Retrospective study to investigate the possible relationship between excess blood loss at caesarean section and reduced intra-operative oxytocin dose.

OBJECTIVE: To investigate a possible relationship between the oxytocin dose at caesarean section and blood loss. STUDY DESIGN: Retrospective analysis of computerised data for all caesarean sections in a UK maternity unit delivering 6000 women annually during 1995-2009 and thus for seven years before and after the 2001 recommended change in oxytocin dose. Validation of computerised and hand-checked clinical data for 1996 and 2006 was performed and annual frequency of blood loss >1000 ml was observed. RESULTS: Validation showed most variables recorded were similar for both acquisition methods. For 17,405 (98.9%) caesarean sections with blood loss recorded, excess or severe loss occurred in 127 (1.6%) of 7177 cases during 1995-2001 compared with 362 (4.0%) of 9035 during 2003-2009 (OR 2.317, CI 1.888-2.843). It was significantly more frequent with multiple than singleton pregnancies (OR 1.946, CI 1.417-2.673), with general than neuraxial anaesthesia (OR 4.296, CI 3.479-5.305) and with non-longitudinal than longitudinal fetal lie (OR 1.942, CI 1.501-2.512). Excluding these three groups, excess blood loss was still more frequent during 2003-2009 than 1995-2001 (OR 3.181, CI 2.374-4.263). Oxytocin given during labour did not influence the frequency of excess blood loss. CONCLUSIONS: The increased rate of excess blood loss at caesarean section during the latter period could be the result of the reduced oxytocin dose. If similar observations are made by others, this possible relationship should be investigated with appropriate objective randomised studies.

Blood loss and blood transfusion at caesarean section: a prospective observational study covering 30 years.

OBJECTIVE: Attitudes to acute blood loss and transfusion have changed during the last 40 years. This study observed the trends in blood loss and transfusion rates at caesarean section during that period to identify any trends between 1976 and 2006. STUDY DESIGN: Prospective analysis of clinical notes of women delivered by caesarean sections in a major district hospital obstetric unit in the UK, delivering around 6000 annually during four 12-month periods every 10 years from 1976 to 2006. Details including demographic, pregnancy, delivery, blood loss, transfusion, and puerperal observations were recorded. RESULTS: 3222 of 22,998 women were delivered by caesarean section during the four study years, increasing from 7.2% in 1976 to 23.4% in 2006 (P<0.001). The median recorded blood loss was 500ml, which did not change significantly over the study years. The rate of excess blood loss however increased in low-risk cases in 2006 compared with 1996 (P<0.001); this increase followed the recommended restricted intra-operative oxytocin dose. Transfusion rates declined significantly from 22% in 1976 to 4-5% in 1996 and 2006 (P<0.001). CONCLUSIONS: Median blood loss remained steady for each of the study years but with an increase in excess blood loss cases in the last study year compared with the two previous study years. The explanation for this is presently uncertain, but was possibly influenced by the 2001 recommendation for a reduced dose of oxytocin at delivery. Transfusion rates declined, probably precipitated by anxieties over infections associated blood products. There was no indication of increased morbidity with the reduced transfusion rates accessed by the surrogate of post-delivery discharge times.

Evidence to support the single-dose over the two-dose protocol for routine antenatal anti-D Rhesus prophylaxis: a prospective observational study.

OBJECTIVE: To assess the potential benefit of the single-dose compared with two-dose protocol for routine antenatal anti-D prophylaxis (RAADP). STUDY DESIGN: Prospective observational study during February to October 2009 in a district teaching hospital delivering 6000 women annually. Obstetric notes of eligible RhD negative women were reviewed following delivery to identify gestation and dose for RAADP injections, with laboratory confirmation when necessary. The primary outcome was compliance administering RAADP at the appropriate gestation with the secondary analysis of compliance for nulliparae and multiparae. RESULTS: For 644 eligible women, there was no statistically significant difference in administering the 28 week injection between the two-dose (95%) and single-dose (98%) protocols (P=0.17, OR 2.47: 95%CI (0.73, 8.34)). Compliance giving the injection at the correct gestation at 28 ± 1 weeks was achieved in 78% and 77% respectively (P=0.72, OR 1.10: 95%CI (0.69, 1.74)). For the two-dose protocol the second injection was given to 96% of those who had received the first injection, but at the correct time point in only 67%. By eliminating the variability of the 34 week injection, the single-dose protocol resulted in significantly greater overall success in giving RAADP at the appropriate times compared with the two-dose protocol (P=0.021, OR 1.68: 95%CI (1.08, 2.64)). There were no differences between nulliparae and multiparae for both protocols. CONCLUSION: The single-dose protocol provides enhanced compliance over the two-dose protocol with the potential for reduced sensitisation combining economic and manpower benefits. There remains room for improvement in administering RAADP at the correct gestation.

