Background

The origin of pro-inflammatory activation in chronic heart failure (HF) remains a matter of debate. Lipopolysaccharide (LPS) may enter the blood stream through the morphologically altered and leaky gut barrier. We hypothesized that lower LPS reactivity would be associated with worse survival as compared to normal or higher LPS reactivity.

Methods

LPS responsiveness was studied in 122 patients with chronic HF (mean ± SD: age 67.3 ± 10.3 years, 24 female, New York Heart Association class [NYHA] class: 2.5 ± 0.8, left ventricular ejection fraction [LVEF]: 33.5 ± 12.5%) and 27 control subjects of similar age (63.7 ± 7.7 years, p > 0.05). Reference LPS was added at increasing doses to ex vivo whole blood samples and necrosis factor-α (TNFα) was measured. Patients were subgrouped into good- and poor-responder status according to their potential to react to increasing doses of LPS (delta TNFα secretion). The optimal cut-off value was calculated by receiver–operator characteristic curve (ROC) analysis.

Results

A total of 56 patients with chronic HF died from any cause during follow-up. At 24 months, cumulative mortality was 16.4% (95% confidence interval 16.0–16.7%). The delta TNFα value representing the optimal cut-off for the prediction of mortality was 1522 pg/mL (24 months) with a sensitivity of 49.3% (95% confidence interval 37.2–61.4%) and specificity of 81.5% (95% confidence interval 61.9–93.6%). LPS responder status remained an independent predictor of death after multivariable adjustment (hazard ratio 0.09 for good- vs. poor-responders, 95% confidence interval 0.01–0.67, p < 0.05).

Conclusions

LPS responsiveness in patients with chronic HF is an independent predictor of death.