Toxicon Vol. 30, No. 5/6, pp, 599-609, 1992. Printed in Great Britain.
0041~)101/92 $5.00 + .00 © 1992 Pergamon Press Ltd
ENVENOMING BY VIPER BITES IN FRANCE: CLINICAL GRADATION A N D BIOLOGICAL QUANTIFICATION BY ELISA FRANt~OISEAUDEBERT,l MARC SORKINE1"2and CASSIANBON1. ~Unit6des Venins, Unit6 associre Institut Pasteur/INSERM 285 Institut Pasteur, 25 Rue du Dr Roux, 75724 Paris Cedex 15, France; and 2Serviced'Anesthrsie-Rranimation,H6pital Henri Mondor, 51 Av. Marrchal de Lattre de Tassigny,94100 Crrteil, France
(Received 16 October 1991; accepted 15 January 1992) F. AUDEBERT,M. SORKINEand C. BoN. Envenoming by viper bites in France: clinical gradation and biological quantification by ELISA. Toxicon 30, 599609, 1992.--Viper bites are frequent in France but the evaluation of the severity of envenomings and consequently patient treatment has not yet been properly evaluated. The purpose of this study was to measure venom antigens in blood and/or urine of bitten patients and to establish a quantitative relationship with clinical observations. A prospective enquiry was conducted in 1990 in France to collect epidemiological, clinical and biological data from hospitals. Urine and blood samples were tested for their content of Vipera aspis venom antigens by a sandwich enzyme linked immunosorbent assay (ELISA). One hundred and two charts were analysed, from patients presenting documented viper bites. Oedema was the prominent local feature (81 cases). Systemic signs consisted of vomiting and/or diarrhoea (22 cases), slight or severe hypotension (15 cases), shock (2 cases) and bleeding (1 case). A relationship was observed between these systemic signs and the extent of the oedema, which permitted the establishment of a grading scale. Grade 0 (no envenoming) was identified by fang marks and absence of oedema and local reaction; grade 1 (mild envenoming) by local oedema and absence of systemic symptoms; grade 2 (moderate envenoming) by regional oedema and moderate systemic symptoms; and grade 3 (severe envenoming) by extensive oedema and severe systemic symptoms. Quantification of venom antigens in blood or urine of patients by ELISA revealed a significant correlation between clinical signs of envenoming and the level of venom antigens in blood or urine. This indicated that the ELISA test is a useful and predictive tool for clinically grading viper envenomings. INTRODUCTION SNAK~mTES are an important public health problem throughout the world, especially in rural areas of tropical and subtropical countries (REID and TX-mAKSTON, 1983). In these countries, the best treatment is serotherapy and in this context efforts have to be made to prepare large quantities of cheap and efficient antisera (TrmAKSTON, 1983). In Europe, the problem of snakebites is often considered to be of less importance, both because snakes * Author to whomcorrespondenceshouldbe addressed. 599
