Pathogen Detection Using Metagenomic Next-Generation Sequencing of Plasma Samples from Patients with Sepsis in Uganda.

Metagenomic sequencing is a promising new method for pathogen detection. We aimed to detect pathogens from archived plasma using metagenomic sequencing in a previously well-characterized cohort of 254 predominantly HIV-infected patients with sepsis in Uganda. We used Illumina sequencing and the Chan Zuckerberg ID metagenomics platform to sequence and identify pathogens. On average, each plasma sample yielded 3,404,737 ± 2,201,997 reads (mean ± standard deviation), of which 220,032 ± 416,691 (6.3% ± 8.6%) were identified as nonhuman reads. Using a background model filter, 414 genus-specific pathogen identifications were found in the 254 samples. Nineteen pathogens were previously detected positive by quantitative PCR (qPCR), compared to sequencing, which demonstrated 30.2% sensitivity and 99.5% specificity. Sensitivity was higher for viral pathogens than nonviral pathogens (37% versus 5%). For example, HIV viremia was detected in 69% of samples using qPCR, and sequencing revealed 70% sensitivity and 92% specificity. There were 75 genus-specific potential pathogens identified by sequencing in this cohort, including hepatitis B and Epstein-Barr virus (EBV), among several others. qPCR showed a prevalence of hepatitis B and EBV viremia of 17% and 45%, respectively. In-hospital mortality was associated with a lower qPCR threshold cycle value for EBV (adjusted odds ratio, 0.85; P < .001) but not for hepatitis B or HIV. In conclusion, a broad range of potential pathogens were identified by metagenomic sequencing in patients with sepsis in Uganda. Unexpectedly high rates of hepatitis B and EBV viremia were found. Whether these viral infections in HIV patients with sepsis are clinically important requires further study. IMPORTANCE The use of next-generation sequencing (NGS) in blood samples is an emerging technology for clinical microbiology labs. In this work, we performed NGS on plasma samples from a well-characterized cohort, where all samples had been previously tested by PCR for 43 pathogens. Therefore, we could compare sequencing performance against that of PCR and identify clinical correlates. A broad range of potential pathogens were identified by metagenomic sequencing in patients with sepsis in Uganda, particularly viruses, which we confirmed by PCR. In addition to HIV viremia, unexpectedly high rates of hepatitis B and EBV viremia were found, which may have important clinical implications.

Biomarkers of endothelial activation/dysfunction distinguish sub-groups of Ugandan patients with sepsis and differing mortality risks.

BACKGROUND: Sepsis is a complex clinical syndrome with substantial heterogeneity. We sought to identify patterns of serum biomarkers of endothelial activation and dysfunction in individuals with sepsis and evaluate subgroup-specific differences in mortality. METHODS: Adult patients with sepsis (n=426) were consecutively recruited from two hospitals in Uganda. Clinical information was collected and serum concentrations of eleven biomarkers involved in the endothelial response to infection were measured in samples from 315 patients. Latent variable models were fit to evaluate whether the endothelial response to sepsis consists of one unified biological process or multiple processes and to identify subgroups of patients with distinct host-response profiles. Differences in survival at day 28 were evaluated using Kaplan-Meier survival curves. RESULTS: We identified three patient subgroups characterized by unique host endothelial response profiles. Patients fitting Profile 2 had significantly worse survival (log-rank p<0.001). Four latent factors (Factor 1-4) were identified, each potentially representing distinct biological processes for the endothelial response to sepsis: Factor 1 (CHI3L1, sTREM1, sFLT1); Factor 2 (ANGPT1, PF4, VEGF); Factor 3 (CXCL10, VWF, sICAM1); and Factor 4 (ANGPT2, sTEK). CONCLUSION: Patient profiles based on patterns of circulating biomarkers of endothelial responses may provide a clinically meaningful way to categorize patients into homogeneous subgroups and may identify patients with a high risk of mortality. Profile 2 may represent dysfunction of the endothelial response to infection. FUNDING: Primary funding: Investigator-Initiated Award provided by Pfizer, Inc (WMS, STJ). Additional support: Canadian Institutes of Health Research (CIHR) Foundation grant (KCK; FDN-148439) and the Canada Research Chair program (KCK).

