Identification of risk factors of severe hypersensitivity reactions in general anaesthesia

Patients

A total of 193 patients (71 males and 122 females, mean age 37.5?±?20.8 years) with
peri-operative hypersensitivity reactions during general anaesthesia were included
in this retrospective observational cohort study. A summary of patients’ demographic
and clinical data are shown in Table 2. A total of 100/193 patients (51.81%; 39 males and 61 females) had reactions of grade
I, 32/193 patients (16.58%; 12 males and 20 females) had grade II reactions, 55/193
patients (28.49%; 20 males and 35 females) had grade III reactions, and 6/193 patients
(3.10%; only females) had grade IV reactions. For the statistical analysis, the patients
were further classified into two groups: group 0 consisting of 132/193 patients (68.40%)
(mean age: 35.16?±?19.58 years) with grade I and II and group 1 consisting of 61/193
patients (31.60%) (mean age: 42.56?±?22.64 years) with grade III and IV reactions.
NMBAs skin tests were performed in 130/193 patients (67.35%, 47 males and 83 females,
mean age 38.95?±?20.49 years), and they were positive in 27/130 patients (20.76%).
In each of the 130 cases, the severity group of the reaction was 0 (grade I?+?II)
in 85/130 patients (65.38%) and 1 (grade III?+?IV) in 45/130 patients (34.61%). The
respective contribution of each neuromuscular blocking drug is shown in Table 3. We carried out skin tests with five different NMBAs, as cross-reactions occurs commonly
among these drugs, in order to identify a safe alternative agent for subsequent anaesthesia.
We did not report our cross-reactivity results because of the small number of positive
skin tests to NMBA in our patients. Serum measurements of succinylcholine specific
IgE levels were performed in 14/16 patients (87.5%) who were exposed to this agent
during general anaesthesia; all of the patients were negative. Skin tests with other
anaesthetic agents, as reported in Table 4, were positive in 3 patients for ketamine, and in 1 patient for atropine. SPTs to
latex were positive in 6/193 patients (3.10%). Serum specific latex-IgE assay was
positive in 16/68 patients (23.52%). Investigations for beta-lactams allergy were
positive in 11/89 patients (12.35%).

Table 2. Demographic and clinical characteristics of patients with a history of reaction during
general anaesthesia

Table 3. Positivity to NMBAs skin tests

Table 4. Skin test results to agents other than NMBAs involved in hypersensitivity reactions
during anaesthesia

Allergic diagnostic work-up

At the end of the allergy work-up, a diagnosis of IgE-mediated hypersensitivity reaction
was established in 55 cases (28.50%), the remaining 138 cases (71.50%) were considered
as non-IgE-mediated reactions. The different clinical symptoms of IgE- and non IgE-mediated
reactions observed during general anaesthesia are reported in Table 5. In our group clinical expression was not more severe in patients with a documented
IgE-mediated reactions to general anaesthetics (OR: 0.906, CI95%: 0.519 to 1.581;
Wald’s test: p?=?0.729). However, cardiovascular symptoms (bradycardia, hypotension,
cardiovascular collapse, cardiac arrest) reported in 29/193 patients (15.03%), were
more frequent in IgE-mediated reactions (27.27% vs 10.14%; Fisher’s exact test: p?=?0.006).
No significant difference was observed between IgE- and non-IgE-mediated reactions
as regards mucocutaneous symptoms (erythema, or urticaria, or angioedema) (67.27%
vs 74.63%; Fisher’s exact: p?=?0.372). Bronchospasm was recorded more often in patients
affected by asthma (30.65% vs. 13.74%; Fisher’s exact test: p?=?0.010), but not in
IgE-mediated reactions in particular (38.18% vs. 29.71%; Fisher’s exact test: p?=?0.306).

Table 5. Patients’ distribution according to clinical symptoms and to mechanism of hypersensitivity
reactions during anaesthesia

History of hypersensitivity to antibiotics (OR: 2.024, CI95%: 1.076 to 3.807; Wald’s
test: p?=?0.029) was more common in IgE- than in non IgE-mediated reactions.

No significant differences were observed between the two groups (IgE- and non-IgE-mediated
reactions) as regards age (OR: 0.999, CI95%: 0.984 to 1.014; Wald’s test: p?=?0.912),
sex (OR: 0.682, CI95%: 0.363 to 1.280; Wald’s test: p?=?0.234), atopy (OR: 0.956,
CI95%: 0.776 to 1.177; Wald’s test: p?=?0.675), asthma (OR: 0.701, CI95%: 0.308 to
1.599; Wald’s test: p?=?0.400) and other respiratory diseases (OR: 0.989, CI95%: 0.246
to 3.969; Wald’s test: p?=?0.988), allergic contact dermatitis (OR: 0.511, CI95%:
0.140 to 1.869; Wald’s test: p?=?0.311), urticaria (OR: 2.093, CI95%: 0.671 to 6.527;
Wald’s test: p?=?0.203), therapy with beta-blockers (OR: 1.485, CI95%: 0.464 to 4,751;
Wald’s test: p?=?0.504), ACE-inhibitors (OR: 1.817, CI95%: 0.600 to 5.496; Wald’s
test: p?=?0.290) or ARBs (OR: 1.154, CI95%: 0.278 to 4.783; Wald’s test: p?=?0.843),
thyroid diseases (OR: 1.485, CI95%: 0.464 to 4.751; Wald’s test: p?=?0.504), serum
tryptase values at baseline (measured in 66/193 patients) (OR: 0.830, CI95%: 0.648
to 1.064; Wald’s test: p?=?0.143), and previous adverse reactions to NSAIDs (OR: 1.164,
CI95%: 0.579 to 2.341; Wald’s test: p?=?0.669).

