SEARCH

SEARCH BY CITATION

Keywords:

  • acute cerebral hemorrhage;
  • emergency treatment of stroke;
  • intracerebral hemorrhage;
  • organ donor;
  • transplantation

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. Disclosure of conflicts of interest
  10. References
  11. Supporting Information

Background and purpose

Intracerebral hemorrhage (ICH) has high acute mortality. The number of potential kidney and liver donors amongst deceased ICH patients was estimated to improve our institutional guidelines on acute care of neurocritical patients to increase organ donation.

Methods

A chart review was carried out by a multi-professional team for consecutive ICH patients admitted to the emergency department at Helsinki University Central Hospital and dying within 14 days between 2005 and 2010.

Results

In all, 955 patients had follow-up data, of whom 254 (27%) died within 14 days and eight ended up as organ donors. An additional 51 potentially suitable donors not different from actual donors were identified: nine suitable for kidney donation, 11 for liver and 31 for both. In 49/51 (96%) cases prognosis seemed non-existent and do-not-resuscitate orders were issued early, which led to refrainment from intensive care in 76.5%. These potential donors differed from those ICH patients surviving a whole year (n = 529) by male preponderance, more severe symptoms (median National Institutes of Health Stroke Scale 25 vs. 6 and Glasgow Coma Scale 7 vs. 15), larger hematoma volumes of 24.8 cm3 (vs. 6.7), and frequent finding of midline shift and intraventricular rupture of the hemorrhage in admission brain CT. Based on the results, our guidelines were revised towards more active treatment including mechanical ventilation for neurocritical patients at the emergency department for at least 48 h, resulting in an increase in organ donations in 2012.

Conclusions

A considerable number of ICH patients are potential organ donors if the evaluation takes place on arrival and organ donation is considered as part of usual end-of-life care.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. Disclosure of conflicts of interest
  10. References
  11. Supporting Information

Most organ donations are from patients with brain catastrophes including subarachnoid hemorrhage (SAH), traumatic brain injury and spontaneous intracerebral hemorrhage (ICH), causing over 80% of brain deaths [1]. A trend towards donation after cardiac death has been predicted if brain dead donations decrease in the future due to better prevention, a decrease in motor vehicle accidents and decreased mortality of cerebral bleedings [2]. Finnish multi-organ donors are almost exclusively heart-beating, i.e. brain dead donors [3], and the average number of harvested organs per donor is 4.1 [4]. Half of the kidney and liver transplants functioning at 1 year from transplantation are predicted to function for >20 years. After the transplantation surgery, the patient's quality of life is equal to that of the general population. However, there is a universal constant shortage of all transplantable solid organs.

The potential of organ donation in ICH patients is under-investigated [5, 6], especially outside the intensive care unit (ICU) setting. The requirement of ICU treatment was even incorporated into suggested criteria for imminent brain death of a potential donor as a deeply comatose patient who is mechanically ventilated and admitted to ICU due to irreversible catastrophic brain damage of known origin [7]. This definition leaves out patients who are not provided with ICU care. Controlled studies evaluating the impact of withdrawal of care on ICH prognostication have been proposed but not completed. The American Stroke Association guidelines from 2010 recommend active treatment for at least 2 days for all patients with ICH [8]. Emergency departments (EDs) and ICUs hold a key role in recognizing neurologically critically ill patients who might become potential donors. Organ procurement offices (OPOs) offer early consultation help on every possible organ donor to prevent losing organs suitable for transplantation.

The aim of the present study was to evaluate the potential of organ donation in our consecutive single-center series of ICH patients admitted to the medical ED. ICH patients admitted to our tertiary center are independent in daily activities and without major debilitating concomitant diseases.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. Disclosure of conflicts of interest
  10. References
  11. Supporting Information

A retrospective chart review was carried out of all consecutive patients with ICH treated from January 2005 to March 2010 at the Department of Neurology, Helsinki University Central Hospital (HUCH), with a catchment population of 1.5 million.

