Thunderstorm asthma outbreak of November 2016: a natural disaster requiring planning

Learning from a tragedy to increase public awareness and improve responses in future thunderstorm asthma events

Thunderstorm asthma is the occurrence of acute asthma either during or immediately after a thunderstorm and it is often characterised by a surge in emergency asthma presentations. The epidemic of thunderstorm asthma in Melbourne, Australia, on 21 November 2016 was the most extreme such event ever worldwide, with nine fatalities currently the subject of a coronial inquiry.1,2 Hospitals and ambulance services were placed under record pressure, and supplies of reliever medications were exhausted at some health services.1 Key tasks for the future are to predict the thunderstorms most likely to lead to asthma outbreaks and to define how best to respond. Further research to better anticipate these outbreaks is crucial and planning for the inevitable recurrence must occur at patient, institutional and state-wide levels.

Mechanisms of thunderstorm asthma outbreaks

While multiple theories have been advanced to explain thunderstorm asthma outbreaks, most agree that thunderstorms and aeroallergens are two necessary, although insufficient, conditions for such events. A widely accepted, though unproven, explanation for events such as those seen in Melbourne is that grass pollen from pastures outside a city are swept into the air by gusty winds on a hot spring day, raising the background concentration of pollen carried towards an affected area.3 Afternoon thunderstorms then develop with cool, descending air pushing out ahead of the storms in a shallow and distinct turbulent outflow, which further concentrates the airborne pollen grains.4 The sudden increase in humidity with the thunderstorm and possibly within the outflow may facilitate osmotic shock of the pollen grains, releasing vast numbers of allergenic starch granules small enough to penetrate into the lower respiratory tract and bring about a much more severe asthmatic response than would result from inhalation of intact pollen grains.3,4 Indeed, previous reports suggest that as many as one-third of people affected in such events had not been diagnosed with asthma previously.5

Patients with seasonal allergic rhinoconjunctivitis (SAR) or allergies to grass pollen appear to be more at risk of thunderstorm asthma.5 The disproportionate frequency of outbreaks in south-eastern Australia is thought to be due to grass pollen,3 and perennial rye grass (Lolium perenne) — the dominant pasture grass in Victoria — is the primary source of outdoor aeroallergens in Melbourne. Treatment of allergic rhinitis is generally associated with a reduction in asthma exacerbations, although specific data for thunderstorm asthma are not available.6 The peak pollen and SAR season in Melbourne falls in November,2,7 and all documented thunderstorm asthma outbreaks in Melbourne have occurred in that month (Box 1).

The immunological mechanisms underpinning SAR due to pollen allergy include aeroallergen interaction with antigen-presenting cells leading to cognate T lymphocyte stimulation.11 This induces the production of allergen-specific IgE, which binds to airway mast cells and other inflammatory cells, readying them to induce inflammation on allergen re-exposure.11 Allergen-induced responses in the lung and nose are closely related, with allergic rhinitis likely priming the lower respiratory tract to increased hyper-responsiveness to allergens.12

High grass pollen counts frequently co-exist with high fungal spore counts. Airway inflammation due to fungal particles may occur through mucosal activation of the innate immune system and may be synergistic with grass pollen exposure. Sensitisation to fungal allergens is more common in people living in the more arid, inland regions of Australia than in coastal dwellers.13 Fungal spores are implicated overseas in epidemic asthma (especially in dry climates), although they have not yet been implicated in asthma outbreaks in Australia.3,14

The day of the storm

On 21 November 2016, the weather was hot (up to 34°C), dry (as low as 20% relative humidity) and windy (northerly at 35–45 km/h) — an archetypal “extreme” grass pollen day (> 100 pollen grains/m³) (Box 1). The average daily grass pollen concentration in Melbourne to 3:30 pm that day was 102–210 grains/m³ (varying with location), with the average fungal spore count of 366 spores/m³ at the University of Melbourne site being composed mostly of spores of Cladosporium, Alternaria and smuts (a group of fungi that often infect plants). Around 5:30 pm, a line of thunderstorms and showers oriented in a north-south direction moved eastwards across the Melbourne region accompanied by a marked outflow with strong wind gusts, a sharp fall in temperature and an equally sharp rise in relative humidity (Box 2). The outflow also caused a short-lived wave of airborne particulate matter.2 There was light rainfall (< 1 mm) in some areas, but lightning activity generally dissipated over the western suburbs before the change reached the city. Pollen concentrations, both grass and non-grass, fell by 30–50% around the time of the weather change and remained low thereafter (Box 2 and Box 3).

Disaster planning

Box 1 shows that, although the 21 November event was without precedent in its impact, it was far from being Melbourne’s first experience of thunderstorm asthma, underscoring the importance of proper diagnosis and management of asthma and allergic rhinitis. With at least six episodes in Melbourne in 33 years, this is unlikely to be the last.

