Neuromuscular Blocking Agents and Reversal Agents Among Hospitalized Children: A Cerner Database Study

Abstract

Background: Sugammadex (Bridion) was approved by the US Food and Drug Administration (FDA) in December 2015 for the reversal of neuromuscular block (NMB) induced by rocuronium and vecuronium bromide in adults undergoing surgery and approved for use in both adults and children in the European Union in 2008. Sugammadex use in children has been reported in the United States, but to what extent is not clear. Aims: The aim was to describe the utilization pattern of NMB agents and factors associated with the use of reversal agents (neostigmine and sugammadex) in US children. Methods: Cross-sectional study of children with exposure to NMB agents between 2015 and 2017 in the Cerner Health Facts^® database, which is an electronic health record (EHR) database across 600 facilities in the United States. Logistic regression estimated factors associated with the use of sugammadex vs neostigmine. Results: A total of 27 094 pediatric clinical encounters were exposed to neuromuscular blocking agents (NMBAs), in which 21 845 were exposed to rocuronium (76%), vecuronium (18%), or both (6%). Among children with exposure to rocuronium and vecuronium, the use of sugammadex was 1.7% in 2016 and 7.6% in 2017. The multivariable logistic model suggested that children who were older (age 12-17 years vs 0-1 year; odds ratio [OR] 1.96; 95% confidence interval [CI], 1.36-2.83), Hispanic or Latino ethnicity and other ethnicities (vs non-Hispanic or Latino; OR 2.03 and 1.56; 95% CI, 1.55-2.67 and 1.15-2.13, respectively), in teaching facilities (OR 1.26; 95% CI, 1.00-1.59), or admitted through emergency departments (OR 1.65; 95% CI, 1.06-2.58) were independently more likely to receive sugammadex than neostigmine after controlling for other covariates. Conclusions: In Cerner Health Facts database 2015 to 2017, among children, rocuronium was more commonly used than vecuronium, and sugammadex use was observed since 2016. Sugammadex and neostigmine users varied by demographic, clinical, and site-level characteristics.

Introduction

Muscle relaxants play an important role in general anesthesia through neuromuscular blockade. Reversal of this effect is often needed to prevent postoperative residual paralysis. In 2008, sugammadex (Bridion) was approved in the European Union for the reversal of muscle relaxants rocuronium or vecuronium; then, in 2010 and 2015, it was approved in Japan and the United States, respectively. Different from neostigmine, a cholinesterase inhibitor, sugammadex, is a thiolated γ-cyclodextrin derivative which selectively encapsulates rocuronium or vecuronium to reverse their neuromuscular blocking action.

The indicated population of sugammadex in the European Union includes both adults and children (2-17 years of age). In the United States, the Food and Drug Administration (FDA) requested a postmarking randomized, controlled trial to evaluate the efficacy, safety, and pharmacokinetics of sugammadex in children aged 0 to 17 years upon the approval of sugammadex in adults. To date, sugammadex has not been approved for children by the FDA but has still been off-label administered in children in the United States. ¹

The current real-world study examines some important aspects of anesthesia management in children, including the use of neuromuscular blocking agents (NMBAs) and neuromuscular block (NMB) reversal agents, the time trend of using sugammadex and neostigmine, and factors associated with the use of sugammadex vs neostigmine among pediatric NMBA users.

Methods

Data Source

We used Cerner Health Facts^® database (Cerner database) for this study. Cerner database is a hospital- and clinic-based electronic medical record database launched in 2000 that contains pharmacy, laboratory, clinical, and billing data from about 50 million patients across over 600 facilities in the United States. As of November 2014, the Cerner database included greater than 47 million unique patients, in whom 8.2 million were inpatients, 15.9 million were emergency patients, and 37.5 million were outpatients. The facilities span the entire United States, of which 27% were in the Midwest, 40% were in the Northeast, 21% were in the South, and 11% were in the West. With regard to facility sizes, 22% of the facilities had less than or equal to 99 beds, 55% had 100 to 499 beds, and 23% had greater than or equal to 500 beds. Sixty percent of the hospitals and clinics were teaching facilities.

