Meta-Analysis
Copyright ©The Author(s) 2022.
World J Crit Care Med. May 9, 2022; 11(3): 178-191
Published online May 9, 2022. doi: 10.5492/wjccm.v11.i3.178
Table 1 Characteristics of included studies
Ref.
Country, design
Inter-vention duration
Major eligibility criteria
Study groups (sample size)
Median daily dose of diuretic (route)
Patients characteristics
Apte et al[21], 2008Australia; RCT72 h(1) Mechanically ventilated; and (2) On continuous IV furosemideFurosemide + Spironolactone (n = 10)97 mg (71-288) (IV); 300 mg (PO)(1) Age: 68 (55-79); (2) Male sex: 7 (70%); (3) SCr, μmol/L: -; (4) Apache II Score: 21 (15-28); and (5) Positive ventilation: 10 (100%)ICU admission for (1) Sepsis: 4 (40%); (2) Cardiovascular: 2 (20%); and (3) COPD/Resp. failure: 2 (20%). In-hospital mortality: -
Furosemide + Placebo (n = 10)168 mg (74-295) (IV)(1) Age: 67 (52-76); (2) Male sex: 6 (60%); (3) SCr, umol/L: -; (4) Apache II Score: 24 (17-26); and (5) Positive ventilation: 10 (100%)ICU admission for (1) Sepsis: 6 (60%); (2) Cardiovascular: 1 (10%); and (3) COPD/Resp. failure: 0 (0%). In-hospital mortality: -
Bihari et al[22], 2016 Australia; RCT24 h(1) Fluid overload (> 10% ICU admission weight); and (2) No prior diuretic last 48 hFurosemide (n = 20)1 mg/kg (IV); Median weight: 78 kg(1) Age: 75 (62-86); (2) Male sex: 12 (60%); (3) SCr, μmol/L: 97 (69-133); (4) Apache III Score: 68 ± 21; and (5) Positive ventilation: 14 (70%)ICU admission for (1) Sepsis: 3 (15%); (2) Cardiovascular: 3 (15%); and (3) COPD/Resp. failure: 12 (60%). In-hospital mortality: 5 (25)
Furosemide + Indapamide (n = 20)1 mg/kg (IV); 5 mg (PO)(1) Age: 70 (53-75); (2) Male sex: 14 (70%); (3) SCr, μmol/L: 91 (63-141); (4) Apache III Score: 74 (29); and (5) Positive ventilation: 10 (50%)ICU admission for (1) Sepsis: 3 (15%); (2) Cardiovascular: 4 (20%); and (3) COPD/Resp. failure: 10 (50%). In-hospital mortality: 5 (25)
Bohn et al[27], 20191United States; Observa-tional (paired groups)24 h(1) ADHF with reduced ejection fraction; and (2) Not responding to furosemide monotherapyFurosemide + Chlorothiazide (n = 34, from 108)1≥ 80 mg (IV); 500 to 1000 mg (IV)(1) Age: 64 (54-69); (2) Male sex: 74 (69%); (3) SCr, umol/L: 132 (90-187); (4) Apache II Score: 12 (9-15); and (5) Positive ventilation: -ICU admission for (1) Sepsis: -; (2) Cardiovascular: 108 (100%); and (3) COPD/Resp. failure: -. In-hospital mortality: 21 (19)
Furosemide (n = 34, from 108)1≥ 80 mg (IV)--
Furosemide + Metolazone (n = 16, from 60)1≥ 80 mg (IV); 5 to 10 mg (PO)(1) Age: 63 (54-74); (2) Male sex: 41 (68%); (3) SCr, umol/L: 142 (102-188); (4) Apache II Score: 10 (7-14); and (5) Positive ventilation: -ICU admission for (1) Sepsis: -; (2) Cardiovascular: 60 (100%); and (3) COPD/Resp. failure: -. In-hospital mortality: 1 (2)
Furosemide (n = 16, from 60)1≥ 80 mg (IV)--
Heming et al[24], 2011France; Observa-tional24 h(1) Mechanically ventilated; and (2) Acute respiratory failureFurosemide + Acetazolamide (n = 29, from 68)280 mg (40-80) (IV); 500 to 1000 mg (PO)(1) Age: 77 (73-83); (2) Male sex: 9 (31%); (3) SCr, umol/L: 66 (57-89); (4) Apache II Score: 25 (20-30); and (5) Positive ventilation: 29 (100%)ICU admission for (1) Sepsis: 6 (21%); (2) Cardiovascular: 5 (17%); and (3) COPD/Resp. failure: 16 (55%). In-hospital mortality: 10 (34)
Imiela and Budaj[25], 2017Poland; RCT96 h(1) ADHF not responding to furosemide; and (2) Significant pulmonary overloadFurosemide3 + Acetazolamide (n = 10)110 mg (± 73) (IV); 250 to 500 mg (PO)(1) Age: 73 (± 8.6); (2) Male sex: 8 (80%); (3) SCr, μmol/L: 137 (± 42); (4) Apache II Score: -; and (5) Positive ventilation: -ICU admission for (1) Sepsis: -; (2) Cardiovascular: 10 (100%); and (3) COPD/Resp. failure: -. In-hospital mortality: -
