Factors associated with death in hospitalized pneumonia patients with 2009 H1N1 influenza in Shenyang, China

In early April 2009, cases of human infection with 2009 pandemic influenza A (H1N1) virus were identified in the United States [1, 2] and Mexico [3], and the virus then spread rapidly to other regions of the world [4, 5]. On December 20, 2009, more than 208 countries and overseas territories or communities had reported laboratory confirmed cases of pandemic influenza A (H1N1), and at least 11,516 deaths had occurred [6]. The first case of 2009 pandemic influenza A (H1N1) virus infection in China was documented on May 10. From May to November of 2009, the number of reported cases in China was 69,160 [China Centre for Disease Control, CDC]. The majority of the early cases reported mild, influenza-like illnesses [7–11], but more severe infections have been reported as the pandemic spreads and the number of infected individuals increases. The greatest number of cases of 2009 H1N1 influenza infection resulting in severe clinical presentations and death have been reported in Mexico and other Western countries. The fatality rate of hospitalized patients with critical illness due to 2009 H1N1 influenza was about 10% [12–14] in United States and Mexico. The clinical features included fever and respiratory symptoms and death most commonly resulted from severe acute respiratory distress syndrome (ARDS) and refractory hypoxemia [7, 15–18]. Obesity, underlying health conditions and delayed neuraminidase inhibitor treatment were the major risk factors for a poor outcome of infection [17, 18]. However, the clinical outcomes and risk factors associated with 2009 H1N1 influenza infection in China remain to be determined and variables that could help clinicians to identify patients at high risk of infection would be valuable. Therefore, this report summarizes the clinical manifestations, clinical outcomes and the risk factors associated with death in hospitalized pneumonia patients associated with 2009 H1N1 influenza infections in Shenyang, China, during November to December 2009.

Here, we report on a series of hospitalized patients with pneumonia caused by 2009 H1N1 influenza infection in Shenyang, China, from November to December 2009. During the period of our study, the pandemic strain of H1N1 virus caused severe illness, including pneumonia and ARDS, and resulted in ICU admissions in 44% of patients and death in 14.7% of patients. These findings were similar to the majority of reports in other countries [14, 17, 19, 20].

In our study, almost 80% of the hospitalizations due to H1N1 influenza infection involved individuals who were between the ages of 18 and 49 years. The male to female ratio was 2.78:1. These age and sex distributions were different from those reported nationally for the normal population in China (male to female ratio was 106.30:100; age distributions were: 15-59 years, 68.70%; 60 years and over, 11.03%; 65 years and over, 7.69%). We concluded that severe illness resulting from H1N1 virus infection was more likely among male adults in Shenyang, which was consistent with other reports [19, 21–25]. Persons aged 50 years or older showed a low incidence of 2009 H1N1 influenza infection, this may be due to "protection" by preexisting immunity resulting from previous exposure to H1N1 influenza infection, along with reduced outdoor activity by these individuals because of the fear of H1N1 infection. A recent study demonstrated that persons aged 50 years or older who were hospitalized with pandemic 2009 influenza A (H1N1) infection were among those most likely to die, despite having lower hospitalization rates [14]. However, our findings did not support this conclusion, possibly due to the small sample size used in our study. Nevertheless, it is appropriate that clinicians should closely monitor elderly patients with pandemic 2009 H1N1 influenza infection and treat them accordingly.

In our study, almost 30% of patients had pre-existing medical conditions, of which chronic pulmonary diseases, diabetes and hypertension were the most common diseases, as found in other studies [19, 24, 25]. The patients included in our study who died had a higher rate of pre-existing diseases. Asthma is also considered a high risk factor for 2009 H1N1 influenza infection, however, our findings did not confirm this as only one patient in our study had asthma as a pre-existing condition. This discrepancy might be caused by the lower morbidity of asthma in China (approximately 2%) compared with Western counties (approximately 10-30%). COPD and heart disease are also considered high risk factors for 2009 H1N1 influenza infection, and in our study about two thirds of the patients with pre-existing medical conditions suffered from COPD or heart disease. Thus, it is appropriate that physicians should pay close attention to patients with COPD or chronic heart disease during the 2009 H1N1 influenza epidemic.

In our study, 32% of patients were obese compared with 7% of adults in the normal population in China, which indicated severe illness from H1N1 virus infection was more likely among obese individuals, as reported in other studies [17, 18, 24, 26, 27]. Almost 80% of patients who died from H1N1 influenza infection were obese and using multivariate logistic-regression models, obesity was found to be a factor associated with death from 2009 H1N1 influenza in our study, despite the wide 95% confidence interval (CI) for the odds ratio (OR) for BMI due to the limitations of using a small sample size.

