Table 2.
Manipulation of fluid output included studies.
| Reference | Study Design | Country | Sample Size | AKI Definition | Population | RRT | Intervention | Comparator | Assessment of Fluid Status | Duration of Intervention | Difference in FB | Outcome Measures |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Non-RRT patients | ||||||||||||
| Bagshaw 2017 | MC Pilot RCT | Canada & Australia | 73 (37/36) | RIFLE | AKI, RIFLE - RISK; ICU; CVL; IDC; SIRS = 2+; ‘Resuscitated' | Non-RRT patients | Furosemide bolus & infusion; titrated to UO | Placebo | Fluid balance | 7 days; until RRT, death, or discharge | 877 mL vs 2407 mL (- 1081 mL; 95% CI, −2697 to 467) | Worsening AKI (43.2% vs. 37.1%, p = 0.6); Kidney recovery (29.7% vs. 42.9%, p = 0.3); RRT (27.0% s. 28.6%, p = 0.8) |
| Cagliani 2021 | SC Retrospective | USA | 126 (39/87) | RIFLE | AKI, RIFLE - R, I or F; ICU; Surgical | Non-RRT patients | Furosemide & fenoldopam | Furosemide alone | Fluid balance | 24 h | R group: 1588 mL [774 mL–3765mL] vs 1074 mL [612 mL–3996mL]; p = 0.07. I or F group: 5179 mL [2121 mL–7233mL] vs 4230 mL [2132 mL–7843 mL]; p = 0.06. | UO (421 mL vs 320 mL; p = 0.22); FB (4230 mL vs 5179 mL; p = 0.06); CrCl (48.5 vs 40.7; p = 0.53). |
| Zhao 2020 | MC Retrospective | USA | 14154 (7885/6269) | KDIGO | ICU; AKI | Non-RRT patients | Furosemide | No diuretic | Fluid balance | Variable | Net FB (−575 mL vs −562 mL; p = 0.5); Positive FB (33.4% vs 33%; p = 0.752). | Increase AKI stage (HR 1.13; 95% CI 1.05–1.21; p < 0.01); RRT (HR 1.5; 95% CI 1.29–1.76; p < 0.01); In-hospital mortality (12.7% vs 21.7%; p < 0.01). |
| RRT Patients | ||||||||||||
| Nuchpramool 2019 | SC RCT | Thailand | 36 (17/19) | On CRRT. | ICU; CRRT | RRT as inclusion criteria | BIVA-guided fluid management | Standard care | BIVA & fluid accumulation (unclear if FB or weight-based) | Not provided | Details not provided. “ … did not provide beneficial effects in the rate of fluid removal indicating by body weight and %FA." | Mortality at day 28 (47% vs 52.6%; p = 0.78) |
| RashidFarokhi 2022 | SC RCT | Iran | 65 (32/33) | Not provided. On CRRT. | ICU; CRRT | RRT as inclusion criteria | BIVA-guided, UFnet prescription | Clinical parameters guided UFnet prescription | Bioimpedance vector analysis (BIVA) | ∼50 h. | BIVA-defined hypervolaemia (31.3 vs 63.6%, p = 0.009) | LBM (80.7 vs. 85.9; OR: 5.2; 95% CI: 0.2 to 10.1; p < 0.05); UO (0.9 mL/kg/hr; OR: 0.6; 95% CI: 0.4–1.1; p = 0.04); Mortality at 30 days (53.15 vs 60.6%; OR: 0.7; 95% CI: 0.3–2; p = 0.54) |
| Shin 2021 | SC Retrospective | South Korea | 216 (88/42) | Not provided | ICU; CRRT | RRT as inclusion criteria | <20% downtime | ≥ 20% downtime | Fluid balance | Not provided | Not provided. Daily fluid balance lower on day 2 (p = 0.046) and day 3 (p = 0.031). | Mortality (p = 0.95). |
