Copyright
©The Author(s) 2019.
World J Gastroenterol. May 21, 2019; 25(19): 2383-2401
Published online May 21, 2019. doi: 10.3748/wjg.v25.i19.2383
Published online May 21, 2019. doi: 10.3748/wjg.v25.i19.2383
Table 1 Characteristics of the studies included in the meta-analysis
Test | Study, Year | Demographic characteristics | CRC | AN | SCL | Exclusion criteria | Symptoms, % | ||||||||||||
N | Age, | Sex, | Area | IBD | OB | AD | AnS | WeL | AbPa | Hem | ChBo | Co | Di | An | |||||
m/md | W% | % | % | % | |||||||||||||||
Mixed cohorts | |||||||||||||||||||
OC-S | Rozen, 2010[17] | 1682 | 63.7 | 49.6 | IL | 1.2 | 8.9 | 0 | yes | yes | yes | 23 | NA | NA | 0 | NA | NA | NA | NA |
OC-S | Mc Donald, 2012[20] | 280 | 631 | 59.6 | UK (S) | 2.1 | NA | 21.4 | NA | no | no | NA | NA | NA | NA | NA | NA | NA | NA |
OC-S | Ou, 2013[21] | 694 | 59.51 | 55.9 | CN | 0.4 | 6.1 | NA | yes | yes | no | NA | NA | NA | NA | NA | NA | NA | NA |
OC-S | van Turenhout, 2014[18] | 3022 | 59.7 | 55.0 | NL | 2.3 | 12.3 | NA | yes | yes | no | 44 | 2.9 | 11.7 | 0 | 18.1 | 3 | 4.2 | 0 |
OC-S | Symonds, 2016[23] | 1381 | 64.11 | 50.6 | AU | 4.8 | 17.2 | NA | no | no | no | 34.8 | NA | NA | NA | NA | NA | NA | NA |
HM-J | Woo, 2005[19] | 85 | 561 | 52.9 | KR | 7.1 | NA | NA | NA | no | no | 49.4 | 0 | 15.3 | 4.7 | 1.2 | 0 | 17.6 | 4.7 |
HM-Ja | Auge, 2016[22] | 208 | 631 | 55.8 | ES | 1.0 | 14.0 | NA | yes | yes | yes | NA | NA | NA | 0 | NA | NA | NA | NA |
FOB Gold® | Auge, 2018[30] | 487 | 62 | 51.2 | ES | 2.5 | 14.6 | NA | no | no | yes | 54.2 | NA | NA | NA | NA | NA | NA | NA |
100% Symptomatic cohorts | |||||||||||||||||||
OC-S | Mowat, 2016[26] | 750 | 641 | 54.7 | UK (S) | 3.7 | NA | 13.6 | no | no | no | 100 | 0.9 | 11 | 34.2 | 42.8 | NA | 16.8 | 8.9 |
OC-S | Rodriguez-Alonso, 2015[25] | 1003 | NA | 46.8 | ES | 3.0 | 13.3 | 23.4 | yes | no | no | 100 | 19 | 36.4 | 34.2 | NA | 12.1 | 23.5 | 8.8 |
OC-S | Cubiella, 2014 (DC)[29] | 1567 | 66.9 | 48.6 | ES | 13.7 | 26.7 | 29.5 | no | no | no | 100 | 24.5 | 43.8 | 59.9 | 57.2 | 14.5 | 22.2 | 34.8 |
OC-S | Cubiella, 2017 (VC)[31] | 715 | 64.4 | 53.3 | ES | 9.4 | 21.1 | 25.3 | no | no | no | 100 | NA | NA | 54 | 47.9 | NA | NA | NA |
HM-J | Parente, 2012[24] | 280 | 67 | 43.9 | IT | 16.8 | 47.2 | 0 | yes | no | no | 100 | 11.1 | 17.9 | 26.1 | 23.9 | NA | NA | 15 |
HM-Ja | Godber, 2016[27] | 484 | 591 | 60.1 | UK (S) | 2.3 | NA | 9.3 | no | no | no | 100 | 1.7 | 18.8 | 15.9 | 39.7 | NA | NA | 4.8 |
