Nationwide and long-term epidemiological research of snakebite envenomation in Taiwan during 2002–2014 based on the use of snake antivenoms: A study utilizing National Health Insurance Database

Abstract

Introduction

In Taiwan, six venomous snake species with medical importance have been found; however, long-term epidemiological data of snakebite envenomation (SBE) is lacking. This study aimed to explore the epidemiology of SBE based on the distribution and use of different antivenoms in different parts of Taiwan to facilitate the development of prevention strategies and resource allocation.

Methods and results

This retrospective study was conducted using the Taiwan National Health Insurance Research Database from 2002 to 2014. A total of 12,542 patients were treated with antivenoms. The directly standardized cumulative incidence was 3.6 cases per 100,000 individuals based on the 2000 World Standard Population. The incidence of SBEs peaked in the summer (35.9%). The relative risk (RR) of male patients versus female patients was 2.5 (p < 0.0001). The RRs of patients aged 18–64 and ≥65 years versus those aged <18 years were 6.0 (p < 0.0001) and 14.3 (p < 0.0001), respectively. Furthermore, the RR of eastern Taiwan versus northern Taiwan was 6.8 (p < 0.0001). The RR of agricultural workers versus laborers was 5.5 (p < 0.0001). Compared with patients envenomed by Trimeresurus stejnegeri stejnegeri or Protobothrops mucrosquamatus, those envenomed by Naja atra or Bungarus multicinctus multicinctus were more likely to occur in central (adjusted odds ratio [aOR] = 2.6, p < 0.0001) or southern (aOR = 3.2, p < 0.0001) Taiwan, but less frequently among agricultural workers (aOR = 0.6, p < 0.0001). The overall case-fatality rate was 0.11%.

Conclusions

Among Asian countries, Taiwan had low incidence and case-fatality rates of SBE. Risk factors included male gender, old age, summer season, being in eastern Taiwan, and being an agricultural worker. Differences of the epidemiological findings between snake species should be focused on when developing strategies for snakebite prevention.

Introduction

In 2017, the World Health Organization (WHO) reinstated snakebite into the list of “neglected tropical diseases” to emphasize its threats to humans [1]. More than 3,000 snake species has been discovered worldwide, around 600 of which are venomous [2]. The annual number of venomous snakebites is about 0.4–2.5 million [2,3]. The top three areas with the highest number of snakebites are South Asia, Sub-Saharan Africa, and Southeast Asia, with incidence rates ranging from 7.8 to 84.7 cases per 100,000 person-years [2]. The case-fatality rate of snakebite envenomation (SBE) due to serious complications or mismanagement is around 5% [2,3].

Snake species, human activities, health-seeking behavior, and appropriateness of the antivenom administration or surgical intervention are among the known factors associated with snakebite complications. Snakes inhabit area at different latitudes and altitudes, and their activities vary depending on environmental factors, such as season, temperature, and humidity [4–8]. People bitten by snakes were more likely to be young men with more outdoor activities [5–15], especially those engaged in agricultural or pastoral labor [9,11,14–16]. According to the literatures, key factors associated with complications and even death from venomous snakebites include delay in seeking medical care [9,17], inappropriate pre-hospital care [9,15], incorrect snake species identification [18], insufficient antivenom administration [19,20], adverse effects of antivenom [12], and lack of designated treatment guidelines [21,22].

Taiwan is an island in southeastern Asia with an area of 33,883 km2. The total population was 23,492,074, and the number of people in agriculture was 2,710,680 (11.5%) in 2022, respectively [23,24]. Six venomous snake species with medical importance were found: Trimeresurus stejnegeri stejnegeri, Protobothrops mucrosquamatus, Deinagkistrodon acutus, and Daboia siamensis in the Viperidae family and Naja atra and Bungarus multicinctus multicinctus in the Elapidae family [25]. The Taiwan Centers for Disease Control (CDC), the only snake antivenom manufacturer, produces four types of antivenom in lyophilized form, which are ammonium sulfate-precipitated F(ab’)2 fragments [26]. There are two bivalent specific antivenoms (one for T. s. stejnegeri and P. mucrosquamatus, and one for N. atra and B. m. multicinctus) and two monovalent antivenoms (one for D. acutus and one for D. siamensis). The diagnosis and management of SBEs followed the guidelines of the Taiwan Poison Control Center as no commercial venom serum test was available during the study period [26]. Briefly, the culprit snake was diagnosed by examining the snake brought or pictured by the patient or their associates. If not available, the diagnosis was made by asking the patient to identify the snake in a picture provided in the emergency department or by identifying typical manifestations through physical examination, serial wound inspection, and relevant medical history [27–33]. The recommended dosage of relevant antivenom to treat a envenomed patient is 1–2 vials for T. s. stejnegeri, 2–4 vials for P. mucrosquamatus, 2–4 vials for D. acutus, 2–4 vials for D. siamensis, 6–10 vials for N. atra, and 2–4 vials for B. m. multicinctus bite envenomings. The incidence rate of SBE decreased from 8.8 cases per 100,000 person-years during 1904–1938 to 4.5 cases per 100,000 person-years during 2005–2009; the case-fatality rate also decreased from 3%–7% to 0.04%–0.08% during the same period, respectively [34,35]. The decline in incidence was related to lifestyle changes, while the large reduction in case-fatality rate might be due to easy access to modern medicine, immediate availability of antivenom in healthcare facilities, and few adverse effects of antivenoms [25,26].

In Taiwan, nationwide and long-term epidemiological research on SBE remains lacking after 1940s [25]. This study aimed to explore the epidemiology of SBE based on the distribution and use of different antivenoms in different parts of Taiwan. The National Health Insurance Research Database (NHIRD) was used to investigate the demographics and clinical features of SBE to facilitate the development of prevention strategies and resource allocation.

Methods

Patient and public involvement

Patients or the public were not involved in the design, or conduct, or reporting, or dissemination plans of our research.

Study population and setting

The NHI system provides uniform insurance coverage for everyone. This retrospective study was conducted using the NHIRD from 2002 to 2014. During this period, the Taiwan CDC distributed four types of antivenoms against the six medically important snakes to medical facilities: bivalent antivenom for T. s. stejnegeri and P. mucrosquamatus (FH), bivalent antivenom for N. atra and B. m. multicinctus (FN), and monovalent antivenom for D. acutus (FA); monovalent antivenom for D. siamensis (FD) was not distributed until 2008 [25]. Patients could only obtain antivenom from the CDC through the NHI system, which is mandatory, and their medical records were detailed in the NHIRD. Those treated with the aforementioned antivenoms during 2002–2014 were included for analysis.

