Avian influenza H5N1 screening of intensive care unit patients with community-acquired pneumonia.

From February 1, 2005, to January 31, 2006, we screened 115 adults for avian influenza (H5N1) and influenza A if admitted to an intensive care unit with pneumonia. Using reverse transcription-PCR, viral culture, and serologic testing for anti-H5 antibody, we identified 8 (7%) patients with influenza A (H3N2); none had H5NI. Estimated costs for H5N1 screening were $7,375.

**********

The ongoing avian influenza (H5N1) pandemic poses risks to both human and animal health (1-5). The potential exists for cross-species transmission of avian influenza to humans and subsequent reassortment of avian and human influenza viruses in coinfected persons (6). Although atypical presentations of avian influenza (H5N1) have been reported (7,8), in most H5N1 case-patients pneumonia was the primary condition (3,4). To assess the prevalence of avian influenza (H5N1) and influenza A pneumonia, we screened adults admitted to a medical intensive care unit (ICU) with community-acquired pneumonia (CAP) for H5N1 and calculated the cost estimates for H5N1 screening in a tertiary care center of an H5N-endemic area in Thailand.

The Study

Thammasat University Hospital is a 450-bed tertiary care center with an 8-bed intensive care unit (ICU) equipped with central air-conditioning and 2 isolation rooms. The hospital serves a 150-km radius referral base in central Thailand and has 980 healthcare workers (HCWs). Annual influenza vaccination was not routinely offered to HCWs. During the study period, 2 confirmed cases of H5N1 occurred within 150 km of our hospital.

All adults admitted to the ICU with CAP between February 1, 2005, and January 31, 2006, were eligible for enrollment. Tracheal aspirates were collected for H5N1 testing, with reverse transcription (RT)-PCR, and viral culture. In patients <60 years with >14 days survival posthospitalization, paired acute-phase and convalescent-phase serum specimens were collected for identifying anti-H5 antibody. Acute-phase serum specimens for determining anti-H5 antibody were obtained within 1 week of symptoms, while convalescent-phase serum specimens were obtained >14 days after the acute-phase specimens were collected. Data collection included demographic characteristics, clinical data, and the costs associated with H5N1 screening. The diagnosis of CAP was defined according to the criteria recommended by the American Thoracic Society (9). Patients who were hospitalized for >2 days and in whom pneumonia developed were excluded from this study. The current Thai national surveillance definition for probable avian influenza (H5N1) included the following: 1) presence of fever (>38[degrees]C), and 2) influenza-like illness, and 3) exposure to sick poultry or residence in the disease-endemic areas with excess poultry death rates, and 4) radiographic evidence of severe CAP without an identified etiologic agent (10).

Viral cultures for H5N1 and influenza A, as part of screening, were incubated in Madin-Darby canine kidney (MDCK) cell monolayers at the Thai National Institute of Health. Tracheal aspirate specimens were tested by an RTPCR assay specific for the hemagglutinin gene of H5 (11). If a specimen yielded a positive H5 band, the specimens were confirmed by different RT-PCR primers and by real-time RT-PCR (12). All serum samples were tested for H5-specific antibody by a microneutralization (micro-NT) test. The reactive samples underwent confirmatory immunofluorescence testing by using H5-transfected 293 T cells as the test antigen (13). Influenza A/Thailand/ l(KAN-1)/2004 (H5N1) was used as the test virus. Acute-phase and convalescent-phase serum samples were serially diluted from 1:20 to 1:80. On the basis of previously established criteria, a positive test was defined as a neutralizing antibody titer >80 with a confirmatory immunofluorescence assay (14). Adults [greater than or equal to] 60 years of age were excluded from the serologic tests because the H5N1 micro-NT was previously reported to be less specific in this population (14).

Laboratory diagnostic costs (RT-PCR for H5NI, viral culture, and paired acute- and convalescent-phase serum samples for anti-H5 antibody) for each patient were obtained from line-item reports of the hospital's fiscal system. All costs in Thai baht currency were converted to US dollars at an exchange rate of 40 bahts per 1 US dollar. The cost for isolation of the index case, if influenza A or avian influenza (H5N1) was detected, were calculated from prior cost estimates (15).

One hundred fifteen of 450 patients (25%) met the definition of CAP and consented to study participation. The patient characteristics are summarized in Table 1. None of the 115 patients had tracheal aspirates positive for H5N1 ; also not positive were any serologic test results from the 42 patients (37%) who were <60 years old and survived >14 days after hospitalization. We were unable to calculate the prevalence of anti-H5 antibody in this sample, given that only 37% of participants underwent complete diagnostic antibody testing.

