Porphyria cutanea tarda (PCT), the commonest porphyria, is caused by the deficiency of the fifth enzyme in heme synthesis, uroporphyrinogen decarboxylase (URO-D), resulting in the accumulation of uroporphyrins in the skin and liver, producing delayed phototoxic skin reactions and asymptomatic chronic hepatopathy.25 Skin lesions consist of tense blisters and fragile skin on light-exposed areas that heal slowly leaving atrophic scars and milia cysts. However, some patients may not have these characteristic lesions, potentially leading to delayed diagnosis, irreversible scarring, and liver damage.
PCT is a complex, multifactorial disorder with 2 main forms: sporadic PCT (S-PCT or type I, 80% of cases) and familial PCT (F-PCT or type II, 20% of cases). Both F-PCT and S-PCT manifest clinically only when hepatic URO-D activity is <30%.10,25 Heterozygous mutations in the URO-D gene (chromosome 1p34) that impair URO-D synthesis or degradation in all types of cells form the genetic substrate of F-PCT. However, heterozygous URO-D mutations reduce URO-D activity only by about 50%, meaning they are necessary, but not sufficient, to cause overt F-PCT.25 This may explain the relatively high percentage of asymptomatic carriers of URO-D gene mutations.2 As a result, additional genetic and environmental factors are necessary to induce F-PCT.6 In S-PCT, reduced URO-D activity occurs only in the liver, by inhibition or inactivation of the hepatic enzyme due to such factors, which are the same for S-PCT and F-PCT and mainly include hemochromatosis gene mutations, hepatitis C virus (HCV) infection, human immunodeficiency virus (HIV) infection, estrogens, and high alcohol intake.25 The distribution of these risk factors varies widely between populations. For example, HCV infection is much more common in PCT patients from Spain13 or the United States5 than in patients from Argentina19 or Sweden.22
We conducted the current study to describe and compare the clinical manifestations, biochemical features, and risk factors of a cohort of 152 patients from Catalonia, Spain, with regard to sex and type of PCT (S-PCT and F-PCT). Our goal was to detect features that may help to prevent delayed diagnosis and to predict F-PCT in newly diagnosed cases.
PATIENTS AND METHODS
We included all PCT patients (n = 152) who consecutively attended the Porphyria Unit of the Hospital Clinic of Barcelona between 2004 and 2008. All patients were Spanish and lived in Catalonia (northeastern Spain), mainly in the Barcelona metropolitan area. PCT had been previously diagnosed in 113 patients (between 1977 and 2004), and was newly diagnosed in 39 patients (between 2004 and 2008). Written consent was obtained from each patient.
We reviewed clinical histories to retrieve epidemiologic information (for example, alcohol intake and estrogen use), PCT signs and symptoms, age of disease onset, delay in diagnosis, and laboratory findings. A clinical index including 16 items was designed to assess the severity of PCT at diagnosis (Table 1). Each item scored 1 point when present. Laboratory results recorded at diagnosis included total urinary porphyrins; serologic testing for hepatitis B virus (HBV), HCV, and HIV; serum alanine aminotransferase (ALT), serum gamma-glutamyl transpeptidase (GGT), serum ferritin, and fasting glucose levels. URO-D erythrocyte activity, and URO-D and hemochromatosis gene mutations were analyzed between 2004 and 2008.
Total porphyrins were analyzed in a 24-hour urine collection by spectrofluorometry (model F-2000; Hitachi). Fifty μL of urine was diluted in 2.7 mL of hydrogen chloride (HCl) 2.7N, and fluorescence was estimated with excitation set at 398 nm and emission at 603 nm. A calibration curve was set with acid dilutions of a Coproporphyrin III standard (Porphyrin Products Inc, Logan, UT; 0.5 μg/mL in 1mol/L HCl). The reference value was <300 nmol/L.
Biochemical confirmation of PCT pattern of porphyrins and isomers in urine and feces was achieved by thin layer chromatography (113 patients diagnosed between 1977 and 2004) or reverse-phase high-performance liquid chromatography of urinary porphyrins according to Lim and Peters16 (39 patients diagnosed between 2004 and 2008).
