UBC Faculty Research and Publications

Inhaled nitric oxide for the adjunctive therapy of severe malaria: Protocol for a randomized controlled… Hawkes, Michael; Opoka, Robert O; Namasopo, Sophie; Miller, Christopher; Thorpe, Kevin E; Lavery, James V; Conroy, Andrea L; Liles, W C; John, Chandy C; Kain, Kevin C Jul 13, 2011

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

Item Metadata


52383-13063_2011_Article_705.pdf [ 385.67kB ]
JSON: 52383-1.0223329.json
JSON-LD: 52383-1.0223329-ld.json
RDF/XML (Pretty): 52383-1.0223329-rdf.xml
RDF/JSON: 52383-1.0223329-rdf.json
Turtle: 52383-1.0223329-turtle.txt
N-Triples: 52383-1.0223329-rdf-ntriples.txt
Original Record: 52383-1.0223329-source.json
Full Text

Full Text

STUDY PROTOCOL Open AccessInhaled nitric oxide for the adjunctive therapy ofsevere malaria: Protocol for a randomizedcontrolled trialMichael Hawkes1,2, Robert O Opoka3, Sophie Namasopo4, Christopher Miller5, Kevin E Thorpe6,7, James V Lavery8,9,Andrea L Conroy10, W Conrad Liles1,11,12,13,14, Chandy C John15 and Kevin C Kain1,10,11,12,13,14*AbstractBackground: Severe malaria remains a major cause of global morbidity and mortality. Despite the use of potentanti-parasitic agents, the mortality rate in severe malaria remains high. Adjunctive therapies that target theunderlying pathophysiology of severe malaria may further reduce morbidity and mortality. Endothelial activationplays a central role in the pathogenesis of severe malaria, of which angiopoietin-2 (Ang-2) has recently beenshown to function as a key regulator. Nitric oxide (NO) is a major inhibitor of Ang-2 release from endothelium andhas been shown to decrease endothelial inflammation and reduce the adhesion of parasitized erythrocytes. Low-flow inhaled nitric oxide (iNO) gas is a US FDA-approved treatment for hypoxic respiratory failure in neonates.Methods/Design: This prospective, parallel arm, randomized, placebo-controlled, blinded clinical trial comparesadjunctive continuous inhaled nitric oxide at 80 ppm to placebo (both arms receiving standard anti-malarialtherapy), among Ugandan children aged 1-10 years of age with severe malaria. The primary endpoint is thelongitudinal change in Ang-2, an objective and quantitative biomarker of malaria severity, which will be analysedusing a mixed-effects linear model. Secondary endpoints include mortality, recovery time, parasite clearance andneurocognitive sequelae.Discussion: Noteworthy aspects of this trial design include its efficient sample size supported by a computersimulation study to evaluate statistical power, meticulous attention to complex ethical issues in a cross-culturalsetting, and innovative strategies for safety monitoring and blinding to treatment allocation in a resource-constrained setting in sub-Saharan Africa.Trial Registration: ClinicalTrials.gov Identifier: NCT01255215BackgroundMalaria is the leading parasitic cause of morbidity andmortality worldwide, causing an estimated 240 millionclinical cases and 800,000 deaths annually [1]. Childrenin sub-Saharan Africa bear the greatest burden of dis-ease, where one in every five childhood deaths is due tomalaria and 25% of survivors of cerebral malaria developlong-term neurocognitive impairment [1,2]. Despite theuse of highly effective anti-malarial medications, 10-30%patients with severe malaria will die [3-5], underscoringthe need for adjunctive therapies that can be applied inendemic areas. To date, effective adjunctive treatmentshave been elusive despite numerous clinical trials [6].New therapies, appropriate for use in endemic areas, aretherefore urgently needed to address the unacceptablyhigh residual mortality associated with severe malaria inpediatric populations.NO is a gaseous, lipid-soluble free radical that is pro-duced in vivo by the enzymatic conversion of L-arginineand molecular oxygen to L-citrulline by members of thenitric oxide synthase (NOS) family of proteins. A freeradical, NO is a highly labile molecule with a half-life ofseveral seconds that reacts with transition metal or thiolgroups of numerous target proteins [7]. NO readily dif-fuses across cell membranes into neighbouring cells, or* Correspondence: kevin.kain@uhn.on.ca1Institute of Medical Sciences, University of Toronto, CanadaFull list of author information is available at the end of the articleHawkes et al. Trials 2011, 12:176http://www.trialsjournal.com/content/12/1/176 TRIALS© 2011 Hawkes et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative CommonsAttribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction inany medium, provided the original work is properly cited.may produce effects distant from its site of productiontransported by vehicles such as low-molecular weightS-nitrosotiols, S-nitrosylated proteins including haemo-globin and albumin, and nitrosyl-metal complexes whichliberate NO spontaneously or after cleavage by ectoen-zymes [8]. NO regulates numerous cellular processesincluding cytoplasmic granule exocytosis, platelet aggre-gation, endothelial-cell activation, apoptosis, inflamma-tion, chemotaxis, neurotransmission and antimicrobialdefense by modulating the activity of regulatory proteins[9]. A well-recognized example is the role of NO as theendothelium-derived relaxation factor that mediatesvasodilation by activating smooth muscle soluble guany-late cyclase (sGC) [10].Evidence from a murine model suggests that reducedNO bioavailability contributes to the pathogenesis ofexperimental cerebral malaria [11]. “Footprint” moleculesof labile nitric oxide including cGMP and nitrite weremarkedly decreased over the course of infection, and NOsupplementation with either a NO donor (dipropylene-triamine NONOate, DPTA/NO) or NO gas providedmarked protection against severe disease [11]. Data fromhuman studies support the hypothesis of reduced bioa-vailable NO in severe malaria: African children withsevere malaria have impaired production of NO [12], lowlevels of mononuclear cell iNOS expression [12], lowlevels of the NOS substrate arginine [13], and elevatedlevels of the NOS inhibitor, asymmetric dimethyl argi-nine [14]. Furthermore, genetic variation in NOS iso-forms that affect plasma and urine levels of NO and itsmetabolites are common in African populations and havebeen shown to influence the risk of cerebral malaria[15-19]. Factors contributing to reduced bioavailable NOin malaria include scavenging of NO by free haemoglobinand superoxide anion, and reduced levels of nitrate, a NOprecursor molecule [11,13,20].The mechanism of action by which NO might improveoutcomes in malaria may involve the vascular endothe-lium, which plays a central role in the pathogenesis ofcerebral malaria. Activation of endothelial cells is charac-terized by increased surface expression of cellular adhe-sion molecules, the exocytosis of Weibel-Palade bodies(WPB), and the breakdown of intracellular tight junctionswith transudation of intravascular fluid producing endorgan dysfunction. In malaria, parasitized erythrocytes(PEs) adhere to the microvascular endothelium, resultingin sequestration and vascular obstruction, impaired per-fusion and tissue hypoxia [21]. Autopsy studies in fatalcerebral malaria reveal sequestration of PEs in the capil-laries and post-capillary venules of multiple organs [22].NO decreases endothelial cell adhesion molecule expres-sion, including intercellular cell adhesion molecule-1(ICAM-1) [23,24] and has been shown to reduce theadherence of PEs to endothelial cells [25]. The release ofintracellular WPB contents from endothelial cells liber-ates von Willebrand factor (vWF) [26,27] and angiopoie-tin-2 (Ang-2) [28,29] into the circulation. Interactions ofvWF with the coagulation cascade may contribute to ves-sel obstruction and may help tether parasitized erythro-cytes to endothelial cells via platelets [30]. Ang-2 acts inan autocrine and paracrine fashion to sensitize theendothelium to the effects of TNF, resulting in increasedadhesion receptor expression [31]. In addition, Ang-2antagonizes the interaction of the Tie-2 receptor withangiopoietin-1 (Ang-1), thereby promoting endothelialpermeability and reducing vessel stability [32-35]. NOinhibits the exocytosis of WPB contents through S-nitro-sylation of critical regulatory factors [9] and may there-fore promote endothelial quiescence, reduce vascularfluid leak, and reduce end-organ damage. A clinical trialdemonstrated that a strategy of NO supplementationusing the NOS substrate L-arginine improved endothelialfunction, as measured by reactive-hyperemia-peripheralarterial tonometry, in Indonesian adults with moderatelysevere malaria [36].Low-flow iNO at a concentration of 5-80 ppm isapproved for use by the US FDA for the treatment ofneonates with hypoxic respiratory failure, in whom itreduces requirements for extracorporeal membrane oxy-genation (ECMO) and improves survival [37]. After adecade of use in clinical practice and in numerous clini-cal trials of iNO in critically ill older children andadults, iNO has a well-established safety profile. Prag-matic considerations, including a theoretically cheapmanufacturing cost and ease of administration by mask,make NO an attractive therapeutic option for unrespon-sive patients in resource-limited settings.In summary, a clinical trial of adjunctive inhaled nitricoxide (iNO) in severe malaria is warranted on the basisof firm proof of concept from animal models [11] and aclinical trial using the NO donor L-arginine [36],together with evidence of safety from clinical experienceand numerous clinical trials of iNO for other conditions[38].MethodsStudy DesignThe study is a prospective, parallel arm, randomized,placebo-controlled, blinded clinical trial of adjunctivecontinuous inhaled nitric oxide at 80 ppm versus pla-cebo (both arms in addition to standard anti-malarialtherapy), among children aged 1-10 years of age withsevere malaria. Figure 1 shows a participant flow dia-gram for the trial, consistent with the ConsolidatedStandards of Reporting Trials (CONSORT) 2010 state-ment [39].Hawkes et al. Trials 2011, 12:176http://www.trialsjournal.com/content/12/1/176Page 2 of 13Assessed for eligibility (n=...)