A novel swine-origin influenza A (H1N1) virus caused worldwide outbreaks starting in April 2009. The aim of this study was to evaluate the clinical characteristics and outcomes of pandemic 2009 H1N1 pneumonia by comparing to community-acquired pneumonia (CAP) of other origin.
The authors conducted a prospective cohort study of consecutive adult (over 15 years old) patients with suspected CAP requiring admission to a tertiary university-affiliated hospital during the second wave of pandemic 2009 H1N1 influenza. Based on the results of real-time reverse transcriptase–polymerase chain reaction (RT-PCR) analysis, the staff completed a standard assessment form and managed the patients according to a uniform protocol. Clinical characteristics, as well as laboratory and radiologic findings, were collected and compared between pandemic 2009 H1N1 pneumonia and CAP of other origin. The primary outcome was in-hospital mortality and secondary outcomes were duration of hospitalization, duration of intensive care unit (ICU) stay, and requirement of mechanical ventilation.
A total of 135 patients with suspected CAP were included in the study. Of these, 59 patients were RT-PCR positive for H1N1 virus, and 76 patients were RT-PCR negative. Patients with H1N1 pneumonia were significantly younger than those with CAP of other origin (46.0 years vs. 68.0 years, p < 0.01) and more frequently had nonspecific symptoms (p < 0.01), initial leukopenia (8.5% vs. 0.0%, p = 0.01), lymphopenia (45.8% vs. 26.3%, p = 0.02), low values of C-reactive protein (CRP; 5.2 mg/dL vs. 13.4 mg/dL, p = 0.02), bilateral abnormalities (57.7% vs. 29.7%, p < 0.01) on chest radiography, ground glass opacities (43.9% vs. 12.8%, p < 0.01) on chest computed tomography, and low values of pneumonia severity index (PSI) score (56.0 vs. 91.0, p < 0.01) than those with CAP of other origin. However, there were no significant differences in infection severity, clinical outcome, length of ICU stay, requirement for mechanical ventilation, and mortality between the two groups.
This study shows that clinical characteristics and outcomes of 2009 H1N1 pneumonia are comparable to those of CAP of other origin. However, some characteristics, including younger age, nonspecific symptoms (including headache, leukopenia, and fatigue), lymphopenia, lower initial CRP and PSI score, and radiologic findings (including bilateral abnormalities and ground glass opacities), may help clinicians to diagnostically differentiate between H1N1 pneumonia and CAP of other origin before the result of RT-PCR are obtained.
Resumen: Comparación de los Hallazgos Clínicos y los Resultados de Pacientes Adultos Ingresados con Neumonía por Influenza A (H1N1) y con Otro Tipo de Neumonía
Un nuevo virus influenza A de origen porcino (H1N1) causó brotes en todo el mundo a partir de abril del 2009. El objetivo de este estudio fue evaluar las características clínicas y los resultados de la pandemia de neumonía H1N1 2009 y compararlos con los de la neumonía adquirida en la comunidad (NAC) de otro origen.
Se llevó a cabo un estudio de cohorte prospectivo de todos los pacientes adultos (mayores de 15 años) incluidos de forma consecutiva con sospecha de NAC que requirieron ingreso en un hospital terciario y universitario durante la segunda ola de pandemia influenza H1N1 2009. En base a los resultados de la reacción en cadena de la polimerasa-transcriptasa inversa (RCP-TI) a tiempo real, los urgenciólogos completaron un documento de valoración estándar y manejaron a los pacientes según el protocolo establecido. Se recogieron las características clínicas, de laboratorio y los hallazgos radiológicos, y se realizó una comparación entre la neumonía H1N1 2009 y la NAC de otro origen. El resultado principal fue la mortalidad intrahospitalaria y los resultados secundarios fueron la duración de la hospitalización, la duración de la estancia en unidades de cuidados intensivos (UCI) y la necesidad de ventilación mecánica.
