вход по аккаунту


Past use of oral contraceptives and the risk of developing systemic lupus erythematosus.

код для вставкиСкачать
Vol. 40,No. 5 , May 1997, pp 804-808
0 1997, American College of Rheumatology
Objective. To examine the relationship between
past use of oral contraceptives (OCs) and development
of systemic lupus erythematosus (SLE).
Methods. Prospective cohort study of 121,645
women who were followed up every 2 years between 1976
and 1990 as part of the Nurses’ Health Study. Women
were classified as never users or past users of OCs
based on self-report. Incidence of SLE was defined by 1)
strict American College of Rheumatology (ACR) classification criteria ( 2 4 ACR criteria), 2) 2 4 ACR criteria
and any physician’s diagnosis, 3) 2 4 ACR criteria and
diagnosis by an ACR-certified rheumatologist, 4) 1 3
ACR criteria, or 5 ) diagnosis by a physician even if the
patient did not meet the ACR criteria.
Results. Compared with never users of OCs, and
after adjusting for age and ever use of postmenopausal
hormones, the relative risk (95% confidence interval
[95% CI]) for the incidence of SLE in the women who
had definite cases of SLE ( 2 4 ACR criteria) (n = 99)
was 1.4 (0.9-2.1) for past users of OCs. Using the most
stringent case definition (ACR criteria plus a diagnosis
of SLE by an ACR member) (n = 58), the relative risk
for past users compared with never users was 1.9 (95%
CI 1.1-33). No relationship was observed between duration of OC use or time since first use and the risk of
developing SLE.
Supported by NIH grants CA-40396 and AR-36308. Dr.
Sanchez-Guerrero’s work was supported by a Research Fellowship
Award, Fogarty International Center, NIH 5FO5-TWO4573-02.
Jorge Sanchez-Guerrero, MD, MSc (current address: Instituto Nacional de la Nutricion Salvador Zubiran, Vasco de Quiroga,
Mexico), Elizabeth W. Karlson, MD, Matthew H. Liang, MD, MPH:
Harvard Medical School, Multipurpose Arthritis and Musculoskeletal
Diseases Center, Boston, Massachusetts; David J. Hunter, MD, DrPH,
Frank E. Speizer, MD, Graham A. Colditz, MD, DrPH: Channing
Laboratory, Brigham and Women’s Hospital, and Harvard School of
Public Health, Boston, Massachusetts.
Address reprint requests to Graham A, Colditz, MD, DrPH,
Nurses’ Health Study, Channing Laboratory, 181 Longwood Avenue,
Boston, MA 02115.
Submitted for publication June 25, 1996; accepted in revised
form November 6, 1996.
Conclusion. Past use of OCs was associated with
a slightly increased risk of developing SLE. The decision
to use hormonal contraception must be individualized,
but the small absolute risk observed for the development of SLE in white women should not be a dominant
factor in the decision.
Systemic lupus erythematosus (SLE) is an autoimmune disease of unknown etiology. Risk factors considered to be important in its pathogenesis include
genetic, environmental, and hormonal factors (1).There
is strong circumstantial clinical, experimental, and epidemiologic evidence to implicate the involvement of sex
steroid hormones in the pathogenesis of SLE (2). We
have previously demonstrated an increased risk of developing SLE in women who have received postmenopausal estrogen replacement (3).
To examine the relationship between oral contraceptive (OC) use and the development of SLE, we
analyzed data from the Nurses’ Health Study cohort.
Based on 14 years of followup, we evaluated the effect of
never use, past use, and the length of use of OCs on the
incidence of SLE.
The Nurses’ Health Study cohort. In June 1976, the
Nurses’ Health Study cohort was assembled from among a
population of female, married registered nurses, ages 30-55
years, living in 11 US states. The questionnaire used to
assemble the group sought information on a variety of health
conditions and exposures, such as use of OCs, use of postmenopausal hormones (PMH), current and past cigarette
smoking habits, and other health practices. Biennial questionnaires have been completed to update the information on
reproductive decisions, cigarette smoking, and the occurrence
of major health conditions.
Ascertainment of past use of OCs. In 1976, the women
were asked whether they had taken OCs, and if so, for how
long. Information on OC use was updated every 2 years
through 1982, with questions about current use and duration of
use. After 1982, few women in the cohort were still taking OCs.
