Evolution of foot-and-mouth disease virus serotype A capsid coding (P1) region on a timescale of three decades in an endemic context

Three decades-long (1977–2013) evolutionary trend of the capsid coding (P1) region of foot-and-mouth disease virus (FMDV) serotype A isolated in India was analysed. The exclusive presence of genotype 18 since 2001 and the dominance of the VP3⁵⁹-deletion group of genotype 18 was evident in the recent years. Clade 18c was found to be currently the only active one among the three clades (18a, 18b and 18c) identified in the deletion group. The rate of evolution of the Indian isolates at the capsid region was found to be 4.96 × 10^− 3 substitutions/site/year. The timescale analysis predicted the most recent common ancestor to have existed during 1962 for Indian FMDV serotype A and around 1998 for the deletion group. The evolutionary pattern of serotype A in India appears to be homogeneous as no spatial or temporal structure was observed. Bayesian skyline plots indicate a sharp decline in the effective number of infections after 2008, which might be a result of mass vaccination or inherent loss of virus fitness. Analyses of variability at 38 known antigenically critical positions in a countrywide longitudinal data set suggested that the substitutions neither followed any specific trend nor remained fixed for a long period since frequent reversions and convergence was noticed. A maximum of 6 different amino acid residues was seen in the gene pool at any antigenically critical site over the decades, suggesting a limited combination of residues being responsible for the observed antigenic variation. Evidence of positive selection at some of the antigenically critical residues and the structurally proximal positions suggest a possible role of pre-existing immunity in the host population in driving evolution. The VP1 C-terminus neither revealed variability nor positive selection, suggesting the possibility that this stretch does not contribute to the antigenic variation and adaptation under immune selection.     

Genetic analysis of human rhinovirus species A to C detected in patients with acute respiratory infection in Kumamoto prefecture,Japan 2011–2012

We performed detailed genetic analysis of the VP4/VP2 coding region in human rhinovirus species A to C (HRV-ABC) strains detected in patients with a variety of acute respiratory infections in Kumamoto, Japan in the period 2011–12. The phylogenetic tree and evolutionary timescale were obtained by the Bayesian Markov chain Monte Carlo method. Phylogenetic analyses showed that the present HRV-A, -B, and -C strains belonged to 25, 4, and 18 genotypes, respectively. Some new genotypes were confirmed as prevalent strains of HRV-C. An ancestor of the present HRV-ABCs could be dated back to about 20,000 years ago. The present HRV-A and -C strains have wide genetic divergence (pairwise distance >0.2) with rapid evolutionary rates (around 7 × 10⁻⁴ to 4 × 10⁻³ substitutions/site/year). Over 100 sites were found to be under negative selection, while no positively selected sites were found in the analyzed region. No evidence of recombination events was found in this region of the present strains. Our results indicate that the present HRV strains have rapidly evolved and subsequently diverged over a long period into multiple genotypes.     

Reconstructing the origin and transmission dynamics of the 1967–68 foot-and-mouth disease epidemic in the United Kingdom

A large epidemic of foot-and-mouth disease (FMD) occurred in the United Kingdom (UK) over a seven month period in Northwest England from late 1967 to the summer of 1968. This was preceded by a number of smaller FMD outbreaks in the country, two in 1967, in Hampshire and Warwickshire and one in Northumberland during 1966. The causative agent of all four events was identified as FMD virus (FMDV) serotype O and the source of the large epidemic was attributed to infected bone marrow in lamb products imported from Argentina. However, the diagnostic tools available at the time were unable to entirely rule out connections with the earlier UK FMD outbreaks, as well as other potential sources from Europe. The aim of this study was to apply molecular sequencing to investigate the likely source of this epidemic using VP1 region and full genome (FG) sequences determined directly from clinical epithelium samples (n = 13) or cell culture isolates (n = 6), from this and contemporary outbreaks in the UK, Europe and South America. Analysis of the VP1 sequences provided evidence for at least three separate incursions of FMDV into the UK including one independent introduction that was responsible for the main 1967/68 epidemic. Analysis of FG sequences from the main 1967/68 outbreak (n = 10) revealed nucleotide substitutions at 94 genomic sites providing evidence for the linear accumulation of nucleotide substitutions (rate = 2.42 × 10⁻⁵ nt substitutions/site/day). However, there were five samples where this linear relationship was absent, indicating evolutional dormancy of the virus, presumably outside a host. These results help define the evolutionary dynamics of FMDV during an epidemic and contribute to the knowledge and understanding from which to base future outbreak control strategies.     

