REVIEW

1
Renal Stress Testing in the Assessment 52
2 of Kidney Disease 53
3 54
4 Q2 Lakhmir S. Chawla1,2 and Claudio Ronco3,4 55
5 1 56
Q21 Department of Medicine, Veterans Affairs Medical Center, Washington, DC, USA; 2Department of Medicine, George Wash-
6 ington University, Washington, DC, USA; 3Department of Nephrology, Dialysis and Transplantation, San Bortolo Hospital, 57
7 Q3 Vicenza, Italy; and 4International Renal Research Institute of Vicenza, Vicenza, Italy 58
8 59
9 60
As part of human evolutionary development, many human organ systems have innate mechanisms to
10 adapt to increased “work demand” or stress. This reserve capacity can be informative and is used 61
11 commonly in cardiology to assess cardiac function (e.g., treadmill test). Similarly, the kidney possesses 62
12 reserve capacity, which can be demonstrated in at least 2 renal domains: glomerular and tubular. When 63
13 appropriate stimulants are used, healthy patients with intact kidneys can significantly increase their 64
glomerular filtration rate and their tubular secretion. This approach has been used to develop diagnostics 65
14 for the assessment of renal function. This article reviews both glomerular and tubular kidney stress tests
15 and their respective diagnostic utility. 66
16 67
KI Reports (2016) -, -–-; http://dx.doi.org/10.1016/j.ekir.2016.04.005
17 KEYWORDS: acute kidney injury; chronic kidney disease; furosemide stress test; glomerular filtration; kidney stress
68
18 Q4 test; maximum GFR; protein; renal functional reserve; subclinical AKI; tubular function 69
19 ª 2016 International Society of Nephrology. Published by Elsevier Inc. This is an open access article under the CC 70
20 BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 71
21 72
22 73
23 74
s part of human evolutionary development, many may diverge. An assessment of both glomerular and
24
25
A human organ systems have innate mechanisms to
adapt to increased “work demand” or stress. At rest,
tubular function may be more informative than just 1
of these domains. Glomerular reserve testing has been
75
76
26 77
organ systems operate at baseline capacity, and this well established but is used infrequently in routine
27 78
capacity can be increased to a certain maximum ca- clinical care. Tubular function diagnostic testing is
28 79
pacity. A familiar example of this concept is cardiac relatively new and in its clinical “infancy.” However,
29 80
function. In a healthy person at rest, cardiac output is tubular assessment appears to hold significant promise
30 81
approximately 5.0 liters/min. However, when a healthy for the assessment of both chronic and acute kidney
31 82
person exercises, the cardiac output can double or even disease.
32 83
triple. Similarly, the kidney has reserve capacity of its
33 Renal Functional Reserve–Glomerular 84
multiple physiological functions (Figure 1). The ability
34 Because of the common use of estimated glomerular 85
to test the reserve of an organ system is often an
35 filtration rate (GFR) equations, there is a tendency for 86
excellent diagnostic tool to uncover subclinical disease
36 non-nephrologists to think that the GFR is a constant. 87
(e.g., treadmill test). Similarly, stress testing of the
37 In fact, the actual GFR changes throughout the day, 88
kidney appears to generate insights into the presence
38 particularly after meals, based on physiological needs.1 89
or absence of kidney disease and parenchymal loss due
39 One of the kidney’s primary roles is to effectively 90
to injury and potentially fibrosis. The 2 main domains
40 remove nitrogenous waste, and as a consequence, the 91
of kidney stress testing are glomerular and tubular. In a
41 consumption and metabolism of protein results in an 92
healthy kidney, these 2 components of the nephron
42 increase in GFR.2 GFR can also be increased through 93
work in concert. However, when the kidney is diseased
43 other mechanisms that work along the protein meta- 94
or injured, the glomerular and tubular function may be
44 bolic pathway. For instance, an i.v. infusion of amino 95
affected equally, or their form and functional capacity
45 acids will result in an increase in GFR.3 This increase in 96
46 GFR over baseline GFR is known as renal functional 97
47 reserve–glomerular (RFR-G).4 Protein ingestion, 98
48 Correspondence: Lakhmir S. Chawla, Veterans Affairs Medical particularly red meat, is a potent stimulant for 99
49 Q5 Q6 Center, 50 Irving Street, Washington, DC, USA. E-mail:
increasing GFR, and the teleologic explanation is likely 100
minkchawla@gmail.com
50 related to an adaptive response to increased protein in 101
Received 8 March 2016; revised 25 April 2016; accepted 26 April
51 2016 the diet.5 102

