Clinical Laboratory Investigations and Antivenom Administration after Malayan Pit Viper (Calloselasma rhodostoma) Envenoming: A Retrospective Study from Southernmost Thailand

ABSTRACT.

The Malayan pit viper (MPV: Calloselasma rhodostoma) is a medically important venomous snake causing numerous envenomations in Thailand. Administration of specific snake antivenom is the only effective treatment for MPV-envenomed patients. However, inappropriate administration or misuse of snake antivenom is problematic in some remote areas of tropical countries where the snakebite envenoming rate is notable. Currently, the indications for administration of MPV antivenom are focused mainly on hematological factors. These include 1) venous clotting time > 20 min, 2) unclotted 20-minute whole-blood clotting time, 3) international normalized ratio > 1.2, 4) platelet count < 50 × 10³/μL, 5) systemic bleeding, and 6) impending compartment syndrome. We aimed to determine the association between laboratory data and antivenom administration in MPV-envenomed patients. A retrospective study of data from 2016 to 2021 in Narathiwat Province, the southernmost province in Thailand, was conducted. A total of 838 MPV-bitten patients were included in this study. Local effects and systemic effects were observed in 58.8% and 27.7% of patients, respectively. Coagulopathies, which range from abnormal blood clotting to systemic bleeding, represented the majority of systemic effects. Acute kidney injury developed in 2.5% of patients. In this study, 57.3% of patients were considered appropriate antivenom recipients. Interestingly, the present study revealed that local bleeding and mild to moderate thrombocytopenia became the independent factors for inappropriate use of MPV antivenom. Reeducation and supervision regarding the rational use of snake antivenom are needed to minimize the misuse of antivenom.

INTRODUCTION

Worldwide, snakebite is a critical health issue in tropical countries, especially Sub-Saharan Africa, South America, South Asia, and Southeast Asia. Based on a report from the WHO, an estimated 5.4 million people are bitten by snakes, with 81,000 to 138,000 deaths each year.¹ In Thailand, although it is recognized that the epidemiology of snakebite has not been sufficiently investigated, there has been a significant decreasing trend in the prevalence of snakebite cases over the past two decades, which might be resulting from environmental changes and people’s lifestyles. However, the prevalence of snakebite cases appears to be clinically significant in some remotely rural areas.² In 2015, it was estimated that the average number of snakebite cases in Thailand was 457 (0.7/100,000).² Interestingly, Malayan pit vipers (MPVs; Calloselasma rhodostoma), which is a hematotoxic venomous snake causing high levels of morbidity and disability, was the major cause of snakebites in Thailand between 2011 and 2016 (38%),² especially in the southern part of the country.^3–6

Clinical manifestations observed after MPV envenoming can be the presence of two needle-stick dot fang marks, local effects, and systemic effects. Local effects range from local inflammatory effects to skin blistering, ecchymoses, hemorrhagic blebs, and necrosis in more severe cases.^3,7 More importantly, venom-induced consumption coagulopathy is a common systemic effect induced by thrombin-like enzymes that change fibrinogen to fibrin and ultimately stimulate fibrinolysis,^8–10 resulting in abnormal blood clotting and systemic bleeding.⁵ These clinical outcomes are induced by the toxic components in snake venom.

Snake venom is a complex mixture of proteins and toxic enzymes used in defensive mechanisms and the digestion of prey. In MPV venom, snake venom phospholipase A2 (svPLA2) and snake venom metalloproteinase (svMP) are two major enzymatic components causing local and systemic effects.¹¹ Moreover, l-amino acid oxidase, α-fibrinogenase, disintegrin, antiplatelet protease, platelet aggregation inducer, and hemorrhagin have also been detected and characterized in MPV venom.^4,5,12 Indeed, each venom contains more than a hundred different proteins.² Hence, combined effects between proteins are not unexpected. For example, it has been revealed that there are several neurotoxins found in hematotoxic snake venom.¹³ Moreover, hepatotoxicity and nephrotoxicity have also been demonstrated in an experimental animal model after MPV envenoming.¹¹ This suggests that in addition to hematology test results, other biochemical laboratory test results should be reviewed for MPV-bitten patients.

Early antivenom administration is the only effective treatment and is essential for the management of systemic MPV envenoming. Currently, the indications for the administration of MPV antivenom are focused mainly on hematological factors. These include 1) venous clotting time (VCT) > 20 min, 2) unclotted 20-minute whole-blood clotting time (20WBCT), 3) international normalized ratio (INR) > 1.2, 4) platelet count < 50 × 10³/μL, 5) systemic bleeding, and 6) impending compartment syndrome.⁷ However, inappropriate use of snake antivenom has been reported in some parts of the world, such as routinely giving antivenom to any patient claiming to have been bitten by a snake without local or systemic effects (“dry bite”) or administering antivenom to the venom of a snake that is different from that which has bitten the patient. These errors lead to ineffective treatment in envenomed victims.¹⁴

Although the number of deaths and disability after MPV envenoming are continuously decreasing in Thailand, the clinical management and antivenom administration criteria after MPV envenoming are not well understood or practiced in some remote areas. In this study, we therefore performed a retrospective analysis of data collected from MPV-envenomed patients in Narathiwat, the southernmost province of Thailand. We aimed to determine the association between laboratory data and antivenom administration in MPV-envenomed patients based on the characteristics of MPV-bitten patients, including demographic data, clinical effects, laboratory investigation results, and treatment, by comparing patients who received monovalent MPV antivenom therapy with those who received nonantivenom therapy. This work will strongly advocate for further clinical approaches to be undertaken to validate antivenom treatment and efficacy across the continent.

