Same-Teams versus Different-Teams for Long Distance Lung Procurement: A Cost Analysis

Abstract

OBJECTIVE:

In an era of broader lung sharing, different-team transplantation (DT, procuring team from non-recipient center) may streamline procurement logistics; however, safety and cost implications of DT remain unclear. To understand whether DT represents a safe means to reduce lung transplant (LTx) costs, we compared post-transplant outcomes and lung procurement and index hospitalization costs among matched DT and same-team transplantation (ST, procuring team from recipient center) cohorts at a single, high-volume institution. We hypothesized that DT reduces costs without compromising outcomes after LTx.

METHODS:

Patients who underwent DT between 1/2016-5/2020 were included. A cohort of patients who underwent ST was matched 1:3 (nearest neighbor) based on recipient age, disease group, lung allocation score, history of prior LTx, and bilateral versus single LTx. Post-transplant outcomes and costs were compared between groups.

RESULTS:

23 DT and 69 matched ST recipients were included. Perioperative outcomes and post-transplant survival were similar between groups. Compared to ST, DT was associated with similar lung procurement and index hospitalization costs (DT versus ST, procurement: median $65,991 versus $58,847, p=0.16; index hospitalization: median $294,346 versus $322,189, p=0.7). On average, procurement costs increased $3,263 less per 100 nautical miles for DT versus ST; DT offered cost-savings when travel distances exceeded approximately 363 nautical miles.

CONCLUSIONS:

At our institution, DT and ST were associated with similar post-LTx outcomes; DT offered cost-savings with increasing procurement travel distance. These findings suggest that DT may mitigate logistical and financial burdens of lung procurement; however, further investigation in a multi-institutional cohort is warranted.

ULTRAMINI-ABSTRACT

To understand whether different-team transplantation (DT, procuring team from non-recipient center) represents a safe means to reduce lung transplant (LTx) costs, we compared post-transplant outcomes and lung procurement and index hospitalization costs among matched DT and same-team transplantation (ST, procuring team from recipient center) cohorts at a single, high-volume institution.

INTRODUCTION

Organ procurement is a relatively inefficient, expensive, and risky aspect of solid organ transplantation. Typically, transplant centers send surgical teams to geographically diverse donor hospitals to inspect, evaluate, and procure allografts for recipients at their own institutions (same-team transplantation, ST). Organ procurement is frequently time-intensive due to unexpected delays,¹ and urgent travel to donor hospitals introduces heightened risks of work-related accidents and fatalities.^2,3 In the US, these logistical and financial burdens prompted recent revision of procurement practices and policies to include requirements for increased liability coverage and more stringent air and ground transportation quality standards.^3,4 Additionally, some have advocated for broader sharing of procurement duties across centers to improve the safety and efficiency of organ procurement for transplant centers and surgeons.^3,4

Organ procurement by surgical teams from institutions other than the recipient transplant center, but located in close proximity to donor hospitals (different-team transplantation, DT), would likely reduce the need for travel by distant recipient teams, minimizing risks to providers and reducing procurement costs.^2,5 DT is commonly employed in abdominal transplantation⁶; however, limited historical data suggest that in lung transplantation (LTx), DT accounts for fewer than 20% of lung procurements.^7,8

In November 2017, US lung allocation policy was emergently revised, replacing donation service area (DSA) with a 250-nautical mile radius around the donor hospital as the first unit of allocation.⁹ Intended to increase access to available donor lungs among high-acuity patients,^10,11 this policy unintentionally reduced local (within-DSA) LTx, and increased allograft ischemic times, procurement travel distances, and costs.^5,10-12 In this new era, a more efficient and cost-effective approach to lung procurement is urgently required. Broader utilization of DT may represent one means to achieve these goals, particularly in the context of preliminary evidence supporting comparable outcomes among ST and DT recipients.¹³ Accordingly, cost is likely to be a critical determinant of the decision to use ST versus DT. We hypothesized that DT is associated with reduced costs and comparable outcomes after LTx. To test these hypotheses, we compared lung procurement and index hospitalization costs and post-transplant outcomes between patients who underwent DT and ST at a single high-volume LTx institution.

