The Journal of Infectious Diseases

MAJOR ARTICLE

Nitric Oxide–Dependent Endothelial Dysfunction and

Reduced Arginine Bioavailability in Plasmodium vivax
Malaria but No Greater Increase in Intravascular
Hemolysis in Severe Disease
Bridget E. Barber,1,3 Timothy William,3,4 Matthew J. Grigg,1,3 Kim A. Piera,1 Youwei Chen,5,6 Hao Wang,1 J. Brice Weinberg,5,6 Tsin W. Yeo,1,3,7,8 and
Nicholas M. Anstey1,2,3
1
Global and Tropical Health Division, Menzies School of Health Research and Charles Darwin University, and 2Department of Infectious Diseases, Royal Darwin Hospital, Darwin, Australia; 3Infectious
Diseases Society Sabah–Menzies School of Health Research Clinical Research Unit, and 4Jesselton Medical Center, Kota Kinabalu, Malaysia; 5Duke University Medical Center, and 6VA Medical
Center, Durham, North Carolina; 7Lee Kong Chian School of Medicine, Nanyang Technological University, and 8Institute of Infectious Disease and Epidemiology, Tan Tock Seng Hospital, Singapore

Background. Pathogenesis of severe Plasmodium vivax malaria is poorly understood. Endothelial dysfunction and reduced
nitric oxide (NO) bioavailability characterize severe falciparum malaria, but have not been assessed in severe vivax malaria.
Methods. In patients with severe vivax malaria (n = 9), patients with nonsevere vivax malaria (n = 58), and healthy controls
(n = 79), we measured NO-dependent endothelial function by using reactive hyperemia–peripheral arterial tonometry (RH-PAT)
and assessed associations with arginine, asymmetric dimethylarginine (ADMA), and hemolysis.
Results. The L-arginine level and the L-arginine to ADMA ratio (a measure of L-arginine bioavailability) were reduced in pa-
tients with severe vivax malaria and those with nonsevere vivax malaria, compared with healthy controls (median L-arginine level,
65, 66, and 98 µmol/mL, respectively [P = .0001]; median L-arginine to ADMA ratio, 115, 125, and 187, respectively [P = .0001]).
Endothelial function was impaired in proportion to disease severity (median RH-PAT index, 1.49, 1.73, and 1.97 in patients with
severe vivax malaria, those with nonsevere vivax malaria, and healthy controls, respectively; P = .018) and was associated with the
L-arginine to ADMA ratio. While the posttreatment fall in hemoglobin level was greater in severe vivax malaria as compared to
nonsevere vivax malaria (2.5 vs 1 g/dL; P = .0001), markers of intravascular hemolysis were not higher in severe disease.
Conclusions. Endothelial function is impaired in nonsevere and severe vivax malaria, is associated with reduced L-arginine bio-
availability, and may contribute to microvascular pathogenesis. Severe disease appears to be more associated with extravascular he-
molysis than with intravascular hemolysis.
Keywords. malaria; Plasmodium vivax; arginine; asymmetric dimethylarginine; endothelial function; nitric oxide; hemolysis.

Plasmodium vivax is the most widespread malaria parasite, caus-
ing an estimated 13.8 million malaria cases per year [1, 2]. Al-
though previously considered benign, P. vivax is now recognized
as a cause of severe and fatal disease [3–8]. Despite this, little is
known about the pathogenesis of disease in severe vivax malaria.
In falciparum malaria, a central process in pathogenesis of se-
vere disease is cytoadherence of parasitized erythrocytes to ac-
tivated endothelium, leading to microvascular sequestration
and obstruction with consequent tissue hypoxia and organ dys-
function [9, 10]. Microvascular dysfunction is an additional pro-
cess associated with tissue hypoxia in severe falciparum malaria
Received 14 July 2016; accepted 6 September 2016; published online 13 September 2016.
Presented in part: 5th International Conference of Research on Plasmodium vivax malaria,
Bali, Indonesia, 19–22 May 2015.
Correspondence: B. E. Barber, Global Health Division, Menzies School of Health Research, PO
Box 41096, Casuarina 0811, Northern Territory, Australia (bridget.barber@menzies.edu.au).
The Journal of Infectious Diseases® 2016;214:1557–64
© The Author 2016. Published by Oxford University Press for the Infectious Diseases Society of
America. All rights reserved. For permissions, e-mail journals.permissions@oup.com.
DOI: 10.1093/infdis/jiw427
[11, 12], with impaired microvascular reactivity associated with
both impaired tissue perfusion and death [11]. Recent analyses
in falciparum malaria have shown that both microvascular se-
questration and endothelial activation are independent predic-
tors of mortality [13], suggesting that both processes contribute
to the pathogenesis of severe falciparum malaria. Reduced endo-
thelial nitric oxide (NO) bioavailability contributes to each of
these processes, with reduced NO levels shown to be associated
with increased parasite cytoadherence [14], increased endotheli-
al activation, and endothelial dysfunction [12, 15, 16]. Several
factors have been shown to contribute to the impairment of
NO bioavailability in falciparum malaria, including reduced ex-
pression of type 2 NO synthase in mononuclear cells [17], re-
duced concentrations of the NO precursor L-arginine [15, 18],
inhibition of NO synthase (NOS) by asymmetric dimethylargi-
nine (ADMA) [19–22], and reduced concentration of the NOS
cofactor tetrahydrobiopterin [23, 24]. Intravascular hemolysis
has also been shown to be increased in severe falciparum malar-
ia, with NO quenching by cell-free hemoglobin further limiting
endothelial NO bioavailability [25].

