Am. J. Trop. Med. Hyg., 62(2), 2000, pp.

277–283
Copyright 䉷 2000 by The American Society of Tropical Medicine and Hygiene

SCORPION ENVENOMATION AND SEROTHERAPY IN MOROCCO

NOREDDINE GHALIM, BOUCHRA EL-HAFNY, FATIMA SEBTI, JAAFAR HEIKEL, NOUREDDINE LAZAR,
RADIA MOUSTANIR, AND ABDELLAH BENSLIMANE
Institut Pasteur du Maroc, Casablanca, Morocco; Laboratoire de Biostatistique et d’Epidémiologie, Hassan II University,
Medical School, Casablanca, Morocco; Ministère de la Santé, Délégation Régionale de Casablanca, Casablanca, Morocco

Abstract. A clinical and biologic study was conducted in Morocco to assess the efficiency of antivenom therapy
for treating victims of scorpion stings. Epidemiologic and clinical data were collected from 275 patients envenomed
by Androctonus mauretanicus mauretanicus and Buthus occitanus scorpions. Patients received antivenom or other
drugs. Blood samples were collected at the time of hospital admission and 1 hr and 3 hr after treatment. Serum
venom levels were quantified by using an ELISA. An association was found between clinical signs of envenoming
and the level of venom in serum. Patients classified as grade II (moderate envenoming) had higher serum levels of
venom level than patients classified as grade I (mild envenoming). At admission to the hospital, the mean venom
concentration was not significantly different between the group not treated with antivenom, the group who received
2–5 ml of antivenom, and the group who received 10 ml of antivenom. A significant decrease in serum venom levels
and an improvement in the clinical conditions were observed in patients administered 10 ml of antivenom. The lower
decrease in serum venom levels in patients who received 2–5 ml of antivenom was due to lower doses of antivenom.
No difference in the venom concentration was observed in patients who were not treated with antivenom. The absence
of administration of antivenom increased the risk of developing clinical signs at the end of the hospitalization period.
However, this risk was much higher when more than 1 hr elapsed between the time of the scorpion sting and the
time of hospital admission. The results demonstrate that antivenom is effective in decreasing circulating venom and
morbidity. Serotherapy is more efficient when given as soon as possible after envenomation and with adequate
quantities of antivenom.

In north Africa, as in numerous tropical countries, enven- measures venom levels after scorpion envenoming has been
omation by scorpion stings is a major public health problem, used for cases stung by Androctonus australis garzonii, B.
particularly in children.1–4 The black scorpion (Androctonus occitanus tunetanus, and Tityus serrulatus,4,21,22 but not for
mauretanicus mauretanicus) and the yellow scorpion (Bu- those stung by A. mauretanicus mauretanicus or B. occitan-
thus occitanus) are the most dangerous scorpions and are us.
responsible for the majority of stings in Morocco. The epi- In this study, an ELISA was developed for detecting the
demiologic data are incomplete, but the number of scorpion toxic fraction of Moroccan scorpion venom in the serum of
stings is estimated to be 40,000 per year in Morocco. sting victims. We conducted a prospective study in the 4
Scorpion venoms are a complex biochemical mixture con- provinces of Morocco (El Kalaa, El-Jadida, Agadir, and Tan-
taining numerous neurotoxic polypeptides.5,6 These polypep- Tan) where scorpion stings are most likely to occur, and also
tides enhance excitability of nerve and muscle cells in scor- analyzed the kinetics of serum venom levels and the evo-
pion sting victims and also cause death, particularly in chil- lution of the clinical signs of envenoming after antivenom
dren.7 The onset of clinical symptoms is rapid (within 5–30 therapy to assess the efficacy of serotherapy. The epidemi-
min) following the sting.8,9 Respiratory failure and cardio- ologic, clinical, and biologic results of the study are report-
vascular manifestations are the usual causes of death.9–11 The ed.
toxicity of A. mauretanicus mauretanicus venom is due to
the presence of neurotoxins that are specifically active on PATIENTS, MATERIALS, AND METHODS
sodium and potassium channels.6,12
The severity of scorpion envenoming and the rapid dif- Patients. A prospective study of scorpion stings was con-
fusion of inoculated venom require that appropriate treat- ducted in 4 provinces of Morocco (El-Kala, El-Jadida, Aga-
ment be started as soon as possible after the sting. Most dir, and Tan-Tan) from July to October 1997. A question-
investigators consider antivenom to be the only specific naire was distributed to physicians in 19 hospitals to collect
treatment for envenoming by scorpion stings.1,3,13–16 How- patient data (name, age, sex, scorpion identification, time of
ever, others have questioned the usefulness of antivenom in sting, treatment applied, clinical manifestations, and the du-
eliminating cardiovascular manifestations of scorpion ration of hospitalization). Scorpion identification was made
stings.17–19 by the patient or the physician. The severity of envenom-
In Morocco, serotherapy has been one of the major ther- ation of each patient was measured by physicians according
apeutic measures used in te treatment of scorpion enven- to a previously defined severity scale described by Krifi and
oming for the last 30 years.20 However, the choice of the others.4 The patients were classified into 1 of 3 clinical
antivenom dose and the administration protocol remain ar- grades. Grade I (mild envenoming) refers to patients pre-
bitrary. No clinical study has been carried out regarding the senting with only local symptoms, local pain, and a burning
efficacy of scorpion antivenom in neutralizing circulating sensation. Grade II (moderate envenoming) refers to patients
venom antigens or its ability to decrease morbidity. The lack presenting with local and general symptoms. Grade III (se-
of such clinical and biologic studies could be due to the vere envenoming) refers to patients presenting with local and
absence of suitable techniques for detecting and quantifying general symptoms, together with cardiocirculatory shock, re-
circulating venom in envenomed patients. An ELISA that spiratory failure, acute pulmonary edema, hyperthermia, and

