Figures
Abstract
Aims
Obesity is an important risk factor for the development of chronic kidney disease. One of the major factors involved in the pathogenesis of obesity-associated kidney disease is glomerular hyperfiltration. Increasing salt-delivery to the macula densa is expected to decrease glomerular filtration rate (GFR) by activating tubuloglomerular feedback. Acetazolamide, a carbonic anhydrase inhibitor which inhibits salt reabsorption in the proximal tubule, increases distal salt delivery. Its effects on obesity-related glomerular hyperfiltration have not previously been studied. The aim of this investigation was to evaluate whether administration of acetazolamide to obese non diabetic subjects reduces glomerular hyperfiltration.
Materials and Methods
The study was performed using a randomized double-blind crossover design. Obese non-diabetic men with glomerular hyperfiltration were randomized to receive intravenously either acetazolamide or furosemide at equipotent doses. Twelve subjects received the allocated medications. Two weeks later, the same subjects received the drug which they had not received during the first study. Inulin clearance, p-aminohippuric acid clearance and fractional lithium excretion were measured before and after medications administration. The primary end point was a decrease in GFR, measured as inulin clearance.
Results
GFR decreased by 21% following acetazolamide and did not decrease following furosemide. Renal vascular resistance increased by 12% following acetazolamide, while it remained unchanged following furosemide administration. Natriuresis increased similarly following acetazolamide and furosemide administration. Sodium balance was similar in both groups.
Citation: Zingerman B, Herman-Edelstein M, Erman A, Bar Sheshet Itach S, Ori Y, Rozen-Zvi B, et al. (2015) Effect of Acetazolamide on Obesity-Induced Glomerular Hyperfiltration: A Randomized Controlled Trial. PLoS ONE 10(9): e0137163. https://doi.org/10.1371/journal.pone.0137163
Editor: Jaap A. Joles, University Medical Center Utrecht, NETHERLANDS
Received: February 7, 2015; Accepted: July 29, 2015; Published: September 14, 2015
Copyright: © 2015 Zingerman et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: The authors have no support or funding to report.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Obesity is an independent risk factor for chronic kidney disease[1]. Obesity-related glomerulopathy is a well-defined entity, characterized by glomerulomegaly with or without focal segmental glomerulosclerosis[2]. In addition, increased adiposity also accelerates the progression of kidney diseases that are not primarily related to obesity, such as IgA nephropathy[3,4], reduced renal mass[5] and possibly renal transplant nephropathy[6]. Obesity and overweight are associated with increased GFR, renal plasma flow and/or filtration fraction[7–15], with central body fat distribution being a more powerful predictor of hyperfiltration than body mass index (BMI)[16]. The improvement of these renal hemodynamic abnormalities following weight loss[17] supports a cause-and-effect relationship between adiposity and glomerular hyperfiltration.
One of the factors involved in the pathogenesis of obesity-associated kidney disease is glomerular hyperfiltration[18–25], which is associated with afferent arteriolar vasodilatation and increased glomerular pressure[11]. Treatment of chronic kidney disease in the obese subject is oriented, among other things, toward decreasing glomerular pressure by antihypertensive treatment, inhibition of the renin angiotensin system and weight loss. The latter decreases arterial hypertension and restores renal hemodynamics toward normal, reducing renal plasma flow, GFR and filtration fraction[7,17]. However, this treatment has its limitations, weight loss being rarely maintained in the long term in obese subjects.
Activation of tubuloglomerular feedback by increased sodium distal delivery and the consequent decrease in GFR is a yet unexplored way of modulating glomerular hyperfiltration in obesity. Tucker et al[26] showed that administration of a carbonic anhydrase inhibitor in normal rats leads to a decrease in single-nephron GFR by inhibiting proximal tubular reabsorption, increasing salt delivery to the macula densa and thus activating tubuloglomerular feedback. This GFR reducing effect of carbonic anhydrase inhibition has also been demonstrated in normal humans[27–33] and in diabetic subjects[31,34]. However, these studies did not use a control group treated with an equipotent natriuretic agent. Recently, Cherney et al showed, in lean subjects with type 1 diabetes mellitus and glomerular hyperfiltration, that increasing solute delivery to the macula densa by inhibition of proximal tubular reabsorption attenuates hyperfiltration[35]. The effects of inhibition of proximal tubular reabsorption have not been previously studied in an obese non-diabetic population.
The present study was designed to test the hypothesis that administration of acetazolamide, a carbonic anhydrase inhibitor, to obese subjects with glomerular hyperfiltration reduces hyperfiltration. The effects of acetazolamide on renal hemodynamics were compared to those of furosemide, a loop diuretic that does not activate tubuloglomerular feedback[28,30,36–38]. We showed that intravenous acetazolamide decreases GFR in non-diabetic obese men with glomerular hyperfiltration, while furosemide, administered at equipotent natriuretic dose, does not affect GFR.
Materials and Methods
Ethics Statement
The protocol of this study was approved by the Institutional Review Board of the Rabin Medical Center. Informed consent was signed by the participants.
Participants
Eligible participants were men aged 18 to 55 years old with a BMI above 30 kg/m2 and a creatinine clearance above 130 ml/min. Creatinine clearance was measured using a 24-hour urine collection and a serum blood test for creatinine. Exclusion criteria were any of the following conditions: pharmacologic treatment for diabetes mellitus, hypertension and cardiac disease; history of kidney disease, heart failure and chronic obstructive lung disease; therapy with corticosteroid, antiepileptic and non steroidal anti inflammatory medications; known allergy to furosemide, acetazolamide, inulin and amino-hippurate.
