An Immunohistochemical Study of the Pathology of Fatal Malaria: Evidence for Widespread Endothelial Activation and a Potential Role for Intercellular Adhesion Molecule-1 in Cerebral Sequestration

The sequestration of parasitized erythrocytes in the microvasculature of vital organs is central to the pathogenesis of severe Plasmodium falciparum malaria. This process is mediated by specific interactions between parasite adherence ligands and host receptors on vascular endothelium such as intercellular adhesion molecule-1 (ICAM-1) and CD36. Using immunohistochemistry we have examined the distribution of putative sequestration receptors in different organs from fatal cases of P.falciparum malaria and noninfected controls. Receptor expression and parasite sequestration in the brain were quantified and correlated. Fatal malaria was associated with widespread induction of endothelial activation markers, with significantly higher levels of ICAM-1 and E-selectin expression on vessels in the brain. In contrast, cerebral endothelial CD36 and thrombospondin staining were sparse, with no evidence for increased expression in malaria. There was highly significant co-localization of sequestration with the expression of ICAM-1, CD36, and E-selectin in cerebral vessels but no cellular inflammatory response. These results suggest that these receptors have a role in sequestration in vivo and indicate that systemic endothelial activation is a feature of fatal malaria.     

Intramuscular loading dose of quinine for falciparum malaria: pharmacokinetics and toxicity

In a study of intramuscular injection of quinine eight adults with moderately severe falciparum malaria resistant to chloroquine were treated with quinine dihydrochloride, being given a loading dose of 20 mg salt (16.7 mg base)/kg followed by three or four eight hourly maintenance doses of 10 mg salt (8.3 mg base)/kg injected into the anterior thigh. All patients responded to treatment. Fever and parasite clearance times (mean (SD) 60 (23) h and 53 (22) h respectively) were comparable with those obtained with intravenous quinine. The mean peak plasma quinine concentration of 11.0 mg/l (34.4 mu mol/l) [corrected] was reached a median of five hours after administration of the loading dose. In all patients plasma quinine concentrations exceeded the high minimum inhibitory concentration for Plasmodium falciparum malaria prevalent in Thailand within four hours of the start of treatment but did not cause toxicity other than mild cinchonism. When intravenous infusion is not possible an intramuscular quinine loading dose is an effective means of starting treatment in patients with moderately severe falciparum malaria who cannot swallow tablets.     

Plasma and whole blood mefloquine concentrations during treatment of chloroquine-resistant falciparum malaria with the combination mefloquine-sulphadoxine-pyrimethamine.

Mefloquine-sulphadoxine-pyrimethamine (MSP) in combination has proved effective against multiple-drug-resistant falciparum malaria, but nothing is known about mefloquine absorption when it is given in this formulation. Nine Thai patients, aged 15-51 years with uncomplicated chloroquine-resistant falciparum malaria, took 11.2-16.7 mg of mefloquine base per kilogram bodyweight as MSP tablets. All patients responded to treatment with fever and parasite clearance times of 61 +/- 29 h (mean +/- s.d.) and 52 +/- 24 h, respectively. The mean apparent absorption half-time (t1/2abs) of mefloquine was 4.89 h (range 2.25-9.72) and mean peak plasma concentration was 1815 ng ml-1 (range 725-3368). Peak plasma mefloquine concentrations in three patients who vomited within 2 h of treatment were 725, 956 and 1972 ng ml-1. There was no significant difference between plasma and whole blood mefloquine concentrations during the first 48 h of treatment. Based on the elimination of parasitaemia, the plasma mefloquine concentrations are adequate for therapy of uncomplicated falciparum malaria although the relationship between plasma concentrations and therapeutic efficacy of mefloquine requires further study.     

Glucose metabolism in quinine-treated patients with uncomplicated falciparum malaria

To investigate host and drug effects on glucose metabolism in acute falciparum malaria, 10 previously untreated, fasting Thai males with uncomplicated infections were given a 2-h intravenous glucose infusion (5 mg/kg ideal body weight min) with an infusion of quinine dihydrochloride (10 mg/kg body weight) during the second hour. Eight patients were restudied in convalescence. Fasting plasma glucose (mean +/- SD) and insulin (geometric mean (-SD to + SD] were higher during acute illness (5.5 +/- 1.0 mmol/l and 6.2 (5.0-7.7) mU/l) than in convalescence (4.2 +/- 0.25 mmol/l and 3.7 (2.1-6.7) mU/l; P less than 0.001 and P = 0.058 respectively). After 1 h, both plasma glucose (9.3 +/- 1.4 vs 7.5 +/- 0.8 mmol/l, P less than 0.001) and insulin (21.2 (13.8-32.5) vs 15.2 (11.2-20.8) mU/l, P = 0.089) remained higher during acute illness; mathematical model (CIGMA) assessment of these values indicated lower tissue insulin sensitivity on admission (97% (71-134] than in convalescence (139% (109-178), P less than 0.025) but normal beta-cell function on both occasions. Two-hour plasma glucose (9.5 +/- 2.0 mmol/l) and insulin (81.8 (51.5-129.9) mU/l) concentrations during acute illness were also significantly higher than in convalescence (7.2 +/- 1.2 mmol/l and 40.1 (23.5-68.4) mU/l, P less than or equal to 0.025) despite similar end-infusion free plasma quinine concentrations (P greater than 0.5). Basal plasma free fatty acid concentrations were increased in acute illness (0.68 +/- 0.24 vs 0.21 +/- 0.12 mmol/l, P less than 0.001) but fell to low levels at 2 h in both studies. These data suggest tissue insulin resistance and augmented quinine-stimulated insulin secretion in acute falciparum malaria, factors which are likely to influence the clinical situation in which malaria-associated hypoglycaemia occurs.     

