Emergence of community-acquired adenovirus type 55 as a cause of community-onset pneumonia.

BACKGROUND: Since 2008, severe cases of emerging human adenovirus (HAdV) type 55 (HAdV-55) were reported sporadically in China. But no comparative studies had been conducted to discern the differences in epidemiologic and clinical abnormalities between HAdV-55 and other types (HAdV-7, HAdV-3, HAdV-14, HAdV-50, and HAdV-C).
METHODS: A multicenter surveillance study for adult and adolescent community-acquired pneumonia (CAP) was conducted prospectively in Beijing and Yan Tai between November 2010 and April 2012. A standardized data form was used to record clinical information. The viral DNA extracted from the clinical samples or adenovirus viral isolates was sequenced.
RESULTS: Among 969 cases, 48 (5%) were identified as adenovirus pneumonia. Six branches were clustered: HAdV-55 in 21, HAdV-7 in 11, HAdV-3 in nine, HAdV-14 in four, HAdV-50 in two, and HAdV-C in one. Most HAdV-55 cases were identified during February and March. All the hypervariable regions of the hexon genes of the 21 HAdV-55 strains were completely identical. Patients who had HAdV-55 were about 10 years older ( P = .027) and had higher pneumonia severity index scores ( P = .030) compared with those with other types (HAdV-7, HAdV-3, HAdV-14, HAdV-50, and HAdV-C). Systemic BP was also higher among patients in the HAdV-55 group ( P = .006). Unilateral or bilateral consolidations were the most common radiologic findings in both patients with HAdV-55 and those with other types (57.9% vs 36%). More than one-half of the patients were admitted to hospital; oxygen therapy was given to 29.2% of the 48 patients, and two needed mechanical ventilation.
CONCLUSIONS: HAdV-55 has established itself as a major pneumonia pathogen in the Chinese population, and further surveillance and monitoring of this agent as a cause of CAP is warranted.

Community-acquired pneumonia (CAP) refers to pneumonia acquired outside of hospital or long-term care facilities. The overall annual incidence of CAP ranges from five to 20 per 1,000 adults. Many microbial pathogens can cause CAP, and the role of viruses may have been underestimated thus far because of a lack of appropriate diagnostic methods.2, 3 Modern molecular techniques have revealed that respiratory viruses account for about 22% of adult CAP cases.4, 5, 6, 7, 8, 9, 10 The most common viruses are influenza, parainfluenza, respiratory syncytial virus, metapneumovirus, and adenovirus.

We previously reported 18 sporadic CAP cases caused by human adenovirus (HAdV) from our single center between August 2008 and April 2011. Polymerase chain reaction (PCR) analysis using type-specific primers targeting the hexon gene revealed that they all belonged to species B (HAdV-11, HAdV-7, HAdV-3, and HAdV-14), and HAdV-11 accounted for 58.8% (10 of 17) of them. However, further genome sequence analysis proved that these 10 HAdV-11 strains were actually HAdV type 55 (HAdV-55). HAdV-55, an intertypic recombinant described originally as genome type 11a, was identified from an outbreak of acute respiratory tract infection in Shanxi Province, China, in 2006. It exhibited a neutralizing antigen epitope of HAdV-11 and the pathogenic properties of HAdV-14.13, 14 The whole-genome sequencing analysis showed that HAdV-55 had an HAdV-14 chassis with a partial HAdV-11 in the hexon gene.15, 16 For this reason, it was renamed HAdV-55.

Our previous case series indicated that HAdV-55 apparently emerged in Beijing. Adenovirus 14 is an emerging agent of concern that has been causing outbreaks of pneumonia not just in China, but worldwide. Adenovirus 55, which is related to adenovirus 14, is now also emerging as an agent of concern. We investigated whether HAdV-55 has a different clinical profile from the profiles of other adenovirus types circulating in China.

