close

Вход

Забыли?

вход по аккаунту

?

Incidence of hydroxychloroquine retinopathy in 1207 patients in a large multicenter outpatient practice.

код для вставкиСкачать
ARTHRITIS & RHEUMATISM
Vol. 40, No. 8, August 1997, pp 1482-1486
0 1997, American College of Rheumatology
1482
INCIDENCE OF HYDROXYCHLOROQUINE RETINOPATHY IN
1,207 PATIENTS IN A LARGE MULTICENTER OUTPATIENT PRACTICE
G. D. LEVY, S. J. MUNZ, J. PASCHAL, H. B. COHEN, K. J. PINCE, and T. PETERSON
Objective. To define the true risk of hydroxychloroquine (HCQ) retinal toxicity by studying the largest
single group yet evaluated.
Methods. Retrospective chart review of all patients in the Kaiser Permanente Medical Care Program,
Southern California Region, who had HCQ prescriptions filled from 1991 through 1993 (1,556 patients in 11
medical centers). Of 1,207 charts of patients who took
HCQ and had documented ophthalmologic examinations, initial screening identified 21 charts (1.7%) that
indicated possible HCQ toxicity.
Results. We identified 1 patient with definite
toxicity (1 of 1,207; 0.08%) and 5 other patients with
indeterminate but probable toxicity (5 of 1,207; 0.4%).
The incidence of definite HCQ retinal toxicity in patients treated with HCQ at <6.5 mg/kg/day was 0.
Conclusion. In HCQ-treated patients whose renal
function is normal, routine ophthalmic screening is not
indicated if the daily dosage is <6.5 mg/kg. In patients
whose daily dosage is >6.5 mg/kg or who have taken
HCQ continuously for > 10 years, annual screening may
be appropriate.
Antimalarial agents have been used for the treatment of rheumatic diseases since the 1950s. Early reports of the ocular toxicity of these agents were prePresented in part at the 24th Western Regional Meeting of
the American College of Rheumatology, Santa Barbara, CA, March
1996, and at the 60th National Scientific Meeting of the American
College of Rheumatology, Orlando, FL, October 1996.
Supported by a grant from the Community Service Program
of Kaiser Foundation Hospitals, Southern California Region.
Gerald D. Levy, MD: Kaiser Permanente Medical Center,
Bellflower, California; Stephen J. Munz, MD, Keith J. Pince, MD:
Kaiser Permanente Medical Center, Anaheim, California; John Paschal, MD: Kaiser Permanente Medical Center, Bakersfield, California:
Howard B. Cohen, MD, Tim Peterson, MD: Kaiser Permanente
Medical Center, San Diego, California.
Address reprint requests to Gerald D. Levy, MD, Department
of Rheumatology, 9400 East Rosencrans, Module 3100, Bellflower,
CA 90706.
Submitted for publication October 4, 1996; accepted in revised form March 21, 1997.
sented by Cambiaggi in 1957 (l),Hobbs et a1 in 1959 (2),
and Bernstein in 1967 (3). All antimalarial agents were
thought to have similar mechanisms of action as well as
similar toxicity. Much early research grouped chloroquine (CQ) and hydroxychloroquine (HCQ) together,
obscuring their different ocular toxicities (4). This difference may be due to the addition of the hydroxy group,
which limits the ability of HCQ to cross the bloodretinal barrier. The elegant study by Raines et a1 used
vitreous fluorophotometry to show a breakdown of the
blood-retinal barrier by CQ, whereas asymptomatic
patients who had received up to 1,067 gm of HCQ had
no breakdown of this barrier (5).
The incidence of retinopathy caused by HCQ is
known to be low. Bernstein’s review of the world biomedical literature found only 18 HCQ-related cases of
retinopathy in 11 reports published from 1963 through
1989 (6). Tobin et a1 reported on 99 patients followed up
for up to 7 years who experienced no permanent loss of
vision (7). Mackenzie’s large study of 900 patients
showed no eye toxicity in patients who received daily
HCQ dosages of ~ 6 . 5mgkg (8). In that study, 50
patients who had received HCQ for >11.7 years
tolerated a mean dosage of 6.49 mglkglday without
developing retinal toxicity. However, in the group that
received a mean dosage of 7.8 mglkglday, there were
mild vision changes. This finding led to the oftenquoted retinal toxicity threshold of 7.8 mglkglday for
HCQ.
