The impact of pandemic (H1N1) 2009 influenza in HIV-infected individuals is unknown. Determining the prevalence of pandemic influenza in this at-risk group will guide vaccination programs. After the first pandemic wave, the seroprevalence rate of pandemic influenza in HIV-infected individuals in western Sydney, New South Wales, Australia, was 34.2%, similar to the rate observed in the general population. However, true seroprevalence is more accurately determined by seroconversion, defined as a 4-fold or greater rise between preexposure and postexposure antibody levels, which was 14.6% in the present study. Seroconversion rates were independent of CD4+ T-lymphocyte count and HIV plasma load. Neither HIV infection, nor severe immunosuppression, was a significant risk factor for pandemic influenza during the first southern hemisphere pandemic wave.
From the *Centre for Infectious Diseases and Microbiology Laboratory Services, Institute of Clinical Pathology and Medical Research, Westmead Hospital, Westmead, New South Wales, Australia; †School of Pediatrics and Child Health, Faculty of Medicine, Dentistry and Health Sciences, The University of Western Australia, Princess Margaret Hospital, Subiaco, Western Australia, Australia; and ‡Department of Infectious Diseases, Wollongong Hospital, Wollongong, New South Wales, Australia.
Received for publication July 28, 2010; accepted October 28, 2010.
The influenza A/California/7/2009 virus antigen for use in HI testing was kindly supplied by the WHO Collaborating Centre for Referral and Research on Influenza, Melbourne, Victoria, Australia.
The authors contribution: J.K. did the study design, performance of hemagglutination inhibition (HI) tests, and preparation of article; K.T. did performance of HI tests; C.C.B. and H.F. did data collection, statistical analysis, and critique of article; L.H. did the study design and conception; D.E.D. did study design and conception and critique of article.
The authors have no conflicts of interest to disclose.
Correspondence to: Dr Jen Kok, Centre for Infectious Diseases and Microbiology Laboratory Services, Level 3, Institute of Clinical Pathology and Medical Research, Westmead Hospital, Westmead, New South Wales 2145, Australia (e-mail: firstname.lastname@example.org).
The pandemic influenza A/H1N1 virus (hereafter H1N109) caused a worldwide outbreak in 2009, with significant activity in Australia. Rates of laboratory-confirmed H1N109 in western Sydney, New South Wales (NSW), were amongst the highest in Australia, but the true community attack rate was underestimated as testing was prioritized for individuals with severe influenza. Initial analysis of hospital and intensive care unit admissions failed to show a strong association between H1N109 and HIV infection in Australia,1 unlike other immunosuppressive conditions that have been associated with severe (measured by hospitalization or death) influenza.2,3 Although fatal H1N109 cases have occurred in HIV-infected individuals,4,5 similar clinical presentation and outcomes were observed in the HIV-infected and the general population.6,7 The seroprevalence rates of H1N109 infection in HIV-infected individuals remain uncertain.
The hemagglutination inhibition (HI) test detects influenza A hemagglutinin subtype-specific antibodies and is commonly used for serologic diagnosis of infection, determining susceptibility to influenza virus in epidemiologic and vaccine studies, measurement of postvaccination antibody responses, and by WHO for global influenza surveillance.8 H1N109 serosurveys and vaccine studies have shown higher baseline HI antibody titers in people aged older than 60 years, potentially explaining the reduced incidence of H1N109 infection in this age group. It is likely that prior H1N109 infection will provide some protection in second or subsequent pandemic waves.
In Australia, the monovalent H1N109 vaccine initially targeting “vulnerable” groups (including HIV-infected individuals) was made available on September 30, 2009.9 The 2010 Pandemic Vaccination Survey estimates that 3.9 million Australians (18.1% of the population) received the monovalent H1N109 vaccine, with the highest uptake observed in those ≥65 years of age.10 Distribution of the current trivalent (containing influenza H1N109, A/H3N2 and B) vaccine commenced in March 2010 in preparation for the southern hemisphere's winter season and is recommended for immunosuppressed patients including those with HIV infection.11 Knowing the prevalence of H1N109 in HIV-infected individuals will assist in determining optimal vaccination strategies in this group.