•Age 1-10 years•Hospitalized for possible severe malariaExcluded (n=...)Not meeting inclusion criteria (n=...)•malaria RDT negative•no features of severe malaria•elevated baseline metHb•chronic illness•severe malnutrition•SMA aloneDeclined to participate (n=...)Other reasons (n=...)Randomized (n=180)Allocated to room air placebo (n≈90)All patients receive standard antimalarial treatmentReceived room air (n=...)Did not receive room air (give reasons) (n=...)Day 14 follow-upLost to follow-up (give reasons) (n=...)Analyzed (n=...)Excluded from analysis (give reasons) (n=...)6-month follow-up (neurocognitive testing)Lost to follow-up (give reasons) (n=...)In-hospital courseLost to follow-up (give reasons) (n=...)Discontinued intervention (give reasons) (n=...)Mortality (n=...)Allocated to iNO (n≈90)All patients receive standard antimalarial treatmentReceived iNO (n=...)Did not receive iNO (give reasons) (n=...)Day 14 follow-upLost to follow-up (give reasons) (n=...)Analyzed (n=...)Excluded from analysis (give reasons) (n=...)6-month follow-up (neurocognitive testing)Lost to follow-up (give reasons) (n=...)In-hospital courseLost to follow-up (give reasons) (n=...)Discontinued intervention (give reasons) (n=...)Mortality (n=...)Figure 1 Participant flow diagram. The participant flow diagram illustrates randomization of 180 children with severe malaria to inhaled nitricoxide (iNO) or placebo, consistent with the Consolidated Standards of Reporting Trials (CONSORT) 2010 statement. RDT: rapid diagnostic test;metHb: methemoglobin; SMA: severe malarial anemia.Hawkes et al. Trials 2011, 12:176http://www.trialsjournal.com/content/12/1/176Page 3 of 13Study ObjectivesThe primary objective of this trial is to compare thelongitudinal change in Ang-2, an objective and quantita-tive biochemical marker of malaria severity, among chil-dren who are randomized to receive inhaled nitric oxidein addition to standard antimalarial therapy comparedto those randomized to placebo in addition to standardantimalarial therapy.Secondary objectives are• To determine the efficacy of adjunctive iNO insevere malaria based on clinical parameters: mortal-ity, recovery times, length of hospital stay.• To assess the effect of adjunctive iNO on labora-tory parameters: parasite clearance, whole blood lac-tate levels, and other biomarkers of malaria severity.• To determine the efficacy of adjunctive nitric oxidein preventing neurocognitive sequelae after severemalaria.• To assess the tolerability and safety of iNO insevere malaria.Study HypothesesThe working hypothesis is that young children hospita-lized with malaria will benefit from adjunctive iNO, asdetermined by more rapid improvement in serum Ang-2levels. We will test this hypothesis by comparing thechange in Ang-2 over the hospital admission between thetwo groups randomized to receive iNO or placebo (roomair) using a mixed-effects linear statistical model.The secondary hypotheses are that adjunctive iNO willreduce mortality, accelerate recovery times, and shortenlength of hospital stay in severe malaria. Furthermore,we hypothesize that iNO will accelerate improvementsin biomarkers of host response to severe malaria, includ-ing whole blood lactate, but will not affect parasiteclearance. We hypothesize that iNO will reduce the rateof adverse neurocognitive sequelae following severemalaria. Finally, we hypothesize that adjunctive iNO willbe safe and well tolerated in children treated for severemalaria.Eligibility criteriaChildren will be eligible for the trial if they meet thefollowing inclusion criteria:1. Age 1-10 years2. Positive malaria rapid diagnostic test (RDT)3. One or more features of severe malaria: repeatedseizures (two or more generalized seizures in 24 h);prostration (in children 1 year and older, the child isunable to sit unsupported or stand although wasable to before the illness); impaired consciousness(Blantyre Coma Score <5 in children 1 to 4 years,GCS <14 for children ≥ 5 years); respiratory distress:age related tachypnea with sustained nasal flaring,deep breathing or subcostal retractions4. Willing and able to complete follow up schedulesfor the study - 14 day and 6 months after hospitaldischarge.The inclusion criteria require timely parasitological con-firmation of malaria infection prior to enrolment, whichposes logistical challenges at our peripheral centre withlimited laboratory resources. For diagnosis, we will usecommercially available immunochromatographic RDTs,complemented where possible with microscopy of periph-eral smears. Despite well-recognized variability in the testperformance characteristics of RDTs under field condi-tions, one commercially available RDT (First ResponseMalaria Ag Combo (pLDH/HRP2), Premier Medical Cor-poration Ltd., India) is highly ranked by the World HealthOrganisation (WHO), with a 95% detection rate even atlow parasitemia and a false positive rate of 0% [40]. ThisRDT includes detection bands for two P. falciparum anti-gens, histidine-rich protein-2 (HRP-2) and parasite lactatedehydrogenase (pLDH), and we will require positivity forboth antigens for trial inclusion. Thus, our parasitologiccriterion is expected to be highly specific, in order toinclude only patients who are truly parasitemic.The exclusion criteria for this study are as follows:1. Baseline methemoglobinemia (>2%)2. Known chronic illness: renal, cardiac, or hepaticdisease, diabetes, epilepsy, cerebral palsy, or clinicalAIDS3. Severe malnutrition, defined as weight-for lengthor height below -3 standard deviations based onWHO reference charts, or symmetrical edema invol-ving at least the feet [41].4. Severe malarial anemia (SMA; Hb <50 g/L) with-out other signs of severe malaria.The latter exclusion criterion was chosen because ofdifferences in the pathophysiology of SMA (increasedclearance of infected and uninfected erythrocytes, anddysregulated hematopoiesis) compared to other malariasyndromes characterised by excessive inflammation andendothelial activation. Based on its postulated mechan-ism of action, it is less clear that iNO would benefitchildren with SMA.Study settingThe study will be conducted at a single pediatric hospi-tal in Jinja, Uganda. Uganda is a low-income countrywith a severely resource-constrained healthcare system.Malaria transmission is moderate and seasonal in Jinja,Hawkes et al. Trials 2011, 12:176http://www.trialsjournal.com/content/12/1/176Page 4 of 13which lies on the northern shore of Lake Victoria in thecentral area of Uganda near the capital, Kampala. JinjaRegional Referral hospital admits at least 175 childrenwith severe malaria annually (excluding cases of severemalarial anemia), representing over 30% of all admis-sions. P. falciparum resistance to chloroquine and sulfa-doxine-pyrimethamine is widespread in the country(34% to 67%) [42].Treatment groupsParticipants in the intervention group will receive iNO at aconcentration of 80 ppm, in addition to Ugandan standardof care of severe malaria, which includes a potent anti-parasitic agent. In light of the recent AQUAMAT trialthat demonstrated a mortality benefit of parenteral artesu-nate over quinine for the treatment of African childrenwith severe malaria [5], artesunate will be used as the anti-malarial of choice for this study. iNO will be administeredcontinuously via non-rebreather face mask for a maximumperiod of 72 hours, but may be discontinued earlier if apatient recovers and no longer tolerates the face mask.The dose of iNO (80 ppm) was chosen as it is the highestapproved dose by the US-FDA for use in neonates, withthe greatest potential for a therapeutic effect. In pre-clini-cal animal studies, a dose of 80 ppm provided greater pro-tection from experimental cerebral malaria than lowerdoses (Serghides, Kim, et al., unpublished data). At leastfive published clinical trials across different age groupsand clinical conditions have safely used this dose [43-47].With careful monitoring for dose-dependent adverseevents and titration of iNO concentration accordingly, adose of 80 ppm can safely be used in our trial.Participants in the control group will receive room air(indistinguishable in odour and appearance from the mix-ture of 80 ppm iNO), in addition