Preoperative Nutritional Conditioning of Crohn’s Patients—Systematic Review of Current Evidence and Practice

Crohn’s disease is an incurable and frequently progressive entity with major impact on affected patients. Up to half of patients require surgery in the first 10 years after diagnosis and over 75% of operated patients require at least one further surgery within lifetime. In order to minimize surgical risk, modifiable risk factors such as nutritional status need to be optimized. This systematic review on preoperative nutritional support in adult Crohn’s patients between 1997 and 2017 aimed to provide an overview on target populations, screening modalities, routes of administration, and expected benefits. Pertinent study characteristics (prospective vs. retrospective, sample size, control group, limitations) were defined a priori. Twenty-nine studies were retained, of which 14 original studies (9 retrospective, 4 prospective, and 1 randomized controlled trial) and 15 reviews. Study heterogeneity was high regarding nutritional regimens and outcome, and meta-analysis could not be performed. Most studies were conducted without matched control group and thus provide modest level of evidence. Consistently, malnutrition was found to be a major risk factor for postoperative complications, and both enteral and parenteral routes were efficient in decreasing postoperative morbidity. Current guidelines for nutrition in general surgery apply also to Crohn’s patients. The route of administration should be chosen according to disease presentation and patients’ condition. Further studies are needed to strengthen the evidence.


Introduction
Nutritional support strategies in malnourished patients became widely accepted tool to decrease postoperative morbidity in major gastrointestinal surgery [1,2]. Up to 85% of patients with Crohn's disease awaiting surgery are malnourished as a consequence of active and disabling disease [3], impeding proper dietary intake and resorption [4][5][6]. Since up to 70% of Crohn's patients at some point requires surgery [7,8], a bowel-sparing attitude is mandatory in order to prevent short bowel syndromes, malnutrition, and anemia [9,10]. Despite advances in non-surgical management of acute flares through multimodal concepts including biologics, antibiotics, and nutritional support, surgery remains a last treatment option for medically exhausted cases [11][12][13]. Interestingly, it should be emphasized that incidence of surgical procedures for Crohn's disease did not decrease after introduction of infliximab [14]. Postoperative complication rates, including most feared intra-abdominal septic complications, reach 30% [15,16]. Nutritional guidelines for Crohn's patients have been published by the European Society for Clinical nutrition and metabolism (ESPEN) in Pertinent study characteristics (prospective vs. retrospective, sample size, control group, matching) were defined a priori and each manuscript was assessed for potential sources of bias. Two authors independently performed the literature search. The search terms were firstly identified in the title, and secondly in the abstract or medical subject heading. All studies of interest were obtained as full text articles and scrutinized thoroughly. Three authors made the final decision on inclusion of a study.
Relevant data were extracted and documented in a database developed ad hoc for all publications. The following items were recorded for each study when available: authors, title, year of publication, disease presentation/surgical indication, details on nutritional regimen (type/formula/duration/timing), and potential limitations of original studies. Postoperative outcomes of interest were complications (overall, infectious/septic, non-infectious), recurrence rates, and changes in different nutritional parameters if available. Data are presented in accordance to the PRISMA statement [18] (Figure 1).
Based on the findings of this study, an algorithm was created for practical guidance. Based on the findings of this study, an algorithm was created for practical guidance.

Data Analysis
Meta-analysis of results was not feasible due to limited and heterogeneous original data. Instead, tables were created with descriptive statistics to display the most relevant findings of each original study and review to give a comprehensive overview of the most relevant results.

Data Analysis
Meta-analysis of results was not feasible due to limited and heterogeneous original data. Instead, tables were created with descriptive statistics to display the most relevant findings of each original study and review to give a comprehensive overview of the most relevant results.

