Open Collections

UBC Faculty Research and Publications

Cell-free hemoglobin mediated oxidative stress is associated with acute kidney injury and renal replacement… Plewes, Katherine; Kingston, Hugh W; Ghose, Aniruddha; Maude, Richard J; Herdman, M. T; Leopold, Stije J; Ishioka, Haruhiko; Hasan, Md. M U; Haider, Md. S; Alam, Shamsul; Piera, Kim A; Charunwatthana, Prakaykaew; Silamut, Kamolrat; Yeo, Tsin W; Faiz, Md. A; Lee, Sue J; Mukaka, Mavuto; Turner, Gareth D; Anstey, Nicholas M; Jackson Roberts, L.; White, Nicholas J; Day, Nicholas P; Hossain, Md. A; Dondorp, Arjen M Apr 27, 2017

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12879_2017_Article_2373.pdf [ 702.9kB ]
JSON: 52383-1.0366869.json
JSON-LD: 52383-1.0366869-ld.json
RDF/XML (Pretty): 52383-1.0366869-rdf.xml
RDF/JSON: 52383-1.0366869-rdf.json
Turtle: 52383-1.0366869-turtle.txt
N-Triples: 52383-1.0366869-rdf-ntriples.txt
Original Record: 52383-1.0366869-source.json
Full Text

Full Text

RESEARCH ARTICLE Open AccessCell-free hemoglobin mediated oxidativestress is associated with acute kidney injuryand renal replacement therapy in severefalciparum malaria: an observational studyKatherine Plewes1,2,3, Hugh W.F. Kingston1,2,4, Aniruddha Ghose5, Richard J. Maude1,2, M. Trent Herdman1,Stije J. Leopold1,2, Haruhiko Ishioka1, Md. Mahtab Uddin Hasan5, Md. Shafiul Haider6, Shamsul Alam7, Kim A. Piera4,Prakaykaew Charunwatthana1, Kamolrat Silamut1, Tsin W. Yeo4,8, Md. Abul Faiz9, Sue J Lee1,2, Mavuto Mukaka1,2,Gareth D.H. Turner1,2, Nicholas M. Anstey4, L. Jackson Roberts II10, Nicholas J. White1,2, Nicholas P.J. Day1,2,Md. Amir Hossain5 and Arjen M. Dondorp1,2*AbstractBackground: Intravascular hemolysis is an intrinsic feature of severe malaria pathophysiology but the pathogenic roleof cell-free hemoglobin-mediated oxidative stress in severe malaria associated acute kidney injury (AKI) is unknown.Methods: As part of a prospective observational study, enrolment plasma cell-free hemoglobin (CFH), lipid peroxidationmarkers (F2-isoprostanes (F2-IsoPs) and isofurans (IsoFs)), red cell deformability, and serum creatinine were quantified inBangladeshi patients with severe falciparum malaria (n = 107), uncomplicated malaria (n = 80) and sepsis (n = 28). Therelationships between these indices and kidney function and clinical outcomes were examined.Results: AKI was diagnosed at enrolment in 58% (62/107) of consecutive patients with severe malaria, defined byan increase in creatinine ≥1.5 times expected baseline. Severe malaria patients with AKI had significantly higherplasma cell-free hemoglobin (geometric mean CFH: 8.8 μM; 95% CI, 6.2–12.3 μM), F2-isoprostane (56.7 pg/ml; 95% CI,45.3–71.0 pg/ml) and isofuran (109.2 pg/ml; 95% CI, 85.1–140.1 pg/ml) concentrations on enrolment compared tothose without AKI (CFH: 5.1 μM; 95% CI, 4.0–6.6 μM; P = 0.018; F2-IsoPs: 27.8 pg/ml; 95% CI, 23.7–32.7 pg/ml; P < 0.001;IsoFs: 41.7 pg/ml; 95% CI, 30.2–57.6 pg/ml; P < 0.001). Cell-free hemoglobin correlated with markers of hemolysis,parasite burden (P. falciparum histidine rich protein 2 (PfHRP2)), and F2-IsoPs. Plasma F2-IsoPs and IsoFs inverselycorrelated with pH, positively correlated with creatinine, PfHRP2 and fractional excretion of sodium, and werehigher in patients later requiring hemodialysis. Plasma F2-IsoP concentrations also inversely correlated with redcell deformability and were higher in fatal cases. Mixed effects modeling including an interaction term for CFHand time showed that F2-IsoPs, IsoFs, PfHRP2, CFH, and red cell rigidity were independently associated withincreasing creatinine over 72 h. Multivariable logistic regression showed that admission F2-IsoPs, IsoFs and red celldeformability were associated with the need for subsequent hemodialysis.(Continued on next page)* Correspondence: arjen@tropmederes.ac1Mahidol Oxford Tropical Medicine Research Unit, Faculty of TropicalMedicine, Mahidol University, Bangkok, Thailand2Centre for Tropical Medicine and Global Health, Nuffield Department ofMedicine, University of Oxford, Oxford, UKFull list of author information is available at the end of the article© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (, which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver( applies to the data made available in this article, unless otherwise stated.Plewes et al. BMC Infectious Diseases  (2017) 17:313 DOI 10.1186/s12879-017-2373-1(Continued from previous page)Conclusions: Cell-free hemoglobin and lipid peroxidation are associated with acute kidney injury and disease severityin falciparum malaria, suggesting a pathophysiological role in renal tubular injury. Evaluation of adjunctive therapiestargeting cell-free hemoglobin-mediated oxidative stress is warranted.Keywords: Acute kidney injury, Pathophysiology, Falciparum malaria, Cell-free hemoglobin, Oxidative stressBackgroundSevere falciparum malaria is characterized by intravascularhemolysis, where cell-free hemoglobin (CFH) increaseswith disease severity [1]. Sources of CFH include ruptureof parasitized red blood cells (RBC) at schizogony, anddestruction of uninfected erythrocytes, most prominentlyin patients with blackwater fever (hemoglobinuria) [2]. In400 BC, Hippocrates first associated blackwater fever withanuria and mortality; findings that consistently resurfacedafter Firth’s report in 1886 [3, 4]. More recently, thiscondition of fulminant hemolysis has been associated withkidney dysfunction in up to 64% of patients [5], but theunderlying mechanisms have not been fully characterized.When the degree of intravascular hemolysis exceedsthe scavenging capacity of plasma haptoglobin forhemoglobin, CFH dimers are filtered by the glomeruliand reabsorbed by the proximal tubule. Once the reab-sorptive capacity is exceeded, hemoglobin appears inthe urine [6]. CFH is independently associated withAKI in patients post-cardiopulmonary bypass, andwith mortality in bacterial sepsis [7–9]. Hemoproteins,hemoglobin and myoglobin, are pathogenic as pro-oxidants when released heme is not scavenged byhemopexin. Heme redox cycling between ferric andferryl states then generates globin radicals inducinglipid peroxidation [10]. In vivo studies on oxidativeinjury have been hampered by the paucity of stableand specific markers of oxidative stress.CFH-mediated non-enzymatic lipid peroxidation ofarachidonic acid generates isomers of prostaglandins,F2-isoprostanes (F2-IsoPs) and isofurans (IsoFs) [11, 12].F2-IsoPs are generated at low oxygen tension whereasIsoFs are generated at higher oxygen tension and togetherare considered robust in vivo measures of oxidative stress[11, 12]. In the current study, the hypothesis was thatCFH-mediated oxidative stress could cause renal damageeither through a direct effect on renal tubules, through areduction in red cell deformability (RCD), or through thevasoconstrictive properties of F2-IsoPs.Arachidonic acids, such as red cell membrane phos-pholipids, are particularly vulnerable to free radical-mediated lipid peroxidation. Oxidative stress-inducedreduction of RCD has been proposed to play