Kidney International Reports (2016) -, -–- 1

REV 5.4.0 DTD
EKIR7_proof
14 May 2016
11:29 am
ce

 REVIEW LS Chawla and C Ronco: Renal Stress Testing in the Assessment of Kidney Disease Q1

103 157
104 158
105 159
106 160
107 161
108 162
109 163
110 164
111 165

web 4C=FPO
112 166
web 4C=FPO

113 167
114 168
115 169
Figure 2. Relationship between glomerular filtration rate (GFR) and
116 170
Figure 1. Comparison of stressors in the heart and kidney. C-R, serum creatinine changes. RFR-G, renal functional reserve–glomerular.
117 Q18 cardiorenal. 171
118 seen in patients who donate a kidney; despite a halving 172
119 of their nephron mass, their serum creatinine and 173
Bosch and colleagues first described glomerular
120 calculated eGFR are “normal.”10 Therefore, when renal 174
functional reserve (RFR-G) in 1983.4 In this seminal
121 disease becomes apparent due to an elevated serum 175
paper, Bosch and colleagues demonstrated that the
122 creatinine, this occurs only after the residual nephrons 176
consumption of protein, not carbohydrates or fat, re-
123 can no longer compensate for the functional loss 177
sults in a substantial increase in GFR in patients with
124 (Figure 3).8 178
healthy kidneys. Multiple subsequent studies have
125 179
confirmed these findings. The clinical implications of Renal Functional Reserve (RFR-G)
126 180
RFR-G will be reviewed. Normal subjects display a significant increase in GFR 1
127 181
128 Baseline (Unstressed) GFR or 2 hours after an acute protein load (1–1.2 g/kg) over 182
129 GFR is normally utilized as a surrogate of kidney their baseline GFR. The difference between peak or 183
130 function in healthy subjects as well as in patients with “maximum” GFR (maxGFR) and baseline GFR describes 184
131 kidney disease. Studies in healthy subjects under the the renal functional reserve of glomerular function 185
132 age of 50 have identified the average baseline normal (RFR-G). Fliser and colleagues11 compared the baseline 186
133 values of GFR to be between 100 and 130 ml/min per and maxGFR in young and elderly healthy subjects and 187
134 1.73 m2.6 Evaluation of population-wide “normal” found that RFR was significantly lower in elderly than 188
135 values is useful, but the concept of “normal” GFR in in young healthy individuals while virtually all base- 189
136 the single individual is more nuanced. It is important to line GFR values of elderly were within the reference 190
137 recognize that a person’s GFR at any given point in range. The renal reserve as assessed by RFR-G is a 191
138 time will vary in relation to the physiological demands measure of the kidney’s capacity to increase GFR by a 192
139 of dietary and hemodynamic conditions. Baseline value combination of nephron recruitment and increases in 193
140 for GFR (bGFR) also depends on age, sex, and body renal blood flow coupled with hyperfiltration.12–15 194
141 size, with considerable variation among healthy in- 195
142 dividuals. Overall, the average daily GFR is remarkably 196
143 stable over years, although there is an age-related 197
144 decline in GFR physiologically by 0.8 ml/min per 198
145 1.73 m2 per year, after the age of 30 years.6,7 199
146 In general, serum creatinine tends to remain rela- 200
147 tively normal even in the presence of kidney damage, 201
148 until approximately 50% of nephrons are lost or 202
149 simply when bGFR approaches 60 ml/min per 1.73 m2 203
150 (Figure 2).8 For this reason serum creatinine cannot be 204
151 considered an accurate marker of renal function when 205
web 4C=FPO

152 GFR is above 60 ml/min per 1.73 m2. Similarly, GFR 206
153 estimation (eGFR) by creatinine-derived equations (e.g., 207
154 MDRD9) cannot be considered a sensitive index for 208
155 early detection of renal disease during the early phases 209
156 of parenchymal damage. A good example of this can be Figure 3. Variation in baseline glomerular filtration rate (GFR). Q19 210
2 Kidney International Reports (2016) -, -–-