600
F. AUDEBERT et al.
appear less poisonous and there are fewer bites. Consequently, antivenom is often not administered and patients tend to be treated symptomatically in general hospitals (CLAUDE et al., 1989). However, viper bites are not a rare medical emergency in France (FONTANELLA et al., 1978; BLETTERY et al., 1984; BEDOCK et al., 1989) and this probably also applies to m a n y other Southern and Central European countries, especially during the w a r m season, from M a y to October. Even if the mortality is low (one or two deaths per year; CrnpPAUX and GOYFFON, 1989) clinicians have problems in evaluating the severity of envenomings and selecting appropriate treatment. Presently in Europe, especially in France, the use of antivenom is considered important in treatment of systemic envenomings. Vipera aspis and Vipera berus are the two main viper species found in France, their venom composition is very similar and consequently any difference in clinical feature of envenoming has never been reported. The symptomatology of European viper bites has been widely described (FAVAREL-GUARRIGUESet al., 1974; REID, 1976; JOUGLARD et al., 1981; PERSSON and IRES~DT, 1981). However, the severity of viper bites may sometimes be underestimated (BEDOCK et al., 1989; LACt-IAUXet al., 1990). It is therefore important to define clinical criteria to evaluate the severity of these accidents. M a n y assay systems have been developed over the past 15 years (TnEAKSTON, 1983) for the detection of venom components in blood or urine of patients. A m o n g them, the enzyme l~nked immunosorbant assay (ELISA) seems to be the most suitable test, its main advantages being sensitivity, reliability and simplicity. So far, it has been mainly used in epidemiological studies for the diagnosis of venomous species (THEAKSTON et al., 1981; MINTON et al., 1984; THEAKSTONand WYATT, 1985; VIRAVAN et al., 1986) and in toxicokinetic analysis of experimental envenoming in animals (BARRAL-NETTO et al., 1990; BARRAL-NETTO and VON SOHSTEN, 1991). A few publications (TI-IEAKSTONand WYATT, 1985; VmAVAN et al., 1986) report the use of E L I S A to measure venom levels after envenoming in humans but none in the case of European viper bites. In the present study, clinical data of patients bitten during 1990 by a viper in France have been analysed to define indicators of the gravity of envenomings. An ELISA has also been developed to determine venom concentrations in the blood and urine of such patients and to attempt to establish a correlation between venom levels and the clinical status of the patients. MATERIALS AND METHODS Materials Vipera aspis venom was obtained from the Unit6 des Venins, Institut Pasteur (Paris, France). Purified F(ab)2 (IPSER EUROPE) from hyperimmunizedhorse serum directed against V. aspis, V. berus and V. ammodytes was provided by Pasteur Mrrieux Srrums et Vaceins (Lyon, France). Aminohexyl-Sepharose4B was purchased from Pharmacia (Uppsala, Sweden). Bovine serum albumin, horseradish peroxidase, normal horse immunoglobulin and o-phenylenediaminedihydrochloride (OPD) were purchased from Sigma (St Louis, MO, U.S.A.). Gelatin and glutaraldehyde were purchased from Merck (Darmstadt, F.R.G.). ELISA plates were obtained from CML (Nemours, France). Viper envenoming and collection of biological samples The inquiry on viper bite envenoming was conducted by the Unit6 des Venins of Institut Pasteur (Paris, France), in France between May and September 1990. A specific document was sent to the major hospitals (emergency departments and intensive care units) and Poison Control Centers in areas where there was a high probability of viper bites. The document included a questionnaire concerning: epidemiologicaldata, the site of bite, clinical and biological manifestations and medical treatment. Blood or serum and urine samples collected from patients during the first 12 hr of hospitalization were sent to Institut Pasteur (Unit6 des Venins). For each
Viper Bites in France
601
[ ] 10 cases or more [ ] 5 to 9 cases II
[ ] 1 to 4 cases
FIG. 1. GEOGRAPHICALDISTRIBUTIONOF RECORDEDCASES.
patient, we calculated the mean value of venom concentrations in the different samples collected as measured by ELISA. Transportation time did not exceed 48 hr and samples were kept at +4°C if analysed within a day, or stored frozen at - 1 8 ° C until assay.
Purification o f specific anti-Vipera venom F( ab ) 2 and preparation o f peroxidase-labelled F ( ab ) : Specific anti-Vipera venom F(ab)2 were purified from hyperimmunized horse serum (IPSER EUROPE) by immunoattinity chromatography on immunosorbent columns prepared by covalent coupling of a mixture of V. aspis, II. berus and II. ammodytes venoms in equal proportions. Purified F(ab)2 were labelled with peroxidase by glutaraldehyde coupling using a two-step procedure (AVRAMEASand TERNIrqCK, 1971).
