Severe maternal morbidity trends over 20 years in a tertiary referral stand-alone maternity unit.

OBJECTIVE: Severe maternal morbidity (SMM) is a better indicator of quality of care than maternal mortality, which is a rare event. Risk factors such as advanced maternal age, caesarean section (CS) and obesity are increasing. The aim of this study was to examine the rate and trends in SMM at our hospital over a 20-year period. STUDY DESIGN: Retrospective review was performed of cases of SMM from January 1st 2000 to December 31st 2019. Yearly rates for SMM and Major Obstetric Haemorrhage (MOH) were calculated (per 1000 maternities) and linear regression analysis was used to model the trends over time. Average SMM and MOH rates were also calculated for the periods 2000-2009 and 2010-2019 and compared using a chi-square test. The patient demographics of the SMM group were compared to the background population delivered at our hospital using a chi-square test. RESULTS: 702 women with SMM were identified out of 162,462 maternities over the study period yielding an incidence of 4.3 per 1000 maternities. When the two time periods (2000-2009 and 2010-2019) are compared, the rate of SMM increased 2.4 vs 6.2 (p < 0.001), largely due to an increase in MOH 1.72 vs 3.86 (p < 0.001) and pulmonary embolus (PE) also increased 0.2 vs 0.5 (p = 0.012). Intensive-care unit (ICU) transfer rates more than doubled 0.19 vs 0.44 (p = 0.006). Eclampsia rates decreased 0.3 vs 0.1 (p = 0.047) but the rate of peripartum hysterectomy 0.39 vs 0.38 (p = 0.495), uterine rupture 0.16 vs 0.14 (p = 0.867), cardiac arrest (0.04 vs 0.04) and cerebrovascular accidents (CVA) (0.04 vs 0.04) remained unchanged. Maternal age > 40 years 9.7% vs 5% (p = 0.005), previous CS 25.7% vs 14.4%; p < 0.001 and multiple pregnancy 8 vs 3.6% (p = 0.002) were more prevalent in the SMM cohort compared to the hospital population. CONCLUSIONS: Overall, rates of SMM have increased threefold and transfer for ICU care has doubled over 20 years in our unit. The main driver is MOH. The rate of eclampsia has decreased and peripartum hysterectomy, uterine rupture, CVA and cardiac arrest remain unchanged. Advanced maternal age, previous caesarean delivery and multiple pregnancy were more prevalent in the SMM cohort compared to the background population.

Analgesia, anaesthesia and obstetric outcome in women with inherited bleeding disorders.

OBJECTIVE: Vertebral canal haematoma (VCH) complicates 1 in 168,000 obstetric epidurals (Ruppen et al., 2006). This risk is increased in women with inherited bleeding disorders (IBD). The impact of a contraindication to regional anaesthesia on pain management and obstetric outcome in these women is unknown. The purpose of this study was to determine anaesthetic use and obstetric outcomes in a cohort of women with IBD. STUDY DESIGN: 97 women with IBD that delivered 130 babies at the CWIUH from Jan 2011 to Dec 2016 were identified from a maternal medicine database. Multidisciplinary planning of peripartum care was communicated to labour ward staff using a simple checklist. The primary bleeding disorders were: Von Willebrands disease (VWD) Type 1 27 (27.8%); VWD Type 2A 3 (3.8%); Low VWF 3 (3.8%); Bleeding disorder of unknown aetiology (BDUA) 19 (19.6%); deficiency of Factors VII, VIII, IX, X, and XI 13 (13.4%); Carriers of Factor VIII, IX, X, XIII deficiency 17 (17.5%); 5 had combined deficiencies (5.2%) and there was one platelet function defect. 9 had a family history of a bleeding disorder (9.3%). Haemostatic support, analgesia, mode of delivery and maternal and fetal outcomes were compared between pregnancies where regional anaesthesia was permitted and those that were not using the Chi-squared test. RESULTS: When pregnancies where regional anaesthesia was not recommended (49) were compared with pregnancies where regional anaesthesia was considered safe (81), the women were more likely to see an anaesthetist before labour 46 (94%) vs 46 (61%): p < 0.001; to require prophylactic haemostatic support for delivery 30 (61%) vs 1 (1%): p < 0.001; to use a remifentanil infusion 15 (31%) vs 0: p < 0.001, and have general anaesthesia for Caesarean Section (CS) 10 (20%) vs 1(1%): p < 0.001. Vaginal birth 35 (71%) vs 53(65%): p = 0.4 and CS rates 14 (29%) vs 26 (32%) p = 0.28 were similar. Postpartum haemorrhage (PPH) was more common 11 (24%) vs 9(12%) vs p = 0.07 but not statistically so. There were no cases of neonatal bleeding or VCH. CONCLUSION: Contraindication to neuraxial blockade in labouring women with IBD does not influence mode of delivery. This information is reassuring to these women who may be anxious about delivery without regional anaesthesia.

Prevention of thrombosis in pregnancy: how practical are consensus derived clinical practice guidelines?