Based on the womens’ responses, we calculated the time since
first use.
Other risk factors. Since use of PMH, cigarette smoking, and body mass index (BMI) may alter endogenous levels
of estrogens, adjustment for these variables was included in the
analysis. Women were classified as never users or ever users of
PMH (3). Cigarette smoking was categorized as never, past, or
current (1-14, 15-24, or 2 2 5 cigarettes per day). BMI was estimated by using the Quetelet’s Index (weight [kglheight [m’]).
Case definition of SLE. Questions regarding the diagnosis of “any rheumatic condition” since 1976 were included
on every questionnaire from 1980 onwards. SLE was included
in the 1982, 1984, 1986, and 1992 questionnaires. In addition,
questions about “other major illnesses diagnosed” were included in every biennial questionnaire.
We mailed a supplementary questionnaire to all participants who reported any rheumatic, musculoskeletal, or
connective tissue disease. The supplementary questionnaire
contained questions about symptoms and/or signs of connective tissue diseases based on the American College of Rheumatology (ACR) classification criteria for SLE (4). These
criteria have a sensitivity of 96% and a specificity of 86% for
SLE (5). The response rate to the supplementary questionnaire was 87%.
For all potential cases of SLE, all available medical
records were obtained and used to determine the case status.
The medical record from each case was reviewed independently by 2 rheumatologists who were blinded to exposure
information. When the rheumatologists disagreed about
whether a specific criterion was present or not, the complete
medical information was reviewed by a third independent
rheumatologist and a final judgment was made.
Deaths were identified from next of kin, postal authorities, and others who returned questionnaires that were mailed
to the deceased cohort member. In addition, we searched state
vital statistics and the National Death Index for nonrespondents in each followup cycle. Deaths were confirmed and cause
of death was documented through medical record review.
Deaths attributable to SLE were included.
A “definite” case was defined according to the ACR
criteria for SLE (4). Patients with definite cases were those
who had 4 or more of the ACR criteria for SLE, as confirmed
by chart review. Among the definite cases, we further classified
them into 2 subgroups of patients using a more stringent case
definition: 1) patients who had 4 or more of the ACR criteria
for SLE and also whose physician recorded a diagnosis of SLE,
and 2) patients who had 4 or more of the ACR criteria and also
received their diagnosis from an ACR member. These were
not mutually exclusive groups.
The use of ACR criteria for diagnosing SLE biases
against including milder or atypical cases or patients who do
not fulfill criteria early in their disease (6). This could underestimate the true incidence of the disease. Therefore, we
performed a further analysis on 2 groups of patients with
“probable” SLE: 1) patients who met 3 or more of the ACR
criteria for SLE, and 2) patients who were diagnosed with SLE
by their physicians, even if they did not meet enough ACR
criteria by record review. These 2 groups were not mutually
Population for analysis. Women for whom information on OC use was missing were excluded from the analysis. A
woman was considered a past user of OCs if she answered
affirmatively any of the questions related to OC use in any
biennial questionnaire between 1976-1982.
In 1976, a total of 121,645 women entered the analysis,
and during the 14-year period through May 31,1990, 1,582,910
person-years of followup were accrued. Only patients with SLE
diagnosed between June 1976 and May 31,1990 were included.
To identify those women who were free from SLE at each
baseline period, any who were diagnosed with SLE at the
beginning of each time period were excluded from subsequent
Statistical analysis. For each participant, the persontime was allocated to the various categories of OC use, based
initially on the information gathered in 1976, and thereafter,
according to the updated information. Person-time was calculated from 1976 until either May 31,1990, the date of diagnosis
of SLE, the date when the participant was lost to followup, or
death, whichever came first. The person-time of followup was
reassigned according to the respondent’s current status at the
beginning of each of the 2-year intervals. If no questionnaire
was returned for a 2-year followup period, the most recent data
were applied to the subsequent followup interval. Incident
cases of SLE were allocated to the exposure status reported at
the beginning of the interval during which they were diagnosed.
The analysis was based on incidence-density rates
using person-years of followup as the denominator. Relative
risk (RR), the measure of association, was defined as the
incidence rate of SLE among women in various categories of
OC use divided by the corresponding rate among women who
never used OCs. Age-specific rates of SLE for users and
nonusers were calculated in 5-year categories of age and used
to compute age-adjusted RR with 95% confidence intervals
(95% CI) (7). The Mantel extension test (7) was used to assess
dose-response relationships. All P values were 2-tailed.