At a patient level, a key message should be to encourage identification of asthma in people with SAR, along with education and the consideration of the prescription of inhaled corticosteroids to patients with both conditions, particularly for anyone who “sneezes or wheezes” in spring, as has been suggested earlier.1 Current asthma treatment guidelines recommend the use of inhaled corticosteroid treatment for patients who have all but intermittent asthma symptoms. Even low doses of inhaled corticosteroids are strongly associated with protection from severe exacerbations and asthma-related death.15

Short-acting β-2 agonist medication alone is only recommended for the management of mild asthma, where symptoms occur less than twice a month and without other indicators of severity.16 In patients with established asthma, especially patients with co-existent seasonal rhinitis, the lead-up to the grass pollen season should prompt review of allergic triggers, asthma preventer adherence for those already prescribed a preventer, asthma action plans and consideration of prescription of an asthma preventer therapy containing inhaled corticosteroid for those not already on a regular preventer. The provision of eformoterol and budesonide combination inhalers for use as needed to patients identified at risk of thunderstorm asthma is conceptually attractive, as they will coincidentally deliver preventer medication to patients finding benefit from the reliever effect, though this is not yet evidence based for the prevention of thunderstorm asthma. Similar theoretical utility and lack of evidence applies to the leukotriene receptor antagonist montelukast.

Allergen immunotherapy has evidence for improved asthma control when used for a range of allergens,17 although given the complexity and expense of this treatment, its prescription is limited to specialists in allergic disease. There is scant evidence supporting the use of immunotherapy for grass pollen-induced asthma and no evidence for the use of allergen immunotherapy in thunderstorm asthma. However, examination of the role of grass pollen immunotherapy in thunderstorm asthma treatment is reasonable given the likelihood of future therapeutic advances, including the availability of sublingual tablet immunotherapy, rendering allergen immunotherapy safer.

The major challenge at an institutional level will be to provide timely, consistent, guideline-based care and ensure appropriate follow-up in emergency departments suddenly affected by high caseloads and acuity. The rapid onset of thunderstorm asthma will likely prevent ambulance services from ever completely meeting the demands of such major events, although warning systems may assist. Increased community awareness of asthma first aid, freely available telephone advice, good rhinitis control and asthma action plans advocating appropriate reliever and preventer use leading to improved self-management practices may further mitigate the impact on patients. General practices, nurse-run hotlines and pharmacies could also receive education and support to allow them to assist with providing a rapid response during a crisis. In addition, pharmacies and hospitals may choose to maintain higher stocks of asthma reliever medication and spacers during spring.

At a state level, there is likely benefit in a public health campaign in mid-spring (at the beginning of the grass pollen season) to raise awareness of the risk of thunderstorm asthma in people with a history of asthma or SAR, and of asthma first aid in the general public. In addition, prediction and public notification of impending thunderstorm conditions occurring during periods of high pollen counts must now be treated akin to other extreme weather events. The Victorian Department of Health and Human Services and the Australian Bureau of Meteorology are developing a thunderstorm asthma prediction and warning system. It is planned that this system will enable warnings to be given to people at risk and will also alert emergency services and health care providers. As part of this system, additional aerobiological measurement sites are being established across Victoria. An important question will be how to avoid high false alarm rates, as it is evident that the coincident presence of thunderstorms and high aeroallergen loads are necessary, but not sufficient, to trigger an outbreak.

Warnings over traditional and social media should suggest at-risk individuals stay inside with windows and doors closed, take their asthma and rhinitis preventer medication, activate action plans and have reliever drugs at hand.1 The Murrumbidgee Local Health District in New South Wales has maintained an active thunderstorm asthma management strategy containing elements of these suggestions over the past 20 years, following an episode of thunderstorm asthma in Wagga Wagga in 1997.1,4,5 Their seasonal public health campaign may serve as an exemplar for the development of a similar system, but on the scale needed to serve the larger populations of capital cities or states.1

This disaster has shaken our health system. Learning from this tragedy, implementing strategies to recognise the finite capacity of our emergency response systems and building a community with good asthma and seasonal allergic rhinitis awareness and control should help us weather the storm.

Box 2 – Meteorological data recorded on 21 November 2016 at Melbourne Airport*

---

*A representative Bureau of Meteorology weather station about 20 km north-northwest of the city. Times shown are in Australian Eastern Daylight Time. The lowest panel shows hourly and bi-hourly counts of total, broken and grass pollen measured at the University of Melbourne.

Box 3 – Air sample collected at the University of Melbourne (Parkville campus) during the 21 November 2016 storm

---

Basic fuchsin staining has been used to enhance pollen grains (pink). Grass pollen grains, broken pollen grains and Cladosporium conidia are indicated (image courtesy of Dr Edwin Lampugnani, University of Melbourne).