HM-Ja | Widlack, 2017[28] | 430 | 671 | 51.0 | UK (E) | 5.6 | NA | NA | no | no | no | 100 | 15.8 | 30 | 43 | 64.2 | NA | NA | 17.2 |
Table 2 Possibility of data extraction on the accuracy of quantitative faecal immunochemical test for haemoglobin for detecting colorectal cancer, advanced neoplasia and significant colonic lesion
Study | CRC | AN | SCL | ||||||||||||||||
Threshold (U) | SxD | SD | Threshold (U) | SxD | Threshold (U) | SxD | |||||||||||||
0 | 10 | 15 | 20 | Other | 0 | 10 | 15 | 20 | Other | 0 | 10 | 15 | 20 | Other | |||||
Rozen, 2010[17] | x | x | x | 25; 30; 40 | x | x | x | 25; 30; 40 | |||||||||||
Van Turenhout, 2014[18] | x | x | x | 40 | x | ||||||||||||||
Mc Donald, 2012[20] | x | HRA | x | ||||||||||||||||
Ou, 2013[21] | 5 | 5; HRA | |||||||||||||||||
Symonds, 2016[23] | x | 60; 80 | x | ||||||||||||||||
Auge, 2016[22] | x | x | x | 30; 40 | x | ||||||||||||||
Woo, 2005[19] | 33 | 3; HRA | |||||||||||||||||
Auge, 2018[30] | x | x | 30; 40; 50; 60 | x | x | x | 30; 40; 50; 60 | x | x | 30; 40; 50; 60 | |||||||||
Cubiella, 2014 (DC)[29] | x | x | x | x | x | x | x | x | x | x | x | x | x | x | x | ||||
Cubiella, 2016 (VC)[31] | x | x | x | x | x | x | x | x | x | x | x | x | x | x | x | ||||
Rodríguez-Alonso 2015[25] | x | x | x | x | x | x | x | x | |||||||||||
Mowat, 2016[26] | x | x | HRA | HRA | x | x | |||||||||||||
Parente, 2012[24] | x | x | |||||||||||||||||
Godber, 2016[27] | x | x | x | 25; 30; 35; 40 | |||||||||||||||
Widlack, 2017[28] | 7; HGD |
Table 3 Colorectal cancer detection: Diagnostic accuracy parameters based on quantitative faecal immunochemical test for haemoglobin threshold concentration and brand (DerSimonian´s method)"
Variable | Studies (n) | Sensitivity1 | I22 | Specificity1 | I22 | Positive LR3 | I22 | Negative LR3 | I22 | Diagnostic OR3 | I22 | Pa | |
OC-Sensor, > LoD µg Hb/g faeces | |||||||||||||
All studies | 4 | 98.2 (96.2-99.3) | 0.0 | 35.8 (34.2-37.3) | 96.1 | 1.55 (1.37-1.75) | 94.2 | 0.07 (0.03-0.14) | 0.0 | 21.41 (10.07-45.5) | 0.0 | 0.6 | |
OC-Sensor, ≥ 10 µg Hb/g faeces | |||||||||||||
All studies | 8 | 90.8 (87.9-93.2) | 69.7 | 79.9 (79.1-80.7) | 99.4 | 4.79 (2.96-7.76) | 99.1 | 0.15 (0.09-0.23) | 52.7 | 31.44 (19.50-50.68) | 44.7 | 0.09 | |
100% Symptomatic | 4 | 94.4 (91.4-96.6) | 0.0 | 65.9 (64.4-67.4) | 99.3 | 2.97 (1.78-4.95) | 99.0 | 0.10 (0.06-0.15) | 0.0 | 28.49 (17.77-45.67) | 0.0 | 0.6 | |
Mixed patients | 4 | 83.2 (76.5-88.6) | 44.5 | 88.2 (87.4-89.0) | 96.7 | 7.78 (4.72-12.82) | 95.1 | 0.21 (0.13-0.33) | 30.7 | 35.36 (14.19-88.10) | 71.0 | 0.6 | |
