Seventy-two patients underwent splenectomy for idiopathic thrombocytopenic purpura between 1979 and 1990. Mean age at splenectomy was 36.4 years (range 11-73). Indications for splenectomy were corticodependence in 21 cases and resistance to steroids in 44 cases. Thirty-five patients had platelet kinetic studies by ⁵¹Cr alloplatelets; 22 of them had splenic sequestration. Hematologic results were evaluated on discharge, at 3 months and in the long term (median follow-up 5.4 years). We had no mortality, morbidity was seen in 7% of the cases. None of the patients suffered from secondary infectious complications. 89% had good results on discharge ( > 120 × 10⁹/1), 72.6% at 3 months and 90% on long-term follow-up. Factors associated with good response to splenectomy included a high postoperative platelet count (more than 120 × 10⁹/1 on discharge), younger age at the time of surgery, preoperative corticodependence and predominantly splenic sequestration.

Simultaneous analysis of DNA and cellular proteins provides information on cell proliferation and metabolism. Cellular protein content coupled with nuclear geometric parameters can be used to evaluate cellular maturation and differentiation. In this study, leucoblasts from 50 cases of adult acute myeloid leukemia were analyzed by flow cytometry, and semiautomatic morphometry was performed on bone marrow smears. Ethanolfixed bone marrow blast cells were stained for DNA with propidium iodide (PI) and for proteins with fluorescein isothiocyanate (FITC). On the resulting FITC versus PI histograms we defined the cells with low protein content which are associated with a nonproliferating subpopulation (LPC fraction). Low protein content fraction and S-phase are correlated (p < 0.01). The LPC fraction values are more dispersed than S-phase values and thus should indicate more clearly eventual differences between cellular populations. This hypothesis has been tested with the prognostic significance of cell-cycle variables: The LPC fraction was significantly higher in the complete remission group than in the other (p < 0.01), while S-phase did not show any difference.The peak value of the protein content histograms is significantly lower in the granulocytic leukemias (Ml, M2, M3) than in the leukemias with a monoblastic component (M4, M5). Furthermore, we showed that the differentiation and the maturation of the myeloid blast cells modify the nuclear size. The combination of these two parameters provides useful information for cytological classification.Key terms: Acute leukemia, cell cycle, analytical cytometry, PI + FITC staining Acute myeloid leukemia (AML) displays a broad spectrum of morphological and functional cellular abnormalities as well as a large diversity in clinical evolution and in therapeutic response. In contrast to acute lymphoid leukemia (ALL), for which immunological markers allow a precise evaluation of the functional differentiation, AML is usually dissected and analysed by means of morphological criteria (7,221, cytokinetic characterization (1,3), or cytogenetics (6).Cytology usually indicates that the differentiation and the degree of maturation of blastic cells can be assessed by cell size and protein content. MATERIALS AND METHODSPatients Fifty patients aged 18 to 75 years with AML were studied before chemotherapy. None of these patients had any record of dysmyelopoiesis and none had received any chemotherapy prior to this study.Acute myeloid leukemias were classified according to differentiation and maturation criteria using the FrenchAmerican-British (FAB) classification (7): M1: 11 cases, M1-M2: 2 cases, M2: 10 cases, M3: 7 cases, M4: 13 cases, Address reprint requests to Dr.

Five patients presenting extremely exaggerated chromatin clumping in leucocytes, associated with a loss of segmentation, are described. Peripheral cytopenia, a high percentage of circulating immature granulocytes (chiefly myelocytes) with variable leucocytosis, marrow hypercellularity with granulocytic hyperplasia and moderate dysplastic changes in erythroblastic and megakaryocytic lines, constitute, together with the granulocytic nuclear anomaly, the characteristic features of the studied cases. Cytogenetic analysis showed a clonal 12p anomaly in one of the three cases available for study. Survival was poor with a median of 15 months, bleeding and infections being responsible for the majority of deaths. All these findings point to abnormal chromatin clumping in leucocytes as the marker of a true entity with both myelodysplastic and myeloproliferative features. We propose that it take its place among myelodysplastic syndromes (MDS) beside chronic myelomonocytic leukaemia (CMML), with which it shares many similarities in clinical and biological behaviour.

