Advertisement

Epidemiological and clinical characteristics of humans with avian influenza A (H7N9) infection in Guangdong, China, 2013–2017

Open AccessPublished:November 08, 2017DOI:https://doi.org/10.1016/j.ijid.2017.07.021

      Highlights

      • The fatality rate of the 256 confirmed cases was 39.0%, and the proportion of ICU admission was 65.6%.
      • The proportions of accepting oseltamivir, zanamivir, and peramivir treatment were 83.1%, 13.6%, and 29.6%, respectively.
      • Earlier oseltamivir and zanamivir treatment were negatively associated with the death of A (H7N9) infected patients.
      • Time from illness onset to confirmation and oseltamivir treatment were positively related to ICU admission.

      Abstract

      Objective

      To describe the demographics and clinical characteristics of patients with A (H7N9) infection, to test the differences in the distribution of demographics and clinical characteristics by clinical severity, and to explore potential factors associated with clinical severity.

      Methods

      This retrospective study was conducted to collect epidemiological and clinical information regarding the confirmed cases in Guangdong through field investigation and review of medical records.

      Results

      Of the 256 cases, 100 (39.0%) patients died, and 168 (65.6%) patients were admitted to ICUs. The male-to-female ratio was approximately 2.0:1, and the median age was 56 years (range, 1 to 88). Among the 215 patients accepting oseltamivir treatment, the median time from the onset of illness to oseltamivir treatment was 5 days (range, 0 to 16); 35 patients received zanamivir treatment after a median of 8 days (range, 0 to 23). The univariable logistic regression models demonstrated that time from the onset of illness to oseltamivir treatment (OR = 1.10, 95% CI = 1.01–1.10) and zanamivir treatment (OR = 1.05, 95% CI = 1.02–1.07) were associated with the death of patients.

      Conclusions

      Preventive measures should focus on high-risk populations, such as the elderly and the groups with high frequency exposure to live poultry. Earlier oseltamivir and zanamivir treatment were recommended.

      Keywords

      Introduction

      Since human infections with avian influenza virus A (H7N9) were first reported by the Chinese Center for Disease Control and Prevention in March 2013 (
      • Gao R.
      • Cao B.
      • Hu Y.
      • Feng Z.
      • Wang D.
      • Hu W.
      • et al.
      Human infection with a novel avian-origin influenza A (H7N9) virus.
      ), mainland China has experienced five influenza A (H7N9) virus epidemics, and the direct medical costs and DALY costs are higher than for most other respiratory infection diseases (
      • Qi X.
      • Jiang D.
      • Wang H.
      • Zhuang D.
      • Ma J.
      • Fu J.
      • et al.
      Calculating the burden of disease of avian-origin H7N9 infections in China.
      ). Although the first wave of A (H7N9) infection was centered on the eastern China, H7N9 viruses have spread from eastern to southern China and shown multiple regionalized lineages with different reassortant genotypes (
      • Lam T.T.
      • Zhou B.
      • Wang J.
      • Chai Y.
      • Shen Y.
      • Chen X.
      • Zhu H.
      • et al.
      Dissemination, divergence and establishment of H7N9 influenza viruses in China.
      ,
      • Zhou P.
      • Ma J.
      • Lai A.
      • Gray G.C.
      • Su S.
      • Li S.
      Avian influenza A(H7N9) virus and mixed live poultry-animal markets in Guangdong province: a perfect storm in the making?.
      ). However, it is still unclear how the H7N9 virus re-emerged and disseminated across the five epidemic waves.
      Previous studies have consistently reported that poultry exposure was one of the most important risk factors for A (H7N9) infection; visiting live poultry markets (LPMs) and raising backyard poultry seemed to be the main source of human infections (
      • Chen Y.
      • Liang W.
      • Yang S.
      • Wu N.
      • Gao H.
      • Sheng J.
      • et al.
      Human infections with the emerging avian influenza A H7N9 virus from wet market poultry: clinical analysis and characterisation of viral genome.
      ,
      • Li J.
      • Chen J.
      • Yang G.
      • Zheng Y.X.
      • Mao S.H.
      • Zhu W.P.
      • et al.
      Case-control study of risk factors for human infection with avian influenza A(H7N9) virus in Shanghai, China, 2013.
      ,
      • Liu B.
      • Havers F.
      • Chen E.
      • Yuan Z.
      • Yuan H.
      • Ou J.
      • et al.
      Risk factors for influenza A(H7N9) disease–China, 2013.
      ). Patients with A (H7N9) infection often experienced at least one medical complication, including progressing pneumonia or acute respiratory distress syndrome (ARDS) (
      • Zhou L.
      • Ren R.
      • Yang L.
      • Bao C.
      • Wu J.
      • Wang D.
      • et al.
      Sudden increase in human infection with avian influenza A(H7N9) virus in China, September–December 2016.
      ); those patients also always showed systematic organic injury, including the circulation, liver, or kidneys (
      • Ji H.
      • Gu Q.
      • Chen L.L.
      • Xu K.
      • Ling X.
      • Bao C.J.
      • et al.
      Epidemiological and clinical characteristics and risk factors for death of patients with avian influenza A H7N9 virus infection from Jiangsu Province, Eastern China.
      ,
      • Wang C.
      • Yu H.
      • Horby P.W.
      • Cao B.
      • Wu P.
      • Yang S.
      • et al.
      Comparison of patients hospitalized with influenza A subtypes H7N9, H5N1, and 2009 pandemic H1N1.
      ). These medical complications or systematic organic injuries prolonged hospitalization, affected treatment and impaired recovery. Prior studies also reported that most patients were admitted to an intensive care unit (ICU), and had severe illness or even death (
      • Wang C.
      • Yu H.
      • Horby P.W.
      • Cao B.
      • Wu P.
      • Yang S.
      • et al.
      Comparison of patients hospitalized with influenza A subtypes H7N9, H5N1, and 2009 pandemic H1N1.
      ,
      • Wang Z.
      • Wan Y.
      • Qiu C.
      • Quinones-Parra S.
      • Zhu Z.
      • Loh L.
      • Tian D.
      • et al.
      Recovery from severe H7N9 disease is associated with diverse response mechanisms dominated by CD8(+) T cells.
      ,
      • Gao H.N.
      • Lu H.Z.
      • Cao B.
      • Du B.
      • Shang H.
      • Gan J.H.
      • et al.
      Clinical findings in 111 cases of influenza A (H7N9) virus infection.
      ). To date, no obvious breakthrough in the clinical treatment of A (H7N9) infection has been identified, but several studies have reported that severe cases show a significantly longer interval from the onset of illness to antiviral treatment and to the negative clinical outcome (
      • Wang Z.
      • Wan Y.
      • Qiu C.
      • Quinones-Parra S.
      • Zhu Z.
      • Loh L.
      • Tian D.
      • et al.
      Recovery from severe H7N9 disease is associated with diverse response mechanisms dominated by CD8(+) T cells.
      ,
      • Leung Y.
      • To M.
      • Lam T.
      • Yau S.
      • Leung O.
      • Chuang S.
      Epidemiology of human influenza A(H7N9) infection in Hong Kong.
      ). Therefore, World Health Organization (WHO) emphasizes that we should strengthen early diagnosis and treatment of A (H7N9) infection, improve treatment of severe cases, and reduce occurrence of severe cases and deaths (
      • World Health Organization
      Human infection with avian influenza A(H7N9) virus – China.
      ).
      Guangdong province is located in the southern part of China, and accounts for approximately 10% of China’s domestic poultry industry, meeting the huge demand (
      • Zhou P.
      • Ma J.
      • Lai A.
      • Gray G.C.
      • Su S.
      • Li S.
      Avian influenza A(H7N9) virus and mixed live poultry-animal markets in Guangdong province: a perfect storm in the making?.
      ,
      • Wu J.
      • Lu J.
      • Faria N.R.
      • Zeng X.
      • Song Y.
      • Zou L.
      • et al.
      Effect of live poultry market interventions on influenza A(H7N9) virus, Guangdong, China.
      ). Although there is no clinical case of human infection reported in Guangdong during the first wave, it has become one of the main epidemic areas of A (H7N9) in the next outbreaks. This present retrospective study was conducted to describe the demographics and clinical characteristics of patients with A (H7N9) infection in Guangdong, to test the differences in the distribution of aforementioned characteristics by clinical severity, and to explore potential factors associated with clinical severity in order to help health-related institutions develop measures to determine the best allocation of medical resources.

