Accepted Manuscript Deceased Organ Donor Management: Does Hospital Volume Matter? Madhukar S. Patel, MD, MBA, ScM, Jahan Mohebali, MD, MPH, Mitchell Sally, MD, FACS, Tahnee Groat, MPH, Parsia A. Vagefi, MD, FACS, David C. Chang, PhD, MPH, MBA, Darren J. Malinoski, MD, FACS PII:
S1072-7515(16)31693-3
DOI:
10.1016/j.jamcollsurg.2016.12.004
Reference:
ACS 8555
To appear in:
Journal of the American College of Surgeons
Received Date: 16 September 2016 Revised Date:
3 December 2016
Accepted Date: 5 December 2016
Please cite this article as: Patel MS, Mohebali J, Sally M, Groat T, Vagefi PA, Chang DC, Malinoski DJ, Deceased Organ Donor Management: Does Hospital Volume Matter?, Journal of the American College of Surgeons (2017), doi: 10.1016/j.jamcollsurg.2016.12.004. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Deceased Organ Donor Management: Does Hospital Volume Matter?
Madhukar S. Patel, MD, MBA, ScM1, Jahan Mohebali, MD, MPH1, Mitchell Sally MD, FACS2,3,
RI PT
Tahnee Groat MPH2, Parsia A. Vagefi MD, FACS1, David C. Chang PhD, MPH, MBA1 , Darren J. Malinoski, MD, FACS2,3
Department of Surgery, Massachusetts General Hospital, Boston, MA
2
Surgical Critical Care Section, Portland Veterans Affairs Medical Center, Portland, OR
3
Department of Surgery, Oregon Health & Science University, Portland, OR
M AN U
SC
1
Disclosure Information: Nothing to disclose.
Disclosures outside the scope of this work: Dr Malinoski received consulting fees and travel
TE D
reimbursement from Symic.
Support: Dr Malinoski received grant funding from the Laura and John Arnold Foundation. This work was supported in part by Grant No. R380T22183 from the Health Resources and Services
EP
Administration, US Department of Health and Human Services and contract 234-2005-37011C. Disclaimer: The content is the responsibility of the authors alone and does not necessarily reflect
AC C
the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the US Government. Presented at the American Transplant Congress, Boston, MA, June 2016.
Correspondence address: Darren J. Malinoski, MD, FACS Portland VA Medical Center PO Box 1034 / P3ANES 1
ACCEPTED MANUSCRIPT
Running Title: Donor Hospital Volume Matters
SC
Keywords: deceased donor; donor management; hospital volume
BMI: body mass index
M AN U
Abbreviations
DCDD: donors after circulatory determination of death DMG: donor management goal
DNDD: donors after neurologic determination of death
TE D
DSA: donor service area ECD: expanded criteria donors
IRB: institutional review board
EP
OPO: organ procurement organization OR: odds ratio
RI PT
Portland, OR 97207 503-220-8262 ext 55613 Fax: 503-721-7859
[email protected] [email protected]
AC C
OTPD: organs transplanted per donor SCD: standard criteria donor SD: standard deviation
UNOS: United Network for Organ Sharing VA: Veterans Affairs
2
ACCEPTED MANUSCRIPT
Abstract Background: Identification of strategies to improve organ donor utilization remains imperative. Despite the association between hospital volume and outcomes for many common disease
RI PT
processes, there have been no studies to assess the impact of organ donor hospital volume on organ yield.