URO-D erythrocyte activity was measured in all patients as a screening method for F-PCT according to McManus et al18 and was reported as pmoles of coproporphyrinogen formed per hour per mg of hemoglobin (U/mgHb). Patients with low or borderline URO-D activity, according to a previously constructed receiver operating characteristic (ROC) curve and a cutoff point of 39 U/mgHb (Badenas et al2), were genotyped. Exons 1-10 of the UROD gene and the associated splice donor and acceptor sites were analyzed by polymerase chain reaction as described previously by our group.2 Hemochromatosis gene mutations (C282Y and H63D) were analyzed as described by Toll et al.26
A descriptive analysis was performed on each variable. The bivariate analysis was performed using the chi-square, ANOVA, and Student t tests and the Pearson correlation. The final multivariate logistic regression analysis included the PCT form (S-PCT/F-PCT) as the dependent variable and several factors, such as HCV infection, sex, and alcohol abuse, as independent variables. The SPSS program (v12.0, SPSS Inc, Chicago, IL) was used, and the level of statistical significance was established as p < 0.05.
Detailed clinical manifestations are shown in Table 1. There were 119 men (78.3%) (male to female ratio, 3.6:1). The mean age was 60 years (range, 24-86 yr). The mean age of clinical onset was 45 years (range, 20-74 yr). The onset of clinical symptoms occurred at age >55 years in 75% of patients. The mean delay in diagnosis was 19.9 months (range, 2-120 mo). The most frequent skin features at diagnosis were skin fragility (86.2%) and vesicles and bullae (78.3%) on the dorsa of the hands; 88.2% of patients presented with at least 1 such skin feature. The mean clinical score of all patients was 4.92 (range, 0-16), and 94% of patients had a score ≥2.
Facial hypertrichosis (84.8% vs. 58%; p = 0.004), involvement of areas other than the hands and head (33.3% vs. 13.4%; p = 0.008), and pruritus (27.3% vs. 12.6%; p = 0.041) were significantly more frequent in women than in men. Skin fragility on the face or scalp was significantly more frequent in men than in women (26.1% vs. 9.1%; p = 0.039). Women had less delay in the diagnosis than men (15.8 vs. 20.9 mo) and a higher clinical score (5.21 vs. 4.84), although the differences were not significant.
Nine patients (5.9%) were diagnosed without presenting skin fragility or blistering. PCT was suspected due to raised transaminases and iron overload (n = 3), generalized hyperpigmentation (n = 2), or periorbital hypertrichosis with facial hyperpigmentation (n = 4). The mean level of urinary porphyrins in these patients was significantly lower than in the rest of the cohort (1692 nmol/L vs. 3859 nmol/L; p = 0.03).
These skin manifestations preceded skin fragility and blistering (mean, 16 mo; range, 3-48 mo) in 9 other patients. Four women were on oral contraceptives. When facial hypertrichosis appeared, they were misdiagnosed with hirsutism, which led their general practitioners to switch them to another contraceptive with a greater estrogen dose. Subsequently, they developed blisters and skin fragility on the dorsa of the hands and were diagnosed with PCT.
In these 18 patients with an atypical disease onset (11.8%), periorbital hypertrichosis was the first symptom in 7.9% of the whole cohort and in 25% of the women.
Mean total urinary porphyrin excretion at diagnosis was 3715 nmol/L (range, 343-19,476 nmol/L), with no significant differences according to sex or the presence or absence of specific risk factors. There was a significant correlation between total urinary porphyrins and the clinical score at diagnosis (Pearson R2 = 0.348; p < 0.001).
The mean ALT value was 96 U/L (range, 21-448; normal value, 5-40 U/L), and 90.7% of patients had raised values. In 54.5% of patients, of whom 96.3% were HCV-positive, liver transaminases remained elevated despite PCT treatment. Mean GGT was 109 U/L (range, 18-900; normal value, 5-40 U/L), with 82.6% of patients having elevated values at diagnosis. GGT mean values were significantly higher in HCV-positive patients compared to HCV-negative individuals (123.6 U/L vs. 80.4 U/L, respectively; p = 0.044).
Serum ferritin was analyzed in 113 patients. The mean value was 520 ng/mL (range, 75-1421 ng/mL; normal values, 20-300 ng/mL in men; and 15-200 ng/mL in premenopausal women). Seventy-four percent of patients had elevated serum ferritin values, and 27% had persistent normal values despite having 1 or more disease relapses. Conversely, 32.5% always had elevated values despite treatment and clinical remission. Comparing serum ferritin at diagnosis according to sex or presence/absence of PCT risk factors showed no significant differences.
Mean fasting glucose was 110.7 mg/dL (range, 75-338; normal value, 65-110 mg/dL). Type 2 diabetes mellitus was diagnosed in 36 patients (23.7%) with basal glucose levels ≥126 mg/dL in 2 determinations. Eighteen additional patients (11.9%) had basal glycemic values between 110 and 125 mg/dL and were considered to have a prediabetic status.
