# Table 2 Summary of cohorts reporting multivariate analyses with a stated hazard ratio, (HR) risk ratio (RR) or odds ratio (OR), 95% confidence intervals, (CI) lower limit (LL) and upper limit (UL).

Author(s) | Year | MVA Type | Mortality | Censored | BIA Marker | BIA Marker MVA | HR/OR for BIA | 95% CI LL | 95% CI UL | Reason for exclusion from MA |
---|---|---|---|---|---|---|---|---|---|---|

End-stage Kidney Failure Cohorts
| ||||||||||

Demirci | 2016 | Cox analysis | 93 |
Y^{1,2,3}
| BIVA | Impedance ratio | HR = 1.13 | 1.04 | 1.23 | Only study using impedance ratio for BIVA analysis. |

Pillon/Chertow | 2004 | Unclear MVA | 361 |
Y^{2,3,4,5}
| BIVA | MVA using vector length (per 100ohm/m change) | RR = 0.75 | 0.57 | 0.88 | Only 1 study using this BIA method |

Chen | 2007 | Cox analysis | 58 |
Y^{1,2,3}
| ECW Ratio | ECW/ICW - for every increase by 0.1 (time dependent) | RR = 1.37 | 1.1 | 1.7 | Only study expressing continuous ECW/ICW variable in 0.1 increments. |

Fan | 2015 | Cox analysis | 37 | N | ECW Ratio | ECW as an absolute value (in litres) | HR = 2.98 | 1.4 | 7.3 | Only study expressing ECW as absolute volume. |

Kim | 2017 | Cox analysis | 24 |
Y^{2}
| ECW Ratio | ECW/ICW - for every increase by 0.01 | HR = 1.12 | 1.01 | 1.25 | Only study expressing continuous ECW/ICW variable in 0.01 increments. |

Paniagua | 2010 | Cox analysis | 182 | N | ECW Ratio | ECW/TBW as continuous variable in CV mortality | OR = 1171.33 | 3.35 | 409899.37 | Only study expressing ECW/TBW as continuous variable (expressed per unit ratio) |

Rhee | 2015 | Cox analysis | 15 | N | ECW Ratio | ECW/TBW > Median | HR = 1.001 | 1.001 | 1.086 | Only study expressing ECW/TBW > median |

Guo/Guo | 2015 | Cox analysis | 52 |
Y^{1,2,3,4}
| ECW Ratio | ECW/TBW > 0.4 | HR = 13.12 | 1.35 | 128 | Only study expressing ECW/TBW > 0.4 as cut off |

Tian | 2016 | Cox analysis | 44 | N | ECW Ratio | ECW Ratio > 1 standard deviation from expected | HR = 2.20 | 0.79 | 6.08 | Only study expressing ECW ratio > 1 standard deviation from expected |

O’Lone | 2014 | Cox analysis | 95 | N | ECW Ratio+OH Index | Two markers: OH/ECW and ECW/TBW (highest 30% each) | HR 2.09 (1.36, 3.20)/HR 2.05 (1.31, 3.22) | Only study expressing both indices with 30% highest decile cut off | ||

Caetano* | 2016 | Cox analysis | 66 |
Y^{1,2}
| OH Index | OH/ECW > 15% | HR = 2.22 | 1.29 | 3.79 | More than 1 study measuring OHI > 15% |

Chazot* | 2012 | Cox analysis | Unclear |
Y^{2,4}
| OH Index | OHI > 15% (dHS/ECW) | HR = 3.41 | 1.62 | 7.17 | More than 1 study measuring OHI > 15% |

Dekker* | 2017 | Cox analysis | Unclear |
Y^{1,2.3,4}
| OH Index | OHI/ECW > 15% (overhydration 2.5–5L) | HR = 2.62 | 2.1 | 3.3 | More than 1 study measuring OHI > 15% |

Hoppe | 2015 | MLR | Unclear | N | OH Index | Continuous variable | OR = 1.12 | 0.92 | 1.37 | Only study expressing OHI as continuous variable in MLR |

Huan-Sheng | 2016 | Cox analysis | 13 | N | OH Index | Absolute OH; BIA-defined protocol linked with episodes of absolute fluid overload to determine management | RCT - testing an intervention vs. control, non comparable design. | |||

Jotterand-Drepper* | 2016 | Cox analysis | 19 |
Y^{1,2,3}
| OH Index | OHI/ECW > 15% | HR = 7.82 | 1.1 | 29.07 | More than 1 study measuring OHI > 15% |