Data collection

In the NHIRD, each drug or procedure was designated a specific code. We retrieved records from the NHIRD regarding the use of FH, FN, FA, and FD. Patient demographics, complications, management, and outcomes associated with the use of such antivenoms were then obtained. The diagnosis was adherent to the International Classification of Diseases (ICD)-9-Clinical Modification or ICD-10.

Definition of the variables

Patient demographics included sex, age, comorbidity, occurrence date, geographic region of bite event, and occupation. The patients were divided into three groups: underage (<18 years), adults (18–65 years), and elderly (>65 years). The severity of the patient’s comorbidity was stratified according to the Charlson Comorbidity Index (CCI) score [36]. Occurrence date was defined as the date of receipt of the first dose of antivenom. The months were categorized into spring (March, April, May), summer (June, July, August), autumn (September, October, November), and winter (December, January, February). Geographic region of bite event was defined as the location of the medical facility where the first dose of antivenom was administered, categorized into northern (Taipei, New Taipei, Keelung, Taoyuan, Hsinchu, Miaoli, Lianjiang, Kinmen), central (Taichung, Changhua, Nantou), southern (Yunlin, Chiayi, Tainan, Kaohsiung, Pingtung, Penghu), and eastern (Yilan, Hualien, Taitung) regions [37–39]. The forest area coverage of the region was defined as the ratio of the forest area to the land area based on aerial survey and field investigation [40]. Occupation was categorized into agricultural worker, laborer, and others.

A single envenomation event was arbitrarily defined as all records within 90 days after the first dose of antivenom. Anaphylactic reaction to antivenom administration was defined as being given epinephrine intravenously for shock, bronchospasm, or angioedema within 1 day following the first dose [41]. Wound infection and antibiotic use were defined as having erysipelas, cellulitis, fasciitis, or myositis and receiving oral or intravenous antibiotics within 7 days, respectively [42]. Surgery was defined as undergoing incision, division, excision, debridement, transplantation, reconstruction, grafts, or amputation within 14 days. Neurological complication was defined as experiencing paralysis, stroke, and encephalopathy within 30 days. Endotracheal intubation was defined as receiving invasive mechanical ventilation within 7 days. Death without other apparent causes within 90 days or during the same visit was defined as SBE related.

Statistical analysis

Cumulative incidence was calculated annually as the total number of occurrences divided by the total number of individuals in the at-risk population. The directly standardized incidence was calculated based on the sex and age distribution in the demographic data from the Taiwan government and the 2000 World Standard Population from the WHO. Due to the skewed count data distributions, the Poisson regression model was used to analyse the relative risks (RRs) of the directly standardized incidence stratified by sex, age, season, geographic region, and occupation. Trend analysis in the general linear model was conducted to observe the annual changes. Spearman’s rank correlation coefficient was calculated to check the agreement on the ranking of the results between the cumulative incidence and forest area coverage in each county. A logistic regression model was used to identify factors related to the antivenom types and dosage, surgery, and death. Variables with a P value ≤0.25 in the univariable analysis were added in a stepwise manner; only those with a two-sided P value ≤0.05 were considered statistically significant and used in the final multivariable analysis. Sex, age, and CCI score were included in the final model regardless of the significance level. Statistical Analysis Software version 9.4 (SAS Institute, Cary, NC, USA) was used for the statistical analyses.

Results

Cumulative incidence

During 2002–2014, 12,542 patients were treated with antivenoms. The average number of cases per year was 965 (range, 871–1,072), and the crude incidence was 4.2 cases per 100,000 individuals (Table 1). The directly standardized incidence was 3.6 cases per 100,000 individuals, with a significant decrease from 4.5 cases per 100,000 individuals in 2002 to 3.1 cases per 100,000 individuals in 2014 (p = 0.0002) (Fig 1).

[Figure omitted. See PDF.]

Fig 1. The annual case number and directly standardized incidence of patients receiving antivenoms in Taiwan (N = 12,542).

*Data is available in S1 Data file.

https://doi.org/10.1371/journal.pntd.0011376.g001

[Figure omitted. See PDF.]

Table 1. Characteristic distributions of SBE cases based on use of snake antivenom, and calculation of incidence and relative risk by characteristics (N = 12,542).

https://doi.org/10.1371/journal.pntd.0011376.t001

Sex and age

Of the 12,542 patients, 70.1% were male, and 28.3% were female. The directly standardized incidences for men and women were 5.1 and 2.0 cases per 100,000 individuals, respectively, and the RR of men versus women was 2.5 (p < 0.0001) (Table 1). Their median age was 53 years (men, 51 years; women, 57 years). The RRs of the elderly and adults versus the underage were 14.3 (p < 0.0001) and 6.0 (p < 0.0001), respectively (Table 1). Age exhibited an increasing trend, with an average increase of 0.5 years per year (p < 0.0001). The proportion of the elderly had gradually increased from 21.4% in 2002 to 27.6% in 2014.

Month of occurrence

The incidence peaked in August, which was summer, accounting for 13.4% of all envenomation events. Contrarily, January had the lowest incidence, which was winter, accounting for 2.5% of all envenomation events. The incidences of patients given FH were evenly distributed during May–November, whereas those of patients given FN, FA, and FD were during July–September, August, and June–July, respectively.

Geographic region

The proportions of cases in northern, central, southern, and eastern Taiwan were 34.5%, 14.9%, 25.7%, and 25.0%, respectively. The directly standardized incidence in eastern Taiwan was the highest, which was 19.6 per 100,000 individuals, whereas those in northern, central, and southern Taiwan were 2.8, 2.8, and 2.9 cases per 100,000 individuals, respectively (Fig 2). The RR of envenomation in eastern Taiwan versus northern Taiwan was 6.8 (p < 0.0001) (Table 1). The geographic variation in incidence was found to be possibly related to the forest area coverage, with a Spearman’s rank correlation of 0.65 between cumulative incidence and forest area coverage in each county (p = 0.0025).

[Figure omitted. See PDF.]

Fig 2. The directly standardized incidence by region in patients receiving antivenoms in Taiwan (N = 12,542).

