| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1,2Department of Medicine1 and Department of Medical Microbiology2, Faculty of Medicine and Health Sciences, UAE University Medical School, PO Box 17666, Al-Ain, United Arab Emirates 3Department of Haematology and Oncology, Tawam Hospital, PO Box 15555, Al-Ain, United Arab Emirates
Correspondence Michael Ellis michael.ellis{at}uaeu.ac.ae
Received April 17, 2005
Accepted August 2, 2005
Serum RANTES (regulated on activation, normal T-cell expressed and secreted) concentrations were measured in 14 patients who had haematological malignancies and developed invasive fungal infections (three of them definite, eight probable and three possible). RANTES levels fell substantially from pre-chemotherapy values at the start of and throughout the fungal infection, and recovered in patients who survived the fungal infection. However, in patients who died from the invasive fungal infection, RANTES levels did not recover. For survivors the mean ± SD levels for RANTES were 7656 ± 877 pg ml–1 on the day prior to chemotherapy, 3723 ± 2443 pg ml–1 on the first day of fungal infection diagnosis (significantly different from baseline; P = 0.001) and 9078 ± 2256 pg ml–1 at recovery from the fungal infection (significantly different from lowest value; P < 0.0001). Platelet counts were closely correlated with the RANTES levels (r = 0.63, P < 0.001). The RANTES concentrations for the three patients who died were similar to those who survived at all equivalent timepoints, but were significantly lower at the time of death (792 ± 877) compared to the values at recovery for survivors (P = 0.005). The finding that patients who died from an invasive fungal infection had very low platelet counts and RANTES concentrations suggests that these could play a role in host response to such infections.
 |
INTRODUCTION |
|---|
 TOP INTRODUCTION METHODSRESULTS AND DISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Non-haematological patients dying with severe sepsis have lower circulating RANTES concentrations compared to survivors (Cavaillon et al., 2003), suggesting a protective role for it; however, there is no information in patients with haematological malignancy. This patient group is characterized by profound thrombocytopenia and also has a high risk for IFI morbidity and mortality. Since platelets are a major source of RANTES, we investigated RANTES changes in such patients who had an IFI.
|
METHODS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Patients admitted with acute malignant haematological disease received standard induction chemotherapy according to international guidelines (Coiffier et al., 2002; Anonymous, 1992; Farag et al., 2005). Patients provided their informed written consent for this study, which was approved by the Tawam Hospital Research Review Committee.
The diagnosis of IFI was based on a modification of the EORTC/MSG (European Organisation for Research and Treatment of Cancer/Mycoses Study Group) criteria (Ascioglu et al., 2002). Patients who were treated empirically with systemic antifungal drugs were also included, i.e. patients with antibiotic-unresponsive neutropenic fever (AUNF). These had 72 h of neutropenic fever, negative blood cultures and absence of focal signs of infection, and were unresponsive to broad-spectrum antibiotic treatment with piperacillin-tazobactam and gentamicin and classified as high risk for overt IFI (Hughes et al., 2002).
At the onset of IFI diagnosis, treatment with liposomal amphotericin B at 5 mg kg–1 day–1 (3 mg kg–1 day–1 for empirical treatment) was commenced, increasing the dose up to 10 mg kg–1 day–1 depending on the response. Granulocyte colony-stimulating factors were given to all patients at the time of clinical onset of IFI.
Mean values of RANTES concentrations and platelet counts were compared by the Mann-Whitney test. The correlation between platelets and RANTES was determined by the Pearson r test. A P value of < 0.05 was considered significant.
|
RESULTS AND DISCUSSION |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Study patients and IFIs
We recruited 14 patients with IFI. Ten of the 14 patients had acute myeloid leukaemia and were treated either with standard induction chemotherapy (nine patients) or with salvage therapy (one patient). Another four patients were diagnosed with acute lymphoblastic leukaemia and received appropriate induction chemotherapy.
Of the 14 patients, nine had invasive pulmonary aspergillosis (IPA), one had invasive sinus aspergillosis, one had candidemia, one had candidal pneumonia and two had antibiotic-unresponsive neutropenic fever. Eleven patients survived the IFI and three died from the IFI. Details of these patients are shown in Table 1.
|
Serum RANTES concentrations
In the 11 patients who survived the IFI, the mean ± SD (95 % confidence interval) RANTES level on the day prior to receiving chemotherapy was 7656 ± 877 (6078 to 9235) pg ml–1 (Fig. 1). The RANTES level progressively fell between baseline and the first day of onset of the IFI, when the level was 3723 ± 2443 (2081 to 5365) pg ml–1 (significantly different from baseline; P = 0.0001). RANTES levels fell further during the course of the fungal infection. The mean minimal daily value recorded during the IFI was 1159 ± 2242 (–347 to 2666) pg ml–1 (significantly different from the start of the IFI; P = 0.005). At recovery from infection the mean RANTES concentration was 9078 ± 2256 (7562 to 10 594) pg ml–1 (significantly different from the minimum; P < 0.0001). In the three patients who died from their IFI the corresponding levels (in pg ml–1) were: baseline, 8422 ± 2573 (2028 to 14 816); start, 1608 ± 1463 (–2027 to 5244); lowest, 16; and death, 792 ± 877 (–1387 to 2972). The values at recovery for the survivors were significantly higher compared to those at death for those who died (P = 0.005) (Table 1, Fig. 1).