Risk factors

We looked for risk factors of severe allergic reactions (i.e. grade III and IV) during
the perioperative period:

1.
Age. Each additional year of age increased the risk of severe reaction by 1.7% (OR: 1.017,
CI95% :1.002 to 1.032; Wald’s test: p?=?0.025)

2.
Asthma and antihypertensive medication. A significant association was observed between the development of a severe reaction
during the general anaesthesia and history of asthma (OR: 2.144, CI95%: 1.028 to 4.471;
Wald’s test: p?=?0.042) and ongoing treatment with ACE inhibitors (OR: 4.326, CI95%:
1.383 to 13.530; Wald’s test: p?=?0.012) or ARBs (OR: 4.703, CI95%: 1.134 to 19.498;
Wald’s test: p?=?0.033). Hypertension was significantly more frequent in patients
of group 1 (severity grade III?+?IV) than in patients of group 0 (severity grade I?+?II)
(29.51% vs. 9; Fisher’s exact test: p?=?0.001).

3.
Basal tryptase: sBT tryptase levels were measured in 66/193 patients: values ranged from 1.5 to 14.5 ng/mL
(mean 4.73?±?2.54 ?g/L); 39/66 patients (59.09%) belonged to group 0 (severity grade
I?+?II), and 27/66 patients (40.91%) to group 1 (severity grade III?+?IV). No patient
had sBT levels above 20 ng/ml, a minor diagnostic criteria for mastocytosis. There
was a significant association between sBT and hypertension: any unitary increase of
sBT levels resulted in an increase in the odds of suffering from hypertension by 30%
(OR:1.296, CI95%: 1.023 to 1.643; Wald’s test: p?=?0.032). In addition, when we randomly
extracted two subjects, one with hypertension, and one without, we found a 74.2% probability
that the first patient showed higher sBT concentrations (Mann–Whitney U test: p?=?0.004).
Furthermore, by one-way ANOVA we found that sBT values increased significantly in
the grade of severity of reactions from grade I to grade III (F test: p?=?0.021),
with Sidak pairwise test respectively giving p?=?0.140 (I vs. II), p?=?0.694 (I vs.
III) and p?=?0.017 (II vs. III). Grade IV was not considered in this analysis because
of the limited number of patients. This finding was confirmed from the observation
that sBT values were significantly higher in patients of group 1 (severity grade III?+?IV)
(mean: 5.62?±?3.14; range: 1.5-14.5) than in patients of group 0 (severity grade I?+?II)
(mean: 4.11?±?1.83; range: 1.5-9.4) (Welch’s test: p?=?0.0304) (Figure 1). In this subgroup of patients age did not affect the severity of the reaction (OR:
1.023; CI95%: 0.996 to 1.050; Wald’s test: p?=?0.102).

Figure 1. Basal serum tryptase levels and hypersensitivity reaction severity in general anaesthesia: Mean basal serum tryptase values?±?standard deviation between group 1 (severity grade
III?+?IV) and group 0 (severity grade I?+?II).

No effect on symptom severity was observed as regards sex (OR: 1.306, CI95%: 0.689
to 2.477, Wald’s test: p?=?0.412), atopy (OR: 0.949, CI95%: 0.762 to 1.181; Wald’s
test: p?=?0.640), history of hypersensitivity to NSAIDs (OR: 0.777; CI95%: 0.381 to
1.584; Wald’s test: p?=?0.489) or antibiotics (OR: 0.984; CI95%: 0.533 to 1.816; Wald’s
test: p?=?0.960), allergic contact dermatitis (OR: 0.974, CI95%: 0.323 to 2.936; Wald’s
test: p?=?0.963), other respiratory diseases (OR: 3.43, CI95%: 0.932 to 12.662; Wald’s
test: p?=?0.064), urticaria (OR: 0.943, CI95%: 0.278 to 3.193; Wald’s test: p?=?0.925),
therapy with beta-blockers (OR: 1.916, CI95%: 0.615 to 5.968; Wald’s test: p?=?0.262),
thyroid diseases (OR: 1.361, CI95%: 0426 to 4.348; Wald’s test: p?=?0.602). Furthermore,
we did not find a significant association between sensitization to NMBAs and the following
variables: sex (15.49% vs 13.11%; Fisher’s exact test: p?=?0.671), higher grade of
severity of the reaction (11,48% vs 15,15%; Fisher’s exact test: p?=?0.656), asthma
(10.81% vs 14.74%; Fisher’s exact test: p?=?0.792), history of hypersensitivity to
NSAIDs and/or antibiotics (16.07% vs 11.11%; Fisher’s exact test: p?=?0.403), atopy
(11.90% vs 14.95%; Fisher’s exact test: p?=?0.672), age (OR: 0.989, CI95%: 0.969 to
1.009; Wald’s test: p?=?0.308), therapy with ACE-inhibitors (14.29% vs 13.97%; Fisher’s
exact test: p??0.999), ARBs (0% vs 14.67%; Fisher’s exact test: p?=?0.365), or beta-blockers
(15.38% vs 13.89%; Fisher’s exact test: p??0.999), allergic contact dermatitis (3.70%
vs 96.30%; Fisher’s exact test: p?=?0.704). We observed a nearly significant association
between allergy to latex and history of food allergy (14.29% vs 4.29%; Fisher’s exact
test: p?=?0.059).