* Fig 2 was generated using Statistical Analysis Software version 9.4 (SAS Institute, Cary, NC, USA). The link to the base layer of the map is: https://data.gov.tw/dataset/7442. Data is available in S1 Data file.

https://doi.org/10.1371/journal.pntd.0011376.g002

Occupation

Of the 12,542 patients, 40.2% were agricultural workers and 24.8% were laborers. The directly standardized incidences for agricultural workers and laborers were 10.9 and 1.9 cases per 100,000 individuals, respectively, and the RR of agricultural workers versus laborers was 5.5 (p < 0.0001) (Table 1). Contrary to the annual decrease of 0.06 cases per 100,000 individuals among laborers (p = 0.0018), the annual incidence among agricultural workers did not significantly change during the study period (p = 0.3524).

Agricultural workers and laborers were found to have different characteristics. Among agricultural workers, the male-to-female ratio was 1.5, whereas among laborers, it was 2.3. Furthermore, the median age of agricultural workers was 60 years, whereas that of laborers was 44 years. Also, the proportions of agricultural workers in northern, central, southern, and eastern Taiwan were 22.6%, 19.7%, 35.9%, and 21.8%, respectively, whereas for laborers, the proportions were 50.9%, 12.0%, 17.8%, and 19.3%, respectively.

Type and dose of antivenom

The proportions of patients given FH, FN, FA, FD, and mixed antivenoms were 73.3%, 17.1%, 0.6%, 0.1%, and 8.8%, respectively (S1 Table); 98.2% of the patients received at least one dose of FH or FN. Compared with patients given FH, those given FN were more likely to be male, bitten during the summer, in central or southern Taiwan, and laborer (Table 2). The average doses of patients given FH, FN, FA, FD, and mixed antivenoms were 2.5, 3.4, 3.0, 2.4, and 5.6 vials, respectively. Furthermore, the proportions of patients given 1, 2, 3–4, and ≥5 vials of antivenom were 42.8%, 22.9%, 17.5%, and 16.9%, respectively. The dose of antivenom in a patient ranged from 1 to 40 vials, with an average of 3.0 vials; moreover, a significant increase of 0.04 vials per person per year was observed (p = 0.0189).

[Figure omitted. See PDF.]

Table 2. Logistic regression analyses of the factors associated with FN antivenom administration with respect to FH antivenom administration (N = 12,542).

https://doi.org/10.1371/journal.pntd.0011376.t002

Complications and management

Of the 12,542 patients, only 0.1% experienced anaphylactic shock and was treated with epinephrine. Although systemic antibiotics were administered to as high as 81.8% of the patients, only 21.7% received antibiotics for a defined infection, and only 1.7% were diagnosed with necrotizing fasciitis. Among those given antibiotics, 70.0% received cephalosporins or penicillins alone, and 21.0% received cephalosporins or penicillins plus aminoglycosides. Surgery was performed in 7.3% of the patients, including incision or division in 2.7%, excision or debridement in 6.1%, transplantation or reconstruction in 2.8%, and amputation in 0.3%. Neurological complications occurred in 1.4%, and 0.7% underwent endotracheal intubation.

In the multivariable analyses, the antivenom type was not associated with anaphylactic shock; however, patients given FN were more likely to have infections (adjusted odds ratio [aOR] = 1.19, p = 0.0036), necrotizing fasciitis (aOR = 7.80, p < 0.0001), surgery (aOR = 3.36, p < 0.0001), neurological complications (aOR = 3.10, p < 0.0001), and endotracheal intubation (aOR = 9.26, p < 0.0001) than those given FH.

Death

A total of 14 patients died, with a case-fatality rate of 0.11%. Among them, eight and six patients died after receiving FH and FN, with case-fatality rates of 0.09% and 0.28%, respectively. Of the 14 patients, 6 died within 7 days, and 8 died within 52 days. The main causes of death were renal failure, intracranial hemorrhage, and respiratory failure. No one died from anaphylactic shock or necrotizing fasciitis.

Discussion

Under-reporting of snakebite incidence and mortality is common in South and Southeast Asian countries and many SBEs land up with alternate therapies. In Nepal, for example, hospital reports suggested that about 20,000 patients were admitted and 1,000 died from snakebites each year, yet figures collected in a community based study suggested that 26,749–37,661 people bitten by snakes and 2,386–3,225 deaths per year across only the Terai region [43]. Similarly, hospital records revealed approximately one third of snakebite incidents found in the community survey in Vietnam [44]. We therefore utilized NHIRD based on antivenom distribution and use to analyze patients envenomed by snakes though the actual epidemiology of SBE would be somewhat different because there is a possibility that some mildly envenomed patients may not receive antivenom, or some patients may have received antivenom inappropriately. However, alternative therapies for SBE do not occur or are extremely rare in Taiwan. Traditional healers are illegal because all medical practices are regulated by laws. Therefore, we believe our study findings are representative for individuals with SBE.

A decrease in the directly standardized incidence of patients given antivenoms was observed during the study period. Patients envenomed by T. s. stejnegeri and P. mucrosquamatus (i.e., treated with FH) were predominant, followed by patients envenomed by N. atra and B. m. multicinctus (i.e., treated with FN). The factors found to be associated with the increased risk of SBE were male gender, old age, summer season, being in eastern Taiwan, and being an agricultural worker. In patients envenomed by N. atra and B. m. multicinctus, the risk factors were male gender, summer season, being in central or southern Taiwan, and being a laborer. Compared with patients envenomed by T. s. stejnegeri and P. mucrosquamatus, those envenomed by N. atra and B. m. multicinctus were more likely to have infections, necrotizing fasciitis, surgery, neurological complications, and endotracheal intubation. However, the case-fatality rate for any venomous snakebite was quite low.

This study demonstrated that the directly standardized incidence in Taiwan was 3.6 cases per 100,000 persons, which was relatively low among Asian countries [2]. There were 2.5 times as many male patients as female patients, similar to Taiwan in the 1940s [34] and other countries in Southeast Asia [10–13], but slightly higher than in East and South Asia [5–9], presumably representing the ratio of males to females engaging in outdoor work or activities. The patients’ median age was 53 years, similar to that in East Asia [6–8] but significantly higher than that in Taiwan in the 1940s [34], Southeast Asia [10–13], and South Asia [5,9]. The increasing age of the patients was consistent with the aging population in Taiwan.

