Value of intravoxel incoherent motion in detecting and staging liver fibrosis: A meta-analysis

Table 1 Baseline characteristics of the included studies and subjects

Ref.	Year	Country	Study design	Sample size	Sex ratio, M/F	mean ± SD age in yr	Patient spectrum (n)
Yoon et al[30]	2019	South Korea	Retro	106	81/25	55.4 ± 11.6	HBV (82), HCV (9), non-B non-C cirrhosis (3), ALD (10), PBC (2)
Fu et al[25]	2019	China	Pros	125	71/54	37.6 ± 9.3 (patients), 31.5 ± 12.9 (volunteers)	HBV (81), HCV (8), NAFLD (3), DIH (2), AIH (6)
Zawada et al[34]	2019	Poland	Pros	40	10/30	22-75	HCV (40)
Sandrasegaran et al[32]	2018	United States	Retro	49	35/14	56.6 (range: 32–73)	HBV (3), HCV (35), ALD (24), AIH (4), Other (7)1
Seo et al[31]	2018	South Korea	Retro	95	70/25	59.5 ± 9.5	HBV (44), HCV (7), ALD (14), no underlying liver disease (30)
Hu et al[29]	2017	China	Retro	56	14/42	47.48 (range: 15–76)	HBV (5), NAFLD (14), ASH (1), PSC (8), AIH (10), overlap syndrome of AIH (5), and DIH (13)
Chung et al[35]	2015	South Korea	Retro	57	35/22	58.7 (range: 32–89)	HBV (34), HCV (1), ALD (2), Other (20)
Wu et al[27]	2015	China (Taiwan)	Pros	49	36/13	62.4 (range: 38–85)	HBV (17), HVC (10), non-B non-C carriers (17), both hepatitis B and C carriers (5)
Ichikawa et al[26]	2015	Japan	Retro	182	127/55	66.4 ± 11.6	HBV (18), HCV (62), ASH (12), NASH (3), AIH (1), PBC (2), unidentified liver disease with an elevated liver enzyme level (12)
Parente et al[28]	2015	Brazil	Pros	59	10/49	54 ± 9	Type 2 diabetic subjects (59)
Chen et al[24]	2014	China	Pros	50	37/13	43.7 ± 1.2 (patients), 38.9 ± 1.3 (volunteers)	HBV (15), HCV (1), ALD (1), unidentified liver disease with an elevated liver enzyme level (8)
Yoon et al[33]	2014	South Korea	Retro	55	42/13	53.9 (range: 18–78)	HBV (45), HCV (1), ALD (1), Other (8)

¹Some patients had more than one etiology of liver disease. M/F: Male/female; SD: Standard deviation; Retro: Retrospective; Pros: Prospective; HBV: Hepatitis B virus; HCV: Hepatitis C virus; ALD: Alcoholic liver disease; PBC: Primary biliary cirrhosis; PSC: Primary sclerotic cholangitis; NAFLD: Nonalcoholic fatty liver disease; NASH: Nonalcoholic steatohepatitis; DIH: Drug-induced hepatitis; AIH: Autoimmune hepatitis; ASH: Alcoholic steatohepatitis.

