Wasp sting mortality in Australia

Despite the importance of fatalities from Hymenoptera stings internationally, little has been published on the circumstances and incidence of such deaths in Australia. For example, an analysis of Australian bee sting fatalities during 1960-1981 was unable to obtain adequate clinical information about most of the fatalities during that period, and coronial records were examined for only one of the 27 fatalities identified in that study.² Despite the increasing abundance of the introduced and aggressive European wasp (V. germanica),⁶ and consequent concerns about the increasing health risks posed by this vespid,³ even less information is available about fatalities from wasp stings in Australia than for those from bee stings. Indeed, only two minimally documented reports of fatal wasp stings exist in the Australian medical literature. In 1913, Cleland noted "oedema of the larynx and subsequent death occurred from a man being stung by a wasp. He had been drinking out of the spout of a waterbag in which the wasp was hiding".⁷ Lee also made reference to a fatal wasp sting in a 47-year-old man in southern Queensland in 1957.⁸

Analysis of fatalities can provide important information on the determinants of severe morbidity, which may be useful in injury prevention. Thus, we investigated, in detail, wasp sting related fatalities that occurred in Australia over the past two decades. This period was coincident with the arrival and dispersal of V. germanica on the Australian mainland.⁶ In the absence of national coronial data for this period, we used State-based coronial information systems. In particular, we aimed to identify the contribution of allergy versus venom toxicity (envenomation), and the likely contribution of V. germanica to the mortality burden.

• the Australian Bureau of Statistics mortality dataset;

• State and Territory Registrars of Births, Deaths and Marriages;

• State and Territory coronial authorities; and

• in one instance, from the family physician of the deceased.

These data allowed us to analyse the circumstances of the wasp stings, the likely wasp species involved, and any history of Hymenoptera sting allergy and its management.

This project was approved by the University of Melbourne Health Sciences human ethics subcommittee.

The seven fatalities identified during the 20-year period correspond to a mean annual incidence rate of 0.02 deaths per million population per year. The deceased were all men living in northern New South Wales, or in Queensland (from the southeast to the far north coast). They were aged between 39 and 73 years (mean age, 54 years). Five of the seven deaths occurred during December and February, and the remaining two deaths occurred in April and May. In no year was there more than one death from wasp sting. All stings occurred in rural areas and four of the deceased were described as farmers in the coronial reports. The details of each case are summarised in Box 1.

In Europe^4,10 and the United States,¹¹ relative proportions of deaths from wasp and bee stings are different from those in Australia. In Sweden, these proportions are 90% for wasps and 10% for bees;⁴ corresponding figures for Denmark are 63% for wasps and 37% for bees,¹⁰ while, in the US, wasp and bee sting fatalities occur with approximately equal frequency.¹¹ By contrast, our finding of a wasp sting fatality rate of 0.02 per million population per year is only a quarter of the reported Australian bee sting fatality rate of 0.086 per million population per year.² This may reflect the relatively recent arrival in Australia of the European wasp⁶ compared with the European honey bee.

Although our study was precipitated by concern about the health impact of European wasps, and we looked at a period that coincided with their arrival and spread on the mainland, we identified no fatalities attributable to this wasp. All of the wasp-sting deaths occurred outside the known range of V. germanica.⁶ This may change as increasing numbers of people become sensitised to its venom. Further research is required to ascertain whether the arrival of V. germanica has altered sting-related morbidity patterns in Australia.

As in previous studies,^2,4,10 in the deaths we examined we found that the reponsible wasp was rarely formally identified. Nonetheless, it seems likely that at least one case was attributable to a Polistes species. This is consistent with reports of severe allergic reactions to stings by Polistes and Ropalidia species in Queensland.¹² Moreover, given the distribution of the various vespid wasps,⁹ it seems likely that all seven fatalities were caused by Polistes and Ropalidia species. This adds impetus to the call for purified Ropalidia venom to become available for immunotherapy,¹² as, currently, only imported Polistes and Vespula venoms are available.

The absence of wasp sting mortality outside Queensland and NSW contrasts with the nationwide occurrence of deaths from bee stings.² It is possible that the wasp species found in Queensland and NSW possess greater allergenic potential than those found elsewhere in Australia. No systematic research has yet been undertaken on this question. The formal identification of the wasp species implicated in future fatalities would facilitate such research.

Consistent with previous analyses of bee and wasp sting mortality internationally,^2,4,5 we found a low number of stings, rapid onset of symptoms, no child fatalities and over-representation of middle-aged men among those who died. It has been proposed that pre-existing heart disease might explain the increased risk for men aged over 40 years;^2,4 certainly, this may have contributed to death in two of our cases. However, our findings differ from these other reports in the high frequency of autopsy examinations and low rates of underlying cardiorespiratory disease.