REV 5.4.0 DTD
EKIR7_proof
14 May 2016
11:29 am
ce

 LS Chawla and C Ronco: Renal Stress Testing in the Assessment of Kidney Disease REVIEW

211 The stimulus to tap into this reserve capacity can overall increase in blood flow is the main mechanism 265
212 arise from adaptive physiological needs like pregnancy rather than a temporary hemodynamic perturbation in 266
213 or the presence of a solitary kidney. Utilization of RFR the afferent/efferent tone and equilibrium (C. Ronco Q7 267
214 in non-disease states is best illustrated by pregnancy. and colleagues, unpublished data). 268
215 In pregnancy, GFR significantly increases during each Stress testing with a protein load is the definitive 269
216 trimester, such that there is a significant rise in bGFR way to assess for the loss of RFR-G, but the significance 270
217 from first to last trimester. Studies done on normal of renal reserve is not just a diagnostic consideration. It 271
218 pregnant women in each trimester have shown a pro- is important to recognize that the loss of renal reserve 272
219 gressive increase of baseline GFR with a parallel may also manifest as a loss in autoregulation capacity in 273
220 reduction of RFR due to its progressive utilization.13 the kidney. This loss of autoregulation may increase 274
221 MaxGFR in normal pregnant women, however, does the vulnerability of those patients with CKD to volume 275
222 not change. However, pathological states can also depletion and certain nephrotoxins (e.g., nonsteroidal 276
223 initiate processes that increase GFR above the normal anti-inflammatory drugs). Population studies suggest 277
224 baseline. Primary hyperfiltration in kidney disease has that increased creatinine variability, which could be 278
225 been shown in patients with diabetes mellitus, poly- due to the loss of autoregulation, predicts progression 279
226 cystic kidney disease, secondary focal segmental glo- to end-stage renal disease.20 280
227 merulosclerosis, sickle cell anemia, high-altitude renal 281
228 syndrome, obesity, hypertension, nephrotic syn- Kidney Stress Test Glomerular in Clinical Q8 282
229 dromes, and glomerulonephritis.16 In physiological Practice and Future Research 283
230 states of diminished RFR, the observed hyperfiltration Assuming RFR-G represents the difference between 284
231 is likely due to recruitment of more nephron units, maximal filtration capacity of the kidney (maxGFR) and 285
232 whereas in pathological states, hyperfiltration is prob- the baseline GFR (bGFR), a protein load is the basis of a 286
233 ably due to an increase in single nephron filtration kidney stress test forcing the kidneys to utilize the 287
234 fraction. This, in part, is the basis of angiotensin II entire filtration capacity. This technique can be used to 288
235 blockade in chronic kidney disease (CKD), and is often “reveal” subclinical kidney disease. MaxGFR and bGFR 289
236 demonstrable by a drop in GFR when angiotensin- assessment with protein loading has been extensively 290
237 converting enzyme inhibitors are given to patients studied and can be used in the clinic to assess RFR-G in 291
238 with CKD. patients with kidney disease.4,8,13,14 Because dietary 292
239 A current limitation on the use of RFR-G assessment protein raises GFR, establishing bGFR is important 293
240 is that these assessments have not been conducted in when attempting to assess RFR-G; developing stan- 294
241 large cross-sectional cohorts, thus the population dardized protocols to accomplish this is an important 295
242 variability of the RFR response is not known. Several research recommendation. There is another approach 296
243 investigators have estimated RFR by measuring the that would allow single GFR assessment instead of 297
244 difference between protein-stimulated GFR and base- having to conduct a baseline and a stimulated stress 298
245 line GFR after a protein load.4 In a separate study,17 test. In this approach, the maxGFR would be assessed 299
246 Bosch and colleagues demonstrated the estimation of among healthy patients across a wide age range and 300
247 RFR by a short-term oral protein loading method. De ethnic range and in both genders. Once these data were 301
248 Nicola and colleagues18 demonstrated that the estima- known, then normative values could be determined for 302
249 tion of RFR can be assessed by amino acid infusion. maxGFR. These data would be used for diagnostic 303
250 Numerous mechanisms have been hypothesized for purposes for patients who underwent a kidney stress 304
251 the increase in GFR after protein load. In their study, test to achieve maxGFR. Those patients who could not 305
252 Woods and colleagues19 hypothesized that protein achieve the appropriate age, gender, and race maxGFR 306
253 loading increases GFR because digested protein raises metrics could be referred for further work-up. 307
254 plasma amino acid levels, which are then filtered at the As part of the future research plan, the safety of 308
255 glomerulus, thereby stimulating proximal tubular ab- repeated protein loading in patients with CKD should 309
256 sorption. In addition, filtered amino acids change the also be assessed. Since protein is a stimulant for GFR, 310
257 sensitivity of macula densa sensing mechanisms, the effects of repeated protein loading in patients with 311
258 causing release of nitric oxide and prostaglandins CKD is unknown. The exposure of repeated high levels 312
259 locally resulting in vasodilation, increasing renal blood of protein in patients with CKD might be deleterious, 313
260 flow and GFR. In our own laboratory we analyzed the but might also “condition” the kidney as well and 314
261 response to acute protein loading, and we detected an stimulate restorative or protective effects—this concept 315
262 increase in GFR proportional to an increase in renal should be studied further. 316
263 blood flow with a constant of filtration fraction. This The idea of assessing renal reserve has been present 317
264 observation seems to support the hypothesis that an for decades, but is infrequently used in clinical 318
Kidney International Reports (2016) -, -–- 3

