Approximately 17% of cases of malignant hyperthermia (MH) occur in children,1 yet the clinical characteristics of MH in the pediatric population have not been fully elucidated. Gaps in knowledge include preoperative risk stratification as well as clinical presentations and responses to treatment that are unique to children based on developmental age.
The primary aim of this study was to use data from the North American Malignant Hyperthermia Registry (NAMHR)2,3 to determine the differences in clinical characteristics of acute MH across pediatric age populations. Given the developmental changes in body structure and composition throughout childhood, we hypothesized that there are differences in clinical presentation, clinical course, and outcomes based on the age group of the patient. A secondary aim was to determine the types of preexisting medical conditions that were associated with pediatric MH cases.
We obtained permission from the IRB of the Stokes Research Institute of The Children’s Hospital of Philadelphia (requirement for written consent was waived by the IRB because the database was de-identified at its source) to conduct a retrospective analysis of all subjects up to and including 18 years of age in the NAMHR with an MH clinical grading scale (CGS) score at or above 35, indicating “very likely” or “almost certain” MH. The CGS represents a qualitative score of the likelihood that an MH event occurred, on the basis of clinical characteristics and laboratory results.4
The NAMHR database is divided into 3 general categories of data that correspond to the data fields of the Adverse Metabolic or Muscular Reaction to Anesthesia (AMRA) Report (Appendix, see Supplemental Digital Content 1, http://links.lww.com/AA/A601): (1) preoperative patient demographics and risk factors; (2) clinical characteristics of the acute MH event; and (3) postevent clinical characteristics and outcomes. The database was supplied to us in an Excel spreadsheet format (Microsoft, Redwood, WA).
To test our hypotheses, we divided eligible subjects into 3 age-based cohorts: 0 to 24 months (youngest age group), 25 months to 12 years (middle age group), and 13 to 18 years (oldest age group). Comparisons of variables of interest among age cohorts were performed with Stata IC 10.1 (College Station, TX), using the Kruskal-Wallis test to analyze continuous variables and χ2 or Fisher exact test to compare categorical variables. Statistical significance was defined as P < 0.05.
There were 351 subjects aged 18 years and younger in the NAMHR, stemming from events that occurred between 1960 and 2011. When these records were subjected to a screen based on the MH CGS score, we obtained 264 (75.2%) records that met the criteria for MH as “very likely” or “almost certain.” Of these, 35 belonged to the youngest age group, 163 in the middle age group, and 66 in the oldest age group.
Preoperative Patient Demographics and Risk Factors
Most subjects in our cohort were male; although this predominance held across age cohorts, it was more pronounced in the oldest age group (P = 0.04) (Table 1). Most subjects were Caucasian; no significant differences were noted in racial composition among age groups. Although the youngest cohort trended towards a higher family history of MH (P = 0.09), there appeared to be no differences between the age groups with regard to family history of MH-related events or illnesses. Forty-nine subjects (18.5%) had a family history of MH or other MH-related illness, but this did not significantly differ by age group. The youngest age group trended toward being more likely to have generalized muscle weakness (P = 0.07) and was noted to have a higher incidence of undescended testes (P = 0.04) and inguinal hernia (P = 0.04). These differences are likely related to a surgical case-based bias. No difference was noted in the number of prior anesthetic procedures, but the MH event was the first procedure for 50% of patients and the second procedure for 20% (not shown in Table 1).
The case type by service was distributed across specialties, with cases most commonly involving ear, nose, and throat (29%) and orthopedics (21.2%). The type of case was variable by age group, with urological cases accounting for nearly 30% of the youngest group events (P ≤ 0.001), ear-nose-throat cases accounting for nearly 40% of the middle group events (P ≤ 0.001), and orthopedic cases accounting for 50% of oldest cohort events (P ≤ 0.001). Cases for the youngest group were nearly all scheduled, while the adolescent cohort had a higher percentage of emergent procedures (P ≤ 0.001). Although denominators for non-MH cases are not available, these case distributions are likely reflective of a typical case distribution in children by age group.
