hemoglobinuria, jaundice, shock, renal failure, severe lactic acidosis, abnormal bleeding, pulmonary edema, and adult respiratory distress syndrome, Kussmaul’s breathing ( ; ).
Compared with adults, children have a higher incidence rate of seizures ( Postels and Birbeck, 2013 ). In children, focal motor and generalized tonic–clonic convulsions are the most common clinically detected seizures; however, subtle or subclinical seizures detected with electroencephalography (EEG) are also common ( Newton et al., 2000 ; Postels and Birbeck, 2013 ). Subtle seizures manifest as nystagmoid eye movements, irregular breathing, excessive salivation, and conjugate eye deviation ( Crawley et al., 1996 ). Most seizures in adult CM patients are generalized seizures; however, focal motor seizures may also occur. Occasionally, the sign of seizure activity is subtle, such as repetitive eye or hand movements, and may be easily overlooked. Subtle seizure activity seems to be more common in children than in adults ( Newton et al., 2000 ). The level of consciousness after a seizure is usually lower than that preceding it. Status epilepticus is unusual in adults, although more than one seizure is frequent ( Vespa et al., 1999 ). Previous studies reported an association between status epilepticus and neurological sequelae among CM patients, which occur in 5-15% of survivors ( Brewster et al., 1990 ), and it has been shown that prolonged seizure activity may damage the brain, causing deficits in both motor and cognitive functions ( Stafstrom et al., 1993 ).
Coma usually develops rapidly after seizures among children living in malaria-endemic areas, and consciousness recovers to normal rapidly (within 24-48 h) ( Genton et al., 1997 ). Different disease processes may affect awareness in children with malaria, including convulsions, hypoglycemia, hyperpyrexia, acidosis, severe anemia, and sedative drugs. Although the cause of impaired consciousness or coma remains unclear, it is likely to result from several interacting mechanisms ( Newton et al., 2000 ). Adhesion of malaria parasite-infected red blood cells (iRBCs) reduces microvascular blood flow ( Kaul et al., 1998 ), which may be the cause of organ tissue dysfunction, such as coma. High concentrations of tumor necrosis factor-α (TNF-α) are associated with coma ( Kwiatkowski et al., 1990 ; Kaul et al., 1998 ). Compared to children, coma gradually develops in adults following drowsiness, disorientation, delirium, and agitation within 2 to 3 days ( Kochar et al., 2002 ). Convulsion leads to the development of a coma and occurs in approximately 15% of adults and 80% of children ( Plewes et al., 2018 ).
Abnormal corneal and oculocephalic reflexes (doll’s eye) are likely to occur in children with deep coma. Abnormal plantar reflexes are also detected, and abdominal reflexes are almost invariably absent. In adults with profound coma, corneal and eyelash reflexes are usually intact unless in a state of deep coma, and the pupils are normal. Forcible jaw closure and teeth grinding (bruxism) are commonly seen in CM. Pout reflex usually indicates a “frontal release”; however, the grasp reflex is frequently absent. In addition, increased muscle tone and tendon reflexes are found. CM may elicit ankle and patellar clonus, and extensor plantar responses. Nevertheless, abdominal and cremasteric reflexes are invariably absent ().
CM affects the CNS, and although most survivors have a full recovery, 3-31% of patients still develop neurological deficits and cognitive sequelae ( Oluwayemi et al., 2013 ). The prevalence of neurological deficits is higher in children than in adults, ranging from 6% to 29% at the time of discharge ( Idro et al., 2004 ; Hawkes et al., 2013 ). Children with CM frequently present long-term neurologic deficits, and episodes of CM imply the development of long-term sequelae in children. In children, the most common sequelae include ataxia, paralysis, paresis, cortical blindness, epilepsy, deafness, behavioral disorders, language disorders, and cognitive impairment ( Brewster et al., 1990 ). Sequelae are less common in adults. During the acute phase of CM, neurologic abnormalities include psychosis, ataxia, transitory cranial nerve palsies or tremor ( Peixoto and Kalei, 2013 ).
The characteristic features of retinopathy due to CM include retinal whitening (macula whitening sparing central fovea and peripheral whitening of the fundus), retinal vessel discoloration to pink–orange or white, retinal hemorrhages, and papilledema ( Hora et al., 2016 ). The first two abnormalities are considered specific symptoms of CM. Commonalities between pediatric and adult patients include retinal hemorrhage, a common manifestation but a less distinctive feature. Retinal hemorrhage correlates with disease severity and cerebral hemorrhage in the microvascular dissection of the brain ( White et al., 2001 ). Papilledema is rare in children and adults. Although it is a nonspecific symptom of CM, it reflects increased intracranial pressure and portends a poor prognosis in children ( Beare et al., 2004 ). A prominent difference between children and adults is vessel discoloration. Orange or white discoloration of the retinal vessels has been attributed to the hemoglobinization of stationary erythrocytes infected with mature parasites ( Beare et al., 2011 ). The degree of retinal microvascular damage is comparable to cerebral damage ( Beare et al., 2004 ; Lewallen et al., 2008 ).
Systemic complications include anemia (20% to 50% incidence), hypoglycemia (30% incidence), hyponatremia (>50% incidence), jaundice (8% incidence), metabolic acidosis characterized by respiratory distress, and hepatosplenomegaly in children living with CM ( White et al., 1987 ; English et al., 1996 ; Idro et al., 2005 ; Maitland and Newton, 2005 ). Renal failure and pulmonary edema are unusual in children ( Newton et al., 1991 ). CM predominantly manifests as CNS dysfunction in children; however, it is mainly present in multisystem and organ (circulatory, hepatic, coagulation, renal, and pulmonary) dysfunctions in adults ( Day et al., 2000 ; Krishnan and Karnad, 2003 ).
In adults, anemia is an inevitable consequence of CM and develops exceptionally rapidly. CM has been reported in patients together with pulmonary edema, adult respiratory distress syndrome and hemoglobinuria, and Kussmaul’s breathing occurs with acute renal failure and severe lactic acidosis ( Newton and Warrell, 1998 ). Hypoglycemia occurs in 8% of patients aged 26 to 28 years ( White et al., 1983 ). Other complications included jaundice, shock, abnormal bleeding, and coagulopathy.
Although the pathophysiology of CM has been extensively investigated, the exact pathogenesis remains unclear. Currently, CM is widely accepted as a multifactorial process related to the adhesion and sequestration of iRBCs, immunological responses, endothelial cell (EC) activation, and loss of BBB integrity ( Idro et al., 2005 ). Nevertheless, any of these mechanisms alone fail to explain the pathogenesis of human CM, and they jointly participate in this potentially fatal infection. A mouse model of experimental cerebral malaria (ECM) has been used to simulate and explain the pathogenesis of human CM ( Figure 1 ).
Schematic of experimental cerebral malaria (ECM) pathogenesis. The ECM is initiated by dendritic cells (DCs) processing and presenting infected red blood cell (iRBC) antigens to CD4 + and CD8 + T cells in the spleen (1). NK cells and macrophages are activated by iRBCs to secrete inflammatory cytokines (2). The iRBCs adhere to endothelial cells (ECs) of the brain microvasculature through the interaction between P. falciparum erythrocyte membrane protein-1 (PfEMP-1) of iRBCs and cell adhesion molecules of ECs (3). The adhesion of iRBCs to the cerebral microvascular endothelium is also further accompanied by agglutination to other iRBCs, platelets, white blood cells (WBCs), and the rosetting effect formed by the adhesion of iRBCs and RBCs. ECs are activated by interactions with iRBCs and responses to inflammatory cytokines. Activated ECs promote the upregulation of cell adhesion molecules (CAMs) on brain ECs and release cytokines and chemokines simultaneously (4). Activated CD8 + T cells express CXCR3 and CCR5 chemokine receptors, which bind to chemokines such as CXCL9, CXCL10, and CXCL4, inducing T-cell migration to the brain (5). Meanwhile, LFA-1 on CD8 + T cells promotes their adhesion to endothelial ICAM-1 (6). Parasitic antigens can be transferred from the vascular lumen to brain ECs. Brain ECs can cross-present parasitic antigens on MHC-1 molecular antigens and bind with antigen receptors (TCRs) on CD8 + T cells (7). The interaction induces apoptosis of ECs, leading to the destruction of the BBB (8). Meanwhile, the iRBCs directly activate platelets and stimulate the release of CXCL4. CXCL4 induces the production of TNF by T cells and macrophages, which causes more platelets to adhere to ECs (9). As leukocytes and platelets are recruited and activated, a local proinflammatory cycle ensues, with a positive feedback loop of EC activation, leukocyte/platelet sequestration, and parasite accumulation (10).
Cerebral iRBCs adherence is an indicative marker of CM in adults and children, and it is considered a starting point during the development of CM. Sequestration is thought to be a specific interaction between iRBCs and vascular ECs, which is not limited to brain tissues but also occurs on ECs in different organs, including the lung, kidney, liver, and intestine.
The adhesion of iRBCs to the vascular endothelium is mediated by P. falciparum erythrocyte membrane protein 1 (PfEMP1) ( Jensen et al., 2020 ), a specific cell-surface ligand expressed by iRBCs. PfEMP1 belongs to the antigen-variant protein family, and the var genes encoding the protein are a large multigene family ( Kim, 2012 ). To date, 60 different var genes have been characterized, and var gene-encoded proteins have shown dual functions in regulating antigen variation and cell adhesion ( Tembo et al., 2014 ). PfEMP1 contains a host molecule binding domain and binds to several cell adhesion molecules (CAMs) on ECs, such as CD36 ( Berendt et al., 1989 ; Ockenhouse et al., 1989 ), intercellular adhesion molecule 1 (ICAM-1) ( Berendt et al., 1989 ), vascular adhesion molecule 1 (VCAM-1) ( Ockenhouse et al., 1989 ; Ockenhouse et al., 1992 ), endothelial protein C receptor (EPCR) ( Mohan Rao et al., 2014 ), thrombospondin, E-selectin ( Turner et al., 1994 ) and chondroitin sulphate A ( Rogerson et al., 1995 ; Fried and Duffy, 1996 ). Adhesion of iRBCs to the cerebral microvascular endothelium is further accompanied by agglutination to other iRBCs, platelets, white blood cells (WBCs), and rosetting produced by adhesion of iRBCs and uninfected erythrocytes ( Fried and Duffy, 1996 ). Sequestration of iRBCs in microvessels may protect iRBCs from clearance by the spleen. In addition, it weakens the capability of iRBCs and RBCs to denature, leading to blood vessel blockage. Previous studies reported a significant correlation between sequestration of iRBCs in cerebral vessels and coma in CM patients ( Silamut et al., 1999 ; Ponsford et al., 2012 ; Storm et al., 2019 ). Taken together, sequestration of iRBCs leads to increased vasoconstriction and vascular obstruction, as well as decreased cerebral blood flow and hypoxia.