Incidence of SLE. During 1,582,910 person-years
of followup, 99 cases of definite SLE were confirmed.
Women who were past users of OCs accounted for 45%
of the total followup time. The number of cases and the
incidence rate (IR) per 100,000 women in the different
OC categories were as follows: never users, 45 cases
(IR = 5.2); past users, 54 cases (IR = 7.5). Current users
accrued 22,873 person-years, with no cases of SLE
occurring in this group. Thus, we analyzed the whole
population based on past use of OCs only.
OC use and the risk of developing SLE (Table 1).
Among the women with definite SLE (n = 99), past
users of OCs had an age- and PMH use-adjusted RR of
developing SLE of 1.4 (95% CI 0.9-2.1) compared with
never users. We further analyzed the group of patients
with definite SLE according to a more stringent case
definition: 1) 4 or more criteria for SLE plus a diagnosis
confirmed by a physician (n = 88), and 2) 4 or more
criteria for SLE plus a diagnosis confirmed by an
ACR-certified rheumatologist (n = 58). The age- and
PMH use-adjusted RR was 1.5 (95% CI 1.0-2.3) for the
former group of past OC users and 1.9 (95% CI 1.1-3.3)
for the latter group of past OC users.
Among the 2 groups of patients having a diagnosis defined as “probable” SLE, there were 155 patients
with 2 3 ACR criteria by chart review. In this group, past
users of OCs had an age- and PMH use-adjusted RR of
1.3 (95% CI 0.9-1.8). A physician’s diagnosis of SLE,
independent of meeting the classification criteria for
SLE, was confirmed by medical chart review for 142
patients. In this group, past users of OCs had an ageand PMH use-adjusted RR of 1.2 (95% CI 0.8-1.6).
Risk of SLE according to different characteristics of OC use (Table 2). We examined whether there
was any association between duration of OC use, time
since first OC use, and time since last OC use and the
risk of SLE among definite cases. Overall, no significant
increase was seen in the risk of developing SLE according to the duration of OC use (test for trend P = 0.43).
Only women who had used OCs for 1-4 years had a
significantlyincreased age-adjusted RR (1.6 [95% CI 1.0Table 1. Past use of oral contraceptives (OCs) and relative risk for
developing systemic lupus erythematosus (SLE)*
SLE category, OC use
Definite cases
2 4 criteria
Past users
Never users
Definite cases with more
stringent case definition
2 4 criteria + MD’s dx
Past users
Never users
2 4 criteria + ACR MDs dx
Past users
Never users
Probable cases
2 3 criteria
Past users
Never users
MD’s diagnosis
Past users
Never users
cases, no.
Relative risk
(95% C1)t
1.4 (0.9-2.1)
1.5 (1.0-2.3)
1.9 (1.1-3.3)
1.3 (0.9-1.8)
1.2 (0.8-1.6)
* Patients with 4 or more American College of Rheumatology (ACR)
criteria were considered to have definite SLE. Among the definite
cases, more stringent case definitions were stated: patients with 4 or
more ACR criteria plus a physician’s diagnosis (MDs dx) or 4 or more
ACR criteria plus an ACR-certified rheumatologist’s diagnosis. Patients with 3 or more criteria or an MDs dx were considered to have
probable SLE. Person-years of followup were the same values for the
5 categories. Past users had 715,541 person-years and never users had
867,369 person-years of followup.
t The relative risk was adjusted for age and postmenopausal hormone
use. 95% CI = 95% confidence interval.
Table 2. Characteristics of oral contraceptive (OC) use and relative
risk for developing systemic lupus erythematosus (SLE) among patients with definite SLE*
Characteristic of OC use
Duration of use
1-4 years
5-10 years
11-14 years
215 years
Total no. of persons
P trend
Time since first use
1-5 years
6-10 years
11-15 years
16-20 years
220 years
Total no. of persons
P trend
Time since last use
1-35 months
36-59 months
60-119 months
120-179 months
180-239 months
2240 months
Total no. of persons
P trend
SLE cases, Person-years Relative risk
of followup
(95% CI)?
