FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2913215572> ?p ?o ?g. }

• W2913215572 endingPage "727" @default.
• W2913215572 startingPage "723" @default.
• W2913215572 abstract "The examination of urine to diagnose disease is an age-old practice, dating back thousands of years in primitive forms, and arguably represents the genesis of laboratory medicine.1Armstrong J.A. Urinalysis in Western culture: a brief history.Kidney Int. 2007; 71: 384-387Abstract Full Text Full Text PDF PubMed Scopus (70) Google Scholar Microscopic evaluation of urine was introduced in the 19th century, and in the 20th century became a cornerstone of the nephrologist’s armamentarium for diagnosing kidney disease. Since the introduction of automated urine analyzers, manual examination of the urine sediment has rapidly fallen out of favor among clinicians.2Becker G.J. Garigali G. Fogazzi G.B. Advances in urine microscopy.Am J Kidney Dis. 2016; 67: 954-964Abstract Full Text Full Text PDF PubMed Scopus (55) Google Scholar, 3Fogazzi G.B. Grignani S. Urine microscopic analysis. An art abandoned by nephrologists?.Nephrol Dial Transplant. 1998; 13: 2485-2487Crossref PubMed Scopus (25) Google Scholar, 4Perazella M.A. The urine sediment as a biomarker of kidney disease.Am J Kidney Dis. 2015; 66: 748-755Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar Despite its central role in the evaluation of patients with kidney disease, limited data are available in the literature on the test performance characteristics of the modern urinalysis (UA) as reported by laboratories using automated analyzers.5Tsai J.J. Yeun J.Y. Kumar V.A. et al.Comparison and interpretation of urinalysis performed by a nephrologist versus a hospital-based clinical laboratory.Am J Kidney Dis. 2005; 46: 820-829Abstract Full Text Full Text PDF PubMed Scopus (63) Google Scholar In a cross-sectional analysis, we studied how well automated UAs distinguished proliferative glomerulonephritis (PGN) from other forms of kidney disease in a cohort of adult patients who had not yet been initiated on immunosuppressive therapy, and for whom clinicopathologic diagnoses were uniformly adjudicated by native kidney biopsies. A total of 512 patients were included in the analysis, 511 of whom had automated urine test strip results available and 421 of whom had urine red blood cell (RBC) counts available within 30 days before undergoing native kidney biopsy (Supplementary Figure S1). Of the 512 patients included in the analysis, 134 had PGN. The most common PGN diagnoses were IgA nephropathy (n = 74), antineutrophil cytoplasmic antibody−associated vasculitis (n = 19), and proliferative forms of lupus nephritis (n = 11). The most common non-PGN diagnoses were diabetic nephropathy (n = 63), membranous nephropathy (n = 38), and secondary focal segmental glomerulosclerosis (n = 35) (Supplementary Table S1). The mean age of the cohort was 53.9 ± 15.9 years, 46.3% were female, and 68.6% were white. The median estimated glomerular filtration rate was 44.9 (interquartile range [IQR] 26.3–76.8) ml/min per 1.73 m2, and median proteinuria was 2.0 (IQR 0.6−5.0) g/g creatinine (Table 1).Table 1Baseline characteristics of study cohortCharacteristicPGN (n = 134)Other kidney disease (n = 378)P valueAge (yr)50.5 ± 17.555.1 ± 15.2<0.01Female (%)46.346.30.99Race (%)<0.01 White66.769.3 Black11.621.4 Other21.79.3Median serum creatinine (μmol/l)138 (97–203)144 (88–214)0.73Median eGFR (ml/min per 1.73 m2)47.6 (29.3–69.2)42.5 (24.8–77.4)0.54Median proteinuria (g/g creatinine)1.8 (0.8–3.7)2.1 (0.5–5.5)0.69Median urine RBC count per HPF18 (6–60)2 (1–10)<0.01Urine dipstick blood (%)<0.01 None or trace8.343.6 1+8.318.4 2+21.817.8 3+61.620.2DM (%)10.528.3<0.01HTN (%)43.356.4<0.01ACEI/ARB (%)37.347.10.05Indications for biopsyaIndividual patients may have more than 1 indication for biopsy. (%)<0.01 Proteinuria67.952.9 Hematuria46.316.9 Abnormal GFR50.854.5Most common primary clinicopathologic diagnosesIgA nephropathy (n = 74)Diabetic nephropathy (n = 63)ANCA-associated vasculitis (n = 19)Membranous nephropathy (n = 38)Proliferative lupus nephritis (n = 11)Secondary FSGS (n = 35)Immune complex GN (n = 11)Advanced chronic changes (n = 29)Cryoglobulinemic GN (n = 4)Vascular sclerosis (n = 26)ACEI, angiotensin-converting enzyme inhibitor; ANCA, antineutrophil cytoplasmic antibody; ARB, angiotensin II receptor blocker; DM, diabetes mellitus; eGFR, estimated glomerular filtration rate; FSGS, focal segmental glomerulosclerosis; GFR, glomerular