Acupuncture for pain relief during induced labour in nulliparae: a randomised controlled study.

OBJECTIVE: To assess the role of acupuncture for analgesia during labour. DESIGN: Double-blind study of manual, electro and sham acupuncture, and single-blind study comparing acupuncture with a control group for analgesia for labour induction. SETTING: A major obstetric unit in the UK. POPULATION: A cohort of 105 nulliparae undergoing labour induction at term. METHODS: Twenty-three subjects needed to be randomised to each group to have an 80% power of detecting a 50% relative reduction in epidural rate with an alpha value of 0.05. MAIN OUTCOME MEASURES: The primary end point was the rate of intrapartum epidural analgesia, and the secondary end points were parenteral analgesia requirement, labour length, delivery mode, neonatal condition and postpartum haemorrhage. RESULTS: There was no difference in epidural analgesia between acupuncture and sham acupuncture, relative risk 1.18 (95% CI 0.8-1.74), or between acupuncture and control, relative risk 0.88 (95% CI 0.66-1.19). There were no significant differences in the secondary end points between the acupuncture groups and the control group. Side effects or complications of acupuncture were not identified. CONCLUSIONS: Using the protocols studied, there was no analgesic benefit with acupuncture for pain relief during induced labour in nulliparae.

A prospective cohort study of menstrual symptoms and morbidity over 15 years following laparoscopic Filshie clip sterilisation.

OBJECTIVE: To observe the incidence of menstrual symptoms and relevant surgery after sterilisation. STUDY DESIGN: 1101 women sterilised with Filshie clips between 1983 and 2002 were assessed prospectively comparing menstrual symptomatology documented immediately before surgery and 5-14 years later by questionnaire. MAIN OUTCOME MEASURES: Prevalence of menstrual dysfunction and subsequent surgery related to pre-operative menstrual symptoms and contraception. RESULTS: After excluding 232 (24%) of the 968 eligible women sent questionnaires whose address had changed, 573 of the remaining 735 women (78%) completed the questionnaire, 223 5-6 years after sterilisation, 175 after 7-9 years and 175 after 10-15 years; the respondents were demographically representative of the total population. Heavy periods increased from 9% before to 35% (P<0.0001) after sterilisation, painful periods from 2% to 21% (P<0.0001) and 6% had undergone hysterectomy or endometrial ablation. These findings were not influenced by the pre-sterilisation method of contraception but were inversely related to advancing age (P<0.0002). The lowest rates of menstrual symptoms were reported 10-15 years after sterilisation. CONCLUSION: Menstrual symptoms increased following Filshie clip sterilisation irrespective of pre-sterilisation symptoms and contraception. Whatever the causative mechanism, the progestogen-loaded intrauterine system (IUS), with similar efficacy but with improved menstrual symptoms, should be considered before sterilisation.

Why remove all cervical polyps and examine them histologically?

A retrospective analysis of 1366 cervical polyps showed that none had malignant features and 67% were removed from asymptomatic women. A policy removing only cervical polyps from symptomatic women or those with abnormal cervical cytology and limiting histological examination to these polyps would result in significant savings and reduce the small risk of morbidity associated with polypectomy.

Delivery after caesarean section: consultant obstetricians' professional advice and personal preferences.

The objective of this study was to determine how obstetricians would wish to be managed in their own pregnancy and their advice to patients delivering after a caesarean section. A questionnaire was sent to 219 consultant obstetricians in two large UK medical deaneries investigating their personal preferences for delivery and management of spontaneous, augmented and induced labour after a caesarean section and how they advise their patients related to the indication for the previous caesarean section. Responses were analysed according to age and gender. The questionnaire yielded a 68% response rate. None would counsel against labour unless there were contraindications. The majority would recommend labour for all indications for the previous caesarean section, although personal preferences were lower (p<0.04): 56% would recommend labour to their patients after a failed instrumental delivery, but only 36% would personally choose that option (p<0.002). Female obstetricians would contemplate and recommend labour more readily than males. Labour augmentation and induction was more frequently recommended (66% and 57%, respectively) than opted for personally (57% and 52%). Reluctance for labour augmentation and induction was greatest among younger consultants. While the majority of consultants favour labour for themselves and recommend this for their patients, it was reassuring to note that patients are encouraged to make the final decision. Informed patient choice is paramount and it is therefore important that full information about risks and benefits is available.

