Effects on organ donation of transition from apnoeic-oxygenation to radioisotope brain scanning to diagnose brain death in children

The task of diagnosing brain death for the purposes of organ donation (DBD) and for guiding cessation of futile treatment for severe brain injury has traditionally been by clinical examination of the state of consciousness and by tests of brainstem function. However, the key test of brain-stem function, the apnoeic oxygenation test, may not always be possible or be confounded by numerous factors. When this occurs, ancillary tests of brain function such as assessment of brain perfusion with lipophilic radioisotopes are recommended by authorities on organ donation such as the Australian and New Zealand Intensive Care Society (ANZICS) [1]. A test of brain perfusion may be a potential routine but more stringent alternative to apnoeic-oxygenation in clinical testing to diagnose brain death. We hypothesised that use of brain scanning in lieu of apnoeic-oxygenation testing would make no difference to the rate of organ donation.


Introduction
The task of diagnosing brain death for the purposes of organ donation (DBD) and for guiding cessation of futile treatment for severe brain injury has traditionally been by clinical examination of the state of consciousness and by tests of brainstem function. However, the key test of brain-stem function, the apnoeic oxygenation test, may not always be possible or be confounded by numerous factors. When this occurs, ancillary tests of brain function such as assessment of brain perfusion with lipophilic radioisotopes are recommended by authorities on organ donation such as the Australian and New Zealand Intensive Care Society (ANZICS) [1]. A test of brain perfusion may be a potential routine but more stringent alternative to apnoeic-oxygenation in clinical testing to diagnose brain death. We hypothesised that use of brain scanning in lieu of apnoeic-oxygenation testing would make no difference to the rate of organ donation.
Lipophilic radionuclide scanning has been used since the late 1980's to diagnose brain death in adults and children [2,3] but the effect of this diagnostic mode on organ donation has not been studied. This study examines an institutional transition from the use of apnoeic-oxygenation testing as part of clinical testing of brain-stem function to the use of Single-Photon Emission Computed Tomography (SPECT) or radiopharmaceutical scan of the brain using Ceretec TM -Technetium Tc 99m Exametazine [(RR,SS)-4.8-diaza-3,6,6,9tetramethylundecane-2, 10-dione bisoxime, formerly known as Tc-99m HMPAO (hexamethylpropylene amine oxime)]. Ceretec TM is a diffusible lipophilic radiopharmaceutical which is rapidly taken up, retained and metabolised to a hydrophilic form by living brain tissue [4]. It attains a maximum concentration in the brain within one minute of intravenous injection followed by a loss of 15% over the ensuing two minutes. The radioactivity of the retained tracer subsequently decays [5].
The premise underlying use of radioisotope scanning to diagnose brain death is that complete absence of radioactivity in the whole brain after intravenous administration signifi es absence of perfusion and/or uptake of the tracer which represents cessation of whole brain function, as legally required in Australia to diagnose brain death.
The objective of the study was to determine the effects of transition from apnoeic-oxygenation to brain radioisotope scanning on organ donation after brain death (DBD) and circulatory death (DCD).

