[Soluble transferrin receptor and erythropoietin in chronic disease anemia with or without iron deficiency]. / Receptor soluble de transferrina y eritropoyetina en la anemia de los procesos crónicos con y sin deficiencia de hierro.

The regulation of transferrin receptor (RTF) is related to intracellular iron stores and with the soluble receptor is present in plasma. It has already been demonstrated that in iron deficiency anemia (IDA), receptor expression increases when iron stores decrease. In anemia of chronic diseases (ACD) it is difficult to establish the real iron status because of the influence exerted by inflammatory or infectious diseases on iron metabolism. We studied 30 healthy normal subjects and 42 anemic patients (hemoglobin less than 120 g/L) affected with ACD divided into two groups with and without iron deficiency, in order to establish the diagnostic value of measuring the soluble transferrin receptor (sRTF). We correlated erythropoietin (EPO) (as an erythropoietic stimulating factor) with the decreased hemoglobin values observed in both groups. The results were analysed with an ANOVA statistic test of one way analysis of variance, and there were no significant differences in sRTF values between the ACD groups with or without iron deficiency. The ratio log EPO vs hemoglobin showed a remarkably significant inverse correlation in both groups. We can conclude that sRTF levels are within the normal reference values in these patients and are not related to organic iron. Consequently, sRTF cannot be considered a good parameter for making a diagnosis of iron deficiency in chronic diseases.

Receptor soluble de transferrina y eritropoyetina en la anemia de los procesos crónicos con y sin deficiencia de hierro. / [Soluble transferrin receptor and erythropoietin in chronic disease anemia with or without iron deficiency]

The regulation of transferrin receptor (RTF) is related to intracellular iron stores and with the soluble receptor is present in plasma. It has already been demonstrated that in iron deficiency anemia (IDA), receptor expression increases when iron stores decrease. In anemia of chronic diseases (ACD) it is difficult to establish the real iron status because of the influence exerted by inflammatory or infectious diseases on iron metabolism. We studied 30 healthy normal subjects and 42 anemic patients (hemoglobin less than 120 g/L) affected with ACD divided into two groups with and without iron deficiency, in order to establish the diagnostic value of measuring the soluble transferrin receptor (sRTF). We correlated erythropoietin (EPO) (as an erythropoietic stimulating factor) with the decreased hemoglobin values observed in both groups. The results were analysed with an ANOVA statistic test of one way analysis of variance, and there were no significant differences in sRTF values between the ACD groups with or without iron deficiency. The ratio log EPO vs hemoglobin showed a remarkably significant inverse correlation in both groups. We can conclude that sRTF levels are within the normal reference values in these patients and are not related to organic iron. Consequently, sRTF cannot be considered a good parameter for making a diagnosis of iron deficiency in chronic diseases.

[Allogeneic hematopoietic transplantation with stem cells extracted from peripheral blood]. / Trasplante hematopoyético alogénico con células progenitoras extraidas de la sangre periférica.

Fifty three patients (pts) received an allogeneic hematopoietic transplant using peripheral blood progenitor cells (PBPC). Diagnosis were acute myeloid leukemia (AML) in 16 pts, acute lymphoblastic leukemia (ALL) in 15, chronic myeloid leukemia (CML) in first chronic phase in 12, aplastic anemia in 4, myelodysplasia in 3 and Hodgkin's disease, major thalasemia and Hunter's syndrome in one each. Mean age was 20 years-old (2-55), 28 males and 25 females. Conditioning regimens were total body irradiation with 1200 cGy and cyclophosphamide 120 mg/kg in 38 pts, busulfan 16 mg/kg and cyclophosphamide 120 mg/kg in 10 pts, total lymphoid irradiation and cyclophosphamide in 3, 2 pts received other chemotherapy based conditionings. PBPC were infused unmanipulated through a central catheter. Graft versus host disease (GVHD) prophylaxis was cyclosporin and short course methotrexate. Donors were 6/6 HLA compatible siblings in 52 cases and 5/6 match in one case. PBPC mobilization was done with G-CSF at a dose of 10 micrograms/kg/day subcutaneously for four days, pheresis started on day 5. Bone marrow harvest was also done in the first thirty cases. Mean cellularities for CD34, CD3, CD4, CD8, CD56, CD19 (cel x 10(6)/kg) were 4.12; 4.59; 2.57; 1.9; 0.55 and 0.68, respectively. Mean recovery of neutrophils > 500/microL was obtained on day +11 and platelets > 20,000/microL on day +13. Patients were hospitalized for a mean period of 26 days (range 18-39) and days with parenteral antibiotics were 12.2 (5-45). Two pts had venoocclusive disease of the liver. Transplant related mortality was 15%. Acute graft versus host disease (GVHD) was observed in 43.4% of pts, only 5 pts had acute GVHD III or IV. Mean time for aGVHD diagnosis was +23 (8-76). Forty three pts were evaluable for chronic GVHD with a mean follow-up of 18 months (4-39). Chronic GVHD was observed in 26.4% by day +240, only 2 pts developed severe cGVHD. The present experience demonstrates an acceptable incidence for cGVHD; however, taking into account recent reports showing an increase of this complication, it seems reasonable not to perform this procedure for non-malignant diseases in which graft versus malignancy effect is not to be expected.

