Research

---

 

Scientific Session

 

 

 

Prevalence and predictors of folic acid use during pregnancy in a large Irish obstetric cohort

McGuire M, Cleary B, Sahm L, Murphy DJ
Department of Pharmacy and Department of Obstetrics & Gynaecology, The Coombe Women and Infants University Hospital, and Trinity College Dublin.

Introduction: Ireland has one of the highest incidences of Neural Tube Defects (NTDs) in Europe, affecting
approximately 1-1.5 per 1,000 births nationally. Neural tube defects (NTDs) are severe abnormalities of the central nervous system that develop in babies during the first weeks of pregnancy. Evidence has led to the belief that deficiency FA is a cause of NTDs. FA may exert its action through correction of a metabolic defect.

Objectives:The objectives of this study were to determine the prevalence and socio-economic predictors of folic acid use during pregnancy in a large obstetric cohort.

Setting: This descriptive study was carried out on the medical records database of the Coombe Women and Infants University Hospital, Dublin covering 2000-2007.

Method: A descriptive cohort study was carried out on the medical records database of the Coombe Women and Infants University Hospital, Dublin covering 2000-2007. Data was analysed using Statistical Package for the Social Sciences Version 15 (SPSS®).

Key findings:

• 85% of women reported taking FA at some time.
• 27.7% of women reported taking FA optimally.
• Being, Irish, of higher SES, nulliparous, married, of older age and having a planned pregnancy was associated with
  higher reported FA uptake and optimal use.
• Smoking, illicit drug and higher alcohol use and social worker referral were associated with lower FA use and optimal use.
• Patients receiving fertility treatment had the highest reported optimal FA use.
  Conclusion: Socio-Economic factors do influence use of FA. Vulnerable groups such as the unemployed, women less than 20 years of age and women with unplanned pregnancies should be specifically targeted in future Public health Campaigns. A multi-pronged approach involving fortification, taking FA supplements and eating a diet rich in FA and a multimedia campaign may be more successful in improving FA uptake in women of child bearing age.

 

Are Fetal Abdominal Subcutaneous Tissue (FAST) measurements using antenatal ultrasound different between male and female fetuses?

E Harrold, N Farah, C Fattah, S Barry, V Donnelly*, G Rafferty*, B Stuart, MJ Turner.
UCD School of Medicine and Medical Sciences, Coombe Women and Infants University Hospital, Mount Carmel Hospital*.

Abnormal fetal growth increases the complications of pregnancy not only for the baby but also for the mother.
Despite the use of more than 50 different formulae to determine EFW, ultrasound has its limitations especially at the extremes of fetal weight. There is emerging interest in fetal body composition. Babies born to diabetic mothers or born growth-restricted have abnormal adiposity. After birth, the pattern and distribution of subcutaneous fat has been assessed by various methods and gender differences have been reported. The purpose of this study was to determine if the FAST measurements using antenatal ultrasound differ between the sexes.

Women who had an ultrasound examination for fetal growth between 20 and 40 weeks gestation were analysed. Women with diabetes mellitus were excluded. The fetal anterior abdominal subcutaneous tissue was measured in a standardised way. The fetal sex was recorded after delivery (EH).

A total of 557 fetuses were measured, 290 male and 267 female. The FAST increased at the same rate for both male and female fetuses and at any given week there was no gender difference. The increased fat composition in females reported after birth was not found in abdominal wall subcutaneous fat measurements using ultrasound during pregnancy. Antenatal centile charts for FAST do not need to be based on gender.

 

Complications related to chorionicity in twin pregnancies.

N Farah, J Hogan, B Stuart, S Daly.
Coombe Women and Infants University Hospital.

Twin pregnancies are associated with higher complication rates when compared to singletons. They are mainly at an increased risk from preterm delivery and perinatal death. We sought to investigate these risks and comparing them between monochorionic and dichorionic pregnancies.

We analysed our twin database from July 1999 till July 2007 and included all twins where one twin weighed more than 500g or was over 24 week’s gestation. Chorionicity was assigned using information on the sex of the babies, the ultrasound findings and the pathology reports.

