Reasons for performing study: Dantrolene sodium is used to prevent exertional rhabdomyolysis in predisposed horses. Food intake might negatively impact dantrolene bioavailability in horses; however, prolonged feed restriction might be detrimental to performance.
Objective: To determine a minimum duration of feed restriction that would optimise plasma dantrolene concentrations in horses after nasogastric administration. It was hypothesised that feed restriction for 4, 8 or 12 h before dantrolene administration would result in higher plasma dantrolene concentrations than achieved with no feed restriction before treatment.
Methods: Five healthy horses were randomly rotated through 4 feed restriction periods of 0, 4, 8 and 12 h duration prior to nasogastric administration of dantrolene sodium (6 mg/kg bwt). Plasma dantrolene concentration was measured by spectrofluorometry at 60, 90, 120, 150, 180 and 210 min after administration. Data were analysed via repeated measures ANOVA.
Results: Peak plasma dantrolene concentration was highest when horses had 0 and 4 h of feed restriction (0.65 ± 0.10 µg/ml at 120 min; 0.66 ± 0.17 at 180 min, respectively) and was lower when horses were restricted from feed for 8 h (0.45 ± 0.15 at 150 min) and 12 h (0.21 ± 0.09 at 180 min). Mean plasma dantrolene concentration did not differ between 0 and 4 h feed restriction at any sample time, but feed restriction for 8 h resulted in significantly lower plasma dantrolene concentration at 60 and 180 min after treatment than when horses were restricted 0 and 4 h, respectively. Plasma dantrolene concentration was significantly lower at all sample times when horses were restricted from feed 12 h compared to 0 or 4 h.
Conclusions: Absorption of nasogastrically administered dantrolene is inhibited by feed restriction before administration. To achieve optimal plasma dantrolene concentrations, feed restriction before oral administration should not exceed 4 h.
Dantrolene sodium is a direct acting muscle relaxant that suppresses calcium release from the sarcoplasmic reticulum of the skeletal muscle cell, subsequently modifying excitation-contraction coupling (Morgan and Bryant 1977; Fruen et al. 1997). Dantrolene has been recommended for the treatment and prevention of exertional rhabdomyolysis in athletic horses (Hodgson 1985; Harris 1997) and also for treatment of anaesthetic complications in horses, including malignant hyperthermia and anaesthetic myopathy (Waldron-Mease 1978; Waldron-Mease et al. 1981).
Among horses, dantrolene has been used most extensively in racing Thoroughbreds (McGowan et al. 2002; Edwards et al. 2003), a population with a reported incidence of exertional rhabdomyolysis of 5 to 7% (MacLeay et al. 1999a; McGowan et al. 2002) and increased susceptibility to post operative myopathy (Klein 1978). A clinical trial demonstrated that Thoroughbred horses with recurrent exertional rhabdomyolysis (RER) had significantly less muscle damage, as evidenced by lower serum creatine kinase activity 4 h after treadmill exercise, when treated with dantrolene (4 mg/kg bwt per os) 90 min before exercise (McKenzie et al. 2004). The therapeutic effects of dantrolene in these horses most likely arise from modification of i.m. calcium regulation, which is dysfunctional in horses with RER (Lentz et al. 1999, 2002) due to an, as yet, unidentified genetic defect (MacLeay et al. 1999b; Dranchak et al. 2005). Additionally, a study of European racing and riding horses identified higher myoplasmic ionised calcium concentrations in intercostal muscle specimens from horses with unspecified acute exertional rhabdomyolysis compared to healthy horses (Lopez et al. 1995). Treatment with dantrolene decreased myoplasmic ionised calcium concentrations in affected horses and hastened recovery (Lopez et al. 1995).
The pharmacokinetics of dantrolene sodium has been studied in horses (Court et al. 1987). The bioavailability of nasogastrically administered dantrolene sodium (4 mg/kg bwt) was 39 ± 10% in horses restricted from feed consumption for 12 h (Court et al. 1987). In another small trial of 3 horses (McKenzie et al. 2004), plasma dantrolene, measured once at 90 min, was nondetectable when horses consumed feed before administration of dantrolene (4, 6 and 8 mg/kg bwt). However, information regarding the impact of feeding on dantrolene absorption in horses is limited.