The clinical features of patients who were hospitalized with 2009 H1N1 influenza infection included fever, cough, myalgia and dyspnea, which were generally similar to other reports. Whereas the incidence of gastrointestinal symptoms such as nausea, vomiting, and diarrhea was much lower than previously reported [7, 15–18, 28, 29]. The results of laboratory tests indicated lymphopenia, hypoproteinemia, elevated LDH and CRP levels, which were consistent with other reports [19, 29, 30]. Abnormalities in the laboratory test results were more significant in patients who were admitted to an ICU and/or died than in patients who were not admitted to an ICU, but these abnormalities were not predictive factors in ICU admission or death. Our results were not consistent with the study in Taiwan which found initial lymphocyte count less 800/microL was associated with the development of respiratory failure [31]. Although lymphopenia was not a risk factor for death in our study, we found lymphopenia was restored after about five days in most surviving patients, whereas this was not observed in patients who ultimately died. The multivariate logistic-regression model results indicated that lymphopenia that did not resolve after five days was a risk factor for death. Furthermore, we determined the CD4 and CD8 T cell counts in most of the patients included in our study. The results indicated a reduction in CD4 and CD8 T cell counts in about half of the patients during the early stage of 2009 H1N1 virus infection, which was similar to previous reports in China [7, 32]. These findings indicated that lymphopenia was mostly caused by T cell reduction, in particular a reduction in CD4 T cells. We were unable to determine whether the low immunity was pre-existing or caused by H1N1 infection because we did not know the basic values for each patient. During the severe acute respiratory syndromes (SARS) epidemic, lymphopenia was considered to be caused by the virus infection but further studies are needed to investigate the precise host immune response to 2009 H1N1 influenza virus. These results suggest the physicians should pay close attention to patients who are infected with 2009 H1N1 influenza virus, who are also obese or have experienced long-term lymphopenia.

In the 1918-1919 influenza A pandemic, most deaths were attributed to concurrent bacterial infection [31]. A report into 2009 pandemic H1N1 influenza also indicated that 29% of the patients displayed bacterial coinfection, which might have contributed to the death rate in the current pandemic [33]. In our study, only 16% patients had positive blood or sputum cultures, and most of these pathogens could potentially be responsible ventilator-associated pneumonia. The majority of bacterial co-infections were drug-resistant bacteria such as A. baumannii and methicillin-resistant S. aureus, whereas non drug-resistant bacteria were found predominantly in other reports [33, 34]. This result may reflect the abuse of antibiotics in China, and Chinese physicians should take measures to ensure protection against respiratory machine-related pneumonia, especially involving drug-resistant bacteria. To achieve this, physicians, nurses and patients should obey the rules of segregation and sterilization strictly and pay attention to bacteria contamination and the rational use of antibiotics. Few bacterial co-infections were detected in patients who did not have mechanical ventilation, which was consistent with most previous studies [14, 26]. However, bacterial diagnostic tests were not performed for all patients, especially those who were not admitted to an ICU, and most patients received antibiotics close to the time of culture collection, which could have reduced the diagnostic sensitivity.

Although antiviral therapy is most beneficial when treatment is initiated within 48 hours after the onset of illness [35], a prospective cohort study of oseltamivir therapy in patients hospitalized with influenza infection indicated a reduction in mortality, even when such therapy was initiated more than 48 hours after illness onset [36]. Recent data from Thailand also showed that oseltamivir therapy was associated with survival in hospitalized patients with influenza pneumonia [37]. Under an Emergency Use Authorization, the FDA recently approved oseltamivir therapy for 2009 H1N1 infection even if it is initiated more than 48 hours after the onset of illness and also approved its use in children under the age of one year [35]. In our study, antiviral drugs were administered to all patients, but such therapy was not initiated within 48 hours of the onset of illness in all patients and there was no difference between surviving patients and those who died in the median number of days from the onset of illness to oseltamivir initiation. Therefore, we were unable to conclude whether or not antiviral therapy in critically ill patients led to better clinical outcomes. No patients undertook the test of the isolated 2009 H1N1 influenza A strains for oseltamivir resistance in our study, so whether oseltamivir resistance affected the outcomes of patients with 2009 H1N1 infection was unclear. Delayed initiation of antiviral therapy may have contributed to an increased severity of illness in our study.

Our study has several limitations. The patients we evaluated represented 30% of the total hospitalizations in Shenyang for 2009 H1N1 infection that were reported to the CDC during the surveillance period that ended in December, 2009. No children or pregnant women were included in our study. Participation in the study was voluntary and was therefore subject to reporting bias. We evaluated only patients with confirmed 2009 H1N1 infection, so the group may not be representative of all hospitalized patients as some may have gone undetected. All diagnostic testing was clinically driven, and tests were not obtained in a standardized fashion. Finally, despite the use of a standardized data-collection form, not all of the required information was collected for all of the patients and the sample size was small.

During the evaluation period in Shenyang, 2009 H1N1 influenza caused severe illness requiring hospitalization. Thirty percent of the patients need an intensive care unit and fifteen percent died. Obesity and lymphopenia, which was not restored after five days of treatment, were factors associated with poor outcomes of 2009 H1N1 influenza infection. Early identification of pneumonia-susceptible patients at high risk of 2009 H1N1 influenza infection may aid clinicians in carrying out effective clinical management of this disease.

Wei Cui is an associate professor in-residence in the Department of Infectious disease, the First affiliated hospital of China Medical University. Her research interests included the pathogenesis of infectious diseases, particularly the pathogenesis of fulminant hepatic failure and intestinal barrier disruption.

Normal values obtained from laboratory tests in adults:

Lymphocyte count: 0.8-4 ×10⁹/L; LDH: 313-618 U/L; CK: 30-135 U/L; AST: 13-33 U/L; ALT: 8-42 U/L; Alb: 35-50 g/L; CRP: 0-0.8 mg/mL; CD4 T cell: 410-1590/μL; CD8 T cell: 190-1140/μL; CD3 T cell: 690-2540/μL.