2.3 Statistical analysis
We extracted ICSRs between January 1, 2010 and December 31, 2019, with
sugammadex as the ”suspect” drug, excluding repetitive reports. The
information component (IC) was used to detect and quantify the
association between the target drug and suspected ADRs. Originating from
Bayesian confidence propagation neural networks, IC can provide a
conservative correlation measure and reduce the risk of highlighting
spurious associations, especially for events with very low expected
frequencies in large databases (such as VigiBase)11.
IC and the corresponding lower end of the 95% credibility interval
(IC025) were used to calculate the disproportionality.
These parameters compare the proportions of ICSRs of a selected ADR
between patients exposed and not exposed to the target drug. If the
proportion in the exposed patients is significantly higher than in the
control group, a signal is detected. An IC025> 0 is the criterion for generating a signal. A positive
IC025 value is the traditionally used statistical
significance threshold for UMC signal detection, indicating that a
particular drug-ADR combination has a higher frequency than expected;
thus, the ADR has a potential association with the
drug12. A signal with a higher IC value indicates a
strong association between ADR and the drug. An IC value >
3 is defined as a strong signal13. The statistical
formula to calculate IC14 is as follows:
IC = log2[(A + 0.5)/(Nexpected +
0.5)]
Nexpected = (A + B) × (A + C)/(A + B + C + D)
IC025 = IC − 3.3 × (A + 0.5)-1/2 − 2 ×
(A + 0.5)-3/2
IC975 = IC + 2.4 × (A + 0.5)-1/2 − 0.5
× (A + 0.5)-3/2,
where A is the number of target ADRs in patients using the target drug,
B is the number of other ADRs in patients using the target drug, C is
the number of target ADR in patients using other drugs, D is the number
of other ADRs in patients using other drugs, and
Nexpected is the number of case reports expected for the
drug-adverse effect combination. A, B, C, and D were obtained as the
frequencies of ICSRs calculated from VigiBase.
Results
Descriptive analysis
A total of 16,219,410 adverse events were reported by patients receiving
any drug treatment and included in VigiBase between January 1, 2010 and
December 31, 2019. After data deduplication and pre-processing, a total
of 2,032 patients with ADRs and sugammadex as the suspected drug were
identified. A summary of the demographic data is presented in Table 1.
Comprehensive spectrum of sugammadex-related ADRs
The disproportionality analysis of ADR reports in the full database
revealed a total of 94 sugammadex-related positive signals (Fig. 1).
They mainly involved the system organ classes (SOCs) of respiratory,
thoracic, and mediastinal disorders; cardiac disorders; injury;
poisoning and procedural complications; and investigations. A high
signal intensity was noted for the recurrence of neuromuscular blockade
(n = 54, IC: 6.74, IC025: 6.33), laryngospasm (n = 53,
IC: 6.05, IC025: 5.64), bronchospasm (n = 119, IC: 5.63,
IC025: 5.36), and bradycardia (n = 169, IC: 5.13,
IC025: 4.90).
Relationship between the sugammadex-related ADRs and
patient age
The ICSR patients were divided into age-based groups (0–17 years,
18–44 years, 45–64 years, and ≥ 65 years), and an IC analysis was
performed for each age group. The IC025 value was
obtained for each group (Figs. 2 and 3); the ANOVA test showed
statistically significant differences between groups (P <
0.01) with the highest overall signal intensity in the 0–17 years
group. The ADRs most commonly reported by different age groups were
dissimilar. In the 0–17 years group: prolonged therapeutic effect (n =
1, IC: 18.00, IC025: 14.22), fixed pupils (n = 1, IC:
16.42, IC025: 12.63), and negative-pressure pulmonary
oedema (n = 1, IC: 16.29, IC025: 12.50); 18–44 years
group: recurrence of neuromuscular blockade (n = 3, IC: 12.16,
IC025: 10.09), alveolar-arterial oxygen gradient
increased (n = 1, IC: 13.53, IC025: 9.74), and
negative-pressure pulmonary oedema (n = 1, IC: 11.94,
IC025: 8.16); 45–64 years group: recurrence of
neuromuscular blockade (n = 11, IC: 12.44, IC025:
11.41), post-resuscitation encephalopathy (n = 1, IC: 13.41,
IC025: 9.62), and increased airway peak pressure (n = 4,
IC: 10.47, IC025: 8.70); ≥ 65 years group: recurrence of
neuromuscular blockade (n = 9, IC: 12.29, IC025: 11.13),
increased airway peak pressure (n = 2, IC: 10.45, IC025:
7.87), and prolonged neuromuscular block (n = 3, IC: 9.24,
IC025: 7.17).
Onset time of sugammadex-related ADRs
The time to onset is the time from the start of medication
administration to an ADR. A total of 1,118 sugammadex-related ADRs were
reported with onset time in the database and 1,996 total ADRs. Among
these, 68.9% (n = 1376) occurred within half an hour, 88.3% (n = 1763)
within 1 day, and < 3% after 7 days. ADRs with significant
positive signals, such as the recurrence of neuromuscular blockade,
laryngospasm, and bronchospasm, mainly occurred within 1 day after the
administration.
Seriousness of sugammadex-related ADRs
One patient may correspond to more than one ADR, resulting in more than
one outcome, and there are a total of 3717 outcomes of 2032 patients.
Among the 3717 ADR outcomes reported, 53 were fatal; the most frequent
fatal ADR was death (9/53, 17.0%), and the SOC with the most frequent
fatal cases was cardiac disorders (21/53, 39.6%). All the fatal ADRs
occurred in 27 patients, 1.33% of sugammadex-related ICSRs; the
majority of these patients were over 65 years of age (12/27, 44.4%),
and the number of men and women was similar.