CRC prevalence ≥ 2.5% | 5 | 91.9 (88.7-94.3) | 76.0 | 69.7 (68.5-71.0) | 99.2 | 3.16 (1.99-5.0) | 98.8 | 0.13 (0.07-0.25) | 66.5 | 23.20 (14.76-36.47) | 20.0 | 0.2 | |
CRC prevalence < 2.5% | 3 | 86.3 (77.7-92.5) | 48.0 | 90.2 (89.4-91.1) | 72.6 | 9.21 (7.23-11.74) | 55.1 | 0.17 (0.09-0.33) | 26.4 | 52.33 (27.23-100.58) | 10.0 | 0.7 | |
OC-Sensor, ≥ 15 µg Hb/g faeces | |||||||||||||
All studies | 5 | 91.0 (87.8-93.6) | 73.3 | 81.8 (80.9-82.7) | 99.7 | 4.77 (2.34-9.71) | 99.4 | 0.15 (0.09-0.25) | 57.3 | 36.64 (20.43-65.71) | 49.7 | 0.04 | |
100% Symptomatic4 | 3 | 93.6 (90.2-96.0) | 32.6 | 65.8 (64.1-67.5) | 99.5 | 2.91 (1.46-5.78) | 99.3 | 0.11 (0.07-0.16) | 0.0 | 29.10 (12.74-66.46) | 35.5 | 0.67 | |
OC-Sensor, ≥ 20 µg Hb/g faeces | |||||||||||||
All studies | 5 | 90.3 (86.9-93.0) | 75.6(86.9-93.0) | 83.4 (82.5-84.2) | 99.7 | 5.30 (2.47-11.34) | 99.4 | 0.15 (0.09-0.27) | 67.2 | 39.02 (21.48-70.88) | 56.1 | 0.04 | |
100% Symptomatic4 | 3 | 92.9 (89.5-95.5) | 26.1 | 68.0 (66.3-69.7) | 99.5 | 3.14 (1.52-6.50) | 99.3 | 0.11 (0.07-0.17) | 0.0 | 29.81 (15.05-59.04) | 29.2 | 0.67 |
Table 4 Diagnostic accuracy parameters for colorectal cancer detection based on quantitative faecal immunochemical test for haemoglobin threshold concentration and brand
Brand | Threshold(µg/g faeces) | Author, year | Sensitivity1 | Specificity1 |
Mixed cohorts | ||||
OC-Sensor® | 25 | Rozen, 2010[17] | 60.0 (36.1-80.9) | 95.2 (94.0-96.2) |
OC-Sensor® | 30 | Rozen, 2010[17] | 55.0 (31.5-76.9) | 95.8 (94.7-96.7) |
OC-Sensor® | 40 | Rozen, 2010[17] | 55.0 (31.5-76.9) | 96.3 (95.3-97.2) |
OC-Sensor® | 40 | van Turenhout, 2014[18] | 75.4 (63.5-84.9) | 94.8 (93.9-95.6) |
OC-Sensor® | 60 | Symonds 2016[23] | 63.6 (50.9-75.1) | 91.9 (90.3-93.3) |
OC-Sensor® | 80 | Symonds 2016[23] | 59.1 (46.3-71.0) | 93.4 (91.9-94.7) |
HM-JACK® | 33 | Woo, 2005[19] | 50.0 (11.8-88.2) | 83.5 (73.5-90.9) |
FOB Gold® | 10 | Auge, 2018[30] | 91.7 (71.5-98.5) | 82.2 (79.6-84.5) |
FOB Gold® | 20 | Auge, 2018[30] | 87.5 (66.5-96.7) | 86.0 (83.6-88.1) |
FOB Gold® | 30 | Auge, 2018[30] | 83.3 (61.8-94.5) | 89.2 (87.0-91.1) |
FOB Gold® | 40 | Auge, 2018[30] | 83.3 (61.8-94.5) | 90.3 (88.2-92.1) |
FOB Gold® | 50 | Auge, 2018[30] | 83.3 (61.8-94.5) | 91.4 (89.4-93.1) |
FOB Gold® | 60 | Auge, 2018[30] | 83.3 (61.8-94.5) | 91.8 (89.9-93.5) |
100% Symptomatic | ||||
HM-JACK® | 20 | Parente, 2012[24] | 61.7 (46.4-75.5) | 88.8 (84.1-92.6) |
HM-JACKarc® | 7 | Widlak, 2017[28] | 84.0 (63.9-95.5) | 93.1 (90.2-95.4) |
