On occasion, a patient’s medical odyssey emerges from obscurity into the public domain. Jahi McMath was an unfortunate 13-y-old girl who suffered a cardiac arrest after surgery and was subsequently declared dead by neurologic criteria (hereafter referred to as brain death [BD]). Her family successfully petitioned the courts to prevent interruption of supportive care. She was maintained on a ventilator for 4.5 y until experiencing cardiopulmonary arrest in June 2018.
Because the profound and protracted legal arguments surrounding Jahi’s medical course resulted in extensive media coverage, many clinical details were disclosed in the public domain, which served as a nidus for editorials and reviews in the medical literature. An article on radionuclide evaluation of BD appeared in this journal in 2016, reviewing the initial course of Jahi McMath’s illness and discussing the role of scintigraphy in the determination of BD (1). Jahi’s entire medical records were released, including images from a radionuclide BD examination (2). This editorial updates the prior report by providing additional clinical history, radionuclide images and their analysis, and a discussion of controversy and questions engendered by this tragic case. Clinical information presented here is in the public domain, either in previously published literature or with permission granted by Jahi’s mother.
CHRONOLOGY
In 2013, Jahi McMath presented with symptoms of obstructive sleep apnea and underwent surgery to resect offending tonsillar tissues (chronology in Supplemental Table 1; supplemental materials are available at http://jnm.snmjournals.org). In the postoperative period, she bled into her airway, suffered a cardiac arrest, was resuscitated, and then was placed on a ventilator. On postoperative day 3, hospital physicians determined that she met the clinical criteria for BD (3); however, the family insisted that supportive therapy be maintained. Electroencephalogram studies at that time were isoelectric. On day 14, a 99mTc-bicisate study was performed, which supported the diagnosis of BD, and a court-appointed pediatric neurologist meticulously repeated the clinical determination of BD, independently confirming the diagnosis. By legal means, the family compelled the hospital to maintain life support until Jahi was released to her mother on day 27. She was then moved to New Jersey, where an exemption to BD can be invoked to accommodate personal religious beliefs, initially to a medical facility and subsequently to a private apartment. She remained on a ventilator until experiencing cardiovascular collapse 4.5 y after the initial BD pronouncement.
SCINTIGRAPHIC IMAGING AND FINDINGS
According to guidelines, ancillary studies are not required to establish the diagnosis of BD unless certain elements of the physical examination cannot be properly performed. Ancillary studies may still have a role to play for social reasons, such as allowing family members to better comprehend the diagnosis (3), which most aptly fits the current circumstance. Brain scintigraphy with lipophilic compounds may also play a separate role in prognosticating potential recovery in patients with catastrophic brain injury, even when not directly related to BD determination, and may be helpful to family and caregivers in decision making.
On day 14 of Jahi’s course, a perfusion study of the brain was performed after injection of 1.0 GBq of 99mTc-bicisate (Neurolite; Lantheus). Twenty scintigraphic imaging files were provided on an eFilm Lite disk (version 4.1.0; Merge Healthcare); these were provided without any editorial stipulations. Representative images are displayed in Figures 1–3; additional images are presented in the supplemental materials. Tomographic information was provided as secondary-capture images, not amenable to scrolling or to triangulation between planes, and lacking the original dynamic range of counts (though the windowing appears grossly adequate). These limitations should not be regarded as consequential because, according to guidelines, tomographic imaging is an optional component of BD studies. The examination was otherwise technically adequate, with performance closely conforming to published guidelines (4,5). No intracranial blood flow or parenchymal perfusion was visualized on dynamic, static, or tomographic phases of the examination.
99mTc-bicisate, 1.0 GBq, was injected via femoral central line with scalp tourniquet in place. Representative dynamic tomographic images were selected from series of 95 rapid sequential images acquired in anterior projection at 1- or 2-s intervals (timing details not specified in images or report). Activity is clearly seen within common carotid arteries; however, no intracranial flow is visualized.
Anterior and right lateral views of 99mTc-bicisate tomographic images obtained several minutes after injection demonstrate activity in extracranial tissues of face and skull. No activity is localized in cerebrum or cerebellum. Ant and A = anterior; Lat = lateral; P = posterior.
Representative 99mTc-bicisate tomographic images displayed in transaxial, coronal, and sagittal planes. No activity is visualized within calvarium. Additional images appear in supplemental materials. ANT = anterior; INF = inferior; LT = left; POST = posterior; RT = right; SUP = superior.
CONTROVERY REGARDING THE MEDICAL COURSE
Although inconsistency in BD determination has been documented in the literature and potentially could lead to a false-positive diagnosis, it is unlikely that both of Jahi’s clinical examinations were invalid, because of the enhanced expertise focused on her high-profile case. Her perfusion study was fully compliant with relevant guidelines, and no intracranial blood flow or parenchymal brain uptake was apparent.
