Kidney Function Reserve Capacity in Early and Later Stage Autosomal Dominant Polycystic Kidney Disease

Abstract

Visual Abstract

Background and objectives 

It is assumed that in autosomal dominant polycystic kidney disease (ADPKD), kidney function remains in the normal range for several decades because of hyperfiltration of remnant nephrons. In this study, we investigate the extent to which patients with ADPKD hyperfilter.

Design, setting, participants, & measurements 

In this cross-sectional study, we measured GFR as urinary clearance using continuous infusion of ¹²⁵I-iothalamate. Kidney function reserve capacity was determined as increase in measured GFR after adding a dopamine infusion of 4.4–6 mg/h. Potential kidney donors were used as healthy controls and matched by age and sex to patients with ADPKD for comparisons across age groups and CKD stages. Hyperfiltration was defined by a loss of kidney function reserve capacity compared with healthy controls.

Results 

A total of 300 participants were studied. In the youngest age group (18–29 years), measured GFR was not different between patients with ADPKD and healthy controls (103±21 versus 111±9 ml/min per 1.73 m²; P=0.14). In this age group kidney function reserve capacity was higher compared with healthy controls (11.1%±8.3% versus 5.3%±6.5%; P=0.04). Moreover, kidney function reserve capacity was similar to healthy controls in patients with ADPKD with early-stage disease (eGFR≥60 ml/min per 1.73 m²), either overall or when divided into fast or slow progressors according to their Mayo height-adjusted total kidney volume class. However, in patients with ADPKD, lower measured GFR was associated with lower kidney function reserve capacity (β=1.0 [95% confidence interval, 0.5 to 1.5] % per 10 ml/min per 1.73 m²; P<0.001). Kidney function reserve capacity was therefore lower compared with healthy controls at older age and later CKD stages.

Conclusions 

Patients with early-stage ADPKD, either classified as having rapidly or slowly progressive disease, are able to increase their GFR in response to dopamine. Hyperfiltration, defined by a loss of kidney function reserve capacity, may therefore not be an early phenomenon in ADPKD.

Introduction

Autosomal dominant polycystic kidney disease (ADPKD) is characterized by the formation and growth of cysts in both kidneys, which results in a decline in GFR. Ultimately, most patients with ADPKD require kidney replacement therapy. It is assumed that the decline in GFR only occurs after several decades, whereas cyst formation and growth already starts in utero. This makes GFR a less sensitive measure for disease severity and prognosis, especially in the early stages of the disease (¹). It has been suggested that this preservation of GFR in early-stage disease is caused by a compensatory mechanism in remnant nephrons that are not yet lost due to disease progression. This phenomenon is called glomerular hyperfiltration (²).

Glomerular hyperfiltration cannot be directly measured in humans. Several measures are therefore used as a surrogate. Glomerular hyperfiltration is sometimes defined as an increased filtration fraction (³). However, measurement of filtration fraction by infusion of exogenous tracers, such as iothalamate and hippuran, may be inaccurate. It may lead to overestimation of filtration fraction, especially when tubular function is compromised, as in ADPKD (⁴). Glomerular hyperfiltration is therefore more commonly defined as the loss of kidney function reserve capacity, i.e., the impairment of the kidney to increase GFR in response to stimuli such as dopamine (⁵,⁶).

If patients with ADPKD hyperfilter in the early stages of their disease, a loss of kidney function reserve capacity is expected to occur before a decline in GFR is detected. Therefore loss of kidney function reserve capacity might be one of the earliest markers of severe disease. Although this is widely assumed, it has never been formally investigated. In this study we first investigated whether individuals hyperfilter across the full spectrum of ADPKD by measuring kidney function reserve capacity. Second, we studied which factors are associated with kidney function reserve capacity. Lastly, we analyzed whether similar results are obtained when hyperfiltration is defined as elevated filtration fraction.

Materials and Methods

Setting and Participants

All adult patients with ADPKD visiting the University Medical Center Groningen from October 2014 until May 2017 were asked to participate in this observational study. The diagnosis of ADPKD was made on the basis of the revised Ravine criteria (⁷). Patients were considered ineligible if they received kidney replacement therapy, had other systemic diseases, or had treatments potentially affecting kidney function. Patients with a wide range of kidney function were included to allow comparisons across early- and later-stage ADPKD. For this study, potential kidney donors were used as healthy controls, and underwent kidney function measurement with iothalamate and hippuran. Only potential kidney donors without a history of cardiovascular or kidney disease and without abnormalities on a routine investigation of blood hematology, chemistry, and urinalysis were included. These healthy controls were age- and sex-matched to patients with ADPKD in a 1:1 ratio. The study was performed in adherence to the Declaration of Helsinki and all participants gave written informed consent.