Table 2 shows the distribution of PCT risk factors according to sex. HCV infection was the most prevalent risk factor (65.8%), especially in men (76.5% of men vs. 27.3% of women; p < 0.001). Hemochromatosis gene mutations C282Y and H63D were analyzed in 134 patients, of whom 64.9% carried at least 1 mutated allele: the most prevalent genotype was heterozygous H63D (40.3%), followed by homozygous H63D (10.4%), heterozygous C282Y (7.5%), and compound heterozygous C282Y/H63D (6.7%). No patient was homozygous for the C282Y mutation. Alcohol abuse (≥50 g/d) was acknowledged by almost 60% of patients (68.9% of men vs. 27.3% of women; p < 0.001). Estrogens were an additional risk factor in 12 of 33 women (36.4%), with oral contraceptives being the source in 7 (58.3%). HIV infection was reported in 14 patients (9.2%). HBV infection and chronic hemodialysis were reported in only 2 patients respectively.
A total of 85.8% of patients had more than 1 putative risk factor for developing PCT, while 13.4% had only 1 factor. The most frequent risk factor combination was HCV infection plus high alcohol intake plus any hemochromatosis allelic mutation (24.6%), especially in men (30%). In women, risk factor patterns were more dispersed, with the combination of hemochromatosis gene mutations and estrogens (14.3%) being the most frequent.
Familial PCT Compared With Sporadic PCT
Twenty-five (16.4%) patients harboring a mutation within the URO-D gene were classified as F-PCT (mean erythrocyte URO-D activity, 36.8 U/mgHb). The remaining 127 patients (83.6%) were classified as S-PCT (mean erythrocyte URO-D activity, 61.2 U/mgHb).
In Table 3 we compare epidemiologic, biochemical, and PCT risk factors in F-PCT and S-PCT patients. The male to female ratio was lower (p < 0.001) in F-PCT than in S-PCT (0.8:1 vs. 5.7:1). The only significant difference between S-PCT and F-PCT with regard to skin features was a lower frequency of hand blisters in F-PCT (81.9% vs. 60%; p = 0.015, data not shown). There were no significant differences in the age of onset. Mean fasting glucose was lower in F-PCT (96.6 mg/dL vs. 113 mg/dL; p = 0.002). HCV infection (16% vs. 75.6%; p < 0.001) and alcohol abuse (40% vs. 63.8%; p = 0.027) were less frequent in F-PCT patients. Of the F-PCT patients, 90% had at least 1 other risk factor (besides URO-D mutations); the main pattern was the combination of high alcohol intake and hemochromatosis gene mutations (35%).
Applying different logistic regression models to predict F-PCT cases showed only 1 significant association: the absence of HCV infection (odds ratio, 0.05; 95% confidence interval, 0.006-0.46).
We present one of the largest reported series of PCT patients, with special emphasis on the clinical manifestations and differences between S-PCT and F-PCT. PCT mostly presented in the fourth or fifth decade of life with skin fragility and blisters on photo-exposed areas, particularly the dorsa of the hands, consistent with the findings of the seminal paper by Grossman et al12 in 1979 that included 40 patients studied between 1960 and 1976. However, the clinical profile of PCT may have changed slightly in recent decades: some manifestations described by Grossman et al,12 such as scleroderma-like changes (18%) or scarring alopecia (33%), were hardly found in our series, perhaps due to earlier diagnosis (even though our delay in diagnosis was 20 months). We believe that is a relevant issue, since PCT reviews3,23 and medical textbooks continue to reference the results of Grossman et al when describing the prevalence of PCT cutaneous features, which might be inaccurate for the current state of PCT. There is a lack of reports providing a detailed analysis of the clinical manifestations of PCT and comparing PCT in men and women.
Periorbital hypertrichosis was more prevalent in women and as frequent as skin fragility on the dorsa of the hands. Women presented more involved skin areas and more pruritus than men, possibly due to clothes that protect fewer areas from sunlight. Skin fragility on the face or head was significantly more prevalent in men, probably due to beard shaving and androgenetic alopecia.
Eighteen "atypical" patients did not report skin fragility or blisters as the initial PCT sign (50%) or never developed them (50%). Facial hypertrichosis with facial hyperpigmentation (12 patients, 7.9%) and generalized hyperpigmentation (3 patients, 2%) were initial or isolated features of PCT in our cohort. Some experts suggest that hypertrichosis is rarely a debut sign of PCT,11 and indeed it was the first sign in only 2 reported cases.4,21 However, in the current series, periorbital hypertrichosis was the initial sign in 25% of women (usually associated with facial hyperpigmentation). Most of these women developed typical skin manifestations after exposure to estrogens, usually after being misdiagnosed with hirsutism. As for generalized hyperpigmentation, we could find only 1 case in the literature where this feature was the earliest cutaneous manifestation of PCT.20 Physicians should therefore be aware of these PCT presentations, which may be more common than previously suggested.