|
Therefore, these results show that chemotherapy results in a progressive fall in serum RANTES concentrations, beginning on the second day of treatment, associated with myeloablative thrombocytopenia and leukopenia. At the beginning of an IFI, RANTES levels fell substantially in all except one patient, reaching concentrations that were approximately 40 % of baseline values. These decreased further in all but one patient during the IFI, reaching minimum levels of only 10 % of baseline. Subsequently, patients who recovered from their IFI had an associated recovery of RANTES levels to at least baseline concentrations. However, patients who died of the IFI showed either no recovery from minimal levels observed during the IFI (one patient) or only partial low concentration recovery (two patients).
Although the number of patients in this study is small, there is consistency in the pattern of RANTES changes for each patient with an IFI. These observations suggest that our findings are accurate.
Platelet counts
Platelet counts changed in a parallel fashion to the RANTES concentrations; the correlation between platelets and RANTES was significant (r = 0.63, P < 0.0001) (Table 1). Platelet count data were available at the recovery stage for 10 surviving patients, seven of whom had platelet counts that had recovered to 
83 x 109 l–1. In all three patients who died from IFI the platelets did not rise to above 26 x 109 l–1.
The correlation that we observed between platelet counts and RANTES concentrations is consistent with other data that suggest that platelets are the major source of this chemokine (Cavaillon et al., 2003). In addition to providing a powerful leukocyte chemoattractant, platelets may also have direct antifungal activity, through a direct hyphal damaging effect (Christin et al., 1998). The role of platelets in protecting against IFI is consistent with the persistent thrombocytopenia observed in each of the three patients who died compared to the recovery of platelet counts seen in the survivors of IFI.
The temporary increases in RANTES concentrations observed on some days, e.g. days 9 and 11 in patient 6 (Fig. 2c), may have arisen from the platelet transfusions given for supportive therapy at those times. It is not possible to determine whether platelet-transfusion-associated RANTES are biologically active and hence whether platelet transfusions have any therapeutic role in managing IFI. It is also notable that in several patients RANTES recovery occurred prior to platelet recovery (Fig. 2c). This suggests not only that there is an additional source for RANTES, e.g. the recovering gut epithelium (Ellis, 2004), but also that the direct antifungal activity of platelets may be different from its indirect activity mediated via RANTES. In addition, the only patient who did not show a substantial fall in RANTES during the IFI (patient 4) might have been able to sustain a normal RANTES level as a result of less damage to extra-platelet sources as he was the only subject to receive a salvage chemotherapy regimen.
|
Neutrophil counts
Among the 11 survivors, the time at which neutrophils began to recover (defined as the first day of progressive rise in neutrophils above 0.1 x 109 l–1) was the same as the time of RANTES recovery (defined as the first day of progressive rise in RANTES from the lowest level) in four patients, preceded RANTES recovery by at least 2 days in five patients and lagged behind RANTES recovery by at least 2 days in two patients. The neutrophil count remained < 0.1 x 109 l–1 in the three patients who died.
The increased RANTES concentrations seen prior to adequate circulating leukocytes suggest that RANTES recovery may be an immunological prerequisite for activating and trafficking neutrophils to fungal infection sites.
Selected patient examples
Fig. 2 illustrates some notable findings. Fig. 2(a) shows the sequence of events in patient 12, who died from pneumonia caused by species other than C. albicans. RANTES levels fell from a baseline value of 10 350 pg ml–1 to values that were generally below 2000 pg ml–1 during the fungal infection. Temporary elevations of RANTES concentrations were noted during the last 7 days, some of which reached approximately baseline values, but were not sustained. They fell dramatically in the last 2 days prior to death. All three patients who died had similar sequential changes in RANTES.
Fig. 2(b) shows the results from patient 8, who had IPA. RANTES levels fell as described for patient 12, but substantial sustained recovery of RANTES was observed by day 15. Neutrophil recovery was seen to lag behind the RANTES recovery by approximately 2 days. Fig. 2(c) illustrates profound thrombocytopenia, neutropenia and severe depletion of RANTES at the time of IPA, as seen in patient 6. Recovery of RANTES levels occurred prior to platelet and neutrophil recovery. Platelet transfusions are associated with unsustained elevations of RANTES concentrations.
Conclusions
Our preliminary observations do not permit us to precisely determine the role that RANTES may have in governing the outcome of an IFI or to distinguish the individual contributions of neutropenia and thrombocytopenia. However, the finding of a substantially depleted RANTES environment in patients with IFI and of persistence of these low concentrations in patients who died from IFI requires further investigation, for example considering the potential antifungal therapeutic dual role that platelet transfusions may have and the effect of recombinant RANTES on the outcome of IFI.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
This study was supported by a grant from the Terry Fox Foundation.
|
REFERENCES |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | J MED MICROBIOL | MICROBIOLOGY | J GEN VIROL | ALL SGM JOURNALS |
, RANTES; x, neutrophils; 
, platelets. Arrows indicate start and end of infection.