Predicting time to asystole following withdrawal of life-sustaining treatment: a systematic review.

The planned withdrawal of life-sustaining treatment is a common practice in the intensive care unit for patients where ongoing organ support is recognised to be futile. Predicting the time to asystole following withdrawal of life-sustaining treatment is crucial for setting expectations, resource utilisation and identifying patients suitable for organ donation after circulatory death. This systematic review evaluates the literature for variables associated with, and predictive models for, time to asystole in patients managed on intensive care units. We conducted a comprehensive structured search of the MEDLINE and Embase databases. Studies evaluating patients managed on adult intensive care units undergoing withdrawal of life-sustaining treatment with recorded time to asystole were included. Data extraction and PROBAST quality assessment were performed and a narrative summary of the literature was provided. Twenty-three studies (7387 patients) met the inclusion criteria. Variables associated with imminent asystole (<60 min) included: deteriorating oxygenation; absence of corneal reflexes; absence of a cough reflex; blood pressure; use of vasopressors; and use of comfort medications. We identified a total of 20 unique predictive models using a wide range of variables and techniques. Many of these models also underwent secondary validation in further studies or were adapted to develop new models. This review identifies variables associated with time to asystole following withdrawal of life-sustaining treatment and summarises existing predictive models. Although several predictive models have been developed, their generalisability and performance varied. Further research and validation are needed to improve the accuracy and widespread adoption of predictive models for patients managed in intensive care units who may be eligible to donate organs following their diagnosis of death by circulatory criteria.

Preparation of vaccines against H5N1 influenza.

In response to the pandemic warning provided by the highly pathogenic H5N1 influenza virus infections in Hong Kong, there were world-wide attempts to develop vaccines. Three strategies were followed and although each was associated with some success, there were also some problems. Pre-clinical vaccine efficacy results are presented from one such strategy, that of using an apathogenic H5N3 avian strain for vaccine production.

Plasmid DNA encoding influenza virus haemagglutinin induces Th1 cells and protection against respiratory infection despite its limited ability to generate antibody responses.

Direct intramuscular injection of plasmid DNA can generate immune responses against encoded antigens. However, the relative ability of DNA vaccines to induce cellular and humoral immunity after a single or booster immunization and the persistence of this response have not been fully elucidated. In this study, induction and maintenance of antibody and T cell subtypes with different doses of naked DNA encoding the haemagglutinin (HA) gene of influenza virus were examined and compared to the immune responses and protection induced by respiratory tract infection and immunization with a killed virus vaccine. Like natural infection, immunization with HA DNA induced potent Th1 responses. Spleen cells from mice immunized once with HA DNA in the dose range 10 ng to 100 microgram secreted significant levels of IFN-gamma, but low or undetectable IL-5, in response to influenza virus in vitro. Furthermore, CD4(+) HA-specific Th1 clones were generated from spleens of immunized mice. Although T cell responses waned 12 weeks after a single immunization, antigen-specific Th1 cells persisted in the spleen for at least 6 months after two booster immunizations. In contrast, influenza virus-specific ELISA IgG titres were low after a single immunization and required two booster immunizations to reach significant levels. Furthermore, haemagglutination inhibition (HI) antibodies were weak or undetectable after two immunizations. Nevertheless, two doses of HA DNA conferred almost complete protection against respiratory challenge with live virus. Thus, despite the limited ability to induce antibodies, DNA vaccines confer protective immunity against influenza virus infection, which appears to be mediated by Th1 cells.

Correlates of informant-rated cognitive decline after stroke.

Cognitive impairment is common after stroke, but measurement is problematic. Six tests of mental ability, unaffected by loss of limb function, were administered to 49 subjects of mean age 74.2 years at a median of 4.3 years (range 0.1-16.8) after stroke together with a depression score and the IQCODE, an informant-rated scale of estimated cognitive decline. Over 90% of stroke patients were able to complete most tests. IQCODE correlated significantly with the HADS depression score (r = 0.35, p = 0.040), the 2-year Barthel score (r = -0.60, p = 0.001) and with a general cognitive factor extracted from the mental ability test scores (r = -0.42, p = 0.016). We conclude that informant-rated methods offer a promising approach to measuring cognitive decline after stroke.

Respiratory syncytial virus infection of gene gun vaccinated mice induces Th2-driven pulmonary eosinophilia even in the absence of sensitisation to the fusion (F) or attachment (G) protein.

