Oculo-respiratory syndrome following influenza vaccination: Review of post-marketing surveillance through four influenza seasons in Canada

Introduction

Influenza vaccines were approved for use in Canada in the late 1970s. The current vaccines are inactivated trivalent split-virus products. The World Health Organization updates these products yearly to provide the best protection for influenza viruses, predicted to circulate in the upcoming influenza season in the Northern hemisphere. Each year, influenza may affect an estimated 10% to 25% of the population and is usually the predominant cause of serious respiratory illness in a community, especially among the elderly and in individuals with chronic medical conditions⁽¹⁾. In Canada, annual immunization programs have focused mainly on those at high risk of serious influenza complications and their contacts, including healthcare workers⁽¹⁾. However, even in healthy children and adults, there is considerable morbidity. Since the 1998-1999 influenza season, the National Advisory Committee on Immunization (NACI) has been encouraging all individuals who wish to protect themselves against influenza to receive the vaccine, even though they are not in a high-risk group⁽²⁾. Since 2000-2001, Ontario has been offering publicly funded vaccine to all age groups.

Serious adverse events (AEs) to influenza vaccination are rare⁽³⁾. Historically, Health Canada (the Public Health Agency of Canada) received, on average, 200 case reports of adverse events following immunization (AEFI) during each influenza season, of which most were local reactions. Early in the 2000-2001 season, however, Health Canada received verbal notification from two provinces - Alberta and British Columbia - of an unusual number of AEFI reports that described a constellation of respiratory symptoms, accompanied by complaints of "red eyes." This phenomenon, given the name "oculo-respiratory syndrome" (ORS), was defined as the presence of at least one of the following symptoms: red eyes, cough, wheeze, chest tightness, difficulty breathing, sore throat, or facial swelling, occurring within 2 to 24 hours of influenza vaccination and resolving within 48 hours of symptom onset⁽⁴⁾. As a result of publicity and enhanced surveillance, a total of 2,450 reports of AEFI vaccine were documented for that season, with 960 (39.2%) meeting that season's established ORS case definition⁽⁵⁾.

Although ORS was eventually linked to all influenza vaccines in use in Canada, the vast majority of cases reported in 2000-2001 involved a single manufacturer's product, in which electron microscopy studies revealed higher proportions of micro-aggregates of unsplit virions⁽⁶⁾. However, in subsequent seasons, after the indicated vaccine was reformulated with the same splitting agent used in other influenza vaccines, AE reports that met the ORS case definition were still reported in association with all influenza vaccines.

This paper describes the experience with ORS in Canada between 2000 and 2004 and the results of a retrospective review of cases reported during previous influenza seasons (1997-2000) that may have met the case definition of ORS.

Methods

Adverse event surveillance system

In Canada, the reporting of AEFI by health care providers (mainly public health nurses and physicians) to local, provincial, and/or territorial public health authorities is the cornerstone of post-marketing surveillance of vaccines. Reports, including those completed by manufacturers, are forwarded to the Agency for aggregation and analysis.

In October 2000, Health Canada, in consultation with provincial/territorial health authorities, initiated intensified surveillance of influenza vaccine safety after being alerted to an increased number of AEs following influenza vaccination. A supplementary questionnaire to collect more detailed information on reported ORS AEs was developed and disseminated during the next three influenza seasons (2000-2001 to 2002-2003). Although it was discontinued in the 2003-2004 influenza season due to decreases in the number of reports, ORS continued to be closely monitored through the standard AEFI form.

Data obtained from standard and supplementary forms were entered into the Canadian AEFI (CAEFI) database. Vaccine distribution data, by province and lot number, were obtained from influenza vaccine manufacturers. These data included the number of returned doses at the end of each influenza season, providing an approximate number of doses administered (net doses distributed) and enabling the estimation of ORS reporting rates.

ORS case definition

During the study period, the ORS case definition underwent modifications. For the 2001-2002 season, NACI revised the case definition to include an expanded clinical spectrum and time frame, because the more restrictive definition may have led to potential reports of ORS being missed in the first season. ORS was redefined as the presence of one or more of the previous signs and symptoms listed in the 2000 definition, but with the following two additions: 1) difficulty swallowing and hoarseness and 2) occurring within 24 hours of influenza vaccination, but with no restriction on symptom duration⁽⁷⁾. NACI recommended that the same case definition be used in the subsequent seasons⁽⁷⁾. We used this definition, the "2001 case definition," for ORS-related data provided in this paper for all influenza seasons. It was also applied to estimate the number of potential cases of ORS during the influenza seasons between 1997 and 2000.

