ABSTRACT
Background
Morphology and morphometric evaluation of lesions beyond conventional parameters can inform the pathophysiology of chronic kidney disease (CKD). We sought to determine whether the occurrence of glomerulotubular neck stenoses associates with progressive CKD.
Methods
We evaluated the normal parenchyma from radical nephrectomies removed for tumor between 2000 and 2021 and analyzed cortex for stenoses of the glomerulotubular neck. Stenosis of the glomerulotubular neck is defined a focal narrowing for which the draining tubule has a greater diameter than at the neck. Progressive CKD was defined as dialysis, kidney transplantation, sustained estimated glomerular filtration rate (eGFR) <10 mL/min/1.73 m2 or sustained 40% decline from the post-nephrectomy eGFR. Each case of progressive CKD was age- and sex-matched to two controls without progressive CKD. Logistic regression models assessed the risk of progressive CKD with stenotic necks adjusting for other histological features, kidney function and CKD risk factors.
Results
There were 65 cases with a mean of 255 glomeruli and 130 controls with a mean of 329 glomeruli. Among both cases and controls, 5% of glomeruli showed visible glomerulotubular necks. The proportion of necks that were stenotic was higher in cases than controls (35% vs 11%, P < .0001). Stenotic necks associated with progressive CKD independent of other histologic and clinical characteristics.
Conclusion
Glomerulotubular neck stenosis is associated with development of progressive CKD.
Keywords: chronic kidney disease, fibrosis, glomerulotubular neck
Graphical Abstract
Graphical Abstract.
KEY LEARNING POINTS.
What was known:
Intrarenal scarring correlates with progressive loss of renal function; its histologic location may provide insight to its mechanistic role.
This study adds:
Stenoses at the glomerulotubular neck predict future loss of renal function.
The present study shows that the prevalence of stenotic necks is a better predictor of progressive chronic renal failure than that afforded by % glomerulosclerosis or interstitial fibrosis/tubular atrophy.
Potential impact:
Prevention or treatment of neck stenoses is likely to slow or stop the progression of chronic renal failure.
INTRODUCTION
The normal glomerulotubular neck is widely patent which ensures an unencumbered flow of glomerular filtrate into the proximal tubule. Narrowing or stenosis of the glomerulotubular neck can occur in animal models and in human chronic kidney disease (CKD) and could cause loss of kidney function by obstruction of flow. The occurrence of stenotic necks has been reported in radiation models of CKD in pigs, rats and non-human primates [1–3], in the unilateral ureteral obstruction (UUO) model of CKD, in a murine model of polycystic kidney disease and in a murine eNOS knockout model of CKD [4–6]. In the murine models, it has been shown that the stenotic necks progress to the formation of atubular
glomeruli, suggesting that glomerular neck stenosis leads to loss of filtering nephrons, and thus loss of kidney function.
Stenotic necks are also a feature of human kidney disease [7, 8]. Glomerulotubular neck abnormalities are well known in cystinosis, and are probably caused by cystine accumulation in proximal tubular epithelial cells, leading to the formation of a narrowed and elongated “swan neck” deformity of the initial proximal tubule [9] resulting in atubular glomeruli and loss of renal function. Scanning electron microscopy studies of the human tubular pole in non-neoplastic kidney of tumor nephrectomies and in explanted human allografts with severe chronic damage have shown significant narrowing of the tubular pole orifices [10]. Stenotic necks have been reported in a large cohort of cases with immunoglobulin A (IgA) nephropathy [8], and also shown in cases of focal segmental glomerulosclerosis using serial section analysis [11].
The presence of stenotic necks at the junction of glomerulus and tubule can provide a precise structural explanation for the known correlation of glomerular filtration rate to tubulo-interstitial disease [7]. In the porcine model, the average reduction in neck diameter was 60% in visible stenotic necks, a reduction that is sufficient to reduce flow and pressure across a stenosis [1]. If occurring in a sufficient number of nephrons, this would lower whole kidney function. However, the quantification of glomerular stenotic necks in human CKD and their effects in comparison with clinical and other histopathological features has not been done. Thus, we performed a nested case–control study within a cohort of patients who underwent a radical nephrectomy. Stenotic necks were characterized on large-wedge kidney sections to determine their association with subsequent progressive CKD.
MATERIALS AND METHODS
Study design
The study participants were from an expansion of a previously published cohort of Mayo Clinic Nephrectomy Registry patients in the Aging Kidney Anatomy study (see Supplemental Methods) [12–14]. These patients had undergone a radical nephrectomy for a renal tumor at the Mayo Clinic, Rochester, Minnesota between 2000 and 2021, with follow-up visits every 3–6 months for the first year and then every 6–12 months as part of a standardized prospective protocol to screen for cancer recurrence and for CKD. Follow-up ended on 1 September 2023. Progressive CKD was defined by a 40% or more decline in estimated glomerular filtration rate (eGFR) from its post-nephrectomy baseline and sustained for at least 3 months, an eGFR <10 mL/min/1.73 m2 and at least 5 mL/min/1.73 m2 below post-nephrectomy baseline that was sustained for at least 3 months, dialysis or kidney transplantation. Progressive CKD was identified before any cancer recurrence. Cases were defined as patients who developed CKD. Two age- and sex-matched controls without progressive CKD at the same follow-up time as each case were identified. These studies were approved by our Institutional Review Board (IRB) with a waiver of consent because data were limited to the medical record (IRB approval number 14-005 798).
Kidney function and risk factors
Pre-nephrectomy age, sex, body mass index (BMI), serum creatinine (corrected to standardized values if assayed pre-standardization), 24-h urine protein, hypertension (as documented in the medical record) and diabetes mellitus (as documented in the medical record), were obtained from the medical records. The eGFR was calculated using the 2021 CKD Epidemiology Collaboration serum creatinine–based equation [15]. When multiple post-nephrectomy serum creatinine levels were available, the one closest to but before 4 months after the nephrectomy was used for the baseline eGFR. A spot urine protein–osmolality ratio was used to estimate the 24-h urine protein excretion [16] and was routinely performed both pre-nephrectomy (up to 1 year before) and post-nephrectomy (up to 4 months after) nephrectomy. If a post-nephrectomy level was not obtained within the first 4 months, a pre-nephrectomy urine protein level was used for baseline.