The purpose of the current study was to determine the impact of feed restriction on plasma concentrations of dantrolene after nasogastric administration to healthy horses. The investigation was designed to assess whether plasma dantrolene concentrations in horses receiving 6 mg/kg bwt of dantrolene nasogastrically wouldbe higher in horses undergoing feed restriction before administration. Our hypothesis was that feed restriction for 12 h prior to treatment would result in higher plasma dantrolene concentrations than those achieved with no feed restriction before treatment. Furthermore, an objective of the study was to determine a minimum duration of feed restriction that would optimise plasma dantrolene concentrations after nasogastric administration.
Materials and methods
Five mature healthy horses were used in the study, including 3 Quarter Horses (1 mare, 2 geldings), one Thoroughbred mare and one Thoroughbred-cross mare. Horses ranged in age from 7–14 years (mean 9 years) and horse bodyweights at the commencement of the study ranged from 488–600 kg (mean 569 kg). Horses were consuming a diet of ad libitum grass hay at the time the study commenced. The study was approved by the Oregon State University Institutional Animal Care and Use Committee.
Experimental design and sample analysis
Horses randomly rotated through a series of 4 feed restriction protocols: no feed restriction (0 h) and 4, 8 and 12 h of total feed restriction before treatment with dantrolene sodium. Treatment periods were separated by a washout period of one week, during which time horses consumed 2 flakes of grass hay twice daily.
To ensure a uniform state of feed consumption amongst horses immediately prior to commencing allocated feed restriction, all horses were weighed and stall confined 18 h prior to each treatment with dantrolene. Horses were then provided with 2 additional flakes of fresh hay 2 h before the allocated period of feed restriction commenced or in the case of 0 h of feed restriction, 2 h before treatment with dantrolene sodium. Horses were prevented from consuming feed throughout the duration of blood sample collection.
Dantrolene sodium (Dantrium 100 mg capsules)1 was administered to all horses via nasogastric tube (6 mg/kg bwt in 2 l of water) within 5 min after sedation with xylazine (0.3 mg/kg bwt i.v.). Capsules were separated and the powder placed into a 60 ml syringe and mixed with 40 ml of water. The resulting solution was injected into the nasogastric tube and 2 l of water was administered to thoroughly flush the tube. Six mls of blood was collected via jugular venipuncture into EDTA at 60, 90, 120, 150, 180 and 210 min after dantrolene administration. Blood samples were protected from light, centrifuged, and the resulting plasma aliquot was collected and frozen at -80°C until the time of analysis. Plasma dantrolene concentration was determined via spectroflourometric analysis2. In summary, 0.1 ml aliquots of plasma were extracted into chloroform after addition of deionised water and saturated ammonium sulphate. The chloroform extractions were analysed by fluorescence spectrophotometry with excitation and emission wavelengths of 395 nm and 530 nm, respectively (Perkin-Elmer MPF-2A spectroflurometer)3. Each analytical batch was independently calibrated and controlled at concentrations of 0, 0.2, 1.0, 2.0 and 4.0 µg/ml and control samples tested at 0.5, 1.0 and 3.0 µg/ml. A straight line calibration model was used and the lower limit of quantification for the method was 0.2 µg/ml and the limit of detection was 0.05 µg/ml. This method had intra-assay precision of 4.6% at 0.10 µg/ml and inter-assay precision of 10.1, 10.9 and 5.9% at 0.5, 1.0 and 3.0 µg/ml, respectively. The inter-assay accuracy of the method was 99, 98 and 99% at 0.5, 1.0 and 3.0 µg/ml, respectively.
Data were analysed by repeated measures ANOVA in a crossover model controlling for horse to horse variability. If the overall treatment effect was significant at a particular value of time, pairwise t tests were performed to separate the means. Significance was established at P<0.05. Results are displayed as mean ± s.e.
Within each feed restriction protocol (0, 4, 8 or 12 h) there were no statistically significant differences in mean plasma dantrolene concentration between 60, 90, 120, 150, 180 and 210 min after treatment (Table 1). However, peak plasma dantrolene concentrations during the study were highest at 120 min after treatment with no feed restriction (0.65 ± 0.10 µg/ml) and at 180 min after treatment following feed restriction for 4 h (0.66 ± 0.17 µg/ml). Peak plasma dantrolene concentrations were lower when horses were feed restricted for 8 h (0.45 ± 0.15 µg/ml at 150 min) and 12 h (0.21 ± 0.09 µg/ml at 180 min) before treatment; however, differences in peak plasma dantrolene concentration between the 4 feed restriction protocols were not statistically significant (Table 1).