METHODS

Data sources and study population

We conducted a single-center retrospective cohort study using institutional and United Network for Organ Sharing (UNOS) data. Patients who underwent primary or redo isolated LTx at Duke University Hospital between January 1, 2016 and May 31, 2020 were included. Patients who underwent multi-organ transplantation, those missing allograft ischemic time, and those for whom lung procurement team or procurement travel distance could not be determined were excluded. The DT cohort was defined by LTx events in which lung procurement teams were affiliated with institutions other than the recipient transplant center (Duke University). A group of patients who underwent ST, defined by LTx events in which lung procurement teams were affiliated with the recipient transplant center, was matched 1:3 (nearest neighbor) based on recipient age, disease group, lung allocation score (LAS), history of prior LTx, and transplant type (bilateral versus single) (Figure S1). This study was initially approved by our Institutional Review Board (Pro00103325) on 9/3/2019 and renewed on 10/5/2021. As this protocol does not involve prospective enrollment of subjects, consent was obtained through a Waiver or Alteration of Consent and HIPAA Authorization and Decedent Research Notification.

Study design

Recipient, operative, donor characteristics and post-transplant outcomes were compared between ST and DT cohorts. Perioperative outcomes of interest included 30-day reintervention (surgical, radiologic, bronchoscopic), 30-day hospital or intensive care unit (ICU) readmission, post-transplant length of stay (LOS), grade 3 primary graft dysfunction (PGD3) at 72 hours post-transplant, need for postoperative extracorporeal membrane oxygenation (ECMO), postoperative date of extubation, tracheostomy within 7 days post-transplant, reintubation or renal replacement therapy during the index hospitalization, 30-day biopsy-proven acute rejection, and 90-day mortality. Rates of textbook outcome achievement were also compared between ST and DT groups. As previously described, textbook outcome included freedom from all of the above listed perioperative endpoints in addition to intraoperative complication, post-transplant LOS >75^th percentile of LTx patients, PGD3 at 48 hours post-transplant, and extubation >48 hours post-transplant.¹⁴ Beyond the perioperative period, survival to death or graft failure, first biopsy-proven acute rejection episode, and onset of chronic lung allograft dysfunction (CLAD) were explored. Bronchoscopic reintervention was defined by need for an intervention such as bronchial dilation or stenting beyond routine post-transplant surveillance. PGD3 and CLAD were defined according to International Society for Heart and Lung Transplantation guidelines.^15,16

Financial implications of same-team versus different-team transplantation

The Duke Transplant Center Office of Finance provided lung procurement and recipient index hospitalization costs for all patients in the study cohort. Donor-specific costs were those related to the lung procurement including air and ground transportation, surgeon fees, and organ procurement fees. Recipient-specific costs were those related to the index hospitalization including intensive and intermediate care, surgeon fees, laboratory fees, and supplies. Donor-specific (lung procurement), recipient-specific (index hospitalization), and total costs were compared between ST and DT cohorts.

Statistical analysis

Recipient, operative, and donor characteristics, perioperative outcomes, and donor-specific, recipient-specific, and total costs were compared between ST and DT cohorts using Wilcoxon rank-sum tests for continuous variables and Chi-squared and Fisher’s exact tests for categorical variables. Unadjusted survival was estimated using the Kaplan-Meier method and compared between groups using log-rank tests.

The association between travel distance and lung procurement cost was estimated separately for ST and DT cohorts using Spearman correlation coefficients. The change in lung procurement cost with increasing travel distance was assessed using linear regression with an interaction between travel distance and procurement team to evaluate whether the change in procurement cost with increasing travel distance differed based on ST versus DT. A two-sided p-value less than 0.05 was considered statistically significant. All analyses were performed using R version 3.6.1 (Vienna, Austria).

RESULTS

Use of same-team and different-team transplantation

Overall, our institution performed 456 LTx during the study period (ST 433 [95.0%] versus DT 23 [5.0%]). The proportion of patients who underwent DT fluctuated throughout the study period (Figure 1). There was no correlation between annual proportion of DT and median procurement travel distance (r=0.36, p=0.6).

Figure 1. Use of different-team transplantation (DT) over time.

Bars represent the percent of all annual lung procurements performed by different teams. The red line represents the median travel distance between donor hospitals and the transplant center for all lung transplants (including both DT and same-team transplantation) performed during each year of the study period.

Recipient, operative, and donor characteristics

After matching, 92 LTx recipients were included of whom 23 underwent DT and 69 underwent ST. Recipient characteristics including age, sex, race/ethnicity, and etiology of respiratory failure were similar between groups (Table 1).

Table 1.

