Skip Navigation

Oculo-respiratory Syndrome: A New Influenza Vaccine-Associated Adverse Event?

  1. Danuta M. Skowronski1,
  2. Barbara Strauss1,4,
  3. Gaston De Serres5,
  4. Diane MacDonald1,
  5. Stephen A. Marion2,
  6. Monika Naus1,
  7. David M. Patrick1, and
  8. Perry Kendall3
  1. 1University of British Columbia Centre for Disease Control, University of British Columbia, Vancouver
  2. 2Department of Health Care and Epidemiology, University of British Columbia, Vancouver
  3. 3Office of the Provincial Health Officer, Ministry of Health Planning, Victoria, British Columbia
  4. 4Health Canada, Population and Public Health Branch, Field Epidemiology Training Program, Ottawa, Ontario
  5. 5Institut National de Santé-Publique de Québec, Quebec, Canada
  1. Reprints or correspondence: Dr. Danuta M. Skowronski, UBC Centre for Disease Control, 655 W. 12th Ave., Vancouver, BC V5Z 4R4, Canada (danuta.skowronski{at}bccdc.ca).
 
Next Section

Abstract

During the 2000–2001 influenza immunization campaign in Canada, a new adverse event, oculo-respiratory syndrome (ORS), was noted in association with administration of vaccine supplied by one manufacturer. The original case definition for ORS specified bilateral conjunctivitis, facial edema, or respiratory symptoms beginning 2–24 h after influenza vaccination and resolving within 48 h after onset. To characterize the spectrum, severity, and impact of ORS, we contacted persons who had reported any influenza vaccine-associated adverse event in British Columbia, Canada, during the 2000–2001 vaccination campaign. With use of a standardized telephone interview, we collected information from 609 (79%) of 769 eligible persons. Thirteen percent of ORS-affected persons reported onset ⩽2 h after vaccination, 27% experienced symptoms for >48 h, and 42% considered the symptoms to be severe. The surveillance case definition for ORS for 2001–2002 was revised to include onset ⩽24 h after vaccination, with no restriction on duration. ORS should be incorporated into annual influenza vaccine safety monitoring.

Three influenza vaccines are supplied to Canadians by 2 manufacturers. These are Fluviral S/F, distributed in Canada by Shire Biologics (formerly Biochem Pharma), and Vaxigrip and Fluzone, both distributed internationally by Aventis Pasteur. All 3 vaccines are egg-based, split, inactivated, and trivalent products administered intramuscularly. During the 2000–2001 immunization campaign, 12.2 million doses of influenza vaccine were distributed across Canada on a provincial basis; 31% of the doses were Fluviral S/F [1, 2]. The province of British Columbia (BC), Canada, purchased 990,000 doses of Fluviral S/F for its public vaccination campaign.

Vaccine-associated adverse events (VAAEs) are reported voluntarily to the BC Centre for Disease Control (BCCDC; Vancouver). Within days after the start of the 2000–2001 vaccination campaign, the BCCDC received reports of bilateral red eyes and respiratory symptoms that occurred within a few hours after influenza vaccination. Most, but not all, reports came from health care workers who were vaccinated early, during an expanded campaign to protect health care providers [3]. Shortly thereafter, similar reports were received from different regions across the province and involving other targeted groups. Notification to Health Canada (Ottawa, Ontario) and national discussion confirmed that this syndrome was occurring in a number of provinces in association with one influenza vaccine. National enhanced surveillance and special investigations were developed to assess this association.

On the basis of cases that emerged early in the vaccination campaign,“oculo-respiratory syndrome” (ORS) was defined in 2000 as bilateral red eyes or respiratory symptoms (including 1 or more of the following symptoms: cough, wheeze, chest tightness, difficulty breathing, or sore throat) or facial edema beginning 2–24 h after influenza vaccination and resolving ⩽48 h after onset [1, 4]. From October 2000 through March 2001, 2450 adverse events associated with influenza vaccination were reported nationally; 1735 persons had reported ocular, respiratory, or facial symptoms, and 960 met the ORS case definition for 2000 (ORS,2000) [1, 4]. Ninety-six percent of cases that met the ORS,2000 case definition occurred after receipt of Fluviral S/F [1]. Epidemiologic investigations confirmed this causal link [5]. Retrospective cohort investigation confirmed an association between ORS and age and sex, with women aged 50–59 years having the highest attack rates (16%) [6]. Association with first-time receipt of influenza vaccine was also noted [5]. National summaries derived from passive surveillance described a generally mild and self-limited syndrome. On the basis of this information, the risks of influenza itself, and concerns about potential vaccine shortage, recommendations supported continued use of all available vaccines [1].