Selected subjects and variables

Case reports meeting the 2001 ORS case definition were retrieved from the CAEFI database. Demographic information - including age at vaccination and sex, as well as other variables, such as con-comitant vaccines given, medical attention sought, and outcome - were extracted. Frequency and reporting rates were obtained for each of the nine ORS-related symptoms. These symptoms were examined in five major groups: 1) only ocular (red eyes), 2) only respiratory, 3) only facial swelling, 4) both ocular and respiratory with or without facial swelling, and 5) other combinations. Reporting rates were estimated per million net doses of vaccine distributed in each season. The total number of reports of influenza AEFI were also extracted and used to estimate the proportion of ORS reports versus all AEs reported following influenza vaccination in each season. The additional data that were analyzed, when available, comprised the following: previous history of ORS, severity of symptoms, and possible risk factors. For the percentage calculation of variables, missing values were included, unless specified otherwise. Results are provided in aggregate form for all four influenza seasons (2000-2001 to 2003-2004), unless important differences were noted between seasons. In addition, we present the most frequent non-ORS symptoms reported concomitantly, and we describe the frequencies and reporting rate estimates of ORS cases uncovered during the seasons between 1997 and 2000.

Finally, we provide a summary of the recent literature on ORS in Canada, including clinical trials, telephone surveys, skin testing, animal study, and prospective and retrospective observational studies.

Some data in this paper regarding the influenza AEFI reported in 2000-2001 and 2001-2002 may differ from those published previously for the following reasons:

• Different case definitions were applied over the various seasons, based on increasing knowledge of the newly identified event.
• The Agency continues to receive and accept reports of AEFI for vaccines administered in previous years, and these reports received after the influenza season and thus not included in previous analyses are captured in this study.
• Additional duplicate reports of AEs have been identified and eliminated.

Results

ORS reported during the influenza seasons 2000-2001 to 2003-2004

Over four influenza seasons, a total of 3,264 reports of influenza AEFI met the case definition for ORS. Most subjects were between age 45 and 64 years (1,424, 44%) (Figure 1). The ages (mean and median) were 49 years (range 6 months to 97 years). Of the total, most were female (2,448, 75%). Female predominance was observed across all age groups, except in children who were < 15 years (representing only 3% of cases), wherein the female-male ratio was 1:1. Of these individuals, 93% had received only influenza vaccine. Among those who received other vaccines simultaneously, pneumococcal vaccine was the one most frequently administered (75%).

Of all individuals who reported ORS (n = 3,264), 986 (30%) sought medical attention; that is, 17 (0.5%) visited emergency rooms; 30 (0.9%) were hospitalized; and 939 (29%) were seen in outpatient settings. Among those admitted to hospital, 15 (50%) had only respiratory symptoms, 10 (33%) had both ocular and respiratory symptoms, and three (10%) reported only facial swelling. More males than females were reported hospitalized (1.5% vs. 0.8%). The percentage and rate (per total number of persons in each age group) of hospitalization were highest among elderly persons (37%, 2% of elderly). There were two fatal case reports among persons with ORS. The first was a 72-year-old male with a previous history of a febrile reaction to influenza vaccine and who developed red eyes and respiratory symptoms, fever, headache, fatigue, and myalgia 12 hours post-vaccination, but did not seek medical attention. He remained febrile for 8 days, was diagnosed with pneumonia and died 54 days post-vaccination due to complications, despite medical care. The second fatal case was a 66-year-old male who developed eye irritation, sore throat, rhinitis, and difficulty breathing within one day of vaccination and died suddenly 8 days later. There was insufficient evidence to determine a causal association for either case.

Figure 1. Age-group distribution (years) of oculo-respiratory syndrome cases for 2000-2001 to 2003-2004 influenza seasons (n = 3,264)

Figure 2. Reporting number and rate per million vaccine doses for cases meeting the oculo-respiratory syndrome definition (1997-1998 to 2003-2004)

A total of 43,195,187 net doses of vaccine were distributed in the public and private sectors across Canada during the four influenza seasons (an overall ORS reporting rate of 76 per million doses [Table 1 and Figure 2]). The frequency and reporting rate per million doses of influenza vaccine were estimated for each of the nine ORS-related symptoms (Table 1). The relative frequency for some of these symptoms varied over the four influenza seasons, but red eyes and cough were the two most commonly reported in all seasons. In addition, the rates of ORS frequency and reporting were assessed in five major groups: only ocular, only respiratory, only facial swelling, oculo-respiratory, and other combinations (Table 1). Only 29% (955) of individuals reported both ocular and respiratory symptoms.