E L I S A f o r Vipera venom antigens Wells of microtitration plates were coated with 100tal of 100 mM sodium carbonate (pH 9.6) containing 5 ~tg/ml of purified antivenom F(ab)2 or 5 Ixg/ml of normal horse immunoglobulin (negative control), for 2 hr at 37°C. Plates were washed with 10mM sodium phosphate buffer pH 7.6 containing 150mM NaCI (PBS) and 0.1% Tween-20 (PBS-Tween) and then incubated with PBS containing 0.5% gelatin, for 45min at 37°C (blocking step). The plates were rewashed with PBS-Tween. One-hundred microlitres of undiluted urine or blood sample, or of V. aspis venom diluted in control human serum or control human urine as a standard, were added to each well and incubated for 1 hr at 37°C. The plates were washed again and incubated with purified horse antivenom F(ab)2 labelled with peroxidase (6~tg/ml and 1001xl/well) for 45min at 37°C. Excess unbound conjugate was eliminated by five washings. Substrate medium (100 ixl of 10mM sodium phosphate, pH 7.3, containing 2 mg/ml of OPD and 0.06% of perhydrol) was added to each well and incubated for 7 min in the dark at room temperature. The reaction was stopped by adding 50 Ixl/well of 2 N sulphuric acid. Optical densities were measured at 490 nm using a microtitration plate spectrophotometer reader. Venom concentrations in the samples were recorded by reference to a standard curve of V. aspis venom (from 1.5 to 100ng/ml) diluted in human control serum (or blood) or urine. Standard curves and controls were identical when performed with whole blood or serum.
RESULTS
Clinical analysis As a result of the inquiry (Fig. 1), information concerning 117 patients was received. Fifteen did not contain enough data to be analysed or were characterized by non-specific ELISA responses and were discarded. All patients entered in the review presented a
602
F. AUDEBERT et al. TABLE 1. CLINICALMANIFESTATIONS
Symptoms or signs Pain Weakness Paresthesia Vomiting and/or diarrhoea Hypotension moderate severe Shock Bleeding Anaphylaxis reactions Local oedema Regional ocdema Extensive oedema Ecchymosis Necrosis Total number of cases
Number of cases 35 6 5 22
li}
17 1
7 48
29 5 16 0 102
confirmed viper bite, i.e. clear evidence of fang marks. The age distribution was from 2 to 87 years (mean 33 years). The patients were predominantly males (66%). The rites of bite were equally distributed between upper (52 cases) and lower (50 cases) limbs. No bite was reported on other parts of the body (face, thorax, abdomen, etc.). Concerning medical treatment, no patient received serotherapy. About a third of the patients (35 cases) complained of local pain, essentially associated with the extent of oedema (Table 1). Weakness was noted only in six cases and paresthesia in five cases. Oedema was the prominent sign of local envenoming: it was present in about 80% of the patients entered in the review (Table 1). Oedema was confined around the bite site (local oedema) in 48 cases, it affected the major part of the limb (regional oedema) in 29 cases, and extended to the ipsilateral or contralateral trunk (extensive oedema) in three and two cases, respectiyely. In these two latter cases, oedema progressively extended to reach a maximum 5 days after the bite. In this series, when oedema was not observed 2 hr after the bite, it never appeared subsequently. Local ecchymosis was noted in 16 patients, but skin necrosis was never reported. Systemic signs of envenoming were also observed. Vomiting and diarrhoea were recorded for more than 2 hr in 22 patients (Table 1). Seventeen patients presented with hypotension (systolic blood pressure lower than 90 mm Hg) not including patients who developed hypotension following manifestations of anaphylaxis. In most cases which had a slight hypotension (12 out of 17), blood pressure was rapidly restored after infusion of crystalloids or colloids (500 ml). Severe hypotension occurred in five other cases which necessitated massive i.v. fluid therapy, requiring more than 1000 ml and the addition of dopamine in two cases, both of whom developed severe shock. One of them, an 87-yearold woman, died following shock, spontaneous bleeding with disseminated intravascular coagulation, and finally pulmonary and renal failure. Of ten patients whose coagulation parameters were investigated, two had thrombocytopenia and three low fibrinogen level; none of these patients presented with clinical signs. Early anaphylactic reactions directed to venom occurred in seven cases, four of which developed anaphylactic or anaphylactoid shock. One of these had a previous history of multiple viper bites.
