Clinical and Pathology Findings Associate Consistently with Larger Glomerular Volume

Abstract

Background Glomerular volume increases when demand exceeds nephron supply, which may lead to glomerulosclerosis. It is unclear if determinants of glomerular volume are consistent between populations that differ by severity of comorbidities.

Methods We studied kidney biopsy specimens from living kidney donors (n=2453) and patients who underwent radical nephrectomy for a renal tumor (n=780). We scanned specimen sections into high-resolution digital images, manually traced glomerular profiles, and calculated mean glomerular volumes using the Weibel–Gomez stereologic formula (separately for nonsclerosed glomeruli and globally sclerosed glomeruli). We then assessed the relationship of glomerular volume with age, clinical characteristics, and nephrosclerosis on biopsy specimen.

Results Compared with kidney donors, patients with tumors were older and more frequently men, obese, diabetic, or hypertensive, had more glomerulosclerosis and interstitial fibrosis on biopsy specimen, and had 12% larger nonsclerosed glomeruli (P<0.001). In both populations, male sex, taller height, obesity, hypertension, and proteinuria associated with larger nonsclerosed glomeruli to a similar extent. In patients with tumors, diabetes, glomerulosclerosis >25%, and interstitial fibrosis >25% also associated with larger nonsclerosed glomeruli. Independent clinical predictors of larger nonsclerotic glomeruli were family history of ESRD, male sex, taller height, obesity, diabetes, and proteinuria. After adjustment for these characteristics, nonsclerotic glomerular volume did not differ between populations and was stable up to age 75 years, after which it decreased with age. Many of these findings were also evident with globally sclerotic glomerular volume.

Conclusions Characteristics associated with glomerular volume are consistent between patient populations with low and high levels of comorbidity.

Glomerular volume is an important structural determinant of overall kidney health. Glomerular size is known to be larger with lower nephron number,^1,2 male sex,³ family history of ESRD,³ obesity,^2,4,5 larger body surface area,⁶ higher serum uric acid level,³ and hypertension.^7,8 Larger glomeruli are evidence of hyperfiltration because they correlate with both higher single nephron GFR⁹ and increased single nephron filtration capacity.¹⁰ There has been a concern that progressive glomerulosclerosis and nephron loss from both aging and disease causes a compensatory enlargement of the remaining functional glomeruli.¹¹ Glomerulomegaly can itself lead to glomerulosclerosis,¹² with further nephron loss hypothesized to cause a vicious cycle that accelerates CKD progression.¹³

Despite the importance of glomerular volume, there has been disagreement regarding the determinants of glomerular volume, particularly with respect to age. Some studies have found that glomerular volume decreases with age,^6,14 others have found that glomerular volume increases with age,^12,15–17 and others have found no change with age.^1,18 Reasons for the discrepancy between prior studies may be due to several factors. First, many studies have not accounted for comorbidities that may influence glomerular size in the analysis. Second, analyses have not accounted for the different types of glomeruli that might skew the interpretation. Globally sclerotic glomeruli (GSG) are smaller than nonsclerotic glomeruli (NSG), and their volumes should be analyzed separately. Further, there are ischemic-appearing NSG (capsule thickening, pericapsular fibrosis, and capillary wrinkling) that may be smaller than nonischemic NSG and require additional analysis. Third, findings may differ if NSG volume is assessed by capsule volume rather than by tuft volume.

To clarify the relationship of glomerular volume with age and comorbidities, we studied two populations: living kidney donors and patients who underwent a radical nephrectomy for a renal tumor. Living kidney donors are selected for health, are relatively young, and have few comorbidities, whereas patients with kidney cancer are selected for disease, are relatively older, and have many more comorbidities. These two populations uniquely allow access to kidney parenchymal tissue without the presence of an overt nephropathy as the indication for a kidney biopsy. The first goal of our study was to characterize the relationship of glomerular volume with clinical and biopsy characteristics in these two disparate populations. The second goal was to determine if any differences in glomerular volume between these two populations could be explained by clinical characteristics and other biopsy findings.