There were 972 twin pairs delivered, that met the inclusion criteria, of which 740 (76%) were dichorionic, 221 (22.7%) monochorionic and in 11 twin pairs we were not able to assign chorionicity. Table 1 outlines the risk of preterm delivery for both monochorionic and dichorionic twins. The risk of perinatal death was 9.7% (43) for monochorionic twins while 2.8% (42) for dichorionic.

Monochorionic twins are more likely to result in perinatal death and preterm delivery. The data presented is useful in counselling patients with twin pregnancies.

Gestation (weeks)	Monochorionic %	Dichorionic %
By 26	3.6 (n 8)	2 (n 15)
By 28	9 (n 20)	2.3 (n 17)
By 32	17.2 (n 38)	6.6 n 49)
By 36	39.4 (n 87)	27.4 (n 203)
By 38	74.7 (n 165)	63.9 (n 473)
By 40	100 (n 221)	99 (n 733)

 

TLR-9 senses human fetal DNA.

Andrea S. Nugent*#, Sean Daly*, John O’Leary* and Luke AJ O’Neill# PhD
#School of Biochemistry and Immunology, Trinity College Dublin.
*The Coombe Women and Infants University Hospital, Dublin, Ireland.

Toll-like receptor 9 (TLR9) senses CpG motifs in DNA. These are more common in bacterial and viral DNA and TLR9 has been shown to have an important role in the sensing of various pathogens during host defence. Recent studies have suggested that abnormal epigenetic changes in CpG-rich islands in fetal mammalian DNA might contribute to the high rate of early pregnancy loss. It is also well known that higher concentrations of free fetal DNA are found in mothers who deliver prematurely. We therefore wished to test the hypothesis that fetal DNA might be sensed by TLR9, and might provoke an inflammatory response, which in turn could lead to preterm labour and early pregnancy loss.

Our investigations have shown that fetal DNA added to the Namalwa B cell line (which expresses high levels of TLR9) or PMBCs could rapidly activate NF-kappaB (as measured by increased I-kappaB degradation) and also p38 activation, both of which are typical TLR9 signals. We also found increased production of the pro-inflammatory cytokines IL6 and TNF. The effects of fetal DNA were more potent that either synthetic CpG – containing oligonucleotides, or adult DNA. We also found that chloroquin, which has been shown to inhibit TLR9 signaling, blocked the effect. Finally, we have found that the effect of fetal DNA on cytokine induction is significantly reduced in TLR9-deficient bone marrowderived macrophages. We have therefore made the novel observation that TLR-9 senses fetal DNA and facilitates an inflammatory reaction. This may then contribute to preterm labour or early pregnancy loss.

 

Real-time PCR analysis for the rapid early detection of Pseudomonas aeruginosa in paediatric Cystic Fibrosis patients.

Logan, Catriona1, Lennon, Grainne1,2, Habbington, Adele1, O’Leary John J.1,3 and O’Sullivan Niamh1,3.

1. Molecular Microbiology Department, Our Lady’s Children’s Hospital, Crumlin, Dublin 12.
2. The Children’s Medical and Research Foundation, Our Lady’s Children’s Hospital, Crumlin, Dublin 12.
3. Pathology Department, Coombe Women and Infants University Hospital, Dolphins Barn, Dublin 8.

Introduction: Recurrent chronic chest infections caused by Pseudomonas aeruginosa are a common complication of Cystic fibrosis (CF). Initial P. aeruginosa infections are intermittent and sensitive to anti-microbials. However mutations in Pseudomonas mucA occur during infection, results in conversion to a ‘mucoid’ phenotype. Mucoid P. aeruginosa cannot be eradicated with antibiotics, resulting in permanent chronic colonization of CF lungs. Conversion to mucoidy can occur as early as 3-months from initial infection. Epidemic spread of P. aeruginosa isolates among CF patients has been reported. Prompt diagnosis and early antibiotic treatment are essential, as P. aeruginosa eradication is only likely prior to conversion mucoidy.

Methods: Real-time PCR assays for P. aeruginosa detection were designed, optimised and validated for the analysis of total nucleic acids extracted from respiratory specimens. More than 2000 respiratory specimens from 175 CF patients received for molecular analysis over an 18-month period were analysed by PCR and results compared with conventional microbiological sputum culture results.