Discussion
To our knowledge, this study is the most extensive safety analysis of
sugammadex in recent years, reporting the characteristics of adverse
reactions associated with sugammadex through a detailed analysis of the
WHO Global Case-safety Reporting Database (VigiBase).
We found 94 positive signals of adverse reactions related to sugammadex
through the disproportionality analysis, mainly involving the
respiratory, cardiovascular, and immune systems. The four ADRs with the
highest signal intensity for sugammadex association were the recurrence
of neuromuscular blockade, laryngospasm, bronchospasm, and bradycardia.
An online app-based study15 in 2018 showed that the
most common ADRs of sugammadex were bradycardia and incomplete
neuromuscular blockade reversal, similar to our findings.
The majority (68.9%) of ADRs occurred within half an hour from
sugammadex administration, suggesting that the patients should be
monitored carefully to detect any adverse reactions during the
anaesthesia recovery period. Nemes et al.16 found that
the incidence of residual postoperative neuromuscular block after
reversal with sugammadex was significantly lower than with neostigmine
or placebo; however, it could still not be avoided entirely. Errando et
al.17 reported that women have a higher incidence of
residual postoperative neuromuscular block than men; furthermore, a
residual block cannot be completely avoided without neuromuscular
monitoring, regardless of the antagonism strategy. Therefore, we
recommend neuromuscular function monitoring throughout anaesthesia and
that the reversal of muscle relaxation should always be driven by the
monitored data. Appropriate reversals in terms of medication, dose, and
timing should never blindly follow established rules.
We found that sugammadex-related ADRs have the highest fatality rate
involving cardiac disorders. At the same time, cardiac disorders are
also the SOC with the highest frequency of ADRs in association with
sugammadex. Several cases18-21 of severe bradycardia
and cardiac arrest have been reported clinically, and Hunter et
al.22 found that since 2016, the number of serious
adverse cardiac events reported after sugammadex in the Food and Drug
Administration (FDA) Adverse Event Reporting System has greatly exceeded
that after neostigmine. At present, the exact mechanism of
sugammadex-induced bradycardia and asystole is still unclear; however,
Kalkan et al. found that both low and high doses of sugammadex can cause
significant histopathological changes in cardiomyocytes and other
harmful effects. Nonetheless, this finding suggests that
anaesthesiologists should use sugammadex cautiously in patients with
underlying cardiovascular diseases during clinical medication. They
should also conduct comprehensive electrocardiography (ECG) and
hemodynamic change monitoring after the medication.
We stratified the reports of adverse events by age and found that
patients aged 0–17 years were the group with the highest risk of ADRs
due to saccharides. However, in contrast to our findings, Gaver et
al.23 reported that sugammadex was as effective and
safe in the paediatric population (age < 19 years) as in
adults, whereas Honing et al.24 showed that elderly
patients were more susceptible to the adverse effects of residual
neuromuscular blockade and had a slower natural recovery. In addition,
it is worth noting that pupil fixation, a common drug toxicity reaction,
had a strong positive signal in the paediatric group, though not in the
entire database, and we suspect that this finding may be related to
renal immaturity and delayed drug metabolism in children. However, there
were some cases where the number of ADR was less than three in the
age-stratified disproportionate analysis, suggesting that the analysis
of the adverse reactions of sugammadex in different age groups requires
a larger amount of data. Currently, sugammadex has not been approved by
the FDA for use in children. There are limited data on the use of
sugammadex in children, especially infants. Further paediatric studies
are required to fully determine the safety of sugammadex in children. We
acknowledge that the VigiBase database has some limitations. Incomplete
reports of ADRs and lack of complete clinical information are common
limitations of pharmacovigilance studies. First, adverse event reporting
is voluntary and performed by various sources (e.g., doctors,
pharmacists, and other clinicians), thus increasing the risk of
incomplete information. However, 130 countries have contributed to the
database, thus ensuring a comprehensive assessment from different
clinical settings. Second, no detailed clinical information and
diagnostic criteria were available; thus, our assessment was limited to
reports from treating clinicians, which may be subject to various
biases, including underreporting (reporting only the most severe or
obvious cases) or overreporting (reporting cases without a clear
diagnosis). Inevitably, reporting of adverse events is more likely soon
after the launch of a new drug than during regular, long-standing use.
In addition, it is important to note that when conducting ADR studies,
the IC value does not indicate a causal relationship between the target
drug and the suspected ADRs; it only shows a quantitative
association13. Therefore, prospective studies and
long-term validation of these findings are required. Despite the
limitations of VigiBase, the analysis of adverse reactions in the
pharmacovigilance database remains an important tool for drug safety
studies and post-marketing drug monitoring. It allows for signal
detection in large populations and can provide significant opportunities
for monitoring drug safety and identifying new, rare signals.
In conclusion, research based on spontaneous ADR reporting is an
important modality for drug safety research and post-marketing drug
monitoring. In the safety profile of sugammadex, our result was similar
to that of the post-marketing clinical trials. We found that adverse
effects of sugammadex were at higher risk in adolescents and more severe
in older patients, for which further research should be performed. In
addition, our findings highlight important concerns about the time onset
of adverse events and cardiac safety, and we suggest that
anesthesiologists should carefully monitor the anesthesia recovery
period, especially for ECG and hemodynamic changes in patients with
underlying cardiovascular diseases.