Two cases of poisoning by raw taro leaf and how a poison control centre, food safety inspectors, and a specialty supermarket chain found a solution

Abstract

Although the taro plant, Colocasia esculenta, is commonly consumed throughout Asia, Africa, the Pacific Islands, and the Caribbean, its consumption is less common in North America. Exposure to raw or improperly prepared taro is associated with oropharyngeal irritation and swelling and, rarely, airway obstruction. Although cases of toxicity in countries where taro is a staple have been reported, cases in North America have not been described. Here, two cases of oral irritation and swelling in BC residents who ate raw taro leaf, were reported to the British Columbia Drug and Poison Information Centre (BC DPIC) and triggered an investigation involving a regional health authority and the Canadian Food Inspection Agency. Investigators found that the vendor, a chain of ethnic supermarkets, offered no point of sale preparation instructions. The vendor responded initially by posting instructional signage and later by voluntary product withdrawal. Analysis of BC DPIC records between 1 November 2011 and 20 December 2013 identified 11 cases of symptomatic taro exposure, five to the leaf and six to the corm. The two index cases and subsequent investigation illustrate how new foods or foods in unfamiliar contexts may present as calls to a poison control centre and that prevention requires collaboration among public and corporate stakeholders.

Key words: taro, food safety, poison control, health protection

Full text

TECHNICAL ARTICLE
Two cases of poisoning by raw taro leaf and how a
poison control centre, food safety inspectors, and a
specialty supermarket chain found a solution
John D. Omura*
$
, Christopher Blake
%
, Lorraine McIntyre
$
, Dorothy Li
§
and Tom Kosatsky
$
*University of British Columbia, School of Population and Public Health, Vancouver, BC.
$
Environmental Health Services, BC Centre for Disease Control, Vancouver, BC.
%
McMaster University, Hamilton, ON.
§
BC Drug and Poison Information Centre, BC Centre for Disease Control, Vancouver, BC.
Abstract: Although the taro plant, Colocasia esculenta, is commonly consumed throughout Asia, Africa, the Pacific Islands,
and the Caribbean, its consumption is less common in North America. Exposure to raw or improperly prepared taro is
associated with oropharyngeal irritation and swelling and, rarely, airway obstruction. Although cases of toxicity in
countries where taro is a staple have been reported, cases in North America have not been described. Here, two cases of
oral irritation and swelling in BC residents who ate raw taro leaf, were reported to the British Columbia Drug and Poison
Information Centre (BC DPIC) and triggered an investigation involving a regional health authority and the Canadian Food
Inspection Agency. Investigators found that the vendor, a chain of ethnic supermarkets, offered no point of sale
preparation instructions. The vendor responded initially by posting instructional signage and later by voluntary product
withdrawal. Analysis of BC DPIC records between 1 November 2011 and 20 December 2013 identified 11 cases of
symptomatic taro exposure, five to the leaf and six to the corm. The two index cases and subsequent investigation
illustrate how new foods or foods in unfamiliar contexts may present as calls to a poison control centre and that
prevention requires collaboration among public and corporate stakeholders.
Key words: taro, food safety, poison control, health protection
The taro plant, Colocasia esculenta, is a common staple
grown and consumed throughout Asia, Africa, the Pacific
Islands, and the Caribbean (Matthews 2004; Rao et al. 2010).
As raw consumption can result in an acrid taste accompanied
by swelling and irritation of the mouth and throat (Savage et al.
2009), corms, leaves, and possibly other parts (e.g., stalks) of the
taro are commonly eaten cooked. The potential for severe
reaction is illustrated by a case report from China that described
progression to airway obstruction following the ingestion of
wild Chinese taro (Alocasia cucullata) (Yuen 2001).
