Novel goose parvovirus as a major risk factor for red skin and bristle feather syndrome in meat ducks: Epidemiological associations with duck circovirus and reovirus Authors: Zhongcheng Sun, Xue Wang, Hao Shi, Hao Zhu, Xue Liu, Tianyao Yang, Qingqing Song, Zhaoyang Li, Chunguo Liu Published: 2026 ABSTRACT In recent years, a disease characterized by dorsal skin reddening and feather development disorder, termed "red skin and bristle feather syndrome" (RSBFS), has emerged in meat duck farms in China, resulting in substantial economic losses. To clarify the major etiological risk factor, their epidemiological patterns, and molecular characteristics, we collected 300 clinical samples from symptomatic and asymptomatic ducks across eight major duck-producing provinces in China between February and July 2025. Multipathogen screening was conducted using quantitative real-time PCR, followed by phylogenetic analysis of the major pathogens. The results identified duck circovirus (DuCV), novel duck reovirus (NDRV), and waterfowl parvovirus (WPV) as the dominant pathogens, with detection rates of 66.44% (194/292), 32.33% (86/266), and 19.18% (56/292), respectively. Among the detected pathogens, monoinfection with WPV—predominantly represented by novel goose parvovirus (NGPV)—showed a strong positive association with RSBFS (OR = 9.85, 95% CI: 4.89–19.84). It was identified as a major etiological risk factor for disease occurrence. Although DuCV or NDRV monoinfection did not exhibit significant pathogenicity, their coinfection with WPV resulted in markedly enhanced interactive effects on disease severity. Epidemiological analysis revealed distinct spatiotemporal clustering, with Shandong province as the main epidemic focus and ducks aged 3 to 5 weeks as the most susceptible group. Genetic evolution analysis indicated that DuCV strains predominantly belonged to genotypes Ib and Ic, the NDRV S1 gene was highly conserved, and NGPV was the dominant subtype of WPV. To our knowledge, this study is the first to identify a strong statistical association of NGPV with RSBFS and to characterize its epidemiological associations with DuCV and NDRV. These findings provide critical epidemiological and molecular evidence for developing targeted vaccines and formulating integrated control strategies in commercial duck production. Introduction In recent years, a condition characterized by cutaneous erythema and abnormal feather development has been frequently observed in meat duck production systems in China, posing a significant challenge to production efficiency and carcass quality. Clinically, the syndrome presents with inflammatory reddening of the dorsal and caudal skin, sparse and disordered feathers shedding easily, and retained feather shafts difficult to remove during processing. This results in so-called "bristle ducks", which severely compromise carcass quality, increase processing costs, and cause substantial economic losses. This condition, collectively referred to in this study as "red skin and bristle feather syndrome" (RSBFS), is pathologically typified by the loss of dorsal and caudal feathers while the follicles and feather shafts remain embedded in the skin. In severe cases, lesions may spread across the body, preventing complete feather removal during slaughter. The exposed skin often becomes erythematous due to friction or inflammation, hence the descriptive name. Various terms have been used for this condition in different regions or reports, including "feather shedding syndrome," "red skin disease," or "bristle syndrome" (Yang et al., 2020; Ting et al., 2021; Soike et al., 2004). Its rising incidence and impact have made it a prominent constraint on the sustainable development and profitability of the intensive duck industry. The etiology of RSBFS is complex and generally considered multifactorial, involving viral infections, secondary bacterial invasions, nutritional or metabolic disturbances, and suboptimal husbandry practices. Among viruses, duck circovirus (DuCV), waterfowl parvovirus (WPV), and novel duck reovirus (NDRV) are the pathogens of major concern and key pathogens associated with this syndrome. Previous studies have linked DuCV infection to feather loss and growth retardation (Huang et al., 2013), while goose parvovirus-related viruses have also been associated with abnormal molting in ducks (Jin et al., 2019). Moreover, other viruses such as reticuloendotheliosis virus are known to induce feather abnormalities in chickens (Koyama et al., 1980), suggesting viral infection as a potential key trigger for feather development disorders. DuCV belongs to the genus Circovirus within the family Circoviridae. It is a non-enveloped, icosahedral virus with a single-stranded circular DNA genome containing six open reading frames, among which ORF C1 encodes the immunogenic capsid protein (Cap) (Hong et al., 2018). DuCV infection can damage immune organs, such as the thymus and bursa of Fabricius, leading to immunosuppression and increased susceptibility to secondary or concurrent infections (Jin et al., 2019; Koyama et al., 1980; Hong et al., 