REV 5.4.0 DTD
EKIR7_proof
14 May 2016
11:29 am
ce

 REVIEW LS Chawla and C Ronco: Renal Stress Testing in the Assessment of Kidney Disease

319 practice, whereas the cardiac stress test is used tubule’s capacity to secrete acid or sodium can be 373
320 routinely. Why is this the case? In our view, the simple assessed via acid or salt loading. The tubule’s concen- 374
321 reason is that cardiologists perceive that they can trating capacity can be assessed via water deprivation 375
322 intervene on patients with diminished cardiac reserve or exogenous administration of desmopressin (DDAVP). 376
323 (e.g., heart failure treatment), whereas the nephrology Among these different techniques, thus far the primary 377
324 community may not feel that an intervention is avail- methodology to assess tubular functional capacity in 378
325 able, and therefore may be unwilling to perform an patients with kidney disease has been via tubular 379
326 extra test. We hypothesize that patients with loss of secretion of either creatinine or an exogenous drug 380
327 renal reserve are at risk for CKD. Future trials should (e.g., furosemide). 381
328 assess whether early identification of diminished renal 382
329 reserve can reliably predict the risk of progression to Tubular Function Assessment in CKD 383
330 CKD. If this can be shown, early screening of renal The first studies of tubular functional capacity in pa- 384
331 reserve may prompt early intervention and forestall the tients with CKD utilized the difference between creati- 385
332 development of CKD. nine clearance and inulin clearance as an assessment of 386
333 tubular function. Herrera and colleagues24 developed an 387
334 Tubular Function Assessment in Kidney Disease elegant study to demonstrate the potential use of tubular 388
335 The renal tubule portion of the nephron is tasked with secretion. In this study, the investigators took 3 cohorts 389
336 an enormous portfolio of responsibilities. Chief among of patients: normal, renal allograft donors (uni-neph- 390
337 those chores are the handling of electrolytes, water, rectomized), and CKD. In these subjects, baseline creat- 391
338 Q9 and amino acids, catabolism of various proteins, and inine clearance and inulin-based GFR were measured and 392
339 the active secretion of endogenous and exogenous then reassessed after a protein meal. They found that 393
340 acids. Tubular function assessment may be more both healthy patients and patients with CKD were able to 394
341 informative than glomerular reserve in patients who increase their inulin-measured GFR in response to a 395
342 already have advanced kidney disease. When patients protein meal: as expected, healthy patients could in- 396
343 do not have obvious kidney disease, the loss of crease their GFR after stimulation much more than CKD 397
344 glomerular reserve (RFR-G) can be an indicator of loss patients. Similarly, healthy patients were able to increase 398
345 of nephron mass.8 However, once a patient has kidney their tubular secretion of creatinine (TScr), but CKD 399
346 injury or disease, glomerular reserve is already sub- patients were unable to increase their TScr. When all 3 400
347 stantially reduced and therefore is less informative. groups of patients were compared, uni-nephrectomized 401
348 In patients with decreased GFR, tubular function patients were able to increase their TScr (but to a lesser 402
349 appears to be more variable. One reason for this degree than normal healthy subjects), while CKD pa- 403
350 observation may be due to renal fibrosis. During the tients were unable to increase their TScr. These data are 404
351 assessment of a kidney disease by tissue biopsy, the consistent with previous studies that show that patients 405
352 level of interstitial fibrosis is one of the strongest pre- with CKD maintain some glomerular renal reserve at all Q11 406
353 dictors of renal survival.21,22 Interstitial fibrosis can levels of baseline GFR.8 In addition, this study demon- 407
354 represent scarred tubules that are fibrosed, or the strated that CKD patients likely operate at near their 408
355 secretion of matrix that fills in between the nephrons, maximum TScr, and thus are less able to increase their 409
356 Q10 or both of these. However, because CKD is generally TScr when challenged with a protein meal. 410
357 marked by a reduction of kidney size, this makes the In a second trial, this same group of investigators 411
358 possibility of “extra” matrix an unlikely sole expla- assessed TScr by infusing i.v. creatinine into normal 412
359 nation for fibrosis (an exception to this would be subjects and kidney donors.25 They found that creati- 413
360 multiple myeloma). In most forms of kidney disease, nine infusion did not increase GFR, and that an infu- 414
361 the kidneys shrink and become more echogenic over sion of i.v. creatinine resulted in an increased TScr in 415
362 time. Based on this observation, we believe that it is healthy patients, but not in kidney donors. Thus, a 416
363 more likely that diseased tubules are replaced by ma- tubular functional assessment with a challenge of i.v. 417
364 trix and fibrosis. In order to test the notion that tubular creatinine had the capacity to reveal the subjects with 418
365 function may identify patients who are at increased decreased nephron mass (i.e., kidney donors) who 419
366 risk for worse outcomes, various studies in patients otherwise had normal serum creatinine levels. 420
367 with both acute and chronic kidney disease have been In aggregate, preliminary studies suggest that 421
368 conducted to determine the utility of tubular secretion tubular stress tests that measure the secretory capacity 422
369 capacity to predict outcomes.23,24 of the renal tubule are informative and predictive of 423
370 Different aspects of tubular function can be inter- outcomes.23–26 However, it should be noted that 424
371 rogated in various ways depending on what feature of tubular stress tests remain research tools and have not 425
372 tubular function is being assessed. For instance, the yet been deployed into the clinic for CKD. 426
4 Kidney International Reports (2016) -, -–-