Receptores histamínicos H2, AMPc y diferenciación celular leucémica / H2 histaminics receptora, AMPc and leukemia celular diferenciation

Con el objeto de explorar su potencial empleo en terapéutica no citotóxicas, se estudió, en células leucémicas humanas, la vía de señalización y probable rol regulatorio del receptor a histamina H2. Mediante ensayos de binding, se detectaron sitios de unión específica de tipo H2, en casi todas las muestras de M.O.y S.P. de pacientes con L.A., con diferentes grados de infiltración. esto sugiere la presencia del receptor H2, en células hemopoyéticas normales y transformadas. La líneas U-937, modelo de célula monoblástica, presenta receptores H2, acoplados a AMPc. Su estímulo no produjo cambios proliferativos, ni deferenciación celular, pero sí un aumento transitorio, vía proteín kinasa A (PKA), en la expresión de Fos y Hun, sin reducción de Myc. Se hipotizó que el fracaso del estímulo H2, para diferenciar las células U-937 podría deberse a que su activación de la PKA es breve. Concordante con lo anterior, los receptores H2, mostraron una veloz de desensibilización homóloga (T. 1/2 = 20´). En cambio la forskolina, un activador directo de la adenil ciclasa, no desensibilizó su estímulo ni aún después de 24 horas de incubación. La forskolina también inhibió la proliferación U-937 a las mismas concentraciones en que estimuló la síntesisde AMPc e indujo su diferenciación fagocitaria, con reducción del NBT y respuesta quimiotáctica al C5a. Conclusiones: 1) La desensibilización veloz de un receptor que transduce una señal diferenciadora, como el H2, en las células U-937, podría ser un mecanismo fisiopatogénico de la malignificación, al bloquear la recepción de estímulos madurativos por la célula neoplásica. 2) Dados estos resultados, y los efectos diferenciadores del dibutril AMPc (DBAMPc) en líneas celulares mieloides, los agentes que elevan el AMPc merecen ser valorados en la terapia de las LMA (AU)

Receptores histamínicos H2, AMPc y diferenciación celular leucémica / H2 histaminics receptora, AMPc and leukemia celular diferenciation

Con el objeto de explorar su potencial empleo en terapéutica no citotóxicas, se estudió, en células leucémicas humanas, la vía de señalización y probable rol regulatorio del receptor a histamina H2. Mediante ensayos de binding, se detectaron sitios de unión específica de tipo H2, en casi todas las muestras de M.O.y S.P. de pacientes con L.A., con diferentes grados de infiltración. esto sugiere la presencia del receptor H2, en células hemopoyéticas normales y transformadas. La líneas U-937, modelo de célula monoblástica, presenta receptores H2, acoplados a AMPc. Su estímulo no produjo cambios proliferativos, ni deferenciación celular, pero sí un aumento transitorio, vía proteín kinasa A (PKA), en la expresión de Fos y Hun, sin reducción de Myc. Se hipotizó que el fracaso del estímulo H2, para diferenciar las células U-937 podría deberse a que su activación de la PKA es breve. Concordante con lo anterior, los receptores H2, mostraron una veloz de desensibilización homóloga (T. 1/2 = 20ï). En cambio la forskolina, un activador directo de la adenil ciclasa, no desensibilizó su estímulo ni aún después de 24 horas de incubación. La forskolina también inhibió la proliferación U-937 a las mismas concentraciones en que estimuló la síntesisde AMPc e indujo su diferenciación fagocitaria, con reducción del NBT y respuesta quimiotáctica al C5a. Conclusiones: 1) La desensibilización veloz de un receptor que transduce una señal diferenciadora, como el H2, en las células U-937, podría ser un mecanismo fisiopatogénico de la malignificación, al bloquear la recepción de estímulos madurativos por la célula neoplásica. 2) Dados estos resultados, y los efectos diferenciadores del dibutril AMPc (DBAMPc) en líneas celulares mieloides, los agentes que elevan el AMPc merecen ser valorados en la terapia de las LMA

