Complications Associated with the Administration of Dantrolene 1987 to 2006: A Report from the North American Malignant Hyperthermia Registry of the Malignant Hyperthermia Association of the United States

Abstract

Background

Dantrolene is the only specific treatment for malignant hyperthermia (MH), a genetic disorder in which life-threatening temperature increase has been induced by inhaled anesthetics and succinylcholine. Because MH presents with nonspecific signs and delay of treatment can be fatal, dantrolene may be given as soon as MH is suspected. We report the complications associated with dantrolene administration as documented in AMRA (adverse metabolic/musculoskeletal reaction to anesthesia) reports submitted to the North American Malignant Hyperthermia Registry.

Methods

AMRA reports were analyzed for differences between subjects with and without complications attributed to dantrolene. Documentation of dantrolene dose and subject weight were inclusion criteria. Because some reported complications were likely due to factors other than dantrolene, a reduced set of cases was also defined. Chi-square and Mann-Whitney tests were used. Logistic regression was applied to describe factors associated with increased risk of complications.

Results

In the full dataset of 368 subjects, the most frequent complications associated with dantrolene were muscle weakness (21.7%), phlebitis (9%), gastrointestinal upset (4.1%) and respiratory failure (3.8%). Logistic regression described a 29% increase in risk of any complication when the total dantrolene dose was doubled, a 144% increase in risk when fluid administration was part of treatment, an 83% decrease in risk in the presence of neurosurgery and a 74% decrease in risk in the presence of oral surgery.

In the dataset reduced by removal ofsome serious complications that were judged likely to have been due to preexisting disease or the MH event, there were 349 subjects. The most frequent complications associated with dantrolene were muscle weakness (14.6%), phlebitis (9.2%) and gastrointestinal upset (4.3%). In this reduced dataset, logistic regression described a 25% increase in risk of any complication when the total dantrolene dose was doubled, a 572% increase in risk in the presence of obstetric or gynecologic surgery, a 56% decrease in risk if furosemide was given and no relationship with fluid administration or other types of surgery.

Conclusions

Complications after dantrolene are common, but rarely life threatening. Unidentified factors in the surgical environment are associated with changes in the risk of complications. Fluid management, as part of the treatment of MH, has an important association with the risk of complications after dantrolene administration and should be monitored closely.

Introduction

Dantrolene, a skeletal muscle relaxant that reduces myoplasmic calcium,^1,2 effectively treats malignant hyperthermia (MH).³ Because MH presents with nonspecific signs⁴ and delay of treatment can be fatal,⁵ dantrolene may be given as soon as MH is suspected. Also, dantrolene has been used to treat spasticity,⁶ reduce the metabolic effects of fever in the presence of neurologic injury,⁷ treat cerebral vasospasm,⁸ neuroleptic malignant syndrome,⁹ and 3,4-methylenedioxymethamphetamine (Ecstasy) intoxication.¹⁰

The medical literature provides limited data on dantrolene’s side effects. In a study of 6 control subjects and 6 MH-susceptible patients, Wedel et al. demonstrated that a 3 mg/kg dose of IV dantrolene produced visual symptoms more commonly in control subjects than in MH-susceptible subjects and that MH-susceptible subjects complained more frequently of muscle weakness, dizziness, and fatigue. For all studied individuals, 7/12 experienced nausea, 3/12 complained of dyspnea, and 1/12 developed a superficial phlebitis.¹¹ Case reports note a possible association of dantrolene and the development of muscle weakness,^12,13 respiratory distress,^14-16 pleural effusion,¹⁷ hepatotoxicity,¹⁸ and hyperkalemia in the presence of concomitant calcium channel blocking drugs.^19,20 There have been no large postmarketing surveillance (Phase 4) studies of dantrolene’s adverse effects.

Therefore, we examined the type and frequency of complications associated with dantrolene as reported to The North American Malignant Hyperthermia Registry via the AMRA (adverse metabolic/musculoskeletal reaction to anesthesia) Report. Additionally, we tested the hypothesis that complications from dantrolene administration were associated with clinical factors or subject characteristics.

This report serves as a benchmark for the experience of the past three decades. During this time Dantrium®, recently manufactured by Proctor and Gamble, was the only formulation of dantrolene that was clinically available. In 2008, the manufacturer sold this product. Reports on the complications of dantrolene after 2008 may have different results specific to the new formulations by new manufacturers.

