Download Chronic urticaria and cancer: an epidemiological study of I 155

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts
no text concepts found
Transcript
British Journal of Dermatology (1990) 123,453-456.
Chronic urticaria and cancer:
an epidemiological study of
I
155 patients
B.LINDELOF, B.SIGURGEIRSSON, C.F .WAHLGREN AND G.EKLUND*
Depanments of Dermatology and .. Cancer Epidemiology, Karolinska Hospital, Stockholm, Sweden
Accepted for publication 25 April II}9O
SUMMARY
To evaluate the possible aSSociation of malignant disease with chronic urticaria
u55
consecutive cases with chronic urticaria were reviewed. The Swedish Cancer Registry,
Stockholm, was searched for records
in the
population (1958-84),
and the expected number of malignancies was calculated on the basis of age- and sexstandardized incidence data. A malignancy was diagnosed in 36
eXlpec:tea number of malignancies was 41. In 23 patients the
with urticaria and the
appeared during the
same year as the onset of urticaria or later. The expected number was 25'6. We conclude that
chronic urticaria is not statistically associated with malignancy in general.
Many dermatoses have been considered to be related to an underlying internal malignancy.1
Urticaria and angio-oedema have been reported mainly in association with lymphoproliferative
disorders.2-16 However, not every dermatosis occurring in a patient
disease is causally related. When the dermatosis is common, as with
from malignant
the possibility
increases that the association has occurred only by chance. We investigated the incidence of
cancer in a
number of patients with chronic urticaria and related the observed number of
malignancies to the number expected, in a contemporary population with the same age and sex
distribution. The incidence of cancers as well as the expected number of cancers were obtained
from the Swedish Cancer Registry, Stockholm.
METHODS
Patients During the years 1968-83, 1155 patients with chronic urticaria were seen at the Department of Dermatology, Karolinska Hospital, Stockholm. All had had symptoms for more than 3 months. The patients consisted of 704 females and 451 males
(Fig.
age 32 years, age range
years) I). The follow-up period was on average 8'2 years. Correspondence: Dr B.LindelOf, Department of Dermatology, Karolinska Hospital, S-104 O! Stockholm, Sweden.
127
\
,
250
200
'"
'E
,,!!
i
150
'0
cl 100
z
50
o
0-9
10-19 20-2930-39 40-4950-5960-69 70-79 80t
Age groups
FIGURE I. Age distribution of I ISS patients with chronic urticaria. Ill, males; 0, females.
Swedish Cancer Registry
Information from the Swedish Cancer Registry, Stockholm, (1958-84) and the 1155 patients
was correlated to identify individuals with malignant tumours. Nationwide information on the
cancer incidence in Sweden is available from 1958, when compulsory registration was begun.
The registry collects information on diagnosed cancers from both clinicians and pathologists so
that most cases are reported by two sources. Each patient is characterized by a unique
identification number composed of six
based on year, month, and day of birth,
supplemented with a registration number
digits) and a check digit. Therefore, the
identification numbers are not affected by the possibility of changes in names. The completeness
re,ns.1trll'hon in the Swedish Cancer Registry has been found to be 96-97% for all cancers. t 7
Statistics
The expected number of malignant tumours was estimated on the basis of incidence data from
the Swedish Cancer Registry, Stockholm. III The age- and
calculated as a national average for
cancer incidence was
The incidence rates were assumed to be constant
within 5-year age and calendar-time intervals. Regional differences were
To allow
for deaths occurring during the observation period, deductions were made on the basis of life
tables for the whole Swedish population. To derive the expected number of malignant tumours,
the number of person years at risk in each group was multiplied by the sex-specific and age­
specific cancer incidence for the relevant 5-year period. The number of person-years of
observation was calculated separately for each sex and amounted to 9400. After the ratio
between observed and expected number of malignancies had been estimated, significance and
confidence interval analysis was performed using the Poisson distribution.
RESULTS
number was 41
A malignant tumour was diagnosed in 36 patients with urticaria. The
and thus the relative risk was 0'88, confidence interval 0,61-1 '22. Of these patients with
malignancy, 23 cancers appeared during
the same year as the onset of urticaria or
128
The
TABLI! I.
Observed and expected number of malignancies
in 1155 patients with chronic urticaria
Number of malignancies
Observed
Expected
All patients (n 1155):
Females (n 23)
Males (n 13)
23
13
26·8 NS
14'2 NS
Total
36
41·oNS
=
=
=
Cancer the same year as onset of urticaria or later
14
Females (n= 14)
15'7 NS
Males (n= 9)
9
9'9 NS
Total
23
NS, not significant.
TABLI!
2. Characteristics of 10 patients with chronic urticaria associated with
malignancy (cancer within 5 years after onset of urticaria)
Age at onset (years) of
Case
Sex
8
M
M
M
M
M
M
F
F
9
10
F
F
2
3
4
5
6
7
Urticaria
Malignancy
Type of malignancy
79
83
73
71
72
63
52
65
60
Rectum
Rectum
Colon
Stomach
Malignant melanoma
Skin, squamous cell carcinoma
Ovary
69
67
69
60
47
61
58
48
78
Corpus uteri
Glioblastoma
Malignant lymphoma
49
79
expected number was 25'6. These differences are not statistically significant (Table I).
