Clinical Nutrition 32 (2013) 368e374
`
`Contents lists available at SciVerse ScienceDirect
`
`Clinical Nutrition
`
`j o u r n a l h o m e p a g e : h t t p : / / w w w . e l s e v i e r . c o m / l o c a t e / c l n u
`
`Original article
`Determinants of home parenteral nutrition dependence and survival of 268
`patients with non-malignant short bowel syndrome
`Aurelien Amiot a,c, Bernard Messing a,c, Olivier Corcos a,c, Yves Panis b,c, Francisca Joly a,c, *
`a Department of Gastroenterology and Nutrition Support, PMAD, Beaujon Hospital, APHP, Clichy, France
`b Department of Colorectal Surgery, PMAD, Beaujon Hospital, APHP, Clichy, France
`c Paris Diderot University, Paris, France
`
`a r t i c l e i n f o
`
`s u m m a r y
`
`Article history:
`Received 9 April 2012
`Accepted 13 August 2012
`
`Keywords:
`Short bowel syndrome
`Home parenteral nutrition
`Intestinal adaptation
`
`Background & aims: Short bowel syndrome (SBS) is a rare and severe condition where home parenteral
`nutrition (HPN) dependence can be either permanent or transient. The timing of HPN discontinuation
`and the survival, according to SBS characteristics, need to be further reported to help plan pre-emptive
`intestinal transplantation and reconstructive surgery.
`Methods: 268 Non-malignant SBS patients have been followed in our institution since 1980. HPN
`dependence and survival rate were studied with univariate and multivariate analysis.
`Results: Median follow-up was 4.4 (0.3e24) years. Actuarial HPN dependence probabilities were 74%,
`64% and 48% at 1, 2 and 5 years, respectively.
`In multivariate analysis, HPN dependence was
`significantly decreased with an early (<6 mo) plasma citrulline concentration >20 mmol/l, a remaining
`colon >57% (4/7) and a remnant small bowel length >75 cm. Among the 124 patients who became HPN
`independent, 26.5% did so more than 2 years after SBS constitution.
`Conclusions: This study indicates that long-term HPN is required in 47% of SBS patients started on this
`therapy. HPN independence is significantly associated with the remnant small bowel length, remaining
`colon and early plasma citrulline concentration. Noteworthy, HPN dependence could be reversed until 5
`years after SBS constitution.
`Ó 2012 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism.
`
`1. Introduction
`
`Intestinal failure (IF) is a highly disabling condition character-
`ized by the inability to maintain protein-energy, fluid, electrolyte
`and micronutrient balances under a normal unrestricted diet.1 IF
`ultimately leads to increasing malnutrition and even death if not
`circumvented by home parenteral nutrition (HPN).2 In adult
`patients, short bowel syndrome (SBS) is the major cause of IF.3
`In SBS, HPN dependence may be either permanent (or deemed
`irreversible) or transient.4 Although it is still debated, the human
`intestine has an inherent ability to adapt
`functionally and
`morphologically after a massive resection, leading to an improve-
`ment in its absorptive capacity overtime.4 These changes are
`mainly related to increased intraluminal absorption and enteral
`neuro-hormonal stimuli but also to behavioural changes, mainly
`
`* Corresponding author. Pôle des maladies de l’appareil digestif, Gastro-
`entérologie et assistance nutritive, Hôpital Beaujon, AP-HP, Université Paris 7, Denis
`Diderot, 100 boulevard du général Leclerc, 92110 Clichy, France. Tel.: þ33 1 40 87 56
`16; fax: þ33 1 40 87 45 74.
`E-mail address: francisca.joly@bjn.aphp.fr (F. Joly).
`
`spontaneous oral hyperalimentation (adaptive hyperphagia).5 All
`these changes aim to ultimately discontinue HPN. Classically, it is
`accepted that the time to discontinue HPN may be of at least 2 years
`in adult patients with a colon in continuity.6 However, late HPN
`discontinuation has also been described.7,8 Therefore, the timing of
`HPN discontinuation, according to the characteristics of
`the
`patients and SBS, needs to be further investigated.
