Risk Factors and Outcomes Stratified by Severity of Acute Kidney Injury in Malaria

Transcription

1 Risk Fctors nd Outcomes Strtified by Severity of Acute Kidney Injury in Mlri Kvith Srvu 1 *, Kumr Rishikesh 1, Chirg R. Prikh 2 * 1 Deprtment of Medicine, Ksturb Medicl College, Mnipl University, Mnipl, Krntk, Indi, 2 Section of Nephrology, Yle University School of Medicine, VA CT Helthcre System, nd the Progrm of Applied Trnsltionl Reserch, New Hven, Connecticut, United Sttes of Americ Abstrct Severe cute kidney injury (AKI) is known to hve prognostic vlue for in-hospitl outcomes in mlri. However, little is known bout the ssocition of AKI of lesser severity with mlril risk fctors nd outcomes nd such gp is becoming incresingly relevnt with the upsurge in the incidence of AKI due to Plsmodium flciprum mlri nd Plsmodium vivx mlri over the lst decde. We imed to identify risk fctors of AKI in mlri nd ssessed in-hospitl outcomes strtified by severity of AKI. We performed n observtionl study of 1,191 hospitlized mlri ptients enrolled between 2007 nd 2011 in tertiry cre cdemic center in Indi. Ptients were ctegorized bsed on pek serum cretinine into one of three groups: no AKI (,1.6 mg/dl), mild AKI ( mg/dl), nd severe AKI (.3 mg/dl). Plsmodium vivx ws the predominnt species (61.41%), followed by Plsmodium flciprum (36.41%) nd mixed infections with both the species (2.18%). Mild nd severe AKI were detected in 12% nd 5.6% of ptients, respectively. Mild AKI due to Plsmodium vivx (49%) nd Plsmodium flciprum (48.5%) ws distributed reltively eqully within the smple popultion; however, cses of severe AKI due to Plsmodium flciprum (80%) nd Plsmodium vivx (13%) ws significntly different (P,0.001). On history nd physicl exmintion, risk fctors for AKI were ge, bsence of fever, higher hert rte, lower distolic blood pressure, icterus, nd heptomegly. The only lbortory prmeter ssocited with risk of AKI on multivrite nlysis ws direct bilirubin. Ptients with mild nd severe AKI hd greter orgn complictions, supportive requirements, longer durtion of hospitl sty nd in-hospitl mortlity in dose-dependent reltionship, thn ptients with no AKI. Mild AKI is ssocited with significnt (P,0.05) morbidity compred to no AKI, nd future studies should ssess strtegies for erly dignosis of AKI nd prevent AKI progression. Cittion: Srvu K, Rishikesh K, Prikh CR (2014) Risk Fctors nd Outcomes Strtified by Severity of Acute Kidney Injury in Mlri. PLoS ONE 9(3): e doi: /journl.pone Editor: Luzi Helen Crvlho, Centro de Pesquis Rene Rchou/Fundção Oswldo Cruz (Fiocruz-Mins), Brzil Received September 10, 2013; Accepted Jnury 30, 2014; Published Mrch 13, 2014 Copyright: ß 2014 Srvu et l. This is n open-ccess rticle distributed under the terms of the Cretive Commons Attribution License, which permits unrestricted use, distribution, nd reproduction in ny medium, provided the originl uthor nd source re credited. Funding: KS ws recipient of the ICMR Interntionl Fellowship for Young Biomedicl Scientists nd is the principl investigtor of n extrmurl reserch project (grnt AMR/46/2011 ECD II) from the Indin Council of Medicl Reserch. CRP ws supported by the mentoring grnt from the Ntionl Institutes of Helth-K24DK The funders hd no role in study design, dt collection nd nlysis, decision to publish, or preprtion of the mnuscript. Competing Interests: The uthors hve declred tht no competing interests exist. * E-mil: kvithsrvu@gmil.com (KS); chirg.prikh@yle.edu (CRP) Introduction There re more thn 3.3 billion people living in countries with ongoing mlri trnsmission t risk of infection [1]. According to the WHO 2012 report, n estimted 219 million cses of mlri nd 600,000 deths occurred in 2010 [2] due to complictions. Acute kidney injury (AKI) is firly common nd serious compliction seen in cute flciprum mlri in dults nd older children. Depending on the definitions used for AKI, intensity of mlri trnsmission, ge ffected, infecting species, nd the cohort studied, incidence of AKI in mlri vries from 0.4% to 60% [3], [4]. Over the lst decde, there hs been n upsurge in the incidence of AKI due to Plsmodium flciprum(p. flciprum) mlri [5], [6], [7] nd reports of AKI due to Plsmodium vivx (P. vivx) mlri [8], [9]. Moreover, in certin prts of the world, AKI ssocited with mlri is the leding cuse of hospitliztion due to AKI [6]. The mortlity of ptients with AKI vries depending on the helth cre ccess nd vilbility nd hs rnged between 10% nd 75% in prior studies [4], [10]. Although the mortlity ssocited with severe AKI in mlri is well estblished, the prognostic importnce of less severe forms of AKI is not known. In view of the high mortlity rtes ssocited with AKI, it is criticl to identify the predictors of AKI in mlri nd dignose renl involvement erly to void the progression to severe AKI. There hs been little work investigting ssocited fctors of severe AKI in mlri. Furthermore, of the published literture, the studies hve either been smll [4], dted bck t lest decde [4], [6], lcking specificity of AKI severity nd independent vribles [4], [6], or devoid of multivrite nlysis [5], [11]. The objective of our study ws to conduct rigorous nlysis of ssocited fctors of AKI nd elucidte the contribution of the severity of AKI to other orgn complictions nd in-hospitl deth in mlri. Methods Study Design nd Ptients We performed retrospective cohort study by extrcting informtion in dt bstrction tool from dult mlri ptients, who were hospitlized t tertiry cre teching hospitl in Indi from Jnury 2007 to December Adults ($18 yers) of either gender nd microscopiclly proven to hve sexul forms of P. vivx, P. flciprum, or both with or without gmetocytes were PLOS ONE 1 Mrch 2014 Volume 9 Issue 3 e90419