The apparent safety of HCQ has led several
authors to question the need for vision screening every 6
months in patients who take the drug (6,9-11). Morsman
et a1 went a step further after their study of 73 patients
and review of the literature, suggesting that routine
screening is not necessary (12). Morand and colleagues
reached a similar conclusion after their prospective
study of 403 patents (13). They found that treatment had
been terminated within 24 months in 54% of patients
with rheumatoid arthritis and 24% of patients with
HYDROXYCHLOROQUINE RETINOPATHY
systemic lupus erythematosus, but none of the terminations were due to ocular toxicity.
The low incidence of HCQ retinal toxicity has
complicated the search for the best method for detecting
such toxicity. Many different tests and screening schedules have been proposed, often without differentiating
between HCQ and CQ toxicity. Currently, there are 2
commonly cited definitions of ocular toxicity. Bernstein’s definition requires the development of persistent
paracentral or central visual field scotomas to suprathreshold white stimuli and a duration of treatment of
>9 months when the daily dosage is >400 mg (6). A
bull’s-eye lesion suggests toxicity. However, mild peripheral visual constriction to suprathreshold stimuli, peripheral constriction or scotoma, or both in response to
red, mild macular stippling, and electroretinogram or
color vision abnormalities do not. The bull’s-eye lesion
had been called pathognomonic, but, as Weise and
Yannuzzi concluded in their 1974 article, patients may
show this macular change without exposure to CQ or
HCQ (14). In the other commonly cited definition,
Easterbrook advocates the use of bilateral, reproducible,
positive field defects that can be shown by 2 different
visual field tests, the Amsler grid test and an automated,
10-degree visual field test, as definitive evidence of
retinal toxicity (9).
The present investigation was undertaken to define the true risk of HCQ retinal toxicity by studying a
large group of treated patients.
PATIENTS AND METHODS
The study protocol was approved by the Institutional
Review Board, Kaiser Permanente Medical Care Program,
Southern California Region. All patients in the Kaiser Permanente Southern California Region who had HCQ prescriptions
filled at Kaiser Permanente pharmacies from 1991 through
1993 (1,556 patients in 11 medical centers) were identified.
Inclusion in the study required that the patients had actually
taken HCQ, had received eye examinations at Kaiser Permanente facilities, and had a chart available for evaluation.
Thirty-seven charts could not be retrieved, and 23 charts
showed no documented HCQ use. In an additional 289 patients, eye examinations were either not done at Kaiser Permanente facilities or not completed within the study period.
Thus, the final study group comprised 1,207 patients. The
initial screening process was done by chart extractors who
reviewed the entire medical record for any evidence of possible
HCQ retinal toxicity. Twenty-one charts with indications of
potential HCQ toxicity (1.7%) were identified.
The charts of the 21 possible cases were then reviewed
by an expert panel of 5 ophthalmologists which included 3
retinal specialists and 2 general ophthalmologists. The ophthalmologists were geographically dispersed throughout the
1483
Table 1. Characteristics of possible cases versus controls in an
outpatient study of the incidence of hydroxychloroquine (HCQ)
retinopathy*
Age, years
Daily HCQ dosage, mg
Total years of HCQ use
Total HCQ dose, grn
Weight, kg
Daily HCQ dosage, mgikg
No. maleino. female
% male/% female
Race, no.
White
Black
Asian
Hispanic
Other
Cases
(n = 21)
Controls
(n = 1,475)
67.4
363
4.1
542
73.5
5.2
2/19
9.5190.5
46.7
354
3.3
402
73.7
5.1
25011,225
16.9183.1
P
0.8
0.704
0.592
0.474
0.044
0.442
0.558
-
0.887
15
3
1
2
0
879
261
75
232
28
-
~~~~~
* Except for sex and race, values are means.
Southern California Region of the Kaiser Permanente system.
They independently reviewed each of the selected records and
assigned them to 1 of 4 possible designations: definite HCQ
toxicity; indeterminate, probable toxicity; indeterminate, probable no toxicity; and no HCQ toxicity. The ophthalmologists
used published criteria from both Bernstein (6) (paracentral or
central visual field scotomas, daily dose >400 mgiday for at
least 9 months) and Easterbrook (9) (presence of bilateral,
reproducible, positive field defects that can be demonstrated
by 2 different tests of the visual field) and their own clinical
experience to determine which category was most appropriate
for each patient.