Endothelial Dysfunction in P. vivax Malaria • JID 2016:214 (15 November) • 1557

 Compared with falciparum malaria, coma in vivax malaria is
very rare [26]. However, other organ dysfunction and hypoten-
sion can occur, and severe disease in vivax malaria is well de-
scribed [7, 8, 27, 28]. Severe vivax malaria occurs at lower
levels of peripheral parasitemia than those usually seen in severe
falciparum malaria, and the pathogenesis of organ dysfunction
in vivax malaria is not well understood [3–8]. We have recently
shown evidence in vivax malaria of a hidden biomass underes-
timated by peripheral parasitemia, with the total P. vivax bio-
mass (as measured by parasite lactate dehydrogenase [ pLDH]
level) associated with systemic inflammation [27]. While there
is little evidence of sequestration of parasitized cells within en-
dothelium-lined vessels [5, 27], P. vivax–parasitized erythro-
cytes may accumulate in nonendothelial cell–lined vessels
and/or extravascular tissues, such as the spleen or bone marrow
[4, 27]. We have also shown that in vivax malaria endothelial
activation and impaired microvascular reactivity are associated
with impaired tissue perfusion and disease severity [27] and are
of comparable magnitude in severe vivax malaria and severe fal-
ciparum malaria. However, the roles of endothelial function and
microvascular NO bioavailability in the pathogenesis of vivax
malaria have not been assessed.
In this study we measured endothelial function in patients
with vivax malaria and assessed factors that have been shown
to correlate with reduced NO bioavailability in severe falcipa-
rum malaria—arginine, ADMA, and markers of hemolysis, in-
cluding cell-free hemoglobin (CFHb). Healthy controls were
included for comparison.
METHODS
Ethics Statement
The study was approved by the ethics committees of the Malay-
sian Ministry of Health (Kota Kinabalu) and the Menzies
School of Health Research (Darwin, Australia).
Study Site and Patients
Patients were enrolled as part of a prospective clinical and epi-
demiological study of all malaria patients admitted to Queen
Elizabeth Hospital, an adult tertiary care referral hospital in
Sabah, Malaysia [29]. Consecutive patients with polymerase
chain reaction (PCR)–confirmed vivax monoinfection were en-
rolled during September 2010–June 2013 if they were nonpreg-
nant, ≥12 years old, had no major comorbidities or concurrent
illness, were within 18 hours of commencing antimalarial treat-
ment, and had not been previously enrolled in the study. Clin-
ical details of these patients have been previously reported [27,
29]. Severe malaria was defined as the presence of ≥1 of the fol-
lowing: unrousable coma (Glasgow coma scale score, <11), mul-
tiple (>2) convulsions, respiratory distress (respiratory rate of
>30 breaths/minute and oxygen saturation of <94%), hypoten-
sion (systolic blood pressure, ≤80 mm Hg), jaundice (bilirubin
level of >43 µmol/L plus parasitemia level of >20 000 parasites/µL
1558 • JID 2016:214 (15 November) • Barber et al
and/or creatinine level of >132 µmol/L), significant abnormal
bleeding, hypoglycemia (blood glucose level, <2.2 mmol/L), met-
abolic acidosis (bicarbonate level of <15 mmol/L or lactate level
of >4 mmol/L), acute kidney injury (AKI; creatinine level, >265
µmol/L), and a hemoglobin level of <7.5 g/dL. Healthy controls
were visitors or relatives of patients with malaria who had no his-
tory of fever in the past 48 hours and a blood film negative for
malaria parasites.
Standardized history and physical examination findings were
documented. Hematologic and biochemical findings, acid-base
parameters, and lactate level (obtained by bedside blood analy-
sis; iSTAT system) were obtained on enrollment. Parasite counts
were determined by microscopy, and parasite species were iden-
tified by PCR [30, 31]. Patients were treated according to hospi-
tal guidelines, as previously described [29].
Laboratory Assays
Venous blood collected in tubes coated with lithium heparin
was centrifuged within 30 minutes of collection, and plasma
was stored at −70°C. ADMA and arginine levels were measured
using reverse-phase high-performance liquid chromatography
with simultaneous fluorescence and UV-visible absorbance de-
tection, as previously described [32]. Plasma concentrations of
the endothelial activation markers intracerebral adhesion mol-
ecule-1 (ICAM-1) and angiopoietin-2 were measured using an
enzyme-linked linked immunosorbent assay (ELISA; R&D Sys-
tems), and interleukin 6 (IL-6) and interleukin 10 (IL-10) levels
were measured by flow cytometry (BD Cytometric Bead Array).
Plasma haptoglobin and CFHb levels were measured by ELISA
according to the manufacturers’ (ICL Laboratories and Bethyl
Laboratories, respectively) instructions. Red blood cell arginase
is released during intravascular hemolysis, and plasma arginase
activity was measured as an additional measure of hemolysis by
using a radiometric assay, as described elsewhere [33]; findings
are reported in micromoles per milliliter per hour. P. vivax bio-
mass was estimated on the basis of the pLDH level, measured by
ELISA [34].
Measurement of Endothelial Function
Endothelial function was measured noninvasively, using pe-
ripheral arterial tonometry, by the change in digital pulse
wave amplitude in response to reactive hyperemia, giving a re-
active hyperemia peripheral arterial tonometry (RH-PAT)
index as previously described [15]. The RH-PAT index is at
least 50% dependent on endothelial NO production [35] and
has been shown to be L-arginine responsive in falciparum ma-
laria [15]. Endothelial function was measured on enrollment
and repeated on day 3 in patients who remained hospitalized.
Measurement of endothelial function was discontinued in pa-
tients with nonsevere malaria, in May 2011.
Statistical Analysis
Statistical analysis was performed with STATA software (ver-
sion 13.0). For continuous variables, intergroup differences
were compared using analysis of variance or Kruskal–Wallis
tests, depending on the distribution of data. Student t tests or
Mann–Whitney tests were used for pair-wise comparisons. Cat-
egorical variables were compared using χ2 or Fisher exact tests.
Associations between continuous variables were assessed initially
using pair-wise correlation to control for disease severity, with
data log transformed to exhibit a normal distribution. The Spear-
man correlation coefficient was used to assess for association be-
tween variables within the severity groups. Multiple linear
regression was used to adjust for confounding variables. The Wil-
coxon signed rank test was used to compare day 0 and day 3 data.
RESULTS
Patients
A total of 67 patients with vivax malaria were enrolled, includ-
ing 58 with nonsevere and 9 with severe disease, in addition to
79 healthy controls. The epidemiological, clinical, and bio-
chemical features, including measures of parasite biomass,
have been previously reported [27, 29]. Baseline characteristics
are shown in Tables 1 and 2. Among the 9 patients with severe
vivax malaria, severity criteria included hypotension (n = 6), re-
spiratory distress (n = 1), jaundice (n = 2), metabolic acidosis
(n = 1), abnormal bleeding (n = 1), and multiple convulsions
(n = 1). Three patients had 2 severity criteria, and 6 had 1 cri-
terion. Cultures of blood specimens obtained before antibiotic
treatment yielded negative results for 4 patients, were positive
for Streptococcus pneumoniae for 1 [29], and were not done
for 4. No deaths occurred. As previously reported [27], the par-
asite count and pLDH level, a marker of P. vivax biomass, were
higher in patients with severe vivax malaria as compared to
those with nonsevere vivax malaria.
Levels of Plasma Arginine and ADMA and Ratio of Arginine to ADMA
Plasma arginine levels were reduced in the severe and nonsevere
vivax malaria groups as compared to the control group (median
level, 65, 66, and 98 µmol/mL, respectively; P = .0001; Table 1
and Figure 1), although there was no difference between malaria
severity groups. There was no difference in plasma ADMA lev-
els on enrollment between patients with vivax malaria and
healthy controls. The arginine to ADMA ratio (reflecting argi-
nine bioavailability) was lower in patients with severe or nonse-
vere vivax malaria, compared with controls (median ratio, 115,
125, and 187, respectively; P = .0001; Figure 1).
The arginine level was strongly correlated with the ADMA
level in all patients with vivax malaria (r = 0.55, P < .0001),
with this relationship remaining significant after adjustment
for disease severity (Table 3). The ADMA level was inversely
correlated with the IL-10 level in all patients with vivax malaria,
after adjustment for disease severity (r = −0.37, P = .006). The
ADMA level was positively associated with the ICAM-1 level
(r = 0.48, P = .0001) but not with the angiopoietin-2 level in
all patients with vivax malaria, after adjustment for disease se-
verity, and in both severity groups. In a multivariate model in-
cluding disease severity, ICAM-1, IL-6, IL-10, and arginine
levels, ADMA levels remained significantly associated with
the arginine level and inversely associated with the IL-10
level. There was no association between ADMA and CFHb lev-
els in patients with vivax malaria.
Endothelial and Microvascular Function
Endothelial function, as measured by the RH-PAT index, was
lower in patients with severe or nonsevere vivax malaria, com-
pared with healthy controls (median RH-PAT index, 1.49, 1.73,
and 1.97, respectively; P = .018; Figure 1). Endothelial function