Materials and Methods

Beijing Network for Adult CAP

The Beijing Network for Adult Community-Acquired Pneumonia (BNACAP), which consists of 11 general hospitals from nine different districts in Beijing and one teaching hospital in Yan Tai, is a clinic-based, multicenter, prospective surveillance system for adults and adolescents with CAP. Yan Tai is a city by the sea in Shan Dong Province, located about 770 km southeast of Beijing. The institutional review board of Beijing Chao-Yang Hospital approved the study (project approval number 10-KE-49). All patients gave their written informed consent.

Study Population

Between November 2010 and April 2012, all adolescent and adult patients (aged 14 years or older) from 12 general hospitals who met the inclusion criteria of CAP were prospectively enrolled during daytime 7 days a week. Patients with HIV infection or neutropenia, those receiving immunosuppressive chemotherapy or prednisone steroids equivalent to 15 mg/d for 30 days, pregnant or breast-feeding women, and those with known or suspected active TB were excluded.

Clinical Data Collection

Clinical information collected by investigators with a standardized data form included the following: age, sex, comorbidities, smoking history, vaccination against influenza and Streptococcus pneumoniae in the past year, symptoms (fever, cough, sputum, dyspnea, chest pain), GI symptoms (nausea, vomiting, diarrhea, and abdominal pain), and neurologic symptoms (headache, dizziness). Clinical signs (body temperature, heart rate, respiratory frequency, BP, and crackles) and treatments (antibiotics, antiviral therapy, or oxygen use) were also recorded. The pneumonia severity index (PSI) was used to assess the severity of illness on the day of enrollment.

Symptoms and signs of all patients were followed up, either during their hospitalization or after discharge, until all symptoms disappeared. For outpatients, the same information was gathered. All the information collected from the patients was input into a computerized database.

Microbiologic Diagnostic Tests Undertaken

The nasal or throat swab specimens collected by the attending physicians were collected in 2-mL viral transport media, transported at 2°C to 8°C, and preserved at −80°C. The viral RNA was extracted from the clinical samples using a QIAamp RNA mini kit (QIAGEN). Following this, a commercially available Seeplex RV 15 ACE Detection kit (Seegene Inc), a multiplex, one-step, reverse transcriptase PCR, was used to screen for 15 different viruses as the cause of the respiratory illness. The kit included assays for adenovirus, influenza A and B viruses, human metapneumovirus, rhinovirus, respiratory syncytial virus (groups A and B), coronavirus (229E, NL63, OC43, and HKU1), parainfluenza virus (type 1, 2, 3, 4), bocavirus, and enterovirus.

Blood cultures were performed for patients presenting with chills and shivering. If pleural fluid and sputum samples were available, Gram stain and culture were performed. Urinary antigen tests for Legionella pneumophila and S pneumoniae (Binax) were also performed on all urine specimens. Acute sera (1-3 days after onset) and convalescent sera (2-4 weeks after onset) were collected for testing of the antibody for HAdV or other respiratory viruses.

Criteria for Viral Pneumonia

Viral pneumonia was diagnosed based on one of the following criteria: (1) the presence of HAdV or other respiratory viruses detected in sputum or throat swab samples by molecular methods or (2) seroconversion, defined as a fourfold or greater increase in titers of antibodies to HAdV or other respiratory viruses.

Cell Culture and Virus Isolation

Nasal or throat swab specimens were inoculated onto Hep-2 cells and cultured in a maintenance medium for detection of a cytopathic effect (CPE). Cells were observed for CPE every 7 days. Cultures exhibiting adenovirus-like CPE were processed again to confirm the presence of the virus.

Extraction of Viral DNA and PCR

Viral DNA was extracted from the clinical samples and adenovirus viral isolates using a QIAamp DNA mini kit (QIAGEN). The hexon and fiber genes were both amplified using primers described previously by Zhu et al. PCR was performed with a 25-μL reaction mixture containing 2 μL of template DNA. Reaction conditions were determined as described previously by Zhu et al.

