Table 5.
Diagnostic Accuracy of the Diagnostic Tests of the Carpal, Metacarpal and Phalangeal Fractures (N=35)
| Author(s) | Index test | Reference test | Fracture | Se % (95% CI) | Sp % (95% CI) | Accuracy % (95% CI) | PPV % (95% CI) | NPV % (95% CI) |
|---|---|---|---|---|---|---|---|---|
| Scaphoid and other carpal bones fractures | ||||||||
| History taking | ||||||||
| Sharifi (2015) [74] | VAS pain score cutt of: 3,0 | MRI | Scaphoid | 100 | 100 | |||
| 4,5 | MRI | Scaphoid | 94 | 92 | ||||
| 5,5 | MRI | Scaphoid | 94 | 82 | ||||
| 6,5 | MRI | Scaphoid | 94 | 72 | ||||
| 7,5 | MRI | Scaphoid | 88 | 43 | ||||
| 8,5 | MRI | Scaphoid | 75 | 28 | ||||
| 9,5 | MRI | Scaphoid | 31 | 13 | ||||
| Physical examination | ||||||||
| Bergh (2014) [44] | Clinical Scaphoid Score ≥4 | MRI 1,5T | Scaphoid | 77 | 56 | 58 | 14 | 96 |
| Gabler (2001) [45] | Repeated clinical and radiological examinations (after 10 days) | MRI 1,0T | Scaphoid | 82 | ||||
| Repeated clinical and radiological examinations (after 38 days) | MRI 1,0T | Scaphoid | 100 | 100 | 100 | 100 | 100 | |
| Repeated clinical and radiological examinations (after 38 days) | MRI 1,0T | Capitate | 100 | |||||
| Repeated clinical and radiological examinations (after 38 days) | MRI 1,0T | Triquetrum | 75 | |||||
| Repeated clinical and radiological examinations (after 38 days) | MRI 1,0T | Hamate | 100 | |||||
| Repeated clinical and radiological examinations (after 38 days) | MRI 1,0T | Lunate | 100 | |||||
| Repeated clinical and radiological examinations (after 38 days) | MRI 1,0T | Trapezoid | 100 | |||||
| Herneth (2001) [47] | Clinical examination | MRI | Scaphoid | 89 | 50 | 73 | 73 | 75 |
| Rhemrev (2010) [63] | Pronation strength ≤10% | Clinical follow-up | Scaphoid | 69 | 65 | |||
| Extension < 50% | Clinical follow-up | Scaphoid | 85 | 59 | ||||
| Supination strength ≤10% | Clinical follow-up | Scaphoid | 85 | 77 | ||||
| Grip strength ≤25% | Clinical follow-up | Scaphoid | 92 | 34 | ||||
| extension <50%, supination strength <10% and presence of a previous fracture of either the involved or uninvolved hand or wrist. | Clinical follow-up | Scaphoid | 15 | 98 | 61 | 85 | ||
| extension <50%, supination strength <10% and presence of a previous fracture of either the involved or uninvolved hand or wrist. | Clinical follow-up | No scaphoid fracture | 46 | 92 | 54 | 89 | ||
| Steenvoorde (2006) [64] | Seven clinical tests (≥ 5 positive tests) | Clinical follow-up | Scaphoid | 100 | 13 | 55 | 52 | 100 |
| Imaging: Radiographs | ||||||||
| Annamalai (2003) [44] | Scaphoid fat stripe on radiography | MRI 0,2T (12-72h) | Scaphoid | 50 | 50 | 50 | 50 | 50 |
| Pronator fat stripe on radiography | Scaphoid | 26 | 70 | 48 | 46 | 49 | ||
| Balci (2015) [71] | Radiographs | MDCT | Scaphoid | 66 | 98 | 77 | 96 | |
| Radiographs | MDCT | Lunate | 20 | 100 | 100 | 97 | ||
| Radiographs | MDCT | Triquetrum | 29 | 100 | 100 | 96 | ||
| Radiographs | MDCT | Pisiform | 0 | 100 | 0 | 99 | ||
| Radiographs | MDCT | Trapezium | 18 | 99 | 33 | 98 | ||
| Radiographs | MDCT | Trapezoid | 0 | 100 | 0 | 99 | ||
| Radiographs | MDCT | Capitate | 8 | 100 | 50 | 98 | ||
| Radiographs | MDCT | Hamata | 41 | 100 | 78 | 98 | ||
| Behzadi (2015) [45] | Radiographs (anterior-posterior, lateral and oblique projections) | MDCT (within 10 days) | Scaphoid | 43 | 81 | 60 | 53 | 73 |
