1

[ Contemporary Reviews in Critical Care Medicine ] 56

2 57
3 58
4 59
5 60
6
7
Continuous Renal Replacement Therapy 61
62
8 Who, When, Why, and How 63
9 64
10 65
Q22 Q1 Srijan Tandukar, MD; and Paul M. Palevsky, MD
11 66
12 67
13 68
Continuous renal replacement therapy (CRRT) is commonly used to provide renal support for
14 69
critically ill patients with acute kidney injury, particularly patients who are hemodynamically
15 70
16 unstable. A variety of techniques that differ in their mode of solute clearance may be used,
71
17 including continuous venovenous hemofiltration with predominantly convective solute clear- 72
18 ance, continuous venovenous hemodialysis with predominantly diffusive solute clearance, and 73
19 continuous venovenous hemodiafiltration, which combines both dialysis and hemofiltration. 74
20 The present article compares CRRT with other modalities of renal support and reviews in- 75
21 dications for initiation of renal replacement therapy, as well as dosing and technical aspects in 76
22 the management of CRRT. CHEST 2018; -(-):--- 77
23 78
24 KEY WORDS: acute kidney injury; continuous renal replacement therapy; dialysis; hemodialysis; 79
25 hemofiltration Q4
80
26 81
27 82
Acute kidney injury (AKI) is a common marked hemodynamic instability, has
28 83
complication in critically ill patients and is become routine. However, substantial
29 84
associated with substantial morbidity and uncertainty remains regarding many of the
30 85
31
risk of death. Approximately 5% to 10% of fundamental aspects of RRT management, 86
32 patients with AKI require renal replacement including the optimal timing of initiation 87
33 therapy (RRT) during their ICU stay,1 with and discontinuation, as well as the selection 88
34 mortality rates of 30% to 70%.2-4 Over the of modality.6 The present article provides an 89
35 past 2 decades, the incidence of RRT- overview of key issues in the management of 90
36 requiring AKI has increased by RRT in the critically ill patient, focused 91
37 approximately 10% per year.5 Risk factors for primarily on the use of continuous renal 92
38 RRT-requiring AKI include older age, male replacement therapy (CRRT). 93
39 sex, African-American race, higher severity 94
40 95
of illness, sepsis, decompensated heart
41 Modalities of RRT 96
failure, cardiac surgery, liver failure, and use
42 97
of mechanical ventilation. While once Multiple modalities of renal support may be
43 98
considered an extraordinary measure, the used in the management of the critically ill
44 99
45 ability to provide RRT, even in the setting of patient with kidney failure. These include 100
46 101
47 102
48 103
ABBREVIATIONS: AKI = acute kidney injury; CRRT = continuous Palevsky), Medical Service, VA Pittsburgh Healthcare System, Pitts- Q2
49 renal replacement therapy; CVVH = continuous venovenous hemo- burgh, PA. 104
50 filtration; CVVHD = continuous venovenous hemodialysis; CORRESPONDENCE TO: Paul M. Palevsky, MD, Room 7E123 (111F- 105
51 CVVHDF = continuous venovenous hemodiafiltration; IHD = inter- U), VA Pittsburgh Healthcare System, University Dr, Pittsburgh, PA 106
mittent hemodialysis; PIRRT = prolonged intermittent renal replace- 15240; e-mail: palevsky@pitt.edu Q3
52 ment therapy; RRT = renal replacement therapy; UFH = 107
Published by Elsevier Inc. under license from the American College of
53 unfractionated heparin Chest Physicians. 108
54 AFFILIATIONS: From the Renal-Electrolyte Division (Drs Tandukar 109
DOI: https://doi.org/10.1016/j.chest.2018.09.004
and Palevsky), Department of Medicine, University of Pittsburgh
55 110
School of Medicine, Pittsburgh, PA; and the Renal Section (Dr

chestjournal.org 1

REV 5.5.0 DTD ! CHEST1962_proof ! 19 October 2018 ! 11:33 am ! EO: CHEST-18-1329

111 CRRT, conventional intermittent hemodialysis (IHD), Although the Kidney Disease: Improving Global 166
112 and the prolonged intermittent renal replacement Outcomes (KDIGO) Clinical Practice Guideline for AKI 167
113 168
therapies (PIRRTs), which are a hybrid of CRRT and recommends the use of CRRT for patients who are
114 169
IHD. All of these use relatively similar extracorporeal hemodynamically unstable,21 the strength of this
115 170
blood circuits and differ primarily with regard to recommendation is low. Observational data, however, do
116 171
duration of therapy and, consequently, the rapidity of suggest that CRRT is more effective in achieving net
117 172
118
net ultrafiltration and solute clearance. In addition, negative fluid balance than IHD.22 In addition, in patients 173
119 dialytic therapies rely predominantly on diffusive solute with fulminant hepatic failure or brain injury with 174
120 clearance, whereas solute removal during hemofiltration increased intracranial pressure, CRRT is associated with 175
121 occurs by convection. better maintenance of cerebral perfusion than IHD.23-26 176
122 177
IHD provides rapid solute clearance and ultrafiltration
123 Selection of CRRT Modality 178
124
during relatively brief (3- to 5-h) treatments; the 179
continuous therapies provide more gradual fluid While initially developed as an arteriovenous therapy,
125 180
removal and solute clearance over prolonged treatment most CRRT is now performed by using pump-driven
126 181
127 times (optimally, 24 h per day but often interrupted due venovenous extracorporeal circuits. Although this 182
128 to system clotting or diagnostic or therapeutic method introduces additional degrees of complexity, 183
129 procedures).1 The multiple forms of PIRRT are including pressure monitors and air detectors, the 184
130 characterized by treatments that are generally between 8 pump-driven venovenous circuit provides higher and 185
131 and 16 h in duration, with slower rates of solute more consistent blood flows and eliminates the hazards 186
132 associated with prolonged arterial cannulation with a 187
clearance and ultrafiltration than IHD but more rapid
133 large-bore catheter. Multiple techniques for delivering 188
than CRRT. PIRRT is most commonly provided by
134 CRRT have been developed. When used solely for 189
using equipment similar to that for IHD but with lower
135 190
blood and dialysate flow rates. It can also be performed volume management, the treatment is known as slow
136 191
by using equipment designed for CRRT but with continuous ultrafiltration. More commonly, when
137 192
augmented dialysate and/or ultrafiltration rates to provided as continuous venovenous hemofiltration
138 193
achieve similar delivered therapy over a shorter (CVVH), continuous venovenous hemodialysis
139 194
140 duration.7 Peritoneal dialysis provides an effective (CVVHD), or continuous venovenous hemodiafiltration 195
141 alternative to the extracorporeal modalities of RRT,8 but (CVVHDF), CRRT provides both solute clearance and 196
142 a detailed discussion of this method is beyond the scope volume removal, with the differences between these 197
143 of this review. modalities related to the mechanisms for solute 198
144 clearance (Fig 1). 199
145 200
Selection of RRT Modality In CVVH, a high rate of ultrafiltration across the semi-
146 201
147 Although CRRT and PIRRT are most commonly used in permeable hemofilter membrane is created by a 202
148 hemodynamically unstable patients, there is marked hydrostatic gradient, and solute transport occurs by 203
149 variation in practice. Some centers use CRRT (or convection. Solutes are entrained in the bulk flow of 204
150 PIRRT) in all ICU patients with renal failure regardless water across the membrane, a process often referred to 205
151 of hemodynamic status, whereas others use IHD, albeit as “solvent drag.”1,27 High ultrafiltration rates are 206
152 with adjustments in prescription, even in vasopressor- needed to achieve sufficient solute clearance, and the 207
153 dependent patients. Although the benefit of a slow, ultrafiltrate volume beyond what is required to achieve 208
154 209
continuous modality of renal support in desired net fluid removal is replaced with balanced IV
155 210
hemodynamically unstable patients may seem self- crystalloid solutions. These replacement solutions may
156 211
evident, randomized trials have failed to show be infused into the extracorporeal circuit either prior to
157 212
differences with regard to either mortality or recovery of or following the hemofilter. Because the high
158 213
159
kidney function comparing CRRT vs either IHD9-17 or ultrafiltration rate hemoconcentrates the blood as it 214
160 PIRRT.18-20 It must be recognized, however, that to passes through the hemofilter fibers, the risk of sludging 215
161 provide IHD in hemodynamically unstable patients, the and fiber occlusion is increased. Prefilter infusion of 216
162 standard prescription may require modification, such as replacement fluid dilutes the blood entering the 217
163 prolongation of treatment time to allow for more hemofilter, mitigating this hemoconcentration. 218
164 gradual ultrafiltration, use of higher dialysate sodium However, prefilter administration of replacement fluid 219
165 concentrations, and reduced dialysate temperatures.12 dilutes the solute content of the blood, reducing effective 220