Viper Bites in France
603
TABLE 2. RELATIONSHIP BETWEEN THE EXTENT OF OEDEMA AND SIGNS OF ENVENOMING
(3edema
Absent
Local
20 0
48 0
29 18
5 4
102 22
0 0 0 0
0 0 0 0
12 0 0 0
0 3 2 1
12) ~ 17
Symptoms or signs Vomiting/diarrhoea Hypotension moderate severe Shock Bleeding
Regional Extensive
Total
1
The severity of systemic signs appeared to be correlated with the extent of oedema, the prominent local feature of envenoming (Table 2). Vomiting and diarrhoea (22 cases) appeared only in patients with regional or extensive oedema. Of 34 patients who had regional or extensive oedema, 22 (about 60%) also presented vomiting and/or diarrhoea, while none of the 68 patients with local or no oedema had these symptoms (Table 2). The 12 patients showing moderate hypotension also had regional oedema. Further, victims with severe hypotension or shock had extensive oedema (Table 2), while no patient with only local or no oedema developed even moderate hypotension. The extent of oedema (regional and extensive) was a good indicator of systemic envenoming. The statistical analysis of the relationship between oedema and systemic manifestations showed a sensitivity of 100% and a specificity of 85%, with a positive predictive value of 65% and a negative predictive value of 100%. The above relationships allowed us to classify envenomings in four grades of increasing severity (Table 3). The signs and symptoms taken into account in our grading were specific for envenoming by V. asp/s and V. berus. The rapid onset of clinical signs within 2-5 hr following the bite was indicative of a severe (grade 3) rather than a moderate (grade 2) envenoming. The 102 cases reported in the inquiry have been graded as described above (Table 4). As expected, all patients presenting signs and symptoms selected to establish the gradation scale were appropriately classified. Pain, which was mentioned in about 30% of the patients, appeared to be associated with oedema rather than with the severity of envenoming. Similarly, weakness, paresthesia, ecchymosis and, more interestingly, anaphylaxic reactions were occasionally noted in mild as well as in moderate or severe envenomings, and thus do not seem to correlate with the gradation of the envenoming TABLE 3. GRADE OF ENVENOMING SUBSEQUENT TO V.
Grade Name 0
No envenoming
1
Minimal envenoming
2
Moderate envenoming
3
Severe envenoming
asp/s
OR V. berus VIPER BITES
Characteristics/symptoms Fang marks No oedema Local oedema around the bite area No systemic symptoms Regional oedema involving a major part of limb Moderate systemic symptoms (slight hypotension, vomiting, diarrhoea) Extensive oedema spreading into the trunk Severe systemic symptoms (prolonged hypotension, shock, bleeding)
F. A U D E B E R T et al.
604 TABLE 4.
DISTRIBUTION OF SIGNS AND SYMPTOMS IN DIFFERENT GRADES OF ENVENOMING
Symptoms or signs
Grade
Envenoming Oedema local regional extensive Vomiting and/or diarrhoea Hypotension slight severe Shock Pain Weakness Parestbesia Ecchymosis Anaphylaxis reactions N u m b e r o f cases
0
1
Absent
2
3
Minimal Moderate
Total
Severe
0 0 0
48 0 0
0 29 0
0 0 5
48 29 5
0
0
18
4
22
0 0 0 0 1 1 0 0 20
0 0 0 11 5 3 7 1 48
12 0 0 20 0 I 7 5 29
0 3 2 4 0 0 2 1 5
12 3 2 35 6 5 16 7 102
(Table 4). Likewise no correlation between the site of bite or the age of the patient and the severity of the envenoming was observed (results not shown).
Biological analysis A sensitive sandwich ELISA was set up to determine the level of Vipera venom antigens in blood (or serum) and urine samples collected from bitten patients during hospitalization. Figure 2 shows typical titration curves obtained by diluting Vipera aspis venom in blood, serum or urine collected from healthy donors. In all cases, accurate determinations could be obtained from 1 to 50 ng/ml.
