Studies

Retrospective studies

Bagnis et al. 1979: retrospective study, N = 3,009, questionnaire, period of investigation 1964–1977. Inclusion criteria: patient history and clinical findings suggest ciguatera poisoning. No data regarding the toxin content of the consumed fish.

Bagnis and Legrand 1987: retrospective study, N = 12,890, standardised 1-page questionnaire covering 25 signs and symptoms, period of investigation 1964–1986. Inclusion criterion: clinical suspicion of ciguatera poisoning. In 25 ciguatera outbreaks (99 patients) estimation of ciguatoxin concentration in fish remains from meals through the use of cat, mouse and mosquito bioassays. This study is an extension of the study of Bagnis et al. 1979, above.

Engleberg et al. 1983: retrospective study, N = 47. Ciguatera outbreaks were determined by identifying patients who visited the emergency department between 25.2. and 20.3.1981 with a combination of gastrointestinal symptoms and one neurological clinical sign after eating fish. The index patients and all persons who had eaten the same fish were contacted and questioned.

Gillespie et al. 1986: retrospective study, N = 527. No data regarding the toxin content of the consumed fish.
Kodama and Hokama 1989: retrospective study, N = 172. Evaluation of ciguatera poisoning protocols. Inclusion criteria: presence of gastrointestinal symptoms plus one or more of the neurological, general or cardiovascular findings considered typical; samples of fish available. No data regarding the toxin content of the consumed fish.

Morris et al. 1982: retrospective study, N = 33. Ciguatera outbreaks were identified by contacting patients who had been diagnosed with fish poisoning in the emergency department of the Knud Hansen Memorial Hospital on the island of St. Thomas. Outbreaks in which patients only had gastrointestinal symptoms and no neurological symptoms were excluded from the study.

Prospective studies

Frenette et al. 1988: prospective study, N = 57. Longitudinal observation until patients were free of symptoms. Data collection with the help of interviews. Inclusion criteria: onset of illness within 48 h, with gastrointestinal and/or neurological symptoms. No data regarding the toxin content of the consumed fish.

Katz et al. 1993: prospective "follow-up" study of 15 cases (of varying age and weight) from a single ciguatera outbreak. Ciguatoxin was detected by ELISA in the remains of the consumed fish.

Lawrence et al. 1980: prospective study, N = 129. Longitudinal observation. No data regarding the toxin content of the consumed fish.

 

See Table 4.3 for a graphic representation of the findings of the above studies.

Case reports

Alcala et al. 1988: crab, Demania reynaudii (palytoxin); Noguchi et al. 1987: Parrotfish, Ypsiscarus ovifrons (palytoxin); Kodama et al. 1989: Mackerel, Decapterus macrosoma (palytoxin).

Information on palytoxin, see below, Comments (Point 2).

Signs & symptoms

Local effects

Nausea, vomiting, diarrhoea, abdominal pain (Table 4.3). Investigations in mice show that there are pathological structural changes not only in the small intestine but also in the colon (Coombe et al. 1987). In contrast, no tissue damage was found in rabbit ileum. It is assumed that the toxin components involved in ciguatera poisoning directly stimulate fluid secretion. Calcium is believed to be the "second messenger" in this process (Fasano et al. 1991).

Time between ingestion and onset of gastrointestinal symptoms on average 5.5 h (1–24 h) (Frenette et al. 1988).

Time between ingestion and onset of symptoms (all symptoms): 24 h (96%), <12 h (77%), <6 h (52%). Nausea is the first symptom of poisoning in 29%, diarrhoea in 11%, abdominal pain and vomiting in <1% each (Bagnis and Legrand 1987).

Neurological effects

• Paraesthesias (extremities), paraesthesias (perioral), burning/painful sensation of the skin upon contact with cold objects;
• Paresis of the skeletal musculature, including the respiratory musculature (dyspnoea, respiratory failure); vertigo, ataxia;
• Myalgias (due to muscle spasms?);
• Arthralgias (due to muscle spasms?);
• Headache, toothache;
• Chills, sweating (autonomic effects of the poison?) (Table 4.3).

