Skip to main content

Table 2 Comparison of the concentration of tryptophan catabolites and prediction of depression by tryptophan catabolites

From: Tryptophan catabolites and depression in the general population: results from the Gutenberg Health Study

 

Not depressed (PHQ-9 < 5) n = 2040

Depressed (PHQ-9 ≥ 10) n = 248

Estimate (diff. 95% CI), p

Test model 1 OR (95% CI), p

Test model 2 OR (95% CI), p

Test model 3 OR (95% CI), p

Tryptophan [nmol/l]

50,188 ± 9121

49,038 ± 10,214

1150 [-186; 2487], p = 0.091

1 (1–1) p = 0.97

1 (1–1) p = 0.82

1 (1–1) p = 0.79

Kynurenine [nmol/l]

2570 (2180/3040)

2445 (2060/2920)

-0.23 [-0.096; -0.36], p = 0.00077

0.905 (0.798–1.025) p = .12

0.905 (0.798–1.026) p = .12

0.861 (0.753–0.984) p = 0.028

Kynurenic acid [nmol/l]

33.90 (27.50/42.36)

32.00 (24.44/38.70)

-0.27 [-0.12; -0.42], p = 0,00,045

0.814 (0.715 -0.925) P = 0.0016

0.827 (0.727–0.939) p = 0.0037

0.834 (0.731–0.951) p = 0.0071

Quinolinic acid [nmol/l]

309.5 (253.0/384.0)

296.5 (240.0/378.3)

-0.097 [-0.046; 0.24], p = 0.18

0.952 (0.836 -1.081) p = 0.46

0.946 (0.830–1.075) p = 0.40

0.917 (0.796–1.053) p = 0.23

  1. Student’s t-test for the comparison of tryptophan catabolite concentration between not depressed and depressed persons. Tryptophan catabolite (trycat) concentrations are shown as means (± standard deviation) or medians (and interquartile range (25th/75th)
  2. The odds ratios reflect change per standard deviation for log-transformed kynurenine, kynurenine acid and quinolinic acid. Tryptophan was normally distributed and thus not log-transformed
  3. Test model 1 = logistic regression analysis with the dependent variable depressed versus not depressed and the independent variables trycat per standard deviation and sex and age. For model 1, n = 248 cases of depression could be used
  4. Test model 2 = logistic regression analysis with the dependent variable depressed versus not depressed and the independent variables trycat per standard deviation and sex, age, SES. For model 2, n = 246 cases of depression could be used
  5. Test model 3: logistic regression analysis with the dependent variable depressed versus not depressed and the independent variables trycat per standard deviation and sex, age, SES, smoking, obesity, excessive alcohol consumption, physical activity score, heart rate, systolic blood pressure [mmHg], CRP, HOMA-IR, LDL/HDL. For model 3, n = 187 cases of depression could be used