Of the initial 1,496 patients (i.e., the original 1,556
minus the 59 whose charts could not be retrieved or showed no
documented HCQ use), the 1,475 whose charts showed no
indication of HCQ toxicity were used as a demographically
matched comparison group for the 21 cases of possible HCQ
toxicity. Table 1 shows the group characteristics of the 21
possible cases and the control group. No significant differences
were identified for age, daily HCQ dosage, total years of
HCQ use, total cumulative dose, daily dosage corrected for
weight (mgikg), sex, or race. The case population ranged in
age from 30 to 77 years (case mean 53.5 years, control mean
46.7 years) and included 18 women and 3 men. The ratio of
women to men was not significantly different from that in
the control group. The duration of therapy ranged from 0.4 to
25.3 years (case mean 4.1 years, control mean 3.3 years); total
HCQ consumption ranged from 61 gm to 3,679 gm (case
mean 557 gm, control mean 402 gm). The mean daily HCQ
dosage for the patients with possible retinopathy was 363
mgiday; control patients consumed a mean of 354 mgiday.
Weight-adjusted dosage ranged from 1.73 to 8.8 mgikglday
(case mean 5.2 mgikgiday, control mean 5.1 mgikglday). All of
the patients with possible retinopathy had normal renal function except for 1 patient with end-stage renal disease who was
receiving kidney dialysis treatment. The characteristics of the
individual patients with possible retinopathy are shown in
Table 2.
LEVY ET AL
1484
Table 2. Characteristics of individual patients screened for possible hydroxychloroquine (HCQ) retinopathy
Patient
Age
Raceisex*
Diagnosis?
Duration,
years
Daily HCQ
dosage, rng
Total HCQ
dose, grn
Daily HCQ
dosage, mgikg
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
48
77
69
58
46
59
64
50
42
58
64
57
34
58
67
45
68
30
33
53
39
WIF
WIF
WIF
WIF
WIM
SLE
RA, TIA
RA
RA
Asthma
RA, T4, HTN
SLE
SLE
SLE, DM
SLE, DM, HTN
RA
SLE, HTN
RA
RA, HTN, T4
CA, vR
RA
RA
Rosacea
SLE, RA
RA, CAD
SLE, HTN, R F
7.3
3.5
4.8
2.8
1.8
2.5
0.6
5.0
1.6
0.9
1.8
5.1
1.8
25.3
3.5
2.3
9.0
1.3
0.4
3.7
1.6
400
400
200
400
200
400
400
400
400
400
400
600
400
400
400
400
200
200
400
400
200
1,066
511
343
413
128
365
73
730
231
131
256
1,113
256
3,679
511
336
657
97
61
535
112
6.98
6.82
3.44
5.43
1.94
6.47
6.38
6.38
3.68
5.18
6.15
6.6
6.11
4.56
8.8
5.64
2.93
1.73
5.99
5.75
2.7
36313.54
5511402
5.215.1
Mean,
caseslcontrols
AIF
WIM
WIF
Wff
BIF
Wff
BIF
HIF
BIF
WIF
WIF
WIF
WIF
HIF
WIM
WIF
53.3146.7
~
~~~
~
~~~
* W = white; A = Asian; B = black; H = Hispanic.
t SLE = systemic lupus erythematosus; RA = rheumatoid arthritis; TIA = transient ischemic attack; T4 = thyroid disease; HTN
DM = diabetes mellitus; CA = cancer; vR = von Recklinhausen disease; CAD = coronary artery disease; R F = renal failure.
RESULTS
Each chart screened for possible HCQ toxicity
was reviewed by the 5 ophthalmologists. Determinations
made based on these chart reviews, and the final designation for each patient based on the determinations
by all 5 ophthalmologists, are shown in Table 3. The
designations were deliberately conservative: patients
were placed in a higher category (ie., indicating more
likely toxicity) if the most appropriate category for them
was ambiguous. We identified only 1 patient with definite toxicity (1 of 1,207; 0.08%) and 5 other patients with
indeterminate but probable toxicity (5 of 1,207; 0.4%).
Eleven patients were classified as having indeterminate
but probably negative HCQ toxicity (11 of 1,207, 0.9%),
and 4 patients were classified by all 5 ophthalmologists
as having no toxicity.