1.0
E
0.8
c (M o>
.r C
0.6
,m It
0.4
0041~)101/92 $5.00 + .00 © 1992 Pergamon Press Ltd
ENVENOMING BY VIPER BITES IN FRANCE: CLINICAL GRADATION A N D BIOLOGICAL QUANTIFICATION BY ELISA FRANt~OISEAUDEBERT,l MARC SORKINE1"2and CASSIANBON1. ~Unit6des Venins, Unit6 associre Institut Pasteur/INSERM 285 Institut Pasteur, 25 Rue du Dr Roux, 75724 Paris Cedex 15, France; and 2Serviced'Anesthrsie-Rranimation,H6pital Henri Mondor, 51 Av. Marrchal de Lattre de Tassigny,94100 Crrteil, France
(Received 16 October 1991; accepted 15 January 1992) F. AUDEBERT,M. SORKINEand C. BoN. Envenoming by viper bites in France: clinical gradation and biological quantification by ELISA. Toxicon 30, 599609, 1992.--Viper bites are frequent in France but the evaluation of the severity of envenomings and consequently patient treatment has not yet been properly evaluated. The purpose of this study was to measure venom antigens in blood and/or urine of bitten patients and to establish a quantitative relationship with clinical observations. A prospective enquiry was conducted in 1990 in France to collect epidemiological, clinical and biological data from hospitals. Urine and blood samples were tested for their content of Vipera aspis venom antigens by a sandwich enzyme linked immunosorbent assay (ELISA). One hundred and two charts were analysed, from patients presenting documented viper bites. Oedema was the prominent local feature (81 cases). Systemic signs consisted of vomiting and/or diarrhoea (22 cases), slight or severe hypotension (15 cases), shock (2 cases) and bleeding (1 case). A relationship was observed between these systemic signs and the extent of the oedema, which permitted the establishment of a grading scale. Grade 0 (no envenoming) was identified by fang marks and absence of oedema and local reaction; grade 1 (mild envenoming) by local oedema and absence of systemic symptoms; grade 2 (moderate envenoming) by regional oedema and moderate systemic symptoms; and grade 3 (severe envenoming) by extensive oedema and severe systemic symptoms. Quantification of venom antigens in blood or urine of patients by ELISA revealed a significant correlation between clinical signs of envenoming and the level of venom antigens in blood or urine. This indicated that the ELISA test is a useful and predictive tool for clinically grading viper envenomings. INTRODUCTION SNAK~mTES are an important public health problem throughout the world, especially in rural areas of tropical and subtropical countries (REID and TX-mAKSTON, 1983). In these countries, the best treatment is serotherapy and in this context efforts have to be made to prepare large quantities of cheap and efficient antisera (TrmAKSTON, 1983). In Europe, the problem of snakebites is often considered to be of less importance, both because snakes * Author to whomcorrespondenceshouldbe addressed. 599
