вход по аккаунту


Shrinking lung syndrome in systemic sclerosis.

код для вставкиСкачать
Vol. 48, No. 10, October 2003, pp 2996–3000
© 2003 American College of Rheumatology
olism were compared by paired t-test and Wilcoxon’s signed
rank test when appropriate.
Sixty-eight patients were included in the study: 54
women and 14 men, with a mean ⫾ SD age of 55, ⫾ 13.8 years
and a median disease duration of 10 years (range ⬍1 year–59
years). Forty-eight patients (71%) were IgM rheumatoid factor
positive (⬎30 IU/ml). Sixty-one patients (90%) took methotrexate (median 15 mg/week), and 5 took other diseasemodifying antirheumatic drugs during the study period. At
baseline, 31 patients (46%) were taking corticosteroids (prednisone [median 10 mg/day]). In 19 of these patients the
prednisone dosage remained stable during the study, whereas
in 12 it was decreased. The mean ⫾ SD DAS28, 5.9 ⫾ 1.4 at
baseline, was decreased significantly after infliximab treatment
(4.6 ⫾ 1.4 [P ⬍ 0.001 versus baseline] at week 2 and 4.1 ⫾ 1.5
[P ⬍ 0.001 versus baseline] at week 6). Fifty-one patients
(75%) were classified as responders and 17 as nonresponders.
OC and PINP levels were significantly increased after
2 weeks and 6 weeks compared with baseline (P ⬍ 0.001),
while BAP levels did not change significantly (Table 1). When
bone formation was analyzed only in the patients who took
prednisone at a stable dosage or in those who did not take
prednisone during the study, we also found a statistically
significant increase in markers of bone formation. ICTP and
␤-CTX levels showed decreases, but only the decrease in the
ICTP level at 6 weeks reached statistical significance (P ⬍
0.001) (Table 1). In the responder group, both markers of bone
resorption (ICTP and ␤-CTX) were decreased significantly at
6 weeks compared with baseline (P ⬍ 0.05). The change in
levels of bone formation markers in the responder group was
similar to that observed in the total group. In the nonresponder
group, none of the markers of bone metabolism showed
significant change. Changes in ␤-CTX levels were slightly
correlated with changes in the DAS28 (r ⫽ 0.348, P ⫽ 0.05),
while changes in the other markers of bone metabolism were
not significantly correlated with changes in disease activity.
Our data suggest that treatment with infliximab has a
favorable effect on bone metabolism in patients with RA.
Several studies have investigated levels of markers of bone
metabolism in RA patients. Active RA was associated with
increased levels of bone resorption in these studies (3,4). Our
findings are in accordance with these data, since the decrease
in disease activity of RA paralleled the decrease in markers of
bone resorption. Reports on markers of bone formation are
DOI 10.1002/art.11292
Early changes in bone metabolism in rheumatoid
arthritis patients treated with infliximab
To the Editor:
Rheumatoid arthritis (RA) is characterized by localized destruction of synovium, cartilage, and bone. Loss of bone
in RA is not only localized in joints but is also generalized, and
the latter (osteoporosis) is recognized as an extraarticular
manifestation of the disease (1). Bone loss in RA is related to
disease activity, immobility and corticosteroid use (2). Bone
remodeling is a continuous process of bone resorption and
bone formation. In RA, bone loss seems to be related to
elevated bone resorption, while data on bone formation are
conflicting (3–5). Infliximab is effective in the treatment of
RA: it causes marked reduction in disease activity and slows
the radiologic progression of localized bone destruction (6).
Therefore, we undertook a study to determine its effect on
markers of bone metabolism in patients with RA.
We studied consecutive patients with active RA, defined as a 28-joint Disease Activity Score (DAS28) (7) of ⱖ3.2
who were treated with infliximab. Infliximab was administered
by intravenous infusion at a dosage of 3 mg/kg at baseline and
2 weeks and 6 weeks thereafter. At each visit, data on disease
activity (erythrocyte sedimentation rate, C-reactive protein
level, disease activity rated on a visual analog scale, number of
swollen and tender joints) and corticosteroid use were recorded. A patient was classified as a responder when the
DAS28 was decreased by at least 1.2 points at 6-week followup.
Blood samples were collected on the morning of each infusion,
and sera were stored immediately at ⫺70°C until assessment.