Three of the 18 atypical cases were totally asymptomatic, but presented with iron overload and increased transaminases and GGT. We suggest urinary porphyrin screening in any patient presenting with iron overload and abnormal liver test function, regardless of HCV status. Atypical cases that were diagnosed without presenting skin fragility and blisters had lower mean levels of urinary porphyrins than stereotypical cases. As the urinary level of porphyrins mirrors tissue deposits of porphyrins, our results suggest that full-blown PCT may develop only when a critical amount of porphyrins is reached in the skin. Furthermore, we found a significant positive correlation between total urinary porphyrin excretion and a 16-item clinical score index in the whole cohort.
Most patients presented with raised levels of serum ALT (91%) and GGT (82%). GGT values were significantly higher in HCV-positive patients than in HCV-negative patients. As we found in a different cohort,13 most HCV-positive patients did not normalize liver transaminases after PCT treatment, suggesting that liver changes in PCT patients with concomitant HCV infection are chiefly related to HCV and not to PCT. Serum ferritin was elevated in most patients (74%) but did not parallel clinical activity and was not associated with any specific risk factor.
We studied the prevalence of well-known risk factors. Most patients had at least 2 putative risk factors for developing PCT, with a high prevalence of HCV infection (65.8%) and alcohol abuse (59.9%). Almost 50% of patients (58.5% of men) had these 2 factors simultaneously. In women, URO-D gene mutations (42.4%) and estrogen exposure (36.4%) were the most frequent risk factors.
Few studies have compared F-PCT and S-PCT patients.1,8,9,14,24 To our knowledge, only 1 study analyzed clinical features,8 and others did not confirm possible F-PCT cases by URO-D genotyping.9,14,24 The frequency of F-PCT in our patients (16.4%) was slightly lower than the reported 24% in Danish8 or 53% in Norwegian patients,1 probably due to the higher prevalence of HCV in Spain.7,13
It is generally established that patients with F-PCT develop overt clinical manifestations earlier than patients with S-PCT.1,11 Furthermore, higher values of total urine porphyrin excretion8 and liver porphyrin crystals24 are reported in F-PCT, suggesting more severe URO-D deficiency. However, we found no significant differences in the age of disease onset, severity of skin manifestations, or total urine porphyrin levels between F-PCT and S-PCT patients. In the bivariate analysis, F-PCT patients had a lower frequency of hand blisters, HCV infection, alcohol abuse, and male sex; and had lower mean fasting glucose. However, in the multivariate analysis, the only factor that remained predictive for F-PCT was negativity for HCV infection, in contrast to the results of Aarsand et al,1 who found that younger age at diagnosis, male sex, high uroporphyrin to heptaporphyrin ratio, and low alcohol consumption were predictors of F-PCT in Norwegian patients.
A total of 35.6% of patients had diabetes mellitus (23.7%) or prediabetes status (11.9%), an association observed in varying frequency in other reports,12,15,17,19,22 although its pathogenetic significance remains unclear. Although we did not find significant differences in the diabetes prevalence between patients with both types of PCT (data not shown), the higher mean fasting glucose in S-PCT patients suggests that impaired glucose metabolism could be linked to some factor overrepresented in the S-PCT group.
In conclusion, in the current study we characterize the clinical and biochemical features of a cohort of 152 Spanish PCT patients with regard to sex, and describe atypical cases that might help to prevent misdiagnosis. Although F-PCT features may vary geographically, we believe that our predictive model could be applied to other countries with a high prevalence of HCV infection. We suggest checking for URO-D erythrocyte activity in any HCV-negative PCT patient from these countries, in order to detect hereditary cases and to prevent the development of clinical manifestations of PCT in asymptomatic famiial carriers by avoidance of risk factors.
