Cost-effectiveness of intestinal transplantation for adult patients with
intestinal failure: a simulation study1–4
Anne Margot Roskott, Henk Groen, Edmond HHM Rings, Jan Willem Haveman, Rutger J Ploeg, Mireille J Serlie,
Geert Wanten, Paul FM Krabbe, and Gerard Dijkstra
ABSTRACT
Background: Home parenteral nutrition (HPN) and intestinal trans-
plantation (ITx) are the 2 treatment options for irreversible intesti-
nal failure (IF).
Objective: This study simulated the disease course of irreversible
IF and both of these treatments—HPN and ITx—to estimate the
cost-effectiveness of ITx.
Design: We simulated IF treatment in adults as a discrete event
model with variables derived from the Dutch Registry of Intestinal
Failure and Intestinal Transplantation, the Intestinal Transplant Reg-
istry, hospital records, the literature, and expert opinions. Simulated
patients were enrolled at a rate of 40/mo for 10 y. The maximum
follow-up was 40 y. Survival was simulated as a probabilistic func-
tion. ITx was offered to 10% of patients with ,12 mo of remaining
life expectancy with HPN if they did not undergo ITx. Costs were
calculated according to Dutch guidelines, with discounting. We
eval- uated the cost-effectiveness of ITx by comparing models
conducted with and without ITx and by calculating the cost
difference per life- year gained [incremental cost-effectiveness ratio
(ICER)].
Results: The average survival was 14.6 y without ITx and 14.9 y
with ITx. HPN costs were V13,276 for treatment introduction, fol-
lowed by V77,652 annually. The costs of ITx were ∼V73,000 dur-
ing the first year and then V13,000 annually. The ICER was
V19,529 per life-year gained.
Conclusion: Our simulations show that ITx slightly improves
survival of patients with IF in comparison with HPN at an
additional cost of V19,529 per life-year gained. Am J Clin Nutr
2015;101:79–86.
Keywords intestinal failure, home parenteral nutrition, intestinal
transplantation, cost-effectiveness, simulation model
INTRODUCTION
Intestinal failure (IF)5 is defined as the reduction in func-
tioning gut mass below the minimal amount necessary to main-
tain an acceptable nutritional status with an enteral diet. This
condition can result from surgery, functional bowel disease, or a
congenital defect.
Short bowel syndrome is the most common cause of IF and
occurs when 70–75% of small bowel length has been lost due to
previous surgery (1–4). IF becomes irreversible when the
gastrointestinal tract is unable to compensate for loss of length,
function, or both. Until the advent of total parenteral (in-
travenous) nutrition (TPN) in the 1960s, irreversible IF was
considered incompatible with life. TPN administered at
home— home parenteral nutrition (HPN)—is the main
treatment of ir- reversible IF.
Irreversible IF is rare, with a reported prevalence in Europe
ranging from 1.1 to 12.7 per million adults. The Dutch
prevalence was reported to be 3.7 per million adults in 1997
(5). Recent results from the Dutch Registration of Intestinal
Failure and Intestinal Transplantation (DRIFT) point to a
prevalence of 10.3 per million on January 2013, which
indicates a considerable increase over the past 16 y.
HPN allows patients with IF to live outside the hospital and
meet their nutritional requirements, although serious complica-
tions often occur, such as parenteral nutrition–associated liver
disease, vascular access problems, sepsis, and metabolic de-
rangement. Patient survival mainly depends on the underlying
disease. IF, including HPN-related complications and
associated social restrictions, often leads to substantial
psychosocial dis- tress and impaired quality of life (QoL) (6,
7).
From a socioeconomic perspective, the health care burden of
irreversible IF is enormous. The estimated annual costs for
maintenance nutritional support differ widely: reported
amounts vary between $75,000 and $300,000 per patient (8, 9).
Intestinal transplantation (ITx) is a relatively new therapeutic
alternative to HPN (10). For patients with IF, the prognosis
with HPN—a 5-y survival rate of 75–85% (11, 12)—is superior
to the
1
From the Departments of Gastroenterology and Hepatology (AMR and
GD), Surgery (AMR, JWH, and RJP), Epidemiology (HG and PFMK), and
Pediatrics (EHHMR), University of Groningen, University Medical Center
Groningen, Groningen, The Netherlands; the Department of Endocrinology
and Metabolism, University of Amsterdam, Academic Medical Center,
Am- sterdam, The Netherlands (MJS); and the Department of
Gastroenterology and Hepatology, University of Nijmegen, Radboud
University Nijmegen Medical Center, Nijmegen, The Netherlands (GW).
2
The authors reported no funding received for this study.
3
AMR and HG contributed equally to this work.