Study Design: A prospective observational study of all deceased organ donors managed by ten
SC
OPOs across UNOS Regions 4, 5, and 6 was conducted from February 2012-June 2015. In order to study the impact of hospital volume on organ yield, each donor was placed into a hospital
M AN U
volume quartile based on the number of donors managed by their hospital. Stepwise logistic regression was used to identify the independent effect of hospital volume on the primary outcome measure of having ≥ 4 organs transplanted per donor (OTPD). Results: Data from 4,427 donors across 384 hospitals were collected and hospitals were assigned
TE D
quartiles based on their volume of deceased donors. Hospitals managed an average of 3.3 ± 5.2 donors per hospital per year. After adjusting for age, ethnicity, donor type, blood type, body mass index, creatinine, and OPO/DSA, being managed in hospitals within the highest volume
EP
quartile remained a positive independent predictor of ≥ 4 OTPD (OR 1.52[1.29-1.79], P<0.001). Conclusions: Deceased organ donor hospital volume impacts organ yield, with the highest
AC C
volume centers being 52% more likely to achieve ≥ 4 OTPD. Efforts should be made to share practices from these higher volume centers and consideration should be given to centralization of donor care.
3
ACCEPTED MANUSCRIPT
Introduction Over the past two decades, hospital volume has been the focus of numerous studies with systematic reviews of the available literature suggesting high volume to be associated with
RI PT
improved outcomes(1, 2). Notwithstanding advancements in surgical care and an increased focus on quality improvement over the ensuing years, recent repeat analysis in the modern era
indicates that there is a strong inverse relationship between hospital volume and mortality for
SC
patients undergoing surgery, suggesting a continued opportunity for improving systems of care (3, 4). This research, which evaluates the impact of hospital level factors on patient outcomes,
M AN U
has largely been enabled by the continued development of standardized approaches for gathering and analyzing data across institutions. Despite the significant amount of investigation that continues to be done on delineating the volume-outcome relationship in both medical and surgical patients, however, there has yet to be a study to assess the impact of organ donor
TE D
hospital volume on organ yield.
Most recently, strategies to improve deceased donor organ utilization have largely focused on standardization and optimization of practices that guide the management of the
EP
individual donor (5-9). Although implementation of these approaches has been shown to increase the number of organs transplanted per donor (OTPD), as well as the quality of grafts available
AC C
for transplantation, the impact of hospital level factors on outcomes remains unknown (6-8, 10, 11). Thus, our objective was to determine the impact of donor hospital volume on OTPD.
Methods
Study Design
4
ACCEPTED MANUSCRIPT
A prospective, observational study of all donors after neurologic determination of death (DNDD) and donors after circulatory determination of death (DCDD) from 10 organ procurement organizations (OPOs) in United Network for Organ Sharing (UNOS) Regions 4, 5,
RI PT
and 6 (covering Oregon, California, Nevada, Utah, New Mexico, Arizona, and Texas), was
performed from February 2012 to June 2015. Among DNDDs, standard criteria donors (SCDs) as well as expanded criteria donors (ECDs) were included. ECDs were donors who were either ≥
SC
60 years old or donors who were 50-59 years old and had at least two of the following:
hypertension, terminal serum creatinine > 1.5 mg/dl, or death caused by a cerebrovascular
M AN U
accident. Although ECD was originally developed as a set of donor criteria that helped identify increased risk of graft failure in renal allograft recipients, it has more generally been used for the classification of marginal deceased organ donors(12). Each OPO had its own process in place for obtaining authorization for donation, and in the majority of cases the OPOs included in this study
TE D
reported that they were responsible for approaching the family. In situations where members from the donor hospital were involved in the process, they were accompanied by trained specialist staff. Of note, OPO/DSA #9 entered the study in November 2013 and OPO/DSA #8 in
EP
January 2014. With regard to Institutional Review Board (IRB) approval, this study was reviewed at the Veterans Affairs (VA) Portland Health Care System IRB and was determined to
AC C
represent non-human subject research. Data collection and outcome measures Donor demographics, blood type, cause of death, OPO and donor service area (DSA), as
well as creatinine prior to procurement were collected prospectively through use of the UNOS Donor Management Goals (DMG) Registry Web Portal (https://nationaldmg.org). These data were entered remotely by the OPOs managing each donor. In order to study the impact of
5
ACCEPTED MANUSCRIPT
hospital volume on organ yield, each donor was placed into a hospital volume quartile based on the number of donors managed by their hospital over the study period. Both DNDDs and DCDDs were included in total donor counts for the designation of hospital volume quartile as it
RI PT
was felt that excluding DCDDs would not accurately reflect the experience of a donor hospital as even the management of DCDDs (though different than that of DNDDs) still requires the
establishment of institutional policies and coordination with OPOs. Hospital quartile 1 was the
SC
lowest volume, whereas hospital quartile 4 was the highest volume. For any OPOs that joined the study after February 2012, volumes were normalized in order to account for late entry. OPOs and
M AN U
their corresponding geographic donation service areas (DSAs) were de-identified and arbitrarily numbered for the purpose of analysis. The primary outcome measure was identification of predictors of ≥ 4 OTPD. This numeric cutoff represents a slightly higher average OTPD goal than the national goal of 3.75 OTPD established by the Donation and Transplantation
TE D
Community of Practice in 2013 (http://healthcarecommunities.org/). It is also one more OTPD than the current national average of 3 (based on OPTN data as of May 6th 2016). By establishing a cutoff above these thresholds the analysis aimed to identify predictors of high performance. A
EP
categorical outcome variable was chosen in order to provide an interpretable endpoint which, in the context of having already established national metrics, was felt to be clinically relevant.
