Fasciotomy worsens the amount of myonecrosis in a porcine model of crotaline envenomation☆☆☆★★★♢

1. Introduction 

Envenomation by North American pit vipers is often associated with progressive swelling and hematologic abnormalities. Significant morbidity is commonly associated with the local effects of an envenomation and may result from tissue necrosis and hemorrhage.1., 2., 3. Crotaline envenomation may lead to increased compartment pressure, which is uncommon, in the affected limb.2., 4., 5. It has been extrapolated from data on crush injuries that prolonged elevation in compartment pressures may lead to tissue necrosis and that surgical decompression is needed to halt this process.6., 7.

In a previous study,8 we demonstrated that Crotalidae polyvalent immune Fab (ovine) (CroFab; FabAV) limited the decrease in perfusion pressure to the affected muscle after an intramuscular crotaline venom injection. Using these same animals in a simultaneous study in a model of intramuscular injection of crotaline venom, our objectives were to evaluate the efficacy of fasciotomy in reducing the amount of myonecrosis, and FabAV in reducing the amount of myonecrosis, at the injection site.

2. Materials and methods 

The study protocol was approved by the US Naval Medical Center Animal Care and Use Committee. The care and handling of animals was in accordance with National Institutes of Health guidelines for ethical animal research.

A porcine model of crotaline venom injection was chosen for this study because swine have previously been used as a model for the human musculoskeletal system. In addition, previous studies have used a swine model to examine the effects of crotaline venom, antivenom, and compartment syndrome.8., 9., 10., 11., 12.

Using methods previously reported,8 20 immature swine weighing 22 to 40 kg were sedated with a 5 mg/kg intramuscular injection of a 1:1 ratio of xylazine (100 mg/kg) and a combination of tiletamine and zolazepam (Telazol: 100 mg/kg). The animals were then endotracheally intubated, and mechanical ventilation was initiated at a rate of 8 breaths/min, with a tidal volume of 15 mL/kg, and adjusted to maintain an end-tidal Pco2 value between 38 and 42 mm Hg. Anesthesia was induced and maintained with 1% isoflurane in 40% nitrous oxide and 60% oxygen, which has been shown to have minimal effect on mean arterial pressure and cardiac output in swine.8., 13.

Central venous access was obtained through the external jugular vein to allow intravenous fluid and FabAV administration. The common carotid artery was cannulated with a 16-gauge catheter and used to measure mean arterial pressure. Pulse rate was continuously monitored by a lead II ECG. Each animal received 15 mL/kg of normal saline solution intravenously over a 30-minute period before baseline variables were obtained, followed by an infusion of normal saline solution at 100 mL/h.

Compartment pressures of the anterior compartments of both hind limbs of each animal were measured continuously with pressure-monitor needles (Stryker Instruments, Kalamazoo, MI) connected to a pressure monitor calibrated (“zeroed”) to the midchest of the animals. The pressure needles were placed at time 0 and left in place until the conclusion of the study.

A 100-mg/mL concentration of venom was prepared from freeze-dried Crotalus atrox venom (Sigma-Aldrich Corporation, St. Louis, MO). Animals were injected intramuscularly with 6 mg/kg of venom (in volumes ranging from 1.3 to 2.4 mL) through a 23-gauge needle into the anterior tibiales muscle of each hind limb at time 0. The dose of antivenom was derived from a dosing curve done before the study to reliably obtain elevation of compartment pressures in this model. The midpoint of the anterior tibiales muscle was marked before injection, and the needle was placed through the skin and muscle until it contacted the tibia and then was withdrawn into the body of the muscle where the venom was injected. Fasciotomy was then performed immediately on the anterior compartment of the right hind limb. Identical fasciotomies were carried out by one investigator on all animals by making a skin incision and then identifying and splitting the fascia along the length of the anterior tibiales muscle with scissors. Compartment pressures were recorded at baseline, time zero, and then hourly until death or the conclusion of the study.

