|
CARIBBEAN RESEARCH INSTITUTE
FISH POISNNG IN THE ES ARIEAN
Robert W. Brody
Ciguatera Study
Caribbean Research Institute
College of the Virgin Islands
St. hxamas, U. S. Virgin Islands 00801
Pe Presentation
24th Annual Sessin
Gulf and Carilxian Fisheries Institute
Miami, Florida
15-18 Novenber, 1971
CONTENTS
Page
I. INTRODUCTION 1
II. TYPES OF FISH POISONING 3
III. SYMPTOMOLOGY AND PUBLIC HEALTH
ASPECTS 10
IV. GEOGRAPHIC DISTRIBUTION 16
V. SPECIES REPORTED AS CIGUATOXIC 21
VI. CHEMISTRY AND PHARMACOLOGY 23
VII. BIOGENESIS AND TRANSMISSION OF
CIGUATOXIN 26
VIII. EFFECTS OF CIGUATERA POISONING ON
THE DEVELOPMENT OF FISHERIES 31
IX. SUMMARY AND CONCLUSIONS 36
X. ACKNOWLEDGEMENTS 38
XI. LITERATURE CITED 39
INTRODUCTION
"When I first got here I thought that I would be able
to practically exist on fish....."; "I was poisoned fifteen
times before I left my father's house. He always liked to buy
big fish because there was more meat..... "; "I can't get a
contract for grouper or snapper with any of the hotels. They
buy the same fish from Santo Domingo for ten cents a pound
more..... "; "Where are all the seafood restaurants ? I thought
this was an island !"
These comment s and others like them represent a
biotoxicological problem which has always plagued the eastern
Caribbean. Ciguatera fish poisoning, scombroid poisoning,
and to a lesser extent "clupeoid", tetraodontoid and elasmo-
branch poisoning have been reported since pre-Columbian time
and the problem shows no sign of lessening.
This paper will present the basic facts of fish poisoning
in the eastern Caribbean area as we presently know them. It
will not attempt to review the voluminous literature from the
Pacific; the reader is referred to excellent s u m mi a r i e s of
-1-
Halstead (1967) and Banner (1971). At the present time we have
very little "hard data" on the chemistry, biogenesis and biology
and pharmacology of the fish poisoning problem in the area.
We have yet to confirm that the most important type of poi-
soning, ciguatera, is in fact identical to the toxin from the
Pacific. The sections on ciguatera are therefore based on our
local observations and inferences from studies in the Pacific
are noted.
-2-
II. TYPES OF FISH POISONING
Fish poisoning in the eastern Caribbean can be broken
down into three major groupings. The endotoxins from the puff-
er-like fishes with the additional rarely reported cases of
clupeoid, elasmobranch and hallucinogenic fish poisoning form
the first group. These incidents represent a very small per-
centage of the total number of cases reported; several years
of casual data-gathering and six months of active research have
only produced two accounts in the last four years, both of
clupeiotoxism.
The biology, chemistry, and pharmacology of puffer
fish poisoning have been accurately summarized else whe re
(Bagnis, 1970). Although the world-wide fatality rate is high
(61%), very few cases oftetraodon poisoning are reported in the
Caribbean, probably because the puffers are not highly esteemed
as a food fish. This may also be the reason for the low in-
cidence of reports of elasmobranch poisoning in the area.
Shark has been harvested recently in an effort to produce a
packaged seafood product ("Sea Flake") with generally good
customer acceptance; there have been no instances of elasmo-
branch poisoning brought to our attention as a result. Hallu-
cinatory fish poisoning (icthyoallyeinotoxism) is also an ap-
parently rare event in the eastern Caribbean; it has been
described to us by an herbalist on St. Thomas but an actual
case history has not been received.
Cleupeoid poisoning is a form of icthyosarcotoxism
caused by ingestion of the flesh of fishes of the order Clu-
peiformes. The symptoms are frequently violent with an ex-
tremely high case fatality rate. The onset of symptoms is
noted with a sharp, metallic taste in the mouth followed rapidly
by severe gastrointestinal upset with distinct indications of
vascular collapse (drop in blood pressure, cyanosis). This
may be accompanied or rapidly followed by neurological dis-
turbances nervousness, dilated pupils, violent headache,
tingling, and in severe cases respiratory distress, convulsions,
coma, and death. The toxin is apparently particularly virulent:
death may occur within fifteen minutes. The literature reports
that persons have died while in the act of eating the fish "part
of the fish was still in the victim's mouth at the time of death"
(Halstead, 1967).
Fish in the families Clupeidae (herrings), Engraulidae
(anchovies), Alb ul idae (bonefishes), and Elopidae tarponss)
have been reported as toxic in the Caribbean; our experience
has been limited to two cases involving the clupeoid fishes
Herengula humeralis (ye llo w b ill e d sprat) and Opisthonema
oglinum (thread herring). One case is reported from a fisher-
man on Tortola, B. V. I. who (with his family of five) ate "yellow
bill sprat" (probably Opisthonema oglinum from his descrip-
tion of a "spine on the back fin") caught in Great Harbor, Peter
Island, B. V. I. He reported "bad fish poisoning" symptoms
generally similar to ciguatera poisoning; the rapid onset re-
ported above was present. He forced all of the members of his
familyto vomit and alltook a "heavy dose of sulphur" (a patent
medicine). He did not contact public health authorities. The
family recovered from acute symptoms within 36 hours. (This
incident took place in late 1967 and was reported in April 1971;
the interview was thus clouded by time and at least two subse-
quent incidents of ciguatera poisoning. ) The second case is
reported byHalstead (1970) from Antigua, W.I. from 1968 (?)
when "some small surface-feeding 'herring-like' fish were
eaten". Two people died in this outbreak. Halstead suggests
that local terminology of "yellow-billed sprat" is applicable to
llerengula humeralis. The violence of the episode suggests
that this case was an example of classic clupeiotoxism while
the Tortola case is not definitely separable from ciguatoxism .