`HPN has become the standard treatment for IF, offering a long-
`term survival demonstrated by several studies.6,9,10 Unfortunately,
`HPN is associated with complications, including progressive stea-
`tohepatitis and further liver failure, catheter-related complications
`and the inability to cope with the HPN regimen.11 Intestinal
`transplantation (ITx) is also an alternative therapy for patients with
`permanent and irreversible HPN dependence.12 However, despite
`its relative safety and efficacy, ITx survival still appears to be lower
`than with HPN.6,12e14 Consequently, based on the American Society
`of Transplantation as well as Medicare and Medicaid, ITx should be
`considered in patients with permanent HPN dependence and in
`cases of failure of HPN, high risk of death, severe SBS, frequent
`hospitalizations and the unwillingness of the patients to accept
`long-term HPN.12 Recently, a European study supporting the choice
`
`0261-5614/$ e see front matter Ó 2012 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism.
`http://dx.doi.org/10.1016/j.clnu.2012.08.007
`
`Page 1
`
`

`
`A. Amiot et al. / Clinical Nutrition 32 (2013) 368e374
`
`369
`
`of HPN showed that the survival of the patients who fulfilled the
`criteria for ITx without HPN-related complications was similar to
`the survival of the patients who did not fulfil the criteria for ITx.14
`Herein, we report a large series of adult SBS patients over a 25-
`year period of follow-up to better define these two important clues
`in the management of SBS patients on HPN: (1) HPN dependence
`and (2) long-term survival of SBS patients on HPN.
`
`nutritional balance monitoring as previously described.6,19 HPN
`characteristics regarding the composition, volume and frequency of
`the infusions were collected at the end of the follow-up. The degree
`of HPN requirement was calculated as the percentage of the daily
`mean of total parenteral energy infused divided by 1.5 times the
`resting energy expenditure.5,20
`
`2. Patients and methods
`
`2.1. Patients
`
`From January 1980 to April 2006, all consecutive adult patients
`with a short bowel syndrome (remnant small bowel length of
`150 cm) that have required HPN in our institution, an approved
`centre for IF and HPN, were included in a retrospective cohort
`study. Patients were recruited from our personal HPN database
`and/or a standardized hospital inpatient dataset diagnosis. We
`excluded the patients with evolving primary malignancies present
`within the first year of the follow-up (n ¼ 8), patients who had
`received treatments other than HPN for intestinal failure, e.g.,
`recombinant human growth hormone or teduglutide and the
`patients that have discontinued HPN within 3 months (n ¼ 21). We
`included patients who had reconstructive surgery of the remnant
`small bowel (segmental reversal small bowel was performed in 28
`cases). Eighty patients from this series have already been included
`in a previous study and 30 in the Pironi’s study.6,14 This work has
`been approved by the local ethical committee according to French
`law. The data records included the demographic data of the
`patients, aetiology of SBS, characteristics of past surgical proce-
`dures, characteristics of intestinal tract, HPN dependence and
`patient mortality including the causes of death. The remnant small
`bowel was evaluated on the basis of operative records and
`expressed as a length. The remaining colon was evaluated accord-
`ing to the method of Cummings et al.15 Briefly, the length of the
`colon was divided into seven sections, each part thus represents
`14% of the colon. These divisions correspond approximately to the
`sites at which jejunocolonic anastomosis were made. The length of
`remaining colon was expressed as a percentage. SBS was classified
`as type 1 (jejunostomy), type 2 (jejunocolonic anastomosis), or type
`3 (jejuno-ileocolonic anastomosis). Special attention was focused
`on the presence (or absence) of chronic intestinal obstruction,
`either functional or mechanical.16 The routine parameters collected
`for nutritional evaluation included weight, height, BMI and resting
`energy expenditure according to the Harris and Benedict equa-
`tion.17 Plasma amino acid sampling were performed after a over-
`night fast, as described previously, in order to determine the post-
`absorptive plasma citrulline concentration.18 Post-absorptive
`plasma citrulline concentration was determined under stable
`conditions in absence of overt dehydration, early, between 1 and 6
`months after SBS constitution and after at least a 2-year adaptive
`period following re-establishment of digestive continuity.