There is disagreement surrounding the exact mechanism by
which raw taro causes oropharyngeal irritation. It is largely
thought to stem from specialized plant cells called idioblasts,
which contain bundles of oxalate crystals called raphides coated
with a proteolytic enzyme. When the plant is crushed or
chewed, the idioblasts forcibly eject the raphides along with the
proteolytic enzyme into the oropharynx causing micro-trauma
and a local reaction (Bradbury and Nixon 1998; Noonan and
Savage 1999; Tagwireyi and Ball 2001; Paull et al. 1999; Savage
et al. 2009; Pang et al. 2010).
Taro contains both insoluble and soluble oxalates. Soluble
oxalates are more readily absorbed from the gastrointestinal
tract compared with the poorly absorbed insoluble oxalates and
are therefore more likely to cause systemic toxicity (Noonan
and Savage 1999). Rarely, renal failure and hypocalcemia (Sanz
and Reig 1992; Noonan and Savage 1999) have been associated
with consumption of plants containing oxalates. Only one case
(suicide attempt) of systemic toxicity attributed to taro leaf
ingestion was identified in the literature, with reported findings
including muscle spasms and renal insufficiency (Tagwireyi and
Ball 2001; Oscarsson and Savage 2007).
Distribution of taro products, particularly its leaves, is
uncommon in North America except through ethnic super-
markets. Consequently, a lack of public awareness regarding
proper preparation and cooking techniques could potentially
result in negative health consequences. The acridity of the taro
corms and leaves varies with species, age of plant, and growing
conditions. In some countries where cultivars are known to
have very low acridity, parts of the plant are eaten raw
(Bradbury and Nixon 1998; Rao et al. 2010). This report
Corresponding author: Tom Kosatsky (email: tom.kosatsky@bc
cdc.ca)
59
EHR Vol. 57(3) 5964 DOI: dx.doi.10.5864/d2014-027 Published on 9 October 2014
Environ. Health Rev. Downloaded from pubs.ciphi.ca by 209.139.228.161 on 11/04/14
For personal use only.

presents a series of taro poisoning cases, the process by which a
provincial poison control centre brought these cases to the
attention of health protection services, and the subsequent
public health response.
Event summary
Identification
In June 2013 the British Columbia Drug and Poison
Information Centre (BC DPIC) received a call from a healthy
middle-aged couple reporting symptoms after ingestion of raw
taro leaves. They had purchased the item the same day from an
Asian supermarket in Vancouver, Canada. Unfamiliar with
appropriate preparation methods, they consumed one to two
raw leaves the size of a hand. Shortly afterwards, both
experienced irritation and burning in the mouth, followed by
swelling at the back of the throat. All symptoms resolved
spontaneously within several hours. The couple did not seek
further medical care.
In November 2013 BC DPIC received another call from a
healthy adult male who had consumed raw taro leaf the evening
prior. The product had been purchased from the same
supermarket chain, but at a different location. He immediately
developed oral paresthesia and swelling, followed by emesis and
pain in his mouth and throat. His symptoms slowly resolved
through the night with the exception of prolonged swelling of
his tongue. BC DPIC staff managed these cases at home by
instructing callers to drink milk and use ice chips to reduce the
oral irritation. Follow-up calls were made to monitor progres-
sion or resolution of symptoms.
Following the occurrence of these relatively uncommon
cases over a short period, BC DPIC’s database (Visual Dotlab
Enterprise, WBM Software, Fresno, CA) was searched for
similar cases between 1 November 2011 and 20 December
2013. Eleven cases of symptomatic taro exposure were
identified over this period, five related to taro leaves and six
to taro corm (Table 1). Of the cases pertaining to leaf
exposure, all occurred between April and November 2013 and
followed ingestion. One of the cases involved partially
Table 1. Symptomatic case reports of taro exposure made to DPIC (1 November 2012 to 31 December 2013).