Materials and methods
This is a retrospective study of cohorts of children who were organ donors and/or had Ceretec TM brain perfusion scan during the period January 1989 to December 2018. The study was conducted by reviewing the records of the Imaging Department and of the Paediatric Intensive Care Unit (PICU) of the Royal Children's Hospital (RCH) and data provided by DonateLife Australia.
Organ Donation after Brain Death (DBD) commenced at RCH in 1989 while organ donation after circulatory death (DCD) commenced in 2011. Brain scanning with Technetium Tc 99m Exametazine (Ceretec TM ) was introduced into clinical practice to diagnose brain death in 2002 after it was found to be a reliable, safe and cost-effective [6,7] and has been used exclusively at RCH to diagnose brain death in lieu of apnoeic- The dose of the radiopharmaceutical was adjusted for weight according to the European Association of Nuclear Medicine paediatric dose guidelines [8]. Scanning was commenced at least 15 minutes after intravenous administration. SPECT images were obtained from a dual head camera with low energy collimation positioned as close as possible to the patient, 128 x 128 matrix, 20 frames/second, 2.81 degrees angular step, 128 steps, zoom 1.46. Images were reconstructed and displayed in three planes using colour scales with 1 pixel thick slices. Brain death was reported if there was no discernible uptake of tracer within brain structures and the brain stem.
Brain death by clinical testing was defi ned as in the ANZICS Statement 1 which essentially is unresponsive coma, the absence of all brain stem refl exes and the absence of respiratory centre function during induced oxygenated hypercapnia (i.e., apnoea) in the clinical setting in which these fi ndings are irreversible.
In addition, there must be defi nite clinical or neuro-imaging evidence of acute brain pathology consistent with the irreversible loss of neurological function. Brain death may also be determined by imaging that demonstrates the absence of intracranial blood fl ow.
We compared the number of donors in whom brain death was diagnosed by clinical testing including apnoeic-oxygenation with the number in whom brain death was diagnosed by Ceretec TM brain perfusion scanning. In addition, we analysed the role of brain perfusion scanning in organ donation after circulatory death. The differences in the rate of organ donation between eras were analysed for signifi cance with Chi-square test (Vassarstats.net). The project was approved by the RCH Human Research Ethics Committee.

PICU admissions and mortality
Throughout the period of study, 43,217 patients were admitted to PICU with 1930 deaths (crude mortality 4.5%). Annual admissions to PICU increased steadily (with minor variation) while the annual mortality decreased initially during this period but remained relatively static during the latter part. For example, annual deaths were approximately 100 and 110 respectively in 1989 and 1990 but declined to 40-60 and remained static over the period 2003-2018 ( Figure 1), resulting in a decreasing number of potential organ donors.

Brain perfusion scanning
Seventy-seven patients (average age 6.6 years, range 1 week-17 years) had scans over the 17-yr period 2002-2018. Scanning was performed to assist in the assessment of severity of brain injury or to diagnose brain death. The indications to perform brain perfusion scanning were unresponsive coma with unreactive pupils in all patients (but other brainstem refl exes were not recorded routinely) and all patients had had a preceding CT scan or MRI scan which had shown severe brain injury. The aetiology of brain injury is shown in Table 1.
Fifty-nine patients were diagnosed as brain dead on an initial Ceretec TM scan, that is, they had no cranial blood fl ow. Eighteen patients (23%) had some brain perfusion on initial scanning and were diagnosed as not brain dead. Of these latter patients, 9 had a second scan after decay of initial radioactivity at a mean interval of 34.7 hours (range 22.5 -71.5 hours). Six of these had absent perfusion on the second scan thus constituting 65 of 77 patients scanned (84%) as diagnosed brain dead, and 12 as not brain dead (16%) ( Figure 2). In one patient who had had 2 scans, residual radioactivity was absent 71.5 hours after the fi rst scan but was present 17.5 hours after a second scan thus precluding a third scan.

Organ donation
Overall, from 1989 to 2018, organs were donated by 95 of 1930 children (4.9%) after brain death or circulatory death at an average of 3.2/year or at a median percentage of 4.6(interquartile range, IQR, 1.8-5.1). Of these, 81 had been diagnosed with brain death and 14 had sustained circulatory death (Figures 3,4). Brain death was diagnosed in 52 donors by apnoeic-oxygenation and in 29 by brain perfusion scanning.

After brain death (DBD)
Organ donation may be considered in eras and according to the mode of diagnosing brain death ( Table 2):

Brain scanning:
Of 520 children who died in the period 2008-2018, when brain scanning was performed to diagnose brain death, 23(4.4%) donated organs at an average 2.1/year.    The annual median percentage DBD rate was 2.6(IQR, 1.9-5.5).
There was an increase of 17% in the donor percentage of deaths compared to the previous era (1989-2001) when apnoeicoxygenation was used to diagnose brain death but was not a signifi cant change (P=0.22).
Overall, there was no signifi cant difference in DBD during the three eras of apnoeic-oxygenation only, apnoeic-oxygenation or brain scanning and brain scanning only (P=0.52).