Sequence Analysis

The PCR products were purified (QIAGEN) and sequenced by a dye terminator method (BigDye Terminator, version 3.1, cycle sequencing kit; Applied Biosystems) with an ABI Prism 3100 genetic analyzer (Applied Biosystems). Sequence data were stored as standard chromatogram format files (.abl) and were analyzed with Sequencher software (version 4.0.5; Gene Codes Corp), the Basic Local Alignment Search Tool program (National Center for Biotechnology Information), BioEdit sequence alignment editor software (version 5.0.9; Tom Hall, North Carolina State University, Raleigh, North Carolina), and the Molecular Evolutionary Genetics Analysis (MEGA) program (Sudhir Kumar, Arizona State University, Phoenix, Arizona).

Statistical Analysis

Data analysis was performed using SPSS 15.0 (IBM). A two-tailed independent-samples t test or a Mann-Whitney U test (in the case of nonnormal distributions) was used to compare continuous variables between the two groups. For the categorical data, univariate analysis was carried out using the χ2 test or Fisher exact test. Significance was fixed at P < .05.

Results

Epidemiology

Between November 2010 and April 2012, 1,013 cases with CAP were enrolled in the BNACAP study. Forty-four cases were ruled out: In 30, no throat/nasal specimen was obtained, and clinical information was missing in 14. Therefore, 969 cases were available for the etiology study. Among them, 393 were positive for at least one pathogen: respiratory viruses in 262, Mycoplasma pneumoniae in 168, typical bacteria in 47, Mycobacterium tuberculosis in 15, and Legionella pneumoniae in four. Dual causes were found in 65 patients (e-Table 1).

Types of HAdV

Forty-eight patients (48 of 969 [5%]) were identified as having adenovirus pneumonia, and 26 of the 48 adenovirus-positive samples showed characteristic adenovirus-like CPE. Basic Local Alignment Search Tool analysis based on the hypervariable region of the hexon genes from all 48 adenovirus-positive samples was performed. Among the 48 samples, 21 (43.8%) were HAdV-55, 11 (22.9%) were HAdV-7, nine (18.8%) were HAdV-3, four (8.3%) were HAdV-14, two (4.2%) were HAdV-50, and one (2.1%) was HAdV-C. Most HAdVs were identified in February and March. No adenovirus was found in November or December (Fig 1 ). The type distribution was similar between Beijing and Yan Tai City (Table 1 ).

Figure 1

Epidemiologic distribution of different types of human adenoviruses. Most human adenovirus type 55 was identified during February and March, and it had epidemiologic characteristics similar to other types. No adenovirus pneumonia was found in November and December, the typical influenza season months.

Table 1

—Epidemiologic and Clinical Characteristics of Patients With CAP Caused by Adenoviruses (Comparison Between HAdV-55 and Other Types)

Characteristic	Total (N = 48)	HAdV-55 (n = 21)	Other Types (n = 27)	P Value
Age, y	38.0 ± 18.5	44.7 ± 21.8	32.9 ± 13.8	.027a
Male (female), No.	33 (15)	16 (5)	17 (10)	.366
From Beijing	29 (60.4)	11 (52.4)	18 (66.7)	.38
Underlying diseases	5 (10.4)	4 (19.1)	1 (3.7)	.153
Current smokers	15 (31.3)	8 (38.1)	7 (25.9)	.531
Influenza vaccination within 1 y, No.	3	1	2	1.0
Streptococcus pneumoniae vaccination within 1 y, No.	1	0	1	1.0
Antibiotics before enrollment, No.	32	13	19	.555
Clinical features
PSI score	41.0 ± 22.8	49.1 ± 27.0	34.8 ± 16.8	.030a
Fever	43 (89.6)	19 (90.5)	24 (88.9)	1.0
Tmax, °C	39.3 ± 0.9	38.9 ± 0.7	39.4 ± 1.0	.070
Cough	45 (93.8)	20 (95.2)	25 (92.6)	1.0
Sputum	33 (68.8)	16 (76.2)	17 (63.0)	.366
Purulent sputum	18 (37.5)	10 (47.6)	8 (29.6)	.380
Dyspnea	9 (18.8)	3 (14.3)	6 (22.2)	.712
Chest pain	4 (8.3)	1 (4.8)	3 (11.1)	.621
GI symptoms	6 (12.5)	4 (19.1)	2 (7.4)	.383
Neurologic symptoms	6 (12.5)	3 (14.3)	3 (11.1)	1.0
Systemic BP, mm Hg	118.3 ± 11.3	123.4 ± 11.7	114.5 ± 9.5	.006a
Heart rate, beats/min	86.6 ± 11.5	84.2 ± 10.1	88.4 ± 12.4	.208
Respiratory rate, breaths/min	20.1 ± 3.2	20.8 ± 4.2	19.6 ± 2.3	.235
Moist rales	17 (35.4)	8 (38.1)	9 (33.3)	.377
Dry rales	2 (4.2)	2 (9.5)	0 (0)	.186
Conjunctival congestion	1 (2.1)	0 (0)	1 (3.7)	1.0
Rashes	2 (4.2)	1 (4.8)	1 (3.7)	1.0

