The changing profile of infant mortality from bacterial, viral and fungal infection over two decades.

AIM: Infection is an important cause of neonatal and infant mortality. We evaluated changes in infant deaths from infections from 1988 to 2008 in the North of England. METHODS: We interrogated a population-based survey and reviewed infant deaths from infection. Proportional contribution to deaths, pathogens identified and risk factors were analysed. RESULTS: Thirteen percentage of 4366 infant deaths from a population of 704 536 livebirths were infectious. The absolute numbers of infant deaths from infection fell over time but the proportion of deaths from infection increased (12.1%, 13.6% and 14.9%). Significantly preterm infants were increasingly represented in successive epochs (14%, 24% and 38%). Infant mortality rate (IMR) from meningococcus and Group B Streptococcus (GBS) fell in the latest epoch, but there was a corresponding increase from Escherichia coli and candida. DISCUSSION: This large study shows that infections have become proportionately more important causes of death especially in very preterm infants. Recent significant reductions in death from meningococcus and GBS are likely to represent successful achievements of vaccination and antibiotic prophylactic policies. Increases in IMR from E. coli may relate to GBS prophylaxis and increases in candida to the increase from preterm populations. Further efforts to understand these changing patterns and develop additional prevention and treatment strategies and vaccines remain an urgent priority.

Viral infections: contributions to late fetal death, stillbirth, and infant death.

OBJECTIVE: To determine the role of viral infections in causing fetal and infant death. STUDY DESIGN: We assessed a well-validated population database of fetal (≥20 weeks gestation) and infant death for infective deaths and deaths from viruses over a 21-year period (1988-2008). We analyzed by specific viral cause, timing (late fetal loss [20-23 weeks], stillbirth [≥24 weeks], neonatal death [0-27 days], and post-neonatal infant death [28-364 days]) and across time. RESULTS: Of the 989 total infective deaths, 108 were attributable to viral causes (6.5% of late fetal losses, 14.5% of stillbirths, 6.5% of neonatal deaths, and 19.4% of postneonatal infant deaths). Global loss (combined fetal and infant losses per 100,000 registerable births) was 139.6 (95% CI, 130.9-148.3) for any infective cause and 15.2 (95% CI, 12.3-18.1) for viral infections. More than one-third (37%) of viral-attributed deaths were before live birth, from parvovirus (63%) or cytomegalovirus (33%). Parvovirus accounted for 26% (28 of 108) of all viral deaths. Cytomegalovirus was associated with a global loss rate of 3.1 (95% CI, 1.8-4.4) and an infant mortality rate of 1.3 (95% CI, 0.4-2.1) per 100,000 live births; 91% of cases were congenital infections. Herpes simplex virus caused death only after live births (infant mortality rate, 1.4; 95% CI, 0.5-2.3). No changes in rates were seen over time. CONCLUSION: We have identified a substantial contribution of viral infections to global fetal and infant losses. More than one-third of these losses occurred before live births. Considering our methodology, our estimates represent the minimum contribution of viral illness. Strategies to reduce this burden are needed.

The North of England Survey of Twin and Multiple Pregnancy.

The population-based Northern Survey of Twin and Multiple Pregnancy (NorSTAMP, formerly the Multiple Pregnancy Register) has collected data since 1998 on all multiple pregnancies in North of England (UK) from the earliest point of ascertainment in pregnancy. This paper updates recent developments to the NorSTAMP and presents some early mortality data from the first 10 years of data collection (1998-2007). Since 2005, mothers have been asked to give explicit consent for their identifiable data to be held by the survey, in line with changing guidance and legal frameworks for identifiable data. In 2009, regional standards of care for multiple pregnancies were developed, agreed, and disseminated. During 1998-2007, 4,865 twin maternities (pregnancies with at least one live birth or stillbirth) were registered, with an average twinning rate of 14.9 per 1,000 maternities. The overall stillbirth and neonatal mortality rates in twins were 18.0/1,000 births and 23.0/1,000 live births respectively. Stillbirth and neonatal mortality rates were significantly higher in monochorionic than dichorionic twins: 44.4 versus 12.2 per 1,000 births (relative risk [RR] 3.6, 95% Confidence Intervals [CI] 2.6-5.1), and 32.4 versus 21.4 per 1,000 live births (RR 1.5, 95% CI 1.04-2.2) respectively. There was no significant improvement during this period in either stillbirth or neonatal mortality rates in either chorionicity group. This population-based survey is an important source of data on multiple pregnancies, which allows monitoring of trends in multiple birth rates and pregnancy losses, providing essential information to support improvements in clinical care and for epidemiological research.

Use of interactive telephone technology for longitudinal data collection in a large trial.

We report here on the use of interactive telephone technology for collecting longitudinal data in a large randomized non-blinded parallel trial. Data were primarily collected via an automated interactive telephone system which enabled data to be downloaded by researchers periodically via a secure website. Alternative methods were used by some participants to provide data; here we analyze the demographic profiles of groups by preferred data provision, and consider the cost-effectiveness and efficiency of such a system. The automated telephone system was used to provide the majority of data obtained (75.7%), however the group preferring to use this system to provide the majority of their data was on the whole older, more likely to be married, university educated, higher income and white compared to participants preferring to submit their data via personal phone call or post. We conclude that interactive telephone technology provides a means by which large quantities of longitudinal data may be collected efficiently. Depending on the target population, however, considerable staff time may be required to manage attrition and consequent data loss, and alternative strategies should be considered to minimize this.