Se incluyeron 135 pacientes con sospecha de NAC. De estos pacientes, 59 tuvieron una RCP-TI positiva para H1N1 y 76 pacientes fueron RCP-TI negativa. Los pacientes con neumonía H1N1 fueron significativamente más jóvenes que los pacientes con NAC de otro origen (46,0 vs. 68,0 años; p < 0,01). Además, los pacientes con neumonía H1N1 tuvieron más frecuentemente síntomas no específicos (p < 0,01), leucopenia inicial (8,5 vs. 0,0%, p = 0.01), linfopenia (45.8 vs. 26.3%, p = 0.02), un valor menor de proteína C reactiva (PCR) (5,2 vs. 13,4 mg/dL, p = 0,02), alteraciones bilaterales en la radiografía de tórax (57.7 vs. 29.7%, p < 0.01), y opacidades en vidrio delustrado en la tomografía computarizada de tórax (43.9 vs. 12.8%, p < 0.01). Teniendo en cuenta la radiografía de tórax, presentaron una puntuación menor del índice de gravedad de neumonía “pneumonia severity index (PSI)” (56,0 vs. 91,0, p < 0,01) que aquéllos con NAC de otro origen. Sin embargo, no hubo diferencias significativas entre los 2 grupos en la gravedad de la infección y en los resultados clínicos (duración de la estancia en UCI, necesidad de ventilación invasiva y mortalidad).
El presente estudio muestra que las características clínicas y los resultados de la neumonía H1N1 2009 son comparables a aquéllos observados en la NAC de otro origen. Sin embargo, algunas características que incluyen la edad más joven, los síntomas no específicos como el dolor de cabeza y la fatiga, la leucopenia, la linfopenia, el menor valor inicial de PCR y de PSI y los hallazgos radiológicos que incluyen alteraciones bilaterales y opacidades en vidrio deslustrado pueden ayudar a los clínicos al diagnóstico diferencial entre la neumonía H1N1 y la NAC de otro origen previamente a la obtención del resultado de la RCP-TI.
A novel swine-origin influenza A (2009 H1N1) was identified in the United States and Mexico in April 2009.[1, 2] To date, worldwide more than 214 countries and overseas territories or communities have reported laboratory-confirmed cases of pandemic H1N1, including 17,919 deaths. In South Korea, a large outbreak of pandemic H1N1 infection began in 2009, peaking in early November 2009. Of 45 million people in South Korea, at least 750,000 patients were confirmed as having pandemic H1N1 infection between April 2009 and April 2010, with 252 patients dying of this disease. Given the limited capacity of emergency departments (EDs) and the limited supplies of antiviral agents, it is important to detect patients who should have confirmatory diagnostic testing performed and adequate treatment given. Uncomplicated influenza typically presents with the abrupt onset of fever, myalgia, headache, and dry cough.[5, 6] According to the initial reports of 2009 H1N1, in contrast to seasonal influenza, severe disease from 2009 H1N1 seems to affect children and young adults and occurs among previously healthy individuals with obesity and respiratory disease.
Although the majority of 2009 H1N1 infection is mild and is a self-limited illness without treatment, pulmonary complications such as pneumonia are common and can cause respiratory failure or acute respiratory distress syndrome. However, the clinical characteristics and outcomes of patients with 2009 H1N1 pneumonia have not been determined. We therefore investigated the clinical characteristics and outcomes of pandemic 2009 H1N1 pneumonia and tried to find unique features of 2009 H1N1 pneumonia by comparing to community-acquired pneumonia (CAP) of other origins.
This study was a prospective cohort study and was approved by the ethics committee of our institution and written informed consent was obtained from each patient's family.
Study Setting and Population
This study was conducted prospectively in isolation rooms of our ED, named the acute care unit (ACU). From August 2009 to December 2009, we conducted a prospective cohort study on consecutive adult (over 15 years old) patients with suspected CAP requiring admission in a tertiary university-affiliated hospital. During the outbreak of pandemic 2009 H1N1 infection, our hospital was designated as a primary care center for patients with suspected pandemic H1N1 infection residing in the district, located in the southwest of the capital city of Seoul, Korea, with a population of 681,476 individuals.
According to the management protocol of our hospital for H1N1 influenza, we perform real-time reverse transcriptase-polymerase chain reaction (RT-PCR) testing in all patients with suspected CAP requiring admission and isolate these patients in the ACU until the results of RT-PCR test are obtained. Based on the results of RT-PCR test, patients with non-H1N1 pneumonia are admitted to ward or intensive care unit (ICU) beds. However, patients with H1N1 pneumonia were admitted continuously to the ACU (Figure 1).
