Swiss Monitoring of Potential Organ Donors ( SwissPOD ) : a prospective 12-month cohort study of all adult ICU deaths in Switzerland

BACKGROUND: The Swiss Monitoring of Potential Organ Donors (SwissPOD) was initiated to investigate the causes of the overall low organ donation rate in Switzerland. The objective of our study was an assessment of the donation after brain death (DBD) process in Swiss adult intensive care units (ICUs), and to provide an overview of the donation efficiency as well as of the reasons for non-donation. METHODS: SwissPOD is a prospective cohort study of all deaths in Swiss ICUs and accident and emergency departments. This study is an analysis of SwissPOD data of all patients who deceased in an adult ICU between 1 September 2011 and 31 August 2012. RESULTS: Out of 3,667 patients who died in one of the 79 adult ICUs participating in SwissPOD, 1,204 were possible, 198 potential, 133 eligible, and 94 utilised DBD donors. The consent rate was 48.0% and the conversion rate 47.5%. In 80.0% of cases, the requests for donation took place before brain death was diagnosed, resulting in a similar proportion of consents and objections as when requests were made after brain death diagnosis.


Introduction
Switzerland's post mortem organ donation rate is one of the lowest in Europe [1].During the last decade, it remained relatively stable with an average of 12.2 donors per million of population (pmp) and year.In the meantime, the number of patients with end-stage organ disease waiting for a transplant almost doubled [2].As a consequence, the lack of donor organs severely impacts the patients' hope for a timely, often lifesaving, transplantation.Moreover, a prolonged waiting time often entails comorbidities in patients (which otherwise might have been prevented), which generally impairs the early outcome.Organ shortage, however, is not unique to Switzerland, as the lack of organs for transplantation and growing waiting lists are issues prevalent in most countries worldwide [3].Two main determinants are often referred to as influencing a country's donation rate, namely the pool of potential donors and consent rate to donation.The number of potential donors seems to be linked to the population age structure, mortality rates from cerebrovascular accident and traumatic brain injury, as well as the availability of intensive care unit (ICU) beds and neurosurgical facilities [4][5][6][7][8].Various studies have shown that consent to donation might be influenced by legislation (explicit or presumed consent), religious beliefs, cultural influences, or the attitude towards and awareness of organ donation in both healthcare professionals and the public [9][10][11][12][13][14][15][16][17][18][19][20][21].Studies in numerous countries have tried to identify their potential for organ donation and to gain insight into what actions would help to increase the donation rate [5,7,8,12,[22][23][24][25][26][27][28].In addition, targeted initiatives to increase donation rates have been undertaken in some countries showing considerable effect [29][30][31][32][33][34][35][36][37].In this context, the Swiss Monitoring of Potential Organ Donors (SwissPOD) was initiated by the Federal Office of Public Health as well as the Swiss university hospitals and transplant centres to investigate the causes of the overall low donation rate in Switzerland.In 2010, Swisstransplant, the Swiss National Foundation for organ donation and transplantation, and the Comité National du Don d'Organes (CNDO) were commissioned with developing the SwissPOD framework and subsequently performing this national study.We present our analysis on deaths in adult ICUs which represented approximately 80% of all deaths registered in the data base.The goal of our study was an assessment of the donation after brain death (DBD) process in Swiss adult ICUs, and to provide an overview of the donation efficiency as well as of the reasons for nondonation.

Methods
SwissPOD was designed as a prospective cohort study of all deaths in Swiss ICUs and accident and emergency departments (A&E).Study sites included all 76 hospitals with an ICU recognised by the Swiss Society of Intensive Care Medicine (SSICM) and their associated A&E.Following approval by all cantonal ethics committees and the Eidgenössische Expertenkommission für das Berufsgeheimnis in der medizinischen Forschung (Federal ethics committee), patient data were collected between 1 September 2011 and 31 August 2012.In SwissPOD all patients who died in an ICU or A&E were registered.All deaths under the age of 44 weeks gestation, and all patients who in life refused to participate in a clinical study were excluded.SwissPOD data were collected using an audit tool which was designed on a hierarchical basis with a series of forms, following the process of organ donation from deceased persons as described in the "Critical Pathway of Deceased Donation" [38].Patient data were gathered and entered into the web-based system database by the local donor coordinator in each hospital.Data monitors at Swisstransplant, validated and archived each form with any queries being resolved directly with the person who completed the form.Treating clinicians were interviewed if the information in the medical record was not clear.Patient data included basic demographic information, detailed information on the causes of brain injury, and medical suitability for organ donation.Further data were collected on brain death testing and diagnosis, whether organ donation was considered, the process of obtaining consent from next of kin (NOK), and finally regarding whether or not organ donation took place.Information on whether organs were offered for transplantation, and if not why, were extracted from the Swiss Organ Allocation System (SOAS).The presented data include SwissPOD data of patients who died in adult ICU only (≈80% of records).Patients who died in paediatric ICUs or A&Es (≈20%) were excluded from the analysis.Patients who were considered for donation after circulatory death (DCD) were not excluded from the analysis.However, they were counted as a reason for the DBD process to stop, as this study focuses on donation after brain death.This explains why 6 utilised DCD donors (Maastricht category III) are not included in the total of utilised donors.The primary endpoint of this study was the number of possible, potential, eligible, and utilised DBD donors.Secondary endpoints included reasons for not progressing in the donation process, donation efficiency (consent rate, conversion rate), and point in time of NOK approaches.

