Incidence and mortality of pemphigus in France.

The incidence of pemphigus varies from 0.5 to 34 cases/million inhabitants/year, with the highest incidence rates in Brazil (Hans-Filho et al., 1996Hans-Filho G. dos Santos V. Katayama J.H. Aoki V. Rivitti E.A. Sampaio S.A. et al.An active focus of high prevalence of fogo selvagem on an Amerindian reservation in Brazil. Cooperative Group on Fogo Selvagem Research.J Invest Dermatol. 1996; 107: 68-75Abstract Full Text PDF PubMed Scopus (69) Google Scholar, Ishii et al., 2008Ishii N. Maeyama Y. Karashima T. Nakama T. Kusuhara M. Yasumoto S. et al.A clinical study of patients with pemphigus vulgaris and pemphigus foliaceous: an 11-year retrospective study (1996–2006).Clin Exp Dermatol. 2008; 33: 641-643Crossref PubMed Scopus (31) Google Scholar, Langan et al., 2008Langan S.M. Smeeth L. Hubbard R. Fleming K.M. Smith C.J. West J. Bullous pemphigoid and pemphigus vulgaris incidence and mortality in the UK: population based cohort study.BMJ. 2008; 337: a180Crossref PubMed Scopus (459) Google Scholar, Meyer and Misery, 2010Meyer N. Misery L. Geoepidemiologic considerations of auto-immune pemphigus.Autoimmun Rev. 2010; 9: A379-A382Crossref PubMed Scopus (78) Google Scholar). Additionally, although the prognosis of pemphigus patients is considered good in the literature, recent findings reported unusually high mortality rates (Almugairen et al., 2013Almugairen N. Hospital V. Bedane C. Duvert-Lehembre S. Picard D. Tronquoy A.F. et al.Assessment of the rate of long-term complete remission off therapy in patients with pemphigus treated with different regimens including medium- and high-dose corticosteroids.J Am Acad Dermatol. 2013; 69: 583-588Abstract Full Text Full Text PDF PubMed Scopus (44) Google Scholar, Langan et al., 2008Langan S.M. Smeeth L. Hubbard R. Fleming K.M. Smith C.J. West J. Bullous pemphigoid and pemphigus vulgaris incidence and mortality in the UK: population based cohort study.BMJ. 2008; 337: a180Crossref PubMed Scopus (459) Google Scholar). We estimated the incidence and mortality of pemphigus among 13 regions in France (Figure 1a) over a 10-year period. Inclusion criteria were: (i) patient living in 1 of the 13 regions and (ii) newly-diagnosed pemphigus. Cases were identified using the computerized databases of the pathology laboratories of the university and general hospitals and private-practice laboratories that perform direct immunofluorescence. Statistical analyses are described in Supplementary Material online. From January 2004 to December 2013, 629 patients were identified in included regions, which corresponded to a population size of 13.75 million inhabitants (Figure 1a). Among them, 380 were excluded: (i) diagnosis of pemphigus not confirmed (n = 74); (ii) patient not domiciled in the selected regions (n = 194), and (iii) diagnosis of pemphigus made before or after the study period (n = 112). A total of 249 incident cases (125 women, 124 men) were included. Mean age at diagnosis was 59.4 ± 18.7 years and was similar between male and female patients (P = 0.93). The age distribution of the population is shown in Figure 1b. Pemphigus types were pemphigus vulgaris (PV) (n = 155 [62%], pemphigus foliaceus (PF) (n = 67 [27%]), paraneoplastic pemphigus (n = 12 [5%]; and other subtypes, n = 15 [6%]) (Table 1).Table 1Baseline characteristics and evolution of the 249 patients with newly diagnosed pemphigusCharacteristicsPemphigus Vulgaris (n = 