Complete protection against respiratory syncytial virus (RSV) infection was induced in mice vaccinated on two occasions with 2.5 microg of DNA, encoding the fusion (F) protein of RSV, precipitated onto gold microbeads. In contrast, immunisation with DNA encoding the attachment (G) protein of RSV resulted in a significant reduction in viral load following infection, but did not afford complete protection. Gene gun delivery of DNA-F elicited a T helper-2 (Th2) biased immune response that could not be modulated by the co-delivery of plasmids encoding IL-2, IL-12 or IFNgamma. Similarly gene gun delivery of DNA-G primed a Th2 response. Thus, all gene gun vaccinated mice produced a predominant Th2 biased pulmonary immune response characterised by the production of IL-4 and IL-5 with little IFNgamma following RSV challenge. Analysis of bronchoalveolar lavage (BAL) cells, 5 days post challenge, indicated that there was only a two-fold increase in the number of inflammatory cells in vaccinated compared with control animals. Despite the strong Th2 cytokine bias of lung lymphocytes and the predominant recruitment of CD4(+) T cells, following challenge, there was not a marked pulmonary eosinophilic response (range from 2 to 7% of BAL). In contrast, the BAL from mice vaccinated with control plasmid contained significantly more eosinophils than any other group.

Sun awareness in school teachers.

School children are important targets for sun awareness education, but the knowledge, attitudes and behaviour of school teachers with respect to sun awareness are poorly understood. A questionnaire-based survey of 76 school teachers was undertaken. Twenty-four per cent of teachers indicated previous experience in teaching sun awareness, but 93% of teachers had no classroom resources for teaching sun awareness. Sun awareness was perceived by respondents as unimportant relative to other health education issues. There was considerable scope for improvement in attitudes and behaviour of teachers with respect to sun awareness. Teachers need more information about environmental factors which affect strength of sunlight, individual risk factors for sunburn and sunscreen strength. Intervention strategies involving school teachers need simple messages, emphasizing the importance of shade, clothing and hats in sun avoidance.

Direct sequencing of the HA gene of clinical equine H3N8 influenza virus and comparison with laboratory derived viruses.

Equine influenza viruses propagated in the laboratory in alternate hosts such as embryonated hens' eggs or mammalian cell culture have been analysed by HA sequencing and antigenically and their sequence compared to the original virus present in clinical material. In contrast to clinically derived human influenza virus which generally grows in MDCK cells without change, the data for equine influenza virus were less clear in that variants of equine virus were derived in both eggs and cells. The study indicated that the current use of eggs for equine influenza virus surveillance and vaccine production is entirely appropriate, but that care should be exercised when equine influenza vaccines are produced in eggs or on mammalian cell cultures.

Assessing the significance of reverse transcriptase activity in chick cell-derived vaccines.

A recent publication reported the detection of low levels of the enzyme reverse transcriptase (RTase) in live viral vaccines prepared in chick embryo cells. The enzyme was detected using an assay with greatly increased sensitivity compared to more conventional methods. The authors have confirmed the observation of RTase activity and demonstrate that the activity is not dependent on the production of viral vaccines in chick cells but is present ubiquitously in chick embryonic fluids. The authors have also been unable to transmit the RTase activity from chick cells to a wide variety of cells of human, monkey, rabbit and turkey origin, suggesting that the activity is not associated with an avian agent capable of infecting these cells. It is concluded that the data available present no cause for concern over the safety of vaccines derived in chick cells and current WHO requirements for such vaccines remain appropriate.

Mixed populations in influenza virus vaccine strains.

Human influenza viruses used for vaccine production have previously been adapted to grow in eggs. During egg adaptation, variants are selected and we have observed that more than one variant may derive in a single egg resulting in a mixed population. We have now investigated the extent of heterogeneity, due to host cell selection, of virus strains used for the manufacture of influenza vaccine for the 1991/1992 and 1992/1993 seasons. The A(H1N1) vaccine virus was homogeneous with respect to substitutions in the haemagglutinin deriving from egg adaptation. However, two A(H3N2) vaccine strains and the influenza B component, B/Yamagata/16/88, consisted of mixed populations, apparently due to their cultivation in eggs. The individual variants within B/Yamagata were isolated and found to be antigenically distinct. The ratios of these variants within different manufacturers' seed stocks varied to the extent that vaccine derived from them could be distinguished antigenically. Furthermore, derivation of high-growth reassortants from the A(H3N2) strains which involves passaging at limit dilution did not necessarily lead to a homogeneous virus population. The significance of these findings for the efficacy of vaccine is not known at present.

