SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Objective

The presence of systemic and/or immune manifestations in myelodysplasia has been currently reported. The influence of these manifestations on the natural outcome of myelodysplastic syndrome has to be considered. We present a multicenter retrospective study (2002–2009) of patients with myelodysplastic syndrome disclosing systemic and/or immune manifestations.

Methods

Forty-six patients with myelodysplasia presenting with systemic and/or immune manifestations were compared in terms of survival with 189 patients with myelodysplasia lacking these features.

Results

The clinical picture in these cases consisted of fever (13%), arthralgia or arthritis (13%), and cutaneous manifestations (67%). Four cases of systemic vasculitis have been reported in our series, and they have a worse prognosis. Immune anomalies were recorded in 29% of the cases, and the presence of cryoglobulins was also associated with a worse prognosis.

Conclusion

A difference in survival between patients with myelodysplastic syndrome with systemic manifestations and patients lacking these manifestations has been observed in the presence of systemic vasculitis and/or cryoglobulins.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Although systemic and/or immune manifestations are common in myelodysplasia, they rarely characterize the initial clinical presentation. Hematologic malignancies are therefore associated with systemic manifestations and the presence of some types of autoantibodies (1–3). The mechanisms of these manifestations are unknown, but they could be related to an immune dysfunction induced by myelodysplastic syndrome (MDS) or by immunosuppressive therapy.

According to data in current literature, the prognosis of MDS in the presence of these manifestations seems to be unmodified (3). Nevertheless, most retrospective and prospective studies have included a limited number of patients. The new diagnostic criteria for MDS highlight the need for further characterization of MDS with systemic and/or immune manifestations in a larger patient population in terms of prognosis and survival.

The objective of this study was to determine the clinical features and outcome of systemic and immune manifestations in patients with MDS, and to search for the prognostic implications of these manifestations in MDS. Here we report a multicenter retrospective study (2002–2009) in patients with myelodysplasia presenting systemic and/or immune manifestations dominating the clinical presentation.

Significance & Innovations

  • There are no cytogenetic and prognosis differences in myelodysplastic syndrome (MDS) patients with systemic and immune manifestations versus MDS patients without these features.

  • Systemic vasculitis and detection of cryoglobulins in MDS patients are associated with a worse prognosis.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Data from 235 patients with a formal diagnosis of myelodysplasia have been analyzed for an 8-year period (2002–2009). All of the data have been retrospectively analyzed using the French national medical electronic record using diagnostic codes. Exclusion criteria consisted of immunosuppressive therapy before the diagnosis of MDS.

Diagnosis of myelodysplasia.

All of the patients included in the study had a formal diagnosis of myelodysplasia based on bone marrow aspirates, peripheral blood diagnostic criteria, percentage of blasts, biopsy samples, and cytogenetic tests. None of the included patients had immunosuppressive therapy before bone marrow aspirates.

The French, American, and British (FAB) classification has been employed in order to classify MDS as low-, intermediate-, or high-risk MDS, according to the International Prognostic Scoring System (IPSS) (4, 5). The IPSS is based on certain criteria such as the number of cytopenias, cytogenetic profile, and the percentage of blasts in the bone marrow, and allows the classification of MDS patients into 1 of 4 prognostic categories: low risk, intermediate 1 risk, intermediate 2 risk, and high risk.

The World Health Organization (WHO) classification–based prognostic scoring system (WPSS) adds some important criteria such as transfusion dependency and cytogenetic risk group in order to classify MDS patients into 5 risk classes: very low risk, low risk, intermediate risk, high risk, and very high risk (6, 7).

Diagnosis of systemic and autoimmune manifestations.

As systemic disease affects a number of organs and tissues or the entire body, systemic manifestations are defined as clinical manifestations concerning one or several organs or systems.

Since the definition of autoimmunity is the failure of an organism to recognize its own constituent parts as self, which allows an immune response against its own cells and tissues, and since autoimmune diseases are defined as those resulting from an aberrant immune response, immune manifestations are defined as clinical and biologic manifestations that result from such an aberrant immune response. The diagnosis of autoimmune manifestations was based on suggestive clinical symptoms together with relevant biologic, anatomopathologic, and/or imagistic results.