The patient with definite toxicity had received an
HCQ dosage of 6.98 mg/kg/day, and 2 of the 5 patients
with probable toxicity had received dosages that also
exceeded 6.5 mg/kg/day (patient 2 received 6.82 mg/kg/
day and patient 12 received 6.60 mg/kg/day). The case of
patient 2 illustrates the difficulty inherent in categorizing some patients due to coexisting morbidities: she had
cataracts, occipital cerebral vascular accidents, and pre-
=
hypertension;
existing macular changes, which made absolute designation impossible. One patient was originally thought to
have retinal toxicity because a classic bull’s-eye lesion
was found; however, fluorescein angiography showed a
scar that was not consistent with HCQ retinal toxicity.
The incidence of definite HCQ retinal toxicity in patients treated with HCQ at <6.5 mg/kg/day was 0.
DISCUSSION
Even in a study of this size, important questions
remain about the true incidence of HCQ retinal toxicity.
All patients did not have the same examination, because
the evaluations were performed by dozens of examining
ophthalmologists at 11 different medical centers. Patients classified as having indeterminate toxicity could
have been easily classified definitively if we simply had
additional information which was unavailable because
the study design was retrospective.
Eye examinations were done at Kaiser Permanente facilities for 1,207 of 1,496 patients (81%). In
addition, there may have been some patients who received examinations at outside facilities not covered by
the health plan, which would have resulted in a higher
overall compliance rate. Patient compliance in this study
HYDROXYCHLOROQUINE RETINOPATHY
1485
Table 3. Summary of ophthalmologists’ findings in review of selected patients taking hydroxychloroquine (HCQ)*
Patient
1
2
3
4
5
6
7
8
Y
10
11
12
13
14
15
16
17
18
19
20
21
Dr. A
Dr. B
Dr. C
Dr. D
Dr. E
Rating of case+
Indeterm+
No toxicity
Indeterm No toxicity
No toxicity
IndetermNo toxicity
No toxicity
IndetermNo toxicity
No toxicity
No toxicity
No toxicity
Indeterm No toxicity
No toxicity
No toxicity
No toxicity
No toxicity
No toxicity
No toxicity
Toxicity
Indeterm +
IndeternIndeterm No toxicity
IndetermNo toxicity
No toxicity
IndetermNo toxicity
No toxicity
IndetermIndeterm No toxicity
IndetermNo toxicity
No toxicity
No toxicity
No toxicity
No toxicity
No toxicity
Indeterm+
IndetermIndetermIndeterm N o toxicity
IndetermIndetermNo toxicity
No toxicity
IndetermNo toxicity
Indeterm+
Indeterm+
No toxicity
N o toxicity
Indeterm +
No toxicity
No toxicity
No toxicity
IndetermIndeterm-
Toxicity
IndetermIndetermNo toxicity
No toxicity
Indeterm+
IndetermNo toxicity
Indeterm+
Indeterm+
No toxicity
Indeterm+
IndetermIndetermIndetermIndeterm+
No toxicity
No toxicity
IndetermNo toxicity
Indeterm-
Toxicity
Toxicity
IndetermIndeterm No toxicity
Indeterm+
Indeterm+
No toxicity
IndetermIndetermNo toxicity
Indeterm+
Indeterm+
No toxicity
No toxicity
No toxicity
IndetermNo toxicity
No toxicity
No toxicity
No toxicity
Toxicity
Probable toxicity
Probable no toxicity
Probable no toxicity
No toxicity
Probable toxicity
Probable no toxicity
No toxicity
Probable no toxicity
Probable no toxicity
No toxicity
Probable toxicity
Probable toxicity
Probable no toxicity
Probable no toxicity
Probable toxicity
Probable no toxicity
No toxicity
Probable no toxicity
Probable no toxicity
Probable no toxicity
* 1ndeterm-t
=
i‘ Toxicity
22 ophthalmologists rated the patient as having HCQ toxicity; probable toxicity = 2 2 ophthalmologists rated the patient as having
=
indeterminate but probable HCQ toxicity; indeterm- = indeterminate but probable no HCQ toxicity.
indeterm+ HCQ toxicity, or at least 1 rated the patient as having indeterm+ HCQ toxicity and at least 1 rated the patient as having HCQ toxicity;
probable no toxicity = 1 ophthalmologist rated the patient as having indeterm- HCQ toxicity; no toxicity = all 5 ophthalmologists agreed that no
HCQ toxicity was present.
was somewhat better than that noted in the only other
published report of HCQ toxicity monitoring in which
compliance was discussed (15). In that study, only 71%
of the patients had 1 or more eye examinations per year.