Failure and regret after laparoscopic filshie clip sterilization under local anesthetic.

OBJECTIVE: To estimate the failure, regret, and reversal rates 5 or more years after laparoscopic Filshie clip sterilization using local anesthesia. METHODS: A total of 1,101 women underwent Filshie clip sterilization between 1983 and 2002. They completed follow-up questionnaires that were analyzed for the following outcomes: failed sterilization, regret after the operation, and sterilization reversal. RESULTS: Two hundred thirty-three of 968 (24%) eligible women sent the questionnaire had moved from their last known address. Of the remaining 735 women, 573 (78%) completed the questionnaire: 223 (39%) 5-6 years after the operation, 175 (30%) after 7-9 years, and 175 (30%) after 10-15 years. One pregnancy occurred 10 months after surgery, and one woman had the procedure repeated when unilateral tubal patency was identified by hysterosalpingography 3 weeks after surgery. Twenty-four (4%) women regretted having the operation; 7 (1.2%) women had a reversal operation, and all subsequently conceived. CONCLUSION: Failure after tubal sterilization using Filshie clips is less than 1:500 operations. Patient selection and surgeons' experience may have influenced these results. Regret occurred in a small proportion. LEVEL OF EVIDENCE: III.

Impact of the severity of fetal distress on decision-to-delivery intervals for assisted vaginal delivery.

To correlate the severity of fetal distress assessed retrospectively using the Dublin intrapartum fetal monitoring classification, with decision-to-delivery intervals (DDI) and neonatal outcome, a prospective 3-month study of 78 consecutive instrumental deliveries was conducted. There were 66 cardiotocographs (CTGs) with an 83% majority agreement on classification: it was agreed 95% of deliveries should be expedited for the 58 interpretable CTGs. Although the more abnormal CTG patterns resulted in shorter DDIs and greater neonatal acidaemia, there were no significant correlations. Acidosis was present in 14% with a suspicious, or 22% an ominous CTG pattern. Delivery as a trial in theatre doubled the DDI at 41.5 +/- 22.7 (mean +/- SD) min, compared with 18.1 +/- 8.1 min in the labour room (p < 0.0001), with one case of severe acidosis in the former group. The prolonged DDI with a trial of delivery in theatre could adversely compromise the already 'distressed' fetus and should be used only when clearly indicated.

Does the surgical approach used for myomectomy influence the morbidity in subsequent pregnancy?

This is a retrospective analysis of post-myomectomy pregnancies beyond 20 weeks' gestation in a teaching hospital for a 16-year period. There were 117 pregnancies, with complete data available for 112 pregnancies, in 81 women. A total of 92 deliveries followed myomectomy at laparotomy, 1 at laparoscopy and 19 at hysteroscopy. Of 72 labours, 22 were induced and 33 were augmented. Ten of those who laboured were delivered by caesarean section. The most common indication for intra-partum caesarean section was poor labour progress. There were no significant differences in neonatal outcome for babies delivered after labour compared with elective caesarean section. Maternal blood loss was however significantly higher with elective caesarean section (OR 2.01; 95% CI 1.96 - 2.06, p < 0.001). We report one uterine rupture at 36 weeks' gestation that followed laparoscopic myomectomy. These data do not support the view that myomectomy performed at laparotomy or hysteroscopy warrant avoiding labour or oxytocic administration. Laparoscopic myomectomy may be associated with a greater risk of scar rupture.

Clinical validation of routine antenatal anti-D prophylaxis questions the modelling predictions adopted by NICE for Rhesus D sensitisation rates: results of a longitudinal study.