Recipient characteristics

Characteristic	Same-Team Transplantation N = 69	Different-Team Transplantation N = 23	Standardized Mean Difference	p-value
Age (years)	56.0 [40.0, 65.0]	56.0 [36.5, 65.0]	0.06	0.8
Sex			0.15	0.5
Female	34 (49.3%)	13 (56.5%)
Male	35 (50.7%)	10 (43.5%)
Race			0.3	0.6
Caucasian/White	63 (91.3%)	20 (87.0%)
Black or African American	5 (7.2%)	3 (13.0%)
Other	1 (1.4%)	0 (0%)
Ethnicity (Hispanic)	1 (1.4%)	0 (0%)	0.17	0.6
Panel reactive antibody at transplant (%)
Class I	0.00 [0.00, 0.00]	0.00 [0.00, 0.00]	0.015	0.3
Class II	0.00 [0.00, 0.00]	0.00 [0.00, 0.00]	0.18	0.6
Etiology of respiratory failure			0.7	0.5
Alpha-1-antitrypsin deficiency	2 (2.9%)	3 (13.0%)
Autoimmune interstitial lung disease	3 (4.3%)	2 (8.7%)
Chronic obstructive pulmonary disease/emphysema	12 (17.4%)	1 (4.3%)
Cystic fibrosis	3 (4.3%)	1 (4.3%)
Hypersensitivity pneumonitis	3 (4.3%)	2 (8.7%)
Idiopathic pulmonary fibrosis	23 (33.3%)	7 (30.4%)
Pulmonary hypertension	3 (4.3%)	2 (8.7%)
Re-transplant	9 (13.0%)	3 (13.0%)
Other interstitial lung disease	3 (4.3%)	1 (4.3%)
Other	8 (11.6%)	1 (4.3%)
COPD/EMPHYSEMA12 (17.4%)1 (4.3%)CYSTIC FIBROSIS3 (4.3%)1 (4.3%)PULMONARY HYPERTENSION3 (4.3%)2 (8.7%)AUTOIMMUNE ILD3 (4.3%)2 (8.7%)ALPHA-1-ANTITRYPSIN DEFICIENCY2 (2.9%)3 (13.0%)OTHER8 (11.6%)1 (4.3%)Disease group			<0.001	>0.9
A	15 (21.7%)	5 (21.7%)
B	6 (8.7%)	2 (8.7%)
C	3 (4.3%)	1 (4.3%)
D	45 (65.2%)	15 (65.2%)
Status at time of transplant			0.04	0.9
Inpatient/hospitalized	11 (15.9%)	4 (17.4%)
Outpatient	58 (84.1%)	19 (82.6%)
Interventions at time of transplant
Extracorporeal membrane oxygenation	6 (8.7%)	4 (17.4%)	0.3	0.2
Intubated	7 (10.1%)	3 (13.0%)	0.09	0.7
History of prior lung transplant	9 (13.0%)	3 (13.0%)	<0.001	>0.9
History of previous thoracic surgery	12 (17.4%)	3 (13.0%)	0.12	0.6
Lung allocation score at time of transplant	44.0 [36.4, 50.4]	43.4 [36.5, 47.6]	0.05	0.9
Most recent 6-minute walk distance (feet)	1400 [1180, 1620]	1420 [1120, 1680]	0.04	0.8
Missing	3 (4.3%)	1 (4.3%)
Cardiac output (L/min)	5.15 [4.50, 6.18]	5.45 [4.83, 6.23]	0.14	0.6
Missing	3 (4.3%)	1 (4.3%)
Arterial pCO2 at transplant (mmHg)	46.0 [41.0, 54.0]	42.0 [39.0, 46.5]	0.4	0.06
Smoking history	29 (42.0%)	16 (69.6%)	0.6	0.02
Pack-years	17.0 [11.0, 30.0]	15.0 [10.0, 34.5]	0.006	0.8
Body mass index at transplant (kg/m2)	24.7 [20.8, 27.2]	24.3 [22.2, 27.2]	0.13	0.8
Pre-transplant pulmonary rehabilitation	63 (91.3%)	20 (87.0%)	0.14	0.5
Pre-transplant pulmonary function
FEV1 (% predicted)	32.0 [19.5, 50.5]	38.0 [26.5, 51.5]	0.04	0.6
Missing	2 (2.9%)	0 (0%)
FVC (% predicted)	47.0 [36.0, 57.0]	52.0 [37.8, 64.5]	0.3	0.15
Missing	2 (2.9%)	1 (4.3%)

Presented as median (interquartile range) for continuous variables and frequency (proportion) for categorical variables.

FEV1, forced expiratory volume in 1 second. FVC, forced vital capacity.

Operative characteristics including use of cardiopulmonary bypass, intraoperative transfusion requirements, and use of ex-vivo lung perfusion (EVLP) were similar between ST and DT cohorts (Table 2). Compared to those who underwent ST, patients who underwent DT received lung allografts with longer ischemic times, however, this difference was not statistically significant (median 462 versus 426 minutes, p=0.16) (Table 2).

Table 2.

Operative characteristics.