a role inrenal insufficiency [13]. In malaria, the high arachidonicacid content of infected RBC membranes reduces withintracellular parasite maturation, suggesting membraneperoxidation occurs during parasite development [14].F2-IsoPs are considered not just bystanders of oxidativeinjury but are bioactive renal vasoconstrictors [12].Both F2-IsoPs and IsoFs have been associated with AKIin patients with rhabdomyolysis and hemolysis post-cardiopulmonary bypass [15–17]. Other plasma andurinary markers of oxidative stress have been shown tobe significantly elevated in severe malaria compared touncomplicated malaria [18–21].The role of plasma CFH-mediated lipid peroxidationin the pathophysiology of severe malaria and AKI hasnot been described. Gaining a better understanding ofthe pathophysiology in malaria will help towards thedevelopment of targeted therapies. This study aimedto examine the generation of CFH-mediated lipidperoxidation and its role in AKI and malaria severityby analyzing the associations between CFH, F2-IsoPs,IsoFs, red cell deformability, and creatinine in patientswith falciparum malaria.MethodsStudy aim, design and settingThe aim of this study was to assess CFH-mediated lipidperoxidation and its role in AKI and disease severity infalciparum malaria. This prospective observational studywas conducted at Chittagong Medical College Hospital,Bangladesh from 2011 to 2014. This tertiary hospitalreceives referrals from malaria hypoendemic areas, andhas basic facilities for intensive care and hemodialysis.Patient characteristicsPatients admitted with slide confirmed severe oruncomplicated P. falciparum malaria were recruitedupon diagnosis. Positive microscopy of peripheral bloodrequired the presence of asexual stages of P. falciparum.Uncomplicated malaria was defined as asexual P. falcip-arum slide positivity without severity criteria. Criteria forsevere malaria were: coma (Glasgow Coma Score < 11),shock (systolic blood pressure (SBP) < 80 mmHg with coolextremities), anemia, (hematocrit <20% plus parasitemia>100,000/μl), jaundice (total bilirubin >51.3 μmol/L plusparasitemia >100,000/μl), hyperparasitemia (asexual parasit-emia >10%), acidosis (bicarbonate <15 mmol/L), hyperlac-tatemia (lactate >4 mmol/L), hypoglycemia (glucose<2.22 mmol/L), convulsions (≥ 2 in 24 h), pulmonaryedema, and/or AKI (serum creatinine >3 mg/dl). PatientsPlewes et al. BMC Infectious Diseases  (2017) 17:313 Page 2 of 12were treated with parenteral artesunate (Guilin No.2Pharmaceuticals, Guangxi, China) followed by arte-mether/lumefantrine (Novartis, Basel, Switzerland)and managed according to WHO treatment guidelines[22]. Hemodialysis was initiated according to local ne-phrologists. Indications for dialysis in this settinginclude: (1) anuria for more than 24 h, (2) severeelectrolyte and acid-base disturbance, (3) serum cre-atinine >3 mg/dl with urine output <0.5 ml/kg/h for12 h, or (4) gradual rise in creatinine despite normal urineoutput. A control group of sepsis patients hospitalizedwith suspected bacterial infection and at least 2/4 systemicinflammatory response criteria (n = 28) was also recruited[23]. Data from non-pregnant patients aged ≥10 years ispresented. Patients were followed until discharge or death;those with AKI were followed in hospital until renalrecovery, if possible, as gauged by local nephrologists.Follow up after hospital discharge was challenging due tothe distances patients travel to the teriary care center.Study proceduresAfter enrolment, a medical history and physical examin-ation were performed. Patients were seen 6-hourly untildischarge or death. Enrolment venous blood samples wereanalyzed for electrolytes, glucose, pH and bicarbonateusing a bedside analyzer (iSTAT, Abbott). Parasitemia wasassessed 6-hourly from thick and thin smears until parasiteclearance. Blood and urine for creatinine measurementwere collected every 24 h for three days. Serum, heparin-ized plasma and urine were frozen in liquid nitrogen withinone hour of collection. Serum creatinine was measuredusing an Olympus analyzer (Beckman Coulter Inc.).AssaysPlasma and urine F2-isoprostane and isofuran concen-trations were determined by gas chromatography-massspectrometry at Vanderbilt University, as described[11, 24]. The 24-h urinary excretion rate of urine F2-isoprostane and isofuran concentrations were calculatedas: concentration x volume × 24 h/time of collection [25].Plasma CFH concentrations were measured by ELISA(Bethyl Laboratories), as described [1]. Plasma Plasmodiumfalciparum histidine rich protein 2 (PfHRP2), a biomarkerof total parasite burden, was quantified using commercialsandwich ELISA (Celisa, Cellabs), as described [26].Red cell deformability (RCD) was measured at enrol-ment using a laser-assisted optical rotational cell analyzer(LORCA, Mechatronic) immediately after blood collection[27, 28]. The deformability was measured by ellipticity ofRBCs and described by the elongation index; (long minusshort axes lengths) divided by (long plus short axeslengths). RCD was assessed at shear stresses ranging from0.3 to 30 Pa. In capillaries, shear stresses of 1.7 Pa andabove are encountered [29].Acute kidney injuryPatients were classified according to AKI status at enrol-ment as defined by an increase in serum creatinine ≥1.5times expected baseline, known or presumed to haveoccurred within the prior seven days (as per the KidneyDisease Improving Global Outcomes (KDIGO) classifi-cation system) [30]. Since urine output was not availablefor all patients, and the duration of illness at presenta-tion was always greater than 48 h, these criteria werenot incorporated for enrolment AKI diagnosis. As pre-admission creatinine values were not available, expectedbaseline creatinine values were calculated as recom-mended using the Modification of Diet in Renal Diseaseformula assuming a glomerular filtration rate (GFR) of75 ml/min/1.73m2 for participants 19 years and older[30] and using the Bedside Schwartz formula assuming aGFR of 100 ml/min/1.73m2 for those 18 years andyounger [31–33]. The highest KDIGO AKI staging wasassessed both on enrolment and during admission inorder to accurately present the heterogeneous AKIstatus at the time of enrolment and subsequent kidneyfunction during admission in those that survived. Stage2 AKI was defined as an increase to ≥ 2.0–2.9 timesexpected baseline; Stage 3 as either an increase to ≥ 3times expected baseline, an increase in creatinine to≥4 mg/dl, initiation of RRT, or in patients <18 years adecrease in GFR to <35 ml/min/1.73 m2 [30]. Creatinineused for enrolment AKI stratification were performedon samples drawn prior to hemodialysis.Statistical analysisGroups were compared using Wilcoxon rank-sum testor Student’s t-tests depending on the distribution of thedata. Correlations were assessed using Spearman’s cor-relation coefficient. As the hypothesis was that F2-IsoPs,IsoFs, and CFH contribute to AKI, these pre-specifiedvariables were assessed in a mixed effects model usingcreatinine as the dependent variable, and in a logisticregression