MATERIALS AND METHODS

Study designs and subjects.

This was a retrospective descriptive data collection study of MPV-bitten patients from November 1, 2016, to October 31, 2021. Patients visiting Ra-Ngae Hospital (120 beds), Rueso Hospital (90 beds), Yi-ngo Hospital 80th Anniversary Commemoration Hospital (60 beds), Takbai Hospital (120 beds), and Naradhiwas Rajanagarindra Hospital (400 beds) were included in this study. These hospitals are located in Narathiwat Province, the southernmost province in Thailand, where Naradhiwas Rajanagarindra Hospital is the main hospital. The inclusion criteria were patients with MPV bites, which were defined through one of the following methods: 1) the snake carcass was brought to the hospital, and 2) the patient could clearly describe or identify the type of snake. Patients with snakebites from other species or unknown species were excluded. For data analysis, patients were categorized into the overall group, nonantivenom therapy group, and antivenom group.

Data collection.

At each hospital, a well-trained registered nurse abstracted medical records using a standardized case report form (CRF). Snakebites were identified using the International Classification of Diseases, Tenth Revision (ICD-10) codes T63.0, as documented in the medical records. Of these, only identified MPV-envenomed patients were specifically included, as described in the inclusion criteria. The CRF was then converted to an electronic dataset that was recorded in a spreadsheet. The collected data comprised the following: demographics; snake type; characteristics of MPV bites and monovalent MPV antivenom administration; bitten areas; fang marks; dry bites; local effects, including pain, swelling, local bleeding, ecchymosis, bleb, necrosis, and impending compartment syndrome; systemic effects, including prolonged VCT, unclotted 20WBCT, INR > 1.2, thrombocytopenia, and systemic bleeding; and other laboratory investigation parameters, including electrolytes, creatine phosphokinase (CPK), creatinine, and white blood cell count. Dry bite was characterized as a snakebite into which snake venom was not injected,¹⁵ and the criteria for dry bite diagnosis included 1) the presence of fang marks, 2) no local or systemic signs and symptoms, and 3) snake identification.^16–18 Local and systemic effects were recorded at the start of admission and during the hospital stay. All laboratory investigations were performed prior to any treatment. Mild, moderate, and severe thrombocytopenia were defined as platelet counts of 101 × 10³/µL to 140 × 10³/µL, 51 × 10³/µL to 100 × 10³/µL, and 21 × 10³/µL to 50 × 10³/µL, respectively.¹⁹ Leukocytosis was defined as a white blood cell count > 11 × 10³/µL.²⁰ Rhabdomyolysis was defined as a serum CPK level > 1 × 10³ units/L.²¹ According to the Kidney Disease: Improving Global Outcomes (KDIGO) guidelines, acute kidney injury (AKI) was defined as an increase in serum creatinine by ≥ 0.3 mg/dL within 48 h.²²

STATISTICAL ANALYSES

The data were analyzed using Stata Statistical Software release 17 (StataCorp LLC, College Station, TX). Demographic characteristics were analyzed with descriptive statistics. The outcomes are presented as numbers and percentages for categorical data and as the medians, interquartile ranges (IQR), and ranges (minimum to maximum) for continuous data. The proportions were calculated proportionally to the total number for each data group obtained. The Mann-Whitney U test was used to test the difference in the medians, whereas the chi-square test or Fisher’s exact test was used to compare the distribution proportions of categorical variables between patients who received antivenom therapy and patients who received nonantivenom therapy. Univariate and multivariate logistic regression analyses were performed to determine the associated factors for inappropriate use of MPV antivenom. Pain, swelling, the presence of fang marks, local bleeding, bleb, the levels of sodium, potassium, chloride, and bicarbonate, white blood cell count, and platelet count were included in the univariate logistic regression analyses. The presence of fang marks, local bleeding, white blood cell count, and platelet count were adjusted in the final model. The magnitudes of the associations obtained from univariate and multivariate analyses were represented as crude odds ratios and adjusted odds ratios, respectively, with the corresponding 95% CIs. The level of statistical significance was set at P <  0.05.

Ethical considerations.

This study was reviewed and approved by the Institutional Review Board Royal Thai Army Medical Department, project code S097h/64_Exp. This study was conducted in accordance with the principles of the Declaration of Helsinki. The requirement for informed consent was waived by the Institutional Review Board Royal Thai Army Medical Department because this is a retrospective study of deidentified data retrieved from medical records.