Clinical and Biochemical Measurements

All participants were scheduled for a 1 day clinical evaluation at our outpatient clinic. Patients with ADPKD collected a 24-hour urine sample 1 day before and a fasting spot urine on the day of this visit. Blood samples were drawn for the measurement of creatinine with an enzymatic assay (Modular; Roche Diagnostics). GFR was estimated using the 2009 CKD Disease Epidemiology equation (⁸). Protein intake (grams per day) was calculated as 24-hour urinary urea excretion ×0.18+15 according to the Maroni formula (⁹), and sodium intake was estimated using 24-hour urinary sodium excretion.

Magnetic resonance imaging was performed to assess total kidney volume in patients with ADPKD only, using a standardized abdominal magnetic resonance imaging protocol, without the use of intravenous contrast. Scanning was performed with a 1.5 (Magnetom Avento; Siemens) or 3.0 Tesla magnetic resonance scanner (Intera; Philips). Total kidney volume was measured on T2-weighted coronal images by an artificial multiobserver deep neural network model for fully automated segmentation (¹⁰) and adjusted for height.

PKD mutation analysis was performed with DNA isolation using PUREGENE nucleic acid purification chemistry on the AUTOPURE LS 98 platform (Qiagen), followed by sequencing of amplified coding exons directly (exon 34–46), or on long-range PCR products (exon 1–33) (¹¹).

Kidney Function Measurements and Kidney Function Reserve Capacity

GFR and effective kidney plasma flow were measured as urinary clearance of ¹²⁵I-iothalamate and ¹³¹I-hippuran, respectively, applying a constant infusion method as described previously (¹²,¹³). Kidney function reserve capacity was determined as increase in measured GFR after adding a dopamine infusion. A more detailed description of this measurement and the physiologic changes of the kidney during dopamine infusion is added to the Supplemental Material. Kidney function reserve capacity was calculated as percentage increase in measured GFR during dopamine infusion: (measured GFR_dopamine−measured GFR)/measured GFR) ×100%; and as absolute increase in measured GFR (ml/min per 1.73 m²): measured GFR_dopamine−measured GFR. A loss of kidney function reserve capacity compared with healthy controls was used as a surrogate for hyperfiltration (⁶). Filtration fraction was expressed as percentage and calculated as (measured GFR/effective kidney plasma flow) ×100%. Kidney blood flow (ml/min per 1.73 m²) was calculated as effective kidney plasma flow/(1−serum hematocrit). Kidney vascular resistance (dynes/s per meter⁵) was calculated as (mean arterial pressure/kidney blood flow) ×80,000.

Statistical Analyses

Normally distributed data are expressed as means±SD, whereas non-normally distributed data are expressed as median with interquartile range.

Patients and healthy controls were first divided into age groups 18–29, 30–39, 40–49, 50–59, and ≥60 years to compare patients with ADPKD in different stages of the disease with healthy controls. Differences between patients with ADPKD and healthy controls were tested using a two-sample t test when normally distributed or a Mann–Whitney U test when not normally distributed. A chi-squared test was used for categorical data. Differences in results of kidney function measurement before and during dopamine infusion were calculated using a paired-sample t test. A P for trend was calculated for baseline characteristics and results of the kidney function measurement across age groups in both patients with ADPKD and healthy controls. Therefore, a one-way ANOVA was used in case of normal distribution, a Kruskal–Wallis H test in case of non-normal distribution, and a linear chi-squared test for categorical data. Second, we divided participants according to CKD stage and compared healthy controls with patients with ADPKD and CKD stages 1–4 with a one-way ANOVA analysis using a post hoc Bonferroni correction.

Univariable and multivariable linear regression analyses were used to investigate possible determinants of kidney function reserve capacity and filtration fraction in patients with ADPKD and healthy controls separately (sex, measured GFR, PKD mutation, height-adjusted total kidney volume [htTKV], renin-angiotensin-aldosterone system [RAAS] inhibitor use, body mass index, protein intake, and sodium intake).

Next, we compared kidney function reserve capacity and filtration fraction between the various risk classes of the Mayo htTKV classification (¹⁴) and between different PKD mutations with a one-way ANOVA with post hoc Bonferroni correction. Lastly, as a sensitivity analysis, we selected patients with CKD stage 1 and 2 (i.e., eGFR≥60 ml/min per 1.73 m²) and divided these patients in fast progressors (Mayo htTKV class 1C–E) or slow progressors (Mayo htTKV class 1A and B). We subsequently tested if there were differences in kidney function reserve capacity or filtration fraction between healthy controls, fast progressors, and slow progressors with a one-way ANOVA with a post hoc Bonferroni correction.