      Methods

      Data collection

      Patients with laboratory-confirmed A (H7N9) infection were defined as those with clinical symptoms (i.e., fever, cough, coryza, and difficulty in breathing) and verified by real-time reverse transcription-PCR or viral isolation (
      • Kang M.
      • He J.
      • Song T.
      • Rutherford S.
      • Wu J.
      • Lin J.
      • et al.
      Environmental sampling for avian influenza A(H7N9) in live-poultry markets in Guangdong, China.
      ,
      • Yang Z.F.
      • He J.F.
      • Li X.B.
      • Guan W.D.
      • Ke C.W.
      • Wu S.G.
      • et al.
      Epidemiological and viral genome characteristics of the first human H7N9 influenza infection in Guangdong Province, China.
      ). All laboratory tests are based on the protocol promoted by WHO (
      • World Health Organization
      Real-time RT-PCR protocol for the detection of avian influenza A(H7N9) virus.
      ), and have been described in detail elsewhere (
      • Wu J.
      • Lu J.
      • Faria N.R.
      • Zeng X.
      • Song Y.
      • Zou L.
      • et al.
      Effect of live poultry market interventions on influenza A(H7N9) virus, Guangdong, China.
      ,
      • Yang Z.F.
      • He J.F.
      • Li X.B.
      • Guan W.D.
      • Ke C.W.
      • Wu S.G.
      • et al.
      Epidemiological and viral genome characteristics of the first human H7N9 influenza infection in Guangdong Province, China.
      ,
      • Zhou J.
      • Wu J.
      • Zeng X.
      • Huang G.
      • Zou L.
      • Song Y.
      • et al.
      Isolation of H5N6, H7N9 and H9N2 avian influenza A viruses from air sampled at live poultry markets in China, 2014 and 2015.
      ). This retrospective study was conducted to analyze the epidemiological and clinical characteristics of 256 confirmed cases of human influenza A (H7N9) infection reported between August 2013 and March 2017 in the web-based surveillance information system of Guangdong Province.
      In this study, investigators were trained to use a structured questionnaire to collect epidemiological information in hospital immediately after the patients were diagnosed in hospital, with face-to-face interviews with the patients, family members, and relevant hospital staff. Epidemiological information collected included the patients’ age, gender, location, occupation, smoking status, drinking status, detailed poultry exposure history (two weeks before the case’s date of illness onset; details are shown in Table 2), and time of the illness onset.
      Additionally, we also used a questionnaire to review medical records and to retrieve clinical data, including underlying disease (including chronic respiratory, cardiovascular, metabolic, kidney, liver, rheumatic, and hematological disease); initial symptoms; acceptance of antiviral (including oseltamivir, zanamivir, and peramivir treatment), glucocorticoid, antimicrobial, or mechanical ventilation treatment; complications; clinical severity (i.e., death and admission to ICUs); and important timelines of medical treatments (i.e., time form illness onset to first medical care, confirmation, hospitalization, oseltamivir treatment, zanamivir treatment, peramivir treatment, and prognosis [i.e., death or recovery]).

      Statistical analysis

      All data were independently entered twice by two investigators using Microsoft Excel 2016 and analyzed using STATA 10.0. A descriptive analysis was performed on the demographics and clinical characteristics to help with the prevention of human influenza A (H7N9) infections. Categorical and continuous data are reported as number (%), means (±SD), or median, as appropriate. Chi-squared tests were used to examine the differences in categorical variables, and Fisher’s exact tests were used when appropriate. T-tests were employed to assess the differences in the normally distributed continuous variables, and Wilcoxon rank-sum tests were applied to examine the differences in the non-normally distributed continuous variables. The univariable logistic regression models without adjusting for other variables were conducted to explore the potential associations between timelines for medical care and clinical severity. All statistical tests were two-sided, and a P-value less than 0.05 was considered significant.

      Results

      As shown in Figure 1, Guangdong province experienced four waves of the A (H7N9) outbreak, and the number of cases were highest in 2014–15. There were seasonal fluctuations in the number of cases, and most A (H7N9) infections in humans occurred during December to March.
      Figure 1
      Figure 1Temporal pattern of confirmed human influenza A (H7N9) infections reported in Guangdong Province (August, 2013 ∼ March, 2017) epidemic curve of cases by week.

      General characteristics of confirmed cases from 2013 to 2017

      Table 1 demonstrated that of the 256 hospitalized patients, 170 (66.4%) were male and 86 (33.6%) were female, and the male-to-female ratio was approximately 2.0:1. The patients’ ages ranged from 1 to 88 years, with a median age of 56 years. The age group with the greatest percentage of patients was the 60- to 79-year-old group (30.9%), followed by the 30- to 49-year-old group (26.6%) and the 50- to 59-year-old-group (25.0%). The most common occupational groups among the confirmed patients were retired (25.8%), enterprise staff (16.8%), unemployed persons (14.8%), and peasants (14.5%), who together accounted for 71.9% of the total. The proportion of patients who reported smoking and drinking is 14.8% and 5.8%, respectively. The fatality rate of the 256 confirmed cases was 39.0%, and the proportion of ICU admission was 65.6% (168/256). Our results revealed that patients in fatal cases were significantly older than in nonfatal cases, and the differences between fatal and nonfatal cases in the distribution of age and occupation were statistically significant (P< 0.05). Additionally, statistically significant differences in patients admitted and not admitted to ICU were observed in the distribution of age, gender, and occupation (P< 0.05).
      Table 1General characteristics of confirmed cases from 2013 to 2017.
      VariableTotal (N = 256)Death, n (%)Admission to ICUs, n (%)
      Fatal (n = 100)Nonfatal (n = 156)P-value
      Chi-squared tests were used for testing the difference in categorical variables, and Wilcoxon rank-sum tests were used for examining the differences in age data.
      Yes (n = 168)No (n = 88)P-value
      Chi-squared tests were used for testing the difference in categorical variables, and Wilcoxon rank-sum tests were used for examining the differences in age data.
      Median age (range), years56 (1–88)61 (19–83)51 (1–88)<0.00157 (1–88)52 (2–88)0.087
      Age group, years
       ≤1822 (8.6)022 (100)<0.0013 (13.6)19 (86.4)<0.001
       18–296 (2.3)1 (16.7)5 (83.2)3 (50.0)3 (50.0)
       30–4968 (26.6)20 (29.4)48 (70.6)48 (70.6)20 (29.4)
       50–5964 (25.0)28 (43.8)36 (56.3)45 (70.3)19 (29.7)
       60-–7979 (30.9)42 (53.2)37 (46.8)58 (73.4)21 (26.6)
       ≥8017 (6.6)9 (52.9)8 (47.1)11 (64.7)6 (35.3)
      Gender
       Male170 (66.4)70 (70.0)100 (64.1)0.330120 (70.6)50 (29.40.019
       Female86 (33.6)30 (30.0)56 (35.9)48 (55.8)38 (44.2)
      Occupation
       Retired66 (25.8)32 (48.5)34 (51.5)0.00246 (69.7)20 (30.3)<0.001
       Enterprise staff
      Enterprise staff means a person who works for an enterprise organization.
      43 (16.8)17 (39.5)26 (60.5)34 (79.1)9 (20.9)
       Unemployed38 (14.8)16 (42.1)22 (57.9)23(60.5)15 (39.5)
       Peasants
      Peasant means a farmer or agricultural worker of low status.
      37 (14.5)21 (56.8)16 (43.2)27 (73.0)10 (27.0)
       Workers
      Worker means an employee in an organization.
      28 (10.9)8 (28.6)20 (71.4)21(75.0)7 (25.0)
       Self-employed
      Self-employed is the act of generating one’s income directly from customers through working, clients or other organizations as opposed to being an employee of a business (or person).
      18 (7.0)5 (27.8)13 (72.2)11 (61.1)7 (38.9)
       Preschool children (age <7 years)18 (7.0)018 (100)2 (11.1)16 (88.9)
       Civil servants
      Civil servants means a person who works for the government or in the civil service.
      4 (1.6)1 (25.0)3 (75.0)1 (25.0)3 (75.0)
       Students4 (1.6)04 (100)1 (2.0)3 (75.0)
      Smoking (yes)38 (14.8)14 (36.8)24 (63.2)0.97420 (52.6)18 (47.4)0.068
      Drinking (yes)15 (5.8)7 (46.7)8 (53.3)0.7919 (60.0)6 (40.0)0.636
      * Chi-squared tests were used for testing the difference in categorical variables, and Wilcoxon rank-sum tests were used for examining the differences in age data.
      a Enterprise staff means a person who works for an enterprise organization.
      b Peasant means a farmer or agricultural worker of low status.
      c Worker means an employee in an organization.
      d Self-employed is the act of generating one’s income directly from customers through working, clients or other organizations as opposed to being an employee of a business (or person).
      e Civil servants means a person who works for the government or in the civil service.