Furosemide3 (n = 10)152 mg (± 97) (IV)(1) Age: 71 (± 14); (2) Male sex: 9 (90%); (3) SCr, umol/L: 141 (± 77); (4) Apache II Score: -; and (5) Positive ventilation: -ICU admission for (1) Sepsis: -; (2) Cardiovascular: 10 (100%); and (3) COPD/Resp. failure: -
Michaud and Mintus[23], 2017United States; Observa-tional (paired groups)24 h(1) Hospitalized at the ICU; and (2) Received IV furosemide + 2nd diuretics for severe fluid overloadFurosemide + Chlorothiazide (n = 58)280 mg (40-720) (IV); 392 mg (± 225) (IV)(1) Age: 61 (± 12); (2) Male sex: 35 (60%); (3) SCr, μmol/L: 124 (± 53); (4) Apache II Score: -; and (5) Positive ventilation: -ICU admission for (1) Sepsis: 4 (6.8%); (2) Cardiovascular: 25 (43%); and (3) COPD/Resp. failure: 15 (26%). In-hospital mortality: 11 (19)
Furosemide (n = 58)193 mg (20-710) (IV)--
Furosemide + Metolazone (n = 64)240 mg (20-960) (IV); 6.8 mg (± 3.3) (PO)(1) Age: 65 (± 14); (2) Male sex: 31 (48%); (3) SCr, μmol/L: 115 (± 44); (4) Apache II Score: -; and (5) Positive ventilation: -ICU admission for (1) Sepsis: 9 (14%); (2) Cardiovascular: 24 (38%); and (3) COPD/Resp. failure: 12 (19%). In-hospital mortality: 17 (27)
Furosemide (n = 64)130 mg (20-750) (IV)--
Ng et al[28], 2013United States; Observa-tional (paired groups)48 h(1) Hospitalized at the ICCU; and (2) Failed to respond to intermittent furosemideFurosemide + Metolazone (n = 42)80 mg (80-160) (IV); 5 mg (2.5-10) (PO)(1) Age: 57 (± 13); (2) Male sex: 22 (52%); (3) SCr, μmol/L: 148 (± 88); (4) Apache II Score: -; and (5) Positive ventilation: -ICU admission for (1) Sepsis: -; (2) Cardiovascular: 42 (100%); and (3) COPD/Resp. failure: -. In-hospital mortality: 0 (0)
Furosemide (n = 42)80 mg (0-160) (IV)--
Shulenberger et al[29], 2016United States; Observa-tional (paired groups)24 h(1) ADHF with loop-diuretic resistance defined as > 160 mg/d of furosemide; and (2) Admitted in the ICUFurosemide + Chlorothiazide (n = 40, from 88)4346 mg (± 144) (IV); 508 mg (± 273) (IV)(1) Age: 59 (± 12); (2) Male sex: 26 (65%); (3) SCr, μmol/L: -; (4) Apache II Score: -; and (5) Positive ventilation: - ICU admission for (1) Sepsis: -; (2) Cardiovascular: 40 (100%); and (3) COPD/Resp. failure: -. In-hospital mortality: 3 (8.5)
Furosemide (n = 40)4351 mg (± 143) (IV)
Furosemide + Metolazone (n = 38, from 89)4261 mg (± 111) (IV); 5.7 mg (± 2.5)(1) Age: 57 (± 13); (2) Male sex: 19 (50%); (3) SCr, umol/L: -; (4) Apache II Score: -; and (5) Positive ventilation: -ICU admission for (1) Sepsis: -; (2) Cardiovascular: 38 (100%); and (3) COPD/Resp. failure: -. In-hospital mortality: 9 (24%)
Furosemide (n = 38)4263 mg (± 102) (IV)
Vánky et al[26], 1997Sweden; n-RCT (unpaired groups)24 h(1) Hospitalized at the ICU post-Cardiac surgery; and (2) Received IV furosemide for severe fluid overloadFurosemide + HCTZ + Amiloride (n = 20)87 mg (± 4) (IV); 50 mg (PO); 5 mg (PO)(1) Age: 70 (± 1.4); (2) Male sex: 15 (75%); (3) SCr, μmol/L: 98 (± 3); (4) Apache II Score: -; and (5) Positive ventilation: -ICU admission for (1) Sepsis: -; (2) Cardiovascular: 20 (100%); and (3) COPD/Resp. failure: -. In-hospital mortality: -
Furosemide (n = 57)117 mg (± 18) (IV)(1) Age: 67 (± 1.2); (2) Male sex: 40 (70%); (3) SCr, μmol/L: 105 (± 4); (4) Apache II Score: -; and (5) Positive ventilation: -ICU admission for (1) Sepsis: -; (2) Cardiovascular: 57 (100%); and (3) COPD/Resp. failure: -. In-hospital mortality: -
1Bohn et al[27]: Baseline characteristics reported are from the whole cohort. However, only critically ill patients receiving vasopressors (Chlorothiazide: 34, Metolazone: 16) were included in aggregated data.
