A comparison of serum antivenom concentrations after intravenous and intramuscular administration of redback (widow) spider antivenom

Abstract

AIMS

There are no studies measuring antivenom concentrations following intramuscular administration. This study aimed to compare antivenom concentrations following intravenous and intramuscular administration of redback spider antivenom (RBSAV).

METHODS

Twenty patients recruited to a controlled trial comparing intramuscular and intravenous administration of antivenom had serial blood samples collected at 30 min intervals for 2 h after the administration of one or two doses of antivenom. Antivenom concentration was measured using an enzyme immunoassay.

RESULTS

Ten patients received intramuscular antivenom but antivenom could not be detected in serum after either one or two vials, at any time point. The median time of the final sample after commencement of antivenom treatment in these patients was 3.2 h (1.8–5 h). Ten patients received intravenous antivenom (three one vial and seven two or more vials) and antivenom was detected in all patients.

CONCLUSIONS

RBS AV given by the intramuscular route is unlikely to be effective in the treatment of redback (widow) spider bite.

WHAT IS ALREADY KNOWN ABOUT THIS SUBJECT

• Widow spider antivenoms, including redback spider antivenom, are often given by the intramuscular route.

• No studies have measured widow spider antivenom following intramuscular or intravenous antivenom.

WHAT THIS STUDY ADDS

• Intramuscular redback spider antivenom is not detectable in serum for at least 3–5 h after treatment. Intravenous antivenom is detectable 30 min after intravenous infusion.

• Intramuscular antivenom may not be an effective administration route.

Introduction

Latrodectism is an important medical condition resulting from widow spider bite [1, 2]. Latrodectus spp. or widow spiders occur on all inhabited continents and result in thousands of bites annually [3]. At least five antivenoms are produced to various widow spider species [4]. However, there is significant controversy about the efficacy, safety and route of administration of widow spider antivenom [4–7].

Unlike snake antivenoms which are almost always used by the intravenous route, widow spider antivenoms have been commonly used by the intramuscular route. In Australia it is recommended by the manufacturer that redback spider antivenom (RBS AV) be given by the intramuscular route; only recently has this been questioned and investigated [6, 8]. Similarly in Brazil the commonest route of administration for black widow antivenom has been the intramuscular route [9]. There is little information on the pharmacokinetics of intramuscular antivenom use and animal studies of snake and scorpion antivenoms show that intramuscular antivenom has delayed and only partial absorption [10, 11].

With widespread use of intramuscular antivenom for widow spider envenoming on at least two continents it is important to investigate whether intramuscular antivenom is absorbed. As part of an ongoing randomized controlled trial of intravenous vs. intramuscular antivenom (Redback spider AntiVenom Evaluation study – RAVE) we measured antivenom concentrations in a subgroup of patients.

Methods

This was an analysis of a subgroup of patients involved in the RAVE study who had serial blood samples collected following the administration of RBS AV. The RAVE study is a double dummy placebo randomized controlled trial of intravenous administration of RBS AV vs. intramuscular administration. Here we only describe timed antivenom concentration data from a subgroup recruited to two of the investigation sites.

Study patients

Patients were recruited from the emergency departments of two major tertiary hospitals, Newcastle Mater Misericordiae Hospital and the John Hunter Hospital in Newcastle, Australia, from January 2003 to September 2006. In addition to the randomized controlled trial, the collection of additional blood samples for the measurement of antivenom concentrations was approved by the Hunter Research Ethics Committee (University of Newcastle) and the Hunter Area Research Ethics Committee for the two sites involved. All patients gave written and informed consent including for repeated blood sampling.

Patients were eligible for inclusion if they had a definite redback spider bite or a clinical syndrome consistent with latrodectism that the treating clinician elected to treat with antivenom. Exclusion criteria were children less than 8 years of age, pregnancy, delay of more than 24 h from bite to treatment and previous hypersensitivity reactions to antivenom.

Patients were randomized to receive either intravenous or intramuscular RBS AV. The patients, treating doctors, investigators and laboratory were blinded to the route of administration. After 16 patients had been recruited for the measurement of antivenom concentration and after discussion between the investigators during a second interim analysis of the main study, the samples were analyzed and allocation unblinded for these patients.