      Poultry exposure history of confirmed cases

      As shown in Table 2, 133 cases (52.0%) reported direct exposure to live poultry in the two weeks preceding illness onset, and the proportions of patients raising backyard poultry, buying live poultry from LPMs, and engaged in selling or slaughtering live poultry were 56.4% (75/133), 30.8% (41/133), and 12.8% (17/133), respectively. A total of 109 patients (42.6%) reported indirect exposure to live poultry, and there was only 1 case reporting close contact exposure (staying with a confirmed H7N9 case in a ward). Among the total 256 cases, 22 (8.6%) had a history of exposure to sick and dead poultry. Additionally, the poultry exposure differences between fatal and nonfatal cases and in patients admitted and not admitted to ICUs were not significant (P> 0.05).
      Table 2Poultry exposure history of confirmed cases.
      VariableTotal (N = 256)Death, n (%)Admission to ICUs, n (%)
      Fatal (N = 100)Nonfatal (N = 156)P-value
      Chi-squared tests were used for testing the difference in categorical variables.
      Yes (n = 168)No (n = 88)P-value
      Chi-squared tests were used for testing the difference in categorical variables.
      Exposure history
      The exposure history means exposures to live poultry in the two weeks preceding illness onset, including direct exposure to live poultry, indirect exposure to live poultry, close contact exposure, and unknown.
      0.4660.882
      Direct exposure to live poultry133 (52.0)47 (35.3)86 (64.7)86 (64.7)47 (35.3)
       Raising backyard poultry75 (75/133)27 (36.0)48 (64.0)46 (61.3)29 (38.7)
       Buying live poultry from LPMs41 (41/133)15 (36.6)26 (63.4)27 (65.9)14 (34.1)
       Engaged in selling or slaughtering live poultry17 (17/133)5 (29.4)12 (70.6)13 (76.5)4 (23.5)
      Indirect exposure to live poultry109 (42.6)48 (44.0)61 (56.0)72 (66.1)37 (33.9)
       Visiting LPMs at least once99 (99/109)44 (44.4)55 (55.6)64 (64.6)35 (35.4)
       Contacting with fresh slaughtered poultry bought from LPMs4 (4/109)2 (50.0)2 (50.0)2 (50.0)2 (50.0)
       Having neighbors raising backyard poultry3 (3/109)03 (100)3 (100)0
       Passing by the live poultry slaughtering area of a restaurant2 (2/109)1 (50.0)1 (50.0)2 (100)0
       Having a household member engaged in selling or slaughtering live poultry1 (1/109)1 (100)01 (100)0
      Close contact exposure1 (0.4)01 (100)1 (100)0
       Staying with a confirmed H7N9 case in a ward1 (1/1)01 (100)1 (100)0
      Unknown13 (5.1)5 (38.5)8 (61.5)9 (69.2)4 (30.8)
      Exposure to sick and dead poultry
       No234 (91.4)90 (38.5)144 (61.5)0.520153 (65.4)81 (34.6)0.792
       Yes22 (8.6)10 (45.5)12 (54.5)15 (68.2)7 (31.8)
      Abbreviations: LPM, live poultry market.
      # The exposure history means exposures to live poultry in the two weeks preceding illness onset, including direct exposure to live poultry, indirect exposure to live poultry, close contact exposure, and unknown.
      * Chi-squared tests were used for testing the difference in categorical variables.

      Clinical characteristics of the confirmed cases

      Figure 2 demonstrated the initial symptoms of the 256 confirmed human influenza A (H7N9) infections. The results showed that the most common initial symptoms were fever (80.2%) and cough (80.2%), and the differences in the frequency of the initial symptoms between fatal and nonfatal groups and in patients admitted and not admitted to ICUs were not significant (P> 0.05). As shown in Table 3, 123 (50.4%) cases had one or more underlying medical conditions, such as chronic cardiovascular disease (28.4%), nutritional and metabolic disease (14.4%), chronic respiratory disease (10.1%), and liver disease (10.1%). The proportions of patients accepting oseltamivir treatment, zanamivir treatment, peramivir treatment, glucocorticoid treatment, and antimicrobial treatment were 83.1% (215/256), 13.6% (35/256), 29.6% (76/256), 40.4% (136/256), and 82.0% (210/256), respectively. The distribution differences in the underlying medical conditions, complications, and treatments between fatal and nonfatal groups were not statistically significant (P> 0.05), and the differences in the distribution of oseltamivir treatment by ICU admission (yes vs. no) were statistically significant (P< 0.05).
      Figure 2
      Figure 2The initial symptoms of 256 confirmed cases of A (H7N9) virus infection in Guangdong Province.
      Table 3Clinical characteristics of the confirmed cases.
      VariableTotal (N = 256)Death, n (%)Admission to ICUs, n (%)
      Fatal (n=100)Nonfatal (n=156)P-value
      Chi-squared tests were used for testing the difference in categorical variables.
      Yes (n=168)No (n = 88)P-value
      Chi-squared tests were used for testing the difference in categorical variables.
      Underlying medical conditions (yes)123 (50.4)50 (40.7)73 (59.3)0.67480 (66.9)40 (33.1)0.563
      Any chronic cardiovascular disease73 (28.4)35 (35.0)38 (72.9)0.26346 (63.0)27 (37.00.623
      Any nutritional and metabolic disease37 (14.4)13 (35.1)24 (64.9)0.24223 (62.2)14 (37.8)0.663
      Any chronic respiratory disease26 (10.1)9 (34.6)17 (65.4)0.55418 (69.2)8 (30.0)0.645
      Any liver disease26 (10.1)13 (50.0)13 (50.0)0.33417 (65.4)9 (34.6)0.993
      Any chronic kidney disease11 (4.3)4 (36.4)7 (63.6)0.2018 (72.7)3 (27.3)0.597
      Any hematological disease4 (1.6)2 (50.0)2 (50.0)0.5662 (50.0)2 (50.0)0.517
      Any rheumatic or autoimmune disease3 (1.2)1 (33.3)2 (66.7)0.3241 (33.3)2 (66.7)0.242
      Oseltamivir treatment (yes)215 (83.1)87 (40.5)128 (59.5)0.540149 (69.3)66 (30.7)0.008
      Zanamivir treatment (yes)35 (13.6)14 (40.0)21 (60.0)0.90221 (60.0)14 (40.0)0.458
      Peramivir treatment (yes)76 (29.6)29 (38.2)47 (61.8)0.88847 (61.8)29 (38.2)0.412
      Glucocorticoid treatment (yes)136 (40.4)48 (35.3)88 (64.7)0.43585 (62.5)51 (37.5)0.252
      Antimicrobial treatment (yes)210 (82.0)83 (39.5)127 (60.5)0.950137 (65.2)73 (34.8)0.784
      Mechanical ventilation
      Invasive134 (52.3)55 (41.0)79 (59.0)0.04196 (71.6)38 (28.4)0.793
      Non-invasive
      Non-invasive mechanical ventilation means using continuous positive airway pressure to provide ventilatory support without an invasive artificial airway.
      69 (27.0)29 (42.0)40 (58.0)44 (63.8)25 (36.2)
      Not required53 (20.7)23 (43.4)30 (56.6)37 (69.8)16 (30.2)
      Complications (yes)209 (81.6)83 (39.7)126 (60.3)0.469137 (65.6)72 (34.4)0.958
      Pneumonia203 (60.2)79 (38.9)124 (61.1)0.925134 (66.0)69 (34.0)0.800
      ARDS156 (46.3)62 (39.7)94 (60.3)0.780100 (64.1)56 (35.9)0.522
      Hepatic insufficiency89 (26.4)32 (36.0)57 (64.0)0.45754 (60.7)35 (39.3)0.223
      Renal inadequacy86 (25.5)35 (40.7)51 (59.3)0.70355 (64.0)31 (36.0)0.689
      Heart failure70 (20.8)25 (35.7)45 (64.3)0.50142 (60.0)28 (40.0)0.245
      DIC10 (3.0)4 (40.0)6 (60.0)0.9516 (60.0)4 (40.0)0.702
      Neurological complications14 (4.2)6 (42.9)8 (57.1)0.7659 (64.3)5 (35.7)0.914
      Septic shock75 (22.3)29 (38.7)46 (61.3)0.93351 (68.0)24 (32.0)0.607
      Rhabdomyolysis3 (0.9)1 (33.3)2 (66.7)0.42603 (100)0.016
      Admitted to ICU (yes)168 (65.6)83 (49.4)85(50.6)<0.001NANANA
      Abbreviations: ARDS, acute respiratory distress syndrome; DIC, disseminated intravascular coagulation; ICU, intensive care unit; NA, not applicable or no data available.
      # Non-invasive mechanical ventilation means using continuous positive airway pressure to provide ventilatory support without an invasive artificial airway.
      * Chi-squared tests were used for testing the difference in categorical variables.