2018). Ducks of all ages are susceptible, with clinical signs including growth retardation, poor feather development, and feather shaft hemorrhage (Yang et al., 2020), often causing serious economic losses. Studies have confirmed a close association between DuCV infection and feather shedding syndrome (Hattermann et al., 2003; Wen et al., 2014; Liu et al., 2009; Kulprasertsri et al., 2024). WPV is a member of the genus Dependoparvovirus within the family Parvoviridae, possessing a linear single-stranded DNA genome. The genome contains inverted terminal repeats and two major ORFs encoding nonstructural proteins (NS1, NS2) and structural proteins (VP1–VP3) (Zádori et al., 1995; Shen et al., 2015). WPV includes classical goose parvovirus (GPV), classical Muscovy duck parvovirus (MDPV), novel goose parvovirus (NGPV), and novel Muscovy duck-goose parvovirus (MDGPV) (Li et al., 2026; Liu et al., 2024; Yu et al., 2016; Zhu et al., 2025). NGPV infection in ducks can induce short beak and dwarfism syndrome (SBDS), characterized by beak atrophy, tongue protrusion, and disordered feather shedding (Yu et al., 2016; Palya et al., 2009). NGPV may directly or indirectly disrupt feather development by interfering with follicular stem cell function and skeletal growth (Li et al., 2018). A parvovirus linked to feather abnormalities has also been isolated from Muscovy ducks in Japan (Takehara et al., 1994). NDRV belongs to the genus Orthoreovirus in the family Reoviridae. The virion is non-enveloped, spherical, and possesses a double-layered capsid. Its genome consists of 10 segments of double-stranded RNA, with the S1 gene encoding the σC protein, which mediates viral attachment and serves as a key antigen for neutralizing antibodies (Du et al., 2020; Zhang et al., 2006; Ma et al., 2012). NDRV infection can cause splenic necrosis and hemorrhage, induce immunosuppression and growth retardation, and exacerbate integumentary and feather lesions, particularly in ducklings (Liu et al., 2011; Wang et al., 2020). Notably, these viruses often co-circulate in mixed infections and can act synergistically with bacterial pathogens such as Salmonella spp. (SM) and Riemerella anatipestifer to aggravate clinical signs, complicating disease control. However, systematic knowledge regarding the predominant pathogens responsible for RSBFS—their epidemiological dynamics, genotypic features, and coinfection mechanisms—remains limited. There is a clear need for targeted prevention and control strategies based on large-scale epidemiological surveys and molecular characterization, which are essential for practical disease management in poultry production. Therefore, from February to July 2025, we collected 300 clinical samples from ducks exhibiting typical RSBFS signs across eight major duck-producing provinces in China, with the aim of clarifying the detection rates, spatiotemporal distribution, coinfection patterns, and molecular characteristics of the major pathogens. Using quantitative real-time PCR (qPCR), we systematically screened for DuCV, WPV, NDRV, and other relevant pathogens. Coupled with detailed epidemiological investigation, risk factor analysis, and phylogenetic analysis, this study aims to clarify the detection rates, spatiotemporal distribution, coinfection patterns, and molecular characteristics of the major pathogens. This study aimed to comprehensively reveal the etiological background of RSBFS and provide a scientific foundation for precise diagnosis and the development of integrated control measures for the duck industry. Materials and methods Sample origin and collection From February to July 2025, a total of 300 samples, comprising 78 tissue samples and 222 swab samples, were collected from commercial meat-type duck flocks across eight provinces in China: Shandong, Anhui, Hebei, Henan, Liaoning, Guangxi, Fujian, and Guangdong (Table 1). These regions represent major areas of intensive duck production in China. Samples were obtained from ducks at various growth stages (1–7 weeks old) exhibiting the typical clinical signs of RSBFS, including dorsal skin reddening, brittle or missing feathers, and retained feather shafts (Fig. 1). A subset of samples was also randomly collected from asymptomatic ducks on the same or neighboring farms as controls for risk factor analysis. The sample set included liver, spleen, and kidney tissues from euthanized birds, as well as oropharyngeal and cloacal swabs from live birds. All samples were placed in sterile containers, immediately transported on ice to the laboratory, and stored at –80°C until nucleic acid extraction.Table 1Distribution of symptomatic and asymptomatic cases and sample type across different regions.Table 1 dummy alt textRegionCasessample typeSymptomaticAsymptomaticTotalOrganSwabTotalShandong42931354194135Anhui3767917879Hebei29154463844Henan2512620626Liaoning6-6336Guangxi5-5325Fujian3-33-3Guangdong2-2112Total11518530078222300Note: “-” indicates no data available.Fig. 1Clinical and pathological manifestations of red skin and bristle feather syndrome (RSBFS) in meat ducks. (A) Clinical signs in an affected flock