2.1 Clinical Variants of GPP
Generalized pustular psoriasis was first reported in 1910 by Leopold von Zumbusch, who described a patient with plaque psoriasis who experienced nine episodes of widespread pustular eruption over a period of 20 years [
14]. Although there have been multiple subsequent case reports and series of GPP, few have included more than 100 patients. The earliest and largest case series with a detailed clinical review included 104 patients and was published in 1968 [
4]. The authors classified GPP into four clinical variants: acute GPP of von Zumbusch, annular pustular psoriasis (APP), localized GPP, and exanthematic GPP.
Annular pustular psoriasis is a less acute variant of GPP and is characterized by a subacute or chronic eruption of annular or serpiginous plaques with erythematous scaly or pustular margins [
15]. Patients are usually constitutionally well, without fever, leukocytosis, arthritis, or other organ involvement. Common aggravating factors are stress, upper respiratory tract infections (URTIs), and menstruation [
15‐
17]. Annular pustular psoriasis is a more benign variant of GPP that responds to topical agents; however, severe juvenile APP that does not respond to conventional systemic therapy and develops into acute GPP has been documented [
15].
Baker and Ryan described localized GPP as pustulosis that occurs in and around discoid psoriatic plaques [
4]. The most common trigger is superimposed bacterial infection and pustules usually resolve promptly with systemic antibiotics. Localized GPP is widely recognized as a variant of plaque psoriasis called psoriasis with pustulosis or “psoriasis cum pustulatione” as it is not uncommon for patients with psoriasis vulgaris to have pustular lesions during periods of heightened disease activity, particularly following a URTI [
18] or the use of irritant topical agents [
19].
Exanthematic GPP was described as a sudden onset of sterile pustules in patients with no previous history of psoriasis [
4]. It usually follows a URTI or use of systemic medications for treating a URTI. Exanthematic GPP is self-limiting and resolves completely within a few weeks. This description is reminiscent of acute generalized exanthematous pustulosis (AGEP); therefore, most dermatologists accept exanthematous GPP as AGEP. The prevalence of psoriasis is higher in patients with AGEP than in the general population [
20]. Both AGEP and GPP are associated with
IL36RN mutations and patients with interleukin-36 receptor antagonist insufficiency have a severe form of AGEP [
21].
Despite being first described 110 years ago, there is still no international consensus on the definition of a GPP flare or the different phenotypes. In 2017, the European Rare And Severe Psoriasis Expert Network (ERASPEN) published their consensus definition of pustular psoriasis, which defined GPP as macroscopic primary sterile pustules occurring on non-acral skin. It excludes cases in which pustules are restricted to within psoriatic plaques [
1]. Generalized pustular psoriasis should only be diagnosed when the condition has relapsed at least once or when it persists for more than 3 months [
1], and AGEP should be actively ruled out. Generalized pustular psoriasis can occur with or without plaque psoriasis, and with or without systemic inflammation. Under ERASPEN criteria, APP is classified as the chronic variant, with pustulation persisting for more than 3 months, and usually occurs without systemic symptoms or associated psoriasis vulgaris. Impetigo herpetiformis, or pregnancy-induced GPP, is now believed to be the same disease as acute GPP because the majority of patients have other triggers in addition to pregnancy [
3,
22,
23].