277
278 GHALIM AND OTHERS

neurologic symptoms such as priapism, convulsions, or labeled with 10 mg of peroxidase by glutaraldehyde cou-
coma. pling using a 2-step procedure.25 Labeled F(ab⬘)2 was titrated
A total of 304 questionnaires were collected, but 29 cases and stored at ⫺20⬚C in 50% glycerol.
were subsequently eliminated because of incomplete data. Enzyme-linked immunosorbent assay. Microtiter plates
Of the 275 patients included in the study, 49% were treated were coated with 5 ug/ml (100 ␮l/well) of F(ab⬘)2 specific
with antivenom, 35% with other drugs (calcium, corticoids, to the toxic fraction of A. mauretanicus mauretanicus venom
antihistamines), and 16% received both treatments. All treat- in 0.1 M sodium carbonate buffer, pH 9.6 and incubated
ment decisions were made by the personal physicians of the overnight at 4⬚C. The plates were washed 4 times with phos-
patients. Of the 179 patients treated with antivenom, 27% phate-buffered saline (PBS), pH 7.4, containing 0.1% Tween
received 2–5 ml and 73% received 10 ml. Antivenom was 20 (PBS-Tween) and coated with PBS containing 5% pow-
injected intramuscularly in 77.6% of the patients, subcuta- dered milk and incubated for 45 min at 37⬚C. The plates
neously in 6.2%, and by both of these routes in 16.2% of were then washed 4 times with PBS-Tween and 100 ␮l of
the patients. serum samples were added to each well and incubated for 1
Venous blood samples used as controls were obtained hr at 37⬚C. The plates were washed with PBS-Tween, 100
from healthy subjects who had never been stung by scorpi- ␮l of peroxidase-labeled F(ab⬘)2 (diluted 1,000-fold in PBS-
ons and who were undergoing routine health examinations Tween containing 5% powdered milk) were added to each
at the Biological Center of the Pasteur Institute (Casablanca, well, and the plates were incubated for 45 min at 37⬚C. The
Morocco). Blood samples of patients stung by scorpions plates were then washed with PBS-Tween and 100 ␮l of o-
were obtained at different hospitals in the 4 selected prov- phenylenediamine solution (2 mg/ml in 0.01 M potassium
inces. Samples were obtained immediately upon admission phosphate buffer, pH 7.3, in the presence of 0.06% hydrogen
of the patients into the hospital (before the initiation of treat- peroxide) were added and the reaction was developed for 10
ment with antivenom and/or other drugs). Additional sam- min at room temperature in the dark. The reaction was
ples were obtained 1 and 3 hr after the start of antivenom stopped by adding 1 M HCl per well and optical density was
therapy and/or other drugs. Blood was collected into vacu- measured at 492 nm with a microtitration plate reader
tainer tubes and immediately centrifuged for at 1,500 ⫻ g (LP400; Diagnostic Pasteur, Paris, France). Venom concen-
for 15 min at 4⬚C. Serum samples were kept frozen in ali- trations in samples were interpolated from a standard curve
quots at ⫺20⬚C until tested. Informed consent was obtained using known venom concentrations diluted in human serum
from all patients who participated in this study. Venous from healthy donors.
blood samples were obtained only from subjects who agreed The specificity of the ELISA was determined by testing
to participate in the study. The protocol used in this study 100 samples of sera from healthy non-envenomed donors.
was reviewed and approved by the Epidemiological Depart- The sensitivity of the assay was analyzed by the constructing
