Procrustes, the traumatic penumbra, and perfusion pressure targets in closed head injury.

THE cerebral perfusion pressure (CPP; mean arterial pressure intracranial pressure) is a major determinant of cerebral perfusion in the injured brain. Several recent studies have attempted to provide data on optimal CPP targets within the context of protocols for the intensive care management of severe traumatic brain injury. The report by Nordström et al. in this issue of ANESTHESIOLOGY adds to our information on this subject. Many centers use an approach based on a conceptual framework popularized by Rosner et al. and attempt to keep CPP above the lower limit of cerebrovascular autoregulation, which is thought to be shifted upwards following traumatic brain injury. This view has significantly influenced current thinking on this topic, and recent expert guidelines have suggested that CPP be maintained above 70 mmHg. However, a different approach, first proposed by clinicians from Lund, Sweden, is based on interventions aimed at reducing intracranial volume and, hence, intracranial pressure. Perhaps the largest perceived difference between the Lund approach and the Rosner et al. approach (and CPP targets of 70 mmHg specified in current published guidelines) is the stated willingness of the Lund group to accept CPP targets as low as 50 mmHg. Although both of these approaches have been shown to result in good clinical outcomes, they have never been directly compared. However, a recent randomized study by Robertson et al. found no benefit, in terms of the Glasgow Outcome Score, from CPP-targeted therapy (aimed at maintaining CPP above 70 mmHg) when compared with conventional intracranial pressure–targeted therapy (in which CPP was maintained above 50 mmHg). This result was attributed, at least in part, to increased cardiorespiratory complications of therapy in the CPP-targeted group. A recent commentary by Robertson in ANESTHESIOLOGY made a persuasive case for targeting a CPP of 60 mmHg in traumatic brain injury protocols. These results provide useful guidance on optimal CPP levels across patient populations. However, they do not address the issue of heterogeneity between or within patients, which could result in some patients, or some areas in the injured brain, benefiting from a higher (or lower) CPP. Such optimization could conceivably result in subtle improvements in neurocognitive outcome that might be missed by the Glasgow Outcome Score. Nordström et al. attempt to address the first of these two issues in their report. They used microdialysis to monitor extracellular fluid metabolite concentrations from perilesional (“worse”) tissue and relatively normal (“better”) tissue in 50 patients with head injury and retrospectively investigated the relationship between CPP and extracellular fluid concentrations of lactate, pyruvate, and glucose. Extracellular fluid glucose concentrations were unrelated to sampling site and were unaffected by CPP levels. However, extracellular fluid lactate concentrations were higher in worse areas than in better areas when the CPP was greater than 70 mmHg or less than 50 mmHg. Although lactate concentrations in better areas did not vary significantly with CPP, lactate concentrations in the worse areas were significantly higher when CPP decreased to less than 50 mmHg. Lactate–pyruvate ratios generally followed these trends, but differences were less robust. The authors infer that perilesional tissues are more sensitive than normal brain tissue to reductions in CPP and do not tolerate CPP levels below 50 mmHg. They also interpret these data to support the use of the Lund protocol, with reductions in CPP to 50 mmHg if needed for intracranial pressure control. While these results are interesting, we need to consider several issues. Clinical management in these patients was based on the Lund concept. Although CPP was initially maintained at 60–70 mmHg, it was allowed to drop to 50 mmHg if this allowed control of intracranial pressure. Most patients received low-dose thiopental (0.5–3 mg · kg 1 · h ) and antihypertensives (metoprolol and clonidine); 11 received dihydroergotamine to reduce cerebral blood volume. These treatment modalities and the variable (but unspecified) levels of hyperventilation that the patients received represent important confounding factors that may have been responsible, at least in part, for some of the metabolic abnormalities seen. Thus, the patients with the lowest CPP values may well have been the ones who received dihydroergotamine for intracranial pressure control or were hyperventilated to moderate hypocapnia—interventions that have either the potential or the documented effect of reducing regional perfusion and causing metabolic deterioration. It is also important to point out that lactate concentrations in the worse areas were significantly higher than those in the better areas, even when the CPP was greater than This Editorial View accompanies the following article: Nordström C-H, Reinstrup P, Xu W, Gärdenfors A, Ungerstedt U: Assessment of the lower limit for cerebral perfusion pressure in severe head injuries by bedside monitoring of regional energy metabolism. ANESTHESIOLOGY 2003; 98:809–14.