| Mishra 2017 | SC RCT | India | 60 (30/30) | KDIGO | ICU; CRRT; AKI; Septic shock | RRT as inclusion criteria | SLED | CRRT | Fluid balance | Median 3.5 days | Fluid balance per 24 h (0.79 L vs 0.68; p = 0.10) | Haemodynamic stability, delta VD (39 vs 42; p = 0.39) |
| McCausland 2016 | Post-hoc analysis of ATN MC RCT | USA | 871 (436/435) | On CRRT. | ICU; AKI; CRRT | RRT as inclusion criteria | Less intense dialysis (IHD or SLED 3x/week or CRRT at 20 mL/kg/hr) | Intensive dialysis (IHD or SLED 6x/week or CRRT at 35 mL/kg/hr) | Fluid balance | 5-6 treatments | Net balance (- 200 mL vs 4 mL; p < 0.01) | UO (159 mL vs 106 mL; p < 0.01); UF volume (1850 mL vs 1700 mL; p = 0.22) |
| Gaudry 2018 | Post-hoc analysis of AKIKI MC RCT | France | 348 (174/174) | On CRRT. | ICU; Septic shock; AKI; | RRT as intervention | Early RRT (post-randomisation) | Late RRT (specific criteria to commence) | Fluid balance | Until CRRT cessation or ICU discharge | FB first 48 h (2.2 L vs 2 L; p = 0.93); Fluid input first 48 h (4.1 L vs 4.1 L; p = 0.55); UO first 48 h (994 mL vs 1881 mL; P < 0.001); UF first 48 h (3.6 L vs 2.1 L; p < 0.01). | RRT dependence at 60 days (3% vs 3%; p = 0.62) |
| Wald 2015 | Pilot MC RCT | Canada | 101 (48/52) | Not provided. On CRRT. | ICU; CRRT | RRT as inclusion criteria | Accelerated RRT (less than 12 h) | Standard Care | Fluid balance | 14 days | FB at day 14 (−1336 mL vs −57 mL) | Mortality at day 90 (38% vs 37%; p = 0.92) |
| Xing 2019 | SC Retrospective | China | 141 (57/84) | RIFLE | ICU; Septic AKI; CVP data | RRT as inclusion criteria | Early initiation (within 12 h of F criteria) | Delay initiation (delay 48 h from F criteria) | Fluid balance | 5 days; death or discharge | FB (1402 mL vs 1543 mL; p = 0.65) | FB (as shown) |
| Wald 2022 | Post-hoc analysis of START-AKI MC RCT | Multinational | 2716 (1366/1350) | KDIGO | ICU; AKI; CRRT | RRT as inclusion criteria | Accelerated strategy for CRRT initiation | Standard strategy | Fluid balance | 14 days | FB at 14 days (4509 mL vs 5646 mL; p = 0.03) | Mortality at 90 days by FB quartiles 1–4 (1 = 40% vs 2 = 45.5% vs 3 = 45.2% vs 4 = 44.9%; p = 0.17) |
| Gaudry 2021 |
MC RCT | France | 278 (137/141) | KDIGO | ICU; AKI; CRRT | RRT as inclusion criteria | Delayed strategy | More delayed strategy | Fluid balance | Until CRRT cessation of ICU discharge | FB at 2 days (1584 mL vs 1581 mL; p 0.99); FB at 7 days (1744 mL vs 2072 mL; p = 0.79) | RRT-free days at 60 days (12 vs 10; p = 0.93) |
| Murugan 2018 | SC Retrospective | USA | 1075 (475/166/434) | On CRRT. | ICU; AKI; CRRT | RRT as inclusion criteria | Net ultrafiltration rate | Different rates (mL/kg/day): 1) <20, 2) 20 - <25 3) >25 | Fluid balance adjusted to body weight | Median 4.7–8.7 days | Cumulative FB at day 7 (10.1 vs 10.5 vs 10.1; p = 0.78); UF (19.5 L vs 27.9 L vs 26.6 L; p < 0.01); FB excluding UF (13.5 L vs 22 L vs 19 L; p < 0.01). | Renal recovery at 1 year in survivors (82.6% vs 72.7% vs 78.4%; p = 0.25); Mortality at 1 year (69.7% vs 60.2% vs 59.4%; p = 0.003) |