Emergence of a novel lineage genetically divergent from the predominant Ind2001 lineage of serotype O foot-and-mouth disease virus in India

In India, emergence of Ind2001 lineage of foot-and-mouth disease virus (FMDV) serotype O was recorded in the year 2001. After causing sporadic incidences, the Ind2001 lineage that re-surged in 2008 out-competed PanAsia from the field during 2009 and continued its dominance during 2010 and 2011 as well. The lineage has diversified in due course of time, leading to two sub-lineages (Ind2001a and Ind2001b). The sub-lineage Ind2001a include isolates collected during 2001–2002 and sub-lineage Ind2001b is constituted largely by isolates collected during 2008–2012. The nucleotide substitution rate of sub-lineage Ind2001b was estimated at 6.58 × 10⁻³ substitutions/site/year. The most stable PanAsia lineage is restricted only to few outbreaks. During 2011, emergence of a new genetic group with >9% nucleotide divergence from rest of the lineages circulating in the country was detected and named as lineage Ind2011. Two specific amino acid substitutions at positions VP1–36 (F) and VP2–133 (T) were observed in the Ind2011 lineage. The new lineage at present is restricted only to southern states of the country. It is uncertain whether the emergence was triggered by immune pressure or due to a bottleneck in transmission or selected for higher fitness value. Six sites (4, 68, 83, 135, 138 and 209) in VP1 protein were identified to undergo episodic diversifying selection in serotype O field isolates. Both emerging and re-emerging lineages had appropriate antigenic match with currently used vaccine strain, INDR2/1975. Irrespective of genetic variability, the field isolates showed remarkable conservation at antigenically critical residues that might contribute to the observed antigenic stability. With the emergence of a new genetic group after a span of 10 years, the overall epidemiological scenario in the region is expected to change in the coming years.     

Tracking the molecular epidemiology of Brazilian Infectious bursal disease virus (IBDV) isolates

Infectious bursal disease is a highly contagious disease of young chickens caused by Infectious bursal disease virus (IBDV). Genome segment A encodes the capsid protein (VP2), while segment B encodes the RNA-dependent RNA polymerase (VP1). In the present study, we trace the molecular epidemiology of IBDV in Brazil by analyzing 29 isolates collected in the major regions of poultry production. To genetically characterize the isolates, phylogenetic and population dynamic analyses were conducted using 68 VP1 (2634 nt) and 102 VP2 (1356 nt) coding sequences from IBDV isolates from different regions of the world. Furthermore, the evolution of IBDV was analyzed by characterizing the selective forces that operated during the diversification of viral isolates. We show that IBDV isolates were introduced into Brazil mainly from the Netherlands and the USA. These introductions were associated with all Brazilian poultry production regions analyzed in this work. In addition, we show that the evolution of IBDV has been shaped by a combination of very low recombination rates and relatively high rates of nucleotide substitution (2.988 × 10⁻⁴ for VP1 and 3.2937 × 10⁻⁴ for VP2), which themselves are a function of purifying selection operating on VP1 and VP2. Furthermore, our extended Bayesian skyline plot suggests that the increase in the effective population size of isolates of IBDV is consistent with its epidemiological history, with a large increase during the emergence of acute outbreaks of IBD in the 1980s.     