600
F. AUDEBERT et al.
appear less poisonous and there are fewer bites. Consequently, antivenom is often not administered and patients tend to be treated symptomatically in general hospitals (CLAUDE et al., 1989). However, viper bites are not a rare medical emergency in France (FONTANELLA et al., 1978; BLETTERY et al., 1984; BEDOCK et al., 1989) and this probably also applies to m a n y other Southern and Central European countries, especially during the w a r m season, from M a y to October. Even if the mortality is low (one or two deaths per year; CrnpPAUX and GOYFFON, 1989) clinicians have problems in evaluating the severity of envenomings and selecting appropriate treatment. Presently in Europe, especially in France, the use of antivenom is considered important in treatment of systemic envenomings. Vipera aspis and Vipera berus are the two main viper species found in France, their venom composition is very similar and consequently any difference in clinical feature of envenoming has never been reported. The symptomatology of European viper bites has been widely described (FAVAREL-GUARRIGUESet al., 1974; REID, 1976; JOUGLARD et al., 1981; PERSSON and IRES~DT, 1981). However, the severity of viper bites may sometimes be underestimated (BEDOCK et al., 1989; LACt-IAUXet al., 1990). It is therefore important to define clinical criteria to evaluate the severity of these accidents. M a n y assay systems have been developed over the past 15 years (TnEAKSTON, 1983) for the detection of venom components in blood or urine of patients. A m o n g them, the enzyme l~nked immunosorbant assay (ELISA) seems to be the most suitable test, its main advantages being sensitivity, reliability and simplicity. So far, it has been mainly used in epidemiological studies for the diagnosis of venomous species (THEAKSTON et al., 1981; MINTON et al., 1984; THEAKSTONand WYATT, 1985; VIRAVAN et al., 1986) and in toxicokinetic analysis of experimental envenoming in animals (BARRAL-NETTO et al., 1990; BARRAL-NETTO and VON SOHSTEN, 1991). A few publications (TI-IEAKSTONand WYATT, 1985; VmAVAN et al., 1986) report the use of E L I S A to measure venom levels after envenoming in humans but none in the case of European viper bites. In the present study, clinical data of patients bitten during 1990 by a viper in France have been analysed to define indicators of the gravity of envenomings. An ELISA has also been developed to determine venom concentrations in the blood and urine of such patients and to attempt to establish a correlation between venom levels and the clinical status of the patients. MATERIALS AND METHODS Materials Vipera aspis venom was obtained from the Unit6 des Venins, Institut Pasteur (Paris, France). Purified F(ab)2 (IPSER EUROPE) from hyperimmunizedhorse serum directed against V. aspis, V. berus and V. ammodytes was provided by Pasteur Mrrieux Srrums et Vaceins (Lyon, France). Aminohexyl-Sepharose4B was purchased from Pharmacia (Uppsala, Sweden). Bovine serum albumin, horseradish peroxidase, normal horse immunoglobulin and o-phenylenediaminedihydrochloride (OPD) were purchased from Sigma (St Louis, MO, U.S.A.). Gelatin and glutaraldehyde were purchased from Merck (Darmstadt, F.R.G.). ELISA plates were obtained from CML (Nemours, France). Viper envenoming and collection of biological samples The inquiry on viper bite envenoming was conducted by the Unit6 des Venins of Institut Pasteur (Paris, France), in France between May and September 1990. A specific document was sent to the major hospitals (emergency departments and intensive care units) and Poison Control Centers in areas where there was a high probability of viper bites. The document included a questionnaire concerning: epidemiologicaldata, the site of bite, clinical and biological manifestations and medical treatment. Blood or serum and urine samples collected from patients during the first 12 hr of hospitalization were sent to Institut Pasteur (Unit6 des Venins). For each
Viper Bites in France
601
[ ] 10 cases or more [ ] 5 to 9 cases II
[ ] 1 to 4 cases
FIG. 1. GEOGRAPHICALDISTRIBUTIONOF RECORDEDCASES.
patient, we calculated the mean value of venom concentrations in the different samples collected as measured by ELISA. Transportation time did not exceed 48 hr and samples were kept at +4°C if analysed within a day, or stored frozen at - 1 8 ° C until assay.
Purification o f specific anti-Vipera venom F( ab ) 2 and preparation o f peroxidase-labelled F ( ab ) : Specific anti-Vipera venom F(ab)2 were purified from hyperimmunized horse serum (IPSER EUROPE) by immunoattinity chromatography on immunosorbent columns prepared by covalent coupling of a mixture of V. aspis, II. berus and II. ammodytes venoms in equal proportions. Purified F(ab)2 were labelled with peroxidase by glutaraldehyde coupling using a two-step procedure (AVRAMEASand TERNIrqCK, 1971).
