Table 2.
Sensitivity, specificity, and positive and negative predictive values.
|
N
|
Cutoff PMN | S (%) | E (%) | PPV (%) | NPV (%) | |
|---|---|---|---|---|---|---|
| Mirra et al. (1976) [8] |
34 | 5 | 100 | 98 | - | - |
| Fehring and McAlister (1994) [9] |
107 | Total | 18 | 89 | - | - |
| Feldman et al. (1995) [10] | 33 | 5 | 100 | 96 | - | - |
| Athanasou et al. (1995) [11] | 106 | 1 | 90 | 96 | 88 | 98 |
| Lonner et al. (1996) [12] | 175 | 5 | 84 | 96 | 70 | 98 |
| Lonner et al. (1996) [12] | 175 | 10 | 84 | 99 | 89 | 98 |
| Pace et al. (1997) [13] | 25 | 5 | 82 | 93 | 90 | 87 |
| Abdul-Karim et al. (1998) [14] | 64 | 5 | 43 | 97 | - | - |
| Spangehl et al. (1999) [3] | 202 | 5 | 80 | 94 | 74 | 96 |
| Pons et al. (1999) [2] | 83 | 5 | 100 | 98 | 94 | 100 |
| Della Valle et al. (1999) [16] | 64∗ | 5 | 25 | 98 | 50 | 95 |
| Banit et al. (2002) [17] | 121 | 10 (knee and hip) | 67 | 93 | 67 | 93 |
| Banit et al. (2002) [17] |
55 | 10 (knee) | 100 | 96 | 82 | 100 |
| Banit et al. (2002) [17] | 63 | 10 (hip) | 45 | 92 | 55 | 88 |
| Musso et al. (2003) [18] |
45 | 5 | 50 | 95 | 60 | 92 |
| Ko et al. (2005) [20] |
40 | 5 | 67 | 97 | 86 | 91 |
| Wong et al. (2005) [21] |
40 | 5 | 93 | 77 | 68 | 95 |
| Wong et al. (2005) [21] |
40 | 10 | 86 | 85 | 75 | 92 |
| Francés Borrego et al. (2006) [22] |
63 | 10 (knee) | 66 | 89 | 81 | 81 |
| Francés Borrego et al. (2006) [22] |
83 | 10 (hip) | 50 | 100 | 100 | 95 |
| Bori et al. (2006) [23] |
61 | 5 | 50 | 81 | 40 | 86 |
| Nuñez et al. (2007) [25] |
136 | 5 | 85 | 87 | 79 | 91 |
| Nilsdotter-Augustinsson et al. (2007) [26] |
85 | 5 | 81 | 100 | 100 | 87 |
| Della Valle et al. (2007) [27] |
105 | 10 (knee) | 88 | 96 | 91 | 93 |
| Bori et al. (2007) [28] |
21 | 5 | 28 | 100 | 100 | 73 |
| Bori et al. (2007) [28] |
21 | 1 | 71 | 64 | 50 | 81 |
| Kanner et al. (2008) [29] |
132 | 5 | 29 | 95 | 40 | 92 |
| Müller et al. (2008) [30] |
37 | Total | 94 | 94 | 97 | 86 |
| Schinsky et al. (2008) [31] |
201 | 10 (hip) | 73 | 94 | 82 | 90 |
| Fink et al. (2008) [32] |
145 | 5 | 90 | 95 | 88 | 96 |
| Savarino et al. (2009) [34] |
31 | 1 | 80 | 100 | 100 | 80 |
| Morawietz et al. (2009) [36] |
147 | 23∗ | 73 | 95 | 91 | 84 |
| Tohtz et al. (2010) [37] |
52 | 23∗ | 86 | 100 | 100 | 94 |
| Miyamae et al. (2013) [39] |
86 | 10 | 71 | 89 | 42 | 97 |
| Ahmadi et al. (2013) [40] |
227 | 5 (elbow) | 51 | 93 | 60 | 90 |
| Muñoz-Mahamud et al. (2013) [41] |
11 | 5 (fracture) | 100 | 55 | 33 | 100 |
| Grosso et al. (2014) [42] |
44 | 5 (shoulder) | 57 | 100 | - | - |
| Grosso et al. (2014) [42] |
44 | 10 (shoulder) | 73 | 100 | - | - |
| Buttaro et al. (2015) [43] |
76 | 5 | 90 | 94 | 87 | 96 |
| Kashima et al. (2015) [44] |
76 | 2 | 94 | 97 | - | - |
| Kashima et al. (2015) [44] |
76 | 5 | 83 | 97 | - | - |
N: number of patients, PMN: polymorphonuclear neutrophil, S: sensitivity, Sp: specificity, PPV: positive predictive value, NPV: negative predictive value; ∗≥23 polymorphonuclear leukocytes in ≥10 HPF (400x). In each HPF, a maximum of 10 polymorphonuclear leukocytes were counted. The sum must be between zero and 100.