AC C
Statistical analysis
A two-part analysis was performed in order to identify predictors of ≥ 4 OTPD. For this
analysis, only DNDDs (both SCD and ECD) were included as DCDDs would inherently be disadvantaged in meeting the primary outcome measure of ≥ 4 OTPD. First, a univariate analysis was conducted to assess age, body mass index (BMI), blood type, ethnicity, donor type, cause of death, OPO, donor hospital volume, and final creatinine prior to organ recovery for donors
6
ACCEPTED MANUSCRIPT
achieving ≥ 4 OTPD. Continuous variables were analyzed using ANOVA and categorical variables were compared using chi-square tests. Univariate logistic regression was used to determine odds ratio estimates for both continuous and categorical variables. Variables with a p
RI PT
< 0.2 on univariate analysis were included in full multivariable model which was pruned using automated, backward, stepwise regression using exit and entrance criteria of p <0.05. Note that for variables that were clinically thought to be inherently associated, collinearity was assessed
SC
using Pearson’s correlation coefficient. In cases where collinearity was noted, the variables were run in separate multivariable models. Variables with a p<0.05 on multivariable analysis were
M AN U
considered to be independent predictors of achieving ≥ 4 OTPD. The final multivariable models were assessed by concordance index (c-statistic; 0.7 to 0.8 represented acceptable discrimination, 0.8 to 0.9 excellent discrimination, and >0.9 outstanding discrimination)(13, 14). Statistical analysis was performed with Stata version 13.1 for Windows (StataCorp, College Station,
TE D
Texas). Within the text and tables, values are reported as mean ± standard deviation (SD) or percent frequency (%) unless stated otherwise.
EP
Results
Over the 40-month study period, data from 4,427 donors were analyzed from 384 donor
AC C
hospitals. There were 3,821 DNDD donors, of which 76% were SCD and 24% were ECD. In addition, there were 605 DCDD donors. Hospitals managed an average of 3.3 ± 5.2 donors per hospital per year. Table 1 demonstrates the characteristics of the four hospital quartiles that were designated by volume, with quartile 1 being those with low volume and quartile 4 being the highest volume quartile. The number of hospitals in each quartile was about the same, ranging from 89-102, whereas the average donors managed per hospital per year ranged from a low of
7
ACCEPTED MANUSCRIPT
0.5 ± 0.2 to a high of 9 ± 7.7 per year. With regards to mean OTPD, quartile 4 had the highest average of 3.3 ± 1.8 (Table 1). Each of the quartiles had approximately the same proportion of SCD, ECD, and DCDD donors (Table 1). OPO/DSA characteristics are provided in Table 2. As
RI PT
can be seen, the number of donors per year ranged from a low of 48.4 in OPO/DSA #3 to a high of 368 in OPO/DSA #4.
