Both traumatic and toxic components are involved in stingray injuries from marine or freshwater stingray punctures, and are common in coastal regions globally1,2. In Brazil, from 2007 to 2013, most of the 4,118 injuries from aquatic animals were due to stingrays (69%) in marine and freshwater environments3. Freshwater stingray injuries are common in the Brazilian Amazon, representing 88.4% of aquatic animal injuries reported from this region3. The number of cases detected officially in the Amazon is likely to be considerably lower than the actual number, as a result of underreporting, given the difficulties faced by riverine and indigenous populations living in remote areas to reach health centers. One survey from the State of Acre, in Brazil, found stingray injuries were common, with 18% of rubber tappers and 23% of Amerindians affected at least once in their lifetime4. Moreover, delays in patient care, along with the use of homemade remedies may impair healing of the wound site and lead to a high frequency of local complications, such as secondary bacterial infections5,6. To assess the health burden due to stingray injuries through population- and hospital-based field studies it is essential to understand the extent of complications following this poorly understood health problem6.
In the Central and Western Brazilian Amazon, freshwater stingrays from the Potamotrygonidae family are more frequent, comprising four genera, Potamotrygon, Paratrygon, Plesiotrygon, and Heliotrygon, and 28 recognized species. Some species are widely distributed, such as Paratrygon aiereba, Potamotrygon motoro, P. scobina and P. orbignyi, and others more restricted, such as Potamotrygon wallacei and other endemic species7–13. The action of a stingray’s venom is necrotizing, edematous, proteolytic, neurotoxic, and myotoxic14–16. Stingray injuries result in severe local signs and symptoms, causing considerable pain and can result in ulcers that may be eventually complicated with necrosis and bacterial infection5,6,17–20. In the Brazilian Amazon, while stingray injuries have been associated with bacterial infection and necrosis, the complication frequency rate is not well known. The most severe cases associated with secondary bacterial infections required antibacterial treatment and a prolonged recovery20 with long-term disabilities21, all associated with social and economic losses. Necrotizing fasciitis due to Vibrio alginolyticus (marine stingrays) and Aeromonas hydrophila (freshwater stingrays)22,23, tetanus24, and invasive mycoses25 secondary to stingray injuries, as well as other severe infections, seem to be less common complications. Small sample sizes and the lack of a standardized clinical protocol for secondary infection definitions limit any precise estimate for these outcomes. Injuries resulting from Amazonian freshwater stingrays have also been reported in non-endemic countries, where these animals are kept as pets, and where physicians are less informed about the management of these types of injuries26–30.
Improved knowledge of epidemiological aspects of freshwater stingray injuries may likely lead to improved case surveillance of this condition in remote localities in the Brazilian Amazon. The aim of this study was to describe the profile of freshwater stingray injuries reported in the State of Amazonas, in the Western Brazilian Amazon, and to identify potential risk factors for secondary bacterial infections.
The State of Amazonas, located in the western Brazilian Amazon, is divided into 62 municipalities and comprises an area of 1,570,946.8km2, with an estimated population of 3,807,921 inhabitants, of whom more than 25% live in rural areas. The capital, Manaus, comprises approximately 45% of the entire population. Vegetation cover mainly consists of a dense evergreen rain forest. The remaining vegetal cover is primarily composed of dense macrothermic ombrophilous forest. The climate, according to the Köppen classification, is Af (super-humid equatorial), with the rainy season occurring from November to April, with pluviometric precipitations above 2,000mm per annum and average temperatures ranging from 26°C to 30°C.
All freshwater stingray injuries in the State of Amazonas that had been reported to the Brazilian Notifiable Diseases Surveillance System [Sistema de Informação de Agravos de Notificação (SINAN)] between 2007 and 2014 were included in this study. Case reports were entered by healthcare providers, including primary, secondary, and tertiary care units at the time of case reporting, usually by the physician or nurse involved in managing the patient (mainly after discharge). The variables retrieved were signs and symptoms, sex, age (in years), anatomical region of the injury, area of occurrence (rural or urban), work-related injury (yes or no), schooling (in years of study), ethnic background, the time that had elapsed from the injury until the time of medical assistance (in hours), and outcome (discharge or death). To identify factors associated with secondary infections from stingray injuries, a cross-sectional study was used where development of secondary bacterial infection was classified as the dependent variable.
Data were aggregated according to the municipality and the year of occurrence to highlight changes in the epidemiological profile considering place and time. The mean incidence, calculated as a ratio of the number of cases and the population of each municipality multiplied by 100,000 was used for mapping. The software ArcMap 10.1 in ArcGIS 10.1 (ESRI, USA) was used for this analysis.