Passive surveillance was largely based on narrative description from numerous sources, and enhanced national surveillance did not systematically assess severity. To gauge the clinical profile, severity, and impact of ORS with use of a standardized check-list approach, the BCCDC recontacted all persons who submitted an influenza VAAE report during the 2000–2001 season. The Provincial Health Officer of British Columbia mandated this review.

Previous SectionNext Section

Methods

Twelve interviewers were trained to conduct a standard telephone survey using a prepiloted questionnaire; interviews were conducted from 19 September through 13 October 2001. Inclusion criteria specified that subjects must be persons aged ∼18 years who made an influenza VAAE report during the 2000–2001 campaign in British Columbia and who provided sufficient locator information.

Data on specific clinical symptoms were solicited, participants were asked about the symptoms' severity, and severity was scored using numeric and Likert scales. The numeric scale was scored from 0 (“no symptoms”) to 10 (“felt life-threatening”). The Likert scale included the categories“mild”,“moderate,” and “severe.” Participants were asked whether they intended to be revaccinated in 2001–2002 using a 5-choice Likert scale (“very likely”,“likely,”“undecided,”“unlikely,” and“very unlikely”).

The χ2 test of significance (defined as P <.05) was used for categorical variables, and the Kruskal-Wallis test was used for numeric variables. Logistic regression assessed independent effects and adjusted ORs for perceived severity and prolonged duration. The variables considered were age; sex; current smoking status; underlying lung-associated, eye-associated, allergic, or chronic conditions; previous influenza vaccination; and onset of symptoms >24 h after vaccination. To assess recall up to 1 year after vaccination, survey responses were compared with the same individual's earlier passive surveillance report with respect to meeting the ORS,2000 case definition; the ϰ statistic was calculated as the measure of agreement.

Previous SectionNext Section

Results

Response rate and baseline characteristics. In 2000–2001, there were 898 influenza VAAE reports to the BCCDC, compared with 49 and 92 in 1998–1999 and 1999–2000, respectively. Of the 898 VAAEs, 769 (86%) met inclusion criteria. Of the 769 affected persons, 609 (79%) completed an interview. Respondents did not differ from nonrespondents with regard to age or sex.

Respondents were grouped into 3 clinical categories: (1) respondents with cases that met the national ORS,2000 case definition (“the ORS,2000 group”; 281 persons), (2) those who reported ORS-defining symptoms outside of the timelines specified by the national ORS,2000 case definition (“the ORS-like group”; 204 persons), and (3) those who reported “any other” symptom (“the AOS group”; 124 persons). Results are provided separately for the ORS,2000 and ORS-like groups, but they are also provided in a single combined category of“ORS” (485 persons).

Persons in all clinical categories had similar baseline characteristics, with the exception of age and underlying condition (table 1). Persons in the ORS-like group were significantly older and more likely to have a chronic condition than were persons in the ORS,2000 group. More persons in the AOS group than the ORS group were in the very old age group (i.e., age of ∼80 years; 6% vs. 1%, respectively; P <.001). Persons in the ORS,2000 group were significantly more likely to have reported their symptoms in association with first-time receipt of influenza vaccine than were persons in the ORS-like and AOS groups (53%, 32%, and 28%, respectively; P <.001).

Figure 1
View larger version:
Figure 1

A, Time from influenza vaccination to onset of oculo-respiratory syndrome (ORS) events. B, Duration of ORS events. ORS-like, persons who reported ORS-defining symptoms outside of the timelines specified by the ORS,2000 case definition; ORS,2000, persons who met the ORS case definition for 2000.

Figure 2
View larger version:
Figure 2

Respondents' intention of being revaccinated based on perceived severity of oculo-respiratory syndrome

Table 1
View larger version:
Table 1

Baseline characteristics of participants in a study of oculo-respiratory syndrome (ORS), according to adverse event category.