Muscle biopsies of approximately 0.5 cm3 were obtained from the muscle of the right hind limb at 0, 4, and 8 hours (or at death) at similar sites in each animal at points surrounding the injection site. The specimens were excised using a surgical blade, and each specimen was at least 2 cm from the previously excised specimen. A similar biopsy was obtained from the nonfasciotomized limb at the conclusion of the study. In a standardized fashion, biopsies were pinned to wooden tongue blades to minimize contraction artifact and then placed in vials of formalin to be processed.14

One hour after envenomation, animals were randomized to receive either 8 vials of FabAV antivenom in 250 mL of normal saline solution or an equal volume of normal saline solution intravenously over a 15-minute period. Eight vials of antivenom, twice the recommended initial human dose, was chosen to maximize effect. The saline or antivenom solutions were prepared by an independent observer, and the investigators were blinded to the therapy. Animals were monitored until death or the completion of the study at 8 hours. The outline of the study design is illustrated in Figure 1.

View Large Image Download to PowerPoint

Figure 1. Study design, FabAV.

Muscle biopsy specimens were received in the histology laboratory in coded formalin bottles. Each biopsy was sectioned in transverse and longitudinal planes and submitted entirely for processing so that 1 hematoxylin and eosin–stained microscopic slide was prepared from each biopsy. Using a previously described measure of myonecrosis after snake envenomation, the biopsies were graded by deriving a percentage from the ratio of necrotic cells to the total number of cells.15 All specimens were read by one of the authors (GAG), a histopathologist blinded to the sample site and treatment group of the animal. To assess intrarater reliability, biopsies were blindly reevaluated after 4 months, and a correlation coefficient between the readings was derived.

Statistical analyses were performed using Statistica 6.0 (StatSoft, Inc., Tulsa, OK). Repeated-measures analysis of variance was used for the analysis of compartment pressures in each hind limb over time. Rank-order methods (signed-rank, rank-sum, and Friedman test) as specified in the text were used for comparison of myonecrosis between groups. Statistical significance was set at P less than or equal to .05.

3. Results 

The weight (mean±SD) of the animals was 27 kg±4 kg. There was no difference detected between groups for pulse rate or mean arterial pressure.8 Mean compartment pressures responded to fasciotomy as expected (Figure 2). Repeated-measures analysis of variance yielded P values less than .001 for overall treatment differences (between limbs receiving fasciotomy versus those that did not).

View Large Image Download to PowerPoint

Figure 2. Comparison of mean compartment pressures over time. Values are mean with 95% confidence intervals.

Biopsies from the hind limbs that underwent fasciotomy revealed a progressive increase in the amount of myonecrosis over time after venom injection. Data result format is median (25th percentile, 75th percentile). The amount of myonecrosis was as follows: time 0, 0 (0, 0); 4 hours, 14.0% (6.5%, 30.5%); final biopsy (8 hours or death of the animal), 14.5% (6.5%, 25.5%). A Friedman test yielded a P value less than .001. Comparison of the amount of myonecrosis in final biopsies revealed that limbs undergoing fasciotomy had a significant increase in myonecrosis compared with limbs not undergoing fasciotomy (14.5% [6.5%, 25.5%] versus 2.5% [0.0%, 15.0%]; P=.048, signed-rank test]. A histogram of the distribution in the amount of myonecrosis in the final biopsies is listed in Figure 3. The correlation coefficient of agreement for the histopathologist was 0.96, and the average difference was less than 0.25%.

View Large Image Download to PowerPoint

Figure 3. Degree of myonecrosis of hind limbs treated with or without fasciotomy.

When final biopsies from animals that received FabAV were compared with those of animals that received saline solution, no difference was detected in the amount of myonecrosis in either hind limb (fasciotomy or no fasciotomy [10.0% (0.0%, 25.0%) versus 10.0% (5.0%, 21.5%)]; P=.64, rank-sum test).