OBJECTIVE: To compare the rates of Rh(D) sensitisations with a policy of restricted routine antenatal anti-D prophylaxis (first pregnancy only) with the rates predicted with universal routine antenatal anti-D prophylaxis (all pregnancies). STUDY DESIGN: A retrospective longitudinal observational study involving 15,500 confinements in Rhesus D negative (Rh(D)-ve) women between 1990 and 2003 in a single health district was conducted. All Rh(D) sensitised pregnancies were identified and evidence for routine antenatal anti-D prophylaxis administration during the first pregnancy was investigated. The rate of Rh(D) sensitisations following a policy of restricted prophylaxis was compared with that predicted with mathematical modelling following universal prophylaxis. RESULTS: There were 50 newly sensitised and 37 previously sensitised pregnancies among 15,596 Rh(D)-ve women. For the calculated 13,575 Rh(D)-ve women whose first confinement was in Oxford and who were eligible for restricted prophylaxis, there were 30 new and 26 previously sensitised pregnancies. Of these 30 new sensitisations, 10 were nulliparae, 12 parity 1, and eight parity 2 or greater (third or later continuing pregnancy); only one of these latter eight women had received routine prophylaxis, four had delivered their first baby before the programme was introduced, and in three documentary evidence could not be confirmed that prophylaxis had been given. There was no difference between a policy of restricted and universal routine antenatal anti-D prophylaxis in the sensitisation rates for women during their third or subsequent pregnancy. CONCLUSIONS: This study has shown that restricted routine antenatal prophylaxis provides continuing protection for subsequent pregnancies although the mechanism for this is unclear. These results challenge the wisdom and expense of a policy of universal prophylaxis and prompt a need for further similar analyses to test the appropriateness of the NICE guideline.

Management of shoulder dystocia: trends in incidence and maternal and neonatal morbidity.

OBJECTIVE: To investigate trends in the incidence of shoulder dystocia, methods used to overcome the obstruction, and rates of maternal and neonatal morbidity. METHODS: Cases of shoulder dystocia and of neonatal brachial plexus injury occurring from 1991 to 2005 in our unit were identified. The obstetric notes of cases were examined, and the management of the shoulder dystocia was recorded. Demographic data, labor management with outcome, and neonatal outcome were also recorded for all vaginal deliveries over the same period. Incidence rates of shoulder dystocia and associated morbidity related to the methods used for overcoming the obstruction to labor were determined. RESULTS: There were 514 cases of shoulder dystocia among 79,781 (0.6%) vaginal deliveries with 44 cases of neonatal brachial plexus injury and 36 asphyxiated neonates; two neonates with cerebral palsy died. The McRoberts' maneuver was used increasingly to overcome the obstruction, from 3% during the first 5 years to 91% during the last 5 years. The incidence of shoulder dystocia, brachial plexus injury, and neonatal asphyxia all increased over the study period without change in maternal morbidity frequency. CONCLUSION: The explanation for the increase in shoulder dystocia is unclear but the introduction of the McRoberts' maneuver has not improved outcomes compared with the earlier results. LEVEL OF EVIDENCE: II.

The impact of a trial of instrumental delivery in theatre on neonatal outcome.

OBJECTIVE: To observe the effect of a trial of instrumental delivery in theatre on outcome for mother and baby. DESIGN: A prospective observational study. METHODS: Relevant maternal and neonatal data were collected for all instrumental deliveries of singleton viable pregnancies delivered over a three month period. POPULATION: Two hundred and twenty nine consecutive deliveries conducted by ventouse or forceps because of fetal distress or dystocia. SETTING: The maternity unit of a teaching hospital delivering around 6000 women annually. MAIN OUTCOME MEASURES: The decision-to-delivery intervals (DDI), mode of delivery and neonatal condition at birth. RESULTS: Sixty (26%) deliveries were managed as a trial in theatre, 46 (77%) because of prolonged second stage, with malposition being a factor in 39, and 14 (23%) because of fetal distress. The mean +/- SD DDI for these 60 deliveries was 59.2 +/- 20.4 minutes (median 58 minutes) compared with 21.2 +/- 9.0 minutes (median 20 minutes) for 169 delivered in the labour room (P < 0.0001). Of these 169 deliveries, 168 were delivered within 46 minutes and 1 delivered by caesarean section at 60 minutes. Nine women (13%) ultimately delivered by caesarean section, eight following a trial in theatre; in seven, there was malposition. Deliveries following a trial had slightly less favourable cord blood gas results. CONCLUSIONS: Trial of instrumental delivery takes two to three times longer than delivery in the labour room; fetal malposition was the major indication for the trial of instrumental delivery and reason for failed delivery. Adopting the recent guidelines of the Royal College of Obstetricians and Gynaecologists, at least 107 (47%) should have been managed as a trial in theatre. The added delay in delivery could be damaging to an already hypoxic fetus, and the use of a trial should be individually assessed.