Molecular evolution of attachment glycoprotein (G) gene in human respiratory syncytial virus detected in Japan 2008–2011

We investigated the evolution of the C-terminal 3rd hypervariable region of G gene in the prevalent human respiratory syncytial virus (RSV) subgroups A (RSV-A) and B (RSV-B) in Japan in 2008–2011. Phylogenetic analysis and the evolutionary timescale was obtained by the Bayesian Markov Chain Monte Carlo method. All 38 RSV-A strains detected were classified into genotype NA1 and the 17 RSV-B strains detected belonged to genotypes BA and GB2. NA1 subdivided around 1998 in the present phylogenetic tree. Genotype BA subdivided around 1994. The evolutionary rates for RSV-A and RSV-B were estimated at 3.63 × 10⁻³ and 4.56 × 10⁻³ substitutions/site/year, respectively. The mean evolutionary rate of RSV-B was significantly faster than that of RSV-A during all seasons. The pairwise distance was relatively short (less than 0.06). In addition, some unique sites under positive selection were found. The results suggested that this region of the RSV strains rapidly evolved with some unique amino acid substitutions due to positive pressure.     

Molecular evolution of fever,thrombocytopenia and leukocytopenia virus (FTLSV) based on whole-genome sequences

FTLSV is a novel bunyavirus that was discovered in 2007 in the Henan province of China and has reported case fatality rates of up to 30%. Despite the high case fatality rate, knowledge of the evolution and molecular epidemiology of FTLSV is limited. In this study, detailed phylogenetic analyses were performed on whole-genome sequences to examine the virus's evolutionary rates, estimate dates of common ancestry, and determine the population dynamics and selection pressure for FTLSV. The evolutionary rates of FTLSV were estimated to be 2.28 × 10^− 4, 2.42 × 10^− 4 and 1.19 × 10^− 4 nucleotide substitutions/site/year for the S, M and L segments, respectively. The most recent ancestor of the viruses existed approximately 182–294 years ago. Evidence of RNA segment reassortment was found in FTLSV. A Bayesian skyline plot showed that after a period of genetic stability following high variability, the FTLSV population appeared to have contracted it. Selection pressures were estimated and revealed an abundance of negatively selected sites and sparse positively selected sites. These data will be valuable in understanding the evolution and molecular epidemiology of FTLSV, eventually helping to determine mechanisms of emergence and pathogenicity and the level of the virus's threat to public health.     

Molecular evolution of respiratory syncytial virus subgroup A genotype NA1 and ON1 attachment glycoprotein (G) gene in central Vietnam

We performed molecular evolutionary analyses of the G gene C-terminal 3rd hypervariable region of RSV-A genotypes NA1 and ON1 strains from the paediatric acute respiratory infection patients in central Vietnam during the 2010–2012 study period. Time-scaled phylogenetic analyses were performed using Bayesian Markov Chain Monte Carlo (MCMC) method, and pairwise distances (p-distances) were calculated. Bayesian Skyline Plot (BSP) was constructed to analyze the time-trend relative genetic diversity of central Vietnam RSV-A strains. We also estimated the N-glycosylation sites within G gene hypervariable region. Amino acid substitutions under positive and negative selection pressure were examined using Conservative Single Likelihood Ancestor Counting (SLAC), Fixed Effects Likelihood (FEL), Internal Fixed Effects Likelihood (IFEL) and Mixed Effects Model for Episodic Diversifying Selection (MEME) models. The majority of central Vietnam ON1 strains detected in 2012 were classified into lineage 1 with few positively selected substitutions. As for the Vietnamese NA1 strains, four lineages were circulating during the study period with a few positive selection sites. Shifting patterns of the predominantly circulating NA1 lineage were observed in each year during the investigation period. Median p-distance of central Vietnam NA1 strains was wider (p-distance = 0.028) than that of ON1 (p-distance = 0.012). The molecular evolutionary rate of central Vietnam ON1 strains was estimated to be 2.55 × 10^− 2 (substitutions/site/year) and was faster than NA1 (7.12 × 10^− 3 (substitutions/site/year)). Interestingly, the evolutionary rates of both genotypes ON1 and NA1 strains from central Vietnam were faster than the global strains respectively. Furthermore, the shifts of N-glycosylation pattern within the G gene 3rd hypervariable region of Vietnamese NA1 strains were observed in each year. BSP analysis indicated the rapid growth of RSV-A effective population size in early 2012. These results suggested that the molecular evolution of RSV-A G gene detected in central Vietnam was fast with unique evolutionary dynamics.     