Definitions
Conforming with the definitions in the "Critical Pathway of Deceased Donation" [38], possible endpoints in the DBD process are possible donor (POSD; a patient with a devastating brain lesion or circulatory failure, apparently medically suitable for organ donation), potential donor (POTD; a patient whose clinical conditions are suspected to fulfil brain death criteria), eligible donor (ELID; a patient medically suitable for donation who has been declared dead based on neurological criteria as stipulated by the Swiss Academy of Medical Sciences), and utilised donor (UTID; an ELID with consent for organ donation from whom at least one organ was transplanted).Absolute contraindications to organ donation as stipulated by the Swiss Transplantation Ordinance are: coma of an unknown origin, unresolved systemic infection or infections from an unknown origin, suspicion or risk of prion disease, suspicion of rabies, degenerative diseases of the nervous system from an unknown origin, and malignancy or <5 year history of treated malignancy (with the exception of carcinoma in situ, primary central nervous system tumours that rarely metastasise outside the nervous system, or low-grade skin tumours with little metastatic capacity such as basocellular carcinoma) [39].Donation efficiency is assessed by calculating the conversion rate (the number of UTID expressed as a percentage of the number of POTD) and the consent rate (the number of consents to organ donation expressed as a percentage of the total consents and objections).In cases where a patient had a donor card, and the NOK were also asked for consent, only one answer was counted.If the answers were diverging, an objection would always override any consent.This holds also if the patient and/or the NOK consented to donation, but there was a formal objection to donation (e.g., by the coroner).As required by law, patients who neither had a donor card nor NOK were excluded from the donation process.

Results
During the one-year study period, 3,667 patients died in one of the 79 adult ICUs participating in SwissPOD.Of these, 1,204 were POSD, 198 POTD, 133 ELID, and 94 UTID.These four groups were analysed for secondary outcomes and characteristics (table 1).Due to the hierarchical structure of the donation process, patient data of each step are included in the data shown for the previous step (i.e., data of POSD are included in the adult ICU deaths group, data of POTD are included in the POSD group, etc.).Percentages in each column refer to the total number of patients in the respective group.
Table 1 shows that, of all ICU adult deaths (n = 3,667), 39.4% were female (n = 1,444) and 60.6% male (n = 2,223).The mean age was 69.8 ± 14.0 years, and average ICU stay was 5.4 ± 9.2 days.An absolute contraindication to organ donation was present in 33.5% of cases (n = 1,227).Of all subjects, 67.3% were ventilated (n = 2,466), and 22.8% were never ventilated (n = 836).Among those who were ventilated, 1,250 had a neurological pathology (34.1% of all deaths, 50.7% of subjects ventilated).In the group of subjects who were never ventilated, 120 had a neurological pathology (3.3% of all deaths, 14.4% of subjects never ventilated).62.6% (n = 2,297) of patients died of a pathology with non-primary cerebral cause.The principal causes of brain injuries were anoxia/hypoxia or cardiac arrest (20.6%, n = 757), intracranial haemorrhage/ ischaemia ( 12  the POSD step, the subjects in the POTD group were on average 10.4 years younger (54. 1 visualises the reduction in the potential for organ donation, from all deaths to UTID.It shows that approximately two thirds of all ICU adult deaths (2,463/3,667) were not considered to be POSD.Except for three deaths who had neither a donor card nor next of kin (which stops the donation process by legal requirement), all deaths which did not qualify for POSD did so because of medical reasons (table 2).On the POSD step, 54.0% of subjects were not expected to fulfil the brain death criteria, being the main reason for the donation process to stop.On the POTD and the ELID steps, objection to donation by the patients (as expressed in the donor card) or their NOK was the most frequent reason for the donation process being stopped (in 50.8% and 74.4% of cases on the respective step).On the POTD step (n = 198), 78.8% of subjects had no donor card (n = 156).Of the 42 patients who had a donor card, 51.1% wanted a person of trust to make the decision (n = 24), 31.0%consented to any donation (n = 13), 7.1% consented to organ donation only (n = 3), and 4.8% objected to any donation (n = 2).As shown in table 2 (POTD column), the comprehensive number of objections by donor card and by the next of kin was actually higher (n = 33), and in 5 cases, the patients had neither a donor card nor next of kin which, by law, excludes them form donation.Consent to organ donation was sought in a total of 250 cases, resulting in 120 consents and 130 objections.This is equivalent to an overall 48.0% consent rate (the number of consents to organ donation expressed as a percentage of the total of consents and objections).As shown in figure 2, 80.0% of requests were made before brain death diagnosis was completed and only 20.0% after.The proportion of consents and objections was very similar regardless whether the requests took place before or after brain death was diagnosed.The conversion rate (the number of UTID expressed as the proportion of the number of POTD) was 47.5%.