Dyserythropoiesis and ineffective erythropoiesis in Plasmodium vivax malaria

Summary. Nine Thai adults with P. vivax malaria were investigated. Light and electron microscope studies of marrow aspirates revealed morphological evidence of dyserythro-poiesis in six of them, Dyserythropoiesis was most marked in the four most anaemic patients. In these four patients the electron microscope also revealed the presence of erythro-blasts at various stages of degradation within the cytoplasm of macrophages. Neither the dyserythropoiesis nor the ineffective erythropoiesis could be attributed to a deficiency of vitamin B₁₂, folate or iron. The abnormalities of erythropoiesis seemed to result from the P. vivax infection itself. Other bone marrow reactions seen in this infection included macrophage hyperplasia, plasmacytosis and increased eosinophil granulocytopoiesis. Unlike in severe P. falciparum malaria, the microvasculature of the marrow was not obstructed by parasitized red cells.     

A study of the factors affecting the metabolic clearance of quinine in malaria

Objective: To assess the factors that contribute to impaired quinine clearance in acute falciparum malaria. Patients: Sixteen adult Thai patients with severe or moderately severe falciparum malaria were studied, and 12 were re-studied during convalescence. Methods: The clearance of quinine, dihydroquinine (an impurity comprising up to 10% of commercial quinine formulations), antipyrine (a measure of hepatic mixed-function oxidase activity), indocyanine green (ICG) (a measure of liver blood flow), and iothalamate (a measure of glomerular filtration rate) were measured simultaneously, and the relationship of these values to the␣biotransformation of quinine to the active metabolite 3-hydroxyquinine was assessed. Results: During acute malaria infection, the systemic clearance of quinine, antipyrine and ICG and the biotransformation of quinine to 3-hydroxyquinine were all reduced significantly when compared with values during convalescence. Iothalamate clearance was not affected significantly and did not correlate with the clearance of any of the other compounds. The clearance of total and free quinine correlated significantly with antipyrine clearance (r _s = 0.70, P = 0.005 and r _s = 0.67, P = 0.013, respectively), but not with ICG clearance (r _s = 0.39 and 0.43 respectively, P > 0.15). In a multiple regression model, antipyrine clearance and plasma protein binding accounted for 71% of the variance in total quinine clearance in acute malaria. The pharmacokinetic properties of dihydroquinine were generally similar to those of quinine, although dihydroquinine clearance was less affected by acute malaria. The mean ratio of quinine to 3-hydroxyquinine area under the plasma concentration-time curve (AUC) values in acute malaria was 12.03 compared with 6.92 during convalescence P=0.01. The mean plasma protein binding of 3-hydroxyquinine was 46%, which was significantly lower than that of quinine (90.5%) or dihydroquinine (90.5%). Conclusion: The reduction in quinine clearance in acute malaria results predominantly from a disease-induced dysfunction in hepatic mixed-function oxidase activity (principally CYP 3A) which impairs the conversion of quinine to its major metabolite, 3-hydroxyquinine. The metabolite contributes approximately 5% of the antimalarial activity of the parent compound in malaria, but up to 10% during convalescence. Received: 20 December 1996 / Accepted in revised form: 9 April 1997     

Coreceptor utilization of HIV type 1 subtype E viral isolates from Thai men with HIV type 1-infected and uninfected wives.