Materials and Methods

IRB review deemed this study exempt. Five hundred eighty-two (582) AMRA reports received by the North American Malignant Hyperthermia Registry as of December 31, 2006 were examined. Inclusion criteria were documentation of weight of the subject, initial or total dantrolene dose, and treatment after December 31, 1986 in the United States or Canada. These criteria were met by 368 reports.

The outcomes of interest, complications attributed by clinicians to dantrolene administration, are specifically queried by checkbox on the AMRA reports. These complications include phlebitis, muscle weakness, respiratory failure, excessive secretions, gastrointestinal upset, and hyperkalemia. Health care providers completing the questionnaire were given the option of writing in additional complications or adding comments (Online Supplement).

Independent variables that were examined for possible association with complications included subject sex, weight, age, total dose, total dose adjusted for weight, initial dose, initial dose adjusted for weight, and whether the subject received more than one dantrolene dose. Other variables noted on the AMRA reports (including the type of surgery during which the MH episode occurred, signs of MH, intervals between events, laboratory tests obtained during or after the MH episode and other procedures used to treat MH) were also examined for possible association with complications attributed to dantrolene. The relationship between the presence of complications attributed to dantrolene and the likelihood of an MH event as documented by the clinical grading scale (CGS) was also examined. The CGS estimates the qualitative likelihood that a MH event occurred, based on observed signs and laboratory tests.²¹

Three datasets were analyzed. The larger set (full dataset) included all cases that met the inclusion criteria and all complications reported on the AMRA reports in association with dantrolene. The smaller subset (reduced dataset) removed complications that were likely due to either underlying medical conditions or the acute MH episode itself rather than to dantrolene administration. These judgments were made by the anesthesiologist investigators with MH expertise (BWB and MGL). Cases were excluded if the only complication associated with dantrolene was grave (respiratory failure, pulmonary edema, hyperkalemia, low cardiac output) and review of the cases identified a potential medical cause other than dantrolene for this complication. If muscle weakness was the only complication associated with dantrolene and review of details by the anesthesiologist investigators with MH expertise (BWB and MGL) concluded that this was likely due to serious preexisting disease, the MH episode itself, or severe rhabdomyolysis, the case was excluded from analysis. Severe rhabdomyolysis was defined by the indicators for the process of muscle breakdown in the CGS; creatine kinase more than 10,000 IU without exposure to succinylcholine and more than 20,000 IU after exposure to succinylcholine. If rhabdomyolysis of this degree was associated with muscle weakness, muscle weakness was not counted as a complication of dantrolene in analysis of the reduced dataset. Hyperkalemia was removed as a complication if the potassium level was ≤6.0 mEq/L, renal disease was present before dantrolene administration, or the MH event was ranked as “very likely” or “almost certain.” For those cases with both a serious complication likely due to factors other than dantrolene and other complications, which may have been due to dantrolene, the case remained in the reduced dataset but without the serious complications. The smallest dataset was created by discounting technical drug administration complications (e.g., difficulty with mixing dantrolene with diluent or the observation of precipitate in the IV tubing) as a dantrolene complication in the reduced dataset.

Several methods were used to test the hypothesis that complications from dantrolene administration were associated with clinical factors or subject characteristics in the full and the reduced datasets. Because many of the variables occurred in very few cases or were not distributed normally, nonparametric or exact tests were used to identify potentially significant factors. For continuous variables, the Mann-Whitney test was used to compare the mean values of the groups in which the complication occurred with that in which it did not occur. For categorical variables, a Fisher exact test was used to assess the difference between the groups. Variables with p<0.2 in these univariate tests that were reported at least 10 times in this dataset became candidates for entry into a multivariate logistic regression model. Stepwise selection into a logistic regression was performed to minimize the -2 log likelihood of the risk of any dantrolene complication. A Wald statistic with p< 0.05 indicated statistical significance. Logistic regressions were developed to assess the association of all variables and their potential interactions with the outcome of any complication reported with dantrolene. Correlation between variables of interest was examined. Analysis was performed using SPSS for Windows release 16.0.1 (SPSS Inc., Chicago, IL).

Odds ratios were calculated from the logistic regression. P values less than 0.05 were accepted as statistically significant. The mean is presented with standard deviation. The median is presented with 1st and 3rd quartiles and range.