Results: Real-time PCR analysis results in increased detection of P. aeruginosa in CF respiratory specimens as compared to conventional microbiological culture.

The UK Manchester and Liverpool epidemic P. aeruginosa strains were not detected in the patient cohort.

Real-time PCR highlights the inadequacy of throat swabs for the determination of P. aeruginosa colonisation status. Analyses identified >5% of specimens as “insufficient for PCR analysis” based on criteria defined for the maximum Ct value for human beta-actin DNA detection. Specimen type was the predominant factor in relation to the observance of specimens “insufficient for PCR analysis” with 14.0% of throat swabs and 6.1% of sputa deemed insufficient. Of specimens deemed “insufficient for PCR analysis”, >95% were reported culture negative for P. aeruginosa.

 

Prevalence and Genotype Distribution of Human Papillomavirus in the Irish Cervical Screening Population.

Mc Inerney J1,2, Pilkington L1, Keegan H1, Ring M1, Bolger N1, Casey M3, Sheils O2, O’Leary JJ 1,2, Martin CM 1,2

1. Department of Pathology, The Coombe Women and Infants University Hospital, Dublin 8, Ireland.
2. Department of Histopathology, University of Dublin, Trinity College, Dublin 2, Ireland.
3. Department of Pathology, University College Hospital Galway, Ireland.

Objectives: Human Papillomavirus (HPV) infection is the causative agent in the development of cervical cancer. HPV types 16 and 18 are detected in greater than 70% of cervical carcinomas. Two vaccines, Cervarix™ and Gardasil®, have recently come on the market targeting these high risk types. With the introduction of a national HPV vaccination program, it is essential that HPV genotype prevalence in the Irish screening population is determined. The aim of this study is to determine prevalence and genotype distribution of HPV in the Irish screening population. This study forms part of the CERVIVA program funded by the Health Research Board, Ireland.

Methods: To date, 1229 Thin Prep® cervical specimens from women attending GP clinics for routine cervical smear tests were recruited to the study through the Department of Cytology, Coombe Women & Infants University Hospital, Dublin and University College Hospital, Galway. Samples recruited to these centres are representative of the screening population in Leinster and Connaught. Cytological diagnosis was made according to BSCC guidelines. High risk HPV DNA was detected using Hybrid Capture II assay (hc2, Digene Ltd., UK). hc2 detects 13 different high risk HPV types (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68) in cervical specimens. Samples positive for HPV DNA were genotyped for 37 HPV types using the Roche Linear Array HPV Genotyping Test.

Conclusions: The prevalence of HPV in this screening population was 18%. Abnormal cytology was present in 10% of samples in this population. HPV16 (31.5%) was the most common type detected, followed by HPV39 (13%), HPV18 (12%), HPV66 (12%) and HPV 31(10.6%). The presence of high risk types other than 16 and 18 highlights the importance of continued cervical screening in a HPV vaccinated population. Further studies are underway to determine prevalence throughout Ireland.

 

Human Papillomavirus Prevalence and Persistence, in the Cervix of Women with Human Immunodeficiency Virus.

Martin C.M.1, Mc Inerney J1, Keegan H1, Pilkington, L1, C. Ni Cheallaigh2, S. Delamere2, M. Griffin3, G. McHugh2, P. Webster3, M. Barrett3, F. Lyons2, C. Bergin2, F. Mulcahy2, S. Clarke2, Sheils O1, O’Leary JJ 1

1. Department of Histopathology, University of Dublin, Trinity College, Dublin 2, Ireland & The Coombe Women and Infants University Hospital.
2. Department of Genitourinary Medicine and Infectious Diseases, St James’s Hospital, Dublin 8, Ireland.
3. Department of Cytology, St James’s Hospital, Dublin 8, Ireland.