Symptom profile. The symptom profiles for persons in the ORS-like group were similar to those for persons in the ORS,2000 group (table 2). Seventy percent of persons affected by ORS experienced ∼2 categories of ocular, respiratory, or facial symptoms—most often oculo-respiratory symptoms.

Table 2
View larger version:
Table 2

Symptom profile, by adverse event category, among participants in a study of oculo-respiratory syndrome (ORS).

Persons who reported ORS overall (i.e., the ORS,2000 and ORS-like groups combined) also reported more systemic (78% vs. 57%; P <.001) and gastrointestinal (46% vs. 23%; P <.001) symptoms than did persons in the AOS group. In particular, persons in the ORS group had significant differences with regard to fever/chills (P <.001), general aches (P <.001), headache (P <.001), fatigue (P <.001), diarrhea (P =.04), stomach pain (P =.04), loss of appetite (P <.001), and difficulty swallowing (P <.001), compared with persons in the AOS group. Conversely, fewer local injection-site complaints were cited by persons with ORS than by those in the AOS group (47% vs. 61%, respectively; P =.006).

The abrupt onset (figure 1A) and resolution (figure 1B) specified by the ORS,2000 case definition are smoothed into more-gradual distributions when time restrictions are relaxed. Both the onset and duration distributions remained skewed toward the left overall, with a median time of onset of 4 h after vaccination and a median duration of 24 h. Only 50 (10%) of 485 persons with ORS reported that the onset of symptoms occurred >24 h after vaccination (24 [5%] reported that onset occurred >48 h after vaccination), but 62 (13%) of the respondents reported that the onset of symptoms occurred <2 h after vaccination, and 132 (27%) reported a duration of symptoms of >48 h (22 [5%] reported a duration of symptoms of >1 month).

As determined by logistic regression (n = 465), prolonged duration of ORS was associated with delayed onset of >24 h (OR, 14.0; 95% CI, 6.64–31.64; P <.001), previous influenza vaccination (OR, 2.9; 95% CI, 1.78–4.63; P <.001), history of lung disease (OR, 1.9; 95% CI, 1.07–3.32; P =.03), and allergies (OR, 1.6; 95% CI, 1.04–2.62; P =.04). There was no interaction noted between sex or age and other variables.

Severity of symptoms. Only a minority (20%) of persons with ORS described symptoms as “mild”; 42% described symptoms as“severe” (table 3). On the numeric scale, 32 persons (7%) with ORS rated their symptoms a 10. Twenty persons who rated their symptoms a 10 provided a rationale: 12 (60%) described difficulty breathing, 5 (25%) described a sensation similar to a heart attack, and 3 (15%) reported collapse.

Table 3
View larger version:
Table 3

Severity of adverse event experience, by adverse event category, among participants in a study of oculo-respiratory syndrome (ORS).

Visits to the emergency department were more common among persons with ORS than among persons in the AOS group (13% vs. 3%; P =.002). The proportion of persons with ORS symptoms rated a 10 ranged from 3% (7 of 216) for persons aged <50 years without a chronic condition to 11% (8 of 73) for persons aged ∼50 years with a chronic condition, and the proportion was up to 13% (3 of 23) for persons aged ∼65 years with a chronic condition. For visits to an emergency department, comparable rates were 12% (26 of 217), 16% (12 of 73), and 21% (5 of 23), respectively; for hospitalization overnight, the rates were 1% (3 of 217), 8% (6 of 73), and 12% (3 of 23), respectively. Overall, among persons with ORS, 8 (2%) of 484 were hospitalized for ∼3 days.

As determined by logistic regression (n = 461), the perception that ORS was severe was associated with a history of lung disease (OR, 2.1; 95% CI, 1.26–3.48; P =.004), being a current smoker (OR, 1.8; 95% CI, 1.09–2.89; P =.02), and age as a continuous variable (OR, 1.02; 95% CI, 1.00–1.03; P =.02). No interaction was noted between sex or age and other variables.