Results

Age and weight of the subjects and the CGS of the events are presented in Table 1. Doses of dantrolene administered are summarized in Table 2.

Table 1.

Subject and Event Characteristics

Full Dataset Characteristic	N	Median	25%	75%	Range
Age, yr	346	20.0	8.0	37.0	0 – 78
Weight, kg	368	68.0	24.3	85.7	3 – 159
CGS Score *	368	48	33	60	0 – 88
Restricted Dataset Characteristic	N	Median	25%	75%	Range
Age, yr	328	19.0	7.0	35.8	0 - 78
Weight, kg	349	66.4	23.7	84.1	3 - 159
CGS Score *	349	48	33	58	0 - 88

^*The clinical grading scale (CGS) score is designed to translate to a malignant hyperthermia (MH) rank that estimates the qualitative likelihood that an episode of MH occurred, based on observed signs and laboratory tests.³³ CGS more than 34 indicates very likely MH. CGS greater than 49 indicates MH is almost certain. There were no significant differences between datasets for the listed characteristics.

Table 2.

Dantrolene Doses

Full Dataset Dose Type	N	Median	25%	75%	Range
Total Dose, mg	368	234.0	100.0	600.0	2.0 – 6,720
Total Dose, mg/kg	368	4.9	2.5	9.0	0.02 – 100
Initial Dose, mg/kg	364	2.4	1.7	2.7	0 – 15.0
Reduced Dataset Dose Type	N	Median	25%	75%	Range
Total Dose, mg	349	222.0	94.5	600.0	2.0 – 6,720
Total Dose, mg/kg	349	4.7	2.5	8.7	0.02 - 100
Initial Dose, mg/kg	345	2.4	1.6	2.6	0 – 15.0

Dantrolene doses were reported in the AMRA (adverse metabolic/musculoskeletal reaction to anesthesia) as initial dose and total dose. There were cases in which one of these two was not reported. Therefore, the Total Dose was defined as the total dose when reported or the initial dose if no other dose was reported. This definition of total dose is the variable used in regression analysis. There was no significant difference between the full and reduced datasets for any of these dantrolene doses.

Full Dataset

Data from the full set of 368, predominantly male (73.4%), subjects are presented in Tables 1 through 4. There were 95 (25.8%) subjects who received a single dose of dantrolene, 258 (70.1%) who received more than one dose, and 15 (4.1%) whose dose frequency was unknown. Because each 20 mg vial of dantrolene requires 60 ml of sterile water diluent, median, 1^st and 3^rd quartile total dantrolene doses required the co-administration of 702 (300, 1,800) ml of sterile water, respectively.

Table 4.

Logistic Regressions: Outcome is the Risk of Any Complication Associated with Dantrolene

Full Dataset (N=368)
Variables	Coefficients	S.E.	P Value	Odds Ratio	(95% CI)
Dose Factor *	0.363	0.101	0.000	1.29	(1.12-1.48)
Fluid Administration	0.893	0.289	0.002	2.44	(1.37-4.35)
Oral Surgery	−1.341	0.581	0.021	0.26	(0.08-0.84)
Neurosurgery	−1.760	0.647	0.006	0.17	(0.05-0.63)
Constant	−3.142	0.599	0.000
Reduced Dataset (N=349)
Dose Factor *	0.317	0.111	0.004	1.25	(1.07- 1.45)
Obstetrics/ Gynecology	1.906	0.608	0.002	6.72	(1.99- 22.7)
Furosemide	−0.822	0.317	0.009	0.44	(0.23- 0.82)
Constant	−2.766	0.614	0.000

^*Dose Factor is the natural log of the total dose of dantrolene or of the initial dose if a total dose was not reported.

The logistic regression for the full dataset describes a 29% increase in risk of any complication when the total dantrolene dose was doubled, a 144% increase in risk when fluid administration was part of treatment, an 83% decrease in risk in the presence of neurosurgery and a 74% decrease in risk in the presence of oral surgery.

The logistic regression for the reduced dataset describes a 25% increase in risk of any complication when the total dantrolene dose was doubled, a 572% increase in risk in the presence of obstetric or gynecologic surgery, a 56% decrease in risk if furosemide was given and no relationship with fluid administration or other types of surgery.

