Lisanne de Koster

123 Diagnostic utility of molecular and imaging biomarkers 2 Table 10. Estimated pooled test performance of RAS mutation analysis in indeterminate thyroid nodules reference ni test + test - test ? nH (%) TP FP FN TN malignant (TP+FN/ni) sensitivity (95% CI), % specificity (95% CI), % LR+ (95% CI) LR- (95% CI) Cantara 2010 41 3 38 0 41 (100%) 2 1 5 33 17% 29 (4-71) 97 (85-100) 9.71 (1.02-92.95) 0.74 (0.46-1.18) Moses 2010 110 8 102 0 110 (100%) 4 4 25 77 26% 14 (4-32) 95 (88-99) 2.79 (0.75-10.45) 0.91 (0.78-1.06) Nikiforov 2011 461 53 408 0 461 (100%) 45 8 48 360 20% 48 (38-59) 98 (96-99) 22.26 (10.87-45.58) 0.53 (0.43-0.64) Mancini 2012 48 8 40 0 38 (79%) 2 3 7 26 19% 22 (3-60) 90 (73-98) 2.15 (0.42-10.91) 0.87 (0.60-1.26) Ohori 2013 424 50 334 40 384 (100%) 46 4 68 266 27% 40 (31-50) 99 (96-100) 27.24 (10.04-73.89) 0.61 (0.52-0.70) BeaudenonHuibregtse 2014 80 12 68 0 41 (51%) 8 2 9 22 21% 47 (23-72) 92 (73-99) 5.65 (1.37-23.35) 0.58 (0.36-0.92) Danilovic 2014 136 0 136 0 136 (100%) 0 0 30 106 22% 0 (0-12) 100 (97-100) 3.45 (0.07-170.43) 0.99 (0.94-1.04) Eszlinger 2014 (1) 164 15 126 23 141 (100%) 2 13 20 106 13% 9 (1-29) 89 (82-94) 0.83 (0.20-3.43) 1.02 (0.88-1.18) Eszlinger 2014 (2) 125 11 114 0 64 (51%) 1 4 8 51 7% 11 (0-48) 93 (82-98) 1.53 (0.19-12.17) 0.96 (0.75-1.22) Liu 2014 52 5 45 2 50 (100%) 1 4 7 38 15% 12 (0-53) 90 (77-97) 1.31 (0.17-10.26) 0.97 (0.73-1.28) Nikiforov 2014 143 24 119 0 143 (100%) 20 4 19 100 27% 51 (35-68) 96 (90-99) 13.33 (4.86-36.55) 0.51 (0.37-0.70) Eszlinger 2015 163 21 137 5 158 (100%) 12 9 33 104 28% 27 (15-42) 92 (85-96) 3.35 (1.52-7.39) 0.80 (0.66-0.96) Gill 2015 60 16 44 0 60 (100%) 8 8 15 29 38% 35 (16-57) 78 (62-90) 1.61 (0.70-3.69) 0.83 (0.59-1.17) Giovanella 2015 61 4 57 0 61 (100%) 3 1 13 44 26% 19 (4-46) 98 (88-100) 8.44 (0.94-75.40) 0.83 (0.65-1.06) Hwang 2015 123 47 76 0 37 (30%) 27 0 8 2 28% 77 (60-90) 100 (16-100) 4.58 (0.36-57.93) 0.28 (0.13-0.62) Le Mercier 2015 34 6 28 0 34 (100%) 4 2 3 25 21% 57 (18-90) 93 (76-99) 7.71 (1.76-33.88) 0.46 (0.20-1.10) Nikiforov 2015 462 17 445 0 95 (21%) 11 3 11 70 5% 50 (28-72) 96 (88-99) 12.17 (3.72-39.76) 0.52 (0.34-0.79) Rossi 2015 118 6 112 0 40 (34%) 1 4 17 18 15% 6 (0-27) 82 (60-95) 0.31 (0.04-2.50) 1.15 (0.92-1.45) Yoon 2015 202 31 167 4 62 (31%) 7 4 39 12 23% 15 (6-29) 75 (48-93) 0.61 (0.20-1.81) 1.13 (0.83-1.54) Valderrabano 2016 105 13 90 2 56 (54%) 8 3 11 34 18% 42 (20-67) 92 (78-98) 5.19 (1.55-17.35) 0.63 (0.42-0.94) POOLED RESULTS 3,112 350 2,686 76 2,212 (72.9%) 212 81 396 1,523 19.5% 27.3 (18.3-38.7) 94.4 (91.5-96.3) 4.87 (2.67-8.90) 0.77 (0.66-0.89) FN: false negative. FP: false positive. LR+: positive likelihood ratio. LR-: negative likelihood ratio. nH: number of indeterminate thyroid nodules with conclusive index test and histopathological follow-up. ni: number of indeterminate thyroid nodules. test+: number of positive index tests. test-: number of negative index tests. test?: number of unknown or nondiagnostic index tests. TN: true negative. TP: true positive.

RkJQdWJsaXNoZXIy MTk4NDMw