[Thrombotic microangiopathies: HUS/TTP. Physiopathological aspects]. / Microangiopatias trombóticas: SUH/PTT. Aspectos fisiopatológicos.

In thrombotic thrombocytopenic purpura (TTP) and in the hemolytic uremic syndrome (HUS) fibrin-platelet thrombi occlude arterioles and capillaries. The mechanism of endothelial cell injury and the mechanism of thrombosis are the most important physiopathological events in this pathology and are largely unknown. In HUS due to the Shiga toxin, the lesion of the endothelial cells is due to penetration of the toxin into the cell via the Gb3 receptor. Endothelial cell death is a consequence of altered protein synthesis at the ribosomal level. Cytokines released during the inflammatory process, possibly enhance the endothelial damage. Genetic and immunologic predisposing factors for the development of HUS have also been postulated. In idiopathic, secondary and familial HUS/TTP the mechanism of endothelial lesion is unknown, but multiple responsible factors have been advocated such as infections, drugs, pregnancy, autoantibodies, apoptosis inducing molecules, etc. and other genetic, hormonal or immunologic predisposing factors may also be involved. Factor H deficiency has been blamed in familiar cases. The most important cause of microcirculation thrombosis is the thrombogenic capacity of endothelial cell "activation" or injury induced by multiple mechanisms. The predominant source of plasma vW factor multimers is apparent in the altered endothelial cell. The unusually large vWF multimers are more effective at binding to platelet glycoprotein Ib-IX and IIb-IIIa complexes and inducing aggregation, as also occurred with the low weight multimers formed with excessive proteolysis, as described in the acute phase of HUS/TTP. The recent report of congenital deficiency of a vWF protease in familial TTP and its functional inhibition by autoantibodies in acquired cases is characteristic of TTP. This protease inhibition has never been described in HUS and might represent pathogenetic differences between TTP and HUS, and contribute to the differential diagnosis, but further confirmation of these findings is needed. We postulate that the abnormal cleavage of the vWF subunit, with formation of different multimers with increased platelet aggregating capacity is an important mechanism to increase the microcirculatory thrombosis, but it is only a partial aspect in a more complex and unknown thrombogenic stimulation secondary to the endothelial lesion or activation. Better knowledge of the endothelial physiology and the genetic polymorphism of the endothelial cell, the clonation of vWF-cleavage protease, etc., will provide valuable tools for the understanding of these fascinating entities.

Microangiopatias trombóticas: SUH/PTT. Aspectos fisiopatológicos. / [Thrombotic microangiopathies: HUS/TTP. Physiopathological aspects]