2 Contemporary Reviews in Critical Care Medicine [ -#- CHEST - 2018 ]

REV 5.5.0 DTD ! CHEST1962_proof ! 19 October 2018 ! 11:33 am ! EO: CHEST-18-1329
221 for IV replacement fluids. Although commonly 276
A
222 considered as a purely diffusive therapy, unmeasured 277
Prefilter Postfilter
223 278
replacement replacement bidirectional filtration into the dialysate compartment
224 fluid fluid 279
and back-filtration from dialysate to blood (driven by
225 From patient To patient 280
variation in the hemodynamic pressure gradient over
226 281
the length of the hemodialysis fibers) result in
227 282
228
significant convective solute transport. CVVHDF is a 283
229 Ultrafiltrate
hybrid, combining the dialysate flow of CVVHD with 284
230 the high ultrafiltration rates and use of replacement 285
231 B fluids of CVVH. 286
232 287
Dialysate The various mechanisms of solute clearance provided
233 288
234
by CVVH and CVVHD result in different profiles of 289
235 From patient To patient solute removal with each modality. Diffusion provides Q5 290
236 efficient clearance of low-molecular-weight solutes 291
237 (< 500-1,500 Daltons); however, diffusive clearance 292
238 declines rapidly as solute molecular weight increases 293
239 Effluent (Fig 2). In contrast, solute movement in convection is 294
240 limited primarily by the size of the pores in the 295
241 hemofilter membrane. Clearances of lower and higher 296
C
242 molecular weight solutes are similar, until the solute 297
Dialysate Replacement
243 298
fluid molecular radius approaches the size of the
244 299
membrane pores.27 Thus, at equivalent effluent flow
245 From patient To patient 300
rates, CVVH provides higher clearances than
246 301
CVVHD for solutes in the range of 1,000 to 20,000
print & web 4C=FPO

247 302
248 Daltons, or even higher if high cutoff membranes 303
Effluent
249 (Ultrafiltrate + spent dialysate)
with larger pores are used. Although it has been 304
250 suggested that the augmented clearance of higher 305
251 Figure 1 – A-C, Schematic diagrams of modalities of continuous renal molecular weight solutes (eg, pro-inflammatory 306
replacement therapy. A, Continuous hemofiltration. Blood flow through
252 the hemofilter is shown from left to right. An ultrafiltrate is generated cytokines) provided by CVVH might be beneficial, 307
253 across the hemofilter membrane, and excess ultrafiltrate above the vol- this theory has not been borne out in clinical 308
254 ume desired for negative fluid balance is replaced with prefilter and/or 309
postfilter replacement solution. B, Continuous hemodialysis. Blood flow
practice.17,28,29 Independent of diffusion and
255 through the hemodialyzer is shown from left to right. Dialysate is convection, adsorption of solutes in the CRRT circuit, 310
256 perfused through the hemodialyzer on the opposite side of the membrane 311
subject to saturation of membrane binding sites, may
257 from the blood countercurrent to the direction of blood flow. The effluent 312
consists of spent dialysate plus the volume of ultrafiltrate desired to also contribute to overall solute clearance.6 Thus,
258 achieve negative fluid balance. C, Continuous hemodiafiltration. Blood 313
choice of CRRT modality (CVVH, CVVHD, or
259 through the hemodiafilter is shown from left to right. As in continuous 314
260 hemodialysis, dialysate is perfused through the hemodialyzer on the CVVHDF) is primarily a function of provider 315
opposite side of the membrane from the blood countercurrent to the preference rather than patient characteristics or
261 Q23 direction of blood flow. The effluent consists of spent dialysate plus ul- 316
262 trafiltrate. As in continuous hemofiltration, excess ultrafiltrate above the objective outcome data. 317
263 volume desired for negative fluid balance is replaced with replacement 318
solution. In the figure, replacement solution is shown being infused
264 319
Q19 postfilter; replacement solution can also be infused prefilter. Indications for Initiation of RRT
265 320
The indications for initiation of CRRT generally
266 321
267
correspond to overall indications for RRT (Table 1), 322
solute clearance at a fixed ultrafiltration rate. Postfilter
268 including volume overload, severe metabolic acidosis 323
infusion has no such effects.
269 and electrolyte disturbances, and overt uremic 324
270 In CVVHD, dialysate is perfused across the external symptoms. Although these indications are well 325
271 surface of the dialysis membrane, and solutes exit ensconced, they are subject to wide interpretation and 326
272 from blood to dialysate by diffusion down their should be considered as only semi-objective. In addition, 327
273 concentration gradient. Ultrafiltration rates are in many patients, RRT is initiated in the setting of 328
274 relatively low compared with those in CVVH, persistent or progressive AKI in the absence of these 329
275 330
permitting net negative fluid balance without the need criteria.

chestjournal.org 3

REV 5.5.0 DTD ! CHEST1962_proof ! 19 October 2018 ! 11:33 am ! EO: CHEST-18-1329

331 strong association between severity of volume overload 386
A
332 at initiation of RRT and mortality risk, causality has not 387
333 Ultrafiltrate 388
been established.22,30,31 Complex interplay exists
334 Blood Flow 389
between underlying severity of illness, development of
335 390
volume overload and mortality, and an absence
336 391
of prospective data showing that initiation of
337 392
338
extracorporeal ultrafiltration at a specific threshold of 393
339 volume overload reduces mortality. 394
340
B 395
Blood Flow Acid-Base Abnormalities
341 396
342 Progressive metabolic acidosis is an inevitable 397
343 consequence of kidney failure, developing due to 398
344 impaired renal acid excretion.32 In patients in whom 399
print & web 4C=FPO

345 severe acidosis is refractory to medical management, 400

346 Dialysate 401
such as the volume overloaded patient who cannot
347 402
tolerate alkali administration, either intermittent or
348 Figure 2 – Convection and diffusion. A, Convection: solute transfer 403
continuous RRT is effective.32-34 Commonly
349 across the membrane occurs via entrainment of solutes in the bulk flow 404
of water during ultrafiltration. Higher molecular weight solutes (light suggested thresholds for initiation of RRT include a
350 405
gray) and lower molecular weight (< 500-1,500 Daltons) solutes (dark pH < 7.1 to 7.2 or serum bicarbonate level < 12 to
351 gray) are transported across the membrane with equal efficiency until 406
352 the molecular radius of the solute exceeds the membrane pore size. B, 15 mmol/L. Earlier initiation of RRT may be 407
353
Diffusion: solute transfer across the membrane occurs by movement necessary in patients with acute lung injury receiving 408
down a concentration gradient from blood to dialysate. Lower molecular
354 weight (< 500-1,500 Daltons) solutes (dark gray) cross the membrane lung-protective ventilation, as severe acidemia can 409
355 Q20 more readily than higher molecular weight solutes (light gray). result from the combination of metabolic and 410
356 respiratory acidosis. Although RRT augments lactate 411
357 clearance, there is scant evidence that initiation of 412
Volume Overload
358 RRT alters clinical outcomes in patients with lactic 413
359 Volume overload in AKI occurs due to the disruption of 414
acidosis not associated with drug toxicity (eg,
360 the kidney’s ability to maintain fluid balance in the face 415
metformin).
361 of administration of IV fluids, blood products, and/or 416
362 other medications required for resuscitation and Severe Electrolyte Abnormalities 417
363 supportive treatment of a critically ill patient; they may 418
364
Multiple electrolyte abnormalities are associated with 419
ensue even in patients who are not oliguric or anuric.1
365 AKI. Severe hyperkalemia is the most life-threatening 420
There are no prospective data establishing specific
366 and requires prompt treatment to prevent 421
thresholds for RRT initiation. RRT is generally indicated
367 cardiotoxicity and arrhythmias. Initiation of RRT is 422
when volume overload compromises organ function and
368 indicated when hyperkalemia is refractory to medical 423
is refractory to diuretic agents. Although observational
369 therapy or recurs following the initial treatment. 424
370
data in both pediatric and adult populations show a 425
Although rigid thresholds based on level of serum
371 potassium cannot be provided, RRT solely for the 426
372 TABLE 1 ] Indications for Initiation of Continuous Renal management of hyperkalemia is rarely appropriate 427
373 Q17 Replacement Therapy 428
when the potassium level is < 6 mmol/L. Conversely,
374 429
Volume overload RRT is generally appropriate in patients in whom the
375 430
Metabolic acidosis potassium level remains > 6.5 mmol/L despite
376 431
377
Electrolyte abnormalities medical management. Although IHD provides more 432
378 Hyperkalemia rapid correction of hyperkalemia and is the preferred 433
379 Hyponatremia modality in this setting, CRRT provides effective, 434
380 Hyperphosphatemia albeit slower, control of the plasma potassium 435
381 Uremia concentration.35 Other electrolyte abnormalities, such 436
382 Encephalopathy as severe hyponatremia or hypernatremia and severe 437
383 hyperphosphatemia, may accompany AKI and should 438
Pericarditis
384 be a factor in the decision to initiate RRT. In patients 439
Persistent/progressive acute kidney injury
385 440
with severe hyponatremia in the setting of AKI, CRRT