Excessive immune responses and the release of a large number of inflammatory factors play important roles in the pathogenesis of CM ( Shikani et al., 2012 ). The humoral response to malaria parasites includes immune activation of macrophages and lymphocytes (CD8 + , CD4 + , natural killer (NK) cells) and activation of monocytes, resulting in accumulation of immune cells in the microvasculature and a systemic inflammatory response secreted by proinflammatory cytokines, including tumor necrosis factor (TNF)-α, interferon (IFN)-γ, and interleukin-1β (IL-1β), which are elevated in an episode of acute CM.
At the early stage of malaria infection, CD4 + and CD8 + T cells are activated by antigen-presenting cells (APCs) to initiate antimalarial protective cellular immune responses. The chemotaxis of T cells to peripheral cerebral vessels is one of the prominent features of CM. Recruitment of CD8 + T cells is the most predominant characteristic ( Riggle et al., 2020 ), and priming of CD4 + and CD8 + T cells initiates CM in the spleen by dendritic cells (DCs) presenting iRBCs antigens. NK cells and macrophages are activated by iRBCs to release inflammatory cytokines, such as TNF-α, IFN-γ, IL-1β, IL-12 and chemokines ( Dunst et al., 2017 ). Adhesion of iRBCs and the release of inflammatory cytokines can activate brain ECs, triggering ECs to produce chemokines and inflammatory cytokines and upregulate CAM expression. Activation of CD8 + T cells results in the expression of chemokine receptors, including CXCR3 and CCR5. Subsequently, chemokine receptors bind to chemokine ligands expressed by ECs to induce CD8 + T-cell migration and infiltration into brain ECs. CD11a (LFA-1) on CD8 + T cells promotes adhesion to endothelial ICAM-1 ( Howland et al., 2015 ; Dunst et al., 2017 ), and upregulated expression of CAMs induces increased recruitment of iRBCs, WBCs, and platelets in brain capillaries, which enhances cerebral microvascular sequestration ( McEver, 2001 ; Shikani et al., 2012 ). The rupture of iRBCs releases merozoites, which are endocytosed by ECs and then cross-presented on major histocompatibility complex class 1 (MHC-1) molecules. MHC-1 binds to antigen receptors (TCRs) on effector CD8 + T cells to activate CD8 + T cells ( Howland et al., 2013 ). Activated CD8 + T cells release perforin, granzyme-B, and chemokines, triggering NK cells and macrophages to migrate toward the brain. Immune cell accumulation and perforin release induce apoptotic signaling in ECs and alter the tight junctions of ECs, resulting in EC dysfunction and increased cerebral vascular permeability ( Yañez et al., 1996 ; Belnoue et al., 2002 ; Haque et al., 2011 ). Disruption of BBB integrity frequently results in perivascular space enlargement, edema formation, and increased intracranial pressure, eventually resulting in death.
Activation of microvascular ECs is a central component of brain microvascular pathology, resulting from the sequestration of iRBCs on the surface of vascular ECs and systematic release of inflammatory cytokines ( Siddiqui et al., 2020 ). Activated ECs are well characterized by aggravation of brain microvascular sequestration, breakdown of tight junctions, and initiation of coagulation cascading reactions.
EPCR, a host receptor involved in anticoagulation and endothelial protection, has been identified as a receptor of PfEMP1 ( Turner et al., 2013 ). It is speculated that EPCR mediates iRBCs sequestration and participates in thrombin-induced disruption of the BBB. EPCR plays a crucial role in stabilizing ECs by activating activated protein C, an inhibitor of thrombin production that prevents EC activation ( Mohan Rao et al., 2014 ). In CM, some variants of the Plasmodium adhesins PfEMP-1 (called DC8 and DC13) preferentially bind to EPCR. Upon binding to EPCR, iRBCs reduce the level of available EPCR binding sites and block the activation of activated protein C by EPCR ( Shabani et al., 2017 ). Induction of the coagulation pathway by reducing the synthesis of EPCR and activated protein C leads to increased thrombin production and EC activation, as well as decreased protective effects of ECs.
Platelets are considered effector cells of the hemostasis system and contribute to CM. It is actively involved in sequestration, inflammation, and coagulation dysfunction and is identified as their joint point ( Cox and McConkey, 2010 ). Platelets bind to iRBCs (agglutination) and ECs via adhesion receptors (CD36, ICAM-1, P-selectin). In addition, platelets promote immune activation by binding Toll-like receptors to parasite-derived molecules, expressing chemokine receptors, and releasing cytokines, chemokines, and other immunomodulatory molecules. All these activated cells (ECs, platelets, monocytes) release microparticles (TNF-α, IFN-γ) ( Combes et al., 2004 ). Taken together, microparticles alter EC functions and are regarded as proinflammatory factors and cellular activation markers.
The BBB is a semipermeable membrane that separates the peripheral blood from the cerebral parenchyma and maintains balance by protecting the brain from potentially harmful blood pathogens and chemicals. The BBB consists of the microvascular endothelium, pericytes, microglia, astrocyte end-feet, neurons, and basement membrane. Microvascular ECs have tight junctions that impede the passive paracellular diffusion of small and large molecules ( Abbott et al., 2010 ; Moura et al., 2017 ).
Binding of PfEMP1 to receptors on ECs, including ICAM-1, VCAM-1, and EPCR, may trigger multiple signaling pathways in ECs, leading to reorganization of the tight junction complex and ultimately resulting in BBB leakage. ICAM-1 induces endothelial cytoskeletal remodeling via Rho-dependent phosphorylation of cytoskeleton-associated proteins, including FAK, paxillin, p130Cas, and cortactin, thereby promoting BBB opening ( Wittchen, 2009 ). In addition, VCAM-1 cross-linking results in the activation of Rac1 signaling, which induces the attenuation of tight junctions through Rho-dependent induction of stress fibers. Binding of PfEMP1 to EPCR fosters activation of tissue factors Va and VIIIa, thereby disrupting the anticoagulant pathway. Activation of these tissue factors results in thrombin generation, leading to fibrin deposition. In addition, PfEMP1 binding to EPCR activates the Rho A and NF-κB pathways through thrombin-mediated cleavage of PAR1, which induces a proinflammatory response, leading to BBB disruption ( Bernabeu and Smith, 2017 ; Kessler et al., 2017 ). Microglia also disrupt the BBB by producing TNF and IL-1β. Adhesion of iRBCs, leukocytes, and platelets to ECs also causes EC damage and irreversible changes ( Nishanth and Schlüter, 2019 ). iRBCs stimulate leukocytes (monocytes, NK cells) to release inflammatory cytokines (TNF-α, IL-1α, IL-1β) by releasing parasitic toxins ( Medana and Turner, 2006 ; Nishanth and Schlüter, 2019 ). TNF-α upregulates miR-155 expression in ECs, leading to dysfunction of BBB integrity by altering tight junctions ( Barker et al., 2017 ). IL-1α and IL-1β activate ECs to release the chemokines CCL2, CCL4, CXCL8, and CXCL10, which promote leukocyte accumulation ( Dunst et al., 2017 ), and infiltrated leukocytes induce EC apoptosis through granzyme-B and perforin-mediated cytotoxicity ( Rénia et al., 2012 ). CD8 + T cells directly induce neuronal cell death through their cytotoxic function and activation of neurons. Due to increased BBB permeability, cytokines, chemokines, immune cells, and plasma factors enter the brain parenchyma and activate neurons and astrocytes, resulting in nerve injury and neurological sequelae ( Schiess et al., 2020 ). Kynurenic acid produced by macrophages and ECs during tryptophan metabolism is further converted into cytotoxic quinoline ( Bosco et al., 2003 ; Medana et al., 2003 ; Guillemin, 2012 ). All othese molecules induce disruption of the BBB ( Figure 2 ).
Molecular mechanisms of blood–brain barrier dysfunction. The binding of P. falciparum erythrocyte membrane protein-1 (PfEMP-1) to the receptors on the ECs, including ICAM-1, VCAM-1, and EPCR, may trigger multiple signaling pathways in ECs, leading to the change to cytoskeleton-associated proteins, ultimately resulting in the disruption of the BBB. Meanwhile, signaling pathways triggered by PfEMP1 lead to activation and injury of astrocytes, microglia, neurons, and perivascular macrophages and initiate the process of neuropathological injury. The binding of PfEMP1 to EPCR fosters the activation of tissue factors Va and VIIIa, thereby disrupting the anticoagulant pathway. Activation of these tissue factors results in thrombin generation, leading to fibrin deposition. Microglia also disrupt the BBB by producing TNF and IL-1β. Astrocytes retract their end feet from ECs, resulting in reduced vascular wrapping. Angiopoietin-2 produced by ECs also leads to reduced vascular wrapping by inducing pericyte dysfunction. The iRBCs stimulate leukocytes to release inflammatory cytokines (TNF-α, IL-1α, IL-1β) by releasing parasitic toxins. These cytokines disrupt BBB integrity by altering tight junctions and activating ECs to release chemokines (CCL2, CCL4, CXCL4, CXCL8, and CXCL10), which promote leukocyte accumulation, including CD4 + T cells and CD8 + T cells. Infiltrated leukocytes induce EC apoptosis through granzyme B and perforin-mediated cytotoxicity. Granzyme B and perforin directly induce neuronal cell death. Adhesion of iRBCs, leukocytes, and platelets to ECs also causes EC damage and irreversible changes. Due to the increased permeability of the BBB, cytokines, chemokines, immune cells, and plasma factors flood into the brain parenchyma and activate neurons and astrocytes, resulting in nerve injury and neurological sequelae. Kynurenic acid produced by macrophages and ECs during tryptophan metabolism is further converted into cytotoxic quinoline, which plays a vital role in stromal cells and microglia. These molecules induce the disruption of the BBB.
