Acute Toxicity Pilot Evaluation of Proliferol in Rats and Swine

Abstract

Proliferol is an investigational new drug containing lidocaine hydrochloride 0.25%, dextrose 12.5%, glycerin 12.5%, and phenol 1.0% in aqueous solution. Despite extensive previous experience with similar drug solutions administered in humans by intraligamentous injection for chronic musculoskeletal conditions for over 50 years, animal toxicity data are unavailable. A pilot study was conducted to assess acute toxic effects prior to undertaking further assessment of this drug. Test animals were four Sprague-Dawley rats and four Yucatan mini-swine. Rats received injections into lumbar paraspinal muscles, whereas swine received injections into lumbosacral ligaments in an attempt to mirror the method of administration in humans. Two doses were studied equivalent to 1× and 5× the typical human dose. Outcomes measured at 24 h and 14 days included clinical observations, clinical chemistry, hematology, urinalysis, local tolerance, and major organ histopathology. In rats and swine, results from clinical chemistry, hematology, and urinalysis were indicative of acute local inflammation. At the high dose, marked (rats) and moderate (swine) short-term above-normal levels in certain liver enzymes were noted. In rats and swine, local tolerance results were indicative of acute local inflammatory changes in the skin, subcutis, and muscle around the injection sites. In rats and swine, major organ histopathology results did not reveal lesions attributable to the drug and clinical observations were within normal limits. In swine, fibroplasia was noted in deeper muscle tissues after 14 days. Injections of Proliferol in lumbar paraspinal muscles in rats and lumbosacral ligaments in swine elicited a modest acute local inflammatory response with no other indications of local or systemic toxicity.

The use of complementary and alternative medical (CAM) therapies is rapidly increasing, prompting additional research efforts to determine their safety and efficacy (Eisenberg et al. 1998; Wolsko et al. 2002). In the United States, any substance administered to humans by injection for medicinal purposes is considered a drug ( Federal Food, Drug, and Cosmetic Act 2004). Research into injectable CAM therapies must therefore conform to drug research regulations, which require animal toxicity data prior to approving human research. However, many CAM therapies have extensive human clinical data but no animal toxicity data. One example of this is prolotherapy, a CAM treatment that involves intraligamentous injection of chemical irritants. Prolotherapy involves precipitating controlled localized acute inflammation to stimulate fibroblastic proliferation, collagen formation, and connective tissue repair (Banks 1991). This is accomplished through repeated injection of compounds such as hypertonic dextrose and glycerin. It is most commonly prescribed for chronic musculoskeletal and spinal pain due to suspected ligamentous injury. Prolotherapy has been used in the United States for over 60 years for a variety of chronic musculoskeletal conditions, and was adapted from sclerosing injections used in conventional medicine to treat connective tissue defects such as varicose veins and hernias (Dagenais, Haldeman, and Wooley 2005; Harris and White 1936).

Various drug solutions have been used in prolotherapy; the most common is a combination of dextrose, glycerin, phenol, and lidocaine (Dorman 1993). According to a recent systematic review, drugs used in prolotherapy that combine dextrose, glycerin, phenol, and a local anesthetic appear most promising for future research, as injections of dextrose alone have failed to demonstrate efficacy (Dagenais, Haldeman, and Wooley 2005). One combination, recently named Proliferol, contains lidocaine hydrochloride 0.25%, dextrose 12.5%, glycerin 12.5%, and phenol 1.0% in aqueous solution. Though not currently approved by the Food and Drug Administration (FDA), this type of drug solution is available from compound pharmacies and may be used clinically by practitioners who deem it appropriate for their patients (Allen 2001). Although no studies have been conducted to examine its pharmacological properties, numerous clinical studies have reported promising results with this type of drug solution for spinal pain. These studies are summarized in a recent systematic review by some of the authors of the present study (SD, SH, JRW) (Dagenais, Haldeman, and Wooley 2005). The indication that will be targeted in the future for Proliferol is chronic low back pain, for which no treatment—whether from conventional medicine or CAM—has demonstrated convincing long-term effectiveness (Dagenais, Haldeman, and Wooley 2005).

Despite extensive prior human use with this type of drug solution, the FDA determined that animal toxicity studies are required prior to undertaking further clinical research. In order to support eventual clinical studies of Proliferol in humans, the acute local and systemic toxicity needs to be assessed in one rodent and one nonrodent species, according to ICH guidelines on preclinical assessment of new drugs. Given the potential for adverse events associated with intraligamentous spinal injections, animal studies should attempt to mirror this method of administration to isolate adverse events associated with the drug solution from those associated with its method of administration. As no data could be located regarding this novel injection method, a pilot study was conducted. The primary objective of this study was to assess the feasibility of experimental methods for future animal studies involving intraligamentous administration of Proliferol. The secondary objective was to conduct a preliminary assessment of acute toxicity of Proliferol in one rodent and one nonrodent species.