Etiology of Sepsis in Uganda Using a Quantitative Polymerase Chain Reaction-based TaqMan Array Card.

Background: Knowledge of causes of sepsis in sub-Saharan Africa is limited. A better understanding of the microbiology of bloodstream infections could improve outcomes. Methods: We used a quantitative polymerase chain reaction (qPCR)-based TaqMan Array Card (TAC) to directly test for 43 targets from whole blood. We analyzed 336 cryopreserved specimens from adult Ugandans with sepsis enrolled in a multisite study; 84% were infected with human immunodeficiency virus. We compared qPCR TAC results with blood culture and determined the association of qPCR with study participant outcomes using logistic regression. Results: The most frequently detected targets were cytomegalovirus (CMV, n = 139, 41%), Mycobacterium tuberculosis (TB, n = 70, 21%), Plasmodium (n = 35, 10%), and Streptococcus pneumoniae (n = 31, 9%). Diagnostic performance varied by target with qPCR sensitivity averaging 61 ± 28% and specificity 98 ± 3% versus culture. In multivariable analysis, independent factors associated with in-hospital mortality included CMV viremia (adjusted odds ratio [aOR] 3.2, 95% confidence interval [CI], 1.8-5.5; p < .01) and TB qPCR-positivity, whether blood culture-positive (aOR 4.6, 95% CI, 2.1-10.0; p < .01) or blood culture-negative (aOR 2.9, 95% CI, 1.2-6.9; p = .02). Conclusions: Using qPCR TAC on direct blood specimens, CMV and TB were the most commonly identified targets and were independently associated with increased in-hospital mortality. qPCR TAC screening of blood for multiple targets may be useful to guide triage and treatment of sepsis in sub-Saharan Africa.

Derivation and validation of a universal vital assessment (UVA) score: a tool for predicting mortality in adult hospitalised patients in sub-Saharan Africa.

BACKGROUND: Critical illness is a leading cause of morbidity and mortality in sub-Saharan Africa (SSA). Identifying patients with the highest risk of death could help with resource allocation and clinical decision making. Accordingly, we derived and validated a universal vital assessment (UVA) score for use in SSA. METHODS: We pooled data from hospital-based cohort studies conducted in six countries in SSA spanning the years 2009-2015. We derived and internally validated a UVA score using decision trees and linear regression and compared its performance with the modified early warning score (MEWS) and the quick sepsis-related organ failure assessment (qSOFA) score. RESULTS: Of 5573 patients included in the analysis, 2829 (50.8%) were female, the median (IQR) age was 36 (27-49) years, 2122 (38.1%) were HIV-infected and 996 (17.3%) died in-hospital. The UVA score included points for temperature, heart and respiratory rates, systolic blood pressure, oxygen saturation, Glasgow Coma Scale score and HIV serostatus, and had an area under the receiver operating characteristic curve (AUC) of 0.77 (95% CI 0.75 to 0.79), which outperformed MEWS (AUC 0.70 (95% CI 0.67 to 0.71)) and qSOFA (AUC 0.69 (95% CI 0.67 to 0.72)). CONCLUSION: We identified predictors of in-hospital mortality irrespective of the underlying condition(s) in a large population of hospitalised patients in SSA and derived and internally validated a UVA score to assist clinicians in risk-stratifying patients for in-hospital mortality. The UVA score could help improve patient triage in resource-limited environments and serve as a standard for mortality risk in future studies.

The Malaria TaqMan Array Card Includes 87 Assays for Plasmodium falciparum Drug Resistance, Identification of Species, and Genotyping in a Single Reaction.