155)Pemphigus Foliaceus (n = 67)Paraneoplastic Pemphigus (n = 12)Other Pemphigus Subtypes1Other pemphigus subtypes: IgA pemphigus (n = 7); pemphigus vegetans (n = 4); pemphigus herpetiformis (n = 3); drug-induced pemphigus (n = 1). (n = 15)Sex ratio, female/male1.07/10.68/12/12/1Age at diagnosis, years, mean ± SD57.5 ± 17.361.3 ± 20.471.1 ± 11.061.3 ± 25.8Distribution by age category, n (%) <40 years30 (19.4)10 (14.9)03 (20.0) 40–59 years50 (32.3)20 (29.9)02 (13.3) 60–74 years50 (32.3)15 (22.4)8 (66.7)3 (20) 75–89 years24 (15.5)21 (31.3)3 (25.0)7 (46.7) ≥90 years1 (0.6)1 (1.5)1 (8.3)0Clinical presentation, n (%) Mucosal and skin lesions105 (67.7)42These four patients were classified as pemphigus foliaceus because they mainly presented skin lesions. The histologic examination of a skin biopsy showed an acantholysis in the upper layers of the epidermis. Three of these four patients had negative serum anti-DSG3 antibodies, and one had very low (22 U) anti-DSG3 antibody ELISA values (n < 20), whereas all these four patients had elevated anti-DSG1 antibody ELISA values. (6.0)10 (83.33)8 (53.3) Exclusive skin lesions9 (5.8)63 (94.0)1 (8.33)7 (46.7) Exclusive mucosal lesions41 (26.5)01 (8.33)0Comorbidities, n (%) Malignancy15 (9.7)5 (7.5)8 (66.7)3 (20.0) Cardiovascular disorder28 (18.1)17 (25.4)5 (41.7)6 (40.0) Hypertension43 (27.7)15 (22.4)8 (66.7)6 (40.0) Pulmonary disorder23 (14.8)7 (10.4)3 (25.0)2 (13.3) Diabetes mellitus20 (12.9)4 (6.0)5 (41.7)4 (26.7) Neurological disorder6 (3.9)5 (7.5)04 (26.7)First-line treatment regimens, n (%) Oral prednisone alone95 (61.3)25 (37.3)7 (58.3)4 (26.7) Oral prednisone + IS36 (23.2)9 (13.4)1 (8.3)1 (6.7) Rituximab + oral prednisone9 (5.8)1 (1.5)01 (6.7) Dapsone without oral prednisone5 (3.2)21 (31.3)05 (33.3) Topical corticosteroid10 (6.5)11 (16.4)4 (33.3)4 (26.7)Follow-up duration, years mean ± SD5.9 ± 3.25.5 ± 3.42.7 ± 2.75.0 ± 4.0Lost to follow-up, n (%)9 (5.8)3 (4.5)01 (6.7)Deaths, n (%)30 (19.4)21 (31.1)7 (58.3)8 (53.3) Males15 (20.0)17 (42.5)1 (25.0)3 (60.0) Females15 (18.8)4 (14.8)6 (75.0)5 (50.0) Oral prednisone alone17 (17.9)8 (32.0)4 (57.1)2 (50.0) Oral prednisone + IS6 (16.7)6 (66.7)1 (100.0)— Rituximab + oral prednisone01 (100.0)01 (100.0) Dapsone without oral prednisone02 (9.5)03 (60.0) Topical corticosteroid7 (70.0)4 (36.4)2 (50.0)2 (50.0)Causes of death, n (%) Malignancy7 (23.3)3 (14.3)4 (57.1)3 (37.5) Cardiovascular disorder8 (26.7)6 (28.6)1 (14.3)1 (12.5) Infectious disorder2 (6.7)4 (19.0)1 (14.3)1 (12.5) Dementia3 (10.0)2 (9.5)02 (25.0) Pulmonary disorder1 (3.3)2 (9.5)1 (14.3)0 Digestive disorder1 (3.3)1 (4.8)00 Alcoholic hepatitis01 (4.8)00 Hyponatremia01 (4.8)00 Diabetes mellitus1 (3.3)—00 Unknown cause7 (23.3)1 (4.8)01 (12.5)Abbreviations: SD, standard deviation; IS, immunosuppressant.1 Other pemphigus subtypes: IgA pemphigus (n = 7); pemphigus vegetans (n = 4); pemphigus herpetiformis (n = 3); drug-induced pemphigus (n = 1).2 These four patients were classified as pemphigus foliaceus because they mainly presented skin lesions. The histologic examination of a skin biopsy showed an acantholysis in the upper layers of the epidermis. Three of these four patients had negative serum anti-DSG3 antibodies, and one had very low (22 U) anti-DSG3 antibody ELISA values (n < 20), whereas all these four patients had elevated anti-DSG1 antibody ELISA values. Open table in a new tab Abbreviations: SD, standard deviation; IS, immunosuppressant. The mean annual crude