In thrombotic thrombocytopenic purpura (TTP) and in the hemolytic uremic syndrome (HUS) fibrin-platelet thrombi occlude arterioles and capillaries. The mechanism of endothelial cell injury and the mechanism of thrombosis are the most important physiopathological events in this pathology and are largely unknown. In HUS due to the Shiga toxin, the lesion of the endothelial cells is due to penetration of the toxin into the cell via the Gb3 receptor. Endothelial cell death is a consequence of altered protein synthesis at the ribosomal level. Cytokines released during the inflammatory process, possibly enhance the endothelial damage. Genetic and immunologic predisposing factors for the development of HUS have also been postulated. In idiopathic, secondary and familial HUS/TTP the mechanism of endothelial lesion is unknown, but multiple responsible factors have been advocated such as infections, drugs, pregnancy, autoantibodies, apoptosis inducing molecules, etc. and other genetic, hormonal or immunologic predisposing factors may also be involved. Factor H deficiency has been blamed in familiar cases. The most important cause of microcirculation thrombosis is the thrombogenic capacity of endothelial cell [quot ]activation[quot ] or injury induced by multiple mechanisms. The predominant source of plasma vW factor multimers is apparent in the altered endothelial cell. The unusually large vWF multimers are more effective at binding to platelet glycoprotein Ib-IX and IIb-IIIa complexes and inducing aggregation, as also occurred with the low weight multimers formed with excessive proteolysis, as described in the acute phase of HUS/TTP. The recent report of congenital deficiency of a vWF protease in familial TTP and its functional inhibition by autoantibodies in acquired cases is characteristic of TTP. This protease inhibition has never been described in HUS and might represent pathogenetic differences between TTP and HUS, and contribute to the differential diagnosis, but further confirmation of these findings is needed. We postulate that the abnormal cleavage of the vWF subunit, with formation of different multimers with increased platelet aggregating capacity is an important mechanism to increase the microcirculatory thrombosis, but it is only a partial aspect in a more complex and unknown thrombogenic stimulation secondary to the endothelial lesion or activation. Better knowledge of the endothelial physiology and the genetic polymorphism of the endothelial cell, the clonation of vWF-cleavage protease, etc., will provide valuable tools for the understanding of these fascinating entities.

Trasplante hematopoyético alogénico con células progenitoras extraidas de la sangre periférica. / [Allogeneic hematopoietic transplantation with stem cells extracted from peripheral blood]

Fifty three patients (pts) received an allogeneic hematopoietic transplant using peripheral blood progenitor cells (PBPC). Diagnosis were acute myeloid leukemia (AML) in 16 pts, acute lymphoblastic leukemia (ALL) in 15, chronic myeloid leukemia (CML) in first chronic phase in 12, aplastic anemia in 4, myelodysplasia in 3 and Hodgkins disease, major thalasemia and Hunters syndrome in one each. Mean age was 20 years-old (2-55), 28 males and 25 females. Conditioning regimens were total body irradiation with 1200 cGy and cyclophosphamide 120 mg/kg in 38 pts, busulfan 16 mg/kg and cyclophosphamide 120 mg/kg in 10 pts, total lymphoid irradiation and cyclophosphamide in 3, 2 pts received other chemotherapy based conditionings. PBPC were infused unmanipulated through a central catheter. Graft versus host disease (GVHD) prophylaxis was cyclosporin and short course methotrexate. Donors were 6/6 HLA compatible siblings in 52 cases and 5/6 match in one case. PBPC mobilization was done with G-CSF at a dose of 10 micrograms/kg/day subcutaneously for four days, pheresis started on day 5. Bone marrow harvest was also done in the first thirty cases. Mean cellularities for CD34, CD3, CD4, CD8, CD56, CD19 (cel x 10(6)/kg) were 4.12; 4.59; 2.57; 1.9; 0.55 and 0.68, respectively. Mean recovery of neutrophils > 500/microL was obtained on day +11 and platelets > 20,000/microL on day +13. Patients were hospitalized for a mean period of 26 days (range 18-39) and days with parenteral antibiotics were 12.2 (5-45). Two pts had venoocclusive disease of the liver. Transplant related mortality was 15

. Acute graft versus host disease (GVHD) was observed in 43.4

of pts, only 5 pts had acute GVHD III or IV. Mean time for aGVHD diagnosis was +23 (8-76). Forty three pts were evaluable for chronic GVHD with a mean follow-up of 18 months (4-39). Chronic GVHD was observed in 26.4

by day +240, only 2 pts developed severe cGVHD. The present experience demonstrates an acceptable incidence for cGVHD; however, taking into account recent reports showing an increase of this complication, it seems reasonable not to perform this procedure for non-malignant diseases in which graft versus malignancy effect is not to be expected.

[Immunocytochemistry techniques for the diagnosis of hematologic neoplasms]. / Aplicación de técnicas inmunocitoquímicas para el diagnóstico de neoplasias hematólogicas.