4
Address correspondence to AM Roskott, Department of Gastroenterol-
ogy and Hepatology, University of Groningen, PO Box 30001, 9700 RB
Groningen, The Netherlands. E-mail: a.m.roskott@umcg.nl.
5
Abbreviations used: DRIFT, Dutch Registry of Intestinal Failure and
In- testinal Transplantation; HPN, home parenteral nutrition; ICER,
incremental cost-effectiveness ratio; IF, intestinal failure; ITR, Intestinal
Transplant Reg- istry; ITx, intestinal transplantation; TPN, total parenteral
nutrition; QALY, quality-adjusted life-year; QoL, quality of life.
Received January 7, 2014. Accepted for publication October 10, 2014.
First published online November 12, 2014; doi: 10.3945/ajcn.114.083303.
Am J Clin Nutr 2015;101:79–86. Printed in USA. © 2015 American Society for Nutrition 79
prognosis after ITx, which yields a 3- to 5-y survival of 69% and
56%, respectively (13). Most recently, however, worldwide 5-y
survival rates from the Intestinal Transplant Registry (ITR) ap-
proached 70%. ITx should be reserved for patients with IF with
life-threatening complications from HPN, the so-called failure of
HPN (14). For these “high risk” patients, survival clearly
improves after ITx in comparison to continued HPN, and these
patients have high mortality rates while on the waiting list for
ITx (14).
Although a few studies discuss the costs of HPN and ITx (9,
15), no study has definitively established the cost-effectiveness
of ITx (8). Extensive and exact data on costs are lacking, but
according to the literature, ITx may become cost-effective after
∼1–2 y in patients who retain their functional grafts (9).
In The Netherlands, the University Medical Center Groningen
is the national center for ITx, where this procedure has been
performed since 2001 in adults and children. HPN is provided
by 2 specialized centers for children and adults (Amsterdam
Medical Center and University Medical Center Nijmegen, St.
Radboud) throughout the country. The aim of this study was to
evaluate the disease course of irreversible IF and the treatment
effects of HPN and ITx and to estimate cost-effectiveness of ITx
on the basis of national and international data by using a
simulation model.
METHODS
Model structure
We simulated the treatment options for irreversible IF in
adults in a cohort-based discrete event model. Discrete event
modeling requires the clinical course of patients to be
represented as a series of consecutive steps or states. The time
spent in the various states is programmed in the model as fixed
durations or as time- dependent probabilities, depending on what
is the best ap- proximation of the actual process being modeled.
In addition, treatment choices can be modeled by fixed
probabilities or by comparing multiple time-dependent
probabilities.
FIGURE 1 States (indicated by boxes) and transitions (indicated by arrows) in the model for treatment of adults with IF consisting of HPN and ITx.
Adult_IF_dead, deceased adult; Adult_IF_HPN, adult requiring home parenteral nutrition; Adult_IF_HPNreq, transition into permanent home parenteral
nutrition requirement state; Adult_IF_ITx, adult undergoing intestinal transplantation; Adult_IF_ITx_fail, adult after graft failure; Adult_IF_postITx, adult
We modeled the treatments HPN and ITx, with subsequent
graft failure or sustained graft function, and death as the final
state (Figure 1). The amount of time that the simulated patients
re- mained in each treatment state (transition time) was
determined by probabilistic Weibull functions. These functions
resemble survival curves, and their shape was based on the
available data. For each patient, the duration of a certain state
was randomly drawn from the distribution of the respective
Weibull function. In case of multiple possible outcomes,
transition times were determined independently for each
outcome and the shortest time determined the outcome.
Selection for ITx was determined by a fixed percentage of
patients with a remaining life expectancy with HPN of ,12 mo.
Maximum simulated follow-up was 40 y, in view of the
average age of adult patients with irreversible IF (median age
of adults in the DRIFT registry of 54 y). During execution of
the model, detailed information was collected about time spent
in the var- ious states in the model. To produce stable estimates,
the model was replicated 500 times. In each replication of the
model, all outcomes were calculated for the entire cohort of
included pa- tients. The model was constructed with the use of
advanced and generic modeling software (AnyLogic
Company).
Model variables
We derived variables for our model from the International
ITR (patient and graft survival), the DRIFT (patient inflow,
HPN survival), hospital records, expert opinions from Dutch
HPN centers and the ITx center (Table 1), and from the
literature (3, 8, 9, 16, 17, 22). Although the number of new IF
patients in The Netherlands was estimated at 4/mo, the
simulations were per- formed with an enrollment rate of 40/mo,
which means that we simulated a larger recruitment area of
perhaps several European countries. This increase in number of
patients does not affect absolute results, because the results are
represented as the dif- ference between the situations of HPN
treatment without ITx and
after intestinal transplantation; HPN, home parenteral nutrition; HPN_dead, transition from home parenteral nutrition to death; HPN_ITx, transition from
home parenteral nutrition to intestinal transplantation; IF, intestinal failure; ITx, intestinal transplantation; ITx_dead, transition from intestinal transplantation
to death; ITx_postITx, transition from intestinal transplantation to post-intestinal transplantation state; postITx_dead, transition from intestinal transplantation
follow-up to death; postITx_fail, transition from post-intestinal transplantation to graft failure.