Analyses were performed with SPSS version 23.0 (SPSS Inc., Chicago, IL). A two-sided P<0.05 was considered statistically significant.

Results

Participant Characteristics

We included 150 patients with ADPKD (59% were women), with a mean age of 46±12 (range 18–75) years. Patients were matched by age and sex with 150 healthy controls. Patients with ADPKD had a similar BP, but were more likely to use antihypertensive medication. As expected, patients with ADPKD had lower eGFR (Table 1). For further analyses, patients with ADPKD and healthy controls were subdivided into five age categories (18–29, 30–39, 40–49, 50–59, and ≥60 years). Supplemental Table 1 shows the baseline characteristics according to these age groups.

Table 1. - Clinical characteristics of patients with ADPKD and matched healthy controls

Characteristics	ADPKD		Healthy Controls
	N		N
Women, n (%)	150	88 (59)	150	80 (53)
Age, yr	150	46±12	150	46±11
Weight, kg	150	83±18	150	81±13
Height, cm	150	176±10	150	176±9
Body mass index, kg/m2	150	26.6±4.8	150	26.2±3.2
Systolic BP, mm Hg	149	127±13	150	123±11
Diastolic BP, mm Hg	149	78±9	150	74±8
Antihypertensive use, n (%)	150	114 (76)	150	12 (8)
RAAS inhibitor use, n (%)	150	104 (69)	150	2 (1)
Protein intake, g/24 h	147	86±23	141	90±29
Sodium intake, mmol/24 h	147	157±60	142	193±76
eGFR, ml/min per 1.73 m2	150	63±31	150	92±15
CKD stage, n (%)	150	27	—	—
1		27 (18)		—
2		52 (35)		—
3A		22 (15)		—
3B		23 (15)		—
4		23 (15)		—
5		3 (2)		—
htTKV, ml/m	143	785 (489–1282)	—	—
Mayo htTKV class, n (%)	143		—
1A		9 (6)		—
1B		31 (21)		—
1C		50 (33)		—
1D		30 (20)		—
1E		16 (11)		—
2		7 (5)		—
PKD mutation, n (%)	130		—
PKD1 truncating		53 (35)		—
PKD1 nontruncating		44 (29)		—
PKD2		27 (18)		—
No mutation detected		6 (4)		—

Variables are presented as mean±SD, or as median (interquartile range) in case of non-normal distribution. ADPKD, autosomal dominant polycystic kidney disease; RAAS, renin-angiotensin-aldosterone system; —, not applicable; htTKV, height-adjusted total kidney volume; PKD, polycystic kidney disease.

Kidney Function According to Age

Overall, eGFR and measured GFR were 63±31 and 66±29 ml/min per 1.73 m² in patients with ADPKD and 92±15 and 102±15 ml/min per 1.73 m² in healthy controls, respectively.

eGFR and measured GFR were lower in patients with ADPKD and healthy controls in older age groups (Table 2). However, in the youngest age group (18–29 years), both eGFR and measured GFR were not different between patients with ADPKD and healthy controls (110±22 versus 113±11 ml/min per 1.73 m²; P=0.64 and 103±21 versus 111±9 ml/min per 1.73 m²; P=0.14, respectively). eGFR as well as measured GFR were lower in patients with ADPKD compared with healthy controls in all other age groups (30 years and older) (Figure 1).