Study Population

This study includes 2 parts. In the first part, we examined the use of NMBAs in children. The study population was children (age 0-17) with at least 1 valid pharmacy record of NMBAs during 2015 to 2017. The following agents were specified as NMBAs: pancuronium, vecuronium, rocuronium, atracurium, cisatracurium, mivacurium, doxacurium, pipecuronium, rapacuronium, tubocurarine, and succinylcholine.

In the second part, we examined the use of NMB reversal agents in children (0-17 years) who received rocuronium or vecuronium (at least 1 valid pharmacy record of rocuronium or vecuronium during 2015-2017). Sugammadex and neostigmine were specified as NMB reversal agents.

The Measurement of Exposure

To define accurately the study population, we excluded pharmacy records of NMBAs or reversal agents if the dose quantity, order strength, or total dispensed doses were zero or missing due to the uncertainty whether these drugs were administered to the patients or not. Patients were defined as “exposed” if having at least 1 NMBA or NMB reversal agent pharmacy record.

Statistical Analysis

In this study, the unit of analysis was encounter, not unique patient, such that 1 patient may have multiple encounters. The proportion of encounters using 1 specific NMBA is calculated as the number of encounters with the use of that specific NMBA divided by the number of encounters with the use of any NMBA. Other proportions are calculated in a similar way. Clinical Classification Software (CCS) developed by Healthcare Cost and Utilization Project (HCUP) was used to group diagnoses and procedures (https://www.hcup-us.ahrq.gov/tools_software.jsp). Multivariate logistic regression was used to evaluate factors associated with use of sugammadex vs use of neostigmine. The factors included NMBA choice (rocuronium vs vecuronium), gender, race (3 categories: Caucasian, African American, and other/unknown), ethnicity (3 categories: not Hispanic or Latino, Hispanic or Latino, and Other/unknown), age (4 age groups: 0-1, 2-5, 6-11, and 12-17), calendar year (2016 vs 2017), facility status (3 categories: teaching facility, nonteaching facility, and unknown status), and hospital admission source (5 categories: referral, emergency, newborn, transfer, and other/unknown); adjusted ORs were reported.

All analyses were completed with SAS 9.4 (Cary, NC).

Results

NMBA Exposure Among Pediatric Encounters

A total of 27 094 encounters had exposure to NMBAs, representing about 0.1% of all pediatric encounters in the 2015 to 2017 Cerner database. The demographic characteristics of the study population are shown in Table 1. The largest groups of pediatric NMB users were adolescents (age 12-17), and about 70% of the encounters occurred in teaching facilities. Although there was a high proportion of missing information, the available data showed that “respiratory intubation and mechanical ventilation” and “appendicitis and other appendiceal conditions” were the most common procedures and diagnoses in pediatric encounters with NMBA use, respectively (Table 1).

Table 1.

Demographic Characteristics of Children Who Used at Least 1 NMBA During Clinical Encounter.

	No.	%
All encounters	27 094	100
Gender
Female	11 854	43.8
Age, y
0-1	8108	29.9
2-5	3855	14.2
6-11	4957	18.3
12-17	10 174	37.6
Race
Caucasian	13 385	49.4
African American	5506	20.3
Other/unknown	8203	30.3
Ethnicity
Hispanic or Latino	3849	14.2
Not Hispanic or Latino	19 978	73.7
Other/unknown	3269	12.1
Hospital/clinic type
Teaching	18 506	68.3
Nonteaching	6817	25.2
Unknown	1771	6.5
Admission source
Referral	14 132	52.2
Other/unknown	6770	25.0
Transfer	4815	17.8
Emergency department	1068	3.9
Newborn	309	1.1
Discharge destination
Home/hospice	22 650	83.6
Transferred to another facility	1518	5.6
Expired	1260	4.6
Other/unknown	1666	6.2
Length of stay (days)
Q1	2
Median	3
Q3	13
Primary procedure at discharge (top 5)
Missing	15 964	58.9
Other diagnostic procedures	502	1.9
Respiratory intubation and mechanical ventilation	1752	6.5
Appendectomy	958	3.5
Dental procedures	564	2.1
Other Operation Room heart procedures	405	1.5
Tonsillectomy and/or adenoidectomy	380	1.4
Primary diagnosis at discharge (Top 5)
Missing	11 133	41.1
Residual codes; unclassified	641	2.4
Appendicitis and other appendiceal conditions	1329	4.9
Cardiac and circulatory congenital anomalies	679	2.5
Disorders of teeth and jaw	636	2.3
Respiratory failure; insufficiency; arrest (adult)	635	2.3
Other congenital anomalies	429	1.6

NMBA = neuromuscular blocking agent.

