LAUSR

Abstract Context: Ciguatera fish poisoning arises primarily from consumption of carnivorous reef fish caught in tropical and sub-tropical waters. Ciguatoxins, a class of tasteless, heat-stable, polycyclic toxins produced by dinoflagellates, accumulate through the food chain and concentrate in various carnivorous fish, such as groupers, barracudas, wrasses, amberjack, kingfishes, and eels. Characteristics of ciguatera fish poisoning include early nausea, vomiting, and diarrhea in the first one to two days post ingestion, followed by the appearance of sensory disturbances. The classic dysaesthesia is cold allodynia, often described as reversal of hot and cold sensation, but a more accurate description is burning pain on exposure to cold. Objective: To discuss and appraise the evidence regarding the use of mannitol or other drugs in treating ciguatera framed in the historical context of the last four decades. Methods: We searched PubMed and Embase for all years from 1966 to March 31, 2017 with search terms “ciguatera”, “mannitol”, and “treatment”. These searches identified 85 articles, of which 36 were relevant to the review question. We searched Google Scholar to supplement the primary search and reviewed the references of articles for sources overlooked in the original searches. These secondary searches identified another 23 references. We excluded six clinical reports (two case series and four case reports) which did not clearly describe ciguatera or which lacked information on treatment or outcome. Fifty-three clinical articles remained for review. We searched PubMed using “ciguatera” AND “treatment” NOT “mannitol” to better identify reports describing other treatments. The search identified 128 articles, of which nine described specific pharmacological treatments and their outcomes. We combined our findings into a consensus review of the evidence both for and against the use of mannitol or other medications for ciguatera fish poisoning. Early human evidence of effectiveness of mannitol: A 1988 report described an unexpected discovery that intravenous mannitol could rapidly and effectively treat ciguatera fish poisoning. Several other uncontrolled case series and case reports appeared to support the use of mannitol. In 2002, a small randomized, controlled trial reported no significant difference between mannitol and normal saline. Subsequent case reports have cited this study as the reason for or to withhold mannitol. Thus, some controversy exists regarding whether mannitol is useful or not for treating ciguatera fish poisoning. Basic science and animal research on ciguatera and mannitol: In vitro experiments of isolated neurons demonstrate that ciguatoxins produce neuronal edema, open certain sodium channels, block potassium channels, cause uncontrolled and repetitive action potentials after a stimulus. Addition of mannitol decreases the edema and reduces the uncommanded action potentials. However, intraperitoneal injection of ciguatoxin in rats increases neuronal refractory period and slows nerve conduction velocity. Treatment with mannitol fails to correct these effects. Comparative trials of mannitol: Evidence supporting mannitol for ciguatera fish poisoning includes four uncontrolled case series, one prospective, unblinded comparative trial and several case reports. Evidence against mannitol consists of one RCT, which has a small sample size and several potential limitations. Empirical human experience with other treatments: Evidence regarding other treatments consists only of ten case reports and three overlapping case series that describe using amitriptyline, fluoxetine, duloxetine, gabapentin, pregabalin, or tocainide. For each of these, a long duration of treatment appears to be necessary to maintain symptomatic improvement. None of these treatments has been shown to be superior to mannitol. Conclusions: It is reasonable to consider using intravenous mannitol in cases of acute ciguatera fish poisoning. Medications used in other neuropathic syndromes appear to suppress the paresthesiae of persistent ciguatera cases. However, the human evidence is of low quality for all treatments.