Table 1. Baseline Characteristics of Patients With Plasmodium vivax Malaria and Healthy Controls

Patients With Vivax Malaria

Characteristic Healthy Controls (n = 79) Nonsevere (n = 58) Severe (n = 9) P Value

Age, y
Median (IQR) 35 (23–44) 24 (18–40) 39 (31–53) .171
Range 14–69 13–61 14–79
Male sex, no. (%) 57 (72) 37 (73) 9 (100) .075
Weight, kg 58 ± 10 56 ± 12 58 ± 8 .584
Current smoker, no. (%) 33 (42) 24 (47) 1 (11) .045
Fever duration, d, median (IQR) 5 (3–7) 4 (3–4) .135
Enrollment systolic blood pressure, mm Hg 123 ± 15 114 ± 16 119 ± 18 .400
Enrollment mean arterial blood pressure, mm Hg 109 ± 13 99 ± 13 104 ± 16 .573
Minimum systolic blood pressure, mm Hg NA 100 ± 9 85 ± 15 <.0001
Pulse rate, beats/min 71 ± 12 88 ± 20 97 ± 14 .197
Respiratory rate, breaths/min 20 ± 3 24 ± 5 26 ± 4 .210
Temperature, °C 36.5 ± 0.4 37.5 ± 1.2 37.0 ± 0.5 .220
Time from malaria treatment to enrollment, h, median (IQR)a NA 4.5 (0–11) 9.7 (6–15) .117

Data are mean ± SD, unless otherwise indicated, and are from all subjects who had endothelial function measured and/or blood analysis performed.
Abbreviations: IQR, interquartile range; NA, not applicable.
a
A total of 14 of 58 patients with nonsevere vivax malaria and 1 of 9 with severe vivax malaria were enrolled prior to commencing antimalarial treatment.