The incidence of ophthalmologic toxicity in the 289
patients from our initial group who did not have documented eye examinations remains unknown, but could
be reasonably assumed to mirror that of the studied
group.
One potential criticism of this type of study is the
possibility that bias is introduced if patients who experience toxicity terminate HCQ treatment on their own
and therefore are not included in the analysis. However,
in our study, all patients initially identified by a search of
Kaiser Permanente pharmacy records for HCQ prescriptions were included in the final analysis. Moreover, any
patient with potential HCQ toxicity recorded in his or
her chart was included regardless of whether treatment
was stopped.
Previous authors (6,8,9,12,16) have commented
that long-term use or high daily doses of HCQ could
potentiate retinal toxicity. Early research (16) suggested
that a daily dosage of <400 mg is safe, whereas Mackenzie (8) recommended a daily dose of <6.5 mg/kg. We
found 302 of 1,496 patients (20%) who received dosages
greater than Mackenzie’s recommended dosage of 6.5
mgJkg/day and 78 of 1,496 (5.2%) who received more
than the “threshold” dosage of 7.8 mgJkg/day. However,
except for the 3 patients in whom definite or possible
retinal toxicity was identified (patients 1, 2, and 12), this
subset of patients did not experience toxicity.
There have been few large studies of patients
receiving HCQ for >10 years, leading several authors
(7,16-18) to recommend limiting HCQ use to < l o years
or increasing the frequency of screening examinations.
We identified 68 patients (70 of 1,496; 4.7%) who took
HCQ for >10 years, and 8 who took HCQ for >20 years.
Our findings call into question whether high daily dosage or long-term use are true potentiators of HCQ
toxicity. Further studies are needed to answer these
questions. Our study, the largest investigation of HCQ
retinal toxicity to date, confirms the low incidence of
HCQ retinal toxicity found by previous investigators
(9,11,12,16). The rarity of this entity has made its study
problematic and has led to widely different recommended screening protocols. Most of the previous studies probably had sample sizes that were too small and
therefore were too statistically weak to resolve the
differences between cases of toxicity and nontoxicity
because of the very low pretest probability of disease.
Ophthalmologists tend to diagnose any HCQtreated patient who has a suspicious lesion as having
LEVY ET AL
possible HCQ toxicity, leading to discontinuation of
HCQ in many cases. Patients may experience a substantial increase in disease activity when HCQ is discontinued (19). Other patients, unnecessarily worried about
vision loss, may elect not to start taking HCQ, depriving
themselves of a useful antirheumatic agent. Our study
identified only 1 patient with HCQ-associated retinal
toxicity and no patients with substantial vision loss.
Many patients with potential retinopathy were ultimately classified as having no toxicity or indeterminate
toxicity which was probably found to be no toxicity after
secondary analysis. This situation reflects the very low
threshold we set for inclusion in the group with possible
toxicity. Even though these patients were thought to
have a low probability of HCQ toxicity, HCQ treatment
was often discontinued. This result illustrates that HCQ
may be stopped without a compelling reason, denying
patients a valuable therapy.
We estimate that >150,000 patients take HCQ
each year,* which, coupled with present screening recommendations, leads to a major, and in our view unnecessary, expenditure of medical resources for ophthalmic
screening. Utilizing the conservative cost figures cited by
MacLean (20), the US expenditure for ophthalmic
screening for HCQ toxicity is estimated to be -$22
million per year. Several investigators (8-10) have recommended that HCQ-related retinopathy screening be
done yearly, and 2 authors (12,13) advocate the elimination of routine screening altogether. Based on our
review of the literature and the very low incidence of
toxicity found in this study, we conclude that, in patients
with normal renal function, routine ophthalmic screening is not indicated if the daily dosage of HCQ is <6.5
mglkgiday. However, pending further study, we continue
to recommend annual screening in patients who have
long-term use, high daily dosage, or renal insufficiency.
