Abstract
Introduction
In recent years, climate change and human activity have modified marine biotopes, including the widening distribution of harmful algal blooms (HABs). Bloom events predominated by microalgae of the genus Ostreopsis have been described on the French Mediterranean coast, but in 2021 an unprecedented bloom occurred on the French Basque coast. The objective of this study is to describe the health impact of the Ostreopsis spp bloom that occurred on the French Basque coast in 2021.
Methods
A historical cohort was conducted, including cases of possible exposure to Ostreopsis spp registered at the Centre Antipoison de Nouvelle-Aquitaine between July 1 and September 30, 2021.
Results
Of 674 patients with possible toxicity due to Ostreopsis spp, 96.9% had bathed in contaminated waters. Most of them developed respiratory tract symptoms (64.4% of patients). The time to the onset of symptoms was <6 h for 73.6% of 174 short-term (<24 h) exposed patients. The median duration of symptoms was 7.5 days for occupational (e.g., lifeguards and surfing instructors) and 3 days for recreational exposures. There were no severe cases. In total, 3% of the cases were of moderate severity, and 97% were of minor severity, according to the Poisoning Severity Score.
Conclusion
Toxic reactions caused by Ostreopsis spp are mostly benign. The clinical picture is similar to that described following exposures to Ostreopsis cf. ovata blooms in the Mediterranean area since the end of the 20th century. Ostreopsis spp are present on the Basque coast. The ecological factors promoting its blooms remain to be clarified.
Introduction
Climate change and human activity are adversely affecting the marine biotope, including the widening of the distribution of harmful algal blooms (HABs). 1 In the 2000s, HABs of the microalgal genus Ostreopsis were recorded in several areas of the Mediterranean coast,2,3 including along the Genoese coast in 2005,4,5 Algerian coast in 2009, 6 Croatian coast, 7 Catalan coast and Balearic Islands, 8 and in the Tyrrhenian Sea, 9 where such events had not been observed previously. Similar observations have been reported in Brazil and New Zealand.10,11 Bloom events predominated by Ostreopsis spp have recently been reported in the Atlantic Ocean.12–14 In 2021, a massive HAB, reported by the present authors, occurred in the waters of the French Basque Country. 15
Microalgae of the genus Ostreopsis are dinoflagellates (Alveolata) of the order Gonyaulacales and are closely related to other toxic genera, including Alexandrium and Gambierdiscus, which are also neurotoxin producers.
In the following, we describe a toxic outbreak of Ostreopsis spp on the French Basque coast during the summer of 2021, including the features of the HAB and the clinical characteristics of the affected individuals. This outbreak was mentioned in a previous paper, reporting the analytical aspect of the 2021 bloom. 15 During this event, the regional health agency opened an information page on its website, where press releases and the standardized reporting form were available. The local press also reported on the issue.
Methods
The study was based on a prospective cohort, using information received during a 3-month span by medical professionals of the Centre Antipoison de Nouvelle-Aquitaine regarding potential cases of exposure to Ostreopsis spp on the Basque coast. Cases of likely exposure to Ostreopsis spp were reported spontaneously by telephone or email to the Regional Poison Control Center (Centre Antipoison de Nouvelle-Aquitaine), the Regional Health Agency, the Emergency Medical Aid Service (Service d’Aide Médicale Urgente [SAMU]), and on-call doctors (SOS Médecins) via the French Public Health Agency (Santé Publique France [SpF]) between July 1 and September 30, 2021. All cases were registered by the Poison Control Center (PCC) and recorded in France's national database of poisoning cases using the common information system of the French PCCs (SICAP), which registers cases of possible exposure collected by the 8 PCCs in the French network. The database is authorized by the French Data Protection Authority (Commission Nationale de l’Informatique et des Libertés [CNIL], accreditation no. 747735). Individual patient consent was not required for this study according to French law regarding retrospective research conforming with MR-003 (Journal Officiel de la République Française [JORF] no. 0160 du 13 juillet 2018. texte no. 109 16 ).
Age, sex, medical history, symptoms, time of arrival at the beach, the beach or beaches frequented by the patient, the nature of the activity (e.g., swimming, walking, or surfing), recreational or occupational exposure, route of exposure (inhalation and/or cutaneous or oral), duration of exposure, ingestion of seafood products, metallic taste of the water, nature and duration of the symptoms, medical consultation, and severity were recorded.