4 Contemporary Reviews in Critical Care Medicine [ -#- CHEST - 2018 ]

REV 5.5.0 DTD ! CHEST1962_proof ! 19 October 2018 ! 11:33 am ! EO: CHEST-18-1329
441 may permit the slower and more controlled correction ammonia levels.42,43 In a retrospective analysis of 496
442 of sodium concentration needed to prevent the registry data, CRRT was associated with improved 21- 497
443 498
neurologic sequelae of osmotic demyelination, day transplant-free survival among patients with acute
444 499
compared with IHD.32 liver failure compared with IHD or no RRT.43 However,
445 500
these data are not sufficient to establish causality, and
446 501
Uremia and Progressive or Persistent Azotemia there are no prospective studies that have specifically
447 502
448 The use of RRT for the management of overt uremic evaluated the use of CRRT for the management of 503
449 symptoms, such as encephalopathy and pericarditis, is hyperammonemia in liver disease. 504
450 well established. Although these are relatively late 505
Timing of Initiation of RRT
451 complications of AKI, other manifestations of uremia, 506
452 such as platelet dysfunction, impaired nutrition, In the absence of specific indications, the optimal timing 507
453 for initiation of RRT in AKI is uncertain. Earlier 508
increased susceptibility to infection and sepsis, heart
454 initiation of AKI allows for optimization of volume 509
failure, and pulmonary edema, may be difficult to
455 status, early correction of acid-base and electrolyte 510
distinguish from other etiologies in the critically ill
456 511
patient with multiple organ dysfunction.36 It is far more disturbances, and control of azotemia prior to the
457 512
common, when specific indications are not present for development of the major metabolic disturbances that
458 513
RRT, to be initiated prophylactically in response to serve as objective indications. However, these potential
459 514
460 persistent or progressive azotemia prior to the benefits of early initiation need to be balanced with the 515
461 development of overt uremic manifestations. The risks and burdens associated with RRT, including 516
462 appropriate timing for such initiation remains a topic of vascular access (eg, hemorrhage, thrombosis, vascular 517
463 debate and is discussed separately later in the article. injury, infection), intradialytic hypotension, and 518
464 resource utilization.44 Furthermore, it is often uncertain 519
465 Drug and Toxin Removal whether an individual patient will have persistent AKI or 520
466 rapid recovery of kidney function, and currently there 521
467
A variety of toxins and drugs, such as toxic alcohols, 522
are no tools to reliably predict the clinical trajectory for
468 lithium, salicylate, valproic acid, and metformin, are 523
the individual patient with AKI.
469 dialyzable, and the timely use of RRT in cases of 524
470 poisoning and drug intoxications with these agents may Multiple observational studies have suggested improved 525
471 be able to avert serious complications. The ability of survival associated with earlier initiation of RRT.2,45-52 526
472 RRT to remove a particular drug or toxin from the However, these studies only included patients who 527
473 circulation is a function of its size, volume of ultimately received RRT and did not account for patients 528
474 distribution, and protein binding. Thus, RRT is effective with AKI who did not undergo early RRT and who 529
475 for the removal of smaller, nonprotein-bound molecules either recovered kidney function or died without 530
476 531
with a volume of distribution < 1 L/kg body weight.37 receiving RRT. Excluding these patients from analysis
477 532
However, because the goal in the treatment of results in potential bias as the actual clinical question is
478 533
intoxications and overdoses is the rapid clearance of the not one of early vs late initiation of RRT but rather early
479 534
480
offending agent, IHD is generally preferred over CRRT vs non-early RRT in patients in whom there is no urgent 535
481
in this setting, even in patients who are indication. 536
482 hemodynamically unstable.38,39 537
Several randomized controlled trials have helped to
483 538
The role of RRT in the management of inform our understanding of this question, although
484 539
hyperammonemia is uncertain. Based on molecular there has been discordance in the results of these trials.
485 540
weight, ammonia is readily cleared by both diffusion and The Effect of Early vs Delayed Initiation of Renal
486 541
487
convection. As with the treatment of poisoning and Replacement Therapy on Mortality in Critically Ill 542
488 intoxication, IHD will provide more rapid reduction in Patients With Acute Kidney Injury (ELAIN) trial was a 543
489 blood ammonia levels. However, in small case series, single-center, unblinded randomized controlled trial of 544
490 high-dose CRRT has been shown to be efficacious for 231 critically ill patients at a university hospital in 545
491 the acute management of severe hyperammonemia (> Germany with stage 2 AKI (doubling of serum 546
492 400 mmol/L) in infants with inborn errors of creatinine level or urine output < 0.5 mL/kg per hour 547
493 metabolism.40,41 The role of CRRT in adults with for 12 h).53 Patients were randomized to receive either 548
494 hyperammonemia complicating liver failure is less immediate initiation of RRT or to a strategy in which 549
495 550
certain. CRRT is associated with reductions in plasma RRT was delayed until an absolute indication was

chestjournal.org 5

REV 5.5.0 DTD ! CHEST1962_proof ! 19 October 2018 ! 11:33 am ! EO: CHEST-18-1329

551 present or AKI had progressed to stage 3 (tripling of During CVVH, the concentration of low-molecular- 606
552 serum creatinine level, urine output < 0.3 mL/kg per weight solutes such as urea in the ultrafiltrate is close to 607
553 608
hour for 24 h, or anuria for 12 h). All 112 patients in the that in plasma water. Similarly, during CVVHD, because
554 609
early arm and 108 of 119 patients (91%) in the delayed the dialysate flow rate is commonly an order of
555 610
arm received CVVHDF, with a median time from stage magnitude lower than the blood flow rate, virtually
556 611
2 AKI to initiation of RRT of 6 h in the early arm complete equilibration of low-molecular-weight solutes
557 612
558
compared with 25.5 h in the delayed arm (median is achieved between plasma and dialysate. Thus, 613
559 difference, 21 h). Ninety-day all-cause mortality was regardless of modality of CRRT, clearance of urea and 614
560 39.3% in the early arm compared with 54.7% in the other low-molecular-weight solutes is approximately 615
561 delayed arm (P ¼ .03). equal to effluent flow. 616
562 617
In contrast, the Artificial Kidney Initiation in Kidney Although several studies published 15 to 20 years ago
563 618
Injury (AKIKI) trial was a multicenter, randomized suggested that higher effluent flow rates were associated
564 619
565 controlled trial across 31 ICUs in France.54 In this trial, with improved survival,3,59 the results were 620
566 619 patients with stage 3 AKI who did not have inconsistent,27,60 and this relationship was not 621
567 emergent indications were randomized to undergo confirmed in two large, multicenter, randomized 622
568 immediate initiation of RRT or a strategy of delayed controlled trials. In the VA/NIH Acute Renal Failure 623
569 initiation based on clinical indications. Of the 308 Trial Network (ATN) study, 1,124 critically ill patients 624
570 patients randomized to undergo the delayed strategy, with AKI were randomized to receive either a more- 625
571 only 157 (51%) ultimately received RRT. For those intensive strategy (CVVHDF at an effluent flow of 626
572 patients receiving RRT, the median time from reaching 35 mL/kg per hour, or IHD or PIRRT six times per 627
573 628
stage 3 AKI to randomization was 4.3 h in the early arm week) or a less-intensive strategy (CVVHDF at an
574 629
compared with 57 h in the delayed arm. Sixty-day effluent flow of 20 mL/kg per hour, or IHD or PIRRT
575 630
mortality did not differ between the two treatment arms three times per week).61 Within each treatment arm,
576 631
(48.5% vs 49.7%; P ¼ .79). No differences in the overall patients switched between CVVHDF or PIRRT and IHD
577 632
578 outcome were observed in subgroup analyses of 348 based on hemodynamic status. More intensive renal 633
579 patients with septic shock and 207 patients with support was not associated with differences in mortality, 634
580 ARDS.55 recovery of kidney function, or the rate of nonrenal 635
581 organ failure across the entire study population or when 636
Important differences between the ELAIN and AKIKI
582 assessed based on the percentage of time receiving 637
trials should be noted. The key entry criterion for the
583 CVVHDF or PIRRT.62 The Randomized Evaluation of 638
584
AKIKI trial (stage 3 AKI) was the criterion for late 639
Normal versus Augmented Level (RENAL) Replacement
585 initiation of RRT in the ELAIN trial. In addition, in the 640
Therapy Study randomized 1,508 patients with AKI
586 AKIKI trial, all modalities of RRT could be used, 641
across 31 ICUs in Australia and New Zealand to
587 whereas in ELAIN, only CVVHDF was used. Critically, 642
CVVHDF at an effluent flow of either 25 or 40 mL/kg
588 the AKIKI trial excluded patients with emergent criteria 643
per hour.63 As in the ATN study, the use of more
589 for initiation of RRT, such as severe hyperkalemia or 644
intensive solute clearance was not associated with
590 pulmonary edema, whereas the majority of patients in 645
591
improved clinical outcomes. On the basis of these data, 646
the ELAIN trial had fluid overload or pulmonary edema
592 the KDIGO Clinical Practice Guidelines recommend a 647
prior to enrollment. Although future clinical trials are
593 target dose for CRRT of 20 to 25 mL/kg per hour, noting 648
needed to address the significant equipoise regarding the
594 that a higher prescribed dose may be required to ensure 649
optimal timing of RRT, we believe that in clinical
595 delivery of this target dose.21 650
practice, in the absence of emergent indications such as
596 651
597
intractable hyperkalemia or severe volume overload, an 652
approach of delayed RRT initiation is not Volume Management
598 653
599 unreasonable.56 A second dimension to the prescription of RRT is 654
600 volume management. Net ultrafiltration may be adjusted 655
601 Q6 Dose of CRRT independently of solute clearance. As previously 656
602 discussed, severity of volume overload is strongly 657
603 Solute Control associated with mortality risk in both children and 658
604 The dose of CRRT is assessed based on the effluent flow adults with RRT-requiring AKI.22,30,31 However, optimal 659
605 660
rate, the sum of dialysate, and total ultrafiltrate flow.57,58 strategies for volume management are uncertain,