Molecular evolution of human respiratory syncytial virus attachment glycoprotein (G) gene of new genotype ON1 and ancestor NA1

We conducted a comprehensive genetic analysis of the C-terminal 3rd hypervariable region of the attachment glycoprotein (G) gene in human respiratory syncytial virus subgroup A (HRSV-A) genotype ON1 (93 strains) and ancestor NA1 (125 strains). Genotype ON1 contains a unique mutation of a 72 nucleotide tandem repeat insertion (corresponding to 24 amino acids) in the hypervariable region. The Bayesian Markov chain Monte Carlo (MCMC) method was used to conduct phylogenetic analysis and a time scale for evolution. We also calculated pairwise distances (p-distances) and estimated the selective pressure. Phylogenetic analysis showed that the analyzed ON1 and NA1 strains formed 4 lineages. A strain belonging to lineage 4 of ON1 showed wide genetic divergence (p-distance, 0.072), which suggests that it might be a candidate new genotype, namely ON2. The emergence of genotype NA1 was estimated to have occurred in 2000 (95% of highest probability density, HPD; 1997–2002) and that of genotype ON1 in 2005 (95% HPD; 2000–2010) based on the time-scaled phylogenetic tree. The evolutionary rate of genotype ON1 was higher than that of ancestral genotype NA1 (6.03 × 10⁻³ vs. 4.61 × 10⁻³ substitutions/site/year, p < 0.05). Some positive and many negative selection sites were found in both ON1 and NA1 strains. The results suggested that the new genotype ON1 is rapidly evolving with antigenic changes, leading to epidemics of HRSV infection in various countries.     

Rapid replacement of human respiratory syncytial virus A with the ON1 genotype having 72 nucleotide duplication in G gene

Human respiratory syncytial virus (HRSV) is the main cause of severe respiratory illness in young children and elderly people. We investigated the genetic characteristics of the circulating HRSV subgroup A (HRSV-A) to determine the distribution of genotype ON1, which has a 72-nucleotide duplication in attachment G gene. We obtained 456 HRSV-A positive samples between October 2008 and February 2013, which were subjected to sequence analysis. The first ON1 genotype was discovered in August 2011 and 273 samples were identified as ON1 up to February 2013. The prevalence of the ON1 genotype increased rapidly from 17.4% in 2011–2012 to 94.6% in 2012–2013. The mean evolutionary rate of G protein was calculated as 3.275 × 10⁻³ nucleotide substitution/site/year and several positively selected sites for amino acid substitutions were located in the predicted epitope region. This basic and important information may facilitate a better understanding of HRSV epidemiology and evolution.     

Acridine derivatives (PANHs,azaarenes) in raw,fried or grilled pork from different origins,and PANH formation during pork thermal processing

The aim of this work was to determine level of azaarenes (PANHs) in raw pork and to investigate their formation during meat frying or grilling, in particular to verify a suggestion that endogenous vitamin E might inhibit production of azaarenes. Azaarene concentration in raw pork samples from various origins ranged from 2.75 ng g⁻¹ to 3.69 ng g⁻¹ (2.75–2.93 ng g⁻¹ in meat of Polish Landrace pigs, 3.00–3.69 ng g^–1 in meat of hybrid Duroc × Polish Landrance pigs). PANH formation during frying of pork meat was not confirmed. On the other hand, PANHs were indeed formed during grilling; their levels ranged from 6.21 ng g⁻¹ to 8.08 ng g^–1. No inhibition influence of vitamin E on formation of PANHs on was found either in fried or grilled pork meat.     