Therefore, when confronted with systemic manifestations, some complementary explorations were performed such as immune and autoimmune parameters, tomodensitometry, biopsy clinics when relevant, and so on.

Laboratory works.

Several biologic parameters have been searched for in this series in all of the participants: blood cell count, hemostasis, standard biochemical parameters, viral serology (hepatitis B virus, hepatitis C virus, cytomegalovirus, Epstein-Barr virus, and human immunodeficiency virus), serum protein electrophoresis, and PNH clone.

Immune parameters such as antinuclear antibodies (ANAs), extractable nuclear antigen antibodies, antineutrophil cytoplasmic antibodies (ANCAs), rheumatoid factor (RF), complement fractions C3 and C4, and total hemolytic complement were analyzed in 77% of the patients, and cryoglobulins were analyzed in 54% of the cases. ANAs and ANCAs were detected by immunofluorescence microscopy and capture enzyme-linked immunosorbent assay. For cryoglobulins, we used a gel diffusion procedure that detects cryoglobulins with greater sensitivity and specificity than the conventional precipitation method (8). RF was detected by agglutination tests detecting IgM-RF, which are the most common methods used in laboratory diagnosis of rheumatoid arthritis, but also with an enzyme-linked immunosorbent assay. A cytogenetic test has been performed in all patients, but only 80% of test results could be obtained.

Other data.

Several other data were retrospectively analyzed as well: treatments, evolution through acute myelogenous leukemia, and survival delay after diagnosis.

Statistical analysis.

Characteristics that were continuous variables were compared using the Mann-Whitney U test. For establishing binary variables, odd ratios (ORs) with corresponding 95% confidence intervals (95% CIs) were calculated. The association between the 2 groups of patients and the IPSS was tested using the chi-square test. The survival difference between the 2 groups was tested using Kaplan-Meier curves and the log-rank test. A P value of less than 0.05 was considered significant. The analysis was performed using the statistical package Statistica, version 6.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

The 235 patients were classified into 2 groups according to the presence or not of autoimmune manifestations. Among them, 46 patients (19.5%) with a mean age of 78 years (range 68–93 years) presented systemic and/or immune manifestations (2002–2009). The remaining 189 MDS patients constitute the control group.

MDS according to the FAB and/or WHO classification were not statistically different in both groups of patients with myelodysplasia, either with or without systemic and immune manifestations (Table 1).

Table 1. Characteristics and classification of MDS patients with and without systemic and autoimmune manifestations*
 MDS with autoimmune manifestationsMDS without autoimmune manifestations
  • *

    MDS = myelodysplasia; WHO = World Health Organization; RA = refractory anemia; RARS = refractory anemia with ringed sideroblasts; RCMD/RS = refractory cytopenia with multilineage dysplasia with or without ringed sideroblasts; RAEB = refractory anemia with blast excess; FAB = French, American, and British; SA = sideroblastic anemia; CML = chronic myeloid leukemia; WBCs = white blood cells.

No. (n = 235)46189
Age, mean (range) years78 (68–93)81 (65–89)
Men, no. (%)25 (54)87 (46)
WHO classification, no. (%)  
 RA/RARS21 (45.6)94 (49.7)
 MDS with del5q313 (6.5)8 (4.23)
 RCMD/RS8 (17)32 (16.9)
 RAEB 110 (21.7)35 (18.5)
 RAEB 23 (6.5)5 (2.6)
FAB classification  
 RA, no. (%)18 (39)102 (54)
 SA, no. (%)6 (13)21 (11)
 RAEB, no. (%)19 (41)51 (27)
 CML, no. (%)2 (4.34)15 (8)
 WBCs, mean (range) × 109/liter2.95 (0.21–22.4)3.1 (0.19–23.8)
 Absolute neutrophil count, mean (range) × 109/liter1.2 (0.02–12.3)1.1 (0.07–14.6)
 Hemoglobin, mean (range) gm/liter8.6 (6.9–11.9)8.9 (6.5–12.4)
 Platelet count, mean (range) × 109/liter53 (4–725)49 (3–681)
 Transfusion dependency, no. (%)21 (45)81 (43)