showing dorsal skin reddening, brittle or missing feathers, and retained feather shafts. (B) Postmortem observation of reddened skin on the dorsal region. (C) Close-up view of residual feather shafts embedded in the dermis.Fig 1 dummy alt text Table 1Distribution of symptomatic and asymptomatic cases and sample type across different regions.Table 1 dummy alt textRegionCasessample typeSymptomaticAsymptomaticTotalOrganSwabTotalShandong42931354194135Anhui3767917879Hebei29154463844Henan2512620626Liaoning6-6336Guangxi5-5325Fujian3-33-3Guangdong2-2112Total11518530078222300Note: “-” indicates no data available. Distribution of symptomatic and asymptomatic cases and sample type across different regions. Note: “-” indicates no data available. Clinical and pathological manifestations of red skin and bristle feather syndrome (RSBFS) in meat ducks. (A) Clinical signs in an affected flock showing dorsal skin reddening, brittle or missing feathers, and retained feather shafts. (B) Postmortem observation of reddened skin on the dorsal region. (C) Close-up view of residual feather shafts embedded in the dermis. Pathogen detection by qPCR Total nucleic acid (DNA and RNA) was extracted from homogenized tissue suspensions or swab eluates using an automated nucleic acid extraction system (Vazyme, Nanjing, China) according to the manufacturer's instructions. Multiplex qPCR assays were performed to simultaneously screen for a panel of pathogens relevant to waterfowl disease. Commercial qPCR detection kits (Xinda Gene, Qingdao, China) were used for DuCV, WPV, NDRV, duck Tembusu virus (DTMUV), duck astrovirus (DAstV) types Ⅰ and Ⅲ, Riemerella anatipestifer, and SM. All reactions were performed using qPCR master mixes (Accurate, Changsha, China) on a Gentier 96R real-time PCR system (Tianlong, Xi'an, China). Amplification was carried out following the respective kit protocols. A sample was considered positive for a specific pathogen if the cycle threshold (Ct) value was ≤ 36 with a characteristic amplification curve. Descriptive statistical analyses The epidemiological characteristics including spatiotemporal distribution, age-specific prevalence, and patterns of coinfection were analyzed using Excel 2016 (Microsoft Corporation, Redmond, WA). For each epidemiological category (month, province, and age group), the detection rate was calculated as: (number of positive samples in category / total number of samples collected and detected in that category) × 100%. This approach accounts for unequal sampling across categories. Risk factor analysis To assess the association between pathogen infection status and RSBFS occurrence, a case-control analysis was performed. Ducks displaying clear clinical signs of RSBFS were defined as cases (n = 115), while asymptomatic ducks from the same or neighboring farms with matched age, breed, and husbandry conditions were defined as controls (n = 185). Univariate analysis was conducted by calculating odds ratios (OR)and 95% confidence intervals (CI)using a 2 × 2 contingency table for each pathogen (in cases of monoinfection) to evaluate the individual association with disease. Bivariate analysis was further performed to evaluate the disease risk associated with coinfections (e.g., WPV and DuCV, WPV and NDRV, DuCV and NDRV, and WPV, DuCV, and NDRV). Statistical significance was inferred if the 95% CI for the OR did not include 1.0. PCR amplification and sequencing For samples positive for DuCV, WPV, or NDRV by qPCR, conventional PCR was conducted with a T100 thermal cycler (Bio-Rad Laboratories, Inc., Hercules, CA) to obtain longer sequences for phylogenetic analysis. Specific primers were designed using Oligo 7 software (Rychlik, 2007) based on conserved regions of the complete genome for DuCV, the VP3 gene for WPV, and the S1 gene for NDRV, using reference sequences retrieved from GenBank (Table 2). Primers were synthesized by Tsingke Biotechnology Co., Ltd. (Qingdao, China). Each RT-PCR reaction (25 µL volume) using the One-step RT-PCR Kit II (Accurate Biology, Changsha, China) contained 12.5 µL of 2 × One-step RT-PCR Master Mix, 1 µL each of forward and reverse primers (10 µM), 2 µL of template DNA or RNA, and 8.5 µL of ddH₂O. Thermal cycling conditions were: reverse transcription at 50°C for 30 min; initial denaturation at 94°C for 5 min; 35 cycles of 94°C for 30 s, primer-specific annealing (temperatures listed in Table 2) for 30 s, and 72°C for 1 min per kb of expected product; followed by a final extension at 72°C for 10 min. PCR products were visualized on 1.5% agarose gels, purified, and subjected to bidirectional Sanger sequencing by Tsingke Biotechnology Co., Ltd. (Qingdao, China).Table 2Primer sequences and conditions used for amplification and sequencing of NDRV, WPV, and DuCV.Table 2 dummy alt textVirusPrimer nameSequence (5′→3′)Reference sequenceTargetAnnealing temp.Product sizeNDRVNDRV-FATGGATCGCAACGAGGTGAMT939507.1σC56°C998 bpNDRV-RATGAATAGCTCTTCTCATCGCWPVWPV-FGGCTCGGCTACGGAAGDQ299421.1VP353°C1127 bpWPV-RCTGGTTCCAATCAATCTATCTTDuCVDuCV-FCATGGACGACTTTTATGGTTGGMT318126.1Partial Cap and Rep55°C882 bpDuCV-RCAACATTTACCAGAARAGCAARAC