Eighteen patients (16%) met the Thai national surveillance definition of probable H5N1, yet tracheal aspirates and serologic test results were negative for H5N1. The median time from initial symptoms to hospitalization was 4 days (range 2-8 days), and all 18 were appropriately placed on contact and droplet isolation; the mean duration of isolation was 9 days (range 4-13 days).

Although 48 (42%) of the 115 participants had no identified etiologic agent associated with CAP, Streptococcus pneumoniae (n = 39; 34%), influenza A (H3N2) (n = 8; 7%), Staphylococcus aureus (n = 7; 6%), and Haemophilus influenzae (n = 6; 5%) were the most common microorganisms detected. In addition, 19 patients (n = 19; 16%) had gram-negative microorganisms detected. All patients with H3N2 pneumonia were promptly transferred to an isolation room; 5 (62.5%) had dual infections of H3N2 and S. aureus (n = 3), Klebsiella pneumoniae (n = 1) and Pseudomonas species (n = 1), while CAP due to H3N2 developed in 3 (37.5%). Of 18 patients who met the definition of probable H5N1, 8 (44.5%) had S. pneumoniae infection, 4 (22.5%) had S. aureus infection, 2 (11%) had H3N2 infection, 2 (11%) had Burkholderia pseudomallei infection, and 2 (11%) had no other agent detected. No CAP patients had anti-H5 antibody seroconversion, although 1 participant had evidence of positive anti-H5 antibody with low titer (10) during the recovery phase. This patient lived in an avian influenza (H5N1)-endemic area without a documented excess poultry death rate, and reported no exposure to sick poultry or persons with suspected avian influenza (H5N1) infection. His tracheal culture yielded S. pneumoniae. All patients with H3N2 pneumonia sought treatment between late March and November, the influenza A season in Thailand, and were in contact and droplet isolation for a mean of 7 days (range 1-12 days). The all-cause mortality rate was 10% (Table 1). The cost estimates were $7,375 for H5N1 screening, $23,328 for subsequent infection control measures, $300 for annual influenza vaccination of ICU HCWs, and $9,800 for annual influenza vaccination of the entire hospital staff (Table 2). The perceived benefits of vaccination of all ICU HCWs included reduced risk for influenza among vaccinated HCWs and reduced risk for influenza transmission to at-risk ICU patients.

Conclusions

Our study findings are relevant to the prevention and control of spread of both H5N1 and H3N2. The relatively high prevalence of H3N2 (7%) among our CAP patients suggests that HCWs in ICUs in disease-endemic regions are at high-risk of acquiring influenza A. An annual influenza vaccination occupational health program, similar to those in developed countries, along with targeted case identification of patients at high risk for influenza pneumonia, may help minimize the clinical and economic consequences of influenza A transmission. Although the importance of a single patient's positive low-titer anti-H5 antibody in this study was uncertain, this finding may represent a false-positive test, given that the patient had no notable exposure to sick poultry or to persons with suspected H5N1 infection. Additionally, the fact that all 18 probable case-patients had negative results for H5N1 suggests that the current Thai surveillance definition may need further refinement. Given the potential for reassortment of H5N1 and influenza A in a coinfected person residing in a disease-endemic setting, additional H5N1 screening, along with cost-effectiveness studies, are warranted before this screening strategy is adapted to H5Nl-endemic areas.

This work was supported by a grant from National Center and Genetic Engineering and Biotechnology, National Science and Technology Development Agency (BT-01-MM-13-4806), and Thai Research Fund to A.A.

Dr Apisarnthanarak is an infectious disease specialist and hospital epidemiologist at Thammasart University Hospital, Thailand, and a member of the Thai Infection Control Work Group. His major research focus is in preventing and controlling nosocomial infections, with a secondary focus on investigating outbreaks and emerging infectious diseases.

References

(1.) Cumulative number of confirmed human cases of avian influenza A (H5N1) reported to WHO, 13 February 2006 [cited 2006 Apr 1]. Available from http://www.who.int/csr/diseases/avian_influenza/ country/cases_table_2006_01_25_/en/index.html

(2.) Abbott A. Pearson H. Fear of human pandemic grows as bird flu sweeps through Asia. Nature. 2004;427:472-3.

(3.) Chotpitayasunondh T, Ungchusak K, Hanshaoworakul W, Chunsuthiwat S, Sawanpanyalert P, Kitphati R, et al. Human disease from influenza A (H5N1), Thailand, 2004. Emerg Infect Dis. 2005;11:201-9.

(4.) Tran TH, Nguyen TL, Nguyen TD, Luong TH, Pham PM, Nguyen VC, et al. Avian influenza A (H5N1) in 10 patients in Vietnam. N Engl J Med. 2004;350:1179-88.

(5.) Apisarnthanarak A, Erb S, Stephenson I, Katz JM, Chittaganpitch M, Sangkitporn S, et al. Seroprevalence of anti-H5 antibody among Thai health care workers after exposure to Avian influenza (H5N1) in a tertiary care center. Clin Infect Dis. 2005;40:e16-8.