2 included. Only cses with body mss index (BMI).16 Kg/M 2 were included becuse ptients with lower BMIs would hve low len body mss nd bseline serum cretinine. Cses with other non-mlril fever etiology, coexisting humn immunodeficiency virus infection, nd pre-existing chronic kidney diseses by history were excluded from this study. All lbortory tests for in-ptients of our tertiry cre hospitl re done in ssocited lbortories within hospitl. Ptients were mnged by the treting physicins s per their clinicl judgement nd ntionl guidelines for the tretment of mlri. All ptients with AKI were mnged with intrvenous fluids, diuretics, systemic lklinistion nd hemodilysis s per the decision of ttending clinicins in consulttion with nephrologists. Ethics Sttement Approvl for retrospective extrction of ll ptients dt from their clinicl records nd further nlysis ws grnted by our prent institutionl (Ksturb Medicl College, Mnipl University, Mnipl, Indi) ethics committee prior to commencement of this study. For our retrospective study, individul ptients informed consent ws not sought by our prent institutionl ethics committee, nd ll ptients records were kept nonymous. We would shre our deidentified study dt upon request fter necessry institutionl pprovls hve been obtined. Vribles Independent vribles. Mlril complictions were defined bsed on the guidelines for mngement of severe mlri by the WHO in 2012 [1]. Cerebrl mlri ws defined s impired consciousness, com, or multiple generlised convulsions within 24 hours. Severe nemi ws defined s Hb,7 g/dl; jundice s totl bilirubin.3 mg/dl; hyperprsitemi s prsite index.5% (prsite index ws expressed s percentge of prsitized erythrocytes on Leishmn s stined peripherl blood smer); pulmonry edem/acute Respirtory Distress Syndrome (ARDS) s bilterl spreding shdows on chest x-rys. Dependent vribles. Our primry outcome ws AKI, defined s serum cretinine $1.6 mg/dl. We divided the cohort into three ctegories bsed on the pek serum cretinine vlue during hospitliztion; serum cretinine,1.6 (No AKI), (Mild AKI) nd.3 mg/dl (Severe AKI). The severe AKI ctegory corresponded to the WHO renl filure definition of cretinine.3 mg/dl (.265 mmol/l) ny time during hospitliztion. Any ptient who required in-ptient renl replcement therpy ws clssified into the severe AKI ctegory. The secondry outcome ws in-hospitl mortlity. Sttisticl nlysis. Ctegoricl dt were summrized s frequency nd proportions by AKI ctegories; proportions were compred by the Chi-squre test or Fisher s exct test s pproprite. Continuous vribles were tested for normlity using the Kolmogorov-Smirnov test. Normlly distributed continuous vribles were reported s mens with stndrd devitions, nd comprisons of groups were performed by nlysis of vrince (ANOVA). Non-normlly distributed dt were summrized s medin s with interqurtile rnge (IQR) nd compred by the Kruskl-Wllis test. Univrite nlysis ws performed with bseline chrcteristics, clinicl vribles s independent vribles, nd AKI s the dependent vrible to determine the possible ssocited fctors for AKI mong mlri ptients. Vribles chieving vlue #0.1 in the univrite logistic regression nlysis were selected nd checked for multicollinerity. If multicollinerity ws present, only one of the fctors would be selected nd included in the multivrite logistic regression nlysis. All tests of significnce were two-sided, with p-vlue of,0.05 indicting sttisticl significnce. The dt were nlysed using the Sttisticl Pckge for the Socil Sciences version 16.0 (SPSS, Chicgo, IL, USA). Results A totl of 1,191 ptients with mlri were hospitlized between Jnury 2007 nd December Medicl records of 109 (9.15%) ptients could not be obtined, nd 232 (19.48%) cses were excluded due to pre-specified criteri s follows: ge, 18yers 96 ptients, microscopiclly negtive but rpid dignostic test (RDT) positive mlri cses 87 ptients, cses with concomitnt febrile illnesses (dengue, heptitis B, heptitis C, HIV, leptospirosis, filrisis, cinetobcter & Stphylococcus ureus sepsis, typhoid fever, prtyphoid fever, tuberculosis, UTI, neurocysticercosis nd chronic obstructive pulmonry disese nd lower respirtory distress syndrome due to Pseudomons eruginos infection 39 ptients, cses with BMI,16 kg/m 2 6 ptients, nd 4 ptients due to chronic kidney diseses. Of the 850 who fulfilled the eligibility criteri, serum cretinine vlues were not vilble in 26 ptients, nd the remining 824 ptients were nlyzed. P. vivx ws the predominnt species cusing mlri, responsible for the cses of 61.41% of the cohort, wheres P. flciprum nd mixed infection with P. vivx nd P. flciprum ccounted for 36.41% nd 2.18%, respectively. Demogrphic, Clinicl, nd Lbortory Chrcteristics cross AKI Ctegories for All Mlri Cses Among the study cohort, 677 (82.16%) ptients hd no AKI, 101 (12.26%) ptients hd mild AKI, nd 46 (5.58%) ptients hd severe AKI. The demogrphic nd clinicl chrcteristics of the groups re given in Tble 1. Mles contributed to.80% of cses in ll the ctegories. Ptients with AKI were older thn those with no AKI. However, there ws no incrementl reltionship of ge between mild AKI nd severe AKI. Significntly fewer ptients with severe AKI hd history of fever. With incresing severity of AKI, there ws correspondingly grded increse of vomiting, pleness, icterus, nd heptomegly. Previous history of mlri ws not ssocited with AKI. There ws grded increse in hert rte nd decrese in distolic blood pressure cross the severity of AKI. The men differences in respirtory rte, systolic blood pressure, nd xillry temperture were not significnt cross severity of AKI. There ws significnt grded decrese in hemoglobin, pltelet count, nd serum sodium, while in contrst, there ws grded increse in white blood cell (WBC) count, erythrocyte sedimenttion rte (ESR), totl nd direct bilirubin, sprtte minotrnsferse (AST), lnine minotrnsferse (ALT), lkline phosphtse (ALP) nd serum potssium levels with worsening severity of AKI. While significntly higher proportions of ptients with mild nd severe AKI hd microscopic hemturi (P = 0.007), levels of urinry protein between these groups were not sttisticlly different (P = 0.113). While P. vivx nd P. flciprum were represented eqully in the mild AKI ctegory, P. flciprum ccounted for 80% of the severe AKI ctegory, which ws highly significnt (Tble 2). The prsite index in the vilble ptients showed n increse in the mild nd severe AKI ctegories compred to the no AKI ctegory. Among totl 58.59% (498/850) study cohort, glucose 6 phosphte dehydrogense (G6PD) ws estimted to be U/gm Hb, U/gm Hb nd U/gm Hb in no AKI, mild AKI nd severe AKI ctegories respectively. Between AKI ctegories no significnt difference in men G6PD vlue ws determined by one wy PLOS ONE 2 Mrch 2014 Volume 9 Issue 3 e90419