RESULTS

Demographic characteristics.

A total of 838 Thai patients with MPV bites were included in this study. The demographic characteristics of the patients are presented in Table 1. Males accounted for 69.9% of all participants. Overall, the median age of patients was 43 years (IQR, 28 to 57 years; range, 2 to 104 years), with a higher median age in the nonantivenom therapy group than in the antivenom therapy group (P = 0.009). In all patients, the median time from the bite to arrival at the emergency department was 30 min (IQR, 20 min to 2 h; range, 0 to 144 h). Among 838 MPV-bitten patients, 377 patients (45.0%) did not receive antivenom, whereas 461 patients (55.0%) received monovalent MPV antivenom. The median time to administer antivenom was 5 h 40 min (IQR, 2 to 14 h; range, 0 to 72 h). Most patients were bitten from April to June. Patients were most commonly bitten during the time period of 6 to 9 pm (Figure 1). On average, patients who were in the antivenom therapy group had a longer hospital stay (median, 2 days; IQR, 1 to 3 days; range, 1 to 33 days) than those in the nonantivenom therapy group (median, 1 day; IQR, 1 to 2 days; range, 1 to 7 days) (P < 0.001) (Figure 2).

Table 1.

Demographic characteristics of patients and local effects after Malayan pit viper bites

Characteristics and local effects	Nonantivenom therapy		Antivenom therapy		P-value	Overall
	n = 377		n = 461			N = 838
	n (% of 838 patients)	Proportions	n (% of 838 patients)	Proportions		n (% of 838 patients)	Proportions
Sex					0.053
Males	250 (29.8)	0.67	336 (40.1)	0.73		586 (69.9)	0.71
Females	122 (14.6)	0.33	122 (14.6)	0.27		244 (29.1)	0.29
Age (years)					0.009*
Median	46		41			43	–
IQR	30–59		27–55			28–57	–
Range	2–104		2–93			2–104	–
Duration from the bite to arrival at the emergency department					0.436
Median	30 min		30 min			30 min	–
IQR	20 min–1 h 38 min		20 min–2 h			20 min–2 h	–
Range	0–72 h		0–144 h			0–144 h	–
Monovalent MPV antivenom administration					NA
No	377 (45.0)	1.00	0	0.00		377 (45.0)	0.45
Yes	0	0.00	461 (55.0)	1.00		461 (55.0)	0.55
Time to administer antivenom
Median	–	–	5 h 40 min			–	–
IQR	–	–	2–14 h			–	–
Range	–	–	0–72 h			–	–
MPV bites by month					0.001*
January to March	86 (10.3)	0.25	112 (13.6)	0.25		198 (23.6)	0.25
April to June	92 (11.0)	0.27	182 (21.7)	0.40		274 (32.7)	0.34
July to September	68 (8.1)	0.20	79 (9.4)	0.18		147 (17.5)	0.19
October to December	94 (11.2)	0.28	78 (9.3)	0.17		172 (20.5)	0.22
Bitten area					0.179
Arm	3 (0.4)	0.01	1 (0.1)	0.00		4 (0.5)	0.01
Hand	121(14.4)	0.33	122 (14.6)	0.27		243 (29.0)	0.29
Leg	33 (3.9)	0.09	56 (6.7)	0.12		89 (10.6)	0.11
Foot	207 (24.7)	0.56	266 (31.7)	0.59		473 (56.4)	0.57
Fang marks					0.236
No	51 (6.1)	0.13	50 (6.0)	0.11		101 (12.0)	0.12
Yes	326 (38.9)	0.87	411 (49.0)	0.89		737 (88.0)	0.88
Dry bite					0.006*
No	172 (20.1)	0.86	372 (44.4)	0.93		544 (64.9)	0.91
Yes	28 (3.3)	0.14	28 (3.3)	0.07		56 (6.6)	0.09
Local effects
Pain					0.128
No	199 (23.7)	0.53	219 (26.0)	0.48		418 (49.9)	0.50
Yes	178 (21.2)	0.47	242 (28.9)	0.52		420 (50.1)	0.50
Swelling					0.083
No	201 (24.0)	0.53	218 (26.0)	0.47		419 (50.0)	0.50
Yes	176 (21.0)	0.47	243 (29.0)	0.53		419 (50.0)	0.50
Local bleeding					0.109
No	286 (34.1)	0.76	327 (39.0)	0.71		613 (73.2)	0.73
Yes	91 (10.9)	0.24	134 (16.0)	0.29		225 (26.8)	0.27
Ecchymosis					0.172
No	364 (43.4)	0.97	436 (52.0)	0.95		800 (95.5)	0.96
Yes	13 (1.6)	0.03	25 (3.0)	0.05		38 (4.5)	0.04
Bleb					0.001*
No	377 (45.0)	1.00	447 (53.3)	0.97		824 (98.3)	0.98
Yes	0	0.00	14 (1.7)	0.03		14 (1.7)	0.02
Necrosis					NA
No	377 (45.0)	1.00	460 (54.9)	1.00		837 (99.9)	1.00
Yes	0	0.00	1 (0.1)	0.00		1 (0.1)	0.00
Impending compartment syndrome					< 0.001*
No	377 (45.0)	1.00	414 (49.4)	0.92		791 (94.4)	0.96
Yes	0	0.00	36 (4.3)	0.08		36 (4.3)	0.04
Overall local effects					0.940
No	146 (17.4)	0.40	189 (22.6)	0.41		335 (40.0)	0.40
Yes	216 (25.8)	0.60	277 (33.0)	0.59		493 (58.8)	0.60