Table 5 OC-Sensor® diagnostic accuracy parameters for colorectal cancer detection (Threshold 10 µg Hb/g faeces) estimated with DerSimonian vs Bivariate methods
Variable | Studies (n) | Sensitivity1 | Specificity1 | Positive LR2 | Negative LR2 | Diagnostic OR2 |
All studies (DS) | 8 | 90.8 (87.9-93.2) | 79.9 (79.1-80.7) | 4.79 (2.96-7.76) | 0.15 (0.09-0.23) | 31.44 (19.50-50.68) |
All studies (Bv) | 8 | 89.6 (82.7-94.0) | 80.2 (67.2-88.9) | 4.52 (2.73-7.50) | 0.13 (0.08-0.20) | 34.85 (20.74-58.57) |
100% Symptomatic (DS) | 4 | 94.4 (91.4-96.6) | 65.9 (64.4-67.4) | 2.97 (1.78-4.95) | 0.10 (0.06-0.15) | 28.49 (17.77-45.67) |
100% Symptomatic (Bv) | 4 | 94.1 (90.0-96.6) | 66.0 (47.1-80.9) | 2.77 (1.69-4.55) | 0.09 (0.06-0.14) | 30.93 (16.09-59.45) |
Mixed patients (DS) | 4 | 83.2 (76.5-88.6) | 88.2 (87.4-89.0) | 7.78 (4.72-12.82) | 0.21 (0.13-0.33) | 35.36 (14.19-88.10) |
Mixed patients (Bv) | 4 | 85.5 (76.5-91.4) | 89.3 (84.1-93.0) | 8.01 (5.07-12.65) | 0.16 (0.10-0.28) | 49.35 (19.88-122.5) |
CRC prevalence ≥ 2.5% (DS) | 5 | 91.9 (88.7-94.3) | 69.7 (68.5-71.0) | 3.16 (1.99-5.0 ) | 0.13 (0.07-0.25) | 23.20 (14.76-36.47) |
CRC prevalence ≥ 2.5% (Bv) | 5 | 91.7 (83.3-96.1) | 69.3 (53.5-81.6) | 2.99 (1.97-4.53) | 0.12 (0.07-0.21) | 24.95 (16.02-38.86) |
CRC prevalence < 2.5% (DS) | 3 | 86.3 (77.7-92.5) | 90.2 (89.4-91.1) | 9.21 (7.23-11.74) | 0.17 (0.09-0.33) | 52.33 (27.23-100.58) |
CRC prevalence < 2.5% (Bv) | 3 | 84.9 (73.4-92.0) | 90.5 (89.0-91.9) | 8.96 (7.63-10.53) | 0.17 (0.09-0.30) | 53.77 (26.99-107.11) |
Table 6 Advanced neoplasia and significant colonic lesion detection: Diagnostic accuracy parameters based on quantitative faecal immunochemical test threshold concentration and brand (DerSimonian´s method)
Variable | Studies(n) | Sensitivity1 | I22 | Specificity1 | I22 | Positive LR3 | I22 | Negative LR3 | I22 | Diagnostic OR3 | I22 | Pa |
Advanced neoplasia | ||||||||||||
OC-Sensor, > LoD µg Hb/g faeces | ||||||||||||
All studies | 3 | 91.0 (88.7-93.0) | 87.9 | 36.9 (35.0-38.8) | 95.2 | 1.40 (1.35-1.45) | 0.0 | 0.26 (0.16-0.44) | 76.6 | 5.44 (3.48-8.48) | 58 | < 0.001 |
OC-Sensor, ≥ 10 µg Hb/g faeces | ||||||||||||
All studies | 5 | 67.9 (65.1-70.5) | 97.4 | 81.0 (80.0-82.0) | 99.5 | 3.42 (1.97-5.94) | 98.8 | 0.41 (0.30-0.57) | 93.4 | 9.43 (8.10-10.98) | 0.0 | < 0.001 |
100% Symptomatic | 3 | 79.7 (76.5-82.6) | 94.6 | 67.3 (65.5-69.1) | 99.4 | 2.43 (1.41-4.17) | 98.5 | 0.32 (0.21-0.49) | 86.5 | 8.67 (6.96-10.80) | 5.8 | < 0.001 |