Antimalarial drug resistance exacerbates the global disease burden and complicates eradication efforts. To facilitate the surveillance of resistance markers in countries of malaria endemicity, we developed a suite of TaqMan assays for known resistance markers and compartmentalized them into a single array card (TaqMan array card, TAC). We included 87 assays for species identification, for the detection of Plasmodium falciparum mutations associated with chloroquine, atovaquone, pyrimethamine, sulfadoxine, and artemisinin resistance, and for neutral single nucleotide polymorphism (SNP) genotyping. Assay performance was first optimized using DNA from common laboratory parasite lines and plasmid controls. The limit of detection was 0.1 to 10 pg of DNA and yielded 100% accuracy compared to sequencing. The tool was then evaluated on 87 clinical blood samples from around the world, and the malaria TAC once again achieved 100% accuracy compared to sequencing and in addition detected the presence of mixed infections in clinical samples. With its streamlined protocol and high accuracy, this malaria TAC should be a useful tool for large-scale antimalarial resistance surveillance.

Design of a multi-arm randomized clinical trial with no control arm.

BACKGROUND: Clinical trial designs that include multiple treatments are currently limited to those that perform pairwise comparisons of each investigational treatment to a single control. However, there are settings, such as the recent Ebola outbreak, in which no treatment has been demonstrated to be effective; and therefore, no standard of care exists which would serve as an appropriate control. METHODS/DESIGN: For illustrative purposes, we focused on the care of patients presenting in austere settings with critically ill 'sepsis-like' syndromes. Our approach involves a novel algorithm for comparing mortality among arms without requiring a single fixed control. The algorithm allows poorly-performing arms to be dropped during interim analyses. Consequently, the study may be completed earlier than planned. We used simulation to determine operating characteristics for the trial and to estimate the required sample size. RESULTS: We present a potential study design targeting a minimal effect size of a 23% relative reduction in mortality between any pair of arms. Using estimated power and spurious significance rates from the simulated scenarios, we show that such a trial would require 2550 participants. Over a range of scenarios, our study has 80 to 99% power to select the optimal treatment. Using a fixed control design, if the control arm is least efficacious, 640 subjects would be enrolled into the least efficacious arm, while our algorithm would enroll between 170 and 430. This simulation method can be easily extended to other settings or other binary outcomes. CONCLUSION: Early dropping of arms is efficient and ethical when conducting clinical trials with multiple arms.

Development of a TaqMan Array Card for Acute-Febrile-Illness Outbreak Investigation and Surveillance of Emerging Pathogens, Including Ebola Virus.

Acute febrile illness (AFI) is associated with substantial morbidity and mortality worldwide, yet an etiologic agent is often not identified. Convalescent-phase serology is impractical, blood culture is slow, and many pathogens are fastidious or impossible to cultivate. We developed a real-time PCR-based TaqMan array card (TAC) that can test six to eight samples within 2.5 h from sample to results and can simultaneously detect 26 AFI-associated organisms, including 15 viruses (chikungunya, Crimean-Congo hemorrhagic fever [CCHF] virus, dengue, Ebola virus, Bundibugyo virus, Sudan virus, hantaviruses [Hantaan and Seoul], hepatitis E, Marburg, Nipah virus, o'nyong-nyong virus, Rift Valley fever virus, West Nile virus, and yellow fever virus), 8 bacteria (Bartonella spp., Brucella spp., Coxiella burnetii, Leptospira spp., Rickettsia spp., Salmonella enterica and Salmonella enterica serovar Typhi, and Yersinia pestis), and 3 protozoa (Leishmania spp., Plasmodium spp., and Trypanosoma brucei). Two extrinsic controls (phocine herpesvirus 1 and bacteriophage MS2) were included to ensure extraction and amplification efficiency. Analytical validation was performed on spiked specimens for linearity, intra-assay precision, interassay precision, limit of detection, and specificity. The performance of the card on clinical specimens was evaluated with 1,050 blood samples by comparison to the individual real-time PCR assays, and the TAC exhibited an overall 88% (278/315; 95% confidence interval [CI], 84% to 92%) sensitivity and a 99% (5,261/5,326, 98% to 99%) specificity. This TaqMan array card can be used in field settings as a rapid screen for outbreak investigation or for the surveillance of pathogens, including Ebola virus.