Time to onset following vaccination for symptoms of ORS ranged from 1 minute to 24 days (mean 8.5 hours, median 5 hours). Duration of symptoms ranged from 1 minute to 3 months (mean 2.7 days, median 1 day).

Additional information, obtained through the supplementary questionnaire during the first three influenza seasons (2000-2001 to 2002-2003), included self-reported severity of symptoms, history of previous influenza vaccination, and risk group (as the main reason for immunization). Of those reporting severity (1,078, 33% of total), 363 (34%) described symptoms as minor (easily tolerated); 408 (38%) as moderate (interfered with regular daily activities); and 307 (28%) as severe (unable to take part in regular daily activities). Thirty-five percent (1,142) reported a history of previous influenza vaccination(s). Of these, 16% (178) had received the vaccine ≥ 2 times in the past. In 51% (1,662) of reports, risk group was recorded; 50% (825) of these were in a high-risk group; 6% (91) were in contact with high-risk individuals; and 34% (569) were healthcare workers. Of the individuals in the high-risk groups, 64% (528) were aged ? 65 years. One-half of the contacts of high-risk groups (46) were between age 25 and 44 years.

Of the other AEs reported concurrently with ORS symptoms, the 10 most frequent were as follows: fever (609), headache (608), fatigue (593), rhinitis (498), myalgia (435), chills (432), allergic reactions (341), local reactions (262), nausea (239), and dizziness (193). The reporting rates for these symptoms ranged from 14 (fever) to four (dizziness) per million net doses of vaccine distributed.

Potential cases of ORS prior to the 2000-2001 season (1997-1998 to 1999-2000)

When the 2001 ORS case definition was applied to cases in the CAEFI database from 1997 to 2000, a total of 21 reports in 1997-1998, 10 in 1998-1999, and 27 in 1999-2000 met the case definition of ORS. The reporting rates per million doses of distributed vaccine were four, two, and five, respectively (Figure 2).

Table 1. Frequency and rate per million vaccine doses for reported AEFI and ORS groups by influenza season

Discussion

Ocular and respiratory symptoms following influenza vaccination occurred in the past in Canada, in the U.S., and in several European countries but remained unrecognized until 2000, when excess reports of such symptoms in Canada resulted in ORS being viewed as a potentially distinct AE. Reports of this syndrome were later found in the Canadian AEFI database prior to the 2000-2001 season. A similar constellation of AEs occurred during the 1995-1996 influenza vaccination campaigns in Italy and several other European countries⁽⁸⁾. At that time, a ten fold increase in AE reports had been detected in Italy in association with two lots of a split influenza vaccine being widely used in Europe. Following this event, a prospective observational study was conducted in Italy to compare the reactogenicity of nine different influenza vaccines (whole, split, or subunit) in 16,714 individuals who were aged ≥ 65 years⁽⁸⁾. Although local reactions and systemic symptoms were most frequent, many reports of conjunctivitis and respiratory symptoms, classified as "allergy-like," were detected. Those who received whole vaccines were at higher risk of developing AEs, compared with those who received split or subunit vaccines. Most AEs, however, were mild or moderate, and no severe reactions were considered vaccine-related.

In the U.S., a search of the Vaccine Adverse Event Reporting System (VAERS) for the period 1990 to 2002, using coding terms mapped to the Canadian case definition, found a number of reports meeting the criteria for ORS⁽⁹⁾. Similar to the experience in Canada - because of the nonspecific and relatively benign nature of its constituent symptoms and the lack of being raised as a signal for investigation - it remained unrecognized as a potentially unique syndrome.

In Canada, ORS reports made up about one-half of all influenza vaccine AE case reports during the past four influenza seasons. Generally, ORS is considered to be mild and self-resolving. Of those who were admitted to hospitals or seen in emergency rooms, most were elderly, and the frequent complaints included difficulty in breathing and chest tightness. Outpatient visitors and individuals who had not sought medical assistance more frequently reported red eyes and cough. Only 29% of all ORS cases reported a combination of red eyes and respiratory symptom(s). Of the other AEs reported by ORS cases, mild constitutional symptoms and local reactions were, as expected, the most common.