 TABLE 1
Model variables for irreversible IF in adults1

Variable Value Source

Annual inflow of IF patients, n 480 DRIFT, HPN centers, expert opinion
Proportion indication for ITx, % 10 DRIFT, expert opinion
Remaining life expectancy with HPN for eligibility for ITx, mo 12 Expert opinion
Costs, V
HPN onset: total 13,276
Hospital admission 6083 HPN centers
Training by nurse 825 HPN centers
CVC placement 1047 Reference 22
HPN administration 1050 HPN centers
Home care 4271 HPN centers
HPN, yearly maintenance, total 77,652
Hospital admission for complications (7 d) 4258 DRIFT, HPN centers
HPN administration 54,750 HPN centers
CVC replacement 3664 Reference 22
Laboratory tests 100 HPN centers
Outpatient visits 390 HPN centers
Home care 14,441 HPN centers
DXA scan 49 HPN centers
ITx, first year, total 72,332
Transplant surgery 14,658 Reference 17
Hospital admission 28,360 Reference 17
Medication 9120 ITx protocol UMCG
Interventions 20,194 Reference 17
ITx, subsequent annual costs, total 15,183 Reference 17
Hospital admission 3650 Reference 9
Medication 9120 ITx protocol Groningen
Interventions/diagnostics 2140 ITx protocol Groningen
Outpatient visits 273 ITx protocol Groningen
1
CVC, central venous catheter; DRIFT, Dutch Registry of Intestinal Failure and Transplantation; DXA, dual-energy
X-ray absorptiometry; HPN, home parenteral nutrition; IF, intestinal failure; ITx, intestinal transplantation; UMCG, Uni-
versity Medical Center Groningen.

HPN with ITx. The larger number of patients does, however, products than does liver transplantation. We modeled these
improve accuracy. Over a simulation period of 10 y, 4800 extra costs. Then, we calculated the costs of follow-up after the
patients were enrolled. After an initial period of HPN, ITx was first year assuming 2 outpatient visits/y (including physician
offered to 10% of patients with a remaining life expectancy with costs) and 6 d of hospital admissions/y (9, 21), in addition to
HPN of evaluations of intestinal function including endoscopies, as
,12 mo. We also evaluated outcomes when higher percentages of well as evalua- tions of kidney function and bone densitometry,
patients (15% and 20%) were offered ITx and in the case of all according to the local University Medical Center Groningen
longer remaining survival expectancy (18, 24, and 36 mo). protocol. Costs of
Survival with HPN was based on data from the literature (11,
12, 18), whereas the Weibull curves for graft and patient
survival after ITx were based on ITR data and extrapolated to 40
y (Figure 2). Survival after graft failure was estimated on the
basis of the difference between graft survival and patient
survival after ITx. In view of the very short waiting list for an
adult donor organ (19), the probability of dying while waiting
was considered negligible.

Costs
We calculated costs according to Dutch national guidelines at
the price level of 2012 (20) and indexed pricing by using the
Dutch consumer price index (www.cbs.nl). The costs of trans-
plantation were based on a recent publication by van der Hilst et
al. (17), which reported the costs of adult liver transplantation
FIGURE 2 Survival curves applied in the simulation model for patient
until 1 y after surgery. The procedure and the costs associated survival with HPN, survival after ITx, graft survival after ITx, and survival
with liver transplantation are comparable to ITx. However, with HPN after graft failure. HPN, home parenteral nutrition; ITx,
according to the expert opinion from surgeons of the Dutch ITx intestinal transplantation.
center, ITx requires longer surgery and more use of blood
immunosuppressive medication were considered to remain the
same after the first year. Costs of HPN onset were calculated
separately from the annual costs of continuous HPN for 7 d/wk.
Costs for onset included a hospital admission for HPN training,
nursing staff time for training, physician costs, and initial HPN
consumables, as well as central venous catheter placement and
initial home care support. We derived these costs from a com-
prehensive evaluation of TPN by Olieman et al (22) with ad-
aptations for the adult situation, mainly the quantity of HPN
products used.