Frozen plasma samples (−20°C) from HIV-infected individuals submitted for HIV load monitoring during the 2009 Australian winter (June to September; “winter” samples) were tested by HI using standard procedures12 with the pandemic strain A/California/7/2009 as antigen.
Plasma samples were treated with receptor-destroying enzyme (RDE; Denka-Seiken Co. Ltd, Tokyo, Japan), 1:4 (vol/vol) to inactivate nonspecific inhibitors. After 16 hours incubation at 37°C, 1.6% trisodium citrate was added to inactivate RDE and reincubated in a 56°C water bath for 30 minutes. In 96-well V-shaped microtiter plates, serial 2-fold dilutions (1:10 to 1:640) of plasma were tested with 4 hemagglutination units of standardized virus and 1% (vol/vol) suspension of human O-negative red blood cells in phosphate-buffered saline. Microtiter plates were then incubated at 22°C for 1 hour after being shaken using a mechanical vibrator. The HI titer is the reciprocal of the highest dilution of plasma that completely inhibited agglutination of red blood cells. Winter samples with HI titers ≥40 were retested in parallel with stored plasma collected from the same patient 3-6 months previously (December 2008 to June 2009; “prewinter” sample). Seroconversion was defined as a 4-fold or greater increase in H1N109-specific HI antibody titers between prewinter and winter samples.
Patient demographics, CD4+ T-lymphocyte count (hereafter CD4+ count), and plasma HIV load were obtained from the local laboratory information system. For statistical analysis, Fisher exact tests were used on categorical variables and Mann-Whitney tests were used on continuous variables; the P values <0.05 were considered statistically significant. This study was approved by the Sydney West Area Health Service Human Research Ethics Committee [HREC2010/3/5.3 (3122) QA].
In NSW, the first laboratory-confirmed case of H1N109 occurred on 20th May, and peak H1N109 activity occurred in mid-July 2009.9 One hundred ninety-nine patients contributed samples during the winter season to measure H1N109-specific HI antibodies. Of 199 samples, HI titers were <40 [n = 131 (65.8%)], 40 [n = 33 (16.6%), 80 [n = 21 (10.6%)], 160 [n = 7 (3.5%)], 320 [n = 5 (2.5%)], and 640 [n = 2 (1%)]. The seroprevalence rate, defined by winter samples with HI titers of ≥40, was 34.2% (68 of 199).
Winter samples with HI titers of ≥40 were then retested in parallel with prewinter samples. In contrast to the seroprevalence rate, the H1N109 seroconversion rate was 14.6% (29 of 199 patients). When stratified according to HI titers, seroconversion was demonstrated in 7 of 33 (21.2%), 11 of 21 (52.4%), 4 of 7 (70%), 5 of 5 (100%), and 2 of 2 (100%) winter samples with HI titers of 40, 80, 160, 320, and 640, respectively.
Table 1 compares patient demographics, CD4+ counts, and plasma HIV loads. H1N109 seroconversion was commonest in the 18-34 age group. Females seroconverted more frequently compared with males (27.3% vs. 11%, P = 0.013).
The mean CD4+ counts were similar in individuals that seroconverted to H1N109 and those that did not [492 (range 93-1562, 95% CI: 359 to 625) vs. 539 (range 12-1353, 95% CI: 501 to 577) cells/μL, P = 0.377], as were the mean log10 plasma HIV loads [2.03 (95% CI: 1.72 to 2.34) vs. 2.02 (95% CI: 1.89 to 2.15), P = 0.94, respectively].
Four of 12 patients (33.3%) with CD4+ count <200 cells seroconverted. In this group of patients, the mean CD4+ count of those that seroconverted compared with those that did not was also similar [139 (range 93-173, 95% CI: 78 to 199) vs. 123 (range 12-181, 95% CI: 73 to 173) cells/μL, P = 0.65].