The patients were classified according to IPSS risk into: low risk in 30.4% of the cases (14 of 46) in the myelodysplasia with autoimmune manifestations group versus 28.6% (54 of 189) in the myelodysplasia control group, intermediate 1 and 2 risk in 28.2% (13 of 46) versus 31.7% (60 of 189), and high risk in 41.3% (19 of 46) versus 39% (75 of 189) of cases.

According to WPSS risk, the patients in our series could be classified as: very low risk in 3 (6.52%) of 46 patients versus 14 (7.4%) of 189 controls, low risk in 11 (23.9%) of 46 patients versus 40 (21%) of 189 controls, intermediate risk in 13 (28.26%) of 46 patients versus 57 (30.15%) of 189 controls, high risk in 15 (32%) of 46 patients versus 68 (36%) of 189 controls, and very high risk in 4 (8.7%) of 46 patients versus 7 (3.7%) of 189 controls.

Noninfectious fever was found in 6 cases (13%) that also presented arthralgia/arthritis (13%). Peripheral neuropathy was present in 5 patients (10%), pulmonary infiltrates in 4 cases (8%), and nephritis in 4 cases (8%). Cutaneous manifestations were present and manifested by purpura in 11 cases (24%), nodules in 10 patients (22%), papules in 5 patients (10%), and necrotic lesions in 5 patients (10%). Ten patients (21%) presented isolated systemic manifestation, 11 (24%) presented 2 manifestations, and 25 (54%) presented more than 2 manifestations. In 4% of the cases, this leukocytoclastic vasculitis occurred mainly in muscular vessels, evoking panarteritis.

Although infrequently, some rare cases of systemic vasculitis have been reported in our series (1 relapsing polychondritis in 2%, 1 case of granulomatosis with polyangiitis [Wegener's] [GPA], 1 case of microscopic polyangiitis, and 1 case of Churg-Strauss syndrome). In all cases, the vasculitis succeeded the MDS diagnosis (mean delay 6 months).

The patient with relapsing polychondritis presented pain, redness, swelling, and tenderness in the ears, nose, throat, and joints (hands, knees, ankles, and wrists); fever; fatigue; and weight loss. Four months after the diagnosis he presented with cutaneous vasculitis. The diagnosis was supported by the clinical presentation and relevant ear biopsy samples. One patient was diagnosed with GPA 5 months after the diagnosis of MDS. The diagnosis was based on pulmonary and otorhinolaryngologic symptoms and radiologic (sinus radiograph, lung and sinus tomodensitometry), immunologic (antimyeloperoxidase ANCA), and morphopathologic findings (lung biopsy and salivary gland biopsy samples revealing granuloma and vasculitis). Corticosteroids in combination with cyclophosphamide (CYC) represented the base of the treatment in both of these cases. In the case of microscopic polyangiitis the patient presented glomerulonephritis, “mononeuritis multiplex,” weight loss, dyspnea, and fever. The following complementary investigations allowed final diagnosis: chest computed tomography (CT) scan, renal biopsy, and electromyography/nerve conduction study. Treatment with steroids (prednisone) in combination with CYC was started and replaced 4 months later by azathioprine.

The diagnosis of Churg-Strauss syndrome was made in 1 patient with a history of asthma who presented fever, weight loss, and sinus inflammation. Cough, dyspnea, and chest pain appeared later on. Hypereosinophilia, CT lung scan, and lung biopsy confirmed the diagnosis. The clinical manifestations were improved with prednisone in combination with CYC.

Immune parameters were searched for in 180 (77%) of 235 patients, and cryoglobulins in 127 (54%) of 235 patients. These parameters were searched for in all 46 patients with systemic/immune manifestations.