Standard protocol approvals, registrations

A detailed description of the study population data collection has been published previously [9]. This study has been approved by our institutional authorities. As a routine observational quality registry with no patient contact, consent for registration is not required by the Finnish legislation.

Patient cohort definition

Patient charts were screened for a diagnosis of ICH by the International Classification of Diseases, 10th Revision, code I61. The World Health Organization definition of stroke that differentiates between stroke ICH and non-stroke ICH was used [10]. Thus patients with primary subdural/epidural hematoma or traumatic ICH or hemorrhage due to a tumor were excluded. Stroke due to primary SAH with or without ICH and hemorrhagic transformation of a cerebral infarction with or without thrombolytic therapy are not initially ICH and were also excluded. The SAH and traumatic intracranial hematoma patients are primarily treated at the Department of Neurosurgery and are not included in the present study.

Description of potential organ donor

Currently there are very few absolute contraindications for organ donation: mainly severe infectious diseases contagious to the host including hepatitis B or C, HIV, tuberculosis, Creutzfeldt−Jacob disease or other prion disease, other systemic diseases affecting organ suitability, and malignancies except cerebral malignancies without systemic distribution. However, a rather conservative algorithm was developed by us (Fig. 1) with strict inclusion/exclusion criteria for identifying only those patients with a high likelihood of becoming donors. Therefore, patients older than 75 years and patients with cancer or other significant systemic or infectious disease were excluded. Diabetes mellitus combined with liver disease was considered an absolute exclusion criterion. Multiple diseases or conditions were also considered an absolute exclusion criterion. Additionally, severe infections, i.e. pneumonia and sepsis during hospitalization, led to exclusion as did uncontrolled hypotension during hospitalization. Patients with uncontrolled prior hypertension and patients with kidney failure or liver transaminase increase on any single measurement during in-hospital care were excluded. Finally, hospital charts of the remaining eligible patients were reviewed case-by-case by a panel of experts in anesthesiology, transplantation surgery and neurology.

image

Figure 1. Algorithm for detecting potential liver or kidney donors who were missed. (1)Cancer (n = 14): prostate (three); breast (two); bone (one); liver (one); maxillary (one); lymphoma (one); lung (two); sigmoid (one); melanoma (one); bladder (one). (2)Systemic diseases or medical conditions (n = 32): diabetes mellitus combined with liver disease and/or other concomitant conditions (four); rheumatoid arthritis (three; one became an actual donor); methicillin-resistant Staphylococcus aureus carrier (three); extended spectrum betalactamase carrier (one); endocarditis with vegetations (one); Eisenmenger syndrome (one); microscopic polyangiitis (one); arteriosclerosis obliterans (one); iv drug abuse (one); multiple concomitant diseases or conditions affecting kidney or liver function (n = 16) including diabetes (n = 10), advanced age, grave obesity, liver cirrhosis, hypertension, prior kidney transplant recipient (two). (3)Organ donation denied: by the patient (two); relative (two); religious or cultural reasons (three). (4)Severe in-hospital infections (n = 27): pneumonia (23); sepsis (three); sepsis and pneumonia (one). ALT, alanine aminotransferase (normal range 10–70 U/l male, 10–45 U/l female). *One patient who was excluded due to ‘Other disease’, i.e. rheumatoid arthritis, became an organ donor.

Download figure to PowerPoint

Data retrieval

Every patient had been examined by a neurologist and data were derived from patient charts and province-wide electronic hospital records and referral notes, as described earlier [9]. All brain scans were evaluated for the presence, location and size of ICH, the presence of midline shift, and the presence of an intraventricular bleeding component. Lesion volumes were estimated with the ABC/2 method [11].

The annual numbers of ICH donors between 2005 and 2010 and in 2012 were retrieved from the local OPO charts.