Interventions
Enrolled patients underwent two renal function studies, one before and after intravenous administration of acetazolamide and the other before and after intravenous administration of furosemide, which was used as control. The order of administration was randomized. The effects of acetazolamide were compared to those of furosemide, since both medications are natriuretic, but the latter increases natriuresis with no activation of the tubuloglomerular feedback[28,30,36–38]. Acetazolamide was administered at a dose of 5 mg/kg BW. Preliminary studies performed in our lab showed that a furosemide dose of 2 mg was equipotent to that of acetazolamide dose, as far as natriuresis was concerned.
A 24-hour urine collection was performed before the two renal function studies. Subjects received 300 mg of lithium carbonate at 22.00 the day before the tests. They were instructed to drink 250 ml of water at bedtime and at 07.00 am. Renal function tests were started at 08.00 am. after a 10-hour fast, excepting the water drink. Intravenous catheters were placed in each upper limb for infusion of clearance markers and blood sampling. After blood sampling for urea, creatinine, proteins, glucose, electrolytes, blood gases, HbA1c and complete blood count, a priming dose of inulin (40 mg/kg) and p-aminohippuric acid (4 mg/kg) was intravenously injected followed by ingestion of 10 ml/kg of water. Thereafter, inulin and p-aminohippuric acid were infused continuously at a rate of 33 and 17 mg/min, respectively. After each voiding, participants drank an amount of water equal to the amount of urine voided. After the first 60 minutes, 4 accurately timed urine collections of 30 minutes each were obtained by spontaneous voiding. Peripheral venous blood was drawn to bracket each urine collection. Arterial pressure was measured by a trained observer, at the end of each urine collection in the supine position, using an electronic oscillometric blood pressure measuring device (Datascope, Accutorr). The cuff was appropriately sized to the diameter of the arm and the arm positioned at the heart level. Each measurement was the mean of 3 readings. After the first 4 timed urine collections, participants received either intravenous acetazolamide at a dose of 5 mg/kg within 5 min or intravenous furosemide at a dose of 2 mg within 5 min. The drugs were injected after dilution in 0.9% sodium chloride as a 20 ml solution. Following the injection, a 0.9% sodium chloride solution was administered intravenously at a rate of 130 ml/hr and two 30-minute urine collections were performed. The second study was performed 14 (13–22) days after the first study, using the drug that had not been administered during the first study. Blood sampled before and after each urine collection was assayed for albumin, protein, sodium, inulin and p-aminohippuric acid. Each of the urine collections was assayed for sodium, inulin and p-aminohippuric acid. Lithium was measured in 2 urine collections preceding diuretic administration and in urine samples collected 30 and 60 min following diuretic administration, as well as in blood samples bracketing these collections. Venous blood gases were measured in blood samples drawn at the end of the two last baseline urine collections and in samples drawn 30 and 60 min following diuretic administration.
The study was performed using a randomized double-blind crossover comparative design. The order of administration was randomized. Fig 1 shows the investigation's flowchart. Nineteen subjects were assessed for eligibility. Twelve subjects received the allocated treatment. Data obtained from the studies of all 12 subjects were analyzed.
Randomization and masking
Participants were assigned to receive either acetazolamide or furosemide during the first renal function test using a simple randomization procedure (computerized random numbers) prepared by a pharmacist not otherwise involved in the study. Medications were prepared by a nurse not otherwise involved in the study and injected by one of the authors (BZ) who was unaware of the medication administered. The participants, the coauthors and everyone involved in the laboratory procedures and data analysis were blinded to the medication administered until completion of data analysis.
Laboratory procedures
Plasma and urinary concentrations of inulin and p-aminohippuric acid were analyzed by colorimetric methods. Lithium in the serum and urine was measured using the ICP-OES (Inductively Coupled Plasma Optical Emission Spectrometer) method. Urine albumin was measured using chemiluminescence. HbA1c, plasma glucose, sodium, creatinine, albumin, protein and bicarbonate and urine sodium, creatinine and urea nitrogen were measured using standard laboratory methods. HbA1c was measured in 11 out of the 12 obese subjects. The subject with missing HbA1c had a fasting blood glucose level of 96 mg/dl.
Calculations
GFR and renal plasma flow were calculated as inulin and p-aminohippuric acid sodium clearance, respectively. Baseline GFR and renal plasma flow were calculated from the average value of the four inulin and p-aminohippuric acid sodium clearance measurements performed before diuretic administration. BMI was calculated as: Body weight/Height2, body weight being expressed in kg and height expressed in m. The fractional lithium excretion was calculated as: lithium clearance/GFR. Baseline fractional lithium excretion was determined as the average value for two measurements performed before diuretic administration. Renal blood flow was calculated as: Renal plasma flow/(1—Hematocrit), flow being expressed in ml/min and hematocrit as a fraction. Mean arterial pressure was calculated as: (Systolic arterial pressure+2xDiastolic arterial pressure)/3, pressure being expressed in mm Hg. Renal vascular resistance, expressed in mm Hg/(ml/min), was calculated as: Mean arterial pressure/ Renal blood flow. Albumin excretion rate was calculated as the mean value from the two 24-hour urine collections. Protein intake was calculated as[39]: (Urine urea nitrogen + [Body weight x 0.031]) x 6.25, protein intake being expressed in g/d, urine urea nitrogen in g/d and body weight in kg. Sodium balance during renal function studies, i.e. from the time of inulin and p-aminohippuric acid injection to that of the end of the last urine collection, was calculated as: Intravenous sodium intake + Oral sodium intake (as drinking water)–Urinary sodium excretion, expressed in mmol.
Statistical Analysis
Normally distributed data are expressed as mean±SD, unless otherwise specified. Variables with skewed distribution are expressed as median (range). Analysis was performed according to the intention-to-treat principle, with all subjects who received the allocated medication being included in the analysis. The significance of differences between groups was evaluated by ANOVA with repeated measures. The effect of administration order of the drugs was evaluated for the primary outcome. When ANOVA showed a significant treatment-by-time interaction, a paired t-test was applied. Regression analysis between variables was performed after log-transformation for non-normally distributed variables. Missing data: serum albumin and serum bicarbonate data were unavailable for one subject each. Analysis for these two variables was performed for 11 out of the 12 subjects.