Kidney microstructure
Each patient had a wedge section of kidney tissue away from the tumor that was scanned into a digital image for annotation (see Supplemental Methods). F.A. identified and scored each patent neck as normal or stenotic in a masked fashion. Each neck, patent or stenotic, was annotated with a colored mark to avoid duplicate counting. Quality control reviews of the majority of necks were done by A.D., in addition to consensus discussions between A.D. and E.P.C. There was a further random sampling of 30 neck images that were read independently and in a masked fashion by A.D. and I.W.G.
Histological evaluation of glomerulotubular necks
A normal glomerulotubular neck was defined by a lumen that had no narrowing or stenosis relative to its origin into the draining proximal tubular segment (Fig. 1A). Stenosis of the glomerulotubular neck was defined as a focal narrowing at or near the neck for which the draining tubule beyond the neck had a greater diameter than at the neck (Fig. 1B–E). Due to the nature of immersion fixation in this study, and inability to accurately measure the luminal diameter from the apex of opposing cells, the diameters of necks were measured by the tubular basement membrane profile at the site of the neck, similar to the previous measurements of tubular diameters [12]. All glomeruli with necks were further graded based on the absence (Fig. 1A–C) or presence of capsule thickening (Fig. 1D and E). Bowman's capsule thickening was defined as thickening with disruption and/or duplication with multilayering of periodic acid–Schiff positive material of the capsule. All stenotic necks were assessed for minimal adjacent interstitium (Fig. 1B), the presence of adjacent interstitium but no tubular atrophy (Fig. 1C and D) or the presence of adjacent interstitial fibrosis and tubular atrophy (IFTA) (Fig. 1E). Finally, hypocellular necks were defined as having flattened, scant or not visible epithelial cells lining the initial segment of the proximal tubule, similar to Bowman's capsule parietal cells and lacking a luminal brush border typical of proximal tubule cells (Fig. 1F).
Figure 1:
Representative images of glomerulotubular necks that are (A) normal (non-stenotic), (B) stenotic with no adjacent interstitium, (C) stenotic with adjacent interstitium but without capsule thickening, (D) stenotic with adjacent interstitium and with capsule thickening, (E) stenotic with adjacent interstitial fibrosis and with capsule thickening, and (F) hypocellular neck segment lined by flattened epithelium similar to Bowman's capsule parietal cells, without columnar proximal tubular cells or luminal brush border. Dotted lines represent the diameter of the glomerulotubular necks. These are from paraffin-embedded 3-µm sections stained with periodic acid–Schiff.
All necks were also graded by the glomerular location in cortex (superficial, middle and deep) as previously reported [12] (Fig. 2). The percentage of overall stenotic necks was obtained by dividing the number of stenotic necks by the total number of all observed necks (including hypocellular necks). The percentage of stenotic necks at each cortex depth was obtained by dividing the number of stenotic necks by the total number of all observed necks at each cortex depth. The percentage of hypocellular necks was obtained by dividing the number of hypocellular necks by the total number of all observed necks. We also calculated mean overall neck diameter and the mean diameter of stenotic necks at their narrowest width. By dividing the total number of glomeruli with any glomerulotubular neck with a total number of (non-sclerosed) glomeruli present on the specimen, we calculated the percentage of visible necks. Finally, we calculated the mean glomerular tuft areas of glomerular profiles with glomerulotubular necks that were not stenotic and of glomerular profiles with stenotic glomerulotubular necks. Histological measures of nephron size and nephrosclerosis were also used in analyses (see Supplemental Methods).
Figure 2:
An example wedge section with glomeruli that have visible glomerulotubular necks identified. From the glomeruli with stenotic necks (red) and glomeruli with both stenotic and normal (blue) necks we obtained the overall percentage of stenotic necks (e.g. 7 stenotic and 17 normal gives 29.2%). The yellow line represents the corticomedullary junction, and two brown lines represent the boundaries between the superficial/middle and middle/deep cortex. These boundaries were drawn at low magnification masked to the glomerular annotations using about 3–4 glomerular diameters to define the width of the superficial depth and the width of the deep depth. Then, from the numbers of stenotic necks and all visible necks at each depth, we calculated the percentage of stenotic necks for that region (33.3% for superficial, 27.8% for middle and 33.3% for deep region in this example). The middle region is thicker than the others because non-sclerosed glomeruli were consistent in size across most of the mid-cortex and smaller just near the capsule and near the medulla [12].
Statistical analyses
Clinical characteristics and kidney histology measures were compared between cases and controls using t-tests for normally distributed data, Wilcoxon rank-sum test for skewed data and chi-square tests for categorical data. The 95th percentile for glomerulotubular neck measures was estimated in controls as an upper reference limit. We used non-parametric Spearman's correlations to compare glomerulotubular neck measures with clinical characteristics, pre-nephrectomy eGFR, 24-h urine protein, glomerular volume and measures of nephrosclerosis [percentage of globally sclerotic glomeruli (%GSG), percentage of IFTA (%IFTA) and IFTA foci density]. We also assessed the correlation between the percentage of stenotic necks and glomerular volume by adjusting for %GSG and %IFTA using partial Spearman correlations [17]. A paired t-test was used to compare the mean glomerular tuft areas of glomeruli with non-stenotic versus stenotic glomerulotubular necks. A logistic regression model was used to assess the risk of progressive CKD with glomerulotubular neck measures. To allow meaningful comparisons of effect sizes (odds ratios), measures of the percentage of stenotic necks were standardized to per five percentage point increments and neck diameters were standardized to per standard deviation (per SD) increments. Models were unadjusted, adjusted for nephron size (glomerular volume and cortex per glomerulus), adjusted for nephrosclerosis (%GSG, %IFTA, IFTA foci density), or adjusted for nephrosclerosis and clinical characteristics (age, sex, BMI, hypertension, systolic and diastolic blood pressure, diabetes, post-nephrectomy baseline eGFR and proteinuria). A sensitivity analysis limited cases and controls to only those with <10% IFTA. Receiver operating characteristic (ROC) curves and area under the curve compared the strength of association of stenotic necks, %GSG and %IFTA with progressive CKD. All statistical analyses used BlueSky Statistics software version 10.3.1 (BlueSky Statistics LLC, Chicago, IL, USA) and R (RStudio) version 4.3.2.