MATERIALS AND METHOD

Animal Welfare Statement

This study was approved by the Institutional Animal Care and Use Committee (IACUC).

Animal Care and Handling

The study was conducted at the Biological Test Center (Irvine, California). Nonrodents were four female Yucatan mini-swine (S&S, Ranchita, California), at least 6 months old and weighing 20 to 30 kg on day 1. Rodents were four male Sprague-Dawley rats (Harlan Sprague-Dawley, San Diego, California), at least 12 weeks old and weighing 350 to 400 g on day 1. Given the complexity of intraligamentous injections, rats and swine were selected for their size. Because the primary objective of the study was to test injection methods in animals for the first time, few animals were needed. Because no difficulties were encountered during the injection process, there was no need to use more animals.

Animals were examined, quarantined, and reexamined prior to the study. Swine were housed in stainless steel pens with raised flooring and fed mini-swine breeder diet 7037 (Harlan Teklad, Madison, Wisconsin). Rats were housed in hanging stainless steel cages with wire mesh floors and fed a global 18% protein rodent diet (Harlan Teklad, Madison, Wisconsin). Animals accessed tap water ad libitum.

Drug Preparation and Testing

Proliferol was prepared by a compound pharmacy (Park West Pharmacy, Irvine, California) using ingredients meeting United States Pharmacopeia (USP) requirements, including lidocaine hydrochloride anhydrous, dextrose anhydrous, glycerin, phenol fused crystal, and sterile water for injection. Drug identity, strength, and sterility were assessed by methods adapted from the USP (USP 2003).

Low and High Doses

The doses used in this study were based on the human dose on a drug volume per body weight basis (ml/kg). Because surface area is not a good indicator of spine size, it was not used to determine dose. Clinical studies with drugs similar to Proliferol for chronic low back pain in humans have used doses ranging from 10 to 30 ml per treatment (Dagenais, Haldeman, and Wooley 2005). The upper limit of this range was selected as the basis for establishing the 1× low dose used in this study, equivalent to 30 ml in a 70-kg human. The high dose was set at 5× the low dose, equivalent to 150 ml in a 70-kg human.

Drug Administration

On day 1, swine were premedicated with intramuscular or subcutaneous glycopyrrolate (0.01 to 0.02 mg/kg), anesthetized with intramuscular Telazol (4 to 6 mg/kg) and xylazine (2.2 mg/kg), intubated, and maintained in anesthesia with isoflurane. Intravenous lactated Ringer’s solution was administered to support blood pressure. While in the prone position, the lumbosacral area was clipped and disinfected with sequential betadine scrubs and 70% isopropyl alcohol, and covered with sterile surgical draping.

To mimic the intraligamentous injection in humans, a medical physician (Robert Green) assisted the veterinarian (Marit Ness-Piacente) with drug administration, using occasional C-arm radiological guidance to confirm needle placement. Injections were given in the lumbar spine and sacroiliac joints with a 20-gauge 2½-inch spinal needle. For the interspinous and supraspinous ligaments, the needle was inserted in interspinous spaces in the lumbar spine at 45° cephalad until contact was made with the inferior surface of the spinous process. The plunger was then withdrawn to confirm the absence of blood in the syringe that could indicate vessel puncture. A drug aliquot was delivered near the ligament-bone junctions. The needle was then redirected 45° caudal until contact was made with superior surface of the spinous process, and a drug aliquot was delivered in a similar manner. For the iliolumbar and sacroiliac ligaments, the needle was inserted medial to the sacroiliac joints at 45° lateral until contact was made with the top of the sacrum (Figure 1). Again, the needle was partially withdrawn and a drug aliquot was delivered near the ligament-bone junction. In high-dose animals, injections were extended cephalad into the lower thoracic interspinous and supraspinous ligaments to administer a larger volume of drug solution.

On day 1, rats were anesthetized with intramuscular ketamine hydrochloride (40 to 90 mg/kg) and xylazine (5 to 10 mg/kg). The lumbosacral area was clipped and disinfected with sequential betadine scrubs and 70% isopropyl alcohol. While in the prone position, multiple intramuscular injections were given bilaterally along the lumbar spine approximately 1 cm from the midline using a 1-inch 23-gauge needle until the target dose was administered. In high-dose animals, injections were extended cephalad into the lower thoracic spine area.