Several clinical studies attempted to better assess the risk of ORS, showing that first-time influenza vaccine recipients developed ORS more often than those who were previously vaccinated⁽¹⁰⁾. Similarly, individuals who were affected previously by ORS had a higher risk of developing the syndrome after revaccination, compared with previously unaffected vaccine recipients^(11,12) , whereas persons who were previously affected tended to develop a higher number of ORS symptoms on revaccination, compared with new cases⁽¹²⁾. The risk of ORS recurrence was as low as 5% in one study⁽¹³⁾ and up to 45% in others^(11,12). Fortunately, most patients described their ORS recurrence as mild⁽¹⁴⁾ or less severe than the original episode⁽¹³⁾. ORS risk appeared higher among those vaccinees who were aged 40 to 59 years^(15,16). Although this middle-aged group reported ORS more frequently, recurrence of ORS occurred more frequently among those aged $ 60 years⁽¹⁴⁾. Prolonged duration of ORS symptoms (> 48 hours) was usually seen among those with a previous history of influenza vaccination, history of lung disease, and allergies⁽¹⁷⁾.

The vaccine implicated during the 2000-2001 season was reformulated to reduce the proportion of unsplit virion. Subsequently, the frequency of ORS was similar to that reported following administration of the other influenza vaccines available in Canada⁽¹²⁾. This suggests that oculo-respiratory symptoms may occur in association with any influenza vaccine^(10,12).

The pathophysiologic mechanisms of ORS are not yet understood, though some studies conducted in Canada suggest the following:

• the complement system may be involved in the red eyes(18) ;
• ORS is not an anaphylactic reaction(18) ; and
• higher aggregate content vaccine may lead to increased immunogenecity of influenza vaccine, compared with those vaccines that contain a lower proportion of aggregates⁽¹⁹⁾.

The results obtained from the surveillance database need to be interpreted with caution. The CAEFI surveillance system is based mainly on voluntary reporting, which differs within each province and territory. Further, the supplementary form, designed to provide additional information on ORS characteristics, was not completed for the majority of the case reports. Among those completed, data were self-reported. In other words, a severe AE may appear mild to some subjects, but severe to others, and recall bias may have affected the results, depending on when the forms were completed. Under-reporting, especially of milder symptoms, is another source of reporting bias. Finally, the non-specific nature of the ORS case definition, although designed to allow sensitive case capture, presents significant limitations. A controlled clinical trial indicated that the frequency in reporting a single ORS AE was higher (though non-significant) among placebo recipients than among vaccine recipients⁽¹⁰⁾. Likewise, the frequency of reporting multiple ORS symptoms was higher in individuals who experienced recurrent ORS (found to be at higher risk of developing ORS) than those reporting for the first time⁽¹²⁾. The use of a more specific case definition - a combination of red eyes and at least one respiratory symptom - that better defines ORS as a syndrome may assist in more targeted investigations of the etiology and pathogenesis of ORS in the future (Table 1).

In conclusion, ORS is likely not a new AE but, rather, one that was underreported in the past, due to its mild nature, and that was unmasked only when its frequency increased during the 2000-2001 influenza season in two provinces. A variation in the process of manufacturing one of the vaccine products in 2000-2001 that resulted in a greater than usual number of unsplit virions in the final product may have led to its increased "reactogenecity" that season. Subsequent publicity and enhanced surveillance efforts have since contributed to the increase in ORS reporting, along with all AE, following influenza vaccination. The impact of this "stimulated reporting" of AEFI after influenza vaccine slowly declined during subsequent seasons, with reporting rates only now returning to the pre-ORS levels.

Acknowledgements

The authors of this paper gratefully acknowledge all individuals who were involved in the surveillance of influenza AEFI, including vaccine recipients, healthcare providers, provincial and territorial public health officials and vaccine manufacturers, and the vaccine safety staff at the Agency. The authors also thank the National and International scientists and experts for their attendance at the ORS-related meetings and workshops and for their guidance. Finally, the authors wish to acknowledge Dr. Barbara Law for review of the manuscript.

References

1. National Advisory Committee on Immunization. Statement on influenza vaccination for the 2004-2005 season. CCDR 2004;30(ACS-3):1-32.