6 Contemporary Reviews in Critical Care Medicine [ -#- CHEST - 2018 ]

REV 5.5.0 DTD ! CHEST1962_proof ! 19 October 2018 ! 11:33 am ! EO: CHEST-18-1329
661 requiring a balance between the provision of net Practice Guidelines for AKI recommend the right 716
662 ultrafiltration to achieve euvolemia, optimization of internal jugular vein as the preferred location for 717
663 718
cardiopulmonary status, and the risk of exacerbating catheter placement, followed by the femoral and the left
664 719
hypotension. Management must be individualized for internal jugular veins.21
665 720
each patient, with frequent reassessment of
666 Proper positioning of the catheter tip is critical for 721
ultrafiltration targets. It should be recognized that short-
667 adequate catheter function. For internal jugular 722
668
term fluctuation in BP is generally unrelated to volume 723
catheters, the catheter tip should be at the junction of the
669 status and that transient hypotension during CRRT 724
superior vena cava and right atrium or in the right
670 needs to be carefully assessed for nonvolume-mediated 725
atrium, depending on catheter design, rather than more
671 factors and often requires management independent of 726
proximally in the superior vena cava.68 To permit proper
672 alterations in ultrafiltration targets. 727
positioning, a longer catheter is required when the left
673 728
Role of CRRT in Sepsis internal jugular vein is cannulated than when the right is
674 729
675
used. Even longer catheters are required for femoral 730
Although cytokine modulation by CVVH has been
676 catheters to permit placement with the catheter tip 731
proposed as an adjunctive therapy in sepsis, clinical
677 within, or as close to, the inferior vena cava. Catheter 732
trials have not shown any meaningful benefit. In a trial
678 malposition is associated with restricted blood flow and 733
of 80 patients randomized to receive isovolemic
679 increased risk of recirculation. In addition, catheter 734
hemofiltration or usual care, there was no improvement
680 malfunction is a frequent cause for interruption in 735
681
Q7 in clinical parameters or mortality with hemofiltration.28 736
circuit flow and filter clotting. Adequacy of catheter
682 Similarly, in the High-Volume Versus Standard-Volume 737
function should be assessed whenever there is frequent
683 Haemofiltration for Septic Shock Patients With Acute 738
system clotting. Although tunneled catheters are not
684 Kidney Injury (IVOIRE) trial, which compared CVVH 739
recommended for routine use, they are associated with
685 at 35 and 70 mL/kg per hour in patients with septic AKI, 740
decreased risks of infection and higher blood flow rates,
686 no benefit was associated with the higher dose of 741
and should be considered when the need for RRT is
687 hemofiltration.29 Thus, current data do not support the 742
688
expected to exceed 1 to 3 weeks.21 743
use of CRRT as an adjunctive therapy in sepsis beyond
689 its role for renal support. 744
690 Anticoagulation for CRRT 745
691 746
Technical Issues in CRRT Management Clotting of the extracorporeal circuit is the most
692 747
common complication during CRRT. Practice Q8
693 748
Vascular Access patterns regarding the use of anticoagulation vary
694 749
695 Initiation of CRRT requires vascular access, which is widely, with estimates of 40% to 60% of patients 750
696 generally established through placement of a large-bore undergoing CRRT without anticoagulation. Although 751
697 double lumen catheter in an internal jugular, femoral, or the use of anticoagulation is often avoided in patients 752
698 subclavian vein. In adults, catheter design and position who are coagulopathic, thrombocytopenic, or are 753
699 must be sufficient to sustain blood flow rates of 200 to having active hemorrhage, anticoagulation-free 754
700 300 mL/min. Cannulation of the right internal jugular treatment may also be successful in the absence of 755
701 vein is generally preferred over the left, given the coagulopathy and thrombocytopenia. Strategies to Q9 756
702 757
straighter course from the right side to the right atrium. minimize the risk of clotting of the extracorporeal
703 758
Although femoral catheters are generally associated with circuit include the following: use of higher blood flow
704 759
higher rates of bacteremia than internal jugular rates; minimization of filtration fraction (the ratio of
705 760
catheters,64 comparable rates of colonization, ultrafiltration to plasma flow) by using CVVHD
706 761
707
bacteremia, and thrombosis were observed with femoral rather than CVVH, or by infusing replacement fluids 762
708 and internal jugular dialysis catheters in a randomized prefilter during CVVH and CVVHDF; ensuring 763
709 controlled trial of 750 patients with RRT-requiring optimal catheter function and responding promptly to 764
710 AKI.65 However, the relative risk for femoral catheter machine alarms to minimize interruptions in blood 765
711 colonization was higher in patients with a BMI > flow; and increasing the frequency of scheduled 766
712 28.4 kg/m2. Subclavian cannulation is generally avoided replacement of the extracorporeal circuit.61,63 In the 767
713 because of the higher risk of insertion complications and absence of anticoagulation, increased vigilance is 768
714 because of the risk of subsequent venous stenosis.66,67 required to ensure there is no compromise of 769
715 770
Based on these considerations, the KDIGO Clinical delivered dose.69

chestjournal.org 7

REV 5.5.0 DTD ! CHEST1962_proof ! 19 October 2018 ! 11:33 am ! EO: CHEST-18-1329

771 When anticoagulation is used, the most common impairment in citrate metabolism, citrate may 826
772 strategies rely on heparin or citrate. Either accumulate and contribute to a high-anion-gap 827
773 828
unfractionated heparin (UFH) or low-molecular-weight metabolic acidosis. Thus, in addition to monitoring the
774 829
Q10 heparin may be used. Dosing protocols for UFH during ionized calcium in the extracorporeal circuit to ensure
775 830
CRRT vary widely, ranging from very low dose (initial adequate calcium chelation, systemic electrolytes,
776 831
bolus of 500-1,000 units followed by infusions of magnesium, total and ionized calcium, and blood pH
777 832
778
300-500 units per hours), designed to minimize systemic need to be monitored frequently. Recommendations are 833
779 effects, to bolus doses of 30 units/kg followed by to check these laboratory readings 1 h after initiating or 834
780 infusions of 5 to 10 units/kg per hours with a target making any change in citrate anticoagulation or the 835
781 systemic activated partial thromboplastin time of 1.5 to CRRT prescription, and then at least every 6 h. Citrate 836
782 Q11 2.0 times the upper limit of normal.70 Low-molecular- accumulation should be suspected if the need for 837
783 weight heparins have been proposed as an alternative to calcium infusion to maintain systemic ionized calcium 838
784 UFH given higher anti-factor Xa activity, a more levels increases, if there is an increasing anion-gap 839
785 consistent anticoagulant response, and a lower metabolic acidosis, or if the ratio of systemic total 840
786 841
incidence of heparin-induced thrombocytopenia. calcium to ionized calcium is > 2.5. If citrate toxicity is
787 842
However, consistent superiority of low-molecular- suspected, the use of citrate should be suspended or
788 843
weight heparin compared with UFH has not been discontinued.
789 844
shown.71,72 In patients with heparin-induced
790 845
thrombocytopenia, all heparin anticoagulation should Drug Dosing During CRRT
791 846
792 be discontinued, and anticoagulation with direct Medication dosing during CRRT can be challenging 847
793 thrombin inhibitors initiated.21 Argatroban is because drug dosing needs to account for multiple 848
794 generally preferred; however, it is hepatically factors beyond the extracorporeal drug removal, 849
795 metabolized. In patients with both liver failure and including nonrenal clearance, residual kidney function, 850
796 AKI, bivalirudin is preferred given its significant and changes in volume of distribution and protein 851
797 nonrenal and nonhepatic metabolism. binding.37,78 Errors in drug dosing can lead to both 852
798 toxicity from inadequate dose reduction and treatment 853
799 The use of citrate as an anticoagulant is based on its 854
failure from underdosing.37 The latter is of particular
800 rapid chelation of calcium in the extracorporeal circuit, 855
importance for antibiotic dosing in patients with sepsis
801 inhibiting multiple calcium-dependent steps in the 856
accompanying their AKI.78,79
802 coagulation cascade. The citrate-calcium complex 857
803 dissociates in the systemic circulation, and the citrate is A detailed discussion of drug dosing during CRRT is 858
804 rapidly metabolized, serving as an alkalinizing agent. beyond the scope of the present review; however, the 859
805 860
Citrate is infused into the extracorporeal circuit with a following broad guidelines are provided. For
806 861
goal of reducing the ionized calcium concentration to < medications such as analgesics, sedatives, and
807 862
0.4 mmol/L. Because some of the citrate-calcium vasopressors, which have an observable clinical effect,
808 863
complex is lost in the effluent, systemic infusion of dosing should be titrated to the desired response. Drugs
809 864
810
calcium is required to prevent hypocalcemia. Regional with high molecular weight, that are highly protein 865
811 citrate anticoagulation is associated with improved bound, or that have very large volumes of distribution 866
812 circuit patency, lower risks of bleeding, and avoidance of will be poorly cleared by CRRT, and dosing does not 867
813 the risk of heparin-induced thrombocytopenia need to be adjusted for RRT. The extracorporeal 868
814 compared with heparin.73-77 However, citrate clearance of low-molecular-weight drugs that are 869
815 anticoagulation is associated with increased risks of nonprotein bound will approximate effluent flow; for 870
816 multiple electrolyte and acid-base disturbances. Close protein-bound drugs, the estimated clearance needs to 871
817 monitoring of both circuit and systemic blood ionized be adjusted for the percent unbound fraction. For all 872
818 873
calcium levels is required to ensure sufficient efficacy medications with readily measurable blood levels, dosing
819 874
without systemic hypocalcemia. Hypernatremia may should be adjusted based on pharmacokinetic
820 875
result when hypertonic citrate solutions are used. monitoring. Finally, it must be recognized that although
821 876
Because metabolism of citrate potentially yields published guides provide estimates for dosing of many
822 877
823
bicarbonate in a 1:3 ratio, citrate anticoagulation may agents, they provide only general parameters that may 878
824 predispose to development of metabolic alkalosis. In not correspond to the specific CRRT mode and dose 879
825 addition, in patients with severe liver disease or other used.80 880