Mycobacterium tuberculosis bacteremia in a cohort of hiv-infected patients hospitalized with severe sepsis in uganda­high frequency, low clinical suspicion [corrected] and derivation of a clinical prediction score.

BACKGROUND: When manifested as Mycobacterium tuberculosis (MTB) bacteremia, disseminated MTB infection clinically mimics other serious blood stream infections often hindering early diagnosis and initiation of potentially life-saving anti-tuberculosis therapy. In a cohort of hospitalized HIV-infected Ugandan patients with severe sepsis, we report the frequency, management and outcomes of patients with MTB bacteremia and propose a risk score based on clinical predictors of MTB bacteremia. METHODS: We prospectively enrolled adult patients with severe sepsis at two Ugandan hospitals and obtained blood cultures for MTB identification. Multivariable logistic regression modeling was used to determine predictors of MTB bacteremia and to inform the stratification of patients into MTB bacteremia risk categories based on relevant patient characteristics. RESULTS: Among 368 HIV-infected patients with a syndrome of severe sepsis, eighty-six (23%) had MTB bacteremia. Patients with MTB bacteremia had a significantly lower median CD4 count (17 vs 64 lymphocytes/mm(3), p<0.001) and a higher 30-day mortality (53% vs 32%, p = 0.001) than patients without MTB bacteremia. A minority of patients with MTB bacteremia underwent standard MTB diagnostic testing (24%) or received empiric anti-tuberculosis therapy (15%). Independent factors associated with MTB bacteremia included male sex, increased heart rate, low CD4 count, absence of highly active anti-retroviral therapy, chief complaint of fever, low serum sodium and low hemoglobin. A risk score derived from a model containing these independent predictors had good predictive accuracy [area under the curve = 0.85, 95% CI 0.80-0.89]. CONCLUSIONS: Nearly 1 in 4 adult HIV-infected patients hospitalized with severe sepsis in 2 Ugandan hospitals had MTB bacteremia. Among patients in whom MTB was suspected, standard tests for diagnosing pulmonary MTB were inaccurate for correctly classifying patients with or without bloodstream MTB infection. A MTB bacteremia risk score can improve early diagnosis of MTB bacteremia particularly in settings with increased HIV and MTB co-infection.

Integrating sepsis management recommendations into clinical care guidelines for district hospitals in resource-limited settings: the necessity to augment new guidelines with future research.

Several factors contribute to the high mortality attributed to severe infections in resource-limited settings. While improvements in survival and processes of care have been made in high-income settings among patients with severe conditions, such as sepsis, guidelines necessary for achieving these improvements may lack applicability or have not been tested in resource-limited settings. The World Health Organization's recent publication of the Integrated Management of Adolescent and Adult Illness District Clinician Manual provides details on how to optimize management of severely ill, hospitalized patients in such settings, including specific guidance on the management of patients with septic shock and respiratory failure without shock. This manuscript provides the context, process and underpinnings of these sepsis guidelines. In light of the current deficits in care and the limitations associated with these guidelines, the authors propose implementing these standardized best practice guidelines while using them as a foundation for sepsis research undertaken in, and directly relevant to, resource-limited settings.

Enrichment of HIV-1 subtype AD recombinants in a Ugandan cohort of severely septic patients.