The chemical nature and biogenesis of clupeiotoxism
is not presently known. Numerous authors have suggested that
fish caught during the summer months are more likely to be
toxic. All reported incidents are from fish caught close to
tropical islands. There are several references in the literature
to planktonic blooms as the causative organism, specifically a
monadd" (dinoflagellate ?) (Halstead, 1967), Skujaella,
(Trichodesmium) (Randall, 1958) or to "swarming of palolo
worms" (Halstead, 1967). It is probable that the toxin is pro-
duced by some planktonic form as the clupeoid fishes are for
the most part plankton feeders, this may add strength to the
position that clupeiotoxism is a separate entity from ciguatera
poisoning. Clupeiotoxin is not thought to be degraded by normal
cooking and the degree of freshness does not seem to have effect
on the toxicity.
Clupeiotoxism may pose a wider threat to public health
than the occasional locally consumed fish. If the toxin is in-
deed caused by "blooming" plankton organism the likelihood of
a large school of toxic fish cannot be overlooked. The sardine
and anchovy groups are frequently thought of as ideal fish for
Fish Protein Concentrate production; we have no data on the
ability of current FPC technology to eliminate the toxin from
-6-
the raw fish.
The second major group of poisonings experienced in
the eastern Caribbean is the result of bacterial decomposition
of fresh fish. In the Virgin Islands and throughout the northern
Leeward Islands fish are typically sold "fresh" from the boat.
Very few fishermen use ice or gut the fish before sale, in fact
there is a strong feeling among older citizens that such pre-
servation is usedto camouflage the true "freshness". Eastern
Caribbean fishermen do not use live-wells and fish frequently
spend a good part of the day in the sun. These conditions ob-
viously tend to promote bacterial decomposition of the fish and
the consequent toxicity problem; it is interesting that the younger
people in St. Thomas, U. S. V. I. who tend to buy fish in the
supermarket report a significantly lower incidence of fish poi-
soning attributable to scombroids.
We presume that an undetermined proportion of the
poisonings reported are the result of some sort of bacterial de-
composition. In many "mild" cases the only complaint is of
gastrointestinal distress; the neurological symptoms specific
to ciguatera and ciguatera-like toxins are not noted. Many of
these cases may be attributable to scombroid poisoning but the
symptoms may not be sufficiently pronounced for a proper
diagnosis to be made.
The distinct histamine-like toxicity caused by bacterial
degradation of the flesh of fish of the family Scombridae is a
relatively common type of fish poisoning throughout the Carib-
bean. The symptomology includes a distinct "sharp" or
"peppery" taste upon eating the fish followed by intense head-
ache, dizziness, a variety of circulatory disfunctions, gastro-
intestinal distress, dryness of the mouth, and inability to
swallow. These symptoms are followed by generalized
erythema, the face becomes swollen and flushed, eyes are
sunken, and an urticarial eruption may develop covering the
entire body. In severe cases there may be additional compli-
cations of shock and respiratory distress. Death has been re-
ported in a few cases but acute symptoms generally dissipate
in eight to twelve hours. This toxic reaction is brought about
by the bacterial degradation of histidine in scombroid muscle
tissue which produces a substance designated as scombrotoxin.
Scombrotox in probably has a combination of chemical con-
stituents including saurine, histidine, and possibly other toxic
compounds. The disease responds well to treatment with anti-
histaminic drugs; this specific treatment has mitigated the
severity of scombroid poisoning as a public health problem in
recent years.
In the eastern Caribbean scombroid poisoning has been
reported from Acanthocybium solandri (wahoo) Scomberomorus
cavalla (kingfish or king mackerel). Scomberomorus regalis
(spanish mackerel), and S. maculatus (cero). We are not aware
of cases reported recently from the eastern Caribbean in the
tunas (Auxis, Euthynnus, Sarda, Scomber, Thunnus) but these
genera may also be incriminated. There probably is no true
seasonality to scombroid poisoning although the incidence in any
one area can be correlated with local "runs" of the particular
species involved. Thus there seem to be more poisonings
during the tourist season when sport fishing pressure is high.
The third general type of fish poisoning is described
as ciguatera fish poisoning. Evidence from the Pacific suggests
that there are at least three (probably more) distinct toxins
capable of producing the ciguatera syndrome. Many authors
(and many physicians in the eastern Caribbean area) have not
separated the diagnosis or treatment of ciguatera from that of
scombroid poisoning and some confusion has resulted. Both
ciguatera and scombroid poisoning have been occasionally re-
ported from the same fish in the Pacific (Halstead, 1967); we
have no such report from the Caribbean in recent years.
-9-
III. SYMPTOMOLOGY AND PUBLIC HEALTH
ASPECTS OF CIGUATERA
In the absence of precise chemical and biogenic data
we have defined ciguatera in terms of its symptomology. The
following symptoms are extracted from Halstead (1967), Bag-
nis (1970), and Banner (1971) and are quoted as they appear in
Teytaud and Brody (1971):
"Ciguatera fish poisoning in its simplest un-
complicated form develops within 3 to 5 hours after
the fish is eaten. There is a sudden onset of abdominal
pain followed by nausea, vomiting, and a watery diar-
rhea. The gastrointestinal symptoms will occur in
about 40 to 75 percent of the cases. The victim feels
weak, generally ill, and may experience muscle aches
throughout the back and thighs in about 10 percent or
more of the cases. Soon after, the victim complains
of numbness and tingling in and about the mouth which
then extends to the extremities (present in about 50
percent or more of the cases). Fever, headache, and
rash are generally absent, and the patient has no de-
sire for food. The acute symptoms usually subside in
about 8 to 10 hours, and within 24 hours after onset
most of the patient's symptoms will have completely
subsided except for a feeling of weakness. However,
the numbness andtingling may continue to lesser ex-
tent for a period of 4 to 7 days. The foregoing re-
sume is typical of the majority of uncomplicated
ciguatoxications that are generally encountered by the
practicing physician in an endemic ciguatoxic locality.