      Timelines for medical care of the confirmed cases

      Table 4 demonstrates that the median time from the onset of illness to first medical care was 1 day (range, 0 to 7), the median time from the onset of illness to confirmation was 8 days (range, 1 to 27), the median time from onset to hospitalization was 4 days (range, 0 to 13), and the median time from onset to prognosis was 20 days (range, 3 to 145). Among the 215 patients accepting oseltamivir treatment, the median time from the onset of illness to oseltamivir treatment was 5 days (range, 0 to 16); 35 patients received zanamivir treatment after a median of 8 days (range, 0 to 23). Of the 79 patients receiving peramivir treatment, the median time from the onset of illness to peramivir treatment was 7 days (range, 0 to 17). Additionally, the univariable logistic regression models for death demonstrated that time from the onset of illness to oseltamivir treatment (OR = 1.10, 95% CI = 1.01–1.10) and zanamivir treatment (OR = 1.05, 95% CI = 1.02–1.07) were associated with the death of patients with A (H7N9) infection. The univariable logistic regression models for admission to ICUs showed that time from illness onset to confirmation (OR = 1.08, 95% CI = 1.00–1.17) and oseltamivir treatment (OR = 1.18, 95% CI = 1.06–1.31) were positively related to ICU admission.
      Table 4Timelines for medical care of the confirmed cases.
      VariableTotal (N = 256)Death, median (range)Admission to ICUs, median (range)
      Fatal (n = 100)Nonfatal (n = 156)OR (95% CI)
      The univariable logistic regression models without adjusting for other variables were conducted to explore the potential associations of timelines for medical care with clinical severity.
      Yes (n = 168)No (n = 88)OR (95% CI)
      The univariable logistic regression models without adjusting for other variables were conducted to explore the potential associations of timelines for medical care with clinical severity.
      Time from illness onset to first medical care (days)1 (0 ∼ 7)1 (0 ∼ 7)1 (0 ∼ 7)0.95 (0.82–1.11)1 (0 ∼ 7)1 (0 ∼ 7)0.99 (0.86–1.17)
      Time from illness onset to confirmation (days)8 (1 ∼ 27)8 (1 ∼ 21)8 (1 ∼ 27)1.03 (0.96–1.11)8 (1 ∼ 27)8 (1 ∼ 19)1.08 (1.00–1.17)
      P<0.05.
      Time from illness onset to hospitalization (days)4 (0 ∼ 13)4 (0 ∼ 11)4 (0 ∼ 13)1.05 (0.94–1.17)4 (0 ∼ 13)4 (0 ∼ 11)1.06 (0.95–1.19)
      Time from illness onset to oseltamivir treatment (days)5 (0 ∼ 16)5 (0 ∼ 13)5 (0 ∼ 16)1.10 (1.01–1.10)
      P<0.05.
      5 (0 ∼ 13)4 (0 ∼ 16)1.18 (1.06–1.31)
      P<0.05.
      Time from illness onset to zanamivir treatment (days)8 (0 ∼ 23)7 (0 ∼ 19)9 (2 ∼ 23)1.05 (1.02–1.07)
      P<0.05.
      7 (0 ∼ 23)9 (1 ∼ 22)0.96 (0.84–1.09)
      Time from illness onset to peramivir treatment (days)7 (0 ∼ 17)7 (0 ∼ 13)7 (2 ∼ 17)0.99 (0.85–1.17)8 (0 ∼ 17)7 (2 ∼ 12)1.03 (0.87–1.21)
      Time from illness onset to prognosis (days)20 (3 ∼ 145)13 (3 ∼ 89)26 (6 ∼ 145)0.95 (0.93–0.97)
      P<0.05.
      22 (3 ∼ 97)17 (3 ∼ 145)1.01 (0.99–1.03)
      # The univariable logistic regression models without adjusting for other variables were conducted to explore the potential associations of timelines for medical care with clinical severity.
      * P< 0.05.