Objectives: HPV is the major aetiological factor in the development of cervical intraepithelial neoplasia (CIN) and cervical cancer. Human immunodeficiency virus (HIV) positive patients are at greater risk for CIN and persistent HPV infections. In this study we examine the prevalence and persistence of HPV in cervical specimens from women with HIV and the effect of Antiviral Retro Therapy (ART) on HPV status and cervical abnormalities. This study forms part of the CERVIVA program funded by the Health Research Board, Ireland

Methods: PreservCyt™ cervical smears from 150 HIV positive women were prospectively recruited through the
Genitourinary and Infectious Disease Clinic at St James’s Hospital Dublin. Samples are taken at time of recruitment and annually where patient is stablised on ART or 3 monthly where patient has commenced ART. CD4 counts and HIV viral load were determined for each patient at baseline. Cytological diagnosis was made according to BSCC guidelines. High risk HPV DNA status was assessed using Hybrid Capture 2 (Digene) (hc2). HPV E6/E7 mRNA expression was detected using PreTect HPV Proofer (Norchip) which detects HPV 16, 18, 31, 33, and 45.

Results: 29% (44/150) of HIV positive women examined in this study had abnormal cytology. The HPV prevalence rate in this population of HIV positive women was 50% (75/150), while the prevalence of HPV E6/E7 mRNA expression in this population was 19% (28/150). Among the women with cytological abnormalities 86% were positive for high risk HPV DNA with 39% positive for HPV E6/E7 mRNA. HPV 45 (11/28) was the most predominant genotype followed by HPV 33 (9/28), 16 (8/28), 18 (7/28) and 31 (2/28). The rate of multiple HPV type infections was 32%. In women with follow up smears on ART, HPV persistence rates for HPV DNA and RNA were 32% and 21% respectively. The rate of HPV positivity correlated with low CD4 counts (<200 x 10 6/L), while no correlation was observed with HIV viral load.

 

Reduction of severity of pruritus after elective caesarean section under spinal anaesthesia with intra-thecal morphine: A comparison of prophylactic administration of Granisetron with Ondansetron.

T. Tan, R. Ojo, S. Immani and M. Carey.
Department of Peri-operative Medicine, The Coombe Women and Infants University Hospital.

The incidence of pruritus in parturients after elective caesarean section under spinal anaesthesia with intrathecal morphine has been reported to be between 60% to 100%, and is a common cause of maternal dissatisfaction. Ondansetron has been shown to reduce pruritus but the effect is short-lived. The objective of this randomised controlled double-blind trial was to evaluate the anti-pruritic efficacy of Granisetron compared with Ondansetron. Eighty ASA I or II women undergoing elective caeserean section under spinal anaesthesia were randomised to one of two groups. Spinal anaesthesia was performed in both groups with 10mg of 0.5% hyperbaric bupivacaine, Fentanyl 25μg plus 150µg preservative free morphine administered intra-thecally. After delivery of the baby and umbilical cord clamped, one group (G) received granisetron 3mg IV; the seocnd group (O) received ondansetron 8mg IV at the equivalent time.

The two groups were similar in terms of age, gestational age, height and weight. According to visual analogue pruritus scores, patients in group G experienced less pruritus at 8 hours (P=0.003) and at 24 hours (P=0.01). Fewer patients in group G (8) required rescue anti-pruritic medication than group O (18) (P=0.03). There was also higher satisfaction scores in group G compared to group O (P=0.0252). There was no difference in the overall incidence of pruritus, nausea and vomiting, and visual analogue pain scores between both groups of patients.

Administration of Granisetron 3mg IV markedly reduces pruritus, use of rescue anti-pruritic medication, and improves satisfaction but does not reduce the overall incidence of pruritus when compared to Ondansetron 8mg IV.

 

Prescribed and Non-prescribed Medication Use in Early Pregnancy in a Prospective Cohort of Women

Brian J Cleary, Hajeera Butt, Judith Strawbridge, Paul Gallagher, Tom Fahey, Deirdre J Murphy.
Department of Pharmacy and Department of Obstetrics & Gynaecology, The Coombe Women and Infants University Hospital, and Trinity College Dublin.

Objectives: To determine the extent and nature of medication use in early pregnancy, exploring inappropriate
prescribing with potential for fetal harm and prescribing for known medical disorders.

Design: Descriptive study.

Setting: Tertiary referral obstetric hospital in Dublin, Ireland.

Participants: All women who booked for antenatal care and had a corresponding delivery record between 2000 and 2007.

Main outcome measures: Prevalence and determinants of medication use for any indication, inappropriate prescribing with potential for fetal harm referenced against FDA pregnancy categories and appropriateness of prescribing for chronic medical disorders referenced against clinical practice guidelines.