Characteristic	Same-Team Transplantation N = 69	Different-Team Transplantation N = 23	Standardized Mean Difference	p-value
Ischemic time (minutes)	426 [366, 570]	462 [396, 579]	0.5	0.16
Cardiopulmonary bypass used	18 (26.1%)	7 (30.4%)	0.1	0.7
Bypass time (minutes)	269 [173, 330]	254 [236, 309]	0.4	0.6
Donor type			0.2	0.4
Donation after brain death donor	62 (89.9%)	22 (95.7%)
Donation after circulatory death donor	7 (10.1%)	1 (4.3%)
Ex-vivo lung perfusion used	4 (5.8%)	2 (8.7%)	0.11	0.6
Transfusion requirement (units)
Packed red blood cells	2.00 [0, 4.00]	1.00 [0.500, 3.00]	0.2	0.8
Fresh frozen plasma	0.00 [0.00, 2.00]	0.00 [0.00, 2.00]	0.2	0.8
Cryoprecipitate	1.00 [0.00, 2.00]	0.00 [0.00, 1.50]	0.14	0.4
Platelets	0.00 [0.00, 2.00]	0.00 [0.00, 1.00]	0.2	0.8
Donor/recipient weight ratio	1.06 [0.893, 1.48]	1.03 [0.940, 1.23]	0.18	0.6
Serologies
Donor Epstein-Barr virus positive	64 (92.8%)	21 (91.3%)	0.12	0.7
Missing	0 (0%)	1 (4.3%)
Donor cytomegalovirus positive	37 (53.6%)	13 (56.5%)	0.058	0.8
Recipient cytomegalovirus positive	48 (69.6%)	10 (43.5%)	0.5	0.02
Cytomegalovirus mismatch (D+/R−)	12 (17.4%)	8 (34.8%)	0.4	0.08
Transplant type			<0.001	>0.9
Bilateral	63 (91.3%)	21 (91.3%)
Single	6 (8.7%)	2 (8.7%)

Presented as median (interquartile range) for continuous variables and frequency (proportion) for categorical variables.

Compared to lung donors in the ST cohort, donors in the DT cohort had lower pre-procurement PaO₂/FiO₂ (P/F) ratios (median 367 versus 456, p<0.01), were located farther from the transplant center (median 438 versus 214 nautical miles, p<0.01) (Table 3), and were less likely to be located in UNOS region 11, the same region as Duke University (30.4% versus 52.2%, p=0.02). Donor characteristics including age, smoking history ≥20 pack-years, and cause of death were similar between groups (Table 3).

Table 3.

Donor characteristics

Characteristic	Same-Team Transplantation N = 69	Different-Team Transplantation N = 23	p-value
Age (years)	38.0 [27.0, 46.0]	37.0 [23.5, 50.0]	>0.9
PHS increased risk	27 (39.1%)	6 (26.1%)	0.3
Smoking history ≥20 pack-years	7 (10.1%)	2 (8.7%)	0.8
Missing	2 (2.9%)	0 (0%)
PaO2/FiO2 ratio	456 [386, 525]	367 [300, 430]	<0.01
Cause of death			0.5
Anoxia	22 (31.9%)	5 (21.7%)
Cerebrovascular accident/stroke	22 (31.9%)	11 (47.8%)
Head trauma	22 (31.9%)	7 (30.4%)
Other	3 (4.3%)	0 (0%)
Mechanism of death			0.3
Asphyxiation	3 (4.3%)	0 (0%)
Blunt injury	9 (13.0%)	6 (26.1%)
Cardiovascular	4 (5.8%)	1 (4.3%)
Drug intoxication	14 (20.3%)	4 (17.4%)
Gunshot wound	14 (20.3%)	1 (4.3%)
Intracranial hemorrhage/stroke	22 (31.9%)	10 (43.5%)
Natural cause	0 (0%)	1 (4.3%)
Seizure	1 (1.4%)	0 (0%)
Other	2 (2.9%)	0 (0%)
Blood infection	12 (17.4%)	1 (4.3%)	0.13
Missing	0 (0%)	1 (4.3%)
Pulmonary infection	54 (78.3%)	14 (60.9%)	0.13
Missing	1 (1.4%)	1 (4.3%)
Distance from transplant center (nautical miles)	214 [116, 430]	438 [338, 695]	<0.01
UNOS region			0.02
1	2 (2.9%)	0 (0%)
2	3 (4.3%)	0 (0%)
3	10 (14.5%)	6 (26.1%)
4	8 (11.6%)	1 (4.3%)
5	0 (0%)	2 (8.7%)
6	0 (0%)	1 (4.3%)
7	4 (5.8%)	1 (4.3%)
8	1 (1.4%)	2 (8.7%)
9	0 (0%)	1 (4.3%)
10	5 (7.2%)	2 (8.7%)
11	36 (52.2%)	7 (30.4%)
Match sequence	24.0 [11.0, 59.0]	41.0 [16.0, 115]	0.09

Presented as median (interquartile range) for continuous variables and frequency (proportion for categorical variables.