      Discussion

      To our knowledge, Guangdong province has become one of the main epidemic areas of A (H7N9) from the second to fifth outbreak in China. The fifth epidemic starting on Oct 1, 2016 is reported to be the largest epidemic wave in mainland China so far (
      • Wang X.
      • Jiang H.
      • Wu P.
      • Uyeki T.M.
      • Feng L.
      • Lai S.
      • et al.
      Epidemiology of avian influenza A H7N9 virus in human beings across five epidemics in mainland China, 2013–17: an epidemiological study of laboratory-confirmed case series.
      ), but this study demonstrated the largest number of human cases in Guangdong was during the second wave. Overall, after the second wave, the number of human cases in both mainland China and Guangdong province reduced during the third and fourth wave (
      • Wang X.
      • Jiang H.
      • Wu P.
      • Uyeki T.M.
      • Feng L.
      • Lai S.
      • et al.
      Epidemiology of avian influenza A H7N9 virus in human beings across five epidemics in mainland China, 2013–17: an epidemiological study of laboratory-confirmed case series.
      ). But in the fifth epidemic wave, the most rigorous control measures in history have been taken in Guangdong province since January 2017, closing all LPMs for at least 3 days once the H7N9-positive LPMs or patients with A (H7N9) infection are detected in the city. Meanwhile, regarding other regions of China, Sichuan and Chongqing provinces were the new provinces reporting confirmed A (H7N9) cases in the fifth wave (as of March 2017). This study was conducted to add the information about demographics and clinical characteristics of patients with A (H7N9) infection in Guangdong province, and the results revealed the fatality rate was 39.0%, which is in accordance with prior studies conducted in China (
      • Ji H.
      • Gu Q.
      • Chen L.L.
      • Xu K.
      • Ling X.
      • Bao C.J.
      • et al.
      Epidemiological and clinical characteristics and risk factors for death of patients with avian influenza A H7N9 virus infection from Jiangsu Province, Eastern China.
      ,
      • Gao H.N.
      • Lu H.Z.
      • Cao B.
      • Du B.
      • Shang H.
      • Gan J.H.
      • et al.
      Clinical findings in 111 cases of influenza A (H7N9) virus infection.
      ,
      • Xiang N.
      • Li X.
      • Ren R.
      • Wang D.
      • Zhou S.
      • Greene C.M.
      • et al.
      Assessing change in avian influenza A(H7N9) virus infections during the fourth epidemic – China, September 2015-August 2016.
      ), and which is much higher than that of seasonal influenza infections (
      • World Health Organization
      Avian and other zoonotic influenza.
      ). Although the Guangdong CDC has conducted the monthly risk assessment of human infections with A (H7N9) and the weekly environmental surveillance of avian H7N9 virus during November to May since 2013, our results also showed that there were seasonal fluctuations in the number of cases, and most A (H7N9) infections in humans occurred during December to March. Similarly, previous studies reported that the A (H7N9) epidemic oscillated seasonally with peaks in China’s spring and winter (
      • Xiang N.
      • Li X.
      • Ren R.
      • Wang D.
      • Zhou S.
      • Greene C.M.
      • et al.
      Assessing change in avian influenza A(H7N9) virus infections during the fourth epidemic – China, September 2015-August 2016.
      ,
      • Lin Q.
      • Lin Z.
      • Chiu A.P.
      • He D.
      Seasonality of influenza A(H7N9) virus in China-fitting simple epidemic models to human cases.
      ).
      In addition, the current study demonstrated that the male-to-female ratio of patients was approximately 2.0:1, and this pattern is consistent with results of previous studies (
      • Sun J.
      • Gong Z.
      • Lv H.
      • Chen Z.
      • Chai C.
      • Liu S.
      • et al.
      Comparison of characteristics between patients with H7N9 living in rural and urban areas of Zhejiang Province, China: a preliminary report.
      ,
      • Chen E.
      • Chen Y.
      • Fu L.
      • Chen Z.
      • Gong Z.
      • Mao H.
      • et al.
      Human infection with avian influenza A(H7N9) virus re-emerges in China in winter 2013.
      ). An explanation for these results is that Chinese men are more likely to engage in selling or slaughtering live poultry than women (
      • Cowling B.J.
      • Jin L.
      • Lau E.H.
      • Liao Q.
      • Wu P.
      • Jiang H.
      • et al.
      Comparative epidemiology of human infections with avian influenza A H7N9 and H5N1 viruses in China: a population-based study of laboratory-confirmed cases.
      ); another explanation may be that most Asian males have bird-related hobbies, and they probably visit LPMs frequently and keep pet birds at home (
      • Ge E.
      • Zhang R.
      • Li D.
      • Wei X.
      • Wang X.
      • Lai P.C.
      Estimating risks of inapparent avian exposure for human infection: avian influenza virus A (H7N9) in Zhejiang Province, China.
      ); all aforementioned behaviors might have increased the potential of unapparent exposure to A (H7N9) infected avian and avian-related environment in males. Furthermore, this study demonstrated that the median age of patients with A (H7N9) infection was 56 (range, 1 to 88) years, most cases occurred in the 60- to 79-year-old group, and the proportion of fatal cases was higher than nonfatal cases in patients more than 60 years old. These results are consistent with those prior studies which found that the elderly have an increased risk for A (H7N9) infections (
      • Zhang Y.
      • Yu Y.S.
      • Tang Z.H.
      • Chen X.H.
      • Zang G.Q.
      Human infection with avian influenza A (H7N9) virus.
      ,
      • Wang H.
      • Xiao X.
      • Lu J.
      • Chen Z.
      • Li K.
      • Liu H.
      • et al.
      Factors associated with clinical outcome in 25 patients with avian influenza A (H7N9) infection in Guangzhou, China.
      ). In contrast, influenza A (H5N1) has infected mainly younger people, but the mechanism for the different age distributions of humans with A (H5N1) and A (H7N9) infections is still unclear. Previous studies considered that these findings may result from the fact that the elderly patients (mostly retired) have more opportunities to be exposed to the environments that are contaminated with H7N9 virus (e.g., LPMs) (
      • Gao H.N.
      • Lu H.Z.
      • Cao B.
      • Du B.
      • Shang H.
      • Gan J.H.
      • et al.
      Clinical findings in 111 cases of influenza A (H7N9) virus infection.
      ), and are more susceptible to contracting avian flu for personal medical conditions (
      • Qin Y.
      • Horby P.W.
      • Tsang T.K.
      • Chen E.
      • Gao L.
      • Ou J.
      • et al.
      Differences in the epidemiology of human cases of avian influenza A(H7N9) and A(H5N1) viruses infection.
      ). In line with previous studies (
      • Zhou L.
      • Ren R.
      • Yang L.
      • Bao C.
      • Wu J.
      • Wang D.
      • et al.
      Sudden increase in human infection with avian influenza A(H7N9) virus in China, September–December 2016.
      ,
      • Zhang W.
      • Wang L.
      • Hu W.
      • Ding F.
      • Sun H.
      • Li S.
      • et al.
      Epidemiologic characteristics of cases for influenza A(H7N9) virus infections in China.
      ), our results also showed that most of the infected patients were retired, peasants, and unemployed; these results might be related to the fact that retired and unemployed people are more probable to shop (at LPMs) and cook for their families, and peasants are more likely to be involved in raising poultry in their back yards or to have neighbors raising live poultry (
      • Khokhar Shahid G.
      • Min Qingfei
      • Su Chao
      Bird flu (H7N9) outbreak and its implications on the supply chain of poultry meat in China.
      ). Moreover, this study revealed that 52.0% of patients with A (H7N9) infection reported direct exposure to live poultry (including being engaged in selling or slaughtering live poultry, raising backyard poultry, and buying live poultry from LPMs), and 42.6% had a history of indirect exposure to live poultry. Similarly, Li et al. analyzed the data of H7N9 cases in China identified as of December 1, 2013, and also showed that 77% of H7N9-infected patients had a history of exposure to living animals, including chickens (76%) (
      • Li Q.
      • Zhou L.
      • Zhou M.
      • Chen Z.
      • Li F.
      • Wu H.
      • et al.
      Epidemiology of human infections with avian influenza A(H7N9) virus in China.
      ). Kang et al. using the data obtained from January 2013 to June 2014 in Guangdong and also found that 94.5% (103/109) of confirmed human H7N9 cases reported a history of recent exposure to poultry (
      • Kang M.
      • He J.
      • Song T.
      • Rutherford S.
      • Wu J.
      • Lin J.
      • et al.
      Environmental sampling for avian influenza A(H7N9) in live-poultry markets in Guangdong, China.
      ). It is well known that exposure to live poultry may be a risk factor for H7N9 infection, and the closure of LPMs has been reported to be effective in controlling the outbreak of human infections with (H7N9) (
      • Wu J.
      • Lu J.
      • Faria N.R.
      • Zeng X.
      • Song Y.
      • Zou L.
      • et al.
      Effect of live poultry market interventions on influenza A(H7N9) virus, Guangdong, China.
      ,
      • Yu H.
      • Wu J.T.
      • Cowling B.J.
      • Liao Q.
      • Fang V.J.
      • Zhou S.
      • et al.
      Effect of closure of live poultry markets on poultry-to-person transmission of avian influenza A H7N9 virus: an ecological study.
      ,
      • Lu J.
      • Liu W.
      • Xia R.
      • Dai Q.
      • Bao C.
      • Tang F.
      • et al.
      Effects of closing and reopening live poultry markets on the epidemic of human infection with avian influenza A virus.
      ). However, buying live poultry is the Chinese traditional diet culture, and closing LPMs permanently can cause an economic burden on individuals and society (
      • Zhou P.
      • Ma J.
      • Lai A.
      • Gray G.C.
      • Su S.
      • Li S.
      Avian influenza A(H7N9) virus and mixed live poultry-animal markets in Guangdong province: a perfect storm in the making?.
      ,
      • Fournie G.
      • Pfeiffer D.U.
      Can closure of live poultry markets halt the spread of H7N9?.
      ). In the long term, we should strengthen the surveillance of LPMs to detect the avain H7N9 virus early, enhance cooperation with the related department of agriculture to improve the standardized management of LPMs, and focus on the groups with high frequency exposure to live poultry (e.g., poultry-related workers and individuals raising backyard poultry) to prevent disease spread.
      The most common initial symptoms among the confirmed cases observed in this study were fever (80.2%) and cough (80.2%), followed by fatigue (33.9%); these results are consistent with previous studies (
      • Gao R.
      • Cao B.
      • Hu Y.
      • Feng Z.
      • Wang D.
      • Hu W.
      • et al.
      Human infection with a novel avian-origin influenza A (H7N9) virus.