(6.) Ungchusak K, Auewarakul P, Dowell SF, Kitphati R, Auwanit W, Puthavathana P, et al. Probable person-to-person transmission of avian influenza A (H5N1). N Engl J Med. 2005;352:333-40.

(7.) Apisarnthanarak A, Kitphati R, Thongphubeth K, Patoomanant P, Anthanont P, Auwanit W, et al. Atypical avian influenza (H5NI). Emerg Infect Dis. 2004;10:1321-4.

(8.) de Jong MD, Bach VC, Phan TO, Vo MH, Tran TT, Smith G J, et al. Fatal avian influenza A (H5NI) in a child presenting with diarrhea followed by coma. N Engl J Med. 2005;352:686-91.

(9.) Niederman MS, Bass JB Jr, Campbell GD, Fein AM, Grossman RF, Mandell LA, et al. Guidelines for the initial management of adults with community-acquired pneumonia: diagnosis, assessment of severity, and initial antimicrobial therapy. Am Rev Respir Dis. 1993;148:1418-26.

(10.) Coordinating system for laboratory testing and surveillance [cited 2006 Apr 1]. Available from http://avianflu.cclts.org/

(11.) Yuen KY, Chan PK, Peiris M, Tsang DN, Que TL, Shortridge KF, et al. Clinical features and rapid viral diagnosis of human disease associated with avian influenza A H5N1 virus. Lancet. 1998;351:467-71.

(12.) Spackman E, Senne DA, Myers T J, Perdue ML, Garber LP, Lohman K, et al. Development of a real-time reverse transcriptase PCR assay for type A influenza virus and the avian H5 and H7 hemagglutinin subtypes. J Clin Microbiol. 2002;40:3256-60.

(13.) Webster R, Cox N, Stohr K. WHO manual on animal influenza diagnosis and surveillance. World Health Organization, Department of Communicable Disease Surveillance and Response. WHO/CDS/CDR/2002.5 Rev. 1.

(14.) Rowe T, Abernathy RA, Hu-Primmer J, Thompson WW, Lu X, Lim W, et al. Detection of antibody to avian influenza A (H5N1) virus in human serum by using a combination of serologic assays. J Clin Microbiol. 1999;37:937-43.

(15.) Apisarnthanarak A, Kitphati R, Tawatsupha P, Thongphubeth K, Apisarnthanarak P, Mundy LM. Varicella-zoster outbreak among Thai healthcare workers. Infect Control Hosp Epidemiol. 2006. In press.

Address for correspondence: Anucha Apisarnthanarak, Division of infectious Diseases, Department of Medicine, Thammasart University Hospital, Pratumthani, Thailand, 12000; email: [e-mail]

Anucha Apisarnthanarak, * Pilaipan Puthavathana, [dagger] Rungrueng Kitphati, [double dagger] Pranee Thavatsupha, [dagger] Malinee Chittaganpitch, [double dagger] Prasert Auewarakul, [dagger] and Linda M. Mundy [subsection]

* Thammasart University Hospital, Pratumthani, Thailand; [dagger] Siriraj Hospital, Bangkok, Thailand; [double dagger] Thai National Institute of Health, Nonthaburi, Thailand; and [subsection] Saint Louis University School of Public Health, Saint Louis, Missouri, USA

Table 1. Demographic and clinical data for 115 hospitalized adults
with severe community-acquired pneumonia at a tertiary care center
in an H5N1-endemic region of Thailand *

                                                            Influenza
                                                 Total        A H3N2
Characteristics                                (N = 115)     (n = 8)

Age, years (mean, range)                      64 (17-82)    72 (55-82)
Sex, male                                       48 (42)       4 (50)
Tobacco smoking                                 21 (18)       1 (12)
No. of comorbid conditions (median, range)      1 (0-4)      3 (1-4)
Underlying diseases ([double dagger])
([section])
  Lung disease                                  48 (42)       4 (50)
  Diabetes                                      25 (22)       2 (25)
  Cardiovascular                                14 (12)       1 (12)
  Cerebrovascular or other neurologic           12 (10)       1 (12)
disease
  Other                                         42 (37)       3 (38)
Initial clinical symptoms ([section])
  Pulmonary ([paragraph])                      108 (94)       7 (87)
  Gastrointestinal (#)                           8 (7)        4 (50)
  Neurologic **                                  9 (8)        1 (12)
  Other                                          2 (2)        0 (0)
APACHE-II score, median (range)                16 (9-22)    17 (9-22)
History of recent travel                           0            0
Met definition of probable H5N1                 18 (16)       2 (25)
History of exposure to index case                  0            0
Outcome
  Death ([dagger][dagger])                      12 (10)       7 (88)
  LOS in MICU                                  14 (1-46)    15 (1-46)
  H5N1 seroconversion                              0            0