Thromboembolic disease (TED) has, for many years, consistently been identified as one of the leading causes of direct maternal mortality. In November 2009, the RCOG published a guideline on the prevention of TED that has been rapidly adopted by hospital trusts in the UK. The aim of our study was to determine the number and profile of women in our population that would require treatment with low molecular weight heparin (LMWH) and the cost implications of such treatment if these guidelines were implemented. A retrospective review of the first 100 women who delivered at the Coombe Women & Infants University Hospital (CWIUH) in 2010 was conducted and risk stratification applied at the relevant time points. A total of 51% were deemed to be at intermediate or high risk of TED at some point during pregnancy. In 35 of the 51 women (70%), this risk was attributable to factors such as age>35 years, parity≥3, BMI>30 kg/m2 or cigarette smoking. In our obstetric population, the percentage of women with these risk factors was: 25.5%, 8.5%, 19% and 16.7%, respectively. Implementation of this guideline would increase the hospital annual expenditure on LMWH by a factor of 17. The strategy of attributing risk by accumulating factors that individually have a low risk of TED and are prevalent in the population needs to be re-visited. The cost of implementation of these guidelines is not inconsiderable in the absence of data to indicate that clinical outcome is improved with their implementation.

The efficacy of fibrinogen concentrate compared with cryoprecipitate in major obstetric haemorrhage--an observational study.

BACKGROUND: Fibrinogen replacement is critical in major obstetric haemorrhage (MOH). Purified, pasteurised fibrinogen concentrate appears to have benefit over cryoprecipitate in ease of administration and safety but is unlicensed in pregnancy. In July 2009, the Irish Blood Transfusion Service replaced cryoprecipitate with fibrinogen. OBJECTIVES: To examine the impact of this externally imposed change on blood product use and clinical outcomes in MOH. METHODS: Women with MOH requiring fibrinogen between 1 January 2009 and 30 June 2011 were identified from an MOH database. Aetiology of MOH, medical treatments, blood product use and clinical outcomes were compared between the cryoprecipitate and fibrinogen groups. RESULTS: Of 21 614 deliveries, 77 cases of MOH were identified. Of the 77 cases, 34 (44%) received cryoprecipitate (n = 14) or fibrinogen concentrate (n = 20). The mean (± SEM) dose utilised was 2.21 ± 0.35 pools of cryoprecipitate and 4 ± 0.8 g of fibrinogen. There was a stronger correlation between the increase in fibrinogen level and dose of fibrinogen (Pearson co-efficient 0.5; P = 0.03) than dose of cryoprecipitate (Pearson co-efficient 0.32; P = 0.3). Mean (± SEM) estimated blood loss (EBL), red cell concentrate (RCC) and Octaplas transfused were greater (but not significantly) in the cryoprecipitate group compared with the fibrinogen group; EBL = 5.2 ± 1.1 vs 3.3 ± 0.5 L (P = 0.1); RCC = 7.2 ± 1.2 vs 5.9 ± 1.0 U (P = 0.4); Octaplas = 4.1 ± 0.7 vs 3.2 ± 0.7 U (P = 0.36), respectively. Haemostasis was secured, and there were no adverse reactions or thrombotic complications. CONCLUSION: Purified virally inactivated fibrinogen concentrate is as efficacious as cryoprecipitate in correcting hypofibrinogenaemia in MOH.

Increasing trends in atonic postpartum haemorrhage in Ireland: an 11-year population-based cohort study.

OBJECTIVE: To derive nationally representative incidence rates of postpartum haemorrhage (PPH), and to investigate trends associated with method of delivery, blood transfusion and morbidly adherent placenta (accreta, percreta and increta). DESIGN: Population-based retrospective cohort study. SETTING: Republic of Ireland. POPULATION: Childbirth hospitalisations during the period 1999-2009. METHODS: International Classification of Diseases (ICD)-9-CM and ICD-10-AM diagnostic codes from hospital discharge records were used to identify cases of PPH. Significant temporal trends in PPH incidence were determined using Cochrane-Armitage tests for trend. Log-binomial regression was conducted to assess annual changes in the risk of PPH diagnosis, with adjustment for potential confounding factors. MAIN OUTCOME MEASURES: PPH, uterine atony, blood transfusion and morbidly adherent placenta. RESULTS: A total of 649,019 childbirth hospitalisations were recorded; 2.6% (n = 16,909) included a diagnosis of PPH. The overall PPH rate increased from 1.5% in 1999 to 4.1% in 2009; atonic PPH rose from 1.0% in 1999 to 3.4% in 2009. Significant increasing trends in atonic PPH rates were observed across vaginal, instrumental, and emergency and elective caesarean deliveries (P < 0.001). The rate of atonic PPH co-diagnosed with blood transfusion also significantly increased (P < 0.001). Relative to 1999, the risk of atonic PPH in 2009 was three-fold increased (adjusted RR 3.03; 95% CI 2.76-3.34). Women diagnosed with a morbidly adherent placenta had a markedly higher risk of total PPH (unadjusted RR 13.14; 95% CI 11.43-15.11). CONCLUSIONS: Increasing rates of atonic PPH highlight the pressing need for research and for clinical audit focusing on aetiological factors, preventative measures and quality of care, to guide current clinical practice.

