**2. A working classification of skin NSAID hypersensitivity based on clinical phenotypes**

Controlled oral challenge is the only definitive way to diagnose the reactions caused by NSAIDs. These challenge-proven immediate responses are a wide group of disorders that includes respiratory, cutaneous and anaphylaxis reactions [1, 3].

The mechanisms of NH are unknown, but two general hypotheses have been proposed [4]. The first one, an enzymatic activity inhibition of at least the cyclooxygenase-1 (COX-1) isoform that may inhibit the prostaglandin synthesis and thus deregulate the 5-lipooxygenase pathway, with cys-leukotriene hyperproduction in some susceptible patients. All NSAIDs that inhibit the COX-1 isoform could precipitate the reaction. For this reason, cross-reactivity among COX-1 inhibitor NSAIDs can be demonstrated in all patients with respiratory reactions and in most patients with urticaria/angioedema reactions (multiple reactors) [1, 3, 4]. The second mechanism can be applied only to a small subset of patients with NSAID hypersensitivity, as those with systemic anaphylaxis [4]. We have previously demonstrated that patients with NSAIDs-induced systemic anaphylaxis can react only to one specific NSAIDs and tolerate other COX-1 inhibitor NSAIDs in controlled oral challenges. Up to 1/3 of patients with NSAID induced systemic anaphylaxis might present immediate acute urticarial previous to anaphylactic episode when taken a specific NSAID (selective reactors) [3–5].

At least, three subsets of these patients with NH may have an associated underlying disease: in fact, around 10 percent of patients with chronic rhinosinusitis with nasal polyps and moderate-to-severe asthma and 30 percent of those with chronic urticaria/angioedema may present a nasoocular/asthmatic or urticarial reaction after NSAID exposure at some times in their lives, respectively [6]. Some clinical phenotypes of skin NSAID hypersensitivity are also definitely associated with allergic respiratory disease [3, 7].

Therefore, the diagnostic approach determines very different clusters of patients [3, 8]. Those who present clinical reactivity between different NSAIDs (multiple reactors) versus those who develop a reaction exclusively to a specific NSAID (selective reactors), and secondly, those in whom there is a very defined concomitant associated disease from the clinical and biological point of view. NSAID exposure exacerbates these diseases, determining a clinical reaction. However, the NSAID withdrawal does not determine a notable modification of the natural history of the disease, which often must continue to be treated despite the avoidance of NSAID [8].

Therefore, the NSAID reaction is an epiphenomenon that together with other clinical features constitutes what we have called the clinical phenotype (formerly called the NSAID-reaction complex) [1, 3]. This classification based on clinical phenotypes is a real and practical approach in the daily clinic, which will allow us to make the best appropriate diagnostic and therapeutic decisions.
