Discussion
In this monocentric retrospective study, 13/237 (5.5%) patients with an anaphylaxis event had a biphasic reaction. The mean time of onset for the biphasic reaction was 8 hr. This observed prevalence, time of onset and range are in line with the current literature (14, 15, 16, 17, 18). However, our results differ from the literature in the distribution of the severity grade according to the Ring and Messmer classification (25), which is described in equal proportions for mild, moderate and severe reactions (47).
Patients with biphasic and monophasic anaphylaxis did not differ in clinical and allergological features, except for loss of consciousness during anaphylaxis and asthma, both more frequent with biphasic than monophasic anaphylaxis. Furthermore, basal tryptase levels were significantly more elevated with biphasic than monophasic reactions.
That we found more loss of consciousness with biphasic than monophasic reactions suggests a more severe reaction with the former. However, the Ring and Messmer classification did not significantly differ between the groups. The association between severe anaphylaxis and biphasic reactions was recently investigated in a large cohort (42): gastrointestinal, skin symptoms, respiratory arrest, cardiac symptoms, and chronic urticaria were associated with the occurrence of biphasic reaction. Such a difference with our results can be explained by the limited number of biphasic patients in our study.
There are no reports of a link between asthma comorbidity and biphasic anaphylaxis (48, 49), which could be explained by recruitment bias because our hospital service also includes an asthma center. However, none of the asthmatic patients with a biphasic reaction were followed by fellow pulmonologists. Moreover, the proportion of asthmatic patients we found is lower than in the literature (18.5% of 237 patients vs 22.5% in a large anaphylaxis registry (3)).
Although asthma was more frequent with biphasic than monophasic anaphylaxis, the former group exhibited no severe asthma, and respiratory symptoms during anaphylaxis were similar in both groups. Asthma is known to increase the severity of anaphylaxis (48, 50, 51), and severe asthma is also associated with a predominance of mast cells positive for tryptase and chymase in the airway submucosa and epithelium (52). Furthermore, severe asthma was recently found associated with elevated basal tryptase level (even independently of type 2 inflammation) (53). We cannot conclude a potential link with asthma treatment because only two patients used inhaled corticosteroids in the biphasic group.
In our study, basal tryptase level was higher with biphasic than monophasic anaphylaxis. To our knowledge, this has never been reported. A potential explanation could be the more frequent history of asthma in the biphasic than monophasic group because basal tryptase level was higher with biphasic than monophasic anaphylaxis for asthmatic patients. Older age and renal failure increase tryptase levels and could be confounding factors, so we examined the distribution of age between our two groups and it was similar. Additionally, none of our biphasic patients had renal failure. Kraft et al (42) found no significant difference in mean basal tryptase level between monophasic (n = 4786) and biphasic anaphylaxis (n = 237). However, the authors included patients with systemic mastocytosis in their analysis, which could perhaps mask a difference in basal tryptase level and explain the discrepancy with our results. Other factors regulating the activation of mast cells, such as the cytokine environmental network or genetic regulation of tryptase production could be involved (54).
We found no significant difference in culprit allergens between monophasic and biphasic anaphylaxis. The most common allergens in our study were drugs and Hymenoptera venom, which agrees with the literature on epidemiological prevalence for the cause of anaphylaxis in adults (51, 55) and could explain the few patients with a food elicitor in our study. In addition, the basal tryptase level in biphasic and monophasic groups with an allergy to Hymenoptera venom was similar but was higher in biphasic patients with drug allergy. This has not been reported before and warrants confirmation in further studies.
The main strength of this study is the detailed information on patient characteristics. Furthermore, symptoms and medical management of biphasic phases were well documented. All patients with an anaphylaxis reaction had documented basal tryptase level, allergy test results and a complementary interview in our center.
The retrospective monocentric design of the study is its main limitation. The lack of power due to a small number of participants with biphasic reactions did not allow for applying a multivariate analysis of risk factors.
In conclusion, we show for the first time that elevated basal tryptase is linked to biphasic reactions. The role of basal tryptase level as a risk factor for biphasic reaction requires further investigations. Additionally, asthma and unconsciousness during the first phase of anaphylaxis could be associated with a biphasic reaction. These findings advocate for prolonged monitoring of these patients during their care.