Results of the univariate analysis of ≥ 4 OTPD are presented in Table 3. All of the
SC
variables screened were found to have a p<0.2. Due to collinearity between donor cause of death and donor hospital volume, a Pearson’s correlation analysis was performed and a strong
M AN U
association was noted between the variables (Pearson chi-squared = 505.1, p<0.001), and these variables were run in two separate multivariable models. After adjusting for age, ethnicity, donor type, blood type, body mass index, creatinine, and OPO/DSA, being managed in hospitals within the highest volume quartile remained a positive independent predictor of ≥ 4 OTPD (OR 1.52
TE D
[1.29-1.79], P<0.001; Table 4). In a separate model adjusting for the same covariates but substituting cause of death for hospital volume, cause of death from head trauma (OR 2.70 [2.253.25], P<0.001; Table 4) or cerebrovascular/stroke (OR 1.56 [1.28-1.92], P<0.001; Table 4)
EP
relative to anoxia were also noted to be a positive independent predictor of ≥ 4 OTPD. Both
AC C
statistical models had acceptable discrimination.
Discussion
Study of the inverse relationship between hospital volume and outcomes in medical and
surgical patients has led to the identification of opportunities for quality improvement. This investigation represents the first evaluation of this association in deceased organ donors, and is notable in having identified that organ donors managed in hospitals within the highest volume
8
ACCEPTED MANUSCRIPT
quartile are 52% more likely to have ≥ 4 OTPD. Further, regional factors, including management by specific OPOs in certain DSAs, were found to independently impact organ yield. Identification of these factors are important as efforts continue to be made in order to optimize
RI PT
the management of organ donors and decrease the imbalance in the supply and demand of organs available for transplantation.
Current approaches for addressing disparities of care based on the volume-outcome
SC
relationship have focused on increasing standardization as well as centralization of care. With regard to the former of these strategies, the implementation of checklists has been found to play
M AN U
an important role in improving postoperative morbidity and mortality in surgical patients(15). Although this has been effective when adopted, challenges include variability in the rigor of checklist implementation, resulting in partial as opposed to full checklist completion as well as potential for discordance between tasks completed on paper but not fully executed in
TE D
actuality(16, 17). The concept of checklists has been extended to care of the potential organ donor as DMGs, representing a checklist of critical care endpoints that DNDDs are managed to meet. In prior studies, meeting a certain proportion of these established goals has been associated
EP
with increased organ yield and improved graft function(6-8). Having now established an association between hospital volume and organ yield using a large database of systematically
AC C
collected data across multiple hospital and OPOs, further investigation to see if hospital volume is associated with the ability to meet DMGs will be the focus of future work. As mentioned, centralization of care represents an additional method for addressing
differences in outcomes noted in high volume hospitals relative to lower volume hospitals. The premise of this approach has been endorsed by large coalitions such as the Leapfrog Group which has developed volume standards for hospitals and surgeons performing certain high-risk
9
ACCEPTED MANUSCRIPT
surgical procedures(18). Although analogous thresholds have not been established for donor hospitals, the concept of establishing organ recovery centers at which donor management and organ procurement is performed has been proposed(19, 20). Specifically, as opposed to
RI PT
performing donor workup, management, and organ procurement occurring at individual donor hospitals, DNDDs are moved to an OPO-based facility wherein these processes are carried
through. In a recent retrospective study of organ donor recovery performed at an OPO based
SC
facility, it was noted that as compared to the traditional method of organ recovery, recovery performed at an OPO facility led to increased efficiency and higher organ yield (3.43 OTPD vs.
M AN U
2.69 OTPD across all donor types, p<0.001), while reducing donor recovery costs(19). An additional advantage of the OPO recovery center model includes the reduction of air travel by the surgical team and the associated risk, time, and cost of this practice (20, 21). Although centralization offers an opportunity to optimize the care of organ donors, establishment of
TE D
required centers relies on high transplant and organ recovery volumes; a cultural shift in donor hospital and OPO management; and the adoption of new processes to accommodate a change in the logistics of donor care. Additionally, significant capital investment is necessary for
EP
establishment of a center, including acquisition costs for equipment so that both donor management as well as recovery operations can be performed(19).