      ,
      • Gong Z.
      • Lv H.
      • Ding H.
      • Han J.
      • Sun J.
      • Chai C.
      • et al.
      Epidemiology of the avian influenza A (H7N9) outbreak in Zhejiang Province, China.
      ). Although Shinde et al. reported that fever, sore throat, and vomiting were significantly more common in fatal A (H5N1) cases than in nonfatal cases (
      • Shinde V.
      • Hanshaoworakul W.
      • Simmerman J.M.
      • Narueponjirakul U.
      • Sanasuttipun W.
      • Kaewchana S.
      • et al.
      A comparison of clinical and epidemiological characteristics of fatal human infections with H5N1 and human influenza viruses in Thailand, 2004–2006.
      ), our study did not find the significant differences in the frequency of the initial symptoms between fatal and nonfatal groups; similar results can be found in Wang’s study about A (H7N9) cases in Guangzhou city (
      • Wang H.
      • Xiao X.
      • Lu J.
      • Chen Z.
      • Li K.
      • Liu H.
      • et al.
      Factors associated with clinical outcome in 25 patients with avian influenza A (H7N9) infection in Guangzhou, China.
      ). Several previous studies considered that the coexisting underlying medical conditions might impair the host’s responses to A (H7N9) infections (
      • Gao R.
      • Cao B.
      • Hu Y.
      • Feng Z.
      • Wang D.
      • Hu W.
      • et al.
      Human infection with a novel avian-origin influenza A (H7N9) virus.
      ,
      • Guan Y.
      • Farooqui A.
      • Zhu H.
      • Dong W.
      • Wang J.
      • Kelvin D.J.
      H7N9 incident, immune status, the elderly and a warning of an influenza pandemic.
      ). In this study, our results showed that 50.4% of patients with medical history data had at least one underlying medical condition (such as chronic cardiovascular disease, nutritional and metabolic disease, and chronic respiratory disease). Similarly, Gao et al. (
      • Gao H.N.
      • Lu H.Z.
      • Cao B.
      • Du B.
      • Shang H.
      • Gan J.H.
      • et al.
      Clinical findings in 111 cases of influenza A (H7N9) virus infection.
      ) also reported that a total of 61.3% (68/111) of the patients with confirmed A (H7N9) infection had one or more coexisting medical conditions. Consistent with previous studies (
      • Gao H.N.
      • Lu H.Z.
      • Cao B.
      • Du B.
      • Shang H.
      • Gan J.H.
      • et al.
      Clinical findings in 111 cases of influenza A (H7N9) virus infection.
      ,
      • Wang X.
      • Jiang H.
      • Wu P.
      • Uyeki T.M.
      • Feng L.
      • Lai S.
      • et al.
      Epidemiology of avian influenza A H7N9 virus in human beings across five epidemics in mainland China, 2013–17: an epidemiological study of laboratory-confirmed case series.
      ,
      • Xiang N.
      • Li X.
      • Ren R.
      • Wang D.
      • Zhou S.
      • Greene C.M.
      • et al.
      Assessing change in avian influenza A(H7N9) virus infections during the fourth epidemic – China, September 2015-August 2016.
      ,
      • Chen X.
      • Yang Z.
      • Lu Y.
      • Xu Q.
      • Wang Q.
      • Chen L.
      Clinical features and factors associated with outcomes of patients infected with a Novel Influenza A (H7N9) virus: a preliminary study.
      ), our study revealed that approximately 81.6% A (H7N9) infected patients experienced one or more complications, and 65.6% were admitted to an ICU. Although this study did not find significant differences in the underlying medical conditions and complications between fatal and nonfatal cases and in patients admitted and not admitted to the ICU, our findings may help clinicians detect patients with suspected A (H7N9) infections earlier and tailor appropriate treatments for patients to prevent a potential epidemic.
      Despite the fact that the use of corticosteroids in patients with A (H7N9) infections is not recommended according to WHO guidelines (
      • World Health Organization
      Avian and other zoonotic influenza.
      ), this study revealed that 136 (40.4%) patients received glucocorticoid treatment; these findings suggested that the standardized treatments of A (H7N9) infections still need be strengthened in Guangdong province to correspond to WHO guidelines. Moreover, the American Public Health Association has suggested that some antiviral drugs (notably oseltamivir) can reduce the duration of viral replication and improve prospects of survival (
      • APHA
      Control of communicable diseases manual.
      ), and the current study revealed that the proportions of patients accepting oseltamivir treatment, zanamivir treatment, and peramivir treatment were 83.1%, 13.6%, and 29.6%, respectively; the median time from the onset of illness to oseltamivir treatment was 5 days (range, 0 to 16), to zanamivir treatment was 8 days (range, 0 to 23), and to peramivir treatment was 7 days (range, 0 to 17). Consistent with previous studies related to A (H7N9) infections (
      • Gao H.N.
      • Lu H.Z.
      • Cao B.
      • Du B.
      • Shang H.
      • Gan J.H.
      • et al.
      Clinical findings in 111 cases of influenza A (H7N9) virus infection.
      ,
      • Chen X.
      • Yang Z.
      • Lu Y.
      • Xu Q.
      • Wang Q.
      • Chen L.
      Clinical features and factors associated with outcomes of patients infected with a Novel Influenza A (H7N9) virus: a preliminary study.
      ), this study revealed some patients have indeed received more than one type of antiviral drug treatment during the period from illness onset to prognosis, and most patients received oseltamivir treatment. It might be due to the fact that oseltamivir is an orally administered antiviral medication (
      • Davies B.E.
      Pharmacokinetics of oseltamivir: an oral antiviral for the treatment and prophylaxis of influenza in diverse populations.
      ), zanamivir (powder) is an orally inhaled antiviral, and peramivir is an intravenously administered antiviral (
      • Alame M.M.
      • Massaad E.
      • Peramivir Zaraket H.
      A novel intravenous neuraminidase inhibitor for treatment of acute influenza infections.
      ); where oseltamivir is unavailable or cannot be used for any reason (e.g., patients who cannot tolerate or absorb oral or enterically administered oseltamivir), the use of inhaled zanamivir and intravenous peramivir should be considered (
      • World Health Organization
      Recommended use of antivirals.
      ,
      • Centers for Disease Control and Prevention
      Influenza antiviral medications: summary for clinicians.
      ). In this study, the results also found that in patients accepting oseltamivir treatment, the proportion of cases admitted to ICUs (69.3%) was significantly higher than those not admitted (30.7%), and these results might be related to a tendency of higher prescription rates in critically ill patients (
      • Herman M.
      • Smieja M.
      • Carruthers S.
      • Loeb M.
      Oseltamivir use amongst hospitalized patients infected with influenza.
      ).
      Although the definitive pathogenic mechanism of the A(H7N9) virus in humans is still unclear, the Chinese government has enhanced measures to strengthen early diagnosis, early antiviral treatment, and treatment of severe cases to reduce occurrence of severe cases and deaths (
      • World Health Organization
      Human infection with avian influenza A(H7N9) virus – China.
      ), and the WHO guidelines recommend that in suspected A (H7N9) infected patients, oseltamivir should be prescribed as soon as possible (ideally, within 48 h following symptom onset) to maximize its therapeutic benefit (
      • World Health Organization
      Avian and other zoonotic influenza.
      ). Correspondingly, our results also demonstrated that time from the onset of illness to oseltamivir treatment and zanamivir treatment were positively associated with the death of patients with A (H7N9) infection, suggesting earlier oseltamivir and zanamivir treatment were negatively associated with the risk of fatality. Moreover, our study also found that time from illness onset to confirmation of A (H7N9) infection and oseltamivir treatment were positively related to ICU admission, suggesting earlier confirmation and oseltamivir treatment were associated with a reduced risk of ICU admission. Similarly, Chan et al. reported that oseltamivir is especially effective for treating H5N1 infection when given early (
      • Chan P.K.
      • Lee N.
      • Zaman M.
      • Adisasmito W.
      • Coker R.
      • Hanshaoworakul W.
      • et al.
      Determinants of antiviral effectiveness in influenza virus A subtype H5N1.
      ), Fry et al., using a randomised placebo-controlled trial, found the duration of influenza illness symptoms was shorter in the osletamivir treatment group than in the placebo group (
      • Fry A.M.
      • Goswami D.
      • Nahar K.
      • Sharmin A.T.
      • Rahman M.
      • Gubareva L.
      • et al.
      Efficacy of oseltamivir treatment started within 5 days of symptom onset to reduce influenza illness duration and virus shedding in an urban setting in Bangladesh: a randomised placebo-controlled trial.
      ), Wu et al. found that delayed confirmation and antivirus therapy were associated with death caused by A (H7N9) infection (
      • Wu Z.Q.
      • Zhang Y.
      • Zhao N.
      • Yu Z.
      • Pan H.
      • Chan T.C.
      • et al.
      Comparative epidemiology of human fatal infections with novel, high (H5N6 and H5N1) and low (H7N9 and H9N2) pathogenicity avian influenza A viruses.
      ).
      Moreover, in the present study, the median time from the onset of illness to first medical care was 1 day (range, 0 to 7), which is similar to those previous reports (
      • Wang H.
      • Xiao X.
      • Lu J.
      • Chen Z.
      • Li K.
      • Liu H.
      • et al.
      Factors associated with clinical outcome in 25 patients with avian influenza A (H7N9) infection in Guangzhou, China.
      ,
      • Li Q.
      • Zhou L.
      • Zhou M.
      • Chen Z.
      • Li F.
      • Wu H.
      • et al.
      Epidemiology of human infections with avian influenza A(H7N9) virus in China.
      ); the median time from onset to prognosis was 20 days (range, 3 to 145), which is longer than the previous study in Guangzhou (10 days, range 2 to 70) and the national case report in 2013 (14 days, range 6 to 58) (
      • Gao H.N.
      • Lu H.Z.
      • Cao B.
      • Du B.
      • Shang H.
      • Gan J.H.
      • et al.
      Clinical findings in 111 cases of influenza A (H7N9) virus infection.
      ). One explanation for these results might be that patients with A (H7N9) infection in Guangdong province had longer periods of hospitalization and rehabilitation.
      Several study limitations should be noted when interpreting the results. First, the data are retrospective. Therefore, the possibility of recall bias cannot be completely ruled out. Second, we have included data for most but not all laboratory data (e.g., lymphocytopenia and aspartate aminotransferase). Third, because of the observed nature of the data, no cause inference can be made regarding the observed associations of the timing of antiviral treatment on clinical outcomes.