In order to identify prognostic factors in angioimmunoblastic lymphadenopathy (AIL), 30 directly diagnosed patients were prospectively followed for more than 42 months. Age and sex distribution, clinical and laboratory findings and evolution were not different from previously reported series. Median duration of survival was 24 months. Parameters associated with a longer survival in our series were localized adenopathies (P = 0.01) and the achievement of a remission (P < 0.0001). Features associated with a shorter survival included drug exposure in relation to the onset of the disease (P = 0.02), rash (P < O.OOOl), lymph node eosinophilia (P = 0.03) and elevated serum lactic dehydrogenase (P = 0.03). Drug exposure and rash were, however, significantly dependent (P = 0.02). In addition, lymphopenia, the presence of circulating immune complexes, and the absence of polyclonal hypergammaglobulinemia may indicate a poor prognosis, although the significance level is not achieved in this short series. None of the parameters tested was significantly related to the lymphomatous transformation of AIL, which occurred in four cases. It is concluded that multicentric prospective studies of AIL are necessary in order to better define this disorder, to find prognostic factors, and to optimize therapy. NGIOIMMUNOBLASTIC LYMPHADENOPATHY (AIL) isA a lymphoid disorder first described in 1974 by Frizzera et al. ' Despite its apparent histologic benignity,',* AIL usually behaves as a rather aggressive disease with frequent organ in~olvement,l-~ and median survival is only 30 months in the largest series published to date.5 Transformation of AIL into a true malignant lymphoma is observed in up to 18% of the case^.^,^ The evolution of the disease remains, however, rather unpredictable and up to 20% of the patients may undergo a spontaneous complete remissi~n.',~+~ The treatment of AIL is still controversial and various therapies have been recommended including initial abstention,'S2 predni~one,',~,~-~ and ~hemotherapy.~.~,'~ It would therefore be of great importance to define precise prognostic factors in order to optimize the therapeutic approach in each case. Due to the rare occurrence of the disease, only one series of more than 30 cases has been published.495 Most of the cases reported in the literature were diagnosed by the retrospective review of lymph node biopsy specimens from We present here the results of a statistical search for prognostic factors in our own series of 30 directly diagnosed and prospectively followed patients with a minimal follow-up of 42 months. Materials and MethodsOur series includes 30 patients in whom AIL was directly diagnosed between 1975 and 198 1. Angioimmunoblastic lymphadenopathy was diagnosed histologically on lymph node biopsy specimens obtained before treatment. All biopsy specimens were reviewed by one of us (P.A.B.) and fulfilled the morphologic criteria established by Frizzera et al.': diffuse obliteration of the nodal architecture by a lymphoid cellular infiltrate including lymphocytes, pl...

It is known that cigarette smoking induces leukocytosis and increased genetic instability in normal individuals. Therefore, a retrospective review was conducted of 173 patients with chronic myelogenous leukemia to detect a possible influence of cigarette smoking on initial characteristics at the time of presentation and on the course of this disease. Thirty-nine patients (23%) were smoking 5 cigarettes/d or more at time of diagnosis. Cigarette smoking was significantly related to male sex (P = 0.0005) and younger age at diagnosis (P = 0.02) and smokers tended to have lower leukocyte counts (P = 0.07) than nonsmokers. Cigarette smoking was significantly associated with early blast crisis (P less than 0.0001) and short survival (P less than 0.0001). Other characteristics associated with a poor prognosis included hepatomegaly, anemia, and a high percentage of peripheral blast cells at time of diagnosis. When studied in a multivariate analysis, cigarette smoking remained the strongest prognostic factor for both occurrence of blast crisis (P = 0.0003) and overall survival (P = 0.0001). Other poor prognosis factors found in the multivariate analysis included a high percentage of blasts in the peripheral blood at time of diagnosis and high platelet count. It is possible that cigarette smoke may act as a promoter or cocarcinogen in the transformation of chronic myelogenous leukemia.