Bone formation was measured by determining serum levels of
osteocalcin (OC), N-propeptide of type I procollagen (PINP),
and bone-specific alkaline phosphatase (BAP). Bone resorption was measured by determining serum levels of
␤-isomerized C-telopeptide of type I collagen (␤-CTX) and
C-propeptide of type I collagen (ICTP). All markers of bone
metabolism were measured using commercial assays according
to the instructions of the manufacturers (OC and ␤-CTX,
measured with an Elecsys 2010 [Roche Diagnostics, Mannheim, Germany], BAP with an enzyme-linked immunosorbent
assay [ELISA; Quidel, San Diego, CA], PINP and ICTP with a
radioimmunoassay and an ELISA respectively [Orion Diagnostica, Helsinki, Finland]). Data on markers of bone metab-
Table 1. Markers of bone metabolism in 68 rheumatoid arthritis patients at baseline and after 2 weeks
and 6 weeks of infliximab treatment*
Formation, mean ⫾ SD
OC, ng/ml
PINP, ␮g/ml
BAP, units/liter
Resorption, median
(interquartile range)
ICTP, ␮g/ml
␤-CTX, ng/ml
Week 2
Week 6
21.2 ⫾ 11.4
43.9 ⫾ 21.3
22.3 ⫾ 3.7
23.0 ⫾ 11.7†
50.6 ⫾ 23.7†
23.2 ⫾ 4.8
23.9 ⫾ 11.4†
50.1 ⫾ 21.1†
22.7 ⫾ 5.7
8.9 (7.1–12.3)
0.32 (0.17–0.44)
8.6 (6.8–11.6)
0.29 (0.18–0.44)
7.8 (6.2–10.2)†
0.29 (0.17–0.46)
* OC ⫽ osteocalcin; PINP ⫽ N-propeptide of type I procollagen; BAP ⫽ bone-specific alkaline
phosphatase; ICTP ⫽ C-propeptide of type I collagen; ␤-CTX ⫽ ␤-isomerized C-telopeptide of type I
† P ⬍ 0.001 versus baseline.
conflicting: both increased and decreased levels of bone formation markers in patients with RA have been reported (4,5).
The observation that markers of bone formation, with the
exception of BAP, increased during infliximab treatment was
somewhat unexpected. Theoretically, the increase in bone
formation could be the result of a decrease in the prednisone
dosage in some patients (8). However, in the present study this
is unlikely: when we analyzed the patients who were not taking
steroids and those who were taking steroids at a stable dosage
during the study period, there were still significant increases in
OC and PINP levels. Another possible explanation is that bone
formation is depressed in active RA, as a result of disease
activity. Thus, the increase in markers of bone formation might
reflect the inhibition of suppressed bone formation in active
RA. Whether long-term treatment with infliximab has a positive effect on bone metabolism, and subsequently on bone
mineral density and fracture rate, will require further investigation.
M. Vis, MD
VU University Medical Center
and Slotervaart Hospital
GJ. Wolbink, MD, PhD
Jan van Breemen Institute
and Slotervaart Hospital
M. C. Lodder, MD
P. J. Kostense, PhD
VU University Medical Center
R. J. van de Stadt, PhD
M. H. M. T. de Koning
Jan van Breemen Institute
B. A. C. Dijkmans, MD, PhD
W. F. Lems, MD, PhD
VU University Medical Center
Jan van Breemen Institute
and Slotervaart Hospital
Amsterdam, The Netherlands
1. Haugeberg G, Uhlig T, Falch JA, Halse JI, Kvien TK. Bone mineral
density and frequency of osteoporosis in female patients with
rheumatoid arthritis: results from 394 patients in the Oslo County
Rheumatoid Arthritis register. Arthritis Rheum 2000;43:522–30.
2. Eggelmeijer F, Papapoulos SE, Westedt ML, van Paassen HC,
Dijkmans BA, Breedveld FC. Bone metabolism in rheumatoid
arthritis: relation to disease activity. Br J Rheumatol 1993;32:387–91.
3. St Clair EW, Moak SA, Wilkinson WE, Sanders L, Lang T,
Greenwald RA. A cross sectional analysis of 5 different markers of
collagen degradation in rheumatoid arthritis. J Rheumatol 1998;25:
4. Cortet B, Flipo RM, Pigny P, Duquesnoy B, Boersma A, Marchandise X, et al. Is bone turnover a determinant of bone mass in
rheumatoid arthritis? J Rheumatol 1998;25:2339–44.