Statistical analyses

Distributions of the continuous variables were tested for normality. Univariate comparisons were performed with the Mann–Whitney U test for continuous variables and with the Pearson χ2 test for categorical variables. The data are given as median, interquartile range and range. Two-sided values of P < 0.05 were considered statistically significant. IBM SPSS 19.0 (SPSS Inc., New York, USA) was used for the analyses.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. Disclosure of conflicts of interest
  10. References
  11. Supporting Information

The cohort includes 1013 ICH occurrences in 980 patients. Twenty-five patients living outside the HUCH catchment area were lost to follow-up before 14 days due to prior discharge and four additional patients after 14 days. Of the 955 patients with follow-up data, 254 (27%) died within 14 days. These patients were evaluated for any contraindication for donating kidney or liver using a predefined exclusion algorithm (Fig. 1). Only eight patients (amongst them, one did not satisfy our strict criteria) had become organ donors. After running the predefined algorithm and re-evaluating the remaining cases in a consensus meeting, another medically fit 51 patients were identified who were considered as potential kidney and/or liver donors but who were mostly treated on a normal ward without considering organ donation on arrival or when transferred to the ward from the ED (Fig. 1).

There were altogether 20 patients in whom brain death was diagnosed during the study period and eight of them became actual donors. Organ harvesting amongst them occurred from none in 2005, from one patient in 2006, one in 2007, two in 2008, one in 2009 and from three patients in 2010. During 2005, organ donation was not considered from six brain dead patients. In later years, decision against donation was based on an OPO consultation (three), refusal by relatives (one), fungal growth in the blood (one) and the need for medico-legal autopsy (one).

The potential (n = 51) and actual (n = 8) donors did not differ in age, sex, cardiovascular risk factors, Glasgow Coma Scale, National Institutes of Health Stroke Scale or radiological characteristics on admission with the exception of larger hematoma volumes of 88.6 cm3 (86.9; 2.8–273.8) for actual donors vs. 24.8 (38.9; 0.8–128) (P = 0.001) (Table S1). A quarter of both groups were treated at the stroke unit (13/51 and 2/8; P = 0.976). A do-not-resuscitate (DNR) order was very frequently issued amongst the potential donors (96%) and less often amongst the actual donors (75%, P = 0.027). These potential donors have been detected retrospectively and were not identified as such during the actual hospital treatment period. In seven of the potential donors brain death diagnosis was made during hospital care. In six of them organ donation was not considered; in one patient donation was considered but deemed contraindicated by the treating physician due to concomitant hypertension and diabetes. For patients with poor prognosis the decision about refrainment of intensive care has to be made before transfer to a non-ICU bed. A DNR order before 24 h took place for 39 potential donors. The potential donors were less often treated in the ICU than actual donors (23.5% vs. 100%; P < 0.0001), despite the need for mechanical ventilation in 33/51 (64.7%). The median time from ICH to death was 1 day (range 0–3 days) for actual donors and 2 days for potential donors (range 0–14) (P = 0.196). Most deaths amongst the potential donors (n = 13, 25.5%) occurred during the second day and 68.6% of all deaths were within 3 days of ICH (Table 1).

Table 1. Time from ICH to death amongst potential donors
ICH to death (days)Number of deceasedPercentageCumulative percentage
0917.618
11325.543
2815.759
359.869
447.877
547.884
635.990
712.092
1023.996
1112.098
1412.0100
Total51100.0 

Compared with the other patients who died within 14 days (n = 195, i.e. 51 potential and eight actual donors subtracted from 254 deceased in Fig. 1), the potential donors (n = 51) were younger and they had previous cardiovascular conditions less often (Table 2). The same held true in comparison with other patients who died either within 30 days (35 patients more, i.e. n = 230) or within 90 days (n = 258) (data not shown). The radiological findings on the admission head CT scan did not differ between the potentially missed donors and other ICH patients who died within 14, 30 or 90 days (data not shown).

Table 2. Demographic characteristics, clinical scores and brain scan findings on admission of the potential donors, other deceased and the survivors
Characteristic Potential donors (n = 51)Other deceased (n = 195)P valueSurvived patients (n = 529)P valuea
  1. Median (interquartile range; range) or n (%). mRS, modified Rankin scale; GCS, Glasgow Coma Scale; NIHSS, National Institutes of Health Stroke Scale. aIn comparison with potential donors; bn = 190; cn = 517.