SBE occurred more frequently during the warm seasons [45], consistent with previous studies [5–8,34,35,37,46–52]. T. s. stejnegeri and P. mucrosquamatus envenomations were frequent during June–November, whereas N. atra and B. m. multicinctus envenomations mainly peaked during June–August. SBE was at least five times more likely to occur among agricultural workers than laborers; snakebite could also be a significant occupational injury associated with agricultural activities [9,11,14,16]. With the industrial transformation in Taiwan since the 1950s, young people were more inclined to engage in non-agricultural work, whereas the elderly who remained as agricultural workers seemed to be at the highest risk of SBE. In eastern Taiwan, which has large forest area coverage, the incidence of SBE was at least six times that in other areas, due to not only increased exposure to snakes because of agricultural work but also the development of sightseeing agriculture and popularity of recreational tourism in forest areas.

In this study, FH and FN were used in majority of the patients, consistent with the results of previous studies in Taiwan; contrarily, patients treated with FA or FD were much fewer than previously reported [34,37]. T. s. stejnegeri and P. mucrosquamatus envenomations were predominant throughout the island, and N. atra and B. m. multicinctus were identified as critical species in central and southern Taiwan [34,37,47]. Although N. atra appeared more frequently in lower-altitude farmlands than T. s. stejnegeri and P. mucrosquamatus [25,34,37,47], agricultural workers were more likely to receive FH than laborers. This might be because agricultural workers who had been working in the farmland for a long time were more familiar with the habits of N. atra and were not easily bitten.

In Taiwan, the main effects of T. s. stejnegeri and P. mucrosquamatus bites were tissue swelling, pain, and local ecchymosis, whereas systemic effects were uncommon [32,33]. N. atra bites could cause local tissue necrosis and secondary wound infection, whereas B. m. multicinctus bites could cause neuromuscular paralysis leading to respiratory failure [27,30]. Although snake species could not be specified based on the use of bivalent antivenom, our study findings indicated that patients given FN were more likely to undergo surgery due to serious wound complications caused by the cytotoxins in N. atra venom [30,31,53,54] or endotracheal intubation due to respiratory failure caused by B. m. multicinctus envenomation [27].

Patients may require surgery due to local tissue necrosis, wound infections such as necrotizing fasciitis, or inappropriate first-aid measures [9,55]. Overall, 7.3% of the patients in this study underwent surgery, which was lower than those in other Asian [7,53,56–58] and sub-Saharan African [15,59,60] countries but similar to those in European and American countries [61–64], although guidelines on the timing of surgical intervention after snakebites are inconclusive [31,41,65–68]. To reduce the need for surgery, patients should receive an adequate amount of antivenom [69,70].

Diagnosing wound infections in SBEs can be difficult and may be overestimated if deep tissue or biopsy cultures are not performed for the wound. However, we observed that 76.8% to 80.9% of N. atra bite patients developed wound infections in studies conducted by Mao et al [30,71]. Out of the 59 and 87 patients, 50 and 73, respectively, had positive wound bacterial cultures [30,71]. In addition, 22 out of 27 and 30 out of 33 had positive deep tissue or biopsy cultures, respectively [30,31]. Furthermore, necrotizing soft tissue infections, including necrotizing fasciitis, were observed in 42.1% to 44.7% of N. atra bites [30,31]. On the other hand, wound infections are less common in T. s. stejnegeri and P. mucrosquamatus (crotaline) bites, with rates of 11.4% and 27.4%, respectively [32,33]. In this study, which utilized the NHIRD, we used criteria similar to those in a recent report to identify SBE wound infections [42]. Our findings indicate that 20.3% of patients who received FH antivenom (i.e., antivenom for T. s. stejnegeri and P. mucrosquamatus) developed wound infections, which is consistent with previous studies conducted by Mao et al. [32,33], and lower than the study conducted by João Victor Soares Coriolano Coutinho et al.[42]. Although the incidence of wound infection might still have been overestimated in this study, we believed the overestimate was likely to be of limited magnitude.

During 2002–2014, the case-fatality rate of SBE in Taiwan was 0.11%, which was lower than that in Taiwan in the early 20th century (3%–7%) [34,37,48–50] or in other countries (0.2%–58%) [5,9–11,17,72–79]. In this study, patients envenomed by N. atra or B. m. multicinctus had a higher case-fatality rate than those envenomed by T. s. stejnegeri or P. mucrosquamatus, as in previous studies [34,37,48–50]. In general, the high number of deaths from SBE was due to delayed or inappropriate antivenom treatment [9,10,17,72,74], and more than half of the deaths were from shock due to bleeding, capillary leak syndrome, or myocardial depression [10,73,74]. In Taiwan, most of the damage caused by SBE was not fatal due to the availability of antivenom and modern medicine; nor did anyone die from complications that could be aggressively managed, such as anaphylactic shock and necrotizing fasciitis.

Limitations

This study has several limitations. First, the drug code of snake antivenom was used to obtain information from the NHIRD, excluding patients who died without seeking medical care, with mild symptoms not requiring antivenom therapy, and envenomed by D. siamensis before 2008. Seven other venomous snakes were found in Taiwan: Ovophis makazayazaya and T. gracilis in the Viperidae family, Sinomicrurus swinhoei and S. sauteri in the Elapidae family, Rhabdophis tigrinus formosanus and Pseudagkistrodon rudis in the Colubridae family, and Myrrophis chinensis in the Homalopsidae family [26,34,80–87]. However, the rarity of bites by these snakes makes it unlikely for them to significantly affect our findings. Second, the individual SBE (i.e., T. s. stejnegeri, P. mucrosquamatus, N. atra, and B. m. multicinctus) could not be determined as the analysis was based on the use of bivalent antivenom against these species. In addition, the decision to use antivenom, or a different type of antivenom, was made by clinicians based on the snake identification procedure, without confirmation of venom in the serum, which could introduce bias. It is also possible that some victims received antivenom for a snakebite, which was later found to be a non-venomous bite or different type of bite. These determinations can only be confirmed through a thorough review of individual case notes. Third, we defined all antivenom records within 90 days as one event and thus might have missed a few cases with repeated SBE. The case number in our study was 5% more than those of the study without definition [35] and 30% less than those of the study defining antivenom records within 14 days as one event [47]. Fourth, the geographic regions of snakebite incidents may be biased. Due to the small land area and convenient transportation in Taiwan, the location of the hospital where patients sought help did not necessarily represent the location where they were envenomed. Furthermore, the patients’ insurance status might not accurately represent their current occupation. Factors affecting the epidemiology of snakebites, such as geographic region and human activities, need to be further studied. Finally, because this study is retrospective, which has certain inherent limitations, the results should be interpreted cautiously.