Necropsy

Necropsies were conducted on day 2 for four animals (i.e., 24-h low-dose swine, 24-h low-dose rat, 24-h high-dose swine, and 24-h high-dose rat) and day 15 for the other four animals (i.e., 14-day low-dose swine, 14-day low-dose rat, 14-day high-dose swine, and 14-day high-dose rat). These time points were based on standard methodology for expanded acute toxicity studies and our on discussions with the FDA regarding the protocol. Rats were euthanized by exsanguination, whereas swine were euthanized by intravenous Euthasol (Delmarva Laboratories, Midlothian, Virginia). Samples were obtained from the major organs (heart, lungs, liver, kidneys, lymph nodes) and injection site (skin, subcutis, muscle) and placed in 10% formalin. At day 15 only, samples were also obtained from control skin and subcutis away from the injection sites.

Outcome Measures

Acute toxicity was assessed by clinical observations, clinical chemistry, hematology, and histopathology. Local tolerance was assessed by injection site appearance and histopathology. In swine, blood for clinical chemistry and hematology was collected from the cranial vena cava prior to drug administration on day 1. In rats, blood for clinical chemistry and hematology was collected via heart puncture at necropsy. Hematology samples (approximately 1 ml each) were collected in vacuum tubes containing EDTA. Coagulation samples were collected in vacuum tubes containing sodium citrate. Blood samples were analyzed by Laboratory Corporation of America (San Diego, California). Normal reference ranges were obtained from the testing laboratory, as well as the literature (Swindle 1998; Swindle et al. 2003). In swine, urine was collected from the bladder at necropsy. In rats, urine was also collected with metabolic cages for 24 hours prior to dosing and prior to necropsy. Levels of C-reactive protein (CRP) in urine were determined by quantitative enzyme-linked immunosorbent assay (ELISA) rat CRP (Hilica, Fullerton, CA). Assay sensitivity was 2.5 ng/ml. CRP was selected based on its use as a marker of acute inflammatory changes in humans. A board-certified veterinary pathologist (Roland Gunther) at Colorado Histo-Prep (Fort Collins, CO) conducted the histopathological evaluation of the tissues collected at necropsy.

Statistical Analysis

Because this was a pilot study with only one animal per group, no statistical analysis was performed.

RESULTS

Clinical Observations

All clinical observations (i.e., food consumption, general appearance, injection site appearance, stool appearance) for swine and rats were within normal limits. Slight elevations in body weight (compared with baseline) were observed in swine and rats necropsied on day 15.

Clinical Chemistry

Baseline clinical chemistry for swine and rats was normal. Clinical chemistry following drug administration is presented in Table 1 (swine) and Table 2 (rats). Values slightly outside the normal reference range included 24-h low-dose swine—above normal triglycerides; 24-h high-dose swine—above normal alanine aminotransferase (ALT), aspartate aminotransferase (AST), and triglycerides; 14-day high-dose swine—above normal ALT. For rats, findings outside the normal reference range included: 24-h low-dose rat—above normal albumin and AST; 24-h high-dose rat—above normal albumin, ALT, and AST; 14-day high-dose rat—above normal AST.

Hematology

Baseline hematology for swine and rats was essentially normal, though blood smears revealed the following: 24-hour low-dose swine—slight hemolysis; 24-hour high-dose swine—above normal platelet count with marked clumps; 14-day high-dose swine—marked platelet clumps and anisocytosis. Hematology following drug administration is presented in Table 3 (swine) and Table 4 (rats). For swine, values outside the normal reference range included 14-day low-dose swine—above normal neutrophil count and slight hemolysis; 24-h high-dose swine—above normal platelet count, slight hemolysis, marked platelet clumps, and giant platelets present; 14-day high-dose swine—mild platelet clumps. For rats, findings outside the normal reference range included 24-h low-dose rat—above normal white blood cell, lymphocyte, and neutrophil counts, and slight hemolysis; 14-day low-dose rat—above normal white blood cell count and marked platelet clumps; 24-h high-dose rat—above normal white blood cell and neutrophil count; 14-day high-dose rat—above normal white blood cell count.