After circulatory death (DCD)
A

After brain (DBD) or circulatory death (DCD)
In the period 2011-2018, when brain scanning was used to diagnose brain death, the combined DBD and DCD rate was

Discussion
This study examined the infl uence of diagnosing brain death by radionuclide scanning in lieu of apnoeic-oxygenation on organ donation. It is not a comparative study of the two methods to diagnose brain death. A notable fi nding of our study was a decrease in the average annual DBD between the two periods when death was diagnosed by apnoeic-oxygenation (3.1/year) and by scanning ( On the other hand, there may have been negative factors on organ donation. Since some brain perfusion may be present when death is diagnosed by clinical absence of brain-stem refl exes [11][12][13] the complete absence of brain blood fl ow is a more stringent condition for diagnosing brain death. It is possible, therefore, that the diagnosis of brain death would be made less frequently by complete absence of perfusion on scanning than by clinical testing, resulting in less brain dead organ donors, but which we did not observe. Another notable fi nding of our study was a signifi cant increase in total organ donors when DBD or DCD were undertaken (9.5% of deaths) compared with the previous era when only DBD was undertaken (3.8% of deaths). Since There are numerous reasons why we abandoned a specifi c aspect of clinical testing, apnoeic-oxygenation, and adopted routine brain radionuclide perfusion scanning to assist in diagnosing brain death. Radionuclide brain perfusion has numerous advantages [17]. We perceive that brain perfusion scanning with lipophilic diffusible radionuclides (Technetium Tc 99m Exametazine or Technetium Tc 99m -ECD [18]) is more practical, simpler to perform, quicker and more easily Citation: Tibballs  accomplished with one test than clinical testing which must be repeated [1]. Scanning is objective, reliable, reproducible, noninvasive, not deleterious to organs and importantly, unlike clinical testing, is not infl uenced by sedative medication, muscle relaxants, poisons, temperature, metabolic and endocrine derangements, electrolyte disturbances, spinal cord injury and spinal automatism all of which may present in patients with severe brain injuries. Moreover, brain scanning can be applied to all ages, including newborns [19] and abolishes the delay in waiting for medication, which may infl uence neurological assessment, to be excreted or metabolised [20] and it does not need to be repeated if no perfusion is evident on an initial scan. However, blood pressure must be adequate to enable supply of the radionuclide to the brain unless opposed by intracranial hypertension. Although it is possible to observe an initial angiographic dynamic phase of perfusion in a dead brain [4] in the absence of intracranial hypertension, there will be no uptake of tracer and no radioactivity detected on delayed scanning. Other imaging techniques with computerised tomography or magnetic resonance, scanning with nonlipophilic radiopharmaceuticals and ultrasound techniques are less suitable for assessing brain function in the context of possible brain death because they can only yield initial angiographic data [4,13].
Importantly, images from brain scanning showing complete lack of tracer uptake are convincing evidence that a condition necessary for brain survival, i.e., brain blood fl ow or brain blood fl ow with uptake of tracer, is not present. It may be thus convincing not only for medical and nursing personnel that the affl icted child is dead and but also facilitate effective communication with family members [13,21] and is helpful in brain death donor counselling [22]. If limited tracer uptake and hence some blood fl ow is present, it reinforces clinical assessment and other imaging results which suggest the futility of continuing active treatment. In the absence of a portable camera, performance of scanning carries the risk of transportation out of the PICU, but in a well-developed system poses very little risk. In contrast, clinical testing of brain-stem refl exes is more diffi cult to explain, to demonstrate and may be less convincing to parents that their child is truly dead when they have a circulation and appear to be breathing when in fact mechanical ventilation is being provided.
According to some investigators the absence of cerebral blood fl ow is the mainstay of neurological determination of death [23] while brain perfusion scanning with a lipophilic radionuclide has become the "gold-test" [6,24,25]. Nonetheless, a diagnosis of brain death made by this technique does not always result in organ donation. In this study, 29 of 65 (45%) children diagnosed with brain death by scanning became organ donors compared with 69% of cases in a study of adults [25]. We were not able to analyse the rate of consent given by parents to organ donation from their child when brain death was diagnosed by clinical testing. We do not know if absent brain perfusion determined by scanning is more or is less convincing than clinical testing for parents that their child is dead. This aspect requires research.