As is common in statistical analysis, some but not all of the assumptions of this technique are met by the data. The outcome variable, presence or absence of complications reported after dantrolene administration, is binary, but it is a composite of all reported complications. The independent variables are not linear combinations of each other. Observations are independent because each case is independent. There may be error in the measurement of independent variables. As described in Methods we used stepwise entry of variables into the regression to judge that no important variables were omitted and no extraneous variables were included. We assume that the logit function is appropriate to describe the outcome variable. For all the above reasons we do not propose that these regressions are sufficient to predict the occurrence of complications. Our goal is to use the available data to alert the clinician to conditions which have been associated with complications after administration of dantrolene. For example, the logistic regression for the full model describes an 18.7% chance of any complication after administration of 100 mg of dantrolene in the absence of fluid administration, neurosurgery or oral surgery. The dose of 100 mg dantrolene is the 1st quartile of the distribution of doses and 600 mg is the 3rd quartile. Increasing the dose to 600 mg is associated with 30.6% chance of any complication and in the presence of fluid administration the chance of any complication increases to 51.8%. Similarly the logistic regression for the reduced model describes a 21.0% chance of any complication after administration of 94.5 mg of dantrolene in the absence of furosemide or obstetric-gynecologic surgery. The dose of 94.5 mg dantrolene is the 1st quartile of the distribution of doses and 600 mg is the 3rd quartile in the reduced dataset. Increasing the dose to 600 mg is associated with 32.3% chance of any complication and in the presence of furosemide the chance of any complication decreases to 17.4%.

In 129 of the 368 subjects (35.1%) in the full dataset, a complication was noted with dantrolene administration (Table 3; Figure 1). Of these, 37 (10.1%) had more than one complication associated with dantrolene administration. In these 129 subjects the median number of complications was 1, as was the 1^st quartile, while the 3rd quartile was 2 and the range 1 to 5.

Table 3.

Complication Frequency

Complication Type	Full Dataset (n=368) Number (%)	Reduced Dataset (n=349) Number (%)
Any Complication	129 (35.1)	83 (23.8)
Muscle Weakness	80 (21.7)	51 (14.6)
Phlebitis	33 (9.0)	32 (9.2)
Gastrointestinal Upset	15 (4.1)	15 (4.3)
Respiratory Failure	14 (3.8)	0 (0)
Hyperkalemia	12 (3.3)	0 (0)
Excessive Secretions	3 (0.8)	3 (0.9)
Other Complications	30 (8.2)*	13 (3.7)**

^*Other complications for the full dataset included: pulmonary edema (N=6), difficulty mixing dantrolene or precipitate in the IV (N=5), pain on injection (N=3), swollen extremity (N=2), concomitant decrease in cardiac output (N=2), prolonged mechanical ventilation (N=2), and hypotension, decreased hearing for 24 hours, blurred vision, difficulty opening eyes, anxiety for several minutes after injection, dysphoria, disorientation, somnolence, generalized weakness, increased alanine aminotransferase (ALT) on the second postoperative day, hyponatremia, each noted once in the full dataset. Some subjects had more than one complication.

^**Other complications for the reduced dataset included: difficulty mixing dantrolene or precipitate in the IV (N=5), pain on injection (N=3) and swollen extremity, decreased hearing for 24 hours, blurred vision, dysphoria, disorientation, somnolence, each noted once in the reduced dataset. Some subjects had more than one complication.

Figure 1.

depicts the number of cases in the full dataset with and without complications stratified by dantrolene dose category. X axis is dantrolene dose category with I = 1 -20 mg, II = 21 -80 mg, III = 81-320 mg, IV = 321-1,280 mg, and V>1,280 mg. Y axis is number of cases. Green shading indicates no dantrolene complications; yellow shading indicates dantrolene complications.

The logistic regression model that best described the risk of any complication associated with dantrolene in these 368 cases included the natural log of the dose of dantrolene in mg, and the binary variables: fluid administration, oral surgery and neurosurgery (Table 4).

Observations that were not part of the logistic regressions include the following: before or during the adverse anesthetic reaction in 101 of 368 cases, either a central venous or pulmonary artery catheter or a transesophageal echocardiogram probe was placed. There was no significant difference in the frequency of dantrolene complications (P=0.46) between those with and without these monitors.

Of the 80 subjects with muscle weakness, 5 were known to have either a history of muscle weakness or a physical finding of generalized muscle weakness or myopathy before the dantrolene administration. The median total dose of dantrolene was larger for those who experienced muscle weakness (400 mg vs. 212 mg), P=0.002. Logistic regression indicated that doubling the total dose of dantrolene increased by 27% the likelihood of a muscle weakness complication, P=0.002. Of the 80 reports of muscle weakness, 33 had a documented duration of dantrolene administration. Of these 33 subjects, 22 had received dantrolene for at least 24 hours.

Twelve subjects were reported to have experienced hyperkalemia as a complication of dantrolene (serum potassium of 5.2 to 9.9 mEq/L). Subjects with hyperkalemia had higher CGS scores (P=0.005). Review of the hyperkalemia cases determined that preoperative renal failure was present in 2 and none of the others received concomitant calcium channel blockers. All subjects with hyperkalemia had experienced a “very likely” or “almost certain” MH event.

There were also 14 cases with respiratory failure, 6 with pulmonary edema and 2 with low cardiac output reported in association with dantrolene administration. One of the 14 subjects with respiratory failure was known to have a myopathy before the dantrolene administration. The 6 patients with pulmonary edema had all received fluid administration, volume unknown, as part of their MH treatment. There were 18 subjects with one or more of these grave complications reported in addition to the other complications associated with dantrolene. In 12 of these patients, there was evidence of serious underlying disease or complex surgery, such as concomitant administration of multiple vasoactive drugs and recent cardiovascular surgery including cardiopulmonary bypass. In 2 of 18 patients there was no evidence of such complicating factors. These two patients had CGS scores of 43 and 58, indicating that the events were “very likely” or “almost very likely or almost certain” MH.

Four subjects were reported to have developed compartment syndrome as a complication of dantrolene. After review of the AMRA reports, the anesthesiologist authors (MGL and BWB) considered that these 4 cases of compartment syndrome were likely secondary to an adverse reaction to anesthesia (including MH) or a preexisting condition (trauma) rather than to dantrolene administration. This complication was still counted in the full dataset, but not in the reduced dataset.

Figure 2 depicts the number of cases with and without fluid administration who experienced a complication stratified by dantrolene dose in comparison with those who did not experience a complication. In the 98 cases in which fluid administration was reported as part of MH treatment, muscle weakness was 2.7 times more likely to be reported (P=0.003) and succinylcholine given two times as often (P=0.005), the median absolute total dantrolene dose was 1.3 times larger (P=0.002), and the median total dantrolene dose/kg was 1.5 times larger (P=0.015) in comparison with cases that did not report fluid administration as part of therapy. Also, the median CGS score was higher in those receiving fluid administration (51, “almost certain MH”) than those who did not (38, “very likely MH”), P<0.0005.

Figure 2.

depicts the number of cases in the full dataset with and without fluid administration and with and without complications, stratified by dantrolene dose category. X axis is the dantrolene dose category with I = 1 -20 mg, II = 21 -80 mg, III = 81-320 mg, IV = 321-1,280 mg, and V>1,280 mg. Y axis is number of cases. Green shading indicates no dantrolene complications; yellow shading indicates dantrolene complications.

Reduced Dataset

Data from the reduced set of 349, predominantly male (72.8%), subjects is presented in Tables 1 - 4. The relationship between dantrolene dose and complications is depicted in Figure 3. There were 88 (25.2%) subjects who received a single dose of dantrolene, 232 (66.5%) who received more than one dose, and 29 (8.3%) whose dose number was unknown. Median, 1^st and 3^rd quartile total dantrolene doses required the co-administration of 666 (284, 1,800) ml of sterile water, respectively.

Figure 3.

depicts the number of cases in the reduced dataset with and without complications stratified by dantrolene dose category. X axis is dantrolene dose category with I = 1 -20 mg, II = 21 -80 mg, III = 81-320 mg, IV = 321-1,280 mg, and V>1,280 mg. Y axis is number of cases. Green shading indicates no dantrolene complications; yellow shading indicates dantrolene complications.

The reduced subset of cases had fewer complications, by definition, but muscle weakness was still common and the incidence of phlebitis and gastrointestinal upset was unchanged (Table 3). In the 83 subjects with at least one complication, the median number of complications was 1, as was the 1st quartile, while the 3rd quartile was 2 and the range 1 to 4. In the reduced dataset, subjects with either a central venous or a pulmonary artery catheter or a transesophageal echocardiography probe were 2.77 times less likely to experience a dantrolene complication than those without such monitors, P=0.002.

Of the 51 reports of muscle weakness, 19 had a documented duration of dantrolene administration. Of these 19 subjects, 13 had received dantrolene for at least 24 hours. Those experiencing muscle weakness had received a significantly higher dose of dantrolene than those who did not experience this complication (360 mg vs. 212 mg), P=0.011. Logistic regression indicated that doubling the dantrolene dose increased by 25% the likelihood of a muscle weakness complication, P=0.012. Figure 4 depicts the number of subjects who received or did not receive furosemide and who did or did not develop a complication stratified by total dantrolene dose. Those subjects treated with furosemide (n=106) received a higher median total dantrolene dose (380 mg vs. 200 mg, P<0.0005) and a higher median total dantrolene dose/kg (6.2 mg/kg vs. 4.0 mg/kg, P=0.002), although their CGS scores did not differ.

Figure 4.

depicts the number of cases in the reduced dataset with and without furosemide and with and without complications, stratified by dantrolene dose category. X axis is the dantrolene dose category with I = 1 -20 mg, II = 21 -80 mg, III = 81-320 mg, IV = 321-1,280 mg, and V>1,280 mg. Y axis is number of cases. Green shading indicates no dantrolene complications; yellow shading indicates dantrolene complications.

The logistic regression model that best described the risk of any complication associated with dantrolene administration in the reduced dataset of 349 cases included the natural log of dose of dantrolene, obstetric or gynecologic surgery, and administration of furosemide (Table 4).

When the reduced dataset is further reduced by discounting problems with mixing or precipitate in the IV tubing as complications of dantrolene, the same number of cases contributes to regression analysis, and the same variables appear in the model. The coefficients of the variables in the logistic regression changed slightly, but not significantly.

Comparison of Models from the Two Datasets

When the statistically significant binary variables, fluid administration, furosemide administration and some types of surgery, were added to the model of dantrolene complications, the odds ratio for dantrolene complications decreased. There was no difference in the effect of increasing the dose of dantrolene on the risk of complications as described by logistic regression between the full and the reduced dataset. The coefficients of the natural logarithm of the dose of dantrolene were not significantly different (Z=1.21).

The binary variables and the incidence of complications are presented in Table 5. The presence of fluid administration was associated with a 144% increase in risk of reported dantrolene complications and administration of dantrolene during oral surgery or neurosurgery was a protective factor only when the most severely ill patients (the full dataset) are part of the model. In the reduced dataset none of these binary variables was significant, however a new protective variable, furosemide administration, was associated with a reduced risk of reported dantrolene complications and administration of dantrolene during obstetric or gynecologic surgery greatly increased the risk of complications.

Table 5.

Number of Complications in Binary Variables of Interest in Both Models

Dataset Type	Variable	Any Complication	No Complications
Full Dataset
	Fluid Administration	109	161
	No Fluid Load	20	78
	Oral Surgery	4	18
	None	125	221
	Neurosurgery	3	20
	None	126	219
Reduced Dataset
	Obstetrics/Gynecology	8	5
	None	75	261
	Furosemide	19	87
	None	64	179

This table presents the data that contributed to significant logistic regressions modeling for the risk of complications associated with dantrolene. The total number in each of these 2 by 2 comparisons differs because the first three variables in Table 5 were significant in the full dataset model and the last two in the reduced dataset model.

In both datasets weight, age, and dantrolene dose unadjusted for weight were very highly correlated (Tables 6 and 7). If weight were forced into a model of dantrolene complications, none of the doses of dantrolene would have sufficient power to enter the model. Age had the same role in this dataset.

Table 6.