Ciguatera, like many other d i s e a s e s, may
vary greatly in its clinical manifestations depending
upon the toxicity of the fish that is eaten, the individ-
ual's sensitivity to the poison, amount of fish ingested,
and other factors. In a broader sense ciguatera fish
poisoning may be characterized as follows: the onset
-10-
of symptoms may vary from almost immediately to
within a period of 30 hours after ingestion of the fish,
but is usually within a period of 6 hours. The initial
symptoms in some cases are gastrointestinal in nature,
consisting of nausea, vomiting, watery diarrhea a,
metallic taste, abdominal cramps, and tenesmus,
whereas in other patients the initial symptoms consist
of tingling and numbness about the lips, tongue, and
throat. This may be accompanied by a sensation of
dryness of the mouth. The muscles of the mouth,
cheeks, and jaws may become drawn and spastic with
an accompanying sensation of numbness throughout.
Generalized symptoms of headache, anxiety, malaise,
prostration, dizziness, pallor, cyanosis, insomnia,
chilly sensations, fever, proiuse sweating, rapid
weak pulse, weight loss, myaigia, and back and joint
aches may be present in varying degrees, or one or
more of the symptoms may be entirely absent. The
victims usually complain of a feeling of profound ex-
haustion and weakness. The feeling of weakness may
become progressively worse until the patient is unable
to walk. Muscle pain are generally described as a
dull, heavy ache, or cramping sensation, but on
occasion may be sharp, shooting, and affect particu-
larly the arms and legs. Victims complain of their
teeth feeling loose and painful in their sockets. Visual
disturbances consisting of blurring, temporary blind-
ness, photophobia, and scotoma are common. Pupils
are usually dilated and the reflexes diminished. Skin
disorders are frequently reported that are usually
initiated by an intense generalized pruritus, accom-
panied by erythema, and maculopapular eruptions,
blisters, extensive areas of desquamation particu-
larly of the hands and feet and occasionally ulcera-
tion. There may also be a loss of hair and nails.
In severe intoxications the neurotoxic com-
ponents are especially pronounced. Paresthesias in-
volve the extremities, and paradoxical sensory dis-
turbances maybe present in which the victim interprets
cold as a "tingling, burning, dry-ice or electric-shock
sensation", or hot objects may give a feeling of cold.
In regard to the paradoxical sensory disturbance (P.
S. D. ), a classic example is that of a naval officer who
was poisoned by an amberjack. Four weeks later he
-11-
was observed subconsciously blowing on his ice cream,
which was "burning his tongue", in order to cool it.
Ataxis and generalized motor incoordination may be-
come progressively worse. The reflexes may be di-
minished, muscular paralyses may develop, accom--
panied by clonic and tonic convulsions, muscular
twitchings, tremors, dysphonia, dysphagia, coma, and
death by respiratory paralysis. The limited morbidity
statistics show a case fatality rate of about 12 percent.
Death may occur within 10 minutes, but generally re-
quire several days. "
Table 1 summarizes the symptoms occurring during the first 24
hours after ingestion as they were reported by 25 persons who
were interviewed following ciguatoxications of minor to moderate
severity in St. Thomas U.S. V.I. during 1971. Several of
these reports represent the symptoms produced in different
individuals by a single fish; they therefore do not represent 25
separate outbreaks.
TABLE 1. Summary of Symptoms
Symptom
abdominal pain
nausea
vomiting
diarrhea
numbness, tingling about mouth
headache
numbness in extremeties
metallic taste
weakness
muscle aches
paradoxical sensory disturbance
itching
Percent Reporting
96 (96)
88 (92)
68 (68)
96 (96)
56 (64)
48 (48)
48 (56)
24 (36)
96 (96)
40 (48)
32 (32)
64 (68)
Summary of symptoms manifested by 25 ciguatoxicated indi-
viduals during first 24 hours after ingestion of fish. (Percent-
ages in parentheses represent change in original descriptions
following questions by the interviewer.)
-12-
Several additional symptoms were reported by three
or fewer (less than 12%) of the persons interviewed. These
symptoms include lack of coordination, muscle spasm, high
fever, visual disturbance s, diminished reflexes, and skin
rash. It is notable that none of these persons required hos-
pitalization and only three reported visiting a physician (several
others contacted a p physician by telephone during the time
period 24 72 hours after ingestion).
Virtually all of the persons interviewed reported no-
ticeable symptoms of ciguatoxication for several days after
the onset of the incident. Most commonly reported was ex-
treme weakness and lethargy lasting up to two weeks. Many
victims reported gastrointestinal symptoms well into the third
day along with itching and/or skin rash. Those persons report-
ing the paradoxical sensory disturbance stated that it persisted
for at least three days, in some cases ten days or two weeks.
The bulk of the other symptoms noted were reported as having
dissipated within the first 24 hours.
We are currently undertaking a more extensive epi-
demiology reporting program in cooperation with local media,
physicians, and public health authorities. Data from this sur-
vey combining questionnaire and interview procedures should
- 13-
be available early in 1972. At this time we have no accurate
estimate of the magnitude of the ciguatera poisoning problem
in the Virgin Islands or for that matter anywhere in the eastern
Caribbean. Outbreaks in Puerto Rico are sufficiently notable
to receive coverage in the major English-language media which
suggests that they are infrequent. The officials responsible
for public health record-keeping in St. Thomas, on the other
hand, estimate three or four cases per week are seen in the
emergency room; if our 25 cases reported in table 1 represent
typical reporting ratios there may be as many as 30 cases per
week in St. Thomas. These figures probably represent the
maxima however and cannot be confirmed. Reports of cigua-
toxication in the British Virgin Islands have stated that virtually
"everybody" has been poisoned at least once (some as many as
fifteen times) but medical advice is almost never sought.
Interestingly the British Virgin Islands are the only demo-
graphic unit mentioned by Halstead (1970) where fish poisoning is
"not regarded as deterrent to the development of the fisheries
programme".
The pattern of sporadic reporting of ciguatera poi-
soning despite the relatively high incidence of the disease is
common throughout the northern Leeward Islands. Information
-14-
from Halstead (1970) and our own contacts with fishermen, in-
ter -island traders, charterboat operators and fishery personnel
in the region suggests that the problem is indeed severe. Vir-
tually every person contacted from St. Kitts, St. Maarten, St.