ment of the Moroccan Ministry of Health. calibration curves using scorpion (A. mauretanicus maure-
Venoms. The venoms of scorpions (A. mauretanicus tanicus and B. occitanus) or snake (N. haje haje, C. cerastes,
mauretanicus and B. occitanus) and snakes (Naja haje haje, and V. lebetina) venoms. The intra-assay variability was es-
Cerastes cerastes, and Vipera lebetina) were obtained by timated by analyzing 12 replicates of sera samples contain-
manual stimulation of animals in captivity at the Experi- ing known venom concentrations (10, 25, and 50 ng/ml) on
mental Center of the Pasteur Institute of Morocco. the same titration plate and with the same reagents. The in-
Antivenom. Scorpion antivenom containing F(ab⬘)2 frag- ter-assay variability was assessed by measuring the venom
ments was prepared in horses hyperimmunized with crude concentration in the same sample for 8 consecutive days on
venom of A. mauretanicus mauretanicus. One milliliter of different titration plates and with different reagents. Sera of
scorpion antivenom will neutralize a minimum of 12.5 50% patients were assayed 2 times in duplicate. The 3 timed col-
lethal doses (LD50s) in mice with a mean weight of 20 grams lections from a given patient were assayed simultaneously
(1 LD50 is the amount of venom that kills 50% of the mouse in the same plate.
population). The LD50 was determined as previously de- Statistical analysis. Data were analyzed with the SPSS
scribed.23 software (SPSS Institute, Chicago, IL) using analysis of var-
Microtiter plates were obtained from CML (Nemours,
iance and appropriate measures of association (odds ratio
France). Horseradish peroxidase and o-phenylenediamine
[OR] and Pearson’s chi-square). Multivariate analysis was
were obtained from Sigma (St. Louis, MO). Sephadex G-50
done using hierarchic logistic regression. All tests were two-
and CNBr-activated Sepharose 4B were obtained from Phar-
tailed. Results are expressed as the mean ⫾ SD. Values were
macia (Uppsala, Sweden). Glutaraldehyde and all other
considered to be significantly different if P ⬍ 0.05.
chemical salts and solvents were obtained from Merck
(Darmstadt, Germany) or Prolabo (Paris, France).
Purification of the F(abⴕ)2 anti toxic-fraction and prep- RESULTS
aration of peroxidase-labeled F(abⴕ)2. The toxic fraction of
A. mauretanicus mauretanicus crude venom was purified by Characteristics of the ELISA. The detection limit of the
Sephadex G-50 chromatography and tested for its toxic ac- assay was 0.78 ng/ml. Venom antigens were not detected in
tivity by subcutaneous injection into mice. Specific anti-tox- sera from healthy non-envenomed donors. The assay specif-
ic fraction F(ab⬘)2 was purified from scorpion antivenom by ically quantified A. mauretanicus mauretanicus venom and
affinity chromatography on a toxic fraction covalently cou- showed cross-reactivity of the F(ab⬘)2 fragments with the B.
pled to a CNBr-activated Sepharose column, as previously occitanus venom. There was no reactivity when snake ven-
described.24 The F(ab⬘)2-specific toxic fraction (5 mg) was oms were tested (Figure 1). The intra-assay coefficients of
 SCORPION ENVENOMATION AND SEROTHERAPY IN MOROCCO 279

TABLE 1
Epidemiologic features of envenomed patients

Patients (%)
Parameters (n ⫽ 275)

Sex
Male 134 (48.7)
Female 141 (51.3)
Age range (years)
0–14 75 (27.3)
15–29 91 (33.1)
30–44 60 (21.8)
⬎44 49 (17.8)
Scorpion species
Androctonus mauretanicus
mauretanicus 229 (83.3)
Buthus occitanus 38 (13.8)
Not specified 8 (2.9)