Data are presented as mean ± SD or No. (%) unless indicated otherwise. CAP = community-acquired pneumonia; HAdV = human adenovirus; PSI = pneumonia severity index; Tmax = maximal temperature.

P < .05.

Demographic Characteristics

The mean age of the 48 cases was 38 years; 12 of the 48 (25%) were aged > 50 years. Patients infected by HAdV-55 were about 10 years older than those infected by other types (P = .027). Men predominated over women, with a sex ratio of about 2:1. More patients infected by HAdV-55 had underlying diseases (19.7% vs 3.7%), although the difference was not significant (Table 1).

Clinical Features: Comparison Between HAdV-55 and Other Types

Most clinical symptoms and signs between patients infected by HAdV-55 and those infected by other types did not differ, except for PSI score and systemic BP; these were significantly higher in patients infected by HAdV-55 (P = .030 and P = .006, respectively) (Table 1).

There was no difference in laboratory findings between HAdV-55-infected cases and those infected by other types (Table 2 ). Eight HAdV cases involved coinfections, including HAdV-55 with M pneumoniae in three, HAdV-55 with parainfluenza virus 3 and influenza virus B in one, HAdV-7 with M pneumoniae in one, HAdV-2 with respiratory syncytial virus A in one, HAdV-14 and parainfluenza virus 4 in one, and HAdV-3 with human coronavirus in one (Table 2).

Table 2

—Laboratory Findings and Chest Radiologic Characteristics of Patients With CAP Caused by Adenoviruses (Comparison Between HAdV-55 and Other Types)