Conclusions

This study described the epidemiologic and local and overall systemic toxin effects of SBE in Taiwan based on data using antivenom. Risk factors such as male gender, old age, summer season, being in eastern Taiwan, and being an agricultural worker. It is worth noting that elapid bites, which caused necrotizing fasciitis and respiratory failure more than crotaline bites, were more likely to occur in central or southern Taiwan and among laborers. Differences of the epidemiological findings between snake species should be focused on when developing strategies for snakebite prevention.

Supporting information

S1 Table. Characteristics of patients who received different antivenoms in Taiwan (N = 12,542).

https://doi.org/10.1371/journal.pntd.0011376.s001

(DOCX)

S1 Data. Fig 1.

The annual case number and directly standardized incidence of patients receiving antivenoms in Taiwan (N = 12,542). Fig 2. The directly standardized incidence by region in patients receiving antivenoms in Taiwan (N = 12,542).

https://doi.org/10.1371/journal.pntd.0011376.s002

(XLSX)

Acknowledgments

The authors thank Jou-Fang Deng, the founder of Taiwan Poison Control Center, for his inspiration of this study.

Citation: Hsu J-Y, Chiang S-O, Yang C-C, Hsieh T-W, Chung C-J, Mao Y-C (2023) Nationwide and long-term epidemiological research of snakebite envenomation in Taiwan during 2002–2014 based on the use of snake antivenoms: A study utilizing National Health Insurance Database. PLoS Negl Trop Dis 17(6): e0011376. https://doi.org/10.1371/journal.pntd.0011376

References

1. Chippaux J-P. Snakebite envenomation turns again into a neglected tropical disease! The journal of venomous animals and toxins including tropical diseases. 2017;23:38–. pmid:28804495.

2. Kasturiratne A, Wickremasinghe AR, de Silva N, Gunawardena NK, Pathmeswaran A, Premaratna R, et al. The global burden of snakebite: a literature analysis and modelling based on regional estimates of envenoming and deaths. PLoS Med. 2008;5(11):e218. Epub 2008/11/07. pmid:18986210; PubMed Central PMCID: PMC2577696.

3. Chippaux JP. Snake-bites: appraisal of the global situation. Bull World Health Organ. 1998;76(5):515–24. Epub 1998/12/30. pmid:9868843; PubMed Central PMCID: PMC2305789.

4. Böhm M, Cook D, Ma H, Davidson AD, García A, Tapley B, et al. Hot and bothered: Using trait-based approaches to assess climate change vulnerability in reptiles. Biological Conservation. 2016;204:32–41.

5. Suraweera W, Warrell D, Whitaker R, Menon G, Rodrigues R, Fu SH, et al. Trends in snakebite deaths in India from 2000 to 2019 in a nationally representative mortality study. eLife. 2020;9:e54076. pmid:32633232.

6. Chen X, He H, Cai H. Clinical and epidemiologic profile of venomous snakebites in Chongqing Wuling mountainous area, western China. Trans R Soc Trop Med Hyg. 2019;113(10):587–9. Epub 2019/07/12. pmid:31294805.

7. Shuang J, Chen Y, Ran Q, Liao X, Lin W, Wu J, et al. A ten-year retrospective review of 1,107 snakebite patients in Sanya, China. Chin Med J (Engl). 2014;127(11):2189–91. Epub 2014/06/04. pmid:24890175.

8. Chan TY, Critchley JA. An epidemiological study of the snake bites in the New Territories East, Hong Kong. Ann Trop Med Parasitol. 1994;88(2):219–21. Epub 1994/04/01. pmid:8067819.

9. Alirol E, Sharma SK, Bawaskar HS, Kuch U, Chappuis F. Snake bite in South Asia: a review. PLoS Negl Trop Dis. 2010;4(1):e603. Epub 2010/02/04. pmid:20126271; PubMed Central PMCID: PMC2811174.

10. Aye KP, Thanachartwet V, Soe C, Desakorn V, Chamnanchanunt S, Sahassananda D, et al. Predictive Factors for Death After Snake Envenomation in Myanmar. Wilderness Environ Med. 2018;29(2):166–75. Epub 2018/03/25. pmid:29572088.

11. Buranasin P. Snakebites at Maharat Nakhon Ratchasima Regional Hospital. Southeast Asian J Trop Med Public Health. 1993;24(1):186–92. Epub 1993/03/01. pmid:8362295.

12. Vongphoumy I, Chanthilat P, Vilayvong P, Blessmann J. Prospective, consecutive case series of 158 snakebite patients treated at Savannakhet provincial hospital, Lao People’s Democratic Republic with high incidence of anaphylactic shock to horse derived F(ab’)2 antivenom. Toxicon. 2016;117:13–21. Epub 2016/03/21. pmid:26995210.

13. Vongphoumy I, Phongmany P, Sydala S, Prasith N, Reintjes R, Blessmann J. Snakebites in Two Rural Districts in Lao PDR: Community-Based Surveys Disclose High Incidence of an Invisible Public Health Problem. PLoS Negl Trop Dis. 2015;9(6):e0003887. Epub 2015/06/27. pmid:26115035; PubMed Central PMCID: PMC4482615.

14. Jayawardana S, Arambepola C, Chang T, Gnanathasan A. Prevalence, vulnerability and epidemiological characteristics of snakebite in agricultural settings in rural Sri Lanka: A population-based study from South Asia. PloS one. 2020;15(12):e0243991–e. pmid:33370325.

15. Chippaux JP. Estimate of the burden of snakebites in sub-Saharan Africa: a meta-analytic approach. Toxicon. 2011;57(4):586–99. Epub 2011/01/13. pmid:21223975.

16. Mahmood MA, Halliday D, Cumming R, Thwin K-T, Kyaw MMZ, White J, et al. Snakebite incidence in two townships in Mandalay Division, Myanmar. PLoS neglected tropical diseases. 2018;12(7):e0006643–e. pmid:29985919.