FIGURE 1. Specificity of the ELISA used in this study. Values Sting site
are the means of 2 assays performed in duplicate. Upper limbs 142 (51.6)
Lower limbs 119 (43.3)
Other parts of the body 14 (5.1)
variation ranged between 1.4% and 4%. The inter-assay co- Time
efficient of variation was 6.3%. Midnight to before 6:00 AM 22 (8)
Epidemiologic and clinical features of envenomation. 6:00 AM to before noon 82 (29.8)
Epidemiologic data and clinical features of 275 envenomed Noon to before 6:00 PM 34 (12.4)
patients are shown in Table 1. The distribution of scorpion 6:00 PM to before midnight 129 (46.9)
Not specified 8 (2.9)
stings indicates that stings occur more often in the first half
of the evening (46.9%) when scorpions search for prey. The Geographic origin of patients
distribution of scorpion stings by geographic origin of the Rural 203 (73.8)
Urban 69 (25.1)
patients indicates that most scorpion stings occurred in rural Not specified 3 (1.1)
areas (73.8%). The clinical manifestations were extremely
Clinical signs
diverse, with mainly local signs (local pain and a burning
Pain (n ⫽ 217) 185 (85.3)
sensation) and rarely general symptoms (sweating, hyper- Burning sensation (n ⫽ 192) 93 (48.4)
thermia, and shivering). Sweating (n ⫽ 256) 30 (11.7)
Influence of the delay between the sting and hospital Shivering (n ⫽ 258) 11 (4.3)
admission on the severity of envenomation. The delay be- Hyperthermia (n ⫽ 249) 8 (3.2)
tween the sting and the hospital admission ranged from 5 Clinical grade
min to 16 hr (1.27 ⫾ 1.6 hr): it was 5 to ⬎30 min in 33.8% Grade I 247 (90)
of the patients, 30 min to ⬎3 hr in 57.4%, and 3 to 16 hr Grade II 28 (10)
in 8.8%. The percentage of patients living in the city and
arriving at a hospital within 30 min of being stung was sig-
nificantly higher than that of patients living in rural areas (17.82 ⫾ 1.9 ng/ml) (F ⫽ 7.30, P ⫽ 0.007). These results
(42% versus 32%; P ⬍ 0.05). Patients from rural areas were indicate that the clinical severity of envenomations by A.
1.57 times more likely to take more than 30 min to get to a mauretanicus mauretanicus and B. occitanus is closely re-
hospital after the scorpion sting than urban patients (OR ⫽ lated to the concentration of venom in serum of the patient.
1.57, P ⬍ 0.05). The time elapsed after the sting was sig- Efficacy of antivenom in neutralizing venom antigens.
nificantly associated with the geographic origin of the patient The kinetics of venom concentrations in serum during the
(P ⬍ 0.05). hospitalization of grade I patients who were treated com-
When the delay in arriving at the hospital was more than pared with those not treated with antivenom (but treated with
30 min, the percentage of patients diagnosed as grade II was other drugs) are shown in Figure 2. Grade I patients were
higher than patients diagnosed as grade I (36% versus divided into 3 groups: those not treated with antivenom,
14.8%). Patients who arrived at the hospital within 30 min those treated with 2–5 ml of antivenom, and those treated
after being stung were 3.2 times more likely to be diagnosed with 10 ml of antivenom. At admission, the venom concen-
as grade I than those taking more than 30 min (OR ⫽ 3.2, tration was not significantly different (F ⫽ 0.96, P ⫽ 0.38)
P ⬍ 0.05). The severity of clinical grading at hospital ad- between the group not treated with antivenom (18.72 ⫾
mission was influenced by the time elapsed after the sting. 24.28 ng/ml), the group who received 2–5 ml of antivenom
Consequently, the delay between being stung and the hos- (12.10 ⫾ 15.70 ng/ml), and the group who received 10 ml
pital admission may be considered a factor of prognosis. of antivenom (19.35 ⫾ 36.07 ng/ml). Thus, all patients
Association between venom levels in serum and sever- showed higher levels of venom antigens at the hospital ad-
ity of envenomation. Patients classified as grade II had a mission. The venom concentrations were not modified either
significantly higher venom concentrations in their serum 1 or 3 hr later in patients not treated with antivenom. How-
(38.86 ⫾ 10.8 ng/ml) than patients classified as grade I ever, 1 hr after administration of antivenom, venom concen-
280 GHALIM AND OTHERS

FIGURE 2. Time course of serum venom levels in grade I (mild envenoming) patients. Semi-logarithmic plots of serum venom concentrations
over time are shown. Serum samples from grade I patients were collected at 3 time intervals: upon admission to the hospital (time ⫽ 0) and
1 and 3 hr later, as described in the Materials and Methods. The mean serum venom concentration was measured in patients treated with 10
ml of antivenom (n ⫽ 113), in patients treated with 2–5 ml (n ⫽ 42), and in patients not treated with antivenom (n ⫽ 76).