Monitoring of the database cleaning and analysis was undertaken using the estimates of internal validity (the extent of errors within the system, for example, coding errors) and completeness of data (underreporting of any surveillance variable). A check of both surveillance attributes was undertaken by two independent researchers prior to analysis to minimize a possible observer-expectancy effect during database handling. The non-parametric Spearman’s correlation coefficient was used to assess the association between the absolute number of cases and the altimetric river levels. Information in regard to the altimetric river levels was provided from the hydrological information system of the National Water Agency [Agência Nacional de Água (ANA)]31. Secondary bacterial infection rates were compared using a chi-square test (corrected using Fisher’s test, if necessary), using individual characteristics from the SINAN database as independent variables. To avoid potential selection bias related to the high frequency of underreporting in the final database, only variables with at least 70% of completeness were considered for this analysis. The crude odds ratio (OR) with its respective 95% confidence interval (CI) was then determined. A backward-stepwise logistic regression was used for the multivariable analyses and the adjusted ORs with 95% CIs were also calculated. All variables associated with outcomes at a significance level of p < 0.2 in the univariate analysis were included in the multivariable analysis. A p-value < 0.05 was considered to be statistically significant using the Hosmer-Lemeshow goodness-of-fit test. Statistical analyses were performed using the Stata statistical package version 13 (Stata Corp. 2013).
This study was approved by the Ethics Review Board (ERB) (approval number 713.140/2014). Data were handled anonymously; therefore, the ERB waivered the informed consent process. The images presented in the manuscript derive from a prospective project held at the Health Surveillance Foundation of Amazonas, and were approved by the same ERB (approval number 713.140/2014).
A total of 476 freshwater stingray injuries were recorded in the State of Amazonas between 2007 and 2014, resulting in an incidence rate of 1.7 cases per 100,000 person/year. Regarding seasonality, there was a higher incidence of cases between August and December (Figure 1). There was a negative correlation between the absolute number of cases and the altimetric river levels (p < 0.001, linear Rs = -0.420). Most of the stingray injuries occurred in males (392 patients; 82.4%). The most affected age groups involved patients between 11 and 20 years old (154 patients; 32.4%) and between 21 and 30 years old (80 patients; 16.8%). Regarding the area of occurrence, 73.8% were reported in rural areas. A total of 43.2% of the patients had up to 4 years of schooling. In terms of ethnicity, a mixed ethnicity was most frequently recorded (85.2%). Work activity-related injuries comprised 26.1% of reported injuries. Maintenance and repair services were the most cited formal occupations of patients (56.7%), followed by agricultural and forestry activities (39%). Most of the injuries occurred to the lower limbs (95.6%). Regarding time elapsed from the sting until receipt of medical assistance, 74.5% of the patients received treatment within the first three hours following injury. No deaths from freshwater stingray injuries had been recorded in the period studied. Aside from occupation, all other variables presented data completeness to at least 70% (Table 1).
|Characteristics (data completeness expressed as percentages)||Number||Percentage|
|Sex (n = 476; 100%)|
|Age group (years; n = 476, 100%)|
|Area of occurrence (n = 466, 97.9%)|
|Years spent at school (n = 333, 70.0%)|
|Ethnicity (n = 467, 98.1%)|
|Work-related accident (n = 452, 94.96%)|
|Occupation (n = 323, 67.9%)|
|maintenance and repair services||183||56.7|
|trade and services employee||6||1.9|
|Anatomical region of the injury (n = 473; 99.4%)|
|Time elapsed from sting to medical assistance (hours; n = 462, 97.06%)|
|Outcome (n = 452, 94.96%)|
Freshwater stingray injuries were unevenly distributed across the study area, with records obtained from 36 of the 62 (58.1%) municipalities of the state. The regions with the highest incidence rates were Alvarães (77.2 cases/100,000 inhabitants), Uarini (51.5/100,000 inhabitants) and Silves (20.4/100,000 inhabitants).
Table 2 presents local and systemic manifestations observed for freshwater stingray injuries. The most frequent local signs and symptoms observed were pain (99.1%), edema (65.3%), ecchymosis (15.6%), bleeding (12%), and erythema (2.9%). Secondary infections and necrosis were observed in 8.9% and 3.8% cases, respectively. Figure 2A, Figure 2B, Figure 2C, Figure 2D, Figure 2E, Figure 2F, Figure G and Figure 2H present some of the clinical cases with local manifestations recorded in the study area.
|Characteristics (n = 452, complete data, 94.3%)||Yes||Percentage|
|Acute local manifestations|
|Chronic local manifestations|
Table 3 summarizes the results of the univariate and multivariate logistic regression analysis evaluating factors associated with secondary infection. Work-related injuries (OR, 4.1; p < 0.001) and a time of >24 hours from being stung until receiving medical assistance (OR, 15.5; p < 0.001) were independently associated with the risk of secondary bacterial infection.