Characteristic	Total (N = 48)	HAdV-55 (n = 21)	Other Types (n = 27)	P Value
WBC, 109/L	7.19 ± 3.59	6.70 ± 3.31	7.33 ± 3.84	.749
Leukocyte < 4,000/mm3, %	4 (8.3)	2 (9.5)	2 (7.4)	1.0
Leukocyte > 10,000/mm3, %	8 (12.5)	3 (14.3)	4 (14.8)	.715
Neutrophil, %	68.7 ± 12.3	69.9 ± 11.8	67.7 ± 12.9	.553
Lymphocyte, %	21.9 ± 9.3	20.9 ± 9.0	22.8 ± 9.8	.484
Hemoglubulin, g/L	141.4 ± 14.8	139.0 ± 15.0	143.3 ± 14.7	.327
Platelet, 109/L	186.9 ± 77.4	196.2 ± 87.6	179.7 ± 69.3	.469
AST, μ/L	26.5 (14-176)	28 (14-130)	26 (17-176)	.975
AST > 40 μ/L	10 (20.8)	4 (19.1)	6 (22.2)	1.0
ALT, μ/L	23 (6-122)	26.5 (9-122)	22 (6-109)	.511
ALT > 40 μ/L	8 (16.7)	3 (14.3)	5 (18.5)	1.0
ALB, g/L	36 (24.5-45.9)	34.6 (30.7-43.7)	36.4 (24.5-45.9)	.600
LDH, μ/L	201 (120-794)	193 (120-467)	217 (129-794)	.771
LDH > 250 μ/L	13 (27.1)	4 (19)	9 (33.3)	.338
CK, μ/L	47 (34-1994)	70.5 (41-345)	87 (34-1994)	.659
CK > 200 μ/L	5 (10.4)	1 (4.8)	4 (14.8)	.369
Tbil, μmol/L	9.9 (4.3-39.4)	11.8 (6.0-39.4)	8.9 (4.3-38.4)	.063
Tbil > 17.1 μmol/L	8 (16.7)	5 (23.8)	3 (11.1)	.272
Cr, μmol/L	68.9 (1.9-148.7)	77.5 (1.9-148.7)	62 (3.6-110.2)	.372
K, mmol/L	3.92 ± 0.42	3.91 ± 0.46	3.93 ± 0.40	.845
Na, mmol/L	136.2 ± 4.1	135.1 ± 4.2	137.0 ± 3.9	.102
Pao2, mm Hg	83.4 ± 23.8	91.8 ± 31.3	77.1 ± 14.7	.167
Paco2, mm Hg	32.6 ± 6.5	32.3 ± 7.3	32.9 ± 6.1	.827
ESR,a mm/h	27 (8-70)	32 (8-66)	23.5 (9-70)	.212
CRP,b mg/L	11.1 (2-147)	11.1 (2-147)	12 (2-104)	.728
CRP > 20 mg/L	10 (43.5)	3 (25)	7 (46.7)
PCT,c ng/mL	0.32 (0.02-2.65)	0.31 (0.05-0.87)	0.32 (0.02-2.65)	.876
PCT > 0.5 ng/mL	4 (33.3)	1 (20)	3 (42.8)	1.0
PCT > 1 ng/mL	2 (16.7)	0 (0)	2 (28.6)	1.0
Chest radiographyd
Bilateral involvement	19 (39.6)	9 (47.4)	10 (40)	.761
Consolidation	20 (45.5)	11 (57.9)	9 (36)	.223
Patchy infiltration	18 (40.9)	7 (36.8)	11 (44)	.760
Ground-grass opacity	12 (27.3)	5 (26.3)	7 (28)	1.0
Pleural effusion	3 (6.8)	1 (5.3)	2 (8)	1.0
Coinfectionse	8 (16.7)	4 (19.1)	4 (14.8)	.715

Data are presented as mean ± SD, No. (%), or median (range). ALB = albumin; ALT = alanine aminotransferase; AST = aspartate aminotransferase; CK = creatine kinase; Cr = creatinine; CRP = C reactive protein; ESR = erythrocyte sedimentation rate; LDH = lactate dehydrogenase; PCT = procalcitonin; Tbil = total bilirubin. See Table 1 legend for expansion of other abbreviations.

n = 27.

n = 23.

n = 12.

Total (N = 44), HAdV-55 (n = 19), and other types (n = 25).

Eight HAdV cases involved coinfections, including HAdV-55 with Mycoplasma pneumoniae in three, HAdV-55 with parainfluenza virus 3 and influenza virus B in one, HAdV-7 with Mycoplasma pneumoniae in one, HAdV-2 with respiratory syncytial virus A in one, HAdV-14 and parainfluenza virus 4 in one, and HAdV-3 with human coronavirus in one.

Forty percent of the patients had bilateral involvement on chest radiography (Table 2). Consolidation, patchy infiltrate, and ground-grass opacity were the most common findings in pneumonia caused by HAdV. Patients infected by HAdV-55 presented consolidation more commonly than did those infected by other types (57.9% vs 36%) (Fig 2 ), but the difference was not significant.