model using hemodialysis as the dependantvariable. As the F2-IsoPs and IsoFs were highly collinear(rp = 0.67; p < 0.001), their association with creatininewas assessed in separate multivariable models. The mul-tivariable models included adjustment for age, SBP,PfHRP2 and RCD as these are considered physiologicallyrelevant in contributing to AKI. In the mixed effectmodel, age, SBP, and time were modeled as fixed effectswhile the rest were treated as random effects. All pre-specified and known AKI risk factor variables that wereassociated with creatinine were included in a multivari-able mixed effects model. The interaction betweenenrolment CFH and time was also assessed. To accountfor hemodialysis confounding the decline in creatinineconcentrations, creatinine values were adjusted at eachtime point following dialysis until time of death by usingPlewes et al. BMC Infectious Diseases  (2017) 17:313 Page 3 of 12a creatinine rise of 1.5 mg/dl per day as proposed foranephric states [34]. Known risk factors for hemodialysis,including additional markers of disease severity (number ofseverity criteria, GCS, and lactate), were assessed usingbackward stepwise logistic regression including variables(Table 5) with a p-value of <0.10 on univariable analysis.Selection of the final model was based on the Akaike infor-mation criteria (AIC). Software used were STATA14.0(Stata), and Prism 6 (Graphpad Software).ResultsBaseline characteristics and clinical course in hospitalA total of 107 consecutive patients with severe malariawere enrolled (Tables 1 and 2), as well as 80 patientswith uncomplicated malaria and 28 with (suspected)bacterial sepsis as comparator control groups. Amongpatients with severe malaria, 58% (62/107) had AKI onenrolment, while another 9% (10/107) subsequently de-veloped AKI during admission (Table 1). The severityof AKI varied with 50% (31/62) meeting the WorldHealth Organization malaria guideline criteria for AKI(creatinine >3 mg/dl) [22], of whom 84% (26/31) wereKDIGO stage 3 and 16% (5/31) had a further progres-sion from KDIGO stage 2 to 3 during admission. In theAKI on enrolment group, 47% (29/62) receivedhemodialysis; of whom 28% (8/29) died, and 53% (33/62)did not receive hemodialysis; of whom 52% (17/33) died(OR for death without dialysis 2.8 (95% CI, 0.9–9.4;P = 0.048). Among the 72% (21/29) survivors in the AKIgroup who received dialysis, the median renal recoverytime was 21 days (IQR, 13–42 days; n = 7). Among the48% (16/33) survivors in the AKI group who did notreceive dialysis, the median renal recovery time was threedays (IQR, 2–4 days; n = 14). Of 10/107 patients whodeveloped AKI after admission, 30% (3/10) receivedhemodialysis and 40% (4/10) died, compared to an overallmortality in the severe malaria cohort of 33% (35/107)(Table 3). The mortality rate among all patients admittedwith or developing AKI but not receiving dialysis wasnearly double (20/40; 50%) that of patients to those whodid (9/32; 28%)(OR 2.6, 95% CI, 0.9–7.8; P = 0.09). Thosewith AKI on enrolment had more severe disease, asTable 1 Baseline demographics and clinical characteristics of patients with severe falciparum malaria by AKI status at enrolmentVariable Total No AKI AKI P(n = 107) (n = 45) (n = 62)DemographicsAge (years) 30 (22–40) 30 (25–45) 27 (18–40) 0.104Males (%)c 75 (70) 35 (78) 40 (65) 0.200Fever prior to admission (days) 7 (6–9) 7 (6–8) 7 (6–9) 0.311History of black or red urinec 20 (20) 5 (11) 15 (26) 0.100Vomiting and diarrhea (days) 5 (1–6) 2.5 (1–5) 5 (3–9) 0.045ComorbiditiesHypertensionc 5 (5) 2 (4) 3 (5) 1.000Cardiovascular diseasec 6 (6) 2 (4) 4 (6) 1.000Type 2 diabetesc 4 (4) 3 (7) 1 (2) 0.307Enrolment clinical parametersGlasgow Coma Score (max 15) 9 (8–14) 10 (9–14) 9 (7–14) 0.354Systolic blood pressure (mmHg) 110 (100–120) 114 (102–120) 107 (99–120) 0.155Mean arterial pressure (mmHg) 82 (72–90) 81 (74–89) 82 (71–94) 0.550Pulse rate (breaths/min) 115 (96–132) 109 (93–131) 116 (97–132) 0.666Respiratory rate (breaths/min) 34 (28–42) 32 (28–37) 36 (28–44) 0.370Hemoglobinuria on enrolment# c 18 (17) 4 (9) 14 (24) 0.124Number of severity criteria 2 (1–3) 1 (1–2) 2 (2–4) <0.001AKI stage at enrolmentStage 1 (≥1.5 × baseline)c 16 (26) – 16 (26) –Stage 2 (≥2.0–2.9 × baseline)c 16 (26) – 16 (26) –Stage 3 (≥3.0 × baseline or ≥4 mg/dl)c 30 (48) – 30 (48) –All values are median (IQR) unless otherwise specified: cnumber (%). P < 0.05 using student t-test or Mann-Whitney U. # hemoglobinuria defined as red, black ordark brown urine on exam with 3/4+ hemoglobin on urine dipstick. Abbreviations: AKI acute kidney injuryPlewes et al. BMC Infectious Diseases  (2017) 17:313 Page 4 of 12defined by a higher median number of WHO severity cri-teria (P < 0.001). No patient reported a history of kidneydisease and there was no difference in comorbidities (hyper-tension, diabetes and cardiovascular disease) between groups(Table 1).Cell-free hemoglobin and oxidative stressPlasma CFH concentrations were significantly higher inpatients with severe malaria (7.0 μM; 95% CI, 5.5-8.7μM; n=105) compared to those with uncomplicated mal-aria (4.3 µM; 95% CI, 3.5-5.2 µM; (n=80)Additional fileTable 2 Baseline admission laboratory parameters of patients with severe falciparum malaria by AKI status at enrolmentVariable Total No AKI AKI P(n = 107) n (n = 45) n (n = 62) nHemoglobin (mg/dl) 9.1 (7.2–11.0) 107 10.6 (8.0–11.3) 45 8.5 (7.1–10.3) 62 0.017Cell-free hemoglobin (μM)b 7.0 (5.5–8.7) 105 5.1 (4.0–6.6) 45 8.8 (6.2–12.3) 60 0.018White blood cells (×103/μl) 9.4 (6.7–12.8) 102 9.2 (6.4–11.7) 44 9.6 (7.0–16.6) 58 0.339Platelets (×103/μl) 30 (19–46) 100 35 (23–53) 44 27 (18–42) 56 0.075Total bilirubin (mg/dl) 2.0 (1.0–5.3) 106 1.5 (0.9–3.2) 45 2.5 (1.3–10.7) 61 0.005Indirect bilirubin (mg/dl) 0.8 (0.3–1.9) 106 0.4 (0.2–1.3) 45 1.3 (0.4–2.7) 61 0.013Lactate dehydrogenase (U/l) 635 (455–886) 107 541(403–643) 45 766 (566–1027) 62 <0.001Creatinine (mg/dl) 1.4 (1.1–3.3) 107 1.2 (1.0–1.3) 45 3.0 (1.6–4.4) 62 <0.001Blood urea nitrogen (mg/dl) 43 (26–75) 107 26 (18–37) 45 66 (44–104) 62 <0.001Potassium (mmol/l) 4.4 (3.9–5.2) 106 4.1 (3.8–4.6) 45 4.7 (4.1–5.5) 61 <0.001Base excess (mmol/l) −8 (−11 to −4) 107 −5 (−7 to −2) 45 −10 (−13 to −7) 62 <0.001Bicarbonate (mmol/l) 17.3 (14.1–20.2) 107 19.2 (17.0–21.9) 45 15.9 (13.3–18.6) 62 <0.001Lactate (mmol/l) 3.85 (2.49–6.28) 107 3.85 (2.58–5.61) 45 3.89 (2.30–6.68) 62 0.852Parasitemia (parasites/μl)b 53,529 (34586–82,846) 107 61,253 (34094–110,048) 45 48,540 (25716–91,619) 62 0.605PfHRP2 (ng/ml) 2584 (1341–7194) 101 1743.8 (1090–3159) 45 3996 (1737–12,382) 56 <0.001Urinary indicesUrine protein:creatinine 0.81 (0.42–1.11) 81 0.81 (0.52–1.15) 34 0.81 (0.41–1.11) 47 0.867Urine albumin:creatinine 9.52 (5.49–19.12) 65 9.52 (5.69–19.12) 25 9.76 (5.19–19.49) 40 0.861pH 5 (5–6) 98 6 (5–6) 42 5 (5–6) 56 <0.001FeNa (%)c 0.67 (0.37–1.35) 94 0.57 (0.22–0.91) 40 0.94 (0.47–2.44) 54 0.012Oxidative stress markersPlasma F2-IsoPs (pg/ml)b 41.1 (34.8–48.5) 64 27.8 (23.7–32.7) 29 56.7 (45.3–70.9) 35 <0.001Plasma IsoFs (pg/ml)b 70.6 (56.2–88.7) 64 41.7 (30.2–57.6) 29 