|Variable||Secondary infection||Crude OR (CI 95%)||p-value||AOR (CI 95%)||p-value|
|yes (%)||no (%)|
|male||37 (9.4)||355 (90.6)||Ref|
|female||3 (3.6)||81 (96.4)||0.36 (0.11-1.18)||0.091||0.36 (0.08-1.64)||0.187|
|Area of occurrence|
|urban||6 (4.9)||116 (95.1)||Ref|
|rural||33 (9.6)||311 (90.4)||2.05 (0.84-5.02)||0.116|
|white||2 (6.3)||30 (93.8)||Ref|
|mixed||34 (8.5)||364 (91.5)||0.98 (0.39-2.43)||0.967|
|Indian||4 (14.8)||23 (85.2)||1.95 (0.64-5.95)||0.240||2.59 (0.63-10.69)||0.187|
|Black||0 (0.0)||7 (100.0)|
|Asian||0 (0.0)||3 (100.0)|
|Years spent in school|
|illiterate||1 (4.8)||20 (95.2)||Ref|
|0-4||15 (12.2)||108 (87.8)||1.94 (0.90-4.18)||0.089||1.39 (0.50-3.84)||0.525|
|5-8||10 (7.4)||126 (92.6)||0.74 (0.33-1.65)||0.467|
|>8||3 (1.0)||304 (99.0)||0.59 (0.17-2.01)||0.396|
|no||18 (5.4)||316 (94.6)||Ref|
|yes||21 (17.8)||97 (82.2)||3.80 (1.95-7.42)||<0.001||4.13 (1.87-9.13)||<0.001|
|Anatomical region of the injury|
|lower limb||36 (8.0)||416 (92.0)||Ref|
|head||2 (50.0)||2 (50.0)||11.34(1.55-82.79)||0.017||11.64(0.46-292.81)||0.136|
|upper limb||2 (12.5)||14 (87.5)||1.57 (0.35-7.19)||0.558|
|body||0 (0.0)||1 (100.0)|
|Time elapsed from sting to medical assistance (hours)|
|0-3||10 (3.1)||312 (96.9)||Ref|
|4-6||5 (9.1)||50 (90.9)||1.12 (0.42-3.00)||0.828|
|7-12||2 (25.0)||6 (75.0)||3.82 (0.74-19.68)||0.109||4.17 (0.31-56.44)||0.283|
|13-24||0 (0.0)||5 (100.0)|
|>24||19 (45.2)||23 (54.8)||18.13 (8.32-39.47)||<0.001||15.48 (6.77-35.40)||<0.001|
Few previous studies have quantified the burden of stingray injuries in the Brazilian Amazon3,4. This study shows that stingray injuries prevailed across all of the study area, with a higher incidence in males living in rural areas. We considered that the lower odds of riverine and indigenous populations reaching health centers most likely resulted in probable underreporting in the region. According to estimates concerning the rate of stingray injuries to people at least once in their lifetime, epidemiological surveillance in rural and indigenous areas would only have a sensitivity of approximately between 5% and 10% in the Western Amazon4. Thus, assessing the burden of stingray injuries using population- and hospital-based field studies in remote areas highlights a large gap in the epidemiology of stingray injuries. Stingray injuries occurred mostly in the younger study population (between 11 and 30 years of age), with a marked seasonality that suggested leisure activities, commonly undertaken in the dry season on beaches that appear along the river banks, were a risk factor for stingray injuries31. Our results showed that 26.1% of the stingray injuries were classified as work-related stingray injuries; therefore, there is a need for public health strategies that aim to reduce the incidence of these injuries, especially for rural workers19,20,31.
Although no deaths or permanent disabilities from freshwater stingray injuries were recorded in the period, severe pain symptoms were commonly observed. Stingray injury victims frequently reported intense pain that seemed out of proportion to the injury1,32,33. Moreover, it has been reported that the intensity of such severe pain has led to disorientation in the injured victim34. Edema, ecchymosis, and local bleeding were also recorded in this case series, consistent with reported literature33. A comparative morphological analysis of the epidermal tissue of the stinger in different marine and freshwater Brazilian stingrays indicates that, in freshwater species, there is a larger number of protein secretory cells spread over the whole epidermal layer of the stinger while, in marine species, the protein secretory cells are located only around or inside the stinger’s ventrolateral grooves35. These differences between the stingers of the two groups help explain the more severe complications following accidents with the freshwater species than with the marine species, especially necrosis19. Most patient injuries occurred on the lower leg or foot. When a stingray is inadvertently disturbed or stepped on, it reflexively swings its barbed tail upwards, and can inflict deep puncture wounds19,36,37. No information regarding the treatment provided to injured patients was available from the official databases. For stingray injuries, immediate first aid treatment involves hot water immersion (45°C) for up to 90 minutes because the venom has a thermolabile nature and, because of the vasodilatory effects of immersion in hot water, symptomatic relief is provided while the limb is immersed. However, pain often returns once the limb is removed from the hot water19,35,36,37. It has been suggested that, in the absence of evidence from controlled clinical trials, when hot water immersion is insufficient in terms of pain relief, oral analgesia and titrated intravenous opioids should be administered while arranging local anesthesia at the wound site or administering a regional nerve block33. All penetrating injuries require irrigation and cleansing, and larger wounds or those containing debris require surgical exploration to extract any remaining imbedded tail fragments as well as wound debridement19,33. In general, our findings showed that systemic signs and symptoms were benign, a finding similar to previous studies5,38.