REV 5.4.0 DTD
EKIR7_proof
14 May 2016
11:29 am
ce

 LS Chawla and C Ronco: Renal Stress Testing in the Assessment of Kidney Disease REVIEW

427 Tubular Assessment in Acute Kidney Injury test to be safe and valid, and any volume losses induced 481
428 The aforementioned studies in CKD used creatinine by the diuresis should be replaced. 482
429 secretion to assess tubular functional assessment. In A version of the FST has also been analyzed in pa- 483
430 patients with acute kidney injury (AKI), many factors, tients with advanced-stage AKI requiring renal 484
431 including lack of steady state, increased catabolism, replacement therapy to determine whether a stan- 485
432 and concurrent medications that interfere with creati- dardized furosemide challenge can predict renal re- 486
433 nine secretion, preclude the use of TScr as a reliable covery. Van der Voort and colleagues reported that a 487
434 measure of tubular secretion. One approach uses i.v. standardized 4-hour infusion of furosemide was also an 488
435 furosemide to assess tubular function. Furosemide, a excellent predictor of renal recovery.26,33 This analysis Q14 489
436 loop diuretic, has pharmacokinetic properties that was a post hoc assessment of a randomized clinical trial, 490
437 make it an appealing functional tool. In contrast to which compared a 4-hour infusion of furosemide to 491
438 other drugs cleared by the kidney, furosemide is not placebo as an intervention to promote renal recovery in 492
439 effectively filtered by the glomerulus. As an organic patients who are on continuous renal replacement 493
440 acid, furosemide is tightly bound to albumin and gains therapy. In this post hoc analysis, the authors assessed 494
441 access to the tubular lumen by active secretion via the the intervention arm of the trial (i.e., the patients 495
442 human organic anion transporter system in the prox- randomized to furosemide) and found that the mean 496
443 imal convoluted tubule.27,28 Once in the tubular lumen, urine output was much higher in patients destined to 497
444 furosemide blocks luminal cation–chloride cotransport recover (654 ml vs. 48 ml, P ¼ 0.007) and had a diag- 498
445 throughout the thick ascending limb of Henle, thereby nostic performance receiver operating characteristic 499
446 preventing sodium reabsorption and resulting in area under the curve of 0.84. These 2 studies demon- 500
447 natriuresis and increased urine flow.29–31 Based on strate that the urine output response to furosemide is 501
448 these properties, furosemide-induced increases in urine informative about renal tubular function throughout 502
449 output represent a methodology to assess the integrity the phases of AKI (progression and recovery). Another 503
450 of the renal tubular function in the setting of AKI. This advantage of the FST is that it does not just measure the 504
451 methodology was developed by Chawla and col- tubule’s secretion capacity, but is actually an assess- 505
452 leagues23 and is referred to as the furosemide stress test ment of integrated renal function34 (Figure 4). In order 506
453 (FST). for furosemide to increase urine output, furosemide 507
454 The FST has been prospectively assessed in a single must be actively secreted into the proximal lumen, and 508
455 cohort study of critically ill patients with AKI and was the thick ascending limb, luminal patency, and col- 509
456 found to have good diagnostic performance. In that lecting duct function must all be intact.35 Because the 510
457 study, Chawla and colleagues23 administered a stan- FST requires an intact nephron for full function, the 511
458 dard dose of i.v. furosemide (1.0–1.5 mg/kg) to criti- FST does readily identify the location of the defect in 512
459 cally ill patients with Kidney Disease: Improving cases in which the FST response is poor. 513
460 Global Outcomes (KDIGO) stage I or stage II AKI and The aforementioned studies of FST are of modest size 514
461 then assessed the urine output response. This study and are currently undergoing larger-scale validation 515
462 showed that the 2-hour urine output response to a (NCT 01275729). However, the FST is based on the 516
463 furosemide challenge was able to predict progression to bedside practice of many clinicians, which involves 517
464 KDIGO stage III within 14 days with a receiver oper- challenging patients with a loop diuretic and assessing 518
465 ating characteristic area under the curve of 0.87 the clinical response. The FST, as currently devised, is 519
466 Q12
3
Q13 (standard error 0.05). At a cutoff of 200 cm at 2 hours, simply a framework around this common bedside 520
467 the sensitivity and specificity of the FST were 87.1% 521
468 and 84.1%, respectively.23 In a follow-up study of the 522
469 same cohort, the same research group showed that FST 523
470 performed better than known AKI biomarkers. 524
471 Importantly, the follow-up study demonstrated that 525
472 the FST performance improves when utilized in pa- 526
473 tients with increased levels of AKI biomarkers.32 These 527
474 data suggest that the combination of AKI biomarkers 528
web 4C=FPO

475 with tubular functional assessment is informative, and 529

476 can be used at the bedside to assist clinicians in 530
477 assessing the severity of AKI. It remains unclear 531
478 whether the FST reveals the severity of AKI, or the loss 532
479 of tubular functional capacity. An important caveat to Figure 4. Furosemide urinary response tests tubular integrity. TAL, 533
480 the FST is that the subject must be euvolemic for the thick ascending limb. Q20 534
Kidney International Reports (2016) -, -–- 5

REV 5.4.0 DTD
EKIR7_proof
14 May 2016
11:29 am
ce

 REVIEW LS Chawla and C Ronco: Renal Stress Testing in the Assessment of Kidney Disease