Cost-effectiveness
To estimate life-years with HPN and after ITx, we added up
the numbers of patients in the HPN and ITx states per year. We
calculated costs per month and per year by using the data on
numbers of patients starting on HPN, numbers of patients re-
ceiving ITx, and numbers of patients in follow-up after the start
of HPN or ITx. To account for the depreciation of future costs
and effects, we applied discounts to the annual data for life-
years and costs on the basis of the economic principle that
benefits are worth more today than in the future because of the
preference to spend now rather than later and because there may
be improvements on interventions in the near future. This
principle can be applied to costs as well as to health outcomes
(23). We discounted future costs and effects by 4.0% and 1.5%,
respectively. For comparison with other countries, we applied
uniform (uniform for costs and effects) discount rates of 3%,
4%, and 5%.
We evaluated the cost-effectiveness of ITx by comparing
life-years and costs of 500 replications of the model with and
without ITx as a treatment option. We combined model outputs
with regard to life-years and numbers of patients with costs to
calculate the incremental cost-effectiveness ratio (ICER)—
defined as the difference in costs divided by the difference in
life-years comparing the alternative intervention (ITx) to the
standard (HPN). The ICER expresses the costs for the gain of a
life-year.
FIGURE 3 Numbers of patients over time in various states in the sim-
ulation model. Black lines represent cumulative numbers of patients starting
HPN (continuous black line) and after ITx (dashed black line). The red
dashed line represents the cumulative number of deceased patients. Blue
lines represent numbers of patients receiving HPN (dashed line) and ITx
(dashed-dotted line); the latter is plotted on the right y axis. HPN, home
parenteral nutrition; ITx, intestinal transplantation.
RESULTS
Survival
All of the patients entered our simulation model over a
period of 10 y. The number of patients receiving HPN was only
slightly smaller than in simulations without ITx (Figure 3).
HPN patient
numbers decreased steadily due to ITx or death. The number of
patients alive after ITx peaked after ∼13 y at 71 patients (16.7%
of the average number of 425 patients undergoing ITx) and
then
also declined (Figure 3, right y axis). The cumulative number
of ITx patients increased over the entire simulation period. At
the end of the simulation period of 40 y, there were still patients
left in both treatment groups: 595 patients remained on HPN
(12.4% of the 4800 enrolled) and 18 patients remained in the
ITx state (4.2%). The average number of deceased patients at
the end of the simulations was 4187 (87% of the total enrolled).
The number of transplants ranged from 362 to 480 between the
model replications. In this model, the average survival was 14.6
y without ITx and 14.9 y with ITx.
Costs
The initial costs of HPN were V13,276, including clinical
training and the placement of a central venous catheter. The
average annual costs for HPN maintenance were V77,652
(Table 1). These annual costs include HPN products, HPN
nurse sup- port, outpatient follow-up, home care, and hospital
admission in case of HPN-related complications. The cost of
ITx was V72,332 during the first year, including screening, the
transplant procedure, and clinical and outpatient follow-up. The
sub- sequent average annual cost was V15,183, which mainly
arose from immunosuppression, rehospitalization, and
outpatient follow- up (Table 1).
Cost-effectiveness
Both total costs (V44.57 million) and life-years (1525 y)
increased with ITx (10%) compared with HPN (Table 2,
undiscounted). Total costs of HPN decreased by just over 11
million euros (0.2% of total costs) when ITx was introduced.
However, in this situation,
total costs of ITx amounted to V55 million. The combination of
these outcomes resulted in an estimated V29,223 for each addi-
tional life-year gained by ITx. When we applied discounts, this
estimate decreased to V19,529 per life-year.
HPN cost calculations were based on an assumed full need
for HPN 7 d/wk. However, because some patients with
irreversible IF only need HPN a few days a week, we may have
overestimated HPN costs. Calculations with HPN of 4 d/wk
instead of 7 resulted
in an increase in the discounted costs per life-year by V2189.
When we selected higher percentages of patients for ITx, the
numbers of transplantations, mean survival, life-years, and total
costs proportionally increased, whereas costs of TPN propor-
tionally decreased (Table 2). The net result was that
undiscounted costs per life-year remained virtually unchanged.
After dis- counting, the cost-effectiveness with increased ITx
percentages slightly improved, probably attributable to the fact
that the life- years were less affected by the discounting
procedure than were the costs.