Four animals died before the conclusion of the study at 2.5, 5, 6, and 7 hours. No animal that received FabAV died during the study period.8 The amount of myonecrosis in the final biopsies done at the death of these animals was not found to differ from that of the animals that survived to the conclusion of the study (fasciotomized limb P=.40, nonfasciotomized limb P=.67; rank-sum test).

4. Discussion 

Pit viper envenomation often causes progressive pain and swelling.1., 2., 3. Because the presentation often mimics the cardinal signs of compartment syndrome (ie, pain, pallor, paresthesias, paralysis, and pulselessness16), compartmental pressure monitoring is frequently done, and there are multiple reports of increased compartment pressures after envenomation.4., 17., 18. Once increased compartmental pressures are documented, it has been argued that a fasciotomy needs to be performed before there is significant myonecrosis resulting from acute compartment syndrome.6., 7. However, although human crotaline envenomation may mimic this syndrome, the pathophysiology remains quite distinct. Snake venom includes a mixture of proteins and digestive enzymes that lead to hematologic abnormalities, capillary leakage, and direct myonecrosis.2., 3., 19.

In our study, although fasciotomy prevented an elevation in compartment pressures over time, it significantly worsened the amount of local myonecrosis seen on histologic samples compared with samples from a nonfasciotomized limb. This result is in concordance with previous animal studies that have demonstrated that crotaline venom directly induced tissue necrosis independently of elevated compartment pressures and that fasciotomy had no effect on abating the local tissue destruction or improving limb function or survival.7., 20., 21. According to this evidence, we feel that fasciotomy may not have a therapeutic role in the treatment of rattlesnake envenomation in human beings.

We failed to detect a change in the amount of myonecrosis at the site of venom injection when animals were treated with FabAV. Previous murine studies have yielded various results in the ability of polyvalent antivenom to reduce myonecrosis and indicate that a significant amount of antivenom may be needed to achieve good results.22., 23. In our study, we used a moderate amount of antivenom despite a large injection of venom. Because of the limited number of animals, additional vials of FabAV may have been needed for us to detect a difference in the amount of myonecrosis. In human beings, it has been reported in postmarketing analysis that up to 47 vials of FabAV have been used to obtain control after a crotaline envenomation.24

There are multiple limitations to this study, including the possibility that our animal model might not reproduce the human effects of rattlesnake envenomation. We chose the swine model because it has been used in previous investigations of crotaline envenomation.8., 11., 12. Observation of the animals was limited to 8 hours, and it is possible that further differences may have been detected if we had observed the animals for a longer period. In particular, the study period was not designed to detect whether the increased myonecrosis seen had any effect on subsequent recovery or limb function.

Another limitation is the total amount of venom used. The total amount of venom averaged 324 mg per animal (12 mg/kg, average weight 27 kg). It has been reported that rattlesnakes inject approximately 210 mg of venom with a large defensive strike.18., 25. Each hind limb therefore received less than a large defensive strike.

The approach to fasciotomy in our study was also substantially different from the therapeutic one used for human crotaline envenomation in which fasciotomy is performed after sustained elevation of compartment pressure. In performing the fasciotomy immediately after venom injection, we sought to maximize its beneficial effect by blunting the elevation of compartment pressure. It is possible that by doing so, we initiated a cascade of effects on the cellular level that worsened the myonecrosis normally seen in crotaline envenomation. However, we believe that if these effects occur, they would occur whether the fasciotomy were performed early or delayed until after the elevation of compartment pressures. It is also possible that the amount of myonecrosis from the fasciotomized limb was exacerbated by the excision of the biopsies, resulting in limited blood supply to the remaining tissue. To limit the potential damage done by the surgery, consecutive biopsies were obtained from sites that were distinct from one another.

In summary, fasciotomy significantly worsens the amount of myonecrosis seen on biopsy in a porcine model of intramuscular crotaline venom injection up to 8 hours after envenomation. No change in the amount of myonecrosis was detected with the use of FabAV treatment at the dosages used in this animal model.