Human sperm acrosome function assays are predictive of fertilization rate in vitro: a retrospective cohort study and meta-analysis

Table 5 Summary of Rs values for AE assay

Subgroup	N	R_s (95% CI)	Test(s) of heterogeneity		Significance test(s) of R_s = 0		P ^a
Subgroup	N	R_s (95% CI)	I² (%)	P	Z	P
Overall	13	0.39 (0.18–0.60)	95.7	0.000	3.70	0.000
Two experiments excluded	11	0.38 (0.27–0.48)	67.1	0.001	7.19	0.000
Geographic area
Asia	4	0.24 (0.05–0.42)	66.7	0.029	2.49	0.013	< 0.001
North America	1	0.54 (0.43–0.65)	–^b	–	9.70	0.000
Europe	6	0.46 (0.37–0.54)	0.0	0.798	11.02	0.000
Preparation method
No preparation	6	0.42 (0.30–0.55)	63.4	0.018	6.72	0.000	0.04
Swim up	3	0.27 (0.01–0.53)	80.6	0.006	2.01	0.044
Discontinuous gradient	2	0.42 (0.23–0.60)	0.0	0.893	4.41	0.000
AE assay method
Fluorometry
MAb 4D4-proacrosin	1	0.49 (0.27–0.70)	–	–	4.43	0.000	< 0.001
Spectrophotometry
Spectrophotometry/BAPNA	5	0.44 (0.35–0.54)	0.0	0.618	9.08	0.000
Accu-Sperm spectrophotometry/BAPNA	2	0.09 (−0.05–0.23)	0.0	0.352	1.22	0.223
Substrate assay
Acrosin target with gelatine substrate assay	2	0.43 (0.30–0.56)	0.0	0.997	6.51	0.000
Hyaluronidase target with agar/hyaluronic acid mixture substrate assay	1	0.54 (0.43–0.65)	–	–	9.70	0.000

Rs spearman correlation coefficient, CI confidence interval, BAPNA N-α-benzoyl-dl-arginine-p-nitroanilide
^aComparison between subgroups with the Q test for heterogeneity
^bNot calculated