| Herneth (2001) [53] | Radiographs | MRI | Scaphoid | 56 | 100 | 73 | 100 | 60 |
| Jorgsholm (2013) [72] | Radiographs | MRI 0.23T (within 3 days) | Scaphoid | 70 (61-78) | 98 (95-100) | 87 | 97 | 82 |
| Radiographs 6-week: DICOM viewer | MRI 0.23T (within 3 days) | Triquetrum | 59 (33-82) | |||||
| Radiographs 6-week: DICOM viewer | MRI 0.23T (within 3 days) | Lunate | 25 (1-81) | |||||
| Radiographs 6-week: DICOM viewer | MRI 0.23T (within 3 days) | Capitate | 7 (0-34) | |||||
| Radiographs 6-week: DICOM viewer | MRI 0.23T (within 3 days) | Hamata | 0 (0-46) | |||||
| Mallee (2016) [57] | Radiographs 6-week: JPEG | MRI | Scaphoid | 42 (37-47) | 56 (54-59) | 53 (51-56) | 20 (17-23) | 79 (76-81) |
| Radiographs 6-week: JPEG | MRI | Scaphoid | 64 (57-71) | 53 (50-57) | 56 (52-59) | 26 (22-30) | 85 (82-88) | |
| Mallee (2016) [57] | Radiographs 6-week: JPEG | CT | Scaphoid | 56 (50-62) | 59 (56-61) | 58 (56-61) | 19 (16-22) | 89 (87-90) |
| Mallee (2016) [57] | Radiographs 6-week: DICOM viewer | CT | Scaphoid | 79 (72-85) | 55 (51-58) | 58 (55-61) | 23 (19-27) | 94 (91-96) |
| Mallee (2016) [57] | Radiographs 6-week: JPEG | MRI + CT | Scaphoid | 52 (45-59) | 58 (55-60) | 57 (55-59) | 14 (12-17) | 90 (88-92) |
| Mallee (2016) [57] | Radiographs 6-week: DICOM viewer | MRI + CT | Scaphoid | 75 (67-83) | 53 (50-56) | 56 (52-59) | 18 (14-21) | 94 (92-96) |
| Ottenin 2012 [60] | Radiographs | Clinical follow-up | Scaphoid | 67ɸ | 93ɸ | 88ɸ | 68ɸ | 92ɸ |
| Ottenin 2012 [60] | Radiographs | Clinical follow-up | Other carpal bones | 40ɸ | 94ɸ | 88ɸ | 44ɸ | 93ɸ |
| Brink (2019) [68] | X-ray | 1-year clinical follow-up | Scaphoid | 25 | 97 | |||
| X-ray | 1-year clinical follow-up | Triquetral | 18 | 100 | ||||
| X-ray | 1-year clinical follow-up | Lunate | 0 | 100 | ||||
| X-ray | 1-year clinical follow-up | Trapezium | 0 | 100 | ||||
| X-ray | 1-year clinical follow-up | Trapezoid | 0 | 100 | ||||
| X-ray | 1-year clinical follow-up | Hamate | 100 | 100 | ||||
| X-ray | 1-year clinical follow-up | Capitate | 100 | 100 | ||||
| Neubauer (2018) [69] | Radiography | Clinical follow-up | Scaphoid | 87 (83-92) | 77 (71-83) | 82 | 80 (75-86) | 84 (80-90) |
| Imaging: MRI | ||||||||
| Beeres (2008) [47] | MRI 1,5T (<24h) | A combination of MRI, bone scintigraphy and when not in agreement, clinical follow-up | Scaphoid | 80 (56-94) | 100 (96-100) | 96 | 100 (74-100) | 95 (88-99) |
| Kumar (2005) [55] | MRI 1,5T (<24h) | MRI in those without fracture at MRI <24h or no clinical signs of fracture | Scaphoid | 100b | 100b | 100b | 100b | 100b |
| Mallee (2011) [56] | MRI 1.0T | Radiographs | Scaphoid | 67 | 89 | 85 |
57 54c |
93 93d |
| Memarsadeghi (2006) [59] | MRI 1,0T | Radiographs obtained 6 weeks after trauma. | All scaphoid | 100 (82-100) | 100 (87-100) | 100 | 100 | 100 |
| Memarsadeghi (2006) [59] | MRI 1,0T | Radiographs obtained 6 weeks after trauma. | Cortical scaphoid fractures | 38 (16-65) | 100 (52-100) | 55 (24-85) | 100 | 27 |
| Memarsadeghi (2006) [59] | MRI 1,0T | Radiographs obtained 6 weeks after trauma. | Other carpal fractures | 85 | 100 | 84 | ||
| de Zwart (2016) [66] | MRI (<72h) | Final diagnosis after MRI, CT, BS and 6-weeks clinical signs | Scaphoid | 67 | 100 (88-100) | 94 | 67 | 97 |
| Imaging: (Multi detector) computed tomography | ||||||||
| Adey (2007) [43] | CT (first round interpretation) | Radiographs 6 weeks after injury | Scaphoid | 89 (84-92) | 91 (86-94) | 89 (89-92) | 28 (23-32) | 99 (97-99) |