The role of amniotic passage in the egg-adaptation of human influenza virus is revealed by haemagglutinin sequence analyses.

Obtaining an isolate of a human influenza virus in the allantoic cavity of the embryonated hen's egg is more efficient if the clinical sample is initially passaged in the amniotic cavity. To investigate the extent to which the variants present after allantoic propagation are also selected by amniotic passage, clinical virus passaged once in the amnion has been subjected to extensive genetic and antigenic analyses. The data indicate that the natural virus can replicate unrestrictedly within the amnion. However, exposure of amniotic virus to the allantois during the incubation period, which will occur through the hole between the amniotic and allantoic cavities caused by the inoculating needle, allows for the possibility of an egg-adapted variant establishing replication within the allantois and returning to the amnion. These observations illustrate why prior passage in the amnion increases the probability of a variant successfully establishing itself during a subsequent allantoic passage.

Genic amplification of the entire coding region of the HEF RNA segment of influenza C virus.

In order to provide an easy and powerful analysis of influenza C viral HEF RNA segment of a recent strain, a combination of reverse transcription and the polymerase chain reaction was used. We amplified the entire coding region of the HEF gene of a laboratory strain of virus called C/Johannesburg/1/66, widely used for binding and esterase activity studies as well as that of a strain isolated in 1991 (C/Paris/145/91) from a patient suffering from severe flu syndrome. The sequences we amplified were about 2 kilobases long. In this work, we show that the forward 'universal primer' Uni1, which has been used for influenza A and B viruses cDNA syntheses can also be used for influenza C virus. The PCR primers were designed to contain restriction sites to make the PCR products ready to be used for further purposes. A restriction analysis of the PCR products combined with analyses of all the human influenza C virus HEF gene sequences published so far permitted the design of sets of oligonucleotides which can prime PCR on cDNA of unknown influenza C virus for cloning.

High growth reassortant influenza vaccine viruses: new approaches to their control.

When a new strain of an influenza virus is required to be incorporated into influenza vaccine, attempts are made to recombine such strains with laboratory adapted viruses, which will grow to high titre in order to improve the yield of the vaccine strain. It is important that such high growth reassortant vaccine strains are not contaminated with genes coding for the antigenic determinants of the high growth laboratory strain. We describe the characterization of two recent high growth reassortants and the application of the polymerase chain reaction to ensure their genetic identity and purity.

Sequence analysis of the haemagglutinin (HA) of influenza A (H1N1) viruses present in clinical material and comparison with the HA of laboratory-derived virus.

We used the polymerase chain reaction to amplify the HA1 coding region of influenza A (H1N1) viruses present in clinical material from recent cases of influenza in the U.K. Previously, we have demonstrated that isolation of human influenza viruses in embryonated hens' eggs selects variants which have amino acid substitutions in their haemagglutinin (HA) clustering around the receptor-binding site. Such egg-selected variants are often antigenically distinct from each other and from corresponding viruses isolated on mammalian cells. Since in general the virus used for vaccine production is an egg-adapted virus, it is important to determine the extent to which these variants are present in the natural virus which causes disease in man. To achieve this, amplified products from clinical material were cloned and many individual clones sequenced. Our results indicate that the HA of the naturally occurring virus is relatively homogeneous and represented by virus isolated in the laboratory on MDCK cells, whereas the variants isolated in eggs are present only at low levels in clinical material.

The hemagglutinin of influenza B virus present in clinical material is a single species identical to that of mammalian cell-grown virus.

When clinical specimens of influenza virus are adapted to grow in embryonated hens' eggs, variants are selected which have specific amino acid substitutions in the hemagglutinin (HA). In contrast, a single virus, distinct from any egg-adapted variant, is selected when virus is isolated on MDCK mammalian cell culture. We have utilized the polymerase chain reaction to determine the nature of the hemagglutinin of influenza B viruses present in clinical material prior to cultivation in the laboratory. Sequence analysis of individual clones of amplified DNA reveals that the HA of clinical virus is essentially homogeneous and identical to the virus derived on MDCK cells. The HA of egg-adapted virus was heterogeneous and nonidentical to that of the clinical material and of the MDCK-derived virus.