RESULTS
Patient characteristics
Of the original 1845 radical nephrectomy patients, 1552 patients matched the inclusion/exclusion criteria. Of these, 65 developed progressive CKD (cases) and 130 were matched controls who did not develop progressive CKD (Fig. 3). There were over 16 000 glomeruli and over 800 stenotic glomerulotubular necks measured and studied in cases with progressive CKD, and over 42 000 glomeruli and over 2000 stenotic glomerulotubular necks studied in cases without progressive CKD. Baseline clinical and histology characteristics of cases and controls are shown in Table 1. Patients with progressive CKD had more prevalent hypertension and diabetes, lower pre-surgery eGFR, lower baseline post-surgery eGFR, higher urine protein, larger glomeruli, higher %GSG and higher %IFTA. Median (interquartile range) follow-up time was 4.1 (2.1, 8.8) years for cases and controls.
Figure 3:
Nested case–control study design.
Table 1:
Summary of baseline clinical and histology characteristics in cases with progressive CKD and controls.
| Characteristics | Progressive CKD (n = 65) | Controls (n = 130) | P-value |
|---|---|---|---|
| Clinical characteristics | |||
| Age, years | 66.6 (10.5) | 66.6 (10.5) | .99 |
| Male, n (%) | 43 (66) | 86 (66) | 1.00 |
| Race, n (%) | .15 | ||
| White | 58 (89.2) | 124 (95.4) | |
| Black | 1 (1.5) | 1 (0.8) | |
| Native American/Alaskan Native | 1 (1.5) | 1 (0.8) | |
| Other/Unknown | 5 (7.7) | 4 (3.1) | |
| BMI, kg/m2 | 31.8 (9.7) | 30.0 (6.0) | .10 |
| Hypertension, n (%) | 52 (80) | 85 (65) | .04 |
| Systolic blood pressure, mm Hg | 138.4 (21.0) | 132.3 (19.7) | .05 |
| Diastolic blood pressure, mm Hg | 75.6 (11.8) | 75.0 (12.8) | .73 |
| Diabetes, n (%) | 19 (29) | 16 (12) | .004 |
| Presurgery eGFR, mL/min/1.73 m2 | 62.6 (23.1) | 71.4 (18.3) | .004 |
| Baseline postsurgery eGFR, mL/min/1.73 m2 | 44.1 (19.6) | 50.0 (12.7) | .01 |
| Estimated 24-h urine protein, mg | 297 (150–875) | 183 (109–355) | .001 |
| Histology characteristics | |||
| Number of non-sclerosed glomeruli | 255 (118) | 329 (142) | .0003 |
| Number of all visible glomerulotubular necks | 12.5 (9.5) | 17.7 (11.6) | .002 |
| At superficial depth | 2.0 (2.2) | 3.5 (3.3) | .002 |
| At middle depth | 5.8 (5.6) | 8.4 (6.4) | .008 |
| At deep depth | 4.6 (3.2) | 5.9 (4.3) | .04 |
| With capsule thickening | 3.5 (4.1) | 3.9 (5.0) | .53 |
| Without capsule thickening | 9.0 (7.2) | 13.8 (9.5) | .0005 |
| Number of stenotic glomerulotubular necks | 4.0 (3.1) | 1.9 (2.0) | <.0001 |
| Without adjacent interstitium | 0.6 (1.1) | 0.3 (0.8) | .12 |
| With adjacent interstitium but no tubular atrophy | 2.2 (2.0) | 1.3 (1.5) | .0004 |
| With adjacent IFTA | 1.2 (1.9) | 0.3 (0.7) | <.0001 |
| Number of hypocellular glomerulotubular necks | 0.4 (0.7) | 0.6 (1.2) | <.0001 |
| Glomerular volume, mm3 | 0.0037 (0.0016) | 0.0028 (0.0010) | <.0001 |
| %GSG | 17.8 (14.7) | 8.8 (8.3) | <.0001 |
| %IFTA | 6.2 (6.6) | 3.2 (5.5) | .001 |
| IFTA foci density, per cm2 | 34.1 (24.5) | 27.6 (20.3) | .05 |
Data shown as mean (SD), median (25%–75%) or n (%).
Histological features
Approximately 5% of glomeruli showed visible glomerulotubular necks across the progressive CKD cases and controls (Table 2). However, cases were three times more likely than controls to have stenotic necks (Table 2 and Fig. 4). The percentage of stenotic necks in the cases compared with controls increased by depth. Furthermore, cases also had a higher percentage of stenotic necks in each subgroup defined by presence or absence of capsule thickening, or by presence or absence of surrounding interstitium or fibrosis. Capsule thickening was associated with stenotic necks (rs = 0.33, P < .0001), but the most common type of stenotic neck in cases and controls had no capsule thickening. There was no difference in the percentage of hypocellular necks between cases and controls. Overall, the 95th percentile (upper reference limit) for the occurrence of stenotic necks in controls was 29% (95% confidence interval 23% to 38%), but this threshold differed depending on the number of necks detected. Table 3 shows the number of stenotic necks that would be abnormal (>95th percentile) for a given number of necks detected on a tissue specimen. There were more cases than controls with an occurrence of stenotic necks of >29% or >20%. The mean diameters of stenotic necks did not differ between cases and controls, but non-stenotic necks were wider in cases than controls (Table 2). However, the mean diameter of all stenotic necks was significantly less than the mean diameter of all normal necks (mean 21 vs 33 μm, P < .0001, Fig. 5). Normal glomerulotubular necks did not have a mean diameter smaller than 20 μm, whereas stenotic necks did not have a mean diameter of more than 45 μm. The mean glomerular tuft area was not significantly different between glomerulotubular necks with versus without stenosis (20 511 vs 21 328 μm2, P = .10, paired t-test). Finally, the reading of 30 random neck images, stenotic or normal, by two masked observers (I.W.G., A.D.) showed agreement in 26 cases (87%, Cohen's kappa 0.72).