Table 2. - Kidney function measurement according to age group

Variables	Age, yr
	18–29		30–39		40–49		50–59		≥60		P for Trend
	ADPKD	Control	ADPKD	Control	ADPKD	Control	ADPKD	Control	ADPKD	Control	ADPKD	Control
N	17	17	26	26	41	41	50	50	16	16	—	—
eGFR, ml/min per 1.73 m2	110±22	113±11	85±20 a	98±13	58±25 a	91±10	50±21 a	86±12	35±21 a	83±13	<0.001	<0.001
Measured GFR, ml/min per 1.73 m2	103±21	111±9	90±18 a	109±10	62±25 a	104±15	53±21 a	97±14	38±19 a	89±19	<0.001	<0.001
Effective kidney plasma flow, ml/min per 1.73 m2	317±78	346±42	260±44 a	346±58	192±76 a	326±51	169±66 a	299±46	123±49 a	282±57	<0.001	<0.001
Kidney blood flow, ml/min per 1.73 m2	548±139	626±74	455±83 a	623±94	311±124 a	577±103	285±111 a	532±87	228±104 a	511±112	<0.001	<0.001
Kidney vascular resistance, dynes/s per centimeter (5)	11,291 (10,174–16,077) b	10,070 (9238–11,092)	13,377 a (12,419–15,777)	9617 (8874–11,100)	21,659 a (17,134–28,200)	11,344 (9574–12,657)	24,723 a (18,214–32,332)	12,378 (11,016–14,530)	27,454 a (20,970–50,899)	13,547 (11,658–16,424)	<0.001	<0.001
Kidney function reserve capacity, %	11.1±8.3 b	5.3±6.5	3.8±5.6 a	9.9±8.2	4.3±9.2 a	11.8±7.6	2.6±11.4 a	9.4±10.2	−0.1±8.8 a	9.8±6.3	0.001	0.18
Filtration fraction, %	33.5±4.4	32.7±4.9	34.5±4.1 b	32.2±4.2	32.2±4.2	32.2±3.6	31.1±3.8 b	32.8±3.8	29.6±4.1	31.9±3.8	<0.001	0.75

Variables are presented as mean±SD, or as median (interquartile range) in case of non-normal distribution. P values are obtained using one-way ANOVA in case of normal distribution and Kruskal–Wallis test in case of non-normal distribution. ADPKD, autosomal dominant polycystic kidney disease; —, not applicable.

^aP<0.001 compared with healthy control same age group.

^bP<0.05 compared with healthy control same age group.

Figure 1.: eGFR (upper panel) and measured GFR (lower panel) are similar in patients with ADPKD compared with healthy controls in the youngest age group and lower in older age groups . Data are expressed as Tukey boxplots with median, interquartile range, minimum and maximum within 1.5 interquartile range, and outliers displayed as black dots; **P<0.001.

Kidney Function Reserve Capacity

Results of the kidney function measurement before and during dopamine infusion are given in Table 3. In healthy controls, infusion of dopamine caused an increase in effective kidney plasma flow and, consequently, an increase in measured GFR in all age groups. However, in patients with ADPKD, although effective kidney plasma flow increased in all age groups, measured GFR did not increase in patients aged 50–59 and ≥60 years.

Table 3. - Kidney function measurement before and during dopamine infusion (predopamine and dopamine, respectively) according to age group

Variables	Age, yr
	18–29		30–39		40–49		50–59		≥60
	Predopamine	Dopamine	Predopamine	Dopamine	Predopamine	Dopamine	Predopamine	Dopamine	Predopamine	Dopamine
ADPKD
N	17		26		41		50		16
Measured GFR, ml/min per 1.73 m2	103±21	114±25 a	90±18	93±19 b	62±25	66±28 a	53±21	54±23	38±19	38±20
Effective kidney plasma flow, ml/min per 1.73 m2	317±78	363±96 a	260±44	284±53 a	192±76	220±95 a	169±66	187±80 a	123±49	131±57 b
Filtration fraction, %	33.5±4.4	32.5±4.6 b	34.5±4.1	33.1±4.3 b	32.2±4.2	30.7±4.1 a	31.1±3.8	29.3±4.2 a	29.6±4.1	28.6±4.7 b
Healthy controls
N	17		26		41		50		16
Measured GFR, ml/min per 1.73 m2	111±9	117±8 b	109±10	120±13 a	104±15	116±18 a	97±14	105±16 a	89±18	97±16 a
Effective kidney plasma flow, ml/min per 1.73 m2	347±42	387±57 a	346±58	403±72 a	326±51	379±69 a	299±46	343±59 a	282±57	319±56 a
Filtration fraction, %	32.7±4.9	31.1±5.1 b	32.2±4.2	30.0±4.3 b	32.2±3.6	31.1±4.1 b	32.8±3.8	31.2±3.8 a	31.9±3.8	30.9±3.7 b

Variables are presented as mean±SD. Differences were tested with a paired t test. ADPKD, autosomal dominant polycystic kidney disease.

^aP<0.001 between predopamine and dopamine.

^bP<0.05 between predopamine and dopamine.