Discussion
Our analysis of one-year SwissPOD data showed that out of 3,667 adult ICU deaths, there were 94 eligible consented donors from whom at least one organ was transplanted.With 94 UTID of 198 POTD, the conversion rate was 47.5% of our study cohort.One of the major impediments to achieve a higher conversion rate was the relatively large proportion of patients or NOK objecting to donation.Our evaluation showed that consent was given in only 48.0% of cases where permission for organ donation was sought.The 47.5% conversion rate is comparable with the rates in other European countries, such as France (47.1%), Germany (47.0%), the United Kingdom (45.0%), or Belgium (44.3%) [5,22,23].It is substantially higher than conversion rates published in a Dutch study (30.0%) as well as for Poland (30.5%) [22,26].Two countries, Finland and Spain, reported conversion rates of more than 50 percent (51.4% and 54.6%, respectively) [22,40].This implies that Swiss *Among the medical reasons for the losses, an absolute contraindication to donation is prioritised (i.e., patients with an absolute contraindication may also have been excluded from the donation process for other medical reasons; however, they are only counted in the absolute contraindication category).† Objection to organ donation is prioritised over other reasons that may have resulted in an exclusion of a patient from the donation process.Swiss law requires explicit consent to donation, therefore the NOK approach is a crucial moment in the donation process.It is well documented that there are various factors that may influence patient and NOK decisions.Among these are the timing and setting of request, national legislation (explicit vs presumed consent), public attitude towards and awareness of organ donation and transplantation (including trust in the health care system and the validity of brain death), the level of education, as well as ethnic or cultural differences and religious beliefs [5, 7, 10, 11, 21, 41-54].Our results -more than half of all patients or NOK objected to organ donation -confirmed that obtaining consent clearly is a hurdle in the donation process.Data published by the Council of Europe reveal that in 2011, refusal rates (the number of families refusing consent expressed as a percentage of number of interviews) ranged from 9.3% in Poland to 52.4% in the Netherlands, with Spain (15.9%),Italy (28.7%), and the United Kingdom (43.4%) in-between [55].Remarkably, data from 28,977 audited ICU and A&E deaths in the UK in 2011/12 showed a major drop in the refusal rate to 34.0% for DBD [56].Refusal rates reported by Roels et al. range from 14.6% (Finland) to 32.7% (France), with Belgium (20.3%), and Poland (23.4%) in-between [22].In a review of published data from a selection of European countries and the USA, Jansen et al. found refusal rates varying from 10.5% (Belgium and Finland) to 59.0% (the Netherlands), with Italy (29.0%),France (34.3%), the UK (41.0%), Germany (42.9%), and USA (46.0%) in-between [57].As consent rates seem dependent on a multitude of factors, improving modifiable components such as the hospital staff being aware of and committed to organ donation, a straightforward communication of the subject, or the quality of interaction with the caregiver appear to be the best way to enhance consent rates.For this reason, the Federal Office of Public Health has defined four action fields to foster organ donation in Switzerland.The action fields cover the training of medical staff, quality and process management, structural and finance improvements in hospitals, and information campaigns aimed at the general public.Except for the last-mentioned, these action fields have been delegated to the CNDO and Swisstransplant for execution.In the hospitals, several measures to establish national standards in the donation process on the educational, structural and process levels are in the implementation phase.Similar to the consent rate, which is subject to heterogeneous influences, several factors may influence the potential for organ donation.Such variables include the population age structure, the availability of ICU beds and/or neurosurgical facilities, as well as mortality rates from cerebrovascular accident (CVA) or traumatic brain injury [4][5][6]58].In our study, the main causes of brain injury on the POTD step were CVA (56.1%), trauma (17.6%), and anoxia/hypoxia or cardiac arrest (24.2%).Even though there is some variability among the data published from different countries, the general picture of the proportions is similar.A study that evaluated the donor potential in the north-east region of Germany found that out of 2,019 deaths with primary or secondary brain damage, the causes were nontraumatic intracranial haemorrhage (51.0%), craniocerebral injury (22.1%), ischaemia (11.9%), and hypoxic damage (10.4%) [23].In a review of medical records for multiple-organ donors in Belgium, the principal causes of brain death were cerebrovascular disease (46.0%), craniocerebral trauma (38.0%), and hypoxia (15.0%) [59].Results from the audit of potential organ donors in the Republic of Ireland showed that the causes of death in patients diagnosed with brain stem death were intracranial haemorrhage (54.1%), traumatic brain injury (28.7%), and hypoxic brain injury (12.1%) [24].In a large US study, the main causes of death were stroke (40.6%), head trauma (36.0%), and anoxia (20.1%) [60].
In order to evaluate the potential for organ donation, two studies compared mortality rates from CVA and traffic accidents with actual donation rates in different countries [7,61].Both studies found considerable differences between the mortality rates in the countries included in their analyses.Since in our study, 83.0% of the UTID died of either CVA (62.8%) or traumatic brain injury (20.2%), we wanted to compare mortality rates from these diagnoses with donation rates in selected European countries.Assuming that traumatic brain injuries were mostly resulting from traffic accidents, we extracted CVA and traffic accident mortality rates from the Eurostat statistics database (ICD-10 I60-69 and V01-99; age standardised, 2010 data or nearest year for EU 15 countries and Switzerland) [62].The donation rates for these selected countries were taken from the International Registry in Organ Donation and Transplantation (IRODaT) [63].Figure 3