Because the expected survival with HPN for ITx candidates
might in fact be longer than 12 mo, analyses using expected
 TABLE 2
Effects and costs of treatment of irreversible IF in adults with and without ITx1
Outcome Without ITx With 10% ITx With 15% ITx With 20% ITx
Patients, n
HPN 4800 4800 4800 4800
ITx 0 425 637 845
Deaths 4202 4187 4177 4168
Life-years (undiscounted)
HPN 70,205 70,082 69,996 69,929
ITx 0 1648 2474 3291
Total 70,205 71,730 72,470 73,220
Mean survival, y 14.6 14.9 15.1 15.3
Costs (undiscounted), V
HPN 5,536,021,048 5,524,971,168 5,518,562,704 5,513,264,042
ITx 0 55,619,183 83,435,940 110,806,446
Total 5,536,021,048 5,580,590,351 5,601,998,644 5,624,070,488
Difference vs. no ITx (undiscounted)
Life-years — 1525 2265 3015
Costs, V — 44,569,304 65,977,596 88,049,440
ICER, V/life-year — 29,223 29,125 29,205
Life-years (discounted at 1.5%)
HPN 55,204 55,099 55,024 54,963
ITx 0 1251 1876 2496
Total 55,204 56,350 56,900 57,459
Costs (discounted at 4%), V
HPN 3,073,275,575 3,066,326,713 3,061,660,272 3,057,583,331
ITx 0 29,329,570 43,976,585 58,379,144
Total 3,073,275,575 3,095,656,283 3,105,636,857 3,115,962,475
Difference vs. no ITx (discounted)
Life-years — 1146 1696 2255
Costs, V — 22,380,709 32,361,283 42,686,901
ICER, V/life-year — 19,529 19,078 18,930
1
Average patient numbers and total costs and life-years of all patients aged .40 y are shown. HPN, home parenteral
nutrition; ICER, incremental cost-effectiveness ratio (costs per life-year gained); IF, intestinal failure; ITx, intestinal
transplantation.

survival times of 18, 24, and 36 mo were also conducted (Table the decrement in survival. With discounting, the increment of
3). Without discounting, the increment of prognosis with HPN prognosis with HPN slightly increased the ICER. The
resulted in lower costs per life-year, and with higher ITx per- decrement of survival relatively outweighs the decrement in
centages. The decrement in costs in this situation outweighed costs in this situation.

TABLE 3
Sensitivity analysis of prognosis of ITx candidates receiving HPN if not transplanted1
With 10% ITx With 15% ITx With 20% ITx

Outcome 12 mo 18 mo 18 mo 24 mo 24 mo 36 mo
Undiscounted
Life-years 71,730 71,694 72,317 72,261 72,774 72,729
Costs, V 5,580,590,351 5,577,636,23 5,589,174,35 5,584,369,096 5,587,351,68 5,582,806,22
1 3 2 6
Difference vs. no ITX
Life-years 1525 1490 2112 2057 2569 2524
Costs, V 44,569,304 41,615,183 53,153,305 48,348,048 51,330,634 46,785,178
ICER, V/life-year 29,223 27,939 25,169 23,508 19,980 18,573
Discounted
Life-years (discount rate 1.5%) 56,350 56,340 56,807 56,804 57,178 57,225
Costs (discount rate 4%), V 3,095,656,283 3,096,049,79 3,101,689,885 3,104,139,627 3,103,762,180 3,111,156,746
3
Difference vs. no ITX
Life-years 1146 1136 1603 1600 1974 2021
Costs, V 22,380,709 22,774,218 28,414,310 30,864,053 30,486,606 37,881,171
ICER, V/life-year 19,529 20,043 17,724 19,284 15,442 18,743
1
HPN, home parenteral nutrition; ICER, incremental cost-effectiveness ratio (costs per life-year gained); ITx, intestinal transplantation.
 When we applied uniform discount rates for costs and effects
of 3%, 4%, and 5%, this resulted in slightly less favorable
ICERs. Applying a higher discount rate to effects clearly
diminished the value of long-term gain in life-years with ITx.
Variations in costs of ITx and HPN had modest effects on the
overall cost-effectiveness (Table 4). Increasing costs of ITx
by
50% increased the costs per life-year by ∼60%, whereas a similar
increase in costs of HPN decreased the cost-effectiveness of ITx
by ∼10%. Simultaneous increases in costs of ITx and HPN by
50% and 20%, respectively, yielded outcomes for costs per life-
year reflecting the counteracting effect of these changes.
DISCUSSION
The introduction of new modalities and treatment regimen in
health care requires not only a demonstration of clinical supe-
riority but also insight into costs for society. Especially in times
of austerity, balancing medical innovation and patient needs
with prioritization of reimbursement has become standard
practice for national health care budgets. The best example in
transplantation, which shows superiority of treatment, enhanced
QoL, and sig- nificant reduction in cost is the comparison
between hemodialysis and kidney transplantation, favoring the
latter. The medical situation concerning patients who receive
either permanent HPN or ITx is comparable but more complex
because of the fact that clinical superiority and the effect on
QoL are actually less clear. To our knowledge, no
comprehensive comparison of the 2 existing treatment
modalities for irreversible IF in adults has been published.
This simulation study enhances insight into the impact of ITx
and its cost-effectiveness compared with HPN. According to our
simulations, ITx improves patient survival by 3 mo over a 40-y
follow-up when restricted to patients with a life expectancy of
12 mo who do not undergo transplantation, which is the actual
situation to date. The absolute survival gain of 3 mo in this
model may seem marginal. However, we have to consider that in
this model the survival benefit of the 10% of transplanted
patients is distributed over a large simulated population of
whom 90% did not undergo ITx. Therefore, the appreciated
survival benefit of an
individual patient undergoing ITx will, in fact, be longer than 3
mo. In our model, the cost per life-year gained was ∼V30,000.