AC C
A number of reasons may account for the increase in OTPD noted in centers with higher
volume as well as the variability in organ utilization noted across OPOs in this study. Specifically, in lower volume centers, which are not as accustomed to donor management, there may be difficulty in obtaining requested investigations for organ workup leading to both suboptimal donor management and decreased utilization by transplant centers. Further, with inherently more contact between high volume donor hospitals and their respective OPO, these
10
ACCEPTED MANUSCRIPT
centers may also be more likely to have processes in place for early OPO referral, implementation of catastrophic brain injury guidelines (CBIGs), and more timely determination of neurologic death, possibly allowing for more optimal donor management after authorization
RI PT
for donation occurs(22, 23). With regards to OPO/DSA volume, it is interesting to note that two
specific OPO/DSA’s (#6 and #8) were less likely to attain ≥ 4 OTPD on adjusted analysis. Although these OPO/DSA’s were in the lowest third of OPO/DSA volume (Table 2), there were other OPO/DSAs with a
SC
similar number of donors per year that were not found to be significant on multivariable analysis. This suggests that other OPO specific factors and practices aside from volume are likely to play a role and
M AN U
warrant further investigation. Lastly, as cause of death and hospital volume were found to be
collinear, due to the fact that >95% of trauma patients were cared for quartile 4 centers, cause of death was independently analyzed. Independent of volume, and after adjusting for covariates, the finding that donors with head trauma were more than twice as likely to have ≥ 4 OTPD suggests
TE D
that additional factors such as the presence of CBIGs at such centers to help guide management may influence outcomes(22).
This study evaluating the impact of donor hospital volume on organ yield has limitations
EP
given the inherent challenges in accurately modeling the complex, and still incompletely understood, interplay between donor hospital, OPO, DSA, and transplant center as well as
AC C
recipient and donor factors. For instance, market competition which has been shown to increase the utilization of allografts by OPOs and the number transplants performed by centers in high competition areas was unable to be directly accounted for (24, 25). Further, additional hospital level factors such as the number of unique primary diagnoses admitted to each hospital, or rather, increased hospital complexity, has been previously associated with lower surgical mortality (26). This variable, which carries additional information that may help explain differences in donor hospitals’ systems of care was unavailable in our current data set. Additionally, as potential donors 11
ACCEPTED MANUSCRIPT
who did not proceed to become organ donors were not tracked in the database used for this study, specific hospital and OPO metrics such as conversion rate were not calculated. Despite these limitations,
however, it should be noted this report represents the only investigation benchmarking the
RI PT
association of donor hospital volume on organ donor yield. Ultimately, as the adoption of the UNOS DMG web portal continues to expand and successful linkage to currently available
transplant registries is performed, further analyses may help explain the noted association in the
SC
current study.
M AN U
Conclusions
Deceased organ donor hospital volume is associated with organ yield, with donors from the highest volume centers being 1.5 times more likely to achieve ≥ 4 OTPD. Future investigations should focus on further delineation of this relationship, including the identification
TE D
of process measures which may account for noted differences. Once identified, practices from high performing centers can be shared with those with lower organ utilization rates. Lastly,
AC C
EP
further consideration should also be given to the centralization of donor care in select DSAs.
12
ACCEPTED MANUSCRIPT
References 1.
Halm EA, Lee C, Chassin MR. Is volume related to outcome in health care? A systematic
review and methodologic critique of the literature. Ann Int Med 2002;137:511-520. Pieper D, Mathes T, Neugebauer E, Eikermann M. State of evidence on the relationship
RI PT
2.
between high-volume hospitals and outcomes in surgery: a systematic review of systematic reviews. J Am Coll Surg 2013;216:1015-1025 e1018.
Reames BN, Ghaferi AA, Birkmeyer JD, Dimick JB. Hospital volume and operative
mortality in the modern era. Ann Surg 2014;260:244-251.