Methods

Study Populations

We analyzed clinical and biopsy data of living kidney donors and patients who underwent a tumor nephrectomy in the Aging Kidney Anatomy study.² All data were obtained from the routine care of patients with institutional review board approval. The kidney donors at three sites (Mayo Clinic Minnesota, Mayo Clinic Arizona, and Cleveland Clinic Ohio) had undergone a biopsy of the donated kidney cortex during transplant surgery from 2000 to 2015. The patients who had a tumor nephrectomy at Mayo Clinic Minnesota had undergone a radical nephrectomy for a renal tumor from 2000 to 2012. Living kidney donors were selected on health as previously described.¹⁹ Acceptable criteria for donation varied by site and era, but in general included 24-hour urine albumin excretion <30 mg and a measured GFR normal for age. Mild hypertension in older donors and moderate obesity (body mass index <35 kg/m²) were allowed. Individuals with diabetes or evident cardiovascular disease were not accepted as donors. We only studied patients who underwent a complete unilateral nephrectomy for a renal tumor (usually renal cancer) with no metastatic lesions or positive lymph nodes at the time of surgery. Partial nephrectomy specimens were not studied because the small rim of renal parenchyma obtained adjacent to the tumor is often distorted by the tumor.

Comorbidities do not generally exclude patients from radical nephrectomy unless severe enough to make the surgical risks outweigh the benefits of treating cancer. Six patients with renal tumor were excluded for specific renal diseases: one each for amyloidosis, ESRD histology, severe chronic lymphocytic pyelonephritis, crescentic GN, severe nodular glomerulosclerosis, and diffuse mesangial sclerosis with FSGS. Diabetic nephropathy occurred with mild to moderate diffuse mesangial expansion in 28 patients and nodular sclerosis in 12 patients with renal tumor; these were included in NSG volume estimates. Segmental sclerosis that was minimal (involving one to four glomeruli) and occurred in 29 patients with renal tumor; these were also included in NSG volume estimates.

Kidney Biopsy Findings

For kidney donors, an 18-gauge needle core biopsy of the cortex was obtained at the time of transplantation, fixed in formalin, and embedded in paraffin. For patients with renal tumor, the stored formalin-fixed whole kidney specimen was retrieved to obtain a large wedge section distal to the tumor, which was then embedded in paraffin. For both populations, a 3-µm thick section was cut from the paraffin-embedded tissue block, stained for periodic acid–Schiff, and scanned into high-resolution digital images (Aperio XT system scanner; Leica Microsystems, Inc., Buffalo Grove, IL; http:/www.aperio.com).

For the kidney donor needle core biopsies sections, we required at least 2 mm² of cortex area present and at least four glomeruli. For patients with renal tumor, the wedge sections had far more cortex than needed for morphometric analysis. Thus, two wedge-shaped regions of interest, between 14 and 15 mm² in area each, were drawn at low magnification to “virtually” biopsy the wedge section at both the most clockwise and the most counterclockwise regions of cortex (Supplemental Figure 1). The mean of the calculated glomerular volumes between both regions of interest was used for all analyses.

Biopsy images were analyzed at Mayo Clinic Minnesota masked to age and other clinical characteristics to minimize bias. The scanned digital images were magnified with ImageScope software (version 12.2.2.5015 Aperio) onto a touchscreen tablet and each glomerular tuft was manually traced with a pen (Supplemental Figure 2, A and B). In the kidney donors, NSG capsules were also traced. The NSG tuft, NSG capsule, and GSG volumes (millimeters cubed) were calculated using the Weibel–Gomez stereological models²⁰ as previously described.² The GSG volume was calculated only in cases with at least two GSG profiles. Nephrosclerosis measures included percentages of ischemic-appearing NSG, GSG, interstitial fibrosis, and luminal stenosis. Ischemic-appearing NSG were identified by periglomerular fibrosis, capsule thickening, and a wrinkled (deflated) tuft (Supplemental Figure 2C). The percentage of NSG that were ischemic-appearing and the percentage of all glomeruli that were GSG were determined on the basis of their two-dimensional profiles,¹⁹ consistent with common clinical practice. Interstitial fibrosis (with tubular atrophy) was characterized by a renal pathologist as a percentage of the cortex (0%, 1%–5%, 6%–10%, 11%–25%, 26%–50%, and >50%). Luminal stenosis (proportion of artery lumen that was intimal thickening) was determined from the most orthogonal small- to medium-sized artery as previously described,¹⁹ and was then dichotomized at >50%.