8 Contemporary Reviews in Critical Care Medicine [ -#- CHEST - 2018 ]

REV 5.5.0 DTD ! CHEST1962_proof ! 19 October 2018 ! 11:33 am ! EO: CHEST-18-1329
881 Nutritional Management exposure to the extracorporeal circuit may trigger 936
882 immediate allergic or delayed immunologic reactions 937
Patients with AKI who are undergoing CRRT are usually
883 938
in substantial negative nitrogen balance due to high secondary to cytokine activation.83 Bradykinin-
884 939
protein catabolic rates. In addition, CRRT results in mediated membrane reactions have been associated
885 940
amino acid loss as well as losses of water-soluble with certain synthetic membranes in conjunction with
886 941
vitamins and other micronutrients. Caloric intake the use of angiotensin-converting enzyme inhibitors.84
887 942
888 should be approximately 35 kcal/kg per day, with a Air embolization may occur at the time of catheter 943
889 target protein intake of 1.5 g/kg per day with insertion or at any time during treatment if air is 944
890 supplementation of water-soluble vitamins.81,82 entrained into the circuit beyond the return line air 945
891 Although enteral feeding is preferred, parenteral support detector. 946
892 may be necessary. 947
893 The most common complication during CRRT is circuit Q12 948
894 clotting, and the most common reason for circuit 949
Complications of CRRT
895 clotting is inadequate catheter function resulting in flow 950
896 As with all medical interventions, CRRT is not without restriction and pressure alarms that interrupt blood 951
897 its risks (Table 2).33 Initiation of CRRT requires flow. Prompt catheter replacement may be necessary if a 952
898 placement of a large-bore central venous catheter that blood flow of 200 to 300 mL/min cannot be sustained. 953
899 may need to be maintained for a prolonged duration. Excessive filtration fraction may lead to 954
900 Well-recognized complications of catheter insertion 955
hemoconcentration within the hemofilter, also
901 include vascular or visceral injury resulting in 956
contributing to filter clotting. If there is no catheter
902 957
hemorrhage, pneumothorax, hemothorax, and arterio- dysfunction, blood flow is maximized, and filtration
903 958
venous fistula formation. Prolonged catheter use is fraction is < 20%, initiation or intensification of
904 959
associated with venous thrombosis or stenosis. Blood anticoagulation should be considered. Complications of
905 960
906 heparin anticoagulation can include bleeding and 961
907 heparin-induced thrombocytopenia. Citrate 962
Q18 TABLE 2 ] Complications Associated With CRRT
908 anticoagulation may result in citrate toxicity from citrate 963
909 Catheter-related complications accumulation, overt hypocalcemia from inadequate 964
910 Hemorrhage calcium replacement, and both metabolic acidosis and 965
911 Infection metabolic alkalosis. 966
912 Venous thrombosis 967
913 Electrolyte abnormalities during CRRT are common. 968
Venous stenosis
914 Hypophosphatemia may result from continuous 969
Traumatic arteriovenous fistula
915 removal in the extracorporeal circuit and can delay 970
Pneumothorax
916 weaning from mechanical ventilation.61,63 971
917 Hemothorax 972
Hypophosphatemia may be avoided by pre-emptive
918 Air embolism 973
enteral or parenteral phosphate supplementation or by
919 Visceral injury 974
the use of phosphate-containing dialysate or
920 Extracorporeal circuit-related complications 975
replacement fluids.85 Other electrolyte abnormalities are
921 Allergic reaction to hemodialyzer/hemofilter or tubing 976
less common, although hyponatremia and
922 977
Circuit thrombosis hypernatremia and hyperkalemia have resulted from
923 978
Hemolysis compounding errors in custom manufactured dialysate
924 979
925 Air embolism and replacement fluids.86 To monitor electrolyte status, 980
926 Hypothermia laboratory monitoring should be performed at least daily 981
927 Hypotension during treatment. 982
928 Electrolyte disturbances 983
929 Unlike conventional IHD, dialysate and replacement 984
Hypophosphatemia
930 fluids are not usually warmed. Modest thermal losses 985
Hypokalemia
931 during CRRT cause vasoconstriction and are believed 986
Hypocalcemia
932 to contribute to increased hemodynamic stability but 987
Hypomagnesemia
933 may mask the onset of fever. If thermal losses are 988
934 Incorrect medication dosing 989
more substantial, significant hypothermia may ensue,
935 necessitating aggressive external warming. 990
CRRT ¼ continuous renal replacement therapy.

chestjournal.org 9

REV 5.5.0 DTD ! CHEST1962_proof ! 19 October 2018 ! 11:33 am ! EO: CHEST-18-1329

991 Hypotension during CRRT is common, occurring in citing continued volume overload as the most common 1046
992 some series in more than one third of patients, but is reason for continuation of RRT. In the ATN study, a 6-h 1047
993 1048
most often unrelated to the CRRT procedure, per timed urine collection was obtained when the urine
994 1049
se.12,87,88 Ultrafiltration exacerbating hemodynamic output was > 750 mL/d.61 RRT was continued if the
995 1050
instability is the most common treatment-related factor measured creatinine clearance was < 12 mL/min, was
996 1051
contributing to hypotension. Negative inotropic and discontinued if > 20 mL/min, and was left to clinician
997 1052
998
vasodilating properties associated with acetate- and judgment if the measured creatinine clearance was 1053
999 lactate-buffered solutions previously contributed to between 12 and 20 mL/min. Although these strategies 1054
1000 treatment-related hypotension; however, with the can inform clinical decision-making, precise criteria for 1055
1001 availability of bicarbonate-buffered dialysate and discontinuation of RRT are lacking. 1056
1002 replacement fluids, this scenario is no longer a 1057
The transition of patients with improved hemodynamic
1003 significant consideration.88 Hypotension may also be 1058
status but persistent AKI to other modalities of RRT is
1004 seen upon initiation of treatment, particularly if the 1059
1005
also highly variable. PIRRT may be used as a transitional 1060
circuit prime is not reinfused; this result has been of
1006 therapy or patients may transition directly to IHD, as 1061
particular concern in pediatric patients and may be
1007 clinical status warrants. Transition from CRRT to 1062
mitigated by using albumin to prime the circuit.89
1008 PIRRT or IHD may facilitate initiation of physical 1063
Although increases in blood flow through the
1009 therapy and mobilization out of bed. In general, patients 1064
extracorporeal circuit were associated with increased
1010 with persistent RRT-dependent AKI must be 1065
1011
circuit volume and hypotension with older dialysis 1066
transitioned to IHD prior to ICU discharge.
1012 systems, there is minimal change in extracorporeal 1067
1013 volume and thus negligible hemodynamic stress 1068
associated with changes in blood flow using current
Ethical Issues in Initiating and Discontinuing
1014 1069
1015 CRRT technology with its hollow fiber filters.88,90,91 RRT 1070
1016 When hypotension is associated with volume depletion, Issues related to the clinical and ethical appropriateness 1071
1017 it should be treated with volume reinfusion and of initiation or continuation of RRT often occur in 1072
1018 adjustment in ultrafiltration targets; in other patients with AKI. Discussions of initiation and 1073
1019 discontinuation of therapy with patients and/or their 1074
circumstances, alternative etiologies should be
1020 family/surrogate decision-makers should be framed in 1075
considered and the hypotension managed with titration
1021 light of the overall prognosis and goals of care, and they 1076
of vasopressor support.
1022 1077
need to consider other life-sustaining treatments in
1023 1078
Discontinuation of CRRT addition to RRT. It is important to ensure that both the
1024 1079
There are no specific criteria for discontinuation of primary managing service and nephrology consultants
1025 1080
CRRT because of recovery of kidney function or managing the RRT provide a consistent assessment of
1026 1081
1027 transition to other modalities of RRT.58,92,93 An initial prognosis and treatment options, presented in a clear 1082
1028 manifestation of recovery of kidney function is increased but sympathetic manner to facilitate the process of 1083
1029 urine output, although specific criteria are sparse. In the shared decision-making.96 The high mortality of AKI in 1084
1030 observational Beginning and Ending Supportive the setting of critical illness and the complex and 1085
1031 Therapy for the Kidney (BEST Kidney) study, a urine emotionally laden aspects of the decisions associated 1086
1032 with initiating or discontinuing renal support suggest 1087
output > 400 mL/d without concomitant diuretic
1033 that early involvement of palliative care services may be 1088
therapy was a predictor of successful CRRT
1034 of benefit.97 Time-limited trials of RRT may be a useful 1089
discontinuation.94 In this observational cohort, patients
1035 1090
who were successfully discontinued from CRRT without strategy in circumstances in which there is uncertainty
1036 1091
requiring re-initiation were more likely to survive to of prognosis or when the patient or family/surrogate
1037 1092
hospital discharge compared with those requiring re- decision-maker are not prepared to make a definite
1038 1093
initiation of CRRT. In another study, urine output > decision regarding initiating or discontinuing RRT and
1039 1094
1040 500 mL/d was proposed as a criterion for other life-sustaining care.98 1095
1041 discontinuation of RRT in a study of initiation and 1096
1042 discontinuation of therapy in patients with AKI.95 The Conclusions 1097
1043 usefulness of this criterion is uncertain, however, as the CRRT has become a mainstay in the management of 1098
1044 treating clinicians continued RRT despite this AKI in critically ill patients. In patients who do not have 1099
1045 1100
recommendation approximately two thirds of the time, objective indications for the emergent initiation of renal