Venom toxicity is rarely reported in Hymenoptera-related fatalities,^2,4,10,11 and we found no fatal massive envenomations. In all cases death was attributable to anaphylaxis (although ideally mast cell tryptase and IgE level measurements should be made to confirm this diagnosis). In contrast with previous international reports, most of the victims in our study had either previously diagnosed allergy to wasp stings or a clear history of systemic reactions indicating severe allergy. Thus, most of these deaths might have been prevented by specific immunotherapy, early treatment with adrenalin or both. However, patients and their families should be aware that vulnerability to venom-induced anaphylaxis persists at least until maintenance doses of immunotherapy are attained.

As with Australian bee sting fatalities,² most deaths occurred in the summer months when human-wasp interaction is most likely. The rural setting of these fatalities highlights concerns about the acute shortage of specialist allergists and clinical immunologists in rural areas,¹³ although it is unclear whether such shortages contributed to the fatal outcomes. Nevertheless, this underservicing, combined with the lack of Ropalidia,¹²Myrmecia¹ and other venoms for immunotherapy, increases the importance of anaphylaxis emergency kits and education¹⁴ for those with life-threatening allergy to insect stings and their families.

The fact that most of the deceased were reported to be farmers is consistent with the fact that rates of work-related deaths in agriculture are among the highest in Australia.¹⁵ Information on bites and stings should therefore be incorporated within injury prevention programs such as those developed by Farmsafe Australia.¹⁵

As noted in previous investigations of bee sting mortality,² dependence on a single data set is likely to lead to an underestimation of the total injury burden. It is possible that additional wasp sting deaths occurred during the study period but were not recorded because of miscertification, misattribution as bee sting, or other inadequate or inaccurate recording. We found that the wasp sting diagnosis was almost wholly dependent on the presence of a witness and absence of retained bee sting on autopsy. Clearly, unwitnessed wasp sting deaths could be attributed to other causes, and our findings should be considered an underestimate of wasp sting mortality.

1. Clarke PS. The natural history of sensitivity to jack jumper ants (Hymenoptera formicidae Myrmecia pilosula) in Tasmania. Med J Aust 1986; 145: 564-566.
2. Harvey P, Sperber S, Kette F, et al. Bee sting mortality in Australia. Med J Aust 1984; 140: 209-211.
3. Levick NR, Winkel KD, Smith GS. European wasps: an emerging hazard in Australia. Med J Aust 1997; 167: 650-651.
4. Johansson B, Eriksson A, Ornehult L. Human fatalities caused by bee and wasp stings in Sweden. Int J Legal Medicine 1991; 104: 99-103.
5. Langley R, Morrow W. Deaths resulting from animal attacks in the United States. Wild Environ Med 1997; 8: 8-16.
6. Spradbery JP, Maywald GF. The distribution of the European or German wasp in Australia, past, present and future. Aust J Zool 1992; 40: 495-510.
7. Cleland JB. Insects and their relationship to disease in man in Australia. Trans 9th Aust Med Congress Sydney 1911; 1: 548-552.
8. Lee DJ. Arthropod bites and stings and other injurious effects. Sydney: School of Public Health and Tropical Medicine. The University of Sydney, 1975.
9. Naumann ID. Hymenoptera. Chapter 42. In: CSIRO. The insects of Australia. 2nd ed. Melbourne: Melbourne University Press, 1991: 916-1000.
10. Mosbech H. Death caused by wasp and bee stings in Denmark 1960-1980. Allergy 1983; 38: 195-200.
11. Barnard J. Studies of 400 Hymenoptera sting deaths in the United States. J Allergy Clin Immunol 1973; 52: 259-264.
12. Solley G. Allergy to stinging and biting insects in Queensland. Med J Aust 1990; 153: 650-654.
13. O'Hehir RE, Douglass JA. Stinging insect allergy. Med J Aust 1999; 171: 649-650.
14. Douglass JA, O'Hehir RE. Peanut allergy. Education, avoidance and adrenaline are the mainstays of management. Med J Aust 1997; 166: 63-64.
15. Fragar L. Agricultural health and safety in Australia. Aust J Rural Health 1996; 4: 200-206.

Research Centre for Injury Studies, Mark Oliphant Building, Flinders University, Bedford Park, SA.
James Harrison, MB BS, MPH, Director.

Reprints will not be available from the authors.
Correspondence: Dr K D Winkel, AVRU, Department of Pharmacology, University of Melbourne, VIC 3010.
k.winkelATpharmacology.unimelb.edu.au

Make a comment