Levels and sources of volatile organic compounds including carbonyls in indoor air of homes of Puertollano,the most industrialized city in central Iberian Peninsula. Estimation of health risk

Twenty nine organic air pollutants including carbonyl compounds, alkanes, aromatic hydrocarbons and terpenes were measured in the indoor environment of different houses together with the corresponding outdoor measurements in Puertollano, the most industrialized city in central Iberian Peninsula. VOCs were sampled during 8 weeks using Radiello^® passive samplers, and a questionnaire on potential VOCs sources was filled out by the occupants. The results show that formaldehyde and hexanal was the most abundant VOCs measured in indoor air, with a median concentration of 55.5 and 46.4 μg m⁻³, respectively followed by butanal (29.1 μg m⁻³), acetone (28.4 μg m⁻³) and acetaldehyde (21.4 μg m⁻³). After carbonyls, n-dodecane (13.1 μg m⁻³) and terpenes (α-pinene, 13.4 μg m⁻³ and limonene, 13.4 μg m⁻³) were the compounds with higher median concentrations. The indoor/outdoor (I/O) ratios demonstrated that sources in the indoor environment are prevailing for most of the investigated VOCs especially for limonene, α-pinene, hexanal, formaldehyde, pentanal, acetaldehyde, o-xylene, n-dodecane and acetone with I/O ratio >6. Multiple linear regressions were applied to investigate the indoor VOC determinants and Spearman correlation coefficients were used to establish common sources between VOCs. Finally, the lifetime cancer risk associated to formaldehyde, acetaldehyde and benzene exposure was estimated and they varied from 7.8 × 10⁻⁵ to 4.1 × 10⁻⁴ for formaldehyde, from 8.6 × 10⁻⁶ to 3.5 × 10⁻⁵ for acetaldehyde and from 2.0 × 10⁻⁶ to 1.5 × 10⁻⁵ for benzene. For formaldehyde, the attributed risk in most sampled homes was two orders of magnitude higher than the one (10⁻⁶) proposed as acceptable by risk management bodies.     

Molecular evolution of hepatitis B vaccine escape variants in China,during 2000–2016

Hepatitis B vaccine escape variants are the main threat to hepatitis B virus (HBV) infection in vaccination era worldwide. With 215 genotype B HBV and 313 genotype C HBV vaccine escape variants isolated from China during 2000–2016, we reported that genotype B HBV vaccine escape strains diverged in ∼1997 (95% HPD; 1987–2005), while genotype C HBV vaccine escape strains diverged in ∼1976 (95% HPD; 1955–2003). Additionally, the p-distance of genotype C HBV vaccine escape strains was 0.0291 ± 0.0169, which was significantly higher than that in the genotype B HBV (t = 131.02, p < 0.05). However, genotype B HBV vaccine escape strains evolved more rapidly than genotype C HBV (2.103 × 10⁻³ vs 1.083 × 10⁻³ substitutions/site/year). Bayesian skyline plot analysis showed that the populations of genotype C HBV vaccine escape strains fluctuated more than those in genotype B HBV. Four sites (A5T/S, L21S, T/A126S and T/N131I/A) and 13 sites (N3S, T5A, G10Q/R/E, L21S, T47K/A/V, L98V/P, I/S126N/V/T, Q129H/R/L, T131P/I/N/A, G145A/R, L175S/F, L213I/S, V224A/G) were found to be under positive selection in genotype B and C HBV vaccine escape strains, respectively. More importantly, N3S, L21S, T47K, L98V, I/S126T and L213I mutations were detected in 1 (2.5%), 1 (2.5%), 1 (2.5%), 3 (7.5%), 1 (2.5%), 1 (2.5%) genotype C HBV infected Chinese younger with neonatal HBV vaccination, respectively. Therefore, our results should be valuable in further understanding the molecular evolution of HBV and providing new ideas for the elimination of HBV infection.     