Two hundred six (42%) of 485 persons reporting ORS, compared with 40 (32%) of 124 in the AOS group, considered themselves “unlikely” or “very unlikely” to receive influenza vaccine in 2001 (P =.04). Willingness to be revaccinated was associated with mild ORS (P <.0001; figure 2). Two-thirds (22 of 32) of the persons who rated ORS symptoms a 10 on the numeric scale were“unlikely” or “very unlikely” to be vaccinated again, compared with one-quarter (3 of 12) of persons who rated symptoms a 1 (P =.01).

Validity assessment. There was good agreement between the findings of this survey and reports to the passive surveillance system from almost 1 year earlier. Of the 609 questionnaires completed, 510 (84%) had responses consistent with the original report for ORS,2000 status, as defined by symptoms, onset, and duration of ORS (ϰ = 0.7). Only 20 reports (4% of participants who reported ORS) were mismatched because they did not cite ORS-defining symptoms, either on the questionnaire (11 reports) or on the original report (9 reports). Removal of these individuals from analyses did not alter results.

Previous SectionNext Section

Discussion

During the 2000–2001 influenza immunization campaign, a newly identified ORS emerged in Canada as an adverse effect associated with one manufacturer's vaccine. This signal was detected within days of starting the immunization campaign—a credit to the sensitivity of vaccine adverse event surveillance in Canada. In part, rapid notification was related to the fact that health care workers were the first to be immunized. Their easy access to the VAAE reporting network brought symptoms to the prompt attention of public health authorities. Many of these health care workers had been vaccinated for the first time—a promotional campaign in British Columbia in 2000 increased immunization coverage among this cohort of ∼127,000 workers from 43% in the 1999–2000 season to 57% in the 2000–2001 season [3].

ORS emerged as a discrete syndrome in Canada in 2000, but conjunctivitis and respiratory symptoms have been linked to influenza vaccine previously. Since 1969, 22 countries have submitted to the World Health Organization ∼3000 adverse event reports of conjunctivitis, respiratory symptoms, or facial edema with an onset ⩽24 h after influenza vaccination (E. Walette, Uppsala Monitoring Centre [Uppsala, Sweden]; personal communication). However, no causality assessment is made of adverse events reported to the Uppsala Monitoring Centre; further investigation and research is needed for their full interpretation. (The opinions and conclusions are not necessarily those of the various reporting centers or the World Health Organization) In 1995–1996, a cluster of adverse events, including conjunctivitis and dyspnea, was identified in several European countries in association with a third manufacturer's influenza vaccine. In Italy, this resulted in a significant 10-fold increase in the number of adverse event reports [7].

Time restrictions set by the case definition early in the 2000 campaign in Canada were ultimately inconsistent with population distributions for onset or duration of symptoms. In particular, we found that restriction of symptom duration resulted in an underestimation of the frequency and severity of ORS. Similar findings were observed in a separate review of reports to the passive surveillance system for influenza vaccine-associated adverse events in Quebec [8]. The National Advisory Committee on Immunization (NACI; Ottawa) has since revised the case definition for ORS in 2001–2002 and beyond to include the same clinical features (and, in addition, hoarseness and difficulty swallowing) but to allow symptom onset any time ⩽24 h after influenza vaccination, with no restriction on symptom duration [5]. Ninety percent of the 485 affected persons in our survey meet this revised NACI definition.

The precise cause of ORS remains unknown, although a plausible explanation exists. Product-related investigations found no contamination of the implicated product, and all vaccines used in Canada in 2000 were completely inactivated. Electron microscopy, however, found a greater proportion of high-molecular weight compounds of unsplit and aggregated virions in the implicated vaccine [5]. Similar morphologic aberrations were found in the implicated vaccine in Europe in 1995–1996 [9]. Modifications were made to the manufacturing process, and a trial that used the reformulated version of the vaccine in Europe for 1996–1997 revealed a return of that vaccine's safety profile to match other influenza vaccines, although excess“allergic-type” reactions persisted [7].

A randomized, double-blind, placebo-controlled trial with reformulated Fluviral S/F for the 2001–2002 season in Canada also showed that the risk of persistent, vaccine-attributable ORS was low (2.8%; 95% CI, 0.5%–5.1%) but that the syndrome occurred significantly more frequently after administration of Fluviral S/F than after administration of placebo [1, 10]. Although ORS has “allergic-type” features, skin test evaluation has shown that ORS is not anaphylactic in nature [11]. Instead, a distinct but as-yet undetermined immunologic mechanism is probably involved [12]. ORS shares clinical features with influenza, an illness that is due to mucosal and systemic immune response to virus that is mostly confined to the respiratory tract, rather than a direct effect of viremia [13]. In addition to respiratory symptoms, 20% of persons with influenza also experience ocular symptoms, including conjunctivitis; gastrointestinal symptoms are also sometimes experienced [13]. This constellation of symptoms may be an important clue to the immunologic mechanism of ORS.