1.6 (1.0-2.6)
0.9 (0.5-1.7)
2.0 (0.7-5.5)
2.9 (0.5-19.3)
2.7 (0.9-8.2)
1.0 (0.2-3.9)
0.8 (0.3-2.0)
1.5 (0.8-2.7)
1.5 (0.9-2.5)
1.8 (0.7-4.8)
1.0 (0.3-3.2)
1.6 (0.9-2.8)
0.9 (0.5-1.9)
1.7 (0.9-3.3)
1.0 (0.1-7.7)
*Definite SLE was defined as 4 or more American College of
Rheumatology criteria. Information on duration of use was missing for
6 participants and 36,242 person-years of followup. Information on
time since first use was missing for 3 participants and 16,654 personyears of followup. Information on time since last use was missing for 7
participants and 53,961 person-years of followup. The never-use
category was the referent in each comparison.
t The relative risk was adjusted for age and postmenopausal hormone
use. 95% CI = 95% confidence interval.
2.61). No significant trend was observed according to the
time since first OC use (P = 0.17). No association was seen
with time since last use of OCs (test for trend P = 0.21).
Similar results were observed in the 2 subgroups
of patients with definite SLE using a more stringent case
definition, as well as in the 2 groups of patients with
probable SLE (data not shown).
Other risk factors. Among definite cases, the
age- and PMH use-adjusted RR of SLE in past users of
OCs, adjusting for cigarette smoking habits, was 1.5
(95% CI 1.0-2.3). Similar results were also seen for the
association of duration of use, time since first use, and
time since last use of OCs and the risk of SLE (data not
shown). Further adjustment for BMI did not change the
magnitude of the association, and its direction remained
unchanged (data not shown).
In this large prospective cohort study of 121,645
women with 1,582,910person-years of followup, past use
of OCs was associated with a slightly increased risk of
SLE compared with never use of OCs. With the 5 case
definitions used, increased relative risks were seen as the
case definition became more stringent. The relative risk
was significantly increased only among definite cases
with the most stringent case definition. No relationship
was noted according to the duration of use, time since
first use, or time since last use of OCs.
This is the first epidemiologic study showing a
slight risk of developing SLE in association with a
history of OC use. Two previous studies have suggested
little or no association. In a case-control study of 195
SLE patients and 143 of their friends (8), OC use was
included among the many factors studied, and no association was present between SLE and either any use or
recent use of OCs (odds ratio [OR] = 0.6, 95% CI
0.2-1.4). A potential limitation of that study, however,
was the choice of controls, since friends may be overmatched regarding personal and environmental exposures to cases. This would tend to bias the results toward
the null hypothesis for these variables. The authors
concluded that if SLE is caused by the use of OCs, the
effect must be extremely modest. A case-control study
published in 1983 (in abstract form) evaluated 74 female
SLE patients and age-, sex-, and race-matched controls, and found no association between OC use and
development of SLE (9). The ORs were not provided,
which makes it difficult to fully evaluate the results of
that study.
Several case reports (10-13), as well as clinical
and laboratory observations, support the biologic plausibility of the finding. The female-to-male ratio with
regard to age at onset of SLE or based on the diagnosis
of SLE after puberty reaches 8:l (2). Abnormal metabolism of estrogens yields an excess production of 16
a-hydroqestrone in SLE patients of both sexes, but
women have the additional elevation of estriol(l4). Low
plasma androgen levels have been reported in women
with active and quiescent SLE (15,16). Flares in SLE
have been reported during periods of major sex hormone changes, such as menses, pregnancy, and postpartum (17). Studies in the New Zealand black X New
Zealand white (NZB X NZW) mouse, a murine model
of lupus, support a role for female hormones (18).
Female (NZB X NZW)F, mice have increased levels of
autoantibodies and die several months earlier than
males. Androgen treatment prolongs survival and reduces immune complex deposition and glomerulonephritis in female (NZB X NZW)F, mice (19), while
prepubertal castration of male (NZB X NZW)F, mice
plus estrogen administration causes an essentially female pattern of disease (20). Overall, the observed
interaction of sex hormones with the immune system has
been consistent.
Several potential limitations of this study require
discussion. Since information on OC use was obtained
by self-report, misclassification of OC use may have
occurred. However, this misclassification should have
been minimal, since the subjects were registered nurses.
