Changes in the total leukocyte and platelet counts in Papuan and non Papuan adults from northeast Papua infected with acute Plasmodium vivax or uncomplicated Plasmodium falciparum malaria

Background There are limited data on the evolution of the leukocyte and platelet counts in malaria patients. Methods In a clinical trial of chloroquine vs. chloroquine plus doxycycline vs. doxycycline alone against Plasmodium vivax (n = 64) or Plasmodium falciparum (n = 98) malaria, the total white cell (WCC) and platelet (PLT) counts were measured on Days 0, 3, 7 and 28 in 57 indigenous Papuans with life long malaria exposure and 105 non Papuan immigrants from other parts of Indonesia with limited malaria exposure. Results The mean Day 0 WCC (n = 152) was 6.492 (range 2.1–13.4) × 109/L and was significantly lower in the Papuans compared to the non Papuans: 5.77 × 109/L vs. 6.86 × 109/L, difference = -1.09 [(95% CI -0.42 to -1.79 × 109/L), P = 0.0018]. 14 (9.2%) and 9 (5.9%) patients had leukopaenia (<4.0 × 109/L) and leukocytosis (>10.0 × 109/L), respectively. By Day 28, the mean WCC increased significantly (P = 0.0003) from 6.37 to 7.47 × 109/L (73 paired values) and was similar between the two groups. Ethnicity was the only WCC explanatory factor and only on Day 0. The mean Day 0 platelet count (n = 151) was 113.0 (range 8.0–313.0) × 109/L and rose significantly to 186.308 × 109/L by Day 28 (P < 0.0001). There was a corresponding fall in patient proportions with thrombocytopaenia (<150 × 109/L): 119/151 (78.81%) vs. 16/73 (21.92%, P < 0.00001). Papuan and non Papuan mean platelet counts were similar at all time points. Only malaria species on Day 0 was a significant platelet count explanatory factor. The mean D0 platelet counts were significantly lower (P = 0.025) in vivax (102.022 × 109/L) vs. falciparum (122.125 × 109/L) patients. Conclusion Changes in leukocytes and platelets were consistent with other malaria studies. The Papuan non Papuan difference in the mean Day 0 WCC was small but might be related to the difference in malaria exposure.


Background
The malaria induced changes in the total white cell (WC) and platelet (PLT) counts have been documented in several clinical series of vivax and falciparum malaria in malaria immune and naïve patients of all ages [1][2][3][4][5][6][7][8][9], describing mostly the WC and PLT counts at presentation.
Studies examining trends in the total WCC are few. One study of experimentally induced falciparum malaria in non immune Americans showed an initial fall in the mean WCC reaching a nadir on Day 3 and rising thereafter to baseline values by study end [2]. Studies in African children have shown modest falls in the mean total WCC over 7 to 14 days whilst another study found that acutely ill falciparum infected children had a significant fall in the mean neutrophil count on Day 3 [29][30][31].
In north east Papua, indigenous Papuans have life long exposure to malaria whereas Indonesians who immigrate to Papua usually contract malaria for the first time in Papua [43]. Whether this difference in malaria exposure affects the white cell and platelet counts in diseased patients is unknown. This paper presents data from a clinical trial on the evolution of the total white cell and platelet counts and explores possible Papuan non Papuan differences.

Materials and methods
Malariometric data from this area of northeast Papua and study conduct are detailed elsewhere [44,45]. Briefly, indigenous adult Papuans (n = 57) and non Papuan (n = 105) adults from other parts of Indonesia were recruited into a hospital based, 28 days, clinical trial comparing chloroquine vs. chloroquine plus doxycycline vs. doxycycline for the treatment of parasitologically proven (Giemsa stained thick and thin blood films) acute uncomplicated falciparum and acute vivax malaria. There was no minimum parasitaemia for either species but a maximum of 150,000/μL asexual falciparum forms/μL for study entry. Recruited patients underwent a history, a detailed physical examination (including an abdominal examination to detect hepatosplenomegaly) and supervised treatment. Giemsa stained blood films were taken and read on Days 0-7, 14, 21, 28 and a blood count (Hb, total WCC, platelet count) was performed on Days 0, 3, 7, and 28; a manual differential white cell count was not done. Double entered, validated data were analysed using Epi Info 6.04b (Centers for Disease Control and Prevention, Atlanta, GA, USA) and Stata v 8 (Stata Corporation, USA). Normally distributed data were compared using the stu-dent's 't' test or ANOVA or the Mann-Whitney U or Kruskall-Wallis for skewed data. Spearman's rank correlation was used to determine the relationship between skewed continuous data. Proportions were compared using uncorrected chi squared values. Multivariate analyses were performed to assess the independence of ethnicity, splenomegaly, gender, malaria species and drug regimen on the total white cell and platelet counts. A P value of ≤ 0.05 was considered significant; no adjustment was made for multiple comparisons. Written informed consent was obtained from all patients. The study was conducted according to the Indonesian Ministry of Health, the Indonesian Navy, and the United States Navy and Army regulations governing the protection of human subjects.