h = hours; IQR = interquartile range; min = minutes; MPV = Malayan pit viper; NA = not available.

^*P < 0.05 indicates that there is a statistically significant difference in proportion of patients between the nonantivenom therapy and antivenom therapy groups.

Figure 1.

Malayan pit viper bites by time of the day.

Figure 2.

Length of hospital stay of patients with Malayan pit viper bites stratified by status of receiving antivenom therapy.

Local and systemic effects in patients with MPV bites.

The most common area bitten was the foot (473 patients, 56.4% of 838 patients) (Table 1). Fang marks were identified in 737 patients (88.0% of 838 patients). Dry bites were observed in 56 patients (6.6% of 838 patients), with a significantly greater proportion in the nonantivenom therapy group than in the antivenom therapy group (P = 0.006).

Four hundred ninety-three patients (58.8% of 838 patients) experienced local effects (Table 1). There was no statistically significant difference in proportion of patients with local effects between the nonantivenom therapy and antivenom therapy groups (P = 0.940). Pain (420 patients, 50.1% of 838 patients) and swelling (419 patients, 50.0% of 838 patients) were the most common local effects, followed by local bleeding (225 patients, 26.8% of 838 patients). Ecchymosis was reported in 38 patients (4.5% of 838 patients). Interestingly, bleb (14 patients, 1.7% of 838 patients) and necrosis (1 patient, 0.1% of 838 patients) were observed only in patients in the antivenom therapy group. Local complications included impending compartment syndrome, which was also found only in the antivenom therapy group (36 patients, 4.3% of 838 patients).

Table 2 presents the systemic effects that were seen in 232 MPV-bitten patients (27.7% of 838 patients). Indeed, of the 232 patients who experienced coagulopathies, 220 patients (26.3% of 838 patients) were in the antivenom therapy group. As a consequence, a significantly greater proportion of patients with systemic effects was observed in the antivenom therapy group than in the nonantivenom therapy group (P < 0.001). Platelet count testing was performed in 608 patients (72.6% of 838 patients). Seventy-nine patients (9.4% of 838 patients), 44 patients (5.3% of 838 patients), and 21 patients (3.5% of 838 patients) had low platelet counts, as categorized by using the upper limit cutoffs of mild (140 × 10³/µL), moderate (100 × 10³/µL), and severe (50 × 10³/µL) thrombocytopenia, respectively.¹⁹

Table 2.

Systemic effects of patients with Malayan pit viper bites

Systemic effects	Nonantivenom therapy		Antivenom therapy		P-value	Overall
	n = 377		n = 461			N = 838
	n (% of 838 patients)	Proportions	n (% of 838 patients)	Proportions		n (% of 838 patients)	Proportions
VCT (min)					<0.001*
≤ 20	124 (14.8)	1.00	131 (15.6)	0.59		255 (30.4)	0.74
> 20	0	0.00	90 (10.7)	0.41		90 (10.7)	0.26
Unclotted 20WBCT					<0.001*
No	69 (8.2)	0.99	86 (10.3)	0.48		155 (18.5)	0.62
Yes	1 (0.1)	0.01	94 (11.2)	0.52		95 (11.3)	0.38
INR					0.008*
≤ 1.2	91 (10.9)	0.91	69 (8.2)	0.61		160 (19.1)	0.75
> 1.2	9 (1.1)	0.09	44 (5.3)	0.39		53 (6.3)	0.25
Platelet count (platelets/µL)					0.019*
< 50 × 103	2 (0.2)	0.01	19 (2.3)	0.05		21 (3.5)	0.04
≥ 50 × 103	200 (23.9)	0.99	387 (46.2)	0.95		587 (70.0)	0.96
Platelet count (platelets/µL)					<0.001*
< 100 × 103	4 (0.5)	0.02	40 (4.8)	0.10		44 (5.3)	0.07
≥ 100 × 103	198 (23.6)	0.98	366 (43.7)	0.90		564 (67.3)	0.93
Platelet count (platelets/µL)					<0.001*
< 140 × 103	7 (0.8)	0.03	72 (8.6)	0.18		79 (9.4)	0.13
≥ 140 × 103	195 (23.3)	0.97	334 (39.9)	0.82		529 (63.1)	0.87
Systemic bleeding					0.016*
No	376 (44.9)	1.00	450 (53.7)	0.98		826 (98.6)	0.99
Yes	1 (0.1)	0.00	10 (1.2)	0.02		11 (1.3)	0.01
Overall systemic effects					<0.001*
No	346 (41.3)	0.97	180 (21.5)	0.45		526 (62.8)	0.69
Yes	12 (1.4)	0.03	220 (26.3)	0.55		232 (27.7)	0.31