All tests were two-sided. P<0.05 was considered as significant. The analyses were carried using SPSS software version 21.
Sample size
Sample size calculation was based on the results of a study by Hannedouche et al31 on the effects of intravenously administrated acetazolamide on renal hemodynamics in healthy and diabetic subjects. GFR decreased by 14±10% following acetazolamide. The SD of the treatment effect was calculated using data appearing in Fig 1 of this publication. Using these data, we calculated that it would be necessary to include 15 subjects in the present investigation in order to demonstrate a 14% decrease in GFR, assuming an alpha of 0.005, a power of 90% and a drop-out of up to 20% after randomization.
Interim Analysis
Due to slow enrollment, an interim analysis was performed after 13 subjects had been randomized and 12 had completed the study. This analysis showed that the primary endpoint was reached. Due to this result and the slow enrollment rate, the study was stopped before completion of the randomization of 15 subjects, as initially planned.
Results
Twelve obese male subjects, aged 36 (32–53) received the allocated medications. Body weight was 122±19 kg and height – 1.78±0.09 m. Body mass index was 38.8±5.7 kg/m2 (range: 31.6–49.5) and waist circumference – 124±12 cm (range: 109–141). Serum creatinine was 69.8±6.2 μmol/L. Fasting blood glucose was 5.05 (4.55–6.49) mmol/L and HbA1c – 5.5 (5.3–6.2)%. Creatinine clearance was 177±27 ml/min. Albumin excretion rate was 12.4 (range: 5.9–39.7; lower quartile: 9.8; upper quartile: 21.9) mg/d. Calculated protein intake, assessed during the day preceding acetazolamide and furosemide studies using 24-hour urine collections, was 115±21 and 111±22 g/d respectively (P NS).
Sodium intake during the day preceding acetazolamide and furosemide studies, as assessed by a 24-hour urine collection, was 248 (105–399) and 262 (123–344) mmol, respectively (P = 0.9). During baseline renal function studies, systolic arterial pressure was 122±11 mm Hg before both acetazolamide and furosemide administration; diastolic arterial pressure was 80±11 and 80±10 mm Hg before acetazolamide and furosemide administration, respectively (P = 0.9).
Renal Hemodynamics
Table 1 and Fig 2 show the renal hemodynamic changes occurring after acetazolamide and furosemide administration. Repeated-measures analysis of variance of GFR (Table 1) showed a significant interaction between treatment and time (P = 0.001), i.e. the effect of acetazolamide on GFR differed from that of furosemide. GFR decreased by 21% following acetazolamide (P<0.001), while it remained unchanged following furosemide administration (Fig 2). Order of administration of the study medications did not affect GFR change. Repeated-measures analysis of variance revealed no significant effect of time on RPF. However, the P value from a paired t-test before, as compared to after, acetazolamide was 0.03, suggesting a trend toward significance. Repeated-measures analysis of variance showed a significant effect of time on renal vascular resistance (P = 0.03) with no significant interaction between treatment and time. Renal vascular resistance increased by 12% following acetazolamide (P = 0.01), while it remained unchanged following furosemide administration (P = 0.1). Fig 3 shows that baseline GFR and the change in GFR following acetazolamide were inversely correlated (r = -0.69, P = 0.01), baseline GFR accounting for 48% of the effect of acetazolamide on GFR. Baseline albuminuria and the change in GFR following acetazolamide were not correlated (r = 0.36, P = 0.25).
GFR decreased by 21% following acetazolamide and remained unchanged following furosemide.
Baseline GFR and the change in GFR following acetazolamide were inversely correlated.
Sodium handling
Table 2 and Fig 4 show renal sodium handling before and after acetazolamide and furosemide injection. Serum sodium remained constant during the whole study. Natriuresis was similar during the two baseline renal function studies preceding medications administration. Repeated-measures analysis of variance revealed no significant interaction between treatment and time for natriuresis, while the effect of time was significant, indicating that acetazolamide and furosemide affected sodium excretion. Urinary sodium excretion increased similarly following administration of the 2 medications. Natriuresis increased by 141% following acetazolamide (P<0.001) and by 151% following furosemide injection (P<0.001). Lithium clearance, a marker of sodium reabsorption by the proximal tubule, was similar during the baseline renal function studies preceding drugs administration. It increased 1.7 times more following acetazolamide than following furosemide (30% vs 18%, P<0.03). Fractional lithium excretion, was similar during the baseline renal function studies preceding drugs administration. It increased 2.3 times more following acetazolamide than following furosemide (61% vs 27%, P = 0.002). Sodium balance during renal function studies was similar following acetazolamide and furosemide, -30±20 mmol and -35±26 mmol, respectively (P = 0.35).
Natriuresis increased similarly following acetazolamide and furosemide administration.
Table 3 shows serum protein and arterial pressure during the study. Systolic and diastolic arterial pressures were similar during baseline studies. Systolic arterial pressure remained constant following both acetazolamide and furosemide administration. Diastolic arterial pressure increased by 3 mm Hg following furosemide injection (P = 0.02) and remained constant following acetazolamide administration (P = 0.9). Serum albumin and total protein remained constant before and after acetazolamide and furosemide administration. Plasma bicarbonate was 24.2±1.4 and 24.7±1.4 mEq/L before and after furosemide (P = 0.1). It decreased from 24.2±1.1 to 23.1±1.1 mEq/L following acetazolamide (P = 0.002).
Adverse effects of allocated medications
Adverse events occurred in 3 subjects following acetazolamide administration: hand paresthesia (two subjects), lip paresthesia (one subject) and alteration of the sense of taste (two subjects). These adverse events resolved spontaneously. No adverse events were recorded following furosemide administration.