Cases were documented using a standardized questionnaire filled in either by the patients or by the physician, lifeguard, surfing instructor, etc. (Figure 1). This questionnaire was drawn up at the very start of the epidemic by the poison control center and the medical services of the regional health agency. It was used to systematically collect the data required for the epidemiological study. It was applied prospectively, when the person filling it in was not from the poison control center, or extensively, when the person contacted the poison control center, as the data in the questionnaire are at least those collected at the time of a call to the poison control center.
Reporting form used during the epidemic.
The severity of the intoxication was estimated according to the Poisoning Severity Score (PSS), a standardized and generally applicable scheme for grading the severity of poisoning that allows a qualitative evaluation of morbidity and facilitates comparability of data. PSS includes 5 severity grades: PSS 0 (no symptoms related to the poisoning), 1 (minor symptoms), 2 (moderate symptoms), 3 (severe symptoms), and 4 (fatal poisoning). 17 All patients were followed up 1–2 months after their intoxication, and there was no report of granted compensation of any kind.
All patients who developed at least one symptom were included. The exclusion criteria were a positive COVID-19 test, symptoms prior to arrival on the coast, duplicates, inquiries, cases without clinical information, and cases unrelated to Ostreopsis spp possible exposure.
We calculated the frequencies of symptoms in the study population, as well as the medians of symptom onset times and duration. Statistical comparisons between different subgroups were made using a Fisher's exact test, χ2 test, and student’s t test.
Results
The Centre Antipoison de Nouvelle-Aquitaine received 830 reports during the 2021 season. A total of 142 patients were excluded: 40 duplicates, 5 nonmedical records, 2 reports concerning the Mediterranean, 1 patient tested positive for COVID-19, 85 reports without information on symptoms, and 9 cases unrelated to possible Ostreopsis spp exposure. Of these 688 reports, 674 were symptomatic (14 asymptomatic), and these are the ones included in the analysis.
For the entire study population and the context of potential exposure, 584 (86.6%) patients were exposed recreationally. In addition, 90 (13.4%) patients were exposed occupationally. When exposed in a professional context, there were 75 lifeguards (93%) and 6 beachfront restaurant staff (7%).
The global sex ratio (m:f) was 1.22 (sex known for 597 patients). More specifically, in the population exposed recreationally, the sex ratio was 1.0 (sex known for 510 patients) and 4.8 for professionals (sex known for 87 patients). The difference between the 2 groups is statistically significant with p < 0.001.
The global median age was 28 years, with an interquartile range [18.0; 44.8] (age known for 492 patients). More precisely, in the population exposed recreationally, the median age was 30.3 [16.0; 46.5] (age known for 411 patients) and 24.0 [21.0; 30.2] for professionals (sex known for 81 patients). The difference between the 2 groups is statistically significant with p < 0.001.
Three hundred seventy-six patients (63.8%) had been swimming, 258 (43.9%) had been resting/walking on the beach or nearby, 142 (24.1%) had been surfing, and other exposures involved 24 patients (4.1%). The activities were known for 588 patients and may be multiple (eg, swimming + surfing). In contrast, 99 patients had a single activity: 52 patients (8.8%) swam, 24 patients (4.1%) surfed, 22 patients (3.7%) rested/walked on the beachor its surroundings, and 1 patient (1%) did other activities.
The exposure was via respiratory and cutaneous routes (including people who swam) in 96.3%, whereas 3.7% were via the respiratory route only (patient with a single exposure circumstance of resting or walking without swimming).
Respiratory tract symptoms affected 64.4% of the patients, who experienced a cough and symptoms of rhinopharyngitis (eg, sore throat and rhinorrhea). Other symptoms included breathing discomfort (32.0%), eye pain (29.7%), fever (27.2%), and/or headache (26.1%). Few of our patients (8.8%) developed cutaneous symptoms of toxicity (Table 1). Associated symptoms included hyperthermia (27%), headaches (26%), myalgia (10.7%), and chills (3.3%).
Time and duration of onset of different types of symptoms (only symptoms that concerned more than 5% of patients).
IQR: interquartile range.
Details of symptom onset time were available for 344 patients, and 174 were acutely exposed (<24 h). For patients with acute exposure,it was less than 6 h for 124 patients (73.6%) and between 6 and 12 h for 16 patients (9.2%) after exposure (Table 1).
Occupational exposures resulted in more symptoms (median of 7) than recreational exposures (median of 5). The duration of symptoms was also significantly longer (p < 0.001) in patients with occupational rather than recreational exposures (median of 7.5 vs 3 days).