INR = international normalized ratio; VCT = venous clotting time; 20WBCT = unclotted 20-minute whole-blood clotting time. *P < 0.05 indicates that there is a statistically significant difference in proportion of patients between the nonantivenom therapy and antivenom therapy groups.

Venous clotting time testing was performed in 345 patients (41.2% of 838 patients), followed by 20WBCT testing in 250 patients (29.8% of 838 patients) and INR testing in 213 patients (25.4% of 838 patients). Ninety patients (10.7% of 838 patients) had a prolonged VCT (> 20 min), whereas 95 patients (11.3% of 838 patients) had an unclotted 20WBCT. Fifty-three patients (6.3% of 838 patients) had a prolonged INR (> 1.2). Eleven patients (1.3% of 838 patients) experienced systemic bleeding, including gingival bleeding (eight patients), hematemesis (two patients), and hematohidrosis (one patient). As expected, neither ptosis nor muscle weakness was found in any patient in this study. Significantly greater proportions of patients with thrombocytopenia, prolonged VCT, unclotted 20WBCT, prolonged INR, and systemic bleeding were observed in the antivenom therapy group than in the nonantivenom therapy group.

Laboratory investigation results of patients with MPV bites.

Data on levels of electrolytes, CPK, and creatinine and white blood cell counts were collected and are presented in Table 3. Sodium and chloride levels were assessed in 414 patients (49.4% of 838 patients). Potassium and bicarbonate levels were assessed in 420 patients (50.1% of 838 patients) and 381 patients (45.5% of 838 patients), respectively. The majority of patients had normal levels of electrolytes. One hundred two patients (12.2% of 838 patients) and 143 patients (17.1% of 838 patients) had hypokalemia and hyperchloremia, respectively, with no statistically significant difference in the proportions of potassium and chloride levels between the nonantivenom therapy and antivenom therapy groups. One hundred thirty-six patients (16.2% of 838 patients) had hyperbicarbonatemia, with a significantly greater proportion in the nonantivenom therapy group than in the antivenom therapy group (P = 0.001).

Table 3.

Additional laboratory investigation results of patients with Malayan pit viper bites

Laboratory investigations	Nonantivenom therapy		Antivenom therapy		P-value	Overall
	N = 377		N = 461			N = 838
	n (% of 838 patients)	Proportions	n (% of 838 patients)	Proportions		n (% of 838 patients)	Proportions
Sodium (mEq/L)					0.792
< 135	3 (0.4)	0.02	8 (1.0)	0.03		11 (1.3)	0.03
135–145	134 (16.0)	0.96	262 (31.3)	0.95		396 (47.3)	0.95
> 145	3 (0.4)	0.02	4 (0.5)	0.02		7 (0.8)	0.02
Potassium (mEq/L)					0.068
< 3.5	42 (5.0)	0.29	60 (7.2)	0.22		102 (12.2)	0.24
3.5–5.5	99 (11.8)	0.70	218 (26.0)	0.78		317 (37.8)	0.76
> 5.5	1 (0.1)	0.01	0	0.00		1 (0.1)	0.00
Chloride (mEq/L)					0.142
< 90	1 (0.1)	0.01	1 (0.1)	0.00		2 (0.2)	0.00
90–105	82 (9.8)	0.59	187 (22.3)	0.68		269 (32.1)	0.65
> 105	57 (6.8)	0.40	86 (10.3)	0.31		143 (17.1)	0.35
Bicarbonate (mEq/L)					0.001*
< 20	6 (0.7)	0.05	8 (1.0)	0.03		14 (1.7)	0.04
20–26	63 (7.5)	0.48	168 (20.0)	0.67		231 (27.6)	0.61
> 26	62 (7.4)	0.47	74 (8.8)	0.30		136 (16.2)	0.35
Creatine phosphokinase (units/L)					0.273
≤ 1 × 103	1 (0.1)	1.00	2 (0.2)	0.40		3 (0.4)	0.50
> 1 × 103	0	0.00	3 (0.4)	0.60		3 (0.4)	0.50
Creatinine rising ≥ 0.3 mg/dL (within 48 h)					0.311
No	1 (0.1)	0.20	1 (0.1)	0.06		2 (0.2)	0.09
Yes	4 (0.5)	0.80	17 (2.0)	0.94		21 (2.5)	0.91
White blood cell count (cells/µL)					<0.001*
< 4 × 103	10 (1.2)	0.05	4 (0.5)	0.01		14 (1.7)	0.02
4 × 103–11 × 103	149 (17.8)	0.73	263 (29.0)	0.59		392 (46.8)	0.64
> 11 × 103	44 (5.3)	0.22	162 (19.3)	0.40		206 (24.6)	0.34

^*P < 0.05 indicates that there is a statistically significant difference in proportion of patients between the nonantivenom therapy and antivenom therapy groups.