Discussion
This randomized controlled investigation shows that acetazolamide reduces GFR in obese non diabetic subjects with glomerular hyperfiltration. A distally acting diuretic injected at an equipotent dose was used as control and showed no effect on GFR.
Previous studies showed that acetazolamide decreases GFR in animal models[40–43], in healthy humans[27–33] and in diabetic subjects[31,34]. The present investigation is the first to compare the effects of acetazolamide to those of an equipotent natriuretic agent and the first to investigate these effects in hyperfiltrating non-diabetic obese subjects.
Acetazolamide, a carbonic anhydrase inhibitor, acts on the proximal tubule by decreasing bicarbonate, sodium and chloride reabsorption. The increased solute delivery to the macula densa activates tubuloglomerular feedback leading to a reciprocal change in single nephron GFR, mostly resulting from changes in glomerular arterioles resistance[44]. The decrease in GFR has also been attributed to an increase in Bowman's space hydrostatic pressure secondary to inhibition of proximal reabsorption[45]. These changes result in a diminished transcapillary pressure gradient and a lowered single nephron GFR. GFR decreased following acetazolamide and remained unchanged following furosemide administration. Sodium balance, slightly negative following diuretic administration, was similar in both groups. Hence, changes in sodium balance did not account for the renal hemodynamic changes. Protein intake, a potential modulator of renal hemodynamics, was similar during the 24 hours preceding the two renal function studies.
The effects of acetazolamide on renal hemodynamics were compared to those of furosemide, a loop diuretic that increases natriuresis by inhibiting the sodium-potassium-2chloride co-transporter in the thick ascending limb of the loop of Henle. Acetazolamide and furosemide both increase solute distal delivery. However, the latter does not activate tubuloglomerular feedback[28,30,36–38] since it inhibits the sodium-potassium-2chloride co-transporter and therefore blocks solute transport into macula densa cells and the consequent decrease in glomerular filtration rate. A possible additional explanation for the lack of effect of furosemide on glomerular hyperfiltration may also be related to the cortical or global renal vasodilation demonstrated by some studies following intravenous administration of furosemide[46–50]. This effect may theoretically contribute to maintaining GFR. In the present study, renal vascular resistance did not change following furosemide. This finding does not support a role for a direct effect of furosemide on the renal vasculature in the settings of the present investigation, performed using low-dose furosemide in water-repleted subjects with high baseline renal plasma flow.
The dose of acetazolamide used in this investigation is similar to that used for clinical indications. Furosemide was administered at a low dose in order to match the natriuretic effect of acetazolamide. Preliminary studies showed that furosemide doses of 10 and 5 mg resulted in a more pronounced natriuretic effect than acetazolamide during the 60 min period following injection. We empirically determined that a 2 mg furosemide dose provides the sought natriuretic effect. A single study by Andreasen et al[51] reported the pharmacodynamics and kinetics of a 5 mg furosemide dose in young healthy adults, along with the effects of higher doses. As far as we know, no published data are available concerning the effects of lower doses. In this publication[51], the results for the 60 min period following injection of the 5 mg dose were provided as urine sodium excretion rate for the 0–15 min period and as fractional excretion of sodium (FENa) for the 15–30 and 30–60 min periods. By backcalculating natriuresis from FENa for the 15–30 and 30–60 min periods in that study, we obtain a urine sodium excretion rate of about 1.4 mmol/min. These findings are consistent with those of the present study, where the 2 mg dose induced about half the natriuresis generated by the 5 mg dose, i.e. 0.6 mmol/min. The authors noted that the median effective dose (ED50) of furosemide is "well below 5 mg", i.e. that the natriuretic response appears at doses much lower than 5 mg. These low doses are not used clinically in adults since the diuretic response wanes rapidly owing to the drug’s short half-life.
Baseline natriuresis was similar before acetazolamide and furosemide administration. Natriuresis increased likewise during the 60 min following administration of the two diuretics. The slightly negative sodium balance was similar in the two groups. During this time period, GFR decreased by 21% following acetazolamide, while it remained constant following furosemide. These findings suggest that the renal hemodynamic changes which occurred after acetazolamide were due to tubuloglomerular feedback activation and not to a systemic hemodynamic effect, nor to differences in natriuretic potency between furosemide and acetazolamide. Previous human studies on the effects of acetazolamide on renal hemodynamics were performed in diabetic and healthy subjects[27–34]. None of them included a control group that received a diuretic with similar effect on sodium excretion as the acetazolamide group. Fractional excretion of lithium was used in this study to evaluate renal sodium handling. Lithium is freely filtrated by the glomeruli, undergoes proximal tubular reabsorption by the same transport system as sodium, and is thereafter excreted without further significant reabsorption or secretion at more distal segments of the nephron[52]. Thus, fractional lithium excretion is a marker of proximal tubular sodium handling and its increase reflects a decrease in proximal sodium reabsorption. An exception to this rule is in the case of subjects under sodium restriction, where this relationship loses its validity[53]. This was not the case in the present investigation, where participants consumed a normal sodium diet. In this study, fractional lithium excretion increased by 61% following acetazolamide, reflecting a marked decrease in proximal sodium reabsorption. This marked increase is accounted for by the combined effect of a 30% increase in lithium clearance and the resulting 21% decrease in GFR. Fractional excretion of lithium increased also following furosemide, albeit more moderately. This probably reflects the fact that some lithium reabsorption occurs in the loop of henle[31,36,54,55]. To a lesser extent, a small inhibitory acute effect of furosemide on proximal tubular sodium reabsorption[55,56] might also account for the increase in lithium clearance. However, the evidence for this effect is conflicting[57]. The change in GFR following acetazolamide was inversely correlated with baseline GFR. Hannedouche et al31 previously showed that acetazolamide decreases GFR in diabetic and healthy subjects, the change in GFR being inversely correlated with baseline GFR. Similarly, Guidi et al[30] showed that the decrease in GFR following administration of acetazolamide to healthy individuals is more pronounced in subjects with high, as compared to those with low, baseline GFR. The present investigation confirms these findings in an obese population.