Overall, kidney function reserve capacity was 3.9±9.7% in patients with ADPKD and 9.7±8.6% in healthy controls (P<0.001). Kidney function reserve capacity was lower in older age groups in patients with ADPKD, but not in healthy controls (Table 2). Surprisingly, in the youngest age group kidney function reserve capacity was higher in patients with ADPKD compared with healthy controls (P=0.04). As expected, in the older age groups kidney function reserve capacity was lower compared with healthy controls (Figure 2, upper panel). Results with absolute kidney function reserve capacity were similar (Supplemental Figure 1). In patients with ADPKD, kidney function reserve capacity was lower at higher CKD stages and was similar to healthy controls in early CKD stages (1–3A) (Figure 2, lower panel).

Figure 2.: Kidney function reserve capacity (as percentage) in patients with ADPKD is slightly higher compared with healthy controls in the youngest age group (upper panel) and similar compared with healthy controls in early CKD stages (lower panel). Data are expressed as Tukey boxplots with median, interquartile range, minimum and maximum within 1.5 interquartile range, and outliers displayed as black dots; *P<0.05; **P<0.001.

We proceeded with testing whether there were determinants of kidney function reserve capacity that could explain the differences we observed in age groups between patients with ADPKD and healthy controls. In patients with ADPKD, measured GFR and htTKV were univariable associated with kidney function reserve capacity (β=1.0 [95% confidence interval (95% CI), 0.5 to 1.5] % per 10 ml/min per 1.73 m²; P<0.001 and β=−2.4 [95% CI, −3.9 to −0.8] % per doubling; P=0.003, respectively). Only measured GFR remained associated in multivariable regression analysis, with higher measured GFR being associated with higher kidney function reserve capacity (β=0.8 [95% CI, 0.05 to 1.6] % per 10 ml/min per 1.73 m²; P=0.04). In healthy controls, measured GFR was also associated with kidney function reserve capacity, but in the opposite direction of what was observed in patients with ADPKD (univariable β=−1.5 [95% CI, −2.3 to −0.6] % per 10 ml/min per 1.73 m²; P=0.001 and multivariable β=−1.5 [95% CI, −2.4 to −0.6] % per 10 ml/min per 1.73 m²; P=0.002) (Table 4).

Table 4. - Possible determinants of kidney function reserve capacity and filtration fraction in patients with ADPKD and healthy controls

Variables		Kidney Function Reserve Capacity (%)					Filtration Fraction (%)
		Univariable		Multivariable		Multivariable
	N	β [95% CI]	P Value	β [95% CI]	P Value	β [95% CI]		P Value	β [95% CI]	P Value
ADPKD
Men versus women	150	−1.5 [−4.7 to 1.7]	0.34	−0.8 [−4.6 to 3.1]	0.70	−0.7 [−2.1 to 0.7]		0.31	−1.1 [−2.7 to 0.5]	0.17
Measured GFR, per 10 ml/min per 1.73 m2	150	1.0 [0.5 to 1.5]	<0.001	0.8 [0.05 to 1.6]	0.04	0.6 [0.4 to 0.9]		<0.001	0.6 [0.3 to 0.9]	<0.001
PKD2 versus PKD1 mutation	124	−0.4 [−4.6 to 3.7]	0.84	−0.4 [−4.8 to 4.0]	0.86	0.1 [−1.8 to 2.0]		0.95	−0.2 [−2.0 to 1.6]	0.85
htTKV (per doubling)	143	−2.4 [−3.9 to −0.8]	0.003	−1.6 [−3.8 to 0.6]	0.15	−0.4 [−1.1 to 0.3]		0.25	0.6 [−0.3 to 1.5]	0.18
RAAS inhibitor use (yes versus no)	150	−1.9 [−5.3 to 1.5]	0.26	0.9 [−3.3 to 5.0]	0.68	−2.4 [−3.9 to −1.0]		0.001	−1.3 [−3.0 to 0.4]	0.12
BMI, per kg/m2	150	0.1 [−0.3 to 0.4]	0.72	0.2 [−0.2 to 0.6]	0.25	0.1 [−0.1 to 0.2]		0.47	0.0 [−0.1 to 0.2]	0.42
Protein intake, per 10 g/24 h	147	0.1 [−0.6 to 0.8]	0.74	0.2 [−0.7 to 1.1]	0.63	0.2 [−0.1 to 0.5]		0.30	−0.0 [−0.4 to 0.3]	0.81
Sodium intake, per 10 mmol/24 h	147	0.1 [−0.2 to 0.3]	0.71	−0.1 [−0.4 to 0.3]	0.73	0.2 [0.1 to 0.3]		<0.001	0.1 [0.0 to 0.3]	0.05
Healthy controls
Men versus women	150	−0.1 [−2.9 to 2.7]	0.96	−0.2 [−3.2 to 2.8]	0.89	0.1 [−1.1 to 1.4]		0.83	0.1 [−1.3 to 1.5]	0.91
Measured GFR, per 10 ml/min per 1.73 m2	150	−1.5 [−2.3 to −0.6]	0.001	−1.5 [−2.4 to −0.6]	0.002	0.5 [0.1 to 0.9]		0.02	0.5 [0.1 to 0.9]	0.03
RAAS inhibitor use (yes versus no)	150	−8.8 [−20.7 to 3.2]	0.15	−7.8 [−20.0 to 4.5]	0.21	−0.4 [−6.0 to 5.1]		0.88	−1.4 [−7.1 to 4.3]	0.63
BMI, per kg/m2	150	0.2 [−0.3 to 0.6]	0.47	0.03 [−0.4 to 0.5]	0.91	0.0 [−0.2 to 0.2]		0.84	0.1 [−0.1 to 0.3]	0.42
Protein intake, per 10 g/24 h	141	0.1 [−0.4 to 0.6]	0.63	0.02 [−0.6 to 0.6]	0.95	−0.2 [−0.4 to 0.1]		0.17	−0.2 [−0.5 to 0.0]	0.07
Sodium intake, per 10 mmol/24 h	142	0.1 [−0.1 to 0.3]	0.53	0.1 [−0.1 to 0.3]	0.31	0.0 [−0.1 to 0.1]		0.75	0.0 [−0.1 to 0.1]	0.50

