Dr. David L. Bowton, FCCP

Perhaps no other concept in modern medicine is surrounded by as much misunderstanding and controversy as that of death by neurologic criteria, or brain death.

Legal challenges to brain death have spanned the gamut of continuing life support for a legally dead patient to denying a family’s request to continue support for their dead child.

Two recent, highly publicized cases highlight the poles of these controversies and concerns. In November 2013, Marlise Muñoz was declared brain dead. She was 33 years old and 14 weeks’ pregnant. Concordant with the patient’s wishes, the husband requested that the hospital remove the patient from mechanical ventilatory support. Citing Texas law that precludes the withholding of life-sustaining treatment to a pregnant patient, the hospital refused, despite the logical fallacy of withholding life support from a person declared dead.

In December 2013, 13-year-old Jahi McMath was declared brain dead. Her parents obtained a court order to prevent the hospital from discontinuing ventilatory support. A better understanding of the development of the concept of brain death and the incorporation of recent evidence into its determination may reduce the confusion and misunderstandings surrounding the determination of brain death.

Three reports in 1959 describe death of the nervous system and brain death (coma dépassé) building upon clinical and laboratory data from the 1930s relating the cessation of brain blood flow and EEG activity to ensuing apnea and subsequent cardiac arrest (Machado et al. J Med Ethics. 2007;33[4]:197). With the increasing sophistication of critical care support, especially mechanical ventilation, the loss of central respiratory drive no longer meant imminent death, and patients could be supported for long periods of time without recovery of brain function.

In 1968, an ad hoc committee at Harvard Medical School proposed a definition of irreversible coma and brain death (Beecher et al. JAMA. 1968;205[6]:337). They posited that any organ that no longer functions and has no possibility of functioning again was "for all practical purposes dead." The goals of the committee were (1) to reduce the suffering of patients and families and the care burden of hospitals; and (2) mitigate the controversies that surrounded obtaining organs for transplantation.

The committee then set out their rationale for their proposed characteristics of the permanently nonfunctioning brain. Subsequently, in 1976, The Conference of the Medical Royal Colleges and their Faculties in the United Kingdom required a deeply comatose state, irremediable structural brain damage, and irreversible cessation of brain stem function to ascertain brain death (Spinello. J Intensive Care Med. May 2014, in press).

In 1981, the National Conference of Commissioners on Uniform State Laws approved the Uniform Determination of Death Act (UDDA), codifying the legal concept of brain death. Thirty-six states and the District of Columbia have enacted the UDDA, but legal precedent and statutes in the remaining states are consistent with the UDDA. New York and New Jersey require that a family’s religious or moral views be considered in the process following the determination of brain death (Gostin. JAMA. 2014;311[9]:903); in all other states, clinicians are not required to consult with family prior to withdrawing ventilatory support from the brain dead patient.

‘No reports of clinical recovery’

However, the UDDA does not define brain death but rather states that "a determination of death must be made with accepted medical standards" (Wijdicks et al. Neurology. 2010;74[23]:1911). The American Academy of Neurology (AAN) published a practice parameter in 1995 to guide clinicians in the determination of brain death; this was updated by their Quality Standards Committee in 2010 (Wijdicks et al. Neurology. 2010;74[23]:1911). In the updated guidance document, the committee stated: "There are no reports of clinical recovery in patients after the clinical diagnosis of brain death has been determined using the AAN practice parameter."

It is striking, however, that there appears to be an inconsistent approach and criteria for the determination of brain death across hospitals, and even amongst specialties within a hospital (Powner et al. Crit Care Med. 2004;32[6]:1284). Further, in most states, any physician is permitted to ascertain brain death, though some states have now added requirements for specific qualifications and/or confirmation by a second physician (Spinello IM. J Intensive Care Med. May 2014, in press).

These inconsistencies in criteria, process, and experience may be responsible for reports in the lay press of recovery following pronouncement of brain death. Every hospital should have a validated, well-defined process for the evaluation of brain death guided by the AAN practice parameter.

A single exam

In most states, a single exam is required to determine brain death. The clinical setting and the results of imaging studies should be used to determine the duration of observation needed to exclude the possibility of recovery. Given that the patient should be normothermic, have a normal systolic blood pressure, have both a history and imaging studies supporting an irreversible cause of coma, and that drugs and toxins must be excluded as a cause of coma, this observation period is uncommonly less than several hours.