| CT (second round interpretation) | Radiographs 6 weeks after injury | Scaphoid | 97 (93-99) | 85 (77-89) | 88 (82-91) | |||
| Breederveld (2004) [49] | CT | Clinical follow-up | Scaphoid | 100 | 100 | 100 | 100 | 100 |
| Cruickshank (2007) [50] | CT (same or next day) | The diagnosis on Day 10 with clinical examination and X-rays, with MRI performed in patients with persistent tenderness but normal X-rays. | Scaphoid and other fractures (Triquetral, Trapezium, Capitate and Lunate) | 94 (72-100) | 100 (87-100) | 98 | 100 (78-100) | 97 (82-100) |
| Ilica (2011) [54] | MDCT | MRI 1,5T | Scaphoid | 86 | 100 | 95 | 100 | 91 |
| Jorgsholm (2013) [72] | CT | MRI 0.23T (within 3 days) | Scaphoid | 95 (91-97) | ||||
| CT | MRI 0.23T (within 3 days) | Capitate | 75 (35-97) | |||||
| CT | MRI 0.23T (within 3 days) | Hamata | 100 (40-100) | |||||
| Mallee (2011) [56] | CT | Radiographs | Scaphoid | 67 | 96 | 91 |
80 76c |
93 94d |
| Mallee (2014) [58] | CT-scaphoid: reformations in planes defined by the long axis of the scaphoid | Radiographs | Scaphoid | 67 | 96 | 91 |
80 76c |
93 94d |
| CT-wrist: reformations made in the anatomic planes of the wrist | Radiographs | Scaphoid | 33 | 89 | 79 |
40 36c |
86 87d |
|
| Memarsadeghi (2006) [59] | MDCT | Radiographs obtained 6 weeks after trauma. | All scaphoid | 73 (48-89) | 100 (87-100) | 89 (78-100) | 100 | 86 |
| Memarsadeghi (2006) [59] | MDCT | Radiographs obtained 6 weeks after trauma. | Cortical scaphoid fractures | 100 (75-100) | 100 (52-100) | 100 | 100 | 100 |
| Ottenin (2012) [60] | MDCT | Clinical follow-up | Scaphoid | 77ɸ | 94ɸ | 91ɸ | 76ɸ | 95ɸ |
| Ottenin (2012) [60] | MDCT | Clinical follow-up | Other carpal bones | 60ɸ | 95ɸ | 91ɸ | 56ɸ | 96ɸ |
| Rhemrev (2007) [63] | MDCT (<24h) | Final diagnosis after CT, BS and, both radiographic (6 weeks after injury) and physical reevaluation. | Scaphoid | 64 | 99 | 94 | 90 | 94 |
| de Zwart (2016) [66] | CT(<72h) | Final diagnosis after MRI, CT, BS and 6-weeks clinical signs | Scaphoid | 33 | 100 (88-100) | 94 | 100 | 94 |
| Brink (2019) [68] | CT | 1-year clinical follow-up | Scaphoid | 100 | 100 | |||
| CT | 1-year clinical follow-up | Triquetral | 100 | 100 | ||||
| CT | 1-year clinical follow-up | Lunate | 100 | 100 | ||||
| CT | 1-year clinical follow-up | Trapezium | 100 | 100 | ||||
| CT | 1-year clinical follow-up | Trapezoid | 100 | 100 | ||||
| CT | 1-year clinical follow-up | Hamate | 100 | 100 | ||||
| CT | 1-year clinical follow-up | Capitate | 100 | 0 | ||||
| Neubauer (2018) [69] | CBCT | Clinical follow-up | Scaphoid | 93 (89-96) | 96 (93-99) | 94 | 96 (93-99) | 92 (89-96) |
| Borel (2017) [70] | CBCT | MRI | Scaphoid cortical fracture | 100 (75-100) | 97 (83-100) | 94 (68-100) | 100 (87-100) | |
| CBCT | MRI | All scaphoid fractures | 94 (68-100) | 97 (83-100) | 94 (68-100) | 97 (82-100) | ||
| CBCT | MRI | Wrist cortical fracture | 100 (83-100) | 95 (75-100) | 96 (78-100) | 100 (83-100) | ||
| CBCT | MRI | All wrist fractures | 89 (70-97) | 95 (75-100) | 96 (78-100) | 88 (67-97) | ||
| Imaging: Bone scintigraphy | ||||||||
| Beeres (2007) [46] | Bone scintigraphy (3-7 days after injury) | Clinical outcome | Scaphoid | 92 | 87 | 88a | 69a | 97 |
| Bone scintigraphy (3-7 days after injury) | Clinical outcome | Scaphoid and other carpal bones | 96 | 59a | 80a | 75 | 93a | |