17. Rahman R, Faiz MA, Selim S, Rahman B, Basher A, Jones A, et al. Annual incidence of snake bite in rural bangladesh. PLoS neglected tropical diseases. 2010;4(10):e860–e. pmid:21049056.

18. Bolon I, Durso AM, Botero Mesa S, Ray N, Alcoba G, Chappuis F, et al. Identifying the snake: First scoping review on practices of communities and healthcare providers confronted with snakebite across the world. PloS one. 2020;15(3):e0229989–e. pmid:32134964.

19. Inthanomchanh V, Reyer JA, Blessmen J, Phrasisombath K, Yamamoto E, Hamajima N. Assessment of knowledge about snakebite management amongst healthcare providers in the provincial and two district hospitals in Savannakhet Province, Lao PDR. Nagoya journal of medical science. 2017;79(3):299–311. pmid:28878435.

20. Blessmann J, Khonesavanh C, Outhaithit P, Manichanh S, Somphanthabansouk K, Siboualipha P. Venomous snake bites in Lao PDR: a retrospective study of 21 snakebite victims in a provincial hospital. Southeast Asian J Trop Med Public Health. 2010;41(1):195–202. Epub 2010/06/29. pmid:20578499.

21. Mong R, Ng VCH, Tse ML. Safety profile of snake antivenom (use) in Hong Kong—a review of 191 cases from 2008 to 2015. Clin Toxicol (Phila). 2017;55(10):1066–71. Epub 2017/06/29. pmid:28657429.

22. Fung HT, Lam SK, Lam KK, Kam CW, Simpson ID. A survey of snakebite management knowledge amongst select physicians in Hong Kong and the implications for snakebite training. Wilderness Environ Med. 2009;20(4):364–70. Epub 2009/12/25. pmid:20030446.

23. Population Data in Taiwan: Department of Household Registration, Ministry of the Interior; 2022 [cited 2023 May 4]. Available from: https://www.ris.gov.tw/app/portal/346.

24. Employment and Unemployment Statistics Enquiry System: Directorate General of Budget, Accounting, and Statistics; 2022 [cited 2023 May 4]. Available from: https://www.stat.gov.tw/cl.aspx?n=2759.

25. Mao Y-C, Hung D-Z. Epidemiology of Snake Envenomation in Taiwan. In: Gopalakrishnakone P, Faiz SMA, Gnanathasan CA, Habib AG, Fernando R, Yang C-C, editors. Toxinology: Clinical Toxinology. Dordrecht: Springer Netherlands; 2013. p. 1–17.

26. Mao Y-C, Hung D-Z. Management of Snake Envenomation in Taiwan. In: Gopalakrishnakone P, Faiz A, Fernando R, Gnanathasan CA, Habib AG, Yang C-C, editors. Clinical Toxinology in Asia Pacific and Africa. Dordrecht: Springer Netherlands; 2015. p. 23–52.

27. Mao Y-C, Liu P-Y, Chiang L-C, Liao S-C, Su H-Y, Hsieh S-Y, et al. Bungarus multicinctus multicinctus Snakebite in Taiwan. The American journal of tropical medicine and hygiene. 2017;96(6):1497–504. pmid:28719273.

28. Cheng CL, Mao YC, Liu PY, Chiang LC, Liao SC, Yang CC. Deinagkistrodon acutus envenomation: a report of three cases. J Venom Anim Toxins Incl Trop Dis. 2017;23:20. Epub 2017/03/28. pmid:28344596; PubMed Central PMCID: PMC5364662.

29. Su H-Y, Huang S-W, Mao Y-C, Liu M-W, Lee K-H, Lai P-F, et al. Clinical and laboratory features distinguishing between Deinagkistrodon acutus and Daboia siamensis envenomation. Journal of Venomous Animals and Toxins including Tropical Diseases. 2018;24(1):43. pmid:30607144

30. Mao YC, Liu PY, Chiang LC, Lai CS, Lai KL, Ho CH, et al. Naja atra snakebite in Taiwan. Clin Toxicol (Phila). 2018;56(4):273–80. Epub 2017/08/24. pmid:28830248.

31. Lai CS, Liu PY, Lee CH, Ho CH, Chen WL, Lai KL, et al. The development of surgical risk score and evaluation of necrotizing soft tissue infection in 161 Naja atra envenomed patients. PLoS Negl Trop Dis. 2022;16(2):e0010066. Epub 2022/02/11. pmid:35143522; PubMed Central PMCID: PMC8830662.

32. Chiang LC, Tsai WJ, Liu PY, Ho CH, Su HY, Lai CS, et al. Envenomation by Trimeresurus stejnegeri stejnegeri: clinical manifestations, treatment and associated factors for wound necrosis. J Venom Anim Toxins Incl Trop Dis. 2020;26:e20200043. Epub 2020/09/29. pmid:32983233; PubMed Central PMCID: PMC7500479.

33. Mao YC, Liu PY, Chiang LC, Lee CH, Lai CS, Lai KL, et al. Clinical manifestations and treatments of Protobothrops mucrosquamatus bite and associated factors for wound necrosis and subsequent debridement and finger or toe amputation surgery. Clin Toxicol (Phila). 2020:1–10. Epub 2020/05/14. pmid:32400229.

34. To S. Statistical studies of poisonous snakebites in Formosan. Journal of the Formosan Medical Association. 1941;40:1477–502; 795–824.

35. Chen C-K, Lin C-C, Shih F-Y, Chaou C-H, Lin JC-C, Lai T-I, et al. Population-based study of venomous snakebite in Taiwan. Journal of Acute Medicine. 2015;5(2):38–42.

36. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373–83. Epub 1987/01/01. pmid:3558716.

37. Sawai Y, Tseng CS. Snakebites on Taiwan. The Snake. 1969;1:9–18.

38. Lin H-C, Li S-H, Fong J, Lin S-M. Ventral coloration differentiation and mitochondrial sequences of the Chinese Cobra (Naja atra) in Taiwan. Conservation Genetics. 2008;9(5):1089–97.

39. Huang HW, Liu BS, Chien KY, Chiang LC, Huang SY, Sung WC, et al. Cobra venom proteome and glycome determined from individual snakes of Naja atra reveal medically important dynamic range and systematic geographic variation. J Proteomics. 2015;128:92–104. Epub 2015/07/22. pmid:26196238.

40. Republic of China. Statistical Information Network of the Republic of China 2021 [cited 2022 March 24]. Available from: https://www.stat.gov.tw/.