Confounded by movement

Movement of the patient with devastating brain injury is often a confounder in the determination of brain death.

This is also a source of confusion and concern for the family. Plantar reflexes have been commonly reported in patients pronounced brain dead. Head turning in response to noxious stimuli, repetitive leg movements, facial myokymia, and other movements have been observed (Wu et al. Crit Care. 2013;17[4]:440; Wijdicks et al. Neurology. 2010;74[23]:1911). When this is observed, knowledge of the potential reflex arcs is needed, and consultation with a neurologist or other physician skilled in brain death determination is usually appropriate.

Movement can sometimes extend to apparent respiratory activity. This can be caused by ventilator autocycling due to variations in circuit pressure (especially in a noncompliant lung or with large tidal volumes) or triggering due to cardiac-induced alterations in transpulmonary pressure. The latter is most common in a hyperdynamic circulatory state.

Apnea is a critical criterion for the determination of brain death and is most reliably assessed when the patient has been removed from the mechanical ventilator. During testing, apnea should be present despite achieving a PaCO₂ of greater than or equal to 60 mm Hg, or 20 mm Hg above the baseline to support the diagnosis of brain death.

Some patients, especially those who are hemodynamically unstable or who require high levels of PEEP to support oxygenation, may not tolerate the apnea test without oxygen desaturation or hypotension.

Most commonly, the apnea test is performed with the patient off the ventilator and 6 L/min of oxygen flowing through an insufflation catheter placed near the carina. We prefer to use a modified Mapleson circuit with enough flow to only partially distend the anesthesia bag in the circuit. With this arrangement, one can look not only at the patient’s chest wall and abdomen for signs of respiratory activity but also look for cyclic changes in the level of bag inflation.

In patients with ARDS, trauma involving the chest or other causes of oxygenation failure, CPAP, with or without an antecedent recruitment maneuver may facilitate successful completion of the apnea test (Hocker et al. Neurocrit Care. 2014;20[2]:298).

Controversy over ancillary testing

There are no well-designed prospective studies examining the accuracy of ancillary tests for the determination of brain death using the appropriate control group of patients with coma but who are not brain dead and with blinding of the interpretation of the study results to the clinical setting.

Brain death is not synonymous with complete neuronal death. While cortical areas exhibit moderate to severe histopathologic ischemic changes in the large majority (but not all) of brain dead patients, the basal ganglia and diencephalon inconsistently demonstrate these changes (Wijdicks et al. Neurology. 2008;70[15]:1234). Thus, it may not be surprising that both false-positive (test positive for brain death – clinically not brain dead) and false-negative (test negative for brain death – clinically brain dead) results have been reported for virtually all tests proposed for confirmatory testing, including CT angiography, transcranial Doppler, and nuclear brain scan. Confirmatory testing is most commonly used in patients who cannot complete an apnea test.

It has been forcefully argued that, in adults, confirmatory tests should not be done (Wijdicks. Neurology. 2010;75[1]:77). There will be some patients in whom brain death cannot be definitively determined. When all criteria of the AAN guidelines cannot be fulfilled, clinicians should err on the side of concluding that the patient is not brain dead and turn their efforts toward counseling the family regarding the likelihood of neurologic recovery and assisting the family in ascertaining the desires of the patient under these circumstances.

It is vital that care providers not lose perspective on the plight of the family in these settings. The symbolic power of a beating heart to a parent, spouse, or loved one cannot be underestimated. Family members who observe resuscitation efforts have been shown to have a lower incidence of posttraumatic stress disorder symptoms (Jabre et al. N Engl J Med. 2013;368[11]:1008).

A recent trial demonstrated that family presence during the brain death determination had an increased understanding of brain death without an adverse impact on emotional well being (Tawil et al. Crit Care Med. 2013;42[4]:934). A consistent process for determination of brain death, engagement of the family in the brain death evaluation and helping them understand the meaning of patient movements that can be distressingly misinterpreted, and honesty when a definitive determination of brain death cannot be determined will not remove the misunderstanding or controversy surrounding a diagnosis of brain death but will serve to ensure its accurate and humane application.

Dr. Bowton is Professor, Section on Critical Care, Department of Anesthesiology, Wake Forest Baptist Health, Winston-Salem, North Carolina. Read previous Critical Care Commentaries online.