Creatinine phosphokinase was assessed in six patients (0.7% of 838 patients). Rhabdomyolysis occurred in three patients (0.4% of 838 patients), all of whom were in the antivenom therapy group. Of 23 patients (2.7% of 838 patients) who were tested for serum creatinine, 21 patients (2.5% of 838 patients) developed AKI according to the criterion of an increase in serum creatinine by ≥ 0.3 mg/dL within 48 h.²² The majority of patients with AKI were in the antivenom therapy group (17 patients, 2.0% of 838 patients). Liver function tests were not performed on any patient in this study. Six hundred twelve patients (73.0% of 838 patients) underwent testing to determine the white blood cell count. The majority of patients had normal white blood cell counts (392 patients, 46.8% of 838 patients). Two hundred six patients (24.6% of 838 patients) had leukocytosis, with a significantly greater proportion in the antivenom therapy group than in the nonantivenom therapy group (P < 0.001).

Evaluation for rational use of MPV antivenom.

After exclusion of patients with missing data, 601 patients were included in the evaluation for rational use of antivenom, with 201 patients in the nonantivenom therapy group and 400 patients in the antivenom therapy group. Using the status of receiving MPV antivenom as a reference, 229 patients (57.3% of 400 patients) were considered appropriate antivenom recipients. Interestingly, 12 patients did not receive antivenom despite the indications of antivenom administration being met (6.0% of 201 patients). In contrast, 171 patients were administered antivenom without the standard indication (42.7% of 400 patients). This led to the idea of identifying the associated factors of inappropriate use of MPV antivenom.

Associated factors of inappropriate use of MPV antivenom.

Focusing on MPV-bitten patients who did not meet the criteria for antivenom administration, after exclusion of missing data, 360 patients were included in univariate and multivariate regression analyses to identify the independent factors contributing to inappropriate use of MPV antivenom. After adjustment for potential confounders, factors associated with inappropriate MPV antivenom utilization were the presence of local bleeding and mild to moderate thrombocytopenia (platelet count > 50 × 10³/µL to < 140 × 10³/µL) (Table 4). In addition, a multicollinearity test was performed to confirm that these two associated factors are independent from each other.

Table 4.

Univariable and multivariable analyses for factors associated with inappropriate use of MPV antivenom

Factors	Total N	Inappropriate use of MPV antivenom, n (%)	Crude odds ratio	95% CI	P-value	Adjusted odds ratio	95% CI	P-value
Pain
No	180	93 (51.7)	1.00	–	–	–	–	–
Yes	180	78 (43.3)	0.72	0.47–1.08	0.114	–	–	–
Swelling
No	183	90 (49.2)	1.00	–	–	–	–	–
Yes	177	81 (45.8)	0.87	0.58–1.32	0.516	–	–	–
Fang marks
No	43	19 (44.2)	1.00	–	–	1.00	–	–
Yes	317	152 (47.9)	1.16	0.61–2.21	0.643	1.11	0.58–2.14	0.752
Local bleeding
No	268	117 (43.7)	1.00	–	–	1.00	–	–
Yes	92	54 (58.7)	1.83	1.14–2.96	0.013*	1.67	1.03–2.74	0.040*
Bleb
No	357	168 (47.1)	–	–	–	–	–	–
Yes	3	3 (100.0)	NA	NA	NA	–	–	–
Sodium (mEq/L)
< 135	7	4 (57.1)	1.57	0.14–2.91	0.562	–	–	–
135–145	232	108 (46.7)	1.00	–	–	–	–	–
> 145	3	0	NA	NA	NA	–	–	–
Potassium (mEq/L)
< 3.5	69	30 (43.5)	0.84	0.48–1.47	0.546	–	–	–
3.5–5.5	177	85 (48.0)	1.00	–	–	–	–	–
> 5.5	1	0	NA	NA	NA	–	–	–
Chloride (mEq/L)
< 90	1	0	NA	NA	NA	–	–	–
90–105	148	73 (49.3)	1.00	–	–	–	–	–
> 105	92	38 (41.3)	0.73	0.43–1.24	0.245	–	–	–
Bicarbonate (mEq/L)
< 20	9	3 (33.3)	0.60	0.15–2.46	0.477	–	–	–
20–26	122	63 (51.6)	1.00	–	–	–	–	–
> 26	89	33 (37.8)	0.58	0.34–1.01	0.052	–	–	–
White blood cell count (cells/µL)
< 4 × 103	10	3 (30.0)	0.46	0.12–1.83	0.272	–	–	–
4 × 103–11 × 103	255	114 (44.7)	1.00	–	–	1.00	–	–
> 11 × 103	95	54 (56.8)	1.67	1.04–2.67	0.034*	1.58	0.98–2.56	0.063
Platelet count (platelets/µL)
50 × 103–139 × 103	19	15 (78.9)	4.45	1.45–13.68	0.009*	3.88	1.25–12.08	0.019*
≥ 140 × 103	341	156 (45.7)	1.00	–	–	1.00	–	–

MPV = Malayan pit viper; NA = not available.