Compliance with routine antenatal rhesus D prophylaxis and the impact on sensitisations: observations over 14 years.

Documented routine antenatal anti-D prophylaxis was given to 90% and 81-87% of eligible women at 28 and 34 weeks of gestation, respectively, during the early 1990s and early 2000s. With increasing experience and education, a significant improvement in the timing of the first (OR 0.26, 95% CI 0.16-0.41: P < 0.0001) and second injections (OR 0.40, 95% CI 0.26-0.61: P < 0.0001) occurred during the latter period. Despite these improvements, there was no reduction in the sensitisation rate at 0.4%. However, this low rate occurred despite significant proportions of women delivering more than 42 days after the second injection. Fifteen of the 16 sensitised women had received routine antenatal prophylaxis.

Induction of labour at the start of the new millennium.

Around 20% of all deliveries are preceded by labour induction, a proportion that has not varied dramatically over recent years. Fetal death was the only indication for labour induction centuries ago, while this is now a very rare indication, with prolonged pregnancy and maternal hypertensive disorders being the major indications for the last 50-60 years. Techniques for inducing labour have also changed from dietary delicacies and verbal threats giving way to physical stimulation mainly achieved by cervical stretching and amniotomy and more recently to sophisticated pharmacological manipulation using oxytocin and prostaglandins, based on our expanding knowledge of the physiological processes involved in spontaneous parturition. Relaxin, antiprogestins, nitric oxide as well as complementary medicines have also been explored in recent years. Successful induction is, however, still not guaranteed and there has been increasing emphasis during the past decade on exploring strategies for identifying the probability of success. Measurement of fetal fibronectin in cervical mucus, maternal serum nitrite/nitrate concentrations, ultrasound delineation of cervical form and electrical impedance measurements across the cervix are all being investigated. Safety, success, and patient satisfaction continue to be the major objectives with economic evaluations now becoming a significant factor in the search for the ideal induction method.

The kinetics of routine antenatal prophylactic intramuscular injections of polyclonal anti-D immunoglobulin.

OBJECTIVE: To observe the pharmacokinetics of intramuscular anti-D immunoglobulin (IgG) given for routine antenatal prophylaxis. DESIGN: Prospective observational study. SETTING: Maternity unit and antenatal serology laboratory in a district teaching hospital. POPULATION: Forty-five rhesus-D-negative pregnant women not sensitised to RhD. METHODS: Serial serum quantitations of anti-D IgG following the intramuscular injections of anti-D IgG 100 microg (500 IU) at 28 and 34 weeks of gestation. Anti-D IgG concentrations were assayed with the RFA-300 continuous flow analyser. MAIN OUTCOME MEASURES: The kinetic profile and duration of detectable anti-D IgG in maternal serum following the first and second injections of anti-D IgG. RESULTS: For the 43 women in whom serial data were collected, there were no detectable differences between pregnancies with an RhD-positive (26) or -negative (17) fetus. Maximum IgG concentrations were detected two to five days following the first anti-D IgG injection and ranged between 0 and 28 ng/mL. Only 30% of women still undelivered at 40 weeks of gestation had detectable IgG at 2 ng/mL or greater. There was a significant relationship between higher maximum values and low maternal surface body area (R2 = 0.204, P = 0.002), but this did not influence duration of persistent IgG. CONCLUSION: Using previously published data, 70% women are not adequately protected with anti-D IgG 12 weeks after the first prophylactic injection. Despite this, previous clinical results suggest that the antenatal prophylaxis schedule used provides adequate protection and that the recommendation for the lowest concentration of protective anti-D IgG antibody levels currently in use is probably overestimated.

Does a first trimester dating scan using crown rump length measurement reduce the rate of induction of labour for prolonged pregnancy? An uncompleted randomised controlled trial of 463 women.

