OUP user menu

Journal of Crohn's and Colitis: 10 (10)

Editor-in-Chief

Laurence J. Egan, Ireland

Associate Editors

Shomron Ben-Horin, IsraelSilvio Danese, ItalyPeter Lakatos, HungaryMiles Parkes, UKJesús Rivera-Nieves, USABritta Siegmund, GermanyGijs van den Brink, NLSéverine Vermeire, Belgium

6.585
5.586

Published on behalf of

Incidence and characteristics of the 2009 influenza (H1N1) infections in inflammatory bowel disease patients

Makoto Naganuma, Toshimitsu Fujii, Reiko Kunisaki, Naoki Yoshimura, Masakazu Takazoe, Yoshiaki Takeuchi, Eiko Saito, Masakazu Nagahori, Keiko Asakura, Toru Takebayashi, Mamoru Watanabe
DOI: http://dx.doi.org/10.1016/j.crohns.2012.06.019 308-313 First published online: 1 May 2013

Abstract

Background: In 2009, influenza A (H1N1) infections spread worldwide. Because the use of immunomodulators is associated with an increased risk of infection, inflammatory bowel disease (IBD) patients who are on immunomodulators might be concerned about H1N1 influenza infections. The aim of this study was to investigate the age distribution and risk factors associated with H1N1 influenza of IBD patients in 2009–2010.

Methods: A multicenter, prospective study was conducted, and 570 IBD patients were enrolled. Patients were followed up for 10 months to identify any new infections. The incidence and age distribution of the H1N1 influenza infections were analyzed. IBD patients with H1N1 influenza infections and 2 matched, noninfected IBD patients were selected to assess the effect of specifying the medication on the incidence of infections.

Results: A total of 38 patients (6.7%) developed H1N1 influenza infections. The incidence of H1N1 influenza infections in patients aged less than 20 years was significantly higher than that among patients in other age groups (p < 0.01). The age distribution for H1N1 influenza infections in IBD patients was comparable to those in the general population. No patients needed hospitalization due to influenza infection. A total of 29 patients (76%) recovered from the H1N1 influenza symptoms within 7 days and 20 patients (53%) received antiviral treatment. The percentage of patients who used steroids or thiopurine was comparable between the cases of H1N1 influenza infection and the control group.

Conclusion: Our prospective study showed that younger IBD patients were frequently infected with the influenza A (H1N1) virus as well as general population. Admission and fatal cases due to H1N1 influenza infections were not observed.

Keywords
  • Inflammatory bowel disease
  • Influenza A (H1N1)
  • Steroids
  • Immunomodulators

1 Introduction

In 2009, infections with influenza A (H1N1) virus spread worldwide.1 Although the most common symptoms are cough and fever, which are uncomplicated influenza-like symptoms, the World Health Organization (WHO) reported that more than 10,000 individuals worldwide died from the infection. Recent studies indicated that fatality ratio in influenza A (H1N1) virus infection seems to be higher than that of seasonal influenza infections.2,3 The WHO identified age < 65 years, immunosuppression, and chronic medical condition as being the specific risks for influenza A (H1N1) infections. These factors are relevant to ulcerative colitis and Crohn's disease4 because majority of patients with IBD are younger and they have often been treated with steroids, thiopurine, tacrolimus, methotrexate, and biological treatments.

In general, the risk factors for the complications of infectious diseases in inflammatory bowel disease (IBD) patients are either therapy-related or non-therapy-related.5,6 Non-therapy-related factors include malnutrition and advanced age, whereas therapy-related factors include operations, immunosuppressive treatment, and leukocytopenia caused by immunomodulators. Patients with IBD are at a higher risk for opportunistic infections711 because of the suppression of their immune systems. A previous study has indicated that the combined use of the immunosuppressants is associated with an increased risk of opportunistic infections.7 Patients who are on immunomodulators and biologics should be considered to be at high risk for complications from influenza A (H1N1) infections.2 However, no data exist on the incidence of H1N1 influenza infections in patients with IBD.12 It is unclear whether IBD patients who are treated with immunomodulators/biologics are more susceptible to infection with H1N1 influenza virus than IBD patients who are not treated with these drugs.