Begg CB, Cramer LD, Hoskins WJ, Brennan MF. Impact of hospital volume on operative
M AN U
4.
SC
3.
mortality for major cancer surgery. JAMA 1998;280:1747-1751. 5.
Kotloff RM, Blosser S, Fulda GJ, et al. Management of the Potential Organ Donor in the
ICU: Society of Critical Care Medicine/American College of Chest Physicians/Association of
6.
TE D
Organ Procurement Organizations Consensus Statement. Crit Care Med 2015;43:1291-1325. Malinoski DJ, Patel MS, Ahmed O, et al. The impact of meeting donor management
goals on the development of delayed graft function in kidney transplant recipients. Am J
7.
EP
Transplant 2013;13:993-1000.
Malinoski DJ, Patel MS, Daly MC, et al. The impact of meeting donor management goals
AC C
on the number of organs transplanted per donor: results from the United Network for Organ Sharing Region 5 prospective donor management goals study. Crit Care Med 2012;40:27732780. 8.
Malinoski DJ, Daly MC, Patel MS, et al. Achieving donor management goals before
deceased donor procurement is associated with more organs transplanted per donor. J Trauma 2011;71:990-995; discussion 996.
13
ACCEPTED MANUSCRIPT
9.
Niemann CU, Feiner J, Swain S, et al. Therapeutic hypothermia in deceased organ donors
and kidney-graft function. N Engl J Med 2015;373:405-414. 10.
Franklin GA, Santos AP, Smith JW, et al. Optimization of donor management goals
11.
RI PT
yields increased organ use. Am Surg 2010;76:587-594.
Westphal GA. A simple bedside approach to therapeutic goals achievement during the
management of deceased organ donors--An adapted version of the "VIP" approach. Clin
12.
SC
Transplant 2016;30:138-144.
Israni AK, Zaun D, Bolch C, et al. Deceased organ donation. Am J Transplant 2016;16
M AN U
Suppl 2:195-215. 13.
LaValley MP. Logistic regression. Circulation 2008;117:2395-2399.
14.
Ohman EM, Granger CB, Harrington RA, Lee KL. Risk stratification and therapeutic
decision making in acute coronary syndromes. JAMA 2000;284:876-878. Haynes AB, Weiser TG, Berry WR, et al. A surgical safety checklist to reduce morbidity
TE D
15.
and mortality in a global population. N Engl J Med 2009;360:491-499. 16.
Mayer EK, Sevdalis N, Rout S, et al. Surgical checklist implementation project: the
EP
impact of variable WHO checklist compliance on risk-adjusted clinical outcomes after national implementation: a longitudinal study. Ann Surg 2016;263:58-63. O'Connor P, Reddin C, O'Sullivan M, et al. Surgical checklists: the human factor. Patient
AC C
17.
Saf Surg 2013;7:14. 18.
Birkmeyer JD, Finlayson EV, Birkmeyer CM. Volume standards for high-risk surgical
procedures: potential benefits of the Leapfrog initiative. Surgery 2001;130:415-422.
14
ACCEPTED MANUSCRIPT
19.
Doyle M, Subramanian V, Vachharajani N, et al. Organ donor recovery performed at an
organ procurement organization-based facility is an effective way to minimize organ recovery costs and increase organ yield. J Am Coll Surg 2016;222:591-600. Doyle MB, Vachharajani N, Wellen JR, et al. A novel organ donor facility: a decade of
experience with liver donors. Am J Transplant 2014;14:615-620.
Englesbe MJ, Merion RM. The riskiest job in medicine: transplant surgeons and organ
procurement travel. Am J Transplant 2009;9:2406-2415. 22.
SC
21.
RI PT
20.
Malinoski DJ, Patel MS, Lush S, et al. Impact of compliance with the American College
Coll Surg 2012;215:186-192. 23.
M AN U
of Surgeons trauma center verification requirements on organ donation-related outcomes. J Am
Kong AP, Barrios C, Salim A, et al. A multidisciplinary organ donor council and
performance improvement initiative can improve donation outcomes. Am Surg 2010;76:1059-
24.