Recently, multiomics platforms, including genomics, transcriptomics, proteomics and metabolomics, have been widely used to unravel the underlying pathogenesis of cancer and design therapeutic strategies ( Nam et al., 2021 ). To date, there has been no combined use of multiomics approaches for CM studies, which has inspired the joint analysis of individual omics data. Analysis of DNA markers, RNA transcripts, proteins, and metabolites generated during the progression of CM contributes to understanding CM pathogenesis, which facilitates the precise diagnosis of CM and the discovery of novel therapeutic targets.
Diagnosis is central to malaria control, and early diagnosis is one of the crucial factors affecting the prognosis of CM. Unfortunately, there is no gold standard for the diagnosis of CM because of its complex and nonspecific clinical manifestations. Currently, the primary clinical symptoms that are available for CM diagnosis include (1) nonarousal coma (no local responses to pain) that persists for more than six hours after experiencing a generalized convulsion; (2) presence of asexual forms of P. falciparum on both thick and thin blood smears; and (3) exclusion of other causes of encephalopathy. To improve the accuracy of CM diagnosis, state-of-the-art cerebral imaging tools are available to assist the diagnosis of CM ( Table 2 ).
Advantages and disadvantages of different approaches for the diagnosis of cerebral malaria.
Diagnostic approaches | Advantages and disadvantages | ||
---|---|---|---|
Imaging approaches | Malaria retinopathy | Fundoscopy | Advantage: relatively low cost and simple, accurate distinction between malarial and nonmalarial comas ( ; ). Disadvantage: requiring trained ophthalmologists and expensive equipment, subject to environmental conditions ( ). |
Optical coherence tomography (OCT) | Advantage: requiring qualitative and quantitative evaluation, noninvasive nature, and high-resolution output ( ). Disadvantage: High cost as well as practical issues ( ). | ||
Teleophthalmology | Inexpensive, portable, require little additional training, and suitable for bedside patients in a variety of settings ( ). | ||
Fluorescein fundus angiography (FFA) | Advantage: Reflect the integrity of retinal blood perfusion and blood–retinal barrier by intraretinal fluorescein, and high-resolution digital imaging ( ). Disadvantage: Large size, bulky and inconvenient to use ( ). | ||
Electroencephalography (EEG) and Micro-EEG | EEG | Advantage: Useful, noninvasive, and relatively inexpensive diagnostic tests make it possible to detect delayed cerebral malaria sequelae ( ). EEG abnormalities in cerebral malaria patients are manifested by diffuse slowing, atypical sleep elements (fusiform and parietal waves), and epileptiform activity ( ). Disadvantage: Require continuous postdischarge follow-up assessment. | |
Micro-EEG | Miniature, portable, easier continuous recording after patient discharge ( ), | ||
Other | Magnetic resonance imaging (MRI) ( ), computed tomography (CT) ( ), intravital microscopy (IVM) ( ), and bioluminescent imaging devices ( ). | ||
Biomarkers | High levels of soluble ICAM-1 ( ), decreased Ang-1 and increased Ang-2 and Ang-2/Ang-1 ( ; ), the elevation of specific smooth muscle proteins in plasma, including carbonic anhydrase III (CA3), creatine kinase (CK), creatine kinase muscle (CKM), and myoglobin (MB) ( ), enhanced plasma levels of CXCL10 and CXCL4 ( ). Hsa-miR-3158-3p represents a promising biomarker candidate for CM prognosis ( ) and the relative expression levels of miR-19a-3p, miR-19b-3p, miR-146a, miR-193b, miR-467a, miR-27a, and miR-146a may be associated with CM ( ; ; ). |
The presence of malarial retinopathy facilitates the improvement in the specificity for the clinical diagnosis of CM and offers strong evidence for CM diagnosis in both adults and children ( MacCormick et al., 2014 ). In pediatric patients, the degree of retinal microcirculation is comparable to that of the brain, making it an easily observable surrogate marker to assess the severity of cerebral pathology during CM ( Bearden, 2012 ). It has been shown that malarial retinopathy presents 100% specificity and 95% sensitivity for the detection of CM, with autopsy as the diagnostic gold standard ( Beare et al., 2006 ).
Fundoscopy is a relatively low-cost and simple technique for the detection of retinopathy, which allows accurate differentiation between malarial and nonmalarial comas. The diagnosis of malarial retinopathy depends on the presence of peripheral retinal whitening, orange and white discoloration of retinal vessels, white-centered hemorrhages, and mild papilledema. The unique retinopathy of patchy retinal whitening and focal changes in vascular color are highly specific for CM diagnosis ( Beare et al., 2006 ; MacCormick et al., 2014 ). In addition, retinal hemorrhage is a common but less distinctive feature, while papilledema is not specific to CM and is unavailable for CM diagnosis alone.
OCT is an in vivo imaging tool that detects retinal changes and is feasible for qualitatively and quantitatively evaluating high-resolution cross-sectional retinal images, papilla of the optic nerve, and even retinal nerve fiber layer thickness ( Spaide et al., 2018 ). OCT is a noninvasive, high-resolution measure; however, this technique fails to diagnose malarial retinopathy.
The introduction of fundoscopy improves the accuracy of CM diagnosis; however, it requires well-trained ophthalmologists and expensive equipment, which restrains its applications in resource-limited settings ( Abu Sayeed et al., 2011 ). To overcome these problems, an innovative approach, teleophthalmology, has emerged for retinal assessment ( Salongcay and Silva, 2018 ). This technique uses a simple and inexpensive portable fundus camera to capture images by well-trained professionals, and then, the images are transferred to ophthalmologists for rapid diagnosis. Teleophthalmology requires little additional training, minimizes healthcare-seeking inconvenience and is feasible in various settings ( Maude et al., 2011 ).
With improvements in optical technology and high-resolution digital imaging, FFA has been extensively used by ophthalmologists across the world. FFA measures the integrity of retinal blood perfusion and the blood–retinal barrier by observing a map of the intraretinal fluorescein. CM patients have nonperfusion in the central retina and extensive nonperfusion in the peripheral retina ( Glover et al., 2010 ). Nevertheless, FFA requires a bulky tabletop retinal camera, whose weight and stillness make it difficult to capture clear images from conscious CM patients.
EEG pulses are recorded by measuring voltage fluctuations caused by ionic currents within the neural tissues. This noninvasive technique has made it possible to detect delayed CM sequelae ( Sahu et al., 2015 ), including neurological disorders such as status epilepticus. CM patients’ EEG abnormalities manifest as diffuse slowing, atypical sleep elements (fusiform and parietal waves), and epileptiform activity ( Postels et al., 2018 ).
Although EEG is a noninvasive and relatively inexpensive diagnostic method, a significant limitation is continuous follow-up assessment of brain activity after discharge from the hospital. To address this concern, micro-EEG, a miniature, wireless, and battery-powered portable headset, was developed, and this device achieved a comparable accuracy for the diagnosis of status epilepticus with standard EEG systems ( Grant et al., 2014 ). This new tool facilitates the recording of brain activity after discharge from the hospital and may provide an option for CM diagnosis.
In addition, other imaging tools, including magnetic resonance imaging (MRI) ( Grant et al., 2014 ; Sahu et al., 2021 ), computed tomography (CT) ( Mohanty et al., 2011 ; Sahu et al., 2021 ), intravital microscopy (IVM) ( Volz, 2013 ), and in vivo bioluminescent imaging ( Franke-Fayard et al., 2006 ), may serve as additional diagnostic approaches for CM.
In addition to imaging tools, biomarkers have been extensively used for the rapid diagnosis of CM. Soluble ICAM-1, which is strongly associated with CM, was reported to be upregulated in the brain microvasculature ( Ramos et al., 2013 ). The soluble EPCR (sEPCR) level at admission is positively correlated with CM and malaria-related mortality, and admission sEPCR was identified as an early biomarker of prognosis among CM patients ( Ramos et al., 2013 ). Angiopoietin-1 (Ang-1) and Ang-2 have been characterized as mediators of endothelial activation and integrity, and Ang-1 maintains vascular quiescence, while Ang-2 displaces Ang-1 upon endothelial activation and sensitizes cells to become responsive to subthreshold concentrations of TNF. Reduced Ang-1 and Ang-2 and increased Ang-2/Ang-1 are detected in patients with CM ( Conroy et al., 2012 ; Eisenhut, 2012 ), which is consistent with the pathophysiological changes of activation and dysfunction of ECs among CM patients. In addition, elevation of specific plasma smooth muscle proteins, including carbonic anhydrase III (CA3), creatine kinase (CK), creatine kinase muscle (CKM), and myoglobin (MB), indicates muscular damage and microvasculature lesions during CM ( Bachmann et al., 2014 ). These proteins may serve as novel biomarkers for predicting CM severity and therapeutic targets for CM.
Previous reports have demonstrated that the expression of circulating microRNAs (miRNAs) is highly sensitive to physiological and pathological stimuli ( Paul et al., 2018 ). As a consequence, their changes in response to P. falciparum infection raise the possibility of new diagnostic and potentially prognostic tools for CM. Hsa-miR-3158-3p was found to be effective for the diagnosis of severe/cerebral malaria across all age groups, and hsa-miR-3158-3p represents a promising biomarker candidate for predicting CM prognosis in all age groups ( Gupta et al., 2021 ). In addition, previous studies have shown associations of the relative expression of miR-19a-3p, miR-19b-3p, miR-146a, miR-193b, miR-467a, miR-27a, and miR-146a with CM ( Martin-Alonso et al., 2018 ; Wah et al., 2019 ; Assis et al., 2020 ).
Spatial metabolomics is an emerging omics tool that provides precise determination of species, contents, and differential spatial distributions of metabolites in animal/plant tissues ( Martin-Alonso et al., 2018 ; Geier et al., 2020 ). In the ECM, both kidney and spleen metabolism are differentially perturbed in CM compared with noncerebral malaria, and lipid metabolism and the TCA cycle are altered in the kidney and spleen ( Ghosh et al., 2012 ). Spatial metabolomics is beneficial for the diagnosis, biomarker discovery, and prognosis prediction of CM.