Of the exposed patients, 150 (22%) performed a COVID-19 test, which in all cases was negative.
We had medical history information (at least one history or no history) for 384 patients (57% of total included patients). Among them, 150 (39%) had at least one medical history (diabetes, hypertension, asthma, etc, alone or in combination), and among these, 12 (3%) were of PSS2 severity (p < 0.05).
A medical consultation related to symptoms of intoxication was sought by 146 patients (22%). Two patients were hospitalized: one for pneumonia and the other for an asthma attack, but both recovered.
The first was a 71-year-old man with a history of atrial fibrillation and a COVID-19 infection in October of the previous year. He had been vaccinated for COVID-19, and his PCR test was negative. The patient was admitted to the hospital for 10 days with radiographic evidence of lobar pneumonia, which progressed favorably on probabilistic antibiotic therapy. No etiology was found. The second patient was a 16-year-old girl with a history of asthma and allergies. She stayed in the hospital for less than a week for a flulike syndrome accompanied by persistent fever. The fever lasted 13 days. The young patient had handled and dissected macroalgae on the beach. A local erythema was noted, and she rapidly presented a clinical picture similar to that of the other patients. No etiology was found at the time of hospitalization, and the chest X-ray was normal, as was Epstein–Barr virus serology. The COVID-19 test was also negative.
There were no cases of high severity (PSS3); in 22 patients (3%), the intoxication was of moderate severity (PSS2), whereas in the majority (n = 652, 97%), it was of minor severity (PSS1).
Discussion
The vast majority of patients were exposed recreationally. They were holidaymakers and residents who frequented the beaches in the study area. The region is very popular, with a large touristic affluence at this time of year. The coast is particularly popular as a surfing spot, which may also explain the high proportion of patients exposed in this way. The portion of the population exposed professionally mainly concerns lifeguards, which is to be expected since the study period overlaps with the beach surveillance period. It should also be noted that employees of seafront restaurants were also probably exposed to Ostreopsis spp; furthermore, this was a population of interest for a study linking the biology of Ostreopsis spp and the epidemiological aspect of exposure to these algae. 18 On the Basque coast, the lifeguard population is currently the focus of interest for the PCC and occupational health services.
The slight male predominance observed in the overall population is, in reality, explained by the proportion of exposed professionals, who are almost 5 times more likely to be men.
The overall age of the patients covers a relatively wide range but is compatible with that of beach users. However, it should be mentioned that the median age of those exposed occupationally is lower, with a very narrow range, because this population is almost entirely made up of lifeguards, who are relatively young and in good physical condition to be able to practice their profession.
Exposure is not easily defined. We do not have environmental analyses for each possible exposure case. Nevertheless, given that Ostreopsis spp was detected in variable concentrations at least once on each beach during this epidemic, we considered that exposure to Ostreopsis spp was likely or at least possible under the conditions of inclusion. Moreover, it is impossible to exclude exposure, given the variability of concentrations during the same day and that the analyses were not daily in 2021. More precisely, for a patient who visited a beach on the Basque coast during the study period, knowing that Ostreopsis spp had been detected there sporadically or periodically, it is not possible to state that he was not potentially exposed.
In many cases, there was more than one risk factor (e.g., surfing and then resting on the beach). Thus, in the vast majority of cases, the exposure routes were combined cutaneous and respiratory routes, excluding the oral route (see Limitations section). Exposures solely via the respiratory route, such as walking on the beach but not swimming, were rare (3.7%), but the resulting symptomatology was the same as that of patients who had been swimming. 2 With the available information, foodborne poisonings are unlikely but cannot be discarded given that toxicological tests were not conducted.
The symptoms involved the nose and the throat—the respiratory system—and the eyes were the most commonly occurring and have been described in previous cases of Ostreopsis spp intoxication in the Mediterranean. The respiratory symptoms could have resulted not only from the toxin but also from the presence of large quantities of heterologous proteins in the respiratory system following the inhalation of contaminated spray. 19 Overall, the symptoms presented by the patients were consistent with a previous Mediterranean outbreak of Ostreopsis spp. 2 Although Ostreopsis spp exposure may cause rhabdomyolysis in humans and mice, 20 no patients underwent creatine phosphokinase testing. Therefore, it was not possible to determine whether any of the 10.7% of patients with myalgia also had rhabdomyolysis.
