Publications

Polymyxin B is a reserve antibiotic, but there has been an upsurge in its use due to a rise in multidrug-resistant gram-negative bacteria. However, nephrotoxicity and resistance concerns persist, with global resistance rates reaching 29%. Effectiveness of Polymyxin B (clinical and microbiological response) and the frequency of nephrotoxicity is not well documented in resource limited settings. Research is essential to guide optimization of Polymyxin B therapy and inform the adoption of measures for early detection of kidney injury to prevent damage. This study aimed to assess the effectiveness of Polymyxin B by evaluating clinical and microbiological responses and determining nephrotoxicity incidence using KDIGO criteria in ICU patients at Moi Teaching and Referral Hospital (MTRH). A prospective observational cohort study was conducted at MTRH ICUs between December 2021 and November 2022, on patients treated with Polymyxin B. Data on demographics, comorbidities, Polymyxin B dosage regimens, clinical responses, and microbiological results were collected. Descriptive statistics summarized patient characteristics, while associations between dosage regimens and outcomes were evaluated using Fisher's exact test and multivariate regression, with a p<0.05 considered statistically significant. Forty-four patients with a mean age of 48 years were included; 66% were male, and cerebrovascular disease was the most common comorbidity. All patients had multidrug-resistant gram-negative infections qualifying for Polymyxin B therapy. Most (89%) received monotherapy, with 86% achieving a good clinical response, 7% experiencing treatment failure, and 7% dying. Doses of 20,000–25,000 IU/Kg/day were associated with microbiological eradication and good clinical response (p<0.001), while 15,000 IU/Kg/day was associated with treatment failure. Acute kidney injury occurred in 48% of patients, with 68% developing hypomagnesemia. Polymyxin B at doses of between 20,000-25,000IU/Kg/day should be considered as a starting dose due to the association with good clinical response, with alternate-day monitoring of serum creatinine levels for early detection of nephrotoxicity.

Background: Dexmedetomidine is the preferred drug for light sedation in intensive care units (ICU) where sedation plays an important role in patient comfort. Its advantages include shorter weaning time and earlier extubating from mechanical ventilation without respiratory depression. However, dexmedetomidine has been associated with over 50% incidence of hemodynamic adverse effects (hypotension and bradycardia) that has led to poor clinical outcomes. This limits its widespread use. Factors such as age, comorbidities, concomitant medications, dosage, baseline mean arterial pressure (MAP) and heart rate (HR) have been associated with adverse effects. Despite the high number of reported adverse effects, there is limited data on incidence and associated factors in resource limited settings. Therefore, its burden at Moi Teaching and Referral Hospital (MTRH) remains unknown. Knowledge on incidence and associated factors may inform future practice on safe use of dexmedetomidine at MTRH. Objective: To determine incidence and clinical factors associated with dexmedetomidine induced adverse effects among patients sedated with dexmedetomidine at MTRH ICU. Methods: This was a prospective observational study done at MTRH ICU. Hemodynamically stable eligible participants on dexmedetomidine were enrolled through census method between mid-April and mid-October 2022.The dependent variable was incidence of dexmedetomidine induced hemodynamic adverse effects. Data on MAP, HR were collected at selected time points within 24 hours. Independent variables which included factors such as age, gender, comorbidities, concomitant medication, renal, liver functions and drug dosages were obtained from patient records. Cutoff for hypotension was a MAP less than 60mmHg or a drop in MAP of 30% within the first hour, while bradycardia was a HR less than 60bpm or a drop in HR of 30% within first hour of drug administration. Continuous, data was summarized using means, medians and categorical data as frequencies and proportions. Fisher’s exact test and Kruskal Wallis test was used to assess for associations between categorical variables and continuous independent variables. The association between the clinical factors and development of dexmedetomidine induced hemodynamic adverse effects was analyzed using logistic regression model. Results: A total of 61 participants were recruited and 41% had traumatic brain injury. The mean age was 37 years and males were 63.9%. All participants had baseline HR>60bpm and MAP>60mmHg during drug initiation. Five patients (8.2%) developed hypotension and one (1.6%) developed bradycardia at the first hour. Mean baseline MAP was 90.49mmHg and mean decline was 5.16mmHg at 1hour. Mean baseline HR was 98.75bpm and mean decline was 0.91bpm within the first hour. Majority of patients received drug doses ranging from 0.2 to 0.7mcg/kg/hr for less than 24 hours. Lower baseline MAP <70mmHg was significantly associated with dexmedetomidine hemodynamic adverse effects (OR 2.17[95%CI 1.08-2.97, p<0.01]). However, there was no significant association between gender (0.21), baseline HR (0.88), comorbidities (0.19), concomitant medications (0.15), dose and duration (1), renal (0.28), liver functions (0.17) and occurrence of hemodynamic adverse effects. Conclusion: This study reported a low incidence of dexmedetomidine induced adverse effects compared to previous studies. Lower baseline MAP<70mmHg was an independent predictor of dexmedetomidine induced hypotension and bradycardia. Recommendations: Patients with lower baseline MAP<70mmHg and on dexmedetomidine should be monitored more frequently within the first hour.