Early standard antimalarial treatment is crucial for CM. In 2011, parenteral artesunate was recommended as the first-line treatment for CM by the World Health Organization (WHO). Although artesunate is effective in clearing malaria parasites, administration with potent artemisinin derivatives alone is insufficient to protect against cell death, nerve damage, and cognitive impairment ( Brejt and Golightly, 2019 ), and the long-term and widespread use of artemisinins alone may lead to the emergence of drug-resistant strains. Artemisinin-based combination therapies (ACTs) are therefore introduced to improve clinical outcomes, reduce mortality, prevent long-term neurocognitive deficits and delay the emergence of artemisinin resistance.
Targeting a single signaling pathway may be insufficient to reduce mortality or improve neurological conditions among CM patients, since CM is a multiprocess disorder. Therefore, adjuvant therapy targeting multiple physiological processes of CM is needed to improve clinical outcomes, prolong survival, and reduce neurological damage in survivors ( John et al., 2010 ). Adjuvant therapy aims to decrease cytoadherence and sequestration, modulate immune responses and improve endothelial functions, with neuroprotection given as a priority, and previous studies have shown the effectiveness of adjuvant therapy in reducing mortality due to CM in ECM models ( Wei et al., 2022 ). However, the results from clinical trials are disappointing.
Clinical episodes of CM are associated with the expression of var genes encoding the specific PfEMP1 protein, while Var genes are independently observed to bind to the brain endothelium in vitro ( Avril et al., 2012 ; Claessens et al., 2012 ). Once the crucial var ligand and its endothelial receptors are identified, high-throughput screening may be used to identify small molecules that block the binding or activation of microvascular endothelium by iRBCs. Levamisole was found to interrupt CD36-dependent binding by inhibiting CD36 dephosphorylation, which is required for high-affinity binding ( Miller et al., 2013 ). It is therefore suggested that blockade of malaria parasite adhesion to the vascular endothelium may be a promising strategy for CM treatment.
Preventive measures prior to malaria may alter the immune system status and delay CM development; therefore, adjuvant therapy targeting immune regulation is difficult. Previous animal studies have identified modulators of host targets as potential adjuvant therapies, opening up new avenues for developing highly selective adjuvant therapies for CM. Targeting mammalian targets of rapamycin (mTOR) with rapamycin has been proven to be effective in suppressing immune responses ( Mejia et al., 2015 ), thus supporting the potential of rapamycin as an adjuvant treatment for CM. 6-Diazo-5-oxo-L-norleucine (DON), a glutamine analog, was found to block the glutaminase-mediated conversion of glutamine to glutamate, thereby inhibiting T-cell activation ( Crunkhorn, 2015 ), and administration of DON resulted in survival from CM and brain recovery in ECM ( Gordon et al., 2015 ). These data demonstrate that regulation of immune balance may be effective for CM treatment.
Several therapeutics have been found to target endothelial dysfunction, including a platelet-activating factor receptor antagonist ( Lacerda-Queiroz et al., 2012 ), statins such as atorvastatin ( Souraud et al., 2012 ) and lovastatin ( Reis et al., 2012 ), activated protein C ( Mohan Rao et al., 2014 ), and erythropoietin ( Kaiser et al., 2006 ). In addition, Ang protein was reported to regulate endothelial barrier integrity and is associated with CM-induced retinopathy and death ( Conroy et al., 2012 ). In response to TNF stimulation, Ang-2 causes destruction of endothelial barrier integrity and triggers endothelial adhesion molecule expression. Secretion of Ang-2 in endothelial Weibel-Palade bodies may lead to vascular leakage, inflammation, and encephaledema associated with CM. Endothelium-targeted therapy that inhibits Weibel-Palade extracellular secretion may block the pathogenic autocrine activity of Ang-2 ( Yeo et al., 2008 ).
CM is a severe neurological syndrome that may cause epilepsy, coma and death, and survivors may present with neurological and cognitive deficits. Protection of nerve cells is therefore highly essential. Among the potential neuroprotective agents, erythropoietin (EPO) is one of the most promising. In addition to stimulating erythropoiesis, EPO has neuroprotective functions and increases the stability of endothelial barriers ( Ghezzi and Brines, 2004 ; Maiese et al., 2005 ). Artesunate plus recombinant human erythropoietin (rhEPO) has been found to reduce endothelial activation and improve BBB integrity in murine ECM models, resulting in faster recovery, increased survival rates, and high neuroprotective effects ( Du et al., 2017 ). Administration of peroxisome proliferator-activated receptor-gamma (PPARγ) has been proven to improve long-term cognitive ability and prolong survival ( Serghides et al., 2014 ). In addition, PPARγ has shown neuroprotective effects via various pathways and promotes neuronal repair, making it an attractive adjuvant therapy. Dysregulation of the limk-1/cofilin-1 pathway might lead to alterations in neuronal morphology and is considered the cause of cognitive defects in patients surviving CM ( Simhadri et al., 2017 ); therefore, the LIMK-1/cofilin-1 pathway is considered a potential therapeutic target for CM. In addition, granzyme-B produced by CD8 + T cells directly kills neurons through cytotoxic function and activation of caspase-3 and calpain1 ( Kaminski et al., 2019 ). Therefore, targeting granzyme-B may be an option to prevent neuronal cell death.
Unfortunately, the clinical efficacy and safety of these adjuvant treatments have not been tested until now. Inclusion of specific PfEMP-1 receptors on the surface of iRBCs may allow its connection with T cells to yield the ability to kill iRBCs, thus inhibiting the downstream pathological reactions initiated by iRBC adhesion. Chimeric antigen receptor T (CAR-T) immune cell therapy is a breakthrough for cancer therapy ( Herzig et al., 2019 ; Bertoletti and Tan, 2020 ). Since iRBC adhesion is the initial step during the development of CM, the efficacy and safety of CAR-T immune cell therapy for CM deserve further investigation.
CM is a multifactorial and multiprocess disorder. Administration of antimalarials alone is effective in clearing malaria parasites; however, such a treatment fails to protect against nerve cell death, neurological damage and cognitive impairment. This urges the development of novel treatment for improved outcomes of CM. In addition, the rapid developments of -omics offer an opportunity for understanding the etiology of CM and provide insights into the clinical diagnosis and therapy of this potentially fatal disorder.
JL, HD, and WWe conceived and designed the study. XS, WW, WC, HZ, WWa, HD, and JL wrote the paper. All the authors read and approved the final manuscript.
This study was supported by the National Natural Science Foundation of China (Grant Number 81802046), the Principle Investigator Program of Hubei University of Medicine (Grant Number HBMUPI202101) and the Advantages Discipline Group (Public health) Project in Higher Education of Hubei Province (2022PHXKQ1).
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Malaria Journal volume 23 , Article number: 253 ( 2024 ) Cite this article
70 Accesses
1 Altmetric
Metrics details
Disordered amino acid metabolism is observed in cerebral malaria (CM). This study sought to determine whether abnormal amino acid concentrations were associated with level of consciousness in children recovering from coma. Twenty-one amino acids and coma scores were quantified longitudinally and the data were analysed for associations.
In a prospective observational study, 42 children with CM were enrolled. Amino acid levels were measured at entry and at frequent intervals thereafter and consciousness was assessed by Blantyre Coma Scores (BCS). Thirty-six healthy children served as controls for in-country normal amino acid ranges. Logistic regression was employed using a generalized linear mixed-effects model to assess associations between out-of-range amino acid levels and BCS.
At entry 16/21 amino acid levels were out-of-range. Longitudinal analysis revealed 10/21 out-of-range amino acids were significantly associated with BCS. Elevated phenylalanine levels showed the highest association with low BCS. This finding held when out-of-normal-range data were analysed at each sampling time.
Longitudinal data is provided for associations between abnormal amino acid levels and recovery from CM. Of 10 amino acids significantly associated with BCS, elevated phenylalanine may be a surrogate for impaired clearance of ether lipid mediators of inflammation and may contribute to CM pathogenesis.
Of the complications associated with Plasmodium falciparum infection, unarousable coma, the hallmark of cerebral malaria (CM), may lead to long-term disability and death [ 1 , 2 ]. Recent research aided by autopsy and imaging indicates that coma may be the result of generalized brain swelling due to accumulation of sequestered infected erythrocytes in the microvasculature [ 3 , 4 , 5 , 6 ]. Sequestration is the physical adherence of infected cells to activated endothelial cells within brain capillaries and venules. Microscopic, sequestered foci with patchy distribution restrict blood flow with downstream consequences including anoxia and acidosis. Leakage of intravascular fluid and, in some cases, erythrocytes into adjacent brain parenchyma through endothelial junctions adds to pathogenic events [ 7 , 8 ].
In addition to the anatomical pathology in the brain, there are systemic abnormalities involving carbohydrate [ 9 , 10 ] lipid [ 11 ] and amino acid [ 12 , 13 ] metabolism. None are specific for CM—indeed, they are also seen in septic syndrome and other severe inflammatory diseases [ 14 , 15 , 16 ]. Metabolic causes of coma in CM have been sought, but no convincing pathogenic mechanism has been found. Nevertheless, these changes may contribute to coma in ways not yet recognized.
Research on the role of nitric oxide in malaria pathogenesis noted abnormalities in the plasma levels of several amino acids in children presenting with CM [ 17 , 18 ]. All abnormal amino acid levels were below the normal ranges, except for one, phenylalanine, which was consistently above 80 micromolar . Healthy controls (HC) rarely exceeded this cutoff concentration. With resolution of coma in treated survivors, phenylalanine levels normalized. This study questioned whether longitudinal measurements of abnormal amino acid levels might be associated with the level of consciousness as children recovered. Such analysis may uncover an amino acid or group of amino acids involved in P. falciparum pathogenesis.
To address this question, a prospective observational study was conducted in Ghana, in which children entering hospital in coma due to malaria were observed frequently (every 12–24 h) for 60 h and their level of consciousness was quantified by Blantyre coma score(s) (BCS) [ 19 ]. At the same time intervals, blood samples were obtained and processed for 21 plasma amino acid levels. The a priori hypothesis was that the levels of one or more amino acid(s) would be most closely associated with the BCS as participants regained consciousness. A healthy control group was enrolled to determine whether amino acid levels in Ghanaian children matched reference laboratory normal ranges for healthy children in the US. The control group also provided disease versus normal comparisons at the time of presentation to hospital.