E L I S A f o r Vipera venom antigens Wells of microtitration plates were coated with 100tal of 100 mM sodium carbonate (pH 9.6) containing 5 ~tg/ml of purified antivenom F(ab)2 or 5 Ixg/ml of normal horse immunoglobulin (negative control), for 2 hr at 37°C. Plates were washed with 10mM sodium phosphate buffer pH 7.6 containing 150mM NaCI (PBS) and 0.1% Tween-20 (PBS-Tween) and then incubated with PBS containing 0.5% gelatin, for 45min at 37°C (blocking step). The plates were rewashed with PBS-Tween. One-hundred microlitres of undiluted urine or blood sample, or of V. aspis venom diluted in control human serum or control human urine as a standard, were added to each well and incubated for 1 hr at 37°C. The plates were washed again and incubated with purified horse antivenom F(ab)2 labelled with peroxidase (6~tg/ml and 1001xl/well) for 45min at 37°C. Excess unbound conjugate was eliminated by five washings. Substrate medium (100 ixl of 10mM sodium phosphate, pH 7.3, containing 2 mg/ml of OPD and 0.06% of perhydrol) was added to each well and incubated for 7 min in the dark at room temperature. The reaction was stopped by adding 50 Ixl/well of 2 N sulphuric acid. Optical densities were measured at 490 nm using a microtitration plate spectrophotometer reader. Venom concentrations in the samples were recorded by reference to a standard curve of V. aspis venom (from 1.5 to 100ng/ml) diluted in human control serum (or blood) or urine. Standard curves and controls were identical when performed with whole blood or serum.
RESULTS
Clinical analysis As a result of the inquiry (Fig. 1), information concerning 117 patients was received. Fifteen did not contain enough data to be analysed or were characterized by non-specific ELISA responses and were discarded. All patients entered in the review presented a
602
F. AUDEBERT et al. TABLE 1. CLINICALMANIFESTATIONS
Symptoms or signs Pain Weakness Paresthesia Vomiting and/or diarrhoea Hypotension moderate severe Shock Bleeding Anaphylaxis reactions Local oedema Regional ocdema Extensive oedema Ecchymosis Necrosis Total number of cases
Number of cases 35 6 5 22
li}
17 1
7 48
29 5 16 0 102
confirmed viper bite, i.e. clear evidence of fang marks. The age distribution was from 2 to 87 years (mean 33 years). The patients were predominantly males (66%). The rites of bite were equally distributed between upper (52 cases) and lower (50 cases) limbs. No bite was reported on other parts of the body (face, thorax, abdomen, etc.). Concerning medical treatment, no patient received serotherapy. About a third of the patients (35 cases) complained of local pain, essentially associated with the extent of oedema (Table 1). Weakness was noted only in six cases and paresthesia in five cases. Oedema was the prominent sign of local envenoming: it was present in about 80% of the patients entered in the review (Table 1). Oedema was confined around the bite site (local oedema) in 48 cases, it affected the major part of the limb (regional oedema) in 29 cases, and extended to the ipsilateral or contralateral trunk (extensive oedema) in three and two cases, respectiyely. In these two latter cases, oedema progressively extended to reach a maximum 5 days after the bite. In this series, when oedema was not observed 2 hr after the bite, it never appeared subsequently. Local ecchymosis was noted in 16 patients, but skin necrosis was never reported. Systemic signs of envenoming were also observed. Vomiting and diarrhoea were recorded for more than 2 hr in 22 patients (Table 1). Seventeen patients presented with hypotension (systolic blood pressure lower than 90 mm Hg) not including patients who developed hypotension following manifestations of anaphylaxis. In most cases which had a slight hypotension (12 out of 17), blood pressure was rapidly restored after infusion of crystalloids or colloids (500 ml). Severe hypotension occurred in five other cases which necessitated massive i.v. fluid therapy, requiring more than 1000 ml and the addition of dopamine in two cases, both of whom developed severe shock. One of them, an 87-yearold woman, died following shock, spontaneous bleeding with disseminated intravascular coagulation, and finally pulmonary and renal failure. Of ten patients whose coagulation parameters were investigated, two had thrombocytopenia and three low fibrinogen level; none of these patients presented with clinical signs. Early anaphylactic reactions directed to venom occurred in seven cases, four of which developed anaphylactic or anaphylactoid shock. One of these had a previous history of multiple viper bites.
