Point of care creatinine testing for early detection of renal dysfunction in Tanzanian HIV patients: a study protocol

Abstract

Background

Renal dysfunction (RD) is more prevalent among people living with HIV (PLHIV) in sub-Saharan Africa (SSA) due to factors such as co-infections, hypertension, diabetes, and nephrotoxic antiretroviral drugs like tenofovir. Early detection of RD is critical but limited by resource constraints in many SSA countries, including Tanzania. Point-of-care (POC) tests, such as the Stat-Sensor Creatinine test by Nova Biomedical, present a cost-effective and non-invasive option for early detection. This study will evaluate the diagnostic performance of the Stat-Sensor creatinine test in detecting renal dysfunction in HIV care and treatment clinics (CTCs) in Tanzania. It will be conducted for a period of six months, from November 2024 to April 2025.

Methodology

The study will measure point-of-care creatinine using a rapid creatinine dipstick against the standard serum creatinine test. The diagnostic performance and agreement to diagnose renal dysfunction will be assessed using a standardized statistical approach: Bland-Altman analysis and linear mixed-effects models to test agreement between creatinine dipstick tests with serum creatinine. Receiver operating characteristics (ROC) statistics will be used to test the diagnostic performance of the creatinine dipstick test to diagnose renal dysfunction.

Discussion

We hypothesize that POC creatinine testing will show strong diagnostic performance, providing a reliable, rapid, and cost-efficient alternative for RD detection, leading to better patient outcomes and integration of POC tests into routine HIV care.

Ethical clearance

The study protocol was reviewed and approved by the Muhimbili University of Health and Allied Sciences Research and Publication committee in June 2024 with reference number, MUHAS-REC-05-2024-2275.

Introduction

With increased access to HIV treatment among people living with HIV (PLHIV) in low- and middle-income countries (LMICs), there is a rising risk of non-communicable diseases (NCDs) within this cohort [1, 2]. Renal dysfunction (RD) is a significant concern among people living with HIV (PLHIV) in Tanzania, with prevalence rates ranging from 7 to 20%, reflecting the substantial burden within this population [3, 4]. This is largely due to hypertension, diabetes, co-infections with hepatitis viruses, and the use of nephrotoxic drugs such as tenofovir [5, 6]. The co-existence of HIV with RD increases the complexity of patient care, leading to poorer health outcomes and higher healthcare costs [7].

Currently, both local and international HIV treatment guidelines recommend screening for RD at baseline for all newly diagnosed HIV patients and every six months after antiretroviral therapy (ART) initiation [8, 9]. The results guide the appropriate treatment regimen for HIV and are done using serum creatinine and urine analysis [8]. However, disparities in equipment availability, skilled personnel, and affordability contribute to varying levels of implementation in most HIV treatment centers in Tanzania and other similar settings [10]. Some patients completely lack access to RD screening, hindering proper clinical decision-making, early detection of RD, and preventive measures for kidney disease despite regular contact with healthcare providers [4]. Even with RD screening, there can be delays in obtaining test results, hence impacting timely clinical decisions [5].

The StatSensor Creatinine Hospital Meter System is a handheld analyzer that uses a miniaturized, single-using multi-enzyme (creatinine amidohydrolase, creatine amidohydrolase, and sarcosine oxidase) with electrochemical biosensor test strip for whole blood creatinine testing, incorporating patented Multi-Well™ technology to enhance accuracy and reliability [11, 12]. Designed for point-of-care testing (POC), it provides a simple, and rapid assessment of renal function using finger stick capillary blood [13]. The test has been reviewed but not approved by the U.S. FDA, and while it is available in India and South Africa for clinical use, it is approved only for research purposes in Tanzania [11]. Currently, it does not have WHO Prequalification (PQ), and there is no mention of CE marking, indicating further regulatory approvals may be pending.

The StatSensor Creatinine Hospital Meter System’s efficacy was evaluated against the laboratory standard serum creatinine test in the United Kingdom and Canada. It showed good agreement in patients with lower creatinine levels but a negative bias in those with higher levels, indicating potential underestimation. However, the studies had a limited number of participants [12, 13]. A multi-country study in Bolivia, Nepal, and Malawi showed that implementing StatSensor point-of-care testing improved the recognition and management of acute kidney injury (AKI) and acute kidney disease (AKD) among high-risk patients. [14]. To reduce the existing infrastructure gap in screening for RD among PLHIV when accessing care, we aim to assess the diagnostic accuracy of point-of-care (POC) creatinine tests in detecting renal dysfunction among PLHIV attending Care and Treatment Clinics (CTC) in Tanzania.