The study was a prospective, observational, longitudinal investigation at Komfo Anokye Teaching Hospital in Kumasi, Ashanti Region, Ghana (6.697479° N 1.631690° W) from 2004 to 2006. Enrolled were children who presented to the hospital with CM, as well as healthy control children who were asymptomatic hospital visitors or attendees at well-clinic check-ups. Children with uncomplicated malaria were not included, as the study was designed to address amino acid levels associated with recovery from coma. Study participants largely resided in Kumasi and its surrounding districts. The study was conducted by senior Ghanaian physicians in the Department of Pediatrics assisted by infectious diseases faculty, fellows, and medical students from the University of Utah. Plasma samples for amino acid analysis were obtained at enrollment (i.e., at admission) and at least every 24 h during hospitalization.
Children were 6 months to 6 years of age (median age, 2.8). WHO case definition for CM was used as inclusion criteria [ 20 ]: (1) any level of P. falciparum parasitaemia on peripheral blood film; (2) unarousable coma as assessed by BCS ≤ 2 not attributable to hypoglycaemia (i.e., blood glucose level < 40 mg/dl); (3) coma persisting more than 60 min after any convulsion; and (4) no other identifiable cause of coma. Exclusion criteria were any of the following: (1) microscopic or culture evidence of bacterial or viral co-infection; (2) oral or intravenous quinine or oral artemisinin-based combination therapy initiated > 18 h prior to enrollment; (3) haemoglobin < 5 mg/dl when blood transfusion was unavailable at the study site.
Similar aged healthy children were prospectively enrolled as HC. Weight was not included as criterion for enrollment and therefore there is significant missing data for this variable (Table 1 ). Eligible HC had no symptoms or signs of active illness, no febrile illness within the past 2 weeks, no history or evidence of an active inflammatory condition, and negative blood film for malaria parasites. Availability of HCs for enrollment at the study site was limited and resulted in fewer than anticipated children in this group.
Demographic information, clinical history, and physical examination were documented using standardized case report forms. History of last food or liquid intake for CM participants was recorded to assess potential confounding influence of protein intake on plasma amino acid concentrations. The severity of CM resulting in anorexia ruled out this possible confounder for the majority of participants. Capillary blood samples were obtained for malaria thick and thin blood films and were prepared by Giemsa staining. Venous samples for routine laboratory analysis included complete blood count, electrolytes, creatinine, and lactate. Because laboratory service was not always available, there are significant missing data (Table 2 ) for these analytes. Urine was obtained for dipstick analysis and culture. Blood and urine laboratory results were immediately available to clinicians. Blood cultures were obtained on all participants with CM. Lumbar puncture was done to investigate possible bacterial meningitis unless it was clear to the presiding clinicians that this diagnosis was unlikely. Thirty-four CM cases received lumbar puncture. Cerebrospinal fluid analysis included: (a) determination of glucose and protein concentrations; (b) cell count with differential performed by personnel trained in microscopy; (c) Gram stain and bacterial and fungal cultures.
Children with CM received anti-malarial therapy and supportive care as per standard Ghanaian Ministry of Health protocols for the years during which the study was conducted (intravenous quinine or intravenous artesunate as recommended by WHO protocols). Treatment was initiated as soon as the diagnosis of malaria was suspected.
BCS was assessed at presentation and at least every 24 h until hospital discharge or death. For some participants (year 1 of the study) BCS was measured at admission and at 12-h intervals for up to 60 h. For years 2 and 3, midnight blood draws became impractical due to transportation and safety concerns, and samples were taken every 24 h. For each child at each interval, three clinicians independently assessed BCS. The assessments were done by a Ghanaian faculty physician, a US infectious diseases faculty or fellow, and a Ghanaian or US medical student. After assessment, the three clinicians met to obtain a consensus for BCS. For some patients this required returning to the bedside to review the findings. In all cases, consensus was reached by the three examiners and one BCS was recorded. At the time of BCS assessments the clinician examiners were unaware of any data pertaining to plasma amino acid concentrations.
Blood samples were collected into heparin tubes, mixed, and immediately centrifuged to sediment blood cells. Supernatant plasma was placed into polypropylene freezer tubes and stored at − 80 °C until shipment. Samples were transported in a liquid nitrogen dry shipper to the US for amino acid analysis. All amino acid analyses for the 3-year study were performed within 1 month after collection for each year. Amino acid analyses were performed at the Biochemical Genetics Section, ARUP Laboratories, University of Utah School of Medicine in collaboration with Dr. Marzia Pasquali. The amino acid analyzer employed ion exchange chromatography for separation and quantification. With two exceptions, all plasma amino acids were quantified. Exceptions were tryptophan, which emerged from the column lastly at a long retention time unsuitable for analysis, and hydroxyproline, which gave frequent values of zero or below the level of sensitivity of the assay. Computer output results from the amino acid analyzer were electronically transferred to Excel spread sheets for subsequent data analysis.
Data was compared for CM cases and healthy controls at entry (Tables 1 , 2 , 3 ). Data was analysed using GraphPad Prism version 10.1.1. Data sets for each variable were tested for normality. Parametric data (Tables 1 , 2 ) for both CM and healthy control groups were compared for significant difference using two-tailed Student’s t-test. For non-parametric data (Table 3 ) the two-tailed Mann–Whitney test was used. A significant difference was defined as P ≤ 0.05.
Association between BCS values and amino acid levels was investigated using statistical modeling. The aim was to assess if an out-of-range level of each amino acid was linked to a low BCS. Since BCS is based on an ordinal discrete scale from 0 to 5, an ordinal logistic regression, which is the most appropriate modeling technique for ordinal variables, was employed [ 21 , 22 ]. The model was formulated using a generalized linear mixed-effects model (GLM-EM) [ 23 ]. It incorporated children as a random effect to account for variations in the sampling period. The ordinal logistic regression model describes the association between the dependent variable and independent variables by estimating odds ratios (OR). Odds ratio was used to identify the association between out-of-range amino acid levels and BCS. An OR significantly below 1.0 indicated that having an amino acid level outside the normal range was associated with low BCS. An OR above 1.0 indicated that having an amino acid level within the normal range was associated with a high BCS. The model equation was the following:
where BCS was children’s BCS expressed as an ordinal variable from 0 to 5, Year was the sampling year, and Child ID random was children’s ID included as a random effect.
A logistic regression GLM-EM was also built to investigate the effect of time on the levels of amino acids [ 24 ]. This model aimed to identify the time threshold after which the levels of amino acids returned to the normal range. An OR significantly above 1.0 indicated a high probability that the amino acid level was normal at a given time, while an OR significantly below 1.0 indicated a high probability that the amino acid level was abnormal at a given time. The equation for the GLM-EM was the following:
where Amino acid level (0,1) was a dichotomous variable reporting if the amino acid level was normal (value = 1) or out of normal range (value = 0), Sampling time was the time intervals of sampling, Year was the sampling year, and Child ID random was children’s ID included as a random effect.
All statistical modelling was performed using BayesX software through R language interface [ 25 , 26 ]. All P values were adjusted using the Bonferroni correction [ 27 ]. A significant association was defined as P ≤ 0.05.
The study was approved by the ethics committee at Komfo Anokye Teaching Hospital and the Institutional Review Board at the University of Utah, USA. Written informed consent was obtained from either parent or guardian of all participants. Consent forms were presented in Twi or in English, depending on the consenting parent or guardian preference. US Department of Health and Human Services guidelines for human subjects research, the University of Utah guidelines, and the guidelines for the Komfo Anokye Teaching Hospital were followed.
Forty-two children were enrolled with CM due to P. falciparum; of CM cases, 6 were enrolled in 2004, 19 in 2005, and 17 in 2006; note: fewer children were enrolled in year 1 due to lack of dedicated study personnel and difficulties for setting up the Study Protocol requirements. Thirty-six healthy controls were enrolled during these same years. The enrollment periods for each year were the same, i.e., during the long rainy season. Two CM participants were excluded from the analyses due to missing data; six children with CM died (14.3%) during hospitalization, all within the first 48 h of admission.
Clinical characteristics for the two groups are listed in Table 1 . The groups were closely matched for age, gender, and weight. Physical findings revealed the degree of illness in the CM group, including significant differences in pulse, respiratory rate, and temperature. Of those CM cases for whom clinical data was available, 68% experienced a witnessed seizure and 59% received diagnostic lumbar puncture. Historical data on most recent food intake indicated that plasma amino acid levels in CM participants were unlikely to be confounded by recent protein ingestion. The length of pre-admission symptoms (mean, 4 days) was consistent with the acute course of CM leading to coma prompting presentation at hospital.
Laboratory findings in CM cases were consistent with CM, including anemia, thrombocytopenia, and metabolic acidosis (Table 2 ). Elevated creatinine was likely due to acute kidney injury [ 28 ]. Hypoglycaemia in CM participants was obscured by immediate intravenous glucose-containing fluid begun on all children presenting in coma.
Plasma samples collected from 40 CM enrollees at various time-points during hospitalization were available for amino acid analysis with one exception: a single sample at time zero. Of the 36 healthy control enrollees there were 6 samples unavailable for analysis because of insufficient venipuncture blood for plasma separation or because samples went missing.
For healthy controls, normal distribution was usually the case. However, all amino acid data for the CM group showed non-parametric distributions. For 5 of the 21 amino acids (Asp, Cys, Leu, Tyr, Val), there were no significant differences between CM versus healthy controls at entry (Table 3 ). In all cases but one, the levels of the remaining amino acids (Ala, Arg, Cit, Glu, Gln, Gly, His, Ilu, Lys, Met, Orn, Pro, Ser, Tau, Thr) were significantly lower in the CM group compared to healthy controls. One amino acid (Phe) was significantly elevated in CM participants compared to healthy controls. Also shown in Table 3 are the normal ranges (mean ± 2SD) for each amino acid, which were established at the ARUP diagnostic laboratory based on a large age-dependent database of healthy US children. For Ghanaian healthy control children, amino acid levels were largely within the US normal ranges.