Table 2Primer sequences and conditions used for amplification and sequencing of NDRV, WPV, and DuCV.Table 2 dummy alt textVirusPrimer nameSequence (5′→3′)Reference sequenceTargetAnnealing temp.Product sizeNDRVNDRV-FATGGATCGCAACGAGGTGAMT939507.1σC56°C998 bpNDRV-RATGAATAGCTCTTCTCATCGCWPVWPV-FGGCTCGGCTACGGAAGDQ299421.1VP353°C1127 bpWPV-RCTGGTTCCAATCAATCTATCTTDuCVDuCV-FCATGGACGACTTTTATGGTTGGMT318126.1Partial Cap and Rep55°C882 bpDuCV-RCAACATTTACCAGAARAGCAARAC Primer sequences and conditions used for amplification and sequencing of NDRV, WPV, and DuCV. Phylogenetic analysis The obtained sequences were assembled and edited using SeqMan software. Assembled and edited sequences were processed using SeqMan software. Sequence identities were confirmed via BLASTn alignment against the GenBank nucleotide database. Multiple sequence alignments were generated with ClustalW. Phylogenetic trees were reconstructed using IQ-TREE 3 under the Maximum Likelihood criterion with the K3Pu+F+G4 model, selected as the best-fit model according to the Bayesian Information Criterion (BIC; Wong, et al., 2026), and node support was assessed with 1,000 bootstrap replicates. Trees were visualized using iTOL v6 (Letunic and Bork, 2024). Nucleotide and amino acid identity was calculated in BioEdit v7.0.9 (http://www.mbio.ncsu.edu/BioEdit/). Reference strains representing known genotypes and subgenotypes (Supplementary Table 1) were included in all analyses. Results Overall pathogen detection profile Multiple pathogens were detected in the 300 clinical samples (Fig. 2). Fig. 2 presents detection profiles for pooled data from symptomatic and asymptomatic ducks. DuCV was the most frequently detected pathogen, with 194 positives and a detection rate of 66.44% (194/292) in the pooled population. In symptomatic versus asymptomatic ducks, DuCV showed comparable rates (67.6% vs 65.8%). NDRV was detected in 86 samples (32.33%, 86/266) overall, with similar rates in symptomatic and asymptomatic groups (34.4% vs 31.0%). WPV was found in 56 samples (19.18%, 56/292) overall, with a detection rate of 44/108 (40.74%) in symptomatic ducks and 12/184 (6.52%) in asymptomatic ducks. DAstV-Ⅲ was detected in 23.58% (29/123) samples in the total population. Bacteria were also common: SM was detected in 50.00% (20/40) samples, and RA was found in 24.07% (13/54) samples. Lower detection rates were observed for DAstV-Ⅰ (9.28%, 9/97), DTMUV (7.41%, 2/27), and classical Muscovy duck reovirus (MDRV, 0.38%, 1/263), with the single MDRV case identified as a coinfection with NDRV.Fig. 2Detection numbers and detection rates of different pathogens in meat ducks. DuCV, duck circovirus; NDRV, novel duck reovirus; WPV, waterfowl parvovirus; DAstV-Ⅲ, Duck astrovirus type Ⅲ; DAstV-Ⅰ, Duck astrovirus type Ⅰ; DTMUV, Duck Tembusu virus; MDRV, Muscovy duck reovirus; SM, Salmonella; RA, Riemerella anatipestifer.Fig 2 dummy alt text Detection numbers and detection rates of different pathogens in meat ducks. DuCV, duck circovirus; NDRV, novel duck reovirus; WPV, waterfowl parvovirus; DAstV-Ⅲ, Duck astrovirus type Ⅲ; DAstV-Ⅰ, Duck astrovirus type Ⅰ; DTMUV, Duck Tembusu virus; MDRV, Muscovy duck reovirus; SM, Salmonella; RA, Riemerella anatipestifer. Epidemiological analysis of major pathogens Temporal distribution: DuCV was detected throughout the 6-month study period, with the highest number of positive cases (n = 118) occurring in June. Its monthly detection rate remained consistently high, exceeding 56% from February to July. NDRV showed a similar peak in detection numbers in June (n = 46), although its highest monthly detection rate (50%) was observed in February. WPV detection exhibited a different pattern, with the highest number of positives (n = 20) and a notable peak in detection rate (66.7%) occurring in the spring months of April and March, respectively (Fig. 3; Supplementary Table 1).Fig. 3Monthly distribution of the three major pathogens, namely waterfowl parvovirus (WPV), duck circovirus (DuCV), and novel duck reovirus (NDRV) associated with red skin and bristle feather syndrome in meat ducks from February to July 2025.Fig 3 dummy alt text Monthly distribution of the three major pathogens, namely waterfowl parvovirus (WPV), duck circovirus (DuCV), and novel duck reovirus (NDRV) associated with red skin and bristle feather syndrome in meat ducks from February to July 2025. Geographical distribution: DuCV-positive samples were found in all eight sampled provinces. Shandong province had the highest number of DuCV positives (n = 87, detection rate 66.4% based on n = 135 total samples). Henan showed the highest DuCV detection rate at 86.96% (20/23). For NDRV, Shandong had the most positives (n = 52, 42.28% of 123 samples), while Hebei recorded the highest rate (50%, 17/34). WPV was most frequently detected in Shandong (n = 37, 28.24% of 131 samples). Notably, Liaoning (n = 6), Guangxi (n = 5), and Fujian (n = 3) showed 100% WPV detection rate, likely reflecting localized outbreaks rather than true regional prevalence, and should be interpreted with extreme caution (Fig. 4; Supplementary Table 2).Fig. 