Histopathology

Histopathology findings are summarized in Table 5 (swine) and Table 6 (rats). For swine, there were no lesions in the heart, liver, or lymph nodes. For swine, chronic, diffuse, mild, nonsuppurative pneumonia of unknown etiology was noted in all animals. The 14-day low-dose swine had acute, disseminated, mild, suppurative glomerulonephritis of unknown etiology. Injection site skin, subcutis, and muscle (superficial) revealed mild hemorrhage and acute necrosis in the 24-h animals, as well as steatitis and fibroplasia in the 14-day high-dose animal. Control skin revealed no lesions. Lumbar and sacroiliac muscle (deep) revealed moderate to severe necrosis in all animals, mild to moderate fibrosis in two animals, and areas of fibroplasia with multinucleated cells in both 14-day animals. For both the low-dose and high-dose animals, the predominant findings were acute inflammatory changes and necrosis after 24 h, followed by fibroplasia after 14 days (see Figures 2 and 3).

For rats, there were no lesions in the heart, lungs, kidneys, or lymph nodes. Chronic, focal, mild, granulomatous hepatitis of unknown etiology was noted in the 24-hour low-dose rat. According to the study pathologist, this finding was likely incidental, though the limited number of animals and absence of controls precludes a definitive conclusion. Injection site skin, subcutis, and muscle (superficial) in the 24-h animals revealed minimal to moderate fasciitis, mild hemorrhage, and mild to moderate necrosis. Control skin, subcutis, and muscle did not reveal any lesions. For both the low-dose and high-dose animals, the acute inflammatory changes observed after 24 h were no longer apparent at 14 days, at which time tissues appeared normal (see Figure 4).

Urinalysis

C-reactive protein levels for rats are presented in Table 7. Each measurement was repeated once and the mean was taken. Baseline levels varied from 0.030 to 0.054 μg/ml, and levels prior to necropsy varied from 0.030 to 0.049 μg/ml, declining between 0% and 8.4%.

DISCUSSION

The primary objective of this pilot study was to assess the feasibility of experimental methods for future Proliferol toxicity studies in animals, attempting to mirror the method of administration currently used in humans. In this study, intraligamentous injection was deemed feasible in sexually mature Yucatan mini-swine based on the size and shape of their spine; no difficulties were encountered during drug administration. Although rats are among the larger rodents, intraligamentous spinal injections were not deemed feasible due to the size of the rats. As an alternative, intramuscular injections were given in the lumbosacral area. For purposes of determining acute systemic toxicity, this route of administration was deemed an appropriate surrogate.

Administration of the high-dose equivalent to 150 ml in a 70-kg human was deemed feasible in both swine and rats. The 5 × margin of safety provided by the high dose would not likely be raised in future studies due to the risk of severe tissue injury to the animals with an increased volume of injections, as well as tissue resistance preventing the injection of additional drug solution. Altering the concentration of the ingredients in Proliferol could affect the ingredients’ pharmacological properties and is not recommended. During necropsies in the 24-h swine it was difficult to determine which tissues should be harvested at the injection sites. Although the skin, subcutis, and superficial muscle at the injections sites were readily identified from skin and needle markings, there were no apparent visual indications of where the injections had been given to deeper spinal tissues. To avoid these difficulties in future studies, the histopathologist could witness the injection process to help determine the tissues of interest. A nontoxic, short-lived dye mixed could also be used to identify deeper injection sites.

Interpreting data from pilot studies is problematic due to small sample sizes and lack of a control group. Nevertheless, it is interesting to examine study findings in light of what is currently known or hypothesized about this drug solution based on clinical experience in humans. A temporary elevation in pain and stiffness is commonly noted following injection of this drug solution in a similar fashion (Dagenais et al. 2005). In the present study, no overt pain behavior (i.e., vocalization, movement, bowel and bladder function, food and water consumption) was observed in animals, even in the high-dose groups. This seemed to indicate that administration of this drug solution by spinal intramuscular (rats) and intraligamentous (swine) injection was well tolerated, though additional pain outcome measures will be considered in future studies.

Interpretation of the clinical chemistry and hematology must occur within the context of the injection procedure. In the present study, the above normal level of AST may be indicative of hepatotoxicity. However, given that administration involved multiple intramuscular or intraligamentous injections with large needles and a large bolus of drug solution, acute trauma to skeletal muscle is a more likely etiology for this observation. This hypothesis is supported by the above-normal level of AST being greater at 24 h than 14 days and more pronounced in the high-dose than low-dose animals. It may be worthwhile to devise control groups in future studies such that the effects of these two factors (i.e., number of injections and volume of drug injected) on AST levels could be evaluated independently. Measurement of additional liver enzymes (e.g., lactate dehydrogenase, gamma-glutamyltransferase) and skeletal muscle enzymes (e.g., muscle creatine phosphokinase) may also help to differentiate potential hepatotoxicity from acute inflammatory reaction to the drug administration process.

