Heat loss from the upper airways and through the skull: studies of direct brain cooling in humans

Abstract

Increased temperature is common after brain trauma and stroke, considered to be detrimental to outcome and usually treated with systemic cooling interventions. However, targeting cooling interventions at the head may be more logical. In addition to arterial blood, the human brain is cooled by heat loss through the skull and heat loss from the upper airways. It is these two mechanisms of heat loss which are the subject of this thesis. The initial research aim was to find out if restoring ‘normal’ airflow through the upper respiratory tracts of intubated, brain-injured patients could reduce brain temperature. Air at room temperature and humidity replicating normal resting minute volume was continuously administered nasally to 15 such patients. After a 30 minute baseline, they were randomised to receive airflow or no airflow for 6 hours and then crossed over for a further 6 hours. The airflow did not produce significant reductions in intracranial temperature (Mean -0.13 °C, SD 0.55 °C, 95% CI -0.43 to 0.17 °C). However, some evidence of heat loss through the skull was serendipitously observed. This was investigated formally in a randomised factorial trial, together with nasal airflow with enhancements (unhumidified air at twice minute volume with 20 ppm nitric oxide gas) intended to overcome some of the possible reasons for the neutral results with ‘normal’ airflow. After a 30 minute baseline, 12 intubated, brain-injured patients received enhanced nasal airflow, bilateral head fanning (8 m/s), both together and no intervention in randomised order. Each intervention was delivered for 30 minutes followed by 30 minutes washout. Mean brain temperature was reduced by 0.15 °C with nasal airflow (p=0.001, 95% CI 0.06 to 0.23 °C) and 0.26 °C with head fanning (p<0.001, 95% CI 0.17 to 0.34 °C). The estimate of the combined effect of airflow and fanning on brain temperature was 0.41 °C. Physiologically, this study demonstrated that heat loss through the upper airways and through the skull can reduce parenchymal brain temperature in brain-injured humans, that the effects are additive and the onset of temperature reduction is rapid. The most promising mechanism appeared to be heat loss through the skull and the final piece of research involved developing and initial (phase I) assessment of a convective head cooling device in healthy volunteers, with intracranial temperature measured non-invasively by magnetic resonance spectroscopy. After a 10 minute baseline, five healthy volunteers received 30 minutes head cooling followed by 30 minutes head and neck cooling via a hood and neck collar delivering 14.5 °C air at 42.5 L/s. The net brain temperature reduction with head cooling was 0.45 °C (SD 0.23 °C, p=0.01, 95% CI 0.17 to 0.74 °C) and with head and neck cooling 0.37 °C (SD 0.30 °C, p=0.049, 95% CI 0.00 to 0.74 °C). There was no significant reduction in cooling with progressive depth into the brain i.e. core brain was cooled. The main relevance of this research is physiological because it adds to knowledge and understanding of mechanisms of heat loss from the upper airways and through the skull in humans. Clinically, factors which enhance or inhibit these mechanisms may have an effect on brain temperature but the therapeutic relevance of head cooling by these methods requires further research.