BACKGROUND: Several population-wide HIV-1 subtype distribution studies in Uganda have evaluated relatively healthy clinic patients. Given the differences in HIV-1 disease progression based on subtype, we examined HIV-1 subtype distribution and disease outcomes among hospitalized patients with severe sepsis. METHODS: Patients with severe sepsis were enrolled at two hospitals in Uganda. Data collected included demographics, Karnofsky scores, highly active antiretroviral therapy (HAART) use, HIV-1 serostatus, CD4+ T cell concentration, whole blood lactate concentration, and blood cultures. HIV-1 subtypes were determined by sequencing parts of the gag and env genes, followed by phylogenetic analysis. RESULTS: Of the 267 patients evaluated, 228 (85.4%) were HIV infected. The predominant HIV-1 subtypes were A (46%), D (17%), and AD recombinants (30%). HIV-1 subtypes B, C, and other recombinants were uncommon. Patients infected with HIV-1 subtypes A, D and AD viruses were similar in demographics, CD4(+) T cell concentration, HAART use, Karnofsky scores, whole blood lactate concentration, and positive blood cultures. There was no difference in 30-day mortality from severe sepsis between the 3 groups (p = 0.99). CONCLUSION: A high proportion of HIV-1 subtypes A and AD recombinants was observed in this cohort of severely septic patients. The proportion of AD recombinants was higher in this cohort than in previous cohorts of Ugandan HIV-1 patients. No difference in baseline demographics, clinical factors or 30-day mortality was seen across HIV-subtypes.

The impact of early monitored management on survival in hospitalized adult Ugandan patients with severe sepsis: a prospective intervention study*.

UNLABELLED: In sub-Saharan Africa, sepsis is an important cause of mortality. Optimal sepsis management including fluid resuscitation, early antibiotic administration, and patient monitoring is limited by lack of supplies and skilled health workers. OBJECTIVE: To evaluate whether early, monitored sepsis management provided by a study medical officer can improve survival among patients with severe sepsis admitted to two public hospitals in Uganda. DESIGN, SETTING, AND PATIENTS: A prospective before and after study of an intervention cohort (n = 426) with severe sepsis receiving early, monitored sepsis management compared to an observation cohort (n = 245) of similarly ill patients with severe sepsis receiving standard management after admission to the medical wards of two Ugandan hospitals. INTERVENTION: Early sepsis management provided by a dedicated study medical officer comprising fluid resuscitation, early antibiotics, and regular monitoring in the first 6 hrs of hospitalization. MEASUREMENTS: Kaplan-Meier survival and unadjusted and adjusted Cox proportional hazards analysis were used to compare the effect of early, monitored sepsis management on 30-day mortality between the intervention cohort (enrolled May 2008 to May 2009) and observation cohort (enrolled July 2006 to November 2006). RESULTS: The majority (86%) of patients in both cohorts were human immuno-deficiency virus-infected. Median fluid volume provided in the first 6 hrs of hospitalization was higher in intervention than observation cohort patients (3000 mL vs. 500 mL, p < .001) and a greater proportion of intervention cohort patients received antibacterial therapy in <1 hr (67% vs. 30.4%, p < .001). Mortality at 30 days was significantly lower in the intervention cohort compared to the observation cohort (33.0% vs. 45.7%, log-rank p = .005). After adjustment for potential confounders, the hazard of 30-day mortality was 26% less in the intervention cohort compared to the observation cohort (adjusted hazards ratio 0.74, 95% confidence interval 0.55-0.98). Mortality among the 13% of intervention patients who developed signs of respiratory distress was associated with baseline illness severity rather than fluid volume administered. CONCLUSION: Early, monitored management of severely septic patients in Uganda improves survival and is feasible and safe even in a busy public referral hospital.

Hypoglycemia at admission is associated with inhospital mortality in Ugandan patients with severe sepsis.