From 2015 to 2017, the NMBAs used in children included rocuronium, vecuronium, succinylcholine, cisatracurium, atracurium, and mivacurium. The most commonly used were rocuronium (more than 60%), followed by vecuronium (about 20%) and succinylcholine (26%-27%; Table 2). The utilization pattern of NMBAs was stable over time, with slight increase in rocuronium (64.3%-66.2%) and vecuronium (18.7%-22.0%) and slight decrease in cisatracurium (4.7%-3.2%) and atracurium (0.8%-0.1%) between 2015 and 2017.

Table 2.

Choices of NMBAs Among the Pediatric Clinical Encounters With Exposure to NMBAs. ^a

Year (No. of patients)	2015 (n = 9804)		2016 (n = 9048)		2017 (n = 8242)
	No.	%	No.	%	No.	%
Rocuronium	6306	64.3	6055	66.9	5456	66.2
Vecuronium	1833	18.7	1663	18.4	1814	22.0
Succinylcholine	2565	26.2	2445	27.0	2184	26.5
Cisatracurium	461	4.7	473	5.2	267	3.2
Atracurium	81	0.8	11	0.1	11	0.1
Pancuronium	2	0.0	2	0.02	2	0.02
Mivacurium	0	0	0	0	1	0.01

NMBAs = neuromuscular blocking agents.

^aNMBA exposures are not mutually exclusive.

We compared the rocuronium and vecuronium users and found that these 2 populations have significantly different age distribution and indications, as predicted from known immaturity of drug excretion mechanisms in infancy. In vecuronium users, 62.2% were infants and 17.1% were adolescents, while in rocuronium users, the proportions were 23.5% and 44.1%, respectively. Vecuronium was more likely to be used for respiratory insufficiency than rocuronium (14.3% vs 5.3%), though there was high proportion of missing information in the primary procedure records. The hospital mortality rate was also higher among vecuronium users (13.4%) than rocuronium users (2.9%). See Supplemental Table 1 on the detailed comparison of the rocuronium and vecuronium users.

Use of NMB Reversal Agents

From 2015 to 2017, the proportion of children receiving neostigmine for rocuronium and/or vecuronium decreased from 32% to 28% and the proportion of children without receiving neostigmine or sugammadex decreased from 68% to 64%. The use of sugammadex increased from 1.7% in 2016 to 7.6% in 2017, regardless of the formulary status (Table 3). We also stratified the analysis by rocuronium vs vecuronium users and found that a much larger proportion of rocuronium users received reversal agents compared to vecuronium users (40.0% vs 7.6%; Table 4). This is expected as vecuronium use probably is for infants intubated in ICUs and just wears off slowly, while rocuronium use mostly is for patients undergoing surgical procedures where a reversal agent is needed for extubation.

Table 3.

Use of Reversal Agents Among Pediatric Clinical Encounters With Exposure to Rocuronium and/or Vecuronium, by Year.

	2015 (n = 7743)		2016 (n = 7315)		2017 (n = 6787)
	No.	%	No.	%	No.	%
Sugammadex only	0	0	124	1.7	518	7.6
Neostigmine only	2496	32.2	1940	26.5	1922	28.3
Both were used	0	0	4	0.1	27	0.4
Neither was used	5247	67.8	5247	71.7	4320	63.7

Table 4.

Use of Reversal Agents Among Pediatric Clinical Encounters With Exposure to Rocuronium and/or Vecuronium, by NMBA.