Eustatius, Anguilla, and Montserrat had either been a victim
himself or knew of a close friend or relative
who had been poisoned within the last five years. Very few
of these cases were brought to the attention of a physician;
most public health officials believe that "only the
very severe cases are brought to the attention of the medical
authorities" (Antigua); Halstead,(1970). One long-time resident
of St. Kitts estimates only about one case in ten receives
medical attention; as might be expected the bulk of the cases
reported involve tourists and non-native residents.
-15-
IV. GEOGRAPHIC DISTRIBUTION
The geographic distribution of ciguatoxic fish in the
Virgin Islands is shown in Figure 1. The island of St. Croix,
on a separate geologic platform, is not reported as producing
ciguatoxic fish in any appreciable quantity and has been omitted
from this figure. A large number of the areas indicated have
been reported as producing toxic fish for centuries. It cannot
be presumed that these are the only localities; toxic fish are
frequently caught in other areas.
In the Virgin Islands there is an extremely strong
feeling among the fishermen that the south side of the Virgin
Bank from Sail Rock east to Peter Island consistently pro-
duces toxic fish. Other fishermen would extend this area east
and north to include most of the coastline of Virgin Gorda,
some would include the Horseshoe Reef and Anegada. Still others
(particularly those who regularly fish this southern Bank) state
that only specific locations in this area produce toxic fish and
that reef areas or "banks" only a few miles away are free
of ciguatera. Virtually all fishermen feel that the entire north
side (the Atlantic side) of the Bank is free of toxicity with the
exception of a very few species. This pattern of geographic
-16-
100 fathn r
/
/
-'.
1
.-
e ..
I.
- .
* -.
Guana(
Jost
TS STrwe
9
Salt .-
-.4r
. itoo fathm.
a Norman
Buck '
/l
OFrenchcap /-B
100 fathcus
FIGURE 1
Map of the Virgin Islands area
-----* 100 fathoms curve
a areas reported as producing
toxic fish
Horsedxs
Reef
distribution of ciguaioxic fishes seems consistent with those
areas reported by earlier authors (Walker, 1922; Arcisz, 1950;
Brown, 1945; de Sylva, 1956, Mann, 1938). Other writers
quoted in Halstead (1967) notably Hill (1868) Rogers (1899) and
Gilman (1942), are contradicted by local fishermen, at least for
the bulk of the species implicated elsewhere in the Virgin Is-
lands.
In almost all reports on the geographic locality where
toxic fish are caught the interviewee was referring to depths
of 30 fathoms or less; the bulk of reports refer to reef areas
in eight fifteen fathoms, but this may be an artifact of fishing
methods rather than biological distribution. The relationship
of depth to capture of ciguatoxic fishes is discussed in a later
section.
The areas reported (by our contacts and by Halstead,
1970) as producing toxic fish in the northern Leeward Islands
are shown in Figure 2. These data are plotted along with areas
mentioned as toxic by various authors in Halstead (1967) and
do not represent an intensive survey. More .specific data will
be presented in a later paper. The more southerly group of
islands in the eastern Caribbean (Martinique south to Trinidad:
the Windward Islands) have not commonly been reported as pro-
ducing ciguatoxic fish in this century. Earlier authors make
-18-
reference to a variety of species and locations but this is not
confirmed by present residents.
Although no quantitative data are available it seems
clear that the majority of ciguatera poisoning outbreaks in the
eastern Caribbean occur in a rather small area from Mont-
serrat north to the Virgin Islands including all of the northern
Leeward Islands and portions of Saba and Anguilla Banks. Be-
yond this area ciguatera poisoning is limited to sporadic out-
breaks which generally involve large specimens of only a few
species.
-19-
FIGURE .2
Map of the Leeward Islands
--- 100 fathoms curve
a areas reported as
producing toxic fish
r '!
1' ~
9*~*~* 9
A
4~ A~*\
--A
-. Anguilla Bank
a'-'' .. -
^., r^ An~guilla
--St. Maarten
'-. ^^ 7 t'tMarten
'-I
/
/'
.'
I
lb
lb.
,\ ;,V St. Baits
I
p.' -
'P..
Saba Bank
I
.V
19
a/
* - *
I '.' -.\
'-,^St. Eustatius
N,
St. Kitts
. I Nevis
*t.*
Mmntserrat
PAGE 20
\. 'p
'>./'
.4
C "-" *
4-
I--
gsaba
/
-S
.- "p.
A
BarbIa
Bazbuxfa W "
/
a.
I
* 4.
* /1
a.,_
.
-''
/
I
i Antigua
i -
.
a'
V
'4. _I
-- *",
* a
'a'"
V. SPECIES REPORTED AS CIGUATOXIC
More than 400 species of fish have been implicated in
ciguaterapoisoning on a global basis (Bagnis, 1970). Of these
400, 91 species could conceivably be found in the eastern Ca-
ribbean. It is possible that an -even larger number could be
associated with ciguatoxications if they were desirable as food
fish. There are also a number of reports of molluscs, crabs,
and lobsters producing the disease. Appendix I lists the 24
fish most frequently reported as toxic in the Virgin Islands.
All of these species are valued as food fish with the exception
of barracuda and amberjack; these two species have such a
bad reputation as ciguateric that only the smallest specimens
can be sold. It is somewhat surprising that the moray eel is
as highly esteemed as it is, considering reports of toxicity
from the Pacific. In St. Thomas eels are typically purchased
by individuals from Spanish Caribbean cultures (Puerto Rico,
Dominican Republic, Cuba), areas where ci gu at e r a is less
frequently reported. The fish at the top of the list tend to be
reported as toxic more frequently than those lower down.