Results obtained from the ordinal GLM-EM showed that BCS was significantly associated with amino acid levels outside the normal range for 10 of the 21 amino acids (Table 4 ). Nine of the 10 amino acids were associated with significant probability for having a low BCS; the lower the odds ratio the greater the probability of having a low BCS. A phenylalanine level outside the normal range was associated with the highest probability of having a low BCS (lowest odds ratio). Conversely, valine out-of-range levels were significantly associated with a high probability of having a high BCS (highest odds ratio). Odds ratios calculated using statistical modeling (GLM-EM) do not specify whether an amino acid association significance is above or below its normal range. In this regard it is notable that only one of the ten amino acids (phenylalanine) with significant association for a low BCS is above its normal range. All other significant amino acid associations are below their normal ranges.
For completeness box plots showing raw data for out-of-range versus normal range individual samples for each amino acid at a given BCS are shown in Supplementary data (Supplementary Fig. 1, panels A–D).
For clarity, the time course was plotted for plasma levels (mean ± SEM) of each of the ten amino acids (Table 4 ) significantly associated with a low BCS (Fig. 1 ). In each panel the right axis reproduces the mean ± SEM BCS. A dashed horizontal line marks the normal range lower limit for nine of the ten amino acids. For these nine all abnormal levels were below the normal range. The one exception is phenylalanine where the dashed line shows the upper limit of normal, commensurate with hyperphenylalaninaemia for this amino acid. Some amino acids (Arg, Cit, Gln) exhibit a time lag before returning to within their normal ranges while BCS rose to a near conscious level (BCS ~ 3–4) by 24 h. Others (Gly, Ilu, Lys, Ser, Thr) showed borderline low levels, but rose to within their normal ranges as time progressed. Exceptionally, high phenylalanine levels promptly fell in concert with rising BCS.
Out-of-range amino acid levels significantly associated with Blantyre Coma Score at sampling time intervals. Plasma levels of each amino acid (circles, mean ± SEM) significantly associated with BCS (Table 4 ) are shown at sampling times. Blantyre Coma Score (triangles, mean ± SEM) at the same sampling times are reproduced for each amino acid panel
A logistic regression model was used to determine the time after which there was a high probability that amino acid levels reached the normal range (Fig. 2 ). Bar graphs for each of the ten amino acids show odds ratios (± 95% CI) at each sampling time. Odds ratios below 1.0 with Asterix denote significant association with levels outside normal ranges. Odds ratios above 1.0 with Asterix denote significant association with levels within normal ranges. Bars without Asterix are without significance. Almost all amino acids normalized by 48–60 h post admission and initiation of treatment. However, the analyses showed marked variability across amino acids: for example, phenylalanine and isoleucine reached their normal range at 36 h, while for other amino acids it took 48 and 60 h to reach the normal range. An outlier was valine, with significant normal range values at zero and 12 h and abnormal range values at 36, 48 and 60 h. The changes were associated with increasing BCS as time elapsed. For reference, Fig. 2 (lower right-most panel) shows BCS data at each sampling time.
Effect of sampling time on having an amino acid level outside of, or within, the normal range. Bar graphs for each amino acid show odds ratios (± 95% CI) at each sampling time. Odds ratios below 1.0 with Asterix denote significant association with levels outside normal ranges. Odds ratios above 1.0 with Asterix denote significant association with levels within normal ranges. Bars without Asterix are without significance. Box plot in the lower right hand panel shows median BCS with variance measures at sampling intervals. Boxes represent interquartile ranges (IQR). Dark lines within boxes are medians. Whiskers indicate minimum and maximum values within 1.5 times the IQR
Abnormal amino acid levels based on age-dependent normal ranges were defined for 21 amino acids as established for healthy US children in the reference laboratory. To determine whether these normal ranges applied to healthy Ghanaian children, an age-matched healthy control group was enrolled and their amino acid plasma levels were measured. The healthy control levels were within normal ranges save for slight median decreases of cystine and glutamine (Table 3 ).
Amino acid levels were compared in CM versus healthy control cohorts at hospital entry. Sixteen of the 21 amino acids were significantly different. The results mirror those reported by others for children [ 13 ] and adults [ 12 ] with severe malaria, including CM. Fifteen of the 16 were below the normal range and one (phenylalanine) was elevated. The hypercatabolic state induced by the inflammatory response in severe malaria, in which amino acids are oxidatively degraded to yield chemical energy likely contributes to the low levels observed [ 29 , 30 ] Additionally, gluconeogenesis in liver consumes glucogenic amino acids. Acute kidney injury associated with amino acid reabsorption abnormalities poses yet another source of loss [ 13 ]. Significantly low levels of glutamine, glutamate, proline, ornithine, citrulline and arginine comprise the pathway of de novo arginine synthesis [ 18 ] a finding consistent with low nitric oxide production [ 31 ]. Longitudinal data found low arginine levels associated with nitric oxide-dependent endothelial dysfunction [ 32 ].
These results extended amino acid abnormalities by longitudinal measurements with analysis for association with BCS. By employing the GLM-EM model the data showed that ten of the sixteen out-of-range amino acids at entry were significantly associated with BCS (P ≤ 0.05, Table 4 ). Six of the ten (arginine, glycine, isoleucine, phenylalanine, threonine, and valine) showed high significance for association with BCS (P ≤ 0.01). Of this set, phenylalanine and valine stood out. Phenylalanine was the only amino acid with levels above the normal range in CM participants compared to healthy controls. Of all significant BCS-associated amino acids, phenylalanine showed the highest association (lowest odds ratio, 0.25). Valine out-of-range values were associated with high BCS.
For the ten amino acids significantly associated with BCS, the out-of-range values at each sampling time were analysed by calculating odds ratios with the GLM-EM model. With time, 9 of 10 amino acids normalized by 60 h post admission and initiation of treatment (odds ratios > 1.0). At 24 to 36 h post admission and initiation of treatment, levels transitioned to normal ranges for isoleucine, lysine, phenylalanine, and threonine. At these same time-points median BCS rose from about 1 to 4–5. Thus, regaining consciousness occurred at the time when these four amino acids normalized. The outlier was valine with significant normal range values at entry and at 12 h with transition to abnormally low levels at 36, 48, and 60 h post admission and initiation of treatment.
Of the ten amino acids associated with BCS, a unique association between BCS and valine was found. Out-of-range valine levels bore significant probability to have high, rather than low, BCS. This finding could relate to the complex catabolism of the branched-chain amino acids (Leu, Ilu, Val). In experimental animal models and possibly in humans, high leucine levels antagonize valine degradation at the decarboxylation (keto acid dehydrogenase) step by competitive inhibition [ 33 , 34 ]. This might delay valine catabolic kinetics resulting in out-of-range levels at later times during recovery, when BCS had risen to 4 or 5.
Like other amino acids, the phenylalanine degradation supplies carbon to the TCA cycle for oxidation [ 35 ]. Why then did investigators find consistently elevated phenylalanine levels in CM? In a previous study, a possible mechanism to explain hyperphenylalaninaemia was found [ 36 ]. The liver enzyme, phenylalanine hydroxylase (PHA), requires the cofactor tetrahydrobiopterin (BH 4 ) for mono-oxygenation of phenylalanine to yield its product, tyrosine [ 37 ]. In health biopterin is poised in the reduced state (BH 4 ) such that with each phenylalanine to tyrosine reaction, the oxidized cofactors, biopterin (B) and dihydrobiopterin (BH 2 ), are reduced back to BH 4 via two pathways (recycling and salvage) [ 37 ]. The reducing equivalents for restoring biopterin to its reduced state are supplied by the nicotinamide adenine dinucleotides NADH and NADPH [ 37 ]. In the systemic compartment (i.e., plasma and liver) about 2/3 of biopterin is poised in the reduced state (BH 4 ) [ 38 ]. This provides sufficient reducing power for tight regulation of plasma phenylalanine below 80 micromolar [ 39 ]. Phenylalanine regulation prevents hyperphenylalaninaemia, which is toxic to the brain (e.g., in the congenital disease, phenylketonuria [ 39 ]). While total biopterins (B + BH2 + BH 4 ) were slightly increased in CM, the percentage of reduced biopterin fell by 75% [ 36 ]. This suggests that hyperphenylalaninaemia in CM results from a diminution of PAH catalysis in the liver due to insufficient BH 4 availability, possibly a result of oxidative stress.
BH 4 is a unique cofactor in that there are only 6 enzymes for which it is required [ 37 , 40 ]. These are the enzymes for (a) synthesis of catecholamines and serotonin in the brain, (b) systemic and CNS synthesis of nitric oxide, (c) PAH in liver, and (d) a little studied liver enzyme for catabolism of membrane ether lipids, alkylglycerol mono-oxygenase (AGMO) [ 40 , 41 , 42 ]. Accumulation of membrane-derived ether lipids from impaired AGMO activity resulting from limited BH 4 availability might adversely affect the CNS [ 43 ]. Blood–brain barrier permeability is markedly increased in experimental animals treated with short chain alkyl (ether) glycerols [ 44 , 45 ]. Limiting AGMO activity could enhance circulating ether lipid mediators such as platelet activation factor (PAF). This possibility is yet to be investigated [ 46 , 47 ]. There is evidence from malaria models that PAF could enhance sequestration of infected red cells if regulation control mechanisms fail [ 48 , 49 ]. The possible role for an ether lipid(s) in malaria pathogenesis should be explored .
Because a group of children with severe malaria without coma was not included, one cannot infer that amino acid abnormalities described here are specific for the depth of coma itself, but rather could reflect broader associations with severe disease states. The cohort of 40 CM cases is relatively small. BCS are subjective, therefore open to observer bias. The linear mixed effects model can only assign an association between out-of-range amino acid levels and BCS, and thus cannot establish cause and effect. Diagnosis of CM carries a significant false positive fraction [ 50 ]. However, 59% of the CM cohort received lumbar puncture to rule out meningitis, thus diminishing the likelihood of non-malarial causes of coma.
Longitudinal data on normalization of amino acid levels with recovery from malarial coma found a highly significant association for phenylalanine. Unlike other coma-associated amino acids, abnormal phenylalanine levels were above, not below, normal range. Hyperphenylalaninaemia likely results from insufficient reducing power normally provided by tetrahydrobiopterin. A previous study found a marked diminution of tetrahydrobiopterin. Therefore, the single other liver tetrahydrobiopterin-dependent enzyme, responsible for degrading biologically active ether lipids (AGMO), may be impaired. Elevated bioactive ether lipids are known to permeabilize the CNS blood–brain barrier and lead to platelet activation, events shown to be critical in CM pathogenesis. If impaired, restoration of AGMO activity could provide efficacious adjunctive therapy.