to Ugandan standard ofcare treatment for severe malaria, including parenteralartesunate. An air compressor will be used to deliver con-tinuous flow of vehicle air in both groups.Randomization and blinding methodsSimple randomization will be employed, using a compu-ter-generated randomization list. Treatment allocationwill be recorded on paper and kept in sequentially num-bered sealed opaque envelopes, which will be drawn foreach randomized participant by an unblinded investigatorwho is not responsible for patient care, laboratory or dataanalysis. We will retain all envelopes and records forquality monitoring purposes.In previous clinical trials using iNO, one of the designand implementation challenges was establishing theblinding procedures while titrating and monitoring con-centrations of iNO, as well as anticipated dose-relatedincreases of methemoglobin and NO2 concentrations[38]. The establishment of two teams, one blinded teammaking all clinical assessments and therapeutic decisions,and another unblinded team monitoring the delivery ofthe treatment gas and assessing the development ofpotential toxicities, allowed iNO to be delivered safelywhile minimizing the possibility that direct knowledge oftreatment allocation would influence the care deliveredto the patient [38]. Thus, an unblinded investigator notinvolved in patient care will be responsible for theadministration, monitoring and recording of iNO, NO2,and methemoglobin levels. Cylinders containing NO willbe attached to the ventilation circuit in all patients, butthe flow of NO will be controlled according to treatmentarm assignment in a manner blinded to patients, care-givers, and healthcare providers. Only the unblindedstudy site investigator will administer, monitor, andtitrate the delivery of NO or placebo (room air), as wellas monitor and record the safety parameters of methe-moglobin and NO2 levels. Laboratory analysis for Ang-2levels (primary outcome) and all other parameter willoccur in a manner blinded to treatment allocation.Outcome measuresThe longitudinal change in serum Ang-2 concentrationover the first 72 hours of hospital admission will be theprimary efficacy endpoint. Elevated Ang-2 levels are asso-ciated with poor clinical outcome in severe malaria [28,29]and Ang-2 has been used to follow disease progressionand recovery in previous studies of malaria [28]. Amongsurvivors of severe malaria, Ang-2 levels have been shownto decrease linearly during recovery at a mean rate of 2700pg/mL per 24 h [28]. Thus, Ang-2 is an objective, quanti-tative marker of disease severity, validated for longitudinalfollow-up of patients with malaria. Ang-2 levels will bemeasured longitudinally at admission (day 0), day 1, day 2and day 3 of hospitalization. Angiopoietin-2 will be mea-sured by enzyme-linked immunosorbent assay (ELISA)from plasma or serum samples, and is readily detectable insamples frozen for storage and later thawed [6-9]. Com-mercially available ELISA kits will be used (DuoSets, R&DSystems, Minneapolis, MN). A mixed-effects linear modelwill be used to compare the change in Ang-2 over timebetween treatment arms.Secondary outcomes of the trial will include relevantclinical, laboratory and neurocognitive endpoints. We willcompare the following clinical endpoints, consistent withother therapeutic trials for malaria [4,48,49]: mortality at48 hours and 14 days after admission; recovery times(time to fever resolution, time to sit unsupported); andlength of hospital stay. Parasitological outcomes includingtime to parasite clearance and parasite recrudescence/re-infection at 14 day follow-up will also be comparedbetween treatment groups. Biomarkers of disease severity(in addition to Ang-2, the primary study endpoint), includ-ing whole blood lactate, will also be followed. Lactate isHawkes et al. Trials 2011, 12:176http://www.trialsjournal.com/content/12/1/176Page 5 of 13produced by the anaerobic metabolism of glucose in theabsence of adequate tissue oxygenation, and elevatedlactate levels represent a final common pathway of tissuehypoxia and decompensated shock, the forerunner of car-diovascular collapse and death. We will measure lactate asan independent biomarker of disease severity during theclinical trial. Biomarkers of endothelial activation, inflam-mation, and coagulopathy will also be followed as theymay provide additional insight into the pathways and pro-cesses altered in cerebral malaria and modulated by iNOdelivery. Finally, neurocognitive outcomes in children withsevere malaria will be followed in order to determine ifadjunctive iNO may have a neuroprotective effect. Theoverall cognitive deficit at 6 months after discharge will beassessed by performing neuropsychological tests as pre-viously described [2].Duration of Study ParticipationAfter admission to hospital for severe malaria, disease sur-vivors will typically be discharged after a week or less.Trial participants will be administered iNO during thefirst 72 hours of admission (or less if a patient recoversand no longer tolerates the mask), which represents theperiod of highest mortality. During hospitalization,detailed data on the interim medical history will be col-lected, with attention to complications like coma, seizuresand hypoglycemia that might affect neurocognitive out-come. After discharge, patients will return for a follow-upvisit at 14 days to test for P. falciparum recrudescence,and at 6-months to undergo neurocognitive testing.SafetyNO is approved by the US FDA for the treatment of neo-nates with hypoxic respiratory failure, where it has beenshown to improve oxygenation, decrease pulmonaryhypertension, reduce the requirement for extracorporealmembrane oxygenation, and improve survival [37]. Inaddition, iNO is widely used in clinical practice acrossNorth America and Europe in older children and adultswith respiratory failure, where it improves oxygenationbut has not been shown to confer a survival benefit [50].As adjunctive therapy, iNO is safe and well tolerated inthese critically ill patient populations and a large numberof randomized controlled trials have demonstrated afavourable safety profile of iNO. One meta-analysis of 12trials including 1237 patients with acute lung injury oracute respiratory distress syndrome demonstrated thatiNO is generally safe, but was associated with a statisti-cally elevated risk of developing renal dysfunction inthese critically ill adults [38]. We anticipate that ourpediatric target population may be less susceptible torenal injury, particularly after exclusion of patients withunderlying chronic renal disease, but we will monitorrenal function in all patients enrolled in our trial withdaily creatinine and urine output measurement.Methemoglobinemia and elevated nitrogen dioxide(NO2) levels in the inspiratory ventilation circuit are well-recognized, dose-dependent, and reversible adverse effectsof nitric oxide administration; however, at doses com-monly used in clinical practice, these are not common orclinically important consequences [38]. Methemoglobine-mia results from the reaction of NO with oxyhemoglobin,thereby reducing oxygen carrying capacity [51]. NO2 isgenerated from the gas phase reaction of NO with mole-cular oxygen, and is a known pulmonary irritant [52,53].In previous clinical trials of iNO, methemoglobin andNO2 were routinely monitored and elevated levels consti-tuted a criterion for NO dose reduction. Among neonatesreceiving iNO at 80 ppm, 35% developed methemoglobi-nemia (>7%) and 19% had elevated NO2 levels (>3%),requiring reduction of the iNO concentration in the venti-lation circuit [44]. Similarly, in our trial, methemoglobinand NO2 will be monitored and the iNO dose will betitrated downward according to these defined thresholds.Furthermore, elevated baseline methemoglobinemia willbe used as an exclusion criterion from study participationas a possible indicator of genetic susceptibility tomethemoglobinemia.Adverse events may occur commonly in a trial involvingchildren with severe malaria, although the majority ofevents are likely due to the clinical course of the infectionand not to study medications. For example, mortalityamong Ugandan children with severe malaria receivingstandard care including potent antimalarial agents was ashigh as 16% among children with impaired consciousnessand 21% among children with deep acidotic breathing aspresenting clinical signs [42]. In order to carefully andrationally monitor the frequency of deaths in our trial fordeviations from the expected baseline level, we plan to usestatistical control charts. The control chart is a commonlyused tool to monitor output of processes in a variety ofsettings, including clinical trials [54]. This method con-tinuously follows a process outcome (e.g., patient mortal-ity), allowing early detection of deviations from a state of“statistical control,” thereby prompting a search for assign-able causes. Further details of the statistical thresholds andthe performance characteristics of the control chart formortality in this trial are given in a supplementary file(Additional File 1).Other adverse events will be monitored using pediatrictoxicity tables modified