Understanding the epidemiology of H1N109 amongst HIV-infected individuals is important when modeling potential benefits of influenza vaccination. In HIV-infected individuals, studies assessing the efficacy of seasonal influenza vaccination showed a relative risk reduction of 41% for developing clinical influenza.13
This study determined that the overall H1N109 seroprevalence rate, defined as H1N109-specific HI titers of ≥40, was 34.2%. This is similar to the postpandemic seroprevalence rate of 28.4% observed in the general community (HIV status unknown) in NSW.14 However, the prepandemic seroprevalence rate of H1N109 in the same study was 12.8%, suggesting a “true” seroprevalence of 15.6%. This mirrors the seroconversion rate of 14.6% in our study, which is similar to the reported seroconversion rate of 13% in the Singaporean general community (HIV status unknown).15
In the present study, the seroprevalence rates between the HIV-infected cohort and the NSW general community were similar in the 18-34 and 35-64 age groups (36.8% vs. 40.1% and 35.1% vs. 26.3%, respectively). The seroconversion rate in HIV-infected individuals was also similar to the true seroprevalence rate in the NSW general population when those with prepandemic H1N109 antibodies were excluded (23.8% vs. 24.3% in the 18-34 and 14.5% vs. 19.6% in the 35-64 age groups, respectively).14 Higher H1N109 seroprevalence rates have been observed in younger populations compared to the elderly worldwide,16,17 where cross-protective antibodies from childhood exposure to 1918 influenza A/H1N1-like viruses may be present.18 In addition, the increased H1N109 seroconversion rate in females in our study is possibly due to increased exposure to children with H1N109 infection.
Although recent studies have defined seroprevalence based on seropositivity to H1N109 as a HI titer of ≥40,16 we propose that true community H1N109 seroprevalence is more accurately measured by seroconversion, defined as a 4-fold or greater antibody rise between preexposure and postexposure samples. Our data suggest that a single postexposure HI titer of 40 or 80 may overestimate true seroprevalence rates as seroconversion was observed in only 21.2% (7 of 33) and 52.4% (11 of 21) of patients with a winter sample HI titer of 40 and 80, respectively.
In our study, H1N109 seroconversion was independent of the degree of HIV-induced immunosuppression. No significant difference in CD4+ counts was observed between patients that did or did not seroconvert to H1N109, consistent with the conclusion that HIV-associated immunosuppression did not increase the risk of H1N109 infection.
Similar to past experience with seasonal influenza vaccination19, the development of significant antibody responses (as defined by a four-fold or greater rise in antibody levels) post H1N109 vaccination is likely to be affected by baseline CD4+ counts. Previous studies using the monovalent split-virion AS03 adjuvanted monovalent H1N109 vaccine (Pandemrix, GSK, Dresden, Germany) or the MF-59 adjuvanted monovalent H1N109 vaccine concluded that HIV-infected individuals with higher CD4+ counts are more likely to seroconvert post vaccination.20,21 The immunogenicity of H1N109 vaccination (as determined by seroconversion) in the HIV-population was reported to be 69%-83%,20,21 although protection against influenza is best measured by clinical effectiveness rather than vaccine immunogenicity.
The limitations of this study include the use of HI assays, which may be less sensitive than microneutralization.22 Plasma rather than sera were used to measure HI antibody titers23 but have been shown in our laboratory to be comparable in detecting seroconversion (data not shown). The seroconversion rate may have been potentially affected by cross-reacting antibodies between different influenza A subtypes, but this would have been unlikely as concurrent reactivity to seasonal influenza A/H1 and A/H3 antigen was low (11.7%-12.4% and 4.5%-9.7%, respectively) in samples positive for H1N109 in the NSW population,14 and H1N109 was the predominant influenza A subtype in Australia during the study period (68% of typed isolates).9
The timing of sample collection may have also influenced the seroconversion rate. The months of June to September 2009 were chosen as this was shortly after the first laboratory-confirmed case of H1N109 in NSW, included the period of peak H1N109 activity and was before the introduction of the monovalent H1N109 vaccine.9 Although possible, it is unlikely that the true seroprevalence rate would have been underestimated by seronegative patients being subsequently infected to H1N109 after sample submission, given the negligible H1N109 activity in NSW after September. Furthermore, we were unable to correlate seroconversion with clinical symptoms or laboratory evidence of H1N109 infection during the study period due to patient confidentiality. Nevertheless, this study provides valuable information about the true prevalence, based on seroconversion, of H1N109 infection in this HIV-infected cohort.