3 Tble 1. Demogrphic nd clinicl chrcteristics of the ptients strtified by AKI ctegories. No AKI Mild AKI Severe AKI Chrcteristics N = 677 N = 101 N = 46 P vlue* Demogrphic Age, yers 33.3 (613.2) 43.2 (615.7) 38.9 (612.6),0.001 Gender, Mle 551 (81.4) 89 (88.1) 40 (87) 0.18 History History of fever 672 (99.4) 101 (100) 40 (88.9),0.001 Dirrhe 39 (5.8) 4 (4) 5 (10.9) 0.25 Vomiting 229 (33.9) 38 (37.6) 25 (54.3) Physicl exmintion Hert rte, bets/min 88.5 (614.1) 92.2 (617.0) 92.7 (615.2) Respirtory rte, breths/min 22.0 (65.9) 23.4 (68.3) 23 (67.6) 0.09 Systolic blood pressure, mm of Hg (613.3) (617.4) (617) 0.16 Distolic blood pressure, mm of Hg 75.7 (69.1) 72.8 (611.4) 71.9 (614.4) Axillry temperture, uf b (65.2) (61.7) 99.5 (61.0) 0.61 Pllor 119 (17.6) 21 (20.8) 20 (43.5),0.001 Icterus 154 (22.8) 43 (42.6) 30 (65.2),0.001 Splenomegly 257 (38.3) 34 (33.7) 30 (66.7),0.001 Heptomegly 188 (28.2) 38 (37.6) 34 (75.6),0.001 Continuous vribles re described s Men (6 Stndrd Devition) nd ctegoricl vribles s n(%). b To convert temperture to uc=[uf 32]*5/9. *P vlues,0.05 re shown in bold. doi: /journl.pone t001 ANOVA [F (2, 495) = 0.45, P = 0.64]. A totl of 6.02% (30/498) ptients were found to hve G6PD deficiency in our study cohort. Out of this, 90% [27/30; men 6 SD = U/gm Hb] were in no AKI ctegory, 6.67% [2/30, 7.6 & 7.8 U/gm Hb] in mild AKI ctegory nd 3.33% [1/30, 7.8 U/gm Hb] were in severe AKI ctegory. All these ptients were treted with weekly dose of primquine to prevent the occurrence of primquine induced hemolysis. Progression of renl filure. At lest two vlues of serum cretinine were vilble in 309 ptients (mong 200 ptients with no AKI, 74 with mild AKI, nd 35 with severe AKI on the dy of dmission). Sixteen ptients progressed from no AKI t dmission to mild AKI during hospitliztion (16/200, 8%). Eight ptients (8/74, 10.81%) progressed from mild AKI to severe AKI, nd two of them required dilysis. No ptients progressed from the no AKI ctegory to severe AKI. Demogrphic, Clinicl, nd Lbortory Chrcteristics cross AKI Ctegories mong P. vivx nd P. flciprum Cohort (Tbles 3 nd 4) P. vivx cohort comprised of 450 ptients with no AKI, 50 ptients with mild AKI nd, 6 ptients with severe AKI. P. flciprum cohort comprised of 214 ptients with no AKI, 49 ptients with mild AKI nd 37 ptients with severe AKI. The demogrphic nd clinicl chrcteristics of both cohorts re depicted in Tble 3 fter excluding mixed infection cses due to reltively very smll cohort (n = 18). Mle prepondernce ($80%) ws noted in both species cross ll AKI ctegories. Among both species, ptients with AKI were noted to be older thn those with no AKI without hving ny incrementl reltionship of ge difference between the mild nd severe AKI ctegories. History of fever ws found to be the sme (,100%) in ll AKI ctegories of P. vivx cohort. However, in the P. flciprum cohort, significntly fewer ptients with severe AKI hd history of fever (P,0.001). In the P. flciprum cohort, there ws significnt grded increse in pllor (P = 0.004), icterus (P,0.001), nd heptomegly (P,0.001) with incresing severity of AKI, while in the P. vivx cohort, there ws significnt grded increse in icterus (P = 0.03) nd splenomegly (P = 0.01) with incresing severity of AKI. In ddition, the P. vivx cohort exhibited significnt grded increse in hert rte (P,0.001) nd significnt decrese in both systolic blood pressure (P = 0.001) nd distolic blood pressure (P,0.001) cross ll AKI severity ctegories. In the P. flciprum cohort, none of the vitl physicl prmeters were significntly different cross ll AKI ctegories. In the P. vivx cohort, there ws significnt grded decrese in pltelet count (P,0.001) nd significnt increse in WBC count (P = 0.008), plsm glucose (P = 0.005), totl nd direct bilirubin (P,0.001), AST (P = 0.002), nd ALP (P = 0.003) with incresing severity of AKI. In P. flciprum, there ws significnt grded decrese in hemoglobin (P,0.001), pltelet count (P = 0.002) nd serum sodium (P = 0.008) with worsening severity of AKI, wheres, significnt grded increse ws noted for ESR (P, 0.001), totl nd direct bilirubin (P,0.001), AST (P,0.001), ALP (P = 0.01) nd proportion of microscopic hemturi (P = 0.008). As evident in Tble 4, in P. flciprum cohort significnt difference in men plsm glucose level ws determined by one wy ANOVA [F (2, 262) = 3.256, P = 0.04]. A Tukey post-hoc test reveled tht there ws significnt (P = 0.047) increse in men glucose level in mild AKI cohort compred with no AKI cohort. Fctors Associted with AKI (Tbles 5 nd 6) We creted two seprte regression nlyses to identify vribles tht re independently ssocited with AKI. In the PLOS ONE 3 Mrch 2014 Volume 9 Issue 3 e90419