Only 4 (8%) of 46 patients had cryoglobulins, 13 (28%) of 46 had ANAs, 2 (4%) of 46 had ANCAs, and 4 (8%) of 46 had RF. Cryoglobulins were found in only 1 patient who presented systemic vasculitis GPA. In the remaining 3 patients, cryoglobulins were isolated in 1 case and associated with ANAs in the other 2 cases.

Patients with cryoglobulinemia presented dyspnea and glomerulonephritis (n = 1) plus purpura (n = 2). One patient also presented arthralgia. Cryoglobulins were of type II (2 cases) and type III (2 cases). RF was positive in 2 patients.

In all patients with cryoglobulinemia, hepatitis C virus as well as hepatitis B virus serologies were negative. Skin biopsy samples revealed in these cases leukocytoclastic vasculitis with dense perivascular infiltrates. In 1 case, thrombus-like deposits were found. Outcome was improved by corticosteroids in 75% of cases.

Cytogenetic findings with good prognosis such as normal, 5q-, 20q-, or -Y presented no statistically significant difference between the 2 groups (OR 0.77, 95% CI 0.25–2.85). No differences were found in age and sex between the 2 groups.

In MDS that succeeded systemic manifestations, immunosuppressive therapy could not be incriminated as the etiologic and/or precipitating factor. In low-risk myelodysplasia patients, the treatments were best supportive care (red blood cells and/or platelet transfusions and/or erythropoietin) in 8 (17%) and 31 (16.4%) patients, respectively, and lenalidomide or 5-azacytidine or decitabine in 6 (13%) versus 23 cases (12%). Low-dose chemotherapy (cytarabine) was used in intermediate-risk patients in 13 patients (28.2%) versus 58 (30.7%). Only 2 patients with intermediate risk in the control group of patients with myelodysplasia had isolated supportive care.

High-risk patients had intensive acute myelogenous leukemia–like chemotherapy (cytarabine and antracyclin) in 10 (21.7%) versus 35 (18.5%), blood cells and/or bone marrow transplantation in 4 (8.7%) versus 15 (7.9%), and isolated supportive care in 5 (11%) versus 25 (13.2%) of the cases. Six and 12 patients, respectively, had bone marrow/blood cell transplantations. A total of 12 (26%) and 41 (21.6%) patients, respectively, had growth factors. Progression to acute myelogenous leukemia occurred in 8 and 32 cases, respectively. Corticosteroids were effective on autoimmune manifestations in more than 90% of cases.

Although there were no differences between the 2 groups of patients in terms of IPSS risk, we could find a higher prevalence of very high risk in myelodysplasia patients with autoimmune manifestations when using the WPSS classification. There were 31 (67%) of 46 deaths in the group of myelodysplasia with autoimmune manifestations versus 130 (69%) of 189 in the control group. Deaths appeared after progression to acute myelogenous leukemia in 8 cases and 32 patients, infections in 15 and 63, respectively, and hemorrhagic complications in 8 versus 35.

No differences in survival were found between the 2 groups of patients (with and without systemic manifestations) (Figures 1 and 2). There were only 2 exceptions: when systemic vasculitis has been diagnosed and/or in the presence of cryoglobulins. Therefore, 2 patients with vasculitis had a fatal outcome at 13 months, 1 at 15 months, and the last at 16 months after MDS diagnosis. The same was true for patients with cryoglobulins: 3 of them had a lethal outcome at 14 months and 1 at 16 months, even though they were classified in intermediate-risk MDS according to both the IPSS and WPSS (Figure 3).

thumbnail image

Figure 1. Survival curves of myelodysplastic syndrome (MDS) patients with and without systemic and autoimmune manifestations (AI) according to the World Health Organization classification–based prognostic scoring system (WPSS) prognosis score.

Download figure to PowerPoint

thumbnail image

Figure 2. Survival curves of myelodysplastic syndrome (MDS) patients with and without systemic and autoimmune manifestations (AI) according to the International Prognostic Scoring System (IPSS) prognosis score.