Kidney Function and Risk Factors

The preoperative evaluations of both kidney donors and patients with renal tumor were reviewed to obtain body mass index, serum creatinine (corrected to standardized values if assayed prestandardization), 24-hour urine protein, diabetes mellitus status (past diagnosis from the patient medical history), and hypertension status. Hypertension was defined by an office BP >140/90 mm Hg or use of antihypertensive medication to lower BP. The GFR was estimated using the CKD Epidemiology Collaboration equation.²¹ The 24-hour urine protein excretion was a timed collection in kidney donors (limited to Mayo Clinic sites) and was estimated from a spot urine protein-to-osmolality ratio in patients with renal tumor.²² Family history of ESRD was defined by being a living related kidney donor. In patients with renal tumor, the tumor volume was estimated from three orthogonal linear dimensions on the preoperative computed tomography or magnetic resonance imaging scan using the ellipsoid formula (V=(1/6)π×a×b×c).²³

Statistical Analyses

Clinical characteristics and biopsy findings were compared between kidney donors and patients with renal tumor with the t test and chi-squared test. Analyses were performed separately with NSG tuft volume, NSG capsule volume (kidney donors only), and GSG volume. Glomerular volumes were log-transformed for homoscedasticity. Smoother spline fits showed any nonlinear trends of glomerular volume with age.²⁴ The mean NSG tuft volume was compared between nonischemic-appearing and ischemic-appearing NSG tufts. The percentage of NSG tufts that were ischemic-appearing was compared by age group (≤60, 61–75, and >75 years). Glomerular volumes were regressed on each clinical characteristic. A test for interaction was used to determine if associations differed between kidney donors and patients with renal tumor and if associations differed between the three kidney donor sites. We also performed a sensitivity analysis limited to biopsies with at least ten NSG profiles.²⁵ Multivariable regression models identified clinical characteristics that independently associated with glomerular volume. Analyses were performed with and without inclusion of urine protein (due to missing data). We assessed whether there was any difference in glomerular volume between kidney donors and tumor patients after adjustment for differences in clinical characteristics. We also assessed whether NSG tuft volume and GSG volume associated with nephrosclerosis on biopsy (percentages of ischemic-appearing NSG, GSG, interstitial fibrosis, and artery luminal stenosis). All statistical analyses were performed using JMP, version 13.0 (SAS Institute, Cary, NC, www.jmp.com).

Results

Kidney Donors Compared with Patients with Renal Tumor

We reviewed the scanned implantation renal biopsy slides of 2677 living kidney donors. After excluding 224 biopsies with insufficient cortex (fewer than four NSG or cortical area <2 mm²), 2453 kidney donors were studied (1832 from Mayo Clinic Minnesota, 318 from Mayo Clinic Arizona, and 303 from Cleveland Clinic Ohio). We also studied 780 tumor nephrectomy patients. The demographic, clinical, and biopsy characteristics of kidney donors and patients with renal tumor are presented in Table 1. Compared with kidney donors, patients with renal tumor were older, more frequently men, obese, diabetic, and hypertensive, with higher 24-hour urine protein and lower eGFR. Patients with renal tumor also had larger NSG tuft volume, larger GSG volume, more ischemic-appearing NSG, higher percentage of GSG, more interstitial fibrosis, and more luminal stenosis than living kidney donors. Among patients with renal tumor, the median tumor volume was 106 cm³ (interquartile range, 40–250 cm³).

Table 1.