Physicians who perform radionuclide BD studies should be aware of wide-ranging discussions in the clinical literature regarding Jahi McMath’s case (2,6). Initial conceptualization of BD frequently included rationalization that lack of central control and coordination precluded continued integration of the organism. Although this dogma was largely abandoned, it is still distinctly unusual for a BD patient to maintain homeostasis over a span of several years, regulate body temperature, develop signs of puberty, and experience several menstrual cycles, as Jahi did. Jahi’s course was also exceptional in that most patients for whom BD is declared will immediately proceed to organ donation or be removed from the ventilator, leaving no opportunity for subsequent observation; under these circumstances, the frequency of delayed return of function cannot be estimated. Despite this paradigm, occasional case reports have surfaced in which neurologic function is noted to return after pronouncement of BD (7), similar to our narrative. Several medical personnel, including a prominent pediatric neurologist, came to believe that Jahi could intermittently respond in a purposeful manner to verbal commands (such as “move your arm”) (2). Additionally, the MRI examination performed 10 mo after the initial cardiac arrest unexpectedly demonstrated large regions of grossly intact brain, including cortex, basal ganglia, thalamus, upper brain stem, and cerebellum, which could serve as a structural basis for intermittent consciousness (8). Return of minimal responsiveness, if present, would directly challenge the presumption that BD is definitive and permanent.
DISCUSSION
Along with broader concerns raised by Jahi McMath’s case, it is also opportune to consider several questions pertaining to nuclear medicine. Thresholds of minimal detectable perfusion have never been determined for scintigraphic studies using either lipophobic or lipophilic radiopharmaceuticals, though these parameters are basic to their interpretation. For lipophobic radiopharmaceuticals, marginal perfusion would certainly be difficult to appreciate on noisy 1- or 2-s dynamic images. For lipophilic compounds, the brain stem and small cortical regions adjacent to the calvarium represent areas where identification of minimal perfusion would be challenging, even with tomography. Coupled with uncertainty regarding the amount of blood flow required to maintain the structural integrity and function of neurologic tissue, it is difficult to assert that lack of visualization of blood flow on a scintigraphic study would predict complete necrosis of the brain or even guarantee absence of function. This concern seemed realized in Jahi’s case, with perfusion not being visible on the 99mTc-bicisate study yet large regions of cerebral cortex remaining intact, and according to some experts, a minimally conscious state emerged. Previous analysis of PET perfusion data in adults has suggested that the minimal regional cerebral blood flow necessary for preservation of tissue integrity is at least 15 mL of blood per 100 g of tissue per minute whereas that required for normal neurologic function is at least 19 mL of blood per 100 g of tissue per minute (9). Could this amount of perfusion have been present but not visible on the 99mTc-bicisate study? Conversely, was perfusion absent during scintigraphy, possibly because of a transient hypotensive episode, but subsequently returned? Our fundamental lack of knowledge regarding the behavior of the 99mTc-bicisate examination prevents us from arriving at definitive answers.
A related issue concerns estimating the specificity of scintigraphy for determination of BD. Characterization of specificity requires evaluation of the examination in a population of subjects similar to that possessing the condition in question but lacking the particular condition itself. The ancillary examination must therefore be studied in a cohort of patients with catastrophic brain injury but without complete loss of function. Such an evaluation has not occurred for many of the modalities, because this group of patients is simply not sent for ancillary study; if the clinical examination reveals any residual neurologic function, the blood flow study is deferred. This fundamental limitation has compromised validation studies in the nuclear medicine literature, which has included only a handful of appropriately non–brain-dead subjects. In one study using lipophobic methods, 10 appropriate non-BD patients were enrolled, but specificity for determining BD was a paltry 50% (10). Presumably, lipophilic radiopharmaceuticals would possess higher specificity, but this has not been rigorously demonstrated.
CONCLUSION
Jahi McMath’s tragic narrative highlights several unanswered questions relevant to the diagnosis of BD. For several decades, radionuclide techniques have served an important ancillary role in complementing incomplete clinical BD examinations though they remain incompletely characterized. Radionuclide BD examinations are invaluable when they demonstrate unexpected intracranial blood flow, preventing erroneous determination of BD; inferences when blood flow is not visualized remain more enigmatic. The threshold of blood flow required for visualization remains unknown. The accepted role of an ancillary test is to supplement but never replace the physical BD examination, as is congruent with an imperfect specificity of these ancillary tests. For this reason, Society of Nuclear Medicine and Molecular Imaging guidelines appropriately recommend that the impression of a positive study conclude “shows no evidence of brain perfusion” rather than “demonstrates BD” (4). The subjective formulation “shows no evidence” is also particularly appropriate in that it avoids the fundamentally unsubstantiated claim that “no blood flow is present.” A more thorough and robust understanding of these tests, combined with an effort to standardize their implementation, could serve to bolster their role in the determination of BD, especially in difficult or controversial circumstances.
DISCLOSURE
No potential conflict of interest relevant to this article was reported.
ACKNOWLEDGMENTS
Appreciation is extended to Dr. D. Alan Shewmon, who provided clinical materials for review; to Jahi’s mother, Nailah Winkfield, for authorizing publication of her daughter’s medical data; and to colleagues who have provided insightful review and comments.
Footnotes
Published online May. 19, 2022.
- © 2022 by the Society of Nuclear Medicine and Molecular Imaging.
REFERENCES
- Received for publication February 3, 2022.
- Revision received May 11, 2022.
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.