Perhaps no other concept in modern medicine is surrounded by as much misunderstanding and controversy as that of death by neurologic criteria, or brain death.

Legal challenges to brain death have spanned the gamut of continuing life support for a legally dead patient to denying a family’s request to continue support for their dead child.

Two recent, highly publicized cases highlight the poles of these controversies and concerns. In November 2013, Marlise Muñoz was declared brain dead. She was 33 years old and 14 weeks’ pregnant. Concordant with the patient’s wishes, the husband requested that the hospital remove the patient from mechanical ventilatory support. Citing Texas law that precludes the withholding of life-sustaining treatment to a pregnant patient, the hospital refused, despite the logical fallacy of withholding life support from a person declared dead.

In December 2013, 13-year-old Jahi McMath was declared brain dead. Her parents obtained a court order to prevent the hospital from discontinuing ventilatory support. A better understanding of the development of the concept of brain death and the incorporation of recent evidence into its determination may reduce the confusion and misunderstandings surrounding the determination of brain death.

Three reports in 1959 describe death of the nervous system and brain death (coma dépassé) building upon clinical and laboratory data from the 1930s relating the cessation of brain blood flow and EEG activity to ensuing apnea and subsequent cardiac arrest (Machado et al. J Med Ethics. 2007;33[4]:197). With the increasing sophistication of critical care support, especially mechanical ventilation, the loss of central respiratory drive no longer meant imminent death, and patients could be supported for long periods of time without recovery of brain function.

In 1968, an ad hoc committee at Harvard Medical School proposed a definition of irreversible coma and brain death (Beecher et al. JAMA. 1968;205[6]:337). They posited that any organ that no longer functions and has no possibility of functioning again was "for all practical purposes dead." The goals of the committee were (1) to reduce the suffering of patients and families and the care burden of hospitals; and (2) mitigate the controversies that surrounded obtaining organs for transplantation.

The committee then set out their rationale for their proposed characteristics of the permanently nonfunctioning brain. Subsequently, in 1976, The Conference of the Medical Royal Colleges and their Faculties in the United Kingdom required a deeply comatose state, irremediable structural brain damage, and irreversible cessation of brain stem function to ascertain brain death (Spinello. J Intensive Care Med. May 2014, in press).

In 1981, the National Conference of Commissioners on Uniform State Laws approved the Uniform Determination of Death Act (UDDA), codifying the legal concept of brain death. Thirty-six states and the District of Columbia have enacted the UDDA, but legal precedent and statutes in the remaining states are consistent with the UDDA. New York and New Jersey require that a family’s religious or moral views be considered in the process following the determination of brain death (Gostin. JAMA. 2014;311[9]:903); in all other states, clinicians are not required to consult with family prior to withdrawing ventilatory support from the brain dead patient.

‘No reports of clinical recovery’

However, the UDDA does not define brain death but rather states that "a determination of death must be made with accepted medical standards" (Wijdicks et al. Neurology. 2010;74[23]:1911). The American Academy of Neurology (AAN) published a practice parameter in 1995 to guide clinicians in the determination of brain death; this was updated by their Quality Standards Committee in 2010 (Wijdicks et al. Neurology. 2010;74[23]:1911). In the updated guidance document, the committee stated: "There are no reports of clinical recovery in patients after the clinical diagnosis of brain death has been determined using the AAN practice parameter."

It is striking, however, that there appears to be an inconsistent approach and criteria for the determination of brain death across hospitals, and even amongst specialties within a hospital (Powner et al. Crit Care Med. 2004;32[6]:1284). Further, in most states, any physician is permitted to ascertain brain death, though some states have now added requirements for specific qualifications and/or confirmation by a second physician (Spinello IM. J Intensive Care Med. May 2014, in press).

These inconsistencies in criteria, process, and experience may be responsible for reports in the lay press of recovery following pronouncement of brain death. Every hospital should have a validated, well-defined process for the evaluation of brain death guided by the AAN practice parameter.

A single exam

In most states, a single exam is required to determine brain death. The clinical setting and the results of imaging studies should be used to determine the duration of observation needed to exclude the possibility of recovery. Given that the patient should be normothermic, have a normal systolic blood pressure, have both a history and imaging studies supporting an irreversible cause of coma, and that drugs and toxins must be excluded as a cause of coma, this observation period is uncommonly less than several hours.