535 practice. The FST is also being assessed by the 0 by 25 7. Davies DF, Shock NW. Age changes in glomerular filtration 589
536 initiative spearheaded by the International Society of rate, effective renal plasma flow, and tubular excretory ca- 590
537 Nephrology. In an austere medical environment, simple pacity in adult males. J Clin Invest. 1950;29:496–507.
591
538 diagnostic tools like serum urea and creatinine are not 8. Barai S, Gambhir S, Prasad N, et al. Functional renal reserve 592
capacity in different stages of chronic kidney disease.
539 readily available. Thus, the use of FST in euvolemic Nephrology. 2010;15:350–353.
593
540 patients with oliguria may allow a thoughtful way to 594
9. Levey AS, Bosch JP, Lewis JB, et al. A more accurate method
541 triage patients who may need more advanced care. to estimate glomerular filtration rate from serum creatinine: a
595
542 Because furosemide is inexpensive and available new prediction equation. Modification of Diet in Renal Dis- 596
543 worldwide, this physiological assessment may allow for ease Study Group. Ann Intern Med. 1999;130:461–470. 597
544 broader use of this diagnostic approach. 10. Tsuda A, Ishimura E, Uedono H, et al. Comparison of the 598
545 estimated glomerular filtration rate (eGFR) in diabetic pa- 599
546 Summary tients, non-diabetic patients and living kidney donors. Kidney 600
547 Kidney stress testing can be accomplished by assessing Blood Press Res. 2016;41:40–47.
601
548 glomerular and tubular domains. These assessments are 11. Fliser D, Zeier M, Nowack R, Ritz E. Renal functional reserve in 602
549 safe and relatively inexpensive and can be done at the healthy elderly subjects. J Am Soc Nephrol. 1993;3:1371–1377.
603
550 bedside or in the clinic. Importantly, these assessments 12. Bosch JP, Lew S, Glabman S, Lauer A. Renal hemodynamic
604
have been shown to be informative in both acute and changes in humans. Response to protein loading in normal
551 and diseased kidneys. Am J Med. 1986;81:809–815.
605
552 chronic kidney disease. However, neither of these 606
stress tests is currently used routinely at the bedside or 13. Ronco C, Brendolan A, Bragantini L, et al. Renal functional
553 reserve in pregnancy. Nephrol Dial Transplant. 1988;3:157–161. 607
554
Q15 the clinic. Assessment of RFR-G can and, in the opinion 608
14. Pecly IM, Genelhu V, Francischetti EA. Renal functional
555 of these authors, should be used to reveal the loss of 609
reserve in obesity hypertension. Int J Clin Pract. 2006;60:
556 RFR in patients at risk for kidney disease. Tubular 1198–1203. 610
557 functional testing has been less developed; early 611
15. Zitta S, Stoschitzky K, Zweiker R, et al. Dynamic renal function
558 studies demonstrate good diagnostic performance, but testing by compartmental analysis: assessment of renal 612
559 large validation studies are still needed. Because functional reserve in essential hypertension. Nephrol Dial 613
560 tubular testing may have the capacity to assess multiple Transplant. 2000;15:1162–1169.
614
561 anatomic domains of the nephron, we believe that 16. Cachat F, Combescure C, Cauderay M, et al. A systematic 615