trations decreased significantly in patients who received an- clinical features after antivenom treatment is showed in Fig-
tivenom. When compared with the group not receiving an- ure 3. Local pain and a burning sensation decreased signif-
tivenom (F ⫽ 9.45, P ⫽ 0.0001), the decrease in venom icantly within 3 hr in patients treated with antivenom and
concentration was greater in patients treated with 10 ml of were lowest at 3 hr. General symptoms (sweating, shivering,
antivenom (6.64 ⫾ 11.12 ng/ml) than in patients receiving and hyperthermia) also decreased after antivenom therapy
2–5 ml (9.86 ⫾ 14.86 ng/ml). Three hours after serotherapy, with time. Patients not treated with antivenom and hospital-
the venom concentration was 4.98 ⫾ 15.25 ng/ml in the ized no more than 4 hr had a 2.7 times greater risk of de-
group who received 10 ml of antivenom and 8.81 ⫾ 15.63 veloping severe envenoming than those treated with anti-
ng/ml in the group who received 2–5 ml of antivenom com- venom (OR ⫽ 2.7, P ⬍ 0.001).
pared with 22.20 ⫾ 27.04 ng/ml in the group not receiving Comparison between the effect of antivenom and
antivenom (F ⫽ 8.49, P ⫽ 0.004). The kinetics of venom symptomatic treatment on the evolution of clinical signs
concentrations in serum of grade II patients treated or not at the end of hospitalization. At the end of their hospital-
treated with antivenom was not determined because of the ization, most (62.8%) of the patients not treated with anti-
small number of grade II patients and because the number venom still showed symptoms of envenoming. The absence
of patients decreased over time. of administration of antivenom increased the risk of devel-
Effectiveness of antivenom according to the sting-ad- oping signs of envenomation at the end of the stay in the
mission interval. Table 2 shows that the patients who ar- hospital (OR ⫽ 2.2, P ⬍ 0.05).
rived at the hospital no more than 1 hr after being stung and Factors influencing the evolution of clinical signs at the
did not received antivenom had a 2.67 times greater risk of end of hospitalization. Table 3 shows that the geographic
developing signs of severe envenoming signs than patients origin of patients and the clinical grade at hospital admission
who received antivenom (OR ⫽ 2.67, P ⬍ 0.005). When the were prognosis factors that influenced the evolution of clin-
delay was more than 1 hr, the risk was 3.67 times higher ical signs of envenomation. The antivenom dose was signif-
(OR ⫽ 3.67, P ⬍ 0.005). These observations indicate that icantly associated with a decrease in morbidity.
antivenom therapy is more efficient when given as soon as
possible after the envenomation.
DISCUSSION
Effect of antivenom on clinical signs. The evolution of
The serotherapy approach for scorpion envenoming is
TABLE 2 generally based on clinical observations.14 Other than the
Effectiveness of antivenom according to the sting admission interval work of de Rezende and others26 in patients envenomed by
the scorpion T. serrulatus, no quantitative clinical studies
% of patients % of patients
with without OR*
have been carried out to determine the role of antivenom in
symptoms symptoms P ⬍ 0.005 neutralization of venom in patients stung by scorpions. We
Sting-admission interval ⱕ1 hr report the first clinical study in Morocco designed to assess
With antivenom 27 73 2.67 (1.36–5.25) venom kinetics in patients envenomed by A. mauretanicus
Without antivenom 50 50 mauretanicus and B. occitanus and to study the evolution of
Sting-admission interval ⬎1 hr venom concentration and clinical signs after serotherapy. In
With antivenom 24 76 3.67 (1.55–8.66) this context, we developed a noncompetitive ELISA to quan-
Without antivenom 54 46 tify toxic antigens from Moroccan scorpion venom in the
* OR ⫽ odds ratio. Values in parenthesis are 95% confidence intervals. serum of patients stung by scorpions.
 SCORPION ENVENOMATION AND SEROTHERAPY IN MOROCCO 281

FIGURE 3. Effect of antivenom on the evolution of clinical signs. The relative frequency of patients for each clinical sign over time (upon
admission to the hospital and 1, 2, and 3 hr later) is shown. The number of patients in each treatment group is local pain (n ⫽ 217), burning
sensation (n ⫽ 192), sweating (n ⫽ 256), shivering (n ⫽ 258), and hyperthermia (n ⫽ 249). The statistical significance of each treatment
group is P ⬍ 0.001.