Download figure to PowerPoint

thumbnail image

Figure 3. Survival curves of myelodysplastic syndrome (MDS) patients with and without systemic and autoimmune manifestations (AI) and with vasculitis and/or cryoglobulinemia according to the World Health Organization classification–based prognostic scoring system (WPSS) prognosis score.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

The association of myelodysplasia and systemic and immune manifestations has been reported for a long time, and was further underlined by many case reports as well as several retrospective and prospective studies (9–23).

The reported incidence of systemic and immune manifestations in MDS patients varies between 10% and 12% (11–13). However, a higher prevalence of 18.5% has been reported in a 4-year prospective study (22). Moreover, in a large Asian study, Okamoto et al observed an incidence of 12% of systemic manifestations associated with MDS and immune biologic abnormalities in 63% of cases (24). We report a high incidence of systemic and autoimmune manifestations in our series of MDS patients (19%), and we further suggest that this incidence might even be underestimated, since we performed a retrospective study with subsequent difficulties in collecting data.

Data in the literature are concordant with regard to clinical systemic manifestations. Therefore, clinical manifestations were rarely part of a typical systemic vasculitis clinical picture such as relapsing polychondritis and/or other different connective tissue diseases (15, 19–21). In most cases, systemic and immune manifestations were represented by cutaneous vasculitis, fever, arthritis, inflammatory bowel disease, pulmonary infiltrates, peripheral polyneuropathy, and nephritis (17, 22).

Immune and autoimmune biologic anomalies have also been reported in MDS, such as the presence of ANAs, ANCAs, RF, and cryoglobulins (3, 21). Systemic manifestations during myelodysplasia have been reported in 46% of cases, and represented a marker of relapse in 29% of the cases.

The systemic and immune manifestations dominated the initial clinical picture of myelodysplasia in 33% of the cases in our series (and preceded myelodysplasia by a mean duration of 6 months). Autoimmunity signs and symptoms rapidly progressed in 11 of 46 cases in the first 6 months of followup, were stable in 22 of 46 cases, and had an insidious progression during several months (more than 6) and/or years in 13 of 46 cases.

Autoimmune phenomena in MDS patients in our retrospective series have responded to steroids, which is in agreement with the literature. A correlation between immunologic abnormalities and prognosis in patients with MDS has been observed by some authors, and data have been contradicted by other reports (22, 24). Moreover, our results revealed a lack of difference in the incidence of clonal alterations between the 2 groups (43% and 39%, respectively). These data are concordant with those from a 4-year prospective study (20), but are in contradiction to other reports that have suggested an abnormal karyotype in MDS associated with systemic and immune abnormalities (13).

These differences could be related to the different populations included in the study by Giannouli et al and in our study (22). Therefore, in the last 2 studies, MDS patients presenting systemic and immune manifestations after immunosuppressive chemotherapy have been excluded. It is now currently admitted that most of the patients with therapy-related MDS have a poorer prognosis and abnormal karyotypes (25).

Even though our report is a retrospective study, it brings some data and raises some questions that add to previous knowledge, mainly because it includes a large number of patients. Although the study by Giannouli et al (22) was a 4-year prospective study, the number of MDS patients included was 70, and only 13 among them had autoimmune manifestations.

When comparing our data with those found by Giannouli and colleagues (22), no differences were found in both studies among the 2 groups of patients (MDS patients with myelodysplasia in the presence of autoimmune disease versus MDS patients without these manifestations) in terms of bone marrow blasts, IPSS, favorable cytogenetic abnormalities, leukemic transformation, and survival. Therefore, both studies report that the prognosis of patients with myelodysplasia in the presence of autoimmune disease is not different versus patients without these manifestations, and is essentially related to risk score (IPSS in their study; IPSS and WPSS in ours).

Our study confirms previous knowledge, but also highlights a new item: the presence of either cryoglobulins or systemic vasculitis such as factor of bad prognosis. Moreover, the parallel use of 2 different diagnosis criteria and prognosis scores strengthens the pertinence of some of our findings in terms of comparative survival.