Correlations Matrix of Variables for the Full Dataset

Full Dataset		Weight (kg)	Age (years)	Initial Dose (mg)	Total Dose (mg)	Total Dose (mg/kg)
Weight (kg)	Pearson Correlation	1.000	.678**	.581**	.470**	.040
Sig (2-tailed)			.000	.000	.000	.445
Age (Years)	Pearson Correlation	.678**	1.000	.399**	.298**	.011
Sig (2-tailed)		.000		.000	.000	.842
Initial Dose	Pearson Correlation	.581**	.399**	1.000	.484**	.188**
Sig (2-tailed)		.000	.000		.000	.000
Total Dose	Pearson Correlation	.470**	.298**	.484**	1.000	.706**
Sig (2-tailed)		.000	.000	.000		.000
Total Dose (mg/kg)	Pearson Correlation	.040	.011	.188**	.706**	1.000
Sig (2-tailed)		.445	.842	.000	.000

^**Correlation is significant (sig) at the 0.01 level (2-tailed).

Table 7.

Correlations Matrix of Variables for the Reduced Dataset

Reduced Dataset		Weight (kg)	Age (years)	Initial Dose (mg)	Total Dose (mg)	Total Dose (mg/kg)
Weight (kg)	Pearson Correlation	1.000	.685**	.571**	.453**	.029
Sig (2-tailed)			.000	.000	.000	.585
Age (Years)	Pearson Correlation	.685**	1.000	.413**	.298**	.009
Sig (2-tailed)		.000		.000	.000	.871
Initial Dose	Pearson Correlation	.571**	.413**	1.000	.480**	.186**
Sig (2-tailed)		.000	.000		.000	.001
Total Dose	Pearson Correlation	.453**	.298**	.480**	1.000	.707**
Sig (2-tailed)		.000	.000	.000		.000
Total Dose (mg/kg)	Pearson Correlation	.029	.009	.186**	.707**	1.000
Sig (2-tailed)		.585	.871	.001	.000

^**Correlation is significant (sig) at the 0.01 level (2-tailed).

Discussion

We found that complications from dantrolene administration were associated with clinical factors. The complications of dantrolene were related to the total dose administered, but not to the initial dose or dose/weight. In these datasets, dantrolene dose, weight and age were correlated. The likelihood of dantrolene-associated complications did not change with increasing severity of the MH event as measured by either the CGS score or rank.

Complications were examined as a function of the subjects’ clinical status to create a reduced dataset that represents a minimum estimate of the factors associated with complications during dantrolene administration. This is in contrast to the full dataset that provides a maximum estimate of the factors associated with complications associated with dantrolene regardless of underlying medical or surgical conditions. While analysis of the reduced dataset reduced the odds ratio for risk of increasing complications with increasing dose of dantrolene, the observed change was not statistically significant.

The small numbers of cases of oral surgery, neurosurgery and obstetric and gynecologic surgery make it difficult to identify any factors that could be expected to be responsible for differences in risk of complications associated with dantrolene administration in these cases. The lower risk of complications during neurosurgery may be related to details of the anesthetic environment during this type of surgery, including fluid restriction. However, we were unable to attribute the lower risk to either more intensive monitoring or shorter inhaled anesthetic exposure. The AMRA report does not request data on the dose of potent inhaled anesthetic or the specific details of fluid management beyond asking whether fluid administration was used as a MH treatment. The low number of neurosurgical, oral surgery and obstetric and gynecologic surgery cases, however, suggests that a larger dataset might not produce the same conclusions.

The finding that administration of a fluid load is associated with increased risk of complications associated with dantrolene when cases with grave cardiovascular, pulmonary and/or renal conditions, including severe MH, are present is not surprising. Mannitol is part of this formulation of dantrolene. There will be fluid shifts due to the pathophysiology of MH and due to mannitol. Also, reconstitution of dantrolene for administration requires 3 ml of sterile water for every 1 mg of dantrolene administered. Documentation of cardiac filling pressures and cardiac output with continuous monitors such as echocardiography, may improve management of critically ill subjects during MH treatment, although we were unable to demonstrate a reduction in dantrolene-associated complications with their use in the full dataset. Fluid management during treatment of MH may be challenging even in the relatively healthy patient, since use of monitoring and administration of furosemide was associated with reduced risk of complications in the reduced dataset of patients without preexisting systemic disease.

