Table 5.

Association between serum sex hormone levels and the proportion of normal sperm morphology^a

Sperm parameters	n	Normal sperm morphology (%)
		Least-squares mean (95% CI)	P	P for trend
Luteinizing hormone
Tertile 1	112	6.8(6.5, 7.5)	Ref	0.01
Tertile 2	114	6.9(6.4, 7.3)	0.74
Tertile 3	112	6.0(5.5, 6.5)	0.01
Luteinizing hormoneb
Tertile 1	111	6.9(6.4, 7.4)	Ref	0.04
Tertile 2	111	6.8(6.4, 7.3)	0.87
Tertile 3	109	6.1(5.6, 6.6)	0.04
Follicle-stimulating hormone
Tertile 1	112	7.0(6.5, 7.5)	Ref	0.01
Tertile 2	114	6.8(6.3, 7.3)	0.46
Tertile 3	112	6.0(5.5, 6.5)	0.01
Follicle-stimulating hormoneb
Tertile 1	109	7.0(6.5, 7.5)	Ref	0.02
Tertile 2	114	6.8(6.3, 7.2)	0.54
Tertile 3	108	6.1(5.6, 6.6)	0.02
Total testosterone
Tertile 1	96	7.1(6.5, 7.6)	Ref	0.06
Tertile 2	97	6.8(6.2, 7.3)	0.41
Tertile 3	97	6.3(5.7, 6.9)	0.06
Free testosterone
Tertile 1	96	6.7(6.1, 7.2)	Ref	0.94
Tertile 2	97	6.7(6.2, 7.2)	0.99
Tertile 3	97	6.7(6.2, 7.3)	0.94
Total estradiol
Tertile 1	113	6.5(6.0, 7.0)	Ref	0.91
Tertile 2	115	6.8(6.3, 7.3)	0.50
Tertile 3	110	6.6(6.1, 7.1)	0.92
Free estradiol
Tertile 1	96	6.5(5.9, 7.0)	Ref	0.49
Tertile 2	97	6.9(6.4, 7.4)	0.27
Tertile 3	97	6.7(6.2, 7.3)	0.50
Sex hormone-binding hormone
Tertile 1	112	7.0(6.5, 7.6)	Ref	0.06
Tertile 2	113	6.5(6.0, 7.0)	0.14
Tertile 3	113	6.3(5.8, 6.8)	0.06

^aGeneral linear models were adjusted for age at sample collection, BMI (continuous), current smoking (yes or no), and current alcohol consumption (yes or no). P for trend was calculated by treating hormone categories as ordinal predictors in multivariate linear regression models

^bMutual adjustment for hormones associated with the proportion of normal sperm morphology in multivariate linear regression models