Age, years61 (12; 24–74)76 (15; 40–93)<0.000162 (18; 22–91)0.119
Male gender36 (71)115 (59)0.129283 (54)0.019
Hypertension26 (51)139 (71)0.006309 (58)0.305
Coronary artery disease1/48 (2)40/191 (21)0.00246/526 (9)0.107
Atrial fibrillation5/48 (10)52/192 (27)0.01537/523 (7)0.396
Heart failure1/48 (2)14/192 (7)0.18213/522 (3)0.862
Diabetes8 (16)32 (16)0.90160 (11)0.357
Dyslipidemia7/48 (15)36/193 (19)0.510106/525 (20)0.350
Prior mRS >24 (8)16 (8)0.93310 (2)0.008
GCS on admission7 (9; 3–15)8 (10; 3–15)0.46715 (2; 3–15)<0.0001
NIHSS on admission25 (19; 1–40)25 (18; 1–40)0.6446 (10; 0–40)<0.0001
ICH volume, cm324.8 (38.9; 0.8–128)34.8 (54.1; 0.3–174)b0.0906.8 (13.8; 0.3–172)c<0.0001
Brain stem ICH11 (22)29 (15)0.24921 (4)<0.0001
Intraventricular hemorrhage 38 (75)131 (67)0.315146 (28)<0.0001
Infratentorial ICH12 (24)43 (22)0.82265 (12)0.02
Midline shift31 (61)128/193 (66)0.460149 (28; n = 524)<0.0001

In comparison with the 529 ICH patients who survived at least 1 year throughout the study period, the missed potential donors (n = 51) had previous cardiovascular conditions equally often (Table 2). There were differences in the admission brain CT scan findings between the surviving patients and the potential donors (Table 2). The patients who survived were more often treated at the stroke unit (58.8%) than the missed potential donors (25.5%; P < 0.0001).

During spring 2012 a multi-professional task force was established in order to improve the detection of potential donors and self-audit the treatment path of organ donors. Members from all hospitals with an ED in the HUCH catchment area were invited. The task force was aimed to detect neurocritical patients as early as possible and continue intensive care with mechanical ventilation if needed for at least 1–2 days. It enabled the ICUs to prioritize these patients and hospital directors stated that ICU beds had to be found for potential donors. This resulted in eight donors with ICH with a similar age and gender distribution in 2012 compared with the total of eight during the years 2005–2010, despite no change in the total number of about 250 ICH admissions yearly. These donors died within 4 days.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. Disclosure of conflicts of interest
  10. References
  11. Supporting Information

Transplantation is the only sustainable treatment for patients with end-stage renal and liver failure, but the constant limited availability of organs restricts transplantations. ICH is associated with high acute mortality and therefore ICH patients with poor prognosis are a potentially significant source of organ transplants. Using strict criteria, more patients suitable for organ donation were identified. These potential donors did not differ from the actual donors in their previous cardiovascular history and could have been medically fit as organ donors, but they were more frequently issued with early DNR orders in the ED and only one-quarter was treated in an ICU. Thus, almost one in five of deceased ICH patients were or could have become donors.

The conversion of eight out of 59 theoretical donors (14%) is far less than the 51% conversion rate amongst mechanically ventilated patients reported in the whole of Finland and other European countries [4], in the USA (40%) [12], UK (40%–45%) [13, 14] and Germany (47%) [5]. The true conversion of eight actual donors amongst the 20 brain dead patients (40%) equals previous audits. In high-achieving organ donation countries such as Spain, it has been verified that a few steps are essential for improving the effectiveness of the referral of possible donors to the ICU [15]. These include the treatment of the neurocritical patient in the ICU, donation as a part of the ICU service portfolio, and an action protocol towards the identification of patients with severe brain damage. Refraining from early DNR decisions is crucial [16] although limitation of life sustaining therapy becomes necessary eventually if there is no evolution to brain death. At present there are on-going educational activities at HUCH, a donor coordinator has been hired, and donor teams have been established for all HUCH area hospitals. During the work of our task force and after changes in the treatment protocol, the number of organ donors started to increase significantly.