Post-transplant outcomes

Patients who underwent ST and DT had similar rates of PGD3 at 72 hours, need for postoperative ECMO, and similar post-transplant LOS; patients who underwent ST were less likely to achieve textbook outcome though this difference was not statistically significant (ST 21.7% versus DT 39.1%, p=0.1) (Table 4). On unadjusted survival analysis, there was no difference in survival to death or graft failure between groups (Figure 2A, log-rank p=0.3). Specifically, 1- and 3-year survival were 84.5% (95% confidence interval [CI] 76.1-93.9%) and 66.5% (95% CI 52.6-84.1%), respectively, in the ST cohort compared to 100% (95% CI 100-100%) and 85.0% (95% CI 67.2-100%), respectively, in the DT cohort. Rejection-free and CLAD-free survival were also similar between groups (Figures 2B and 2C).

Table 4.

Post-transplant perioperative complications

Characteristic	Same-Team Transplantation N = 69	Different-Team Transplantation N = 23	p-value
Achieved a textbook outcome a	15 (21.7%)	9 (39.1%)	0.1
Reintervention within 30 days	19 (27.5%)	8 (34.8%)	0.5
Surgical	18 (26.1%)	8 (34.8%)
Bronchoscopic	1 (1.4%)	0 (0%)
Radiologic	1 (1.4%)	0 (0%)
Grade 3 primary graft dysfunction at 48 hours b	19/69 (27.5%)	5/22 (22.7%)	0.6
Ungradablec	0/69 (0%)	1/23 (4.3%)
Grade 3 primary graft dysfunction at 72 hours b	14/69 (20.3%)	4/21 (19.0%)	0.9
Ungradablec	0/69 (0%)	2/23 (8.7%)
Post-operative extracorporeal membrane oxygenation	13 (18.8%)	5 (21.7%)	0.8
Extubated in >48 hours	24 (34.8%)	9 (39.1%)	0.7
Tracheostomy within 7 days	19 (27.5%)	6 (26.1%)	0.9
Reintubated during index hospitalization	16 (23.2%)	3 (13.0%)	0.3
Renal replacement therapy during index hospitalization	12 (17.4%)	5 (21.7%)	0.6
ICU readmission within 30 days	13 (18.8%)	2 (8.7%)	0.3
Hospital readmission within 30 days	17 (24.6%)	2 (8.7%)	0.1
Post-transplant length of stay (days)	25.0 [19.0, 38.0]	25.0 [14.0, 50.5]	0.6
Length of stay >75th percentile	16 (23.2%)	7 (30.4%)	0.5
Acute rejection within 30 days	13 (18.8%)	3 (13.0%)	0.5
Mortality within 90 days	4 (5.8%)	0 (0%)	0.2
In-hospital mortality	7 (10.1%)	0 (0%)	0.2
Mortality within 1 year	10 (14.5%)	0 (0%)	0.06

Presented as median (interquartile range) for continuous variables and frequency (proportion) for categorical variables.

^aTextbook outcome was defined as freedom from the following perioperative events: intraoperative complication, reintervention within 30 days, grade 3 primary graft dysfunction at 48- or 72-hours post-transplant, postoperative extracorporeal membrane oxygenation, extubation >48 hours post-transplant, tracheostomy within 7 days post-transplant, reintubation during the index hospitalization, renal replacement therapy during the index hospitalization, intensive care unit readmission with 30 days, hospital readmission within 30 days, biopsy-proven acute rejection within 30 days, mortality within 90 days, or hospital length of stay >75^th percentile of lung transplant patients. Any one complication was sufficient to define textbook outcome failure, but patients who failed to achieve a textbook outcome could have multiple events.¹⁴

^bFor analyses of grade 3 primary graft dysfunction, patients for whom primary graft dysfunction could not be graded were excluded (1 patient at 48 hours, 2 patients at 72 hours; see footnote c). Patients who were on extracorporeal membrane oxygenation support and had radiographic evidence of pulmonary edema classified as having grade 3 primary graft dysfunction per International Society for Heart and Lung Transplantation guidelines.

^cIn accordance with International Society for Heart and Lung Transplantation guidelines, patients who were on extracorporeal membrane oxygenation support and had no radiographic evidence of pulmonary edema were designated as “ungradable” and excluded from analyses of primary graft dysfunction.

Figure 2. Kaplan-Meier survival analysis stratified by same-team transplantation (ST) versus different-team transplantation (DT).