We studied 22 patients with hematological neoplasias which included: 12 patients with a diagnosis of Acute Myeloblastic Leukemia (AML) following the morphology and cytochemistry criteria established by FAB (French, American and British Committee), a Myeloblastic Leukemia secondary to MDS (Myelodysplastic Syndromes) and a biphenotypic acute leukemia where we established the relationship between the traditional peroxidase reaction with the anti-MPO by APAAP. We also carried out the nonspecific esterase reaction and determined the immunologic phenotype by FACS technology. The same procedure was used for the cellular analysis of the light chains kappa (kappa) and lambda (lambda) in 3 cases of hairy cell leukemia, one lymphoma and 4 cases of plasma cell neoplasia and reactive plasma cell disease. We conclude that immunocytochemical reactions must be used when morphology and traditional cytochemical reactions need to be confirmed in order to establish a correct diagnosis and this is specially important for B and T lymphomas. Their prognostic value is restricted and the results are useful as a complement to morphology, cytochemistry and immunological determinations.

Hemolytic uremic syndrome in children: platelet aggregation and membrane glycoproteins.

PURPOSE: Several mechanisms have been proposed to explain the fibrin-platelet thrombosis at the microcirculation level in the different clinical conditions of hemolytic uremic syndrome (HUS). The relationships between platelet structure and function during the first 4 weeks of evolution of the disease were studied to understand the mechanism of platelet alteration. PATIENTS AND METHODS: Coagulation parameters, platelet counts, and aggregation were studied in 49 children, and membrane glycoproteins (GPs) in 20 of the 49 children (mean age, 17 months) with HUS (Group 2) were studied during the first 4 weeks of evolution of the disease. RESULTS: No disseminated intravascular coagulation was found in patients with recurrent or persistent thrombocytopenia. Platelet aggregation was sequentially performed during the first weeks of evolution. All patients had a functional decrease in the acute period of HUS. Platelet GPs GPIb, GPIIbIIIa, GPIIb, and GPIIIa were evaluated. GPIIbIIIa complex presented low level and never reached normal values during the first 4 weeks of disease. CONCLUSIONS: Platelet alterations are probably caused by multiple mechanisms: "exhausted" platelets, structural membrane alterations caused by arginine-glycine-aspartic peptide blockade, or diminished or nonfunctional membrane GPIb and GPIIbIIIa complexes.

Aplicación de técnicas inmunocitoquímicas para el diagnóstico de neoplasias hematólogicas. / [Immunocytochemistry techniques for the diagnosis of hematologic neoplasms]

We studied 22 patients with hematological neoplasias which included: 12 patients with a diagnosis of Acute Myeloblastic Leukemia (AML) following the morphology and cytochemistry criteria established by FAB (French, American and British Committee), a Myeloblastic Leukemia secondary to MDS (Myelodysplastic Syndromes) and a biphenotypic acute leukemia where we established the relationship between the traditional peroxidase reaction with the anti-MPO by APAAP. We also carried out the nonspecific esterase reaction and determined the immunologic phenotype by FACS technology. The same procedure was used for the cellular analysis of the light chains kappa (kappa) and lambda (lambda) in 3 cases of hairy cell leukemia, one lymphoma and 4 cases of plasma cell neoplasia and reactive plasma cell disease. We conclude that immunocytochemical reactions must be used when morphology and traditional cytochemical reactions need to be confirmed in order to establish a correct diagnosis and this is specially important for B and T lymphomas. Their prognostic value is restricted and the results are useful as a complement to morphology, cytochemistry and immunological determinations.

Forskolin induces U937 cell line differentiation as a result of a sustained cAMP elevation.

The present study examines the effects of forskolin on U937 cell differentiation. We recently reported that dibutyryl cAMP (dbcAMP), but not cAMP-elevating agents such as histamine, promotes U937 cell differentiation. cAMP production elicited by stimulation of histamine H2 receptors showed a rapid, homologous desensitization, which might explain the dissimilar responses to histamine and dbcAMP. Forskolin induced an increase in cAMP levels in a concentration-dependent manner (EC50=30 microM) for an extended period of at least 24 h. Forskolin but not histamine (up to 100 microM), also inhibited cell growth in a dose-dependent fashion (EC50=22 microM). After 3 days of incubation, 75 microM forskolin induced U937 cell differentiation as judged by an increased rate of reduction of nitrobluetetrazolium (mean+/-S.E.M.: 21.3+/-6.6% in treated cells vs. 3.2+/-1.9% in the control group, P < 0.001) and an augmented chemotactic response to complement 5a (C5a) (33.2+/-5.9% in forskolin-treated vs. 0.34+/-0.12% in control cells, P < 0.01). Furthermore, c-Myc levels decreased following forskolin treatment, while the histamine H2 receptor agonist dimaprit had no effect. We conclude that forskolin induces U937 cell differentiation through a sustained rise in cAMP levels.