Table 2:
Comparison of glomerulotubular neck characteristics between progressive CKD cases and controls.
| Biopsy characteristic | CKD (n = 65) | Controls (n = 130) | P-value |
|---|---|---|---|
| Percent of non-sclerosed glomeruli with visible necks | 5% (3%) | 5% (2%) | .48 |
| Percent of normal necks with capsule thickening | 26.7% (29.6%) | 18.8% (19.5%) | .03 |
| Necks that are stenotic among | |||
| All necks | 35% (20%) | 11% (10%) | <.0001 |
| Necks in superficial cortex | 26% (37%) | 15% (24%) | .007 |
| Necks in middle cortex | 30% (28%) | 11% (14%) | <.0001 |
| Necks in deep cortex | 34% (29%) | 10% (19%) | <.0001 |
| Necks without capsule thickening | 22% (19%) | 6% (7%) | <.0001 |
| Necks with capsule thickening | 13% (16%) | 5% (7%) | <.0001 |
| Stenotic necks without adjacent interstitium among all necks | 5% (10%) | 2% (4%) | .002 |
| Stenotic necks with adjacent interstitium but no tubular atrophy among all necks | 19% (16%) | 7% (7%) | <.0001 |
| Stenotic necks with adjacent IFTA among all necks | 11% (16%) | 3% (6%) | <.0001 |
| Percent of necks that are hypocellular | 4% (7%) | 4% (9%) | .78 |
| Patients with percentage of stenotic necks above a threshold, n (%) | |||
| >20% | 51 (79%) | 18 (14%) | <.0001 |
| >29% (95th percentile in controls) | 43 (66%) | 6 (5%) | <.0001 |
| Mean neck diameter, μm | |||
| Overall | 31.0 (5.7) | 30.8 (5.3) | .81 |
| Non-stenotic only | 36.2 (7.0) | 32.1 (5.8) | <.0001 |
| Stenotic only | 20.4 (5.6) | 20.8 (5.4) | .64 |
Data shown as mean (SD) or n (%).
Figure 4:
Percent of stenotic necks is significantly higher in progressive CKD cases than controls (P < .0001). Graph shows values for each patient as well as mean ± 1 SD.
Table 3:
The 95th percentile for stenotic necks based on total glomerular necks available on biopsy for controls (n = 130).
| Number of glomerular necks | Number of stenotic necks >95th percentile | Number of controls |
|---|---|---|
| 1–2 | 1 or more | 4 |
| 3–4 | 2 or more | 9 |
| 5–8 | 3 or more | 13 |
| 9–12 | 4 or more | 22 |
| 13–16 | 5 or more | 23 |
| 17–24 | 6 or more | 32 |
| 25–32 | 7 or more | 12 |
| 33–64 | 9 or more | 15 |
The 95th percentile for number of stenotic necks was calculated using quantile regression in controls with number of glomerular necks as the model predictor.
Figure 5:
Distribution of non-stenotic versus stenotic necks measured at the glomerulotubular junction. Stenosis of the glomerulotubular neck was defined as a focal narrowing for which the draining tubule had a greater diameter than at the neck. As a group, stenotic necks are narrower than normal necks, without there being a specific cut-off minimum.
Correlations of stenotic necks, histological and clinical features
The correlations of glomerulotubular neck characteristics with clinical characteristics and histological measures of glomerular size and nephrosclerosis that we had identified in earlier studies are shown in Table 4 and Supplementary data, Table S1. A lesser occurrence of visible necks among non-sclerosed glomeruli was correlated with older age, hypertension, lower pre-nephrectomy eGFR, larger glomeruli, higher %GSG, higher %IFTA, and higher IFTA foci density. Capsule thickening in glomeruli with normal necks correlated with hypertension, higher proteinuria, larger glomeruli, higher %GSG, higher %IFTA, and higher IFTA foci density. A higher proportion of stenotic necks among all necks correlated with lower eGFR, higher proteinuria, larger glomerular volume, and higher %GSG and %IFTA. In an additional analysis using partial Spearman correlation, a higher proportion of stenotic necks still associated with larger glomerular volume after adjusting for %GSG and %IFTA (rs = 0.27, P = .0002). By subcategory, a high proportion of stenotic necks with capsule thickening or with adjacent IFTA correlated with higher proteinuria, larger glomeruli, higher %IFTA and higher IFTA foci density. However, a high proportion of stenotic necks without capsule thickening or without adjacent interstitium only correlated with younger age. Wider mean diameters of all necks (stenotic or normal) correlated with larger glomerular volumes, less %GSG, less %IFTA and less IFTA foci density.
Table 4:
Spearman's correlations of glomerulotubular neck characteristics with clinical and other histological characteristics among 195 combined cases and controls.
| Glomerulotubular neck characteristic | Age | HTN | eGFR | Estimated 24-h protein | Glom. volume | %GSG | %IFTA | IFTA foci density |
|---|---|---|---|---|---|---|---|---|
| % Non-sclerosed glomeruli with visible necks among all glomeruli | −0.16 (.02) | −0.19 (.007) | 0.18 (.01) | −0.07 (.36) | −0.15 (.03) | −0.29 (<.001) | −0.16 (.03) | −0.19 (.007) |
| % Capsule thickening among all normal necks | 0.03 (.69) | 0.15 (.04) | −0.07 (.31) | 0.22 (.002) | 0.30 (<.001) | 0.20 (.006) | 0.28 (<.001) | 0.15 (.03) |
| % Stenotic necks among all necks | −0.07 (.36) | 0.13 (.07) | −0.15 (.04) | 0.20 (.005) | 0.30 (<.001) | 0.25 (<.001) | 0.31 (<.001) | 0.13 (.08) |
| % Stenotic necks among necks without capsule thickening | −0.15 (.03) | 0.01 (.94) | 0.02 (.78) | 0.06 (.40) | 0.07 (.32) | 0.02 (.78) | 0.13 (.07) | 0.02 (.79) |
| % Stenotic necks among necks with capsule thickening | 0.04 (.61) | 0.14 (.05) | −0.13 (.07) | 0.24 (<.001) | 0.28 (.007) | 0.22 (.003) | 0.24 (<.001) | 0.14 (.06) |
| % Stenotic necks without adjacent interstitium among all necks | −0.20 (.004) | −0.04 (.61) | 0.06 (.43) | −0.05 (.50) | 0.03 (.71) | −0.05 (.52) | −0.16 (.03) | −0.16 (.03) |
| % Stenotic necks with adjacent interstitium but no tubular atrophy among all necks | −0.10 (.16) | 0.05 (.83) | −0.05 (.49) | 0.13 (.07) | 0.17 (.02) | 0.05 (.45) | 0.13 (.06) | 0.01 (.93) |
| % Stenotic necks with adjacent IFTA among all necks | 0.06 (.44) | 0.16 (.03) | −0.09 (.21) | 0.25 (<.001) | 0.18 (.01) | 0.27 (<.001) | 0.42 (<.001) | 0.26 (<.001) |
| % Hypocellular necks among all necks | 0.03 (.69) | −0.05 (.48) | −0.05 (.47) | 0.05 (.49) | −0.03 (.72) | 0.05 (.50) | 0.10 (.15) | 0.05 (.49) |
| Mean overall neck diameter | −0.13 (.06) | −0.05 (.47) | 0.10 (.15) | 0.00 (.96) | 0.23 (.001) | −0.20 (.004) | −0.22 (.002) | −0.23 (.001) |
| Mean stenotic neck diameter | −0.11 (.17) | 0.02 (.78) | −0.06 (.43) | 0.00 (.97) | 0.11 (.16) | −0.01 (.88) | −0.06 (.47) | −0.09 (.26) |
The coefficients are recorded, with P-values in parentheses.