βs with 95% CIs and P values were calculated using linear regression analysis with pairwise exclusion of missing data. Dependent variable is kidney function reserve capacity or filtration fraction; independent variables are sex, measured GFR, PKD mutation, htTKV, use of RAAS inhibitors, BMI, protein intake, and sodium intake. ADPKD, autosomal dominant polycystic kidney disease; 95% CI, 95% confidence interval; PKD, polycystic kidney disease; htTKV, height-adjusted total kidney volume; RAAS, renin-angiotensin-aldosterone system; BMI, body mass index.

We investigated kidney function reserve capacity across the various Mayo htTKV risk classes and across different PKD mutations. Although we found differences in kidney function reserve capacity between patients with ADPKD and healthy controls, we found no differences in kidney function reserve capacity across these risk classes and mutation types in patients with ADPKD (Supplemental Figure 2).

When including only patients with a preserved kidney function (CKD stage 1 and 2, i.e., eGFR≥60 ml/min per 1.73 m², n=72), we also found no differences in kidney function reserve capacity between fast progressors (Mayo htTKV class 1C–E, n=47), slow progressors (Mayo htTKV class 1A and B, n=25) and healthy controls (Figure 3).

Figure 3.: In patients with CKD stage 1 and 2 ( i.e. , eGFR≥60 ml/min per 1.73m ² ) kidney function reserve capacity (upper panel) and filtration fraction (lower panel) are comparable in fast progressors (Mayo htTKV class 1C–E) and slow progressors (Mayo htTKV class 1A and B) compared with healthy controls. Data are expressed as Tukey boxplots with median, interquartile range, minimum and maximum within 1.5 interquartile range, and outliers displayed as black dots.

Effective Kidney Plasma Flow and Filtration Fraction

We repeated the analyses with our secondary measure of glomerular hyperfiltration: filtration fraction, defined as measured GFR/effective kidney plasma flow x100%. Overall, effective kidney plasma flow was 203±86 ml/min per 1.73 m² in patients with ADPKD and 318±55 ml/min per 1.73 m² in healthy controls (P<0.001). Filtration fraction was comparable between patients with ADPKD and healthy controls (32.1±4.3 versus 32.4±3.9%; P=0.51). In older age groups, effective kidney plasma flow was lower in patients with ADPKD and healthy controls, but in patients with ADPKD the decrease in effective kidney plasma flow was more considerable. In patients with ADPKD, we observed a lower filtration fraction at older age, whereas filtration fraction was similar at older age in healthy controls (Table 2). In the youngest age group and those with CKD stage 1, effective kidney plasma flow was similar in patients with ADPKD compared with healthy controls (Supplemental Figure 3). Results with kidney blood flow and kidney vascular resistance were similar (Supplemental Figures 4 and 5). Because measured GFR was also similar compared with healthy controls in the youngest age group, there was no difference in filtration fraction (33.5±4.4% versus 32.7±4.9%; P=0.62). In the other age groups, filtration fraction in patients with ADPKD was also similar to healthy controls, with the age group 30–39 year filtration fraction being slightly higher, and in the age group 50–59 years filtration fraction being slightly lower than in healthy controls (Figure 4, upper panel). Filtration fraction decreased in patients with ADPKD at higher CKD stages (Figure 4, lower panel).