Clinical characteristics and biopsy findings of living kidney donors and patients who underwent a tumor nephrectomy

Characteristic	Kidney Donors (n=2453)	Patients with Renal Tumor (n=780)	P Value
Demographic
Age, yr	43.7 (11.8)	63.6 (12.2)	<0.001
≤60	2229 (90.9%)	306 (39.2%)
61–75	221 (9.0%)	344 (44.1%)
>75	3 (0.1%)	130 (16.7%)
Men, %	1019 (41.5%)	499 (64.0%)	<0.001
Height, cm	170.7 (9.5)	172.4 (9.9)	<0.001
Risk factors
Diabetes mellitus, %	0 (0%)	117 (15.0%)	<0.001
Hypertension, %	238 (9.7%)	505 (64.7%)	<0.001
Known family history of ESRD, %	1259 (51.3%)	0 (0.0%)	<0.001
Body mass index, kg/m2	27.5 (4.8)	30.0 (6.5)	<0.001
Kidney function
eGFR, ml/min per 1.73 m2	92.4 (16.2)	70.9 (19.8)	<0.001
<60 ml/min per 1.73 m2	24 (1.0%)	211 (27.1%)	<0.001
24-h urine protein, mga	46.3 (35.8)	584 (2157)	<0.001
>500 mg	0 (0.0%)	103 (18.9%)	<0.001
Biopsy measures
Biopsy area, mm2	6.7 (2.9)	28.9 (0.4)	<0.001
NSG tuft volume, mm3	0.0026 (0.0010)	0.0029 (0.0012)	<0.001
Number of NSG on biopsy	17.8 (9.9)	75.8 (23.5)	<0.001
GSG volume, mm3b	0.0008 (0.0003)	0.0011 (0.0005)	<0.001
Number of GSG on biopsy	0.6 (1.0)	8.9 (15.3)	<0.001
Ischemic-appearing NSG, %			<0.001
0–5	2323 (94.7%)	642 (82.3%)
6–10	88 (3.6%)	89 (11.4%)
>10	42 (1.7%)	49 (6.3%)
GSG of all glomeruli, %			<0.001
0–10	2184 (89.0%)	515 (66.0%)
11–25	230 (9.4%)	199 (25.5%)
>25	39 (1.6%)	66 (8.5%)
Interstitial fibrosis, %			<0.001
0–10	2436 (99.3%)	717 (91.9%)
11–25	17 (0.7%)	22 (2.8%)
>25	0 (0%)	41 (5.3%)
Luminal stenosis, %c			<0.001
≤50	1709 (78.3%)	241 (30.9%)
>50	473 (21.7%)	539 (69.1%)

Data are presented as mean (SD) or n %.

^an=2003 for donors, n=545 for patients with renal tumor.

^bAnalysis was limited to 268 donors and 632 patients with renal tumor with two or more GSG profiles, although means were the same for the 763 donors and 709 patients with renal tumor with one or more GSG profile.

^cn=2182 donors (89%) and all patients with renal tumor had an artery present on the biopsy specimen.

Age Trends in Glomerular Volume

The NSG tuft volume did not change with age in the kidney donors (−0.1% per 10 years; P=0.84), whereas NSG tuft volume declined with older age in patients with renal tumor (−2.6% per 10 years; P=0.01); this decline was more evident after age 60 years (Figure 1). Compared with age ≤60 years, the percentage of ischemic-appearing NSG was higher for ages 61 years and older in kidney donors (Figure 2A) and patients with renal tumor (Figure 2B). Ischemic-appearing NSG tuft volumes were smaller than the nonischemic NSG tuft volumes (Figure 2C). This had trivial effect on the overall NSG tuft volume because ischemic-appearing NSG tufts were a small fraction. There was no evident change in GSG volume with age in either population (Supplemental Figure 3) or in the NSG capsule volume with age among the living kidney donors (Supplemental Figure 4).

Figure 1.

NSG tuft volume in living kidney donors and patients who underwent a tumor nephrectomy does not increase with age. The log NSG volume for living kidney donors and patients who underwent a tumor nephrectomy was plotted against age. Although NSG volume remained stable with older age in living kidney donors (blue smooth-fit curve), there was a trend toward smaller NSG volume in patients with renal tumor as they got older (green smooth-fit curve). With a linear fit, NSG tuft volume declined −0.1% per 10 years in kidney donors (P=0.84), and −2.6% per 10 years in patients with renal tumor (P=0.01).