In general the larger specimens of these species are
frequently incriminated in ciguatera poisoning incidents. The
fish generally can be considered "shor e-fi she s" or "reef -
-21-
fishes" and for the most part conform to the pattern noted by
other authors (Randall, 1958; Halstead, 1967; Banner, 1971):
toxic fishes are not common at depths greater than 80 100
fathoms. Two notable exceptions in the Virgin Islands are the
blackfin snapper, Lutjanus buccanella and misty grouper,
Epinephalus mystacinus. We have data on two outbreaks af-
fecting five people from blackfin snapper caught in deep water
and three additional outbreaks implicating misty grouper (which
has not been taken at depths of less than 55 fathoms) involving
at least ten persons during mid-1971. These data appear in
Appendix II. Additional data on ciguatera from deep-living
species is noted in a later section of this report. Although
Banner (1971) states "true ciguateric fishes appear to be only
those fishes tied directly to the flora and fauna of coral reef.. "
there is excellent clinical data to support these outbreak re-
ports; several members of our staff were among the victims.
Samples have been retained for extraction and bioassay to
quantify the toxicity of .these fish (see also section on fisheries
development).
-22-
VI. CHEMISTRY AND PHARMACOLOGY
Our knowledge of ciguatera poisoning in the eastern
Caribbean is presently based upon clinical reports and is only
beginning to be quantified by chemical and biological assays.
On the basis of symptomology and from the species implicated
it is highly probable that ciguatera poisoning in the eastern
Caribbean is produced by very similar (if not identical) com-
pounds to those known from the Pacific. Scheuer and other
workers at the University of Hawaii have isolated what they
consider to be the primary toxin and, in cooperation with
Hashimoto and his colleagues at the University of Tokyo;
several secondary toxins. The primary toxin (deemed cigua-
toxin) is insoluable in water, soluable in polar organic sol-
vents, heat stable to 1000C, stable below 0C as a crude toxin
but unstable in the semipurified or purified form unless ex-
tracted, purified, and stored in an inert atmosphere at low
temperatures. The non-crystalline product has the empirical
formula (C35H65NO3)n and the molecule has indications of a
quaternary nitrogen atom, one or more hydroxyl groups and a
carbonyl function. It is not a phospholipid. A c-r ystalline
product currently undergoing analysis to determine its structure -
al formula (Banner 1971).
-23-
Present evidence (again from the University of Hawaii
group) suggests that ciguatoxin acts upon excitable membranes
to increase the permeability of Na + ions, upsetting the ionic
balance of the membrane. Ciguatoxin is not an anticholinesterase
in vivo, despite earlier papers and various therapies for cigua -
toxications based on stimulation of cholinesterase cannot be
endorsed at the present time. The toxin is active at the level
of 0. 025 mg/kg when injected into mice with a toxin yield of
5 10 mg/kg from highly toxic flesh. The toxin is carried at
a uniform level throughout the musculature of toxic fish but
may be 50 to 100 times as concentrated in the viscera, par-
ticularly the liver.
Our laboratory in St. Thomas is currently using an
acetone diethyl ether extraction with purification by column
chromatography developed by Scheuer (per. comm.) and bio-
assay using intraperitoneal injection into 20+ gram Charles
River CD-1 mice. We have previously used other extraction
techniques including crude aqueous extraction with emulsifiers
and a variety of experimental bioassay techniques. Our con-
clusions are basically the same as the workers in Hawaii al-
though based on much less experience: careful solvent ex -
traction and purification are necessary, rigidly cont r oled
bioassay procedures are required, and experienced laboratory
-94-
personnel are an absolute requirement. To date there has
been no rapid, simple colorimetric or other chemical test for
ciguatoxin developed. Screening programs are exceedingly ex-
pensive and difficult to manage and are only in operation in
Japan on a limited basis for s e 1 e c t e d samples from highly
suspect areas. As much as we might desire it, we are not
very close to a rapid means of identifying ciguatoxic fish in
the laboratory and even further from a simple test which might
be part of a housewife's shopping kit.
The traditional West Indian methods of determining if
a fish is ciguatoxic have been discussed at length by previous
authors. Appendix III lists these methods as reported to us by
natives of the Virgin and Leeward Islands. Many housewives
swear by some particular method utilizing visual inspection of
external characters of the fish. Most admit that in practice
both the visual methods and those requiring addition of some
indicator are unreliable. We have submitted each of these
methods to an assay with at least two known toxic fish and two
non-toxic fish and have not found them reliable.
-25-
VII, BIOGENESIS AND TRANSMISSION OF CIGUATOXIN
At the present time ,we have no accurate data on the
mechanisms of biogenesis of ciguatoxin (or its related compounds)
nor information on its transmission through the food chain.
Banner, Helfrich, Randall, and others at the University of
Hawaii have concentrated a good deal of effort on these prob -
lems in the Pacific and their findings to date are summarized
below (from Banner, 1971):
1. No causative agent or organism has yet been
identified as producing toxins similar to ciguatoxin.
2. No definite evidence has been found to suggest
that: a) copper or other metallic ions act as chelators, trace
minerals or catalysts in the formation of the toxin; b) no
demonstrable increase in ciguateric fishes was noted in areas
where new surfaces were exposed by natural disasters,
dredging, blasting, predation by Acanthaster; c) contamination
of the marine environment by pollutants (specifically wax esters
at Wake Island) have no effect on ciguatera.
3. Normally non-toxic omnivor es can be made
toxic when fed small amounts of toxic fish over a period of
time.
-26-
4. Toxic Lutjanus bohar retain toxicity for up to
30 months when fed a non-toxic diet.
5. A detrital feeding acanthurid ( Ctenochaetus
striatus) has demonstrable ciguatoxin in the flesh, viscera,
and gut contents.
6. Most carnivores seem to carry the same toxin
(ciguatoxin) although additional toxins (eg: Aluterin, ciguaterin)
may also be present.
It should be noted that Dr. Banner will present a paper entitled
"Biological Origin and Transmission of Ciguatoxin" tomorrow
(18 November 1971), which could shed some new light on this
subject.
Given the similar symptoms and species distribution
reported in Pacific and eastern Caribbean ciguatera poisonings
it is reasonable to assume that similar biogenesis and trans-
mission of the toxin can be expected. There are several per-
sistent beliefs among eastern Caribbean fishermen which will
be repeated here although we have beenunable to confirm them.