The time to onset of symptoms was less than 6 h in the clear majority of cases (nearly three-quarters), although a few patients had a later onset. It should be pointed out, however, that the sample of patients for whom we had this information was relatively small. Data from patients with subacute exposure (more than 24 h and less than 90 days) were not used for these calculations because they were generally difficult to use—in particular, because these patients generally reported later, generating a significant memory bias for the time of onset of symptoms. In other series, symptoms appeared within <4 h of exposure in the vast majority of patients. However, we used the time of arrival at the beach as a reference for dating exposure; although, some patients may have reached the beach and started by resting before swimming or surfing, for example. These activities may have their own levels of exposure, which could modify the delay in the onset of symptoms. In addition, the concentrations of Ostreopsis spp vary greatly during the course of a single day, particularly in relation to the tides. The duration of exposure was difficult to define. While the duration was typically very short (one swim), in a few cases, there may have been multiple exposures during the same day, in which case the daily variation in algal concentrations may have played a role. 21 As cases of subacute exposure mostly involved patients with occupational exposures, for the latter group, the duration was defined as equivalent to the contractual working time.
The number of symptoms reported by occupationally exposed patients was greater than that of recreationally exposed patients. This seems to be compatible with the fact that these patients are potentially exposed to higher doses of algae because they are exposed longer. With the same observations, the median duration of the 4 most common symptoms was about 2 days. However, for subacute exposures (eg, those in the occupational setting), the duration of symptoms was longer, ranging from twice as long for the most common symptoms and up to 7 days for breathing discomfort. This longer duration of symptoms could be explained by the fact that these patients are, by nature, exposed for more extended periods.
Information on the patient's medical history was collected for the majority of patients. It should be highlighted that patients who had at least one medical history (no matter which one), compared to those who had none, had significantly more severe clinical pictures (p < 0.05).
A medical consultation was necessary in 22% of cases, a proportion that still seems significant given the potential number of people exposed on Basque beaches during the summer. It is important to be aware of the pressure this can put on health professionals in the area and on hospitals, which already have to manage a higher flow of patients due to the seasonal influx of tourists.
To date, 11 Ostreopsis species have been taxonomically described, but molecular data suggest cryptic diversity, and some ribotypes are as yet undescribed.22,23 Among the 11 species are those able to synthesize palytoxin (PLTX) analogs, notably ovatoxins (OVTXs), which are synthesized by Ostreopsis cf. ovata, and ostreocins, which are synthesized by Ostreopsis siamensis.24–26 The link between taxonomy and toxicity is unclear. For example, strains belonging to Ostreopsis. cf. ovata, a species detected worldwide, vary from nontoxic to highly toxic. 1
The genus Ostreopsis is epiphytic and grows on macroalgae attached to rocky substrates, but cells are sometimes found in the water column, a result of their detachment from the substrate due to strong wave action. Depending on the substrate, temperature, wind conditions, and water agitation, Ostreopsis spp abundances can increase locally, ranging from a mucilaginous covering on rocks and macroalgae to an extensive brown foam on the water surface. 15 Blooms of Ostreopsis spp are mainly associated with calm waters, higher temperatures, and high nutrient availability. 27 Basque country and North of Spain have rocky coasts and increasing surface (and bottom) water temperatures above 19.5 °C for several months, which is suspected to trigger the blooms. 15 Contamination of sea spray by the microalga can cause respiratory symptoms in beachgoers, 28 including those who do not enter the water. Many patients with symptoms describe a metallic taste to the water. An additional probable route of Ostreopsis spp toxicity is the consumption of contaminated fish and shellfish. 29
The bloom in August was found to contain a mixture of 2 Ostreopsis species, O. cf siamensis and O. cf ovata, with the latter being more abundant. 15 Previously, however, the only species present on the European Atlantic coast north of Portugal was O. cf. siamensis. Toxin profiles were also analyzed from samples collected on site and showed a mixture of the PLTX analogs: OVTX-a and OVTX-b. Analyses of a collected and cultured strain of O. cf. ovata indicated that it was capable of producing OVTX-a, -b, -c, -d, and -e. This toxin profile is similar to the profiles of species found in the Mediterranean. Two other strains of O. cf. siamensis were collected and cultured, but both were negative for the 20 tested PLTX analogs. 15 It is currently not possible to make the link between patients’ symptoms and the toxins produced by O. cf. ovata, especially because it is not certain whether the symptoms are solely or partly linked to the toxins produced.