Confounded by movement

Movement of the patient with devastating brain injury is often a confounder in the determination of brain death.

This is also a source of confusion and concern for the family. Plantar reflexes have been commonly reported in patients pronounced brain dead. Head turning in response to noxious stimuli, repetitive leg movements, facial myokymia, and other movements have been observed (Wu et al. Crit Care. 2013;17[4]:440; Wijdicks et al. Neurology. 2010;74[23]:1911). When this is observed, knowledge of the potential reflex arcs is needed, and consultation with a neurologist or other physician skilled in brain death determination is usually appropriate.

Movement can sometimes extend to apparent respiratory activity. This can be caused by ventilator autocycling due to variations in circuit pressure (especially in a noncompliant lung or with large tidal volumes) or triggering due to cardiac-induced alterations in transpulmonary pressure. The latter is most common in a hyperdynamic circulatory state.

Apnea is a critical criterion for the determination of brain death and is most reliably assessed when the patient has been removed from the mechanical ventilator. During testing, apnea should be present despite achieving a PaCO₂ of greater than or equal to 60 mm Hg, or 20 mm Hg above the baseline to support the diagnosis of brain death.

Some patients, especially those who are hemodynamically unstable or who require high levels of PEEP to support oxygenation, may not tolerate the apnea test without oxygen desaturation or hypotension.

Most commonly, the apnea test is performed with the patient off the ventilator and 6 L/min of oxygen flowing through an insufflation catheter placed near the carina. We prefer to use a modified Mapleson circuit with enough flow to only partially distend the anesthesia bag in the circuit. With this arrangement, one can look not only at the patient’s chest wall and abdomen for signs of respiratory activity but also look for cyclic changes in the level of bag inflation.

In patients with ARDS, trauma involving the chest or other causes of oxygenation failure, CPAP, with or without an antecedent recruitment maneuver may facilitate successful completion of the apnea test (Hocker et al. Neurocrit Care. 2014;20[2]:298).

Controversy over ancillary testing

There are no well-designed prospective studies examining the accuracy of ancillary tests for the determination of brain death using the appropriate control group of patients with coma but who are not brain dead and with blinding of the interpretation of the study results to the clinical setting.

Brain death is not synonymous with complete neuronal death. While cortical areas exhibit moderate to severe histopathologic ischemic changes in the large majority (but not all) of brain dead patients, the basal ganglia and diencephalon inconsistently demonstrate these changes (Wijdicks et al. Neurology. 2008;70[15]:1234). Thus, it may not be surprising that both false-positive (test positive for brain death – clinically not brain dead) and false-negative (test negative for brain death – clinically brain dead) results have been reported for virtually all tests proposed for confirmatory testing, including CT angiography, transcranial Doppler, and nuclear brain scan. Confirmatory testing is most commonly used in patients who cannot complete an apnea test.

It has been forcefully argued that, in adults, confirmatory tests should not be done (Wijdicks. Neurology. 2010;75[1]:77). There will be some patients in whom brain death cannot be definitively determined. When all criteria of the AAN guidelines cannot be fulfilled, clinicians should err on the side of concluding that the patient is not brain dead and turn their efforts toward counseling the family regarding the likelihood of neurologic recovery and assisting the family in ascertaining the desires of the patient under these circumstances.

It is vital that care providers not lose perspective on the plight of the family in these settings. The symbolic power of a beating heart to a parent, spouse, or loved one cannot be underestimated. Family members who observe resuscitation efforts have been shown to have a lower incidence of posttraumatic stress disorder symptoms (Jabre et al. N Engl J Med. 2013;368[11]:1008).

A recent trial demonstrated that family presence during the brain death determination had an increased understanding of brain death without an adverse impact on emotional well being (Tawil et al. Crit Care Med. 2013;42[4]:934). A consistent process for determination of brain death, engagement of the family in the brain death evaluation and helping them understand the meaning of patient movements that can be distressingly misinterpreted, and honesty when a definitive determination of brain death cannot be determined will not remove the misunderstanding or controversy surrounding a diagnosis of brain death but will serve to ensure its accurate and humane application.

Dr. Bowton is Professor, Section on Critical Care, Department of Anesthesiology, Wake Forest Baptist Health, Winston-Salem, North Carolina. Read previous Critical Care Commentaries online.