Nasopharyngeal specimens were retrieved from all patients using flock swabs (Flexible Minitip 503CS01 flock swabs, Diagnostic Hybrids, Athens, OH). Samples were placed in 3-mL Universal Transport Medium (Diagnostic Hybrids, Athens, OH) and stored at 2 to 4°C. The specimens were vortexed and nucleic acids were automatically extracted with a NucliSENS easyMAG instrument (bioMérieux, l'Etoile, France). RT-PCR test was performed within 12 hours after the sample was collected, without freezing. This was done by using RealTime Ready Inf A/H1N1 detection kits (Roche Diagnostics, Basel, Switzerland) and multiwall-plate real-time PCR platforms (LightCycler 480 real-time PCR system, Roche Diagnostics). In our hospital during the study period, actual average turnaround time of RT-PCR test was more than 24 hours.
Diagnosis of pneumonia was based on Infectious Diseases Society of America/American Thoracic Society consensus guidelines. Briefly, patients showing symptoms and/or signs compatible with respiratory infection (fever ≥ 38°C; tachypnea; cough; and/or findings of crackles, bronchial breathing, or diminished breath sounds on auscultation) and radiologic infiltrations consistent with pneumonia were included, while patients who were under 15 years old, patients with hospital-acquired pneumonia, or patients transferred from other hospitals were excluded. To identify the causative microorganisms, all patients suspected of CAP underwent blood cultures and sputum cultures if sputum was present. In some cases, pleural fluid cultures, cultures of endotracheal aspirates, and Mycoplasma pneumoniae antibody testing was done.
Clinical data were recorded with a standard reporting form including age, sex, medical comorbidities (hypertension, diabetes mellitus, respiratory disorder, malignancy, and other diseases), main presenting symptoms, vital signs, laboratory findings, and findings of chest radiography and chest computed tomography (CT). On admission day, vital signs, arterial blood gas analysis (ABGA), and laboratory data were collected and pneumonia severity index (PSI) was calculated as a prognostic index for CAP. The primary outcome was in-hospital mortality, and secondary outcomes were duration of hospitalization, duration of ICU stay, and the need for mechanical ventilation. Clinical characteristics, as well as laboratory and radiologic findings, were compared between pandemic 2009 H1N1 pneumonia and CAP of other origin.
Categorical variables are expressed as absolute or relative frequencies, and continuous variables as mean ± standard deviation (±SD) or median and interquartile range (IQR). We used an independent-sample (unpaired) Student's t-test to compare the means of normally distributed continuous variables in those patients with H1N1 pneumonia and non-H1N1 pneumonia. The Mann-Whitney U-test was used to compare the means for nonnormally distributed variables. Categorical variables were compared using Fisher's exact test or chi-square test. A p-value < 0.05 was considered statistically significant, with no adjustment for multiple comparisons. All statistical analyses were performed using SPSS for Windows version 12.0 (SPSS Inc., Chicago, IL).
Demographic and Clinical Characteristics
During the study period, 135 patients who were admitted with a diagnosis of CAP were included in this study. Of these patients, 59 patients were RT-PCR positive for the 2009 pandemic influenza A (H1N1) virus (defined as the H1N1 pneumonia group), and 76 patients were RT-PCR negative (defined as the non-H1N1 pneumonia group; Figure 1). Etiologic findings are demonstrated in Table 1. Of patients with H1N1 pneumonia, 15 patients (25.4%) had bacterial coinfection. The clinical characteristics of patients with H1N1 pneumonia and with non-H1N1 pneumonia are shown in Table 2. The median age of all patients was 60 (IQR = 42.0 to 72.0) years, and 77 (57.0%) patients were male. Patients with H1N1 pneumonia were significantly younger than those with non-H1N1 pneumonia (46.0 years vs. 68.0 years, p < 0.01), but the sex distribution and body mass index (BMI) did not differ significantly between two groups. Thirty-five (59.3%) patients in the H1N1 pneumonia group had underlying medical conditions, while 63 (82.9%) patients in the non-H1N1 pneumonia group had at least one comorbidity (p < 0.01). Of comorbidities, asthma and cardiovascular disease were more frequently seen in patients with non-H1N1 pneumonia. The median duration from onset of symptoms to hospital admission in H1N1 pneumonia group was shorter than that of non-H1N1 pneumonia group (p < 0.01). Cough was the most common presenting symptom in both groups. However, atypical symptoms such as headache, myalgia, and fatigue were more common in H1N1 pneumonia group. On physical examination, the most common abnormal breath sound in both groups was crackles, and abnormal breath sounds were more frequently found in the non-H1N1 pneumonia group.