 

Paraesthesias (sensation of numbness, tingling, itching) are frequently localised to the palms of the hands and the soles of the feet, as well as the lips and oral mucosa. These are the same body regions in which contact with cold objects elicits an unpleasant, often acutely painful tingling or burning sensation. This sensitivity to cold also appears in the form of dysaesthesia to cold air and when swallowing cold food or liquid. The sensitivity to cold is a highly sensitive and specific sign of ciguatera poisoning (sensitivity 0.78, specificity 1.00) and to date has never been seen in a neurological illness that was not associated with fish or shellfish poisoning (Frenette et al. 1988). Heat contact does not normally elicit the opposite reaction (Bagnis et al. 1979). Thus this phenomenon is not a reversal of temperature perception in the strict sense.

Ciguatera during pregnancy: movements such as those that occur during cerebral seizures have been observed in foetuses (Pearn et al. 1982).

Time between ingestion and onset of neurological symptoms on average 20.5 h (3–72 h) (Frenette et al. 1988).

Time between ingestion and onset of symptoms (all symptoms): 24 h (96%), <12 h (77%), <6 h (52%). The first symptoms of poisoning were perioral paraesthesias and paraesthesias of the extremities in 56%, myalgias and arthralgias in 2% and vertigo in <1% (Bagnis and Legrand 1987).

Muscular effects

Myalgias (consequence of effects of toxins on ion channels in muscle cells?) (Table 4.3).

Time between ingestion and onset of symptoms (all symptoms): 24 h (96%), <12 h (77%), <6 h (52%) (Bagnis and. Legrand 1987).

 

Cardiac effects

Bradycardia (effects of toxin on the heart conduction system?), arterial hypotension (differential diagnosis of other causes: dehydration) (Table 4.3).

Table 4.3 Signs and symptoms of ciguatera poisoning: comparison of various studies (comments on the reasons for the variability of the symptoms of ciguatera poisoning, see below)

Country/Region	Study	n	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20	21	22
French Polynesia	Bagnis et al. 1979	3,009
French Polynesia	Bagnis and Legrand 1987	12,890
Australia (Queensland)	Gillespie et al. 1986	527
US Virgin Islands	Morris et al. 1982	33
Cuba	Frenette et al. 1988	57 p
Hawaii	Katz et al. 1993	15 p
USA (Florida)	Lawrence et al. 1980	129 p

 

 

1	Nausea	12	Vertigo
2	Vomiting	13	Impaired coordination
3	Diarrhoea	14	Pruritus
4	Abdominal pain	15	Skin rash
5	Bradycardia	16	Myalgias
6	Hypotension	17	Arthralgias
7	Dyspnoea	18	Headache
8	Paraesthesias (extremities)	19	Toothache
9	Paraesthesias (perioral)	20	Chills
10	Burning/painful sensation of the skin upon contact with cold objects	21	Sweating
11	Muscle weakness (extremities)	22	Lack of energy/generalised weakness

 

	<10%	p = prospective study n = number of sick people who were included in documentation of the symptoms of poisoning
	10–25%
	26–50%
	51–75%
	76–90%
	>90%
	not mentioned

 

 

Table 4.4 Duration of signs and symptoms after ciguatera poisoning

Country	Study	n		1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20	21	22
Cuba	Frenette et al. 1988	57 p	mean duration in days	0.8	1.8	6.8	4.0				38	35	53–63				33		45	33	32	28			53
			range in days	0.1–3	0.1–3	0.2–45	0.2-28				1–126	1–102	1–180				1–102		1–149	1–146	1-126	0,1-156			3-180
Hawaii	Katz et al. 1993	15 p	range of duration of illness in days	0–6 days 3/15 7–20 days 4/15 21–34 days 1/15 35–48 days 2/15 49–62 days 0/15												63–76 days 0/15 77–90 days 0/15 91–104 days 2/15 105–118 days 2/15 >119 days 1/15