2Heming et al[24]: Only 29 participants from the whole cohort (n = 68) received a loop-diuretic in combination with acetazolamide. All aggregated data were re-analysed using the original dataset shared by the authors.
3Some patients received torsemide. The dose was converted to furosemide equivalent.
4Shulenberger et al[29]: Only intensive care unit patients (Chlorothiazide: 40, Metolazone: 38) were included in aggregated data, after re-analysis based on the original dataset shared by the authors.
RCT: Randomized Controlled Trial; ADHF: Acute decompensated heart failure; SCr: Baseline Serum creatinine; ICU: Intensive care unit; ICCU: Intensive cardiac care unit.
Table 2 Safety events and change in serum creatinine and electrolytes at 24-h for all included studies
Ref.Treatment group24-h biochemical changes1
Safety events, n (%)
Creatinine, μmol/L
Sodium, mmol/L
Potassium, mmol/L
Bicarbonate, mmol/L
Hypona-tremia
Hypo-kalemia
Mineralocortocoid-antagonist
Apte et al[21], 2008Spironolactone + Furosemide (n = 10)+4.8 (4.1-6.9)-1.0 (?)+0.13 (?)---
Furosemide (n = 10)+23 (-4.4-39)+3.0 (?)+0.13 (?)---
Thiazides
Bihari et al[22], 2016Indapamide + Furosemide (n = 20)-5.2 ± 380 ± 0-0.4 ± 1.8+1.4 ± 6.30 (0)0 (0)
Furosemide (n = 20) -2.3 ± 14+2.0 ± 4.0-0.2 ± 0.6+0.9 ± 2.50 (0)0 (0)
Bohn et al[27], 2019CTZ + Furosemide (n = 34) -----8 (24)
MTZ + Furosemide (n = 16) -----3 (19)
Michaud and Mintus[23], 2017CTZ + Furosemide (n = 58) -18 ± 57+0.5 ± 5.6-0.4 ± 0.6+3.3 ± 5.115 (26)10 (17)
MTZ + Furosemide (n = 64) -18 ± 73-1.2 ± 5.3-0.3 ± 0.6+2.6 ± 5.625 (39)11 (17)
Ng et al[28], 2013MTZ + Furosemide (n = 42) +2.7 ± 28---18 (43)15 (35)
Shulenberger et al[29], 2016CTZ + Furosemide (n = 40) +8.8 ± 27+0.1 ± 2.3--2 (5)234 (85)3
MTZ + Furosemide (n = 38) +18 ± 35-0.7 ± 3.1--2 (5)227 (71)3
Vánky et al[26], 1997HCTZ + Amiloride + Furosemide (n = 20)-2.0 ± 7.1-----
Furosemide (n = 57) -2.0 ± 7.6-----
Carbonic anhydrase inhibitor
Heming et al[24], 2011Acetazo + Furosemide (n = 29) +4.3 ± 9.4--0.3 ± 0.4-3.6 ± 5.1-9 (31)
Imiela and Budaj[25], 2017Acetazo + Furosemide (n = 10)------
Furosemide (n = 10)------
1Results are presented in median (IQR), or mean ± SD change within 24-h, from baseline.
2Only severe hyponatremia event (Na+ < 125 mmol/L) were reported.
3Hypokalemia was defined as K+ < 4.0 mmol/L, instead of 3.5 mmol/L for all other studies.
CTZ: Chlorothiazide; MTZ: Metolazone; HCTZ: Hydrochlorothiazide; Acetazo: Acetazolamide.