2.2 Clinical Characteristics of GPP
Generalized pustular psoriasis is a severe disease that affects both children and adults. Table
1 shows clinical characteristics of patients with GPP from notable case series and analyses of healthcare systems and records. Approximately 31–78% of patients with GPP have preceding psoriasis [
17,
24], with a pre-pustular duration in some patients of ≥ 20 years [
4]. Onset of GPP occurs earlier in patients without preceding psoriasis [
4,
17,
25]. Generalized pustular psoriasis flares are often provoked. Common precipitating factors include infections (particularly URTIs), sunlight, pregnancy, menstruation, hypocalcemia, and a long list of systemic drugs [
26]. Excessive use of irritant topical preparations, such as coal tar, may induce GPP and the use of potent topical corticosteroids has also been implicated [
26]. However, the most common precipitating factor is the withdrawal of systemic corticosteroids [
27], which is reported in up to 20% of patients with GPP [
28]. Other systemic drugs reported to provoke GPP include, but are not limited to, lithium, progesterone, phenylbutazone, antimalarials, fluoxetine, ustekinumab, infliximab, adalimumab, and apremilast [
26,
29]
Table 1
Comparison of patient demographics and GPP clinical characteristics and outcomes in studies reporting > 25 patients with adult-onset GPP and > 15 patients with juvenile GPP, and with sufficient and/or relevant data to tabulate results
Male:female | 34:68 | 43:61 | 88:120 | 31:32 | 12:16 | 21:13 | 15:18 | 21:23 | 23:16 | 10:8 | 11:16 |
Mean age of GPP onset, years | 40.9 | > 40c | 33.6 | 43.0 | 37.5 | 58d | 40.7 | 20.5 | 6.4e | 3.8 | 6.5d |
History of plaque psoriasis, n (%) | 79 (77.5) | 62 (59.6) | 65 (31.3) | 32 (50.8) | 3/7f (42.9) | 22 (64.7) | 10f (40) | 14 (31.8) | 12 (31) | 7 (38.9) | 10 (37.0) |
Mean pre-pustular duration, years | 11.7 | g | 6 | 12h | | 11.2 | | | | | |
Patients with triggers identified, n (%) | 87 (85.3) | 54 (51.9) | | 16/35f (45.7) | 6/7f (85.7) | | 20 (60.6) | | 16 (41) | | 16/21f (76.1) |
Fever, n (%) | 94 (92.2) | | | | 7/7f (100) | | 8 (24.2) | | i | | 17/21 (80.9) |
Malaise/fatigue, n (%) | | | | | 7/7f (100) | | | | | | |
Leukocytosis, n (%) | 68 (66.7) | | | | 2/7f (28.6) | | 21/29 (72.4) | | i | | 13/18f (72.2) |
Arthritis, n (%) | 35 (34.3) | 33 (32) | | | | | | 7 (31.8) | | 0 | |
Mortality, n (%) | 7 (7) | 17 (16)j | | 2 (3) | 1 (4) | 2 (6) | 1 (4) | 1 (2.3) | | 0 | 0 |
Rebound of psoriasis with novel pustulation is well documented following treatment withdrawal of systemic corticosteroid therapy as well as non-corticosteroid therapy in patients with classic plaque psoriasis [
26,
27,
29‐
31]. For example, Khemis et al. [
31] reported that 13.7% of 139 patients taking brodalumab experienced a rebound of psoriasis 12 weeks after stopping brodalumab. All 19 patients had pustular lesions, but only two patients had extensive lesions with associated fever, fatigue, elevated CRP, and leukocytosis. Pustular flares on withdrawal of brodalumab may be attributable to heightened disease activity following rebound of psoriasis on treatment cessation and this may also partly explain the occurrence of GPP following corticosteroid withdrawal.
Generalized pustular psoriasis is characterized by recurrent flares of widespread, painful, fiery erythema studded with sterile pustules that may coalesce to form lakes of pus. Mucosal involvement may also occur, and manifests as cheilitis, geographic tongues or fissured tongues, and genital lesions [
32]. Severity can range from mild to severe, and an individual patient can experience different disease extent, severity, and cutaneous and extracutaneous symptoms with each flare. Generalized pustular psoriasis flares may or may not be accompanied by systemic symptoms, including fever, general malaise, fatigue, and disabling edema [
26]. Common laboratory abnormalities observed in GPP flares include leukocytosis and neutrophilia, elevated CRP levels, hypocalcemia, hypoalbuminemia, and abnormal liver function tests [
3,
6‐
8].