incidence of pemphigus was 1.85 cases/million inhabitants/year (95% confidence interval [CI] = 1.63–2.08). The mean world-population standardized incidence was 1.45 cases/million inhabitants/year (95% CI = 1.11–1.79). Mean annual French population-standardized incidence rates ranged from 0.99 cases/million inhabitants/year (95% CI = 0.64–1.45) in the North region of France to 3.39 cases/million inhabitants/year (95% CI = 1.94–5.58) in the Haute-Vienne region. The linear North-South gradient was significant (P = 0.004) with an incidence rate ratio estimated to 1.11 (95% CI = 1.04–1.18) for a move of 100 km to the South (Figure 1a and Supplementary Material online). The annual incidence rate ratio per calendar year over the study period was 1.05 (95% CI = 0.97–1.06), showing no significant increase of incidence over the study period (P = 0.45). Otherwise, the incidence rate of pemphigus increased with age (Figure 1c). The median follow-up duration of the cohort was 5.4 years (interquartile range [IQR] = 3.0–8.1 years). Thirteen patients were lost to follow-up (5.2%). A total of 66 (27%) patients died during the study period (Table 1). The 1-, 2-, and 5-year overall survival rates were 92% (95% CI = 88–95%), 88% (95% CI = 83–91%), and 77% (95% CI = 71–82%), respectively, in the whole population; 97% (95% CI = 92–99%), 93% (95% CI = 87–96%), and 83% (95% CI = 76–89%), respectively, in PV patients; 88% (95% CI = 78–94%), 84% (95% CI = 72–91%), and 74% (95% CI = 61–83%), respectively, in PF patients; and 67% (95% CI = 34–86%), 58% (95% CI = 27–80%), and 39% (95% CI = 8–70%), respectively, in paraneoplastic pemphigus patients (Figure 1d). Relative to expected age-, sex-, and region-specific overall death rates in the general population in France, the standardized mortality ratio of pemphigus patients was 1.67 (95% CI = 1.46–1.93). The median age of death was 82.4 years (IQR = 76.9–87.5 years) in the whole pemphigus population, corresponding to 82.3 years (IQR = 76.6–86.0 years) in PV patients, 87.4 years (IQR = 81.6-88.5 years) in PF patients, 74.9 years (IQR = 67.7–80.7 years) in paraneoplastic pemphigus patients, and 80.7 years (IQR = 77.4–82.9 years) in patients with other pemphigus subtypes. All deaths were observed in patients older than 60 years at diagnosis. Interestingly, the proportion of PF patients older than 75 years at diagnosis (22 of 67 [32.8%]) was twofold higher than that of PV patients (25 of 155 [16.1%]). The cause of death could be recorded in 57 of the 66 deceased patients. Main causes of death were malignancy (n = 17 [29.8%]), cardiovascular disease (n = 16 [28.1%]), infection (n = 8 [14.0%]), and dementia (n = 7 [12.3%]) (Table 1). Older age at diagnosis and association with neoplasia were statistically associated with mortality. Indeed, the risk of mortality in patients older than 75 years corresponded to a hazard ratio of 16.3 (95% CI = 9.4–28.3) relative to younger patients, and the risk of mortality in patients with neoplasia adjusted on age by left truncation from birth to diagnosis, corresponded to a hazard ratio of 2.44 (95% CI = 1.35–4.40; P = 0.005). The left-truncated age-adjusted mortality rate of PF was not significantly higher than that of PV mortality (hazard ratio = 1.55; 95% CI = 0.84–2.84; P = 0.16). The crude incidence rate of PV was estimated at 1.15/million inhabitants/year, which is more than sixfold lower than the crude incidence rate of PV reported by Langan et al., 