Table 2 Methodological and imaging protocol characteristics of included studies

Ref.	Reference standard	Scoring system	F0/F1/F2/F3/F4	Mean intervals	Double blindness	Scanner	FS	Trigger methods	b values	Scan time1
Yoon et al[30], 2019	S/B	METAVIR	13/6/19/18/50	6 (4-38) d	Yes	GE	1.5 T	FB	0, 15, 25.4, 42.9, 72.5, 122.5, 207, 350, 592, 1000	4.5
Fu et al[25], 2019	S/B	METAVIR	26/17/33/25/24	< 1 mo	Yes	GE	3.0 T	RT	0, 50, 100, 150, 200, 300, 500, 800, 1000, 1300, 1500, 1700, 2000	10
Zawada et al[34], 2019	B	Scheuer scale	7/17/10/2/2	NA	Unclear	GE	1.5 T	RT	0, 20, 50, 100, 200, 400, 600, 800, 1000	2.6
Sandrasegaran et al[32], 2018	B	METAVIR	4/8/2/9/26	< 3 mo	Unclear	Siemens	1.5 T	RT	0, 50, 100, 300, 600, 800	12-15
Seo et al[31], 2018	S/B/T	Batts-Ludwig	30/14/4/18/29	81.2 d	Yes	Philips	3.0 T	FB	0, 10, 25, 50, 75, 100, 200, 500, 800	4.25
Hu et al[29], 2017	B	METAVIR	6/19/13/8/10	6 d	Unclear	Siemens	3.0 T	NA	0, 25, 50, 75, 100, 150, 200, 500, 800	5
Chung et al[35], 2015	B/I	METAVIR	21/1/6/7/22	15.9 (2–43) d	Unclear	Siemens	1.5 T	RT	0, 30, 60, 100, 150, 200, 400, 600, 900	7
Wu et al[27], 2015	S	METAVIR	6/16/10/10/7	< 7 d	Unclear	Siemens	3.0 T	RT	0, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, 300, 400, 500, 1000	< 7
Ichikawa et al[26], 2015	S/B	METAVIR	72/13/14/19/64	< 3 mo	Unclear	GE	3.0 T	RT	0, 10, 20, 30, 40, 50, 80, 100, 200, 500, 1000	2-3
Parente et al[28], 2015	B	METAVIR	43/9/5/1/1	< 3 mo	Unclear	Philips	3.0 T	RT	0, 10, 20, 40, 80, 160, 200, 400, 800, 1000	3
Chen et al[24], 2014	B	METAVIR	25/4/9/11/1	< 1 mo	Unclear	GE	3.0 T	RT	0, 50, 100, 200, 400, 600, 800	NA
Yoon et al[33], 2014	S/B/T	METAVIR	11/7/7/9/21	NA	Unclear	Siemens	3.0 T	FB	0, 25, 50, 75, 100, 200, 500, 800	6.5

¹The unit of scan time is min. NA: Not available; S: Liver surgery; B: Percutaneous liver biopsy; T: Liver transplantation; I: Imaging findings; FS: Field strength; FB: Free breathing; RT: Respiratory triggered.

Table 3 Diagnostic raw data of intravoxel incoherent motion in each liver fibrosis group

Study	Index	Threshold1	TP	FP	FN	TN	SEN	SPE
LF ≥ F1
Chen et al[24], 2014	D*	17.52	20	7	5	18	0.80	0.72
Fu et al[25], 2019	Dt	0.63	75	3	24	23	0.76	0.88
Ichikawa et al[26], 2015	D*	72.15	96	10	14	62	0.87	0.86
Parente et al[28], 2015	D*	37.75	10	13	6	30	0.63	0.70
Wu et al[27], 2015	D*	51.30	27	0	16	6	0.63	1.00
LF ≥ F2
Fu et al[25], 2019	Dt	0.62	64	7	18	36	0.78	0.84
Hu et al[29], 2017	D*	32.10	30	8	1	17	0.97	0.68
Ichikawa et al[26], 2015	D*	71.72	91	14	6	71	0.94	0.84
Sandrasegaran et al[32], 2018	D*	23.40	31	4	6	8	0.84	0.67
Seo et al[31], 2018	D*	77.64	44	17	7	27	0.86	0.61
Wu et al[27], 2015	D*	40.90	16	3	11	19	0.59	0.86
Yoon et al[33], 2014	D*	43.33	31	0	6	18	0.84	1.00
Yoon et al[30], 2019	D*	70.70	67	2	20	17	0.77	0.89
Zawada et al[34], 2019	Dt	1.00	7	5	7	21	0.50	0.81
LF ≥ F3
Chung et al[35], 2015	f	0.28	20	5	9	23	0.69	0.82
Fu et al[25], 2019	Dt	0.58	40	16	9	60	0.82	0.79
Ichikawa et al[26], 2015	D*	65.04	78	14	5	85	0.94	0.86
Wu et al[27], 2015	D*	40.90	13	6	4	26	0.76	0.81
LF = F4
Fu et al[25], 2019	Dt	0.58	23	31	1	70	0.96	0.69
Hu et al[29], 2017	D*	14.44	8	3	2	43	0.80	0.93
Ichikawa et al[26], 2015	D*	61.97	58	24	6	94	0.91	0.80
Seo et al[31], 2018	D*	54.42	22	17	7	49	0.76	0.74
Wu et al[27], 2015	D*	40.30	7	15	0	27	1.00	0.64
Yoon et al[33], 2014	D*	44.17	21	12	0	22	1.00	0.65

¹The unit of Dt and D* is 10⁻³ mm²/s. LF: Liver fibrosis; TP: True positive; FP: False positive; FN: False negative; TN: True negative; SEN: Sensitivity; SPE: Specificity; Dt: True diffusion coefficient; D*: Pseudo-diffusion coefficient; f: Perfusion fraction.