                                                Without
                                              concomitant
                                               influenza
                                                A H3N2       p value
Characteristics                                (n = 107)    ([dagger])

Age, years (mean, range)                      64 (17-74)       0.06
Sex, male                                       44 (41)         NS
Tobacco smoking                                 20 (19)         NS
No. of comorbid conditions (median, range)      1 (0-3)       <0.001
Underlying diseases ([double dagger])
([section])
  Lung disease                                  44 (41)         NS
  Diabetes                                      23 (21)         NS
  Cardiovascular                                13 (12)         NS
  Cerebrovascular or other neurologic           11 (10)         NS
disease
  Other                                         39 (36)         NS
Initial clinical symptoms ([section])
  Pulmonary ([paragraph])                      101 (94)         NS
  Gastrointestinal (#)                           4 (4)        0.001
  Neurologic **                                  8 (8)          NS
  Other                                          2 (l)         INS
APACHE-II score, median (range)                15 (9-22)        NS
History of recent travel                           0            NA
Met definition of probable H5N1                 16 (15)         NS
History of exposure to index case                  0            NA
Outcome
  Death ([dagger][dagger])                       5 (5)        <0.001
  LOS in MICU                                  14 (2-42)        NS
  H5N1 seroconversion                              0            NA

* Data are no. (%) of patients, unless otherwise indicated; NS,
nonsignificant; NA, nonapplicable; APACHE-II score, Acute Physiology
and Chronic Health Evaluation Score II; LOS, length of stay; MICU,
medical intensive care unit.

([dagger]) Categorical variables were compared using xz or Fisher exact
test, as appropriate; Continuous variables were compared using the
Wilcoxon rank sum test or t test, as appropriate. All p values were
2-tailed; p<0.05 was considered significant.

([double dagger]) Included those considered by the Advisory Committee
on Immunization Practices of the Centers for Disease Control and
Prevention to be associated with an increased risk of complication
from influenza infection.

([section]) Most patients had multiple underlying diseases and
initial clinical symptoms so the sums of all percentages are > 100%.

([paragraph]) Included cough, dyspnea or tachypnea, rigor and/or
chills, pleuritic chest pain, purulent sputum, or changes in the
characteristics of sputum, and auscultatory findings.

(#) Included diarrhea, and/or nausea or vomiting, abdominal tenderness.

** Included drowsiness, confusion, coma.

([paragraph][paragraph]) All patients did not receive antiviral
therapy.

Table 2. Cost estimates for routine avian influenza (H5N1)
surveillance, laboratory diagnostics, and infection control measures
in the ICU, February 1, 2005-January 31, 2006 *

                                     No.             Estimated
Category                           measures          cost (US$)

Cost associated with H5N1
routine screening
  Diagnostic testing
    RT-PCR                           115             $25 x 115
    Viral culture                    115             $30 x 115
    Paired acute- and                 42              $25 x 42
    convalescent-phase
    serology for anti-H5
    antibody ([dagger])
  Isolation for probable H5N1
  (n = 18) ([double dagger])
    Gowns/d                         1,800      $1/gown x 1,800 x 9 d
    Gloves/d                        1,800     $0.05/pair x 1,800 x 9 d
    Surgical masks/d                1,800     $0.25/mask x 1,800 x 9 d
    Staff time (min/d) to put       1,800        $1.26/hour x 1,800
    on/take off gloves, gowns,
    and mask
Cost of universal influenza
vaccination
  ICU HCWs                            30              $10 x 30
  HCWs, entire hospital              980             $10 x 980

Category                                            Total (US$)

Cost associated with H5N1
routine screening
  Diagnostic testing
    RT-PCR                                             2,875
    Viral culture                                      3,450
    Paired acute- and                                  1,050
    convalescent-phase
    serology for anti-H5
    antibody ([dagger])
  Isolation for probable H5N1
  (n = 18) ([double dagger])
    Gowns/d                                            16,200
    Gloves/d                                            810
    Surgical masks/d                                   4,050
    Staff time (min/d) to put                          2,268
    on/take off gloves, gowns,
    and mask
Cost of universal influenza
vaccination
  ICU HCWs                                              300
  HCWs, entire hospital                                9,800

* ICU, intensive care unit; RT-PCR, reverse transcriptase polymerase
chain reaction; HCWs, healthcare workers.

([dagger]) All 115 patients had acute phase serum samples tested for
anti-H5 antibody; 42 (37%) were <60 years old and survived [greater
than or equal to] 14 days after hospitalization.

([double dagger]) Estimated 1 min to put on and take off the protection
gear with 100 encounters per day (15).