[Markers of thrombin generation in kidney failure of children with epidemic hemolytic uremic syndrome]. / Marcadores de generación de trombina en insuficiencia renal de niños con sindrome urémico hemolítico epidémico.

We present herein the levels of the early markers of blood coagulation activation and TNF-alpha in 12 children with the epidemic form of the hemolytic-uremic syndrome, median age 16 months, range 12-18. All patients recovered from the disease within 2 to 4 weeks. Four blood samples were collected: at admission, 1 week and 2 weeks later and on remission. Prothrombin fragments 1 + 2 (F1 + 2), thrombin-antithrombin complex (TAT) and tumor necrosis factor alpha (TN-alpha) were assayed by commercial ELISA techniques, while von Willebrand factor (vWf) was measured by Laurell's method. At admission, F1 + 2 and TAT levels were 7.8 nM (3.7-12.3) and 22.7 ng/ml (8-76), respectively. Besides, significant correlations were obtained for F1 + 2 levels vs blood creatinine, r: 0.57 p < 0.001; F1 + 2 vs urea, r: 0.66 p < 0.001; TAT vs blood creatinine, r: 0.77 p < 0.001; TAT vs blood urea, r: 0.59 p < 0.001. Median vWf value at admission in 11/12 children was 260% (170-420), correlating with F1 + 2, r: 0.77 p < 0.001 and with TAT, r: 0.41 p < 0.01. Such values tended to normalize with the improvement of the disease. A negative correlation was found for platelet count vs F1 + 2, r: -0.64 p < 0.001. TNF-alpha levels were increased in 5/12 children, 22.2 pg/ml (17.2-53.7). These results may be attributable to similar stimuli on endothelial cells.

Oxidative stress and membrane fluidity in erythrocytes from patients with hemolytic uremic syndrome.

In order to investigate the implications of oxidative disturbances in the hemolysis associated with the Hemolytic Uremic Syndrome (HUS), basal levels of lipid peroxidation products, the response to t-butyl hydroperoxide induced damage and membrane fluidity were assayed by the technique of electron spin resonance in erythrocytes spin labeled with 5-Doxyl stearic acid obtained from eight children with HUS, during the 1st, 2nd, 4th and 12th weeks after diagnosis. During the acute phase of the disease, red blood cells (RBC) showed increased initial lipid peroxidation products, a higher susceptibility to oxidative insult and a lower membrane fluidity. All parameters reached control values the 12th week after diagnosis. The results suggest that in the acute phase of HUS, RBCs are exposed to an oxidative imbalance that could contribute to hemolysis directly through oxidative damage and/or by decreasing membrane fluidity.

[Prognostic factors and mortality among adult patients with acute leukemia and infection]. / Factores pronósticos y mortalidad en pacientes adultos con leucemia aguda e infección.

In order to define mortality rates and clinical findings with prognostic value in febrile infections among our adult patients with acute leukemia we prospectively studied--during a 34 months period--177 episodes of clinical suspected infection which occurred in 49 patients. By means of univariate analysis and a subsequent multiple logistic regression study, the association between 27 clinical and microbiological data and febrile episode survival rates were evaluated. Both the overall mortality rate and the specific one for febrile episodes were high (44.9% and 12.7% respectively). An age over 30 years old (p = 0.025), fever lasting more than five days (p = 0.025), lung involvement (p = 0.001) and fungal isolation in a culture specimen (p = 0.005) were all associated with a higher episode mortality. However, only an age older than 30 years (adjusted odds ratio, A.O.R. = 1.118; 95% confidence interval, C.I.95% = 1.015-1.232; p = 0.025) and pneumonia (A.O.R. = 1.454; C.I.95% = 1.288-1.642; p < 0.001) remained as independent predictors of a greater mortality in the multivariate analysis. Although fever of unknown origin was associated with a better prognosis (p = 0.024) two other variables--viral infections (A.O.R. = 0.642; C.I.95% = 0.421-0.979; P = 0.041) and the isolation of two or more etiologic agents (A.O.R. = 0.795; C.I.95% = 0.651-0.972; p = 0.027)--had a protective value with the multiple regression analysis.