from the US National Institute ofAllergy and Infectious Diseases [55]. Using this compre-hensive checklist of potential adverse events, investigatorswill grade the severity of the event and the likelihood thatthe event is causally associated with the study gas accord-ing to scales defined a priori [55,56].Hawkes et al. Trials 2011, 12:176http://www.trialsjournal.com/content/12/1/176Page 6 of 13In addition to the ethical oversight provided by boththe Ugandan and North American institutions, an inde-pendent Data and Safety Monitoring Board (DSMB) hasbeen convened to supervise the trial. The DSMB is com-posed of medical and biostatistical experts with repre-sentation from Uganda and North America who willmeet periodically and as necessary to review trial pro-gress, safety data, and indices of study quality. Severeadverse events, including all deaths in the trial, will bereported in a timely fashion to the DSMB and to theethics boards that approved the study.Ethical considerationsEthical approval has been obtained from the MakerereUniversity Research and Ethics Committee (Kampala,Uganda), the Uganda National Council on Science andTechnology, and the Research Ethics Board of the Uni-versity Health Network, Toronto, Canada (REB# 10-0607-B). The research is being conducted in accordancewith the Declaration of Helsinki.Informed consent will be obtained from the parent/guardian of all children that participate in the study.Trained study team personnel will seek consent after acomprehensive discussion with the parent/guardian of aprospective participant in the local language (Lusoga) atan education-appropriate level. Assent will also be soughtfrom those children who are alert and able to understandthe trial (age 8 and above), and sustained dissent on thepart of children will be honoured [57]. Specific aspectsincluded in the consent discussion include: the acuityand potential lethality of severe malaria, residual mortal-ity despite antiparasitic treatment, the absence of proveneffective adjunctive therapies, potential benefits andharms of iNO, the concept of randomization and poten-tial allocation to placebo control (although all patientswill receive standard care including potent antimalarials),blinding of treatment allocation, blood samples requiredfor the trial above what is necessary for clinical care, andthe distinction between the experimental interventionand clinical care.International collaborative research may face complexcommunity challenges [58,59]. Community engagement(CE), a participatory process of collaboration andexchange between the various key stakeholders in theresearch process, may mitigate risks with respect to trialsuccess, optimize participant retention, and minimizesocial disruption by providing a platform to seek inputfrom and provide ongoing feedback to community mem-bers [60]. There is currently no consensus on what CEactivities are required in clinical trials, but a recently pub-lished model of CE in global health research provides auseful framework of key CE activities and their ethicalimplications [60].We will use CE to improve awareness of our trial andits findings in the catchment area of the Jinja RegionalReferral Hospital. We also hope the CE activities willcontribute to key ethical objectives for the trial, includ-ing respect for communities, fairness, transparency andaccountability of the trial overall. We will focus on thefollowing CE activities and their associated aims: (1)understanding the relevant community by consciouslyreaching out beyond the hospital to its catchment areasand listening to their issues and concerns; (2) providinginformation about the trial, including the pre-clinicalevidence behind iNO, timeframe, procedures, and whatwill happen if the trial is successful; (3) building rela-tionships and trust with local frontline healthcare work-ers; (4) specific educational/training activities, based onconsultation with the nurses and/or frontline healthcareworkers to ascertain what would be most relevant andbeneficial for them; and (5) feedback of trial results,guided by the community itself as to how and whattypes of feedback activities would be most appropriate.Several levels of community will be targeted, includingparents and primary caregivers of children, who com-prise the group at highest risk of malaria, as well ashealthcare professionals within the hospital catchmentarea. These activities do not constitute a mechanism forrecruitment of participants to the trial, since only chil-dren with severe malaria will be eligible. Instead thecommunity engagement process is intended to buildtrust and avoid misunderstandings through a dynamicexchange of information and ideas between trial scien-tists and community members.Sample Size CalculationWe will enrol 180 patients (approximately 90 in eachtreatment arm). To arrive at this sample size estimate,we began with a preliminary calculation based on datafrom a recent clinical data in severe malaria, in whichAng-2 decreased by 2700 pg/mL/day (95% CI 1800-3600pg/mL/day) [28]. We assume that a 50% change in thisparameter would represent a clinically significant thera-peutic effect. By standard calculations for normally dis-tributed data, 80 patients per group will provide 80%power to detect a difference between two treatmentarms of 1350 pg/mL/day at p = 0.05 (two-sided). Toaccount for possible dropout, loss to follow-up, and/ornon-evaluable data of 10% of patients, approximately 90patients per study arm are required.To validate this preliminary sample size estimate, con-sistent with our analytic plan (mixed-effects linearmodel), a simulation study was performed. A number ofassumptions needed to be made, such as within-patientcorrelation and Ang-2 variability. Patient Ang-2 datawere simulated using a multivariate normal distribution.Hawkes et al. Trials 2011, 12:176http://www.trialsjournal.com/content/12/1/176Page 7 of 13A simple autoregressive correlation structure was usedwith correlations of 0.75, 0.5 and 0.25 for time lags of 1,2 and 3 days, respectively. In previous studies, variabilityappears proportional to the mean [28]. Simulations wererun with three different relationships where the standarddeviation was taken to be 40%, 50% and 80% of theAng-2 mean at each time point. Both groups wereassumed to start at Ang-2 levels of 15,000 pg/mL [28]and the average values at each of days 1, 2 and 3 werebased on the hypothesized slopes of Ang-2. One thou-sand replications were performed at each standarddeviation relationship and treatment effect. The mixedeffects models were fit and the likelihood ratio test wasused to test the hypothesis of no time-by-treatmentinteraction at the 5% level (two-tailed). Table 1 showsthe results of the power simulations.Interim AnalysisAn interim analysis for efficacy, safety and trial quality isplanned at the midpoint of patient enrolment (approxi-mately 45 patients per group). There is no plan to stop thetrial prematurely for efficacy or futility based on the pri-mary endpoint (Ang-2), other biochemical parameters,clinical recovery times or parasitological outcomes, giventhe modest size of the trial and limited statistical power atthe midpoint. Data at the time of the interim analysis willbe presented to the DSMB members for review, who mayadvise that the trial continue without modification, con-tinue with changes to the protocol, or be discontinuedprematurely. With respect to the statistical interpretationof safety data, the DSMB may recommend termination ormodification of the trial if mortality rates exceed statisticalthresholds as described above. However, we do not pro-pose that the DSMB be strictly bound by pre-specified cri-teria, because of the complexity of the trade-offs betweensafety, efficacy, and the possibility that new informationwill change considerations. Rather, consideration of stop-ping guidelines requires a reasoned judgment based on allinformation that is available at the time of data review.Primary AnalysisThe primary focus is whether or not the rate of reduc-tion in Ang-2 differs between the treatment groups. Instatistical terms, this is a time-by-treatment interaction.Given that we will have repeated measurements, possi-bly incomplete, over time, a linear mixed-effects modelwill be used to estimate and test the magnitude of thetime-by-treatment interaction. A linear time trend willnot be assumed by treating “day” as a categorical vari-able in the model. The primary analysis will be by inten-tion-to-treat (ITT). That is, patients will be analysed inthe group to which they were randomized, regardless ofdeviations from study protocol. A secondary per-protocolanalysis may be considered if important deviations fromthe protocol compromise the validity of the ITTanalysis.Secondary outcomesMortalityAnalysis will follow standard methods in other clinicaltrials for malaria [4,48,49]. Mortality at 48 h and 14days will be coded as a binary variable. Absolute andrelative risk reduction will be reported with binomial95% confidence intervals. Analysis will be by c2 test orFisher’s exact test. In addition, we will present Kaplan-Meier survival curves comparing patients treated withiNO and placebo. Time