Mitochondrial dysfunction and ischemia in critical illness: an adipose tissue microdialysis study in 203 ICU patients

Introduction

Ischemia and mitochondrial dysfunction have been implicated in critical illness. The potential of MD to diagnose and separate ischemia and mitochondrial dysfunction in ICU patients remains currently unknown.

Methods

A retrospective, observational study of 203 mechanically ventilated patients studied over a 6-year period with MD including medical, surgical and trauma patients. Sepsis stages: SIRS (n = 24), severe sepsis (n = 46) and septic shock (n = 133). Median age 67 years (range: 17 to 92 years). Mortality was 53%. All subjects had a MD catheter placed in femoral adipose tissue upon admission to the ICU. Interstitial fluid samples were collected six times per day, for 3 consecutive days, and were analyzed for glucose, lactate, pyruvate, and glycerol levels. The lactate to pyruvate (LP) ratio was calculated. Blood lactate was measured. Ischemia was defined as LP ratio >30 and pyruvate level <70 mmol, while mitochondrial dysfunction was defined as LP ratio >30 and pyruvate >70 mmol.

Results

Analysis during the course of the 3-day period revealed three distinct patterns: no ischemia/mitochondrial dysfunction (n = 150 or 74%), ischemia (n = 27 or 13%) and mitochondrial dysfunction (n = 26 or 13%). On day 1, median blood lactate was higher in mitochondrial dysfunction (2.2 mmol/l) compared with both ischemia (1.3 mmol/l) and with no ischemia/mitochondrial dysfunction (1.3 mmol/l) (P = 0.004). Again on day 1, median interstitial fluid lactate was higher in mitochondrial dysfunction (8.4 mmol/l), in comparison with ischemia (1.4 mmol/l) and with the group without ischemia/mitochondrial dysfunction (2.5 mmol/l) (P < 0.001). Similar results were obtained with interstitial fluid glycerol levels (P = 0.009). Median LP ratio was higher in ischemia (LP = 36), and mitochondrial dysfunction (LP = 33) compared with those without ischemia/mitochondrial dysfunction (LP = 17) (P < 0.001). Median interstitial fluid glucose was lower in ischemia (2 mmol/l) compared with both mitochondrial dysfunction (4 mmol/l) and with no ischemia/mitochondrial dysfunction (5 mmol/l) (P < 0.001). ICU mortality was 77% in mitochondrial dysfunction, 52% in ischemia and 49% in the group without ischemia/mitochondrial dysfunction (P = 0.033).

Conclusion

Bedside subcutaneous adipose tissue MD is possible to diagnose and separate ischemia and mitochondrial dysfunction in general ICU patients. These two conditions are not so common; however, mitochondrial dysfunction seems to be associated with higher mortality rates.