5. Hall GM, Spector TD, Delmas PD. Markers of bone metabolism in
postmenopausal women with rheumatoid arthritis: effects of corticosteroids and hormone replacement therapy. Arthritis Rheum 1995;
6. Lipsky PE, van der Heijde DM, St. Clair EW, Furst DE, Breedveld
EC, Kalden JR, et al, for the Anti–Tumor Necrosis Factor Trial in
Rheumatoid Arthritis with Concomitant Therapy Study Group.
Infliximab and methotrexate in the treatment of rheumatoid arthritis. N Engl J Med 2000;343:1594–602.
7. Prevoo MLL, van’t Hol MA, Kuper HH, van Leeuwen MA, van de
Putte LBA, van Riel PLCM. Modified Disease Activity Scores that
include twenty-eight–joint counts: development and validation in a
prospective longitudinal study of patients with rheumatoid arthritis.
Arthritis Rheum 1995;38:44–8.
8. Lems WF, Gerrits MI, Jacobs JWG. van Vugt RM, van Rijn HJ,
Bijlsma JW. Changes in (markers of) bone metabolism during high
dose corticosteroid pulse treatment in patients with rheumatoid
arthritis. Ann Rheum Dis 1996;55:288–94.
DOI 10.1002/art.11291
Correlation between clinical response to treatment
and radiographic progression: comment on the article
by Strand and Sharp
To the Editor:
I read with great interest the recent article by Strand
and Sharp (1), which is a comprehensive review of radiographic data from randomized, controlled trials of leflunomide
and biologic disease-modifying antirheumatic drugs
(DMARDs). I wish to comment only on the poor correlation
of radiographic progression and clinical response, as described
by the authors.
Strand and Sharp apparently comment on the American College of Rheumatology 20% response as well as the
change in the C-reactive protein (CRP) level and the erythrocyte sedimentation rate. Many studies have shown a correlation of the level of acute-phase reactants with radiographic
progression, but this can be seen only after a time lag of 6–12
months (2,3). Therefore, it is not the change or the rate of
improvement that is important but rather the level of disease
activity that is reached. For example, a decrease of the CRP
level from 4 mg/dl to 3 mg/dl represents a 25% improvement.
However, the level of 3 mg/dl still indicates very active disease,
meaning that a continuing high rate of radiographic progression can be expected. Moreover, taking into account the fact
that radiographic progression correlates very well with CRP
levels observed 6–12 months prior, we cannot expect a significant decrease in radiographic progression during the first 6
months of a trial, even if the CRP level decreases from 4 mg/dl
to 1 mg/dl within a year. This is also the reason for the
observation that inhibition of progression was much more
pronounced during the second 6 or 12 months of treatment.
Another factor that may lead to the mistaken assumption of a
reduction of the progression rate, also mentioned by the
authors, may be the decreasing number of patients caused by
dropouts of nonresponders; this is a serious statistical problem
that is difficult to solve.
There is still another possible reason for the lack of
correlation between clinical response and radiographic progression: the mean progression rate of 0.5–2.5 Sharp units is
very small and represents a rate of only ⬃0.125–0.6 on a
normalized 0–100 scale. Under these circumstances, only very
small differences in progression inhibition can be expected.
Inclusion of patients for whom the potential for radiographic
progression is high would improve the chance to be able to
demonstrate differences between treatments as well as between responders and nonresponders.
The article stresses the good agreement between the
so-called estimated yearly progression rates and the progres-
sion rates in the placebo group. This was not always the case.
For example, in the US301 trial, the estimated rate of progression at baseline was 3.7, while the observed yearly progression
rate in the placebo group was 2.2. Related to the estimated
progression rate, not only treatment with biologic agents and
leflunomide but also treatment with methotrexate (MTX)
resulted in a profound inhibition of progression (74% in
US301 and 76% in MN302). In the Early Rheumatoid Arthritis
(ERA) trial (4), radiographic progression decreased by ⬃85%
during the first year of MTX treatment when compared with
the predicted progression. This would have been considered
impossible a few years ago. Other trials of conventional
DMARD treatment (i.e., MTX) have shown that the progression rate during the first 6 or 12 months approximates the
estimated rate of progression, while a distinct inhibition of
progression could be observed only during the second 6 or 12
months of treatment (5,6). This may indicate that the predicted
progression rate may have been overestimated in some of the
recent randomized trials (i.e., the ERA trial). The rate of
radiographic progression observed in the patients before their
inclusion in a trial (if previous radiographs are available)
would be more reliable.