1. Ciguatera is produced by fish which eat the fruit
or leaves of manchineel (Hippomane mancinella). This theory
has been proposed since 1511 (by Peter Martyr of Anghera;
-27-
Halstead, 1967) and is probably based on advise to early ex-
plorers from Caribbean Indians. Hippomane is certainly toxic
but its pharmacological action is quite different. It is doubtful
that this theory is correct.
2. The most persistent theory in the Vi r gin and
Leeward Islands involves copper. Natural copper deposits
("copper banks") are presumed to exist and fish which feed on
these banks become toxic. Some of the more sophisticated
fishermen suggest that it is not actually the copper metal but a
small "sea moss" (which grown in areas where copper con-
centrations are high) which actually manufactures the toxin or
a precursor. The "sea moss" responsible has been pointed
out to us by several fishermen (actually) three species:
Enteromorphialingulata from shallow water at Buck Island, St.
Thomas; Cladophora sp. from fish pot warps of Flanagan Is-
land, U. S. V. I.; Chaetomorphia sp. from rocky s u b t i d a 1 at
Buck Island, St. Thomas). None of these algae showed toxic
activity when extracted with Tween and injected I. P. into mice;
we plan to repeat this experiment with solvent solvent ex-
traction and column chromatography when these algae can be
collected from historically toxic areas. Most fishermen
suggest that the production of toxic "sea moss" is seasonal
-28-
with peak growth in late spring or early summer. The asso-
ciation of ciguatoxin with copper is not limited to copper banks
by fishermen; it is proposed that shipwrecks (particularly older
wrecks with copper-sheathed bottoms) and copper antifouling
paints supply all of the copper needed. We can in no way con-
firm the theory of copper-induced ciguatoxin at this time.
3. A theory proposed by a few fishermen and com-
pletely unproven at this time attempts to explain the high con-
centrations of toxic fish on the south of the Virgin Bank (with
the lower toxicity reported from the same species on the north
side of the Bank) and in the area from Antigua north to the
Anegada Passage. It is proposed that the toxin is produced by
some organism (presumably a primary producer) which is found
only in areas where deep, cold, nutrient-laden water is upwelling.
The theory is reasonable when applied to the southern Virgin
Bank and the southeastern portions of Saba Bank, both noted as
producing toxic fish, as there is good evidence that upwelling
does indeed occur in the Anegada Passage. The upwelling pro-
cess cannot be confirmed in the St. Kitts St. Eustatius -
Redondo area at the present time due to lack of data. The
specific organisms(s) responsible and the mechanism of toxin
production are not known by the proponents of this theory.
-29-
We are therefore no closer to an accurate description
of the biogenesis or transmission mechanisms of ciguatoxin than
purely theoretical considerations. We have proposed a series
of studies similar to those undertaken by the Univer sity of
Hawaii group including chemical, ecological, and epidemiological
programs for the next three years to attempt a better under-
standing of this problem.
VIII. EFFECTS OF CIGUATERA POISONING ON
THE DEVELOPMENT OF FISHERIES
We have every reason to believe that ciguatera poisoning
is a major impediment to the sale of local finfish in the Virgin
Islands and thus is a strong deterrent to expansion of the com-
mercial fisheries. Interviews by Halstead (1970) suggest that
this is true throughout the Leeward Islands too, although residents
of Antigua, St. Kitts, and St. Maarteo express the opinion that
there is little alternative to continued buying of local fish and
risk of intoxication.
InDammann's 1967-68 survey of commercial fisheries
of the Virgin Islands slightly more than half of total finfish con-
sumption was from local (U.S. and British V.I.) sources
(1, 672,400 of 3, 084, 373). We have no data on how much of the
fish imports could be replaced by local production if ciguatera
were not a problem. Interviews with fishermen suggest that
very few hotels and restaurants catering to the tourist trade
would purchase locally caught grouper, snapper, jacks, and
kingfish because of fear of fish poisoning. Dammann's Table 9
"Fisherman-reported problems in the Virgin Islands commercial
fishery" does not include any data on this subject, however Table
12 indicates that only two percent (two of 79) of the fishermen
-31-
contacted felt that there were "no fish" (commonly reported as
ciguatoxic) so one might assume that ciguatera was indeed con-
sidered a problem.
Two investigations of fisheries development potential
in the Virgin Islands area have recently been completed. The
first (Dammann et al., 1970) developed lines of approach carried
out in the second project (Brownell and Rainey, 1971) for ex-
panding the Virgin Islands fisheries through exploitation of deep
water stocks. This effort was motivated by several natural
limiting factors on the shallow water fish populations, among
them the risk of ciguatera poisoning. It now appears that even
species previously considered non-toxic because their normal
depth-range is greater than 100 fathoms are implicated in
ciguatoxications. Brownell and Rainey (1971) report three out-
breaks from mity grouper Epinephelus mystacinus taken at
130-135 fathoms and the only documented case of c iguat era
from a silk snapper Lutjanus vivanus from 110 fathoms. Two
questionable outbreaks are reported by Dammann et al. (1970)
from Epinephelus nigri tus (actually E. mystacinus). In addition
to the outbreaks reported for E. mystacinus and Lutjanus
buccanella in Appendix II, we are aware of several outbreaks
from L. buccanella attributed to fish caught during the exploratory
-32-
fishing projects of the UN/FAO Caribbean Fisheries Development
Programme (CFDP)in 1970 and 71. We are currently extracting
and bioassaying about two tons of fish caught on UN/FAO cruises
from areas where toxicity is reported. These data will be re-
ported in early 1972. Although these data will provide us with a
more precise estimate of the proportion of ciguatoxic fishes in
the deep shelf shelf-slope-populations, we have already
ascertained that this resource is not free of ciguatoxin.