	Rocuronium only (n = 16 535)		Vecuronium only (n = 4028)		Both rocuronium and vecuronium (n = 1282)
	No.	%	No.	%	No.	%
Sugammadex only	576	3.5	40	1.0	26	2.0
Neostigmine only	6030	36.5	257	6.4	71	5.5
Both were used	26	0.2	3	0.1	2	0.2
Neither was used	9903	59.9	3728	92.6	1183	92.3

NMBA = neuromuscular blocking agent.

Factors Associated With the Use of Sugammadex

We conducted a multivariable logistic regression to assess factors associated with sugammadex vs neostigmine use for clinical encounters in 2016 to 2017 (Table 5). Compared to 2016, the odds ratio (OR) for 2017 was as high as 4.96 (95% confidence interval [CI] 4.01, 6.15), which is consistent with the trend analysis shown in Table 3 and not surprising, given the timing of FDA approval (December 15, 2015). The choice of NMBA (rocuronium vs vecuronium) was not associated with the use of sugammadex. Children who were Hispanic or Latino, older (12-17 years), admitted to a teaching facility, or admitted from the emergency department (ED) were more likely to receive sugammadex than neostigmine, while children who were neither Caucasian nor African American (ie, those who were unknown or other races) were less likely to receive sugammadex than neostigmine as the reversal agent. The c-statistic of the logistic regression is 0.753.

Table 5.

Odds Ratio of Receiving Sugammadex vs Neostigmine, Results From Multivariable Logistic Regression. ^a

	Odds ratio	95% Confidence interval
NMBA
Rocuronium	1
Vecuronium	1.25	0.92-1.70
Gender
Male	1
Female	0.95	0.8-1.14
Race
Caucasian	1
African American	0.94	0.73-1.20
Other/unknown	0.49	0.39-0.63
Ethnicity
Not Hispanic or Latino	1
Hispanic or Latino	2.03	1.55-2.67
Other/unknown	1.56	1.15-2.13
Age, y
0-1	1
2-5	0.71	0.43-1.17
6-11	0.84	0.56-1.27
12-17	1.96	1.36-2.83
Year
2016	1
2017	4.96	4.01-6.15
Teaching facility
No	1
Yes	1.26	1-1.59
Unknown	0.22	0.08-0.63
Admission source
Referral admission	1
Emergency department admission	1.65	1.06-2.58
Newborn admission	2.58	0.66-10.02
Transfer admission	1.31	0.97-1.76
Other/unknown	2.15	1.77-2.61

NMBA = neuromuscular blocking agent.

^aThe multivariable logistic regression model included all the variables listed in the table.

The demographic and clinical characteristics of neostigmine users and sugammadex users in 2016 to 2017 were also compared numerically (see Supplemental Table 2). Sugammadex users were older (proportion for age group 12-17: 77% vs 58% in 2016, 74% vs 56% in 2017), more likely to be in teaching facilities (75% vs 68% in 2016, 81% vs 72% in 2017), less likely to be discharged to home (71% vs 84% in 2016, 70% vs 79% in 2017), and were more likely to be discharged to unknown destinations (22% vs 8% in 2016, 12% vs 3%). In 2017, sugammadex users had longer length of stay (LOS) than neostigmine users (in 2017, Q1 and Q3 were 1 and 4 days for sugammadex vs 1 and 3 days for neostigmine, respectively). Among those with available information, the most commonly recorded diagnosis/procedure was appendicitis/appendectomy for both sugammadex users and neostigmine users.

Discussion

This drug utilization study described the use of NMBAs and reversal agents in children. We have 2 key findings: First, rocuronium is the most commonly used NMBA among children; second, sugammadex is being used in children in US hospitals and clinics: In 2017, 7.6% of pediatric rocuronium or vecuronium encounters had sugammadex as the reversal agent.

This study agrees with a previous one in demonstrating more frequent use of rocuronium than vecuronium in children. ² Vecuronium is a popular choice for use in critically ill children because of minimal cardiovascular effects and histamine release, and rocuronium is a desacetoxy analog of vecuronium with a more rapid onset of action.^3,4 We found notable differences between rocuronium and vecuronium users in our study: The higher proportion of patients with mechanical ventilation may lead to the lower proportion of using reversal agents in vecuronium than rocuronium users, as reversal agents are used less frequently in ICU patients being mechanically ventilated than in surgical patients. However, we cannot confirm this in our study because the Cerner database does not have information whether the use of NMB was in ICU or in the operating room.