DISCUSSION

The Malayan pit viper is a clinically significant venomous snake that inhabits agricultural farming areas across Southeast Asia, especially in South Vietnam, Cambodia, Thailand, the Malaysian mainland, and the Indonesian Island of Java.² The bites of these vipers lead to clinically significant outcomes in victims that can be treated through the administration of specific antivenom. In this work, we performed a retrospective study using data from the standardized CRF of MPV-envenomed patients in Thailand’s southernmost region to evaluate and identify the association between clinical laboratory investigation parameters in envenomed patients. We determined clinical and laboratory data of 838 patients who enrolled in a general hospital (Naradhiwas Rajanagarindra Hospital) and community hospitals (Ra-Ngae Hospital, Rueso Hospital, Yi-ngo Hospital 80th Anniversary Commemoration Hospital, and Takbai Hospital) in Narathiwat Province. This study provides necessary insights into the clinical approach for MPV-envenomed patients in rural areas of Southeast Asia.

Over the past 60 years, a few studies have described a limited number of patients with MPV envenomation.^3,4,23–26 To the best of our knowledge, the present work is the largest retrospective study in Southeast Asia to focus on MPV bites. The present study revealed that the peak season of MPV bites was from April to June, which is the early monsoon season,⁴ and patients were most commonly bitten in the time period of approximately 6 to 9 pm. These findings were consistent with a study conducted in Thailand reporting that the highest number of MPV bites occurred in May,^3,4 and the peak incidence was observed at 8 pm.³ The current study showed that 6.6% of MPV-bitten patients exhibited dry bites. This was slightly higher than the proportions reported by Tangtrongchitr et al.³ (3.0%) and the proportions described in the study conducted in Savannakhet Province, Lao People’s Democratic Republic (2.5%).²⁴ Our study showed that 58.8% of MPV-bitten patients experienced local effects, which was lower than the proportions reported by Kraisawat and Promwang²⁷ (86.9%) and Tangtrongchitr et al.³ (93.4%). Impending compartment syndrome was observed in 4.3% of patients in this study, compared with 7.2% reported by Tangtrongchitr et al.³ As expected, impending compartment syndrome was reported only in patients in the antivenom therapy group, as it is one of the criteria for MPV antivenom administration.⁷ Interestingly, bleb and necrosis were also seen only in patients in the antivenom therapy group. Nonetheless, an insufficient number of cases could lead to overinterpretation. Consequently, it is inappropriate to draw any conclusions at present, with a larger number of patients needed to fully evaluate the findings.

In the present study, 27.7% of patients with MPV envenomation exhibited coagulopathies. However, there have been variations in the proportions of patients who experienced systemic effects described in each study, including those reported by Viravan et al.²⁸ (33.8%), Wongthongkam et al.⁴ (35.4–52.5%), Kraisawat and Promwang²⁷ (38.6%), and Tangtrongchitr et al.³ (88.0%). Systemic bleeding was observed in 8.4% of patients in a previous study,³ which was higher than in our report (1.3%). Our study demonstrated that platelet count testing was the most common test performed (72.6%), followed by VCT (41.2%), 20WBCT (29.8%), and INR (25.4%) testing. However, the highest proportions of abnormal laboratory test results were those for unclotted 20WBCT (11.3%; proportion, 0.38), followed by prolonged VCT (10.7%; proportion, 0.26), prolonged INR (6.3%; proportion, 0.25), and platelet count < 50 × 10³/μL (3.5%; proportion, 0.04). In contrast, in a previous study by Tangtrongchitr et al.,³ platelet count and VCT testing were performed in more than 90% of patients. It was reported that a prolonged VCT was the most common abnormal laboratory test result (73.7%; proportion, 0.81), followed by an unclotted 20WBCT (34.1%; proportion, 0.58), prolonged INR (30.5%; proportion, 0.44), and platelet count < 50 × 10³/μL (17.4%; proportion, 0.18).³ Of note, among laboratory tests for snakebite-induced coagulopathy, VCT testing is more complicated and subject to analytical errors.²⁹ The cutoff also varied; for example, Wongtongkam et al.⁴ and Pongpit et al.³⁰ used a period of 30 min as a threshold for prolonged VCT, whereas a period of 20 min was used in our study and most of the previous studies.^3,7,27,31 Therefore, other methods are recommended, including 20WBCT and INR testing.^2,32