4Geographical distribution of the three major pathogens across eight provinces in China. DuCV, duck circovirus; NDRV, novel duck reovirus; WPV, waterfowl parvovirus. Prevalence estimates from provinces with n < 10 have limited statistical power and should be interpreted with caution.Fig 4 dummy alt text Geographical distribution of the three major pathogens across eight provinces in China. DuCV, duck circovirus; NDRV, novel duck reovirus; WPV, waterfowl parvovirus. Prevalence estimates from provinces with n < 10 have limited statistical power and should be interpreted with caution. Age distribution: Analysis of infection rates by week of age revealed distinct patterns for the three viruses (Fig. 5; Supplementary Table 3). DuCV infection was most prevalent in ducks aged 3 to 5 weeks, peaking at 37.11% at 3 weeks. NDRV infection was predominantly found in younger ducks, with higher proportions (27.91% and 25.58%, respectively) at 3 and 4 weeks of age and a sharp decline after 5 weeks. WPV infection also showed a clear peak in susceptibility during the 3-to-5-weeks period, with the highest rate (41.07%) at 4 weeks, followed by a decrease in older ducks.Fig. 5Age distribution (by week) of the three major pathogens, namely waterfowl parvovirus (WPV), duck circovirus (DuCV), and novel duck reovirus (NDRV) in meat ducks with red skin and bristle feather syndrome. Values represent the percentage of positive samples within each age group.Fig 5 dummy alt text Age distribution (by week) of the three major pathogens, namely waterfowl parvovirus (WPV), duck circovirus (DuCV), and novel duck reovirus (NDRV) in meat ducks with red skin and bristle feather syndrome. Values represent the percentage of positive samples within each age group. Coinfection analysis: Coinfection patterns were complex and frequent (Fig. 6A–C). Among the 194 DuCV-positive samples, the most frequent co-infecting pathogen was SM (41.7%), followed by WPV, which was present in 24.2% (47/194) of DuCV-positive cases. Conversely, in the 56 WPV-positive samples, coinfection with DuCV was extremely common, occurring in 83.9% (47/56) of cases. For the 86 NDRV-positive samples, the highest coinfection rate was with SM (61.1%), but the most numerically significant viral coinfection was with DuCV (41/86, 47.7%).Fig. 6Coinfection profiles of major pathogens. (A) Pathogens co-detected with duck circovirus (DuCV). (B) Pathogens co-detected with waterfowl parvovirus (WPV). (C) Pathogens co-detected with novel duck reovirus (NDRV). Bars represent the coinfection rates, and numerical values adjacent to bars indicate the number of co-positive samples. For each coinfection rate, the denominator comprised only those index-positive samples also tested for the specific co-pathogen. DAstV-Ⅲ, Duck astrovirus type Ⅲ; DAstV-Ⅰ, Duck astrovirus type Ⅰ; DTMUV, Duck Tembusu virus; MDRV, Muscovy duck reovirus; SM, Salmonella; RA, Riemerella anatipestifer.Fig 6 dummy alt text Coinfection profiles of major pathogens. (A) Pathogens co-detected with duck circovirus (DuCV). (B) Pathogens co-detected with waterfowl parvovirus (WPV). (C) Pathogens co-detected with novel duck reovirus (NDRV). Bars represent the coinfection rates, and numerical values adjacent to bars indicate the number of co-positive samples. For each coinfection rate, the denominator comprised only those index-positive samples also tested for the specific co-pathogen. DAstV-Ⅲ, Duck astrovirus type Ⅲ; DAstV-Ⅰ, Duck astrovirus type Ⅰ; DTMUV, Duck Tembusu virus; MDRV, Muscovy duck reovirus; SM, Salmonella; RA, Riemerella anatipestifer. Regarding specific viral combinations (Fig. 7), the DuCV and WPV coinfection had the highest relative rate among dual infections (24.2% of DuCV-positive samples), while NDRV and WPV coinfection was less frequent (15.3% of NDRV-positive samples). Notably, triple coinfection involving all three major viruses (DuCV, NDRV, and WPV) was detected in 19.5% (8/41) of the samples that were positive for at least two of these viruses.Fig. 7Detection rates and coinfection patterns among novel duck reovirus (NDRV), waterfowl parvovirus (WPV), and duck circovirus (DuCV).Fig 7 dummy alt text Detection rates and coinfection patterns among novel duck reovirus (NDRV), waterfowl parvovirus (WPV), and duck circovirus (DuCV). Risk factor analysis Univariate analysis (Table 3) revealed that monoinfection with WPV was strongly and significantly associated with clinical RSBFS (OR = 9.85; 95% CI: 4.89–19.84). The positive rate for WPV was 40.7% in cases versus only 6.5% in controls. In contrast, monoinfection with either DuCV (OR = 1.09; 95% CI: 0.66–1.80) or NDRV (OR = 1.17; 95% CI: 0.69–1.98) showed no significant association, as their detection rates were similar between case and control groups.Table 3Risk factor analysis: Association between pathogen infection status and red skin and bristle feather syndrome.Table 3 dummy alt textInfection statusCase: positive/total (%)Control: positive/total (%)Odds ratio95% confidence intervalWPV mono-infection44/108 (40.74)12/184 (6.52)9.854.89–19.84DuCV mono-infection73/108 (67.59)121/184 (65.76)1.090.66–1.80NDRV