Correlations with progressive CKD
Detection of visible necks was not associated with progressive CKD (Table 5). Capsule thickening of normal necks was associated with progressive CKD, but not after adjusted analysis. A higher proportion of stenotic necks was associated with progressive CKD, even after analysis adjusting for nephron size, nephrosclerosis and clinical characteristics. There were 69% to 77% higher odds of progressive CKD per each 5% increase in stenotic necks among all glomerulotubular necks (Table 5). This association was linear on spline plots (not shown). This association was also evident in unadjusted and adjusted analyses for neck subgroups defined by depth (superficial, middle and deep). The risk of progressive CKD with stenotic necks among glomeruli without capsule thickening was stronger than that in necks with capsule thickening, and this difference increased in adjusted analysis. In adjusted analysis, the risk of progressive CKD with stenotic necks was stronger among those without adjacent IFTA. Hypocellular necks, overall neck diameters and stenotic neck diameters were not associated with progressive CKD. In a sensitivity analysis limited to 168 patients (51 cases and 117 controls) with IFTA <10%, the overall findings were similar (Table 6). However, in this latter analysis, wider overall neck diameter associated with progressive CKD after adjusting for nephrosclerosis but not after adjusting for glomerular volume. Figure 6 shows the ROC curves for histological morphometric measures; the proportion of stenotic necks is superior to glomerular volume, %GSG and %IFTA as a classifier for predicting progressive CKD.
Table 5:
Risk of progressive CKD (vs control) with glomerulotubular neck characteristics.
| Unadjusted | Adjusted for nephron sizea | Adjusted for nephrosclerosisb | Further adjusted for clinical characteristicsc | |||||
|---|---|---|---|---|---|---|---|---|
| Biopsy characteristic | OR (95% CI) | P-value | OR (95% CI) | P-value | OR (95% CI) | P-value | OR (95% CI) | P-value |
| Percent of non-sclerosed glomeruli with visible necks, per 1% | 0.96 (0.85–1.07) | .48 | 0.99 (0.87–1.11) | .82 | 1.03 (0.91–1.16) | .61 | 1.08 (0.95–2.23) | .24 |
| Percent of normal necks with capsule thickening, per 5% | 1.07 (1.01–1.14) | .04 | 0.98 (0.91–1.06) | .69 | 1.02 (0.95–1.10) | .55 | 0.99 (0.91–1.07) | .79 |
| Percent stenotic necks, per 5%, among | ||||||||
| All necks | 1.73 (1.50–2.05) | <.0001 | 1.70 (1.46–2.02) | <.0001 | 1.69 (1.46–2.01) | <.0001 | 1.77 (1.49–2.17) | <.0001 |
| Superficial necks | 1.07 (1.02–1.12) | .01 | 1.08 (1.02–1.14) | .009 | 1.08 (1.02–1.14) | .008 | 1.10 (1.04–1.18) | .003 |
| Middle necks | 1.24 (1.15–1.36) | <.0001 | 1.23 (1.13–1.34) | <.0001 | 1.21 (1.12–1.33) | <.0001 | 1.22 (1.11–1.34) | <.0001 |
| Deep necks | 1.23 (1.15–1.34) | <.0001 | 1.22 (1.13–1.33) | <.0001 | 1.21 (1.13–1.33) | <.0001 | 1.21 (1.11–1.33) | <.0001 |
| Necks without capsule thickening | 1.70 (1.44–2.05) | <.0001 | 1.94 (1.59–2.46) | <.0001 | 1.80 (1.50–2.25) | <.0001 | 1.90 (1.54–2.46) | <.0001 |
| Necks with capsule thickening | 1.42 (1.21–1.69) | <.0001 | 1.27 (1.06–1.53) | .01 | 1.31 (1.11–1.58) | .003 | 1.31 (1.09–1.60) | .005 |
| Percent stenotic necks without adjacent interstitium among all necks, per 5% | 1.43 (1.13–1.88) | .006 | 1.59 (1.22–2.15) | .001 | 1.55 (1.18–2.11) | .003 | 1.60 (1.19–2.26) | .004 |
| Percent stenotic necks with adjacent interstitium but no tubular atrophy among all necks, per 5% | 1.62 (1.38–1.95) | <.0001 | 1.68 (1.39–2.09) | <.0001 | 1.64 (1.38–2.01) | <.0001 | 1.70 (1.40–2.12) | <.0001 |
| Percent stenotic necks with adjacent IFTA among all necks, per 5% | 1.48 (1.25–1.80) | <.0001 | 1.37 (1.14–1.67) | .001 | 1.37 (1.13–1.70) | .002 | 1.34 (1.09–1.67) | .006 |
| Percent of hypocellular necks, per 5% | 1.03 (0.85–1.23) | .78 | 1.02 (0.83–1.23) | .83 | 1.00 (0.81–1.20) | .98 | 0.99 (0.78–1.22) | .91 |
| Mean overall neck diameter, per SD | 1.04 (0.77–1.40) | .81 | 0.87 (0.61–1.22) | .43 | 1.21 (0.87–1.71) | .26 | 1.17 (0.84–1.67) | .36 |
| Mean stenotic neck diameter, per SD | 0.92 (0.66–1.28) | .64 | 0.77 (0.52–1.12) | .18 | 0.95 (0.65–1.36) | .77 | 1.95 (0.62–1.43) | .79 |
Glomerular volume, cortex per glomerulus.
%IFTA, IFTA foci density, %GSG.
Age, sex, BMI, hypertension, systolic blood pressure, diastolic blood pressure, diabetes mellitus, post-surgery baseline eGFR, proteinuria.
