This is the abstract of a free paper presented at a scientific conference
- Critically appraise this abstract.
- You will be asked questions regarding validation of a diagnostic / screening test.
A comparison of the fetal acoustic stimulation test (FAST) and the non stress test (NST) in detecting fetal hypoxia.
Introduction: A healthy fetus, if exposed to external sound stimulation, often responds with vigorous movements which can be felt by the mother. A hypoxic fetus usually does not show this response.
Objective: To evaluate the fetal acoustic stimulation test (FAST) and the non stress test (NST) in detecting fetal hypoxia.
Design and Setting: A randomized clinical trial carried out at the Teaching Hospital, Colombo
Method: During the three month period commencing 1st June 2004, 123 consecutive pregnant women who had normal viable, singleton fetuses, but complained of reduced or absent fetal movements after 37 weeks of gestation, were randomly allocated to have either a FAST(n=63) or NST (n=60).
All the women who either did not feel fetal movements after the FAST (n=8) or had a non reactive NST (n=23) were delivered by emergency Caesarean section (CS).
The results of the FAST and the NST were compared with the 5 min Apgar scores in these babies delivered by CS (n=31).
Outcome measures:
Maternal perception of fetal movements after FAST, results of NST and the babies’ 5 minute Apgar scores at delivery.
Results:
Of the babies delivered by CS (n=31), six of the babies in the FAST group (n= 8) and 10 of the babies in the NST group (n=23) had neonatal asphyxia.
Therefore the FAST had a better sensitivity (75%) compared to the NST (43%) in detecting fetal hypoxia (p < 0.001)
Conclusion: The FAST is a reliable and a more sensitive test than the NST for detecting fetal hypoxia.
It significantly reduces the need for emergency CS in women with reduced or absent fetal movements.
- Indicate the errors in this study
a) RCT is inappropriate 15
(i) FAST & NST diagnostic tests Pos / Neg. Not “interventions” outcome / result
(ii) All 123 should have had the FAST, NST and Apgar scores) 15
2. Comment on the gold standard
i)5 min Apgar score cut off point (gold standard for neonatal asphyxia in this study)
should be defined eg Apgar <7 10
- Umbilical artery pH better 05
Other factors during CS eg. position of patient, type of anaesthesia,
uterine incision to delivery interval etc can affect pH and Apgar
(other than antepartum hypoxia) 10
iii) 5 min Apgar scores of the Positive FAST and Reactive NST babies delivered within
24 hrs not recorded (some would have been in labour and may have delivered vaginally)
3. Comment on the conclusions
i} Conclusions totally invalid – wrong design : Impossible to conclude 20
No meaning in the conclusion regarding CS (two different populations!)
4. What statistical test would have been used to compare the sensitivity of FAST
& NST and get p < 0.001?
Chi square test 10
5. Other than sensitivity, what other indices should be calculated when validating a diagnostic test?
Specificity
Negative Predictive Value
Positive Predictive Value
Accuracy
Likelihood Ratio of a Positive test
Likelihood Ratio of a Negative test
3. This table is to be shown to the candidate
| Disease (+) | Disease (-) | Total | |
| Test (+) | 35 | 05 | 40 |
| Test (-) | 15 | 445 | 460 |
| Total | 50 | 450 | 500 |
- Please demonstrate how the following indices are calculated.
Sensitivity = 35/ 50 (20)
Specificity = 445/ 450 (20)
Positive Predictive Value = 35/ 40 (20)
Negative Predictive Value = 445/ 460 (20)
Accuracy = 35 +445 /500 (20)
(100)
About the Author
