Findings and significance of Multistix urinalysis testing in a population survey

INTRODUCTION Two main functions of the kidneys are to clean body fluids from several waste metabolites and keep the hydro-electrolytic balance by producing urine which contains body endogenic and exogenic wastes (1). Testing urine can help screen, diagnose and monitor a variety of diseases and metabolic conditions such as diabetes or liver diseases (2). The urine dipstick test is a simple diagnostic tool commonly performed in standard urinalysis in medical practice to assess selected metabolites (e.g. urobilinogen, glucose, etc.), markers (erythrocytes, leucocytes, etc.) and other characteristics of urines (e.g. pH, density, etc.) in urine (3). To realize the test, a strip with small squared colored fields (up to ten and each field is dedicated for one substance) is dripped into the collected clean midstream urine sample for a few seconds. Results can be inferred from the color changes occurring within seconds or minutes, which are compared with a reference color table, which usually appears on the urine test package (4). Dipstick urinalysis is often performed as part of a routine medical examination. It can be done by any health professional at the doctor's office, in the hospital or at home by patients for self-monitoring (2). Dipstick urinalysis is simple to perform, results are obtained quickly (about 1 minute), at a low cost (less than 1 US $ for one test in most instances) and the test, which is non-invasive, is not associated with side effects (5). Dipstick testing can be specific enough to detect a variety of abnormal conditions (e.g. proteinuria, glycosuria, ketonuria, microscopic hematuria) that correspond to actual diseases (diabetes, kidney disease). However, it remains generally useful to confirm dipstick results with other testing and clinical information (4). Like most screening tests, dipstick testing can also result in falsely positive or falsely negative results(4). False positives may lead to labeling a healthy person as sick (6), including doing subsequent unnecessary investigations such as kidney biopsy, cystoscopy, unnecessary antibiotic treatment, long term follow-up for inconsequential abnormalities and psychological stress in healthy persons (7). The sensitivity of the dipstick test is variable according to the tested substance, from 19-48% for the nitrite to 91-100% for the hematuria (Table1), and many other conditions may lead to false negative results (22). False negative results may lead to missed diagnosis and delayed management of underlying conditions. The adequate use of the dipstick test is further complicated because it assesses multiple substances (up to 10), which increases the potential detection of both false positives and false negatives. A number of studies have been conducted to analyze the benefit of urine dipstick screening (Table 1 and Table 2). It has been recommended to screen for protein or albumin for persons with certain risk factors, such as hypertension or diabetes (4,8). In contrast, the Canadian Task Force on the Periodic Health Examination, the American Cancer Society and other authoritative organizations do not recommend the screening for hematuria and proteinuria in asymptomatic persons, but screening for hematuria may be appropriate for men and women over 60 years of age (5). Few studies have assessed the usefulness of screening for hemoglobin, protein or glucose in urines and even fewer have assessed the benefits and harms to test for pH, density, leukocytes, urobilinogen and nitrite (7). Some studies suggest that the clinical benefits of multistix testing are small (7). The question arises, then, whether the test is useful for mass screening in healthy individuals. In this study, we examine 1) the frequency of abnormal conditions detected by the multistix when applying the test to a random sample of adults; 2) whether positive tests inter-correlate; and 3) the association of positive tests with selected socio-demographic and clinical variables. We also reviewed the literature in relation to recommendations for the use of multistix testing. The study is based on results of a multistix test that was applied to all participants of a national population survey in the Republic of Seychelles (i.e. a random sample of the general adult population). Implicit to the objectives of this study described above is the question of whether it can be useful to use a multistix screening test in participants to population-based surveys, which are generally attended mainly by apparently healthy individuals. Methods and analysis We used data from a population-based health survey conducted in the Seychelles in 2013-2014. Selection of the eligible participants was made using a random selection of all adults aged 25-64 years, from the electronic register of the whole population aged 25-64 years. A total of 1240 men and women participated in the survey (participation rate of 73%). Lifestyle and health behaviors were assessed by questionnaire and several clinical variables (e.g. blood pressure, body mass index, etc.) and laboratory variables were measured (e.g. fasting blood glucose, lipids, CRP, creatinine). A spot sample of urine was collected between 7 and 10 am (hence most often from a second morning urine) and immediately tested with a multistix urinalysis strip (Bayer, Mutistix 10 SG), which tests for the presence of glucose, protein, ketones, blood, leucocytes, nitrites, bilirubin, urobilinogen, pH and density. And the results were read by trained health personnel. We tabulated the distribution of the tests. We analyzed results two times: once considering “trace” and “1+” results as a positive test (i.e. a stringent criterion for assessing a test as positive) and the second time by considering “trace” and “1+” results as a negative tests (noise). We preformed correlation analysis between all the urinary tests. We then examined the multivariate associations of dichotomized multistix tests according to age, sex, smoking, alcohol drinking, BMI, blood pressure, blood glucose, CRP, creatinine using logistic regression. We used the statistical software STATA version 14. RESULTS In total, 1240 men and women aged 25-64 years participated in the study (a 73% participation rate). Table 3 shows that the large majority of the participants had negative (normal) results for each of the dipsticks parameters with a frequency varying from 90-99.7 %. Considering trace as positive augment the frequency of about 5% for ketone, protein and blood. In descending order of positive results, we found protein, blood and ketone. Table 4 shows several correlations between the parameters. The strongest correlation is between glucose and protein and between nitrite and bilirubin. There are also correlations between leucocyte and protein, and between blood and glucose. Table 5 shows several statistically significant associations between urine tests and the selected variables, including between: - blood and elevated CRP and creatinine, - leukocyte and physical activity, hospital admission in the last 12 months and high CRP and female sex. - nitrite and female sex. - glucose with diabetes. - ketone with daily smoking and diabetes. - protein with high blood pressure and diabetes. DISCUSSION The findings show that most of the participants in a population-based survey have normal results using a dipstick test. The test most often found to be positive was protein. We found several correlations between the dipstick parameters. We also found several associations between dipstick tests and selected socio-demographic and biological conditions, particularly with glucose, ketones, and proteins.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.