Prevalence CRC ≥ 2.5% | 4 | 71.7 (68.8-74.5) | 97.1 | 76.7 (75.4-77.9) | 99.5 | 2.96 (1.65-5.30) | 98.9 | 0.36 (0.25-0.53) | 91.8 | 9.29 (7.79-11.09) | 9.3 | < 0.001 |
OC-Sensor, ≥ 15 µg Hb/g faeces | ||||||||||||
All studies | 5 | 65.0 (62.2-67.8) | 97.6 | 83.5 (82.5-84.4) | 99.6 | 3.90 (2.04-7.47) | 99.0 | 0.43 (0.31-0.60) | 94.2 | 10.06 (8.14-12.44) | 42.4 | < 0.001 |
100% Symptomatic | 3 | 76.8 (73.5-79.9) | 95.6 | 70.3 (68.5-72.1) | 99.4 | 2.63 (1.39-4.97) | 98.8 | 0.34 (0.22-0.52) | 88.4 | 8.88 (7.23-10.91) | 0.0 | < 0.001 |
Prevalence CRC ≥ 2.5% | 4 | 69.3 (66.3-72.1) | 97.1 | 79.4 (78.2-80.6) | 99.6 | 3.33 (1.67-6.66) | 99.1 | 0.38 (0.27-0.54) | 91.1 | 9.72 (7.46-12.66) | 55.4 | < 0.001 |
OC-Sensor, ≥ 20 µg Hb/g faeces | ||||||||||||
All studies | 5 | 62.9 (60.1-65.7) | 97.8 | 85.1 (84.2-86.0) | 99.6 | 4.34 (2.16-8.73) | 99.0 | 0.45 (0.32-0.62) | 95.3 | 10.62 (8.24-13.67) | 57.4 | < 0.001 |
100% Symptomatic | 3 | 75.1 (71.7-78.3) | 96.1 | 72.4 (70.7-74.1) | 99.5 | 2.85 (1.43-5.65) | 98.8 | 0.35 (0.23-0.55) | 90.4 | 9.19 (7.47-11.32) | 1.8 | < 0.001 |
Prevalence CRC ≥ 2.5% | 4 | 67.5 (64.5-70.4) | 97.3 | 81.2 (80.0-82.3) | 99.6 | 3.66 (1.74-7.71) | 99.1 | 0.40 (0.29-0.55) | 91.7 | 10.19 (7.49-13.86) | 66.5 | < 0.001 |
Significant colonic lesion | ||||||||||||
OC-Sensor, LoD µg Hb/g faeces | ||||||||||||
All studies | 3 | 91.7 (89.5-93.6) | 0.0 | 36.9 (35.0-39.0) | 94.2 | 1.45 (1.32-1.59) | 80.4 | 0.24 (0.19-0.30) | 0.0 | 6.01 (4.57-7.92) | 0.0 | < 0.001 |
OC-Sensor, ≥ 10 µg Hb/g faeces | ||||||||||||
All studies | 4 | 78.6 (75.6-81.4) | 91.5 | 69.8 (67.9-71.6) | 99.2 | 3.75 (2.08-6.76) | 98.3 | 0.34 (0.27-0.42) | 59.6 | 11.72 (6.41-21.45) | 82.8 | < 0.001 |
100% Symptomatic4 | 3 | 80.4 (77.4-83.2) | 89.6 | 67.0 (65.0-68.9) | 99.2 | 2.54 (1.45-4.46) | 98.5 | 0.31 (0.26-0.37) | 23.7 | 8.56 (6.18-11.86) | 49.8 | < 0.001 |
Table 7 Diagnostic accuracy parameters for advanced neoplasia detection based on quantitative faecal immunochemical test for haemoglobin threshold concentration and brand
Brand | Threshold(µg/g faeces) | Author, year | Sensitivity1 | Specificity1 |
Mixed cohorts | ||||
OC-Sensor® | 5 | Ou, 2013[21] | 56.8 (39.5-72.9) | 88.7 (85.7-91.2) |
OC-Sensor® | 25 | Rozen, 2010[17] | 27.5 (20.3-34.7) | 96.7 (95.8-97.6) |
OC-Sensor® | 25 | Terhaar sive Droste, 2010[32] | 48.3 (42.6-53.9) | 94.3 (93.2-95.3) |
OC-Sensor® | 30 | Rozen, 2010[17] | 26.8 (19.9-34.7) | 97.3 (96.5-98.1) |
OC-Sensor® | 30 | Terhaar sive Droste, 2010[32] | 46.0 (40.4-51.7) | 95.1 (94.1-96.1) |
OC-Sensor® | 40 | Rozen, 2010[17] | 26.2 (19.1-33.2) | 97.8 (97.0-98.5) |