| Beeres (2008) [47] | Bone scintigraphy (between 3 and 5 days) | A combination of MRI, bone scintigraphy and when not in agreement, clinical follow-up | Scaphoid | 100 (83-100) | 90 (81-96) | 92 | 71 (52-87) | 100 (95-100) |
| Breederveld (2004) [49] | Bone scintigraphy (three-fase) | Clinical follow-up | Scaphoid | 78 | 90 | 86 | 78 | 90 |
| Rhemrev (2010) [62] | Bone scintigraphy (3-5 days) | Final diagnosis after CT, BS and, both radiographic (6 weeks after injury) and physical reevaluation. | Scaphoid | 93 | 91 | 91 | 62 | 99 |
| de Zwart (2016) [66] | Bone Scintigraphy (between 3 and5 days) | l diagnosis after MRI, CT, BS and 6-weeks clinical signs | Scaphoid | 100 | 97 (83-100) | 97 | 75 | 100 |
| Imaging: Ultrasonography | ||||||||
| Fusetti (2005) [51] | HSR-S global evaluation | CT (immediately after HSR-S performed) | Scaphoid | 100 | 79 | 83 | 56 | 100 |
| HSR-S scaphoid cortical disruption | CT (immediately after HSR-S performed) | Scaphoid | 100 | 95 | 96 | 83 | 100 | |
| HSR-S radioarpal (RS) effusion | CT (immediately after HSR-S performed) | Scaphoid | 100 | 42 | 54 | 31 | 100 | |
| HSR-S scapho-trapezium-trapezoid (STT) effusion | CT (immediately after HSR-S performed) | Scaphoid | 100 | 84 | 88 | 62 | 100 | |
| HSR-S cortical disruption with RS and STT effusion (high index of suspicion) | CT (immediately after HSR-S performed) | Scaphoid | 100 | 100 | 100 | 100 | 100 | |
| Herneth (2001) [53] | US | MRI | Scaphoid | 78 | 100 | 87 | 100 | 75 |
| Javadzadeh (2014) [74] | BUS | Radiographs | Carpal bones | 42 (23-64) | 87 (74-94) | 74 (62-83) | 57 (33-79) | 78 (65-88) |
| Javadzadeh (2014) [74] | WBT ultrasonography | Radiographs | Carpal bones | 47 (27-68) | 87 (74-94) | 75 (64-84) | 60 (36-80) | 80 (67-89) |
| Platon (2011) [61] | US | CT | Scaphoid | 92 | 71 | 76 | 46 | 97 |
| US | CT | Scaphoid fracture with a high potential of complication | 100 | 67 | 71 | 30 | 100 | |
| Yildirim (2013) [65] | BUS | MRI (<24h) | Scaphoid | 100 (69-100) | 34 (19-52) | 49 | 30 (16-49) | 100 (74-100) |
| Imaging: Tomosynthesis | ||||||||
| Ottenin (2012) [60] | Tomosynthesis | Clinical follow-up | Scaphoid | 91ɸ | 98ɸ | 96ɸ | 90ɸ | 98ɸ |
| Ottenin (2012) [60] | Tomosynthesis | Clinical follow-up | Other carpal bones | 80ɸ | 98ɸ | 96ɸ | 83ɸ | 98ɸ |
| Scaphoid, other carpal bones and/or metacarpal fractures | ||||||||
| Physical examination | ||||||||
| Nikken (2005) [73] | Anatomic snuffbox tenderness | Additional treatment need | Scaphoid and other carpal bones. Metacarpal bones II–IV | 39 | 78 | 62 | 56 | 65 |
| Imaging: Radiographs | ||||||||
| Balci (2015) [71] | Radiographs | MDCT | Metacarpal | 67 | 99 | 82 | 98 | |
| Jorgsholm (2013) [72] | Radiographs | MRI 0.23T (within 3 days) | Metacarpal | 30 (7-65) | ||||
| Nikken (2005) [73] | Radiographs | Additional treatment need | Scaphoid and other carpal bones. Metacarpal bones II–IV | 72 | 92 | 84 | 87 | 82 |
| Brink (2019) [68] | X-ray | 1-year clinical follow-up | Metacarpal | 67 | 100 | |||
| Imaging: MRI | ||||||||
| Nikken (2005) [73] | MRI | Additional treatment need | Scaphoid and other carpal bones. Metacarpal bones II–IV | 67 | 76 | 73 | 63 | 79 |
| Imaging: CT | ||||||||
| Brink (2019) [68] | CT | 1-year clinical follow-up | Metacarpal | 100 | 100 | |||
| Metacarpal bones and finger fractures | ||||||||
| Physical examination | ||||||||