We found that older persons and those with chronic conditions (particularly lung disease) were more likely to report severe and/or prolonged ORS. These persons are also known to be at increased risk for the severe complications of influenza itself. Clinically significant exacerbation of underlying chronic respiratory disease has previously been cited in association with influenza vaccination, but this link has recently been questioned [14, 15]. ORS may reinforce the biologic plausibility of this association. We cannot rule out the possibility that the subgroup of patients with underlying chronic conditions experienced coincident exacerbation of underlying illness mimicking ORS that was temporally, but not causally, related to vaccination. It is not surprising, however, to learn that older and more frail persons reported more difficulty and a longer time to recovery from a syndrome that causes significant respiratory symptoms even in healthy persons.

Some caution should be applied in interpreting the results of this survey. Reported adverse events often represent the worst cases, and not all reports in temporal association with vaccine are causally related. Other etiologies should be considered, particularly because individual symptoms of ORS are nonspecific. Our survey may provide an overly negative perspective of the true adverse event experience; it should be emphasized that most persons who received vaccine during the 2000–2001 season did not report any symptoms. Although the number of reported cases is an underestimation of the true rate of ORS occurrence [6], unreported cases may have been much milder. Our survey was conducted almost 1 year after vaccination, and this may have affected symptom recall. Strong agreement between the responses to our survey and the original reports to the passive surveillance system is reassuring in this regard. Finally, this was a cross-sectional survey conducted via telephone; more-rigorous methods of assessing ORS during the problematic season would have been preferred. Randomized, double-blind, placebo-controlled trials using vaccine reformulated for the 2001–2002 season confirmed the persistence of a statistically significant association between influenza vaccination and ORS, although the impact was noted to be milder than in 2000–2001 [1, 10, 16].

Nearly one-half of the respondents in our survey of persons reporting an influenza vaccine-associated adverse event that qualified as ORS said that they would self-exclude themselves from revaccination. A review of vaccine uptake by this cohort conducted during the 2001–2002 campaign has confirmed a reluctance to be reimmunized that has important implications [17]. ORS should be incorporated into routine vaccine safety surveillance in all countries, and manufacturing safeguards should be considered to limit the aggregate and unsplit virion content of all influenza vaccines administered annually, to reduce the risk of ORS and to improve acceptance of this important and potentially life-saving vaccine.

Previous SectionNext Section

Acknowledgements

We would like to acknowledge the contributions of public health authorities throughout British Columbia—in particular, Terry Dickson, Linda Poirier, and Larry Gustafson of South Fraser Health Region (Langley, BC). We acknowledge the national centers that contributed data to the World Health Organization (WHO) International Drug Monitoring Programme-Uppsala Monitoring Centre (Uppsala, Sweden). The opinions and conclusions, however, are not necessarily those of the various centres or the WHO.

Previous SectionNext Section

Footnotes

  • Financial support: University of British Columbia Centre for Disease Control.