Results: Excluding folic acid, use of at least one medication was reported in 23989 (39.2%) pregnancies. 11970 (19.5%), 545 (0.9%) and 352 (0.4%) women reported taking OTC medications, illicit drugs (including methadone) and herbal medicines / supplements respectively. FDA category D and X medications were reported by 1913 (3.1%) and 1987 (3.2%) women. Factors associated with reporting use of any medication included being over 35 years of age, having an unplanned pregnancy, being single, smoking during pregnancy, drinking alcohol prior to pregnancy and being a private patient. Medications with potential for fetal harm were reported more often among those with unplanned pregnancies, the unemployed, single women, smokers and publicly-funded patients. Asthma, depression and hypertension were among the most commonly reported chronic medical disorders. Medications known to cause fetal harm were reported by 86 (15.7%) women treated for depression and 68 (20%) women treated for hypertension.

Conclusions: A significant proportion of women report using medications before booking for antenatal care. Women and prescribers need to be aware of the lack of safety data for many medications that are frequently taken during pregnancy and the need for pre-pregnancy planning. Prescribers should exercise caution when prescribing medications with potential to cause fetal harm to women of childbearing potential.

 

POSTER PRESENTATIONS

 

 

 

ECSSIT – Elective Caesarean Section Syntocinon® Infusion Trial: The first 100 patients.

Sharon Sheehan, Michael Carey, Deirdre Murphy, on behalf of the ECSSIT Group.

Objective: The aim of ECSSIT is to compare the blood loss at elective lower segment caesarean section with
administration of oxytocin 5IU bolus versus oxytocin 5IU bolus and oxytocin 40IU infusion. We will analyse the first 100 patients recruited at the Coombe Women and Infants University Hospital to this trial. The baseline characteristics, operative factors and peri-operative outcomes of these patients will be assessed, blinded.

Study design: Women booked for elective caesarean section are recruited to the randomised controlled trial and randomised to receive either oxytocin bolus and placebo infusion or oxytocin bolus and oxytocin 40 IU infusion. The outcome measures are measured blood loss at caesarean section and the need for an additional uterotonic agent.

Results: Of the first 100 women recruited, 70% had a previous caesarean section. The total mean measured blood loss was 822mls. 27 women had a major haemorrhage (measured blood loss >1000ml). An additional uterotonic agent was required in 15 women. 18 women had pre-op anaemia (Hb<10.5d/dl), whilst 55 women had post-operative anaemia. The trial infusion was used in 96 women and discontinued in 12.

Conclusion: It is interesting to see the high incidence of major obstetric haemorrhage (27%) among this population. We eagerly await the conclusion of the trial and to see the results in each arm.

The study aims to recruit 2000 women, 1000 in each arm. To date (beginning September 2008), 400 have been recruited across the 3 Dublin maternity hospitals. The target end date is Summer 2010.

 

Caesarean section surgical-site infection: incidence and risk factors

Authors: Adeeb Khalifeh, Anne-Marie Meenan*, Rosena Hannify*, Niamh O’Sullivan*, Chris Fitzpatrick, Michael J Turner UCD School of Medicine and Medical Science & the Microbiology Department*, Coombe Women and Infants University Hospital, Dublin 8, Ireland.

Objective: To determine the incidence of wound infection and independent risk factors following caesarean section. Patients and Methods: A prospective audit was conducted on 99 women who delivered by caesarean section in July- August 2008. All women received prophylactic intravenous cefuroxime (1.5 g intravenously) given at cord clamping. All wounds, belonging to the clean-contaminated class, were reviewed before discharge. Their microbiological results were also reviewed if they presented to the hospital after discharge. Different clinical parameters were recorded using a protocol adapted from Northern Ireland on caesarean section surveillance (See Table 1). The definition of surgical site infection was standardised.

Results: The overall wound infection incidence was 7% (7 patients); 5 were diagnosed as inpatients and 2 as
outpatients. Three were elective caesarean sections and four were emergency caesarean sections, one of which not in labour. All sections lasted less than one hour. There were no independent risk factors identified, including Diabetes Mellitus and immunosuppression. The organisms isolated are shown in Table 2. Wound infection did not prolong length of stay as inpatients.