The toxic blackfin snappers caught by the CFDP came
from Saba and Anguilla Banks; fishermen in Montserrat report
that most known poisonous fish had been captured in deep water
- up to 250 fathoms. It is highly probable that the abrupt drop-
off to depths of 200 fathoms or more surrounding many of the
Leeward Islands harbor excellent stocks of food fish but it is
quite likely that some of these species carry ciguatera poisoning.
Halstead's 1970 survey found that fishermen, fisheries
officers, and public health officials were almost unanimous that
ciguatera was a deterrent to de velopment of the commercial
fisheries. Most Islands reported ciguatera in fish from depths
of 0-60 fathoms and the most frequently toxic are all among the
first ten species listed in Appendix I. At least two large com-
mercial fishing operations in St. Maarten have given up shallow
-33-
water fishing because of repeated ciguatoxications by their catch;
several fishermen have reported having to discard large catches
of jacks and grouper because their previous catches had caused
poisoning. In the small communities of the Virgin and Leeward
Islands an individual fishermen is occasionally completely boy-
cotted because of his reputation for regularly landing toxic fish.
Fishermen are expected to be able to determine whether or not
a particular fish is toxic; an occasional instance is accepted,
however.
The fisheries of the Virgin and Leeward Islands do not
lend themselves to exploitation by large vessels with modern
ground fishing gear. The pelagic stocks are apparently not
sufficient to support a much larger fishing effort than is currently
in progress. There are probably not sufficient stocks in the
shelf-edge populations to withstand intensive fishing pressure
equivalent to the Gulf of Mexico Florida Straits snapper in-
dustry. The majority of fishermen in the eastern Caribbean are
owner-operators of small boats (20 feet or less) who rarely go
more than ten miles from their home port. These fishermen
could be trained and proper gear could be utilized for exploiting
the area's natural stocks both in shallow and deep water but fish
poisoning cases would be likely to increase. A thorough under-
-34-
standing of the ciguatera problem must be developed before ex-
pansion of the fi shery can be effectively accomplished.
-35-
IX. SUMMARY AND CONCLUSIONS
1. Fish poisoning in the eastern Caribbean is reported
from all of the islands of the northern Virgin and Leeward Islands
group. Puerto Rico, Hispaniola and St. Croix have a much lower
incidence rate as do the Windward Islands (Trinidad to Martinique).
2. Although c 1 u p e oi d, elasmobranch, tetraodontoid,
and hallucinatory fish poisoning are reported from the eastern
Caribbean, scombroid poisoning and ciguatera poisoning are con-
sidered to be most important. Because scombroid poisoning can
be prevented bymodern preservationtechniques and treatment of
the disease is specific and effective, it is considered a less se-
vere problem than ciguatera poisoning.
3. Epidemiological reporting of ciguatoxications has
only been begun in the last month throughout the Virgin Islands
and a careful survey of the Leeward Islands must await additional
funding. Ciguatera is presently reported as a severe public health
problem with only a fraction of the cases reaching medical atten-
tion. The problem seems most severe in the area from Mont-
serrat north to the British Virgin Islands including the south-
east portion of Saba Bank and the southern shelf of the Virgin Is-
lands pl at; au.
-36-
4. The chemistry, pharmacology, and ecology of
ciguatoxin and closely allied compounds are at present poorly
understood. The symptomology and species distribution of the
toxins in the eastern Caribbean strongly suggest that a situation
exists which is very similar to that described from the Pacific
islands by the Marine Biotoxin group at the University of Hawaii
over the past sixteen years.
5. Toxicity in eastern Caribbean fishes seems to be
more prevalent among the large carnivores of reef or reef-re-
lated habitats. There are a number of data which suggest that
ciguatoxin(s) are produced by some organism in the reef food web
and that the toxin is passed through the food web without a signi-
ficant modification and concentrated by the larger carnivores.
6. Development of the commercial fisheries in the
eastern Caribbean is severely impeded by the prevalence of
ciguatoxin in commercially desirable species. There is good
evidence that the shelf-edge stocks of snapper and grouper are
not free from ciguatera poisoning as previously proposed and
that exploitation of this presently underutilized resource may be
impeded by this toxicity.
3-7 -
X. ACKNOWLEDGEMENTS
This report was prepared with the cooperation of the
staff of the Ciguatera Project, Caribbean Research Institute,
College of the Virgin Islands. The Ciguatera Project is supported
by a Sea Grant (1-35368) from the National Oceanic and Atmospheric
Agency of the U. S. Department of Commerce with matching con-
tributions from the United Nations Food and Agriculture Agency's
Caribbean Fisheries Development Programme, the College of the
Virgin Islands, and the G o ve rnment of the U. S. Virgin Islands.
-38-
XI. LITERATURE CITED
Arcisz. W. 1950. U. S. Fish & Wildlife Serv. Sp. Sci. Rep. no.
27, 1-27
Bagnis, R. et al. 1970. Bull. World Health Org. 42:69-88.
Banner, A. H. 1971. Tech. Report #1 (Final Report) ONR
Contract #N00014-67-C-0127.
Brown H. H. 1945. Develop: Welfare West Indies, Bull. 20:
34-37
Brownell. W. N. & W. E. Rainey 1971. Carib. Res. Inst. Rept.
VIERS no. 3.
Dammann, A. E., W. Swingle & J. Yntema 1969. Carib. Res.
Inst. Spec. Rept. VIERS no. 1
et al. 1970. Carib. Res. Inst. Spec. Rept. VIERS
no. 2.
de Sylva, D. 1956. Mar. Lab. Univ. Miami. Spec. Serv. Bull.,
v. 13:1-9.
Gilman, C. L. 1942. U. S. Nay. Med. Bull. 40(l):19-27.
Halstead, B. W. 1967. Poisonous and Venomous Marine Animals
U. S. G. P. O.
1970. UN/FAO CFDP working paper FI:SF/CAR Reg.
189.
Hill, R. 1868. Proc. Sci. Assoc. Trinidad 1 (5)210-223.
Mann, W. L. 1938. U. S. Nay. Med. Bull. 36(4):631-634.
Randall, J. E. 1958. Bull. Mar. Sci. Gulf Caribbean 8(3):236-267.