`
`2.2. HPN management
`
`The conditions under which HPN was administered have been
`outlined elsewhere.6,19 The dietary program was composed of
`unrestricted free solid food ingestion and, especially in patients
`with end-enterostomy, the use of either oral rehydration solution
`or mineral Vichy-St-Yorre and tap water restriction. All patients
`were educated by trained dietetician to promote spontaneous oral
`hyperphagia. The composition and volume of solutions given to the
`patients and the number of infusions per week were adjusted to
`individual needs. The attempt was always made to set up at
`a minimum the level of HPN regimen program according to the
`
`2.3. Absorption study
`
`SBS patients underwent a metabolic study to evaluate the
`intestinal macronutrient absorption at various times during the
`course of their HPN.5,20 Briefly, over a 6-day period, the first 3-day
`equilibrium period was used to confirm that patients were
`continuing their spontaneous intake of energy, carbohydrates,
`lipids, proteins, and fibres, while absorption was measured the last
`3 days. Unrestricted intake was measured by trained dietetician
`(food diary) by amount or weight and calculated with Bilnut soft-
`ware (Bourgerette P, Rolshansen M; BILNUT 4.0; SCDA Nutrisoft,
`Cerelles, France). Stool was collected daily and frozen at 20 C.
`Protein,
`lipid, and total energy were determined by nitrogen
`elemental analysis (N analyser Flash EA1112; Thermo Scientific,
`Waltham, MA, the method of Van de Kamer and bomb calorimetry,
`PARR 1351 bomb calorimeter; Parr Instrument, Moline, IL). Quan-
`tification of carbohydrate-derived energy was calculated by sub-
`tracting the energy associated with the protein and lipid
`components from the total energy. The calorie-conversion factors
`used were 4.2, 9.35, and 5.65 kcal/g for carbohydrates, lipids, and
`proteins, respectively. The coefficient of net intestinal absorption
`represented the proportion of ingested energy not recovered in the
`stool output. The degree of nutritional oral autonomy was calcu-
`lated as the total calories ingested daily multiplied by the net total
`caloric absorbed fraction and divided by 1.5 times the resting
`energy expenditure.5,20
`
`2.4. Statistical analysis
`
`All patients were followed-up until death or June 2006 when
`the data were collected. HPN duration was calculated from the date
`of HPN start, up to the death, definitive HPN cessation, or end of the
`follow-up. HPN dependence was defined as unability of patients to
`be definitively weaned off HPN at the time of data collection. The
`survival time was calculated from the date of HPN start to the death
`or end of the follow-up. For statistical analysis of survival, patients
`who have undergone ITx were presumed dead at the time of the
`transplantation. HPN dependence and survival probabilities were
`calculated using the KaplaneMeier method. HPN dependence and
`survival variables were compared using univariate and multivariate
`analyses. HPN dependence and survival distributions were
`compared using the log rank test, with p values of 0.05 considered
`statistically significant, according to the concomitant variables of
`survival and HPN dependence. Qualitative categories of values were
`defined by dichotomy from median value in two distinct groups of
`equal size. To identify the independent factors contributing to HPN
`dependence and survival, Cox proportional hazard models were
`then adjusted to the above-mentioned variables with an ascending
`stepwise procedure using SPSS statistical software (SPSS Inc, Chi-
`cago, Illinois). Quantitative variables were expressed as medians
`(ranges) or as means  standard deviation; both probabilities of
`survival and Hazard risks (RRs) were provided with 95% confidence
`intervals (CIs). Spearman’s two-tailed test was used to assess
`correlations. To determine early and postadaptive plasma citrulline
`concentrations performance,
`receiver operating characteristic
`(ROC) curves were constructed and area under curves (AUROCs)
`calculated with an empirical non parametric method. The charac-
`teristics of the patients at different time points were compared by
`
`Page 2
`
`

`
`370
`
`A. Amiot et al. / Clinical Nutrition 32 (2013) 368e374
`
`the Wilcoxon test and by Student t-test for HPN dependent and
`independent patients.
`
`Table 2
`Digestive characteristics of 268 adult patients with non-malignant short bowel
`syndrome.
`
`3. Results
`
`3.1. Patient’s characteristics
`
`(Tables 1 and 2) Two hundred and sixty-eight patients (139
`females, median age 52.5 (18e89) years) fulfilled the inclusion
`criteria. Past history of cancer was noted in 95 (35%) patients
`(chronic radiation enteritis in 61 cases): gynaecologic carcinoma in
`42 cases, digestive carcinoma in 39, urologic carcinoma in 6,
`Hodgkin’s or non-Hodgkin’s lymphoma in 4, head and neck carci-
`noma in 2 and myelodysplastic syndrome in 2. None of these
`malignancies was active at time of SBS constitution. Median delay
`between past cancer and SBS constitution was 3.8 (0.3e30.6) years.