Background: Dyslipidemia is the presence of abnormal blood lipid parameters, characterized by increased LDL-C, triglycerides, and total cholesterol but reduced HDL-C. It is a common finding in patients with T2DM, occurring at a prevalence rate of between 70% and 85%, and promotes the development of long-term cardiovascular complications, which are the leading cause of mortality in this population. Statins are the first-line drugs, but lipid control varies from patient to patient despite being widely used.Objective: To assess lipid control and the factors associated with LDL-C control in patients with type 2 DM who are on statins at a national referral hospital in Western Kenya.Methods: A retrospective study on 211 patients with type 2 DM who had been on a statin for at least three months. Data was obtained from patient records and lipid measures categorized as controlled or uncontrolled based on the Kenya National Guidelines for the Management of Diabetes Mellitus, 2018. Chi-square and Fischer’s exact test determined the association between variables. A multivariate logistic regression model was fit for variables significant at the bivariate level, and a P value of <0.05 was considered significant.Results: Most (99%) were on a single lipid-lowering drug, mainly atorvastatin, and 92% were on moderate-intensity dosing. Regarding lipid control, 50.3% had uncontrolled LDL-C, 30% had uncontrolled HDL-C, and 47% had uncontrolled triglyceride levels. Being on a high-intensity statin increased the likelihood of LDL-C control compared to moderate-intensity dosing (OR 8.57 [95% CI 4.3-16.9, P<0.001]).Conclusion: LDL-C was the most poorly controlled parameter. Patients on high-intensity statins had better LDL-C control; therefore, high-intensity statin therapy should be initiated in diabetic patients who do not achieve their LDL-C targets.

Background: Pregnancy poses specific challenges in the diagnosis of Plasmodium falciparum infection due to parasite sequestration in the placenta. The diagnosis of Plasmodium falciparum infection in pregnant mothers therefore requires highly sensitive methods in order to detect the presence of parasites. These include those that detect the presence of antigens and those that detect and quantify the presence of the malaria parasites.Objective: The study assessed the performance of mRDT diagnostic test ((PfHRP2-RDT) in the detection of malaria infection in blood samples from nulliparous pregnant women within the first trimester of pregnancy in Western Kenya.Methods: This was a prospective study on blood specimens collected from pregnant women in a malaria-endemic region in Kenya. m-polymerase chain reaction (mPCR) and mRDT tests were performed. The diagnostic accuracy of m-RDT was compared with mPCR as the gold standard for the purpose of this study.Setting: Twelve primary health facilities in Busia, Bungoma and Kakamega Counties in KenyaResults: Out of 264 mPCR positive samples, 130 were mRDT positive (true positives) while 134 were mRDT negative (false negative). And out of 441 mPCR negative samples, 41 were positive on mRDT (false positive). Thus, in comparison with mPCR, the sensitivity and specificity of mRDT to detect malaria infection in nulliparous pregnant mothers in first trimester was 49.2% and 88.9% respectivelyConclusions: The sensitivity of mRDT to detect Plasmodium falciparum infections in nulliparous pregnant mothers in the first trimester was not satisfactory compared to mPCR tests.