Endothelial Dysfunction in P. vivax Malaria • JID 2016:214 (15 November) • 1559

Table 2. Laboratory and Physiological Measurements Among Patients With Plasmodium vivax Malaria and Healthy Controls

Patients With Vivax Malaria P Value

Malaria
Characteristic Healthy Controls Nonsevere (n = 58) Severe (n = 9) All Groups Groups

Parasite count, parasites/µL . . . 4055 (1599–10 165) 10 243 (4387–19 520) .020
Parasite LDH level,a ng/mL . . . 44.6 (7.8–143) 307 (258–619) .016
(n = 53)
Hemoglobin level, g/dL, mean ± SD
Overall . . . 12.2 (2.04) 13.9 (1.22) .015
Nadir . . . 11.2 (1.98) 11.4 (0.8) .847
Absolute decrease . . . 1 (0.4–1.5) 2.5 (1.7–3.4) .0001
CFHb level, µM 15 146 (9641–25 256) 32 498 (20 068–45 189) 31 338 (9889–35 824) .0001 .161
(n = 50) (n = 48)
Plasma LDH level, µL 213 (174–290) 311 (254–420) 295 (272–344) .0001 .550
(n = 50) (n = 57)
Haptoglobin levela, g/L 1.44 (1.01–1.72) 0.27 (0.07–1.09) 0.28 (0.035–0.530) .0001 .846
(n = 60) (n = 53)
Plasma arginase clearance rate, µM/mL/h 0.115 (0.078–0.192) 0.121 (0.074–0.168) 0.073 (0.039–0.094) .069 .035
(n = 48) (n = 51) (n = 8)
Creatinine level, μmol/L . . . 82 (67–99) 128 (94–136) .011
(n = 57)
Bilirubin level, μmol/L . . . 17 (11.6–25.1) 25.5 (13–41) .132
(n = 55)
Lactate level, mmol/L . . . 1.14 (0.78–1.47) 1.28 (1–2.23) .162
(n = 51)
Angiopoietin-2 level, pg/mL 1183 (875–1597) 4557 (3463–6197) 8857 (6547–9743) .0001 .001
(n = 50) (n = 53)
ICAM-1 level, pg/mL 149 (123–167) 505 (416–639) 686 (682–832) .0001 .077
(n = 50) (n = 53)
IL-6 level, pg/mL Below detection limit in 27/30 patients 27.8 (9.1–93.6) 49.5 (36.1–54.0) .0001 .369
(n = 46)
IL-10 level, pg/mL Below detection limit in 29/30 patients 185.5 (41.3–504.1) 173 (107–319) .0001 .733
(n = 46)
L-arginine level, µmol/mL 98.4 (80.1–112.8) 67.4 (54–82.5) 65.1 (55.6–71) .0001 .520
(n = 51)
ADMA level, µM 0.540 (0.490–0.595) 0.553 (0.446–0.642) 0.533 (0.504–0.541) .792 .526
(n = 51)
L-arginine:ADMA ratio 187 (153–213) 125 (109–142) 115 (103–127) .0001 .270
Endothelial functionb 1.97 (1.64–2.27) 1.73 (1.46–2.05) 1.49 (1.37–1.88) .018 .237
(n = 79) (n = 30) (n = 8)

Data are median value (interquartile range), unless otherwise indicated. Investigations are performed on enrollment, unless otherwise stated. Where a result is below the detection limit, half the
lower limit of detection is substituted.
Abbreviations: ADMA, asymmetric dimethylarginine; CFHb, cell-free hemoglobin; ICAM-1, intracerebral adhesion molecule-1; IL-6, interleukin 6; IL-10, interleukin 10; LDH, lactate
dehydrogenase.
a
Levels were below the detection limit in 1 of 60 controls, 11 of 53 patients with nonsevere vivax malaria, and 3 of 9 patients with severe vivax malaria.
b
Determined on the basis of the reactive hyperemia peripheral arterial tonometry index.

was inversely correlated with the arginine level and the arginine
to ADMA ratio in all patients with vivax malaria, after adjust-
ment for disease severity (Table 3), with the association with the
arginine to ADMA ratio remaining significant on multivariate
analysis. There was no correlation between endothelial function
and levels of endothelial activation markers or lactate.
Measures of Intravascular Hemolysis
CFHb and plasma LDH levels were higher and the haptoglobin
level lower in patients with severe or nonsevere vivax malaria,
compared with controls (Table 2). However, there was no evi-
dence that intravascular hemolysis increased with the severity of
vivax malaria, with no difference in CFHb, haptoglobin, or
LDH levels between severity groups. There was no difference
in levels of plasma arginase (released from hemolyzed red
blood cells) between healthy controls (0.115 µM/mL/hour)
and patients with nonsevere vivax malaria (0.121 µM/mL/
hour), and levels were unexpectedly lower in patients with se-
vere disease (0.073 µM/mL/hour), compared with healthy con-
trols (P = .026) and patients with nonsevere disease (P = .035).
Arginase can also be released from alternatively activated
(M2) monocytes or macrophages [36]. The M2 phenotype has
previously been shown to be associated with IL-10 [36], and in
the current study arginase level was correlated with the IL-10
level in patients with nonsevere (r = 0.29, P = .05) but not in
those with severe (r = −0.35, P = ns) vivax malaria. The

1560 • JID 2016:214 (15 November) • Barber et al

Figure 1. Plasma concentrations of arginine and asymmetric dimethylarginine
(ADMA), the arginine to ADMA ratio (a measure of arginine bioavailability), and en-
dothelial function (determined on the basis of the reactive hyperemia peripheral ar-
terial tonometry [RH-PAT] index) in patients with severe Plasmodium vivax malaria,
those with nonsevere P. vivax malaria, and healthy controls.
Figure 2. Plasma concentrations of arginine and asymmetric dimethylarginine
(ADMA), the arginine to ADMA ratio (a measure of arginine bioavailability), and
endothelial function (determined on the basis of the reactive hyperemia peripheral
arterial tonometry [RH-PAT] index) in patients with severe Plasmodium vivax
malaria.