10 Contemporary Reviews in Critical Care Medicine [ -#- CHEST - 2018 ]

REV 5.5.0 DTD ! CHEST1962_proof ! 19 October 2018 ! 11:33 am ! EO: CHEST-18-1329
1101 support, the optimal timing of RRT remains 13. Lins RL, Elseviers MM, Van der Niepen P, et al. Intermittent versus 1156
1102 continuous renal replacement therapy for acute kidney injury 1157
controversial. Although the use of continuous therapies patients admitted to the intensive care unit: results of a randomized
1103 clinical trial. Nephrol Dial Transplant. 2009;24(2):512-518. 1158
may facilitate management in hemodynamically
1104 14. Schefold JC, von Haehling S, Pschowski R, et al. The effect of 1159
unstable patients, existing data do not show that use of
1105 continuous versus intermittent renal replacement therapy on the 1160
CRRT results in improved survival or recovery of kidney outcome of critically ill patients with acute renal failure
1106 1161
function compared with alternatives such as (CONVINT): a prospective randomized controlled trial. Crit Care.
1107 2014;18(1):R11. 1162
1108
conventional IHD and PIRRT. Large, well-designed 1163
15. Bagshaw SM, Berthiaume LR, Delaney A, Bellomo R. Continuous
1109 clinical trials have established that for the majority of versus intermittent renal replacement therapy for critically ill 1164
patients, augmenting solute clearance using effluent flow patients with acute kidney injury: a meta-analysis. Crit Care Med.
1110 2008;36(2):610-617. 1165
1111 rates > 20 to 25 mL/kg per hour is not associated with 1166
16. Pannu N, Klarenbach S, Wiebe N, Manns B, Tonelli M. Alberta
1112 improved outcomes; however, optimal strategies for Kidney Disease N. Renal replacement therapy in patients with acute 1167
1113 volume management still must be defined. Similarly, renal failure: a systematic review. JAMA. 2008;299(7):793-805. 1168
1114 other aspects of the management of CRRT are subject to 17. Friedrich JO, Wald R, Bagshaw SM, Burns KEA, Adhikari NK. 1169
Hemofiltration compared to hemodialysis for acute kidney injury:
1115 substantial variations in practice, including strategies for systematic review and meta-analysis. Critical Care. 2012;16(4):R146. 1170
1116 1171
anticoagulation. Finally, the role of CRRT needs to be 18. Zhang L, Yang J, Eastwood GM, Zhu G, Tanaka A, Bellomo R.
1117 Extended daily dialysis versus continuous renal replacement therapy 1172
considered in the setting of overall goals of care and the for acute kidney injury: a meta-analysis. Am J Kidney Dis.
1118 1173
use of other life-sustaining treatments. 2015;66(2):322-330.
1119 1174
19. Kielstein JT, Kretschmer U, Ernst T, et al. Efficacy and
1120 1175
Acknowledgments cardiovascular tolerability of extended dialysis in critically ill
1121 patients: a randomized controlled study. Am J Kidney Dis. 1176
Financial/nonfinancial disclosures: The authors have reported to 2004;43(2):342-349.
1122 CHEST the following: P. M. P. has received research support from 1177
1123 BioPorto and consulting fees from Novartis, GE Healthcare, Baxter, 20. Schwenger V, Weigand MA, Hoffmann O, et al. Sustained low 1178
Q21 HealthSpan Dx, and Durect. None declared (S. T.). efficiency dialysis using a single-pass batch system in acute kidney
1124 injury—a randomized interventional trial: the REnal Replacement 1179
1125 Therapy Study in Intensive Care Unit PatiEnts. Crit Care. 1180
1126
Q13 References 2012;16(4):R140.
1181
1127 1. Tolwani A. Continuous renal-replacement therapy for acute kidney 21. Kidney Disease: Improving Global Outcomes (KDIGO) Acute 1182
injury. N Engl J Med. 2012;367(26):2505-2514. Kidney Injury Work Group. KDIGO Clinical Practice Guideline for
1128 Acute Kidney Injury. Kidney Int. 2012;2012(suppl):1-138. 1183
2. Liu KD, Himmelfarb J, Paganini E, et al. Timing of initiation of
1129 dialysis in critically ill patients with acute kidney injury. Clin J Am 22. Bouchard J, Soroko SB, Chertow GM, et al. Fluid accumulation, 1184
1130 Soc Nephrol. 2006;1(5):915-919. survival and recovery of kidney function in critically ill patients with 1185
acute kidney injury. Kidney Int. 2009;76(4):422-427.
1131 3. Saudan P, Niederberger M, De Seigneux S, et al. Adding a dialysis 1186
dose to continuous hemofiltration increases survival in patients with 23. Davenport A. Continuous renal replacement therapies in patients
1132 1187
acute renal failure. Kidney Int. 2006;70(7):1312-1317. with liver disease. Semin Dial. 2009;22(2):169-172.
1133 1188
4. Uchino S, Kellum JA, Bellomo R, et al. Acute renal failure in 24. Davenport A. Continuous renal replacement therapies in patients
1134 critically ill patients: a multinational, multicenter study. JAMA. with acute neurological injury. Semin Dial. 2009;22(2):165-168. 1189
1135 2005;294(7):813-818. 1190
25. Lin CM, Lin JW, Tsai JT, et al. Intracranial pressure fluctuation
1136 5. Hsu RK, McCulloch CE, Dudley RA, Lo LJ, Hsu CY. Temporal during hemodialysis in renal failure patients with intracranial 1191
changes in incidence of dialysis-requiring AKI. J Am Soc Nephrol. hemorrhage. Acta Neurochir Suppl. 2008;101:141-144.
1137 1192
2013;24(1):37-42.
1138 26. Ronco C, Bellomo R, Brendolan A, Pinna V, La Greca G. Brain 1193
6. Macedo E, Mehta RL. Continuous dialysis therapies: core curriculum density changes during renal replacement in critically ill patients
1139 2016. Am J Kidney Dis. 2016;68(4):645-657. with acute renal failure. Continuous hemofiltration versus 1194
1140 7. Marshall MR, Ma T, Galler D, Rankin APN, Williams AB. Sustained intermittent hemodialysis. J Nephrol. 1999;12(3):173-178. 1195
1141 low-efficiency daily diafiltration (SLEDD-f) for critically ill patients 27. Tolwani AJ, Campbell RC, Stofan BS, Lai KR, Oster RA, Wille KM. 1196
requiring renal replacement therapy: towards an adequate therapy. Standard versus high-dose CVVHDF for ICU-related acute renal
1142 Nephrol Dial Transplant. 2004;19(4):877-884. 1197
failure. J Am Soc Nephrol. 2008;19(6):1233-1238.
1143 1198
8. Burdmann EA, Chakravarthi R. Peritoneal dialysis in acute kidney 28. Payen D, Mateo J, Cavaillon JM, et al. Impact of continuous
1144 injury: lessons learned and applied. Semin Dial. 2011;24(2):149-156. venovenous hemofiltration on organ failure during the early phase of 1199
1145 9. Mehta RL, McDonald B, Gabbai FB, et al. A randomized clinical trial severe sepsis: a randomized controlled trial. Crit Care Med. 1200
1146 of continuous versus intermittent dialysis for acute renal failure. 2009;37(3):803-810. 1201
Kidney Int. 2001;60(3):1154-1163. 29. Joannes-Boyau O, Honore PM, Perez P, et al. High-volume versus
1147 1202
10. Augustine JJ, Sandy D, Seifert TH, Paganini EP. A randomized standard-volume haemofiltration for septic shock patients with acute
1148 controlled trial comparing intermittent with continuous dialysis in kidney injury (IVOIRE study): a multicentre randomized controlled 1203
1149 patients with ARF. Am J Kidney Dis. 2004;44(6):1000-1007. trial. Intensive Care Med. 2013;39(9):1535-1546. 1204
1150 11. Uehlinger DE, Jakob SM, Ferrari P, et al. Comparison of continuous 30. Sutherland SM, Zappitelli M, Alexander SR, et al. Fluid overload and 1205
1151 and intermittent renal replacement therapy for acute renal failure. mortality in children receiving continuous renal replacement 1206
Nephrol Dial Transplant. 2005;20(8):1630-1637. therapy: the prospective pediatric continuous renal replacement
1152 therapy registry. Am J Kidney Dis. 2010;55(2):316-325. 1207
12. Vinsonneau C, Camus C, Combes A, et al. Continuous
1153 venovenous haemodiafiltration versus intermittent 31. Vaara ST, Korhonen AM, Kaukonen KM, et al. Fluid overload is 1208
1154 haemodialysis for acute renal failure in patients with multiple- associated with an increased risk for 90-day mortality in critically ill 1209
organ dysfunction syndrome: a multicentre randomised trial. patients with renal replacement therapy: data from the prospective
1155 1210
Lancet. 2006;368(9533):379-385. FINNAKI study. Crit Care. 2012;16(5):R197.