41. Regional Office for South-East Asia WHO. Guidelines for the management of snakebites. 2nd ed ed. New Delhi: WHO Regional Office for South-East Asia; 2016 2016–08.

42. Soares Coriolano Coutinho JV, Fraga Guimarães T, Borges Valente B, Gomes Martins de Moura Tomich L. Epidemiology of secondary infection after snakebites in center-west Brazil. PLoS Negl Trop Dis. 2023;17(3):e0011167. Epub 2023/03/07. pmid:36877732; PubMed Central PMCID: PMC10019779.

43. Alcoba G, Sharma SK, Bolon I, Ochoa C, Babo Martins S, Subedi M, et al. Snakebite epidemiology in humans and domestic animals across the Terai region in Nepal: a multicluster random survey. Lancet Glob Health. 2022;10(3):e398–e408. Epub 2022/02/19. pmid:35180421.

44. Thang VV, Bao TQQ, Tuyen HD, Krumkamp R, Hai LH, Dang NH, et al. Incidence of snakebites in Can Tho Municipality, Mekong Delta, South Vietnam-Evaluation of the responsible snake species and treatment of snakebite envenoming. PLoS Negl Trop Dis. 2020;14(6):e0008430. Epub 2020/06/20. pmid:32555599; PubMed Central PMCID: PMC7323996.

45. Chaves LF, Chuang TW, Sasa M, Gutiérrez JM. Snakebites are associated with poverty, weather fluctuations, and El Niño. Sci Adv. 2015;1(8):e1500249. Epub 2015/11/26. pmid:26601254; PubMed Central PMCID: PMC4643785.

46. Liu C-H, Jiang DD-S, Lian W-C, Liu D-P. Epidemiological Analysis of Antivenin Usage in Taiwan Area, 2002–2005. 2009;25(7):510–25.

47. Liu C-H, Hsieh W-C. Analysis of Antivenin Usage in Taiwan from National Health Insurance Research Database, 2008–2012. 2017;33(9):57–.

48. Sawai Y, Tseng CS, Kuo TP, Wu CS. Snakebites in Kaohsiung Prefecture, Taiwan. The Snake. 1970;2:13–7.

49. Sawai Y, et al. An epidemiological study of snakebites in the Southeast Asia. Jikken Igaku Zasshi = Japanese Journal of Experimental Medicine. 1972;42(3):283–307. pmid:4538156

50. Kuo TP, Wu CS. Clinico-pathological studies on snakebites in Taiwan. Journal of the Formosan Medical Association. 1972;71(7):447–66. pmid:4507795

51. Liao WB, Lee CW, Tsai YS, Liu BM, Chung KJ. Influential factors affecting prognosis of snakebite patients management: Kaohsiung Chang Gung Memorial Hospital experience. Chang Gung Med J. 2000;23(10):577–83. Epub 2000/12/29. pmid:11126148.

52. Hung DZ. Taiwan’s venomous snakebite: epidemiological, evolution and geographic differences. Trans R Soc Trop Med Hyg. 2004;98(2):96–101. Epub 2004/02/18. pmid:14964809.

53. Laohawiriyakamol S, Sangkhathat S, Chiengkriwate P, Patrapinyokul S. Surgery in management of snake envenomation in children. World J Pediatr. 2011;7(4):361–4. Epub 2011/08/31. pmid:21877258.

54. Ho CH, Chiang LC, Mao YC, Lan KC, Tsai SH, Shih YJ, et al. Analysis of the Necrosis-Inducing Components of the Venom of Naja atra and Assessment of the Neutralization Ability of Freeze-Dried Antivenom. Toxins (Basel). 2021;13(9). Epub 2021/09/27. pmid:34564623; PubMed Central PMCID: PMC8473173.

55. Brenes-Chacon H, Gutierrez JM, Camacho-Badilla K, Soriano-Fallas A, Ulloa-Gutierrez R, Valverde K, et al. Long-term sequelae secondary to snakebite envenoming: a single centre retrospective study in a Costa Rican paediatric hospital. BMJ Paediatr Open. 2020;4(1):e000735. Epub 2020/10/01. pmid:32995568; PubMed Central PMCID: PMC7497519.

56. Chattopadhyay A, Patra RD, Shenoy V, Kumar V, Nagendhar Y. Surgical implications of snakebites. Indian J Pediatr. 2004;71(5):397–9. Epub 2004/05/28. pmid:15163866.

57. Rha JH, Kwon SM, Oh JR, Han BK, Lee KH, Kim JH. Snakebite in Korea: A Guideline to Primary Surgical Management. Yonsei Med J. 2015;56(5):1443–8. Epub 2015/08/11. pmid:26256993; PubMed Central PMCID: PMC4541680.

58. Kim YH, Choi JH, Kim J, Chung YK. Fasciotomy in compartment syndrome from snakebite. Arch Plast Surg. 2019;46(1):69–74. Epub 2019/01/29. pmid:30685944; PubMed Central PMCID: PMC6369054.

59. Wagener M, Naidoo M, Aldous C. Wound infection secondary to snakebite. S Afr Med J. 2017;107(4):315–9. Epub 2017/04/12. pmid:28395682.

60. Halilu S, Iliyasu G, Hamza M, Chippaux JP, Kuznik A, Habib AG. Snakebite burden in Sub-Saharan Africa: estimates from 41 countries. Toxicon. 2019;159:1–4. Epub 2018/12/31. pmid:30594637.

61. Paolino G, Di Nicola MR, Pontara A, Didona D, Moliterni E, Mercuri SR, et al. Vipera snakebite in Europe: a systematic review of a neglected disease. J Eur Acad Dermatol Venereol. 2020;34(10):2247–60. Epub 2020/06/13. pmid:32530549.

62. Bucaretchi F, Herrera SR, Hyslop S, Baracat EC, Vieira RJ. Snakebites by Bothrops spp in children in Campinas, São Paulo, Brazil. Rev Inst Med Trop Sao Paulo. 2001;43(6):329–33. Epub 2002/01/10. pmid:11781603.

63. Irion VH, Barnes J, Montgomery BE, Suva LJ, Montgomery CO. Presentation and Management of Venomous Snakebites: Should All Patients Be Transferred to a Tertiary Referral Hospital? J Surg Orthop Adv. 2016;25(2):69–73. Epub 2016/08/16. pmid:27518288.