Kim* | 2015 | Cox analysis | 50 |
Y^{2}
| OH Index | OH/ECW > 15% | HR = 2.58 | 1.16 | 5.75 | More than 1 study measuring OHI > 15% |

Mathew | 2015 | MLR | 41 |
Y^{2,3}
| OH Index | Absolute OH > Median (3.1L) | OR = 2.96 | 1.04 | 8.46 | Only study expressing OHI > median |

Onofriescu | 2014 | Cox analysis | 9 |
Y^{2,3}
| OH Index | OH/ECW > 15% used to define BIA-defined overhydration in RCT of BIA-driven vs standard care. | RCT - testing an intervention vs. control, non comparable design. | |||

Onofriescu* | 2015 | Cox analysis | 66 |
Y^{1,2,3,4}
| OH Index | RFO (OH/ECW) > 15% and > 17.4% | 15%: HR 1.87 | 1.12 | 3.13 | More than 1 study measuring OHI > 15% |

Siriopol/Siriopol | 2015 | Cox analysis | 31 |
Y^{1,2,4}
| OH Index | OH/ECW > 6.68% | HR = 2.93 | 1.3 | 6.58 | Only study expressing OHI > 6.68% |

Siriopol | 2017 | Cox analysis | 89 |
Y^{1,2,4}
| OH Index | OH/ECW > 6.9% | HR = 1.34 | 0.67 | 2.68 | Only study expressing OHI > 6.9% |

Tangvorap-honkchai | 2016 | Cox analysis | 110 | N | OH Index | OH as a continuous variable | HR = 1.15 | 1.03 | 1.28 | Only study expressing OHI as a continuous variable in Cox regression. |

Wizemann* | 2009 | Cox analysis | 86 |
Y^{2,4}
| OH Index | OH/ECW > 15% | HR = 2.10 | 1.39 | 3.18 | More than 1 study measuring OHI > 15% |

Zoccali* | 2017 | Cox analysis | 5866 | N | OH Index | OH/ECW > 15% in males and > 13% in females | HR = 1.26 | 1.19 | 1.33 | More than 1 study measuring OHI > 15% |

Bebera-shvili** | 2014 | Cox analysis | 38 | N | PA | 1 degree increase PA (time varying risk) | HR = 0.61 | 0.53 | 0.71 | More than 1 study expressing PA as continuous variable |

Bebera-shvili** | 2014 | Cox analysis | 64 |
Y^{2,3}
| PA | PA - continuous variable in MVA | HR = 0.72 | 0.54 | 0.96 | More than 1 study expressing PA as continuous variable |

Fiedler | 2009 | Cox analysis | 36 |
Y^{2}
| PA | PA < 4 | HR = 2.34 | 1.O6 | 5.14 | Only study expressing PA < 4 |

Koh** | 2011 | Cox analysis | 35 | N | PA | PA - continuous variable in MVA | HR = 0.39 | 0.27 | 0.57 | More than 1 study expressing PA as continuous variable |

Segall/Segall | 2014 | Cox analysis | 11 | N | PA | PA < 5.58 | HR = 2.15 | 1.16 | 3.99 | Only study expressing PA < 5.58 |

Shin** | 2017 | Cox analysis | 15 | N | PA | PA - continuous variable in MVA | HR = 0.56 | 0.33 | 0.97 | More than 1 study expressing PA as continuous variable |

de Araujo | 2013 | Cox analysis | 13 |
Y^{1,2,4,6}
| PA+ECW Ratio | Stratified for diabetic status - PA predictive in nonDM/not predictive in DM | Stratified for diabetic status with two separate analyses | |||

Heart Failure Cohorts
| ||||||||||

Trejo-Velasco | 2016 | Cox analysis | 19 | N | BIVA | BIVA Hyperhydration (defined as > 74.3%) | HR = 2.60 | 1.10 | 6.40 | Only study expressing BIVA |

Sakaguchi | 2015 | Cox analysis | 37 | N | ECW Ratio | ECW Ratio (measured/predicted) | HR = 1.48 | 1.20 | 1.83 | Only study expressing ECW ratio (measured/predicted) |

Alves | 2016 | Cox analysis | 34 | N | PA | PA < 4.8 | HR = 2.67 | 1.21 | 5.89 | Only study expressing PA < 4.8 |

Colin-Ramirez | 2012 | Cox analysis | 66 | N | PA | PA < 4.2 | HR = 3.08 | 1.06 | 8.99 | Only study expressing PA < 4.2 |