Table 3.
Adjusted odds ratios for progression to dengue hemorrhagic fever/dengue shock syndrome according to baseline characteristics and serum fatty acids %
| Characteristic | Controls: Cases | Unadjusted OR (95% CI)1 | Adjusted OR (95% CI)2 | Adjusted P3 |
|---|---|---|---|---|
| Sex | 0.85 | |||
| Female | 117:56 | 1.00 | 1.00 | |
| Male | 118:53 | 0.94 (0.60, 1.48) | 0.95 (0.57, 1.58) | |
| Age, y | ||||
| <15 | 165:93 | 1.00 | 1.00 | 0.003 |
| ≥15 | 70:16 | 2.47 (1.35, 4.49) | 2.75 (1.41, 5.34) | |
| Hours with fever before consultation | 0.02 | |||
| <72 | 81:49 | 1.00 | 1.00 | |
| 72 to <96 | 154:60 | 0.64 (0.41, 1.02) | 0.52 (0.31, 0.89) | |
| Interleukin-10, per 200 pg/ml | - | 1.25 (1.03, 1.53) | 1.23 (0.97, 1.57) | 0.09 |
| Tumor necrosis factor-α, per 10 pg/ml | - | 1.22 (0.97, 1.55) | 1.31 (0.98, 1.74) | 0.07 |
| 15:0 pentadecanoic acid quintile4, % (median) | 0.002 | |||
| Q1 (0.11) | 46:27 | 1.00 | 1.00 | |
| Q2 (0.14) | 47:27 | 0.98 (0.50, 1.91) | 0.67 (0.32, 1.40) | |
| Q3 (0.16) | 47:23 | 0.83 (0.42, 1.66) | 0.55 (0.26, 1.19) | |
| Q4 (0.19) | 47:22 | 0.80 (0.40, 1.60) | 0.53 (0.24, 1.15) | |
| Q5 (0.24) | 47:9 | 0.33 (0.14, 0.77) | 0.24 (0.09, 0.60) | |
| 22:6n-3 DHA quintile4, % (median) | 0.007 | |||
| Q1 (0.81) | 47:7 | 1.00 | 1.00 | |
| Q2 (1.11) | 47:26 | 3.71 (1.47, 9.39) | 4.44 (1.67, 11.8) | |
| Q3 (1.37) | 47:20 | 2.86 (1.10, 7.39) | 4.57 (1.64, 12.7) | |
| Q4 (1.68) | 47:24 | 3.43 (1.35, 8.72) | 5.58 (2.01, 15.4) | |
| Q5 (2.26) | 47:32 | 4.57 (1.84, 11.4) | 5.34 (2.03, 14.1) | |
| 20:3n-6 dihomo-γ-linolenic acid quintile4, % (median) | 0.007 | |||
| Q1 (0.91) | 47:31 | 1.00 | 1.00 | |
| Q2 (1.14) | 47:22 | 0.71 (0.36, 1.40) | 0.51 (0.24, 1.09) | |
| Q3 (1.34) | 47:23 | 0.74 (0.38, 1.46) | 0.58 (0.27, 1.24) | |
| Q4 (1.55) | 47:17 | 0.55 (0.27, 1.12) | 0.43 (0.19, 0.96) | |
| Q5 (1.98) | 47:16 | 0.52 (0.25, 1.07) | 0.30 (0.13, 0.69) | |
| Δ5-desaturase5 quintile (median) | 0.0004 | |||
| Q1 (2.89) | 47:15 | 1.00 | 1.00 | |
| Q2 (3.80) | 47:12 | 0.80 (0.34, 1.89) | 0.97 (0.38, 2.50) | |
| Q3 (4.50) | 47:14 | 0.93 (0.41, 2.15) | 0.85 (0.34, 2.12) | |
| Q4 (5.16) | 47:22 | 1.47 (0.68, 3.17) | 1.40 (0.60, 3.28) | |
| Q5 (6.77) | 47:46 | 3.07 (1.51, 6.23) | 3.28 (1.42, 7.58) |
From unconditional logistic regression models with case status (dengue hemorrhagic fever/dengue shock syndrome) as the outcome and each covariate presented as predictor.
From an unconditional logistic regression model with case status as the outcome and predictors that included sex (1 indicator for male), age (1 indicator for < 15 y), hours with fever before consultation (1 indicator for ≥ 72 h), interleukin-10 (continuous), tumor necrosis factor-α (continuous), 15:0 pentadecanoic acid quintile (4 indicators), 22:6n-3 DHA quintile (4 indicators), and 20:3n-6 dihomo-γ-linolenic acid quintile (4 indicators). Estimates for ∆5-desaturase were obtained from a model that excluded 20:3n-6 dihomo-γ-linolenic acid due to collinearity.
For sex, age, hours with fever before consultation and cytokines, Wald test. For FA, test for linear trend when a variable representing the median of each quartile was introduced into the logistic regression model as a continuous predictor.
% of total serum FA. DHA, docosahexaenoic acid.
Ratio of 20:4n-6 / 20:3n-6.