Figure 2

Radiographic findings of four patients infected with different types of adenoviruses. A, An 18-year-old young man infected with human adenovirus (HAdV)-14. Chest CT scan showed patchy infiltrate, ground-grass opacity, and partial consolidation of left lower lobe. B, A 14-year-old boy infected with HAdV-55. Chest CT scan showed consolidation and ground-grass opacity of left lower lobe. C, A 15-year-old girl infected with HAdV-7. Chest CT scan showed patchy infiltrate and ground-grass opacity of left lower lobe. D, A 24-year-old man infected with HAdV-3. Chest CT scan showed consolidation and ground-grass opacity of right lower lobe. ys = years old.

Complications, Management, and Prognosis of Patients With HAdV Pneumonia

More than one-half of the patients were admitted to hospital, but there was no difference between HAdV-55 and other types (Table 3 ). No case was proved to have a coinfection with bacteria, but coinfections with other respiratory viruses (25%) or M pneumoniae (12.5%) were common. Oxygen therapy was given to 29.2% of the patients, and only two needed mechanical ventilation. Antibiotics were given to all the patients, but only four were prescribed antiviral drugs (all from Beijing Chao-Yang Hospital). The clinical outcomes, including duration of fever and other respiratory symptoms, length of stay in hospital, and hospitalization expenses, were similar between HAdV-55 and other types (Table 3).

Table 3

—Complications, Management, and Prognosis of Patients With Adenoviral Pneumonia

Characteristics	Total (N = 48)	HAdV-55 (n = 21)	Other Types (n = 27)	P Value
Hospitalization	26 (54.2)	11 (52.4)	15 (55.6)	1.0
Complications
Coinfection with bacteria	0 (0)	0 (0)	0 (0)	…
Coinfection with Mycoplasma pneumoniae	6 (12.5)	3 (14.3)	3 (11.1)	1.0
Coinfection with other respiratory virus	12 (25)	4 (19.1)	7 (25.9)	.726
ARDS	2 (4.2)	0 (0)	2 (7.4)	.497
Management
Oxygen therapy	14 (29.2)	7 (33.3)	7 (25.9)	.750
Mechanical ventilation	2 (4.2)	0 (0)	2 (7.4)	.497
Antibiotics	48 (100)	21 (100)	27 (100)	1.0
Antivirala	4 (8.3)	4 (19.1)	0 (0)	.031
Outcomes
Duration of fever, d	6 (2-20)	7 (2-20)	5 (2-9)	.201
Duration of cough, d	14.5 (4-37)	14.5 (5-37)	14 (4-28)	.810
Duration of sputum, d	13.5 (4-37)	13.5 (4-37)	12 (6-29)	1.0
Duration of dyspnea, d	6 (1-21)	6 (2-11)	6 (1-21)	.905
Duration of chest pain, d	5 (5-7)	5 (5-5)	6 (5-7)	.667
LOS in hospital, d	7.5 (3-26)	9 (5-21)	6 (3-26)	.087
Hospitalization expenses, RMB	6,020 (1,200-49,947)	7,608 (4,229-32,684)	4,865 (1,200-49,947)	.234

Data are presented as No. (%) or median (range). LOS = length of stay; RMB = Chinese Yuan. See Table 1 legend for expansion of other abbreviations.

Three of the patients were prescribed acyclovir and the fourth was given ribavirin. All the patients were from Beijing Chao-Yang Hospital.

Genome Sequence and Analysis of HAdV-55

A phylogenetic analysis was conducted based on the hypervariable region of the hexon gene to demonstrate the genetic relationship between HAdVs strains and the other seven HAdVs species (A-G); 47 of 48 strains belonged to HAdV species B and only one was HAdV-C (e-Fig 1A). Further partial hexon gene of HAdV strains were compared with the sequences of HAdV-B species in GenBank (e-Fig 1B); 21 of the 47 HAdV-Bs formed a dependent branch and revealed 100%, 96.7%, and 80% homologies with HAdV-55 (FJ643676), HAdV-11 (AF532578), and HAdV-14 (AY803294), respectively. Another phylogenetic tree was then conducted based on the partial fiber gene of the 15 HAdVs, hexon genes which were homologous to HAdV-55 (e-Fig 1C). The partial fiber gene had 99.8% to 100%, 99.4% to 99.5%, and 94.2% to 94.3% nucleotide identity with HAdV-55 (FJ643676), HAdV-14 (AY803294), and HAdV-11 (AF532578), respectively. (The amplification or sequencing of the other six strains failed, and the real-time PCR indicated that their hexon genes were HAdV-14 and their fiber genes were HAdV-14 [data not show]).