Self-report has been shown to be highly accurate in this
cohort (21). Furthermore, any random misclassification
in OC use would underestimate the true association.
The prospective design of this study precludes the
possibility that these results were explained by recall
bias, since information on OC use was collected prior to
the diagnosis of SLE.
Another potential bias was the increased opportunity for the diagnosis of SLE among the women using
OCs, since these women were under medical care. One
way to examine this potential bias would be to restrict
the analysis to those women who had reported at least 1
physician visit during the period 1988-1990 (87% of the
cohort). However, there were not enough patients with
SLE to perform the analysis. Since the members of this
cohort were registered nurses, the opportunity for diagnosis should have been equivalent irrespective of OC
use. Furthermore, SLE is a chronic illness with periods
of remission and exacerbation, and it is improbable that
a woman would have remained undiagnosed once
Given the difficulty of SLE diagnosis, symptoms
may be present for some time before the diagnosis is
made. Thus, OCs might precipitate more manifestations
rather than cause the disease. Although this is a possibility, nurses with rheumatic symptoms might not be
prescribed OCs. We believe that both possibiiities were
likely, but were isolated, random, and probably did not
influence the results. Whether or not OCs induce SLE
flares cannot be answered with the results of this study,
but is the subject of an ongoing clinical trial.
A major concern in any population study of SLE
is the complexity of diagnosis and the necessity to use
standardized criteria in a heterogeneous disease. Using
the ACR criteria for SLE underestimates the true
incidence of disease. However, the incidence rate of SLE
in this study, 5.2 cases and 7.5 cases per 100,000persons,
was similar to rates reported in other studies (range
3.8-15.9 per 100,000persons) (22-25).
In a rare disease, a small underestimate of incidence is less important in a study of etiology than is
classifying nondiseased subjects as diseased (21).Adding
a small number of true cases to the very large number of
true noncases will have almost no influence on estimates
of the distribution of exposures among the noncases. In
contrast, use of less strict criteria will add noncases to
the cases, and, since they are relatively small, could
make up an appreciable proportion of the total cases,
with the potential for materially disturbing the apparent
distribution of exposures among cases. If the misclassification is random or nondifferential with respect to exposure,
the risk estimate will be driven toward the null value.
We analyzed the data using a variety of case
definitions, and the results held. An increase in the risk
of SLE was observed only with the more stringent case
definition. This could be due either to the exclusion of
noncases or to the inclusion of more severely affected
cases after excluding milder cases, atypical cases, or
cases at an early stage of the disease. If the latter is true,
OCs may be associated with more severe disease, not
with increasing the risk of developing SLE.
It is also possible that SLE might have been
classified as another connective tissue disease or that
other diseases were misclassified as SLE, since these
diseases may have overlapping manifestations. This misclassification would have been random and should not
bias the results.
Our study did not have enough power to analyze
the risk of SLE according to current use or type of OCs,
or to family history of connective tissue disease. Because
the cohort was 95% white, we could not study the effect
of race.
The benefits and risks of using OCs represent a
complicated decision for the individual patient (26). Our
results should be considered in the choice of a contraceptive method for healthy white women, but since the
increase in the absolute risk for developing SLE is very
small, it will usually not be a dominant factor.
T h e authors gratefully acknowledge the continuing
participation of the nurses in this study, the helpful comments
of Dr. Sue E. Hankinson on this manuscript, and the expert
assistance of Karen Corsano and Barbara Egan.