No datasets were generated or analysed during the current study.
Marsh K, Forster D, Waruiru C, Mwangi I, Winstanley M, Marsh V, et al. Indicators of life-threatening malaria in African children. N Engl J Med. 1995;332:1399–404.
Article CAS PubMed Google Scholar
WHO. World Malaria Report 2021. Geneva: World Health Organization; 2021.
Google Scholar
Seydel KB, Kampondeni SD, Valim C, Potchen MJ, Milner DA, Muwalo FW, et al. Brain swelling and death in children with cerebral malaria. N Engl J Med. 2015;372:1126–37.
Article CAS PubMed PubMed Central Google Scholar
Potchen MJ, Kampondeni SD, Seydel KB, Haacke EM, Sinyangwe SS, Mwenechanya M, et al. 1.5 Tesla magnetic resonance imaging to investigate potential etiologies of brain swelling in pediatric cerebral malaria. Am J Trop Med Hyg. 2018;98:497–504.
Kampondeni S, Seydel KB, Zhang B, Small DS, Birbeck GL, Hammond CA, et al. Amount of brain edema correlates with neurologic recovery in pediatric cerebral malaria. Pediatr Infect Dis J. 2020;39:277–82.
Article PubMed Google Scholar
Poespoprodjo JR, Douglas NM, Ansong D, Kho S, Anstey NM. Malaria. Lancet. 2023;402:2328–45.
Darling TK, Mimche PN, Bray C, Umaru B, Brady LM, Stone C, et al. EphA2 contributes to disruption of the blood-brain barrier in cerebral malaria. PLoS Pathog. 2020;16: e1008261.
Moxon C, Alhamdi Y, Storm J, Toh J, Ko JY, Murphy G, et al. Parasite histones mediate blood-brain barrier disruption in cerebral malaria. Clin Med. 2020;20(Suppl 2):s96–7.
Article Google Scholar
White NJ, Warrell DA, Chanthavanich P, Looareesuwan S, Warrell MJ, Krishna S, et al. Severe hypoglycemia and hyperinsulinemia in falciparum malaria. N Engl J Med. 1983;309:61–6.
Planche T, Krishna S. Severe malaria: metabolic complications. Curr Mol Med. 2006;6:141–53.
Herdman MT, Sriboonvorakul N, Leopold SJ, Douthwaite S, Mohanty S, Hassan MM, et al. The role of previously unmeasured organic acids in the pathogenesis of severe malaria. Crit Care. 2015;19:317.
Article PubMed PubMed Central Google Scholar
Leopold SJ, Apinan S, Ghose A, Kingston HW, Plewes KA, Hossain A, et al. Amino acid derangements in adults with severe falciparum malaria. Sci Rep. 2019;9:6602.
Conroy AL, Tran TM, Bond C, Opoka RO, Datta D, Liechty EA, et al. Plasma amino acid concentrations in children with severe malaria are associated with mortality and worse long-term kidney and cognitive outcomes. J Infect Dis. 2022;226:2215–25.
Dinarello CA. Interleukin-1 and the pathogenesis of the acute-phase response. N Engl J Med. 1984;311:1413–8.
Lundblad RL. Biotechnology of plasma proteins. Ist. Boca Raton: CRC Press; 2012. p. 460.
Book Google Scholar
van Gassel RJJ, Baggerman MR, van de Poll MCG. Metabolic aspects of muscle wasting during critical illness. Curr Opin Clin Nutr Metab Care. 2020;23:96–101.
Lopansri BK, Anstey NM, Stoddard GJ, Mwaikambo ED, Boutlis CS, Tjitra E, et al. Elevated plasma phenylalanine in severe malaria and implications for pathophysiology of neurological complications. Infect Immun. 2006;74:3355–9.
Rubach MP, Zhang H, Florence SM, Mukemba JP, Kalingonji AR, Anstey NM, et al. Kinetic and cross-sectional studies on the genesis of hypoargininemia in severe pediatric Plasmodium falciparum malaria. Infect Immun. 2019;87:e00655-e718.
Molyneux ME, Taylor TE, Wirima JJ, Borgstein A. Clinical features and prognostic indicators in paediatric cerebral malaria: a study of 131 comatose Malawian children. Q J Med. 1989;71:441–59.
CAS PubMed Google Scholar
WHO. Severe falciparum malaria. Trans R Soc Trop Med Hyg. 2000;94:S1–90.
Armstrong BG, Sloan M. Ordinal regression models for epidemiologic data. Am J Epidemiol. 1989;129:191–204.
Norris CM, Ghali WA, Saunders LD, Brant R, Galbraith D, Faris P, et al. Ordinal regression model and the linear regression model were superior to the logistic regression models. J Clin Epidemiol. 2006;59:448–56.
Hedeker D. A mixed-effects multinomial logistic regression model. Stat Med. 2003;22:1433–46.
Ugwu CLJ, Zewotir TT. Using mixed effects logistic regression models for complex survey data on malaria rapid diagnostic test results. Malar J. 2018;17:453.
Umlauf N, Adler D, Kneib T, Lang S, Zeileis A. Structured additive regression models: an R interface to BayesX. Stat Softw. 2015;63:1–46.
Team RC. R: a language and environment for statistical computing. Vienna: R Foundation for Statistical Computing; 2021.
Cabin RJ, Mitchell RJ. To Bonferroni or not to Bonferroni: when and how are the questions. Ecol Soc Am. 2000;81:246–8.
Batte A, Berrens Z, Murphy K, Mufumba I, Sarangam ML, Hawkes MT, et al. Malaria-associated acute kidney injury in African children: prevalence, pathophysiology, impact, and management challenges. Int J Nephrol Renovasc Dis. 2021;14:235–53.
Basler T, Meier-Hellmann A, Bredle D, Reinhart K. Amino acid imbalance early in septic encephalopathy. Intensive Care Med. 2002;28:293–8.
Bröer S, Bröer A. Amino acid homeostasis and signalling in mammalian cells and organisms. Biochem J. 2017;474:1935–63.
Anstey NM, Weinberg JB, Hassanali MY, Mwaikambo ED, Manyenga D, Misukonis MA, et al. Nitric oxide in Tanzanian children with malaria: inverse relationship between malaria severity and nitric oxide production/nitric oxide synthase type 2 expression. J Exp Med. 1996;184:557–67.
Yeo TW, Lampah DA, Gitawati R, Tjitra E, Kenangalem E, McNeil YR, et al. Recovery of endothelial function in severe falciparum malaria: relationship with improvement in plasma L-arginine and blood lactate concentrations. J Infect Dis. 2008;198:602–8.
Harper AE, Miller RH, Block KP. Branched-chain amino acid metabolism. Annu Rev Nutr. 1984;4:409–54.
Staten MA, Bier DM, Matthews DE. Regulation of valine metabolism in man: a stable isotope study. Am J Clin Nutr. 1984;40:1224–34.
Castillo L, Yu YM, Marchini JS, Chapman TE, Sanchez M, Young VR, et al. Phenylalanine and tyrosine kinetics in critically ill children with sepsis. Pediatr Res. 1994;35:580–8.
Rubach MP, Mukemba J, Florence S, Lopansri BK, Hyland K, Volkheimer AD, et al. Impaired systemic tetrahydrobiopterin bioavailability and increased oxidized biopterins in pediatric falciparum malaria: association with disease severity. PLoS Pathog. 2015;11: e1004655.
Blau N, Thony B, Cotton RGH, Hyland K. Disorders of tetrahydrobiopterin and related biogenic amines. In: CR Scriver AB, WS Sly, D Valle, (eds.) The Metabolic and molecular bases of inherited disease. 8 edn. Vol II. New York: McGraw-Hill; 2001.
Hyland K. Estimation of tetrahydro, dihydro and fully oxidised pterins by high-performance liquid chromatography using sequential electrochemical and fluorometric detection. J Chromatogr. 1985;343:35–41.
Kaufman CS. Hyperphenylalaninemia: phenylalanine hydroxylase deficiency. In: CR Scriver AB, WS Sly, D Valle, (eds). The metabolic and molecular bases of inherited disease. 8 th edn VII. New York: McGraw-Hill; 2001, 1667–724.
Werner ER. Three classes of tetrahydrobiopterin-dependent enzymes. Pteridines. 2013;24:7–11.
Article CAS Google Scholar
Watschinger K, Werner ER. Alkylglycerol monooxygenase. IUBMB Life. 2013;65:366–72.
Sailer S, Keller MA, Werner ER, Watschinger K. The emerging physiological role of AGMO 10 years after its gene identification. Life. 2021;11:88.
Dorninger F, Forss-Petter S, Wimmer I, Berger J. Plasmalogens, platelet-activating factor and beyond - ether lipids in signaling and neurodegeneration. Neurobiol Dis. 2020;145: 105061.
Erdlenbruch B, Alipour M, Fricker G, Miller DS, Kugler W, Eibl H, et al. Alkylglycerol opening of the blood-brain barrier to small and large fluorescence markers in normal and C6 glioma-bearing rats and isolated rat brain capillaries. Br J Pharmacol. 2003;140:1201–10.
Erdlenbruch B, Jendrossek V, Eibl H, Lakomek M. Transient and controllable opening of the blood-brain barrier to cytostatic and antibiotic agents by alkylglycerols in rats. Exp Brain Res. 2000;135:417–22.
Stafforini DM, McIntyre TM, Zimmerman GA, Prescott SM. Platelet-activating factor, a pleiotrophic mediator of physiological and pathological processes. Crit Rev Clin Lab Sci. 2003;40:643–72.
Gupta S, Seydel K, Miranda-Roman MA, Feintuch CM, Saidi A, Kim RS, et al. Extensive alterations of blood metabolites in pediatric cerebral malaria. PLoS ONE. 2017;12: e0175686.
Lacerda-Queiroz N, Rodrigues DH, Vilela MC, Rachid MA, Soriani FM, Sousa LP, et al. Platelet-activating factor receptor is essential for the development of experimental cerebral malaria. Am J Pathol. 2012;180:246–55.