filtration rate; GN, glomerulonephritis; HPF, high-power field; HTN, hypertension; PGN, proliferative glomerulonephritis.a Individual patients may have more than 1 indication for biopsy. Open table in a new tab ACEI, angiotensin-converting enzyme inhibitor; ANCA, antineutrophil cytoplasmic antibody; ARB, angiotensin II receptor blocker; DM, diabetes mellitus; eGFR, estimated glomerular filtration rate; FSGS, focal segmental glomerulosclerosis; GFR, glomerular filtration rate; GN, glomerulonephritis; HPF, high-power field; HTN, hypertension; PGN, proliferative glomerulonephritis. Patients with PGN had a median urine RBC count of 18 (IQR 6−60) per high-power field (HPF), whereas those with other forms of kidney disease had a median urine RBC count of 2 (IQR 1−10) per HPF (P < 0.01). Moreover, among the patients with PGN, we found a trend toward higher RBC counts in patients with crescentic disease compared to those without glomerular crescents (median [IQR] 23.5 (11.5−92.5) vs. 15 (4−60) RBCs/HPF, respectively, P = 0.06). Of the patients with PGN, 8.3% had less than 1+ blood on their test strip compared to 43.6% of those with other forms of kidney disease. The Spearman correlation coefficient between test strip blood measurements and the urine RBC count was 0.66. Table 2 demonstrates the performance characteristics of automated urine test strip protein and blood at different thresholds for diagnosis of PGN versus other causes of kidney disease. Table 3 shows the same performance characteristics for quantitative proteinuria measurements and automated urine RBC counts. Figure 1 shows receiver operating characteristic (ROC) curves for diagnosis of PGN versus other causes of kidney disease using test strip blood or the automated urine RBC count as predictors. The areas under these ROC curves were 0.77 and 0.75, respectively. The difference in the ROC curves was not significant when compared among patients who had both tests performed (P = 0.15). Using the laboratory’s conventional threshold of >2 RBCs/HPF to define abnormal hematuria, the RBC count had 86% sensitivity, 51% specificity, 39% positive predictive value (PPV), and 91% negative predictive value (NPV) for PGN. Among patients with proteinuria <0.5 g/g creatinine, NPV increased to 96%. Analogously, a negative test strip for blood had 95% sensitivity, 29% specificity, 32% PPV, and 94% NPV. The NPV increased to 96% when restricted to patients with proteniuria <0.5 g/g creatinine.Table 2Sensitivity (Sens), specificity (Spec), positive predictive value (PPV), negative predictive value (NPV), postive likelihood ratio (LR+), and negative likelihood ratio (LR−) at different thresholds of test strip protein and blood for diagnosis of proliferative glomerulonephritisBlood ≥0Blood ≥TRBlood ≥1+Blood ≥2+Blood 3+Protein ≥0 Sens/Spec (%)94.7/29.091.7/43.683.5/62.061.7/79.8 PPV/NPV (%)32.1/94.036.5/93.743.7/91.451.9/85.5 LR+/LR−1.33/0.181.71/0.182.18/0.273.05/0.48Protein ≥TR Sens/Spec (%)94.7/14.690.9/37.088.6/50.081.1/65.459.9/82.2 PPV/NPV (%)28.1/88.733.6/92.138.4/92.645.2/90.854.1/85.4 LR+/LR−1.11/0.361.44/0.251.77/0.232.34/0.293.37/0.49Protein ≥1+ Sens/Spec (%)85.7/24.481.8/43.181.1/54.874.2/68.955.3/84.3 PPV/NPV (%)28.6/82.933.5/87.138.6/89.245.6/88.455.3/84.3 LR+/LR−1.13/0.591.44/0.421.79/0.342.39/0.373.52/0.53Protein ≥2+ Sens/Spec (%)73.7/36.170.5/49.569.7/59.663.6/71.847.7/85.9 PPV/NPV (%)28.9/79.532.9/82.737.7/84.944.2/84.954.3/82.4 LR+/LR−1.15/0.731.40/0.601.73/0.512.26/0.513.38/0.61Protein ≥3+ Sens/Spec (%)39.1/61.336.4/66.236.4/72.633.3/80.325.0/91.2 PPV/NPV (%)26.3/74.027.4/74.831.8/76.537.3/77.450.0/77.6 LR+/LR−1.01/0.991.08/0.961.33/0.881.69/0.832.84/0.82TR, trace. Open table in a new tab Table 3Sensitivity (Sens), specificity (Spec), positive predictive value (PPV), negative predictive value (NPV), postive likelihood ratio (LR+), and negative likelihood ratio (LR−) for different thresholds of quantitative proteinuria and urine RBC counts for diagnosis of proliferative glomerulonephritisRBCs ≥0/HPFRBCs >2/HPFRBCs >5/HPFRBCs >10/HPFRBCs >15/HPFProtein ≥0 g/g Sens/Spec (%)85.7/51.476.2/62.361.0/75.750.5/81.2 PPV/NPV (%)38.8/90.942.1/87.947.4/84.449.1/82.0 LR+/LR−1.76/0.282.02/0.382.51/0.522.69/0.61Protein ≥0.5 g/g Sens/Spec (%)86.7/23.374.3/61.665.7/71.651.4/80.141.9/84.9 PPV/NPV (%)28.9/82.941.1/87.045.4/85.348.2/82.150.0/80.3 LR+/LR−1.13/0.571.93/0.422.31/0.482.58/0.612.77/0.68Protein ≥1.0 g/g Sens/Spec (%)72.4/35.361.9/66.454.3/75.043.8/82.935.2/87.3 PPV/NPV (%)28.7/78.039.9/82.943.9/82.047.9/80.450.0/79.0 LR+/LR−1.12/0.781.84/0.572.17/0.612.56/0.682.77/0.74Protein ≥2.0 g/g Sens/Spec (%)48.6/47.641.9/72.340.0/78.830.5/86.623.8/90.4 PPV/NPV (%)25.0/72.035.2/77.640.4/78.545.1/77.647.2/76.4 LR+/LR−0.93/1.081.51/0.801.89/0.762.28/0.802.48/0.84Protein ≥3.5 g/g Sens/Spec (%)28.6/62.026.7/78.425.7/83.920.0/90.116.2/93.8 PPV/NPV (%)21.3/70.730.8/74.836.5/75.942.0/75.848.6/75.7 LR+/LR−0.75/1.151.24/0.931.60/0.892.02/0.892.61/0.89HPF, high-power field; RBC, red blood cell. Open table in a new tab TR, trace. HPF, high-power field; RBC, red blood cell. When test strip blood was added to a clinical prediction model of PGN which included age, sex, race (black vs. nonblack), proteinuria (<1 vs. ≥1 g/g creatinine), estimated glomerular filtration rate, acute kidney injury as the reason for biopsy, diabetes mellitus, hypertension, systemic lupus erythematosus, vasculitis, and hepatitis C, the area under the ROC curve increased from 0.71 for the base model alone up to 0.81 (P < 0.01). Among patients for whom automated urine RBC counts were available, the addition of the RBC count instead of test strip blood to the same clinical prediction model increased the area under the ROC curve from 0.71 to 0.76 (P < 0.01) (Figure 2). Our data show that the automated UA has fair ability to differentiate PGN from other kidney diseases. A negative urine RBC count or dipstick for blood had high negative predictive value, especially among patients with low levels of proteinuria, despite our cohort having a high proportion of patients with PGN. Both tests, however, had limited specificity and positive predictive value, and similar performance overall. It is possible that specificity and PPV would be higher in nonbiopsied chronic kidney disease cohorts, as most patients with diseases such as diabetic nephropathy or hypertensive nephrosclerosis do not undergo biopsy, and our cohort is thus likely to be enriched for atypical presentations of these diseases. Although this could affect the generalizability of our findings, the inclusion of only patients with biopsy-proven disease is a fundamental strength of our analysis. Without being able to compare our automated UA results to the results of the gold standard diagnostic method, ascertainment bias could lead to invalid estimates of the test performance characteristics. Despite its limitations as a diagnostic test, the automated UA added significantly to a basic clinical prediction model of PGN. Among patients with PGN, we furthermore found a trend toward an association between higher levels of hematuria and more severe disease, as indicated by the presence of glomerular crescents. Although this finding was not statistically significant, our study may not have been adequately powered to detect a statistically significant difference. Taken together, our data demonstrate quantitatively how the automated UA may aid clinicians, albeit imperfectly, when determining the appropriateness of additional workup for kidney disease, such as serological studies and, ultimately, a kidney biopsy. Interestingly, we found no difference between the performance characteristics of the urine RBC count and the semiquantitative test strip blood measurements for diagnosis of PGN. Because automated analyzers do not reliably detect less common but more specific features of PGN in the urine sediment, such as acanthocytes and RBC casts,6Köhler H. Wandel E. Brunck B. Acanthocyturia–a characteristic marker for glomerular bleeding.Kidney Int. 1991; 40: 115-120Abstract Full Text PDF PubMed Scopus (170) Google Scholar, 7Fogazzi G.B. Saglimbeni L. Banfi G. et al.Urinary sediment features in proliferative and non-proliferative glomerular diseases.J Nephrol. 2005; 18: 703-710PubMed Google Scholar further studies are needed to compare the performance characteristics of the manual sediment examination when carried out by trained nephrologists to those of modern laboratory-based automated analyzers. All the authors declared no competing interests. This study was supported by NIH R01DK093574 (SSW). RP was supported by an ASN Ben J. Lipps Research Fellowship Grant. We thank all members of the Waikar Lab for their work on the Boston Kidney Biopsy Cohort. Download .pdf (.07 MB) Help with pdf files Supplementary Methods Download .pdf (.05 MB) Help with pdf files Figure S1Overview of analysis cohort. Download .pdf (.07 MB) Help with pdf files Table S1Primary clinicopathologic diagnoses of patients in the analysis cohort." @default.