4 Tble 2. Lbortory chrcteristics of the ptients strtified by AKI ctegories. Chrcteristics No AKI Mild AKI Severe AKI P vlue* N = 677 N = 101 N = 46 Hemtologicl prmeters Hemoglobin, g/dl [11.1, 14.4] [10, 14.6] [8.6, 12.6],0.001 WBC count, cells/ml 5200 [4200, 6700] 6100 [5100, 8200] 8500 [5400, 11650],0.001 Pltelet count, cells/ml [51000, ] [30000, ] [20000, 55500],0.001 ESR, mm in 1 hour 28 [14, 51] 35 [16, 82] 68 [33, 107],0.001 Plsm glucose, mg/dl 113 [97, 135] 121 [100, 160] 113 [75, 161] Liver function tests Totl bilirubin, mg/dl 1.5 [1.0, 2.6] 2.4 [1.3, 5.0] 15.6 [5.7, 25.9],0.001 Direct bilirubin, mg/dl 0.5 [0.3, 1.0] 0.9 [0.5, 3.1] 11.8[4.5, 20.9],0.001 AST, U/L 38 [27, 58] 47 [33, 73] 93 [56, 148],0.001 ALT, U/L 37 [24, 57] 45 [29, 63] 51 [27, 70] ALP, U/L 88 [67, 115] 98 [75, 145] 148 [97, 200],0.001 Renl function tests nd electrolytes Blood ure, mg/dl 28 [23, 37] 52 [36, 71] 176 [117, 229],0.001 Serum cretinine, mg/dl 1.1 [1, 1.3] 1.8[1.6, 2.2] 5.2 [4.2, 7.8],0.001 Serum sodium, meq/l 135 [131, 137] 133 [129, 136] 131 [128, 133],0.001 Serum potssium, meq/l 3.9 [3.6, 4.3] 4 [3.5, 4.6] 4.6 [4, 5.4],0.001 Urine exmintion Dipstick protein $1+ (30 mg/dl) 204 (59%) 45 (69.2%) 22 (73.3%) Urine RBC $0 10/hpf 27 (8%) 10 (15.4%) 7 (24.1%) Prsite relted Prsite Index, % b 0.3 [0.2, 0.8] 0.65 [0.2, 2.0] 0.6 [0.3, 2.2] Species of Plsmodium P. vivx 450 (66.5%) 50 (49%) 6 (13%),0.001 c P. flciprum 214 (31.6%) 49 (48.5%) 37 (80.4%) Both 13 (1.9%) 2 (2%) 3 (6.5%) Abbrevitions: WBC, White blood cell; ESR, erythrocyte sedimenttion rte; AST, sprtte minotrnsferse; ALT, lnine minotrnsferse; ALP, lkline phosphtse; RBC, red blood cell; hpf, high power field. All continuous vribles re described s Medin [Inter Qurtile Rnge] nd ctegoricl vribles s n(%). b Prsite index ws vilble in 122, 26 nd 22 cses with no AKI, mild AKI nd severe AKI respectively. c P vlue is for Chi squre test compring P.flciprum with P.vivx. *P vlues,0.05 re shown in bold. doi: /journl.pone t002 first nlysis, we included vribles from ptient histories nd physicl exmintions (before ny lbortory tests were performed) to identify signs nd symptoms tht would be hrbingers of AKI. In the second nlysis, we wnted to ssocite other lbortory bnormlities tht would be present with AKI. On univrite nlysis, incresing ge, bsence of fever, history of vomiting, higher hert rte, lower distolic blood pressure, nd higher respirtory rte, presence of pllor, icterus, nd heptomegly were ssocited with AKI. However, with multivrite nlysis of fctors from history nd physicl exmintion, only ge, bsence of fever, higher hert rte, lower distolic blood pressure, icterus, nd heptomegly were independently ssocited with AKI (Tble 5). Icterus ws the strongest ssocited fctor for AKI, followed by heptomegly nd incresing ge. Lbortory prmeters like lower hemoglobin, higher WBC count, lower pltelets, elevted ESR, totl nd direct bilirubin, AST, ALP, lower serum sodium, higher serum potssium, nd infection with P. flciprum were significnt on univrite nlysis. But multivrite nlysis of lbortory fctors demonstrted tht only direct bilirubin ws independently ssocited with AKI (Tble 6). Fctors Associted with AKI by Mlri Species (Tbles 7 nd 8) As described bove for Tble 5 nd 6, we creted two seprte regression nlyses ech for P. vivx nd P. flciprum to identify vribles those re independently ssocited with AKI. On univrite nlysis in the P. vivx cohort, ge, hert rte, respirtory rte, systolic nd distolic blood pressure, nd icterus were ssocited with AKI; however, on multivrite nlysis of fctors from history nd physicl exmintion for the P. vivx cohort (Tble 7), only ge, hert rte, respirtory rte nd distolic blood pressure were independently ssocited with AKI. Univrite nlysis of the P. flciprum cohort reveled tht ge, pllor, icterus nd heptomegly were ssocited with AKI, while in multivrite nlysis, only bsence of fever, icterus nd heptomegly were independently ssocited with AKI. Among lbortory prmeters, univrite nlysis showed higher WBC count, PLOS ONE 4 Mrch 2014 Volume 9 Issue 3 e90419

5 Tble 3. Mlri species wise demogrphic nd clinicl chrcteristics of the ptients strtified by AKI ctegories. Plsmodium vivx (N = 506) Plsmodium flciprum (N = 300) No AKI Mild AKI Severe AKI No AKI Mild AKI Severe AKI Chrcteristics N = 450 N = 50 N = 06 P vlue* N = 214 N = 49 N = 37 P vlue* Demogrphic Age, yers 32.3(613.0) 45.9(616.2) 40.7(618.3), Gender, mle 367(81.6) 48 (96.0) 06 (100.0) (80.4) 39 (79.6) 31 (83.8) 0.89 History History of fever 446 (99.3) 50 (100.0) 06 (100.0) (99.5) 49 (100.0) 31 (86.1),0.001 Dirrhe 22 (4.9) 02 (4.0) (7.5) 02 (4.1) 05 (13.5) 0.27 Vomiting 131 (29.2) 14 (28.0) 04 (66.7) (43.5) 23 (46.9) 20 (54.1) 0.48 Physicl exmintion Hert rte, bets/min , Respirtory rte, breths/min Systolic blood pressure, mm of Hg Distolic blood pressure, mm of Hg , Axillry temperture, uf b Pllor 65 (14.5) 04 (8.0) 01 (16.7) (23.5) 17 (34.7) 18 (48.6) Icterus 68 (15.1) 14 (28.0) 02 (33.3) (37.6) 28 (57.1) 26 (70.3),0.001 Splenomegly 161 (35.9) 12 (24.0) 05 (83.3) (43.3) 20 (40.8) 22 (61.1) 0.12 Heptomegly 104 (23.2) 12 (24.0) 04 (66.7) (39.1) 24 (49.0) 27 (75.0),0.001 Continuous vribles re described s Men (6 Stndrd Devition) nd ctegoricl vribles s n(%). b To convert temperture to uc=[uf 32]*5/9. *P vlues,0.05 re shown in bold. doi: /journl.pone t003 PLOS ONE 5 Mrch 2014 Volume 9 Issue 3 e90419

6 Tble 4. Mlri species wise Lbortory chrcteristics of the ptients strtified by AKI ctegories. Plsmodium vivx (N = 506) Plsmodium flciprum (N = 300) No AKI Mild AKI Severe AKI No AKI Mild AKI Severe AKI Chrcteristics N = 450 N = 50 N = 06 P vlue* N = 214 N = 49 N = 37 P vlue* Hemtologicl prmeters Hemoglobin, g/dl 13.3 [11.6, 14.6] 13.5 [11.6, 14.6] 12.6 [10.8, 14.4] [10.1, 13.5] 11.4 [7.6, 14.4] 10.3 [8, 11.7],0.001 WBC count, cells/ml 5200 [4200, 6500] 6100 [5000, 8000] 7000 [3700, 8950] [4300, 7275] 6100 [5200, 8800] 8900 [5750, 12050] 0.13 Pltelet count, cells/ml [56000, ] [23250, ] [18250, 50500], [42500, ] [30500, ] [21500, 70000] ESR, mm in 1 hour 22 [12.5, 45] 25 [14, 50] 40 [10, 87] [18, 63.5] 57 [17, 106] 82.5 [32.7, 120],0.001 Plsm glucose, mg/dl 114 [97.2, 137] [105, 153.5] 187 [87, 308] [95, 132] 121 [96, 162] 105 [71, 142] 0.04 Liver function tests Totl bilirubin, mg/dl 1.4 [0.9, 2.1] 2.0 [1.2, 3.4] 7.9 [2.1, 20.2], [1.1, 3.7] 2.8 [1.2, 6.6] 17.4 [5.6, 29],0.001 Direct bilirubin, mg/dl 0.5 [0.3, 0.7] 0.7 [0.5, 1.7] 6.4 [1.3, 18.2], [0.4, 2.1] 1.5 [0.6, 5.6] 14.1 [4.3, 23.8],0.001 AST, U/L 35 [26, 49.7] 46 [32.7, 65.5] 64 [27.5, 150.5] [29, 84] 50 [32.7, 80] 94 [59, 138],0.001 ALT, U/L 33.5 [23, 51.2] 46 [28, 66.5] 39 [12.5, 70.2] [29, 75] 42 [30, 56] 45 [30, 67] 0.54 ALP, U/L 82 [64, 107] [79.5, 155.7] 123 [103.7, 240.7] [75, 142.7] 95 [74, 119] 148 [89, 182] 0.01 Renl function tests nd electrolytes Blood ure, mg/dl 27 [21, 35] 51 [37, 