The Newcastle Mater Misericordiae Hospital pharmacy in conjunction with Pharmaland Pty Ltd produced prepacked kits for the trial. Vials of 250 units of RBS AV were purchased from CSL Ltd and Pharmaland provided identical vials containing normal saline. Each pack contained two vials, each vial randomly assigned to either IM Prep containing RBS AV and IV Prep containing normal saline, or IV Prep containing RBS AV and IM Prep containing normal saline. To allow a second treatment by the same route of administration each prepacked kit contained a second pack with two vials that were identical to the first pack.

After patients had given informed consent they were given both an intramuscular injection of one vial and an intravenous infusion of the other vial made up in 200 ml of normal saline over 20 min. After 2 h the patient was reassessed and a decision was made by the treating doctor as to whether a second vial of antivenom was required. A second study treatment with identical allocation was then given if required.

In addition to clinical data collection, the subgroup of patients reported here had blood samples collected prior to the administration of antivenom, then 30 min, 60 min, 90 min and 120 min after completion of the intravenous infusion. If a further dose was administered then a further four samples were collected after the second dose. It was intended that the second dose be given 2 h after the completion of the first but delays were expected due to clinical decision making.

Sample analysis

For quantification of antivenom in serum an enzyme immunoassay (EIA) was developed based on similar assays for snake antivenom [12, 13]. Redback spiders were collected from three different geographical regions of Australia. Spiders were frozen and the venom glands were dissected out under a light microscope using a previously developed method [14]. The venom glands from 35 spiders were ruptured using a pestle in distilled water and centrifuged (5000 rev min⁻¹ for 2 min) to remove debris. The supernatant was removed, re-centrifuged and the remaining supernatant freeze dried, weighed and stored at −80°C. RBS AV is a horse derived F(ab′)₂ antivenom and was donated by CSL Ltd, Parkville, Victoria, Australia and used for standards. Each vial is a liquid that contains 250 units of antivenom activity. However, the mass of F(ab′)₂ fragments in each vial differs for each batch of antivenoms and is indicated on the label. The variation in mass between batches of vials meant that the standard curves were constructed using Units of activity. A horseradish peroxidase conjugate of antihorse IgG, tetramethylbenzidine (TMB) and polyoxyethylene-sorbitan monolaurate (TWEEN 20) were purchased from Sigma. Greiner Microlon 96-well high-binding plates were used and read at 450 nm on a Bio-rad Microplate Manager. All procedures were carried out at room temperature.

The freeze dried venom gland extract was reconstituted in phosphate buffer saline (PBS) to a concentration of 1 mg ml⁻¹ and this solution was aliquoted and stored at −20°C. The plate wells were coated with a solution of redback spider venom gland extract, 5 µg ml⁻¹ in sodium bicarbonate buffer (50 mm pH 9.5) for 1 h at room temperature then at 4°C overnight. The wells were then washed three times with PBS containing 0.02% TWEEN 20, and 300 µl of a blocking solution of 1% bovine serum albumin was applied. After 1 h the plates were washed again, and 100 µl of sample solution as a 20% dilution in PBS was applied. The plates were allowed to stand for 1 h and then washed. Anti-Horse IgG-horse radish peroxidase conjugate (100 µl) at a concentration of 500 ng ml⁻¹ in blocking solution was then applied. After standing a further 1 h the plates were washed again and 100 µl of TMB applied. Colour was allowed to develop for 10 min and the reaction was stopped by the addition of 50 µl 1 m H₂SO₄.

A sample background absorbance was determined by applying a sample to wells which had been coated with blocking solution only. This background absorbance was subtracted from the measured absorbance in each sample. Samples from one patient could not be used because of very high background absorbance outside the range of the standard curves. All samples were tested in duplicate, with duplicate wells having a coefficient of variation of <10% for low and high absorption. A quality control sample was included on each plate and the maximum variation from the known concentration was 17.5%.

One unit (1 U) of antivenom activity is defined as the ability to neutralize 0.01 mg of venom from the species of venomous creature against which the antivenom has been raised [15]. Standard curves for antivenom in serum were made with a range of concentrations from 4.4 to 280 U l⁻¹. Serum concentrations were interpolated from the standard curve using a three parameter logistic regression in PRISM 4. The minimum level of detection corresponds to 5 U l⁻¹ in undiluted serum. Pre-antivenom samples from the 20 patients where assayed to determine the background in normal samples: the mean value was 6.5 (95% range 0–14.8 U l⁻¹). Therefore 15 U l⁻¹ was taken as the cut-off value for antivenom or the limit of quantification.