• W2913215572 created "2019-02-21" @default.
• W2913215572 creator A5011689889 @default.
• W2913215572 creator A5064518665 @default.
• W2913215572 creator A5076486975 @default.
• W2913215572 date "2019-05-01" @default.
• W2913215572 modified "2023-09-24" @default.
• W2913215572 title "Performance of the Automated Urinalysis in Diagnosis of Proliferative Glomerulonephritis" @default.
• W2913215572 cites W1755481809 @default.
• W2913215572 cites W1973492162 @default.
• W2913215572 cites W1979754756 @default.
• W2913215572 cites W1984805222 @default.
• W2913215572 cites W1988276346 @default.
• W2913215572 cites W2167846507 @default.
• W2913215572 cites W2257335168 @default.
• W2913215572 doi "https://doi.org/10.1016/j.ekir.2019.02.001" @default.
• W2913215572 hasPubMedCentralId "https://www.ncbi.nlm.nih.gov/pmc/articles/6506710" @default.
• W2913215572 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/31080928" @default.
• W2913215572 hasPublicationYear "2019" @default.
• W2913215572 type Work @default.
• W2913215572 sameAs 2913215572 @default.
• W2913215572 citedByCount "3" @default.
• W2913215572 countsByYear W29132155722020 @default.
• W2913215572 countsByYear W29132155722021 @default.
• W2913215572 countsByYear W29132155722023 @default.
• W2913215572 crossrefType "journal-article" @default.
• W2913215572 hasAuthorship W2913215572A5011689889 @default.
• W2913215572 hasAuthorship W2913215572A5064518665 @default.
• W2913215572 hasAuthorship W2913215572A5076486975 @default.
• W2913215572 hasBestOaLocation W29132155721 @default.
• W2913215572 hasConcept C126322002 @default.
• W2913215572 hasConcept C2779225474 @default.
• W2913215572 hasConcept C2780026642 @default.
• W2913215572 hasConcept C2780091579 @default.
• W2913215572 hasConcept C2780368995 @default.
• W2913215572 hasConcept C71924100 @default.
• W2913215572 hasConceptScore W2913215572C126322002 @default.
• W2913215572 hasConceptScore W2913215572C2779225474 @default.
• W2913215572 hasConceptScore W2913215572C2780026642 @default.
• W2913215572 hasConceptScore W2913215572C2780091579 @default.
• W2913215572 hasConceptScore W2913215572C2780368995 @default.
• W2913215572 hasConceptScore W2913215572C71924100 @default.
• W2913215572 hasIssue "5" @default.
• W2913215572 hasLocation W29132155721 @default.
• W2913215572 hasLocation W29132155722 @default.
• W2913215572 hasLocation W29132155723 @default.
• W2913215572 hasLocation W29132155724 @default.
• W2913215572 hasOpenAccess W2913215572 @default.
• W2913215572 hasPrimaryLocation W29132155721 @default.
• W2913215572 hasRelatedWork W1999927105 @default.
• W2913215572 hasRelatedWork W2056021467 @default.
• W2913215572 hasRelatedWork W2147267039 @default.
• W2913215572 hasRelatedWork W2408636602 @default.
• W2913215572 hasRelatedWork W2413098285 @default.
• W2913215572 hasRelatedWork W2440796701 @default.
• W2913215572 hasRelatedWork W2761376155 @default.
• W2913215572 hasRelatedWork W2913215572 @default.
• W2913215572 hasRelatedWork W4300667862 @default.
• W2913215572 hasRelatedWork W4308150526 @default.
• W2913215572 hasVolume "4" @default.
• W2913215572 isParatext "false" @default.
• W2913215572 isRetracted "false" @default.
• W2913215572 magId "2913215572" @default.
• W2913215572 workType "article" @default.