      Conclusions

      In summary, our study comprehensively estimated the epidemiological and clinical characteristics of patients with A (H7N9) infection in Guangdong province, and provided some clues about treatments reducing the clinical severity. In this study, A (H7N9) infection was more common in male, elderly, retired, peasant class, and unemployed patients, and patients with high-frequency exposure to live poultry; the fatality rate and the proportion of ICU admissions were higher; and earlier oseltamivir treatment was associated with a decreased risk of fatality and ICU admission. To minimize the public health risk of human A (H7N9) infection, preventive measures are suggested to focus on high-risk populations (such as the elderly and men), and earlier guideline-based antiviral treatment was recommended to reduce clinical severity.

      Conflict of interest

      The authors have declared that no competing interests exist.

      Contributions

      Yuwei Yang, Min Kang, and Haojie Zhong searched the literature, conceived the study, designed the study, analyzed the data, interpreted the results, and drafted the report. Yuwei Yang and Min Kang organized the study, collected the data and analyzed the data. Lan Guo analyzed the data. Haojie Zhong, Tie Song, Jianfeng He, Xiaohua Tan, and Guofeng Huang collected the data, interpreted the results, and obtained funding.

      Funding

      This study was supported by Chinese Center for Disease Control and Prevention and Guangdong Provincial Health and Family Planning Commission. This study was also supported by grants from Science and Technology Program of Guangdong, China (No. 2013B020307007).

      Role of the sponsors

      The sponsors had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

      Ethical approval

      The study was approved by the Guangdong CDC Institutional Review Board. All the participants were fully informed of the purpose of the survey, and were informed A (H7N9) virus infection in humans included in statutory infectious disease surveillance reporting system. The institutional review boards of the sampled hospitals also approved the review process and waived the requirement to obtain patients’ written informed consent.

      Statement

      No data has been published before.

      Acknowledgements

      We gratefully acknowledge the assistance of 21 prefecture-level CDCs and related hospitals in collecting epidemiological information and medical records. In addition, we express our thanks to all the participants in the study, without whom the study would not have been possible.

