 0 1 0 1 0 0 1 0 0
1 2 1 0 0 0 2 0 0
0 0
0 1 0 1
12
10
Missed
Totals
0 0 0
11 2 4 1 2 6 5 14 1 1 1
1 1 2 0 10 0 0 1 1 0
3
0
3 2 1 2 1 1 1
16
62
0 0 0
0 0
Booking: detected on ultrasound scan at booking; A F P detected because of raised maternal serum alpha-feloprotein; detailed: detected on detailed ultrasound scan generated on previous history; early AN: detected on ultrasound scan generated for other reasons prior to 28 weeks; late AN: detected on ultrasound scan generated for other reasons after 28 weeks; missed: abnormalitiesdetected only in early neonatal period; MCA: multiple congenital abnormalities; CCAM: congenital cystic adenomatoid malformation of the lung.
RESULTS During 1988 and 1989,8549 women who were booked to deliver at the Birmingham Maternity Hospital fulfilled the above criteria. Of this total, 4984 booked under consultants who routinely performed a mid-trimester scan, whilst 3565 booked under consultants who requested only a scan at the booking appointment together with further scans if clinically indicated. In the mid-trimester group, the mean gestation of scanning was 18-5 weeks (SD 4, range 15-37, median 17), whilst in the booking scan group the corresponding figure was mean 15 weeks (SD 5.8, range 7-38, median 13). This difference is significant at p < 0.001. The total abnormalities detected in the two groups, both prenatally and postnatally, are shown in Tables 1 and 2. Within the booking group, 62 fetuses with anomalies were present, which included 16 with choroid plexus cysts or minor degrees of dilatation of the renal pelves which resolved. Excluding these, the overall anomaly rate was 12.9/1000 bookers. In this group, 16 fetuses with abnormalities were not detected prenatally,
908
G. CONSTANTINE AND J. McCORMACK
Table 2. Abnormalities in mid-trimester scan group
Abnormality
Early AN
Late AN
Missed
Totals
0 0 0 0 0 0 1 1 2 2 0 0 1 0 1 0 0
0 0 0 0 0 1 1 0 11 1 0 2 0 1 1 0 0
19
0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 1 1 0 0 0 0 0 1 0 0
2
3
8
18
MTS
AFP
Detailed
Choroid plexus cyst Anencephaly Spina bifida Encephalocele Isolated hydrocephaly Other CNS Major renal Dilated renal pelvis Cardiac Gut atresia Abdominal wall defects Diaphragmatic hernia Volvulus MCA H ydrops Jeune’s syndrome Cystic hygroma
19 8 4 1 2 0 1 11 0 0 2 0 0 0 0 1 1
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 1 0
Totals
50
0
0 0 1 0
8 5 1 2 1
4 13 14 3 2 2 1 1
3 1 1 81
For abbreviationssee Table 1. MTS: Abnormalitiesnoted at mid-trimester scan.
Table 3. Sensitivitiesand kappa values for detection of abnormalities
Booking scan group All major abnormalities
All major abnormalities excluding cardiac Mid-trimester scan group All major abnormalities
A11major abnormalities excluding cardiac
Sensitivity (%)
Kappa
(a)