In conclusion, the rates of H1N109 seroconversion in HIV-infected individuals, irrespective of CD4+ count or HIV plasma load, were comparable to the general population in NSW, Australia. This may not reflect the experience in other countries with higher rates of HIV infection, although data is lacking. H1N109 seroprevalence rates in different populations based on a single sample are likely to be significantly higher than rates determined by seroconversion. This may impact on modeling studies and vaccination programs.
1. The ANZIC Influenza Investigators. Critical care services and 2009 H1N1 influenza in Australia and New Zealand. N Engl J Med
2. Centers for Disease Control and Prevention (CDC). Hospitalized patients with novel influenza A (H1N1) virus infection-California, April-May, 2009. MMWR Morb Mortal Wkly Rep
3. Writing Committee of the WHO Consultation on Clinical Aspects of Pandemic (H1N1) 2009 Influenza. Clinical aspects of pandemic (H1N1) 2009 influenza infection. N Engl J Med
4. Archer BN, Cohen C, Naidoo D, et al. Interim report on pandemic H1N1 influenza virus infections in South Africa, April to October 2009: epidemiology and factors associated with fatal cases. Euro Surveill
. 2009;14:pii 19369. Available at: http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=19369
. Accessed June 21, 2010.
5. Klein NC, Chak A, Chengot M, et al. Fatal case of pneumonia associated with pandemic (H1N1) 2009 in HIV-positive patient. Emerg Infect Dis
6. Isais F, Lye D, Llorin R, et al. Pandemic (H1N1) 2009 influenza in HIV-infected adults: clinical features, severity, and outcome. J Infect
. 2010; doi:10.1016/j.jinf.2010.08.002.
7. Campos-Loza A, Soto-Ramírez L, Sierra-Madero J, et al. New swine-origin influenza A (H1N1)v in HIV-infected patients during the 2009 outbreak in Mexico City. 801. Presented at: 17th Conference on Retroviruses and Opportunistic Infections; 2010; San Francisco, CA.
12. Kendal AP, Pereira MS, Skehel JJ. Concepts and Procedures for Laboratory-Based Influenza Surveillance
. Atlanta, GA: US Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention; 1982.
13. Tasker SA, Treanor JJ, Paxton WB, et al. Efficacy of influenza vaccination in HIV-infected persons. A randomized, double-blind, placebo-controlled trial. Ann Intern Med
14. Gilbert GL, Cretikos MA, Hueston L, et al. Influenza A (H1N1) 2009 antibodies in residents of new South Wales, Australia, after the first pandemic wave in the 2009 southern hemisphere winter. PLoS ONE
. 2010;5:e12562. doi:10.1371/journal.pone.0012562.
15. Chen MI, Lee VJ, Lim W, et al. 2009 influenza A(H1N1) seroconversion rates and risk factors among distinct adult cohorts in Singapore. JAMA
17. Greenberg ME, Lai MH, Hartel GF, et al. Response to a monovalent 2009 influenza A (H1N1) vaccine. N Engl J Med
18. Hancock K, Veguilla V, Lu X, et al. Cross-reactive antibody responses to the 2009 pandemic H1N1 influenza virus. N Engl J Med
19. Benne CA, Kroon FP, Harmsen M, et al. Comparison of neutralizing and hemagglutination-inhibiting antibody responses to influenza A virus vaccination of human immunodeficiency virus-infected individuals. Clin Diagn Lab Immunol
20. Bickel M, Wieters I, Khaykin P, et al. Low rate of seroconversion after vaccination with a split virion, adjuvanted pandemic H1N1 influenza vaccine in HIV-1-infected patients. AIDS
21. Orlando G, Pariani E, Mazza F, et al. Pandemic influenza vaccine in adult HIV-1-infected patients. AIDS
22. Bachmann MF, Ecabert B, Kopf M. Influenza virus: a novel method to assess viral and neutralizing antibody titers in vitro. J Immunol Methods
23. Fuller JD, Craven DE, Steger KA, et al. Influenza vaccination of human immunodeficiency virus (HIV)-infected adults: impact on plasma levels of HIV Type 1 RNA and determinants of antibody response. Clin Infect Dis