Ventilating Brain-Injured Patients

Rep. Gabrielle Giffords (D-Ariz.), who suffered a gunshot wound to the head at point blank range early this year, continues to make a remarkable recovery. Giffords was left in critical condition and on mechanical ventilation at The University of Arizona Medical Center in Tucson after a deranged gunman opened fire at a public gathering in January, killing six and wounding her and 12 others.

All medical evidence suggests that Giffords, as an ICU patient with traumatic brain injury (TBI), was at heightened risk for acute lung injury and acute respiratory distress syndrome.

“Rates of ALI in these patients run as high as 30 percent, and there is beautiful data to show that the bigger the brain injury, the more likely the patient is to develop ALI,” said Jeffrey Singh, MD, FRCPC, MSc., of the neurovascular therapeutics program at Toronto Western Hospital, University of Toronto, Ontario, Canada.

Archive ImageA

Yet Giffords was extubated after spending just one week on mechanical ventilation. In fact, she had begun breathing on her own a few days earlier. Her dramatically successful outcome illustrates, among other things, the progress researchers have made in recent years on how to properly manage TBI patients’ airways.

Intracranial pressure risk

The debate on how to mechanically ventilate patients with TBI has largely focused on the dangers posed by CO2, a potent cerebral vasodilator. Increased cerebral blood flow increases intracranial pressure (ICP). In the past, TBI patients were given large tidal volumes, high PaCO2, and very low or no positive end-expiratory pressure to avoid worsening ICP, according to Dr. Singh.

“Twenty years ago, the goals of mechanical ventilation in TBI were to maintain good oxygenation and maintain mild hypocapnia to reduce cerebral hyperemia,” he said. “Now, the thinking is toward strategies that limit ventilator-induced lung injury. These are conflicting paradigms. The old way is probably the opposite of what we want to do for lung-protection. Ventilating this way may increase your risk of developing ALI as time goes on.”

Most patients, he said, require low-normal physiologic volumes and reasonable PEEP to prevent expiratory collapse without risking over-distension: pressure-limited, tidal volume-limited ventilation. “Why is there reluctance to follow this strategy in brain-injured patients?” Dr. Singh asked. “Most comes from a fear of ICP, that high levels of PEEP are bad for ICP, or can invoke permissive hypercapnia. Those are the two big clouds preventing us from always delivering the best lung-protective mechanical ventilation to patients with TBI.”

Brain swelling danger

Most patients with brain injury do not have problems with ICP and should receive lung-protective mechanical ventilaion, Dr. Singh continued. However, some patients are extremely sensitive, and any small change to CO2 will result in a huge increase in ICP.

“If you see on a scan that the brain is very swollen, with no gaps filled with cerebrospinal fluid because the fluid has been squeezed out by all the swelling, that is cause for concern,” he said. “Blood flow to the brain is tightly associated with blood CO2 level. With more CO2, blood volume increases and, in these ‘tight’ brains, ICP may shoot up. Every mmHg that CO2 tension goes up causes about a 1.5 percent increase in cerebral blood flow. If a swollen brain has no more room to swell, and you increase that blood flow, things get very dangerous very quickly.”

Hyperventilating these patients will reduce cerebral blood flow – but, in TBI, large portions of the brain are starving for blood. The clinician must take great care: Increase CO2 in small portions and you put some at risk of ICP. Hyperventilate and you risk starving many of blood. “We now know that hyperventilation strangulates parts of the brain,” observed Peter Rhee, MD, director of trauma and critical care at The University Arizona Medical Center in Tucson, and the physician who directed Rep. Giffords’ care.

“For me, if it was my loved one, I would probably ventilate to a PCO2 of 39 and no less. Because the intracranial compartment and the lungs are interrelated, care must be taken for a patient with moderate ICP. There are a multitude of other tools available to help minimize ICP.”

Dr. Singh added: “Aim for a normal CO2. Don’t hyperventilate the patient. Mild hypercapnea is well tolerated and not a problem. Too much CO2 is less often a problem but low CO2 is a problem for almost everyone.”

Careful use of hypocapnea

Existing literature on this point testifies to the fine line clinicians must tread when mechanically ventilating TBI patients.

Proponents of hyperventilation claim it effectively reduces ICP and that, despite a concomitant reduction in cerebral blood flow (CBF) levels, poses no threat of ischemia. Adversaries counter that reduced CBF levels do have deleterious effects on cerebral oxygenation, and that hyperventilation’s benefits lack corroborating evidence. Maybe the optimal strategy lies somewhere in between.

“Conflicting data may support a range of therapeutic options, from the enthusiastic overuse of hyperventilation to the avoidance of hyperventilation,” researchers wrote in 2005. “It is our opinion that the careful use of hypocapnea for the short-term control of raised ICP remains a useful therapeutic tool.”1

Daniel Davis, MD, an expert in resuscitation at the University of California, San Diego’s department of emergency medicine, noted in his retrospective analysis of the literature in 2008: “Recent clinical and experimental data suggest that hyperventilation is an important contributor to these adverse outcomes in TBI patients. Various mechanisms appear to be responsible for the worsened outcomes, including hemodynamic, cerebrovascular, immunologic, and cellular effects.”2

In 2009, researchers from Brazil urged a judicious use of aggressive ventilation: “Prophylactic hyperventilation in the first 24 hours to achieve a decrease of ICP by cerebral vasoconstriction may lead to an increase of the injured cerebral area due to tissue hypo-perfusion,” they wrote. “Prolonged hyperventilation must be avoided if ICP is not high . However, optimized hyperventilation in short periods seems to be the most promising technique for control of ICP and cerebral perfusion pressure (CPP).”3


1. Stocchetti N, Maas A, Chieregato A, et al. Hyperventilation in Head Injury: A Review. CHEST. 2005;127(5):1812-27.

2. Davis D. Early ventilation in traumatic brain injury. Resuscitation. 2008 Mar;76(3):333-40. Epub 2007 Sep 17.

3. Oliveira-Abreu M, Lajana de Almeida M. Management of mechanical ventilation in brain injury: hyperventilation and positive end-expiratory pressure. Rev Bras Ter Intensiva. 2009;21(1):72-9.

Contact Michael Gibbons at [email protected]