`In one case, a patient with head and neck cancer unrelated to
`digestive history and arterial mesenteric infarction had a delay
`between past cancer and SBS constitution less than one year.
`Median follow-up was 4.4 (0.3e24) years. SBS was classified as type
`1 in 48 (18%) cases, type 2 in 179 (67%) and type 3 in 41 (15%) cases.
`The median length of post-duodenal remnant small bowel was 84
`(0e150), and 65 (0e150), and 70 (5e150) cm in type 1, 2, and 3 SBS,
`respectively (NS). The median percentage of remaining colon was
`61 (0e100) %. Twenty eight patients underwent segmental reversal
`small bowel procedure at time of the re-establishment of digestive
`continuity (jejunocolonic anastomosis in 27 cases and jejunoileal
`anastomosis in 1). Chronic intestinal obstruction was noticed in 35
`(13%) cases, including mechanical in 29 cases (surgical complica-
`tions in 9 cases, chronic radiation enteritis in 9, chronic intestinal
`pseudoobstruction in 6, soft tissue tumour in 5, arterial mesenteric
`infarction in 3 and volvulus, Crohn’s disease and venous mesenteric
`infarction in 1).
`
`3.2. Nutritional status
`
`At the end of the follow-up, the mean BMI had decreased
`compared to baseline (20.7  3.9 kg/m2 vs. 24  4.4 kg/m2,
`p < 0.001). An adaptive hyperphagia (>1.5 times the resting energy
`expenditure) was observed in 177 (66%) patients. Oral intake was
`significantly lower in patients with a chronic intestinal obstruction
`(1903 kcal/d  60 vs. 2327 kcal/d  60, p ¼ 0.01). Post-absorptive
`20.8  12.5
`plasma
`citrulline
`concentration was
`and
`23.7  12.5 mmol/l, in the early course (n ¼ 130) and after at least
`a 2-year adaptive period (n ¼ 146), respectively (p ¼ 0.07). The
`
`Table 1
`Demographic characteristics of 268 adult patients with non-malignant short bowel
`syndrome.
`
`Characteristics
`Gender
`Male
`Female
`Age at time of short bowel constitution, yrs (SD, range)
` 40 yrs
`41e59 yrs
` 60 yrs
`Causes of bowel resection
`Mesenteric infarction
`Arterial/venous
`Radiation enteritis
`Surgical complications
`Volvulus and traumatism
`Crohn’s disease
`Chronic intestinal pseudo-obstruction
`Soft tissue tumour
`Others
`
`No of patients (%)
`
`129 (47)
`139 (53)
`
`61 (23)
`124 (46)
`83 (31)
`
`115 (43)
`93 (35)/22 (8)
`61 (23)
`33 (12)
`14 (5)
`15 (6)
`11 (4)
`16 (6)
`3 (1)
`
`Characteristics
`Remnant small bowel
`< 50 cm
`50e99 cm
`100e150 cm
`Digestive circuit anastomosis
`Jejunostomy (type 1)
`Jejunocolic anastomosis (type 2)
`Jejunoileocolic anastomosis (type 3)
`Other digestive features
`Left colostomy
`Duodenopancreatectomy
`Segmental reversal small bowel
`
`No of patients (%)
`
`87 (32)
`95 (36)
`86 (32)
`
`48 (18)
`179 (67)
`41 (15)
`
`35 (13)
`5 (2)
`28 (10)
`
`median duration of HPN was 1.7 (0.3e20) years. The mean number
`of infusions per week was 5.4  1.7 with a volume per infusion of
`2449  766 ml and an energy value of 1362  351 kcal per infusion.
`The mean degree of HPN requirement was 57%  25% i.e. HPN
`represents 57%  25% of the total energy expenditure in patients
`that are still dependent of HPN at the end of the follow-up.