A rapid and sensitive method for the determination of dibutyl phthalate in wine by flow-injection chemiluminescence analysis

A sensitive method for the determination of picogram level dibutyl phthalate (DBP) in wine by flow-injection chemiluminescence (FI–CL) analysis is presented for the first time, which was based on the quenching effect of DBP on the luminol–myoglobin (Mb) CL system. The decrement of CL intensity was linearly proportional to the logarithm of DBP concentration in the range of 0.1–100 pg mL⁻¹ with the detection limit of 0.03 pg mL⁻¹ (3σ). At a flow rate of 2.0 mL min⁻¹, a complete determination of DBP including sampling and washing could be accomplished in 0.5 min, giving the maximum sample throughput of 120 h⁻¹. The proposed method was successfully applied to the determination of DBP in wine, human serum and urine samples with the relative standard deviations (RSDs) of less than 3.0% (n = 5). The molecule docking results showed that DBP interacted with the amino acid residues near the heme moiety of Mb. The possible CL mechanism of luminol–Mb–DBP reaction should be that the binding of Mb with DBP forming a 1:1 complex (binding constant K = 1.55 × 10⁴ L mol⁻¹) led to the conformational change of Mb and resulted in the quenching of CL intensity.     

Application of non-supervised pattern recognition techniques to classify Cabernet Sauvignon wines from the Balkan region based on individual phenolic compounds

Phenolic compounds in sixteen Cabernet Sauvignon wines from different wine-growing sub-regions in the Balkan region were investigated using HPLC with DAD and fluorescence detector and spectroscopic analysis, as well as statistical PC/F and cluster analysis. The HPLC analysis of investigated red wines showed that the content of total hydroxybenzoic acids, detected at 280 nm, was the highest in wines from Tikveš wine-growing subregion, Macedonia (127–140 mg L⁻¹). Total hydroxycinnamic acids, detected at 320 nm, were the highest in wines from Župa wine-growing subregion, Serbia (43–45 mg L⁻¹). The concentration of total flavonoids (flavan-3-ols, flavonols, flavons and flavanon), detected at 280, 360 and 322/275 nm, respectively, was the highest in wine from Katarzyna Estate wine-growing subregion, Bulgaria (167 mg L⁻¹). Finally, the concentration of total anthocyanins, detected at 520 nm, was the highest in wine from Šumadija wine-growing subregion, Serbia (1463 mg L⁻¹). The results of PCA and cluster analysis together confirmed that the content of phenolic compounds in Cabernet Sauvignon wines depends on agro-climatic factors, oenological practice in different wineries and the growing season in the Balkan region that were investigated. The areas in the Balkan region in this study with similar agro-climatic characteristics showed shorter clustering distance, indicating similar phenol profiling in the red wines tested.     

Interferon-γ (IFNG) microsatellite repeat and single nucleotide polymorphism haplotypes of IFN-α receptor (IFNAR1) associated with enhanced malaria susceptibility in Indian populations

Pro-inflammatory cytokines IFNγ and IFNα function through their cellular receptors IFNγR1 and IFNαR1, respectively to mediate immune processes during malaria infection. A total of 21 SNPs, 2 ins/del polymorphisms and a microsatellite repeat, selected on the basis of their reported association with infectious diseases including malaria in world populations, were analysed for association with Plasmodium falciparum malaria susceptibility in a case-control study with adult patients and ethnically-matched controls drawn from a disease meso- to hyperendemic and a nonendemic region of India. Among the five IFNG SNPs tested, an intron 3 and a 3′UTR SNP associated with disease in the endemic region. In addition, large (CA)_n repeats of IFNG intron 1 associated with protection from severe malaria in the endemic region (severe vs. control, odds ratio = 0.21, 95% CI = 0.08–0.52, P = 1.3 × 10⁻⁴). The TA11CAG haplotype (rs2069705 T/C, rs2430561 A/T, rs3138557 (CA)_n, rs2069718 T/C, rs2069727 A/G, rs2069728 G/A) carrying a short CA₁₁ repeat also exhibited very strong association with severe malaria, particularly in the endemic region (severe vs. control, OR = 14.56, 95% CI = 3.39–85.81, P = 3 × 10⁻⁵). One SNP each from the IFNA8 and IFNA17 of IFNA gene cluster had a protective effect in the non-endemic region but not in the endemic region. A promoter and an intron 2 SNP of IFNAR1 were risk factors for disease and the IFNAR1 haplotype GCCAGG (rs2843710 C/G, rs2850015 C/T, +6993 C/T, rs2243594 A/G, rs1012335 G/C, rs2257167 G/C) carrying both the risk alleles strikingly associated with disease manifestation in the endemic region (severe vs. control, OR = 27.14, 95% CI = 3.12–1254, P = 2 × 10⁻⁵; non-severe vs. control, OR = 61.87, 95% CI = 10.08–2521, P = 1 × 10⁻⁸). The data indicates dissimilar contribution of cytokine and cytokine receptor variants to disease in populations residing in areas of differential malaria endemicity.     