chestjournal.org 11

REV 5.5.0 DTD ! CHEST1962_proof ! 19 October 2018 ! 11:33 am ! EO: CHEST-18-1329

1211 32. Yessayan L, Yee J, Frinak S, Szamosfalvi B. Continuous renal 54. Gaudry S, Hajage D, Schortgen F, et al. Initiation strategies for renal- 1266
1212 replacement therapy for the management of acid-base and replacement therapy in the intensive care unit. N Engl J Med. 1267
electrolyte imbalances in acute kidney injury. Adv Chronic Kidney 2016;375(2):122-133.
1213 Dis. 2016;23(3):203-210. 1268
55. Gaudry S, Hajage D, Schortgen F, et al. Timing of renal support and
1214 33. Finkel KW, Podoll AS. Complications of continuous renal outcome of septic shock and acute respiratory distress syndrome. A 1269
1215 replacement therapy. Seminars in Dialysis. 2009;22(2):155-159. post hoc analysis of the AKIKI randomized clinical trial. Am J Respir 1270
Crit Care Med. 2018;198(1):58-66. Q16
1216 34. Cerda J, Tolwani AJ, Warnock DG. Critical care nephrology: 1271
management of acid-base disorders with CRRT. Kidney Int. 56. Vanmassenhove J, Vanholder R, Van Biesen W, Lameire N. Haste
1217 1272
2012;82(1):9-18. makes waste-Should current guideline recommendations for
1218 initiation of renal replacement therapy for acute kidney injury be 1273
35. Kellum JA, Murugan R, Nadim MK. Indications, timing, and patient
1219 selection. In: Kellum JA, Bellomo R, Ronco C, eds. Continuous Renal changed? Semin Dial. 2018;31(3):204-208. 1274
1220 Q14 Replacement Therapy. Oxford University Press; 2016:48-56. 57. Van Wert R, Friedrich JO, Scales DC, Wald R, Adhikari NKJ; 1275
36. Meyer TW, Hostetter TH. Uremia. N Engl J Med. 2007;357(13): University of Toronto Acute Kidney Injury Research Group. High-
1221 dose renal replacement therapy for acute kidney injury: systematic 1276
1316-1325.
1222 review and meta-analysis. Crit Care Med. 2010;38(5):1360-1369. 1277
37. Mueller BA, Pasko DA, Sowinski KM. Higher renal replacement
1223 therapy dose delivery influences on drug therapy. Artif Organs. 58. Bouchard J, Weidemann C, Mehta RL. Renal replacement therapy in 1278
1224 2003;27(9):808-814. acute kidney injury: intermittent versus continuous? How much is 1279
enough? Adv Chronic Kidney Dis. 2008;15(3):235-247.
1225 38. Ghannoum M, Nolin TD, Lavergne V, Hoffman RS; EXTRIP 1280
Workgroup. Blood purification in toxicology: nephrology’s ugly 59. Ronco C, Bellomo R, Homel P, et al. Effects of different doses in
1226 continuous veno-venous haemofiltration on outcomes of acute 1281
duckling. Adv Chronic Kidney Dis. 2011;18(3):160-166.
1227 renal failure: a prospective randomised trial. Lancet. 1282
39. Patel N, Bayliss GP. Developments in extracorporeal therapy for the 2000;356(9223):26-30.
1228 poisoned patient. Adv Drug Deliv Rev. 2015;90:3-11. 1283
60. Bouman CS, Oudemans-Van Straaten HM, Tijssen JG, Zandstra DF,
1229 40. Spinale JM, Laskin BL, Sondheimer N, Swartz SJ, Goldstein SL. 1284
Kesecioglu J. Effects of early high-volume continuous venovenous
1230 High-dose continuous renal replacement therapy for neonatal hemofiltration on survival and recovery of renal function in 1285
hyperammonemia. Pediatr Nephrol. 2013;28(6):983-986. intensive care patients with acute renal failure: a prospective,
1231 1286
41. Hanudel M, Avasare S, Tsai E, Yadin O, Zaritsky J. A biphasic randomized trial. Crit Care Med. 2002;30(10):2205-2211.
1232 1287
dialytic strategy for the treatment of neonatal hyperammonemia. 61. VA/NIH Acute Renal Failure Trial Network, Palevsky PM,
1233 Pediatr Nephrol. 2014;29(2):315-320. Zhang JH, et al. Intensity of renal support in critically ill patients 1288
1234 42. Slack AJ, Auzinger G, Willars C, et al. Ammonia clearance with with acute kidney injury. N Engl J Med. 2008;359(1):7-20. 1289
1235 haemofiltration in adults with liver disease. Liver Int. 2014;34(1):42- 62. Palevsky PM, O’Connor TZ, Chertow GM, et al. Intensity of renal 1290
48. replacement therapy in acute kidney injury: perspective from within
1236 1291
43. Cardoso FS, Gottfried M, Tujios S, Olson JC, Karvellas CJ, US Acute the Acute Renal Failure Trial Network Study. Crit Care. 2009;13(4):
1237 Live Failure Study Group. Continuous renal replacement therapy is 310. 1292
1238 associated with reduced serum ammonia levels and mortality in 63. RENAL Replacement Therapy Study Investigators, Bellomo R, 1293
1239 acute liver failure [published online ahead of print August 31, 2017]. Cass A, et al. Intensity of continuous renal-replacement therapy in 1294
Q15 Hepatology. https://doi.org/10.1002/hep.29488. critically ill patients. N Engl J Med. 2009;361(17):1627-1638.
1240 1295
44. Wald R, Adhikari NKJ, Smith OM, et al. Comparison of standard 64. Oliver MJ, Callery SM, Thorpe KE, Schwab SJ, Churchill DN. Risk of
1241 and accelerated initiation of renal replacement therapy in acute 1296
bacteremia from temporary hemodialysis catheters by site of
1242 kidney injury. Kidney Int. 2015;88(4):897-904. insertion and duration of use: a prospective study. Kidney Int. 1297
1243 45. Gettings LG, Reynolds HN, Scalea T. Outcome in post-traumatic 2000;58(6):2543-2545. 1298
acute renal failure when continuous renal replacement therapy is 65. Parienti JJ, Thirion M, Megarbane B, et al. Femoral vs jugular venous
1244 applied early vs. late. Intensive Care Med. 1999;25(8):805-813. 1299
catheterization and risk of nosocomial events in adults requiring
1245 acute renal replacement therapy: a randomized controlled trial. 1300
46. Durmaz I, Yagdi T, Calkavur T, et al. Prophylactic dialysis in
1246 patients with renal dysfunction undergoing on-pump coronary JAMA. 2008;299(20):2413-2422. 1301
1247 artery bypass surgery. Ann Thorac Surg. 2003;75(3):859-864. 66. el-Nachef MW, Rashad F, Ricanati ES. Occlusion of the subclavian 1302
1248 47. Sugahara S, Suzuki H. Early start on continuous hemodialysis vein: a complication of indwelling subclavian venous catheters for 1303
therapy improves survival rate in patients with acute renal failure hemodialysis. Clin Nephrol. 1985;24(1):42-46.
1249 following coronary bypass surgery. Hemodial Int. 2004;8(4):320-325. 1304
67. Schwab SJ, Quarles LD, Middleton JP, Cohan RH, Saeed M,
1250 48. Elahi MM, Lim MY, Joseph RN, Dhannapuneni RR, Spyt TJ. Early Dennis VW. Hemodialysis-associated subclavian vein stenosis. 1305
1251 hemofiltration improves survival in post-cardiotomy patients with Kidney Int. 1988;33(6):1156-1159. 1306
acute renal failure. Eur J Cardiothorac Surg. 2004;26(5):1027-1031.
1252 68. Morgan D, Ho K, Murray C, Davies H, Louw J. A randomized trial 1307
49. Demirkilic U, Kuralay E, Yenicesu M, et al. Timing of replacement of catheters of different lengths to achieve right atrium versus
1253 1308
therapy for acute renal failure after cardiac surgery. J Card Surg. superior vena cava placement for continuous renal replacement
1254 2004;19(1):17-20. therapy. Am J Kidney Dis. 2012;60(2):272-279. 1309
1255 50. Manche A, Casha A, Rychter J, Farrugia E, Debono M. Early dialysis 69. Vesconi S, Cruz DN, Fumagalli R, et al. Delivered dose of renal 1310
1256 in acute kidney injury after cardiac surgery. Interact Cardiovasc replacement therapy and mortality in critically ill patients with acute 1311
Thorac Surg. 2008;7(5):829-832. kidney injury. Critical Care. 2009;13(2):R57.
1257 1312
51. Karvellas CJ, Farhat MR, Sajjad I, et al. A comparison of early versus 70. van de Wetering J, Westendorp RG, van der Hoeven JG, Stolk B,
1258 late initiation of renal replacement therapy in critically ill patients Feuth JD, Chang PC. Heparin use in continuous renal replacement 1313
1259 with acute kidney injury: a systematic review and meta-analysis. Crit procedures: the struggle between filter coagulation and patient 1314
1260 Care. 2011;15(1):R72. hemorrhage. J Am Soc Nephrol. 1996;7(1):145-150. 1315
1261 52. Wang X, Jie Yuan W. Timing of initiation of renal replacement 71. Joannidis M, Kountchev J, Rauchenzauner M, et al. Enoxaparin vs. 1316
therapy in acute kidney injury: a systematic review and meta- unfractionated heparin for anticoagulation during continuous veno-
1262 analysis. Ren Fail. 2012;34(3):396-402. venous hemofiltration: a randomized controlled crossover study. 1317
1263 Intensive Care Med. 2007;33(9):1571-1579. 1318
53. Zarbock A, Kellum JA, Schmidt C, et al. Effect of early vs delayed
1264 initiation of renal replacement therapy on mortality in critically ill 72. Reeves JH, Cumming AR, Gallagher L, O’Brien JL, Santamaria JD. 1319
patients with acute kidney injury: the ELAIN randomized clinical A controlled trial of low-molecular-weight heparin (dalteparin)
1265 1320
trial. JAMA. 2016;315(20):2190-2199. versus unfractionated heparin as anticoagulant during continuous