Study strengths and limitations
This study is the first comprehensive, nationwide assessment of the donation process and its outcomes for patients who died in an adult ICU.In view of the fact that the detection and referral of potential donors is required by law, SwissPOD had a 100% participation rate from the ICUs accredited by the SSICM.Since our assessment was based on the first year of SwissPOD data, we were not able to perform any benchmarking, and consequently refrained from doing statistical analyses, except for descriptive statistics.Furthermore, one should keep in mind that the number of patients included in this study represents only approximately 6% of all deaths in Switzerland during the investigation period.Thus, there may exist an additional potential for DBD or donation after circulatory death from paediatric ICU deaths, from patients who died in A&E, intermediate care units, on general hospital wards, or out-of-hospital.However, we assume that the additional potential of organ donors, especially in the last patient group, is relatively small.

Conclusions
Our analysis of SwissPOD data showed that despite the undeniably low donation rate, Swiss adult ICUs are generally doing an excellent job in considering the option of donation at end of life care.This is reflected by the fact that the Swiss conversion rate is comparable to data published from major European countries such as France, Germany, or the United Kingdom.This finding is supported by an international comparison of CVA and traffic accident mortality rates with donation rates.It showed that the Swiss conversion value is comparable with countries that have lower refusal rates as well as higher donation rates.Our study also showed that in Switzerland, the rate of NOK refusal to organ donation is among the highest in Europe.Unsurprisingly, this fact clearly has a negative impact on the donation rate.The reasons for the high percentage of patients and NOK refusing consent to organ donation require further study.From our findings we conclude that optimising the NOK approach seems to be the most effective means of increasing the post mortem organ donation rate in Switzerland.

Figure 1 2
Figure 1 Losses by step in the donation process.ELID = eligible donor; POSD = possible donor; POTD = potential donor; UTID = utilised donor ELID = eligible donor; ICU = intensive care unit; POSD = possible donor; POTD = potential donor; UTID = utilised donor Original article Swiss Med Wkly.2014;144:w14045 Swiss Medical Weekly • PDF of the online version • www.smw.chadult ICUs are predominantly doing an excellent job in the detection and referral of potential donors.

Figure 3 CVA
Figure 3 CVA and traffic accident mortality rates vs donation rates in EU 15 countries and Switzerland.Source: Eurostat database, © European Union, 1995-2012; the International Registry in Organ Donation and Transplantation (IRODaT).pmp = per million of population Figures (large format)

Figure 2
Figure 2Consent vs objection to donation.

Table 1 :
Demographic and clinical data.

Table 2 :
Summary of the reasons why the DBD process stopped.
CVA and traffic mortality rate (292 pmp).Average values (EU 15 countries plus Switzerland) are 462 pmp combined mortality, and 17.3 pmp deceased donors.In the diagram, the "x"-mark indicates the conversion value (numerals displayed on the right side of the mark).It is calculated from the donation rate divided by the total mortality rate, multiplied by 1000.Example for the calculation of the value for Switzerland: 12.6 / (252 + 40) x 1000 = 43.The Swiss conversion value is slightly above the average, being 40.In a ±5 range from the average, it is similar to Italy (44), displays a comparison of the CVA and traffic accident mortality rates with the donation rate (all rates are shown as pmp).It shows that Switzerland has the lowest Swiss Medical Weekly • PDF of the online version • www.smw.chcombined