OBJECTIVE: Dysglycemia during sepsis is associated with poor outcomes in resource-rich settings. In resource-limited settings, hypoglycemia is often diagnosed clinically without the benefit of laboratory support. We studied the utility of point-of-care glucose monitoring to predict mortality in severely septic patients in Uganda. DESIGN: Prospective observational study. SETTING: One national and two regional referral hospitals in Uganda. PATIENTS: We enrolled 532 patients with sepsis at three hospitals in Uganda. The analysis included 418 patients from the three sites with inhospital mortality data, a documented admission blood glucose concentration, and evidence of organ dysfunction at admission (systolic blood pressure≤100 mm Hg, lactate>4 mmol/L, platelet number<100,000/µL, or altered mental status). INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: We evaluated the association between admission point-of-care blood glucose concentration and inhospital mortality. We also assessed the accuracy of altered mental status as a predictor of hypoglycemia. Euglycemia occurred in 33.5% (140 of 418) of patients, whereas 16.3% (68 of 418) of patients were hypoglycemic and 50.2% (210 of 418) were hyperglycemic. Univariate Cox regression analyses comparing in-hospital mortality among hypoglycemic (35.3% [24 of 68], hazard ratio 2.0, 95% confidence interval 1.2-3.6, p=.013) and hyperglycemic (29.5% [62 of 210], hazard ratio 1.5, 95% confidence interval 0.96-2.4, p=.08) patients to euglycemic (19.3% [27 of 140]) patients showed statistically significantly higher rates of inhospital mortality for patients with hypoglycemia. Hypoglycemia (adjusted hazard ratio 1.9, 95% confidence interval 1.1-3.3, p=.03) remained significantly and independently associated with inhospital mortality in the multivariate model. The sensitivity and specificity of altered mental status for hypoglycemia were 25% and 86%, respectively. CONCLUSION: Hypoglycemia is an independent risk factor for inhospital mortality in patients with severe sepsis and cannot be adequately assessed by clinical examination. Correction of hypoglycemia may improve outcomes of critically ill patients in resource-limited settings.

Fitting a square peg into a round hole: are the current Surviving Sepsis Campaign guidelines feasible for Africa?

In their article, Baelani and colleagues surveyed anesthesia providers from African low- and middle-income countries (LMICs) to evaluate whether or not the current Surviving Sepsis Campaign (SSC) guidelines are feasible in such resource-constrained settings. The authors report that an alarmingly low percentage of hospitals have the capacity to implement the SSC guidelines in their entirety but a higher percentage are able to implement the majority of SSC guidelines and grade 1 recommendations. In reality, the probability of adherence to SSC guidelines for septic management is even lower than reported, given that the majority of sepsis management in African LMICs is likely performed by non-intensivists outside of intensive care units. Efforts to address the challenges of managing severely ill patients in LMICs have recently been taken on by the World Health Organization. After reviewing available evidence for sepsis management predominantly from high-income countries, a panel of experts developed a consensus-based strategy tailored for resource-limited settings. However, more research that can evaluate the challenges specific to sepsis management in LMICs and not currently addressed by the SSC guidelines is needed. Comprehensive, evidence-based guidelines combined with innovative approaches to sepsis management in LMICs are required to make a meaningful impact on worldwide sepsis survival.

Severe sepsis in two Ugandan hospitals: a prospective observational study of management and outcomes in a predominantly HIV-1 infected population.

BACKGROUND: Sepsis likely contributes to the high burden of infectious disease morbidity and mortality in low income countries. Data regarding sepsis management in sub-Saharan Africa are limited. We conducted a prospective observational study reporting the management and outcomes of severely septic patients in two Ugandan hospitals. We describe their epidemiology, management, and clinical correlates for mortality. METHODOLOGY/RESULTS: Three-hundred eighty-two patients fulfilled enrollment criteria for a severe sepsis syndrome. Vital signs, management and laboratory results were recorded. Outcomes measured included in-hospital and post-discharge mortality. Most patients were HIV-infected (320/377, 84.9%) with a median CD4+ T cell (CD4) count of 52 cells/mm(3) (IQR, 16-131 cells/mm(3)). Overall mortality was 43.0%, with 23.7% in-hospital mortality (90/380) and 22.3% post-discharge mortality (55/247). Significant predictors of in-hospital mortality included admission Glasgow Coma Scale and Karnofsky Performance Scale (KPS), tachypnea, leukocytosis and thrombocytopenia. Discharge KPS and early fluid resuscitation were significant predictors of post-discharge mortality. Among HIV-infected patients, CD4 count was a significant predictor of post-discharge mortality. Median volume of fluid resuscitation within the first 6 hours of presentation was 500 mLs (IQR 250-1000 mls). Fifty-two different empiric antibacterial regimens were used during the study. Bacteremic patients were more likely to die in hospital than non-bacteremic patients (OR 1.83, 95% CI = 1.01-3.33). Patients with Mycobacterium tuberculosis (MTB) bacteremia (25/249) had higher in-hospital mortality (OR 1.97, 95% CI = 1.19-327) and lower median CD4 counts (p = 0.001) than patients without MTB bacteremia. CONCLUSION: Patients presenting with sepsis syndromes to two Ugandan hospitals had late stage HIV infection and high mortality. Bacteremia, especially from MTB, was associated with increased in-hospital mortality. Most clinical predictors of in-hospital mortality were easily measurable and can be used for triaging patients in resource-constrained settings. Procurement of low cost and high impact treatments like intravenous fluids and empiric antibiotics may help decrease sepsis-associated mortality in resource-constrained settings.