Figure 2.

Ischemic-appearing NSG were smaller and more common with older age. Percentage of ischemic-appearing NSG is significantly higher in (A) kidney donors and (B) patients with renal tumor older than 60 years of age (P<0.001 and P=0.007, respectively), and (C) among 414 patients with at least one ischemic-appearing NSG, ischemic-appearing NSG are significantly smaller than nonischemic-appearing NSG. Error bars represent 95% confidence intervals.

Clinical Characteristics Associated with NSG Volume

Larger NSG tuft volume was associated with male sex, taller height, obesity, hypertension, and higher 24-hour urine protein in both kidney donors and patients with renal tumor, and the magnitude of these associations was similar between both populations (Table 2). Among persons up to age 75 years, the NSG tuft volume did not change with age in either population (0.0% per 10 years in living donors; P=0.99 and −1.6% per 10 years in patients with renal tumor; P=0.24). Among patients with renal tumor aged >75 years, NSG tuft volume declined with age (−21.1% per 10 years; P=0.009). In patients with renal tumor, larger NSG tuft volume was also associated with diabetes but not with tumor size. eGFR did not associate with NSG tuft volume in either population. Associations with NSG tuft volume did not change substantively in an analysis limited to those with at least ten glomeruli (Supplemental Table 1). Associations with NSG capsule volume were not substantively different from associations with NSG tuft volume (Supplemental Table 2). Associations of NSG tuft volume with clinical characteristics did not differ between the three kidney donor sites except for family history of ESRD (P=0.04 for test for interaction; Supplemental Table 3).

Table 2.

Characteristics associated with NSG tuft volume in living kidney donors compared with patients who underwent a tumor nephrectomy

Clinical Characteristic	Unadjusted				Test for Interaction
	Kidney Donors (n=2453)		Patients with Renal Tumor (n=780)
	% Difference	P Value	% Difference	P Value
Age per 10 yr (up to age 75 yr)	0.0	0.99	−1.6	0.24	0.30
Age per 10 yr (>75 yr)a			−21.1	0.009
Men	13.6	<0.001	19.3	<0.001	0.12
Height per SD	5.9	<0.001	7.6	<0.001	0.27
BMI per SD	10.8	<0.001	9.3	<0.001	0.30
Hypertension	8.6	0.001	10.7	<0.001	0.61
Diabetes			22.6	<0.001
eGFR per SD	−1.4	0.12	0.7	0.58	0.18
Family history of ESRD	7.4	<0.001
24-h urine protein per doubling, mgb	2.4	<0.001	4.1	<0.001	0.11
Tumor volume per doubling			0.9	0.11

SD: 9.6 cm for height, 5.3 kg/m² for body mass index (BMI), and 19.5 ml/min for eGFR.

^aFor patients with renal tumor, the change in the % difference for age up to 75 years compared with age >75 years was significant (P=0.04).

^bTotal N does not add up to 2453 or 780 because of missing data (n=2003 for donors, n=545 for patients with renal tumor).

Differences in NSG Volume between Kidney Donors and Patients with Renal Tumor

The NSG tuft volume was larger in patients with renal tumor than kidney donors (12.7%; P<0.001), but this association was no longer evident after adjustment for differences in clinical characteristics between the two populations (1.8%; P=0.49). In the combined sample up to age 75 years, NSG tuft volume increased with age (1.8%; P<0.001) due to the larger NSG tuft volumes of the older patients with renal tumor compared with the younger kidney donors (Figure 1). Male sex, taller height, obesity, diabetes mellitus, family history of ESRD, and urine protein were also independently associated with larger NSG tuft volume (Table 3). Hypertension was also independently associated with larger NSG tuft volume if urine protein was not included. The NSG tuft volume did not change with age after adjusting for these clinical characteristics that differed between older patients with renal tumor and younger kidney donors (0.1%; P=0.93).

Table 3.