(A) Patient or graft survival. (B) Rejection-free survival. (C) Chronic lung allograft dysfunction-free survival. Shaded regions represent the 95% confidence limits for ST (blue) and DT (red).

Financial implications of same-team and different-team transplantation

Donor-specific (lung procurement), recipient-specific (index hospitalization), and total costs There were no statistical differences in donor-specific (lung procurement), recipient-specific (index hospitalization), or total (donor-specific + recipient-specific) LTx costs between ST and DT cohorts (all p>0.05) (Table 5). Specifically, median total LTx costs were $380,093 versus $383,194 among ST and DT cohorts, respectively (p=0.8) (Table 5, Figure S2). Median lung procurement costs were $65,991 [DT] versus $58,847 [ST] (p=0.16), and median index hospitalization costs were $294,346 [DT] versus $322,189 [ST] (p=0.7) (Table 5). Detailed comparisons of categorized donor-specific and recipient-specific costs are shown in Tables S1 and S2, respectively. There were several differences in procurement charges between ST and DT groups: specifically, DT was associated with lower surgical, ground transportation, supply, and flyout fees compared to ST (Table S1).

Table 5.

Cost comparison between same-team and different-team transplantation.

Cost	Same-Team Transplantation N = 69	Different-Team Transplantation N = 23	p-value
Donor-specific (total lung procurement cost)	$58847 (51257, 66506), [12545, 94152]	$65991 (56173, 71495) [9905, 88849]	0.16
Recipient-specific (total index hospitalization cost)	$322189 (266083, 446866), [202048, 2493569]	$294346 (244641, 580605), [222690, 983048]	0.7
Total cost	$380093 (315965, 501121), [264853, 2545498]	$383194 (298128, 630880), [267186, 1061829]	0.8

Presented as median (interquartile range) [full range]

Association between lung procurement cost and travel distance

There was a strong positive correlation between travel distance and total lung procurement cost for both cohorts (ST: r=0.81, p<0.01; DT: r=0.50, p=0.016). For both DT and ST, total lung procurement costs increased with increasing travel distance; however, the rate of cost increase was lower with DT compared to ST. On average, lung procurement costs increased $3,263 less per 100 nautical miles for DT versus ST (interaction p<0.01) (Figure 3). To estimate the travel distance at which DT becomes less expensive than ST, we determined the intersection of the cost trajectories for DT and ST predicted using a linear regression model, and found that DT was cost-saving for our center when lung procurement travel distances exceeded approximately 363 nautical miles (Figure 3). In our cohort, 17 (73.9%) DT and 22 (31.9%) ST lung allografts were procured from donors located >363 nautical miles from the transplant center.

Figure 3. Change in lung procurement cost over increasing travel distance.

Points correspond to individual lung procurements and lines correspond to the best fit determined from a simple linear regression model with an interaction between procurement team and travel distance. The best fit lines intersect at approximately 363 nautical miles.

Across all patients in the study cohort, approximately $5,461,187 were spent on lung procurements. Theoretically, if all patients had undergone ST, our institution would have incurred approximately $263,462 (+4.8%) more in lung procurement costs across all included patients (total: $5,724,649). Conversely, if all patients who received lungs procured more than 363 nautical miles from the transplant center had undergone DT, our institution would have incurred approximately $211,054 (−3.9%) less in lung procurement costs across all included patients (total: $5,250,133).

DISCUSSION

In this single-institution study, we examined our contemporary experience using ST and DT, and compared outcomes and costs among corresponding LTx recipients. We found that ST and DT were associated with similar perioperative outcomes and post-transplant survival. Lung procurement and index hospitalization costs were also similar between groups; however, lung procurement costs increased less with increasing travel distance for DT versus ST, rising $3,263 less per 100 nautical miles on average for DT versus ST, and leading to significant cost-savings when procurement travel distances exceeded approximately 363 nautical miles. These findings suggest that increased use of DT may mitigate logistical and financial burdens of lung procurement without compromising outcomes in an era of broader lung sharing (Figure 4).

Figure 4. Graphical abstract.

Compared to same-team transplantation (procuring team from recipient center), different-team transplantation (procuring team from non-recipient center) may reduce costs associated with long distance lung procurement without compromising recipient outcomes in an era of broader lung sharing.