Differential diagnosis

Acroparaesthesias, accompanied by dysaesthesia upon contact with cold, and myalgias, which occur concurrently with gastrointestinal symptoms (in particular nausea and diarrhoea) or subsequent to these symptoms following consumption of fish, distinguish ciguatera poisoning from other forms of food poisoning. Neurotoxic shellfish poisoning is characterised by comparable clinical symptoms, but these are less marked and self-limiting within a short period of time. Apart from this, the patient history regarding what type of seafood has been eaten is also of assistance (see "Diagnosis & treatment: General practitioner/health post").

Morbidity

Ciguatera poisoning is unique amongst the illnesses caused by poisonous animals due to the persistence of sometimes bizarre residual symptoms.

Reliable morbidity data are rare. Only a few prospective studies have been carried out over a sufficiently long period of time (Lawrence et al. 1980, Frenette et al. 1988). Frenette et al. (1988) indicate the mean and median values of the duration of gastrointestinal and neurological symptoms (Table 4.4). In total, 90% of the patients studied by Frenette et al. (1988) were free of symptoms after 18 weeks.

However, residual neurological symptoms can persist for several months and appear to be aggravated by certain influences, e.g. consumption of alcohol. In contrast, gastrointestinal symptoms generally last only a few days.

Around one third of patients with ciguatera poisoning are bedridden, approx. 30% of those at home and approx. 3% in hospital. Gastrointestinal symptoms generally disappear after 24–30 h, frequently with subsequent dehydration if the patient was not sufficiently rehydrated. Cardiovascular symptoms generally take 2–5 days to resolve, while the neurological symptoms, in particular paraesthesias, can persist for weeks, months and in extreme cases years (Bagnis and Legrand 1987).

Case fatality rate

3/3,009 (0.1%) (Bagnis et al. 1979).

<0.1% of a total of 12,890 of observed cases of ciguatera (Bagnis and Legrand 1987), although thorough documentation of the fatalities in these studies was not possible due to local circumstances.

In the USA there have been no cases of ciguatera poisoning with a fatal outcome reported to date, although between 1972 and 1986, 206 ciguatera outbreaks (7.5% of all outbreaks of food poisoning) were reported to the CDC.

Mortality is possibly confined by the fact that the lethal effect of the ciguatoxins on the fish themselves limits the dose of toxins that they can accumulate (Lewis 1992).

Mortality from certain types of poisoning that are connected with ciguatera, e.g. palytoxin poisoning, can be considerably higher (data on palytoxin, see below, Comments, Point 2).

Standardised clinical diagnosis

Evaluation of patients with ciguatera poisoning is extremely difficult due to the variability of the symptomatology (see above). Two scales ('symptom check list rating scales') have been developed to help quantify the severity of acute illness and to monitor the success of treatment in chronic cases. The validity and reliability of these scales have been tested with positive results. They may prove to be useful and accurate methods to record the severity of illness and therapeutic efficacy for clinical use and in epidemiological studies (Lange 1993).

Laboratory and physical investigations

1. Detection of toxins

There is no specific laboratory method to confirm ciguatera poisoning in humans.

Detection of ciguatoxin and related polyethers in fish tissue

Indirect evidence of ciguatera poisoning: Review (Hokama 1992), Enzyme Immunoassay Stick Test (laboratory) (Hokama 1985, Hokama et al. 1987), simplified Solid-Phase Immunobead Assay (field test, able to be used by laypersons, but not yet commercially available) (Hokama 1990; Katz et al. 1993).

Detection of palytoxin in fish tissue

Indirection evidence of poisoning associated with ciguatera fish poisoning (Hokama 1992). Data on palytoxin, see below (Comments, Point 2).