Rogers. J. M. 1899. Pop. Sci. Monthly 55:680-685.
Teytaud, A.R. & R.W. Brody 1971. Carib. Res. Inst. Spec. Pub.
Cig. no. 1.
Walker, F.D. 1922. U.S. Nay. Med. Bull. 17(2):193-202.
-39-
APPENDIX I
Species Most Commonly Reported as Toxic in the Virgin Islands
Area
Species
Sphyraena barracuda
Seriola dumerili
Caranx latus
Caranx ruber
Caranx hippos
Lutjanus jocu
Mycteroperca venemosa
Scomberomorus cavalla
Caranx fusus
Gymnothorax funebris
Epinephelus adscensionis
Mycteroperca bonaci
Scomberomorus vegalis
Caranx lugubris
Lutjanus griseus
Lutjanus buccanella
Epinephelus mystacinus
Epinephelus guttatus
Seriola rivoliana
Caranx bartholomaei
Apsilus dentatus
Epinephelus morio
Lachinolaimus maximus
Belistes vetula
Common Name
Great barracuda; barra
Amberjack; amber
Horse-eye jack
Bar jack; carang
Crevalle jack
Dog Snapper; dog tooth
Yellowfin grouper; gramminix
Kingfish; king mackerel
Blue runner; hard nose
Green moray (probably Conger
or Congo eel)
Rock hind
Black grouper
Cero (often mistaken for
"spanish mackerel")
Black jack
Grey snapper
Blackfin snapper
Misty grouper (often erroneously
called Warsaw grouper)
Red hind
Almaco jack
Yellow jack
Black snapper
Red grouper
Hogfish
Queen triggerfish; old wife
APPENDIX II
Outbreak reports from fishes caught at depths greater than 500
feet during 1970-71 (data from interviews; Ciguater a Case
Repository numbers refer to CRI files)
CCR-71-011. Blackfin snapper Lutjanus buccanella about
four lbs. Fish caught south of Frenchcap Cay, U. S. V. I.
at about 40 fathoms bylocal fishermen. Fishwas eaten
by three people all of whom reported abdominal pain,
bowels, nausea, vomiting in that order. Onset between
three and six hours after ingestion. Secondary symptoms
included extreme weakness, listlessness, itching which
developed 12-24 hours after ingestion and lasted for
several five seven days. None of the victims reported
previous intoxication.
CCR-71-018. Blackfin snapper Lutjanus buccanella about
three lbs. Fish was purchased on the waterfront at St.
Thomas by young couple visiting relatives on the island.
Fish was broiled with sauce, no symptoms developed
until about eight hours after ingestion. Vomiting, diar-
rhea, weakness in the knees, dull headache persisted all
the following day. Late afternoon produced P. S. D. for
man but not wife. Returned to mainland three days after
ingestion, no followup available.
CCR-71-008. Misty grouper Epinephelus mystacinus 36
lbs. Fish purchased at the dock caught (apparently)
south of St. John. Victims brought five lbs. (tw, large
steaks); refrigerated them and cooked fish next evening.
Four persons had dinner of this fish; three young men
and a young lady one man and the lady reported nausea,
vomiting, and weakness within six hours; headache,
nausea, weakness persisted for "three or four days".
The third victim had no violent symptoms of gastro-
intestinal origin but was lethargic and felt "weak in the
joints" next day. The fourth person did not report any
illness.
CCR-71-021. Misty grouper Epinephelus mystacinus 56
lbs. Fish caught bylocal sport-commercial fishermen
at the "Warsaw Pocket" (misnamed since the area pro-
duce misty groupers) about 3-1/2 miles south of Nor-
man Island, B. V. I. at depth of about 120 fathoms. Fish
was filleted and headed; at least six persons ate fillets
with no ill effects. Two more people made soup of the
head; they reported some intestinal discomfort and weak-
ness, tingling sensations and lethargy the following day.
Five other persons fried a small section of the liver:
each reported eating "not more than a few bites" that
night. All awoke within three hours with violent abdominal
cramps, vomiting, and violent headache.
Severity of gastrointestinal symptoms continued for six
hours or so, then extreme '.'.akness, sinus-like head-
ache, and watery bowels persisted for two four days.
P. S. D. and tingling and numbness in the lips were re-
ported about sixteen hours after ingestion by three of the
five. All reported persistent symptoms of weakness and
soreness in all body joints for seven ten days. P. S. D.
persisted for at least a week in two victims.
CCR-71-023. Misty grouper Epinephelus mystacinus a-
bout 30 lbs. Steaks were sold to about four persons
none of who apparently developed ciguatera symptoms.
A soup was made of the head and eaten by three persons.
All described gastrointestinal distress, diarrhea, and
nausea within three six hours; apparently the symptoms
disappeared within about 24 hours for two of the victims;
the third reported listlessness, weakness, and achy
joints which lessened by the third day after ingestion.
APPENDIX III
Methods for identifying ciguatoxic fish as described in the folk-
lore of the Virgin and Leeward Islands (from interviews; Dam-
mann et al., 1969; Halstead, 1967)
I. External characters of the fish or fish flesh which indicate
toxicity:
-More yellow or brassy color, especially about the head
-Stripes (in species where they are not normally obvious)
-Darker coloration
-Red coloration to the eyes
-Yellow mucus on inner lining of gullet
-Green tint to raw flesh
-Tiny black "veins" running through the musculature
-Brassy or coppery odor to the flesh
-Teeth are black
-Suspect specifies with roe is toxic
-Enlarged or bloated stoifnach
-Flesh tastes bitter or hot in mouth
II. Indicator organisms which suggest toxicity:
-Worms in the flesh (particularly jacks and mackerel)
(Worms in the stomach indicate a non-toxic fish)
-Isopod ectoparasites ("cockroach") are not found on toxic
fish (jacks)
-Flies will not land on flesh
-Ants will not eat
III. Methods employing an indicator:
-Silver turns black when boiled with toxic fish
-Sweet potato turns black when boiled with toxic fish
|
RSS
HIDE