66] 96 [63.2, 190.2], [25, 45] 63 [34.2, 91.2] 178 [129, 244.5],0.001 Serum cretinine, mg/dl 1.1 [1.1, 1.3] 1.8 [1.6, 2.0] 4.0 [3.3, 9.5], [1.0, 1.3] 1.8 [1.6, 2.3] 5.2 [4.3, 7.4],0.001 Serum sodium, meq/l 135 [132, 137] [130.2, 136] 139 [131, 144] [130, 137] 132 [128, 135] 131 [128, 132.5] Serum potssium, meq/l 3.8 [3.5, 4.2] 3.9 [3.6, 4.4] 4.2 [3.8, 4.7] [3.6, 4.4] 4.1 [3.5, 4.7] 4.8 [4.1, 5.4] 0.35 Urine exmintion Dipstick protein $1+ (30 mg/dl) 118 (54.1) 23 (71.9) 03 (60) (67.5) 22 (66.7) 16 (72.7) 0.91 Urine RBC $0 10/hpf 18 (8.5%) 05 (15.6%) (6.6%) 05 (15.6%) 06 (28.6%) Prsite relted Prsite Index, % b 0.2 [0.2, 0.6] 0.3 [0.1, 1.3] [0.2, 1.1] 1.3 [0.4, 3.5] 0.4 [0.2, 2] 0.06 Abbrevitions: WBC, White blood cell; ESR, erythrocyte sedimenttion rte; AST, sprtte minotrnsferse; ALT, lnine minotrnsferse; ALP, lkline phosphtse; RBC, red blood cell; hpf, high power field. All continuous vribles re described s Medin [Inter Qurtile Rnge] nd ctegoricl vribles s n(%). b Prsite index ws vilble in 51, 09 nd 00 cses of P. vivx with no AKI, mild AKI nd severe AKI respectively; 69, 16 nd 19 cses of P. flciprum with no AKI, mild AKI nd severe AKI respectively. *P vlues,0.05 re shown in bold. doi: /journl.pone t004 PLOS ONE 6 Mrch 2014 Volume 9 Issue 3 e90419

7 Tble 5. Univrite nd Multivrite Anlysis for History nd Physicl Exmintion Fctors Associted with AKI. Chrcteristic OR (95% CI)* Adjusted OR (95% CI)* Age 1.04 ( ) 1.05 (1.03, 1.06) Fever 0.17 (0.05, 0.63) 0.2 (0.05, 0.8) Vomiting 1.46 (1.02, 2.11) 1.13 (0.75, 1.71) Dirrhoe 1.07 (0.50, 2.25) Hert rte 1.02 (1.01, 1.03) 1.02 (1.003, 1.030) Respirtory rte 1.03 (1.00, 1.05) 1.03 (0.997, 1.06) Systolic blood pressure 0.99 (0.98, 1.01) b Distolic blood pressure 0.97 (0.95, 0.99) 0.97 (0.95, 0.99) Axillry temperture 0.99 (0.96, 1.03) Pllor 1.81 (1.2, 2.73) 1.04 (0.64, 1.68) Icterus 3.34 (2.31, 4.83) 2.45 (1.58, 3.82) Splenomegly 1.26 (0.88, 1.81) Heptomegly 2.48 (1.72, 3.57) 1.62 (1.06, 2.46) Spontneous bleeding 2.95 (0.95, 9.13) Abbrevitions: OR, odds rtio; CI, confidence intervl. All insignificnt vribles by univrite model were not included in the multivrite model. b Systolic blood pressure hd multicollinerity with distolic blood pressure. * Confidence intervls tht do not overlp the null vlue of OR = 1 re shown in bold font. doi: /journl.pone t005 lower pltelets, rised totl nd direct bilirubin, nd elevted AST s significnt fctors for AKI in the P. vivx cohort. Furthermore, multivrite nlysis (Tble 8) showed independent ssocition of higher WBC count nd decresed pltelets with AKI in the P. vivx cohort. Univrite nlysis showed decresed hemoglobin, rised ESR, rised totl nd direct bilirubin, elevted AST, nd decresed serum sodium being significnt in P. flciprum cohort; multivrite nlysis showed independent ssocition of decresed hemoglobin nd elevted direct bilirubin with AKI in the P. flciprum cohort. Associted Complictions with AKI (Tbles 9 nd 10) Tble 9 shows the ssocition of AKI with other complictions, wherein complictions hve been tken s dependent vrible nd AKI ctegory s independent vrible. Ptients with severe AKI hd higher odds of cerebrl mlri, jundice, severe nemi, nd spontneous bleeding compred to ptients with no AKI; even the mild AKI ctegory hd significnt risk of cerebrl mlri, jundice, severe nemi, nd pulmonry edem/ards compred to the no AKI ctegory. In totl mlri cohort, cerebrl mlri, jundice, spontneous bleeding, severe nemi nd deth were significntly (P,0.04) different mong AKI ctegories. In intr Tble 6. Univrite nd Multivrite Anlysis for Lbortory Fctors Associted with AKI. Chrcteristics OR (95%CI)* Adjusted OR (95% CI)* Hemoglobin 0.87 (0.81, 0.93) 0.92 (0.84, 1.00) WBC count/ (1.03, 1.16)) 1.01 (0.97, 1.04) Pltelet count/ (0.88, 0.95) 0.97 (0.93, 1.01) Erythrocyte sedimenttion rte 1.01 ( ) Totl bilirubin 1.17 (1.12, 1.21) Direct bilirubin 1.21 (1.15, 1.26) 1.14 (1.07, 1.20) Asprtte minotrnsferse 1.01 (1.01, 1.01) Alnine minotrnsferse 1.00 (0.999, 1.00) Alkline phosphtse 1.01 (1.00, 1.01) 1.00 (0.996, 1.00) Serum sodium 0.96 (0.92, 0.99) 0.96 (0.92, 1.00) Serum potssium 1.08 (0.99, 1.17) 0.99 (0.88, 1.12) P. flciprum mlri b 3.23 (2.22, 4.69) 1.55 (0.94, 2.55) Abbrevitions: OR, odds rtio; CI, confidence intervl. Erythrocyte sedimenttion rte ws not included becuse of multicollinerity with hemoglobin; totl bilirubin nd sprtte minotrnsferse were not included becuse of multicollinerity with direct bilirubin; lnine minotrnsferse ws not included due to insignificnt OR (95% CI) by univrite nlysis. b Comprison with P.vivx. * Confidence intervls tht do not overlp the null vlue of OR = 1 re shown in bold. doi: /journl.pone t006 PLOS ONE 7 Mrch 2014 Volume 9 Issue 3 e90419

8 Tble 7. Univrite nd Multivrite Anlysis for History nd Physicl Exmintion Fctors Associted with AKI by Mlri Species. Plsmodium vivx Plsmodium flciprum Chrcteristics OR (95% CI)* Adjusted OR (95% CI)* OR (95% CI)* Adjusted OR (95% CI)* Age 1.06 (1.04, 1.08) 1.06 (1.04, 1.08) 1.02 (1.00, 1.04) 1.02 (1.00, 1.04) Fever 0.07 (0.01, 0.65) 0.07 (0.01, 0.69) Dirrhoe 0.72 (0.16, 3.14) 1.09 (0.43, 2.77) Vomiting 1.15 (0.63, 2.08) 1.30 (0.79, 2.15) Hert rte 1.03 (1.01, 1.05) 1.03 (1.01, 1.05) 1.01 (0.99, 1.02) Respirtory rte 1.07 (1.01, 1.13) 1.08 (1.02, 1.15) 0.99 (0.96, 1.03) Systolic blood pressure 0.97 (0.95, 0.99) b 1.00 (0.99, 1.02) Distolic blood pressure 0.96 (0.94, 0.98) 0.95 (0.92, 0.98) 0.98 (0.95, 1.00) 0.98 (0.95, 1.01) Axillry temperture 0.99 (0.95, 1.05) 0.99 (0.95, 1.05) Pllor 0.58 (0.22, 1.51) 2.24 (1.31, 3.82) 1.77 (0.97, 3.19) Icterus 2.24 (1.19, 4.23) 1.99 (0.98, 4.05) 2.81 (1.67, 4.71) 1.94 (1.07, 3.49) Splenomegly 0.78 (0.43, 1.42) 1.28 (0.77, 2.12) Heptomegly 1.32 (0.71, 2.46) 2.33 (1.39, 3.91) 1.92 (1.08, 3.42) Abbrevitions: OR, odds rtio; CI, confidence intervl. Univrite nlysis did not yield significnt difference. b Vrible not included ws systolic blood pressure becuse of multicollinerity with distolic blood pressure in P. vivx. * Confidence intervls tht