TE D
1062.
Adler JT, Yeh H, Markmann JF, Nguyen LL. Market competition and density in liver
transplantation: relationship to volume and outcomes. J Am Coll Surg 2015;221:524-531. Adler JT, Yeh H, Markmann JF, Axelrod DA. Is donor service area market competition
EP
25.
1355. 26.
AC C
associated with organ procurement organization performance? Transplantation 2016;100:1349-
McCrum ML, Lipsitz SR, Berry WR, et al. Beyond volume: does hospital complexity
matter?: an analysis of inpatient surgical mortality in the United States. Med Care 2014;52:235242.
15
ACCEPTED MANUSCRIPT
Table 1. Hospital Quartile Characteristics
No. of hospitals
Mean overall OTPD, mean±SD
Quartile 1 (n=140) Quartile 2 (n=365)
97 102
0.5 ± 0.2 1.1 ± 0.5
2.9 ± 1.6 2.8 ± 1.6
Quartile 3 (n=817)
89
2.9 ± 1.1
2.9 ± 1.7
RI PT
Mean OTPD by donor type, mean (% within quartile)
Donors per hospital per year, mean±SD
96 384
9 ± 7.7 3.3 ± 5.2
3.3 ± 1.8 3.1 ± 1.8
Quartile 4 (n=3105) Total
SCD
ECD
DCDD
3.2 ± 1.7 (65) 3.3 ± 1.6 (65)
2.4 ± 1.5 (13) 1.9 ± 1.1 (17)
2 ± 0.9 (22) 1.9 ± 1.1 (17)
3.4 ± 1.6 (65)
1.8 ± 1.4 (10)
1.9 ± 1.1(24)
3.9 ± 1.7 (66) 3.8 ± 1.7 (66)
2.1 ± 1.4 (14) 2.1 ± 1.4 (14)
1.9 ± 0.9(20) 1.9 ± 0.9 (20)
SC
Variable Donor hospital volume
M AN U
SCD, standard criteria donor; ECD, expanded criteria donor; DCDD, donation after circulatory determination of death; OTPD, organs
AC C
EP
TE D
transplanted per donor.
16
ACCEPTED MANUSCRIPT
Table 2. Organ Procurement Organization and Donor Service Area Characteristics Donors per year 89.6 276 48.4 368 71.9 32.9 78 62.5 217.5
#10 (n=222)
67.6
SC
#4 (n=1,208) #5 (n=236) #6 (n=108) #7 (n=256) #8 (n=88) #9 (n=345)
RI PT
Variable, OPO/DSA #1 (n=294) #2 (n=906) #3 (n=159)
AC C
EP
TE D
M AN U
DSA, donor service area; OPO, organ procurement organization
17
ACCEPTED MANUSCRIPT
Table 3. Univariate Analysis of Variables Associated with ≥ 4 Organs Transplanted per Donor
Odds ratio
95% Confidence interval for odds ratio
Age (n=3,822) Body Mass Index (n=3,822)
0.95 0.95
0.94-0.95 0.94-0.96
<0.001 <0.001
Blood type A (ref; n=1,349) AB (n=139) B (n=463) O (n=1,871) Ethnicity
1 0.53 0.98 1.19
0.36-0.77 0.79-1.21 1.03-1.37
0.001 0.86 0.017
N/A 0.69-1.10 1.06-1.42 0.65-1.11 0.80-1.95
N/A 0.26 0.005 0.232 0.323
1
N/A
N/A
0.11
0.09-0.14
<0.001
1 0.64 0.99 2.82
N/A 0.54-0.75 0.32-3.04 2.39-3.34
N/A <0.001 0.984 <0.001
0.79
0.46-1.35
0.392
1 0.93 1.38
N/A 0.71-1.21 0.94-2.04
N/A 0.59 0.10
1.03 0.96 0.54 0.84 0.84 1.18
0.80-1.33 0.68-1.36 0.33-0.86 0.60-1.18 0.52-1.37 0.86-1.61
0.81 0.83 0.01 0.32 0.48 0.30
1.60
1.12-2.27
0.009
1 0.99
N/A 0.63-1.54
N/A 0.95
1 0.87 1.23 0.85 1.25
Donor type SCD (ref; n=2,910)
TE D