109.2 (85.1–140.1) 35 <0.001All values are median (IQR) unless otherwise specified: bgeometric mean (95% CI), cnumber (%). P < 0.05 using student t-test, Mann U Whitney or Fischer’s exacttests; significant in bold. Abbreviations: AKI acute kidney injury, PfHRP2 P falciparum histidine rich protein 2, FeNa fractional excretion of sodium, F2-IsoPs plasmaF2-isoprostanes, IsoFs plasma isofuransTable 3 Outcomes by AKI status on enrolmentOutcome Total No AKI AKI P(n = 107) (n = 45) (n = 62)AKI stage during admission†Stage 1 (≥1.5 × baseline)c 11 (12) 4 (10) 7 (15) 0.458Stage 2 (≥2.0–2.9 × baseline)c 11 (12) 1 (2) 10 (21) 0.021Stage 3 (≥3.0 × baseline or ≥4 mg/dl)c 36 (40) 5 (12) 31 (65) <0.001Received RRT (%)c 32 (30) 3 (7) 29 (47) <0.001Length of hospital stay (days) 7.6 (5.6–12.9) 5.9 (4.9–8.3) 10.6 (6.6–18.0) <0.001Death (%)c 35 (33) 10 (22) 25 (40) 0.061Study hours to death (%)c 22.5 (12.5–51.0) 40.1 (19.0–115.0) 21.0 (10.0–38.0) 0.074All values are median (IQR) unless otherwise specified: cnumber (%). P < 0.05 using Mann U Whitney or Fischer’s exact test; significant in bold. †Highest KDIGOstage during admission in those that survived more than 24 h. Abbreviations: AKI = acute kidney injury; RRT = renal replacement therapyPlewes et al. BMC Infectious Diseases  (2017) 17:313 Page 5 of 121: Figure S1; P = 0.002) or sepsis (2.5 μM; 95% CI, 1.4–4.3 μM; n = 28; P < 0.001). Plasma F2-IsoPs and IsoFswere also higher in patients with severe malaria com-pared to uncomplicated malaria (Additional file 1: Fig-ure S1; F2-IsoPs, P < 0.001; IsoFs, P = 0.005). Amongthose with severe malaria, plasma CFH, F2-IsoPs andIsoFs correlated with other measures of hemolysis, in-cluding LDH, total bilirubin, and indirect bilirubin.PfHRP2, [26] but not peripheral blood parasitemia,correlated positively with CFH (rs = 0.55, P < 0.001;rs = 0.18, P = 0.08) (Fig. 1A). Similarly, F2-IsoPs andIsoFs correlated with PfHRP2 (rs = 0.34, P = 0.008;rs = 0.31, P = 0.017) but neither correlated with para-sitemia (rs = 0.02, P = 0.89; rs = −0.03, P = 0.98).CFH weakly correlated with plasma F2-IsoPs (rs = 0.30,P = 0.018), as a measure of oxidative stress (Fig. 1B);but the correlation with IsoFs was not significant(rs = 0.16, P = 0.22). Both plasma F2-IsoPs and IsoFswere inversely associated with pH (rs = − 0.51,P < 0.001; rs = − 0.49; P < 0.001) (Fig. 1C, D) andpositively correlated with base deficit (rs = 0.59,P < 0.001; rs = 0.63; P < 0.001). In a multivariableregression model adjusting for disease severity, CFHand decreasing pH (acidosis) were positive predictorsof (log) F2-IsoPs (β coefficient 0.13; 95% CI, 0.02 to0.25; P = 0.026; −2.94; 95% CI, −4.24 to −1.64;P < 0.001, respectively).Cell-free hemoglobin, oxidative stress, and renal functionIn severe malaria, CFH was higher in patients with AKIcompared to those without AKI (P = 0.018) (Fig. 2;Table 2). CFH in patients with severe malaria-associated AKI was also higher compared to patientswith sepsis-related AKI (0.8 μM; 95% CI, 0.1–3.9 μM;n = 4; P = 0.001). Patients with hemoglobinuria atenrolment had significantly higher CFH (geometricmean: 15.6 μM; 95% CI, 6.9–35.6 μM; n = 18), PfHRP2(median: 10,411 ng/ml; IQR, 2909–14,504 ng/ml;n = 18), and serum creatinine (median: 2.9 mg/dl; IQR,1.3–4.7 mg/dl; n = 18) compared to those withouthemoglobinuria (CFH: 5.4 μM; 95% CI, 4.3–6.7 μM;n = 74; P < 0.001; PfHRP2: 2146 ng/ml; IQR, 1266–4216 ng/ml; n = 73; P = 0.001; creatinine: 1.4 mg/dl;IQR, 1.1–2.7 mg/dl; n = 76; P = 0.040). In severe100 101 102 103 104 10510-1100101102103PfHRP2 (ng/ml)Cell-free hemoglobin  (µM)rs = 0.55P  < 0.0016.9 7.0 7.1 7.2 7.3 7.4 7.5 7.6 7.7101102103pHPlasma F2-isoprostanes (pg/ml) rs = - 0.51P  < 0.00110-1 100 101 102 103101102Cell-free hemoglobin (µM)Plasma F2-isoprostanes (pg/ml) rs = 0.30P  = 0.0186.9 7.0 7.1 7.2 7.3 7.4 7.5 7.6 7.7101102103pHPlasma Isofurans (pg/ml)rs = - 0.49P  < 0.001a bc dFig. 1 Correlation of oxidative stress markers in severe malaria. a Parasite burden, as measured by PfHRP2 concentration, positively correlatedwith plasma cell-free hemoglobin concentration (n = 96), b cell-free hemoglobin positively correlated with plasma F2-isoprostanes (n = 62).Acidemia (venous pH) was inversely correlated with (c) plasma F2-isoprostanes, and d plasma isofurans. All values are from enrolment assess-ments. Abbreviations: PfHRP2, Plasmodium falciparum histidine rich protein 2; rs, Spearman’s correlation coefficientPlewes et al. BMC Infectious Diseases  (2017) 17:313 Page 6 of 12malaria, plasma F2-IsoPs and IsoFs were significantlyhigher in patients with AKI compared to those without(both P < 0.001) (Fig. 2; Table 2). Enrolment plasmaF2-IsoPs and IsoFs strongly correlated with enrolmentserum creatinine (rs = 0.71, P < 0.001; rs = 0.71,P < 0.001) (Fig. 3A, B) and with the fractional excre-tion of sodium (rs = 0.49, P < 0.001; rs = 0.37,P = 0.006). The 24-h urine F2-IsoP excretion concen-tration was lower in the AKI group (median: 0.9 ng;IQR, 0.6–2.4 ng; n = 9) compared to those withoutAKI (2.9 ng; IQR, 1.1–4.0 ng; n = 11; P = 0.037).However, the 24-h urine IsoF excretion concentrationwas similar between groups (AKI median: 9.7 ng;IQR, 8.1–15.2 ng; n = 9; No AKI: 9.2 ng; IQR, 4.3–53.2 ng;n = 12; P = 0.72).In patients without AKI on enrolment but subse-quently developing AKI, initial plasma IsoFs were highercompared to those who did not develop AKI (geometricmean: 81.6 pg/ml; 95% CI, 28.6–232.9 pg/ml versus35.0 pg/ml; 95% CI, 25.6–48.0 pg/ml; P = 0.027). Fur-thermore, in this group (excluding 3/10 patients whoreceived hemodialysis) peak plasma concentrations ofF2-IsoPs and IsoFs at 24 h were higher than in patientsnot developing AKI (F2-IsoPs: 43.5 pg/ml; 95% CI 26.4–71.8 pg/ml; versus 24.8 pg/ml; 95% CI 19.1–32.1 pg/ml;P = 0.020; IsoF: geometric mean 220.0 pg/ml; 95% CI,52.9–914.8 pg/ml versus 48.0 pg/ml; 95% CI, 32.8–70.2 pg/ml; P = 0.003). The peak creatinine in theformer group was reached at 48 h (median: 4.5 mg/dl;IQR, 1.1–4.6 mg/dl).Plasma F2-IsoPs and IsoFs on enrolment, prior tohemodialysis, were higher among those who receivedhemodialysis (F2-IsoPs: 58 pg/ml; 95% CI, 46–72 pg/ml;n = 19; versus 36 pg/ml; 95% CI, 29–44 pg/ml; n = 45;P = 0.006; IsoFs: 131 pg/ml; 95% CI, 99–173 pg/ml;n = 19; versus 54 pg/ml; 95% CI, 41–71 pg/ml; n = 45;P < 0.002). Plasma CFH at enrolment was not higher inthose who received hemodialysis (CFH: 9.0 μM; 95% CI,5.2–15.4 μM; n = 31; versus 6.2 μM; 95% CI, 4.9–7.9 μM; n = 74; P = 0.15).Red cell deformability, renal function and oxidative stressRCD at enrolment was lower at shear stresses between1.69 and 30.00 Pa in patients with AKI on enrolmentcompared to those without AKI (Figure 4). DecreasedRCD correlated with higher creatinine at all these shearstresses and most strongly at 9.49 Pa (rs = − 0.31,P = 0.019). RCD was also lower at shear stressesbetween 1.69 and 9.49 Pa in patients who subsequentlyrequired hemodialysis (all P < 0.05). RCD at low shearstress was inversely associated with plasma F2-IsoPs(0.3 Pa: rs = − 0.46, P = 0.003; n = 39) but not with IsoFs(1.69 Pa: rs = − 