Rolf Rau, MD, PhD
Ratingen, Germany
1. Strand V, Sharp JT. Radiographic data from recent randomized
controlled trials in rheumatoid arthritis: what have we learned?
Arthritis Rheum 2003;48:21–34.
2. Matsuda Y, Yamanaka H, Higami K, Kashiwazaki S. Time lag
between active joint inflammation and radiological progression in
patients with early rheumatoid arthritis. J Rheumatol 1998;25:427–32.
3. Van Leeuwen MA, van Rijswijk MH, van der Heijde DM, Te
Meerman GJ, van Riel PL, Houtman PM, et al. The acute-phase
response in relation to radiographic progression in early rheumatoid arthritis: a prospective study during the first three years of the
disease. Br J Rheumatol 1993;32 Suppl 3:9–13.
4. Bathon JM, Martin RW, Fleischmann RM, Tesser JR, Schiff MH,
Keystone EC, et al. A comparison of etanercept and methotrexate
in patients with early rheumatoid arthritis. N Engl J Med 2000;343:
5. Rau R, Herborn G, Menninger H, Sangha O. Progression in early
erosive rheumatoid arthritis: 12 month results from a randomized
controlled trial comparing methotrexate and gold sodium thiomalate. Br J Rheumatol 1998;37:1220–6.
6. Rau R, Wassenberg S, Zeidler H. Low dose prednisolone therapy
(LDPT) retards radiographically detectable destruction in early
rheumatoid arthritis: preliminary results of a multicenter, randomized, parallel, double-blind study. Z Rheumatol 2000;59 Suppl
DOI 10.1002/art.11392
To the Editor:
Rau has commented on 2 aspects of radiographic
analysis that were discussed in our review article. First, he
mentioned several possible explanations for the poor correlation between radiographic progression and many clinical measures of disease activity observed in several of the randomized
controlled trials (RCTs), focusing on the erythrocyte sedimentation rate and the CRP level. We agree that he has identified
possible explanations. Perhaps we did not emphasize enough
the notion of how difficult it is to establish a relationship in
short-term trials (12 months for most) in view of the limited
change in radiographic scores. As stated in the article, “. . . the
range of change [in radiographic progression] was small compared with the error of ascertainment.”
With regard to the comments on the estimated yearly
rates of progression, we reiterate that estimated progression
rates are, of course, only estimates. Such estimations are
limited by errors in dates of disease onset and are less valid in
patients with short disease duration. Rau’s suggestion that
“. . . the predicted progression rate may have been overestimated in some of the recent . . . trials” is theoretically possible,
and neither overestimation nor underestimation can be ruled
out precisely because these are estimates. He specifically
quotes the discrepancy observed in the placebo group in the
US301 trial but overlooks the fact that in this intent-to-treat
analysis, 60% of the patients who began treatment with
placebo received either leflunomide or MTX for an average of
6.5 months when they demonstrated no response at or after 4
months of treatment, which may account for some or all of the
observed discrepancy. Similarly, in the adalimumab RCT, the
estimated yearly progression rate well exceeded the rate of
progression in the placebo group, in part because patients were
required to have received only 3 months of MTX treatment at
the time of enrollment, and DMARD rescue was offered at 4
months as well (Keystone E, Kavanaugh AF, Sharp J, Hua Y,
Teoh L, Fischkoff S, et al. Adalimumab [D2E7], a fully human
anti-TNF-␣ monoclonal antibody, inhibits the progression of
structural joint damage in patients with active RA despite
concomitant methotrexate therapy. Arthritis Rheum 2002;46
Suppl 9:S205). In the recently reported anakinra RCT, although patients were required to have received 6 months of
MTX therapy, again, rescue treatment was offered at 4
months, and the estimated yearly progression rate well exceeded the rate in the placebo group (Shergy WJ, Cohen S,
Greenwald M, Codding C, Nash P, Haraoui B, et al. Anakinra
[Kineret威] inhibits the progression of radiographically measured joint destruction in rheumatoid arthritis [abstract]. Arthritis Rheum 2002;46:3420).