`
`3.3. HPN dependence
`
`One hundred and twenty four (46%) patients were permanently
`dependent on HPN. Of the 124 patients who were permanent HPN
`dependent, 17 (14%) were able to discontinue HPN but had to restart
`definitively HPN after a median delay of 12 (4e123) months. HPN
`independent patients had a longer remnant small bowel length,
`greater percentage of remaining colon, and higher plasma citrulline
`concentration, BMI, oral
`intake and degree of nutritional oral
`autonomy (p < 0,05) than HPN dependent patients (Table 3). HPN
`independent patients had respectively a higher and a lower
`percentage of type 3 and type 1 SBS (p < 0.001). Actuarial HPN
`dependence probabilities were 74%, 64% and 48% at 1, 2 and 5 years,
`respectively (Fig. 1). In multivariate analysis, HPN dependence was
`significantly decreased when associated with an early plasma
`citrulline concentration greater than 20 mmol/l [RR ¼ 0.23 (0.12e
`0.48), p < 0.001], a remaining colon greater than 57% (4/7)
`[RR¼ 0.41 (0.26e0.83), p¼ 0.001] and a remnant small bowel length
`greater than 75 cm [RR ¼ 0.47 (0.26e0.85), p ¼ 0.001]. Fig. 2 shows
`the actuarial HPN dependence probabilities of SBS patients accord-
`ing to the remnant small bowel length and type of anastomosis.
`Significant correlations were found between postadaptive
`plasma citrulline concentrations and the HPN independence
`(p ¼ 0.01), but not with early postoperative plasma citrulline
`concentrations. To predict the HPN independence, postadaptive
`plasma citrulline AUROCs were 0.85 (0.78e0.92, p < 0.001).
`Among the 124 patients that became HPN independent, 54.5% of
`patients discontinued HPN before 12 months, 19% between 12 and
`24 months and 26.5% after 24 months after the re-establishment of
`digestive continuity. The comparison between the two groups of
`patients that have been discontinued HPN before (n ¼ 91) and after
`2 years (n ¼ 33) showed that the latest group (after 2 years) was
`characterized by younger age at SBS constitution (47.1  2.9 yrs vs.
`53.6  1.6 yrs, p ¼ 0.05).
`
`3.4. Survival, ITx and causes of death
`
`One hundred and five (39%) patients died during the follow-up
`(Table 4). HPN- and SBS-related complications accounted for 13
`(13%) and 14 (13%) deaths, respectively. The main causes of death
`were due to cardiovascular disease in 21 (20%), underlying disease
`
`Page 3
`
`

`
`Table 3
`Comparison of the characteristics of 268 SBS patients according to their HPN status:
`independent (n ¼ 144) or dependent (n ¼ 124).
`
`Characteristics of the patients
`
`HPN
`independent
`
`HPN
`dependent
`
`p
`
`A. Amiot et al. / Clinical Nutrition 32 (2013) 368e374
`
`371
`
`in 22 (21%) and second primary malignancy in 18 (17%). There was
`no significant difference between the causes of death within the
`first two years compared to the rest of the follow-up. However,
`trends for a higher rate of death related to an underlying disease in
`the first two years (28% vs. 17%, p ¼ 0.19) was observed. The actu-
`arial survival probabilities were 94%, 70%, % and 52% at 1, 5 and 10
`years, respectively (Fig. 3). In multivariate analysis, the survival was
`significantly increased in patients with an age <60 years [RR ¼ 0.65
`(0.4e0.97), p ¼ 0.04] whereas the survival was significantly
`decreased in cases with an arterial mesenteric infarction [RR ¼ 3.4
`(2.1e5.4), p < 0.001], a past cancer history [RR ¼ 2.8 (1.8e4.6),
`p < 0.001] and the presence of an end ostomy [RR ¼ 1.8 (1.2e
`2.8), p ¼ 0.006]. Furthermore, the 10-year survival was signifi-
`cantly higher in patients that became HPN-free compared to those
`who remained HPN dependent (67.0%  0.6% vs. 40.7%  0.5%,
`p < 0.001) (Fig. 4).
`
`3.5. Absorption study
`
`The degree of nutritional oral autonomy was available in 224
`(84%) patients. Its evaluation was performed 9.0 (0.5e243) months
`after re-establishment of digestive continuity. The mean degree of
`nutritional oral autonomy was 69%  40% and it was significantly
`lower in type 1 anastomosis (53%  7%) than in type 2 (72%  3%)
`and type 3 (77%  7%) anastomosis (p ¼ 0.02) although the remnant
`small bowel length did not differ between each group (p ¼ 0.29).
`The mean degree of nutritional oral autonomy was also lower in
`patients with chronic intestinal obstruction (52%  8% vs. 72%  3%,
`p ¼ 0.01). Univariate analysis revealed a degree of nutritional
`autonomy greater than 70% significantly associated with a higher
`survival (p < 0.001) and a lower HPN dependence (p < 0.001) but it
`was not included in the multivariate analysis because it was eval-
`uated at various time points during the follow-up.