The aim of this study was to prospectively assess the incidence of H1N1 influenza infections from 2009 to 2010. We also assessed the age distribution and risk factors associated with H1N1 virus infections in IBD patients.

2 Patients and methods

2.1 Study design and patient populations

This was a prospective study involving IBD patients at 4 institutions in Japan (Tokyo Medical and Dental University Hospital, Yokohama City University General Hospital, Social Insurance Center General Hospital, and Showa University Hospital). No in-patients were included in this study. A total of 570 patients were enrolled in this study from September to December 2009 (Table 1). The numbers of ulcerative colitis (UC) and Crohn's disease (CD) patients were 371 and 187, respectively. Indeterminate colitis patients (n = 11) were also enrolled. Patients were followed up for new influenza infections during the study period, which ended during September 2010. The protocols in this study were approved by the ethical review boards of all of the institutions, and patients aged < 20 years old were allowed to be enrolled in the study if the patients and their parents agreed.

View this table:
Table 1

Clinical features of IBD patients.

N% (or range)
Gender (male:female)325:245
Median age (range)38.7(14–82)
Age groups (years)≦ 19457.9%
20–2914826.0%
30–3915527.2%
40–4910117.7%
50–595910.4%
60 ≦6110.7%
IBD subtypeUC37165.1%
CD18732.8%
Indeterminate colitis122.1%
Current smoking8314.6%
Medication
Mesalamine/SASP43175.6%
Steroids9616.8%
Elemental diets8214.4%
Thiopurine15026.3%
Methotrexate61.1%
Infliximab8114.2%
Tacrolimus346.0%

2.2 Case–control study

To assess the effect of medication on the rate of H1N1 influenza infections, a nested, case–control study was performed. For the case group, patients with H1N1 influenza infections were selected. At every clinic visit, the patients were asked whether they had developed H1N1 influenza infections. All influenza-infected patients were diagnosed based on both clinical symptoms (such as fever, fatigue, sore throat, runny nose, muscleache, and headache) and laboratory tests. Rapid diagnostic tests can detect influenza A and influenza B viruses and distinguish between them. Specimens were obtained using nasopharyngeal swabs. Influenza infections were mostly diagnosed at each institution and these were also diagnosed at other hospital/clinics except 4 institutions (e.g. general practitioner).

Two uninfected patients were selected as controls and were matched based on age (with an interval of 10 years; ≦ 19, 20–29, 30–39, 40–49, and ≧ 50 years-old). Controls were required to visit their clinics within 2 weeks before or after the infections were observed in the matched patients.

2.3 Data acquisition

The patients' demographic data included their current age, age at diagnosis, current smoking habits, type of disease (UC or CD) and medical treatments during the follow-up period. All patients were classified as UC or CD according to previously established criteria.13 In the present study, patients were asked whether H1N1 influenza vaccinations were administered at each clinic. We also asked whether patients with H1N1 influenza infections had used antiviral treatments. The duration of H1N1 influenza symptoms (which were less than 7 days, 7–14 days, or more than 14 days) and hospital admissions due to H1N1 influenza infections was also assessed.

The use of infliximab was recorded when patients were treated with it within 8 weeks of the infectious episode or the index date. Because health insurance did not cover the expenses associated with adalimumab and certolizumab, these treatments were not studied (although adalimumab could be used after October 2010). The use of 5-aminosalicylates, corticosteroids, and tacrolimus was recorded when patients were administered these agents within 2 weeks of the infectious episode or the index date. The use of other medications (including azathioprine, 6-mercaptopurine (6-MP) and methotrexate) was recorded when these agents had been used within 12 weeks of the H1N1 influenza episode or the index date. For infected and control cases, each medical treatment was recorded. For example, if a H1N1 influenza infection was observed in December 2009 and the patient had used 5-ASA and corticosteroids within 2 weeks prior to the infection, it was recorded that 5-ASA and corticosteroids were used for that patient. The use of medications in matched controls was also recorded during the same period.