This amount is similar to liver transplantation and lower than
for
heart and lung transplantation (24, 25). Studies reporting on the
costs of HPN/TPN and ITx are scarce, although these therapies
account for a great impact on the economic health care burden
TABLE 4
Sensitivity analysis of changes in costs of ITx and HPN (undiscounted)1
ITx 150%
Costs, V
HPN 5,524,971,168
ITx 83,428,775
Total 5,608,399,943
Difference vs. HPN
Life-years 1525
Costs, V 72,378,895
ICER, V/life-year 47,457
in The Netherlands, as well as in other developed countries
worldwide. Our study confirms the high costs estimated for
HPN therapy. Costs for HPN are composed of the continual
need for high-priced nutritional infusion products and (to a
lesser extent) by the costs for the onset of this therapy and for
hospitalization in case of complications of HPN therapy. Total
costs for ITx are composed mainly of hospitalization during the
first year after the procedure, of the costs the procedure itself,
and reinterventions, as well as lifelong continuous costs of
expensive immunosup- pressive medication.
For this study, we assumed that 10% of patients met the in-
dications for ITx and would not survive beyond 1 y without
ITx. Therefore, in our model, this percentage of patients
progressed from the HPN state into the ITx state. This
relatively small percentage reflects a strict handling of
indications for ITx, which is the case in The Netherlands. But
because indication rates have been increasing during the past
decade, especially in high-volume centers, we also evaluated
outcomes with higher percentages (14, 16, 19, 26). When 15%
and 20% of patients were selected for ITx, it improved survival
and incurred higher costs with no substantial change in
cost-effectiveness. This finding economically supports the
further widening of the indication for ITx.
It is difficult to estimate the prognosis of an ITx candidate
receiving HPN: allowing patients with a better prognosis
results in higher percentages of selected ITx candidates. This
prognosis could be longer than 12 mo. The sensitivity analysis
of the es- timated prognosis of the ITx candidate shows a
decrease in ICER with extension of the prognosis without
discounting and a mar- ginal increase in the ICER with
discounting. Therefore, the percentage of candidates selected
for ITx (and their prognosis with HPN) seems not to have a
major impact on the cost-effectiveness of ITx.
Although average costs of HPN exceed those of ITx within 2
y after the start of treatment, we cannot simply compare the 2
treatment options for the individual patient. There are 3 main
reasons for this. First, ITx is now only offered to patients who
would otherwise die within 12 mo, which would obviously be
far most beneficial financially. Second, for each patient, the
costs made for HPN before the ITx procedure is performed
have to be taken into account. In this model, a percentage of
patients went into the ITx status right from the start of a 40-y
follow-up period. However, most patients will be in the more
expensive TPN state before ITx is offered. Third, we cannot
limit our comparison to the pure economic aspect of IF; the
value of QoL with a certain treatment has to be taken into
account. However, it is very
HPN 150% ITx + HPN 150% ITx + HPN 120%
8.287.456.752 8,287,456,752 6,629,965,402
55.619.183 83,428,775 66,743,020
8.343.075.936 8,370,885,527 6,696,708,422
1525 1525 1525
39,044,364 66,853,955 53,483,164
25,600 43,834 35,068
 1
HPN, home parenteral nutrition; ICER, incremental cost-effectiveness ratio (costs per life-year gained); ITx, intestinal
transplantation.
 pated in writing the manuscript; and PFMK: participated in research design,
difficult—if not impossible—to compare the personal objective
performing the research, data analysis, and writing the manuscript. All au-
valuation of one patient with the opinion of another. The opinion
on QoL of a patient with failure of HPN treatment (awareness of
dying soon without transplantation) is not comparable to the
valuation of QoL by a patient having successful HPN treatment.
Including the aspect of QoL (index type of measure) in our
modulation was impossible because of the lack of available data
in the HPN/ITx population.
In this study, we made calculations on the basis of direct costs
only. From our own studies reporting on QoL, we know that
43% of HPN patients are unable to work and only 14% are eco-
nomically active in society (6). Most survivors after ITx re-
integrate into society, with a self-sustained socioeconomic status
(16). However, frameworks for incorporating indirect costs of
work absence or resumption are limited in this situation of IF
with long-term absence. The friction cost method only allows
costs up
to a maximum period of ∼6 mo. The human capital method
assumes lifelong friction costs and is therefore generally con-
sidered to overestimate costs of productivity loss (23). We did
include the costs for home care in the calculation of costs on
HPN on the basis of the expert opinion that 50% of patients
receive home care, which contributed considerably to total costs
(27). Finally, our results remain an assumption because they are
based on a model that can only approximate reality.