Year Month Region Gender
Taro
component
Method of
exposure Amount of exposure Symptoms
2012 January Lower
mainland
Female Corm Dermal
contact
Raw Dermal irritation
2012 May Lower
mainland
Female Corm Ingestion 30 g, raw Oral irritation within 5
minutes
2012 July Vancouver
Island
Male Corm Ingestion 30 g, raw Mouth burning
immediately
2013 February Lower
mainland
Male Corm Ingestion 34 bites raw Sore throat within 30
minutes
2013 March Vancouver
Island
Male Corm Ingestion Small piece raw Oral irritation within 20
minutes
2013 April Lower
mainland
Female Leaf Ingestion 2 leaves partially
cooked (no details)
Oral paresthesia within 30
minutes
2013 June Lower
mainland
Female Leaf Ingestion Few bites of food
cooked in taro leaf
Oral paresthesia
immediately
2013 June Lower
mainland
Male Leaf Ingestion 1 leaf (size of hand)
raw
Burning in mouth, oral
swelling
2013 June Lower
mainland
Female Leaf Ingestion 2 leaves (size of hand)
raw
Burning in mouth, oral
swelling
2013 November Lower
mainland
Male Leaf Ingestion 2 bites raw Oral irritation, swelling,
immediately; then
vomiting, sore throat,
lethargy
2013 December Lower
mainland
Female Corm Ingestion 4 bites of dried corm,
cooked (method not
mentioned)
Bitter taste and tingling
tongue soon after
60 EHR, Vol. 57, Issue 3
Environ. Health Rev. Downloaded from pubs.ciphi.ca by 209.139.228.161 on 11/04/14
For personal use only.

cooked taro leaf and one case involved ingestion of food
cooked in taro leaf with no details provided on the method of
cooking. Of the cases pertaining to corm exposure, five were
related to ingestion and one to dermal exposure. The five oral
exposures occurred between May 2012 and December 2013.
All five cases pertaining to leaf exposure were received from
BC’s lower mainland. This region is home to the largest
population of immigrants in the province (WelcomeBC
2008). Of the corm-related cases, four were received from
the lower mainland (Vancouver and surrounding region) and
two were from Vancouver Island. In addition, five asympto-
matic calls all pertaining to raw corm exposure were identified
(Table 2). All symptomatic and asymptomatic cases were
adults.
Results
Field investigation and communication
Following the initial report in June, BC DPIC staff contacted
food safety specialists at the BC Centre for Disease Control
(BCCDC). Finding an apparent lack of instructional in-
formation around taro leaf toxicity and safety practices on
government and regulatory websites, the BCCDC contacted
Health Canada and subsequently the Canadian Food Inspec-
tion Agency (CFIA). The CFIA referred the issue of taro
toxicity to their Fresh Fruits and Vegetables Program. Health
Canada staff brought the issue to the attention of the
Chemical Health Hazard Assessment Division of the Bureau
of Chemical Safety which recommended that a fact sheet be
developed.
With the emergence of the additional case in November,
discussions resumed between BCCDC and the CFIA. This
case identified the name and location of the supermarket store
where the product had been purchased. BCCDC notified and
requested assistance from the regional health authority, which
sent a food safety inspector to the store. At the time of
inspection, no cooking instructions were observed near the
product or on the product packaging (Figures 1, 2, and 3).
Further, employees who were interviewed provided conflicting
information regarding appropriate preparation methods: one
employee advised that taro leaf should only be used for display
or to infuse flavour into food and was not edible even after
cooking, whereas another employee advised that taro leaf was
safe to consume after cooking.
Mitigation
Six days after the initial inspection, the supermarket manager
voluntarily posted a sign instructing customers to cook taro
leaves prior to consumption (Figure 4). Two weeks later the
CFIA also sent an inspector to the supermarket and contacted
the chain’s head office to inquire about the product. At the end
of November the chain voluntarily discontinued sale of the
product. By December the CFIA began developing a consumer
fact sheet on natural toxins in taro leaves for the Natural
Toxins in Fresh Fruit and Vegetables section of their website.
