| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Nevertheless, platelets have been implicated in animal, clinical and experimental studies of malaria pathogenesis. Platelets mediate a syndrome of cerebral malaria in an animal model of malaria infection.4 Histopathological studies of children who have died of severe malaria showed that platelet clumps with and without infected erythrocytes are frequently found in the vasculature.5 Finally, infected erythrocytes may adhere to platelets, and the clumps of infected erythrocytes and platelets have been associated with severe disease.6
Therefore, there remains a paradox that, while thrombocytopenia is associated with infection and platelets have been implicated in the pathogenesis of severe disease, most studies suggest that low counts of platelets are not associated with an adverse outcome.2,7 Without understanding this paradox, it is not possible to evaluate the true contribution of platelet to the pathogenesis of severe malaria.
The causes of thrombocytopenia in acute malaria are poorly understood. Certainly, increased platelet destruction is significant during malaria infection. The platelet lifespan is reduced during malaria,8 which seems to be associated with a diffuse pattern of platelet sequestration rather than a predominant splenic or hepatic clearance.9 Thrombocytopenia is often associated with palpable splenomegaly10 and circulating immune complexes.1
The contribution of abnormalities of platelet production to thrombocytopenia during malaria infection is less clear. Early studies in malaria patients described dysmorphic megakaryocytes in the bone marrow of patients with acute falciparum malaria.11 Other investigators showed less lobulated and immature megakaryocytes in Gambian children with acute malaria.12 Moreover, the biosynthesis and regulation of thrombopoietin (TPO), the main growth factor for megakaryocytes and thrombocytopoiesis, seems to be normal in patients with malaria.8
The cytokines released during an acute inflammatory response contribute to the pathogenesis of thrombocytopenia. Tumor necrosis factor-
(TNF-) has been associated with platelet consumption in mice,13 and the administration of interleukin (IL)-10 to healthy volunteers has been shown to cause thrombocytopenia.14 TNF- and IL-10 have been reported to play an important role in the development of Plasmodium falciparum malarial anemia,15 but the role of these cytokines has not been studied in the development of thrombocytopenia in patients with acute malaria. We therefore examined the association of these cytokines with thrombocytopenia to elucidate the relationship between thrombocytopenia and severe disease.
Laboratory investigations.
A 3-mL venous blood sample was collected into EDTA and processed immediately. A full blood count was obtained by a hematology analyzer (Coulter MD II; Coulter Corp., Miami, FL). Peripheral blood films were stained with 3% May-Grünwald-Giemsa. Absolute parasite counts were calculated for each child using either thin films (counting the number of parasitized RBCs/500 RBCs) or from thick films (counting the number of parasitized RBCs/200 white blood cells [WBCs]) stained with 10% Giemsa and with reference to the data from the full blood count. Plasma concentrations of TNF-, IL-10, interferon (IFN)-
, IL-12, and erythropoietin (Epo) were measured by ELISA (R&D Systems, Abingdon, United Kingdom).
Statistical analyses.
The values of parasitemia, TNF-, IL-10, IL-12, IFN-, and platelet counts were normally distributed when log-transformed. We used Pearson’s correlation to measure the association between the variables studied. Groups were compared using the non-parametric Mann-Whitney test. We used logistic regression to study the association of independent variables with thrombocytopenia (dependent variable). For binary logistic regression, thrombocytopenia was treated as a dichotomized variable using a cut-off for platelet counts of 150 x 109/L (SPSS 11.0; SPSS, Chicago, IL).
The subgroup of children with thrombocytopenia (N = 41) was compared with children with platelet counts > 150 x 109/L (N = 78; Table 1
). No differences in age, sex ratio, or nutritional status (weight-for-age Z-score) were found when children with thrombocytopenia were compared with children with normal platelet counts. Children with thrombocytopenia had lower Hb values and higher P. falciparum parasitemias. However, these differences did not reach statistical significance. Children with acute malaria and thrombocytopenia had significantly higher levels of plasma IL-10 (P = 0.002) and lower levels of IL-12 (P = 0.012).
|
, IL-10, IFN-, IL-12, nutritional status, and age as independent variables, the only variable significantly associated with thrombocytopenia was IL-10 (Exp [B], –3.1; 95%CI, –1.7 to –5.7; P < 0.01). The same association was confirmed when platelet counts on admission were used as a continuous dependent variable in a regression model (slope [B], –39.6; 95%CI, –59 to –20; P < 0.01; Figure 1).