1. Aarsand AK, Boman H, Sandberg S. Familial and sporadic porphyria cutanea tarda: characterization and diagnostic strategies. Clin Chem
2. Badenas C, To-Figueras J, Phillips JD, Warby CA, Munoz C, Herrero C. Identification and characterization of novel uroporphyrinogen decarboxylase gene mutations in a large series of porphyria cutanea tarda patients and relatives. Clin Genet
3. Bleasel NR, Varigos GA. Porphyria cutanea tarda. Australas J Dermatol
4. Boffa MJ, Reed P, Weinkove C, Ead RD. Hypertrichosis as the presenting feature of porphyria cutanea tarda. Clin Exp Dermatol
5. Bonkovsky HL, Poh-Fitzpatrick M, Pimstone N, Obando J, Di Bisceglie A, Tattrie C, Tortorelli K, LeClair P, Mercurio MG, Lambrecht RW. Porphyria cutanea tarda, hepatitis C, and HFE gene mutations in North America. Hepatology
6. Brady JJ, Jackson HA, Roberts AG, Morgan RR, Whatley SD, Rowlands GL, Darby C, Shudell E, Watson R, Paiker J, Worwood MW, Elder GH. Co-inheritance of mutations in the uroporphyrinogen decarboxylase and hemochromatosis genes accelerates the onset of porphyria cutanea tarda. J Invest Dermatol
7. Bruguera M, Forns X. [Hepatitis C in Spain.] Med Clin (Barc)
8. Bygum A, Christiansen L, Petersen NE, Horder M, Thomsen K, Brandrup F. Familial and sporadic porphyria cutanea tarda: clinical, biochemical and genetic features with emphasis on iron status. Acta Derm Venereol
9. Cruz-Rojo J, Fontanellas A, Moran-Jimenez MJ, Navarro-Ordonez S, Garcia-Bravo M, Mendez M, Munoz-Rivero MC, de Salamanca RE. Precipitating/aggravating factors of porphyria cutanea tarda in Spanish patients. Cell Mol Biol (Noisy-le-grand)
10. Elder GH. Porphyria cutanea tarda. Semin Liver Dis
11. Elder GH. Porphyria cutanea tarda and related disorders. In: Kadish KM, Smith KM, Guilard R, eds. The Porphyrin Handbook: Medical Aspects of Porphyria
. Vol 14. San Diego, CA: Academic Press; 2003:67-92.
12. Grossman ME, Bickers DR, Poh-Fitzpatrick MB, Deleo VA, Harber LC. Porphyria cutanea tarda. Clinical features and laboratory findings in 40 patients. Am J Med
13. Herrero C, Vicente A, Bruguera M, Ercilla MG, Barrera JM, Vidal J, Teres J, Mascaro JM, Rodes J. Is hepatitis C virus infection a trigger of porphyria cutanea tarda? Lancet
14. Koszo F, Morvay M, Dobozy A, Simon N. Erythrocyte uroporphyrinogen decarboxylase activity in 80 unrelated patients with porphyria cutanea tarda. Br J Dermatol
15. Lamon JM, Frykholm BC. Pregnancy and porphyria cutanea tarda. Johns Hopkins Med J
16. Lim CK, Peters TJ. Urine and faecal porphyrin profiles by reversed-phase high-performance liquid chromatography in the porphyrias. Clin Chim Acta
17. Lundvall O, Weinfeld A, Lundin P. Iron storage in porphyria cutanea tarda. Acta Med Scand
18. McManus J, Blake D, Ratnaike S. An assay of uroporphyrinogen decarboxylase in erythrocytes. Clin Chem
19. Mendez M, Rossetti MV, Batlle AMDC, Parera VE. The role of inherited and acquired factors in the development of porphyria cutanea tarda in the Argentinean population. J Am Acad Dermatol
20. Pascual C, Garcia-Patos V, Bartralot R, Pedragosa R, Capdevila M, Barbera J, Castells A. Cutaneous pigmentation, only manifestation of porphyria cutanea tarda in a HIV-1 positive patient. Ann Dermatol Venereol
21. Ramsay CA, Magnus IA, Turnbull A, Baker H. The treatment of porphyria cutanea tarda by venesection. Q J Med
22. Rossmann-Ringdahl I, Olsson R. Porphyria cutanea tarda in a Swedish population: risk factors and complications. Acta Derm Venereol
23. Sarkany RP. The management of porphyria cutanea tarda. Clin Exp Dermatol
24. Siersema PD, Rademakers LH, Cleton MI, ten Kate FJ, de Bruijn WC, Marx JJ, Wilson JH. The difference in liver pathology between sporadic and familial forms of porphyria cutanea tarda: the role of iron. J Hepatol
25. Thunell S, Harper P. Porphyrins, porphyrin metabolism, porphyrias. III. Diagnosis, care and monitoring in porphyria cutanea tarda-suggestions for a handling programme. Scand J Clin Lab Invest
26. Toll A, Celis R, Ozalla MD, Bruguera M, Herrero C, Ercilla MG. The prevalence of HFE C282Y gene mutation is increased in Spanish patients with porphyria cutanea tarda without hepatitis C virus infection. J Eur Acad Dermatol Venereol