Table 4:
Hazard ratios and 95% confidence intervals of the association between pineal parenchyma volume, pineal gland composition, melatonin, and total prostate cancer risk (n = 861), AGES-Reykjavik Study
| Parenchyma volume† | Pineal calcifications and cysts | Melatonin level‡ | Total prostate cancer cases/participants | HR (95% CI) Age-and creatinine-adjusted |
HR (95% CI) Multivariable-adjusted§ |
|---|---|---|---|---|---|
| Low | No | Low | 30/193 | Ref. | Ref. |
| Low | No | High | 25/124 | 1.15 (0.66, 2.00) | 1.02 (0.58, 1.80) |
| Low | Yes | Low | 7/60 | 0.77 (0.34, 1.76) | 0.83 (0.36, 1.90) |
| Low | Yes | High | 7/54 | 0.74 (0.32, 1.70) | 0.67 (0.29, 1.54) |
| High | No | Low | 23/127 | 1.22 (0.71, 2.11) | 1.21 (0.70, 2.10) |
| High | No | High | 26/173 | 0.84 (0.49, 1.45) | 0.79 (0.45, 1.38) |
| High | Yes | Low | 13/50 | 1.97 (1.03, 3.80)* | 1.98 (1.02, 3.84)* |
| High | Yes | High | 11/80 | 0.79 (0.39, 1.59) | 0.79 (0.39, 1.61) |
HR = hazard ratio; 95% CI = 95% confidence interval
Pineal parenchyma volumes were dichotomized by the median for the subsample of 122.6 mm3: low is below or equal to the median and high is above the median.
Urinary 6-sulfatoxymelatonin levels were dichotomized by the median for the subsample 17.0 ng/mL: low is below or equal to the median and high is above the median.
Multivariable-adjusted model adjusted for age, family history of prostate cancer, smoking status, physical activity (never, rarely/occasionally, moderate/high), history of depression, current beta-blocker use, annual doctor’s exam, and creatinine (ng/mL).
p<0.05