The above normal level of ALT observed in the 24-h high-dose animals may also be indicative of hepatotoxicity. However, this elevation was temporary, did not result in any histopathologic evidence of lesions to the liver attributable to the drug solution, and returned to normal after 14 days. Given the drug solution’s phenol content, the most likely explanation of this observation is that administration of 5× the maximum previously reported clinical human dose resulted in temporary hepatic insult without alteration of hepatic function or decreased animal well-being as noted by clinical observations. Because no such results were noted in low-dose animals, the currently used human clinical dose is not likely to pose a risk for hepatotoxicity. This latter point will be one of the objectives of additional pre-clinical as well as early clinical studies of Proliferol.

Although ALT and AST may be good indicators of liver damage in rats, they may not be helpful in swine. The ALT and AST findings of the present study may not be a species difference issue related to liver damage. Most likely, however, these findings are associated with skeletal muscle breakdown from the trauma of the muscular injection, because both ALT and AST have muscle components.

The above normal white blood cell and neutrophil, and below normal lymphocyte counts noted in the present study may be indicative of infection. Given that most of these values returned to normal after 14 days, they may also have represented a localized response to acute inflammation following the injection procedure. Further, the observed neutrophilia and lymphopenia is often the result of stress, rather than infection. Infection usually results in significant neutrophilia, left shift, and toxicity. Future research to evaluate toxicity and left shift is needed before the conclusion of infection is made. In future studies, daily monitoring of body temperature may also help distinguish infection from acute inflammation as the etiology of these observations.

The finding of slight hemolysis in the 24-h low-dose rat and swine and 14-day high-dose swine may indicate anemia. However, we do not believe this is the case, because the data do not appear to support this. The hemolysis may be due to a reaction to the “injectibles” or to collection and transport issues, such as prolonged contact of the plasma with the cells during transport or mechanical movement—shaking motion—of the samples. Moreover, the finding of platelet clumping in 24-h and 14-day high-dose swine and 14-day low-dose rat is probably an artifact, representing the natural tendency of the platelets to stick or adhere together. Because platelet clumps do not circulate in the body, this observation may not be significant in the context of the study’s primary findings.

Because urine C-reactive protein is often used as a marker of acute inflammation, the finding that C-reactive protein levels in rats were slightly below normal prior to necropsy appears counterintuitive given the drug solution’s proposed mechanism of action. However, this observation may indicate that acute inflammatory changes were localized rather than systemic or that C-reactive protein is not the best marker of acute inflammation in rats (Giffen et al. 2003). Along with completely analyzing urine for specific gravity and urine sediment for casts, crystals, and other biochemical parameters (e.g., protein, bilirubin, white blood cell, and red blood cell), it would be beneficial for future researchers to assess haptoglobin levels, in addition to or as a substitute for C-reactive protein, to monitor acute inflammation rats.

The histopathologic changes observed following the injection procedure and drug administration are consistent with the drug solution’s proposed mechanism of action (i.e., eliciting an acute inflammatory reaction to promote connective tissue repair). The significance of the fibroplasia observed in the low-dose and high-dose 14-day swine is unknown and requires further research. Due to the absence of control group, it is not very clear if these changes are a result of the drug or the injection procedure. A previous study using this drug solution in humans reported increased cellularity, increased active fibroblasts with plump and conspicuous nuclei, and a 60% increase in average fiber diameter in the posterior sacroiliac ligaments 3 months post treatment in three patients (Klein, Dorman, and Johnson 1989). The clinical implications of such observations are unknown at this time.

There appear to be discrepancies in the histopathology findings (suppurative glomerlonephritis) and clinical chemistry findings (no azotemia) in the 14-day low-dose swine. Because no azotemia was noted (i.e., blood urea nitrogen is not elevated), the observed mild glomerulonephritis may be a nonevent and may have been present before the onset of the study. The clinical chemistry does not reflect elevated blood urea nitrogen because the glomerulonephritis was mild and it takes some time for blood urea nitrogen to mobilize beyond the normal reference range.

CONCLUSION

The methods used in this pilot study to evaluate the acute toxicity of an injectable CAM therapy drug solution containing lidocaine hydrochloride 0.25%, dextrose 12.5%, glycerin 12.5%, and phenol 1.0% in rats and swine were deemed feasible for use in future animal toxicity studies with minor modifications. Based on clinical chemistry, hematology, and histopathologic analysis of injection site soft tissues, this drug solution appears to produce acute inflammatory changes at the injection site, without apparent systemic toxicity. Future research is needed with adequate control groups to determine if these changes are the result of the drug or the injection procedure.