Characteristics of those 10 patients who had their malignancy the same year or within 5 years
after onset of urticaria are given in Table 2. The study population is shown in Figure I.
DISCUSSION
This study strongly suggests that chronic urticaria is not statistically associated with
malignancy. The study population is well defined and the number of observed and expected
cancers are reliable owing to the accurate population statistics used in Sweden.
The link between skin markers and underlying malignant disease has always interested
dermatologists. If
an
association between a skin lesion and internal malignancy has been
129
suggested, the dermatologist has to decide whether there is a need for investigation. If so, how
extensive should a search be and how often should the search be repeated? The patient may
undergo unnecessary procedures with both socioeconomic and psychological consequences,
particularly if the association is false. If the suggested cutaneous sign is common and chronic,
because internal malignant diseases are often chronic, it is easy to find patients with both
conditions at the same time, as illustrated in Table 2. Therefore, the only way to prove an
association of a common skin condition with internal malignancy is to perform a large and
controlled study. If, however, the dermatosis is uncommon, e.g. erythema gyratum repens, case
reports are the basis for the association as large studies cannot be carried out. We conclude that
the hypothesis that urticaria is associated with internal malignant disease was based on case
reports only, and this large-scale study has shown that this association probably occurred by
chance.
REFERENCES
I
Haynes HA, Curth HO. Cutaneous manifestations associated with malignant internal disease. In: Dermatology in
General Medicine (Fitzpatrick TB et al., eds), 2nd edn. New York: McGraw-HiIl Company, 1979.
2 Hills EA. Aaenocarcinoma of the bronchus with Cushing's syndrome, carcinoid syndrome, neuromyopathy and
urticaria. Br J Dis Chest 1968; 62: 88-92.
3 Ross TF, Coleman DL, Naughton JL. Angioedema and smaIl-cell carcinoma of the lung. Thorax 1982; 37: 950-14 Lewkonia
RM, Pineo GF. Angioedema-like skin lesions associated with Iymphoproliferative disease. Can Med
Assoc J 1982; 127: 867:-8.
5 Gottlieb M, Campbell K, Peizmann, K, Hu C-H. Long-standing angioedema with CI-esterase inhibitor deficiency
associated with occult lymphoma. West J Med 1983; 1)8: 258-&1.
6 Fiechtner 11,
Marx 11, Wolski KP, Schloesaer LL. Acquired angioedema, autoimmune hemolytic anemia and
lymphoma: resolution after therapy. Clin Immunollmnnmopath 1980; IS: 642-5.
7 Vallent K, Feher J, Somogyi A, JOOs A. Angioneurotic oedema and hypocomplementaemia in patients with thyroid
cancer. Acta Chir Acad Scim HU1I(faricae 1982; 23: 163-170 (Ger).
8 Highet AS. Urticaria vasculitis and IgA myeloma. Eastern Counties Dermatological Society March 1979. Society
Proceedi ngs. Br J DermatoI1980; 102: 355-7.
9 Jones L, Bubel MA, Newland, AC. Cl esterase inhibitor as a possible tumour marker in malignant lymphoma. Clin
.
Lab Haemat 1985; 7: 279-81.
10 Stockley RJ, Daneshmend
TK. Painful urticaria heralding the appearance of Philadelphia chromosome positive
acute myeloid leukaemia. Clin Lab Haemat 1986; 8: 161-4.
11 Gordon EH, BeaU GN, Klaustermeyer
WB Angioedema and multiple myeloma. Ann Allergy 1983; SI: 349-50.
.
12 Hasegawa J, Ke M. Cold urticaria, cryoglobulinemia and? myeloma. Arch Dermatol 1970; 101: 239-41.
13 Ben-Chetrit E, Stem Z, Polliack A. Cold urticaria: an unusual manifestation of histiocytic lymphoma. Acta Haemat
1982; 68: 343-4.
14 Rawnsley HM, Shelley
WB Cold urticaria with cryoglobulinemia in a patient with chronic lymphocytic leukemia.
.
Arch Dermatol 1968;,s: 12-17.
15 Bonvalet D, Foldes C, Civatte J. Cutaneous manifestations in chronic lymphocytic leukemia.J Dermatol Surg Oncol
1984; 10: 278-82.
16 Lindelof B, Wahlgren C-F. Chronic urticaria associated with internal malignancy. Int J Dermatol 1990; 29: 384.
17 Mattsson B. Cancer registration in Sweden. Studies on completeness and validity of incidence and mortality
regislers. Thesis. Karolinska Institute, Stockholm, 1984; 1-33.
18 Adami H-O, Gunnarson T, Sparen P, Eklund G. The prevalence of cancer in Sweden 1984. Acta Oncologica 1989;
38: 463-70.
130