562 noninvasive kidney stress testing may allow clinicians review of glomerular hyperfiltration assessment and defini-
616
to phenotype, prognosticate, and better follow patients tion in the medical literature. Clin J Am Soc Nephrol. 2015;10:
563 382–389. 617
564 with kidney disease. Further research into the appro- 618
priate use of these diagnostic techniques is warranted. 17. Bosch JP, Lauer A, Glabman S. Short-term protein loading in
565 assessment of patients with renal disease. Am J Med. 619
566 1984;77:873–879. 620
567
Q16 DISCLOSURE 621
18. De Nicola L, Blantz RC, Gabbai FB. Renal functional reserve in
568 All the authors declared no competing interests. treated and untreated hypertensive rats. Kidney Int. 1991;40: 622
569 406–412. 623
570 REFERENCES 19. Woods LL. Mechanisms of renal hemodynamic regulation in 624
571 1. Koopman MG, Koomen GC, Krediet RT, et al. Circadian response to protein feeding. Kidney Int. 1993;44:659–675. 625
rhythm of glomerular filtration rate in normal individuals. Clin
572 20. O’Hare AM, Batten A, Burrows NR, et al. Trajectories of kid- 626
Sci. 1989;77:105–111.
573 ney function decline in the 2 years before initiation of long-
627
2. Brenner BM, Meyer TW, Hostetter TH. Dietary protein intake term dialysis. Am J Kidney Dis. 2012;59:513–522.
574 and the progressive nature of kidney disease: the role of
628
21. Bohle A, Grund KE, Mackensen S, Tolon M. Correlations
575 hemodynamically mediated glomerular injury in the patho- 629
between renal interstitium and level of serum creatinine.
576 genesis of progressive glomerular sclerosis in aging, renal Morphometric investigations of biopsies in perimembranous 630
577 ablation, and intrinsic renal disease. N Engl J Med. 1982;307: glomerulonephritis. Virchows Arch A Pathol Anat Histol. 631
578 652–659. 1977;373:15–22. 632
579 3. Graf H, Stummvoll HK, Luger A, Prager R. Effect of amino acid 22. Mackensen-Haen S, Bader R, Grund KE, Bohle A. Correlations 633
580 infusion on glomerular filtration rate. N Engl J Med. 1983;308: between renal cortical interstitial fibrosis, atrophy of the 634
159–160. proximal tubules and impairment of the glomerular filtration
581 635
4. Bosch JP, Saccaggi A, Lauer A, et al. Renal functional reserve rate. Clin Nephrol. 1981;15:167–171.
582 636
in humans. Effect of protein intake on glomerular filtration
583 rate. Am J Med. 1983;75:943–950.
23. Chawla LS, Davison DL, Brasha-Mitchell E, et al. Development
637
and standardization of a furosemide stress test to predict the
584 5. Hostetter TH. Human renal response to meat meal. Am J severity of acute kidney injury. Crit Care. 2013;17:R207.
638
585 Physiol. 1986;250:F613–F618. 639
24. Herrera J, Rodriguez-Iturbe B. Stimulation of tubular secre-
586 6. Delanaye P, Schaeffner E, Ebert N, et al. Normal reference tion of creatinine in health and in conditions associated with
640
587 values for glomerular filtration rate: what do we really know? reduced nephron mass. Evidence for a tubular functional 641
588 Nephrol Dial Transplant. 2012;27:2664–2672. reserve. Nephrol Dial Transplant. 1998;13:623–629. 642
6 Kidney International Reports (2016) -, -–-