In our study, the survival curves show that the prognosis is not different between MDS patients with and without autoimmune manifestations, no matter the category risk score. Therefore, according to the IPSS, no differences were found between MDS patients with and without autoimmune manifestations for each prognosis category. The same results were observed in both groups of patients when referring to WPSS prognostic scores, except for very high-risk categories. Meanwhile, these data are suggestive of a worse prognosis in the presence of autoimmune manifestations in this risk category, but could not be formal because of the limited number of patients in both groups.

We did not find any difference in the prognosis between the 2 groups of MDS patients, either for systemic manifestations or for immune abnormalities, except for the detection of cryoglobulins. The use of IPSS and WPSS scores in our study allowed for a better knowledge of prognosis in MDS with systemic manifestations. Only a few studies evaluated the IPSS prognostic category and the clinical significance of the association between systemic and immune manifestations with MDS (20).

To the best of our knowledge, our report is one of the studies that included a large number of patients. Even though we could not find differences in terms of survival between the 2 groups of patients (with and without systemic manifestations), there were 2 exceptions: when systemic vasculitis has been diagnosed and/or in the presence of cryoglobulins.

Therefore, in the 4 cases reported in our study, the presence of systemic vasculitis during MDS was a factor of adverse prognosis. All of these 4 patients presented an evolution of MDS through acute myelogenous leukemia in a mean interval of 6 months. These results are consistent with the data in the literature that report a worse prognosis in the case of MDS with acute systemic vasculitic syndrome, which is associated with rapid clinical deterioration and high mortality (17). The physiopathologic explanation of this worse prognosis is probably based on the presence of several immune abnormalities in MDS. However, the presence of cryoglobulins has never been reported as a worse prognostic factor in MDS. Therefore, alterations of some structural and functional components of the immune system such as decreased CD4 cell population and impaired immunoglobulin production, diminished activity of natural killer cells, and alterations in the mitogenic answer could explain not only the presence of some clinical and biologic systemic manifestations in MDS, but also some features of MDS such as the ineffective hematopoiesis (3, 24–26).

Interferon regulatory factor 1 (IRF-1) is a transcriptional factor that is involved in the development of the immune system and interferon signaling. Myelodysplasia patients without IRF-1 expression have a decreased incidence of immune manifestations, suggesting that the absence of IRF-1 transcriptional factors confers protection against the development of autoimmunity in MDS (27).

However, our retrospective study failed to record all immune and cytogenetic data in our population. As already mentioned, ANAs, ANCAs, RF, and total hemolytic complement tests have been performed only in 77% of cases, and cryoglobulins in 54%. Due to this lack of homogeneity, it could be suggested that cryoglobulins might have an underestimated influence on the natural outcome in MDS. Further prospective studies are thus needed in order to evaluate this issue. Moreover, the lack of some other clinical or biologic information in the medical records supports the need for prospective studies in the field of MDS associated with systemic and immune manifestations.

Our results failed to show a difference in MDS patients with systemic and immune manifestations versus MDS patients without these features, especially in terms of cytogenetic characteristics and prognosis. Although systemic vasculitis in patients with MDS is a rare phenomenon, when present it is associated with a worse prognosis and rapid clinical deterioration. Similarly, the detection of cryoglobulin, although rare, is a worse prognostic factor associated with low survival rates. The pathogenesis of the vasculitis in MDS remains speculative, in spite of some reports that suggest an immunologic mechanism. The association of MDS with systemic and immune manifestations does not seem to be infrequent, therefore suggesting a potential common primary immune deregulation. Larger prospective studies should be further performed in order to progress in the understanding of underlying mechanisms as well as their clinical and biologic significance.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Belizna had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. De Hollanda, Beucher, Henrion, Ghali, Lavigne, Lévesque, Hamidou, Subra, Ifrah, Belizna.

Acquisition of data. De Hollanda, Beucher, Henrion, Ghali, Subra, Ifrah, Belizna.

Analysis and interpretation of data. De Hollanda, Beucher, Henrion, Ghali, Subra, Ifrah, Belizna.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES