Table 3.
Performance of general and customized SAPS III equations for major geographic regions
| Study | Calibration for general SAPS III | Calibration for customized SAPS III | aROC (95% CI) | SMR (95% CI) for general SAPS III | SMR (95% CI) for customized SAPS III |
|---|---|---|---|---|---|
|
Central and South America |
|
||||
| Soares, 2006 [15] |
C: 13.637 (P = 0.092) |
C: 9.132 (P = 0.331) |
0.87 (0.85 to 0.90) |
1.19 (1.04 to 1.37) |
0.95 (0.84 to 1.07) |
| Serrato, 2007 [16] |
C: 6.64 (P >0.1) |
NR |
0.86 (0.825 to 0.895) |
0.81 |
NR |
| Alves, 2009 [17] |
H: 16.42 (P = 0.037) |
H: 16.66 (P = 0.034) |
0.881 (0.843 to 0.913) |
1.10 (0.90 to 1.33) |
0.86 (0.70 to 1.04) |
|
C: 17.57 (P = 0.025) |
C: 15.95 (P = 0.047) |
||||
| Maccariello, 2010 [18] |
C: 10.16 (P = 0.254) |
C: 9.33 (P = 0.315) |
0.82 (0.76 to 0.88) |
1.26 (1.10 to 1.46) |
1.04 (0.92 to 1.18) |
| Silva Junior, 2010 [19] |
C: 10.47 (P = 0.234) |
NR |
0.86 (0.83 to 0.88) |
1.04 (1.03 to 1.07) |
NR |
| Soares, 2010 [20] |
C: 15.804 (P = 0.045) |
C: 12.607 (P = 0.126) |
0.84 (0.81 to 0.87) |
1.29 (1.09 to 1.53) |
1.02 (0.87 to 1.19) |
| Costa e Silva, 2011 [21] |
H (DD): 6.86 (P = 0.551) |
H (DD): 6.33 (P = 0.610) |
DD: 0.73 (0.67 to 0.78) |
DD: 1.35 (1.07 to 1.63) |
DD: 1.09 (0.83 to 1.35) |
|
H (NCD): 10.47 (P = 0.163) |
H (NCD): 13.22 (P = 0.113) |
NCD: 0.80 (0.73 to 0.86) |
NCD: 1.15 (0.75 to 1.55) |
NCD: 1.00 (0.61 to 1.39) |
|
| Nassar Junior, 2012 [22] |
C: 226.6 (P <0.001) |
NR |
0.855 (0.846 to 0.864) |
0.46 (0.37 to 0.54) |
NR |
| Nassar Junior, 2013 [23] |
C: 51.8 (P <0.001) |
NR |
0.804 (0.779 to 0.828) |
0.31 (0.11 to 0.50) |
NR |
| De Oliveira, 2013 [24] |
H: 59.41 (P <0.001) |
C: 123.49 (P <0.001) |
0.696 (0.607 to 0.786) |
1.94 (1.38 to 2.64) |
1.88 (1.34 to 2.56) |
|
C: 155.57 (P <0.001) |
H: 45.6 (P <0.001) |
||||
|
North America |
|
||||
| Keegan, 2012 [25] |
C: 36.6 (P <0.05) |
NR |
0.801 (0.785 to 0.816) |
0.66 (0.69 to 0.75) |
NR |
|
Australasia |
|
||||
| Duke, 2008 [26] |
H: 36.15 (P = 0.009) |
H: 27.37 (P = 0,06) |
0.88 (0.85 to 0.90) |
0.84 (0.724 to 0.969) |
0.92 (0.80 to 1.06) |
|
C: ? (P = 0.019) | |||||
| Tsai, 2008 [31] |
H: 5.445 (P = 0.71) |
NR |
0.73 |
NR |
NR |
| Khwannimit, 2010 [27] |
H: 106.7 (P <0.001) |
H: 98.2 (P <0.001) |
0.933 (0.921 to 0.944) |
0.86 (0.79 to 0.93) |
0.92 (0.85 to 0.99) |
| C: 101.2 (P <0.001) |
C: 96.2 (P <0.001) |
||||
| Khwannimit, 2011 [28] |
H: 101.6 (P <0.001) |
H: 79.9 (P <0.001) |
0.916 (0.902 to 0.929) |
0.81 (0.74 to 0.88) |
0.88 (0.80 to 0.96) |
|
C: 176.3 (P <0.001) |
C: 170 (P <0.001) |
||||
| Lim, 2011 [32] |
C: 3.174 (P = 0.923) |
C: 3.286 (P = 0.915) |
0.78 (0.75 to 0.81) |
0.72 (0.62 to 0.83) |
0.78 (0.67 to 0.89) |
| Juneja, 2012 [29] |
HL(?): 13.12 (P = 0.108) |
NR |
0.901 (0.871 to 0.932) |
0.763 (0.628 to 0.918) |
NR |
| Khwannimit, 2013 [30] |
H: 39.4 (P <0.001) |
NR |
0.817 (0.790 to 0.845) |
0.97 (0.89 to 1.06) |
NR |
|
C: 49.6 (P <0.001) | |||||
| Lim, 2013 [33] |
H: 123.06 (P <0.001) |
H: 73.53 (P <0.001) |
0.829 (0.82 to 0.86) |
0.72 (0.65 to 0.78) |
0.78 (0.71 to 0.85) |
|
C: 118.45 (P <0.001) |
C: 70.52 (P <0.001) |
||||
|
Central and Western Europe |
|
||||
| Ledoux, 2008 [34] |
C: 16.59 (P = 0.035) |
C: 8.30 (P = 0.405) |
0.85 (0.82 to 0.89) |
0.82 (0.70 to 0.93) |
0.96 (0.84 to 1.08) |
| Sakr, 2008 [35] |
H: 211.84 (P <0.001) |
H: 177.37 (P <0.001) |
0.84 (0.81 to 0.88) |
NR |
NR |
|
C: 208.49 (P <0.001) |
C: 126.79 (P <0.001) |
||||
| Metnitz, 2009 [36] |
H: 100.18 (P <0.001) |
H: 51.56 (P <0.001) |
0.82 |
0.79 (0.74 to 0.85) |
0.86 (0.80 to 0.92) |
|
C: 90.29 (P <0.001) |
C: 45.61 (P <0.001) |
||||
|
Northern Europe |
|
||||
| Strand, 2009 [37] |
C: 17.40 (P = 0.066) |
C: 18.25 (P = 0.051) |
0.81 (0.79 to 0.93) |
0.71 (0.65 to 0.78) |
0.74 (0.68 to 0.81) |
| Christensen, 2011 [38] |
HL(?): 9.23 (P = 0.51) |
NR |
0.69 (0.63 to 0.75) |
NR |
NR |
|
Southern Europe and Mediterranean countries |
|
||||
| Capuzzo, 2009 [39] |
H: 23.36 (P = 0.002) |
H: 25.73 (P = 0.001) |
0.835 (0.794 to 0.876) |
0.83 (0.77 to 0.89) |
0.81 (0.75 to 0.87) |
|
C: 22.47 (P = 0.004) |
C: 26.19 (P = 0.001) |
||||
| Mbongo, 2009 [40] |
C: 8.57 (P = 0.38) |
C: 7.5 (P = 0.48) |
0.917 (0.880 to 0.954) |
0.71 (0.56 to 0.90) |
0.69 (0.55 to 0.87) |
| Poole, 2009 [41] |
Ua: 2,035,9 (P <0.001) |
Ua: 1,929.2 (P <0.001) |
0.855 (0.851 to 0.860) |
0.73 (0.72 to 0.75) |
0.73 (0.72 to 0.75) |
| López-Caler, 2013 [42] | HL(?): 20.05 (P <0.05) | 0.855 (0.851 to 0.860) | 0.845 (0.821 to 0.869) | 0.89 (0.80 to 0.98) | 0.86 (0.77 to 0.95) |
aROC, area under the receiver operator characteristic curve; SAP, simplified acute physiology score; SMR, standardized mortality ratios; H, Hosmer-Lemeshow H-statistic; C, Hosmer-Lemeshow C-statistic; HL(?), Hosmer-Lemeshow statistic unspecified; NR, not reported; DD, acute kidney injury diagnosis-day; NCD, day of nephrology consultation. aU statistics quantifies the difference from the null hypothesis, that is, a good fit of the model in the studied population when Cox calibration test is used.