Table 1. Plasma dantrolene concentrations (mean ± s.e.; µg/ml) in 5 horses restricted from feed consumption for 0, 4, 8 and 12 h prior to nasogastric administration of 6 mg/kg of dantrolene. Range and median displayed in brackets
Hours of feed restriction
Different superscript letters indicate significantly (P<0.05) different values within a row.
There were no significant differences in mean plasma dantrolene concentration at any time between the 0 and 4 h feed restriction protocols (Table 1). Feed restriction for 8 h resulted in significantly lower mean plasma dantrolene concentration compared to 0 and 4 h of feed restriction at 60 and 180 min after treatment, respectively. Feed restriction for 12 h resulted in significantly (P = 0.0004–0.0015) lower mean plasma dantrolene concentration compared to 0 and 4 h of feed restriction at all times except for 120 min after treatment following 4 h of feed restriction. Feed restriction for 12 h before treatment also resulted in significantly lower mean plasma dantrolene concentration at 90 and 210 min after treatment than occurred after 8 h of feed restriction.
Considerable interindividual variation in plasma dantrolene concentrations was apparent for all feeding protocols (Figs 1–4). Plasma dantrolene was not detected at any time in one horse fasted for 8 h (Fig 3) or in one horse fasted for 12 h (Fig 4). Median plasma dantrolene concentration was greatest at all sample times in horses that were not feed restricted except for 210 min after treatment when the median value was slightly higher for 4 h of feed restriction. Median plasma dantrolene was 0 µg/ml for 4 of the 6 sampling times in horses restricted from feed for 12 h.
In the current study, hay consumption by horses in the 2 h preceding nasogastric dantrolene administration resulted in significantly higher and more uniform plasma concentrations than when feed intake was restricted before treatment. Plasma dantrolene concentrations observed after 0 and 4 h of feed restriction were comparable to those considered protective against exertional rhabdomyolysis in Thoroughbred horses (McKenzie et al. 2004). However, our findings conflict with our proposed hypothesis and also somewhat with the findings of 2 previous studies (Court et al. 1987; McKenzie et al. 2004), in which oral or nasogastric administration of dantrolene (4 mg/kg bwt) to horses restricted from feed for 12 h resulted in peak concentrations of 0.6 µg/ml and 1.4 µg/ml, respectively, 90 min after treatment. In contrast, in the current study, nasogastric administration of 6 mg/kg bwt of dantrolene to similarly deprived horses resulted in nondetectable plasma concentrations in 4 of 5 horses at 90 min and an approximate peak plasma dantrolene concentration of only 0.21 µg/ml at 180 min. Although peak plasma dantrolene concentration was considerably higher after 0 and 4 h of feed restriction than 12 h of feed restriction, it remained less than 50% of the peak value reported by Court et al. (1987).
The reason for these differences is not clear, however, several factors might contribute, including methods of drug administration, sample processing and analysis, and dietary composition. In the study by Court et al. (1987), dantrolene was mixed with 500 ml of physiological saline prior to nasogastric administration, whereas a larger volume of water was used in the current study. Differences in the chemical composition, pH and volume of the diluent in which dantrolene is administered could potentially alter gastrointestinal absorption and subsequent plasma concentrations. Additionally, Court et al. (1987), determined dantrolene concentrations on heparinised, previously frozen whole blood, using a combination of solvent extraction, column chromatography and fluorometry, whereas the current study utilised a less sensitive spectrofluorometric assay on plasma. However, in a study of 7 horses receiving dantrolene sodium nasogastrically (6 mg/kg bwt in 2 l of water) after feed restriction for 12 h (McKenzie et al. 2009; McKenzie and Mosley 2009), HPLC analysis with a sensitive limit of detection (2 ng/ml) identified a similar peak plasma dantrolene concentration (0.33 µg/ml at 120 min) to that observed in horses in the current study undergoing feed restriction for 12 h.
Horses in this study consumed a ration of ad libitum grass hay before and during the entire study period. In a previous study (McKenzie et al. 2004), horses consuming a diet containing 6.4 kg of grain per day had similar peak plasma dantrolene concentrations when treated after 12 h of feed restriction to horses in the current study treated after 0 and 4 h of feed restriction. Although the impact of dietary composition on dantrolene absorption in horses is unknown, it is not implausible that dietary induced differences in gastric pH, gastric emptying rate, or gastrointestinal flora might affect gastrointestinal absorption of the drug (Inotsume et al. 1986).