2.4 Data analysis and statistical methods

The incidence and age distribution of patients with H1N1 influenza infections were assessed. The age distribution of H1N1 influenza infections in the general population was obtained from a report of the Japanese Infectious Disease Surveillance Center. In the case–control study, the gender, duration of disease, type of disease, and medication use (including mesalamine, corticosteroids, thiopurine, infliximab, methotrexate and tacrolimus) during the infection or on the index date were compared between patients with infections (case) and those without infections (control) to assess the effects of medication use on the incidence of infections. The percentage of patients who used any immunomodulators (any IM) or 2 types of IM was also compared between the case and control groups. Any IM was defined as the use of at least one of the following medications: steroids, infliximab, thiopurine, tacrolimus or methotrexate. A total of 36 cases and 2 controls per case were required to detect an odds ratio of 2.0 if 30% of patients in the control group were estimated to use immunomodulators.

For statistical analyses, the differences in ages between two groups were tested using a t-test, and the differences in the incidence of other factors (e.g., sex, and type of disease) between two groups were tested using chi-squared tests and Fisher's exact test. The significance level in all tests was < 5%. All data were analyzed using SPSS version 18 (SPSS Corp., Tokyo, Japan).

3 Results

3.1 Younger age was associated with an increased risk of H1N1 influenza infections in IBD patients

The patients' clinical features are provided in Table 1. A total of 38 patients ((6.7%) 95% Confidence Interval [CI] (4.6–8.8%)) were infected with H1N1 influenza virus. Type B influenza virus infections were found in 2 patients, respectively. Antiviral treatments were used in 20 patients (53%), whereas other patients' symptoms improved without antiviral treatments. In all cases, hospital admission was not required because of influenza infections. Twenty-nine patients (76%) recovered from the influenza symptoms within 7 days. Seven patients continued to have symptoms for more than 7 days. As expected, most of these infections occurred in young patients (Fig. 1). Almost quarter of the IBD patients (12 of 45 patients, 26.7%) who were less than 20 years old were infected with the H1N1 influenza virus. The incidence of H1N1 influenza infections in patients who were less than 20 years old was significantly higher than that of other patients (Fig. 1) (P < 0.01, OR: 6.9, 95% CI: 3.2–14.9).

Figure 1

H1N1 influenza virus infection in patients with IBD (n = 38). Patients whose ages were < 20 years were more frequently affected with H1N1 influenza infections than other age groups.

3.2 Risk factors for H1N1 influenza infections in IBD patients

For H1N1 influenza infections, the patient's clinical features and the use of medications are provided in Table 2. Because H1N1 influenza virus infection was mainly investigated in this study, 2 patients who were infected with Type B influenza virus were excluded. Although the patients in the control group were not matched to cases based on gender, duration of disease, type of disease (UC or CD), or the use of medications for IBD, these factors were comparable in both groups (Table 2). In our study, only 13% of the cases (5 of 38 patients) received H1N1 vaccines before they developed influenza infections. These 5 patients received a dose of H1N1 vaccination and all vaccinations were given within 2 weeks prior to H1N1 influenza infections. In the control group, 12 of 76 patients (15.7%) received H1N1 vaccines before the index dates. This indicated that most of the H1N1 influenza infections were spread before the H1N1 vaccine was available from 2009 to 2010.

View this table:
Table 2

Risk factors for H1N1 influenza infections in IBD patients.