do not overlp the null vlue of OR = 1 re shown in bold. doi: /journl.pone t007 species nlysis only jundice nd pulmonry edem/ards differed significntly (P,0.05) mong AKI ctegories in P. vivx, wheres in P. flciprum cohort cerebrl mlri, jundice, severe nemi nd deth were significntly different (P,0.05) (Figure 1). Severe AKI s single orgn dysfunction occurred in 8.7% (4/46) of ptients nd s prt of multi-orgn dysfunction in 91.3% (42/ 46). Even in ptients with mild AKI, 51.5% hd evidence of other orgn dysfunction. Both in totl mlri cohort nd intr species cohort, ptients with mild nd severe AKI hd incrementl requirements for blood products, mechnicl ventiltion, ICU cre nd longer hospitl sty, except no requirement for mechnicl ventiltion in severe AKI of P. vivx cohort (Figure 2). Hemodilysis ws required in 47.8% of cses of severe AKI. No ptient with mild AKI hd undergone hemodilysis. Ptients received medin dilysis of 2.5 (IQR 1.75, 5.52). The severe AKI ptients were more likely to die nd sty longer in the hospitl thn those with mild AKI nd no AKI. Ptients with mild AKI were lmost three times s likely to sty longer thn 7 dys, nd those with severe AKI were eight times s likely to sty longer thn 7 dys compred to ptients with no AKI. The medin durtion of hospitliztion were 5 (IQR 4, 7), 7 (IQR 5, 8.5) nd 9 (IQR 5.5, 15) dys for ptients with no, mild, nd severe AKI, respectively. Tble 8. Univrite nd Multivrite Anlysis for Lbortory Fctors Associted with AKI by Mlri Species. Plsmodium vivx Plsmodium flciprum Chrcteristics OR (95% CI)* Adjusted OR (95% CI)* OR (95% CI)* Adjusted OR (95% CI)* Hemoglobin 0.99 (0.88, 1.13) 0.86 (0.78, 0.94) 0.86 (0.77, 0.97) WBC count/ (1.05, 1.30) 1.21 (1.06, 1.38) 1.03 (0.98, 1.07) Pltelet count/ (0.82, 0.94) 0.88 (0.81, 0.96) 0.95 (0.91, 0.99) 0.99 (0.94, 1.04) Erythrocyte sedimenttion rte 1.02 (1.01, 1.02) b Glucose 1.01 (1.00, 1.01) 1.01 (1.00, 1.01) 1.01 (1.00, 1.01) 1.01 (1.00, 1.01) Totl bilirubin 1.12 (1.04, 1.20) b 1.14 (1.09, 1.19) b Direct bilirubin 1.14 (1.05, 1.24) 1.06 (0.96, 1.17) 1.18 (1.11, 1.25) 1.15 (1.07, 1.23) Asprtte minotrnsferse 1.01 (1.00, 1.01) 1.01 (0.99, 1.01) 1.01 (1.00, 1.01) 0.99 (0.99, 1.01) Alnine minotrnsferse 1.00 (0.99, 1.01) 1.00 (0.99, 1.00) Alkline phosphtse 1.00 (1.00, 1.01) 1.00 (0.99, 1.00) Serum sodium 0.99 (0.96, 1.04) 0.93 (0.88, 0.97) 0.95 (0.89, 1.01) Abbrevitions: OR, odds rtio; CI, confidence intervl. Univrite nlysis did not yield significnt difference. b Prmeters not included in multivrite nlysis due to multicollinerity: For P. vivx totl bilirubin with direct bilirubin; For P. flciprum totl bilirubin with direct bilirubin nd ESR with hemoglobin. * Confidence intervls tht do not overlp the null vlue of OR = 1 re shown in bold. doi: /journl.pone t008 PLOS ONE 8 Mrch 2014 Volume 9 Issue 3 e90419

9 The in-hospitl mortlity ws 0.3%, 2%, nd 15.2% for ptients with no, mild, nd severe AKI, respectively. In mild AKI ctegory, pulmonry edem/ards, requirement for mechnicl ventiltion, nd deth were eqully distributed mong P.vivx nd P. flciprum, wheres cerebrl mlri, jundice, ICU cre, nd durtion of hospitliztion were non-significntly (P.0.05) incresed in P. flciprum except requirement for blood products (OR (95% CI) = 2.8 (1.1, 7.3) nd severe nemi (OR (95% CI) = 9.2 (1.1, 76.8) (Tble 10). In the severe AKI ctegory, ll complictions nd supportive requirements s mentioned bove were non-significntly (P.0.05) incresed in P. flciprum, except durtion of hospitliztion nd spontneous bleeding which were incresed in P. vivx ctegory. Mortlity ws seen in totl nine ptients with AKI. One P. vivx ptients with mild AKI who died lso hd PE/ARDS. One P. flciprum ptient with mild AKI who died lso hd cerebrl mlri, PE/ARDS, severe nemi nd jundice. Among six P. flciprum ptients with severe AKI who died, one hd PE/ARDS, nother hd both severe nemi nd jundice, nd other four hd jundice. Single ptient with mixed infection nd severe AKI who died lso hd PE/ARDS nd jundice. Out of seven ptients with severe AKI (six P. flciprum nd one mixed infection) hemodilysis ws dministered to three severe AKI ptients. Discussion Most previous publictions of AKI in mlri hve described the clinicl course of severe AKI in the setting of mlri, nd severl of these only focused on the P.flciprum species [6], [10], [12], [13], [14]. Our objective ws to determine whether less severe forms of AKI hd n impct on morbidity nd mortlity in ptients hospitlized for mlri. Our results illustrte tht even ptients with mild AKI hd higher odds of complictions, such s cerebrl mlri, jundice, severe nemi, nd pulmonry edem/ ARDS, compred to ptients who did not develop AKI. Similrly, mild AKI incresed the need for supportive requirements nd mrkedly incresed the length of hospitl sty. However, mild AKI ws not ssocited with higher mortlity. The occurrence of severe AKI in mlri is recognized to imprt grim prognosis [10]. The incidence of severe AKI in the present cohort of cute mlri ws 5.58%, nd 47.8% of the severe AKI cses required hemodilysis. The low incidence of severe AKI compred to other studies cn be explined by our inclusion of the full spectrum of hospitlized mlri ptients nd the predominnce of P.vivx species [12]. The requirement of dilysis ws similr to tht of Thnchrtwet et l. [12], lthough lower thn the reported 73 80% by other studies [6], [13], [15]. The 15.2% severe AKI mortlity rte ws in concordnce with some studies, [4], [6] but lower thn the rnge of 26 32% reported by others [10], [12]. Severe AKI ws ssocited with higher odds of complictions, requirement of supportive cre, nd durtion of hospitliztion in conformity with other studies [10], [13]. In view of the high mortlity ssocited with severe AKI nd ssocited morbidity even with mild AKI, it is prmount to detect mild AKI erly to prevent progression to severe AKI. Hence, we decided to identify the fctors ssocited with mild nd severe AKI. Fctors Associted with AKI in Mlri On histories nd physicl exms, incresing ge, hert