2. National Advisory Committee on Immunization. Statement on influenza vaccination for the 1998-1999 season. CCDR 1998;24(ACS-2):1-8.

3. National Advisory Committee on Immunization. Influenza vaccine. In: Canadian Immunization Guide. 6^th Edition. Ottawa: Health Canada, 2002:120-7.

4. National Advisory Committee on Immunization. Supplementary Statement on influenza vaccination: Continued use of Fluviral® influenza vaccine in the 2000-2001 season. CCDR 2001;27(ACS-1, 2):1-3.

5. National Advisory Committee on Immunization. Supplementary statement for the 2001-2002 season: Influenza vaccination of persons who experienced oculo-respiratory syndrome following previous influenza vaccination. CCDR 2001;27(ACS-7):1-8.

6. Choudhri Y, Walop W. Influenza vaccine-associated adverse events: Results of passive surveillance, Canada 2001-2002. CCDR 2002;28(23):1-8.

7. National Advisory Committee on Immunization. Supplementary statement for the 2002-2003 influenza season: Update on oculo-respiratory syndrome in association with influenza vaccination. CCDR 2002;28(ACS-6):1-8.

8. Spila-Alegiani S, Salmaso S, Rota MC, et al. Reactogenicity in the elderly of nine commercial influenza vaccines: Results from the Italian SVEVA study. Study for the evaluation of adverse events of influenza vaccination. Vaccine 1999; 17(15-16):1898-1904.

9. Khromava A, Pool V, Chen R. Oculo-respiratory syndrome following influenza vaccine - United States, 1990-2002: New or previously unrecognized ?1 st international conference on therapeutic risk management & 19th international conference on pharmacoepidemiology August 21-24, 2003, Philadelphia, PA, USA. Pharmacoepidemiology and Drug Safety 2003;12(1):S59.

10. Scheifele DW, Duval B, Russell ML, et coll. Ocular and respiratory symptoms attributable to inactivated split influenza vaccine: Evidence from a controlled trial involving adults. Clin Infect Dis 2003;36(7):850-7.

11. SkowronskiDM,DeSerresG,HebertJ,etcoll. Skin testing to evaluate oculo-respiratory syndrome (ORS) associated with influenza vaccination during the 2000-2001 season. Vaccine 2002;20(21-22):2713-9.

12. De Serres G, Boulianne N, Duval B, et coll. Oculo-respiratory syndrome following influenza vaccination: Evidence for occurrence with more than one influenza vaccine. Vaccine 2003;21(19-20):2346-53.

13. Skowronski DM, Strauss B, Kendall P, et coll. Low risk of recurrence of oculo-respiratory syndrome following influenza revaccination. CMAJ 2002;167(8):853-8.

14. Skowronski DM, De Serres G, Scheifele D, et coll. Randomized, double-blind, placebo-controlled trial to assess the rate of recurrence of oculorespiratory syndrome following influenza vaccination among persons previously affected. Clin Infect Dis 2003;37(8):1059-66.

15. Boulianne N, De Serres G, Duval B, et coll. Manifestations cliniques et incidence du syndrome oculo-respiratoire survenant après la vaccination contre l'influenza - Québec, 2000. RMTC 2001;27(10):85-90.

16. DeSerres G,Grenier JL, Toth E, et coll. The clinical spectrum of the oculo-respiratory syndrome after influenza vaccination. Vaccine 2003; 21:2354-61.

17. Skowronski DM, Strauss B, De Serres G, et coll. Oculo-respiratory syndrome: A new influenza vaccine-associated adverse event? Clin Infect Dis 2003;36(6):705-13.

18. Fredette MJ, De Serres G, Malenfant M. Ophthalmological and biological features of the oculorespiratory syndrome after influenza vaccination. Clin Infect Dis 2003;37(8):1136-8.

19. Babiuk S, Skowronski DM, De Serres G, et coll. Aggregate content influences the Th1/Th2 immune response to influenza vaccine: Evidence from a mouse model. J Med Virol 2004;72(1):138-42.

Source: Nooshin Ahmadipour, MD, MSC, Theresa Tam, MD, FRCPC, Wikke Walop, PhD, Arlene King, MD, MHSC, Immunization and Respiratory Infections Division, Robert Pless, MD, MSc, Public Health Training and Applications Division, Public Health Agency of Canada, Ottawa, Ontario.