to death will be analysed usingsurvival analysis (log-rank test for difference betweentreatment arms).Time to recoveryAmong survivors (a subgroup of randomized partici-pants), recovery times will be analysed by survival analy-sis (log-rank test). Time to sit unsupported, time to comaresolution (in the subset of patients with coma at studyadmission) and time to discharge will be documented bytreating clinicians blinded to treatment allocation. Timeto fever resolution, defined as the time required toachieve a temperature <38°C and the time to mainte-nance of temperature <38°C, will be determined from fre-quent vital sign monitoring. The time required to achievea reduction in parasite density of 50%, 90% and to unde-tectable levels will be determined from daily bloodsmears. Results will be expressed as the median time toeach event, with 95% confidence intervals.Additional biomarkers (continuous variables, repeatedlongitudinal measurements) will be analysed usingTable 1 Statistical power (1-b) of a trial comparing nitric oxide with placebo (n = 90 patients per group) for theadjunctive therapy of severe malaria, using longitudinal Ang-2 levels as primary outcomeStandard Deviation of Ang-2 levels at admission (% of mean) Difference in rate of change of Ang-2 between groups30% change(810 pg/mL/day)40% change(1080 pg/mL/day)50% change(1350 pg/mL/day)40 82% 97% 99%50 66% 89% 98%80 42% 60% 80%Hawkes et al. Trials 2011, 12:176http://www.trialsjournal.com/content/12/1/176Page 8 of 13mixed-effects linear models, with the raw value or log-transformed value of the biomarker level as the depen-dent variable, as appropriate.Neurocognitive outcomeAs in previous studies involving children in sub-SaharanAfrica [2], standardized instruments will be used for neu-rocognitive testing: Kaufman Assessment Battery forChildren (working memory), the visual form of the com-puterized Test of Variables of Attention (executive atten-tion), and the Tactual Performance Test (tactile-basedlearning). Summary variables from each test, convertedto age-specific standardized (z) scores, will provide quan-titative measures of these three cognitive domains (work-ing memory, attention and tactile learning). Details ofthese tests are described elsewhere [2,61]. Frequencies ofoverall cognitive and neurologic deficits in children trea-ted with iNO and children receiving placebo will be com-pared by c2 test or Fisher’s exact test. Differences incognitive areas affected in the two age groups (18 mo-4years, 5-10 years) will be assessed by comparing fre-quency of individuals with deficits in each area by c2 testor Fisher’s exact test. This will also serve as the best sur-rogate of impairment in a particular area in one agegroup versus another, since the type of testing for eachcognitive area will be different for the two age groups, sono direct comparison of level of impairment will be pos-sible across age groups. For individual tests, age-adjustedz-scores (determined from normative data in previousstudies among healthy community controls) will be ana-lyzed by means of mixed-effects models to examine studygroup differences in relation to neurocognitive outcomes.The models will provide estimated mean differencesbetween children treated with iNO to placebo controls.Subgroup AnalysesAlthough our sample size is modest, we will explore sub-group effects by examining interaction terms in themixed-effects linear model. Subgroup analyses related toimportant prognostic factors will be performed: age < 5years or ≥5 years; HIV seropositivity; and bacterial co-infection.Quality managementQuality management (QM), both quality control andassurance, is a continuous, ongoing process of evaluationof the quality of the conduct and documentation of stu-dies. The first step in QM will be training/re-training ofthe research staff to ensure consistency in clinical man-agement, sample processing and data collection. Standardoperating procedures (SOPs) have been developed for allstudy related procedures and protocols, and study per-sonnel will document any deviation from SOPs togetherwith the reason for the deviation. The next step for QMwill be monitoring of collected data on a prospectivebasis, with daily review of source documents for comple-teness, accuracy and consistency. Next, data entry will beverified periodically, and discrepancies will be reviewedwith the nurses and medical officers to discover the rea-son for errors, take corrective measures and preventfuture errors. Collection, storage and transport of clinicalsamples will also be monitored on a regular basis. Theethical conduct of the study will be monitored throughinitial training in research ethics for study staff, and doc-umentation, mediation and resolution of any perceivedviolations of ethical standards by participants, their par-ents/guardians, or members of the community at large.Measures to minimize bias in the trial will be subjectedto formal evaluation. Quality of randomization and allo-cation concealment will be evaluated by keeping sequen-tially numbered opaque envelopes containing therandomization code, which will be opened, signed anddated at the time of randomization. Quality of trial blind-ing will be evaluated by asking key trial persons (partici-pants, parents/guardians, medical officers, and nurses) toguess patients’ treatments at the end of their trial partici-pation, and compare the answers with the actual treat-ments, as previously described [62]. External independentoversight of trial quality will be performed by the DSMB,who will review trial quality indices periodically, as wellas an external trial auditor.DiscussionNitric oxide is an attractive candidate for an adjunctivetherapeutic agent for severe malaria given pre-clinical dataon its efficacy in animal models and an established trackrecord of safety in clinical practice and previous trials.Unlike other NO donor molecules, iNO has not beenreported to cause systemic vasodilation and hypotension[38]. Furthermore, unlike the NO precursor L-arginine,iNO does not require functional endothelial cell NOS,which may be compromised in patients with severe dis-ease. It is routinely used in clinical practice as an approvedagent for hypoxic respiratory failure in neonates, and hasan established track record of safety in critically ill patientpopulations. NO has been used in a wide variety of clinicalsettings in older children and adults including acuterespiratory distress syndrome, pulmonary hypertension,and pregnancy-induced hypertension [63].Outside the existing patent, iNO is relatively inexpen-sive [64], and can feasibly delivered by mask [47] in areaswith minimal health infrastructure. As currently mar-keted, INOmax from INO Therapeutics is cost-effectivein high-income countries for the treatment of respiratoryfailure in neonates [65], but may be prohibitively expen-sive in low- or middle-income countries. The real cost ofiNO (not just the price from a single company) is muchlower (medical grade iNO $1.99/h compared to $125/hfor INOmax) [66]. The patent for INOmax expires inHawkes et al. Trials 2011, 12:176http://www.trialsjournal.com/content/12/1/176Page 9 of 132013, leaving open the possibility for cheap manufactur-ing and commercialization of iNO in low-income set-tings, should iNO prove to be beneficial in severemalaria.Other adjunctive treatment strategies for severe malariapreviously tested in randomized controlled trials includeimmunomodulation, iron chelation, reduction of oxidativestress, anti-coagulation, volume expansion, reduction ofintracranial pressure, and prevention of seizure activity [6].Only one agent (albumin) was associated with a mortalitybenefit [6,67,68], although poor methodologic quality insome of these trials may have limited their ability to detectmeaningful treatment effects [69]. The search for an effec-tive adjuvant in severe malaria remains a worthwhile goalgiven the significant residual mortality with primary anti-parasitic treatment [4,5]. In this context, iNO, if demon-strated to be effective, would represent an importantadvance in malaria therapeutics.Our hypothesis that children with severe malaria willbenefit from adjunctive iNO can be answered using amodest sample size (n = 180). This parsimonious designwas made possible by selecting a quantitative biomarker ofmalaria severity, Ang-2, as the primary endpoint and byusing a powerful statistical technique (mixed-effects linearmodel). In contrast, a study using mortality (dichotomousvariable) as the primary endpoint would require a total ofover 1,000 patients to detect a 30% reduction from thebaseline mortality of 20% with 80% power, likely necessi-tating multicentre recruitment over several years andmonumental resources. Ang-2 is a surrogate but well vali-dated measure of malaria severity, appropriate for an earlyefficacy trial in a human population. Repeated longitudinalmeasurement of Ang-2 allows for increased precision inthe quantitative outcome variable, thereby reducing thenecessary sample size. The analytic plan, a mixed-effectslinear statistical model, includes random-effect terms,appropriate for representing clustered and thereforedependent data arising when