Univariate analysis
H1N1 influenza (+)H1N1 influenza (−)p valueOR95% CI
(n = 38)(n = 76)
Mean age (yrs)30.0 +/− 13.230.7 +/− 13.20.79
Gender (male: female)25:1349:270.891.060.47–2.40
Duration of disease (yrs)6.3 +/− 6.47.8 +/− 6.70.36
Current smoking2 (5.3%)11(14.4%)0.120.330.07–1.56
Disease0.711.170.51–2.70
UC2650
CD1226
H1N1 influenza vaccinations5 (13.2%)12 (15.7%)0.470.810.26–2.49
Treatment
5-ASA31(81.6%)58 (76.3%)0.521.370.52–3.65
Steroids6 (15.8%)16 (21.0%)0.500.700.25–1.97
Thiopurine8 (21.0%)23 (30.2%)0.300.610.25–1.54
Infliximab6 (15.7%)13 (17.1%)0.860.910.31–2.61
Any IM22 (57.9%)46 (60.5%)0.780.900.41–1.98
Any two IM7 (18.4%)17 (22.3%)0.630.780.29–2.09
  • IM: Immunomodulators included any use of steroids, thiopurine, infliximab, tacrolimus, or methotrexate.

To assess the effect of immunomodulators on H1N1 influenza infections, we considered the use of thiopurine, infliximab, tacrolimus, and methotrexate to be the use of any immunomodulators (any IM). The percentage of patients who used steroids or thiopurine was comparable between the cases and the control group (Table 2). We also confirmed that the percentage of patients who used any 2 IM (e.g., infliximab and 6-MP) was also comparable between both groups. In the present study, no patients needed hospitalization due to H1N1 influenza infection, although the steroids, thiopurine, and infliximab were used. The percentage of the need for antiviral treatments and duration of H1N1 influenza symptoms in patients with or without any IM has been compared (Table 3).

View this table:
Table 3

Duration of H1N1 influenza symptoms and rate of use of antiviral treatments.

Duration of influenza symptomsNeed for antiviral treatments
< 7 days≧ 8 days or 8 days ≦(−)(+)
(n = 29)(n = 7)(n = 17)(n = 20)
Mean age (years)28.8 +/− 12.431.4 +/− 8.128.1 +/− 10.229.7 +/− 13.1
Use of any IM18 (62%)3 (43%)11 (65%)11 (55%)
  • IM: immunomodulators.

  • Information regarding both duration of influenza symptoms in 2 patients and that regarding need for antiviral treatment was lacking in a patient.

4 Discussion

In the spring of 2009, 2 cases of human infections with influenza A (H1N1) were reported, and the WHO declared a pandemic phase level 6.1 In Japan, the first patients were reported at the end of April. The Japanese Ministry of Health, Labour and Welfare reported that approximately 140,000 persons were infected and 127 patients died. This terrible pandemic also caused IBD patients to become anxious, especially the IBD patients who were on immunomodulators and biologics. In this study, the incidence of H1N1 influenza infections in IBD patients was investigated. Most of the H1N1 influenza infections were observed in younger patients. Our study indicated that the incidence of H1N1 influenza infections during the 2009–2010 season was 6.7% in IBD patients. National Infectious Disease Surveillance Center reported that estimation of H1N1 incidence was 8.4% (approximately 10.7 million patients) in general population. The incidence of H1N1 influenza infections in our patients (6.7%, 95% CI 4.6–8.8%) was comparable to that in Japanese general population.

For decades, influenza infections occurred predominantly in older and immunocompromised patients. However, our study indicated that the incidence of H1N1 influenza infections was significantly higher in patients who were < 20 years than that among other patients. Several studies reported that most patients for H1N1 influenza infection were aged from 5 to 18 years in general population.1416 The vast majority of H1N1 influenza cases in all country had occurred among adolescents and young adults. The median age of pandemic (H1N1) 2009 influenza-infected patients was lower than that of seasonal influenza A (H3N2)-infected patients.14 Another study also reported that the percentage of positivity of pandemic (H1N1) 2009 influenza virus was highest in 5–18 year age groups (OR 2.5 for > 35 years age group) when compared to seasonal influenza.13 Data compiled by WHO indicated that approximately 25% of cases were aged 0–9 years, 36% were aged 10–19 years, 17% were aged 20–29 years, 9% were aged 30–39 years, 7% were aged 40–49%, and 5% were aged > 50 years.17 These results consisted with our result indicating that age distribution of H1N1 influenza infections between IBD patients and the general population was comparable (Fig. 2).