Lacerda-Queiroz N, Rachid MA, Teixeira MM, Teixeira AL. The role of platelet-activating factor receptor (PAFR) in lung pathology during experimental malaria. Int J Parasitol. 2013;43:11–5.
Taylor TE, Fu WJ, Carr RA, Whitten RO, Mueller JS, Fosiko NG, et al. Differentiating the pathologies of cerebral malaria by postmortem parasite counts. Nat Med. 2004;10:143–5.
Download references
Research reported in this publication was supported by the National Center for Advancing Translational Sciences of the National Institutes of Health under Award Number UL1TR002538. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or their employers. John Hibbs Jr., J. Brice Weinberg, and Guy Zimmerman (deceased) provided useful comments. Tsin Yeo and Nick Anstey reviewed the manuscript and raised important suggestions for statistical analysis. Mary MacFarland aided in the preparation of the manuscript. Thanks to Will J. Fennelly for data entry and to Arthur L. Granger for arranging statistical assistance. Special appreciations go to Justice Sylverken, Clinical Assistant, and the pediatric nursing staff at Komfo Anokye Teaching Hospital for their dedicated patient care and service to the study.
This work has not been presented in any form at a scientific meeting.
NIH/NIAID: K23 AI 116869 BKL (awardee), DLG (sponsor).
Authors and affiliations.
Division of Infectious Diseases, Department of Internal Medicine, University of Utah Spencer Fox Eccles School of Medicine, 2761 E. Swasont Way, Holladay, Salt Lake City, UT, 84117, USA
Donald L. Granger & Devon C. Hale
George H. Wahlen Veterans Affairs Medical Center, Salt Lake City, UT, USA
Donald L. Granger
Department of Pediatrics, Komfo Anokye Teaching Hospital, Kumasi, Ghana
Daniel Ansong & Tsiri Agbenyega
Intermountain Health Care, Salt Lake City, UT, USA
Melinda S. Liddle & Bert K. Lopansri
Department of Psychiatry, North Shore University Hospital, Glen Oaks, NY, USA
Benjamin A. Brinton
Research Triangle Institute International, Washington, DC, USA
Richard Reithinger & Donal Bisanzio
You can also search for this author in PubMed Google Scholar
D.L.G wrote the main manuscript text. B.K.L., R.R., and D.B provided edits to the manuscript. D.L.G. and D.B. assembled the Tables and Figures. D.A., T.A., M.S.L., B.A.B., D.C.H., and B.K.L. carried out the study protocol in Kumasi, Ghana. D.B., R.R., and D.L.G. analyzed the data. D.B. and R.R. provided statistical expertise. All authors reviewed and approved the manuscript.
Correspondence to Donald L. Granger .
Competing interests.
The authors declare no competing interests.
Publisher's note.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
12936_2024_5077_moesm1_esm.zip.
Additional file1Supplementary Fig. 1. Box plots show out-of-range versus normal-range amino acid plasma levels at given BCS. Fig. 1 A–1D show results for all 21 amino acids analysed. Coloured circles represent individual amino acid levels for participants with cerebral malaria. An Asterix denotes significant association between amino acid levels within or outside the normal ranges at given Blantyre Coma Scores as determined by the generalized linear mixed-effects model
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/ .
Reprints and permissions
Cite this article.
Granger, D.L., Ansong, D., Agbenyega, T. et al. Longitudinal associations of plasma amino acid levels with recovery from malarial coma. Malar J 23 , 253 (2024). https://doi.org/10.1186/s12936-024-05077-9
Download citation
Received : 14 May 2024
Accepted : 13 August 2024
Published : 23 August 2024
DOI : https://doi.org/10.1186/s12936-024-05077-9
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative
ISSN: 1475-2875
IMAGES
COMMENTS
The clinical presentation of cerebral malaria is diffuse symmetrical encephalopathy with fever and absent or few focal neurological signs. In children, coma can rapidly develop after fever onset (mean, 2 days) . In adults, coma is typically gradual with increasing drowsiness, confusion, obtundation, and high fevers (mean duration, 5 days).
Cerebral malaria is the most severe neurological manifestation of severe malaria. With an incidence of 1,120/100,000/year in the endemic areas of Africa, children in this region bear the brunt. Peak incidence is in pre-school children and at a minimum, 575,000 children in Africa develop cerebral malaria annually( 2 ).
Cerebral malaria is a medical emergency. All patients with Plasmodium falciparum malaria with neurologic manifestations of any degree should be urgently treated as cases of cerebral malaria. Pathogenesis of cerebral malaria is due to damaged vascular endothelium by parasite sequestration, inflammatory cytokine production and vascular leakage, which result in brain hypoxia, as indicated by ...
Treatment of Severe and Cerebral Malaria. Artemisinins, which have the fastest parasite clearing time of all anti-malarial drugs, have become the drugs of the moment, and artesunate is the first-line therapy for treating severe and cerebral malaria in both children and adults ().. The main difference between the treatment of uncomplicated and severe malaria is the route of administration of ...
Cerebral malaria (CM), results from Plasmodium falciparum infection, and has a high mortality rate. CM survivors can retain life-long post CM sequelae, including seizures and neurocognitive deficits profoundly affecting their quality of life. As the Plasmodium parasite does not enter the brain, but resides inside erythrocytes and are confined to the lumen of the brain's vasculature, the ...
Cerebral malaria may be the most common non-traumatic encephalopathy in the world. The pathogenesis is heterogenous and the neurological complications are often part of a multisystem dysfunction. The clinical presentation and pathophysiology differs between adults and children. Recent studies have elucidated the molecular mechanisms of pathogenesis and raised possible interventions ...
Cerebral malaria is the most severe neurological complication of infection with Plasmodium falciparum. With >575,000 cases annually, children in sub-Saharan Africa are the most affected.
In mice with cerebral malaria, cross-presentation of Plasmodium antigens by dendritic cells results in the activation of cytotoxic CD8 + T cells 277,278,279,280.
treatment targets in cerebral malaria ... presentation Decreased intercellular contacts Increased Degranulation cytoadhesion C3a C4a C5a Cytokines Cytoadherence Haemoglobin MHC-I Blood vessel lumen
P. falciparum caused malaria is the most dreaded form of this infectious disease that is responsible for significant global mortality. One of its most common clinical manifestations is cerebral malaria, which is common in children under 5 years of age. Though curable, drug action is limited by timely diagnosis and hospitalization.
Cerebral malaria is a life-threatening complication of P. falciparum infestation that occurs in approximately 2% of patients. Pathogenesis may be explained by 2 mechanisms: vascular sequestration of parasitized erythrocytes and the potential cerebral toxicity by cytokines. Progressive clinical changes occur, along with high fever and chills.
cerebral malaria is functionally grossly intact. 35 ,36 Studies in African children with cerebral malaria do show a subtle increase in BBB permeability with a disruption of endothelial intercellular tight junctions on autopsy.37 ,38 Imaging studies reveal that most adults with cerebral malaria have no evidence of cerebral oedema.39 ,40 In African
Cerebral malaria (CM), results from Plasmodium falciparum infection, and has a high mortality rate. CM survivors can retain life‐long post CM sequelae, including seizures and neurocognitive deficits profoundly afecting their quality of life. As the Plasmodium parasite does not enter the brain, but resides inside erythrocytes and are confined ...
These variances in CM disease presentation may arise due to differences in the immature brain, including differences in host responses of the cerebral vasculature in different brain regions to sequestration and the magnitude of inflammation. ... Cerebral malaria pathology manifests itself differently in white matter and gray matter of the brain ...
Clinical presentation. Infection with malaria parasites may result in a wide variety of symptoms, ranging from absent or very mild symptoms to severe disease and even death. ... occurs in the vessels of the brain it is believed to be a factor in causing the severe disease syndrome known as cerebral malaria, which is associated with high mortality.
Introduction: Cerebral malaria (CM) is the most lethal form of severe malaria with high case fatality rates. Overtime, there is an inherent risk in changing pattern of presentation of CM which, if the diagnosis is missed due to these changing factors, may portend a poor outcome.
Cerebral malaria is a serious neurological complication of severe malaria that affects about 1% of children under the age of 5 who have been infected with Plasmodium falciparum.
Malaria is a disease caused by a parasite. The parasite is spread to humans through the bites of infected mosquitoes. People who have malaria usually feel very sick with a high fever and shaking chills. ... Cerebral malaria. If parasite-filled blood cells block small blood vessels to your brain (cerebral malaria), swelling of your brain or ...
The clinical manifestations of malaria vary with parasite species, epidemiology, immunity, and age. Issues related to clinical manifestations and diagnosis of malaria will be reviewed here. Technical aspects of laboratory tools for diagnosis of malaria are discussed further separately. The epidemiology, pathogenesis, diagnosis, and treatment of ...
Cerebral malaria is the most severe pathology caused by the malaria parasite, Plasmodium falciparum. The pathogenic mechanisms leading to cerebral malaria are still poorly defined as studies have been hampered by limited accessibility to human tissues. Nevertheless, histopathology of post-mortem human tissues and mouse models of cerebral ...
Cerebral malaria (CM) is a fatal neurological complication of P. falciparum malaria (Luzolo and Ngoyi, 2019), and children aged under 3 years and pregnant women are most susceptible ... Brain microvessel cross-presentation is a hallmark of experimental cerebral malaria. EMBO Mol. Med. 5 (7), ...
Guidelines for the treatment of malaria - 3rd edition. 1.Malaria - drug therapy. 2.Malaria - diagnosis. 3.Antimalarials - administration and dosage. ... The designations employed and the presentation of the material in this publication do not ... Cerebral malaria. Severe P. falciparum malaria with coma (Glasgow coma scale < 11,
Disordered amino acid metabolism is observed in cerebral malaria (CM). This study sought to determine whether abnormal amino acid concentrations were associated with level of consciousness in children recovering from coma. Twenty-one amino acids and coma scores were quantified longitudinally and the data were analysed for associations. In a prospective observational study, 42 children with CM ...
Even though this type of malaria is most common in children living in sub-Saharan Africa, it should be considered in anybody with impaired consciousness that has recently travelled in a malaria-endemic area. Cerebral malaria has few specific features, but there are differences in clinical presentation between African children and non-immune adults.