0.30, P = 0.062; n = 40).05101520µMP = 0.018CFH pF2-IsoP pIsoF050100150200No AKI AKIP < 0.001P < 0.001(pg/ml)  Fig. 2 Cell-free hemoglobin, and oxidative stress measures atenrolment in patients with severe malaria by AKI status. Plasmacell-free hemoglobin (n = 105), F2-isoprostanes (n = 64) and isofurans(n = 64) were significantly more elevated on enrolment in those withacute kidney injury. Geometric mean and 95% CI are shown.Abbreviations: AKI, acute kidney injury; CFH, cell-free hemoglobin;pF2-IsoP, plasma F2-isoprostanes; pIsoF, plasma isofurans101 102 10310-1100101Plasma F2-isoprostanes  (pg/ml)Creatinine (mg/dL)rs = 0.71P  < 0.001101 102 10310-1100101Plasma isofurans (pg/ml)Creatinine (mg/dL)rs = 0.71P  < 0.001a bFig. 3 Correlations with creatinine in patients with severe malaria. Creatinine on enrolment positively correlated with (a) plasma F2-isoprostanes(n = 63), and (b) plasma isofurans (n = 64). Abbreviations: rs, Spearman’s correlation coefficientPlewes et al. BMC Infectious Diseases  (2017) 17:313 Page 7 of 12Predictors of renal functionSince plasma F2-IsoPs and IsoFs were collinear, two differ-ent multivariable mixed effects models were considered toassess the associations of each of them with change inserum creatinine over time (Table 4, F2-IsoP model andIsoF model, respectively). There was a significant inter-action between (log)CFH and time in both multivariablemodels, P = 0.010 and P = 0.012 for the model that in-cluded (log)F2-IsoPs and (log)IsoFs, respectively (Table 4).The effect of enrolment CFH on creatinine had a lag time.In the model with F2-IsoPs, increased enrolment plasmaF2-IsoPs, PfHRP2 and decreased red cell deformabilitywere also independently associated with increasing serumcreatinine over the first 72 h of admission. In the modelwith IsoFs, increased enrolment plasma IsoFs, andPfHRP2 were independently associated with an increase inserum creatinine over 72 h. In a logistic regression modeladjusted for disease severity, plasma F2-IsoPs (OR = 7.37,95% CI, 1.86–29.23) was independently associated withthe need for hemodialysis during admission, but RCD wasnot (overall model fit R2 = 0.20; Table 5). In the logisticregression model with IsoFs (rather than F2-IsoPs) as in-dependent variable, elevated IsoFs (OR = 5.5, 95% CI,1.76–17.24) at enrolment was also independently1.69 3.00 5.33 9.49 16.87 (elongation index)Shear stress (pascals)******No AKI AKI Fig. 4 Red cell deformability at enrolment in patients with severe malaria by AKI status. Red cell deformability at shear stresses from 1.69to 30.0 Pa were significantly lower in the AKI group (n = 51) compared to the no AKI group (n = 33). Abbreviations: AKI, acute kidneyinjury; RCD, red cell deformability. Asterisks represent P < 0.05 significanceTable 4 Association of variables with change in absolute creatinine over 72 h in patients with severe malariaUnivariable analysis Multivariable F2-IsoP model Multivariable IsoF modelVariable β (95% CI)a P β (95% CI)a P β (95% CI)a PLog CFH§ 0.27 (−0.19 to 0.74) 0.249 — — — —Log F2-IsoP# 2.35 (1.27 to 3.43) <0.001 3.01 (1.92 to 4.12) <0.001 — —Log IsoF 2.06 (1.46 to 2.65) <0.001 — — 1.83 (1.26 to 2.40) <0.001LogPfHRP2 0.61 (0.25 to 0.98) 0.001 0.70 (0.02 to 1.38) 0.045 0.69 (0.05 to 1.32) 0.034RCD at SS 1.69 Pa −1.67 (−14.93 to 11.59) 0.805 — — — —RCD at SS 9.49 Pa −6.42 (−15.10 to 2.26) 0.147 −4.26 (−7.86 to −0.67) 0.020 −2.46 (−5.92 to 1.00) 0.163Age 0.01 (−0.03 to 0.05) 0.624 — — — —SBP 0.03 (−0.01 to 0.06) 0.094 — —Visit (time) 0.65 (0.52 to 0.78) <0.001 — — — —Log CFH × Log F2-IsoP 0.31 (−0.19 to 0.86) 0.221 — — — —Log CFH × Log IsoF 0.62 (0.27 to 0.98) 0.001 — — — —Log F2-IsoP × visit 0.85 (0.55 to 1.14) <0.001 — — — —Log IsoF × visit 0.62 (0.47 to 0.76) <0.001 — — — —Log CFH × visit 0.10 (−0.10 to 0.21) 0.076 0.21 (0.05 to 0.36) 0.010 0.20 (0.04 to 0.36) 0.012aRegression coefficient (β) with 95% confidence intervals (CIs) showing the estimated decrease in creatinine predicted by a 1-U change in the independent(predictor) variables. Interaction terms improved the model fit. Abbreviations: CFH cell free hemoglobin, F2-IsoP plasma F2-isoprostanes, IsoF plasma isofurans,CFH × F2-IsoP cell-free hemoglobin and plasma F2-isoprostane interaction term, CFH × IsoF cell-free hemoglobin and plasma isofuran interaction term, PfHRP2P. falciparum histidine rich protein 2, RCD red cell deformability at shear stress 1.69 and 9.49 Pa. # Plasma IsoFs collinear with plasma F2- IsoPs. P-values initalics denote statistical significancePlewes et al. BMC Infectious Diseases  (2017) 17:313 Page 8 of 12associated with the subsequent need for hemodialysis,but RCD and CFH were not (overall model fitR2 = 0.41; Table 5).Cell-free hemoglobin, oxidative stress and survivalThose who died had higher enrolment plasma F2-IsoPs(geometric mean: 55 pg/ml; 95%CI, 38–79; n = 21) andIsoFs (geometric mean: 97 pg/ml; 95%CI, 64–145;n = 21) compared to survivors (F2-IsoPs: 36 pg/ml;95%CI, 30–42; n = 43; P = 0.014; IsoFs: 61 pg/ml;95%CI, 46–80; n = 43; P = 0.054). There was no differ-ence in CFH between those who died and survived(P = 0.143).DiscussionIn this study, plasma cell-free hemoglobin and oxidativestress markers (F2-IsoPs and IsoFs) were strongly associ-ated with the presence of AKI on admission, and oxida-tive stress markers predicted subsequent creatinineelevation and hemodialysis requirement during admis-sion in adult patients with severe falciparum malaria.Earlier studies have shown that urine F2-IsoPs andother urine and plasma oxidative stress markers are ele-vated in severe malaria compared to uncomplicated mal-aria [18–21]. CFH-mediated oxidative stress has beenshown to contribute to AKI in diseases and medical pro-cedures inducing hemolysis [15–17, 35, 36]. In malaria,intravascular hemolysis involves both parasitized andnon-parasitized red blood cells [2]. In the current study,plasma CFH was associated with higher parasite burden,as measured by PfHRP2, consistent with a previousstudy [1]. This suggests that the hemoglobin releasedat schizont rupture, the end of the 48-h intra-erythrocytic lifecycle, contributes to plasma CFHconcentration.Redox cycling of hemoglobin forms a radical speciesthat can initiate lipid peroxidation of arachidonic acid togenerate F2-IsoPs and IsoFs [11, 12, 37]. In this study,CFH concentrations correlated with increased levels ofplasma F2-IsoPs, a marker of oxidative stress. Severalfactors influence the generation of lipid peroxidationmarkers. The oxidative capacity of plasma CFH islargely dependent on its redox state and the fate ofthe heme moiety. Heme is water-insoluble; it binds tohemopexin, albumin, lipoproteins, and cell mem-branes [38], and its oxidative capacity differs consid-erably between these fractions. In addition, othersources of free radicals, such as those created duringphagocytic oxidative burst, may initiate lipid peroxida-tion to generate plasma F2-IsoPs and IsoFs. Indeed,renal histopathology of AKI in severe malaria showsaccumulation of host monocytes (in addition to para-sitized red blood cells) in the renal