One primary concern surrounding DT is the inherent inability of recipient surgeons to directly inspect and assess the quality of donor allografts.^17,18 Assurance of acceptable outcomes among corresponding LTx recipients is therefore necessary to support and encourage use of DT. Several historical single-institution studies found similar post-transplant outcomes in ST and DT cohorts.^19,20,8,21 More recently, Yang and colleagues analyzed Scientific Registry of Transplant Recipients data and demonstrated comparable risks of PGD3 and 1-year graft failure after LTx among recipients of ST versus DT.¹³ In our study, patients in both ST and DT cohorts had similar rates of PGD3 at 72 hours, postoperative ECMO, and reintubation, and similar overall, rejection-free, and CLAD-free survival up to 4 years post-transplant. Our findings may thus provide further assurance that DT supports acceptable outcomes for LTx recipients in the modern era. It is conceivable that pre-existing professional relationships with surgical teams at distant institutions may influence the decision of when to proceed with DT versus ST. In our experience, prior experience or relationships with procuring teams were not considered a requirement for proceeding with DT. Regardless, reliance on DT to aid in organ procurement requires effective communication and trust,¹³ aspects that may be otherwise enhanced by certifying procuring surgeons in some fashion and developing an OPO-based procuring surgeon registry. Likewise, as the frequency of geographically distant lung procurements continues to rise,^5,12 use of telehealth technology may allow recipient surgeons to evaluate organs remotely and engage in collaborative decision-making when DT is used,^22,23 increasing provider comfort and ensuring consistent procurement of high-quality allografts. Moreover, as LTx evolves amidst the ongoing Coronavirus Disease 2019 pandemic, our findings may encourage integration of DT into conventional transplant practices. At our institution, travel restrictions early in the pandemic necessitated increased DT to minimize exposure risks for patients and providers.²⁴ While an increase in DT may have initially represented a necessary action to support LTx during the pandemic’s first wave, increasing evidence supporting acceptable outcomes in DT cohorts may encourage more liberal use of collaborative procurement services beyond this global challenge.

DT may also represent a critical means to reduce growing financial barriers and travel risks associated with lung procurement.^{5,10-12,22,23} Costs related to procurement transport, organ acquisition, and supplies have increased since introduction of the new lung allocation policy, yet DT remains infrequently utilized.^5,12,22,23 In our study, lung procurement, index hospitalization, and total LTx costs were similar in ST and DT cohorts. However, we observed a strong positive correlation between procurement travel distance and cost, and noted that DT became increasingly valuable with increasing travel distance, offering cost-savings when travel distances exceeded approximately 363 nautical miles. The particular distance at which DT becomes less expensive than ST likely differs for individual transplant centers; nonetheless, this finding offers immediately actionable information for our center, providing an opportunity to critically evaluate our procurement practices as we strive to adopt the most cost-effective strategy. Unfortunately, the single-center nature of our study precluded evaluation of the range of distances over which transplant centers may find DT most cost-effective. However, Organ Procurement and Transplantation Network data from the 2-year monitoring report of the lung allocation policy change demonstrated an increase in the number of LTx that used donor lungs procured more than 500 nautical miles from the transplant center, and a significant increase in the maximum lung procurement distance from 2,327 to 4,137 nautical miles after the policy change.¹¹ In this context, the 363 nautical mile travel distance at which our institution found DT cost-saving is notably modest, suggesting that there is likely tremendous potential for transplant centers nationwide to reduce costs and streamline LTx logistics through increased use of DT. As procurement travel is frequently urgent and dangerous,^4,25,26 the value of DT may be further amplified by the potential to reduce costs and improve safety with broader use. Further investigation is warranted to understand the overall financial and safety advantages of DT versus ST to guide targeted allocation of available resources.

In addition to broader use of DT, one emerging model that may further mitigate logistical and financial burdens of lung procurement is the specialized donor care facility (SDCF). SDCFs represent dedicated facilities in close proximity to donor hospitals that are designed to support donor management by OPOs separate from donor hospital and transplant center teams.^26,27 SDCFs may promote efficient donor management and organ procurement, optimize allograft quality, and increase organ utilization, while simultaneously reducing costs and travel risks.^26-29 While use of SDCFs still requires procurement teams to travel to these facilities rather than donor hospitals, there is a trend of increasing utilization of procurement teams located in close proximity to SDCFs, offering another opportunity to reduce or avoid procurement travel.²⁶ In our study, available data precluded identification of OPOs that may have had access to SDCFs, and we were therefore unable to compare outcomes and costs among same-team, different-team, and SDCF team procurements. Future studies should seek to elucidate the relative cost-benefit of different-team versus SDCF team procurements to understand the most financially sustainable means to facilitate safe and efficient procurement of acceptable donor lungs for transplantation.