Table 1. Etiologic Findings in Patients With H1N1 Pneumonia and With Non-H1N1 Pneumonia
H1N1 Pneumonia Group (n = 59)
Non-H1N1 Pneumonia Group (n = 76)
Pathogens were identified from blood cultures, sputum cultures, pleural fluid cultures, cultures of endotracheal aspirates, or M. pneumoniae antibody test.
Table 2. Baseline Clinical Characteristics of Patients With H1N1 Pneumonia and With Non-H1N1 Pneumonia
H1N1 Pneumonia Group (n = 59)
Non-H1N1 Pneumonia Group (n = 76)
BMI = body mass index; BP = blood pressure; IQR = interquartile range (i.e., 25th and 75th percentiles).
Age, median (IQR), yr
Male sex, n (%)
BMI, median (IQR), kg/m2
Comorbid conditions, n (%)
Chronic lung disease
Chronic kidney disease
Duration of symptoms, median (IQR), days
Symptoms, n (%)
Initial vital signs, median (IQR)
Systolic BP, mm Hg
Diastolic BP, mm Hg
Pulse rate, beats/min
Respiration rate, breaths/min
Abnormal breathing sounds, n (%)
Crackles and wheezing
Laboratory and Radiologic Findings
Laboratory and radiologic findings are shown in Table 3. There were statistically significant differences in initial leukopenia (p = 0.01), initial lymphopenia (p = 0.02), and initial C-reactive protein (CRP; p = 0.02) between two groups. However, there was no significant difference in initial procalcitonin values between the two groups (p = 0.21; Table 3).
Table 3. Laboratory and Radiologic Findings in Patients With H1N1 Pneumonia and Non-H1N1 Pneumonia
H1N1 Pneumonia Group (n = 59)
Non-H1N1 Pneumonia Group (n = 76)
Leukopenia was defined as a WBC count less than 4 × 109 cells/L.
Lymphopenia was defined as ANC of less than 800 × 106 cells/L.
ANC = absolute neutrophil count; ALT = alanine aminotransferase; AST = aspartate aminotransferase; ABGA = arterial blood gas analysis; CRP = C-reactive protein; FiO2 = fraction of inspired oxygen; GGO = ground glass opacity; IQR = interquartile range (i.e., 25th and 75th percentiles); PaO2 = partial oxygen tension in arterial blood; WBC = white blood cell.
Laboratory findings, median (IQR)
WBCs, ×109 cells/L
ANC, ×106 cells/L
Leukopenia, initial, n (%)
Lymphocyte, ×103 cells/L
Lymphopenia, initial, n (%)
Platelets, ×109 cells/L
Blood urea nitrogen, mg/dL
CRP, initial, mg/dL
Procalcitonin, initial, ng/mL
Initial < 300, n (%)
Worst < 300, n (%)
Radiologic findings, n (%)
Infiltration and consolidation
GGO and consolidation
Chest radiographs were obtained from all patients in both groups. As seen in Table 3, 52 (88.1%) patients in the H1N1 pneumonia group and 74 (97.4%) patients in the non-H1N1 pneumonia group had abnormal findings. The most common abnormal finding in the H1N1 pneumonia group was lung infiltrates (67.3%), while lung consolidations (56.8%) were most common in the non-H1N1 pneumonia group. Chest CT was performed in 88 patients. Ground glass opacities and bilateral abnormalities were more frequently found in the H1N1 pneumonia group, while consolidation patterns and pleural effusions were more frequently found in the non-H1N1 pneumonia group. Findings of chest CT in nine patients with normal x-rays included ground glass opacity in five patients, consolidation in two patients, and a combination of consolidation and ground glass opacity in two patients. Of patients with normal chest x-rays in the H1N1 pneumonia group, five patients had ground glass opacity, one had consolidation, and one had a combination of consolidation and ground glass opacity. Most of patients with normal chest x-rays had bilateral lower zone involvement on chest CT.