Table 6:
Risk of progressive CKD (vs control) with glomerulotubular neck characteristics limited to 168 patients with <10% IFTA.
| Unadjusted | Adjusted for nephron sizea | Adjusted for nephrosclerosisb | Further adjusted for clinical characteristicsc | |||||
|---|---|---|---|---|---|---|---|---|
| Biopsy characteristic | OR (95% CI) | P-value | OR (95% CI) | P-value | OR (95% CI) | P-value | OR (95% CI) | P-value |
| Percent of non-sclerosed glomeruli with visible necks, per 5% | 0.63 (0.30–1.25) | .20 | 0.69 (0.31–1.45) | .34 | 0.88 (0.40–1.85) | .74 | 1.14 (0.48–2.65) | .76 |
| Percent of normal necks with capsule thickening, per 5% | 1.04 (0.96–1.12) | .34 | 0.98 (0.87–1.05) | .37 | 1.02 (0.89–1.06) | .58 | 0.99 (0.87–1.06) | .46 |
| Percent stenotic necks, per 5%, among: | ||||||||
| All necks | 1.69 (1.45–2.01) | <.0001 | 1.69 (1.44–2.04) | <.0001 | 1.64 (1.40–1.98) | <.0001 | 1.78 (1.47–2.25) | <.0001 |
| Superficial necks | 1.08 (1.02–1.14) | .007 | 1.08 (1.02–1.15) | .01 | 1.10 (1.03–1.17) | .004 | 1.13 (1.05–1.23) | .001 |
| Middle necks | 1.23 (1.13–1.35) | <.0001 | 1.20 (1.10–1.32) | .0001 | 1.18 (1.08–1.31) | .0005 | 1.20 (1.09–1.34) | .0006 |
| Deep necks | 1.22 (1.13–1.33) | <.0001 | 1.22 (1.12–1.34) | <.0001 | 1.19 (1.10–1.31) | .0001 | 1.20 (1.10–1.34) | .0002 |
| Necks without capsule thickening | 1.72 (1.44–2.11) | <.0001 | 1.90 (1.55–2.44) | <.0001 | 1.80 (1.47–2.29) | <.0001 | 2.02 (1.57–2.75) | <.0001 |
| Necks with capsule thickening | 1.37 (1.14–1.68) | .001 | 1.25 (1.02–1.54) | .03 | 1.24 (1.01–1.54) | .04 | 1.31 (1.04–1.68) | .03 |
| Percent stenotic necks without adjacent interstitium among all necks, per 5% | 1.59 (1.22–2.16) | .002 | 1.72 (1.29–2.40) | .0007 | 1.68 (1.25–2.35) | .001 | 1.82 (1.28–2.76) | .002 |
| Percent stenotic necks with adjacent interstitium but no tubular atrophy among all necks, per 5% | 1.66 (1.38–2.05) | <.0001 | 1.72 (1.39–2.19) | <.0001 | 1.67 (1.37–2.11) | <.0001 | 1.84 (1.46–2.42) | <.0001 |
| Percent stenotic necks with adjacent IFTA among all necks, per 5% | 1.41 (1.16–1.76) | .001 | 1.33 (1.08–1.67) | .008 | 1.20 (0.96–1.53) | .12 | 1.14 (0.91–1.49) | .23 |
| Percent of hypocellular necks, per 5% | 1.06 (0.87–1.28) | .54 | 1.03 (0.83–1.26) | .77 | 0.96 (0.76–1.19) | .73 | 0.96 (0.72–1.23) | .77 |
| Mean overall neck diameter, per SD | 1.15 (0.82–1.63) | .42 | 0.90 (0.61–1.34) | .62 | 1.50 (1.02–2.25) | .04 | 1.58 (1.02–2.49) | .04 |
| Mean stenotic neck diameter, per SD | 0.93 (0.63–1.35) | .71 | 0.77 (0.50–1.17) | .22 | 0.93 (0.59–1.41) | .72 | 1.11 (0.68–1.81) | .67 |
Glomerular volume, cortex per glomerulus.
IFTA, IFTA foci density, %GSG.
Age, sex, BMI, hypertension, systolic blood pressure, diastolic blood pressure, diabetes mellitus, post-surgery baseline eGFR, proteinuria.
Figure 6:
ROC curves for histological pathology by morphometry as a classifier for subsequent progressive CKD among patients who underwent radical nephrectomy for tumor. The area under the curve (AUC) was 0.847 [95% confidence interval (CI) 0.776 to 0.917] for percentage of stenotic glomerulotubular necks, 0.715 (95% CI 0.638 to 0.793) for glomerular volume, 0.707 (95% CI 0.623 to 0.790) for %GSG and 0.695 (95% CI 0.617 to 0.772) for %IFTA.
DISCUSSION
These data show that the frequency of stenosis of glomerulotubular necks was strongly associated with the development of progressive CKD independent of other chronic histological changes, kidney function or common CKD risk factors. Our data show that 35% of visible glomerular necks are stenotic in progressive CKD cases versus 11% in the controls. This is in line with the features of a radiation model of CKD in which 38% and 12% of visible glomerular necks are stenotic in irradiated and non-irradiated rats, respectively [2]. Notably, stenotic necks without Bowman's capsule thickening or without adjacent IFTA associated with younger age and strongly predicted progressive CKD after adjusting for other histological and clinical factors associated with progressive CKD. This suggests that stenotic necks are an early nephron lesion that may precede more commonly recognized chronic changes detected on histology or by kidney function. The generalizability of this study to most progressive human CKD is likely, because of the known occurrence of stenotic necks in diverse forms of CKD [7, 8, 11, 18].
Our prior studies have shown the association of nephrosclerosis and enlarged glomeruli with subsequent progressive CKD. The present study adds to this literature by showing that glomerulotubular neck stenosis is a strong predictor for the subsequent development of progressive CKD, even after accounting for nephrosclerosis, nephron enlargement or clinical characteristics. That we found a better ROC for neck stenoses is probably in part because neck stenoses may be both fibrotic and non-fibrotic. That is, the feature of neck stenoses goes beyond the %IFTA as a correlate and predictor of CKD.
Our finding that neck stenoses add to the predictive power of histology relates also to the size of the specimens used. In contrast to kidney biopsies, the availability of wedge sections from radical nephrectomy specimens allows for the evaluation of hundreds of glomeruli in a single section, even though only 5% of glomeruli are sectioned at a level where the neck is visible. With percutaneous biopsies obtained in clinical practice, far fewer necks will be detected than with wedge sections, even for biopsies containing high numbers of glomeruli. However, if even a few necks are detected in a core biopsy with at least half having stenotic necks, this would be abnormal (>95th percentile in controls) and would identify patients at risk for progressive CKD.