Evolution and dispersal of St. Louis encephalitis virus in the Americas

Using a Bayesian coalescent approach on a dataset of 73 envelope gene sequences we estimated substitution rates and dates of divergence for St. Louis encephalitis virus (SLEV) in the Americas. We found significant rate heterogeneity among lineages, such that “relaxed” molecular clock models were much better supported than a strict molecular clock. The mean substitution rate estimated for all SLEV was 4.1 × 10^?4 substitutions/site/year (95% HPD 2.5–5.7)—higher than previous estimates that relied on the less well-suited strict clock. Mean substitution rates for individual lineages varied from 3.7 × 10^?4 to 7.2 × 10^?4 substitutions/site/year. For the first time we also assessed the magnitude and direction of viral gene flow within the Americas. The overall direction of gene flow during the period represented by the phylogeny is from South to North, and the region between 15°N and 30°N latitude appears to be the major source of virus for the rest of North America, which is consistent with migratory birds returning to their northern breeding grounds having acquired infection while wintering in the region of the Gulf of Mexico.     

Cost-effectiveness of rabies post-exposure prophylaxis in the context of very low rabies risk: A decision-tree model based on the experience of France

IntroductionBenefit-risk of different anti-rabies post-exposure prophylaxis (PEP) strategies after scratches or bites from dogs with unknown rabies status is unknown in very low rabies risk settings.Design and settingA cost-effectiveness analysis in metropolitan France using a decision-tree model and input data from 2001 to 2011.PopulationA cohort of 2807 patients, based on the mean annual number of patients exposed to category CII (minor scratches) or CIII (transdermal bite) dog attacks in metropolitan France between 2001 and 2011.InterventionsFive PEP strategies: (A) no PEP for CII and CIII; (B) vaccine only for CIII; (C) vaccine for CII and CIII; (D) vaccine+ rabies immunoglobulin (RIG) only for CIII; and (E) vaccine for CII and vaccine+ RIG for CIII.Main outcomes measuresThe number of deaths related to rabies and to traffic accidents on the way to anti-rabies centers (ARC), effectiveness in terms of years of life gained by reducing rabies cases and avoiding traffic accidents, costs, and incremental cost-effectiveness ratios (ICER) associated with each strategy.ResultsStrategy E led to the fewest rabies cases (3.6 × 10⁻⁸) and the highest costs (€1,606,000) but also to 1.7 × 10⁻³ lethal traffic accidents. Strategy A was associated with the most rabies cases (4.8 × 10⁻⁶), but the risk of traffic accidents and costs were null; therefore, strategy A was the most effective and the least costly. The sensitivity analysis showed that, when the probability that a given dog is rabid a given day (P_A) was >1.4 × 10⁻⁶, strategy D was more effective than strategy A; strategy B became cost-effective (i.e. ICER vs strategy A <3 × French Gross Domestic Product per capita) when P_A was > 1.4 × 10⁻⁴.ConclusionsIn the metropolitan France's very low rabies prevalence context, PEP with rabies vaccine, administered alone or with RIG, is associated with significant and unnecessary costs and unfavourable benefit-risk ratios regardless to exposure category.