At the most recent International Small Bowel Transplant
Symposium in Oxford, United Kingdom, June 2013, it became
clear that QoL, patient valuation, and an active patient role in
the discussion about ITx are being considered and studied. We
were not able to adjust life-years for quality of life in our model
because no data exist. As an approximation, we used data from a
recent evaluation of cost-effectiveness of preservation methods
in kidney transplantation (28). Assuming a utility of 0.66 on
HPN and 0.9 after ITx, the overall utility of a life-year would be
,0.9, so with costs remaining unchanged, the costs per
quality-adjusted life-year (QALY) gained would be higher than
the costs per life-
year (V31,787 undiscounted). However, due to the favorable
gain of QoL after ITx, QALYs after ITx would constitute a
higher proportion of total QALYs (3.1%) than without adjust-
ment for QoL (2.3%). In the near future, we can imagine the
broadening of indications if the expected improved QoL after
ITx compared with HPN (16, 29) can be confirmed and if
patient valuation and opinion are allowed to play a role in
indicating ITx. Taking this into account and a possible further
improve- ment of outcome after ITx, one could envisage
offering ITx as a rehabilitative long-term treatment option to
all patients with irreversible IF. In conclusion, our simulation
model shows that ITx slightly improves life-years in patients
with irreversible IF at relatively low cost.
We thank Cora Jonkers (Nutritional Support Team, Amsterdam Medical
Center, The Netherlands) for sharing her expert opinion on nutritional
support and HPN specifically; Max Marquez (Secretary of the ITR) for
providing the ITR data; and Robert Borgers (Manager, Department of
Gastroenterology, University Medica Center Groningen, The Netherlands)
for his help and data on clinical costs.
The authors’ responsibilities were as follows—AMR, HG, EHHMR, and
GD: participated in research design, collection of data, data analysis, per-
forming the research, and writing the manuscript; JWH, MJS, and GW:
participated in collection of data and writing the manuscript; RJP: partici-
thors read and approved the final manuscript. None of the authors had Interdisciplinary management of infantile short bowel syndrome:
a conflict of interest to declare. resource consumption, growth, and nutrition. J Pediatr Surg
2010;45:490–8.

REFERENCES
1. Goulet O. Recent studies on small intestinal transplantation. Curr
Opin Gastroenterol 1997;13:500–9.
2. Bines JE. Intestinal failure: a new era in clinical management. J Gas-
troenterol Hepatol 2009;24(Suppl 3):S86–92.
3. Wanten G, Calder PC, Forbes A. Managing adult patients who need
home parenteral nutrition. BMJ 2011;342:d1447.
4. Schurink M, Nieuwenhuijs V, Hulscher JB, Heineman E, Rings
EHHM. Outcome of an intestinal rehabilitation program for children
with short bowel syndrome. Ned Tijdschr Geneeskd 2012;156:A4690
5. Bakker H, Bozzetti F, Staun M, Leon-Sanz M, Hebuterme X,
Pertkiewicz M, Shaffer J, Thul P; ESPEN-Home Artificial Nutrition
Working Group. Home parenteral nutrition in adults: a European
multicentre survey in 1997. Clin Nutr 1999;18:135–40.
6. Roskott AMC, Huisman-de Waal G, Wanten GJ, Jonkers-Schuitema
C, Serlie MJ, Baxter JP, Hoekstra-Weebers JEHM. Screening for psy-
chosocial distress in patients with long-term home parenteral
nutrition. Clin Nutr 2013;32:396–403.
7. Huisman-de Waal G, Schoonhoven L, Jansen J, Wanten G, van
Achterberg
T. The impact of home parenteral nutrition on daily life-a review. Clin
Nutr 2007;26(3):275–88.
8. Evans RW. Financial, economic and insurance issues pertaining to
intestinal transplantation: when is too much not enough? In: Langnas
A, Goulet O, Eamonn M, editors. Intestinal failure: diagnosis, man-
agement and transplantation. Oxford (United Kingdom): Blackwell
Publishing; 2008. p. 363–77.
9. Sudan D. Cost and quality of life after intestinal transplantation.
Gastroenterology 2006;130(2):S158–62.
10. Fishbein TM. Intestinal transplantation. N Engl J Med 2009;361(10):
998–1008.
11. Pironi L, Joly F, Forbes A, Colomb V, Lyszkowska M, Baxter J, Gabe
S, He´buterne X, Gambarara M, Gottrand F, et al. Long-term
follow-up of patients on home parenteral nutrition in Europe:
implications for intestinal transplantation. Gut 2011;60:17–25.