':This conservative estimate was derived by 2 methods. Thc
first utilizes the number of tablets sold in the US in 1995 by Sanofi
Winthrop Pharmaceuticals (100,000,000 [personal communication]),
divided by the average daily pill consumption derived from our study
(642 pillsiyearipatient). This gives an cstimate of 156,000patients. The
second method extrapolates our Kaiser Permanente experience of
-1,500 patients out of a population of 2,250,000 enrollees in the
Southern California Region. When this ratio (0.0007) is multiplied by
the US population in 1995 of 261,000,000 (US Census Bureau), an
estimate of 174,000 patients is derived.
ACKNOWLEDGMENTS
We wish t o thank Wansu Chen, MS, for statistical
analysis, an d Debra Escoe, BA, an d Elaine Berman for chart
abstracting. The Medical Editing Department, Kaiser Foundation Research Institute, provided editorial assistance.
REFERENCES
1. Cambiaggi A Unusual ocular lesions in a case of systemic lupus
erythemdtosus. Arch Ophthalmol 57:451-457, 1957
2. Hobbs HE, Sorsby A, Freedman A Retinopathy following chloroquine therapy. Lancet 2:478-480, 1959
3. Bernstein HN: Chloroquine ocular toxicity. Sum Ophthalmol
12:415-443, 1967
4. Finbloom DS, Silver K, Newsome DA, Gunkel R: Comparison of
hydroxychloroquine and chloroquine use and the development of
retinal toxicity. J Rheumatol 12:692-694, 1985
5. Raines MF, Bhargava SK, Rosen ES: The blood-retinal barrier in
chloroquine retinopathy. Invest Ophthalmol Vis Sci 30: 1726-1731,
1989
6. Bernstein HN: Ocular safety of hydroxychloroquine sulfate
(Plaquenil). South Med J 85:274-279, 1992
7. Tobin DR, Krohel GB, Rynes RI: Hydroxychloroquine: sevenyear experience. Arch Ophthalmol 100:81-83, 1982
8. Mackenzie AH: Dose refinements in long-term therapy of rheumatoid arthritis with antimalarials. Am J Med 75:40-45, 1983
9. Easterbrook M: The ocular safety of hydroxychloroquine. Semin
Arthritis Rheum 23 (suppl 1):62-67, 1993
10. Spalton DJ, Verdon Roe GM, Hughes GRV: Hydroxychloroquine, dosage parameters and retinopathy. Lupus 2:355-358, 1993
11. Wallace DJ: Antimalarial agents and lupus. Rheum Dis Clin North
Am 20:243-263, 1994
12. Morsman CDG, Livesey SJ, Richards IM, Jessop JD, Mills P V
Screening for hydroxychloroquine retinal toxicity: is it necessary?
Eye 4:572-576, 1990
13. Morand EF, McCloud PI, Littlejohn GO: Continuation of long
term treatment with hydroxychloroquine in systemic lupus erythematosus and rheumatoid arthritis. Ann Rheum Dis 51:13181321, 1992
14. Weise EE, Yannuzzi L A Ring maculopathies mimicking chloroquine retinopathy. A m J Opthalmol 78:204-210, 1974
15. Mazzuca SA, Yung R, Brandt KD, Yee RD, Katz BP: Current
practices for monitoring ocular toxicity related to hydroxychloroquine (Plaquenil) therapy. J Rheumatol 21:59-63, 1994
16. Johnson MW, Vine A K Hydroxychloroquine therapy in massive
total doses without retinal toxicity. Am J Ophthalmol 104:139-144,
1987
17. Shearer R, Dubois E: Ocular changes induced by long-term
hydroxychloroquine (Plaquenil) therapy. Am J Ophthalmol 64:
245-252, 1967
18. Rynes RI: Ophthalmologic safety of long term hydroxychloroquine sulfate treatment. A m J Med 75:35-39, 1983
19. The Canadian Hydroxychloroquine Study Group: A randomized
study of the effect of withdrawing hydroxychloroquine sulfate in
systemic lupus erythematosus. N Engl J Med 324:150-154, 1991
20. MacLean CH, Lee PP, Keeler EB: Cost analysis of routine
ophthalmologic screening for hydroxychloroquine retinal toxicity
(abstract). Arthritis Rheum 39 (suppl 9):S72, 1996
Документ
Категория
Без категории
Просмотров
1
Размер файла
520 Кб
Теги
outpatient, incidence, large, patients, retinopathy, practice, multicenter, hydroxychloroquine, 1207
1/--страниц
Пожаловаться на содержимое документа