The patterns of association between glomerulotubular necks and progressive CKD provides insight into potential mechanisms by which stenotic necks are pathological. As confirmed by the sensitivity analysis, stenotic necks without capsule thickening or without adjacent IFTA were strongly associated with progressive CKD in analyses adjusting for nephron size, nephrosclerosis and clinical characteristics. This could indicate that stenotic necks lacking capsule thickening or adjacent IFTA are an early histological lesion among functioning glomeruli that will eventually lead to capsule thickening and IFTA. In addition, other chronic changes may mask stenotic necks. Stenotic necks may represent a very early stage of a process that eventually leads to nephron destruction. With time, misdirected filtration may occur, resulting in thickening and multilayering of the Bowman's capsule, periglomerular fibrosis, atubular glomeruli, IFTA, and eventually global glomerulosclerosis of the involved nephron.
In a unilateral ureteral obstruction model in which there is formation of glomerulotubular neck stenosis and atubular glomeruli, the glomerulotubular neck region has cellular positivity for alpha smooth muscle actin, raising the possibility of contractility at the neck as the mechanism for non-fibrotic stenotic necks [4].
Indeed, proteinuria, enlarged glomeruli, higher %GSG, and higher %IFTA all associated with stenotic necks that had capsule thickening or adjacent IFTA but not stenotic necks lacking capsule thickening or lacking adjacent IFTA. If stenotic necks with capsule thickening are a more advanced lesion, this would explain their weaker association with progressive CKD. These glomeruli may have already lost significant filtration function and have less potential for further loss of function.
As glomerular hyperfiltration demands an increase in reabsorptive capacity of the proximal tubule, neck stenosis may be a “protective mechanism” that is not itself pathogenic but reflects nephrons that are in some way resisting stress on the proximal tubule from glomerular hyperfiltration. Glomerular hyperfiltration may both enlarge the glomerulus and stretch the neck via increase intraglomerular pressure (larger glomerulotubular neck diameter did correlate with larger glomerular volume) in a manner that counteracts stenotic necks when capsule thickening and IFTA begin to develop. Thus, glomerular hyperfiltration may increase neck diameters and proximal tubule diameters in response to neck stenosis. It does not appear that glomeruli without stenotic necks enlarge to compensate for glomeruli with stenotic necks as there was no detectable difference in glomerular tuft profile areas with versus without stenotic necks.
The mechanisms of formation of stenotic glomerulotubular necks could be internal or external to the neck [19]. An internal pathogenesis could occur because of glomerular ultrafiltrate causing tubular damage at the initial tubular segment, the glomerulotubular junction. This mechanism would be supported by a direct correlation between magnitude of proteinuria and the prevalence of stenotic necks. While we found a direct correlation of proteinuria with stenotic necks with capsule thickening, this correlation was not present with necks that lacked capsule thickening or stenotic necks that lacked adjacent IFTA. Likewise, damage from the glomerular ultrafiltrate to initial proximal tubule epithelium could lead to hypocellular necks, but hypocellular necks were not associated with progressive CKD.
An external pathogenesis would occur because of periglomerular fibrosis that extended around the glomerulus, leading to envelopment of the neck. The progression of glomerular to tubular injury was shown by careful morphological analysis of stenotic necks in two rat models of chronic renal failure [20]. An external and fibrotic pathogenesis of stenotic neck formation in human kidney disease is supported by the biopsy studies of Sato et al., who linked periglomerular fibrosis to the formation of stenotic necks in 340 biopsies of IgA nephropathy [8]. It is also supported by Bonsib's 1999 report in which narrowed necks appeared to be associated with fibrosis surrounding the glomerulotubular junction [11]. In the present studies, %IFTA directly and significantly correlated with the proportion of necks that were stenotic and with narrower overall neck diameter. However, there was no correlation between %IFTA and stenotic necks without capsule thickening. Further, stenotic necks lacking capsule thickening or lacking adjacent IFTA or interstitium were associated with progressive CKD. Thus, these data did not support periglomerular fibrosis as the only mechanism of progressive CKD. Thus, it is possible that IFTA is simply an eventual sequela of stenotic necks rather than causal mechanism as previously discussed.
Our studies support the presence of atubular glomeruli as a feature of CKD. Since we only had access to one tissue section, we could not directly detect atubular glomeruli in this study; nonetheless, there was indirect evidence of their existence. Measures of other chronic changes (hypertension, lower eGFR, enlarged glomeruli and nephrosclerosis) associated with lower detection of glomerulotubular necks among non-sclerosed glomeruli. Atubular glomeruli are also a chronic change and by definition these glomeruli lack a neck. Thus, their increased frequency would explain a lower frequency of detected necks when related chronic changes are present.
There are limitations to this study. The evidence of glomerulotubular neck patency or stenosis is ideally assessed in perfusion-fixed rather than the standard clinical formalin-fixed specimens used in this study. Because immersion fixation may decrease the sensitivity of detecting glomerulotubular neck stenosis, associations with neck stenosis may have been biased toward the null hypothesis. That we found the strong predictive value of neck stenoses in immersion-fixed specimens is also closer to real-life clinical specimens. In addition, in our radiation nephropathy study the kidneys were perfusion-fixed and the prevalence of stenotic necks in the controls and the CKD subjects of that study were close to the percentage prevalence of stenotic necks in the present, immersion-fixed, cases [2]. The present work used two-dimensional section and the true three-dimensional orientation of the proximal tubule and glomerular tubular neck to the plane of this two-dimensional section may add noise to the detection of neck stenosis. Two-dimensional artifact cannot, however, explain the significant excess of stenotic necks in subjects that developed progressive CKD compared with those that did not. Also, by the masked method used for this study, we show that stenotic necks were significantly smaller in diameter than ones read as normal.
We conclude that an increased prevalence of glomerulotubular neck stenoses predicts the progressive loss of renal function.
Supplementary Material
ACKNOWLEDGEMENTS
We thank Miloš Denić for assistance with computer algorithms for processing of biopsy annotations data.
Contributor Information
Eric P Cohen, Division of Nephrology, Department of Medicine, NYU Grossman School of Medicine, New York, NY, USA.
Aleksandar Denic, Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN, USA.
Fnu Aperna, Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN, USA.
Aidan F Mullan, Division of Biomedical Statistics and Informatics, Department of Medicine, Mayo Clinic, Rochester, MN, USA.
