The objective of this study is to determine the effect of thiamine therapy on oxygen consumption (VO2) in critically-ill patients. We will evaluate this by measuring VO2 before and after thiamine or placebo administration in patients admitted to the ICU and requiring mechanical ventilation. A secondary aim is to evaluate the effect of thiamine vs. placebo on the metabolic profile of the patients.

Extensive research has been done over the past two decades looking at the role of oxygen delivery (DO2) and oxygen utilization (VO2) in critical illness. VO2 depends on cardiac output, arterial oxygen content, and the body's ability to extract oxygen effectively from the blood. Oxygen demand rises in critical illness as the body goes into a catabolic state, and lower VO2 has been associated with higher lactate levels and with poorer outcomes. Although increasing DO2 was shown in past studies to raise VO2 in some patients, other investigators have found that many critically-ill patients failed to demonstrate a rise in VO2 in spite of achieving supranormal values of cardiac index (CI) and DO2. This group, in contrast to patients whose VO2 rose with the increase in CI and DO2, had exceedingly poor outcomes, suggesting that an inability to extract oxygen from the blood confers a poorer prognosis.

Thiamine deficiency can manifest in several ways, but the syndrome of wet beriberi, caused by thiamine deficiency, includes lactic acidosis, cardiac decompensation and vasodilatory shock, similar to sepsis and other forms of critical illness. The mechanism by which thiamine deficiency causes dysfunction rests upon the vitamin's essential role in the Krebs cycle and Pentose Phosphate Pathway. Lack of adequate thiamine results in the failure of pyruvate to enter the Krebs Cycle, thus preventing aerobic metabolism. The resulting decrease in aerobic metabolism and increase in anaerobic metabolism leads to decreased oxygen consumption by the tissues and increased lactic acid production.

Our group has found previously that upwards of 20% of critically ill patients with sepsis are thiamine deficient within 72 hours of presentation. In a dog model of septic shock, Lindenbaum et al have shown that, regardless of thiamine levels, supplementation with thiamine improved not only lactate clearance and mean arterial pressure, but increased VO2 as well. An increase in VO2 max after administration of thiamine to healthy volunteers has also been described. In our prior open-label study, we found that the administration of a single dose of 200mg of intravenous thiamine to critically ill patients led to a statistically significant increase in VO2 in those with normal or elevated cardiac output, suggesting that thiamine may increase the extraction component of VO2, even in the absence of absolute thiamine deficiency. This effect was not seen in patients with low cardiac output.

VO2 is known to rise in inflammatory states, reflecting increased energy expenditure. Prior studies have shown that VO2 will decrease with interventions such as fever control. In spite of VO2 being higher than normal in critically-ill patients, however, the end-organ damage and lactic acidosis suggest that it is not high enough to meet the metabolic demands of the critically-ill body. If we are able to increase VO2 further in critically-ill patients, we could potentially help maintain aerobic metabolism and decrease tissue hypoxia and the resulting end-organ damage. Our hypothesis is that administering thiamine intravenously to critically-ill patients who do not have abnormally low cardiac index will increase VO2.

We will use an anesthesia monitor with a gas exchange module to measure VO2 continuously over a 9 hour period. After 3 hours of baseline VO2 data are collected, baseline thiamine level, lactate, and central venous O2 saturation will be obtained. A single dose of 200mg of IV thiamine will then be given, and 6 hours of post-thiamine data will then be collected. We will screen all consenting patients for whom we do not know the cardiac index with a non-invasive cardiac index measurement using the Cheetah non-invasive cardiac output monitor (NICOM). We will not include patients with a cardiac index less than or equal to 2.4L/min/m2, due to our preliminary data showing these patients did not increase VO2 in response to thiamine. All patients enrolled will have cardiac index monitored continuously during the study by the NICOM, in order to assess whether or not there is any relationship between VO2 and cardiac index. Patients will also have blood drawn for a metabolomic panel before and after thiamine or placebo to assess whether thiamine has an effect on the metabolome.

Inclusion Criteria:

1. Adult patients (age > 18 yrs) admitted to an ICU
2. Mechanically ventilated for an acute illness, with stable respiratory status (no changes in ventilator settings in the 3 hours prior to enrollment)
3. Cardiac index >2.4L/min/m2 as measured by Noninvasive Cardiac Output Monitor(NICOM) by Cheetah Medical or, if being used clinically, by PA catheter or Vigileo device.
4. Upper central venous line in place

Exclusion Criteria:

1. Unstable ventilator settings during measurement of VO2
2. Temperature >100.5
3. FIO2>60%
4. Endotracheal cuff leak, chest tube, or other evident source of air leak
5. Positive end expiratory pressure > 12cmH2O
6. Intravenous thiamine supplementation within 2 weeks of enrollment, or oral supplementation more than that found in a multivitamin.
7. Protected populations (pregnant woman, prisoners, cognitively impaired)

All patients were followed for outcomes, including adverse events, until hospital discharge

We collected data on mortality, but this was not attributed to study drug in any instance.

We found differences in variability in VO2 after study drug in the subgroup of patients (35%) who were thiamine deficient.