mono-infection33/96 (34.38)53/171 (30.99)1.170.69–1.98WPV + DuCV co-infection36/63 (57.14)11/73 (15.07)7.523.33–16.94WPV + NDRV co-infection10/43 (23.26)3/115 (2.61)11.312.94–43.52DuCV+NDRV co-infection12/24 (50.00)29/58 (50.00)10.39–2.59WPV+DuCV+NDRV co-infection5/15 (33.33)3/32 (9.38)4.830.97–23.97Note: WPV, waterfowl parvovirus; DuCV, duck circovirus; NDRV, novel duck reovirus. Table 3Risk factor analysis: Association between pathogen infection status and red skin and bristle feather syndrome.Table 3 dummy alt textInfection statusCase: positive/total (%)Control: positive/total (%)Odds ratio95% confidence intervalWPV mono-infection44/108 (40.74)12/184 (6.52)9.854.89–19.84DuCV mono-infection73/108 (67.59)121/184 (65.76)1.090.66–1.80NDRV mono-infection33/96 (34.38)53/171 (30.99)1.170.69–1.98WPV + DuCV co-infection36/63 (57.14)11/73 (15.07)7.523.33–16.94WPV + NDRV co-infection10/43 (23.26)3/115 (2.61)11.312.94–43.52DuCV+NDRV co-infection12/24 (50.00)29/58 (50.00)10.39–2.59WPV+DuCV+NDRV co-infection5/15 (33.33)3/32 (9.38)4.830.97–23.97Note: WPV, waterfowl parvovirus; DuCV, duck circovirus; NDRV, novel duck reovirus. Risk factor analysis: Association between pathogen infection status and red skin and bristle feather syndrome. Note: WPV, waterfowl parvovirus; DuCV, duck circovirus; NDRV, novel duck reovirus. Bivariate analysis demonstrated that coinfections involving WPV significantly increased disease risk. WPV and DuCV coinfection was associated with an OR of 7.52 (95% CI: 3.33–16.94), and WPV and NDRV coinfection showed an even stronger association (OR = 11.31; 95% CI: 2.94–43.52). DuCV and NDRV coinfection was associated with an OR of 1.00 (95% CI: 0.39–2.59). Although the triple coinfection (DuCV, NDRV, and WPV) showed a higher positive rate in cases (33.3%) than controls (9.4%), the OR (4.83; 95% CI: 0.97-23.97) did not reach statistical significance. The wide CI including 1.0 indicates insufficient statistical power (n = 15 cases, n = 32 controls) for definitive conclusions. Phylogenetic analysis Genetic evolution of DuCV: Forty-three DuCV partial genomes were obtained from positive samples (GenBank accession numbers PZ143892 to PZ143934). Phylogenetic analysis (Fig. 8) classified all strains within genotype I. Subgenotype Ib was the most dominant, comprising 30 strains, followed by subgenotype Ic with 12 strains. Only one strain clustered within subgenotype Ia. The sequenced strains shared 94.8 to 99.8% nucleotide sequence identity with the reference strain OR134465.1 (DuCV-GX08-2019) and were closely related to recently circulating Chinese strains (Supplementary Table 4).Fig. 8Phylogenetic tree based on partial Cap and Rep gene sequences of duck circoviruses (DuCVs) using Maximum Likelihood method with K3Pu+F+G4 model and 1,000 bootstrap replicates. Bootstrap values >70% are indicated at key nodes. Reference strains representing known genotypes (Ia, Ib, and Ic) are included for comparison.Fig 8 dummy alt text Phylogenetic tree based on partial Cap and Rep gene sequences of duck circoviruses (DuCVs) using Maximum Likelihood method with K3Pu+F+G4 model and 1,000 bootstrap replicates. Bootstrap values >70% are indicated at key nodes. Reference strains representing known genotypes (Ia, Ib, and Ic) are included for comparison. Genetic evolution of NDRV: Nineteen complete S1 gene sequences of NDRV were acquired (GenBank accession numbers PZ105720 to PZ105738). These strains exhibited high conservation, with 97.1 to 98.8% nucleotide identity and 96.9 to 99.2% amino acid identity to the contemporary Chinese reference strain ON040913.1 (Duck/N-DRV-LY20/China/2021) (Supplementary Table 4). In the phylogenetic tree (Fig. 9), all study strains formed a tight cluster with other recent NDRV isolates from China, distinct from classical MDRV.Fig. 9Phylogenetic tree based on the S1 gene sequences encoding σC protein of novel duck reoviruses (NDRVs) using Maximum Likelihood method with K3Pu+F+G4 model and 1,000 bootstrap replicates. Bootstrap values >70% are indicated at key nodes. Reference strains representing classical and novel genotypes are included for comparison.Fig 9 dummy alt text Phylogenetic tree based on the S1 gene sequences encoding σC protein of novel duck reoviruses (NDRVs) using Maximum Likelihood method with K3Pu+F+G4 model and 1,000 bootstrap replicates. Bootstrap values >70% are indicated at key nodes. Reference strains representing classical and novel genotypes are included for comparison. Genetic evolution of WPV: Twenty-seven WPV VP3 gene sequences were obtained (GenBank accession numbers PZ105739 to PZ105765). Twenty-six of these were identified as NGPV, confirming NGPV as the overwhelmingly dominant WPV subtype in this syndrome. One strain (WYL250722-H10) was identified as MDGPV. The 26 NGPV strains shared 98.5 to 99.8% nucleotide identity and 98.4 to 99.6% amino acid identity with the reference strain MT174597.1 (DSDJX1810). Strain WYL250722-H10 (MDGPV) showed 98.8% nucleotide identity to reference OQ301813.1. The nucleotide identity between this MDGPV strain and the NGPV strains ranged from 89.7 to 91.0% (Fig. 10).Fig. 