|
There is evidence that IL-10 may directly induce thrombocytopenia. The administration of a low dose of recombinant human IL-10 (8 µg/kg/d) decreased platelet production in healthy adult volunteers.14 In the same study, there was a corresponding reduction in splenic sequestration of platelets in the IL-10–treated group compared with the placebo-treated subjects. In the IL-10–treated group, there was a trend toward lower numbers of megakaryocyte colony-forming units (CFU-MKs) compared with volunteers who received placebo. This single study suggests that IL-10–induced reduction in platelet count is caused, at least in part, by a reduction in platelet production.
Thrombocytopenia is a common finding in patients with HIV.17 Therefore, the interpretation of thrombocytopenia as a marker of clinical outcome in malaria-endemic areas should take into account the HIV status of the population studied. In our study, only 2 of 120 children were positive for HIV, and it is therefore unlikely that HIV status confounded our results.
How do these findings explain the relationship between thrombocytopenia and severe disease? The clinical relevance of thrombocytopenia associated to P. falciparum malaria has been addressed in two large studies. In a clinical study in Kenyan children, platelet counts < 150 x 109/L were found in 57% of children with acute malaria and were associated with age, prostration, and parasite density, but not with bleeding problems or mortality.2 On the other hand, a recent study carried out in a hypoendemic area in Senegal identified thrombocytopenia as an independent predictor of death (odds ratio = 13.3).18 However, other data from Southeast Asia do not support a relationship between thrombocytopenia and outcome.7 These inconsistent relationships between disease severity or outcome and thrombocytopenia may reflect the differences in clinical epidemiology and the corresponding differences immune status and immunopathology in patients with severe malaria. Our study could not include death as a dependent variable because none of the 120 children studied died during the follow-up period. However, platelet counts in the subgroup of children with severe malarial anemia (N = 29) were not significantly different from those with mild or moderate anemia. Therefore, our results do not support a relationship between thrombocytopenia and disease severity, in relation to severe malarial anemia, in P. falciparum malaria.
Indeed, the association between thrombocytopenia and IL-10 levels in malaria infection may explain, at least in part, the failure to find consistent associations of thrombocytopenia and poor outcome in malaria infection despite the evidence from clinical experimental and animal studies for a role of platelets in the pathophysiology of malaria. IL-10 itself is associated with less severe forms of clinical malaria,15 and this relationship would confound any simple relationship between thrombocytopenia and outcome. It is possible that a reduction in platelet count mediated by IL-10 reduces the pathologic interaction between infected erythrocytes and platelets, associated with severe disease,5,6 and is therefore one mechanism whereby IL-10 improves the outcome of severe malaria.
Our results suggest that the significance of platelets in the pathophysiology of falciparum malaria cannot be gauged by determining the simple relationship between thrombocytopenia and outcome. Further studies to define the role of platelets in the pathophysiology of severe malaria will require more detailed functional and histopathological approaches to measure the activation and fate of platelets during malaria infection.
Received March 2, 2006. Accepted for publication May 18, 2006.
Acknowledgments: C.C-P has been supported by the University of Oxford and a BEFI grant (ISCIII, Ministry of Health, Spain). D.J.R is supported by the National Blood Service, the Howard Hughes Memorial Institute, and EuBioMalPar NoE. The research was carried out in part at the NBS-Oxford Centre and benefits from NHS R&D funding. The Centre for Geographic Medicine Coast is supported by core funds from the Wellcome Trust, UK and the Kenyan Institute of Medical Research.
* Address correspondence to David J. Roberts, National Blood Service Oxford Centre, John Radcliffe Hospital, Headington, Oxford OX39BQ, UK. E-mail: david.roberts{at}ndcls.ox.ac.uk 
Authors’ addresses: Climent Casals-Pascual and David J. Roberts, Blood Research Laboratory, National Blood Service Oxford Centre, John Radcliffe Hospital, Headington, Oxford OX39BQ, UK. Oscar Kai, Charles RJC Newton, and Norbert Peshu, KEMRI Centre for Geographic Medicine Coast, PO Box 230, Kilifi, Kenya.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