Table 4 Summary diagnostic performance and subgroup analysis

Characteristics	No.	Sensitivity	Specificity	Positive LR	Negative LR	DOR	AUC
LF ≥ F1	5	0.78 (0.73-0.82)	0.81 (0.74-0.86)	3.93 (2.12-7.27)	0.30 (0.19-0.46)	15.37 (5.74-41.16)	0.862 (0.811-0.914)
LF ≥ F2	9	0.82 (0.79-0.86)	0.80 (0.75-0.84)	3.69 (2.53-5.37)	0.24 (0.16-0.37)	19.35 (9.51-39.34)	0.883 (0.856-0.909)
LF ≥ F3	4	0.85 (0.79-0.90)	0.83 (0.77-0.87)	4.70 (3.51-6.28)	0.21 (0.10-0.44)	22.61 (8.24-62.05)	0.886 (0.865-0.907)
LF = F4	6	0.90 (0.84-0.94)	0.75 (0.70-0.79)	3.36 (2.57-4.38)	0.16 (0.08-0.31)	26.99 (12.88-56.58)	0.899 (0.866-0.932)
Subgroup analysis in LF ≥ F2
Study design
Retrospective	6	0.86 (0.82-0.90)	0.78 (0.71-0.83)	3.73 (2.19-6.34)	0.17 (0.10-0.29)	29.82 (11.65-76.32)	0.921 (0.891-0.952)
Prospective	3	0.71 (0.62-0.79)	0.84 (0.74-0.90)	3.97 (2.42-6.52)	0.42 (0.25-0.69)	10.20 (4.30-24.16)	0.905 (0.862-0.948)
Double blindness
Yes	3	0.80 (0.74-0.85)	0.75 (0.66-0.83)	3.75 (1.69-8.32)	0.25 (0.19-0.33)	15.46 (8.20-29.14)	0.874 (0.848-0.901)
Unclear	6	0.85 (0.80-0.89)	0.82 (0.76-0.87)	3.80 (2.47-5.84)	0.22 (0.10-0.49)	22.39 (6.91-72.51)	0.892 (0.855-0.930)
Field strength
1.5 T	3	0.76 (0.68-0.83)	0.81 (0.68-0.90)	3.16 (1.67-5.99)	0.34 (0.18-0.64)	10.40 (3.51-30.87)	0.839 (0.758-0.919)
3.0 T	6	0.85 (0.81-0.89)	0.79 (0.74-0.84)	3.92 (2.42-6.37)	0.19 (0.10-0.36)	26.17 (10.81-63.36)	0.904 (0.872-0.935)
Low b-values (0-50 s/mm2)
< 2	5	0.81 (0.75-0.86)	0.81 (0.73-0.87)	3.40 (2.17-5.35)	0.26 (0.14-0.47)	17.12 (6.11-47.98)	0.877 (0.834-0.921)
≥ 2	4	0.83 (0.78-0.88)	0.79 (0.72-0.85)	4.11 (2.07-8.17)	0.22 (0.10-0.47)	21.73 (7.22-65.42)	0.890 (0.855-0.935)
Trigger method
RT	5	0.82 (0.77-0.86)	0.79 (0.73-0.84)	3.66 (2.24-5.97)	0.27 (0.14-0.54)	15.05 (5.78-39.22)	0.867 (0.828-0.905)
Non-RT	4	0.83 (0.77-0.88)	0.81 (0.70-0.89)	4.02 (1.85-8.76)	0.21 (0.13-0.34)	29.43 (9.89-87.59)	0.919 (0.883-0.956)

No.: Number of studies; LR: Likelihood ratio; DOR: Diagnostic odds ratio; AUC: Area under summary receiver operating characteristic curves; RT: Respiratory triggered.