Most enrolled subjects had a urinary albumin excretion rate spanning within a range associated with increased risk for cardiovascular mortality[58]. The effects of acetazolamide and furosemide on albuminuria could not be estimated owing to the diluted state of the urine during the renal function studies, resulting in undetectable urinary albumin levels. Thus we were unable to determine whether acetazolamide affects this risk marker, in addition to its effects on glomerular hyperfiltration. The effects of acetazolamide on albumin excretion rate may not necessarily match those on glomerular filtration rate. It is of interest that baseline GFR predicted the effects of acetazolamide on glomerular hyperfiltration, while albuminuria did not. Thus, further investigations are required in order to assess the effects of acetazolamide on the latter.
Glomerular hyperfiltration is one of the factors responsible for the increased prevalence of chronic kidney disease in obese sujects[18–24,59]. Thus, decreasing glomerular pressure and single nephron GFR may protect the kidney from hyperfiltration-mediated injury. The available therapeutic tools aimed at decreasing glomerular filtration pressure are antihypertensive treatment, inhibition of the renin angiotensin, weight loss[7,17] and low salt diet[13]. Ogna et al[60] recently showed in a population-based survey that overweight and obesity, and to a lesser extent salt intake, are independent predictors of glomerular hyperfiltration in the general population. These findings raise the hypothesis that decreasing salt intake may attenuate hyperfiltration. Krikken et al[13], measuring glomerular function in healthy lean and overweight subjects on a low- and a high-sodium diet, demonstrated that the change from low- to high-sodium diet led to an increase in filtration fraction in the overweight group, while no change was found in the lean group, suggesting that a low salt diet may attenuate glomerular hyperfiltration in subjects with increased adiposity. It should, however, be noted that this study was performed in normotensive subjects and that the glomerular hemodynamic response to a change in salt intake may differ in hypertensive and in normotensive subjects[61], irrespective of the obesity status.
Abating hyperfiltration through tubuloglomerular feedback activation provides a new approach to this issue, by directly acting on one of the mechanisms causing obesity-induced hyperfiltration. Non diabetic obese subjects have increased fractional sodium reabsorption in the proximal tubule[14,62,63]. This abnormal tubular sodium handling also occurs in diabetic murine models and diabetic humans[64–66]. Vallon et al[66,67] presented data obtained in a murine diabetic model suggesting that a primary proximal tubular factor leads to enhanced proximal tubular reabsorption, decreased solute delivery to the macula densa and deactivation of the tubuloglomerular feedback, with consequent glomerular hyperfiltration. The results of the present study are compatible with this concept, although do not disprove the involvement of primary vascular factors[24] in the pathogenesis of glomerular hyperfiltration. The present study is the first to investigate the effects of tubuloglomerular feedback manipulation in obese non diabetic subjects. This effect, which was previously demonstrated in diabetic and healthy subjects[27–34], is not specific to the obese population.
The short term design of the present study does not allow inferring about the effectiveness of acetazolamide in the long-term in obese subjects. Recently, Cherney et al[35] studied in type 1 diabetic subjects, the effects of an eight-week treatment with empagliflozin, a sodium glucose co-transporter 2 inhibitor that decreases proximal tubular glucose and sodium reabsorption and thus increases solute delivery to the macula densa. The authors showed that this treatment leads to a decrease in GFR in type 1 diabetic subjects with glomerular hyperfiltration. This result suggests that in this population, attenuation of glomerular hyperfiltration by tubuloglomerular feedback manipulation using a sodium glucose co-transporter 2 inhibitor is feasible during a 2-month period. Is the hypofiltrating effect of acetazolamide maintained in the long-term? Acetazolamide is rarely used as a diuretic since its long-term natriuretic effect is modest. The main factors accounting for this are the increased reabsorption of the excess solute and fluid delivered distally[68], the decrease in GFR, consequent to tubuloglomerular feedback activation, leading to a decline in the filtered sodium load[45,68] and the decrease in plasma bicarbonate and consequent reduction in glomerular filtrate bicarbonate concentration to a level lower than the maximal reabsorptive capacity of the nephron, thus abating distal delivery of sodium and bicarbonate. Does the short-term diuretic effect of acetazolamide imply an ephemeral effect on GFR? Lorenz et al68 studied knockout mice for the Na/H exchanger isoform 3 which mediates 60% of salt reabsorption in the proximal tubule. Single nephron GFR of these knockout mice was lowered owing to tubuloglomerular feedback activation. This long-term hypofiltrating effect in animals with Na/H exchanger isoform 3 deficiency suggests that chronic carbohydrase inhibition would similarly result in long-term glomerular hypofiltration.
Skott et al[69] showed that a 3-day treatment course of oral acetazolamide decreases GFR in diabetic patients with mild chronic renal insufficiency and in healthy subjects. The 24-hour urine sodium excretion, which was markedly increased at Day 1 of the treatment, as compared to the pre-treatment level, decreased at Day 3 to a level similar to the pre-treatment level. This suggests that the hypofiltrating effect of acetazolamide still persists at a time when the diuretic effect has vanished. However, no conclusive data are available regarding its long-term effect on glomerular filtration. Therefore, the long-term effects of natriuretic agents acting on the proximal tubule, such as carbohydrase inhibitors and sodium glucose co-transporter 2 inhibitors, should be investigated in hyperfiltrating subjects at risk for advanced chronic kidney disease. Of note, if the long-term hypofiltrating effect of acetazolamide was demonstrated, its side effects[70]— nephrolithiasis, lethargy, paresthesia and mainly the ineluctable metabolic acidosis—may preclude its use as a chronic treatment for glomerular hyperfiltration.