Viper Bites in France
603
TABLE 2. RELATIONSHIP BETWEEN THE EXTENT OF OEDEMA AND SIGNS OF ENVENOMING
(3edema
Absent
Local
20 0
48 0
29 18
5 4
102 22
0 0 0 0
0 0 0 0
12 0 0 0
0 3 2 1
12) ~ 17
Symptoms or signs Vomiting/diarrhoea Hypotension moderate severe Shock Bleeding
Regional Extensive
Total
1
The severity of systemic signs appeared to be correlated with the extent of oedema, the prominent local feature of envenoming (Table 2). Vomiting and diarrhoea (22 cases) appeared only in patients with regional or extensive oedema. Of 34 patients who had regional or extensive oedema, 22 (about 60%) also presented vomiting and/or diarrhoea, while none of the 68 patients with local or no oedema had these symptoms (Table 2). The 12 patients showing moderate hypotension also had regional oedema. Further, victims with severe hypotension or shock had extensive oedema (Table 2), while no patient with only local or no oedema developed even moderate hypotension. The extent of oedema (regional and extensive) was a good indicator of systemic envenoming. The statistical analysis of the relationship between oedema and systemic manifestations showed a sensitivity of 100% and a specificity of 85%, with a positive predictive value of 65% and a negative predictive value of 100%. The above relationships allowed us to classify envenomings in four grades of increasing severity (Table 3). The signs and symptoms taken into account in our grading were specific for envenoming by V. asp/s and V. berus. The rapid onset of clinical signs within 2-5 hr following the bite was indicative of a severe (grade 3) rather than a moderate (grade 2) envenoming. The 102 cases reported in the inquiry have been graded as described above (Table 4). As expected, all patients presenting signs and symptoms selected to establish the gradation scale were appropriately classified. Pain, which was mentioned in about 30% of the patients, appeared to be associated with oedema rather than with the severity of envenoming. Similarly, weakness, paresthesia, ecchymosis and, more interestingly, anaphylaxic reactions were occasionally noted in mild as well as in moderate or severe envenomings, and thus do not seem to correlate with the gradation of the envenoming TABLE 3. GRADE OF ENVENOMING SUBSEQUENT TO V.
Grade Name 0
No envenoming
1
Minimal envenoming
2
Moderate envenoming
3
Severe envenoming
asp/s
OR V. berus VIPER BITES
Characteristics/symptoms Fang marks No oedema Local oedema around the bite area No systemic symptoms Regional oedema involving a major part of limb Moderate systemic symptoms (slight hypotension, vomiting, diarrhoea) Extensive oedema spreading into the trunk Severe systemic symptoms (prolonged hypotension, shock, bleeding)
F. A U D E B E R T et al.
604 TABLE 4.
DISTRIBUTION OF SIGNS AND SYMPTOMS IN DIFFERENT GRADES OF ENVENOMING
Symptoms or signs
Grade
Envenoming Oedema local regional extensive Vomiting and/or diarrhoea Hypotension slight severe Shock Pain Weakness Parestbesia Ecchymosis Anaphylaxis reactions N u m b e r o f cases
0
1
Absent
2
3
Minimal Moderate
Total
Severe
0 0 0
48 0 0
0 29 0
0 0 5
48 29 5
0
0
18
4
22
0 0 0 0 1 1 0 0 20
0 0 0 11 5 3 7 1 48
12 0 0 20 0 I 7 5 29
0 3 2 4 0 0 2 1 5
12 3 2 35 6 5 16 7 102
(Table 4). Likewise no correlation between the site of bite or the age of the patient and the severity of the envenoming was observed (results not shown).
Biological analysis A sensitive sandwich ELISA was set up to determine the level of Vipera venom antigens in blood (or serum) and urine samples collected from bitten patients during hospitalization. Figure 2 shows typical titration curves obtained by diluting Vipera aspis venom in blood, serum or urine collected from healthy donors. In all cases, accurate determinations could be obtained from 1 to 50 ng/ml.
1.0
E
0.8
c (M o>
.r C
0.6
,m It
0.4