The ELISA we developed was simple, reproducible, and om. It has been reported that acute allergic reactions together
very sensitive. This assay showed that antivenom produced with anaphylactic shock may be caused by crude antivenom
against A. mauretanicus mauretanicus specifically recogniz- preparations.29 The reduction of adverse reactions due to an-
es the venom of this scorpion and cross-reacts with the ven- tivenom has been shown to be due to deletion of the Fc part
om of another scorpion (B. occitanus). This result confirms of the antivenom immunoglobulin.30
previous findings that showed that the antivenom produced This study showed that antivenom was efficient in reduc-
by Institute Pasteur of Morocco is able to neutralize almost ing the levels of circulating venom antigens and decreasing
all venoms of dangerous north African scorpions.27 clinical symptoms. Similarly, the efficacy of antivenom in
To examine the relationship between the severity of scor- neutralizing circulating venom and systemic symptoms was
pion envenoming and serum venom levels, we measured the reported in patients stung by T. serrulatus.26,31
concentration of venom in serum samples of patients upon The efficacy of serotherapy is probably dose-dependent.
hospital admission. The results showed a correlation be- Indeed, the decrease in the circulating venom concentration
tween the severity of envenoming and serum venom levels. was significant after treatment with 10 ml of antivenom.
In envenomed patients classified as grade II, venom concen- Doses less than 10 ml were insufficient for producing good
trations were higher than in envenomed patients classified as clinical improvement. Similar results were reported in rab-
grade I. Our results confirmed those of others reports.4,21,28 bits; these showed that low doses of antivenom could not
Indeed, it has been shown that Brazilian patients with sys- completely negate the electrocardiographic effects caused by
temic manifestations after T. serrulatus envenoming had sig- the venom.9 Thus, administration of high doses of antivenom
nificantly higher plasma venom concentrations than patients was recommended in some countries.15,16 For example, the
with only local pain at the site of the sting.21,28 In envenomed recommended dose in Brazil is 20 ml.16 In Saudi Arabia,
Tunisian patients, a correlation between clinical symptoms higher doses (5–20 ml compared with 0.5–1 ml) reduced the
of envenoming and the level of scorpion venom antigens in mortality rate from an 4–6.8% to less than 0.05%. 3,7,15 Thus,
serum was reported.4 Therefore, the quantification of circu- the ineffectiveness of serotherapy reported by some re-
lating toxin concentrations should be of clinical interest in searchers18,19,32 could be partially explained by the use of low
assessing the severity of envenomation and, consequently, potency and insufficient doses of antivenom. The use of high
improving the management of patients stung by scorpions. potency and adequate doses of antivenom is an important
In this study, in which an antivenom composed of F(ab⬘)2 factor to be considered.
fragments was used, we did not observe any allergic reac- The clinical grade of the patients at hospital admission
tions or anaphylactic shock after administration of antiven- was influenced by the sting-admission delay, which was sig-
282 GHALIM AND OTHERS

TABLE 3 venoming, particularly in rural areas, where the means to

Factors influencing the evolution of clinical signs at the end of hos- treat victims in intensive care units are absent. Neutralization
pitalization of venom by antivenom is intimately dependent on both dose
% of patients % of patients and time of administration of F(ab⬘)2.
with without Odds
symptoms symptoms ratio P