microvasculature[39]. Heme-containing myoglobin concentrations arealso increased in severe malaria, although to a muchlesser extent than CFH [1]. The oxidative capacity ofmyoglobin increases at low pH because of increasedpseudoperoxidase activity [10]. This could also applyto CFH, given that both plasma F2-IsoPs and IsoFswere associated with reduced venous blood pH in thecurrent study.F2-IsoPs are potent renal vasoconstrictors, which re-duce renal blood flow and GFR [12]. In this study,concentrations of plasma CFH, F2-IsoPs, and IsoFs atenrolment were all higher in patients with AKI, whereasF2-IsoPs and IsoFs correlated with the fractional excre-tion of sodium (an indirect measure of renal tubularTable 5 Association of variables with subsequent hemodialysis requirement in patients with severe malariaUnivariable analysis Multivariable F2-IsoP model Multivariable IsoF modelVariable OR (95% CI)a P OR (95% CI)a P OR (95% CI)a PCFH 1.01 (0.99 to 1.03) 0.063 — — 1.06 (0.98 to 1.14) 0.156Log F2-IsoP# 3.45 (1.30 to 9.16) 0.013 7.37 (1.86 to 29.23) 0.005 — —Log IsoF 3.48 (1.62 to 7.49) 0.001 — — 5.50 (1.76 to 17.24) 0.003LogPfHRP2 1.92 (1.27 to 2.91) 0.002 — — — —LogRCD at SS 1.69 Pa 0.29 (0.08 to 1.09) 0.066 — — — —LogRCD at SS 9.49 Pa 0.10 (0.01 to 0.75) 0.025 0.057 (0.002 to 1.33) 0.075 0.031 (0.001 to 1.44) 0.076Age 1.00 (0.97 to 1.03) 0.940 — — — —SBP 1.02 (0.99 to 1.04) 0.260 — —Number of severity criteria 1.90 (1.38 to 2.62) <0.001 — — — —GCS 0.95 (0.85 to 1.07) 0.393 — — — —LogLactate 1.22 (0.64 to 2.30) 0.548 — — — —aOdds ratio (OR) with 95% confidence intervals (CIs) showing the hemodialysis requirement predicted by a 1-U change in the independent (predictor) variables. Abackward stepwise multivariable model included all variables in univariable analyses which were removed on the basis of P ≥ 0.05. Abbreviations: CFH cell freehemoglobin, F2-IsoP plasma F2-isoprostanes, IsoF plasma isofurans, PfHRP2 P. falciparum histidine rich protein 2, RCD red cell deformability at shear stress 1.69 and9.49 Pa, SBP systolic blood pressure, GCS Glasgow Coma Score. # Plasma IsoFs collinear with plasma F2- IsoPs. P-values in italics denote statistical significancePlewes et al. BMC Infectious Diseases  (2017) 17:313 Page 9 of 12injury) and were independently associated with increasedcreatinine over 72 h and the subsequent requirement forhemodialysis. A common problem of AKI biomarkers isthat their concentrations can be increased as a result ofrenal dysfunction, rather than be a cause of it. Indeed,the lower 24-h urine F2-IsoP excretion concentrations inpatients with established AKI suggest that hemoglobindimers and/or F2-IsoPs are not filtered well at lowcreatinine clearances. However, the group of patientswithout AKI on enrolment that subsequently developedAKI also showed elevated enrolment plasma IsoF con-centrations. This was then followed by concomitantincreases in 24 h plasma F2-IsoPs and IsoF with a subse-quent peak in serum creatinine at 48 h. This time coursesuggests a role of plasma F2-IsoPs and IsoFs in thepathogenesis of AKI. The delayed effect of CFH-mediated oxidative stress on renal function is alsosuggested by the multivariable mixed effects model, inwhich time modified the effect of enrolment CFH oncreatinine, where the slope of change in creatinineincreased with time.Plasma F2-IsoPs were also associated with reduced RBCdeformability. Red cell rigidity has been well-described insevere malaria [27]. It is thought to be caused by oxidativedamage to RBC membranes, and contribute to micro-vascular flow obstruction [27, 28]. RBC membranes arerich in arachidonic acid. Heme-mediated lipid peroxida-tion of RBC membranes could be an important cause ofreduced RCD, and a significant source of plasma F2-IsoPsand IsoFs in severe malaria. RCD at low shear stresses ismainly determined by membrane properties, whereas athigh shear stresses surface-to-volume relationships be-come more important [40, 41]. In this study, the correl-ation between plasma F2-IsoPs and decreased RCD wasstrongest at low shear stress, suggesting RBC membranedamage. In the current study, RCD was lower in patientswith AKI, inversely correlated with serum creatinine, andindependently associated with an increase in serumcreatinine over 72 h. Reduced RCD could furthercompromise renal medullary perfusion concomitant withF2-IsoP- and IsoF-induced vasoconstriction and obstruct-ing microvascular sequestered parasitized red blood cells,all contributing to renal tubular damage [39].This study has some limitations. The sample size of thisdetailed study was relatively small. Additional biomarkersof AKI were not assessed. However, both changes inserum creatinine and hemodialysis are well establishedindicators of kidney dysfunction and injury. Living orpost-mortem renal biopsy as a standard to evaluate theunderlying kidney pathology was not ethical or feasible inthis study. Quantification of plasma and urine hemoglobinoxidation states was not performed. However, studies haveshown that methemoglobin is the predominant form inpatients with malaria and blackwater fever [42]. Myoglobinas an alternative source of heme-mediated oxidativedamage was not assessed, but other studies have shown a100-fold less increase in plasma myoglobin compared toCFH in patients with severe malaria [1].ConclusionsIn summary, the findings of this study suggest that CFHand systemic F2-IsoPs and IsoFs contribute to the patho-genesis of AKI in severe malaria and these markers ofoxidative stress are associated with need for hemodialysisand in-hospital mortality. CFH can have a direct oxida-tive damaging effect on renal tubules. In addition, CFHinduced lipid peroxidation may have a dual effect in themechanism of AKI in malaria. Firstly, the lipid peroxi-dation metabolites, plasma F2-IsoPs, may cause directrenal vasoconstriction. Secondly, lipid peroxidation ofRBC membranes may result in reduced RCD, which inturn could exacerbate ischemia in the renal medulla.Therapies targeted at reducing hemoprotein-mediatedoxidative stress have been shown to improve renal func-tion and mortality [9, 17, 35]. Treatments that reduceCFH levels or CFH-mediated oxidative stress, such ashaptoglobin or paracetamol respectively, may have po-tential as an adjunctive therapy to improve kidneyfunction and survival in severe malaria.Additional fileAdditional file 1: Figure S1. Supplement to Plewes K, Kingston HWF,Ghose A, et al. Cell-free hemoglobin mediated oxidative stress is associatedwith acute kidney injury and renal replacement therapy in severe falciparummalaria: an observational study. Cell-free hemoglobin, and oxidativestress measures at enrolment stratified by malaria severity. Plasmacell-free hemoglobin (n = 185), F2-isoprostanes (n = 82) and isofurans(n = 82) were significantly more elevated on enrolment in those withsevere malaria compared to uncomplicated malaria. Geometric mean and 95%CIs shown. Abbreviations: AKI, acute kidney injury; CFH, cell-free hemoglobin;pF2-IsoP, plasma F2-isoprostanes; pIsoF, plasma isofurans. (DOCX 1707 kb)AbbreviationsAKI: acute kidney injury; CFH: cell-free hemoglobin; F2-IsoPs: F2-isoprostanes;GFR: glomerular filtration rate; IsoF: Isofurans; KDIGO: Kidney DiseasesImproving Global Outcomes; PfHRP2: Plasmodium falciparum histidinerich protein 2; RBC: red blood cell; RCD: red cell deformability; RRT: renalreplacement therapyAcknowledgmentsWe thank the patients, relatives, research assistants, attending physicians, andsupport staff at Chittagong Medical College Hospital for their assistanceand collaboration with the Mahidol-Oxford Tropical Medicine ResearchUnit. Benjamas Intharabut, Ketsanee Srinamon, Md Safiqul MostafaChoudury, Sanjib Kanti Paul, and Sumon Sharma for their instrumentalassistance. We would also like to acknowledge Bill Zachert for performingquantification of F2-isoprostanes and isofurans.FundingThis work was supported by the Wellcome Trust of Great Britain (grant number089275/Z/09/Z); the Australian National Health and Medical Research Council(grant number 605807, and Fellowships to NMA and TWY), and the NationalInstitutes of Health (grant GM15431 to LJR). KP was supported by the InfectiousDiseases Society of America, Education and Research Foundation and thePlewes et al. BMC Infectious Diseases  (2017) 17:313 Page 10 of 12National Foundation for Infectious Diseases, Young Investigator recipient of theMerle A. Sande/Pfizer Fellowship in International Infectious Diseases; and theClinician Investigator Program at the University of British Columbia, Canada.Availability of data and materialsThe datasets generated and analyzed during the current study are not publiclyavailable. The Mahidol Oxford Tropical Medicine Research Unit has a DataAccess Committee that reviews all data requests on a case by case basis. MORUis committed to ensuring that data sharing is planned for at the inception ofa study: including during negotiations with funders and collaborating sites,during evaluation of compliance with local and international ethics andregulatory requirements, and during the design and conduct of consentprocesses. Queries and applications for datasets should be directed to thecorresponding author who will discuss with the MORU Data Access committee.For further information please refer to the MORU Data Sharing Policy(’ contributionsLJR and AMD conceived of the study. KP contributed to study design,patient enrollment, acquisition of patient samples, performed the statisticalanalysis and wrote the manuscript. HWK, RJM, MTH, SJL, HI, TWY, PC, and KScontributed to the study design, patient enrolment, sample collection andmanuscript revision. AG, MMUH, MSH, SA and MAH contributed to studydesign, clinical care of the patients and manuscript revision. KAP performedthe cell-free hemoglobin measurements, and provided interpretation of thedata and revision of the manuscript. SJL, MM and AMD advised and assistedwith statistical analysis, and critically revised the manuscript for intellectualcontent. LJR provided the F2-IsoP and IsoF quantification. MAF, GDHT, TWY,NMA, NPJD, NJW, LJR, and AMD provided interpretation of the data and criticalmanuscript revisions. All authors read and approved of the final manuscript,and agreed to be accountable for all aspects of the work.Competing interestsAll authors declare that they have no competing interests.Consent for publicationInformed written consent for publication was obtained from each patient orlegally acceptable representative.Ethics approval and consent to participateInformed written consent to participate was obtained from each patient orlegally acceptable representative. All procedures were in accordance withthe ethical standards of Declaration of Helsinki 2008. Ethical approval wasobtained from Chittagong Medical College Ethical Review Committee, andOxford Tropical Research Ethics Committee (reference 21-11).Publisher’s NoteSpringer nature remains neutral with regard jurisdictional claims in publishedmaps and institutional affiliations.Author details1Mahidol Oxford Tropical Medicine Research Unit, Faculty of TropicalMedicine, Mahidol University, Bangkok, Thailand. 2Centre for TropicalMedicine and Global Health, Nuffield Department of Medicine, University ofOxford, Oxford, UK. 3Department of Medicine and Vancouver Coastal Health,University of British Columbia Clinical Investigator program, Vancouver,Canada. 4Global Health Division, Menzies School of Health Research andCharles Darwin University, Darwin, Northern Territory, Australia. 5Departmentof Medicine, Chittagong Medical College and Hospital, Chittagong,Bangladesh. 6Department of Nephrology, Chittagong Medical College andHospital, Chittagong, Bangladesh. 7Department of Anesthesiology,Chittagong Medical College and Hospital, Chittagong, Bangladesh. 8LeeKong Chian School of Medicine, Nanyang Technological University,Singapore, Singapore. 9Dev Care Foundation, Dhaka, Bangladesh.10Department of Pharmacology, Vanderbilt University, Nashville, TN, USA.Received: 19 October 2016 Accepted: 30 March 2017References1. Yeo TW, Lampah DA, Tjitra E, Gitawati R, Kenangalem E, Piera K, Granger DL,Lopansri BK, Weinberg JB, Price RN, Duffull SB, Celermajer DS, Anstey NM.Relationship of cell-free hemoglobin to impaired endothelial nitric oxidebioavailability and perfusion in severe falciparum malaria. J Infect Dis. 2009;200:1522–9.2. Foy H, Kondi A, Rebelo A, Soeiro A. Survival of transfused red cells inblackwater fever circulation and of blackwater red cells in normal circulation(preliminary report). Trans R Soc Trop Med Hyg. 1945;38:271–86.3. Hippocrates. Of the epidemics. 430BC. Accessed 12 February 2015.4. Stephens JWW. Blackwater fever: a historical survey and summary ofobservations made over a century. London: Liverpool University Press;Hodder & Stoughton; 1937.5. Tran TH, Day NP, Ly VC, Nguyen TH, Pham PL, Nguyen HP, Bethell DB, DihnXS, White NJ. Blackwater fever in southern Vietnam: a prospectivedescriptive study of 50 cases. Clin Infect Dis. 1996;23:1274–81.6. Bunn HF, Jandl JH. The renal handling of hemoglobin. II catabolism. J ExpMed. 1969;129:925–34.7. Mamikonian LS, Mamo LB, Smith PB, Koo J, Lodge AJ, Turi JL.Cardiopulmonary bypass is associated with hemolysis and acute kidneyinjury in neonates, infants, and children. Pediatr Crit Care Med. 2014;15:e111–9.8. Vermeulen Windsant IC, Snoeijs MG, Hanssen SJ, Altintas S, Heijmans JH,Koeppel TA, Schurink GW, Buurman WA, Jacobs MJ. Hemolysis is associatedwith acute kidney injury during major aortic surgery. Kidney Int. 2010;77:913–20.9. Janz DR, Bastarache JA, Peterson JF, Sills G, Wickersham N, May AK, Roberts2nd LJ, Ware LB. Association between cell-free hemoglobin,acetaminophen, and mortality in patients with sepsis: an observationalstudy. Crit Care Med. 2013;41:784–90.10. Reeder BJ, Wilson MT. Hemoglobin and myoglobin associated oxidativestress: from molecular mechanisms to disease states. Curr Med Chem. 2005;12:2741–51.11. Fessel JP, Porter NA, Moore KP, Sheller JR, Roberts 2nd LJ. Discovery of lipidperoxidation products formed in vivo with a substituted tetrahydrofuranring (isofurans) that are favored by increased oxygen tension. Proc NatlAcad Sci U S A. 2002;99:16713–8.12. Morrow JD, Hill KE, Burk RF, Nammour TM, Badr KF, Roberts 2nd LJ. A seriesof prostaglandin F2-like compounds are produced in vivo in humans by anon-cyclooxygenase, free radical-catalyzed mechanism. Proc Natl Acad Sci US A. 1990;87:9383–7.13. Brown CD, Ghali HS, Zhao Z, Thomas LL, Friedman EA. Association ofreduced red blood cell deformability and diabetic nephropathy. Kidney Int.2005;67:295–300.14. Hsiao LL, Howard RJ, Aikawa M, Taraschi TF. Modification of host cellmembrane lipid composition by the intra-erythrocytic human malariaparasite plasmodium falciparum. Biochem J. 1991;274:121–32.15. Billings FT, Ball SK, Roberts LJ, Pretorius M. Postoperative acute kidney injuryis associated with hemoglobinemia and an enhanced oxidative stressresponse. Free Radic Biol Med. 2011;50:1480–7.16. Holt S, Reeder B, Wilson M, Harvey S, Morrow JD, Roberts LJ, Moore K.Increased lipid peroxidation in patients with rhabdomyolysis. Lancet. 