Interindividual variability in plasma dantrolene concentrations was a prominent feature of our findings, similar to other studies of orally administered dantrolene in horses (Todi et al. 1988; McKenzie et al. 2004). This phenomenon has also been reported in man, despite the significantly greater bioavailability of orally administered dantrolene in humans (Meyler et al. 1981; Inotsume et al. 1986). Hence, the absorption of dantrolene by some individuals may not be as greatly affected by feeding or feed restriction, as demonstrated by Horse 5 in the current study, which displayed very similar plasma dantrolene concentrations after 0, 4 and 8 h of feed restriction. The cause of this interindividual variability is likely multifactorial and has not been specifically determined, but may include substantial individual variability in dantrolene clearance.
It is possible that repeated oral or i.v. administration of dantrolene to horses would produce more predictable blood concentrations (Court et al. 1987). However, this suggestion is complicated by the considerable cost of both forms of the drug and the possibility of side effects (Waldron-Mease 1978; Allen et al. 1988). Side effects reported after i.v. dantrolene administration to horses include weakness, ataxia and urination (Court et al. 1987). Prolonged post anaesthetic recumbency was reported in one horse administered 9 mg/kg bwt per os before anaesthesia (Valverde et al. 1990); however, horses receiving 4 mg/kg bwt per os 90 min before treadmill exercise displayed no weakness or neurological deficits and adequately completed their exercise task (McKenzie et al. 2004). Although hepatotoxicity is a feature of dantrolene administration in man (Chan 1990), no evidence of hepatotoxicity was identified in Thoroughbred horses receiving 4 mg/kg bwt per os daily for 3 weeks (McKenzie et al. 2004).
The major finding of this study, that adequate absorption of dantrolene occurs in horses without prolonged feed restriction, is likely to benefit athletic horses treated with dantrolene for control of exertional rhabdomyolysis, since extended feed deprivation might detrimentally impact performance. Management alterations involving exercise and dietary changes are currently the most convenient and effective methods for control of chronic exertional rhabdomyolysis in horses (MacLeay et al. 2000; Firshman et al. 2003; McKenzie et al. 2003); however, dantrolene may be useful in horses that do not respond to these measures or that require treatment for severe rhabdomyolysis. Additionally, since feed deprivation is routinely practiced before elective general anaesthesia of horses, the findings of this study also have implications regarding the use of dantrolene for the purpose of preventing post operative myopathy. Our findings suggest that dantrolene administration to horses fasted for 12 h or more prior to general anaesthesia may result in nondetectable or very low blood dantrolene concentrations, which would potentially negate any anticipated therapeutic benefits.
A weakness of this study was the relatively infrequent measurement of plasma dantrolene concentrations in treated horses. This possibly contributed to the considerable fluctuations in plasma dantrolene concentration occasionally observed between consecutive sample times in the same horse. The sampling regime was based on the expected pharmacokinetics of nasogastrically administered dantrolene (Court et al. 1987) and was also influenced by the considerable expense of the commercially available spectrofluorometric assay. Pronounced fluctuations in plasma dantrolene concentration might reflect enterohepatic cycling of the drug and might also have been exacerbated by the relatively high minimum level of detection of the spectrofluorometric assay. A previous study (Court et al. 1987) determined that excretion of intact dantrolene into bile is minimal and concluded the enterohepatic cycling is unlikely to occur in the horses. However, this data was derived in ponies after i.v. administration of dantrolene and it is not certain that oral or nasogastric administration of dantrolene to horses would emulate these findings.
In conclusion, although our results differ somewhat to previous findings, the current study represents a considerably more comprehensive investigation of the impact of feed restriction on gastrointestinal dantrolene absorption in horses. The progressive decline in mean and median plasma dantrolene concentrations with increasing duration of feed restriction indicates that, at least in horses in the current study, feed restriction inhibited dantrolene absorption. Our findings suggest that where possible, feed restriction before nasogastric dantrolene administration should be avoided or should not exceed 4 h.
Conflicts of interest
SV receives royalty payments from Kentucky Equine Research and the patent for type 1 PSSM. The remaining authors have no potential conflicts to declare.
This study was kindly funded by the American Quarter Horse Association. Ms Garrett was funded by the Merck Merical Summer Scholar Program.
1 Proctor and Gamble Pharmaceuticals, Cincinnati, Ohio, USA.
2 NMS Laboratories, Willow Grove, Pennsylvania, USA.