The strengths of the present study are its randomized double blind design, the use of a diuretic injected at equipotent dose, the exclusion of diabetic subjects and attainment of a similar sodium balance in the acetazolamide and furosemide groups. Its limitations are the small number of subjects involved and its short term design. The cross over design of the study enabled us to demonstrate the effect of the study drug despite a small number of participants.
In summary, manipulating tubuloglomerular feedback with acetazolamide is effective in acutely abating glomerular hyperfiltration in obese non diabetic subjects.
Acknowledgments
We thank Judith Hirsch and Iskra David for their technical assistance during the preliminary studies and Ruth Miller for her linguistic assistance in the preparation of this manuscript.
Author Contributions
Conceived and designed the experiments: AC BZ. Performed the experiments: BZ MHE YO AE SBSI. Analyzed the data: BZ AC YO BRZ UG. Wrote the paper: AC BZ MHE YO AE BRZ UG SBSI.
References
- 1. Wang Y, Chen X, Song Y, Caballero B, Cheskin LJ. (2008) Association between obesity and kidney disease: a systematic review and meta-analysis. Kidney Int 2008; 73: 19–33. pmid:17928825
- 2. Kambham N, Markowitz GS, Valeri AM, Lin J, D’Agati VD. (2001) Obesity-related glomerulopathy: an emerging epidemic. Kidney Int 59:1498–509. pmid:11260414
- 3. Bonnet F, Deprele C, Sassolas A, Moulin P, Alamartine E, Berthezène F et al. (2001) Excessive body weight as a new independent risk factor for clinical and pathological progression in primary IgA nephritis. Am J Kidney Dis 37:720–7. pmid:11273871
- 4. Kataoka H, Ohara M, Shibui K, Sato M, Suzuki T, Amemiya N et al. (2012) Overweight and obesity accelerate the progression of IgA nephropathy: prognostic utility of a combination of BMI and histopathological parameters. Clin Exp Nephrol 16:706–12. pmid:22350469
- 5. González E, Gutiérrez E, Morales E, Hernández E, Andres A, Bello I et al. (2005) Factors influencing the progression of renal damage in patients with unilateral renal agenesis and remnant kidney. Kidney Int 68:263–70. pmid:15954916
- 6. Meier-Kriesche H-U, Arndorfer JA, Kaplan B. (2002) The impact of body mass index on renal transplant outcomes: a significant independent risk factor for graft failure and patient death. Transplantation 73:70–4. pmid:11792981
- 7. Brøchner-Mortensen J, Rickers H, Balslev I. (1980) Renal function and body composition before and after intestinal bypass operation in obese patients. Scand J Clin Lab Invest 40:695–702. pmid:7280549
- 8. Reisin E, Messerli FG, Ventura HO, Frohlich ED. (1987) Renal haemodynamic studies in obesity hypertension. J Hypertens 5:397–400. pmid:3668243
- 9. Ribstein J, du Cailar G, Mimran A. (1995) Combined renal effects of overweight and hypertension. Hypertension 26:610–5. pmid:7558220
- 10. Porter LE, Hollenberg NK. (1998) Obesity, salt intake, and renal perfusion in healthy humans. Hypertension 32:144–8. pmid:9674651
- 11. Chagnac A, Weinstein T, Korzets A, Ramadan E, Hirsch J, Gafter U. (2000) Glomerular hemodynamics in severe obesity. Am J Physiol Renal Physiol 278:F817–22. pmid:10807594
- 12. Pecly IMD, Genelhu V, Francischetti EA. (2006) Renal functional reserve in obesity hypertension. Int J Clin Pract 60:1198–203. pmid:16981964
- 13. Krikken JA, Lely AT, Bakker SJL, Navis G. (2007) The effect of a shift in sodium intake on renal hemodynamics is determined by body mass index in healthy young men. Kidney Int 71:260–5. pmid:17091123
- 14. Chagnac A, Herman M, Zingerman B, Erman A, Rozen-Zvi B, Hirsh J et al. (2008) Obesity-induced glomerular hyperfiltration: its involvement in the pathogenesis of tubular sodium reabsorption. Nephrol Dial Transplant 23:3946–52. pmid:18622024
- 15. Wuerzner G, Pruijm M, Maillard M, Bovet P, Renaud C, Burnier M et al. (2010) Marked association between obesity and glomerular hyperfiltration: a cross-sectional study in an African population. Am J Kidney Dis 56: 303–12. pmid:20538392
- 16. Kwakernaak Arjan J., Zelle Dorien M., Bakker Stephan J.L., and Navis Gerjan. Central Body Fat Distribution Associates with Unfavorable Renal Hemodynamics Independent of Body Mass Index. J Am Soc Nephrol 24: 987–994, 2013. pmid:23578944
- 17. Chagnac A, Weinstein T, Herman M, Hirsh J, Gafter U, Ori Y (2003) The effects of weight loss on renal function in patients with severe obesity. J Am Soc Nephrol 14:1480–6. pmid:12761248
- 18. Praga M. Synergy of low nephron number and obesity: a new focus on hyperfiltration nephropathy. (2005) Nephrol Dial Transplant 20:2594–7. pmid:16223782
- 19. Ritz E, Koleganova N, Piecha G. (2011) Is there an obesity-metabolic syndrome related glomerulopathy? Curr Opin Nephrol Hypertens 20:44–9. pmid:21088574