Evolution of the total white cell counts
The changes in the mean white cell counts during follow up were similar for both malaria species and ethnic groups (Figures 1 and 2). Compared to baseline, the fall in the mean total WCC on Day 3 was significant in the non Papuans: (i) Pf: -0.898 (5.735 -6.633) × 10 9 /L (P = 0.0015), (ii) Pv: -1.069 (6.132 -7.201) × 10 9 /L (P = 0.0012). The mean Day 3 Day 0 difference between the non Papuans and Papuans was also significantly different The mean (95% CI) total white cell counts in falciparum infected Papuan and non Papuan patients  The mean (95% CI) total white cell counts in vivax infected Papuan (5.66 × 10 9 /L) and non Papuan (7.20 × 10 9 /L) patients

Discussion
These analyses have shown that at presentation the majority of malaria infected patients had total white cell counts within the normal range and most were thrombocytopaenic. By Day 28, both mean counts had increased but thrombocytopaenia remained in just over 20% of patients. These data complement and are broadly consistent with those of other clinical series.
Data from this study were from two dissimilar groups with respect to ethnicity and the degree of malaria acquired immunity. The Papuans had life long malaria exposure whilst the non Papuans had limited or no malaria exposure. The Papuans had lower mean Day 0 white counts than the non Papuans and there was a notable difference in the WCC changes over time. They rose steadily in the Papuans but fell significantly on Day 3 in the non Papuans, consistent with a study of experimen- The mean (95% CI) platelet counts in falciparum infected Papuan and non Papuan patients Pf-P Pf-nonP tally induced falciparum malaria in malaria naïve American volunteers [2]. By Day 28, the mean WCCs in the non Papuans had caught up and both groups had similar mean WCCs. This difference in the WCC progressions between the two ethnic groups is difficult to explain purely on the basis of a redistribution of white cells and/ or splenic sequestration, given that the malaria naïve non Papuans had significantly less splenomegaly. These data might indicate that a lesser degree of malaria exposure, thus acquired immunity, in the non Papuans resulted in a more robust inflammatory response, reflected crudely by a higher, baseline total WCC [27]. The initial fall in the total WCC in the non Papuans might reflect a difference in response to dying parasitized red cells consequent to treatment.

Mean (95% CI) platelet counts of malaria infected Papuans and non Papuans as a function of a palpable spleen
Thrombocytopaenia was common at presentation in both groups of patients. There were no differences between the Papuans and non Papuans but the mean platelet count was lower in vivax compared to falciparum infected patients. There were negative correlations between the Day 0 platelet count and the falciparum parasite count and the Day 0 temperature in the vivax patients. These findings are consistent with some but not other reports and are of limited clinical value [2,4,8,31,34]. Over time, the mean platelet count increased, consistent with the findings of others but two reports document an initial fall in the platelet count followed by recovery [2,4]. The mean platelet count appeared higher on Day 7 compared to Day 28 but there are no intervening platelet counts, so the day of the mean peak platelet count cannot be determined. Some 20% of patients were still thrombocytopaenic by study end, suggesting a number of patients need longer than four weeks for full platelet recovery. The multivariate analysis found drug arm to be significant factor for the Day 7 platelet count with the combination arm having a higher mean platelet count than the doxycycline arm. This is of little clinical significance.
Our study had limitations. The study sample was small and powered for a clinical trial. The data presented were secondary analyses to look for possible Papuan non Papuan differences. Many patients failed treatment before Day 28, given rescue treatment and withdrawn from the study; this may have introduced a statistical bias. Most patients were male, thus limiting the applicability of the findings. There were many statistical comparisons and some significant results may have occurred by chance. The differential white cell count was not measured and could have provided additional interesting data.

Conclusion
To conclude, these analyses have compared and contrasted the changes in the total white cell and platelet counts in Papuans and non Papuans with different degrees of malaria exposure that may partly explain the difference in the mean total white cell count at presentation between the two groups.
The mean (95% CI) platelet counts in vivax infected Papuan and non Papuan patients