Endothelial Dysfunction in P. vivax Malaria • JID 2016:214 (15 November) • 1561

Table 3. Correlations With Endothelial Function and Asymmetric Table 4. Longitudinal Results in Patients With Severe or Nonsevere
Dimethylarginine (ADMA)–Associated Values in Patients With Nonsevere Plasmodium vivax Malaria
or Severe Plasmodium vivax Malaria
Vivax Malaria
Patients With Vivax Malaria, by Severity Severity, Factor Day 0 Days 2–4 P Value

Severe
Overalla Nonsevere Severe
Arginine level, 65 (60–68) 127 (122–128) .043
P P P µmol/L
Correlation Rb Value Rb Value Rb Value (n = 5)
c
ADMA level, 0.506 (0.504–0.533) 0.653 (0.576–0.823) .043
Between RH-PAT index and µM (n = 5)
Arginine level 0.407 .011 0.438 .015 0.017 .968 Arginine: 118 (105–127) 156 (147–186) .043
ADMA level − 0.106 .527 ADMA ratio
Arginine: 0.505 .001 0.568 .001 −0.217 .606 RH-PAT index 1.44 (1.31–1.54) 2.04 (1.76–2.24) .046
ADMA ratio (n = 6)
CFHb level 0.396 .017 0.435 .018 0.237 .572 CFHb level, µM 35 434 (18 672–40 986) 23 656 (17 281–36 700) .128
(n = 7)
Angiopoietin-2 −0.203 .221
level Nonsevere
ICAM-1 level −0.004 .979 Arginine level, 76 (61–94) 127 (122–128) .026
µmol/L
Between ADMA leveld and
(n = 16)
Arginine level 0.551 <.0001 0.555 <.0001 0.489 .182
ADMA level, 0.610 (0.529–0.664) 0.717 (0.638–0.864) .0009
CFHb level 0.048 .722 µM (n = 16)
IL-6 level −0.234 .085 −0.267 .073 0.149 .702 Arginine: 131 (110–168) 165 (135–194) .179
IL-10 level −0.365 .006 −0.421 .004 0.194 .617 ADMA ratio
ICAM-1 level 0.479 .0001 0.430 .001 0.853 .004 RH-PAT index 1.84 (1.50–2.22) 1.95 (1.67–2.00) .975
(n = 14)
Angiopoietin-2 0.204 .112
level CFHb level, µM 36 130 (33 551–57 546) 27 918 (19 960–46 522) .092
(n = 24)
Abbreviations: ADMA, asymmetric dimethylarginine; CFHb, cell-free hemoglobin; ICAM-1,
intracerebral adhesion molecule-1; IL-6, interleukin 6; IL-10, interleukin 10; RH-PAT, Data are median value (interquartile range).
reactive hyperemia peripheral arterial tonometry index. Abbreviations: ADMA, asymmetric dimethylarginine; CFHb, cell-free hemoglobin; RH-PAT,
a
Bivariate model adjusting for disease severity. reactive hyperemia peripheral arterial tonometry.
b
Pair-wise correlation coefficient, with data for all variables log transformed to normality.
c
The reactive hyperemia peripheral arterial tonometry (RH-PAT) index was determined as a
measure of endothelial function on 30 patients with nonsevere vivax malaria and 8 patients
with severe vivax malaria. The index remained significantly correlated with the arginine to
ADMA ratio (P = .019) in a multivariate model including disease severity and CFHb. Figure 2). In both patient groups, ADMA levels also increased,
d
The ADMA level was measured in 58 patients with nonsevere vivax malaria and 9 patients with day 3 levels being above those of healthy controls. Despite
with severe vivax malaria. The level remained significantly associated with the arginine level
(P = .049) and inversely associated with the IL-10 level (P = .005) in a multivariate model the increase in ADMA levels, the arginine to ADMA ratio in-
including disease severity, IL-6 level, and ICAM-1 level. creased in all patient groups from day 0 to days 2–4.
In patients with vivax malaria who had the RH-PAT index
hemoglobin level on admission was higher in patients with severe measured on day 3, median endothelial function improved, sig-
vivax malaria as compared to those with nonsevere vivax malaria nificantly so in patients with severe malaria (from 1.49 to 2.04
(13.9 vs 12.2 g/dL, P = .015), but the fall in hemoglobin level to [n = 6]; P = .046) but not in those with nonsevere malaria (from
the nadir level was greater in severe as compared to nonsevere 1.73 to 1.95 [n = 14]; P = .975). In all patients with vivax malar-
disease (2.5 vs 1 g/dL; P = .0001). The CFHb level fell following ia, the improvement in endothelial function from baseline to
treatment in patients with severe or nonsevere vivax malaria, day 3 was correlated with an increase in the arginine to
with the decrease similar between both groups (Table 4). ADMA ratio over the same period (r = 0.64, P = .035 [n = 11]).
The median CFHb level was higher in the 3 patients with se-
DISCUSSION
vere vivax malaria without hypotension, compared with those
with hypotension (35 824 µM and 14 372 µM, respectively); In this study, we show that in vivax malaria arginine bioavail-
however, this was not statistically significant, and there was ability (as measured by the arginine to ADMA ratio) is reduced
no correlation between CFHb level and mean arterial pressure and is associated with NO-dependent endothelial dysfunction.
in patients with vivax malaria. There was also no correlation be- The degree of impairment of endothelial function in severe
tween the CFHb level and levels of creatinine, lactate, or mark- vivax malaria is comparable to that in patients with severe fal-
ers of endothelial activation (ICAM-1 and angiopoietin-2). ciparum malaria from the same study cohort (median RHPAT
index, 1.58 [IQR, 1.26–1.96] [21]. As has been demonstrated in
Longitudinal Course of Arginine Level, ADMA Level, Arginine to ADMA falciparum malaria, endothelial function improved with recov-
Ratio, and Endothelial Function ery from vivax malaria, with improvement in endothelial func-
Arginine levels increased significantly from baseline to days 2–4 tion correlating with an increase in the arginine to ADMA ratio.
in patients with severe or nonsevere vivax malaria (Table 4 and Taken together, these results suggest that impaired L-arginine