Most children had induction of general anesthesia with IV drugs, although the youngest group was more likely to receive an inhaled induction (P = 0.003). As expected, IV induction was significantly associated with emergent cases (97.4% vs 80.9%, P = 0.009). The youngest age group was less likely to receive IV succinylcholine (P ≤ 0.001), vecuronium (P = 0.001), propofol (P ≤ 0.001), or fentanyl (P < 0.001) but was more likely to receive halothane (P < 0.001) or sevoflurane (P = 0.013) compared with the older cohorts. These results are also likely explained by normal age-based clinical practice.
Sinus tachycardia, hypercarbia, and rapid temperature increase were the most commonly observed physical examination findings (observed in 73.1%, 68.6%, and 48.5%, respectively) (Table 2); the oldest cohort was more likely to develop these findings. The oldest age group took longer to reach their maximum end-tidal CO2 value, had higher peak potassium values, and was more likely to demonstrate sweating during the event. The middle age cohort had a lower maximum end-tidal CO2 and blood gas PCO2 compared with the youngest and oldest age groups. The middle age cohort was more likely to develop masseter spasm, and the youngest age cohort was more likely to develop skin mottling. Although no significant differences were noted in the incidence of generalized muscle rigidity, dark colored urine, tachypnea, cyanosis, ventricular arrhythmias, or overall temperature increase among age groups, the youngest age cohort was approximately half as likely to demonstrate muscle rigidity.
Masseter spasm was significantly more common in children who received succinylcholine (75/125 vs 11/128, P < 0.001). However, there were differences when analyzed separately by age group. For example, although infrequently used in subjects in the youngest age group, masseter spasm and subsequent MH occurred in 4 of the 5 subjects in this age group who received succinylcholine. In the oldest age group, there was no significant difference in appearance of masseter spasm between the subjects who did and did not receive succinylcholine (P = 0.17).
Subjects in the oldest age cohort reached a higher maximum temperature compared with those in the youngest and middle cohorts; subjects in the youngest age group tended to take longer to reach their maximum value (P = 0.07), and they tended to have a lower blood pH (P = 0.06). Subjects in the youngest age group also demonstrated significantly higher peak lactic acid levels (P = 0.0025) and lower peak creatine kinase (P ≤ 0.001) values.
Treatments and Outcomes
For nearly all treatments, regimens were similar across age cohorts. The oldest age cohort was more likely to receive active cooling (they had higher temperatures) and fluid administration (Table 3). The older cohort more commonly received glucose and insulin in their treatment, which likely related to their higher potassium values. Dantrolene was administered in >73% of events, with the average initial and total doses being 2.4 mg/kg and 5.9 mg/kg, respectively. These doses were similar across groups. Overall, 21.6% of patients reported side effects after dantrolene administration, the most common of which was muscle weakness, which was more commonly reported in the oldest age cohort (the youngest group would not have been able to report such an effect).
There were 10 MH-associated deaths in the entire cohort. Six of these deaths occurred in the middle age group (3.7%), and 4 occurred in the oldest age group (6.1%). Five of these patients (50%) received dantrolene during the course of their treatment. In examining these cases further, the deaths occurred between 1990 and 2010. Factors associated with death, such as drugs, are impossible to determine because of the way knowledge of MH and its treatments changed over time. Other major complications were uncommon, with cardiac dysfunction having the highest incidence at 4.5%. Such events tended to be more common in the oldest age group (P = 0.06).
Recrudescence of symptoms after initial treatment occurred in 14.4% of cases, with no difference across age cohorts. Two of these cases were fatal despite treatment with dantrolene: one in the middle age group (4-year-old, recrudesced 6 hours after the initial event), and 1 in the oldest age group (14-year-old, recrudesced 24 hours after the initial event).
In this retrospective analysis of pediatric MH cases reported to the NAMHR, we examined the predisposing factors and clinical characteristics of acute MH episodes and contrasted these findings among 3 broad age groups across the spectrum of childhood. Our main hypothesis, that there are age-based differences in clinical characteristics of acute MH, was based on knowledge of differences in body composition, specifically percentage of skeletal muscle, throughout childhood. During development, the most significant periods of skeletal muscle growth are after the stage of infancy and then again at around the age of puberty.5,6 Since MH is primarily a disease of muscle that predisposes to MH susceptibility and determines severity of the acute MH response, we postulated that older children would have more severe effects of acute MH than younger children. Our results appear to be consistent with this hypothesis.