Methods and analysis

Study hypothesis: Based on the existing evidence [12, 13, 15], we hypothesize that, Point of care (POC) creatinine (StatSensor Creatinine) test will demonstrate a high diagnostic accuracy compared to standard serum creatinine in detecting renal dysfunction among PLHIV.

Specific aims

• 1. To compare the diagnostic performance of the StatSensor creatinine test with standard serum creatinine tests in identifying renal dysfunction.

• 2. To evaluate the association between early detection of RD using POC creatinine test and clinical outcomes (change in drug regimen, referral to higher facility) in PLHIV.

Study design and settings

This will be a diagnostic accuracy cross-sectional study conducted at three HIV care and treatment clinics (CTCs) at Regional Referral Hospitals (RRHs) in Dar es Salaam, Tanzania (Mwananyamala, Temeke, and Ilala). These hospitals were chosen because they are key regional referral centers in Dar es Salaam, providing care and treatment services to a large number of people living with HIV (PLHIV). This will ensure a diverse and representative sample of the target population. Additionally, due to the large number of patients attending these facilities, the targeted sample size can be achieved within the allocated timeline, making them ideal for this study.

Recruitment Procedures and Sampling Technique: Participants will be randomly selected from patients attending the Care and Treatment Clinics (CTC). Each patient will be assigned a unique identification number from the clinic appointment book. These identification numbers will be written on pieces of paper, and a Probability Proportional to Size (PPS) technique will be utilized to select participants based on the patient volume at each clinic. Recruitment will continue until the desired total sample size of 717 participants is achieved, reflecting the distribution of patients across the three study centers.

A study physician at each site will ensure that participants meet the predefined inclusion criteria. Data collection will utilize electronic clinical report forms integrated with REDCap software, ensuring that participant information is kept separate from data processing and analysis. All recruitment and procedural activities will adhere strictly to good clinical practice principles and will be carried out by personnel specifically trained for the study.

Case Definition, Risk Groups, and Inclusion/Exclusion Criteria: Our research focuses on PLHIV who are either newly diagnosed or undergoing follow-up care in HIV clinics (CTCs). Renal dysfunction will be defined as eGFR of < 60mL/min/1.73m², calculated by CKD-EPI 2021 [9, 16].

Inclusion criteria

• PLHIV aged 18 years and above attending CTC clinics in the three regional hospitals.

Exclusion criteria

• PLHIV unable to provide blood samples for analysis.

Primary outcomes measure – Accuracy of point of care (POC) creatinine dipstick in detecting renal dysfunction among PLHIV compared to standard serum creatinine test and the association between early detection of renal dysfunction using the POC creatinine test (StatSensor creatinine by Nova Biomedical) and clinical outcomes (change in drug regimen, referral to a higher facility) in PLHIV.

Clinical procedures and measurements

All sampled patients will be informed about the study and those patients that consented will be recruited in a study. The data will be collected by using a well-structured interviewer administered electronic questionnaire which will include socio-demographic factors, clinical findings as well as selected laboratory results.

Physical examination: Blood pressure will be measured using standard techniques, calibrated blood pressure machine ‘Omron M1 Plus’. This will be taken after the patient had taken rest of about five minutes; the average of two readings will be taken as the blood pressure of the patient. Weight and height will be measured by the research assistant, who will then calculate the Body Mass Index (BMI) based on these readings. Blood pressure was categorized as normal (< 120/80 mmHg), elevated (120–129/<80 mmHg), and high (≥ 130/80 mmHg and/or use of antihypertensive therapy) according to the American Heart Association (AHA) hypertension guidelines [17, 3]. Diabetes mellitus will be defined as an RBG value of ≥ 11.1 mmol/L and/or use of antidiabetic medication(s).