Severity and Clinical Outcome
Severity and clinical outcomes are shown in Table 4. Infection severity and clinical outcome had no statistically significant differences between two groups. However, duration of hospitalization was statistically significant different between two groups (p = 0.01). The median PSI score was significantly lower in the H1N1 pneumonia group than in the non-H1N1 pneumonia group. However, PSI scores of less than 70 in patients who required mechanical ventilation were 12.5% (1/8) in the H1N1 pneumonia group and 11.1% (1/9) in the non-H1N1 pneumonia group. The mortalities in patients with PSI scores less than 70 were 2.7% (1/37) in the H1N1 pneumonia group and 0.0% (0/26) in the non-H1N1 pneumonia group. There were no differences between groups in the frequency of a PSI score ≤ 70 in patients who required mechanical ventilation or in in-hospital mortality in patients with a PSI score ≤ 70.
Table 4. Severity at Admission and Clinical Outcome of Patients With H1N1 Pneumonia and With Non-H1N1 Pneumonia
H1N1 Pneumonia Group (n = 59)
Non-H1N1 Pneumonia Group (n = 76)
ICU = intensive care unit; IQR = interquartile range (i.e., 25th and 75th percentiles); MV = mechanical ventilation; PSI = pneumonia severity index.
Infection severity, n (%)
Severe sepsis/septic shock
In-hospital mortality, n (%)
Duration of hospitalization, median (IQR), days
Duration of ICU stay, median (IQR), days
Required MV, n (%)
PSI score, median (IQR)
PSI score ≤ 70, n (%)
PSI score ≤ 70 in patients who required MV, n (%)
In-hospital mortality in patients with a PSI score ≤ 70, n (%)
To our knowledge, this is the first study to prospectively investigate the clinical characteristics and outcomes of 2009 pandemic influenza A (H1N1) pneumonia by comparing it to CAP of other origin. The main results of this study were that some characteristics including young age, atypical symptoms, leukopenia, lymphopenia, low value of initial CRP, and low PSI scores were significantly more frequent in patients with H1N1 pneumonia compared to those with CAP of other origin.
A strength of this study was that all patients with 2009 pandemic influenza A (H1N1) pneumonia (all diagnosed with RT-PCR of H1N1 and chest x-rays and/or chest CT) during the study period were included. Therefore, there was no selection bias into the study.
Previous reports of 2009 H1N1 in the United States have shown that this novel virus was more pathogenic than seasonal influenza virus, and children and young adults were more susceptible than older persons.[7, 12-14] We have also found that patients with H1N1 pneumonia were significantly younger than those with CAP of other origin. Our analysis showed that the presence of underlying medical conditions known to predispose to complications of influenza was common in both groups, but more common in patients with CAP of other origin than in those with H1N1 pneumonia. The association of obesity with 2009 pandemic influenza A (H1N1) infection has been reported. There has been a report that 90% of patients with H1N1 infection had a BMI ≥ 30, and 70% were extremely obese with a BMI ≥ 40.[15, 16] However, Kumar and colleagues reported that there was no significant difference in BMI between the survivors and nonsurvivors, despite the fact that obesity was more common in patients with confirmed or probable 2009 influenza A (H1N1) infection. The median BMI of our patients with H1N1 pneumonia was 23, and only four patients had a BMI greater than 30. There was no significant difference in BMI between patients with H1N1 pneumonia and those with CAP of other origin. This difference may be because South Korea has a relatively low prevalence of obesity compared with Western countries.
H1N1 infection can cause a variety of symptoms including fever, cough, sore throat, and headache. In this study, patients with H1N1 pneumonia, compared to those with CAP of other origin, had more nonspecific symptoms such as headache, myalgia, and fatigue. However, the incidence of sore throat, which was 66% in another study, was lower in patients with H1N1 pneumonia than in those with CAP of other origin. This difference is thought to be because this study focused on patients with H1N1 pneumonia rather than those with simple H1N1 infection.
Data from the United States and China reveal that a complete blood count in patients with H1N1 infection showed abnormal findings, including leukopenia, lymphopenia, thrombocytopenia, and anemia.[13, 18] In this study, a significant difference was observed between two groups in terms of leukopenia and lymphopenia. Leukopenia, especially lymphopenia, on admission may be an important finding for the tentative diagnosis of H1N1 pneumonia before the results of RT-PCR testing are obtained.