`
`50
`50.9  1.3
`23.6  0.3
`20.0  0.3a
`
`0.54
`0.48
`0.17
`0.002
`
`0.27
`0.01
`0.02
`0.17
`0.005
`0.79
`0.79
`0.01
`0.30
`
`46
`42a
`4a
`20
`7a
`5
`6
`7a
`8
`53  3a
`<0.001
`51  3a
`<0.001
`16.5  1.2a <0.001
`17.3  1.0a <0.001
`
`26a
`67
`7a
`
`12
`12
`
`<0.001
`0.90
`<0.001
`
`0.86
`0.32
`
`97
`0.13
`1964  78a <0.001
`66
`0.14
`53  3a
`<0.001
`71  5
`46a
`
`0.79
`<0.001
`
`54
`52.3  1.3
`24.4  0.4
`21.5  0.4
`
`Demographic characteristics
`Female(%)
`Age at time of SBS constitution (yrs)
`Initial BMI(kg/m2)
`Final BMI(kg/m2)
`Causes of SBS (%)
`39
`Mesenteric infarction
`27
`Arterial
`12
`Venous
`26
`Radiation enteritis
`18.5
`Surgical complications
`6
`Volvulus and traumatism
`5
`Crohn’s disease
`Chronic intestinal pseudo-obstruction 1
`Soft tissue tumour
`4
`Remnant bowel
`91  3
`Small bowel length (cm)
`69  3
`Large bowel (%)
`Post-absorptive plasma citrulline (mmol/l)
`27.4  1.3
`Postoperative
`29.5  1.4
`Postadaptive
`Digestive circuit (%)
`Jejunostomy (type 1)
`Jejunocolic anastomosis (type 2)
`Jejunoileocolic anastomosis (type 3)
`Other digestive features (%)
`Left colostomy
`Segmental reversal small bowel
`Oral intake
`Free oral intake (%)
`Oral intake (Kcal/d)
`Adaptivehyperphagiaa (%)
`Oral autonomy degree (%)
`Outcome
`Follow-up (mo)
`Survival (%)
`
`8
`66
`23
`
`14
`8
`
`100
`2625  68
`75
`90  3
`73  5
`78
`
`a Categories statistically significant using chi-2 test and student t-test.
`
`4. Discussion
`
`This study focuses on a cohort of 268 adult SBS patients
`followed-up over a 25-year period of which a part has been
`
`Fig. 1. Actuarial probability of home parenteral nutrition dependence of adult short bowel syndrome patients (n ¼ 268), for the overall study population and according to the
`remnant small bowel length, the remaining colon percentage and the postoperative plasma citrulline concentration.
`
`Page 4
`
`

`
`372
`
`A. Amiot et al. / Clinical Nutrition 32 (2013) 368e374
`
`Fig. 2. Actuarial probability of home parenteral nutrition dependence of adult short bowel syndrome patients (n ¼ 268), according to the remnant small bowel length and the type
`of anastomosis.
`
`In this study, We confirmed that the remnant small bowel
`length has a high predictive value to identified transient or
`permanent HPN dependence.6,16 However, the remnant small
`bowel
`length cut-off observed here was lower than in our
`previous study (75 cm vs. 100 cm).6 The re-establishment of
`colonic continuity or segmental reversal of the small bowel may
`have played a role in this decrease.19,21 We also confirmed that
`the presence of a significant part (at least 4/7) of remaining colon
`was a predictor of HPN discontinuation. This observation coin-
`cides with previous reports that demonstrate the important role
`of the colon following massive small bowel resection. This role
`included the improved absorption of fluid and electrolytes,
`mucosal hyperplasia, hyperfermentation of malabsorbed carbo-
`hydrates attributable to changes in the bacterial microbiota and
`secretion of enteric hormones involved in the intestinal adapta-
`tion (GLP-2, PYY).5,22e24 Although previous studies have failed to
`find a significant correlation between the amount of remaining
`colon and HPN dependence, we demonstrate its benefit in this
`very large series.6,20,24,25
`Citrulline is a non essential amino acid produced by the small
`bowel mucosa. In patients with SBS, the postadaptive plasma
`citrulline concentration was reported a marker of HPN dependence,
`under stable conditions and in the absence of renal failure.18
`However, the early assessment of the plasma citrulline concentra-
`tion has never been evaluated to predict a permanent or transient
`HPN dependence. In this study, we showed that the early plasma
`citrulline concentration (evaluated between 1 and 6 months,
`following the re-establishment of digestive continuity, when oral
`feeding is fully resumed) was a significant predictor of HPN inde-
`pendence. The role of early plasma citrulline concentration to
`predict
`further HPN dependence should be confirmed in
`a prospective study.