1562 • JID 2016:214 (15 November) • Barber et al

bioavailability contributes to endothelial dysfunction in vivax
malaria.
In severe falciparum malaria, reduced endothelial NO bioa-
vailability is partly accounted for by a reduced concentration of
the NO precursor L-arginine [15]. In this study, the concentra-
tion of L-arginine was also low in patients with vivax malaria;
however, arginine levels were similar between patients with se-
vere disease and those with nonsevere disease. This is consistent
with previous studies of Indonesian [15] and Malaysian [21]
adults with falciparum malaria, in which arginine levels were
also similar between patients with severe or nonsevere falcipa-
rum malaria. Nevertheless, arginine bioavailability (measured
as the arginine to ADMA ratio) was lowest in severe falciparum
malaria [19, 21] and thought contributory to the greater endo-
thelial dysfunction found in those with severe disease [15].
In this study, although ADMA levels did not differ between
patients with vivax malaria and healthy controls, the reduction
in arginine bioavailability (ie, the arginine to ADMA ratio) in
both uncomplicated and severe vivax malaria suggests that,
as with falciparum malaria, ADMA levels are inappropriately
high. Also, consistent with previous studies in falciparum ma-
laria [21, 22], the ADMA level was strongly correlated with
the arginine level. This association has been demonstrated
in other acute inflammatory conditions [34, 37] and may
reflect arginine-mediated inhibition of dimethylarginine-
dimethylaminohydrolase [38] or competitive transport through
cationic amino acid transporters [39]. As with falciparum
malaria [21], in vivax malaria the ADMA level was inversely
correlated with the IL-10 level, suggesting a role of inflamma-
tory cytokines in ADMA regulation and consequent NO
bioavailability.
In severe falciparum, reduced NO bioavailability also occurs
as a result of increased intravascular hemolysis, leading to NO
quenching from CFHb, degradation of arginine by erythrocyte-
derived arginase, and inhibition of NOS by erythrocyte-derived
ADMA [10]. In contrast, in the current study we did not find
evidence that intravascular hemolysis was increased in severe
as compared to nonsevere vivax malaria or that it was associated
with the endothelial dysfunction in vivax malaria. Measures of
intravascular hemolysis were not increased in severe compared
to nonsevere vivax malaria, despite the higher circulating para-
sitemia level and higher total parasite biomass seen in patients
with severe relative to those with nonsevere vivax malaria. In-
deed, the level of plasma arginase, released from red cells during
intravascular hemolysis, was lower in patients with severe ma-
laria, compared with that in patients with nonsevere vivax ma-
laria. Importantly, however, the fall in hemoglobin level was
significantly greater during severe as compared to nonsevere
vivax malaria, despite a posttreatment reduction in CFHb
level in both groups. Taken together, these findings suggest
that in severe vivax malaria there is a greater proportion of ex-
travascular infected red cell loss, compared with nonsevere
Endothelial Dysfunction in P. vivax Malaria
disease [40]. This notion is also consistent with our previous
findings suggesting a greater accumulation of infected red
blood cells in non–endothelial-lined and/or extravascular tissues
in severe vivax malaria than in nonsevere vivax malaria, possibly
in the spleen and bone marrow [27]. This is also further support-
ed by the recent finding that P. vivax merozoites preferentially
infect immature reticulocytes [41]. The lack of association of he-
molysis with disease severity also suggests that heme-mediated
toxicity may not be an important contributor to mechanisms
of severe disease in vivax malaria.
In vivax malaria, hypotension occurs more commonly than
in falciparum malaria [3, 28] and, in the current study, occurred
in 6 of 9 patients. It is possible that this increased frequency of
hypotension is due in part to a lesser degree of intravascular he-
molysis and vascular NO quenching in severe vivax malaria
than in severe falciparum malaria, potentially allowing greater
vasodilation. Although we did not detect an association between
the CFHb level and enrollment mean arterial pressure (MAP) in
this study, the level of CFHb was nonsignificantly higher in the
3 patients without shock as compared to the 6 with shock. Larg-
er series will be required to further evaluate the association be-
tween the CFHb level and MAP/systemic vascular resistance in
severe vivax malaria.
A limitation of this study was the small number of patients
with severe vivax malaria, which limited our ability to assess
for associations with disease severity and other variables. In ad-
dition, a proportion of patients were enrolled following com-
mencement of antimalarial treatment, which may have
affected baseline measurements of ADMA levels, arginine lev-
els, and/or endothelial function.
In conclusion, our study demonstrated that, as with falcipa-