OBJECTIVE: To evaluate the effect of a first trimester ultrasound dating scan on the rate of induction of labour for prolonged pregnancy. DESIGN: Randomised controlled trial to include 400 women in each arm of the trial. SETTING: Participating general practices and a district general teaching hospital. POPULATION: Women attending their general practitioner in the first trimester to confirm pregnancy, in whom a first trimester ultrasound scan was not indicated. METHODS: Women randomised to the study group (scan group) underwent an ultrasound dating scan between 8 and 12 weeks, measuring crown-rump length. The estimated date of delivery (EDD) was changed if there was a discrepancy of more than 5 days from the gestation, calculated from the last menstrual period (LMP). For the remaining women (no-scan group), gestation was determined using the LMP. MAIN OUTCOME MEASURES: The rate of induction of labour for prolonged pregnancy. RESULTS: Due to circumstances beyond the researchers' control, recruitment was abandoned when 463 women had been enrolled. The EDD was adjusted in 13 (5.7%) women in the scan group and in 2 (0.9%) in the no-scan group. There was no difference in the rate of induction for prolonged pregnancy between the scan (19 [8.2%]) and the no-scan (17 [7.4%]) groups (relative risk 1.10; 95% CI 0.59-2.07). CONCLUSIONS: Acknowledging the reduced numbers recruited for study, it is concluded that there is no evidence that a first trimester ultrasound dating scan reduces the rate of induction of labour for prolonged pregnancy and may result in a more expensive healthcare strategy.

Does discussion of possible scar rupture influence preferred mode of delivery after a caesarean section?

Using retrospective and prospective analyses of antenatal records, it was found that by 2003, discussions about the options for delivery after one lower segment caesarean section (LSCS) were almost always documented in antenatal notes, compared with only rare entries in 1993; specific mention of the risks of scar rupture were made in just under 50% in 2003. There was a considerable reduction in the proportion of women whose preference was to labour in 2003 compared with 1993 and also in the number who ultimately delivered vaginally. There was, however, no evidence that those women with whom possible scar rupture had been discussed were discouraged from attempting vaginal delivery. These data suggest that, contrary to expectations, increasing patients' involvement in their management in this situation does not result in fewer caesarean sections.

Efficacy of follow-up and contact tracing of women who test positive for genital tract chlamydia trachomatis prior to pregnancy termination.

We examined the efficacy of follow-up, contact tracing and the need for retreatment in women who were screen-positive for genital tract Chlamydia trachomatis prior to termination of pregnancy. Eighty-six of 1363 (6.3%) women screened positive. These women were significantly younger than those who screened negative (P < 0.0001). The genitourinary medicine (GUM) clinic was notified of 73 (84.9%) screen-positive women and 41 (47.7%) attended for follow-up. Contact tracing was undertaken in 38 (92.7%) women who attended and 29 (70.7%) women who attended required retreatment for Chlamydia. The median duration between pregnancy termination and GUM clinic attendance was significantly longer in women who required retreatment compared to those who did not require retreatment (P = 0.003). In conclusion, follow-up and contact-tracing of women who screen positive for genital tract C. trachomatis was incomplete. This may substantially compromise the cost-effectiveness of a screen-and-treat programme.

The usefulness of ultrasound assessment of amniotic fluid in predicting adverse outcome in prolonged pregnancy: a prospective blinded observational study.

OBJECTIVE: To determine whether a single ultrasound scan at or beyond 40 weeks of gestation to detect a single deepest pool of amniotic fluid <2 cm and amniotic fluid index (AFI) <5 cm is clinically useful in the prediction of subsequent adverse pregnancy outcome. DESIGN: A prospective double blind cohort study. SETTING: A university teaching hospital delivering approximately 6000 women annually. POPULATION: One thousand and five hundred and eighty-four pregnant women at or beyond 40 weeks of gestation. METHODS: Ultrasound assessment of liquor to detect the single deepest pool of amniotic fluid and derive the AFI at or after 40 weeks of gestation. MAIN OUTCOME MEASURES: Perinatal death, meconium aspiration, birth asphyxia, intervention in labour for fetal distress, a cord arterial pH <7 and admission to the neonatal unit. RESULTS: An AFI <5 cm but not a single deepest pool <2 cm was significantly associated with birth asphyxia or meconium aspiration. An AFI <5 cm was also significantly associated with caesarean section for fetal distress in labour, a cord arterial pH <7 at delivery and low Apgar scores. Despite there being a statistically significant association with adverse outcomes the sensitivity of AFI was low at 28.6%, 12% and 11.5% for major adverse outcome, fetal distress in labour or admission to the neonatal unit, respectively. CONCLUSIONS: The AFI is superior to a measure of the single deepest pool as an assessment of the fetus at or after 40 weeks but has a poor sensitivity for adverse pregnancy outcome. Routine use is likely to lead to increased obstetric intervention without improvement in perinatal outcomes.