10Phylogenetic tree based on VP3 gene sequences of waterfowl parvoviruses (WPVs) using Maximum Likelihood method with K3Pu+F+G4 model and 1,000 bootstrap replicates. Bootstrap values >70% are indicated at key nodes. Reference strains representing classical goose parvovirus (GPV), Muscovy duck parvovirus (MDPV), novel goose parvovirus (NGPV), and Muscovy duck-goose parvovirus (MDGPV) genotypes are included for comparison.Fig 10 dummy alt text Phylogenetic tree based on VP3 gene sequences of waterfowl parvoviruses (WPVs) using Maximum Likelihood method with K3Pu+F+G4 model and 1,000 bootstrap replicates. Bootstrap values >70% are indicated at key nodes. Reference strains representing classical goose parvovirus (GPV), Muscovy duck parvovirus (MDPV), novel goose parvovirus (NGPV), and Muscovy duck-goose parvovirus (MDGPV) genotypes are included for comparison. Discussion This large-scale epidemiological and molecular investigation identifies DuCV, NDRV, and WPV (predominantly as NGPV) as the three core viral pathogens circulating in Chinese meat duck flocks affected by RSBFS. The high detection rate of DuCV (66.4%), often considered a primary immunosuppressive pathogen in ducks, suggests it may play a foundational role in the disease complex by compromising host defenses and creating an opportunity for other pathogens (Hong et al., 2018). However, our risk factor analysis provides a crucial nuanced insight: While DuCV is ubiquitous, monoinfection with DuCV alone was not a significant risk factor for clinical RSBFS. This finding is critical for directing control efforts and suggests that the presence of DuCV may be a necessary but insufficient condition for the full expression of the syndrome. A key finding of this study was the identification of WPV (specifically NGPV) as the primary risk factor for RSBFS. The strong association (OR > 9) between NGPV monoinfection and disease, coupled with its known pathogenicity in causing feather disorders like short beak and dwarfism syndrome (Shen et al., 2015; Zádori et al., 1995; Zhu et al., 2025), identifies NGPV as the dominant infectious risk factor for RSBFS. These findings have direct implications for poultry production: surveillance and control programs for RSBFS must prioritize NGPV detection and immunity. The observed interactive effects between NGPV and DuCV/NDRV may involve several plausible mechanisms, though these remain to be experimentally validated: (i) DuCV-induced immunosuppression through thymic and bursal damage (Hong et al., 2018) may compromise innate immune responses to NGPV, facilitating enhanced viral replication; (ii) NDRV-induced splenic necrosis and cytokine dysregulation (Wang et al., 2020) may create a pro-inflammatory environment that exacerbates NGPV-mediated feather follicle damage; (iii) concurrent infection may overwhelm host repair mechanisms during the critical feather growth period (3-5 weeks). The term "interactive effects" is used here to describe the statistical observation of higher disease odds in coinfected birds; biological synergy requires experimental proof. For producers, this underscores the danger of concurrent infections and highlights the importance of comprehensive flock health management that addresses multiple pathogens, not just one. Our pathogen screening was restricted to a predefined panel of known waterfowl pathogens using targeted qPCR. While sensitive and specific for known pathogens, this approach cannot detect novel or emerging pathogens. Metagenomic next-generation sequencing (mNGS) of clinical samples would provide an unbiased assessment of the complete virome and microbiome, potentially revealing previously unrecognized etiological contributors to RSBFS. We have preserved a subset of clinical samples for future mNGS analysis. The epidemiological patterns offer practical guidance for on-farm management. The concentration of cases in major duck-producing provinces like Shandong, Anhui, and Henan points to high stocking density and frequent animal movement as key risk amplifiers. Strengthening biosecurity, implementing strict all-in/all-out systems, and controlling farm-to-farm transmission in these hubs should be a regional priority. The peak susceptibility age of 3 to 5 weeks coincides with a period of rapid growth, high metabolic demand for feather development, and potential decline in maternal antibodies. This identifies a critical window in the production cycle where nutritional support, stress reduction, and targeted vaccination (if available) could be most beneficial. The temporal pattern—increased WPV detection in spring and DuCV/NDRV in summer—suggests seasonal stressors (climatic shifts, heat, humidity) may influence virus dynamics, prompting the need for seasonally adjusted management practices. From a genetic perspective, the high conservation of the NDRV S1 gene and the dominance of specific DuCV (Ib/Ic) and WPV (NGPV) genotypes are encouraging for vaccine development, with the predominance of DuCV Ib being consistent with our previous longitudinal surveillance (Li et al., 2025) and the overwhelming dominance of NGPV corroborated by