Laura Barisoni, Department of Pathology, Division of AI & Computational Pathology, and Department of Medicine, Division of Nephrology, Duke University, Durham, NC, USA; Department of Medicine, Division of Nephrology, Duke University, Durham, NC, USA.
Vidit Sharma, Division of Urology, Mayo Clinic, Rochester, MN, USA.
Ian W Gibson, Department of Pathology, University of Manitoba, Winnipeg, MB, Canada.
Andrew D Rule, Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN, USA; Division of Epidemiology, Mayo Clinic, Rochester, MN, USA.
FUNDING
This study was supported in part by the National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, PI A.D.R. (R01 DK090358), and by the resources and facilities at the Manhattan VAMC, New York City, NY.
AUTHORS’ CONTRIBUTIONS
Study design, data collection, data analysis, writing: E.P.C., A.D., A.D.R. Histology, data collection, analysis: A.D., F.A., I.W.G. Statistical analysis: A.D., A.F.M., A.D.R. Editing, revisions: E.P.C., L.B., V.S., I.W.G., A.D.R.
DATA AVAILABILITY STATEMENT
The data used for the present studies are from patient records. Their use requires IRB approval of the requestor and the Mayo Clinic IRB, in addition to a data use agreement for use in research. We are open to collaborations from investigators who submit a research proposal to us.
CONFLICT OF INTEREST STATEMENT
All authors have nothing to disclose and there are no conflicts of interest. Artificial Intelligence was not used.
REFERENCES
- 1.Cohen EP, Robbins ME, Whitehouse Eet al. Stenosis of the tubular neck: a possible mechanism for progressive renal failure. J Lab Clin Med 1997;129:567–73. 10.1016/S0022-2143(97)90011-1 [DOI] [PubMed] [Google Scholar]
- 2.Cohen EP, Regner K, Fish BLet al. Stenotic glomerulotubular necks in radiation nephropathy. J Pathol 2000;190:484–8. [DOI] [PubMed] [Google Scholar]
- 3.Parker GA, Cohen EP, Li Net al. Radiation nephropathy in a nonhuman primate model of partial-body irradiation with minimal bone marrow sparing-part 2: histopathology, mediators, and mechanisms. Health Phys 2019;116:409–25. 10.1097/HP.0000000000000935 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Forbes MS, Thornhill BA, Chevalier RL.. Proximal tubular injury and rapid formation of atubular glomeruli in mice with unilateral ureteral obstruction: a new look at an old model. Am J Physiol Renal Physiol 2011;301:110. 10.1152/ajprenal.00022.2011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Galarreta CI, Grantham JJ, Forbes MSet al. Tubular obstruction leads to progressive proximal tubular injury and atubular glomeruli in polycystic kidney disease. Am J Pathol 2014;184:1957–66. 10.1016/j.ajpath.2014.03.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Forbes MS, Thornhill BA, Park MHet al. Lack of endothelial nitric-oxide synthase leads to progressive focal renal injury. Am J Pathol 2007;170:87–99. 10.2353/ajpath.2007.060610 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Cohen EP. Glomerulotubular neck stenosis: the link between interstitial scarring and reduced GFR? Nephrol Dial Transplant 1998;13:2961–2. 10.1093/oxfordjournals.ndt.a027802 [DOI] [PubMed] [Google Scholar]
- 8.Sato M, Hotta O, Taguma Y.. Glomerulo-tubular junction stenosis as a factor contributing to glomerular obsolescence in IgA nephropathy. J Pathol 2002;197:14–19. 10.1002/path.1085 [DOI] [PubMed] [Google Scholar]
- 9.Larsen CP, Walker PD, Thoene JG.. The incidence of atubular glomeruli in nephropathic cystinosis renal biopsies. Mol Genet Metab 2010;101:417–20. 10.1016/j.ymgme.2010.08.015 [DOI] [PubMed] [Google Scholar]
- 10.Gibson I. Scanning Electron Microscopy of the Mammalian Renal Glomerulus in Health and Disease. Glasgow, Scotland: University of Glasgow, 1994. [Google Scholar]
- 11.Bonsib SM. Focal-segmental glomerulosclerosis. The relationship between tubular atrophy and segmental sclerosis. Am J Clin Pathol 1999;111:343–8. 10.1093/ajcp/111.3.343 [DOI] [PubMed] [Google Scholar]
- 12.Denic A, Ricaurte L, Lopez CLet al. Glomerular volume and glomerulosclerosis at different depths within the human kidney. J Am Soc Nephrol 2019;30:1471–80. 10.1681/ASN.2019020183 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Denic A, Elsherbiny H, Mullan AFet al. Larger nephron size and nephrosclerosis predict progressive CKD and mortality after radical nephrectomy for tumor and independent of kidney function. J Am Soc Nephrol 2020;31:2642–52. 10.1681/ASN.2020040449 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ricaurte Archila L, Denic A, Mullan AFet al. A higher foci density of interstitial fibrosis and tubular atrophy predicts progressive CKD after a radical nephrectomy for tumor. J Am Soc Nephrol 2021;32:2623–33. 10.1681/ASN.2021020267 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Inker LA, Eneanya ND, Coresh Jet al. New creatinine- and cystatin C-based equations to estimate GFR without race. N Engl J Med 2021;385:1737–49. 10.1056/NEJMoa2102953 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Wilson DM, Anderson RL.. Protein-osmolality ratio for the quantitative assessment of proteinuria from a random urinalysis sample. Am J Clin Pathol 1993;100:419–24. 10.1093/ajcp/100.4.419 [DOI] [PubMed] [Google Scholar]
- 17.Conover WJ. Practical Nonparametric Statistics, 3rd edn. New York: Wiley, 1999. [Google Scholar]
- 18.Oliver J. Architecture of the Kidney in Chronic Bright's Disease. New York: Paul Hoeber, Inc., 1939. [Google Scholar]
- 19.Cohen EP. Stenosis of the glomerulotubular neck. Nephron 2002;92:259–62. 10.1159/000063323 [DOI] [PubMed] [Google Scholar]
- 20.Kriz W, Hosser H, Hahnel Bet al. From segmental glomerulosclerosis to total nephron degeneration and interstitial fibrosis: a histopathological study in rat models and human glomerulopathies. Nephrol Dial Transplant 1998;13:2781–98. 10.1093/ndt/13.11.2781 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data used for the present studies are from patient records. Their use requires IRB approval of the requestor and the Mayo Clinic IRB, in addition to a data use agreement for use in research. We are open to collaborations from investigators who submit a research proposal to us.