CRP and procalcitonin are indicators of bacterial infection.[19-21] The diagnostic utility of procalcitonin compared to CRP is disputed. A recent meta-analysis concluded that procalcitonin was more accurate than CRP for differentiating between viral and bacterial infections. However, clinical performance may vary depending on microbial factors, clinical syndrome, and cutoff values used.[22, 23] In this study, the value of initial CRP was significantly lower in patients with H1N1 pneumonia than in those with CAP of other origin. However, there was no significant difference in the value of initial procalcitonin between two groups. These results are consistent with the previous report indicating that procalcitonin was less effective than CRP in severe 2009 H1N1 influenza infection. Previous reports revealed that a PaO2/FiO2 ratio below 150 may be a poor prognostic factor in seasonal influenza pneumonia.[25, 26] In a Canadian study, however, no significant difference was found in PaO2/FiO2 ratio between survivors and nonsurvivors in critically ill patients with H1N1 infection. In our study, PaO2/FiO2 ratio also had no significant difference between two groups. From the above results, CRP may be a marker for differentiating H1N1 pneumonia from CAP of other origin, while procalcitonin and the PaO2/FiO2 ratio seem to be irrelevant.
There have been some reports describing the radiologic findings of 2009 pandemic influenza A (H1N1) infection. They reported that the most common radiologic and CT findings were unilateral or bilateral ground glass opacities with or without associated focal or multifocal areas of consolidation. In this study, 67.3% of patients with H1N1 had lung infiltrates, while 56.8% of patients with CAP of other origin had lung consolidation. In the H1N1 pneumonia group, ground grass opacity and bilateral abnormalities were more frequently found, compared with in CAP of other origin. These findings may help to differentiate patients with 2009 H1N1 pneumonia from CAP of other origin.
Bacterial coinfection in 2009 H1N1 infection was only present in 0.4% in a Mexican study and 4% in a California population.[28, 29] However, in our study, bacterial coinfection in 2009 H1N1 was present in 25.4%, and S. pneumoniae was the predominant pathogen of bacterial coinfection. These results correspond to those of a recent study of fatal cases of confirmed 2009 H1N1 infection in United States. According to that study, evidence of bacterial coinfection was found in 29%, including 10 cases caused by S. pneumoniae. These findings appear to suggest that for patients with 2009 H1N1 pneumonia requiring admission, initial empirical antibiotic treatment for coverage of S. pneumoniae may be appropriate, although further studies are needed to confirm this.
In this study, although patients with H1N1 pneumonia were younger and healthier than those with non-H1N1 pneumonia, infection severity, duration of ICU stay, requirement of mechanical ventilation, and in-hospital mortality had no significant differences between the two groups. These data may suggest that H1N1 pneumonia can be as dangerous as CAP of other origin.
Commons and Denholm reported that PSI had insufficient predictive ability to safely identify low-risk patients with 2009 H1N1 influenza. However, in our study, in-hospital mortality in patients with a PSI score ≤ 70 was 2.7% in the H1N1 pneumonia group and 0.0% in the non-H1N1 group. Therefore, PSI score may be a predictive factor to identify patients with H1N1 pneumonia at low risk of death, as in CAP of other origin.
Only adult patients were included in this study. Therefore, clinical characteristics and outcomes of pediatric patients with H1N1 pneumonia cannot be inferred from this study. The other limitation is that multivariate analysis was not performed. Therefore, confounding factors may influence the results of our study. We need to be cautious when interpreting any statistically significant outcomes close to a p-value of 0.05.
Our study shows that clinical characteristics and outcomes of 2009 H1N1 pneumonia are generally comparable to those of community-acquired pneumonia of other origin. However, some characteristics, including younger age; nonspecific symptoms including headache, leukopenia, fatigue, lymphopenia, and lower initial CRP and PSI score; and radiologic findings including bilateral abnormalities and ground glass opacities, may help the clinicians to diagnostically differentiate between H1N1 pneumonia and CAP of other origin before the results of RT-PCR are obtained and make early administration of ostelmivir possible in patients with 2009 H1N1 pneumonia.