data are gathered over timeon the same individuals [70]. We performed a computersimulation study to validate our sample size estimate usingassumptions based on previous studies of Ang-2 in severemalaria. In 9,000 simulated trial outcomes under differentassumptions for the treatment effect size and baselinevariability in Ang-2, the mixed-effects linear modeldetected a significant treatment effect with >80% powerunder most plausible scenarios. This a priori power calcu-lation provides further refinement on a crude sample sizecalculation and provides additional evidence that theplanned number of patients is adequate to test ourhypothesis. This sample size validation is particularlyimportant in light of a review highlighting that numerousprevious trials of adjunctive treatments for cerebralmalaria had insufficient statistical power to detect evenlarge treatment effects [69].Some unique aspects of our trial relate to its setting in aperipheral, resource-constrained pediatric hospital in sub-Saharan Africa. Standard clinical investigations includingquality-controlled microscopy, biochemistry and micro-biology, as well as equipment to monitor gas delivery (NO,NO2 and methemoglobin monitoring) need to be intro-duced to the facility for the trial. A commercially availableportable biochemistry instrument (i-STAT®, Abbott Pointof Care Inc., Princeton, NJ) and a pulse CO-oximeter fornon-invasive methemoglobin monitoring (Masimo Rad-57™, Masimo Corporation, Irvine, CA) will allow foronsite monitoring of critical investigations, with outsour-cing of other clinical testing to reference laboratories inKampala. Objective qualitative determination of parasite-mia at presentation for the purposes of trial enrolmentusing commercially available lateral flow immunochroma-tographic tests for parasite antigen detection (rapid diag-nostic tests) will be used to supplement local microscopy,which may be subject to error in the absence of rigorouslaboratory quality control [71]. While upgrading hospitalcapacity for our trial requires the infusion of considerableresources, it is hoped that this will result sustained localcapacity development in clinical management, laboratorydiagnostics, modern therapeutics, and innovation at theJinja Regional Referral Hospital, consistent with the ethicalobjectives of our trial.The cross-cultural setting of the trial poses certain ethi-cal challenges, demanding a sensitive approach to theinformed consent process. While some of these ethicalconsiderations are of a universal nature, others may bemore specific to the sub-Saharan African context. First,the trial involves a vulnerable pediatric population withsurrogate decision-makers. In addition to parent/guardianconsent in the local and education-appropriate language,we have built in a second assent process for competentparticipants, given that children may legitimately andautonomously participate in decisions related to their ownhealthcare [57,72]. Next, the acuity and lethality of theunderlying infection demand that consent be obtainedearly after admission for maximal treatment benefit, yetmust not interfere with the emergency management of cri-tically ill participants and must give adequate time forstressed parents/guardians to give due considerationbefore consenting. The complexity of the scientific design(randomized, placebo-controlled adjunctive therapy, withblinding of treatment allocation) together with variableeducation level and familiarity with biomedical research ofparents/guardians presents difficulties requiring carefulexplanation during the consent discussion (e.g., conveyingthe position of scientific equipoise in order to accept pos-sible randomization to placebo). Additional ethical consid-erations which may be more specific to the Africancontext include the “therapeutic misconception,” (partici-pants may not clearly distinguish research from healthHawkes et al. Trials 2011, 12:176http://www.trialsjournal.com/content/12/1/176Page 10 of 13interventions) [73,74]. Furthermore, African parents fre-quently express concern about blood taking, includingfears about the misuse of the blood, unauthorized testingfor HIV, long-term storage, genetic testing, and harm tothe child from excessive blood loss [73]. Our consent pro-cess explicitly addresses the distinction between study andclinical interventions, as well as detailed descriptions ofvolumes, frequency and subsequent handling of bloodsamples in the trial. Finally, we have incorporated a com-munity engagement (CE) plan into our trial design to fos-ter a trusting relationship with the surrounding catchmentpopulation.Monitoring patient safety in a trial involving critically illchildren in a resource-limited environment poses addi-tional challenges. The administration of study gas will betightly regulated and monitored with state-of-the-art tech-nology, with particular attention to two dose-dependent,reversible adverse effects: elevated inspired NO2 andmethemoglobinemia. Strict criteria for study gas disconti-nuation have been established. Standardized pediatric toxi-city tables will be used to monitor for other adverse eventsin a blinded and objective manner. We will also monitormortality using statistical control charts, in order torationally detect deviations from the expected baselinemortality of 20% [42,75]. This approach involves striking abalance between the earliest detection of elevated mortal-ity in order to institute corrective measures (“true alarm”),and the risk of halting the trial unnecessarily for variationsin mortality due to chance alone (“false alarm”). Theupper control limit of mortality that will prompt a safetyreview has thus been defined using statistical principles,together with the predicted performance of this surveil-lance strategy (Additional File 1). Trial safety oversightwill ultimately be provided by an independent DSMB whowill meet regularly to review recruitment, interim evidenceof efficacy, safety and trial quality.Limitations of this trial design include its use of a sur-rogate marker, Ang-2, as the primary endpoint. Thisallows for an efficient design but may be less compellingto clinicians than would a demonstrable mortality bene-fit (of note, mortality has been included as a secondarytrial endpoint). On the other hand, Ang-2 has beenextensively validated as an objective and quantitativebiomarker of malaria severity [28,29,76,77]. Thus, Ang-2is an appropriate endpoint for an initial efficacy study ofiNO in a human population, but a promising treatmenteffect would require validation in costly, large, multicen-tre trials. Blinding presents a challenge for a gaseoustherapy requiring monitoring for dose-dependent toxici-ties, which we have addressed by separating study tasksbetween a blinded team (responsible for clinical care,assessment of endpoints and laboratory testing) and anunblinded team (responsible for administering treatmentand monitoring safety parameters) to minimize potentialbias. Risks and costs related to conducting the study ina resource-constrained setting are balanced with poten-tial benefits in terms of local capacity building, the highincidence of severe malaria allowing timely completionof the study, and applicability of the therapy in othermalaria-endemic areas.In summary, based on compelling data supporting theefficacy of iNO in experimental cerebral malaria in ani-mal models, coupled with the documented safety ofiNO in clinical practice and trials for other diseases, wehave outlined a protocol for a randomized clinical trialof iNO for the adjunctive treatment of severe malaria inUgandan children. If our study demonstrates a signifi-cant treatment effect, this would represent a major andimportant advance in the treatment of severe malariawith broad potential for global public health impact.Additional materialAdditional file 1: Details of a process control chart to monitormortality. This Microsoft Word file provides details of the statisticalthresholds and the performance characteristics of a statistical controlchart to monitor mortality in this trial of critically ill children.List of abbreviationsAng-1: Angiopoietin-1; Ang-2: Angiopoietin-2; CE: community engagement;CONSORT: Consolidated Standards of Reporting Trials; DPTA/NO:dipropylenetriamine NONOate; DSMB: Data and Safety Monitoring Board;ECMO: extracorporeal membrane oxygenation; HRP-2: histidine-rich protein-2; ICAM-1: intercellular cell adhesion molecule-1; iNO: inhaled nitric oxide;NO: nitric oxide; NO2:nitrogen dioxide; NOS: nitric oxide synthase; PE:parasitized erythrocyte; pLDH: parasite lactate dehydrogenase; QM: qualitymanagement; RDT: rapid diagnostic test; sGC: soluble guanylate cyclise; SMA:severe malarial anemia; SOP: standard operating procedure; vWF: vonWillebrand Factor; WPB: Weibel-Palade bodies.AcknowledgementsWe thank Jeremy Dabor, Laura Erdman, Lena Serghides, Hani Kim, ArisaGoldstone, Renaud Boulanger, and Susan Graham for their help throughoutthe planning of this trial. This work was supported by the CanadianInstitutes of Health Research (CIHR) MOP-244701 and 13721 [KCK], CIHRTeam grant in malaria, Canada Research Chair [KCK], Doctoral ResearchAward [ALC] and Clinician-scientist Award [MH] as well as kind donationfrom Mr David S. “Kim” Kertland.Author details1Institute of Medical Sciences, University of Toronto, Canada. 