Preoperative risk assessment for MH susceptibility has been traditionally based on knowledge of case reports of MH-associated diseases, and more recently, genetic linkage studies.7 For individual patients, risk assessment is based on family history of possible MH events. Although subjects in the youngest age cohort at baseline tended to have more generalized muscle weakness than the older cohorts, most (> 80%) of all MH events occurred in phenotypically normal children without a family history of MH-related comorbidities.
For the entire cohort, the most common initial presenting signs of acute MH were tachycardia and hypercarbia. However, we observed significant differences in additional findings among age groups. In particular, the oldest aged subjects were more likely to develop a higher maximum temperature, a rapidly increasing temperature, higher peak creatine kinase levels, and higher potassium levels. These findings, taken together, indicate greater degrees of rhabdomyolysis, which can be explained by the greater overall percentage of baseline muscle composition in older children. Inherent thermoregulatory differences during development (i.e., age-related differences in rate of heat gain and dissipation) may have also accounted for the significant difference in temperature levels observed across age groups.8,9 The youngest aged subjects developed lower blood gas pH values and higher peak lactic acid levels than older subjects, which may indicate less muscle reserve to buffer the occurrence of anaerobic metabolism during an acute MH event in a population that is known to have higher resting metabolic rates. The middle aged cohort of patients more often displayed masseter spasm but less commonly displayed the standard common findings of the oldest cohort. This heterogeneous response may hinder diagnosis of MH in these cases.
Treatments of acute MH were similar across age groups, except for some differences related to age-based physical findings, such as higher temperatures and higher potassium levels in older subjects, who were also more likely to receive active cooling and glucose/insulin, respectively.
The incidence of death from acute MH in pediatric patients is not well known. Rosero et al.1 reported 3 deaths of 454 pediatric MH cases (0.66%), which was less than their findings of an overall nationwide incidence (including adults) of 11.7%. We demonstrated a higher overall death rate (4.5%) of pediatric patients contained in the NAMHR. Differences in methodological data collection as well as reporting bias among studies are the most likely causes of these inconsistent rates of death. Although we found that the NAMHR contained no pediatric MH deaths before 1990, this was likely biased by differences in recognition of MH before that time. Therefore, our death rate among highly suspected subjects is probably an underestimate.
Recrudescence after treatment was observed in 14.4% of subjects and did not differ by age group. This is slightly less than the overall rate of 20% (19.1% in subjects <12 years) previously reported by Burkman et al.10 Although this latter study also examined subjects included in the NAMHR, our cohorts included additional years of collection and focused on a larger group of pediatric subjects.
There are important limitations of data interpretation when using databases such as the NAMHR. The NAMHR is populated using cases from various different sources, and most of MH cases that occur in North America are not contained in the NAMHR. Furthermore, a number of cases in the database occurred before the creation of the registry (1987) and were entered into the current database in a retrospective fashion. However, we are aware of no plausible reason why these inherent biases would affect one age group over the other.
In summary, our main hypothesis, that there are age-related differences in clinical characteristics of acute MH among different age cohorts during childhood, was confirmed by analysis of subjects in the NAMHR. In general, older subjects with presumed greater muscle mass demonstrated a greater likelihood of higher body temperatures and higher potassium levels, while the youngest subjects had greater levels of metabolic acidosis. Our secondary hypothesis, that we would be able to identify specific groups of children at risk for development of acute MH, was unproven, because nearly all children in all age groups were phenotypically normal before developing MH.
Name: Priscilla Nelson, MD.
Contribution: This author helped design and conduct the study, analyze the data, and write the manuscript.
Attestation: Priscilla Nelson has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Ronald S. Litman, DO.
Contribution: This author helped design and conduct the study, analyze the data, and write the manuscript.
Attestation: Ronald S. Litman has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.
This manuscript was handled by: Peter J. Davis, MD.