REV 5.4.0 DTD
EKIR7_proof
14 May 2016
11:29 am
ce

 LS Chawla and C Ronco: Renal Stress Testing in the Assessment of Kidney Disease REVIEW

643 25. Rodriguez-Iturbe B, Herrera J, Marin C, Manalich R. Tubular 31. Brater DC, Anderson SA, Strowig S. Azosemide, a “loop” 697
644 stress test detects subclinical reduction in renal functioning diuretic, and furosemide. Clin Pharmacol Ther. 1979;25: 698
645 mass. Kidney Int. 2001;59:1094–1102. 435–439.
699
646 26. van der Voort PH, Boerma EC, Pickkers P. The furosemide 32. Koyner JL, Davison DL, Brasha-Mitchell E, et al. Furosemide 700
stress test to predict renal function after continuous renal
647 replacement therapy. Crit Care. 2014;18:429.
stress test and biomarkers for the prediction of AKI severity.
701
J Am Soc Nephrol. 2015;26:2023–2031.
648 702
27. Hasannejad H, Takeda M, Taki K, et al. Interactions of human 33. van der Voort PH, Boerma EC, Koopmans M, et al. Furose-
649 organic anion transporters with diuretics. J Pharmacol Exp
703
mide does not improve renal recovery after hemofiltration for
650 Ther. 2004;308:1021–1029. acute renal failure in critically ill patients: a double blind
704
651 28. Bowman RH. Renal secretion of [35-S]furosemide and randomized controlled trial. Crit Care Med. 2009;37:533–538. 705
652 depression by albumin binding. Am J Physiol. 1975;229: 34. Powell TC, Warnock DG. The furosemide stress test and
706
653 93–98. predicting AKI outcomes. J Am Soc Nephrol. 2015;26: 707
654 29. Burg M, Stoner L, Cardinal J, Green N. Furosemide effect on 1762–1764. 708
655 isolated perfused tubules. Am J Physiol. 1973;225:119–124. 35. Berger BE, Warnock DG. Mechanisms of action and clinical 709
656 30. Dirks JH, Seely JF. Effect of saline infusions and furosemide uses of diuretics. In: Brenner BM RF, eds. The Kidney. 3rd ed. 710
657 on the dog distal nephron. Am J Physiol. 1970;219:114–121. W.B. Saunders; 1986:433–456. Q17
711
658 712
659 713
660 714
661 715
662 716
663 717
664 718
665 719
666 720
667 721
668 722
669 723
670 724
671 725
672 726
673 727
674 728
675 729
676 730
677 731
678 732
679 733
680 734
681 735
682 736
683 737
684 738
685 739
686 740
687 741
688 742
689 743
690 744
691 745
692 746
693 747
694 748
695 749
696 750
Kidney International Reports (2016) -, -–- 7

REV 5.4.0 DTD
EKIR7_proof
14 May 2016
11:29 am
ce