  • Received October 7, 2002.
  • Accepted November 25, 2002.
Previous Section
 

References

  1. ↵
    1. 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. Can Commun Dis Rep 2001;27:1-7.
    Medline
  2. ↵
    1. Health Canada Bureau of Infectious Diseases
    . Oculo-respiratory syndrome in association with the influenza vaccine: Canada, October-November 2000. Can Commun Dis Rep 2000;26:1-2.
    Medline
  3. ↵
    1. British Columbia Ministry of Health and Ministry Responsible for Seniors
    . Influenza report for British Columbians: 2000/2001. Victoria, BC: British Columbia Ministry of Health and Ministry Responsible for Seniors; 2001. p. 6-7.
  4. ↵
    1. National Advisory Committee on Immunization
    . Statement on influenza vaccination: continued use of Fluviral influenza vaccine in the 2000–2001 season. Can Commun Dis Rep 2001;27:1-3.
    Medline
  5. ↵
    1. National Advisory Committee on Immunization
    . Statement on influenza vaccination for the 2001–2002 season. Can Commun Dis Rep 2001;27:13-8.
    Medline
  6. ↵
    1. Boulianne N,
    2. De Serres G,
    3. Duval B,
    4. Shadmani R,
    5. Rochette L
    . Clinical manifestations and incidence of oculo-respiratory syndrome following influenza vaccination— Quebec, 2000. Can Commun Dis Rep 2001;27:85-90.
    Medline
  7. ↵
    1. Spila-Alegiani S,
    2. Samaso S,
    3. Rota MC,
    4. Tozzi AE,
    5. Raschetti R
    . Reactogenicity in the elderly of nine commercial influenza vaccines: results from the Italian SVEVA study. The Italian SVEVA Group. Vaccine 1999;17:1898-904.
    CrossRefMedlineWeb of Science
  8. ↵
    1. De Serres G,
    2. Grenier JL,
    3. Toth E,
    4. et al
    . The clinical spectrum of the oculo-respiratory syndrome after influenza vaccination. Vaccine. in press.
  9. ↵
    1. Tumova B,
    2. Schramlova J,
    3. Vitkova E,
    4. Sulova V,
    5. Radovnicky V
    . Programs and abstracts of Options for the Control of Influenza III (Cairns, North Queensland, Australia). 1996. Is the release control of influenza vaccine sufficient [poster P5–14; abstract p82]?; p. 82.
  10. ↵
    1. Scheifele DW,
    2. Duval B,
    3. Russell M,
    4. et al
    . Ocular and respiratory symptoms attributable to inactivated influenza vaccine: evidence from a controlled trial in adults. Clin Infect Dis. (in press).
  11. ↵
    1. Skowronski DM,
    2. De Serres G,
    3. Hebert J,
    4. et al
    . Skin testing to evaluate oculo-respiratory syndrome (ORS) associated with influenza vaccination during the 2000–2001 season. Vaccine 2002;20:2713-9.
    CrossRefMedlineWeb of Science
  12. ↵
    1. Skowronski DM,
    2. Lu H,
    3. Warrington R,
    4. et al
    . Does antigen-specific cytokine response correlate with the experience of oculo-respiratory syndrome following influenza vaccine? J Infect Dis 2003;187:495-9.
    Abstract/FREE Full Text
  13. ↵
    1. Nicholson KG,
    2. Webster RG,
    3. Hay AJ
    1. Nicholson KG
    . Human influenza. In: Nicholson KG, Webster RG, Hay AJ, editors. Textbook of influenza. London: Blackwell Science; 1998. p. 219-64.
  14. ↵
    1. Nicholson KG,
    2. Webster RG,
    3. Hay AJ
    1. Wiselka MJ
    . Vaccine safety. In: Nicholson KG, Webster RG, Hay AJ, editors. Textbook of influenza. London: Blackwell Science; 1998. p. 350-1.
  15. ↵
    1. American Lung Association Asthma Clinical Research Centers
    . The safety of inactivated influenza vaccine in adults and children with asthma. N Engl J Med 2001;345:1529-36.
    CrossRefMedlineWeb of Science
  16. ↵
    1. Skowronski DM,
    2. De Serres G,
    3. Warrington R,
    4. et al
    . Final program and abstract book of the 5th Annual Conference on Vaccine Research (Baltimore). Bethesda, MD: National Foundation for Infectious Diseases; 2002. Oculo-respiratory syndrome following influenza vaccine: recurrence risk with re-formulated vaccine for 2001–2002 [abstract S43]; p. 61.
  17. ↵
    1. Skowronski DM,
    2. Strauss B,
    3. Kendall P,
    4. Duval B,
    5. De Serres G
    . Low risk of recurrence for persons previously affected by oculo-respiratory syndrome (ORS): implications for influenza re-vaccination. CMAJ 2002;167:853-8.
    Abstract/FREE Full Text
| Table of Contents

This Article

  1. Clin Infect Dis. (2003) 36 (6): 705-713. doi: 10.1086/367667
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 54 (5)
  1. Clinical Infectious Diseases
  1. Alert me to new issues

Most