The main symptoms observed in our population were oropharyngeal pain, rhinitis, and cough (Table 1). Other authors have also described these symptoms in the Mediterranean,2,3,18,30 although their proportions in their populations were different. Gallitelli et al. 30 reported a significantly shorter delay in the onset of symptoms than that observed here. However, we considered the start of exposure to be the time of arrival at the beach, which is perhaps a difference in method with this study. Regarding the duration of symptoms, the data seem to be consistent in general with (although a little longer than) the Tichadou et al. study 2 but are much longer than those found by Gallitelli et al., who reported that symptoms generally lasted less than 24 h. Overall, it can be said that the clinical pictures observed in our series of patients are similar to those observed to date in the Mediterranean; however, the onset and duration of symptoms differ somewhat. It is possible that methodological differences in data collection could explain these differences. Interestingly, in the group with occupational exposure, the clinical picture was both longer and more marked.
During the summer of 2022, O. cf ovata was again detected on the Atlantic coast, resulting in a large number of calls to the PCC and reports to the Regional Health Agency. Causes of algal blooms in areas where they previously were unknown may include the presence of suitable substrate, sea conditions (warming of deep waters, changes in turbulence, salinity, and pH), or meteorological conditions (ambient temperature, light). 1
This toxic outbreak linked to Ostreopsis spp is an emerging situation and is likely to recur, given the persistence of the identified species in Basque waters. In health terms, an assay of toxins in marine aerosol would be of great interest.
Limitations
Our study had several limitations. First, the number of cases was determined based on voluntary reports. Thus, even if the number of cases is significant, it was likely to have been underestimated; the real number of cases was probably much higher. For example, the large community of surfers along the Basque coast is highly likely to be exposed, but they accounted for fewer than the expected cases, although the local surfing associations reported many cases. Occupational exposures were also probably underreported, as these seasonal positions are highly sought-after. Some affected workers do not provide information about their health status, likely to avoid losing their jobs. For example, some of them were reporting that their colleagues were sick, although we did not have their reports. Moreover, exposure is largely inevitable for seaside workers (presence on the beach for surveillance and entry into the water for rescue). It is possible that the relatively longer duration of the clinical pictures of occupationally exposed persons can be explained by a more important and repeated exposure.
Second, the main differential diagnosis in this period was COVID-19, although many patients tested negative. For those who were not tested, the occurrence of clustered and simultaneous cases was sufficient to suggest Ostreopsis spp intoxication. The number of cases per beach was also recorded in our study, although this was not a reliable parameter because the total number of people visiting a particular beach varied; the time of year was noted as well. In addition, the weather at different beaches may not be the same, and people who visited a beach with a lifeguard or surfing instructor may have been encouraged by that person and thus more likely to report Ostreopsis spp intoxication.
Moreover, as shown in Table 1, some of the symptoms were not described using medical terminology since many surveys were filled out secondarily by lifeguards, surfing instructors, or the patients themselves. For example, “unspecified breathing discomfort” rather than “dyspnea” was recorded when the latter was not described by a physician on-site. We chose not to over-interpret these symptoms.
Besides, the description of this toxic epidemic is based on voluntary reports with symptoms chronologically and comparatively compatible with Ostreopsis spp intoxication. It is true that these clinical pictures are not specific.
The route of oral exposure was difficult to determine. Indeed, patients who had been swimming might have ingested small amounts of seawater, but this seemed too inconsistent to consider it an oral route. Nevertheless, we considered that only patients who had eaten seafood or indicated that they had involuntarily ingested a significant amount of seawater (eg, surfers) were orally exposed.
Also, given the study design, there may have been recall bias during the follow-up by telephone, conducted at least 4 weeks after the initial exposure.
Conclusions
The vast majority of Ostreopsis spp possible exposure and intoxications are mild, resulting in upper and lower respiratory tract symptoms. Symptoms are generally mild, rarely require hospital care, and generally resolve within a few days. Occupational exposures in seaside workers may be more severe or more persistent. To date, this is the first description of a toxic epidemic potentially caused by Ostreopsis spp on the French Atlantic coast. Additional studies are needed to assess the health impact of Ostreopsis spp exposures more accurately and to understand the mechanisms behind the blooms.
Footnotes
Acknowledgments
The authors would like to thank the staff of the French poison control centers, Ifremer and hospital services, and local health professionals who contributed to the development of this work.
Author Contribution(s)
Study concept and design (CP, NC, ML); data acquisition (CP, EA, PL, MR, ML); data analysis (CP, JAV); drafting and critical revision of the manuscript (CP, NC, ML); approval of final manuscript (CP, NC, JAV, EA, PL, MR, ML).
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.