2. ECG
Reversible T wave changes (Van der Sar and Glatter 1982), bradycardia.

3. CK, LDH, GOT, myoglobinuria
CK, LDH, GOT increased in serum, myoglobinuria in palytoxin poisoning. Data on palytoxin, see below (Comments, Point 2).

First aid

Induction of vomiting within the first hours of ciguatera poisoning, if the patient cannot vomit spontaneously, and use laxatives. However, the effect of these measures on the course of poisoning has not been documented in controlled clinical studies.

Treatment (symptomatic)

Fluid balance
Rehydration i.v. or p.o.

Mannitol

The mechanism of action of mannitol has not been clarified (Lewis et al. 1993). Palafox et al. (1988) report on 24 patients who all improved dramatically under mannitol treatment, 2 of whom were comatose and 1 who was in shock. The neurological and muscular symptoms improved within minutes, while the gastrointestinal symptoms responded more slowly. 21/24 received mannitol within 10 h, 3 within 82 h after the onset of symptoms (Palafox et al. 1988).

In the pilot study by Pearn et al. (1989), 5 patients who received 1–2 doses of mannitol in the acute phase of poisoning improved significantly. In 3 of these patients the symptoms disappeared completely. Treatment was commenced a maximum of 15 h after the onset of symptoms. If the symptoms had already been present longer than 24 h, mannitol treatment appeared not to be effective (Pearn et al. 1989).

A double-blind, randomized trial of manitol therapy of 50 patients with ciguatera poisoning on Raratonga, Cook Island, found non difference between manitol and normal saline (Schnorf et al. 2002). With the results of this study there is no longer convincing evidence for manitol as specific treatment for ciguatera poisoning.

Adverse reactions

If mannitol is administered to a dehydrated patient who has not yet been sufficiently rehydrated, there may be sequelae that can be explained by the pharmacological mode of action of mannitol (e.g. arterial hypotension, acute kidney failure).

Hypersensitivity reactions to mannitol have been described, but are rare (McNeill 1985).

Endotracheal intubation and artificial respiration

Endotracheal intubation and artificial respiration if respiratory failure occurs. This is very rare.

 

Treatment of shock
Atropine (0.01 mg/kg BW) and dopamine (5–20 µg/kg BW/min), if bradycardia and arterial hypotension become symptomatic (shock).

Other forms of drug treatment

Calcium gluconate, amitriptyline (Bowman 1984, Davis and Villar 1986, Calvert et al. 1987), lidocaine analogues, such as tocainide (Lange et al. 1988) and countless other substances have been suggested. In practice, none of these substances has proved successful to date, and there are no controlled clinical studies available.

Diet

No consumption of fish or fish products, seeds or seed products, nuts or products made from nuts. No alcohol, opiates or barbiturates. Avoid exposure to solvents, herbicides, insecticides, adhesives, ethers, artificial resins and cosmetics. This should be adhered to for 3–6 months, and possibly even 12 months, after symptoms have disappeared (Sims 1985, 1987). These statements have not been verified in controlled clinical studies! They assume a sensitisation to ciguatoxins, an assumption which is based on the observation that fish that do not elicit any symptoms in persons not previously exposed do lead to the development of symptoms in persons who had previously suffered from ciguatera. Moreover, there appears to be an intolerance to substances that belong to the ciguatoxin group (polycyclic ethers) (Sims 1987). The concept of sensitisation to explain certain phenomena in ciguatera poisoning is still entirely hypothetical.

Comments
Reasons for the great variability in the signs and symptoms of documented ciguatera cases:

1.To date there is still no clinical diagnostic test that can confirm a diagnosis of ciguatera fish poisoning based solely on anamnestic-clinical criteria. Thus it is not possible to include a patient in a study based on objective criteria. Detection of ciguatoxin and related polyethers in fish tissue only supports the clinical diagnosis indirectly, and, besides, this is not available in many studies.