Perhaps no other concept in modern medicine is surrounded by as much misunderstanding and controversy as that of death by neurologic criteria, or brain death.

Legal challenges to brain death have spanned the gamut of continuing life support for a legally dead patient to denying a family’s request to continue support for their dead child.

Two recent, highly publicized cases highlight the poles of these controversies and concerns. In November 2013, Marlise Muñoz was declared brain dead. She was 33 years old and 14 weeks’ pregnant. Concordant with the patient’s wishes, the husband requested that the hospital remove the patient from mechanical ventilatory support. Citing Texas law that precludes the withholding of life-sustaining treatment to a pregnant patient, the hospital refused, despite the logical fallacy of withholding life support from a person declared dead.

In December 2013, 13-year-old Jahi McMath was declared brain dead. Her parents obtained a court order to prevent the hospital from discontinuing ventilatory support. A better understanding of the development of the concept of brain death and the incorporation of recent evidence into its determination may reduce the confusion and misunderstandings surrounding the determination of brain death.

Three reports in 1959 describe death of the nervous system and brain death (coma dépassé) building upon clinical and laboratory data from the 1930s relating the cessation of brain blood flow and EEG activity to ensuing apnea and subsequent cardiac arrest (Machado et al. J Med Ethics. 2007;33[4]:197). With the increasing sophistication of critical care support, especially mechanical ventilation, the loss of central respiratory drive no longer meant imminent death, and patients could be supported for long periods of time without recovery of brain function.

In 1968, an ad hoc committee at Harvard Medical School proposed a definition of irreversible coma and brain death (Beecher et al. JAMA. 1968;205[6]:337). They posited that any organ that no longer functions and has no possibility of functioning again was "for all practical purposes dead." The goals of the committee were (1) to reduce the suffering of patients and families and the care burden of hospitals; and (2) mitigate the controversies that surrounded obtaining organs for transplantation.

The committee then set out their rationale for their proposed characteristics of the permanently nonfunctioning brain. Subsequently, in 1976, The Conference of the Medical Royal Colleges and their Faculties in the United Kingdom required a deeply comatose state, irremediable structural brain damage, and irreversible cessation of brain stem function to ascertain brain death (Spinello. J Intensive Care Med. May 2014, in press).

In 1981, the National Conference of Commissioners on Uniform State Laws approved the Uniform Determination of Death Act (UDDA), codifying the legal concept of brain death. Thirty-six states and the District of Columbia have enacted the UDDA, but legal precedent and statutes in the remaining states are consistent with the UDDA. New York and New Jersey require that a family’s religious or moral views be considered in the process following the determination of brain death (Gostin. JAMA. 2014;311[9]:903); in all other states, clinicians are not required to consult with family prior to withdrawing ventilatory support from the brain dead patient.

‘No reports of clinical recovery’

However, the UDDA does not define brain death but rather states that "a determination of death must be made with accepted medical standards" (Wijdicks et al. Neurology. 2010;74[23]:1911). The American Academy of Neurology (AAN) published a practice parameter in 1995 to guide clinicians in the determination of brain death; this was updated by their Quality Standards Committee in 2010 (Wijdicks et al. Neurology. 2010;74[23]:1911). In the updated guidance document, the committee stated: "There are no reports of clinical recovery in patients after the clinical diagnosis of brain death has been determined using the AAN practice parameter."

It is striking, however, that there appears to be an inconsistent approach and criteria for the determination of brain death across hospitals, and even amongst specialties within a hospital (Powner et al. Crit Care Med. 2004;32[6]:1284). Further, in most states, any physician is permitted to ascertain brain death, though some states have now added requirements for specific qualifications and/or confirmation by a second physician (Spinello IM. J Intensive Care Med. May 2014, in press).

These inconsistencies in criteria, process, and experience may be responsible for reports in the lay press of recovery following pronouncement of brain death. Every hospital should have a validated, well-defined process for the evaluation of brain death guided by the AAN practice parameter.

A single exam

In most states, a single exam is required to determine brain death. The clinical setting and the results of imaging studies should be used to determine the duration of observation needed to exclude the possibility of recovery. Given that the patient should be normothermic, have a normal systolic blood pressure, have both a history and imaging studies supporting an irreversible cause of coma, and that drugs and toxins must be excluded as a cause of coma, this observation period is uncommonly less than several hours.