All the hypervariable regions of the hexon genes (18281-19247nt, 967bp) of the 21 HAdV-55 strains were completely identical. Two of the 15 partial fiber genes (30689-31818nt, 1130bp) had one nucleotide difference from the others. The nucleotide sequences data from these 48 HAdVs were submitted to GenBank, and accession numbers were allocated as KC510700 to KC510747 (Hexon genes) and KC510748 to KC510762 (fiber genes).

Discussion

To our knowledge, this study is the first large cohort on the epidemiology and clinical features of CAP associated with HAdV-55, the emerging pathogen among immunocompetent adolescents and adults. Our data showed clearly that HAdV-55 has established itself as a major pneumonia pathogen in the Chinese population and that further surveillance and monitoring of this agent as a cause of CAP is warranted.

HAdV has been recognized as an important viral cause of ARDS. The HAdV types most frequently associated with ARDSs include subspecies B1 HAdV-3, HAdV-7, and HAdV-21 and species E HAdV-4. The association of subspecies B2 HAdV (HAdV-11, HAdV-14, HAdV-34, HAdV-35) infection with ARDS has been rarely reported historically, with some of that documentation covering military trainees.15, 18, 19

In China, HAdV-3 and HAdV-7 were the most common types of pathogens.20, 21, 22 HAdV-11a associated with ARDSs can be traced back to the 1980s and it reemerged as an ARDS pathogen in 2006.12, 16 HAdV-55 (formerly known as HAdV-11a) was renamed HAdV-55 based on complete genomic sequence data, which clearly showed that HAdV-55 was a recombinant between the HAdV-11 and HAdV-14 ancestral strains.

Adenovirus 14 is an emerging agent of concern that has been causing outbreaks of pneumonia not just in China but worldwide.24, 25 Tate et al reported an outbreak of severe respiratory disease associated with HAdV-14 in a US Air Force training facility. Five hundred fifty-one of 1,147 trainees (48%) with febrile respiratory illness were infected with HAdV-14; 23 trainees were hospitalized with pneumonia; four of those required admission to an ICU, and one died. Subsequently, outbreaks associated with HAdV-14 were reported in other states, such as Oregon, Alaska, and so forth.27, 28

Today, adenovirus 55, which is related to adenovirus 14, is also emerging as an agent of concern. Because of the absence of cases caused by other HAdV types, Vento et al could only compare pneumonia caused by HAdV-14 with HAdV-14-negative cases. Our study had the advantage of being able to compare the differences in clinical, laboratory, or radiographic abnormalities caused by HAdV-55 and other types of pathogens. We proved that patients with diseases due to HAdV-55 were about 10 years older (P = .027) and had higher PSI scores (P = .030)

Our study has several limitations. First, our case series mainly represents the relatively mild and moderate end of the disease. We believe a wider surveillance study is needed to evaluate the spectrum of the disease caused by this emerging pathogen in an affected area in China. In addition, virus isolates were typed only by amplification and sequencing of the hexon gene, and no seroneutralization was performed. More laboratory tests should be carried out to understand the genomics and pathogenic characteristics.

Conclusions

In conclusion, our data provide new insight into the epidemiology of HAdV-55 infection in China. Patients with HAdV-55 infection were about 10 years older and had higher PSI scores than did patients infected by other types (HAdV-3, HAdV-7, HAdV-14). Furthermore, because it is difficult to discern HAdV pneumonia from clinical symptoms and signs, viral cause determination and a good surveillance system are important.