Penetrating stingray wounds may present with delayed healing and secondary infection, with these wounds having worsened due to the dermonecrotic effects of the venom33,38. Secondary infection has been reported to be the most important complication of stingray injuries19,33,36,37, and occurred in 9% of the patients in this study. Given the size, penetrating nature, and slow healing of stingray injuries, there is a significant risk for secondary infection39. One study considered secondary infections were more likely to occur in unclean and larger wounds33, but risk factors for secondary infections are poorly understood. In this study, work-related injuries and a time >24 hours from being stung until obtaining medical assistance were independently and significantly associated with the risk of secondary infection. This association between work-related injuries and secondary infection prevalence was not expected. Since reporting of work-related injuries in Brazil is compulsory, stingray injuries, especially those that progress to complications such as secondary infection, are more likely to be reported than accidents occurring within other groups, and reporting bias may have elevated this association. However, specific hygiene behaviors, including the use of homemade medicines and environmental workplace exposures predisposing to secondary infection, cannot be discarded21,22,40.
In this study, the rate of secondary infection was 9%. Estimates of secondary infection rates range from rare cases in one observational study33. Delayed patient care at a health unit was found to increase the risk of secondary infection up to 15 times in the Brazilian Amazon region. In this region, poor access to health centers is a reality due to long distances and a lack of transportation in remote areas. The implementation of a rapid transport system for stingray injury patients, integrated with providing assistance for other health issues, is likely to significantly reduce morbidity and probable disability rates related to secondary infections. Immediate first aid for stingray injuries is simple, involving hot water immersion and wound cleansing, and this information should be incorporated in travel medicine and occupational health programs within endemic areas.
Medical management of secondary infection from stingray injuries remains controversial, partly because of the limited number of evidence-based management protocols available6,33,38. Bacterial strains and their antimicrobial resistance profile in general are not determined for stingray wounds in the Amazon, particularly in remote settings, and the treatment for secondary infection is generally empirical. The most common agents in P. motoro stingray mucus are Gram-negative rods, namely Aeromonas spp., including B-lactam-resistant bacterial strains with the potential to cause severe secondary infection in wounds acquired during stingray envenoming40,41. Trimethoprim/sulfamethoxazole, ciprofloxacin, or tetracycline have been suggested for the treatment of fish wound infections36. Treatment failure has previously been reported in Central Brazil for ciprofloxacin21 and in California, United States, regarding trimethoprim/sulfamethoxazole42 for patients presenting with stingray wound infections.
In this study, analysis was limited to the existing fields of the SINAN reporting forms. This may have impaired the identification of risk factors for secondary bacterial infections. Information on first aid and management post-injury were not evaluated in this study as this information was not available from SINAN. The heterogeneity of the observers and the incomplete data concerning the natural history of the patients must be considered as limitations when analyzing these data. We highlight that estimates of independent risk factors for complications, using a more complete set of independent variables, should be assessed in future multicentered prospective studies. Moreover, the nature of the surveillance system may have influenced record keeping, for example, patients in remote areas with mild stings may not have reported their injuries to health services. However, the broad population coverage of the official surveillance and the low cost for data collection allowed us to obtain valuable information for the primary care health system due to the large sample size available for analysis. Furthermore, our results indicate a sufficiently strong association to suggest a likely causal connection.
In conclusion, our study showed a wide distribution of stingray injuries, with a higher incidence in males residing in rural areas. The frequency rate for secondary infections following stingray injuries was 9%. If the time from being stung until receiving medical care was >24 hours, this delay in treatment was the major factor associated with the risk of secondary bacterial infection. The possibility of reducing local effects and sequelae through the use of early antibiotic therapy for secondary infection, anti-inflammatory medication, and complementary treatments needs to be further investigated in line with good clinical practice. Cooperative efforts towards the control of this poorly recognized health problem through research and surveillance partnerships, particularly in the Amazon region, are imperative. Moreover, training multidisciplinary teams in stingray injury management, case monitoring, and surveillance is needed in Amazonian health services.