Randomised trial of sidecar crib use on breastfeeding duration (NECOT).

OBJECTIVE: To determine whether the use of sidecar cribs on the postnatal ward affects breastfeeding duration. DESIGN: A randomised non-blinded parallel trial comparing sidecar cribs with standalone cots. SETTING: Postnatal wards of the Royal Victoria Infirmary, Newcastle upon Tyne. PARTICIPANTS: 1204 pregnant women intending to breastfeed were recruited at 20 weeks' gestation and randomised at 34 weeks to use either a sidecar crib attached to their bed (n=601) or a standalone cot adjacent to their bed (n=603). MAIN OUTCOME MEASURES: Duration of any, and exclusive, breastfeeding up to 26 weeks obtained by telephone follow-up. RESULTS: 334 mothers were withdrawn or lost to follow-up from the trial; infant feeding data were therefore obtained for 870 mothers (433 intervention; 437 controls). Using an intention-to-treat Cox regression analysis, no significant difference was found between the two groups for duration of any breastfeeding (sidecar crib vs cot, hazard ratio (HR) 0.96, 95% CI 0.79 to 1.18), or exclusive breastfeeding (HR 0.99, 95% CI 0.85 to 1.16) adjusting for maternal age, education, previous breastfeeding and delivery type. Bed sharing was not significantly more common in mothers randomised to sidecar cribs (67% vs 64%, adjusted difference 2.8%, 95% CI -3.5% to 9.0%). There were no adverse events. CONCLUSION: The use of sidecar cribs for mothers and infants did not improve the duration of any or exclusive breastfeeding, or frequency of bed sharing at home.

Difficulties in selecting an appropriate neonatal thyroid stimulating hormone (TSH) screening threshold.

BACKGROUND: The UK Newborn Screening Programme Centre recommends that a blood spot thyroid stimulating hormone (TSH) cut-off of 10 mU/l is used to detect congenital hypothyroidism (CHT). As the value used varies from 5 to 10 mU/l, we examined the implications of altering this threshold. METHODS: Our regional blood spot TSH cut-off is 6 mU/l. Positive or suspected cases were defined as a TSH >6 mU/l throughout the study period (1 April 2005 to 1 March 2007). All term infants (>35 weeks) whose first TSH was 6-20 mU/l had a second TSH measured. The biochemical details of infants with a TSH between 6.1 and 10.0 mU/l and then >6 mU/l on second sampling were sent to paediatric endocrinologists to determine approaches to management. RESULTS: 148 of 65 446 infants (0.23%) had a first blood spot TSH >6.0 mU/l. 120 were term infants with 67 of these (0.1% of all infants tested) having a TSH between 6.1 and 10.0 mU/l and 53 a TSH >10.0 mU/l. Of the 67 term infants with a TSH between 6.1 and 10.0 mU/l on initial testing, four continued to have a TSH >6 mU/l. One with a TSH >10 mU/l and one infant with a TSH <10 mU/l on the second blood spot have been diagnosed with CHT. The survey of endocrinologists highlighted significant differences in practice. CONCLUSIONS: A reduced threshold of 6 mU/l will increase the number of false positive term infants by 126%, but abnormalities of thyroid function requiring treatment will be detected. We suspect that the additional expense involved in setting a lower threshold is justified.

The North of England Multiple Pregnancy Register: five-year results of data collection.

From 1998 the population-based North of England Multiple Pregnancy Register (MPR) has collected data on all multiple pregnancies in the region from the earliest point of ascertainment in the pregnancy. This article describes the development of the MPR and the findings of the first 5 years of data collection. Mothers now give explicit consent for their inclusion with named data, in accordance with section 60 of the Health and Social Care Act 2001. During 1998 to 2002, 2310 twin pregnancies were registered, with an increasing twinning rate of 13.6 to 16.6 per 1000 maternities. Chorionicity ascertainment in twin maternities with at least one stillbirth or live birth has improved from 81% in 1998 to 91% in 2002. Before 24 weeks of gestation, 8.4% (359/4620) of fetuses were lost either spontaneously or as a result of termination of pregnancy. The perinatal mortality rate was much higher in monochorionic than dichorionic twins, mainly due to differences in stillbirth rates (49.0 vs. 11.5 per 1000 maternities respectively, risk rate = 4.2; 95% confidence intervals 2.7-6.6). The gestational-age-specific neonatal mortality rates were similar in twins and singletons, except in the group of term births (> or weeks' gestation) when compared by conventional gestational age categories. For stillbirths, the rates were even lower than in singletons in gestational age categories of less than 32 weeks. The register is an important resource of data on multiple pregnancies, which allows monitoring of trends in multiple birth rates and pregnancy losses and provides a unique opportunity for etiological and long-term follow-up studies.