12. Messing B, Crenn P, Beau P, Boutron-Ruault MC, Rambaud JC, Ma-
tuchansky C. Long-term survival and parenteral nutrition dependence
in adult patients with the short bowel syndrome. Gastroenterology
1999;117(5):1043–50.
13. Garg M, Jones RM, Vaughan RB, Testro AG. Intestinal
transplantation: current status and future directions. J Gastroenterol
Hepatol 2011;26:1221–8.
14. Abu-Elmagd K, Bond G, Reyes J, Fung J. Intestinal transplantation:
a coming of age. Adv Surg 2002;36:65–101.
15. Longworth L, Young T, Beath SV, Kelly DA, Mistry H, Protheroe
SM, Ratcliffe J, Buxton MJ. An economic evaluation of pediatric
small bowel transplantation in the United Kingdom. Transplantation
2006;82:508–15.
16. Abu-Elmagd KM, Kosmach-Park B, Costa G, Zenati M, Martin L,
Koritsky DA, Emerling M, Murase N, Bond GJ, Soltys K, et al.
Long- term survival, nutritional autonomy, and quality of life after
intestinal and multivisceral transplantation. Ann Surg
2012;256:494–508.
17. van der Hilst CS, Ijtsma AJ, Bottema JT, van Hoek B, Dubbeld J,
Metselaar HJ, Kazemier G, van den Berg AP, Porte RJ, Slooff MJ.
The price of donation after cardiac death in liver transplantation: a
pro- spective cost-effectiveness study. Transpl Int 2013;26:411–8.
18. Messing B, Lemann M, Landais P, Gouttebel MC,
Ge´rard-Boncompain M, Saudin F, Vangossum A, Beau P,
Gue´don C, Barnoud D, et al. Prognosis of patients with
nonmalignant chronic intestinal failure re- ceiving long-term home
parenteral nutrition. Gastroenterology 1995; 108:1005–10.
19. Smith JM, Skeans MA, Thompson B, Horslen SP, Edwards EB,
Harper AM, Snyder JJ, Israni AK, Kasiske BL. OPTN/SRTR 2011
annual data report: intestine. Am J Transplant 2013;13(Suppl
1):103–18.
20. Oostenbrink J, Koopmanschap M, Rutten F. Manual for cost studies,
methods and standard prices for economic evaluation in health care.
Amstelveen (The Netherlands): Dutch Health Care Council; 2004.
21. Andersen DA, Horslen S. An analysis of the long-term complications
of intestine transplant recipients. Prog Transplant 2004;14:277–82.
22. Olieman JF, Poley MJ, Gischler SJ, Penning C, Escher JC, van den
Hoonaard TL, van Goudoever JB, Bax NM, Tibboel D, IJsselstijn H.

23. Drummond MF, Sculpher MJ, Torrance GW, O’Brien BJ, Stoddart GL.
Methods for the economic evaluation of health care programmes. 3rd
ed. Oxford (United Kingdom): Oxford University Press; 2005.
24. Ouwens JP, van Enckevort PJ, TenVergert EM, Bonsel GJ, van der Bij
W, Haagsma EB, Rutten FF, Slooff MJ, Koe¨ter GH. The cost
effec- tiveness of lung transplantation compared with that of heart and
liver transplantation in the Netherlands. Transpl Int 2003;16:123–7.
25. Groen H, van der Bij W, Koeter GH, TenVergert EM. Cost-effective-
ness of lung transplantation in relation to type of end-stage pulmonary
disease. Am J Transplant 2004;4:1155–62.
26. Abu-Elmagd KM, Costa G, Bond GJ, Soltys K, Sindhi R, Wu T,
Koritsky DA, Schuster B, Martin L, Cruz RJ, et al. Five hundred in-
testinal and multivisceral transplantations at a single center: major
advances with new challenges. Ann Surg 2009;250:567–81.
27. Pironi L, Goulet O, Buchman A, Messing B, Gabe S, Candusso M,
Bond G, Gupte G, Pertkiewicz M, Steiger E. Outcome on home par-
enteral nutrition for benign intestinal failure: a review of the literature
and benchmarking with the European prospective survey of ESPEN.
Clin Nutr 2012;31:831–45.
28. Groen H, Moers C, Smits JM, Treckmann J, Monbaliu D, Rahmel A,
Paul A, Pirenne J, Ploeg RJ, Buskens E. Cost- effectiveness of
hypothermic machine preservation versus static cold storage in renal
transplantation. Am J Transplant 2012;7: 1824–30.
29. Pironi L, Baxter JP, Lauro A, Guidetti M, Agostini F, Zanfi C, Pinna
AD. Assessment of quality of life on home parenteral nutrition and
after intestinal transplantation using treatment-specific questionnaires.
Am J Transplant 2012;12(Suppl 4):S60–6.