- 20. Wahba IM, Mak RH. (2007) Obesity and obesity-initiated metabolic syndrome: mechanistic links to chronic kidney disease. Clin J Am Soc Nephrol 2:550–62. pmid:17699463
- 21. D’Agati VD, Markowitz GS. (2008) Supersized kidneys: Lessons from the preclinical obese kidney. Kidney Int 73:909–10. pmid:18379527
- 22. Bayliss G, Weinrauch LA, D’Elia JA. (2012) Pathophysiology of obesity-related renal dysfunction contributes to diabetic nephropathy. Curr Diab Rep 12: 440–6. pmid:22638939
- 23. Palatini P. (2012) Glomerular hyperfiltration: a marker of early renal damage in pre-diabetes and pre-hypertension. Nephrol Dial Transplant 27: 1708–14. pmid:22431709
- 24. Sasson AN, Cherney DZ. (2012) Renal hyperfiltration related to diabetes mellitus and obesity in human disease. World J Diabetes 3: 1–6. pmid:22253940
- 25. Praga M. (2005) Synergy of low nephron number and obesity: a new focus on hyperfiltration nephropathy. Nephrol Dial Transplant 20: 2594–7. pmid:16223782
- 26. Tucker BJ, Steiner RW, Gushwa LC, Blantz RC. (1978) Studies on the tubuloglomerular feedback system in the rat. The mechanism of reduction in filtration rate with benzolamide. J Clin Invest 62: 993–1004. pmid:711863
- 27. Puschett JB, Goldberg M. (1969) The relationship between the renal handling of phosphate and bicarbonate in man. J Lab Clin Med 73: 956–69. pmid:5786461
- 28. Colussi G, Rombolà G, Surian M, De Ferrari ME, Airaghi C, Benazzi E et al. (1989) Effects of acute administration of acetazolamide and frusemide on lithium clearance in humans. Nephrol Dial Transplant 4: 707–12. pmid:2510078
- 29. Skøtt P, Hommel E, Bruun NE, Arnold-Larsen S, Parving HH. (1989) The acute effect of acetazolamide on and proximal tubular reabsorption of sodium and water in normal man. Scand J Clin Lab Invest 49: 583–7. pmid:2595250
- 30. Guidi E, Colussi G, Rombolà G, Airaghi C, Minetti E, Malberti F et al. (1995) Evaluation of the tubuloglomerular feedback system in human subjects. Exp Nephrol 3: 61–4. pmid:7712144
- 31. Hannedouche T, Lazaro M, Delgado AG, Boitard C, Lacour B, Grünfeld JP. (1991) Feedback-mediated reduction in glomerular filtration during acetazolamide infusion in insulin-dependent diabetic patients. Clin Sci (Lond) 81: 457–64.
- 32. Steele TH, Manuel MA, Boner G. (1975) Diuretics, urate excretion and sodium reabsorption: effect of acetazolamide and urinary alkalinization. Nephron 14: 49–61. pmid:235746
- 33. Guidi E, Cozzi MG, Minetti EE, Civati G, Busnach G, Brando B. (2001) Effect of familial hypertension on glomerular hemodynamics and tubulo-glomerular feedback after uninephrectomy. Am J Hypertens 14: 121–128 pmid:11243302
- 34. Slomowitz LA, Bergamo R, Hirschberg R, Grosvenor M, Kopple JD. (1996) Enalapril attenuates the renal hemodynamic effect of acetazolamide in patients with diabetes mellitus: possible implications for tubuloglomerular feedback. Am J Nephrol 16: 315–9. pmid:8739285
- 35. Cherney DZI, Perkins BA, Soleymanlou N, Maione M, Lai V, Lee A et al. (2014) Renal hemodynamic effect of sodium-glucose cotransporter 2 inhibition in patients with type 1 diabetes mellitus. Circulation 129: 587–97. pmid:24334175
- 36. Wright FS, Schnermann J. (1974) Interference with feedback control of by furosemide, triflocin, and cyanide. J Clin Invest 53: 1695–708. pmid:4830232
- 37. Gutsche HU, Brunkhorst R, Müller-Ott K, Franke H, Niedermayer W. (1984) Effect of diuretics on the tubuloglomerular feedback response. Can J Physiol Pharmacol 62: 412–7. pmid:6733586
- 38. Johnston PA, Kau ST. (1992) The effect of loop of Henle diuretics on the tubuloglomerular feedback mechanism. Methods Find Exp Clin Pharmacol 14: 523–9. pmid:1287376
- 39. Masud T, Manatunga A, Cotsonis G, Mitch WE. (2002) The precision of estimating protein intake of patients with chronic renal failure. Kidney Int 62: 1750–6. pmid:12371976
- 40. Persson AE, Wright FS. (1982) Evidence for feedback mediated reduction of during infusion of acetazolamide. Acta Physiol Scand 114: 1–7. pmid:7136739
- 41. Mathisen O, Raeder M, Sejersted OM, Kiil F. (1976) Effect of acetazolamide on glomerular balance and renal metabolic rate. Scand J Clin Lab Invest 36: 617–25. pmid:1019572
- 42. Cogan MG, Rector FC. (1982) Determinants of proximal bicarbonate, chloride, and water reabsorption during carbonic anhydrase inhibition. Am J Physiol 242: F274–84. pmid:7065180
- 43. Yeyati NL, Altenberg GA, Adrogué HJ. (1992) Mechanism of acetazolamide-induced rise in renal vascular resistance assessed in the dog whole kidney. Ren Physiol Biochem 15: 99–105. pmid:1375769
- 44.
Munger KA, Kost CK, Brenner BM M DA. The renal circulations and glomerular ultrafiltration. In: Taal MW, Chertow GM, Marsden PA, Skorecki K, Yu ASL, Brenner BM, editors. Brenner and Rector’s the Kidney, Ninth Edition. Philadelphia, Saunders-Elsevier Inc, 2012; pp 94–137.