The generally good agreement between the estimated
and the observed rate of progression of radiographic abnormalities was unexpected, and, frankly, it would not surprise us
if this “estimation” of yearly progression is widely used in
future trials to find that it does not always show a close
relationship. Nevertheless, it is one more baseline characteristic to test comparability of patients in the treatment groups
assigned by randomization. Considering the importance of
early radiographic abnormalities in predicting the long-term
course of rheumatoid arthritis, we think that further evaluation
of this estimate in future trials as a measure to assure
appropriate randomization is desirable. Finally, use of this
estimate can help in defining differences between protocol
populations and facilitating comparisons across therapies.
Vibeke Strand, MD
Stanford University
Palo Alto, CA
John Sharp, MD
University of Washington
Seattle, WA
DOI 10.1002/art.11393
Shrinking lung syndrome in systemic sclerosis
To the Editor:
Shrinking lung syndrome (SLS) is a rare pulmonary
disease characterized by dyspnea, diaphragmatic elevation,
and a restrictive pattern on pulmonary function test results,
without evidence of significant pleuroparenchymal disease.
SLS is a rare complication in patients with systemic lupus
erythematosus (SLE) (1–3). Two reports in Arthritis & Rheumatism have described the occurrence of SLS in other autoimmune diseases, in particular 1 case in a patient with Sjögren’s syndrome (4), 1 case in a patient with rheumatoid
arthritis, and 1 case in a patient with undifferentiated connective tissue disease (5).
We now report the occurrence of SLS in a patient with
systemic sclerosis (SSc; scleroderma). This may be of particular
interest because pulmonary involvement is a highly prevalent
feature of SSc and because, to the best of our knowledge, this
is the first report to describe SLS in SSc. The patient, a
48-year-old woman, had a history of Raynaud’s phenomenon
since 1994 and fingertip ulcerations since 1996. In October
2000, she reported exertional dyspnea; results of chest radiography were normal. Moderate pulmonary hypertension (PH)
was suggested by echocardiography and confirmed by cardiac
catheterization (mean pulmonary arterial pressure 37 mm Hg).
She also reported esophageal dysphagia. The autoantibody
profile showed a positive assay for antinuclear antibody (high
titer, anticentromere) and a negative assay for anti–extractable
nuclear antigens (including anti-SSA/SSB) and anti–doublestranded DNA. Nailfold capillaroscopy showed a scleroderma
pattern. A diagnosis of limited SSc was made (6).
In July 2001, the patient developed acute respiratory
failure requiring hospitalization and mechanical ventilation
therapy. Chest radiography showed small lung fields and a
raised right hemidiaphragm (Figure 1). A new high-resolution
computed tomography scan showed no sign of parenchymal
lung disease. Echocardiography confirmed moderate PH; spiral computed tomography revealed no signs of chronic pulmonary embolism, and bronchoalveolar lavage findings were
We performed classic measurements of respiratory
mechanics in a patient who was paralyzed and was receiving
controlled mechanical ventilation with constant inspiratory
flow by means of end-inspiratory and end-expiratory occlusion
maneuvers. Total respiratory system static compliance appeared markedly decreased (23 ml/cm H2O; 27% of normal
value), with a high degree of lung stiffness (lung static compliance ⫽ 38 ml/cm H2O; 19% of normal value), and moderate
chest wall stiffness (chest wall static compliance ⫽ 70 ml/cm
H2O; 46% of normal value). End respiratory resistance was
normal (8 cm H2O/liter/second). Maximal inspiratory pressure
was very low (⫺20 cm H2O), suggesting marked respiratory
muscle weakness.
Laboratory tests revealed normal values of serum
creatinine phosphokinase, aldolase, urea, creatinine, C3, and
C4. There was no proximal muscular weakness, and a muscle
biopsy (right deltoid) did not show any feature of myositis.
Phrenic nerve impairment was also excluded. SLS was then
suspected, and high-dose corticosteroid therapy (prednisone 1
mg/kg/day) was started. After a week we could observe a
Figure 1. Chest radiograph showing elevation of the right hemidiaphragm.
progressive improvement of clinical parameters; 2 weeks later
it was possible to discontinue support with mechanical ventilation, and O2 therapy could be discontinued 1 week later.