| Tayal (2007) [77] | Physical examination: deformity | Radiographs and surgical findings | Metacarpal bones and phalanx | 55 (44-66) | 89 (83-96) | 76 | 77 (68-87) | 75 (65-85) |
| Physical examination: swelling | Radiographs and surgical findings | Metacarpal bones and phalanx | 94 (88-99) | 13 (5-20) | 45 | 41 (30-52) | 75 (65-85) | |
| Physical examination: erythema | Radiographs and surgical findings | Metacarpal bones and phalanx | 26 (16-36) | 85 (77-93) | 62 | 53 (42-54) | 63 (53-74) | |
| Imaging: Ultrasonography | ||||||||
| Tayal (2007) [77] | US | Radiographs and surgical findings | Metacarpal bones and phalanx | 90 (74-97) | 98 (95-100) | 95 | 97 (93-100) | 94 (89-99) |
| Javadzadeh (2014) [74] | BUS | Radiographs | Metacarpal bones | 73 (43-90) | 78 (45-94) | 70 (48-85) | 80 (49-94) | 70 (40-89) |
| BUS | Radiographs | Phalanx | 83 (61-94) | 90 (78-96) | 88 (78-94) | 79 (57-91) | 93 (81-97) | |
| WBT ultrasonography | Radiographs | Metacarpal bones | 82 (52-95) | 89 (57-98) | 70 (48-85) | 90 (60-98) | 80 (49-94) | |
| WBT ultrasonography | Radiographs | Phalanx | 94 (74-99) | 95 (84-99) | 95 (86-98) | 89 (87-100) | 98 (87-100) | |
| Kocaoglu (2016) [76] | US | Radiographs | Metacarpal bones | 93 (79-98) | 98 (90-100) | 96 | 97 (85-100) | 95 (85-98) |
| Imaging: CBCT | ||||||||
| Faccioli (2010) [75] | CBCT | MSCT | Articular involvement of the phalanx | 100 | 100 | 100 | 100 | 100 |
| CBCT | MSCT | Phalangeal bone fragments | 87 | 100 | 92 | 100 | 82 | |
BUS Bedside Ultra Sonography, CBCT Cone Beam Computed tomography arthrography, MDCT Multidetector Computed tomography, MRI Magnetic resonance imaging, T Tesla, US Ultra Sonography, HSR-S High Spatial Resolution sonography, VAS Visual Analogue Scale, Se Sensitivity, Sp Specificity, PPV Positive predictive value, NPV Negative predictive value, LR Likelihood ratio
aOne patient had a physical examination matching with another carpal fracture instead of a scaphoid fracture at both 2 and 6 weeks after injury
bFour patient did not receive MRI during follow-up (reference standard)
cPositive predictive value accounting for prevalence and incidence
dNegative predictive value accounting for prevalence and incidence
c/dThe positive predictive value and negative predictive value were determined with use of the Bayes theorem, which requires an a priori estimate of the prevalence (pretest probability) of the presence of scaphoid fractures. The positive predictive value is the patient’s probability of having a scaphoid fracture when the test is positive, and the negative predictive value is the probability of a patient not having a scaphoid fracture when the test is negative. The predictive values of any imaging modality depend critically on the prevalence of the characteristic in the patients being tested; hence the use of the appropriate Bayesian analysis is important. For the determination of positive and negative predictive values, we estimated an average prevalence of scaphoid fractures of 16% on the basis of the best available data. The positive predictive value was calculated as sensitivity · prevalence/(sensitivity · prevalence) 1 [(1 – specificity) · (1 – prevalence)], and the negative predictive value was calculated as specificity · (1 – prevalence)/[(1 – sensitivity) · prevalence] 1 [specificity · (1 – prevalence)].54,60
ɸ Average between presented individual values of three readers (junior radiologist, junior orthopedic surgeon and senior radiologist)