Discussion
Given that consumption of raw taro is uncommon, literature
regarding exposure to the raw form is limited. Despite this,
several case reports highlight the medical significance of raw
taro exposure. For example a 45-year-old man in Hong Kong
developed mouth and throat pain and later upper airway
obstruction after consuming raw wild Chinese taro (Alocasia
cucullata or Colocasia cucullata) (Yuen 2001). It was not
specified whether he consumed taro corm or leaf. Consump-
tion of taro leaves is typically less common than corms, and raw
leaf consumption is presumably even less common. However, a
case report from Greece demonstrates the potential health
consequences of raw taro leaf exposure: a 2-year-old boy who
presumably consumed a portion of the houseplant leaf known
as Elephant’s Ear (Colocasia esculenta) developed acute drool-
ing, speech difficulties, and dysphagia (Mihailidou et al. 2002).
The patient recovered without any treatment.
In Zimbabwe, Elephant Ear is known as Madhumbe and is a
staple food in certain regions. A retrospective review of hospital
Table 2. Asymptomatic calls of taro exposure made to DPIC (1 November 2012 to 31 December 2013)
Year Month Region Gender
Taro
component
Method of
exposure Amount of exposure Symptoms
2011 December Lower
mainland
Female Corm Ingestion 1 mouthful raw Asymptomatic at 5
minutes
2011 December Lower
mainland
Male Corm Ingestion 1 mouthful raw Asymptomatic at 5
minutes
2012 October Lower
Mainland
Female Corm Ingestion 2 large raw; juiced with
vegetables and fruits
Asymptomatic at 15
minutes
2012 October Lower
Mainland
Male Corm Ingestion 2 large raw; juiced with
vegetables and fruits
Asymptomatic at 15
minutes
2013 February Lower
mainland
Female Corm Ingestion Small piece raw Asymptomatic at 30
minutes
Omura et al. 61
Environ. Health Rev. Downloaded from pubs.ciphi.ca by 209.139.228.161 on 11/04/14
For personal use only.

admissions in six referral hospitals in Zimbabwe from 1980
1989 due to food poisoning found 47 cases of taro leaf
poisoning, making it the third most commonly reported type
of food poisoning (Kasilo and Nhachi 1994). Moreover, the
majority of cases were among children, with 33 of the 47 (70%)
cases being under five years of age. Similarly, a retrospective
review from Hong Kong identified 21 adult cases of taro
poisoning between 2005 and 2009 (Pang et al. 2010),
constituting two-thirds of poison control calls for oral mucosal
irritation following plant ingestions. The part of the plant
consumed, corm or leaf, was not specified. Although both
reviews came from regions more commonly consuming taro
than North America, the frequency of events suggests that even
in areas of widespread taro availability and consumption,
symptomatic exposures to raw taro occur due to improper
preparation and consumption practices.
Lessons learned
Given the risk of problematic exposure to raw taro leaf
resulting from lack of awareness and increased availability of
Fig. 2. Taro leaf in product packaging as observed during
the grocery store inspection.
Fig. 3. Labelling of taro leaf product as observed during
the grocery store inspection.
Fig. 1. Taro leaf as observed during the grocery store
inspection.
Fig. 4. Point of sale signage as subsequently observed in
the grocery store.
62 EHR, Vol. 57, Issue 3
Environ. Health Rev. Downloaded from pubs.ciphi.ca by 209.139.228.161 on 11/04/14
For personal use only.

the product, mitigating exposure appears necessary as demon-
strated by the frequency of reported cases and successful actions
described here. In fact, no new cases of taro leaf exposure were
reported to BC DPIC from 21 December 2013 to 26 May
2014. Although many food preparation techniques for detox-
ifying taro leaves have been described, including grating and
macerating, mashing and pounding, baking, boiling, steaming,
washing in cold water, adding salt, and soaking and cooking in
milk (Matthews 2004; Oscarsson and Savage 2007; Kaushal
et al. 2013), the most often recommended method is cooking
the leaves in boiling water for at least 30 minutes (Savage and
Dubois 2006). The mechanism by which cooking decreases the
acridity of the plant is not fully understood, though it has been
suggested that applying heat denatures the raphide protease
thought to be responsible for the plant’s toxicity (Paull et al.