After 9 months the patient is still experiencing spontaneous
ventilation and is receiving low doses of prednisone (5 mg/day)
as maintenance therapy. There was no obvious clinical or
serologic evidence of SLE and/or other connective tissue
disease throughout the course of the disease. The diagnosis of
SLS is based essentially on clinical findings, and no pathologic
or laboratory changes are specific for this condition.
This case of SLS, which was suspected on the basis of
right hemidiaphragmatic elevation and restrictive lung disease,
was particularly severe, requiring mechanical ventilation. The
severity was probably related to the concomitant PH. The
treatment strategy for SLS is mainly based on steroids, given as
daily doses of prednisone ranging from 30 mg to 60 mg, with
clinical improvement and normalization of functional tests in
most patients, as we observed in our patient.
The pathogenesis of SLS is still obscure (7,8). The
evaluation of respiratory mechanics during mechanical ventilation, which was performed for the first time in a patient with
SLS, showed sharply decreased pulmonary compliance, with a
mildly decreased thoracic compliance. Maximal inspiratory
pressure was also decreased, suggesting muscular weakness,
probably subsequent to pleural and/or pulmonary stiffness.
Our observations suggest that SLS can occur even in patients
with SSc and should be considered when other causes of
scleroderma lung disease have been excluded.
Carlo Alberto Scirè, MD
Roberto Caporali, MD
Marinella Zanierato, MD
Francesco Mojoli, MD
Antonio Braschi, MD
Carlomaurizio Montecucco, MD
University of Pavia
Pavia, Italy
1. Hoffbrand BI, Beck ER. Unexplained dyspnoea and shrinking lung
in systemic lupus erythematosus. Br Med J 1965;1:1273–7.
2. Karim MY, Miranda LC, Tench CM, Gordon PA, D’Cruz DP,
Khamashta MA, et al. Presentation and prognosis of the shrinking
lung syndrome in systemic lupus erythematosus. Semin Arthritis
Rheum 2002;31:289–98.
3. Warrington KJ, Moder KG, Brutinel WM. The shrinking lungs
syndrome in systemic lupus erythematosus. Mayo Clin Proc 2000;
4. Tavoni A, Vitali C, Cirigliano G, Frigelli S, Stampacchia G,
Bombardieri S. Shrinking lung in primary Sjögren’s syndrome.
Arthritis Rheum 1999;10:2249–50.
5. Ahmed S, Herrick A, O’Driscoll BR. Shrinking lung syndrome in
patients without systemic lupus erythematosus: comment on the
concise communication by Tavoni et al. Arthritis Rheum 2001;44:
6. LeRoy EC, Medsger TA. Criteria for the classification of early
systemic sclerosis. J Rheumatol 2001;28:1573–6.
7. Hardy K, Herry I, Attali V, Cadranel J, Simulowski T. Bilateral
phrenic paralysis in a patient with systemic lupus erythematosus.
Chest 2001;119:1274–7.
8. Wilcox PG, Stein HB, Clarke SD, Pare PD, Pardy RL. Phrenic
nerve function in patients with diaphragmatic weakness and systemic lupus erythematosus. Chest 1988;93:352–8.
DOI 10.1002/art.11288
Clinical Image: Pseudoleukocytosis in cryoglobulinemia
The patient, a 75-year-old man with type I cryoglobulinemia associated with multiple myeloma, presented with a leukocyte count
of 30,000/mm3. Previous counts had ranged between 7,000/mm3 and 10,000/mm3. A peripheral smear revealed numerous amorphous
cryoglobulin aggregates (arrows). These precipitates had been erroneously counted as leukocytes with the Coulter counter, resulting
in pseudoleukocytosis: the manual white blood cell count was only 10,000/mm3. Pseudoleukocytosis has classically been reported in
patients with cryoglobulinemia whose serum was allowed to cool below 37°C. As expected, these cryoglobulin deposits disappeared
on the peripheral smear when the patient’s serum was rewarmed. Despite the use of newer Coulter counters, this benign entity
continues to be a clinically relevant consideration in patients with cryoglobulinemia. In cryoglobulinemia patients who have apparent
leukocytosis, a peripheral smear should be reviewed to exclude this interesting phenomenon.
Jefferson R. Roberts, MD
David R. Finger, MD
Tripler Army Medical Center
Honolulu, HI
Без категории
Размер файла
143 Кб
syndrome, shrinking, lung, systemic, sclerosis
Пожаловаться на содержимое документа