Agreement between physicians and non-physician clinicians in starting antiretroviral therapy in rural Uganda.

BACKGROUND: The scarcity of physicians in sub-Saharan Africa - particularly in rural clinics staffed only by non-physician health workers - is constraining access to HIV treatment, as only they are legally allowed to start antiretroviral therapy in the HIV-positive patient. Here we present a pilot study from Uganda assessing agreement between non-physician clinicians (nurses and clinical officers) and physicians in their decisions as to whether to start therapy. METHODS: We conducted the study at 12 government antiretroviral therapy sites in three regions of Uganda, all of which had staff trained in delivery of antiretroviral therapy using the WHO Integrated Management of Adult and Adolescent Illness guidelines for chronic HIV care. We collected seven key variables to measure patient assessment and the decision as to whether to start antiretroviral therapy, the primary variable of interest being the Final Antiretroviral Therapy Recommendation. Patients saw either a clinical officer or nurse first, and then were screened identically by a blinded physician during the same clinic visit. We measured inter-rater agreement between the decisions of the non-physician health workers and physicians in the antiretroviral therapy assessment variables using simple and weighted Kappa analysis. RESULTS: Two hundred fifty-four patients were seen by a nurse and physician, while 267 were seen by a clinical officer and physician. The majority (>50%) in each arm of the study were in World Health Organization Clinical Stages I and II and therefore not currently eligible for antiretroviral therapy according to national antiretroviral therapy guidelines. Nurses and clinical officers both showed moderate to almost perfect agreement with physicians in their Final Antiretroviral Therapy Recommendation (unweighted kappa=0.59 and kappa=0.91, respectively). Agreement was also substantial for nurses versus physicians for assigning World Health Organization Clinical Stage (weighted kappa=0.65), but moderate for clinical officers versus physicians (kappa=0.44). CONCLUSION: Both nurses and clinical officers demonstrated strong agreement with physicians in deciding whether to initiate antiretroviral therapy in the HIV patient. This could lead to immediate benefits with respect to antiretroviral therapy scale-up and decentralization to rural areas in Uganda, as non-physician clinicians--particularly clinical officers--demonstrated the capacity to make correct clinical decisions to start antiretroviral therapy. These preliminary data warrant more detailed and multicountry investigation into decision-making of non-physician clinicians in the management of HIV disease with antiretroviral therapy, and should lead policy-makers to more carefully explore task-shifting as a shorter-term response to addressing the human resource crisis in HIV care and treatment.

Treatment of severe sepsis with artemether-lumefantrine is associated with decreased mortality in Ugandan patients without malaria.

We enrolled 382 patients at two hospitals in Uganda in a prospective observational study of severe sepsis. Because artemisinins improve survival in murine sepsis models, we performed a post hoc analysis of the association between the use of artemether-lumefantrine (A-L) and mortality in patients with or without malaria. In patients with negative malaria smears (N = 328 of 379), Kaplan-Meier curves revealed decreased combined inpatient and 30-day mortality among patients receiving A-L versus those who did not (20.6%, SE = 10.6 versus 48.8%, SE = 3.2; Log rank chi(2) = 3.93, P = 0.048). The decrease in mortality associated with A-L was maintained in the most clinically ill patients determined by Karnofsky Performance Scores