1999). Individuals new to using taro leaf may not be familiar
with these food preparation techniques. Compared with taro
leaves, taro corm availability and consumption are more
common in North America. Taro corm (Figure 5), when
cooked, has an appealing purple colour and is used in many
popular Asian dishes such as taro cake (named yu
`tou or yu
`na
˘i
in China) and in bubble teas. As with taro leaves, taro corm
preparation may involve washing in water, baking, boiling, or
other techniques prior to further preparation and consump-
tion. And just as consumers are unfamiliar with taro leaves,
they may also be unaware that taro corms need to be detoxified
before handling and consumption. As such, issues with
handling taro corms may also be of importance particularly
given the potential dermal reactions to skin contact with raw
corms. In general, additional interventions through public
health protection services are likely necessary to ensure that
consumers and food handlers are adequately informed regard-
ing preparation instructions, including information at the
point of sale. Instructions on the label packaging should
include the use of gloves while preparing these foods to prevent
dermal exposure, a warning not to consume raw taro leaves or
corms, and to details on how to thoroughly cook before
consumption.
In the case of oral exposure to raw taro leaf or corm, patients
should first be assessed for airway compromise. Should
significant oropharyngeal swelling or airway compromise occur,
patients should obtain emergent medical attention at a health
care facility (Yuen 2001). In the absence of severe symptoms,
patients should be instructed to immediately remove any plant
debris from the mouth by wiping with a wet cloth, rinsing
the mouth with water to flush out the crystals, then drinking a
full glass of milk or ingesting another calcium-containing food
or supplement (Yuen 2001). The calcium is thought to limit
absorption of soluble oxalates by forming insoluble calcium
salts (Hossain 2003; Brogren and Savage 2003). Cold items such
as popsicles, ice cream, or ice chips may provide symptomatic
relief for the local irritation. Patients or caregivers can further
monitor for worsening of symptoms during the first few hours
after oral exposure.
Future challenges
In the setting of ever changing urban Canadian environments,
this report demonstrates the potential health impact of
emerging food products with which health protection services
may have limited awareness or experience. The described cases
likely under represent the actual number of taro plant
poisonings in the province because of the mild and self-
resolving nature of most reactions. Health protection services
must remain vigilant to changing population demographics,
food trends and preferences, and the retail food market. This
can be particularly challenging in multicultural cities where
such food products may be sold in a limited set of specialty
stores.
Poison control centres offer a potential mechanism for
monitoring such trends, therefore providing an important
service to the general public. BC DPIC receives over 26,000
calls annually and thus receives and holds valuable and
substantial data. When monitored in real time these data
have the potential to serve as an important surveillance tool,
and in doing so poison control centres can trigger and inform
health protection practices.
Acknowledgements
The authors would like to thank the BC Drug and Poison
Information Centre, the Canadian Food Inspection Agency,
Health Canada, and the Fraser Health Authority for their
assistance with this investigation.
The authors would like to thank Daniel Mosquin, Research
Manager, UBC Botanical Garden, Vancouver, BC, for his
assistance with plant terminology and identification.
The photographs used in this report are courtesy of the
Fraser Health Authority.
References
Bradbury, J., and Nixon, R. 1998. The acridity of raphides from the
edible aroids. J. Sci. Food Agric, 76: 608616. doi: 10.1002/(SICI)
1097-0010(199804)76:4B608::AID-JSFA9963.0.CO;2-2.
Brogren, M., and Savage, G. 2003. Bioavailability of soluble oxalate
from spinach eaten with and without milk products. Asia Pac. J.
Clin. Nutr, 12(2): 219224.
Fig. 5. Taro corm.
Omura et al. 63
Environ. Health Rev. Downloaded from pubs.ciphi.ca by 209.139.228.161 on 11/04/14
For personal use only.