Independent clinical characteristics associated with NGS tuft volume in both kidney donors and patients with renal tumor up to age 75 years

Clinical Characteristic	Multivariable Adjusteda
	Kidney Donors and Patients with Renal Tumor (n=2447)
	% Difference	P Value
Age per 10 yr (up to age 75 yr)	0.1	0.93
Men	9.5	<0.001
Height per SD	2.1	0.04
BMI per SD	8.4	<0.001
Hypertension	3.8	0.09
Diabetes	13.6	0.008
eGFR per SD	1.9	0.06
Family history of ESRD	7.3	<0.001
24-h urine protein per doubling, mgb	2.5	<0.001
Patients with renal tumor versus kidney donors	1.8	0.49

The mean SD across both populations was used (9.6 cm for height, 5.3 kg/m² for body mass index [BMI], 19.5 ml/min for eGFR).

^aAdjusted for each other characteristic.

^bIf 24-hour urine protein was excluded, findings were similar except hypertension was also associated with larger NSG tuft volume (6.2% larger; P=0.002).

Biopsy Findings Associated with GSG Volume

There were 268 kidney donors and 632 patients with renal tumor with at least two GSG profiles on their biopsy specimen section. GSG volume correlated with NSG volume in kidney donors (r_s=0.25; P<0.001) and in patients with renal tumor (r_s=0.39; P<0.001). Male sex was associated with larger GSG volume in both living donors and patients with renal tumor (Table 4). Among patients with renal tumor, GSG volume was also associated with taller height, obesity, hypertension, diabetes, lower eGFR, and higher proteinuria. Only proteinuria significantly differed in its association with GSG volume between kidney donors and patients with renal tumor, but proteinuria levels were also markedly higher in patients with renal tumor than in kidney donors (Table 1). In the combined sample, patients with renal tumor had larger GSG volumes than kidney donors (38.3%; P<0.001), even after adjusting for other characteristics (17.7%; P<0.001). Male sex, diabetes, and lower eGFR were also independently associated with larger GSG (Table 5).

Table 4.

Characteristics associated with GSG volume in living kidney donors compared with patients who underwent a tumor nephrectomy

Clinical Characteristic	Unadjusted				Test for Interaction
	Kidney Donors (n=268)		Patients with Renal Tumor (n=632)
	% Difference	P Value	% Difference	P Value
Age per 10 yr (up to age 75 yr)	1.9	0.45	3.2	0.18	0.70
Age per 10 yr (>75 yr)a			−14.8	0.09
Men	19.6	0.001	17.3	<0.001	0.78
Height per SD	4.6	0.11	5.9	0.001	0.70
BMI per SD	1.7	0.60	6.7	<0.001	0.19
Hypertension	−3.0	0.64	9.1	0.03	0.13
Diabetes			40.2	<0.001
eGFR per SD	−4.2	0.20	−5.9	0.001	0.66
Family history of ESRD	6.1	0.27
24-h urine protein per doubling, mgb	−1.3	0.46	5.9	<0.001	<0.001
Tumor volume per doubling			1.1	0.16

Analysis was limited to kidney donors and patients with renal tumor with at least two GSG profiles to analyze. SD: 9.6 cm for height, 5.3 kg/m² for body mass index (BMI), and 19.5 ml/min for eGFR.

^aBecause of those older than 75 years (n=2), analysis was not performed for the donors.

^bTotal N does not add up to 268 or 632 because of missing urine protein data (n=225 for kidney donors, n=439 for patients with renal tumor).

Table 5.

Independent clinical characteristics associated with GSG volume in both kidney donors and patients with renal tumor up to age 75 years

Clinical Characteristic	Multivariable Adjusteda
	Kidney Donors and Patients with Renal Tumor (n=567)
	% Difference	P Value
Age per 10 yr (up to age 75 yr)	0.0	0.99
Men	12.7	0.04
Height per SD	−0.1	0.98
BMI per SD	2.6	0.15
Hypertension	2.6	0.57
Diabetes	30.0	0.001
eGFR per SD	−6.6	0.003
Family history of ESRD	9.0	0.17
24-h urine protein per doubling, mgb	1.6	0.17
Patients with renal tumor versus donors	17.7	0.007

Analysis was limited to kidney donors and patients with renal tumor with at least two GSG profiles to analyze. The mean SD across both populations was used (9.6 cm for height, 5.3 kg/m² for body mass index [BMI], 19.5 ml/min for eGFR).