2008Langan S.M. Smeeth L. Hubbard R. Fleming K.M. Smith C.J. West J. Bullous pemphigoid and pemphigus vulgaris incidence and mortality in the UK: population based cohort study.BMJ. 2008; 337: a180Crossref PubMed Scopus (459) Google Scholar in the United Kingdom and lower than that reported from Southern European countries, ranging from 4 to 4.4 cases/million inhabitants/year (Baican et al., 2010Baican A. Baican C. Chiriac G. Chiriac M.T. Macovei V. Zillikens D. et al.Pemphigus vulgaris is the most common autoimmune bullous disease in Northwestern Romania.Int J Dermatol. 2010; 49: 768-774Crossref PubMed Scopus (59) Google Scholar, V’lckova-Laskoska et al., 2007V’lckova-Laskoska M.T. Laskoski D.S. Kamberova S. Caca-Biljanovska N. Volckova N. Epidemiology of pemphigus in Macedonia: a 15-year retrospective study (1990–2004).Int J Dermatol. 2007; 46: 253-258Crossref PubMed Scopus (42) Google Scholar). Interestingly, we observed higher incidence rates of pemphigus in Southern regions of France compared to Northern regions, which is in accordance with the North to South gradient of pemphigus incidence in Europe. We observed an increasing incidence of pemphigus with age, with the highest incidence in people aged older than 80 years. Importantly, the 92%, 88%, and 77% 1-, 2- and 5-year survival rates calculated in the present study suggest that the prognosis of pemphigus is worse than the 5% mortality rate usually reported in general reviews (Bystryn and Rudolph, 2005Bystryn J.-C. Rudolph J.L. Pemphigus.Lancet. 2005; 366: 61-73Abstract Full Text Full Text PDF PubMed Scopus (333) Google Scholar; Chams-Davatchi et al., 2005Chams-Davatchi C. Valikhani M. Daneshpazhooh M. Esmaili N. Balighi K. Hallaji Z. et al.Pemphigus: analysis of 1209 cases.Int J Dermatol. 2005; 44: 470-476Crossref PubMed Scopus (185) Google Scholar). Our data suggest that the high mortality rate observed in our population of pemphigus patients was mainly related to the old age of patients because the median age at death of PF and PV patients was 87.4 and 82.4 years, respectively. Indeed these elderly patients poorly tolerated corticosteroids and immunosuppressant side effects, as demonstrated by their main causes of death. We observed an unexpectedly high mortality rate in PF patients, which was likely related to the high proportion (32.8%) of patients older than 75 years among PF patients. The second prognostic factor was the association with neoplasia, although the 58% mortality rate of paraneoplastic pemphigus patients in the present series was lower than the 75–90% mortality rate reported in the literature (Wieczorek and Czernik, 2016Wieczorek M. Czernik A. Paraneoplastic pemphigus: a short review.Clin Cosmet Investig Dermatol. 2016; 9: 291-295Crossref PubMed Scopus (59) Google Scholar). In conclusion, this study highlights the high mortality of pemphigus in elderly patients, including PF, which is often presented as a more benign subtype than PV (Bystryn and Rudolph, 2005Bystryn J.-C. Rudolph J.L. Pemphigus.Lancet. 2005; 366: 61-73Abstract Full Text Full Text PDF PubMed Scopus (333) Google Scholar). The authors state no conflict of interest. We are indebted to Nikki Sabourin-Gibbs, Rouen University Hospital, for her help in editing the manuscript and all the contributors who provided help during the study. The study was supported by a grant from the French Society of Dermatology. Download .pdf (.14 MB) Help with pdf files Supplementary Data