Conclusion: The incidence of 7% is similar to previous studies elsewhere. The datasheet used in our analysis will be rolled out nationally by the Health Protection Surveillance Centre (HSE) as part of the strategy for the control of Antimicrobial Resistance in Ireland (SARI). The next step is to assess the timing of prophylactic antibiotics on a larger number of patients; would that influence the incidence of infection after caesarean section?

	BMI (Average)	Membrane rupture	Previous CS	Diabetes Mellitus	Immunosuppression
Total n=99	25.3kg/m2	4.6hrs	38	3	3
Infected wounds n=7	26.0 kg/m2	2.6hrs	1	1	0

 

Organisms	Number of cases
Staphylococcus Aureus	3
E Coli	2
Mixed Anaerobes	4
Enterococcus faecalis	3
Proteus Mirabilis	1
Group B Streptococcus	1

 

Vaginal delivery for the second twin, is it safe?

N Farah, J Hogan, B Stuart, S Daly.

Coombe Women and Infants University Hospital.
The mode of delivery in twins is controversial. Some institutions suggest that all twin pregnancies are delivered by caesarean section as there is significant morbidity in delivering the second twin by caesarean section. We analysed our twin database from July 1999 till July 2007 and included all twins where one twin weighed more than 500g or over 24 weeks gestation. There were 972 sets of twins identified that met the criteria, in 11 sets the chorionicity could not be assigned and these were not included in the analysis.
In 509 sets both twins delivered vaginally, in 17 cases the first twin delivered vaginally while the 2nd twin was
delivered by an emergency caesarean section. Of those delivered by emergency caesarean section there was only one case were the apgar score at 5 minutes was less than 7. In 8 cases twin 2 was heavier than twin 1, and in 13 cases the consultant was present at the delivery.

Vaginal delivery should be anticipated in twin pregnancies as in our population it occurred in 53%. The need to
perform an emergency caesarean section for the second twin is rare (3.2%). If the second twin is delivered by
caesarean section, this is not associated with any significant morbidity.

 

Does the Placenta help to identify Identical Twins?

J Hogan, N Farah, Z Galvin, A Radomski, J O’Leary, B Stuart, S Daly.
Coombe Women and Infants University Hospital, Dublin.

The placenta from twin pregnancies are routinely sent for histopathology in some institutions. Pathological
assessment to determine chorionicity is considered by many to be the gold standard. We sought to investigate if this is actually the case.

A retrospective analysis of our hospital’s database of all twins (delivered at or after 24 weeks gestation or weighing equal to or greater than 500g) from July 1999 to July 2007 was undertaken. The sex of the infants, the placental pathology and the ultrasound reports were reviewed.

Using the above criteria, there were 972 twin pairs. Of these, 806 pairs had the placenta sent for histological
examination (82.9%). Table 1 displays the classification of chorionicity by pathology. There were 198 classified
as monochorionic but 16 of these had different sex infants. This means that 8% had been wrongly classified as monochorionic when they were in fact divhorionic twins. In 12 of these cases, ultrasound concurred with the fetal sex and these twin pairs had been classified as dichorionic during their pregnancy.

Our study suggests that the placental pathological assessment of chorionicity is accurate in most but not all cases. If these results are replicated in other institutions, radiological plus histological determination of chorionicity may become the gold standard in the future.

Chorionicity	n
Dichorionic	583
Monochorionic	198
Unknown	25

 

Ultrasound Placental Localisation After One Previous Caesarean Section

J Hogan, N Farah, B Stuart, MJ Turner
UCD School of Medicine and Medical Science, Coombe Women and Infants University Hospital, Dublin.
A prior lower segment caesarean section has been reported as a risk factor for placenta praevia. Women with a prior caesarean section are more likely to require repeat section and abnormal placental localisation makes it technically more challenging. To our knowledge, there are no studies reported which identify the placental site in women with one prior caesarean section.

We reviewed women with one prior caesarean section who delivered over a four month period and who had the placenta localised during a formal ultrasound in the second or third trimester. The smoking history, previous uterine curettage and parity was also recorded.

Of 262 patients delivered with one prior caesarean section, 170 had a departmental ultrasound. Three patients had placenta praevia and there was no c