^aAdjusted for each other characteristic.

Glomerular Volume and Nephrosclerosis

We then assessed whether glomerular volume was associated with nephrosclerosis on kidney biopsy specimens. Adjusting for clinical characteristics, larger NSG tuft volume was associated with >25% GSG, whereas conversely, smaller NSG tuft volume was associated with >25% interstitial fibrosis among patients with renal tumor (Supplemental Table 4). Consistent with this findings, the percentage of smaller ischemic-appearing NSG correlated with interstitial fibrosis in both patients with renal tumor (r_s=0.46; P<0.001) and kidney donors (r_s=0.23; P<0.001). Larger GSG volume was associated with ischemic-appearing NSG and >10% GSG after adjusting for clinical characteristics in patients with renal tumor (Supplemental Table 5). There was no evidence these associations differed in kidney donors from patients with renal tumor (P>0.05 for all tests for interaction).

Discussion

In a large sample of living kidney donors and patients who underwent tumor nephrectomy, we found that clinical and biopsy factors that associated with NSG volume were generally consistent between groups. After accounting for differences in demographics and comorbidities, there was no detected difference in NSG volume between these two populations. Larger NSG volume was associated with male sex, tall height, obesity, hypertension, and proteinuria, and these associations were similar in both populations to the extent they overlapped. Differences reflected characteristics present in one population but absent or rare in the other population. In kidney donors, known family history of ESRD (living related donor) was associated with larger NSG volume. In patients with renal tumor, diabetes mellitus, interstitial fibrosis >25%, or glomerulosclerosis >25% was associated with larger NSG volume, and age >75 years was associated with smaller NSG volume. It appears that a consistent and reproducible set of clinical and biopsy characteristics influences NSG volume.

The stable NSG volume until age 75 years found in both kidney donors and patients with renal tumor is consistent with stable NSG basement membrane width,²⁶ percentage mesangial volume of NSG volume,²⁶ single nephron GFR,⁹ and glomerular filtration capacity¹⁰ in kidney donors with age. Despite humans losing half of their nephrons from onset of adulthood to early seventies, compensatory hypertrophy of remaining glomeruli is not seen.² This study further documents that NSG volume does not increase with age in the elderly (>75 years), but actually decreases. Why NSG volume decreased after age after 75 years is unclear. Although ischemic-appearing NSG are about half the volume of nonischemic-appearing NSG, ischemic-appearing NSG were still too rare to explain overall trends in NSG volume.

There were several clinical characteristics that were associated with larger NSG volume in a consistent manner between both kidney donors and patients with renal tumor. These associations were also consistent whether NSG volume was assessed by tuft or by capsule. Men had larger NSG volumes than women. The thicker glomerular basement membranes observed in men compared with women²⁶ could theoretically protect against injury from the increased glomerular wall tension with larger glomeruli.²⁷ However, larger glomeruli could still play a role in the higher risk of CKD in men compared with women.²⁸ Taller height and obesity likely increase the demand for glomerular function from increased metabolic waste generated with a larger body size. Systemic hypertension may directly contribute to glomerular hypertension, with subsequent glomerular enlargement.^29–31 Diabetes and obesity may cause glomerular hyperfiltration and glomerulomegaly from reduced afferent arteriole resistance via macula densa feedback with increased proximal tubule reabsorption of glucose and sodium and from other vasoactive factors.³² Proteinuria results from glomerular hypertrophy due to disorganized glomerular structure that is unable to efficiently prevent protein leaking.¹²