- 45. Leyssac PP, Karlsen FM, Skøtt O. (1991) Dynamics of intrarenal pressures and glomerular filtration rate after acetazolamide. Am J Physiol. 261(1 Pt 2): F169–78. pmid:1858899
- 46. Birtch AG. (1967) Redistribution of renal blood flow produced by furosemide and ethacrynic acid. Circ Res 21: 869–78. pmid:6078147
- 47. Puschett JB and Goldberg M. (1968) The acute effects of furosemide on acid and electrolyte excretion in man. J Lab Clin Med 71: 666–77. pmid:5651365
- 48. Ludens JH. (1968) Enhancement of renal blood flow by furosemide. J Pharmacol Exp Ther 163: 456–460. pmid:5681183
- 49. Gerber JG and Nies AS. (1980) Furosemide-induced vasodilation—importance of the state of hydration and filtration. Kidney Int 18: 454–459. pmid:7230610
- 50. Gerber JG. (1983) Role of prostaglandins in the hemodynamic and tubular effects of furosemide. Fed Proc 42: 1707–10. pmid:6403387
- 51. Andreasen F, Lauridsen IN, Hansen FA, Christensen S, Steiness E. (1989) Dose dependency of furosemide-induced sodium excretion. J Pharmacol Exp Ther 248: 1182–8. pmid:2703969
- 52. Boer WH, Fransen R, Shirley DG, Walter SJ, Boer P, Koomans HA. (1995) Evaluation of the lithium clearance method: direct analysis of tubular lithium handling by micropuncture. Kidney Int 47: 1023–30. pmid:7783399
- 53. Boer WH, Koomans HA, Beutler JJ, Gaillard CA, Rabelink AJ, Dorhout Mees EJ. (1989) Small intra- and large inter-individual variability in lithium clearance in humans. Kidney Int 35: 1183–8. pmid:2504986
- 54. Koomans HA, Boer WH, Dorhout Mees EJ. (1989) Evaluation of lithium clearance as a marker of proximal tubule sodium handling. Kidney Int 36:2–12. pmid:2681925
- 55. Dorhout Mees EJ, Beutler JJ, Boer WH, Koomans HA. (1990) Does lithium clearance reflects distal delivery in humans? Analysis with furosemide infusion. Am J Physiol 258: F1100–F1104. pmid:2330975
- 56. Radtke HW, Rumrich G, Kinne-Saffran E and Ullrich KJ. (1972) Dual action of acetazolamide and furosemide on proximal volume absorption in the rat kidney. Kidney Int 1: 100–105. pmid:4275896
- 57. Seely JF, Dirks JH. (1977) Site of action of diuretic drugs. Kidney Int 11: 1–8. pmid:320370
- 58. Hillege HL, Fidler V, Diercks GF, van Gilst WH, de Zeeuw D, van Veldhuisen DJ et al; Prevention of Renal and Vascular End Stage Disease (PREVEND) Study Group. (2002) Urinary albumin excretion predict cardiovascular and noncardiovascular mortality in general population. Circulation: 106: 1777–82. pmid:12356629
- 59. Wickman C, Kramer H. (2013) Obesity and kidney disease: potential mechanisms. Semin Nephrol 33: 14–22. pmid:23374890
- 60. Ogna A, Forni Ogna V, Bochud M, Guessous I, Paccaud F, Burnier M et al. (2015) Association between obesity and glomerular hyperfiltration: the confounding effect of smoking and sodium and protein intakes. Eur J Nutr 2015 May14. [Epub ahead of print].
- 61. Chiolero A, Maillard M, Nussberger J, Brunner HR, Burnier M. (2000) Proximal sodium reabsorption: An independent determinant of blood pressure response to salt. Hypertension 36: 631–7. pmid:11040249
- 62. Strazzullo P, Barba G, Cappuccio FP, Siani A, Trevisan M, Farinaro E et al. (2001) Altered renal sodium handling in men with abdominal adiposity: a link to hypertension. J Hypertens 19: 2157–64. pmid:11725158
- 63. Strazzullo P, Barbato A, Galletti F, Barba G, Siani A, Iacone R et al. (2006) Abnormalities of renal sodium handling in the metabolic syndrome. Results of the Olivetti Heart Study. J Hypertens 24: 1633–9. pmid:16877967
- 64. Bank N, Aynedjian HS. (1990) Progressive increases in luminal glucose stimulate proximal sodium absorption in normal and diabetic rats. J Clin Invest 86: 309–16. pmid:2365820
- 65. Hannedouche TP, Delgado AG, Gnionsahe DA, Boitard C, Lacour B, Grunfeld JP. (1990) Renal hemodynamics and segmental tubular reabsorption in early type 1 diabetes. Kidney Int 37: 1126–33. pmid:2188030
- 66. Vallon V, Richter K, Blantz RC, Thomson S, Osswald H. (1999) Glomerular hyperfiltration in experimental diabetes mellitus: potential role of tubular reabsorption. J Am Soc Nephrol 10: 2569–76. pmid:10589696
- 67. Vallon V, Blantz RC, Thomson S. (2003) Glomerular hyperfiltration and the salt paradox in early [corrected] type 1 diabetes mellitus: a tubulo-centric view. J Am Soc Nephrol 14: 530–7. pmid:12538755
- 68. Lorenz JN, Schultheis PJ, Traynor T, Shull GE, Schnermann J. (1999) Micropuncture analysis of single-nephron function in NHE3-deficient mice. Am J Physiol. 277(3 Pt 2): F447–53. pmid:10484528
- 69. Skøtt P, Hommel E, Bruun NE, Arnold-Larsen S, Parving HH. (1988) Effects of acetazolamide on kidney function in type 1 (insulin-dependent) diabetic patients with diabetic nephropathy. Diabetologia 31: 806–10. pmid:3234635
- 70. Reiss WG, Oles KS. (1996) Acetazolamide in the treatment of seizures. Ann Pharmacother. 30: 514–9. pmid:8740334