2Division ofInfectious Diseases, Department of Pediatrics, The Hospital for Sick Children,Toronto, Canada. 3Department of Paediatrics and Child Health, MulagoHospital and Makerere University, Kampala, Uganda. 4Department ofPaediatrics, Jinja Regional Referral Hospital, Jinja, Uganda. 5Department ofRespiratory Medicine, Faculty of Medicine, University of British Columbia,Vancouver, Canada. 6Dalla Lana School of Public Health, University ofToronto, Canada. 7Applied Health Research Centre, St Michael’s Hospital,Toronto, Ontario, Canada. 8Centre for Global Health Research, The KeenanResearch Centre, Li Ka Shing Knowledge Institute, St. Michael’s Hospital,Toronto, Canada. 9Department of Public Health Sciences and Joint Centrefor Bioethics at the University of Toronto, Canada. 10Department ofLaboratory Medicine and Pathobiology, University of Toronto, Canada.11Department of Medicine, University of Toronto, Toronto, Canada. 12SandraA. Rotman Laboratories, McLaughlin-Rotman Centre for Global Health,Toronto, Canada. 13McLaughlin Centre for Molecular Medicine, Toronto,Hawkes et al. Trials 2011, 12:176http://www.trialsjournal.com/content/12/1/176Page 11 of 13Canada. 14Tropical Disease Unit, Toronto General Hospital, Toronto, Canada.15Division of Global Pediatrics, Department of Pediatrics, University ofMinnesota, Minnesota, USA.Authors’ contributionsMH designed the study and wrote the manuscript. ROO participated instudy design and logistical planning, and obtained ethical approval from theUgandan bodies. SN participated in study design and logistical planning. CMparticipated in the study design and provided expertise related to iNOadministration and safety monitoring. KET provided statistical input,designed the primary statistical analysis, and performed the computersimulation experiment for validation of the sample size. JVL providedexpertise in ethics and community engagement. ALC participated in thestudy design and helped to draft the manuscript. WCL participated in thestudy design and helped to draft the manuscript. CCJ participated in thestudy design and helped to draft the manuscript. KCK conceived the study,participated in the study design and helped to draft the manuscript. Allauthors read and approved the final manuscript.Competing interestsCM is Chief Scientific Officer of Nitric Solutions Inc., developer of nitric oxide(NO) based medical products.KCK, WCL, and ALC are listed as inventors on a patent owned by UniversityHealth Network (Toronto) related to Ang-2 as a biomarker for infectiousdiseases that compromise endothelial integrity.All other authors: no conflicts.Received: 24 January 2011 Accepted: 13 July 2011Published: 13 July 2011References1. WHO: World malaria report 2008. WHO, Geneva, Switzerland; 2008.2. John CC, Bangirana P, Byarugaba J, Opoka RO, Idro R, Jurek AM, Wu B,Boivin MJ: Cerebral malaria in children is associated with long-termcognitive impairment. Pediatrics 2008, 122:e92-99.3. Newton CR, Krishna S: Severe falciparum malaria in children: currentunderstanding of pathophysiology and supportive treatment. PharmacolTher 1998, 79:1-53.4. Dondorp A, Nosten F, Stepniewska K, Day N, White N: Artesunate versusquinine for treatment of severe falciparum malaria: a randomised trial.Lancet 2005, 366:717-725.5. Dondorp AM, Fanello CI, Hendriksen IC, Gomes E, Seni A, Chhaganlal KD,Bojang K, Olaosebikan R, Anunobi N, Maitland K, et al: Artesunate versusquinine in the treatment of severe falciparum malaria in African children(AQUAMAT): an open-label, randomised trial. Lancet 376:1647-1657.6. John CC, Kutamba E, Mugarura K, Opoka RO: Adjunctive therapy forcerebral malaria and other severe forms of Plasmodium falciparummalaria. Expert Rev Anti Infect Ther 8:997-1008.7. Korhonen R, Lahti A, Kankaanranta H, Moilanen E: Nitric oxide productionand signaling in inflammation. Curr Drug Targets Inflamm Allergy 2005,4:471-479.8. Bogdan C: Nitric oxide and the immune response. Nat Immunol 2001,2:907-916.9. Matsushita K, Morrell CN, Cambien B, Yang SX, Yamakuchi M, Bao C,Hara MR, Quick RA, Cao W, O’Rourke B, et al: Nitric oxide regulatesexocytosis by S-nitrosylation of N-ethylmaleimide-sensitive factor. Cell2003, 115:139-150.10. Palmer RM, Ferrige AG, Moncada S: Nitric oxide release accounts for thebiological activity of endothelium-derived relaxing factor. Nature 1987,327:524-526.11. Gramaglia I, Sobolewski P, Meays D, Contreras R, Nolan JP, Frangos JA,Intaglietta M, van der Heyde HC: Low nitric oxide bioavailabilitycontributes to the genesis of experimental cerebral malaria. Nat Med2006, 12:1417-1422.12. Anstey NM, Weinberg JB, Hassanali MY, Mwaikambo ED, Manyenga D,Misukonis MA, Arnelle DR, Hollis D, McDonald MI, Granger DL: Nitric oxidein Tanzanian children with malaria: inverse relationship between malariaseverity and nitric oxide production/nitric oxide synthase type 2expression. J Exp Med 1996, 184:557-567.13. Lopansri BK, Anstey NM, Weinberg JB, Stoddard GJ, Hobbs MR,Levesque MC, Mwaikambo ED, Granger DL: Low plasma arginineconcentrations in children with cerebral malaria and decreased nitricoxide production. Lancet 2003, 361:676-678.14. Yeo TW, Lampah DA, Tjitra E, Gitawati R, Darcy CJ, Jones C, Kenangalem E,McNeil YR, Granger DL, Lopansri BK, et al: Increased asymmetricdimethylarginine in severe falciparum malaria: association with impairednitric oxide bioavailability and fatal outcome. PLoS Pathog 6:e1000868.15. Kun JF, Mordmuller B, Perkins DJ, May J, Mercereau-Puijalon O, Alpers M,Weinberg JB, Kremsner PG: Nitric oxide synthase 2(Lambarene) (G-954C),increased nitric oxide production, and protection against malaria. J InfectDis 2001, 184:330-336.16. Dhangadamajhi G, Mohapatra BN, Kar SK, Ranjit MR: The CCTTTpentanucleotide microsatellite in iNOS promoter influences the clinicaloutcome in P. falciparum infection. Parasitol Res 2009, 104:1315-1320.17. Boutlis CS, Hobbs MR, Marsh RL, Misukonis MA, Tkachuk AN, Lagog M,Booth J, Granger DL, Bockarie MJ, Mgone CS, et al: Inducible nitric oxidesynthase (NOS2) promoter CCTTT repeat polymorphism: relationship toin vivo nitric oxide production/NOS activity in an asymptomatic malaria-endemic population. Am J Trop Med Hyg 2003, 69:569-573.18. Ohashi J, Naka I, Patarapotikul J, Hananantachai H, Looareesuwan S,Tokunaga K: Significant association of longer forms of CCTTTMicrosatellite repeat in the inducible nitric oxide synthase promoterwith severe malaria in Thailand. J Infect Dis 2002, 186:578-581.19. Kun JF, Mordmuller B, Lell B, Lehman LG, Luckner D, Kremsner PG:Polymorphism in promoter region of inducible nitric oxide synthasegene and protection against malaria. Lancet 1998, 351:265-266.20. Sobolewski P, Gramaglia I, Frangos J, Intaglietta M, van der Heyde HC:Nitric oxide bioavailability in malaria. Trends Parasitol 2005, 21:415-422.21. Day NP, Phu NH, Mai NT, Chau TT, Loc PP, Chuong LV, Sinh DX,Holloway P, Hien TT, White NJ: The pathophysiologic and prognosticsignificance of acidosis in severe adult malaria. Crit Care Med 2000,28:1833-1840.22. MacPherson GG, Warrell MJ, White NJ, Looareesuwan S, Warrell DA: Humancerebral malaria. A quantitative ultrastructural analysis of parasitizederythrocyte sequestration. Am J Pathol 1985, 119:385-401.23. Turner GD, Morrison H, Jones M, Davis TM, Looareesuwan S, Buley ID,Gatter KC, Newbold CI, Pukritayakamee S, Nagachinta B, et al: Animmunohistochemical study of the pathology of fatal malaria. Evidencefor widespread endothelial activation and a potential role forintercellular adhesion molecule-1 in cerebral sequestration. Am J Pathol1994, 145:1057-1069.24. De Caterina R, Libby P, Peng HB, Thannickal VJ, Rajavashisth TB,Gimbrone MA Jr, Shin WS, Liao JK: Nitric oxide decreases cytokine-induced endothelial activation. Nitric oxide selectively reducesendothelial expression of adhesion molecules and proinflammatorycytokines. J Clin Invest 1995, 96:60-68.25. Jakobsen PH, Morris-Jones S, Ronn A, Hviid L, Theander TG, Elhassan IM,Bygbjerg IC, Greenwood BM: Increased plasma concentrations of sICAM-1,sVCAM-1 and sELAM-1 in patients with Plasmodium falciparum or P.vivax malaria and association with disease severity. Immunology 1994,83:665-669.26. de Mast Q, Groot E, Lenting PJ, de Groot PG, McCall M, Sauerwein RW,Fijnheer R, van der Ven A: Thrombocytopenia and release of activatedvon Willebrand Factor during early Plasmodium falciparum malaria.J Infect Dis 2007, 196:622-628.27. Hollestelle MJ, Donkor C, Mantey EA, Chakravorty SJ, Craig A, Akoto AO,O’Donnell J, van Mourik JA, Bunn J: von Willebrand factor propeptide inmalaria: evidence of acute endothelial cell activation. Br J Haematol 2006,133:562-569.28. Yeo TW, Lampah DA, Gitawati R, Tjitra E, Kenangalem E, Piera K, Price RN,Duffull SB, Celermajer DS, Anstey NM: Angiopoietin-2 is associated withdecreased endothelial nitric oxide and poor clinical outcome in severefalciparum malaria. Proc Natl Acad Sci USA 2008, 105:17097-17102.29. Lovegrove FE, Tangpukdee N, Opoka RO, Lafferty EI, Rajwans N, Hawkes M,Krudsood S, Looareesuwan S, John CC, Liles