To our knowledge, this study is the first to study associations with GSG volume in addition to NSG volume. We found some of the characteristics that associate with larger NSG volume also associate with larger GSG volume, specifically, male sex and diabetes. Several investigators have suggested that enlargement of NSG precedes glomerular collapse, leading to glomerulosclerosis.^12,31 However, substantial glomerulosclerosis without NSG enlargement also occurs with aging alone.³³ Thus, there are likely two pathways that lead to glomerulosclerosis. First, an “ischemic pathway” characterized by NSG that collapse into smaller GSG as seen with aging. Second, a “glomerular hypertrophy pathway” characterized by both inherent and acquired factors that result in glomerular enlargement, podocyte detachment, tuft collapse, and GSG that are larger.¹² Genetic variants (specifically APOL1) have been linked to larger glomeruli and nephron loss.³⁴ The glomerular hypertrophy pathway may dominate parenchymal injury beyond aging alone (“real CKD”) as larger GSG associated with the severity of nephrosclerosis on biopsy specimen. The reason for larger GSG volumes in patients with renal tumor compared with living kidney donors is unclear, but may be related to other unmeasured comorbidities, or differences in the severity of comorbidities.

A conceptual model for the factors that influence glomerular volume is shown in Figure 3. With aging alone, there is no change in NSG volume until age 75 years, when NSG volume decreases. With glomerulosclerosis from aging alone, NSG collapse into small GSG. The added presence of intrinsic risk factors (family history of ESRD, height, and male sex) and acquired risk factors (obesity, diabetes, and hypertension) lead to hypertrophy of NSG. These larger NSG can leak protein and collapse into larger GSG, resulting in a GFR decline beyond that expected with aging. This nephron loss with larger GSG both increases demand on the remaining NSG³⁵ and leads to atrophy of the attached tubule with increased interstitial fibrosis. Interestingly, we found interstitial fibrosis and tubular atrophy to associate with smaller rather than larger NSG. Interstitial fibrosis and tubular atrophy may result in ischemic effects on adjacent NSG, preventing compensatory enlargement (Supplemental Figure 5).

Figure 3.

Conceptual model for glomerular volume that includes both age-related and risk factor related pathways. Aging alone does not lead to change in NSG volume until 75 years of age. Beyond 75 years of age, NSG volume shrinks. Intrinsic risk factors (family history of ESRD, male sex, and taller height), and acquired risk factors (obesity, diabetes, and hypertension) lead to larger NSG volume. As a consequence of glomerulomegaly, there is more glomerular leakage and proteinuria. Larger NSG degenerate into larger GSG with nephron loss and interstitial fibrosis. This feedbacks into glomerulomegaly of remaining NSG and larger GSG become the dominant contributors to glomerulosclerosis (%GSG). However, interstitial fibrosis also impairs compensatory enlargement of adjacent NSG possibly via ischemia.

There are several potential limitations in this study. The Weibel–Gomez²⁰ stereologic model to calculate glomerular volumes uses a coefficient of 1.382, with an assumption that glomeruli are spheres and a coefficient 1.01 that assumes 10% variation in the glomerular volume distributions within biopsy samples.³⁶ These stereologic assumptions may have caused a small amount of bias that is unlikely to explain the reported associations. Lack of perfusion, formalin fixation, and paraffin embedding leads to glomerular volume shrinkage, but these factors were applied to all tissue specimens. Ischemic-appearing NSG appear to be on a continuous spectrum between nonischemic-appearing NSG and GSG. Thus, there may be some degree of misclassification between glomerular categories, but the morphometry was performed masked to clinical characteristics. Some of the NSG may be atubular, particularly if ischemic-appearing³⁷; however, we lacked serial section to identify atubular glomeruli. Even with 2453 biopsy specimens, we were likely underpowered to detect associations with GSG volume in kidney donors because only 225 had at least two GSG profiles to analyze and 24-hour urine protein data. Further, GSG volume estimates did not account for the shrinkage and disappearance of GSG over time.² Assessment of family history of ESRD is somewhat biased as some living unrelated donors and patients with renal tumor do have relatives with ESRD. The studied populations were predominantly white, and we were not able to meaningfully assess race differences.

In conclusion, only very select patients with an overt nephropathy undergo a kidney biopsy for clinical indications. However, there are substantial structural changes with glomeruli that occur with aging and in early kidney disease settings where a biopsy would not be indicated. Given the absence of kidney biopsy data in the general population, this combined study of living kidney donors and patients who underwent a nephrectomy for a renal tumor provides the clearest picture yet of the factors influencing glomerular size among patients without a specific kidney disease.

Disclosures

None.