ECD (n=911) Cause of death Anoxia (ref ; n=1,092) Cerebrovascular/stroke (n=1,406) CNS Tumor (n=13) Head trauma (n=1,248)
AC C
#1 (n=294) #2 (n=906) #3 (n=159)
EP
Other (n=63) OPO/DSA
#4 (n=1208) #5 (n=236) #6 (n=108) #7 (n=256) #8 (n=88) #9 (n=345)
#10 (n=222) Donor hospital volume Quartile 1 (ref; n=122) Quartile 2 (n=302)
18
p Value*
RI PT
M AN U
White (ref; n=1,917) Black (n=348) Hispanic (n=1,218) Asian (n=257) Other (n=82)
SC
Variable
ACCEPTED MANUSCRIPT
Quartile 3 (n=732) Quartile 4 (n=2,666) Creatinine prior to organ recovery (n=3,816)
0.99 1.70
0.66-1.48 1.16-2.49
0.95 0.006
0.73
0.69-0.77
<0.001
*p Value calculated using univariate logistic regression
AC C
EP
TE D
M AN U
SC
procurement organization; DSA, donor service area
RI PT
SCD, standard criteria donor; ECD, expanded criteria donor; CNS, central nervous system; OPO, organ
19
ACCEPTED MANUSCRIPT
Table 4. Multivariable Analyses: Independent Predictors of ≥ 4 Organs Transplanted per Donor
p Value*
0.96 0.99
0.96-0.97 0.98-1.00
<0.001 0.029
0.49 1.27 0.73 0.31 0.40 0.54 0.77
0.32-0.76 1.10-1.48 0.56-0.96 0.24-0.40 0.25-0.65 0.33-0.88 0.73-0.81
1.52
1.29-1.79
<0.001
0.96
0.96-0.97
<0.001
0.32-0.75 1.07-1.45 0.57-0.97 0.25-0.44 0.25-0.65 0.35-0.96 0.74-0.82 2.25-3.25 1.28-1.92
0.001 0.005 0.031 <0.001 <0.001 0.033 <0.001 <0.001 <0.001
EP
AC C
SC
RI PT
95% Confidence interval for odds ratio
TE D
Model 1† hospital volume Donor age, per y Body Mass Index, per point Blood type AB vs A O vs A Ethnicity, Black vs White Donor type, ECD vs SCD OPO/DSA #6 vs OPO/DSA #1 OPO/DSA #8 vs OPO/DSA #1 Creatinine prior to organ recovery, per mg/dL Donor hospital volume, quartile 4, highest volume vs lowest volume quartile 1 Model 2‡ cause of death Donor age, per y Blood type AB vs A O vs A Ethnicity, Black vs White Donor type, ECD vs SCD OPO/DSA #6 vs OPO/DSA #1 OPO/DSA #8 vs OPO/DSA #1 Creatinine prior to organ recovery, per mg/dL Cause of death, head trauma vs anoxia Cause of death, cerebrovascular/stroke vs anoxia
Odds ratio
M AN U
Variable, model
0.49 1.24 0.74 0.33 0.41 0.58 0.78 2.70 1.56
0.001 0.001 0.023 <0.001 <0.001 0.014 <0.001
*pValue calculated using a stepwise backward selection logistic regression model. †
Concordance Index C = 0.793.
‡
Concordance Index C = 0.799.
ECD, expanded criteria donor; OPO, organ procurement organization; DSA, donor service area 20
ACCEPTED MANUSCRIPT
Precis Highest volume centers were more likely to achieve ≥4 organs transplanted per donor. This suggests efforts should be made to share practices from higher volume centers and consideration
AC C
EP
TE D
M AN U
SC
RI PT
should be given to centralization of donor care.
21