OC-Sensor® | 40 | Terhaar sive Droste, 2010[32] | 43.2 (37.6-48.9) | 95.8 (94.8-96.7) |
HM-JACKarc® | LoD | Auge, 2016[22] | 96.6 (82.8-93.4) | 10.6 (6.9-15.9) |
HM-JACKarc® | 10 | Auge, 2016[22] | 34.5 (19.9-52.7) | 87.2 (81.6-91.3) |
HM-JACKarc® | 20 | Auge, 2016[22] | 31.0 (17.3-49.2) | 92.8 (88.0-95.7) |
HM-JACKarc® | 30 | Auge, 2016[22] | 31.0 (17.3-49.2) | 93.3 (88.7-96.1) |
HM-JACKarc® | 40 | Auge, 2016[22] | 27.6 (14.7-45.7) | 93.9 (89.4-96.6) |
FOB Gold® | 10 | Auge, 2018[30] | 45.7 (33.7-58.1) | 84.7 (80.8-88.0) |
FOB Gold® | 20 | Auge, 2018[30] | 37.1 (26.1-49.6) | 87.9 (84.2-90.8) |
FOB Gold® | 30 | Auge, 2018[30] | 35.7 (24.6-48.1) | 90.3 (87.0-93.1) |
FOB Gold® | 40 | Auge, 2018[30] | 32.9 (22.4-45.2) | 91.1 (87.8-93.6) |
FOB Gold® | 50 | Auge, 2018[30] | 31.4 (20.9-43.6) | 92.3 (89.3-94.7) |
FOB Gold® | 60 | Auge, 2018[30] | 30.0 (19.9-42.3) | 92.3 (89.2-94.6) |
100% symptomatic | ||||
HM-JACK® | 20 | Parente, 2012[24] | 35.6 (27.9-44.1) | 94.5 (89.7-97.2) |
Table 8 Diagnostic accuracy parameters for significant colonic lesion detection based on quantitative faecal immunochemical test for haemoglobin threshold concentration and brand
Brand | Threshold(µg/g faeces) | Author, year | Sensitivity1 | Specificity1 |
Significant colonic lesion (100% symptomatic) | ||||
OC-Sensor® | 15 | Cubiella (DC), 2014[29] | 76.2 (72.0-80.0) | 74.4 (71.7-76.9) |
OC-Sensor® | 15 | Cubiella (VC), 2017[31] | 89.5 (84.1-93.6) | 40.6 (36.4-44.9) |
OC-Sensor® | 20 | Cubiella (DC), 2014[29] | 74.7 (70.5-78.6) | 76.1 (73.5-78.6) |
OC-Sensor® | 20 | Cubiella (VC), 2017[31] | 87.8 (82.2-92.2) | 42.1 (37.9-46.5) |
HM-JACKarc® | 10 | Godber, 2016[27] | 68.9 (53.2-81.4) | 80.2 (76.1-83.7) |
HM-JACKarc® | 15 | Godber, 2016[27] | 66.7 (50.9-79.6) | 83.1 (79.2-86.5) |
HM-JACKarc® | 20 | Godber, 2016[27] | 64.4 (48.7-77.7) | 85.7 (81.9-88.7) |
HM-JACKarc® | 25 | Godber, 2016[27] | 64.4 (48.7-77.7) | 87.5 (83.9-90.3) |
HM-JACKarc® | 30 | Godber, 2016[27] | 64.4 (48.7-77.7) | 88.6 (85.2-91.4) |
HM-JACKarc® | 35 | Godber, 2016[27] | 64.4 (48.7-77.7) | 89.2 (85.9-92.0) |
HM-JACKarc® | 40 | Godber, 2016[27] | 64.4 (48.7-77.7) | 90.0 (86.7-92.5) |
- Citation: Pin Vieito N, Zarraquiños S, Cubiella J. High-risk symptoms and quantitative faecal immunochemical test accuracy: Systematic review and meta-analysis. World J Gastroenterol 2019; 25(19): 2383-2401
- URL: https://www.wjgnet.com/1007-9327/full/v25/i19/2383.htm
- DOI: https://dx.doi.org/10.3748/wjg.v25.i19.2383