Outcome
Five studies [26,29,32,33,35] showed significantly better results in terms of overall and infectious complications in groups undergoing preoperative nutritional therapy compared to control groups ( Table 2). Among these studies, four [26,29,32,33] used EEN formulas and one [35] TPN. In the study of Heerasing et al. [26], 25% of patients could avoid surgery due to EEN induced remission and were bridged to alternative immunosuppressant therapy, but follow-up was limited to one year. Further effects of EEN were a significant decrease in CRP levels, surgical complications (8% vs. 32%) and infectious complications (abscess, collection, or leak, 3% vs. 20%) [4]. Wang et al. [29] showed an effect of EEN on different nutritional parameters (significant improvement of BMI, anemia and CRP levels), significantly lower infectious (21% vs. 44%) and non-infectious (26% vs. 51%) complication rates and less recurrence at six months (7% vs. 26%). Lower incidences of total (19% vs. 29%) and specific infectious complications (wound infection, abscess, and leak) were observed in the study of Li et al. [32] when comparing the steroid-weaned EEN group with the steroid-weaned control group. Further, supplemented patients needed less emergency surgeries compared to the different control groups. Li et al. [33] demonstrated a significant improvement of albumin and CRP levels after EEN therapy and at three months postoperatively, intra-abdominal septic complications were significantly lower (4% vs. 18%). In the cohort of Jacobson [35], patients pre-treated with TPN showed clinical remission and improved nutritional status (albumin, weight) at the time of surgery and no serious early (30 days) postoperative complications were observed in these 15 consecutive patients, contrarily to 28% in the matched control group.
Three studies [28,30,37] could not demonstrate differences in outcome due to unequal nutritional baseline conditions between nutritional and control groups, as detailed in Table 2. Three studies [31,36,39] without control groups compared their results to historical controls of other authors and found overall complication rates of 14% [39] and 18% [31]. However, these two studies were designed to compare nutritional endpoints [31] and anastomotic techniques [39] rather than focusing on impact of nutritional support. The preoperatively optimized cohort of Zerbib et al. [36] with 64% of patients receiving nutritional support (in combination with bowel rest, weaning of steroids, abscess drainage, and antibiotics) presented an overall morbidity of 18% and a low rate of fecal diversion. Another study without a control group by Guo et al. [27] identified EEN of <3 months, preoperative anemia and bacteria in fistula tract as independent risk factors for surgical site infection (31%), while preoperative abscess drainage represented a protective factor. Another study did not compare nutritional regimens [34], but reported TPN in combination with antibiotics, drainage, and postponed surgery in patients with penetrating disease which led to similar complication rates compared to patients with non-perforating disease (13% vs. 11%). The cohort of Yao et al. [38] was severely malnourished and half of patients were supplement by TPN one week before surgery and continued two weeks postoperatively. IgM levels decreased and BMI increased significantly in the study group, while no changes were observed in the control group. No difference was found regarding overall postoperative complications between the two groups (27% each), but a six month follow-up showed that the rate of resuming work was higher in the study group [38].
All reviews are consistent among each other regarding conclusions and agree on the importance of perioperative nutritional support. Further, they provide recommendations in line with current guidelines: if compared, enteral nutrition should be the preferred route of administration. All reviews underline the importance of a multimodal approach (preoperative optimization). Evidence-based recommendations however are scarce, since no solid evidence is available, and all authors agree that more high quality studies are needed to establish solid recommendations. Further, the impact of specific components of nutritional supplements should be studied to provide further evidence-based formulas [5,41]. Table 3. Reviews on preoperative nutritional support in Crohn's disease patients.

Author
Year Design Aim/Conclusions Forbes [4] 2016 Guidelines 64 recommendations to guide nutritional support in IBD patients.
Nguyen [5] 2016 N. Review Preoperative optimization by enteral and parenteral nutrition mandatory. Timing, route of administration, type, duration debated.
Schwartz [41] 2016 N. Review Evidence in favour of PN, but larger trials needed.
Montgomery [42] 2015 N. Review Recommendations for nutritional assessment and preoperative optimization.
Spinelli [45] 2014 N. Review Preoperative optimization crucial for surgical outcome, preoperative enteral nutrition for at least 10-14 days to prefer over TPN.
Triantafillidis [19] 2014 N. Review Indications for TPN are the same as in every major surgical patient.
Efron [49] 2007 N. Review Perioperative TPN might be beneficial, more high quality studies needed.
Lochs [17] 2006 Guidelines No specifics for Crohn's patients, perioperative nutrition as in general GI surgery.
Husain [50] 1998 N. Review Nutrition has a critical benefit in postoperative Crohn's disease.

Discussion
This systematic review scrutinized available evidence over the last 20 years to provide evidence-based guidelines for perioperative nutritional support in patients suffering from Crohn's disease. Fourteen original studies evaluated nutritional support in mostly severely ill patients, and a large heterogeneity was observed among studies regarding type, formula, and timing of nutrition. Only few prospective studies were available, and a randomized controlled study comparing different nutritional strategies was not to date. Hence, comparison between studies is delicate, and conclusions should be drawn cautiously. Even though nutritional support strategies were different, all studies presented encouraging results and emphasized the importance of nutritional support within a multimodal preoperative optimization concept. Some general principles in patients suffering from Crohn's disease must be discussed, including particularities of Crohn's patients, screening modalities, and current guidelines, which are discussed and compared to the evidence provided by this systematic review.