Figure 4.: Filtration fraction is comparable in patients with ADPKD and healthy controls according to age group (upper panel) and decreased in patients with ADPKD at higher CKD stages (lower panel). Data are expressed as Tukey boxplots with median, interquartile range, minimum and maximum within 1.5 interquartile range, and outliers displayed as black dots. *P<0.05; **P<0.001.

In patients with ADPKD, measured GFR, use of RAAS inhibitors, and sodium intake were univariably associated with filtration fraction (β=0.6 [95% CI, 0.4 to 0.9] % per 10 ml/min per 1.73 m²; P<0.001, β=−2.4 [95% CI, −3.9 to −1.0] %; P=0.001 and β=0.2 [95% CI, 0.1 to 0.3] % per 10 mmol/24 h; P<0.001, respectively). In the multivariable regression analysis, measured GFR remained significantly associated with filtration fraction and the association of sodium intake with filtration fraction was borderline significant (β=0.6 [95% CI, 0.3 to 0.9] % per 10 ml/min per 1.73 m²; P<0.001 and β=0.1 [95% CI, 0.0 to 0.3] % per 10 mmol/24 h; P=0.05, respectively) (Table 4). In the healthy controls, only measured GFR was associated with filtration fraction in the univariable (β=0.5 [95% CI, 0.1 to 0.9] % per 10 ml/min per 1.73 m²; P=0.02) as well as the multivariable regression analysis (β=0.5 [95% CI, 0.1 to 0.9] % per 10 ml/min per 1.73 m²; P=0.03) (Table 4).

As with kidney function reserve capacity, we found no differences in filtration fraction across risk classes of the Mayo htTKV classification and across PKD mutations in patients with ADPKD. In addition, filtration fraction was not different between healthy controls and patients with ADPKD according to Mayo htTKV class or PKD mutation (Supplemental Figure 6).

When including only patients with a preserved kidney function (CKD stage 1 and 2, i.e., eGFR≥60 ml/min per 1.73 m², n=72), we also found no differences in filtration fraction between fast progressors (Mayo htTKV class 1C–E, n=47), slow progressors (Mayo htTKV class 1A and B, n=25), and healthy controls Figure 3.

Discussion

This study showed that young patients with ADPKD have a GFR that is comparable with that of healthy controls at similar age, despite having enlarged kidneys. Remarkably, patients with ADPKD in this age group had a normal level of kidney function reserve capacity, as did patients in early CKD stages. In older age groups and at later CKD stages, kidney function reserve capacity was lower compared with healthy controls. These results indicate that loss of kidney function reserve capacity is not an early phenomenon in ADPKD.

Franz and Reubi (¹) were the first to observe, in a small group of patients with ADPKD (n=44), that kidney function remains stable for decades before it deteriorates. Later, Grantham et al. (²) hypothesized that this phenomenon is caused by compensatory hyperfiltration of the kidneys. There are, however, few studies that have sought to confirm or deny this hypothesis. To date, these studies have been small and used an unstimulated elevated GFR above a certain value as definition for hyperfiltration in patients with ADPKD. This is a definition used in patients with a healthy kidney function (³,¹⁵,¹⁶). Whether this definition can be used in patients with kidney disease is debatable. It is assumed that such patients hyperfilter to compensate for the loss of nephrons. In that case one, would not expect GFR to become higher than in healthy controls. In addition, this definition makes it impossible to study hyperfiltration in later stages of the disease. A seminal study, performed in 180 children with ADPKD, found that hyperfiltration (defined as a creatinine clearance of ≥140 ml/min per 1.73 m²), present in 20% of children, was associated with higher rates of growth in htTKV and decline in creatinine clearance (¹⁷). In this study, kidney function was estimated using creatinine clearance, which entails glomerular filtration as well as tubular secretion of creatinine. It has previously been shown that the tubular secretion of creatinine is higher in patients with ADPKD compared with healthy controls, especially in patients in the normal GFR range (¹⁸). Therefore it cannot be excluded that the elevated creatinine clearance in these children represents tubular dysfunction rather than glomerular hyperfiltration.