The shortage of organ donations is a global problem and the brain dead donor rates vary markedly amongst countries [17]. Finland was the first nation in which brain death was legally accepted as death of a person, making heart-beating donation possible [18]. The Finnish legislation on organ and tissue donation changed on 1 August 2010 introducing a presumed consent instead of the previous consenting of the patient when capable to do so or next-of-kin consent after brain death. So far constant donor numbers for years 2009 and 2010 have been recorded for Finland in the aIRODaT registry [3]. In our algorithm on potential donors, only few were excluded because of denial, but a majority of brain dead patients were not considered as donors. Thus, a large group of donors was missed because of a poor in-hospital process of potential donor recognition. The presence of an in-house coordinator from an OPO increases organ donations [19-21] and this should be introduced in every ED.

Our study has certain limitations and strengths. First, this is a retrospective study with a relatively low 14-day mortality (n = 254). Secondly, in order to gain the whole potential of transplantations, a maximum of 2 weeks of ICU care time was hypothesized, yet every extra day yielded one to three potential donors after the third in-hospital day (Table 1). Thirdly, a potential donor is not equal to a successful organ transplantation and it is impossible to predict how many of the missed potential donors would have ended up brain dead. The study population did not include patients with SAH or traumatic ICH. Our study also has strengths. Very detailed data on our individual patients in a consecutive ICH population are available. Further, the hospital chart of each potential case was reviewed by a panel of experts to simulate a real-life situation. Also our algorithm was modeled using very strict criteria which probably rather underestimated than overestimated the number of potential donors. Some of the excluded patients were not eligible for donations due to severe infections or hypotension. These in-hospital complications could be avoided by more intensive care, thus leading to more potential donors.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. Disclosure of conflicts of interest
  10. References
  11. Supporting Information

A fairly large number of ICH patients dying in the acute phase may become actual organ donors if adequate in-hospital guidelines, education and resources together with streamlined processes are applied as was the case at our teaching hospital. The key elements were recognizing the critically ill neurological patient, securing the intensive care, and considering organ donation as part of usual end-of-life care. Of utmost importance is the very delicate handling required. Most relatives are shocked by the acute deterioration. If non-prognosis is expected very sensitive guidance and information to the relatives by an experienced physician is necessary to provide an understanding and minimize emotional pain in this hyperacute setting (<2 days after onset). Whereas only eight ICH patients had become donors in the six previous years, the same number of ICH patients became donors in one single year in 2012, making up 40% of all donations in our hospital that year.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. Disclosure of conflicts of interest
  10. References
  11. Supporting Information

This study was funded by the HUCH Research Funds (TYH2013313), Academy of Finland (AM), Finnish Medical Foundation (TS, AM, JP), Biomedicum Helsinki Foundation (AM) and Sigrid Juselius Foundation (AM, TT).

Disclosure of conflicts of interest

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. Disclosure of conflicts of interest
  10. References
  11. Supporting Information

AK, A-MK, KR, SM, AM, JP, DS and HI have no disclosures. TS: grant outside the submitted work from the Finnish Medical Foundation; editorial board for Experimental and Translational Stroke Medicine. AM: consultancy fee Boehringer-Ingelheim outside the submitted work. TT: advisory board Boehringer Ingelheim; consultancy fees as national coordinator; principal investigator: H Lundbeck A/S, Sanofi Aventis, PhotoThera Inc., Mitsubishi Pharma (steering committee chairman), BrainsGate Inc.; grants outside the present work from multiple foundations; payments for lectures from multiple organizations. MK: board member Mitsubishi Pharma Europe Ltd, Lundbeck A/S; consultancy fee Mitsubishi Pharma Europe Ltd, Lundbeck A/S; lecture payment Siemens AG.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. Disclosure of conflicts of interest
  10. References
  11. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. Disclosure of conflicts of interest
  10. References
  11. Supporting Information
FilenameFormatSizeDescription
ene12288-sup-0001-TableS1.docWord document43KTable S1. Demographic characteristics, clinical scores, and brain scan findings on admission of actual and potential donors.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.