`
`previously been published.6 The HPN dependence probability was
`47% at 5 years and was significantly associated with a remnant
`small bowel length greater than 75 cm, a remaining colon greater
`than 4/7 and a postoperative citrulline concentration greater than
`20 mmol/l. These factors should be considered prospectively to
`guide earlier SBS patients at high risk of permanent HPN depen-
`dence on surgical rehabilitation, pharmacological therapy and/or
`ITx. The survival probabilities were 73% and 56% at 5 and 10 years,
`respectively. Furthermore, the mortality was mainly associated
`with underlying diseases and to a lesser extend with SBS- or HPN-
`related complications, supporting the safety of long-term HPN in
`approved HPN centres.16
`We used the same criteria from our previous study, excluding
`patients with a remnant small bowel length >150 cm and/or
`evolving primary malignancies.6 In contrast to the previous study,
`we also excluded patients that have discontinued HPN within the
`first 3 months (n ¼ 21). Hence, analysis of patients’ prognosis and
`HPN dependence was allowed without interference from con-
`founding variables despite the analysis of numerous variables has
`inherent bias. Although characterized by lower oral intake, we
`chose not to exclude chronic intestinal pseudoobstruction- and
`chronic radiation enteritis-related SBS as well as patients with
`persistent chronic intestinal obstruction, because we thought they
`are also representative of the SBS population. Excluding these
`patients would have also removed patients with chronic peritoneal
`adhesions, narcotic use and associated extra digestive comorbidity.
`
`Table 4
`Causes of deaths in a cohort of 268 adult patients with non-malignant short bowel
`syndrome.
`
`Characteristics
`Related to primary disease
`Radiation enteritis or uropathy
`Cancer relapse
`Arterial mesenteric infarction
`Postsurgical complications
`Desmoid tumour
`Related to HPN
`Sepsis
`Liver failure
`Related to short bowel
`Cachexia
`Metabolic complication
`Miscellaneous
`Cardiovascular disease
`Secondary primary malignanciesa
`Unknown
`Sepsis
`HBV recurrence-relatedfulminant hepatitis
`
`No of patients (%)
`22 (21)
`6
`6
`4
`4
`2
`13 (13)
`10
`3
`14 (13)
`9
`4
`56 (53)
`21
`18
`9
`7
`1
`
`a Second primary malignancies included lung carcinoma in 4 cases, prostatic
`carcinoma in 2, non-Hodgkin’s lymphoma in 2, colorectal carcinoma in 2, urological
`carcinoma in 2 and leukaemia, head and neck cancer, and small bowel carcinoma in
`1 (median delay of occurrence: 4.2 (0.3e15.4) yrs after SBS constitution).
`
`Fig. 3. Actuarial survival probability of adult short bowel syndrome patients on home
`parenteral nutrition (n ¼ 268).
`
`Page 5
`
`

`
`A. Amiot et al. / Clinical Nutrition 32 (2013) 368e374
`
`373
`
`improvement in the management of HPN and SBS patients and also
`small bowel surgical rehabilitation may be an explanation.19,21
`The impact of our results on ITx indication needs further
`considerations. According to the American United Network for
`Organ Sharing report, the 1-, 5- and 10-year survival rates after ITx
`were 78.3%, 57.1% and 43.5%, respectively.28 Despite ongoing
`progress, ITx survival remained lower compared to survival on
`HPN. We can figure that a subgroup of patients could benefit from
`ITx because of a high risk of death on HPN. Recently, Pironi has
`shown that the 5-year survival rate of ITx candidates who did not
`undergo ITx were lower than those who underwent ITx, only in
`case of HPN-related liver failure and desmoid tumours.14 These
`latter data support a direct referral for a pre-emptive ITx in patients
`with permanent HPN dependence and HPN failure and/or aggres-
`sive desmoid tumours. It is conceivable that we can use survival
`predictors in SBS patients as criteria for pre-emptive ITx screening.