Confounded by movement

Movement of the patient with devastating brain injury is often a confounder in the determination of brain death.

This is also a source of confusion and concern for the family. Plantar reflexes have been commonly reported in patients pronounced brain dead. Head turning in response to noxious stimuli, repetitive leg movements, facial myokymia, and other movements have been observed (Wu et al. Crit Care. 2013;17[4]:440; Wijdicks et al. Neurology. 2010;74[23]:1911). When this is observed, knowledge of the potential reflex arcs is needed, and consultation with a neurologist or other physician skilled in brain death determination is usually appropriate.

Movement can sometimes extend to apparent respiratory activity. This can be caused by ventilator autocycling due to variations in circuit pressure (especially in a noncompliant lung or with large tidal volumes) or triggering due to cardiac-induced alterations in transpulmonary pressure. The latter is most common in a hyperdynamic circulatory state.

Apnea is a critical criterion for the determination of brain death and is most reliably assessed when the patient has been removed from the mechanical ventilator. During testing, apnea should be present despite achieving a PaCO₂ of greater than or equal to 60 mm Hg, or 20 mm Hg above the baseline to support the diagnosis of brain death.

Some patients, especially those who are hemodynamically unstable or who require high levels of PEEP to support oxygenation, may not tolerate the apnea test without oxygen desaturation or hypotension.

Most commonly, the apnea test is performed with the patient off the ventilator and 6 L/min of oxygen flowing through an insufflation catheter placed near the carina. We prefer to use a modified Mapleson circuit with enough flow to only partially distend the anesthesia bag in the circuit. With this arrangement, one can look not only at the patient’s chest wall and abdomen for signs of respiratory activity but also look for cyclic changes in the level of bag inflation.

In patients with ARDS, trauma involving the chest or other causes of oxygenation failure, CPAP, with or without an antecedent recruitment maneuver may facilitate successful completion of the apnea test (Hocker et al. Neurocrit Care. 2014;20[2]:298).

Controversy over ancillary testing

There are no well-designed prospective studies examining the accuracy of ancillary tests for the determination of brain death using the appropriate control group of patients with coma but who are not brain dead and with blinding of the interpretation of the study results to the clinical setting.

Brain death is not synonymous with complete neuronal death. While cortical areas exhibit moderate to severe histopathologic ischemic changes in the large majority (but not all) of brain dead patients, the basal ganglia and diencephalon inconsistently demonstrate these changes (Wijdicks et al. Neurology. 2008;70[15]:1234). Thus, it may not be surprising that both false-positive (test positive for brain death – clinically not brain dead) and false-negative (test negative for brain death – clinically brain dead) results have been reported for virtually all tests proposed for confirmatory testing, including CT angiography, transcranial Doppler, and nuclear brain scan. Confirmatory testing is most commonly used in patients who cannot complete an apnea test.

It has been forcefully argued that, in adults, confirmatory tests should not be done (Wijdicks. Neurology. 2010;75[1]:77). There will be some patients in whom brain death cannot be definitively determined. When all criteria of the AAN guidelines cannot be fulfilled, clinicians should err on the side of concluding that the patient is not brain dead and turn their efforts toward counseling the family regarding the likelihood of neurologic recovery and assisting the family in ascertaining the desires of the patient under these circumstances.

It is vital that care providers not lose perspective on the plight of the family in these settings. The symbolic power of a beating heart to a parent, spouse, or loved one cannot be underestimated. Family members who observe resuscitation efforts have been shown to have a lower incidence of posttraumatic stress disorder symptoms (Jabre et al. N Engl J Med. 2013;368[11]:1008).

A recent trial demonstrated that family presence during the brain death determination had an increased understanding of brain death without an adverse impact on emotional well being (Tawil et al. Crit Care Med. 2013;42[4]:934). A consistent process for determination of brain death, engagement of the family in the brain death evaluation and helping them understand the meaning of patient movements that can be distressingly misinterpreted, and honesty when a definitive determination of brain death cannot be determined will not remove the misunderstanding or controversy surrounding a diagnosis of brain death but will serve to ensure its accurate and humane application.

Dr. Bowton is Professor, Section on Critical Care, Department of Anesthesiology, Wake Forest Baptist Health, Winston-Salem, North Carolina. Read previous Critical Care Commentaries online.