Figure 2

Distribution of H1N1 influenza infections among age groups (15–19, 20–29, 30–39, 40–49, 50 ≦ years-old ) in IBD patients and general population. This was cited from a report from the Japanese Infectious Disease Surveillance Center. Patients or children aged < 15 were excluded in the figure.

We failed to find that the incidence of H1N1 influenza infections in the 2009–2010 season was higher in patients who were on any immunomodulators than in patients who were not on any immunomodulators. These results may indicate that having a less healthy immune system does not imply an increased risk for H1N1 influenza infection during pandemic exposure while patients who were on immunomodulators might increase the risk of complications of H1N1 influenza infection.

In this study, 5 of 38 patients received the H1N1 vaccines before the influenza infections. One possibility of non-response to the vaccine in these patients was due to short duration between vaccination and H1N1 influenza infections.

There were limitations to this study. First, we could not distinguish H1N1 infections from seasonal influenza infections. From August 2009 to March 2010, the Infectious Disease Surveillance Center reported that influenza AH1 (Russian type), AH3 (Hong Kong Type), B and H1N1 infections had occurred in 18 (0.06%), 153 (0.52%), 35 (0.12%) and 28,939 (99.29%) patients, respectively. Thus, most of the influenza type A viruses in our study may have been H1N1 viruses. Second, we only included cases of H1N1 influenza infections that occurred after enrollment in this study to obtain more exact information from our prospective observations. It is assumed that some H1N1 influenza infections that occurred prior to enrollment were missed. Finally, some patients with H1N1 influenza infections who only showed symptoms may also have been missed. However, only influenza-like symptoms misled our diagnoses because it is difficult to distinguish between influenza and influenza-like bronchitis or pharyngitis.

Our study indicated that the incidence of H1N1 influenza infections in 2009–2010 was 7 times higher in IBD patients aged < 20 years than other IBD patients. Younger IBD patients would increase the probability of transmission due to the close proximity at school and university, compared to other populations.14 Furthermore, older patients may have cross-immunoreactivity for H1N1 pandemic influenza virus.14,18,19 Previous study indicated that pandemic 2009–2010 H1N1 influenza infections were associated with a higher percentage of severe cases and mortality.20 Other report also showed that 28% of H1N1 influenza infected-IBD patients needed to be hospitalized and 3 (12%) of 25 patients were sent to intensive care unit.21 On the other hand, our study showed that it was relatively mild among IBD patients and no hospitalization was required. The probable explanation for this result is frequent visits to the hospital among IBD patients. As a result, early diagnosis could be done in most cases and antiviral treatment might be given at the appropriate time.

Conflict of interest statement

Potential competing interest: None.

Financial support: This work was supported in part by Health and Labour Sciences Research Grants for research on intractable diseases from the Ministry of Health, Labour and Welfare of Japan.

Acknowledgments

The authors would like to express their thanks to Nature Publishing Editing for editing the manuscript for English language readability.

Specific author contributors

Makoto Naganuama (MN) and TF have equally contributed to this study and were involved in conceptualizing the study. MN and TF designed the main part of this study and drafted the manuscript. MN, TF, RK, NY, YT, and Masakazu Nagahori were involved in the acquisition of data. K.A. and T.T. advised us on the statistical analyses. MW organized the study and provided specific comments. All authors read and approved the final manuscript.

References

View Abstract