Muscle weakness was the most common complication of dantrolene administration and was associated with total dose received. Flewellen et al. reported that maximum muscle weakness in normal volunteers occurs at a dantrolene dose of 2.4 mg/kg, which produces a plasma concentration near the therapeutic concentration of the drug.²² Therefore, it is not surprising that muscle weakness was frequently reported in our subjects who received a therapeutic dose of dantrolene. Since 80% of recrudescence events occurred within 16 hours²³ of the initial MH treatment, it seems reasonable to suggest that if a patient receiving dantrolene has muscle weakness and is metabolically stable 24 hours after initial therapy, dantrolene could be administered at longer intervals or at a reduced infusion rate and then stopped. Patients who experience acute MH may also experience muscle weakness as a result of muscle injury that can be part of this syndrome. The clinician treating an MH episode should request repeated measurement of creatine kinase until it returns to normal levels. Patients with muscle weakness may require longer periods of postoperative positive pressure ventilation, which has its own associated increase in morbidity and mortality.

The 9% incidence of phlebitis is unsurprising given dantrolene’s high pH (9.5) and relative insolubility. The frequency of phlebitis does suggest that once an acute MH event is controlled, efforts should be made to administer dantrolene via a large bore IV and all IV sites should be carefully monitored. Gastrointestinal upset was reported in 4% of our subjects. This side effect has been observed in humans.^{11, 22, 24} Dantrolene has been demonstrated to interfere with excitation-contraction coupling of murine intestinal smooth muscle cells²⁵ and rat gastric fundus, and colon.²⁶

The benefit of logistic regression is that it identified variables that contributed to a statistically significant description of the risk of complications after administration of dantrolene. It is noteworthy that the total dose of dantrolene is identified by logistic regression as an important risk factor. However, in clinical practice dantrolene is usually administered in a dose based on patient weight. Therefore heavier patients will usually receive a larger total dose. There was not sufficient power to identify a significant interaction between weight and dose in these logistic regression models.

Because the dose of dantrolene was related to the reported complications, we reviewed the dantrolene dose administered by clinicians in this case series and what might have been their rationale for this dosing. The initial dose of dantrolene per kilogram in this report was not different from the dantrolene dose that has been recommended by the Malignant Hyperthermia Association of the United States. This recommendation was based on the results of the study by Kolb et al.³ Because recrudescence of MH occurs in a substantial number of patients²³ it is important to continue dantrolene administration after initial therapy. There is a plateau phase in the plasma decay of dantrolene which lasts between four and six hours.^{22, 27} Therefore, the recommendation is to administer an additional 1 mg/kg every six hours after initial dosing to treat an acute MH episode, assuming that the patient is clinically stable. Others have suggested administration of a continuous infusion of dantrolene to produce more stable plasma concentrations.²⁸ If the patient is not stable, more dantrolene should be administered until all signs of MH have abated. There is no formal recommendation as to the maximal dose of dantrolene that should be given. Recrudescence is a serious condition with possibly fatal outcome. It is expected that administration of dantrolene for at least 24 hours will decrease the frequency of recrudescence.

We chose to examine all cases in which dantrolene was administered, regardless of the likelihood that MH was occurring in that case, because our goal was to describe potential complications associated with administration of this drug. A limitation of this study is the voluntary nature of reporting to this database. Another limitation of this study is incomplete and possibly biased reporting on AMRA reports. Often the individual who reported the identification and initial treatment of an MH episode transferred the patient to the care of an intensivist. This may have resulted in underreporting of the total dose of dantrolene given and underreporting of the late complications of MH treatment.

To evaluate complications that may be associated with new formulations of dantrolene, anesthesia providers should document in detail the anesthetics that they suspect are complicated by MH by completing an AMRA report for suspected MH events and sending it to the North American Malignant Hyperthermia Registry (www.mhreg.org). The data supplied by such reports are necessary for improving our ability to diagnose and treat this anesthetic-induced disease.

In summary, we report a 24% rate of complications associated with the administration of dantrolene in 349 cases. The most common complications observed were muscle weakness, phlebitis and gastrointestinal upset. In cases where serious underlying medical conditions were also present, respiratory failure, pulmonary edema, and hyperkalemia were additionally noted as complications. These severe complications were most likely due to the patient’s underlying medical conditions rather than dantrolene administration itself. The likelihood of a complication associated with dantrolene administration increased as the dose of dantrolene increased, was variably affected by the type of surgery, and could be greatly increased by fluid administration (full dataset) and decreased by administration of furosemide (reduced dataset). When dantrolene is administered, clinicians should exercise vigilance for changes in intravascular fluid volume and subsequent development of cardiorespiratory complications.