      References

        • APHA
        Control of communicable diseases manual.
        20th ed. APHA Press, Washington DC2015
        • Alame M.M.
        • Massaad E.
        • Peramivir Zaraket H.
        A novel intravenous neuraminidase inhibitor for treatment of acute influenza infections.
        Front Microbiol. 2016; 7: 450
        • Centers for Disease Control and Prevention
        Influenza antiviral medications: summary for clinicians.
        U.S. Department of Health and Human Services, 2017
        • Chan P.K.
        • Lee N.
        • Zaman M.
        • Adisasmito W.
        • Coker R.
        • Hanshaoworakul W.
        • et al.
        Determinants of antiviral effectiveness in influenza virus A subtype H5N1.
        J Infect Dis. 2012; 206: 1359-1366
        • Chen Y.
        • Liang W.
        • Yang S.
        • Wu N.
        • Gao H.
        • Sheng J.
        • et al.
        Human infections with the emerging avian influenza A H7N9 virus from wet market poultry: clinical analysis and characterisation of viral genome.
        Lancet. 2013; 381: 1916-1925
        • Chen X.
        • Yang Z.
        • Lu Y.
        • Xu Q.
        • Wang Q.
        • Chen L.
        Clinical features and factors associated with outcomes of patients infected with a Novel Influenza A (H7N9) virus: a preliminary study.
        PLoS One. 2013; 8: e73362
        • Chen E.
        • Chen Y.
        • Fu L.
        • Chen Z.
        • Gong Z.
        • Mao H.
        • et al.
        Human infection with avian influenza A(H7N9) virus re-emerges in China in winter 2013.
        Euro Surveill. 2013; : 18
        • Cowling B.J.
        • Jin L.
        • Lau E.H.
        • Liao Q.
        • Wu P.
        • Jiang H.
        • et al.
        Comparative epidemiology of human infections with avian influenza A H7N9 and H5N1 viruses in China: a population-based study of laboratory-confirmed cases.
        Lancet. 2013; 382: 129-137
        • Davies B.E.
        Pharmacokinetics of oseltamivir: an oral antiviral for the treatment and prophylaxis of influenza in diverse populations.
        J Antimicrob Chemother. 2010; 65: i5-10
        • Fournie G.
        • Pfeiffer D.U.
        Can closure of live poultry markets halt the spread of H7N9?.
        Lancet. 2014; 383: 496-497
        • Fry A.M.
        • Goswami D.
        • Nahar K.
        • Sharmin A.T.
        • Rahman M.
        • Gubareva L.
        • et al.
        Efficacy of oseltamivir treatment started within 5 days of symptom onset to reduce influenza illness duration and virus shedding in an urban setting in Bangladesh: a randomised placebo-controlled trial.
        Lancet Infect Dis. 2014; 14: 109-118
        • Gao R.
        • Cao B.
        • Hu Y.
        • Feng Z.
        • Wang D.
        • Hu W.
        • et al.
        Human infection with a novel avian-origin influenza A (H7N9) virus.
        N Engl J Med. 2013; 368: 1888-1897
        • Gao H.N.
        • Lu H.Z.
        • Cao B.
        • Du B.
        • Shang H.
        • Gan J.H.
        • et al.
        Clinical findings in 111 cases of influenza A (H7N9) virus infection.
        N Engl J Med. 2013; 368: 2277-2285
        • Ge E.
        • Zhang R.
        • Li D.
        • Wei X.
        • Wang X.
        • Lai P.C.
        Estimating risks of inapparent avian exposure for human infection: avian influenza virus A (H7N9) in Zhejiang Province, China.
        Sci Rep. 2017; 7: 40016
        • Gong Z.
        • Lv H.
        • Ding H.
        • Han J.
        • Sun J.
        • Chai C.
        • et al.
        Epidemiology of the avian influenza A (H7N9) outbreak in Zhejiang Province, China.
        BMC Infect Dis. 2014; 14: 244
        • Guan Y.
        • Farooqui A.
        • Zhu H.
        • Dong W.
        • Wang J.
        • Kelvin D.J.
        H7N9 incident, immune status, the elderly and a warning of an influenza pandemic.
        J Infect Dev Ctries. 2013; 7: 302-307
        • Herman M.
        • Smieja M.
        • Carruthers S.
        • Loeb M.
        Oseltamivir use amongst hospitalized patients infected with influenza.
        Influenza Other Respir Viruses. 2014; 8: 547-548
        • Ji H.
        • Gu Q.
        • Chen L.L.
        • Xu K.
        • Ling X.
        • Bao C.J.
        • et al.
        Epidemiological and clinical characteristics and risk factors for death of patients with avian influenza A H7N9 virus infection from Jiangsu Province, Eastern China.
        PLoS One. 2014; 9: e89581
        • Kang M.
        • He J.
        • Song T.
        • Rutherford S.
        • Wu J.
        • Lin J.
        • et al.
        Environmental sampling for avian influenza A(H7N9) in live-poultry markets in Guangdong, China.
        PLoS One. 2015; 10: e126335
        • Khokhar Shahid G.
        • Min Qingfei
        • Su Chao
        Bird flu (H7N9) outbreak and its implications on the supply chain of poultry meat in China.
        J Appl Poult Res. 2015; 2: 215-221
        • Lam T.T.
        • Zhou B.
        • Wang J.
        • Chai Y.
        • Shen Y.
        • Chen X.
        • Zhu H.
        • et al.
        Dissemination, divergence and establishment of H7N9 influenza viruses in China.
        Nature. 2015; 522: 102-105
        • Leung Y.
        • To M.
        • Lam T.
        • Yau S.
        • Leung O.
        • Chuang S.
        Epidemiology of human influenza A(H7N9) infection in Hong Kong.
        J Microbiol Immunol Infect. 2017; 50: 183-188
        • Li Q.
        • Zhou L.
        • Zhou M.
        • Chen Z.
        • Li F.
        • Wu H.
        • et al.
        Epidemiology of human infections with avian influenza A(H7N9) virus in China.
        N Engl J Med. 2014; 370: 520-532
        • Li J.
        • Chen J.
        • Yang G.
        • Zheng Y.X.
        • Mao S.H.
        • Zhu W.P.
        • et al.
        Case-control study of risk factors for human infection with avian influenza A(H7N9) virus in Shanghai, China, 2013.
        Epidemiol Infect. 2015; 143: 1826-1832
        • Lin Q.
        • Lin Z.
        • Chiu A.P.
        • He D.
        Seasonality of influenza A(H7N9) virus in China-fitting simple epidemic models to human cases.
        PLoS One. 2016; 11: e151333
        • Liu B.
        • Havers F.
        • Chen E.
        • Yuan Z.
        • Yuan H.
        • Ou J.
        • et al.
        Risk factors for influenza A(H7N9) disease–China, 2013.
        Clin Infect Dis. 2014; 59: 787-794
        • Lu J.
        • Liu W.
        • Xia R.
        • Dai Q.
        • Bao C.
        • Tang F.
        • et al.
        Effects of closing and reopening live poultry markets on the epidemic of human infection with avian influenza A virus.
        J Biomed Res. 2016; 30: 112-119
        • Qi X.
        • Jiang D.
        • Wang H.
        • Zhuang D.
        • Ma J.
        • Fu J.
        • et al.
        Calculating the burden of disease of avian-origin H7N9 infections in China.
        BMJ Open. 2014; 4: e4189
        • Qin Y.
        • Horby P.W.
        • Tsang T.K.
        • Chen E.
        • Gao L.
        • Ou J.
        • et al.
        Differences in the epidemiology of human cases of avian influenza A(H7N9) and A(H5N1) viruses infection.
        Clin Infect Dis. 2015; 61: 563-571
        • Shinde V.
        • Hanshaoworakul W.
        • Simmerman J.M.
        • Narueponjirakul U.
        • Sanasuttipun W.
        • Kaewchana S.
        • et al.
        A comparison of clinical and epidemiological characteristics of fatal human infections with H5N1 and human influenza viruses in Thailand, 2004–2006.
        PLoS One. 2011; 6: e14809
        • Sun J.
        • Gong Z.
        • Lv H.
        • Chen Z.
        • Chai C.
        • Liu S.
        • et al.
        Comparison of characteristics between patients with H7N9 living in rural and urban areas of Zhejiang Province, China: a preliminary report.
        PLoS One. 2014; 9: e93775
        • Wang C.
        • Yu H.
        • Horby P.W.
        • Cao B.
        • Wu P.
        • Yang S.
        • et al.
        Comparison of patients hospitalized with influenza A subtypes H7N9, H5N1, and 2009 pandemic H1N1.
        Clin Infect Dis. 2014; 58: 1095-1103
        • Wang Z.
        • Wan Y.
        • Qiu C.
        • Quinones-Parra S.
        • Zhu Z.
        • Loh L.
        • Tian D.
        • et al.
        Recovery from severe H7N9 disease is associated with diverse response mechanisms dominated by CD8(+) T cells.
        Nat Commun. 2015; 6: 6833
        • Wang H.
        • Xiao X.
        • Lu J.
        • Chen Z.
        • Li K.
        • Liu H.
        • et al.
        Factors associated with clinical outcome in 25 patients with avian influenza A (H7N9) infection in Guangzhou, China.
        BMC Infect Dis. 2016; : 16
        • Wang X.
        • Jiang H.
        • Wu P.
        • Uyeki T.M.
        • Feng L.
        • Lai S.
        • et al.
        Epidemiology of avian influenza A H7N9 virus in human beings across five epidemics in mainland China, 2013–17: an epidemiological study of laboratory-confirmed case series.
        Lancet Infect Dis. 2017; 17: 822-832
        • World Health Organization
        Recommended use of antivirals.
        World Health Organization, Geneva2009
        • World Health Organization
        Real-time RT-PCR protocol for the detection of avian influenza A(H7N9) virus.
        WHO Collaborating Center for Reference and Research on Influenza at the Chinese National Influenza Center, Beijing2013
        • World Health Organization
        Avian and other zoonotic influenza.
        World Health Organization, Geneva2016
        • World Health Organization
        Human infection with avian influenza A(H7N9) virus – China.
        World Health Organization, 2017
        • Wu J.
        • Lu J.
        • Faria N.R.
        • Zeng X.
        • Song Y.
        • Zou L.
        • et al.
        Effect of live poultry market interventions on influenza A(H7N9) virus, Guangdong, China.
        Emerg Infect Dis. 2016; 22: 2104-2112
        • Wu Z.Q.
        • Zhang Y.
        • Zhao N.
        • Yu Z.
        • Pan H.
        • Chan T.C.
        • et al.
        Comparative epidemiology of human fatal infections with novel, high (H5N6 and H5N1) and low (H7N9 and H9N2) pathogenicity avian influenza A viruses.
        Int J Environ Res Public Health. 2017; : 14
        • Xiang N.
        • Li X.
        • Ren R.
        • Wang D.
        • Zhou S.
        • Greene C.M.
        • et al.
        Assessing change in avian influenza A(H7N9) virus infections during the fourth epidemic – China, September 2015-August 2016.
        MMWR Morb Mortal Wkly Rep. 2016; 65: 1390-1394
        • Yang Z.F.
        • He J.F.
        • Li X.B.
        • Guan W.D.
        • Ke C.W.
        • Wu S.G.
        • et al.
        Epidemiological and viral genome characteristics of the first human H7N9 influenza infection in Guangdong Province, China.
        J Thorac Dis. 2014; 6: 1785-1793
        • Yu H.
        • Wu J.T.
        • Cowling B.J.
        • Liao Q.
        • Fang V.J.
        • Zhou S.
        • et al.
        Effect of closure of live poultry markets on poultry-to-person transmission of avian influenza A H7N9 virus: an ecological study.
        Lancet. 2014; 383: 541-548
        • Zhang W.
        • Wang L.
        • Hu W.
        • Ding F.
        • Sun H.
        • Li S.
        • et al.
        Epidemiologic characteristics of cases for influenza A(H7N9) virus infections in China.
        Clin Infect Dis. 2013; 57: 619-620
        • Zhang Y.
        • Yu Y.S.
        • Tang Z.H.
        • Chen X.H.
        • Zang G.Q.
        Human infection with avian influenza A (H7N9) virus.
        Rev Inst Med Trop Sao Paulo. 2014; 56: 367-368
        • Zhou P.
        • Ma J.
        • Lai A.
        • Gray G.C.
        • Su S.
        • Li S.
        Avian influenza A(H7N9) virus and mixed live poultry-animal markets in Guangdong province: a perfect storm in the making?.
        Emerg Microbes Infect. 2015; 4: e63
        • Zhou J.
        • Wu J.
        • Zeng X.
        • Huang G.
        • Zou L.
        • Song Y.
        • et al.
        Isolation of H5N6, H7N9 and H9N2 avian influenza A viruses from air sampled at live poultry markets in China, 2014 and 2015.
        Euro Surveill. 2016; : 21
        • Zhou L.
        • Ren R.
        • Yang L.
        • Bao C.
        • Wu J.
        • Wang D.
        • et al.
        Sudden increase in human infection with avian influenza A(H7N9) virus in China, September–December 2016.
        West Pac Surveill Response J. 2017; 8: 6-14