HIV-1 coreceptors CCR5 and CXCR4 play an important role in viral entry and pathogenesis. To better understand the role of viral tropism in HIV-1 transmission, we examined the coreceptor utilization of viral isolates obtained from men enrolled in a study of heterosexual transmission in northern Thailand. Viral isolates were obtained from HIV-1-positive males who had either HIV-1-infected spouses (RM; n = 5) or HIV-1-uninfected spouses (HM; n = 10). Viral isolates from 1 of the 5 RM males and 2 of the 10 HM males were CCR5 tropic, whereas isolates from 3 RM males and 6 of the HM male isolates were CXCR4 tropic. Of the nine X4-tropic isolates, seven also used at least one of the following coreceptors: CCR8, CCR1, CCR2b, or CX3CR1, and none employed CCR5 as an additional coreceptor. More importantly, three isolates, RM-15, HM-13, and HM-16 (one from a transmitter and two from nontransmitter), did not infect GHOST4.cl.34 cells expressing any of the known coreceptors. Further analysis using MAGI-plaque assays, which allow visualization of infected cells, revealed that RM-15 had low numbers of infected cells in MAGI-R5 and MAGI-X4 cultures, whereas HM-13 and HM-16 had high levels of plaques in MAGI-X4 cultures. Replication kinetics using activated lymphocytes revealed that these three isolates replicated in CCR5(+/+) as well as CCR5(-/-) peripheral blood mononuclear cells, suggesting that these isolates did not have an absolute requirement of CCR5 for viral entry. All three isolates were sensitive to the X4-antagonistic compounds T-22 and AMD3100. Analysis of the C2V3 region did not reveal any significant structural differences between any of the Thai subtype E isolates. Thus, there was no association between the pattern of coreceptor usage and transmissibility among these subtype E HIV-1 isolates.     

HIV-1 seroprevalence in an inner-city public hospital.

In a hospital-based seroprevalence survey for human immunodeficiency virus type 1 (HIV-1) infection, a stratified sampling method based on age and gender was used to collect 5429 blood samples at an inner-city hospital. Sentinel Hospital Surveillance System (SHSS) criteria developed by the Centers for Disease Control and Prevention were used to classify patient diagnoses into two categories by the likelihood of being associated with HIV-1 infection. The two categories were those with high likelihood of association with HIV-1 (SHSS-ineligible) and those with low likelihood of association with HIV-1 infection (SHSS-eligible). Of the 5429 blood samples, 4262 were SHSS-eligible and 1167 were SHSS-ineligible. After personal identifies were removed, specimens were tested by ELISA and confirmed by Western blot analysis. The overall prevalence rate of HIV-1 infection was 0.98%. The seroprevalence rate was almost 2.6 times higher in high-association patients compared with low-association patients (1.89% versus 0.73%, P < .001). Results from this study indicate a high unsuspected HIV-1 seroprevalence rate in a subpopulation (SHSS-eligible) considered to have diagnoses with low likelihood of association with HIV-1 infection. These patients may better approximate HIV-1 seroprevalence in the general population of the area served by the hospital than would a sample of all patients. Monitoring HIV-1 seroprevalence in the SHSS-eligible group will be a useful measure for community serosurveillance for HIV-1 infection.     

Risk factors for surgical-wound infection following cardiac surgery

In a prospective study of 1,009 adult patients undergoing elective cardiac surgery at The Johns Hopkins Hospital, we determined the association between a variety of preoperative and operative parameters and the risk of postoperative sternal- or mediastinal-wound infection. Of the parameters reflecting nutritional state, only one, reduced level of albumin in serum, was significantly associated with sternal- or mediastinal-wound infection by univariate analysis. The final multiple logistic regression analysis indicated that four variables were significant (P less than .05) independent predictors of sternal- or mediastinal-wound infection: obesity (relative odds = 3.8; 95% confidence limits = 1.9-7.5), diabetes mellitus (relative odds = 2.6; 95% confidence limits = 1.4-4.8), length of hospital stay before surgery greater than five days (relative odds = 2.0; 95% confidence limits = 1.2-3.5), and current cigarette smoking (relative odds = 1.8; 95% confidence limits = 1.1-3.1). Of these variables, perhaps only smoking will lend itself routinely to attempts at intervention.     

High rates of transmission of subtype E human immunodeficiency virus type 1 among heterosexual couples in Northern Thailand: role of sexually transmitted diseases and immune compromise.

The heterosexual transmission of subtype E human immunodeficiency virus type 1 (HIV-1) infection was evaluated in 467 couples in Thailand in whom the man was HIV-1 positive and the woman had no risk factors for HIV other than sex with her infected partner. At baseline, 216 (46.3%) of the 467 women were positive for HIV-1; prevalence of HIV among women was 52.2% when their male partners had CD4+ lymphocyte counts of <200 cells/microL, 45.9% in women whose partners had counts of 200-499 micro/L, and 39.2% in women whose partners had counts of >/=500/microL. Women were twice as likely to be HIV positive if their partners had a history of a sexually transmitted disease (STD); however, their HIV prevalence was 29% among couples who had no STD history. It appears that female partners of men infected with subtype E HIV-1 are at high risk of infection even when the man's CD4+ cell count is relatively high. A high rate of STDs may contribute significantly to this risk [corrected].