2.Ciguatera poisoning is not caused by a single substance, but rather by different components that contribute in varying amounts to the composition of the poison (ciguatoxin, maitotoxin, ciguaterin, scaritoxin, palytoxin). It is thus conceivable that the types of poisoning that are currently included under the term ciguatera syndrome or ciguatera fish poisoning could be further subdivided if there were better correlation between individual toxins and their corresponding signs and symptoms of poisoning.

There are already 2 cases in which such differentiation appears to be justified. Poisoning caused by Scarus gibbus of the Gambier Islands, which in the initial stage displays the clinical characteristics of ciguatera syndrome, has a second phase that occurs after 5–10 days and which distinguishes it clearly from the normal picture of ciguatera poisoning. Marked cerebellar symptoms occur during this phase, last for about 1 week and require a further 4 weeks to resolve completely (Chungue et al. 1977, Bagnis et al. 1974). The toxin responsible for this second phase of the illness is also a polyether and is most probably scaritoxin.

A further toxin that stands out clearly from the group of poison components that cause ciguatera poisoning is palytoxin. This toxin is also related to ciguatoxin and is likewise a polyether. It has been held responsible for causing particularly serious cases of poisoning. These patients suffer from severe dyspnoea, which can progress to respiratory failure, and severe muscle spasms that lead to damage of the musculature (CK, LDH, GOT increased in serum, myoglobinuria) (Kodama et al. 1989, Noguchi et al. 1987). One of 2 patients died after consuming Parrotfish (Noguchi et al. 1987). In contrast to ciguatoxin and related polyethers, palytoxin has also been found in crabs (*Demania reynaudii*) and has caused fatalities in the Philippines (Alcala et al. 1988).

One of the most important reasons for the characterisation of subgroups of ciguatera poisoning or ciguatera syndrome on the basis of the main or sole toxin involved is the area of differential therapeutics. Although the toxins associated with ciguatera poisoning that have been isolated to date are all polyethers and all affect ion channels in membranes, they do differ considerably with regard to high specificity for particular ion channels.

3. Different species of fish appear to accumulate the poison components that are considered responsible for the ciguatera syndrome in variable compositions, so that the clinical picture may contain a species-specific component (Kodama and Hokama 1989). In contrast to herbivorous fish, carnivorous fish appear to cause ciguatera poisoning that has a severe course and causes significantly more cardiovascular signs and symptoms (Glaziou and Martin 1993, Kodama and Hokama 1989). Moreover, the toxins responsible for ciguatera poisoning are not evenly distributed in individual fish. The liver and gonads contain considerably higher toxin concentrations than other internal organs or muscle tissue.

4. Interindividual differences in the amount of venom ingested, with a corresponding dose-effect relationship.

5. Ethnic differences with regard to the clinical picture, which may be due to different diets or susceptibility to particular poison effects (Bagnis et al. 1979).

6. Differences in the severity of recurrent ciguatera poisoning. Recurrent ciguatera poisoning is said to be more severe (Glaziou and Martin 1993; Katz et al. 1993).

7. Differences in the severity of poisoning with greater age and body weight of the patients. Severity and duration of symptoms reported to increase significantly with age and body weight (Glaziou and Martin 1993, Katz et al. 1993).

8. Problems distinguishing the ciguatera syndrome from other differential diagnoses (other forms of fish and shellfish poisoning, infectious illnesses that can be transmitted via fish, shellfish and crabs, including type E botulism, organophosphate-containing insecticides, eosinophilic meningitis).

 

9. Differences in study design. Most studies are retrospective and based on evaluation of questionnaires. The few prospective studies available had only short observation periods, with 3 exceptions (Frenette et al. 1988, Lawrence et al. 1980, Katz et al. 1993).

 

Thus the terms ciguatera poisoning or ciguatera syndrome represent umbrella terms. It is not yet clear to what extent the toxicological, biological and geographical variables will make a subdivision possible.