12 Contemporary Reviews in Critical Care Medicine [ -#- CHEST - 2018 ]

REV 5.5.0 DTD ! CHEST1962_proof ! 19 October 2018 ! 11:33 am ! EO: CHEST-18-1329
1321 venovenous hemodialysis with filtration. Crit Care Med. 1999;27(10): 85. Troyanov S, Geadah D, Ghannoum M, Cardinal J, Leblanc M. 1376
1322 2224-2228. Phosphate addition to hemodiafiltration solutions during 1377
continuous renal replacement therapy. Intensive Care Med.
1323 73. Oudemans-van Straaten HM, Bosman RJ, Koopmans M, et al. 1378
2004;30(8):1662-1665.
Citrate anticoagulation for continuous venovenous hemofiltration.
1324 Crit Care Med. 2009;37(2):545-552. 86. Culley CM, Bernardo JF, Gross PR, et al. Implementing a 1379
1325 74. Oudemans-van Straaten HM, Kellum JA, Bellomo R. Clinical review:
standardized safety procedure for continuous renal replacement 1380
therapy solutions. Am J Health Syst Pharm. 2006;63(8):756-763.
1326 anticoagulation for continuous renal replacement therapy—heparin 1381
1327 or citrate? Crit Care. 2011;15(1):202. 87. Akhoundi A, Singh B, Vela M, et al. Incidence of adverse events 1382
during continuous renal replacement therapy. Blood Purif.
1328 75. Wu MY, Hsu YH, Bai CH, Lin YF, Wu CH, Tam KW. Regional 2015;39(4):333-339. 1383
citrate versus heparin anticoagulation for continuous renal
1329 replacement therapy: a meta-analysis of randomized controlled 88. Sigwalt F, Bouteleux A, Dambricourt F, Asselborn T, Moriceau F, 1384
1330 trials. Am J Kidney Dis. 2012;59(6):810-818. Rimmele T. Clinical complications of continuous renal replacement 1385
therapy. Contrib Nephrol. 2018;194:109-117.
1331 76. Tolwani AJ, Prendergast MB, Speer RR, Stofan BS, Wille KM. 1386
A practical citrate anticoagulation continuous venovenous 89. Fernandez S, Santiago MJ, Gonzalez R, et al. Hemodynamic impact
1332 1387
hemodiafiltration protocol for metabolic control and high solute of the connection to continuous renal replacement therapy in
1333 clearance. Clin J Am Soc Nephrol. 2006;1(1):79-87. critically ill children. Pediatr Nephrol. 2018. 1388
1334 77. Tolwani A, Wille KM. Advances in continuous renal replacement 90. Schytz PA, Mace ML, Soja AM, et al. Impact of extracorporeal blood 1389
1335 therapy: citrate anticoagulation update. Blood Purif. 2012;34(2):88- flow rate on blood pressure, pulse rate and cardiac output during 1390
93. haemodialysis. Nephrol Dial Transplant. 2015;30(12):2075-2079.
1336 1391
78. Pea F, Viale P, Pavan F, Furlanut M. Pharmacokinetic considerations 91. Sherman RA. We lower blood flow for intradialytic hypotension.
1337 Semin Dial. 2016;29(4):295-296. 1392
for antimicrobial therapy in patients receiving renal replacement
1338 therapy. Clin Pharmacokinet. 2007;46(12):997-1038. 1393
92. Gibney N, Hoste E, Burdmann EA, et al. Timing of initiation and
1339 79. Wong WT, Choi G, Gomersall CD, Lipman J. To increase or discontinuation of renal replacement therapy in AKI: unanswered 1394
1340 decrease dosage of antimicrobials in septic patients during key questions. Clin J Am Soc Nephrol. 2008;3(3):876-880. 1395
continuous renal replacement therapy: the eternal doubt. Curr Opin 93. Gibney RTN, Bagshaw SM, Kutsogiannis DJ, Johnston C. When
1341 1396
Pharmacol. 2015;24:68-78. should renal replacement therapy for acute kidney injury be initiated
1342 and discontinued? Blood Purification. 2008;26(5):473-484. 1397
80. Lewis SJ, Mueller BA. Antibiotic dosing in patients with acute kidney
1343 injury: “enough but not too much”. J Intensive Care Med. 2016;31(3): 94. Uchino S, Bellomo R, Morimatsu H, et al. Discontinuation of 1398
1344 164-176. continuous renal replacement therapy: a post hoc analysis of a 1399
1345 81. Macias WL, Alaka KJ, Murphy MH, Miller ME, Clark WR, prospective multicenter observational study. Crit Care Med. 1400
Mueller BA. Impact of the nutritional regimen on protein catabolism 2009;37(9):2576-2582.
1346 1401
and nitrogen balance in patients with acute renal failure. JPEN J 95. Mendu ML, Ciociolo GR, McLaughlin SR, et al. A Decision-making
1347 Parenter Enteral Nutr. 1996;20(1):56-62. algorithm for initiation and discontinuation of RRT in severe AKI. 1402
1348 82. McClave SA, Taylor BE, Martindale RG, et al. Guidelines for the Clin J Am Soc Nephrol. 2017;12(2):228-236. 1403
1349 provision and assessment of nutrition support therapy in the adult 96. Bosslet GT, Pope TM, Rubenfeld GD, et al. An official ATS/AACN/ 1404
critically ill patient: Society of Critical Care Medicine (SCCM) and ACCP/ESICM/SCCM policy statement: responding to requests for
1350 American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.). 1405
potentially inappropriate treatments in intensive care units. Am J
1351 JPEN J Parenter Enteral Nutr. 2016;40(2):159-211. Respir Crit Care Med. 2015;191(11):1318-1330. 1406
1352 83. Ebo DG, Bosmans JL, Couttenye MM, Stevens WJ. Haemodialysis- 97. Okon TR, Vats HS, Dart RA. Palliative medicine referral in patients 1407
1353 associated anaphylactic and anaphylactoid reactions. Allergy. undergoing continuous renal replacement therapy for acute kidney 1408
2006;61(2):211-220. injury. Ren Fail. 2011;33(7):707-717.
1354 1409
84. Krieter DH, Grude M, Lemke HD, et al. Anaphylactoid reactions 98. Scherer JS, Holley JL. The Role of time-limited trials in dialysis
1355 during hemodialysis in sheep are ACE inhibitor dose-dependent and decision making in critically ill patients. Clin J Am Soc Nephrol. 1410
1356 mediated by bradykinin. Kidney Int. 1998;53(4):1026-1035. 2016;11(2):344-353. 1411
1357 1412
1358 1413
1359 1414
1360 1415
1361 1416
1362 1417
1363 1418
1364 1419
1365 1420
1366 1421
1367 1422
1368 1423
1369 1424
1370 1425
1371 1426
1372 1427
1373 1428
1374 1429
1375 1430

chestjournal.org 13

REV 5.5.0 DTD ! CHEST1962_proof ! 19 October 2018 ! 11:33 am ! EO: CHEST-18-1329