64. Tanen D, Ruha A, Graeme K, Curry S. Epidemiology and hospital course of rattlesnake envenomations cared for at a tertiary referral center in Central Arizona. Acad Emerg Med. 2001;8(2):177–82. Epub 2001/02/07. pmid:11157295.

65. Pietrangiolillo Z, Frassoldati R, Leonelli V, Freschi R, Russomando A, Lucaccioni L, et al. Compartment syndrome after viper-bite in toddler: case report and review of literature. Acta Biomed. 2012;83(1):44–50. Epub 2012/09/18. pmid:22978057.

66. Ho CH, Ismail AK, Liu SH, Tzeng YS, Li LY, Pai FC, et al. The role of a point-of-care ultrasound protocol in facilitating clinical decisions for snakebite envenomation in Taiwan: a pilot study. Clin Toxicol (Phila). 2021:1–14. Epub 2021/02/20. pmid:33605805.

67. Türkmen A, Temel M. Algorithmic approach to the prevention of unnecessary fasciotomy in extremity snake bite. Injury. 2016;47(12):2822–7. Epub 2016/11/05. pmid:27810154.

68. Cumpston KL. Is there a role for fasciotomy in Crotalinae envenomations in North America? Clin Toxicol (Phila). 2011;49(5):351–65. Epub 2011/07/12. pmid:21740134.

69. Hall EL. Role of surgical intervention in the management of crotaline snake envenomation. Annals of Emergency Medicine. 2001;37(2):175–80. pmid:11174236

70. Toschlog EA, Bauer CR, Hall EL, Dart RC, Khatri V, Lavonas EJ. Surgical considerations in the management of pit viper snake envenomation. J Am Coll Surg. 2013;217(4):726–35. Epub 2013/07/31. pmid:23891068.

71. Mao YC, Liu PY, Hung DZ, Lai WC, Huang ST, Hung YM, et al. Bacteriology of Naja atra Snakebite Wound and Its Implications for Antibiotic Therapy. Am J Trop Med Hyg. 2016;94(5):1129–35. Epub 2016/03/16. pmid:26976881; PubMed Central PMCID: PMC4856614.

72. Hannan Wan Ibadullah WA, Azmi MF, Abas MI, Syed Abdul Rahim SS, Jeffree MS, Azhar ZI, et al. Determinants of snakebite mortality in Asia: A systematic review. Ann Med Surg (Lond). 2021;62:16–20. Epub 2021/01/26. pmid:33489110; PubMed Central PMCID: PMC7809159.

73. Gopalakrishnan M, Vinod KV, Dutta TK, Shaha KK, Sridhar MG, Saurabh S. Exploring circulatory shock and mortality in viper envenomation: a prospective observational study from India. Qjm. 2018;111(11):799–806. Epub 2018/08/15. pmid:30107433.

74. Looareesuwan S, Viravan C, Warrell DA. Factors contributing to fatal snake bite in the rural tropics: analysis of 46 cases in Thailand. Trans R Soc Trop Med Hyg. 1988;82(6):930–4. Epub 1988/01/01. pmid:3257001.

75. Habib AG, Warrell DA. Antivenom therapy of carpet viper (Echis ocellatus) envenoming: effectiveness and strategies for delivery in West Africa. Toxicon. 2013;69:82–9. Epub 2013/01/24. pmid:23339853.

76. Seneviratne SL, de Silva CE, Fonseka MM, Pathmeswaran A, Gunatilake SB, de Silva HJ. Envenoming due to snake bite during pregnancy. Trans R Soc Trop Med Hyg. 2002;96(3):272–4. Epub 2002/08/15. pmid:12174776.

77. Dunnihoo DR, Rush BM, Wise RB, Brooks GG, Otterson WN. Snake bite poisoning in pregnancy. A review of the literature. J Reprod Med. 1992;37(7):653–8. Epub 1992/07/01. pmid:1522574.

78. Langley RL. A review of venomous animal bites and stings in pregnant patients. Wilderness Environ Med. 2004;15(3):207–15. Epub 2004/10/12. pmid:15473462.

79. Langley RL. Snakebite during pregnancy: a literature review. Wilderness Environ Med. 2010;21(1):54–60. Epub 2010/07/02. pmid:20591355.

80. Chen YC, Yen DH, Chen YW, Huang MS, Huang CI, Chen MH. Toxin ophthalmia caused by nuchal gland secretion of the Taiwan tiger keelback (Rhabdophis tigrinus formosanus). J Formos Med Assoc. 2014;113(10):750–3. Epub 2014/09/23. pmid:25240304.

81. Chen MH, Wu SH, Chen YW, Lee YL, Chen YC. A case report of Ovophis makazayazaya envenoming. Clin Toxicol (Phila). 2021;59(7):679–80. Epub 2020/11/03. pmid:33135496.

82. Tsai TS, Chan YY, Huang SM, Chuang PC. Case Report: Symptoms and Prognosis of Trimeresurus gracilis Envenomation. Am J Trop Med Hyg. 2022. Epub 2022/02/08. pmid:35130484.

83. Uetz P, Freed, P, Aguilar, R., Reyes, F. & Hošek, J. The Reptile Database 2022 [cited 2022 Jan 12]. Available from: http://www.reptile-database.org.

84. Weinstein SA WD, Keyler DE. “Venomous” Bites from Non-Venomous Snakes: A Critical Analysis of Risk and Management of “Colubrid” Snake Bites. London: Elsevier; 2011.

85. Smart U, Ingrasci MJ, Sarker GC, Lalremsanga H, Murphy RW, Ota H, et al. A comprehensive appraisal of evolutionary diversity in venomous Asian coralsnakes of the genus Sinomicrurus (Serpentes: Elapidae) using Bayesian coalescent inference and supervised machine learning. Journal of Zoological Systematics and Evolutionary Research. 2021;59(8):2212–77. https://doi.org/10.1111/jzs.12547.

86. Zhang Z, Zhang X, Hu T, Zhou W, Cui Q, Tian J, et al. Discovery of toxin-encoding genes from the false viper Macropisthodon rudis, a rear-fanged snake, by transcriptome analysis of venom gland. Toxicon. 2015;106:72–8. pmid:26403866

87. Weinstein SA. Snake venoms: A brief treatise on etymology, origins of terminology, and definitions. Toxicon. 2015;103:188–95. Epub 2015/07/15. pmid:26166305.