rum malaria, NO-dependent endothelial function is impaired
in vivax malaria, is related to arginine bioavailability, and im-
proves following commencement of antimalarial treatment. In
the absence of significant endothelial cytoadherence of
P. vivax–infected red blood cells, these processes may contrib-
ute to impaired microvascular perfusion and disease pathogen-
esis. In contrast to falciparum malaria, however, intravascular
hemolysis was not increased in severe disease, suggesting that
this is not a significant cause of impaired NO bioavailability
in severe vivax malaria. Furthermore, the greater fall in hemo-
globin level during severe disease in the absence of greater intra-
vascular hemolysis suggests that extravascular hemolysis may be
a more important contributor to loss of infected and uninfected
red cells in severe vivax malaria.
Notes
Acknowledgments. We thank all the patients enrolled in the prospective
study at Queen Elizabeth Hospital and the clinical staff involved in their
care; Uma Paramaswaran, Rita Wong, Beatrice Wong, Ann Wee, and
Kelly Nestor, for assistance with clinical and laboratory study procedures;
Sarah Auburn and Jutta Marfurt, for supervising the PCR assays; the Clin-
ical Research Centre, Sabah, for logistical support; and the Director General
of Health, Malaysia, for permission to publish this study.
• JID 2016:214 (15 November) • 1563
 Financial support. This work was supported by the Australian National
Health and Medical Research Council (Program Grants 496600 and
1037304, project grant 1045156, and fellowships to N. M. A., T. W. Y.,
and B. E. B. and scholarship to M. J. G.).
Potential conflicts of interest. All authors: No reported conflicts. All
authors have submitted the ICMJE Form for Disclosure of Potential Con-
flicts of Interest. Conflicts that the editors consider relevant to the content
of the manuscript have been disclosed.
References
1. Price RN, Douglas NM, Anstey NM. New developments in Plasmodium vivax ma-
laria: severe disease and the rise of chloroquine resistance. Curr Opin Infect Dis
2009; 22:430–5.
2. World Health Organization. World Malaria Report 2015. Geneva: World Health
Organizatioin, 2015.
3. Anstey NM, Douglas NM, Poespoprodjo JR, Price R. Plasmodium vivax: clinical
spectrum, risk factors and pathogenesis. Adv Parasitol 2012; 80:151–201.
4. Baird JK. Evidence and implications of mortality associated with acute Plasmodi-
um vivax malaria. Clin Microbiol Rev 2013; 26:36–57.
5. Lacerda MVG, Fragoso SCP, Alecrim MGC, et al. Postmortem characterization of
patients with clinical diagnosis of Plasmodium vivax malaria: to what extent does
this parasite kill? Clin Infect Dis 2012; 55:e67–74.
6. Tjitra E, Anstey NM, Sugiarto P, et al. Multidrug-resistant Plasmodium vivax as-
sociated with severe and fatal malaria: a prospective study in Papua, Indonesia.
PLoS Med 2008; 5:e128.
7. Kochar DK, Das A, Kochar SK, et al. Severe Plasmodium vivax malaria: a report on
serial cases from Bikaner in northwestern India. Am J Trop Med Hyg 2009;
80:194–8.
8. Lança EFC, Magalhães BML, Vitor-Silva S, et al. Risk factors and characterization
of Plasmodium vivax-associated admissions to pediatric intensive care units in the
Brazilian Amazon. PLoS One 2012; 7:e35406.
9. Turner GDH, Morrison H, Jones M, et al. An immunohistochemical study of the
pathology of fatal malaria: evidence for widespread endothelial activation and a
potential role for intercellular adhesion molecule-1 in cerebral sequestration.
Am J Pathol 1994; 145:1057.
10. Miller LH, Ackerman HC, Su X-z, Wellems TE. Malaria biology and disease path-
ogenesis: insights for new treatments. Nat Med 2013; 19:156–67.
11. Yeo TW, Lampah DA, Kenangalem E, Tjitra E, Price RN, Anstey NM. Impaired
skeletal muscle microvascular function and increased skeletal muscle oxygen con-
sumption in severe falciparum malaria. J Infect Dis 2013; 207:528–36.
12. Yeo TW, Lampah DA, Kenangalem E, et al. Decreased endothelial nitric oxide bi-
oavailability, impaired microvascular function, and increased tissue oxygen con-
sumption in children with falciparum malaria. J Infect Dis 2014; 210:1627–32.
13. Hanson J, Lee SJ, Hossain MA, et al. Microvascular obstruction and endothelial
activation are independently associated with the clinical manifestations of severe
falciparum malaria in adults: an observational study. BMC Med 2015; 13:122.
14. Serirom S, Raharjo WH, Chotivanich K, Loareesuwan S, Kubes P, Ho M. Anti-ad-
hesive effect of nitric oxide on Plasmodium falciparum cytoadherence under flow.
Am J Pathol 2003; 162:1651–60.
15. Yeo TW, Lampah DA, Gitawati R, et al. Impaired nitric oxide bioavailability and
L-arginine–reversible endothelial dysfunction in adults with falciparum malaria.
J Exp Med 2007; 204:2693–704.
16. Yeo TW, Lampah DA, Gitawati R, et al. Angiopoietin-2 is associated with de-
creased endothelial nitric oxide and poor clinical outcome in severe falciparum
malaria. Proc Natl Acad Sci U S A 2008; 105:17097–102.
17. Anstey N, Weinberg J, Hassanali M, et al. Nitric oxide in Tanzanian children with