We used a gold-standard technique with continuous infusion of ¹²⁵I-iothalamate to measure GFR, and infused dopamine to determine kidney function reserve capacity (^19–23). Several studies have shown a loss of kidney function reserve capacity, measured using infusion of a low dose of dopamine, in conditions where hyperfiltration is expected, such as in diabetes mellitus (²⁰) and after unilateral nephrectomy (²¹). Loss of kidney function reserve capacity is therefore an accepted definition of hyperfiltration, and allows for the study of hyperfiltration at all stages of CKD. In our study, GFR was similar in patients with ADPKD compared with healthy controls in the youngest age group, despite considerable cyst burden. These results suggest that there is a compensation to account for the loss in kidney function that is assumed to occur because of cyst formation. Yet, there was no loss of kidney function reserve capacity in the youngest age group of patients with ADPKD when compared with healthy age- and sex-matched controls as well as in patients with ADPKD with CKD stages 1 and 2. Therefore, one might conclude that there is no hyperfiltration in early-stage ADPKD. This conclusion is corroborated by the fact that filtration fraction, another surrogate measure of hyperfiltration, was also not elevated at young age and early CKD stages. At older age and in later CKD stages, kidney function reserve capacity was impaired, suggesting that there is indeed hyperfiltration at later-stage ADPKD. Taken together, these results suggest that there is no hyperfiltration in early-stage ADPKD and that cyst formation in this stage does not lead to significant nephron loss. Therefore, there may be a possibility that treatments that influence vascular tone, like RAAS inhibitors, may not have a benefit over other BP-lowering agents in early-stage ADPKD to prevent future kidney function decline. It may be that only when disease progresses, cyst burden becomes even more prominent, and other disease mechanisms come into play (like inflammation and fibrosis) that nephrons are lost and compensatory hyperfiltration occurs.

As expected, a lower measured GFR was associated with a lower kidney function reserve capacity in patients with ADPKD. Because we observed differences in kidney function reserve capacity between age groups, we tested whether there were determinants other than disease severity (i.e., measured GFR or htTKV) that could explain differences in kidney function reserve capacity and filtration fraction between patients with ADPKD and healthy controls in the different age groups. Adjustment for RAAS inhibitor use, body mass index (²⁴,²⁵), and sodium and urea excretion (as surrogates for sodium intake and protein intake, respectively) did not materially change our results with respect to kidney function reserve capacity. However, we did observe that a higher sodium intake was associated with a higher filtration fraction in patients with ADPKD, even after adjustment for sex, measured GFR, PKD mutation, and htTKV. This is in line with the literature, which shows that sodium intake affects kidney hemodynamics in patients with CKD (²⁶,²⁷). No association was found between sodium excretion and kidney function reserve capacity in healthy controls nor in patients with ADPKD.

Importantly, because kidney function reserve capacity is not impaired early in the disease course, it cannot serve as an early biomarker for disease severity and risk of future disease progression in ADPKD. This is corroborated by the fact that kidney function reserve capacity was not different across Mayo htTKV risk classes for future disease progression across the entire study population, nor in patients with early CKD stages or across different types of PKD mutation.

Our study has some limitations. First, our study is cross-sectional in nature and we cannot draw conclusions about associations with disease progression. We used therefore a risk classification for future disease progression (on the basis of kidney volume indexed for age) and investigated differences between PKD mutations to overcome this limitation. Second, it is not clear whether maximal kidney function reserve capacity is reached with dopamine alone. An increase in measured GFR can also be obtained with infusion of amino acids (²⁸). However, other authors found that simultaneous infusion of dopamine and amino acids in patients with ADPKD did not lead to a significant increase in kidney function reserve capacity compared with infusion of dopamine alone (²⁹). Third, our participants did not consume a standardized diet before the kidney function measurement. In our multivariable analyses, we therefore adjusted for sodium and urea excretion, reliable measures for sodium and protein intake. These adjustments did not change the results. The strengths of our study are that we performed extensive kidney hemodynamic measurements with gold-standard methods to measure GFR, effective kidney plasma flow, kidney function reserve capacity, and filtration fraction. Although the sample size of this study may seem small, a study with such extensive measurements has, to the best of our knowledge, never been performed. Therefore this study entails rather a relatively large population of patients with ADPKD and healthy controls. In addition, information was available on other disease parameters, like total kidney volume and PKD mutation analysis.

In conclusion, patients with ADPKD at young adult age or with early CKD stages have a GFR in the normal range, and are still able to increase their GFR in response to dopamine. Hyperfiltration, measured as loss of kidney function reserve capacity, can therefore not be used as an early biomarker of disease severity. Filtration fraction was also not elevated. Taken together, these results suggest that there may be no hyperfiltration in early-stage ADPKD.

Disclosures

None.