1. Alarcon-Segovia D: The pathogenesis of immune dysregulation in
systemic lupus erythematosus: a troika. J Rheumatol 11:588-590,
2. Masi AT, Kaslow RA: Sex effects in systemic lupus erythematosus:
a clue to pathogenesis. Arthritis Rheum 21:480-484, 1978
3. Sanchez-GuerreroJ, Liang MH, Karlson EW, Hunter DJ, Colditz
G A Postmenopausal estrogen therapy and the risk for developing
systemic lupus erythematosus. Ann Intern Med 122430-433, 1995
4. Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield
NF, Schaller JG, Talal N, Winchester RJ: The 1982 revised criteria
for the classification of systemic lupus erythematosus. Arthritis
Rheum 25:1271-1277, 1982
5. Karlson EW, Sanchez-Guerrero J, Wright EA, Lew RA, Daltroy
LH, Katz JN, Liang MH: A connective tissue disease screening
questionnaire (CSQ) for population studies. Ann Epidemiol
6. Lam-Orta H, Alarcon-SegoviaD, Diaz-Jouanen E Systemic lupus
erythematosus: differences between patients who do, and who do
not, fulfill classification criteria at the time of diagnosis. J Rheumatol 7831-837, 1980
7. Rothman KJ, Boice J D Epidemiologic Analysis with a Programmable Calculator (NIH publication no. 79-1649). Washington DC,
Public Health Service, 1979
8. Strom BL, Reidenberg MM, West S, Snyder ES, Freundlich B,
Stolley PD: Shingles, allergies, family medical history, oral contraceptives, and other potential risk factors for systemic lupus erythematosus. Am J Epidemiol 140:632-642, 1994
9. Hochberg MC, Kaslow RA: Risk factors for the development of
systemic lupus erythematosus: a case-control study (abstract). Clin
Res 31:732A, 1983
10. Spiera H, Plotz CM: Rheumatic symptoms and oral contraceptives
(letter). Lancet i:571, 1969
11. Garovich M, Agudelo C, Pisko E: Oral contraceptives and systemic lupus erythematosus. Arthritis Rheum 23:1396-1398,1980
12. Chapel TA, Burns R E Oral contraceptives and exacerbation of
lupus erythematosus. Am J Obstet Gynecol 110:366-369, 1971
13. Schleicher EM: LE cells after oral contraceptives. Lancet k821822, 1968
14. Lahita RG, Bradlow HL, Kunkel HG, Fishman J: Increased 16-(Y
hydroxylation of estradiol in systemic lupus erythematosus. J Clin
Endocrinol Metab 53:174-178, 1981
15. Jungers P, Nahoul K, Pelissier C, Dougados M, Tron F, Bach J - F
Low plasma androgens in women with active or quiescent systemic
lupus erythematosus. Arthritis Rheum 25:454-457, 1982
16. Lahita RG, Bradlow HL, Ginzler E, Pang S, New M: Low plasma
androgens in women with systemic lupus erythematosus. Arthritis
Rheum 30:241-248, 1987
17. Dubois EL, Strain L, Mita E, Bernstein G, Friou G J LE cells after
oral contraceptives (letter). Lancet ii:679, 1968
18. Kelley VE, Winkelstein A Age- and sex-related glomerulonephritis in New Zealand white mice. Clin Immunol Immunopathol
16:142-150, 1980
19. Roubinian JR, Talal N, Greenspan JS, Goodman JR, Siiteri P K
Delayed androgen treatment prolongs survival in murine lupus.
J Clin Invest 63:902-911, 1979
20. Roubinian JR, Talal N, Greenspan JS, Goodman JR, Siiteri P K
Effect of castration and sex hormone treatment on survival,
anti-nucleic acid antibodies, and glomerulonephritisin NZB/NZW
F1 mice. J Exp Med 147:1568-1583, 1978
21. Colditz GA, Martin P, Stampfer MJ, Willett WC, Sampson L,
Rosner B, Hennekens CH, Speizer FE. Validation of questionnaire
information on risk factors and disease outcomes in a prospective
cohort study of women. Am J Epidemiol 123:894-900, 1986
22. Michet CJ Jr, McKenna CH, Elveback LR, Kaslow RA, Kurland
L T Epidemiology of SLE and other connective tissue diseases in
Rochester, Minnesota, 1950 through 1979. Mayo Clin Proc 6 0
105-113, 1985
23. Hochberg M C The incidence of systemic lupus erythematosus in
Baltimore, Maryland, 1970-1977. Arthritis Rheum 28230-86, 1985
24. Nived 0,Sturfelt G, Wollheim F Systemiclupus erythematosus in
an adult population in southern Sweden: incidence, prevalence
and validity of ARA revised classification criteria. Br J Rheumatol
24~147-154, 1985
25. Siege1 M, Lee S L The epidemiology of systemic lupus erythematosus. Semin Arthritis Rheum 3:l-54, 1973
26. Baird DT, Glasier AF: Hormonal contraception. N Engl J Med
328:1543-1549. 1993
Без категории
Размер файла
576 Кб
past, contraceptive, lupus, developing, systemic, erythematosus, oral, use, risk
Пожаловаться на содержимое документа