If reassessment of brain perfusion is required with a second scan, it is necessary to wait suffi cient time until residual radioactivity in the brain from the fi rst scan has decayed. That interval is largely determined by the half-life of technetium Tc 99m which is 6 hours. After 24 hours, the fraction of remaining radioactivity is less than 0.063 [5]. In 9 of the patients in this cohort who presented for a second scan, no residual radioactivity from the fi rst scan was detected after a mean of 35 hours whereas radioactivity in one patient was present after 17.5 hours. Thus, a delay of at least 24 hours after a fi rst scan is probably required before a repeat scan is performed to avoid a spurious result attributable to residual radioactivity, but we cannot specify a minimum necessary delay based on this data. If performed to diagnose brain death, and to avoid delay required for a second scan, it is recommended that scanning be performed only when all the clinical conditions to diagnose brain death (except testing with apnoeic-oxygenation) are met. If performed too early a fi nding of cerebral perfusion may delay rather than expedite the diagnosis of brain death [26].
Although the presence of bilateral unreactive pupils is the best predictor of non-survival [27], the absence of the pupillary refl ex does not necessarily signify brain death [28]. In this study, some brain perfusion on brain scanning was observed in approximately one quarter of patients who had no pupillary refl exes and were subsequently defi ned as not brain dead.
In contrast to radionuclide scanning, the key test of brainstem function, indeed the sine qua non, the apnoeic-oxygenation test, has disadvantages. Even without confounding factors it can be argued to be unscientifi c and unreliable [29,30], principally because in infants and children [31][32][33][34][35], (otherwise fulfi lling the criteria for brain death), the onset of spontaneous ventilation has been observed at blood levels of partial pressures of carbon dioxide well above those at which brain death may be diagnosed (60mmHg in Australia, Canada, USA; 50mmHg in Britain) and moreover hypercarbia itself may depress brain function, which the ANZICS Statement concedes [1]. Thus the test may provide a false positive result, that is of diagnosing death when it is not present. Importantly, the test itself may be theoretically harmful -it may cause brain death since the test requires creation of hypercapnia which, in the functioning brain, is known to cause cerebral vasodilation and intracranial hypertension and consequently may cause brain death if not already established. The test has thus been criticised as a selffulfi lling one: that of potentially causing the condition, brain death, which it purports to diagnose [29, 30,36]. Moreover, the clinical testing, including apnoeic-oxygenation testing must be performed twice, at intervals, and by different examiners even if the fi rst test is diagnostic of brain death and may thus prolong the process and jeopardise the viability of organs.
Lastly, although the apnoeic-oxygenation test is widely used, it is not applied uniformly, is diffi cult to perform safely with frequent reports of hypotension and hypoxaemia [29,30].
On the other hand, the test has the advantage of permitting family members to observe in their affl icted child the lack of respiration [37], a universally recognised cardinal sign of death.
In our current practice, possibly unique, when a child shows signs consistent with brain death such as unresponsiveness, unreactive pupils or loss of other brain stem refl exes (apnoeic-Citation: Tibballs J, Raman S, Francis P (2021) Effects on organ donation of transition from apnoeic-oxygenation to radioisotope brain scanning to diagnose brain death in children. Arch Organ Transplant 6(1): 001-007. DOI: https://dx.doi.org/10.17352/2640-7973.000017 oxygenation is not tested), and who would be a suitable organ donor, a brain perfusion scan is performed. However, if any refl ex is present scanning is deferred to avoid foregoing the possibility of DBD. Organ procurement under the condition of brain death is then considered if perfusion is absent while organ procurement after circulatory death is considered if any perfusion remains but when further treatment would be futile.

Conclusion
In our institution, the routine adoption of radionuclide brain scanning to diagnose brain death has not adversely affected organ donation after brain death, which has remained constant. We suggest that lipophilic radionuclide brain scanning could be adopted as a routine test in lieu of apnoeic-oxygenation when other clinical examination including unresponsive coma, dilated unreactive pupils and radiological imaging are suggestive but not confi rmatory of the diagnosis of brain death, rather than be regarded as a confi rmatory test or ancillary test used to circumvent factors confounding clinical testing.