rte, bsence of fever, lower distolic blood pressure, nd presence of icterus nd heptomegly were independently ssocited with presence of AKI. This informtion cn be pplied in ny helthcre setting to identify ptients who hve higher likelihood of AKI in the presence of mlri. Our nlysis, which included lbortory prmeters, confirmed tht heptic dysfunction such s direct bilirubinemi is independently ssocited with AKI. Thus, independent ssocition of heptomegly nd bilirubin suggest tht heptic complictions re closely tied to development of AKI. With our study design, it is impossible to discern if AKI preceded liver bnormlities or vice vers. Future studies should investigte the temporl ssocition using more sensitive mesures such s biomrkers of AKI tht re elevted much before serum cretinine [16]. Hyperbilirubinemi plys n importnt role in the pthogenesis of AKI. It lters the renl hemodynmics [17], [18] nd depresses the crdic function [19]. Bilirubin is known to be toxic for renl tubulr cells s shown in obstructive jundice ptients in the presence of dditionl fctors like hypovolemi nd hypoxi [20]. Ptients with history of fever probbly sought medicl ttention erlier thn those without history of fever nd thereby were protected ginst severe AKI. However, we cnnot rule out probble recll bis bout fever by the ptients. We found distolic blood pressure to be negtively ssocited with AKI. Lower Figure 1. Assocition of AKI with complictions by mlri species. *= P,0.05 cross AKI ctegories. doi: /journl.pone g001 PLOS ONE 9 Mrch 2014 Volume 9 Issue 3 e90419

10 Figure 2. Assocition of AKI with supportive requirements by mlri species. *= P,0.05 cross AKI ctegories. doi: /journl.pone g002 distolic blood pressure is reflection of systemic vsodilttion which results in decresed effective blood volume nd reduced blood flow to the kidney [21], [22]. This would suggest tht mjority of the erly mild AKI is due to pre-renl zotemi, providing n opportunity to intervene erly to prevent progression of mild AKI to severe AKI. Hert rte ws positively ssocited with AKI which is reflection of shock. We lso did not find ny ssocition with prsite index. Moreover, Ncher et l. hs demonstrted tht schizont count nd their reltive proportion to ring forms, were not risk fctors for severe AKI, implying sequestrtion ws not the most importnt determinnt of severe AKI in mlri [23]. Fctors Associted with AKI by Mlri Species Plsmodium vivx. Tble 7 shows ssocition of ge, hert rte, nd respirtory rte with AKI, but negtive ssocition with distolic blood pressure. Thus, it ppers tht AKI in P. vivx is primrily due to hemodynmic ltertion in kidney. Plsmodium flciprum. Hyperbilirubinemi is ssocited with AKI in P. flciprum. This is supported by ssocition of AKI with icterus, heptomegly (Tble 7), nd direct bilirubin. Negtive ssocition of AKI with hemoglobin (Tble 8) reflects hemolysis contributing to hyperbilirubinemi. We would like to emphsize tht lthough there re incresing reports of P. vivx ssocited AKI, we observed tht the P. vivx contributed to hlf of the mild AKI ctegory but only to 13% of the severe AKI ctegory [8], [9]. However, becuse P. vivx is the Tble 9. Associtions of AKI with complictions nd supportive requirements. Mild AKI (N = 101) Severe AKI (N = 46) Chrcteristics N (%) Odds Rtio (95% CI) * N (%) Odds Rtio (95% CI) * Associted complictions Cerebrl mlri 6 (5.9) 7.1 (2.2, 22.3) 6 (13.0) 16.8 (5.2, 54.4) Jundice 37 (38.5) 2.5 (1.6, 4.0) 36 (78.3) 14.6 (7.1, 30.2) Spontneous bleeding 1(1.0) 0.8 (0.1, 6.8) 4 (8.7) 7.9 (2.3, 27.5) Severe nemi 9 (9.3) 3.9 (1.7, 9.3) 6 (14.0) 6.3 (2.3, 17.1) Pulmonry edem/ards 7 (6.9) 2.4 (1.01, 5.9) 2 (4.3) 1.5 (0.3, 6.6) Supportive requirements Blood/Blood product 25(24.8) 3.9 (2.3, 6.7) 30 (65.2) 22.5 (11.5, 43.9) Mechnicl Ventiltion 6 (6.0) 4.7 (1.6, 13.6) 12 (26.1) 26.2 (10.3, 66.4) Inotropes (4.3) 6.1 (1.2, 32.4) ICU cre 21 (19.8) 6.2 (3.3, 11.6) 21 (46.7) 21.9 (10.8, 44.3).7 dys of hospitliztion 37 (36.6) 2.7 (1.7, 4.3) 28 (62.2) 7.7 (4.1, 14.6) Deth 2 (2.0) 6.8 (0.9, 48.9) 7 (15.2) 60.5 (12.2, 300.9) Abbrevitions: CI, confidence intervl; ARDS, cute respirtory distress syndrome; ICU, intensive cre unit. Reference ctegory ws No AKI. *Confidence intervls tht do not overlp the null vlue of OR = 1 re shown in bold. doi: /journl.pone t009 PLOS ONE 10 Mrch 2014 Volume 9 Issue 3 e90419

11 Tble 10. Assocition of AKI, with complictions nd supportive requirements by mlri species. Mild AKI (N = 99) Severe AKI (N = 43) Chrcteristics P. vivx N=50 P. flciprum N=49 P. vivx N=06 P. flciprum N=37 Associted Complictions N (%) N (%) OR (95% CI) # N (%) N (%) OR (95% CI) # Cerebrl Mlri 01 (2.0) 05 (10.2) 5.6 (0.6, 49.5) (16.2) * Jundice 13 (28.3) 22 (45.8) 2.1 (0.9, 5.1) 04 (66.7) 29 (78.4) 1.8 (0.3, 11.7) Spontneous bleeding 01 (2.0) 00 * 01 (16.7) 01 (2.7) 0.1 (0.0, 2.6) Severe nemi 01 (2.1%) 08 (16.7) 9.2 (1.1, 76.8) (17.6) * PE/ARDS 04 (8.0) 03 (6.1) 0.7 (0.1, 3.5) (2.7) * Supportive requirements Blood/Blood product 08 (16) 17 (34.7) 2.8 (1.1, 7.3) 02 (33.3) 25 (67.6) 4.2 (0.7, 26.0) Mechnicl Ventiltion 03 (06) 03 (6.2) 1.0 (0.2, 5.4) (27) * Inotropes * 01 (16.7) 00 * ICU cre 07 (14) 12 (24.5) 1.9 (0.7, 5.6) 01 (16.7) 17 (47.2) 4.5 (0.5, 42.2).7 dys of hospitliztion 14 (28) 23 (46.9) 2.3 (0.9, 5.2) 05 (83.3) 21 (58.3) 0.3 (0.0, 2.6) Deth 01 (2.0) 01 (2.0) 1.0 (0.1, 16.8) (16.2) * Abbrevitions: OR, Odds rtio; CI, confidence intervl; PE, pulmonry edem; ARDS, cute respirtory distress syndrome; ICU, intensive cre unit. *Sttistics could not be computed. # Confidence intervls tht do not overlp the null vlue of OR = 1 re shown in bold. doi: /journl.pone t010 predominnt cuse of mlri in Southest Asi, the bsolute contribution to severe AKI my still be substntil. Thus, erly detection nd tretment of mild AKI cses with P. vivx my in fct significntly reduce the prevlence of severe AKI in these regions. Strengths nd Weknesses of the Study The present study hs significnt strengths. It is the first report of AKI strtified by