malaria: inverse relationship between malaria severity and nitric oxide production/
nitric oxide synthase type 2 expression. J Exp Med 1996; 184:557–67.
18. Lopansri BK, Anstey NM, Weinberg JB, et al. Low plasma arginine concentrations
in children with cerebral malaria and decreased nitric oxide production. Lancet
2003; 361:676–8.
19. Yeo T, Lampah D, Tjitra E, et al. Increased asymmetric dimethylarginine in severe
falciparum malaria: association with impaired nitric oxide bioavailability and fatal
outcome. PLoS Pathog 2010; 6:214–7.
1564 • JID 2016:214 (15 November) • Barber et al
20. Weinberg JB, Yeo TW, Mukemba JP, et al. Dimethylarginines: endogenous inhib-
itors of nitric oxide synthesis in children with falciparum malaria. J Infect Dis
2014; 210:913–22.
21. Barber BE, William T, Grigg MJ, et al. Asymmetric Dimethylarginine (ADMA) in
adult falciparum malaria: relationships with disease severity, antimalarial treat-
ment, haemolysis and inflammation. Open Forum Infect Dis 2016; 3:ofw027.
22. Chertow JH, Alkaitis MS, Nardone G, et al. Plasmodium Infection Is Associated
with Impaired Hepatic Dimethylarginine Dimethylaminohydrolase Activity and
Disruption of Nitric Oxide Synthase Inhibitor/Substrate Homeostasis. PLoS
Pathog 2015; 11:e1005119.
23. Yeo TW, Lampah DA, Kenangalem E, et al. Impaired systemic tetrahydrobiopterin
bioavailability and increased dihydrobiopterin in adult falciparum malaria: asso-
ciation with disease severity, impaired microvascular function and increased endo-
thelial activation. PLoS Pathog 2015; 11:e1004667-e.
24. Rubach MP, Mukemba J, Florence S, et al. Impaired systemic tetrahydrobiopterin
bioavailability and increased oxidized biopterins in pediatric falciparum malaria:
association with disease severity. PLoS Pathog 2015; 11:e1004655.
25. Yeo TW, Lampah DA, Tjitra E, et al. Relationship of cell-free haemoglobin to im-
paired nitric oxide bioavailability and perfusion in severe falciparum malaria. J In-
fect Dis 2009; 200:1522–9.
26. Lampah D, Yeo TW, Hardianto SO, et al. Coma associated with microscopy-diag-
nosed Plasmodium vivax: a prospective study in Papua, Indonesia. PLoS Negl Trop
Dis 2011; 5:e1032.
27. Barber BE, William T, Grigg MJ, et al. Parasite biomass-related inflammation, en-
dothelial activation, microvascular dysfunction and disease severity in vivax ma-
laria. PLoS Pathog 2015; 11:e1004558.
28. Andrade BB, Reis-Filho A, Souza-Neto SM, et al. Severe Plasmodium vivax malar-
ia exhibits marked inflammatory imbalance. Malar J 2010; 9:13.
29. Barber BE, William T, Grigg MJ, et al. A prospective comparative study of knowl-
esi, falciparum and vivax malaria in Sabah, Malaysia: high proportion with severe
disease from Plasmodium knowlesi and P. vivax but no mortality with early referral
and artesunate therapy. Clin Infect Dis 2013; 56:383–97.
30. Padley D, Moody A, Chiodini P, Saldanha J. Use of a rapid, single-round, multi-
plex PCR to detect malarial parasites and identify the species present. Ann Trop
Med Parasitol 2003; 97:131–7.
31. Imwong M, Tanomsing N, Pukrittayakamee S, Day NPJ, White NJ, Snounou G.
Spurious amplification of a Plasmodium vivax small-subunit RNA gene by use of
primers currently used to detect P. knowlesi. J Clin Microbiol 2009; 47:4173.
32. Jones CE, Darcy CJ, Woodberry T, Anstey NM, McNeil YR. HPLC analysis of
asymmetric dimethylarginine, symmetric dimethylarginine, homoarginine and
arginine in small plasma volumes using a Gemini-NX column at high pH. J Chro-
matogr B Analyt Technol Biomed Life Sci 2010; 878:8–12.
33. Morris CR, Kato GJ, Poljakovic M, et al. Dysregulated arginine metabolism, hemo-
lysis-associated pulmonary hypertension, and mortality in sickle cell disease.
JAMA 2005; 294:81–90.
34. Nijveldt R, Teerlink T, Van Der Hoven B, et al. Asymmetrical dimethylarginine
(ADMA) in critically ill patients: high plasma ADMA concentration is an inde-
pendent risk factor of ICU mortality. Clin Nutr 2003; 22:23–30.
35. Nohria A, Gerhard-Herman M, Creager M, Hurley S, Mitra D, Ganz P. Role of
nitric oxide in the regulation of digital pulse volume amplitude in humans.
J Appl Physiol 2006; 101:545.
36. Weinberg JB, Volkheimer AD, Rubach MP, et al. Monocyte polarization in chil-
dren with falciparum malaria: relationship to nitric oxide insufficiency and disease
severity. Sci Rep 2016; 6:29151.
37. Iapichino G, Umbrello M, Albicini M, et al. Time course of endogenous nitric
oxide inhibitors in severe sepsis in humans. Minerva Anestesiol 2010; 76:325–33.
38. Wang J, Sim A, Wang X, Wilcken D. L-arginine regulates asymmetric dimethylar-
ginine metabolism by inhibiting dimethylarginine dimethylaminohydrolase activ-
ity in hepatic (HepG2) cells. Cell Mol Life Sci 2006; 63:2838–46.
39. Teerlink T, Luo Z, Palm F, Wilcox CS. Cellular ADMA: regulation and action.
Pharmacol Res 2009; 60:448–60.
40. Douglas N, Anstey N, Buffet P, et al. The anaemia of Plasmodium vivax malaria.
Malar J 2012; 11:135.
41. Malleret B, Li A, Zhang R, et al. Plasmodium vivax: restricted tropism and rapid
remodeling of CD71-positive reticulocytes. Blood 2015; 125:1314–24.