Geographic origin of patient Acknowledgments: We are grateful to the medical and nursing staffs
Urban 50.8 49.2 8.80 0.003 of the different hospitals where clinical data and blood were col-
Rural 30.5 69.5 lected. We thank Drs. Hassan Rich and Noureddine Dersi, and Fat-
ima Chgoury for participation in this study.
Scorpion*
A.m.m. 36.4 63.6 0.62 0.73 Financial support: This study was supported by the Pasteur Institute
B.o. 29.7 70.3 of Morocco.
Sex Authors’ addresses: Noreddine Ghalim, Bouchra El-Hafny, Fatima
Female 39.8 60.2 2.20 0.13 Sebti, Noureddine Lazar, Radia Moustanir, and Abdellah Bensli-
Male 31.1 68.9 mane, Institut Pasteur du Maroc, 1, Place Abou Kacem Ez-Zahraoui,
BP 120, Casablanca, Morocco. Jaafar Heikel, Laboratoire de Bio-
Age statistique et d’Epidémiologie, Faculté de Médecine, Université Has-
Children 38.4 61.6 0.36 0.54 san II, Casablanca, Morocco and Ministère de la Santé, Délégation
Adults 34.4 65.6 Régionale de Casablanca, Casablanca, Morocco.
Sting site
Upper pelvis 28.2 71.8 8.00 0.004 REFERENCES
Lower pelvis 45 55
Clinical grade at admission 1. Goyffon M, Vachon M, Broglio N, 1982. Epidemiological and
Grade I 30.5 69.5 25.84 0.0000 clinical characteristics of the scorpion envenomation in Tu-
Grade II 80.8 19.2 nisia. Toxicon 20: 337–344.
2. Dehesa-Davila M, Possani LD, 1994. Scorpionism and sero-
Sting admission interval
therapy in Mexico. Toxicon 32: 1015–1018.
ⱕ1 hr 35.8 64.2 0.02 0.88 3. Ismail M, 1995. Review article: the scorpion envenoming syn-
⬎1 hr 35 65 drome. Toxicon 33: 825–858.
Antivenom dose 4. Krifi MN, Kharrat H, Zghal K, Abdouli M, Abroug F, Bouch-
10 ml 20.5 79.5 23.29 0.0001 oucha S, Dellagi K, El-Ayeb M, 1998. Development of an
2–5 ml 40.4 59.6 ELISA for the detection of scorpion venom in sera of humans
No antivenom 51.6 48.4 envenomed by Androctonus australis garzonii (AAG) and Bu-
thus occitanus tunetanus (BOT): correlation with clinical se-
* A.m.m. ⫽ Androctonus mauretanicus mauretanicus; B.o ⫽ Buthus occitanus.
verity of envenoming in Tunisia. Toxicon 36: 887–900.
5. Watt D, Simard JM, 1984. Neurotoxic proteins in scorpion ven-
om. J Toxicol Toxin Rev 3: 181–221.
nificantly associated with the geographic origin of the pa- 6. Zerrouk H, Bougis PE, Cdard B, Benslimane A, Martin-Eau-
tient. When the time elapsed between sting and arrival at the claire MF, 1991. Analysis by high-performance liquid chro-
matography of Androctonus mauretanicus mauretanicus
hospital increased, the risk of be classified as a severe
(Black scorpion) venom. Toxicon 29: 951–960.
grade(s) also increases. Thus, the elapsed time is a prognosis 7. El Aminn EO, Berair R, 1995. Piqûres de scorpion chez
factor. The risk of developing signs of envenoming at the l’enfant. Expérience saoudienne. Arch Pediatr 2: 766–773.
end of hospitalization is high when patients arrived later at 8. Ismail M, Abdullah ME, Morad AM, Ageel AM, 1980. Phar-
the hospital. This risk exists also in patients treated with macokinetics of I-125-labeled venom from the scorpion An-
droctonus amoreuxi, AUD and SAV. Toxicon 18: 301–308.
serotherapy. Therefore, the effectiveness of serotherapy de- 9. Ismail M, Fatani AJY, Dabees TT, 1992. Experimental treatment
pends not only on the antivenom dose, but also on the time protocols for scorpion envenomation: a review of common
elapsed between scorpion sting and administration of therapeutics and effect of kallikrein-kinin inhibitors. Toxicon
antivenom. This is consistent with results reported in Saudi 30: 1257–1279.
10. Amaral CFS, de Rezende NA, Freire-Maia L, 1992. Acute pul-
Arabian patients.3 Similar results were reported in mice by
monary edema after Tityus serrulatus scorpion sting in chil-
Revelo and others.33 These investigators showed that sero- dren. Am J Cardiol 71: 242–245.
therapy is particularly effective when antivenom is admin- 11. Gueron M, Stan J, Cohan W, 1967. Severe myocardial damage
istered immediately after venom injection, and that the se- and heart failure in severe scorpion sting. Am Heart J 19:
rotherapy is less effective when administered hours after en- 719–726.
12. Zerrouk H, Mansuelle P, Benslimane A, Rochat H, Martin-Eau-
venomation. However, antivenom must be administered de- claire MF, 1993. Characterization of a new leiurotoxin I-like
spite the long delay between the sting and administration of scorpion toxin P05 from Androctonus mauretanicus maure-
serotherapy. Despite the decrease in the curative effect of tanicus. FEBS Lett 320: 189–192.
antivenom when the sting-serotherapy delay increases, its 13. Freire-Maia L, Campos JA, 1987. Response on the letter to the
editor by Gueron and Ovsyshcher on the treatment of the
preventive effect must still be considered in the treatment of
cardiovascular manifestations on scorpion envenomation.
envenomed patients. Indeed, free F(ab⬘)2 not complexed to Toxicon 25: 125–130.
venom removes toxins fixed on their receptors and prevents 14. Ismail M, 1993. Serotherapy of the scorpion envenoming syn-
their binding.34 In this context, it has been suggested to use drome is irrationally convicted without trial. Toxicon 31:
a large quantity of antivenom when patients arrive later at a 1077–1088.
15. Ismail M, 1994. The treatment of scorpion envenoming syn-
hospital after being stung.2 drome: The Saudi experience with serotherapy. Toxicon 32:
In conclusion, this study shows that serotherapy is the 1019–1026.
most important therapeutic measure to use after scorpion en- 16. Freire-Maia L, Cwnpos JA, Amaral CFS, 1994. Approaches to
 SCORPION ENVENOMATION AND SEROTHERAPY IN MOROCCO 283