`Although an age > 60 years and a previous cancer history may be
`considered as contraindications to ITx, an arterial mesenteric
`infarction and the presence of an end ostomy should be evaluated
`for pre-emptive ITx criteria in the early years of SBS constitution.
`
`5. Conclusion
`
`HPN is a safe therapy for SBS with a reasonably low related
`mortality when managed in a tertiary care centre. Long-term HPN
`is indicated in 47% of SBS after 5 years of follow-up because of
`irreversible IF. This study indicates that HPN dependence is
`significantly related to the SBS anatomy, early citrulline concen-
`tration and potentially to the net intestinal absorption. Noteworthy,
`HPN discontinuation significantly occurs until 5 years post SBS
`constitution, suggesting a longer intestinal and/or behavioural
`adaptation process than previously reported.
`
`Conflict of Interest
`
`None of the authors have any potential conflicts to disclose.
`
`Funding
`
`None.
`
`Authors contributions
`
`Study concept and design: AA, BM, FJ, YP; Acquisition of data:
`AA; Analysis and interpretation of data: AA, FJ, OC, YP, BM; Drafting
`of the manuscript: AA, OC, FJ, BM; Statistical analysis: AA; Study
`supervision: BM.
`
`References
`
`1. O’Keefe SJ, Buchman AL, Fishbein TM, Jeejeebhoy KN, Jeppesen PB, Shaffer J.
`Short bowel syndrome and intestinal failure: consensus definitions and over-
`view. Clinical Gastroenterology and Hepatology 2006;4:6e10.
`2. Jeejeebhoy KN. Management of short bowel syndrome: avoidance of total
`parenteral nutrition. Gastroenterology 2006;130:S60e6.
`3. Nightingale JM. The Sir David Cuthbertson Medal Lecture. Clinical problems of
`a short bowel and their treatment. The Proceedings of the Nutrition Society
`1994;53:373e91.
`4. Ziegler TR, Leader LM. Parenteral nutrition: transient or permanent therapy in
`intestinal failure? Gastroenterology 2006;130:S37e42.
`5. Joly F, Mayeur C, Messing B, Lavergne-Slove A, Cazals-Hatem D, Noordine ML,
`et al. Morphological adaptation with preserved proliferation/transporter
`content in the colon of patients with short bowel syndrome. American Journal
`of Physiology 2009;297:G116e23.
`6. Messing B, Crenn P, Beau P, Boutron-Ruault MC, Rambaud JC, Matuchansky C.
`Long-term survival and parenteral nutrition dependence in adult patients with
`the short bowel syndrome. Gastroenterology 1999;117:1043e50.
`7. Vanderhoof JA, Langnas AN. Short-bowel syndrome in children and adults.
`Gastroenterology 1997;113:1767e78.
`
`Fig. 4. Actuarial survival probability of adult short bowel syndrome patients (n ¼ 268),
`according to HPN dependence or independence.
`
`Following a massive resection, the remaining small bowel adapts,
`enhancing its ability to recover the loss of its absorptive func-
`tions.2,5,6,20 These adaptive changes are dependent on enteral
`neuro-hormonal stimuli as well as behavioural (mostly adaptive
`hyperphagia with high spontaneous oral alimentation), motility and
`microbiota changes.5,20,22 These changes ultimately permitted to
`SBS patients to restore oral autonomy and discontinue HPN. It is
`accepted that the time to discontinue HPN may be of at least 2 years
`in adult patients with a colon in continuity.6 However, Pironi et al.
`has recently reported, in a mixed cohort of children and adult SBS
`patients, a noticeable amount of patients discontinued HPN after
`more than 3 years.8 In our series, 76% of the patients were able to
`discontinue HPN in the first two years while 24% were able to stop
`HPN after two years following re-establishment of digestive conti-
`nuity. Taking together these results suggest a longer and ongoing
`intestinal adaptation process than previously reported. Others
`factors such as ability to restore oral feeding, food aversion, ageing,
`chronic narcotic use, psychological factors, underlying comorbid-
`ities and social factors, could also explain delayed HPN discontinu-
`ation.2,26 Indeed, the group of patients with a delayed HPN
`discontinuation was characterized by a younger age at SBS consti-
`tution. Furthermore, our absorption study, that takes into account
`digestive anatomy and absorption capacity as well as the amount of
`oral intake, showed that a high degree of oral autonomy (>70%) was
`significantly