More laboratory parameters were collected in our study. In this work, we also showed laboratory data for electrolyte levels after MPV bites. The majority of patients had normal levels of electrolytes. Patients in the nonantivenom therapy group were considered references based on the assumption that they did not have or had only a scant amount of MPV venom in their bloodstream. It can be implied that the hypokalemia and hyperchloremia observed were not specific to the exposure to MPV venom, because there was no statistically significant difference in levels between the nonantivenom therapy and antivenom therapy groups. Nevertheless, the reason for the greater proportion of patients with hyperbicarbonatemia in the nonantivenom therapy group remains unclear. In the present study, rhabdomyolysis was observed in three patients (0.4%), compared with two patients (1.2%) in the study by Tangtrongchitr et al.³ Interestingly, rhabdomyolysis was observed only in patients with impending compartment syndrome. However, the available data are too limited to draw any conclusions. Indeed, compartment syndrome was reported to be associated with rhabdomyolysis because of increased pressure inside an enclosed anatomical space, which reduces blood supply and subsequent tissue necrosis.³³ Of note, this study indicated that not every patient with impending compartment syndrome develops rhabdomyolysis. The most plausible explanation is that the presence of rhabdomyolysis depends on the severity of compartment syndrome.

According to data showing that most patients in this study were admitted to the hospital for a few days (IQR, 1 to 3 days), rationally, an increase in serum creatinine by ≥ 0.3 mg/dl within 48 h was used to define AKI based on the KDIGO guidelines.²² The criteria of an increase in serum creatinine of ≥ 1.5 times baseline within 7 days or in urine volume of < 0.5 mL/kg/h for 6 h²² are not applicable in this study because of data unavailability. Interestingly, the current study showed that 21 patients (2.5%; proportion, 0.91) had AKI, compared with only one patient (0.7%) in the study by Kraisawat and Promwang.²⁷ Although a high proportion of patients with AKI after MPV bite was noted, the very limited number of cases could lead to overinterpretation. A prospective study with a larger number of patients is needed to determine the relevance of the phenomenon observed at the clinical level. In fact, the appearance of AKI in our work can be supported by an animal model showing histopathological changes in renal tissues after the experimental envenomation of MPV, suggesting the presence of nephrotoxic substances, such as svPLA2 and svMP.¹¹ The nephrotoxicity observed after viper envenoming can be minimized or prevented through prophylactic administration of the recommended dose of antivenom.^11,34 Leukocytosis with neutrophilia and left shift are common findings in snakebite envenoming.³⁵ In this preliminary study, leukocytosis was reported in 24.6% of patients, with a significantly greater proportion in the antivenom therapy group than in the nonantivenom therapy group.

In our study, antivenom was administered in 55.0% of patients, compared with 88.0% and 58.0% in the studies by Tangtrongchitr et al.³ and Vongphoumy et al.,²⁴ respectively. Of note, the criteria for antivenom administration in the Laos study by Vongphoumy et al.²⁴ did not include unclotted 20WBCT or prolonged VCT, and a different cutoff for prolonged INR was used (INR > 5). In terms of rational use of antivenom, 57.3% of patients in this study were considered appropriate antivenom recipients, compared with 85.4% in the study by Tangtrongchitr et al.³ Using univariate and multivariate regression analyses for MPV-bitten patients who did not meet the criteria for antivenom administration, the present study revealed that the presence of local bleeding, without any systemic hematological effects, was the independent factor for inappropriate use of MPV antivenom. In addition, we found that antivenom was wrongly used in patients with only mild to moderate thrombocytopenia rather than adhering to the criteria which indicate that it should be administered to those with severe thrombocytopenia. Misuse of snake antivenom potentially occurs in inexperienced practices. Training in snake antivenom application and close supervision by senior staffs are recommended for inexperienced clinicians to minimize possible inappropriate antivenom use.⁶

There were some limitations in our study. As a retrospective study, there was a potential for missing data. Information regarding clinical manifestations and diagnosis (an ICD-10 code) was based on that from primary clinicians. There was only a small number of patients for some laboratory investigations, which makes drawing a conclusion improper at present. This study was conducted only in Narathiwat Province, located in the southern part of Thailand, and was based on Thai patients; therefore, the results may not be generalizable to the entire country or other regions because of variations in practices, laboratory methods, and cutoff values. The strength of our study was that it was a large study focusing on MPV bites. Our study may have a high impact as a resource for comparison with studies with similar populations.

CONCLUSION

We identified local and systemic effects after MPV bites. Coagulopathies, which range from abnormal blood clotting to systemic bleeding, were the major systemic effects, with a significantly greater proportion of affected patients in the antivenom therapy group than in the nonantivenom therapy group. Acute kidney injury developed in some patients. Rational use of antivenom was evaluated. In this study, 57.3% of patients were considered appropriate antivenom recipients. Interestingly, we found that the presence of local bleeding and mild to moderate thrombocytopenia were the independent factors for inappropriate use of MPV antivenom. Hence, reeducation regarding snake antivenom administration should be of greater concern.

Data Availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.