Laboratory measurements

Blood samples will be collected by trained phlebotomists using standard aseptic techniques. A total of 5 ml of whole blood will be drawn from each study participant. Following collection, a drop of whole blood will be used immediately for point-of-care (POC) creatinine testing using the StatSensor Creatinine test. The remaining blood sample will be processed for serum creatinine analysis. To do this, the collected blood will be centrifuged at 3,000 RPM for 10 min within two hours of collection. The serum will then be carefully separated and transferred into appropriate containers, which will be labeled with unique study identifiers to prevent contamination. All samples will be transported in a temperature-controlled cooler to a certified laboratory within one hour of collection, where they will be analyzed using the Absorbance Assay method on the Mindray BS-230 chemistry analyzer. Proper chain of custody protocols will be followed to ensure sample integrity throughout the process. Laboratory professionals will oversee the processing of serum samples to maintain high standards of quality and accuracy.

The study will assess renal function among these participants using the estimated glomerular filtration rate (eGFR) calculated from serum creatinine levels, applying the 2021 CKD-EPI equation. In our analysis, eGFR will serve as a continuous outcome variable, enabling detailed regression analysis. Additionally, eGFR values will be categorized into several groups as defined by the KDIGO guideline (18): normal/high (≥ 90), mildly decreased (60–89), mild–moderately decreased (45–59), moderately–severely decreased (30–44), severely decreased (15–29), and kidney failure (< 15) mL/min/1.73 m²[18, 19] These categories will provide a comprehensive understanding of renal health across different stages of kidney function in our study population.

Research assistants who are medical doctors will undergo a training program, including both theoretical and practical components, to ensure proficiency in conducting point-of-care tests and handling study data. The training will cover the study protocol, proper use of the rapid creatinine analyzer, ethical considerations, and data collection techniques. Competency assessments will be conducted at the end of the training to ensure that all research assistants are adequately prepared. Continuous supervision and refresher training sessions will be provided throughout the study to maintain high standards of performance.

Sample size calculation and analysis plan

The minimum sample size required is 717, calculated based on a sensitivity of 96% and a specificity of 87% as described in the previous studies of stat sensor creatinine diagnostic accuracy (details attached in supplementary material) [20, 21].

For aim 1

Descriptive statistics will include measures of central tendency (mean, median) and spread (standard deviation; quartiles, interquartile range). Point-of-care (POC) creatinine measurements will be correlated with serum creatinine levels using Spearman’s rank test. We will then evaluate the sensitivity and specificity of the POC creatinine using Receiver Operating Characteristic (ROC) curve analysis with serum creatinine values as the gold standard. Additionally, we will conduct sensitivity testing to assess the robustness of the POC creatinine under different conditions. Likelihood ratios will be calculated to further evaluate the diagnostic accuracy of the POC creatinine. Simple linear regression will be employed to analyze the relationship between the POC creatinine with serum creatinine respectively. We will control for potential confounders such as ART regimens, baseline health status, comorbid conditions, and demographic variables by collecting comprehensive baseline data on these factors. Statistical adjustments will be made in the analysis to account for these variables. A p-value less than 0.05 will be considered statistically significant.

For aim 2

Chi-square tests for bivariate analysis. Logistic regression will be employed to adjust for confounders like age, sex, and HIV treatment duration. Descriptive statistics will be presented to summarize the distribution of early RD detection and corresponding clinical outcomes.

Time frame and milestones

The study will be conducted over six months, from November 2024 to April 2025, with key milestones including recruitment, data collection, analysis, and dissemination.

Potential risk and mitigation

No major risk are expected from the study or during sample collection. There is minimal risk of pain or difficulty in collecting blood samples from the participants. Also, there is a risk of needle prick injury during sample collection, exposing the phlebotomist to the risk of blood-borne infection. To minimize these risks, qualified phlebotomists with experience in HIV clinics will be involved with sample collection, adhering to all safety measures.

Patient involvement

PLHIV attending selected CTC clinics will receive basic information on the association of HIV and NCDs, specifically kidney diseases. To further engage the patients, a detailed explanation of all the procedures involved will be provided to all patients before and during data collection. In addition, all enrolled patients will be given their examination findings and serum creatinine test results. The results will also be shared with the attending physicians to facilitate proper clinical decision-making.

Discussion

Expected Outcomes and Potential Impact: Currently, the majority of people living with HIV (PLHIV) who attend care primarily in the Care and Treatment Clinics (CTCs) at primary facilities do not routinely receive standard care testing for non-communicable diseases (NCDs), including screening for renal dysfunction, as indicated in the standard care guidelines. We expect that our study will demonstrate the diagnostic accuracy of point-of-care (POC) creatinine testing. This could facilitate its integration into routine HIV care, thereby improving the detection and management of renal dysfunction among PLHIV.