There are several limitations to our study. As we examined LTx performed at a single, high-volume institution, our findings may not generalize to other programs. However, use of institutional data allowed for granular examination of perioperative outcomes and costs that are unavailable in national datasets. Our center’s donor-related procurement costs highlight several differences in procurement charges between ST and DT groups (Table S1). While this itemization of costs offers valuable insight into factors contributing to differences in procurement charges between ST and DT groups, it is important to note that billing by OPOs is varied, with some charging a lump sum amount and others providing itemized or categorized charges, and may introduce unmeasured bias into the charge comparisons presented. As a large transplant center with extensive procurement resources, the distance at which DT became cost-effective for our center may not apply to other programs. Multi-institutional studies are required to understand the relative travel distances at which centers of variable size and location may find DT most valuable. Additionally, our DT cohort was small, likely limiting our power to detect statistical differences in outcomes and costs, as well as overall comparisons, between DT and ST groups, necessitating critical evaluation of all relationships regardless of statistical significance. Three (12.0%) procurements in the DT cohort were performed within 100 nautical miles of the transplant center, a relatively short distance over which ST may have been less expensive. It is conceivable that given the small size of our DT cohort, unnecessary expense associated with this subset of procurements skewed the donor-specific costs within this group and limited our ability to detect a cost difference between DT and ST groups. The small number of patients in our DT cohort also precluded multivariable analysis to understand the independent effect of procurement team on outcomes and costs. Absence of comprehensive procurement data, including detailed reasons for organ acceptance or decline as well as center-specific reasons for use of DT, precluded our center’s analysis of rates of organ acceptance and decline (dry runs) with ST versus DT. Similarly, costs associated with donor lungs procured by DT or ST that were discarded by recipient center or deemed insufficient quality on arrival were not included in our study’s cost analysis and limits the ability to draw broader conclusions on cost savings. Future studies should seek to obtain comprehensive procurement data to assess rates of organ acceptance and decline (dry runs) for ST versus DT as well as specific reasons a center utilizes DT to further understand how procurement team may impact organ utilization and the financial burden associated with dry runs by same versus different procurement teams. Donor characteristics were not included in the matching algorithm in order to elucidate differences between donors for whom our institution used ST versus DT. Follow up studies in larger patient cohorts should seek to understand how donor characteristics known prior to procurement influence institutional and surgeon-specific decisions to use same versus different procurement teams. Future multi-institutional studies should make provisions for appropriate risk adjustment to understand the financial implications of DT for centers with variable donor and recipient populations and further elucidate the impacts of team and travel logistics on cost differences between ST and DT; use of multi-institutional data may also facilitate development of a predictive model to guide procurement team selection based on travel distance and estimated costs to determine the most fiscally solvent paradigm for individual transplant programs and patients.

CONCLUSIONS

In this single-institution analysis, we found that ST and DT were associated with similar post-transplant outcomes and lung procurement, index hospitalization, and total costs in LTx. Importantly, DT offered significant cost-savings when procurement travel distances exceeded approximately 363 nautical miles. Amidst a national increase in lung procurement travel distances and costs, our findings may offer LTx programs preliminary understanding of the financial implications of DT, and inform the most cost-effective use of these resources. Further investigation in a multi-institutional study is warranted to better understand how DT improves the efficiency, cost-effectiveness, and safety of lung procurement in an era of broader sharing.

Supplementary Material

1

Figure S1. Flow diagram of LTx recipients stratified by same-team transplantation (ST) versus different-team transplantation (DT) and matched.

2

Figure S2. Total lung transplant costs stratified by same-team transplantation (ST) versus different-team transplantation (DT). Total costs represent the sum of donor-specific (lung procurement) and recipient-specific (total index hospitalization) costs.

CENTRAL MESSAGE

At our single high-volume institution, DT and ST were associated with similar post-lung transplant outcomes and DT offered cost-savings with increasing procurement travel distance.

PERSPECTIVE STATEMENT

To understand whether different-team transplantation (DT, procuring team from non-recipient center) represents a safe means to reduce lung transplant (LTx) costs, we compared post-transplant outcomes and lung procurement and index hospitalization costs among matched DT and same-team transplantation (ST, procuring team from recipient center) cohorts at a single, high-volume institution.

CENTRAL PICTURE

Compared to ST, DT may reduce costs associated with long distance lung procurement.

ABBREVIATIONS

CI

confidence interval

CLAD

chronic lung allograft dysfunction

DSA

donation service area

DT

different-team transplantation

ECMO

extracorporeal membrane oxygenation

FEV1

forced expiratory volume in 1 second

FVC

forced vital capacity

ICU

intensive care unit

LTx

lung transplantation

LAS

lung allocation score

LOS

length of stay

OPO

organ procurement organization

P/F

PaO₂/FiO₂

PGD3

grade 3 primary graft dysfunction

PHS

Public Health Service

SDCF

specialized donor care facility

ST

same-team transplantation

UNOS

United Network for Organ Sharing