WC, Kain KC: Serumangiopoietin-1 and -2 levels discriminate cerebral malaria fromuncomplicated malaria and predict clinical outcome in African children.PLoS One 2009, 4:e4912.30. Bridges DJ, Bunn J, van Mourik JA, Grau G, Preston RJ, Molyneux M,Combes V, O’Donnell JS, de Laat B, Craig A: Rapid activation ofendothelial cells enables Plasmodium falciparum adhesion to platelet-decorated von Willebrand factor strings. Blood 2010, 115:1472-1474.Hawkes et al. Trials 2011, 12:176http://www.trialsjournal.com/content/12/1/176Page 12 of 1331. Fiedler U, Reiss Y, Scharpfenecker M, Grunow V, Koidl S, Thurston G,Gale NW, Witzenrath M, Rosseau S, Suttorp N, et al: Angiopoietin-2sensitizes endothelial cells to TNF-alpha and has a crucial role in theinduction of inflammation. Nat Med 2006, 12:235-239.32. Parikh SM, Mammoto T, Schultz A, Yuan HT, Christiani D, Karumanchi SA,Sukhatme VP: Excess circulating angiopoietin-2 may contribute topulmonary vascular leak in sepsis in humans. PLoS Med 2006, 3:e46.33. Saharinen P, Eklund L, Miettinen J, Wirkkala R, Anisimov A, Winderlich M,Nottebaum A, Vestweber D, Deutsch U, Koh GY, et al: Angiopoietinsassemble distinct Tie2 signalling complexes in endothelial cell-cell andcell-matrix contacts. Nat Cell Biol 2008, 10:527-537.34. Thurston G, Suri C, Smith K, McClain J, Sato TN, Yancopoulos GD,McDonald DM: Leakage-resistant blood vessels in mice transgenicallyoverexpressing angiopoietin-1. Science 1999, 286:2511-2514.35. Fukuhara S, Sako K, Minami T, Noda K, Kim HZ, Kodama T, Shibuya M,Takakura N, Koh GY, Mochizuki N: Differential function of Tie2 at cell-cellcontacts and cell-substratum contacts regulated by angiopoietin-1. NatCell Biol 2008, 10:513-526.36. Yeo TW, Lampah DA, Gitawati R, Tjitra E, Kenangalem E, McNeil YR,Darcy CJ, Granger DL, Weinberg JB, Lopansri BK, et al: Impaired nitric oxidebioavailability and L-arginine reversible endothelial dysfunction in adultswith falciparum malaria. J Exp Med 2007, 204:2693-2704.37. Finer NN, Barrington KJ: Nitric oxide for respiratory failure in infants bornat or near term. Cochrane Database Syst Rev 2006, CD000399.38. Adhikari NK, Burns KE, Friedrich JO, Granton JT, Cook DJ, Meade MO: Effectof nitric oxide on oxygenation and mortality in acute lung injury:systematic review and meta-analysis. BMJ 2007, 334:779.39. Schulz KF, Altman DG, Moher D: CONSORT 2010 statement: updatedguidelines for reporting parallel group randomised trials. Bmj 340:c332.40. WHO: Malaria rapid diagnostic test performance: results of WHO producttesting of malaria RDTs: round 1. Geneva, Switzerland: WHO; 2008.41. WHO: Management of severe malnutrition: a manual for physicians andother senior health workers. Geneva, Switzerland: WHO; 1999.42. Idro R, Aloyo J, Mayende L, Bitarakwate E, John CC, Kivumbi GW: Severemalaria in children in areas with low, moderate and high transmissionintensity in Uganda. Trop Med Int Health 2006, 11:115-124.43. Inhaled nitric oxide in full-term and nearly full-term infants with hypoxicrespiratory failure. The Neonatal Inhaled Nitric Oxide Study Group. NEngl J Med 1997, 336:597-604.44. Davidson D, Barefield ES, Kattwinkel J, Dudell G, Damask M, Straube R,Rhines J, Chang CT: Inhaled nitric oxide for the early treatment ofpersistent pulmonary hypertension of the term newborn: a randomized,double-masked, placebo-controlled, dose-response, multicenter study.The I-NO/PPHN Study Group. Pediatrics 1998, 101:325-334.45. Head CA, Swerdlow P, McDade WA, Joshi RM, Ikuta T, Cooper ML,Eckman JR: Beneficial effects of nitric oxide breathing in adult patientswith sickle cell crisis. Am J Hematol 85:800-802.46. Weiner DL, Hibberd PL, Betit P, Cooper AB, Botelho CA, Brugnara C:Preliminary assessment of inhaled nitric oxide for acute vaso-occlusivecrisis in pediatric patients with sickle cell disease. Jama 2003,289:1136-1142.47. Long R, Jones R, Talbot J, Mayers I, Barrie J, Hoskinson M, Light B: Inhalednitric oxide treatment of patients with pulmonary tuberculosisevidenced by positive sputum smears. Antimicrob Agents Chemother 2005,49:1209-1212.48. Tran TH, Day NP, Nguyen HP, Nguyen TH, Pham PL, Dinh XS, Ly VC, Ha V,Waller D, Peto TE, White NJ: A controlled trial of artemether or quinine inVietnamese adults with severe falciparum malaria. N Engl J Med 1996,335:76-83.49. van Hensbroek MB, Onyiorah E, Jaffar S, Schneider G, Palmer A, Frenkel J,Enwere G, Forck S, Nusmeijer A, Bennett S, et al: A trial of artemether orquinine in children with cerebral malaria. N Engl J Med 1996, 335:69-75.50. Troncy E, Collet JP, Shapiro S, Guimond JG, Blair L, Ducruet T, Francoeur M,Charbonneau M, Blaise G: Inhaled nitric oxide in acute respiratory distresssyndrome: a pilot randomized controlled study. Am J Respir Crit Care Med1998, 157:1483-1488.51. Griffiths MJ, Evans TW: Inhaled nitric oxide therapy in adults. N Engl J Med2005, 353:2683-2695.52. Leavey JF, Dubin RL, Singh N, Kaminsky DA: Silo-Filler’s disease, the acuterespiratory distress syndrome, and oxides of nitrogen. Ann Intern Med2004, 141:410-411.53. Elsayed NM: Toxicity of nitrogen dioxide: an introduction. Toxicology 1994,89:161-174.54. Svolba G, Bauer P: Statistical quality control in clinical trials. Control ClinTrials 1999, 20:519-530.55. NIAID: Division of Microbiology and Infectious Diseases Pediatric ToxicityTables (revised 11.21.07)..56. WHO: WHO standard definitions for adverse event [http://www.who.int/vaccines-documents/DocsPDF05/815.pdf.].57. Wendler DS: Assent in paediatric research: theoretical and practicalconsiderations. J Med Ethics 2006, 32:229-234.58. Newman PA: Towards a science of community engagement. Lancet 2006,367:302.59. Tindana PO, Singh JA, Tracy CS, Upshur RE, Daar AS, Singer PA, Frohlich J,Lavery JV: Grand challenges in global health: community engagement inresearch in developing countries. PLoS Med 2007, 4:e273.60. Lavery JV, Tinadana PO, Scott TW, Harrington LC, Ramsey JM, Ytuarte-Nunez C, James AA: Towards a framework for community engagement inglobal health research. Trends Parasitol 26:279-283.61. Boivin MJ, Bangirana P, Byarugaba J, Opoka RO, Idro R, Jurek AM, John CC:Cognitive impairment after cerebral malaria in children: a prospectivestudy. Pediatrics 2007, 119:e360-366.62. Hrobjartsson A, Forfang E, Haahr MT, Als-Nielsen B, Brorson S: Blinded trialstaken to the test: an analysis of randomized clinical trials that reporttests for the success of blinding. Int J Epidemiol 2007, 36:654-663.63. Sokol J, Jacobs SE, Bohn D: Inhaled nitric oxide for acute hypoxemicrespiratory failure in children and adults. Cochrane Database Syst Rev2003, CD002787.64. Hansen TW: Inhaled nitric oxide and the societal perspective. Pediatrics2004, 113:1849-1851; author reply 1849-1851.65. Angus DC, Clermont G, Watson RS, Linde-Zwirble WT, Clark RH, Roberts MS:Cost-effectiveness of inhaled nitric oxide in the treatment of neonatalrespiratory failure in the United States. Pediatrics 2003, 112:1351-1360.66. Pierce CM, Peters MJ, Cohen G, Goldman AP, Petros AJ: Cost of nitric oxideis exorbitant. Bmj 2002, 325:336.67. Maitland K, Nadel S, Pollard AJ, Williams TN, Newton CR, Levin M:Management of severe malaria in children: proposed guidelines for theUnited Kingdom. Bmj 2005, 331:337-343.68. Akech S, Gwer S, Idro R, Fegan G, Eziefula AC, Newton CR, Levin M,Maitland K: Volume expansion with albumin compared to gelofusine inchildren with severe malaria: results of a controlled trial. PLoS Clin Trials2006, 1:e21.69. Enwere G: A review of the quality of randomized clinical trials ofadjunctive therapy for the treatment of cerebral malaria. Trop Med IntHealth 2005, 10:1171-1175.70. Laird NM, Ware JH: Random-effects models for longitudinal data.Biometrics 1982, 38:963-974.71. Hawkes M, Kain KC: Advances in malaria diagnosis. Expert Rev Anti InfectTher 2007, 5:485-495.72. Henschel AD, Rothenberger LG, Boos J: Randomized clinical trials inchildren–ethical and methodological issues. Curr Pharm Des 16:2407-2415.73. Molyneux CS, Peshu N, Marsh K: Trust and informed consent: insightsfrom community members on the Kenyan coast. Soc Sci Med 2005,61:1463-1473.74. Lynoe N, Hyder Z, Chowdhury M, Ekstrom L: Obtaining informed consentin Bangladesh. N Engl J Med 2001, 344:460-461.75. Kyu HH, Fernandez E: Artemisinin derivatives versus quinine for cerebralmalaria in African children: a systematic review. Bull World Health Organ2009, 87:896-904.76. Conroy AL, Lafferty EI, Lovegrove FE, Krudsood S, Tangpukdee N, Liles WC,Kain KC: Whole blood angiopoietin-1 and -2 levels discriminate cerebraland severe (non-cerebral) malaria from uncomplicated malaria. Malar J2009, 8:295.77. Conroy AL, Phiri H, Hawkes M, Glover S, Mallewa M, Seydel KB, Taylor TE,Molyneux ME, Kain KC: Endothelium-based biomarkers are associatedwith cerebral malaria in malawian children: a retrospective case-controlstudy. PLoS One 5:e15291.doi:10.1186/1745-6215-12-176Cite this article as: Hawkes et al.: Inhaled nitric oxide for the adjunctivetherapy of severe malaria: Protocol for a randomized controlled trial.Trials 2011 12:176.Hawkes et al. Trials 2011, 12:176http://www.trialsjournal.com/content/12/1/176Page 13 of 13


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