| Murugan 2019 | Post-hoc analysis of RENAL MC RCT | Australia & New Zealand | 1434 (477/479/478) | On CRRT. | ICU; CRRT | RRT as inclusion criteria | Net ultrafiltration rate | Different rates (mL/kg/hr): 1) <1.01, 2) 1.01–1.75 3) >1.75 | Fluid balance | Until CRRT cessation or ICU discharge | Cumulative FB (2.3 L vs −0.4 L vs −3.6 L; p < 0.01). NUF (1.7 L vs 8.5 L vs 16.5 L; p < 0.01); FB excluding NUF (4.6 L vs 8.5 L vs16.5 L; p < 0.01) | RRT dependence at day 90 (3.8% vs 5.8% vs 6.9%; p = 0.28); Mortality at day 90 (44.9% vs 39.2% vs 48.6%; p = 0.01) |
| Naorungroj 2020 | SC Retrospective | Thailand | 347 (159/102/86) | On CRRT | ICU; CRRT | RRT as inclusion | Net ultrafiltration | Different rates (mL/kg/hr): 1) <1.01, 2) 1.01–1.75 3) >1.75 | Fluid balance | Until CRRT cessation or ICU discharge | Cumulative FB (527.0 mL vs −657 mL vs −1751 mL; p < 0.01); | Unadjusted RRT dependence at day 28 (21.3% vs 17.9% vs 18.4%; p = 0.66); Unadjusted mortality at day 28 (25.8% vs 32.7% vs 18.4%; p = 0.066). |
| Wu 2021 | MC Retrospective | USA | 911 (165/369/377) | Not provided. On CRRT. | ICU; Sepsis; CRRT | RRT as inclusion criteria | Net ultrafiltration rate | Different rates (mL/kg/hr): 1) <1.6, 2) 1.6–3.1, 3) >3.1 | Fluid overload expressed as percentage | Until CRRT cessation or ICU discharge | Cumulative fluid overload in 48 h (−0.6% vs −2.8% vs −4.8%; p < 0.01) | In-hospital mortality (53.3% vs 40.4% vs 44.3%; p = 0.021); Renal recovery (55.2% vs 71.0% vs 69.5%; p < 0.01). |
| Non-RRT and RRT Patients | ||||||||||||
| Berthelsen 2018 | MC Pilot RCT | Denmark | 20 (7/13) | Renal Recovery Score (RRS) | RRS, Mod-High Risk (<60%); ICU <24 hrs; 10% fluid accumulation | RRT commenced as per protocol | Forced fluid removal to achieve CFB <1000 mL. Furosemide bolus & infusion then CRRT. | Standard care | Fluid balance | 5 days; death or discharge | - 8434 mL vs - 641 mL (5814 mL; 95% CI: 2063–9565, P = 0.003) | CFB (as shown); Mean DFB (−1269 mL vs 133 mL; p < 0.01); Achieve neutral FB (86% vs 30%; p = 0.06) |
Abbreviations: AKI = acute kidney injury; SC = single centre; MC = multicentre; RCT = randomised controlled trial; CRRT = continuous renal replacement therapy; RRT = renal replacement therapy; ICU = intensive care unit; FB = fluid balance; UO = urine output; CFB = cumulative fluid balance; CVL = central venous line; IDC = indwelling catheter; SIRS = systemic inflammatory response syndrome; CI = confidence interval; NUF = net ultrafiltration rate; STARRT-AKI = STandard versus Accelerated initiation of Renal Replacement Therapy in Acute Kidney Injury; ATN = Acute Renal Failure Trial Network; AKIKI = Artificial Kidney Initiation in Kidney Injury; RENAL = Randomized Evaluation of Normal versus Augmented Level; BIVA = bioimpedance vector analysis.