1999;353:1241.17. Moore KP, Holt SG, Patel RP, Svistunenko DA, Zackert W, Goodier D, ReederBJ, Clozel M, Anand R, Cooper CE, Morrow JD, Wilson MT, Darley-Usmar V,Roberts LJ. A causative role for Redox cycling of Myoglobin and itsinhibition by Alkalinization in the pathogenesis and treatment ofRhabdomyolysis-induced renal failure. J Biol Chem. 1998;273:31731–7.18. Charunwatthana P, Abul Faiz M, Ruangveerayut R, Maude RJ, Rahman MR,Roberts 2nd LJ, Moore K, Bin Yunus E, Hoque MG, Hasan MU, Lee SJ,Pukrittayakamee S, Newton PN, White NJ, Day NP, Dondorp AM. N-acetylcysteine as adjunctive treatment in severe malaria: a randomized,double-blinded placebo-controlled clinical trial. Crit Care Med. 2009;37:516–22.19. Das BS, Patnaik JK, Mohanty S, Mishra SK, Mohanty D, Satpathy SK, Bose TK.Plasma antioxidants and lipid peroxidation products in falciparum malaria.AmJTrop Med Hyg. 1993;49:720–5.Plewes et al. BMC Infectious Diseases  (2017) 17:313 Page 11 of 1220. Rubach MP, Mukemba J, Florence S, Lopansri BK, Hyland K, Volkheimer AD,Yeo TW, Anstey NM, Weinberg JB, Mwaikambo ED. Impaired systemictetrahydrobiopterin bioavailability and increased oxidized biopterins inpediatric falciparum malaria: association with disease severity. PLoS Pathog.2015;11:e1004655.21. Yeo TW, Lampah DA, Kenangalem E, Tjitra E, Price RN, Weinberg JB, HylandK, Granger DL, Anstey NM. Impaired systemic tetrahydrobiopterinbioavailability and increased dihydrobiopterin in adult falciparum malaria:association with disease severity, impaired microvascular function andincreased endothelial activation. PLoS Pathog. 2015;11:e1004667.22. World Health Organization. Severe Malaria. Trop Med Int Health. 2014;19(Suppl I).23. Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA, Schein RM,Sibbald WJ. Definitions for sepsis and organ failure and guidelines for theuse of innovative therapies in sepsis. The ACCP/SCCM consensusconference committee. American College of Chest Physicians/Society ofCritical Care Medicine. Chest. 1992;101:1644–55.24. Milne GL, Yin H, Brooks JD, Sanchez S, Jackson Roberts LJ, Morrow JD.Quantification of F2-isoprostanes in biological fluids and tissues as ameasure of oxidant stress. Methods Enzymol. 2007;433:113–26.25. Waikar SS, Sabbisetti VS, Bonventre JV. Normalization of urinary biomarkers tocreatinine during changes in glomerular filtration rate. Kidney Int. 2010;78:486–94.26. Dondorp AM, Desakorn V, Pongtavornpinyo W, Sahassananda D, Silamut K,Chotivanich K, Newton PN, Pitisuttithum P, Smithyman AM, White NJ, DayNP. Estimation of the total parasite biomass in acute falciparum malariafrom plasma PfHRP2. PLoS Med. 2005;2:e204.27. Dondorp AM, Angus BJ, Chotivanich K, Silamut K, Ruangveerayuth R,Hardeman MR, Kager PA, Vreeken J, White NJ. Red blood cell deformabilityas a predictor of anemia in severe falciparum malaria. AmJTrop Med Hyg.1999;60:733–7.28. Nuchsongsin F, Chotivanich K, Charunwatthana P, Omodeo-Sale F, TaramelliD, Day NP, White NJ, Dondorp AM. Effects of malaria heme products on redblood cell deformability. AmJTrop Med Hyg. 2007;77:617–22.29. Hardeman MR, Dobbe JG, Ince C. The laser-assisted optical rotational cellanalyzer (LORCA) as red blood cell aggregometer. Clin Hemorheol Microcirc.2001;25:1–11.30. Kidney Disease: Improving Global Outcomes (KDIGO) Acute Kidney InjuryWork Group. KDIGO clinical practice guideline for Acute kidney injury.Kidney Int. Suppl 2012. 2:1–138.31. Schwartz GJ, Munoz A, Schneider MF, Mak RH, Kaskel F, Warady BA, FurthSL. New equations to estimate GFR in children with CKD. J Am Soc Nephrol.2009;20:629–37.32. Staples A, LeBlond R, Watkins S, Wong C, Brandt J. Validation of the revisedSchwartz estimating equation in a predominantly non-CKD population.Pediatr Nephrol. 2010;25:2321–6.33. Zappitelli M, Parikh CR, Akcan-Arikan A, Washburn KK, Moffett BS, GoldsteinSL. Ascertainment and epidemiology of acute kidney injury varies withdefinition interpretation. Clin J Am Soc Nephrol. 2008;3:948–54.34. Chen S. Retooling the creatinine clearance equation to estimate kinetic GFRwhen the plasma creatinine is changing acutely. J Am Soc Nephrol.2013;24:877–88.35. Boutaud O, Moore KP, Reeder BJ, Harry D, Howie AJ, Wang S, Carney CK,Masterson TS, Amin T, Wright DW, Wilson MT, Oates JA, Roberts LJ.Acetaminophen inhibits hemoprotein-catalyzed lipid peroxidation andattenuates rhabdomyolysis-induced renal failure. Proc Natl Acad Sci USA.2010;107:2699–704.36. Spinelli SL, Lannan KL, Casey AE, Croasdell A, Curran TM, Henrichs KF,Pollock SJ, Milne GA, Refaai MA, Francis CW, Phipps RP, Blumberg N.Isoprostane and isofuran lipid mediators accumulate in stored red bloodcells and influence platelet function in vitro. Transfusion. 2014;54:1569–79.37. Morrow JD, Harris TM, Roberts LJ. Noncyclooxygenase oxidative formationof a series of novel prostaglandins: analytical ramifications for measurementof eicosanoids. Anal Biochem. 1990;184:1–10.38. Ferreira A, Balla J, Jeney V, Balla G, Soares MP. A central role for free hemein the pathogenesis of severe malaria: the missing link? J Mol Med (Berl).2008;86:1097–111.39. Nguansangiam S, Day NP, Hien TT, Mai NT, Chaisri U, Riganti M, DondorpAM, Lee SJ, Phu NH, Turner GD, White NJ, Ferguson DJ, Pongponratn E. Aquantitative ultrastructural study of renal pathology in fatal plasmodiumfalciparum malaria. Tropical Med Int Health. 2007;12:1037–50.40. Pfafferott C, Meiselman HJ, Hochstein P. The effect of malonyldialdehyde onerythrocyte deformability. Blood. 1982;59:12–5.41. Pfafferott C, Nash GB, Meiselman HJ. Red blood cell deformation in shearflow. Effects of internal and external phase viscosity and of in vivo aging.Biophys J. 1985;47:695–704.42. Fairley NH, Bromfield RJ. Laboratory studies in malaria and blackwater fever.Part II. Blackwater fever. Trans R Soc Trop Med Hyg. 1934;28:141–56.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript your next manuscript to BioMed Central and we will help you at every step:Plewes et al. BMC Infectious Diseases  (2017) 17:313 Page 12 of 12


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items