the treatment of scorpion envenoming. Toxicon 32: 1009– Diniz CR, Amaral CFS, 1995. Efficacy of antivenom therapy
1014. for neutralizing circulating venom antigens in patients stung
17. Gueron M, Ovsyshcher I, 1987. What is the treatment for the by Tityus serrulatus scorpions. Am J Trop Med Hyg 52: 277–
cardiovascular manifestations of scorpion envenomation? 280.
Toxicon 25: 121–124. 27. Delori P, van Rietschoten J, Rochat H, 1981. Scorpion venoms
18. Gueron M, Ilia R, Sofer S, 1992. The cardiovascular system and neurotoxins: an immunological study. Toxicon 19: 493–
after scorpion envenomation. A review. J Toxicol Clin Toxicol 407.
30: 245–258. 28. de Rezende NA, Chavez-Olorlegui C, Amaral CFS, 1996. Is the
19. Sofer S, Shahak E, Gueron M, 1994. Scorpion envenomation severity of Tityus serrulatus scorpion envenoming related to
and antivenom therapy. J Pediatr 124: 973–978. plasma a venom concentration? Toxicon 34: 820–823.
20. Louis J, 1976. Venins et antivenin du scorpion marocain An- 29. Bond GR, 1992. Antivenin administration of centruroides scor-
droctonus mauretanicus. Me Armees 4: 429–434. pion stings: risks and benefits. Ann Emerg Med 21: 788–791.
21. Chavez-Olortegui C, Fonseca SCG, Campolina D, Amaral CFS, 30. Morais JF, de Freitas MCW, Yamaguchi IK, dos Santos MC,
Diniz CR, 1994. ELISA for the detection of toxic antigens in Dias da Silva W, 1994. Snake antivenoms from hyperimmu-
experimental and clinical envenoming by Tityus serrulatus nized horses. Comparison of the antivenom activity and bio-
scorpion venom. Toxicon 32: 1649–1656. logical properties of their whole IgG and F(ab⬘)2 fragments.
22. de Rezende NA, Dias MB, Cwnpolina D, Chavez-Olortegui C, Toxicon 32: 725–734.
Amaral CFS, 1995. Standardization of an enzyme linked im- 31. de Rezende NA, Amaral CFS, Freire-Maia L, 1998. Immuno-
munosorbent assay (ELISA) for detection circulating toxic therapy for scorpion envenoming in Brazil. Toxicon 36:
venom antigens in patients stung by the scorpion Tityus ser- 1507–1513.
rulatus. Rev Inst Med Trop Sao Paulo 37: 71–74. 32. Gueron M, Margulis G, Ilias R, Sofer S, 1993. The management
23. Pharmacopée Européenne, 1994. Deuxième édition. Conseil de of scorpion envenomation. Toxicon 31: 1071–1076.
l’Europe, ed. Sainte-Ruffine, France: Maisonneuve SA. 33. Revelo MP, Bambirra EA, Ferrera AP, Diniz CR, Chavez-Olor-
24. Avraméas S, Temynck T, 1969. The crosslinking of proteins tegui C, 1996. Body distribution of Tityus serrulatus scorpion
with glutaraldehyde and its use for the preparation of immu- venom in mice and effects of scorpion antivenom. Toxicon
noadsorbents. Immunochemistry 6: 53–66. 34: 1119–1125.
25. Avraméas S, Temynck T, 1971. Peroxidase-labeled antibody and 34. Pépin-Covatta S, Lutsch C, Grandgeorge M, Lang J, Scherr-
Fab conjugate with enhanced intra cellular penetration. Im- mann JM, 1996. Immunoreactivity and pharmacokinetics of
munochemistry 8: 1175–1179. horse anti-scorpion venom F(ab⬘)2-scorpion venom interac-
26. de Rezende NA, Dias MB, Campolina D, Chavez-Olortegul C, tions. Toxicol Appl Pharmacol 141: 272–277.