Hossain, R.Z., Ogawa, Y., Morozumi, M., Hokama, S., and Sugaya, K.
2003. Milk and calcium prevent gastrointestinal absorption and
urinary excretion of oxalate in rats. Front Biosci. 8: a117125.
Kasilo, O.M.J., and Nhachi, C.F.B. 1994. Food Poisoning Admissions
in Referral Hospitals in Zimbabwe: A Retrospective Study. Hum.
Exp. Toxicol, 13(2): 7782.doi: 10.1177/096032719401300203.
Kaushal, P., Kumar, V., and Sharma, H. 2013. Utilization of taro
(Colocasia esculenta): a review. J. Food Sci. Technol, 50(1): 94
100. doi: 10.1007/s13197-011-0236-0.
Matthews, P.J. 2004. Genetic Diversity in Taro, and the Preservation
of Culinary Knowledge. Ethnobotany Research & Applications, 2:
5571.
Mihailidou, H., Galanakis, E., Paspalaki, P., Borgia, P., and Man-
tzouranis, E. 2002. Pica and the Elephant’s Ear. J. Child Neurol,
17(11): 855856. doi: 10.1177/08830738020170111705.
Noonan, S.C., and Savage, G.P. 1999. Oxalate Content of Foods and
Effect on Humans. Asia Pac. J. Clin. Nutr, 8(1): 6474. doi: 10.
1046/j.1440-6047.1999.00038.x.
Oscarsson, K.V., and Savage, G.P. 2007. Composition and availability
of soluble and insoluble oxalates in raw and cooked taro (Colocasia
esculenta var. Schott) leaves. Food Chem, 101(2): 559562. doi:
10.1016/j.foodchem.2006.02.014.
Pang, C.T., Ng, H.W., and Lau, F.L. 2010. Oral mucosal irritating
plant ingestion in Hong Kong: epidemiolog y and its clinical
presentation. Hong Kong J. Emerg. Med, 17(5): 477481.
Paull, R., Tang, C., Gross, K., and Uruu, G. 1999. The Nature of the
Taro Acridity Factor. Postharvest Biol. Technol, 16:7178. doi:
10.1016/S0925-5214(98)00099-4.
Rao, V., Matthews, P., Eyzaguirre, P., and Hunter, D. 2010. The
Global Diversity of Taro: Ethnobotany and Conservation.
Sanz, P., and Reig, R. 1992. Clinical and Pathological Findings in
Fatal Plant Oxalosis. Am. J. Forensic Med. Pathol, 13(4): 342345.
doi: 10.1097/00000433-199212000-00016.
Savage, G.P., and Dubois, M. 2006. The effect of soaking and cooking
on the oxalate content of taro leaves. Int. J. Food Sci. Nutr,
57(56): 376381. doi: 10.1080/09637480600855239.
Savage, G.P., Ma
˚rtensson, L., and Sedcole, J.R. 2009. Composition of
oxalates in baked taro (Colocasia esculenta var. Schott) leaves
cooked alone or with additions of cows milk or coconut milk. J.
Food Compos. Anal. 22(1): 8386. doi: 10.1016/j.jfca.2008.
05.011.
Tagwireyi, D., and Ball, D.E. 2001. The management of Elephant’s
Ear poisoning. Hum. Exp. Toxicol. 20(4): 189192. doi: 10.1191/
096032701678766822.
WelcomeBC. 2008. Census fact sheet Regional settlement of
immigrants in British Columbia. [Online] Available: http://www.
welcomebc.ca/welcome_bc/media/Media-Gallery/docs/communit
ies/regional_immigrant_settlement_in_bc.pdf.
Yuen, E. 2001. Upper airway obstruction as a presentation of Taro
poisoning. Hong Kong J. Emerg. Med, 8(3): 163165.
64 EHR, Vol. 57, Issue 3
Environ. Health Rev. Downloaded from pubs.ciphi.ca by 209.139.228.161 on 11/04/14
For personal use only.