Free fetal DNA is not increased before 20 weeks in intrauterine growth restriction or pre-eclampsia.

BACKGROUND: The aim of this study was to quantify maternal plasma fetal DNA and total DNA in early pregnancy in intrauterine growth restriction (IUGR) or pre-eclampsia (PET). METHODS: A nested case control study was carried out in a University Teaching Hospital. Plasma samples were obtained from 1993 women before 20 weeks of gestation. Pregnancies complicated by IUGR or PET were identified and compared to controls. DNA was extracted and real-time quantitative PCR applied for the SRY and beta-actin genes. IUGR or PET groups were compared to controls using the chi(2) and Wilcoxon rank sum tests. RESULTS: SRY was detected in 86% of IUGR (31/36), 94% of PET (15/16) and 78% of controls (56/72). The median SRY was similar in women with IUGR (28 GE/mL) or PET (30.5 GE/mL) and controls (27.5 GE/mL). beta-actin was increased in the IUGR group (3975 GE/mL) compared to controls (1835 GE/mL) (p = 0.045). Cigarette consumption was greater in the IUGR group compared to controls (p = 0.004). CONCLUSIONS: Fetal DNA quantitation in maternal plasma before 20 weeks is not a useful predictor of IUGR or PET. beta-actin levels were elevated before 20 weeks in women with IUGR and may be a marker of maternal susceptibility to this condition.

Obstetric and neonatal outcome of babies weighing more than 4.5 kg: an analysis by parity.

OBJECTIVES: To analyse by parity the obstetric and neonatal outcome of babies delivered weighing more than 4.5 kg. METHODS: All deliveries resulting in a baby weighing more than 4.5 kg, in the 5 years from 1991 to 1995, were identified using a computerised database. The following variables confined to singleton, cephalic pregnancies were recorded: mode of delivery, duration of labour, incidence of shoulder dystocia and admission to the neonatal centre. Outcome measures in primigravidae and multigravidae were compared using the Epi Info package (WHO, Version 6.0b January 1997). RESULTS: There were 32,834 deliveries over the study period and 828 (2.5%) weighed more than 4.5 kg. Birthweight more than 4.5 kg occurred in 1.6% (n=198) of primigravidae and 3.1% (n=630) of multigravidae (P<0.05). Primigravidae had a higher risk of prolonged labour (27.7% vs. 4.9%), operative vaginal delivery (32% vs.9%) and emergency caesarean section (24.2% vs. 5.7%) compared to multigravidae. When delivering a macrosomic baby, primigravidae had a higher incidence of prolonged labour (27% vs. 7.9%), operative vaginal delivery (32% vs.25%) and emergency caesarean section (24.2% vs. 5.7%) compared to normal weight babies. The incidence of shoulder dystocia and elective caesarean section were similar in both primigravidae and multigravidae. CONCLUSIONS: Macrosomic infants have an increased incidence of prolonged labour, operative vaginal delivery and emergency caesarean section compared with normal weight babies and these complications are more pronounced in primigravidae compared to multigravidae. Shoulder dystocia occurs with equal frequency in primigravidae and multigravidae. The poor antenatal predictability of macrosomia, the high rate of vaginal delivery and the low incidence of shoulder dystocia would not support the use of elective caesarean section for delivery of the macrosomic infant either in primigravidae or multigravidae.

Antenatal detection of abnormal liver function tests - a marker for poor perinatal outcome.

The purpose of this study was to examine (a) the incidence of liver disease diagnosed in our antenatal population, (b) the diagnostic value of initial symptoms and liver function tests (LFTs), (c) the adequacy of investigation and management of the liver disorder and (d) the obstetric and neonatal outcome in this group of patients. Women with abnormal LFTs that delivered at our hospital over a 2-year period were identified from computerised hospital records and data was obtained from chart review. Forty-six out of a total of 13 181 (0.35%) women had liver disease diagnosed in pregnancy: Diagnoses included intrahepatic cholestasis of pregnancy (13), pre-eclampsia and the HELLP syndrome (eight), acute fatty liver of pregnancy (three), hyperemesis gravidarum (one), hepatitis C (13), B (four) and hepatitis A (one), cholelithiasis (two) and hepatitis of unknown aetiology (one). Symptoms at presentation were more predictive of the final diagnosis than the initial LFT profile. Investigation of the liver disorder was incomplete in 50% of cases. One mother required intensive care for 6 weeks postpartum and three others had significant postpartum haemorrhage. There was one neonatal death and 24 neonates were admitted to the special care baby unit. Eighteen women attended for their postnatal check up at 6 weeks. Eight of these women were referred to a hepatologist. Detection of liver disease in pregnancy identifies a group at risk of poor neonatal and maternal outcome. Structured guidelines should be implemented in obstetric units to facilitate appropriate investigation, treatment and referral patterns for these women.