Results

Twenty patients were included with a median age of 39 years (interquarile range 27–49 years) and 10 were female. Ten patients received intramuscular antivenom (five got one vial and five got two or more vials) and 10 received intravenous antivenom (three got one vial and seven got two or more vials). The median dose of antivenom was two vials (range: 1–4).

Following the intramuscular administration of RBS AV, no antivenom could be detected in serum after either one or two vials, at any time point. The median time of the final sample after commencement of antivenom treatment in these patients was 3.2 h (1.8–5 h). Antivenom was detected in all patients following intravenous RBS AV and the time course of the serum concentrations is shown in Figure 1.

Figure 1.

Timed serum concentrations for 10 patients receiving intravenous redback spider antivenom (RBS AV): three patients received one vial (thick line) and seven received two or more vials (thin line). Time zero was the commencement of the first antivenom infusion and the infusion was given over 20 min. Infusion of a second vial began approximately 2.5 h after the first vial

Discussion

We found that RBS AV was not detectable in serum after intramuscular administration for at least 5 h after administration of the first vial of antivenom, whereas intravenous antivenom was detectable in the first sample 30 min after completing the infusion. These findings are consistent with previous human and animal studies of snake and scorpion antivenoms administered intramuscularly [10, 16, 17], and show that intramuscular antivenom does not reach the systemic circulation and so may not be effective.

No previous studies have measured antivenom (Fab, F(ab′)₂ or IgG) concentration in humans after intramuscular administration [18]. One previous study in children measured venom concentration after intramuscular scorpion antivenom. This study demonstrated that intramuscular antivenom alone did not cause a significant reduction in detectable free venom compared with intravenous antivenom which resulted in venom becoming rapidly undetectable [19]. Another study from Morocco also measured venom concentrations following intramuscular antivenom. Again there was no convincing decrease in venom concentrations in the 4 h period after antivenom administration compared with no treatment [20].

Previous animal studies have shown that intramuscular antivenom has little effect on venom concentrations [10, 16]. Antivenom has been measured in rabbits following intramuscular administration [18]. In these rabbit studies of intramuscular antivenom administration, F(ab′)₂ and IgG had a poor bioavailability (36–42%) and delayed time to peak concentration of 48–96 h [10, 18]. This is consistent with our findings that intramuscular antivenom is not going to be detectable in sufficient concentrations within the first few hours of administration.

For antivenom to be effective it must either reach the site of venom action or remain at high concentrations in the systemic circulation to trap venom components, so they can then be eliminated from the body. RBS AV consists of horse F(ab′)₂ fragments which are unlikely to distribute rapidly to peripheral sites where redback spider venom appears to act [18, 21]. Therefore it is likely that high systemic antivenom concentrations, well within the range of the assay that we have developed, are required for any clinical effect.

There is ongoing controversy over the effectiveness of RBS AV via the intramuscular route. One small study of 33 patients found no difference in the primary endpoint between RBS AV given by intramuscular vs. intravenous routes [6]. The clinical results of our larger study are not yet available. As well as investigating the effectiveness of intramuscular antivenom in clinical trials, pharmacokinetic studies are required to demonstrate plausible mechanisms for observed clinical findings.

Blood sampling was only continued until 2 h after the administration of the first or second vial of antivenom, depending on whether the patient received a second vial or not. Therefore delayed appearance of antivenom in the serum would be missed by the sampling design in this study. However, for optimal clinical utility RBS AV needs to be absorbed rapidly into the systemic circulation so that the symptoms and signs of envenoming begin to resolve within 1–2 h of administration. Previous studies of intramuscular antivenom have shown a significant delay in intramuscular absorption [18].

High background absorbance that occurred in one patient limits the specificity of this assay if it was used as a diagnostic test. High background absorbance is commonly encountered in EIAs and is immunological in nature [22]. It is due to naturally occurring antibodies in sera that bind to antibodies from other species. These are not present in all sera and therefore cause this unpredictable high background in some patients [22]. This is unlikely to be a problem for our use of the assay except that this patient had to be excluded from the analysis. The assay is not used diagnostically to determine if the patient has been given redback spider antivenom so specificity is less important. Specificity is far more important for assays used to detect venom in patient sera.

In conclusion, RBS AV is not detectable by enzyme immunoassay in the 3–5-h period following the administration of antivenom by the intramuscular route compared to intravenous administration. This finding indicates that RBS AV given by the intramuscular route is unlikely to be as rapidly clinically effective as the intravenous route.