High fever has been reported in approximately 24–96% of patients during a GPP flare (Table
1) [
3,
8,
11] and approximately 30–70% of patients have leukocytosis with neutrophilia recorded [
3,
11,
33]. There is no international consensus on the definition of a severe GPP flare. In a single-center study of 95 patients with acute GPP affecting > 30% of the body surface area (BSA), pain, fever, and leukocytosis were prevalent and documented in 96.8%, 95.8%, and 82.1% of patients, respectively [
3]. Hence, a GPP flare affecting > 30% BSA may be classified as a severe or significant flare because of its association with systemic inflammation. Systemic inflammation, defined as a fever > 38 °C with leukocytosis > 12 × 10
9/L, is more prevalent in patients with
IL36RN variants [
20,
22]. Patients with homozygous
IL36RN variants have a more severe phenotype that is characterized by early onset, low prevalence of plaque psoriasis, and a high risk of systemic inflammation [
34]. Extracutaneous manifestations of a GPP flare include arthritis and arthralgia, uveitis and conjunctivitis, and neutrophilic cholangitis [
3,
4,
8,
35]. Generalized pustular psoriasis is also associated with multiple comorbidities, such as hyperlipidemia, hypertension, diabetes mellitus, obesity, anxiety, and depression [
3,
32,
36].
2.3 Clinical Course of GPP
Generalized pustular psoriasis is an unpredictable disease with a highly variable course. While some patients exhibit multiple flares per year, others may only experience a flare every few years. In the single-center study of 95 patients with a GPP flare affecting > 30% BSA, 56.8% of patients had only one flare, 29.5% had two to five flares, and 13.7% had more than five flares over an average follow-up duration of approximately 5 years [
3]. However, 70% of the patients with only one severe flare experienced multiple mild flares on attempts to taper off systemic therapy or when exposed to classic triggers such as URTI, stress, and menstruation, despite receiving systemic therapy [
3]. Furthermore, 30% of these patients had persistent pustular lesions and fewer than 5% had clear or almost clear skin between GPP flares. A study using the French National Health Data System (SNDS) database also showed that most patients with GPP had no flare or one flare per year during the study period (between 2010 and 2018) [
9]. In this study, most (79.4%) of the 569 patients with incident GPP had only one flare, 17.6% had two to four flares, and 3% had five or more flares. This observation was confirmed by a survey of dermatologists contributing data to the Corrona registry, which showed that most (69%) of the 29 dermatologists surveyed estimated that their patients had an average of no flare or one flare per year and 28% estimated that their patients had an average of two to three flares per year [
37].
Generalized pustular psoriasis flares typically last 2–5 weeks but may persist longer than 3 months [
3,
11,
37]. Approximately one-third of patients have persistent pustular lesions, even when receiving systemic therapy that resolved the acute symptoms [
3]. Persistence of pustular manifestations appears to be greater among patients with APP than those with GPP. In a Korean study, pustular lesions persisted beyond 2 months in 60% of five patients with APP compared with 27% of 15 patients with GPP [
11]. It is estimated that up to 10% of patients may not respond to treatment and experience persistent flare [
24]. Furthermore, few patients achieve clear skin between flares, despite receiving various treatments [
3,
27,
37]. Patients may have psoriasis vulgaris, inverse or erythrodermic psoriasis, APP, and acrodermatitis continua of Hallopeau between GPP flares [
3,
11,
25].
Generalized pustular psoriasis flares can be severe and potentially life threatening. An estimated 50% of GPP flares require hospitalization [
38] and a patient with GPP will require hospitalization for a flare at least once every 5 years [
3]. The duration of inpatient stay for a GPP flare is reported to be 10–14 days [
39]. The mortality rate directly attributable to GPP or its associated treatment, specifically with the use of systemic corticosteroids, has been reported to be 2–16% [
3,
4,
10,
38]. A recent study from Japan using a national inpatient database showed an overall in-hospital mortality rate of 4.2% among 1513 patients with GPP [
38]. The mortality rate was highest in patients receiving systemic corticosteroid monotherapy and lowest in patients receiving biologics, at 9.1% and 1.0%, respectively [
38].