severity in lrge cohort of consecutive ptients hospitlized with mlri. The severe AKI definition corresponds to WHO criteri for cute renl filure nd enbles comprison with published literture. This study is the lrgest cohort of cute mlri ptients with 60% prevlence of P.vivx tht hs provided rigorous nlysis of ssocited fctors of AKI. However, the retrospective nture of the study hs its inherent limittions. Although we found bsence of fever to be ssocited with AKI, probble recll bis resulting in this finding could not be ruled out due to lck of dt on durtion of illness/dely in presenttion. We did not hve historicl or in-hospitl urine output dt systemticlly recorded. Hence, the potentil of oliguri s risk fctor for AKI could not be ssessed. Outptient bseline serum cretinine vlues were not vilble, nd t lest two vlues of serum cretinine were vilble only in bout third of the cohort. Thus, we could not define AKI by contemporry risk, injury, filure, loss, nd end stge kidney disese (RIFLE) or cute kidney injury network (AKIN) criteri. With our study design, s the timing of the AKI occurrence or the mechnism of AKI could not be discerned, these should be further evluted by prospective design with timely ssessment of pnel of most sensitive nd specific renl biomrkers. Hemodilysis done on ptients ws bsed on clinicins judgement in consulttion with nephrologists nd ws not bsed on pre-defined hrd end points. Additionlly, we do not hve dt on hemodilysis initition References 1. World Helth Orgniztion (2012) Mngement of severe mlri: A prcticl hndbook. Itly: World Helth Orgniztion Press. 89 p. timings for ny ptient. Dt regrding metbolic cidosis, hemoglobinuri, reticulocyte count, cogultion prmeters, nd mesure of depth of com in cerebrl mlri-like Glsgow com scle, nd empiric ntibiotics used were not vilble, nd prsite density ws vilble only in few ptients. Specition ws done microscopiclly nd not by more sensitive polymerse chin rection tests. Conclusion Age, bsence of fever, lower distolic blood pressure, hert rte, icterus, heptomegly, nd direct bilirubin were fctors found to be ssocited with AKI in our study. The results confirm tht severity of AKI hs prognostic vlue in morbidity nd mortlity in cute mlri. Future studies should investigte if even erlier detection with novel biomrkers, before increse in serum cretinine, would lter the epidemiology of this preventble compliction. It remins possibility tht erly recognition of AKI nd prompt, ggressive tretment to vert disese progression my improve the outcomes in the setting of mlri. Acknowledgments The uthors thnk Prof. Brkur Annthkrishn Shstry (Ksturb Medicl College, Mnipl University, Mnipl) nd Prof. Sudh Vidysgr (Ksturb Medicl College, Mnipl University, Mnipl) for their encourgement. Author Contributions Conceived nd designed the experiments: KS KR CRP. Anlyzed the dt: KS CRP. Wrote the pper: KS KR CRP. Acquisition of dt: KR KS. Criticl revision of the mnuscript for importnt intellectul content: KS CRP. 2. World Helth Orgniztion (2012) World Mlri Report: World Helth Orgniztion Press. 195 p. PLOS ONE 11 Mrch 2014 Volume 9 Issue 3 e90419

12 3. Sheehy TW, Reb RC (1967) Complictions of flciprum mlri nd their tretment. Ann Intern Med 66: Sitprij V (1988) Nephropthy in flciprum mlri. Kidney Int 34: Prksh J, Singh A, Gujrti S, Mheshwri A (2002) Acute renl filure in Mlri: Chnging trends. Indin J Nephrol 12: Pnd S, Ds M, Meher L, Rthod P (2003) Risk fctor for cute renl filure in severe flciprum mlri. Indin J Nephrol 13: Kochr D, Kochr S, Agrwl R, Sbir M, Nyk K, et l. (2006) The chnging spectrum of severe flciprum mlri: clinicl study from Bikner (northwest Indi). J Vector Borne Dis 43: Prksh J, Singh A, Kumr N, Sxen R (2003) Acute renl filure in Plsmodium vivx mlri. J Assoc Physicins Indi 51: Kochr DK, Ds A, Kochr SK, Sxen V, Sirohi P, et l. (2009) Severe Plsmodium vivx mlri: report on seril cses from Bikner in northwestern Indi. Am J Trop Med Hyg 80: Trng TT, Phu NH, Vinh H, Hien TT, Cuong BM, et l. (1992) Acute renl filure in ptients with severe flciprum mlri. Clin Infect Dis 15: Vnnphn S, Wlters N, Sengnedswng T, Tngpukdee N, Khm-In P, et l. (2010) Fctors ssocited with cute renl filure in severe flciprum mlri pteints. Southest Asin J Trop Med Public Helth 41: Thnchrtwet V, Deskorn V, Shssnnd D, Win K, Yzr KK, et l. (2013) Acute Renl Filure in Ptients with Severe Flciprum Mlri: Using the WHO 2006 nd RIFLE Criteri. Int J Nephrol 2013: Tngpukdee N, Omr Elshiekh S, Phumrtnprpin W, Krudsood S, Wilirtn P (2011) Fctors ssocited with cute renl filure in flciprum mlri infected ptients. outhest Asin J Trop Med Public Helth 42: Win KK, Thnchrtwet V, Wttngoon Y, Jerrksuwn S, Rungweeryut R, et l. (2012) Fctors ssocited with cute renl filure in dults with severe flciprum mlri. Southest Asin J Trop Med Public Helth 43: Nqvi R, Ahmd E, Akhtr F, Nqvi A, Rizvi A (2003) Outcome in severe cute renl filure ssocited with mlri. Nephrol Dil Trnsplnt 18: Prikh CR, Devrjn P (2008) New biomrkers of cute kidney injury. Crit Cre Med 36: S159 S Bloom D, McClden T, Rosendorff C (1975) Effects of jundiced plsm on vsculr sensitivity to nordrenlin. Kidney int 8: Cioffi W, DeMeules J, Khng K, Wit R (1986) Renl vsculr rectivity in jundice. Surgery 100: Better OA (1986) Nephrology Forum: renl nd crdiovsculr dysfunction in liver disese. Kidney Int 29: Bum M, Stirling G, Dwson J (1969) Further study into obstructive jundice nd ischemic renl dmge. BMJ 2: Bruneel F, Gchot B, Timsit J, Wolff M, Bedos J, et l. (1997) Shock complicting severe flciprum mlri in Europen dults. Intensive cre med 23: Sitprij V, Npthorn S, Lorptnskul S, Suithichiykul T, Moollor P, et l. (1996) Renl nd systemic hemodynmics in flciprum mlri. Am J Nephrol 16: Ncher M, Treeprsertsuk S, Singhsivnon P, Silchmroon U, Vnnphn S, et l. (2001) Assocition of heptomegly nd jundice with cute renl filure but not with cerebrl mlri in severe flciprum mlri in Thilnd. Am J Trop Med Hyg 65: PLOS ONE 12 Mrch 2014 Volume 9 Issue 3 e90419