our laboratory's ongoing surveillance (unpublished data). The molecular data provided here can directly inform the design of matched, regionally effective vaccines against these key viruses. The detection of a diverse WPV strain (MDGPV) also serves as a reminder for ongoing surveillance to monitor for emerging variants that could evade immunity. It is vital to view RSBFS through a multifactorial lens, particularly in the context of intensive production. While this study focuses on viral triggers, the high detection rates of bacteria like SM and RA indicate they contribute to secondary inflammation and worsening lesions. Furthermore, non-infectious factors prevalent in commercial settings—such as nutritional deficiencies (sulfur amino acids, biotin), skin abrasions from housing, and environmental toxins—are known to cause or exacerbate feathering problems (Yang et al., 2020; Hoffman, 2002; Gan et al., 2023). Therefore, a successful control strategy must be integrated, combining: (1) pathogen-specific measures (vaccination against NGPV and DuCV, antimicrobial control of bacteria); (2) nutritional optimization to support skin and feather integrity; (3) enhanced husbandry to reduce stress and physical damage; and (4) environmental monitoring for contaminants. Several limitations of this study should be acknowledged. First, as a cross-sectional survey, our findings reveal epidemiological associations but do not experimentally confirm causation. NGPV pathogenicity has been demonstrated in SPF ducks (Zhu et al., 2025), including feather development disorders, but definitive confirmation specifically for the RSBFS phenotype requires controlled challenge studies. Second, the 6-month sampling window (February to July 2025) may not capture full seasonal dynamics; longitudinal surveillance spanning ≥12 months is warranted. Third, sample size imbalances across provinces limit the robustness of regional prevalence estimates from underrepresented regions (Liaoning n = 6, Guangxi n = 5, Fujian n = 3, Guangdong n = 2), which should be interpreted as exploratory only. Fourth, while our case-control design matched cases and controls for farm, age, breed, and husbandry conditions, we did not systematically measure feed composition, mycotoxin levels, or environmental parameters; non-infectious factors (nutritional deficiencies, toxins, stocking density) are established causes of feathering disorders and may confound the observed associations. Fifth, the proposed mechanisms (DuCV-induced immunosuppression, NDRV-mediated inflammation, host repair overwhelm during critical feather growth) remain hypothetical pending experimental validation. Sixth, our predefined qPCR panel, while sensitive for known pathogens, cannot detect novel pathogens. Finally, the non-significant triple-coinfection association likely reflects insufficient statistical power (post-hoc analysis indicates ∼45 cases and 90 controls with triple infection would be needed for 80% power at α=0.05 to detect OR=4.83) rather than absence of effect. Despite these constraints, our findings establish a robust epidemiological risk factor and molecular framework for understanding RSBFS. This framework identifies NGPV as the primary statistical risk factor, quantifies the interactive effects of coinfections, characterizes the predominant circulating genotypes, and maps the spatiotemporal distribution of disease risk. These foundational data provide clear priorities for future research: SPF duck challenge studies to validate pathogenicity, expanded longitudinal surveillance to capture seasonal dynamics, comprehensive nutritional and environmental assessments to disentangle multifactorial contributions, mechanistic investigations of viral synergy, metagenomic exploration of the complete pathogenome, and multivalent vaccine development targeting the characterized genotypes (DuCV Ib/Ic, NDRV S1, NGPV). In conclusion, this study provides actionable guidance for the poultry industry: prioritize NGPV surveillance and control, manage immunosuppressive viruses (DuCV and NDRV) to prevent additive disease effects, and integrate these measures within holistic farm management programs that address the identified spatial, temporal, and host-age risk factors. While our findings establish strong statistical associations, experimental confirmation through controlled infection studies remains the critical next step to validate pathogenesis and inform the development of effective multivalent vaccines based on the characterized strains. CRediT authorship contribution statement Zhongcheng Sun: Writing – original draft, Visualization, Software, Methodology, Investigation, Formal analysis, Data curation. Xue Wang: Data curation. Hao Shi: Resources. Hao Zhu: Resources, Investigation. Xue Liu: Investigation. Tianyao Yang: Resources. Qingqing Song: Resources, Funding acquisition. Zhaoyang Li: Resources, Funding acquisition. Chunguo Liu: Writing – review & editing, Writing – original draft, Validation, Supervision, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Disclosures The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.