Compared to neostigmine users, sugammadex users were older and were more likely to have been admitted to teaching facilities. As sugammadex is a new drug for NMB reversal (and not indicated for pediatric use in the United States) and younger children are more vulnerable to adverse events and potential risks, physicians tend to take a more conservative approach for younger children. ⁵ Therefore, physicians might be more likely to prescribe sugammadex for older children. Moreover, sugammadex was more likely to be used in children aged 12 to 17 years, the age for which sugammadex is indicated and approved for use in most places outside the United States. Also, teaching facilities may adopt new therapies faster than nonteaching facilities;^6,7 thus, most sugammadex uses were found in teaching facilities in this study. In crude comparisons, we observed similar trends to those in multivariable adjusted analyses. In addition, we found sugammadex patients had a slightly longer LOS, with identical medians and 1 extra day at the 75th percentile in 2017. This might be related to the difference in procedure types and/or disease severity between sugammadex and neostigmine users. We speculate that patients who were sicker and underwent more complex procedures might be more likely to receive sugammadex due to its more rapid onset of action and the perception of a better safety profile than neostigmine. ⁸ The large degree and disparity in missingness of diagnosis and procedure codes data thwart efforts to probe this possibility further. A recent US study in a children’s hospital also reported that children who received sugammadex were older and sicker (higher American Society of Anesthesiologists [ASA] score) than children who received neostigmine. ¹

There have been systematic reviews which summarized results of randomized clinical trials on the efficacy and safety of sugammadex for reversing postoperative residual neuromuscular blockade in children,^9,10 suggesting sugammadex may be fast and effective in reversing rocuronium/vecuronium-induced NMB in children. The FDA required postmarking randomized controlled trials among children are currently ongoing and will provide further evidence on the efficacy, safety, and pharmacokinetics of sugammadex in children. In real world, there have been various case reports documenting the use of sugammadex in difficult anesthetic conditions in children, including cardiac transplantation, ¹¹ Duchenne muscular dystrophy,^12,13 Sjögren-Larsson syndrome, ¹⁴ phosphomannomutase-2 congenital disorder of glycosylation (PMM2-CDG), ¹⁵ after bronchial foreign body removal, ¹⁶ type I mucopolysaccharidosis, ¹⁷ neonate, ¹⁸ renal transplantation, ¹⁹ “cannot ventilate, cannot intubate” scenario, ²⁰ congenital myotonic dystrophy type 1, ²¹ and myasthenia gravis. ²² Despite challenges, most of the cases were managed successfully. These case reports highlight that the pediatric population receiving sugammadex as the NMB reversal agent may be very broad and varied greatly in diseases and clinical conditions. A comparative effectiveness study reviewed medical records from a children’s hospital and reported that the sugammadex group had fewer cases of bradycardia than the neostigmine group after matching on age and procedure category. ¹

The strengths of this study include large electronic health record (EHR)-level database and multiple utilization indicators of NMBAs and NMB reversal agents. There are limitations of this study. First, medications from floor stock or from a crash cart might not appear in the pharmacy order data, and therefore, the overall prevalence of NMBAs and NMB reversal agents may be underestimated. Second, some important variables have a high proportion of missing data, for example, the diagnosis and procedure records, which restricted further analyses. The missingness could be due to the restriction of records to a third-party billing system that Cerner may not have access to, or an outpatient reoccurring encounter. Finally, the data from Cerner database may not be nationally representative, and the results may not be generalizable to hospitals that were not included in the Cerner database.

Conclusion

In Cerner 2015 to 2017, among children, rocuronium was more commonly used than vecuronium, and sugammadex use was observed since 2016. In 2017, 7.6% of children hospitalizations exposed to rocuronium or vecuronium used sugammadex as the reversal agent. Sugammadex and neostigmine users varied by demographic, clinical, and site-level characteristics.