Particularities in Surgery for Crohn's Disease
Patients suffering from Crohn's disease are a particular subset of patients in many ways. At the time of surgery, most patients are treated by immunomodulating drugs, present with intra-abdominal infections, and are anemic and malnourished [55]. In a recent meta-analysis, steroid use, low albumin level, preoperative surgical history, and preoperative abscess were retained as risk factors for adverse surgical outcome [56]. Besides steroids and thiopurines, biologics such as anti-TNF provide new treatment options for disease control [57]. However, the influence of these drugs on postoperative outcome is matter of debate. While Fumery et al. described an increased risk of complications [15], a recent meta-analysis did not find any association between immunomodulating therapy and postoperative outcome [58]. Malnutrition on the other hand is common (up to 85%) among Crohn's patients awaiting surgery and is a well-known risk factor for adverse postoperative outcome in surgical patients in general [59]. For Crohn's patients needing surgery, anastomotic dehiscence, intraabdominal abscess, and fistula, regrouped as intraabdominal septic complications, represent most feared complications [60,61]. Intraabdominal septic complications hinder the postoperative course in up to 20% of patients with potentially severe consequences [15,62] and either reoperation or percutaneous drainage is needed in most cases. Hence, efforts to improve modifiable risk factors before surgery are of utmost importance.

Guidelines for Perioperative Nutrition and Preoperative Optimization
By the time of official ESPEN guidelines publication in 2006 [17], specific data on the effect of perioperative nutrition was lacking. Considerable evidence on nutritional support in general gastrointestinal surgery and in critically ill patients led by extrapolation to the recommendation to treat Crohn's patients accordingly [17]. This message was reinforced 10 years later by revised guidelines [4]. Hence, the ESPEN guidelines on enteral nutrition for surgery [53], published in 2006 and 2017 [54], do apply for Crohn's patients if they tolerate nutritional supplements to meet their metabolic needs. Most recommendations concerning enteral nutrition were elaborated on firm evidence and are hence highly recommended [53,54]. Whenever possible, the route of administration should be enteral, which is also advocated for patients with Crohn's disease [17]. This was previously emphasized by a review on nutritional support strategies in Crohn's disease [63]. Guidelines for parenteral nutrition [64] in the perioperative phase do likewise apply for Crohn's patients if metabolic needs are not met by enteral nutrition alone or if disease presentation at the time of scheduled surgery impedes enteral nutrition (e.g., intestinal obstruction or high output fistula). Nutritional guidelines merge with enhanced recovery after surgery (ERAS) guidelines [65], which are beneficial for surgical patients regarding outcome, length of stay, and costs. Recent reports suggested that enhanced recovery combined with minimally invasive techniques may lead to further improvements in surgical outcomes of Crohn's patients [66,67]. Whenever possible, elective surgical patients should be treated according to the ERAS protocol: avoidance of long term fasting, integration of nutritional strategies into the overall management of the patient, metabolic homeostasis, and early mobilization [4].
Nutritional strategies need to be part of a concept called preoperative optimization, including weaning of steroids if possible, drainage of percutaneous abscesses if applicable, and intravenous antibiotics if indicated [68]. Several of the studies retained for the present analysis presented promising data within such a multimodal approach [27,36,39,40,42]. Thus, surgery has to be delayed if possible in order to ensure best conditions. In case of emergency, EN or PN should start postoperatively [4].

Nutritional Screening
Several original studies [30,37,38] retained for the present analysis reported on nutritional screening tools or markers to provide nutritional support, especially by emphasizing the importance of body mass index (BMI). They all identified BMI as a follow-up tool of nutritional status during parenteral nutritional therapy. Recent guidelines advocate BMI <18.5 kg/m 2 , weight loss >10-15% within six months and serum albumin <30 g/L as best reflectors of severe undernutrition in Crohn's disease [4]. Concerning screening, ESPEN guidelines for Inflammatory Bowel Disease [4] recommend that Crohn's patients should be screened for malnutrition as patients undergoing general surgery [53,54] through validated screening tools. Particularly recommended are the Nutritional Risk Score (NRS) [69] and the Malnutrition Universal Screening Tool (MUST) [70]. Patients with a NRS ≥3 are considered to be at risk for gastrointestinal surgery [71].

Further Nutritional Strategies for Perioperative Support in Crohn's Patients
Two concepts need special consideration as result of this systematic review: EEN and TPN. Interestingly, most included studies reported on either EEN or TPN or a combination of both. EEN, either in elemental or polymeric form, has a direct anti-inflammatory effect [72], promotes mucosal healing [73], modifies intestinal microflora [74], and might decrease the antigenic load through bowel rest. EEN can induce clinical remission in pediatric and adult patients [75,76], as observed by Heerasing et al. [26] in 25% of patients awaiting surgery. In a recent review analyzing EEN in non-surgical Crohn's patients [77], EEN has been associated with remission rates of up to 80%. Wang et al. [29] observed decreased recurrence rates at six months in the EEN group, however, clinical recurrence was similar two years after surgery in both groups. Li et al. [32] and Smedh et al. [39] further presented interesting data on EEN allowing subsequent steroid-weaning, contributing to lower complication rates in EEN-groups in these studies. Disease presentations were severe in all studies with EEN [26,28,32,33,39] but one, [29], and might thus be particularly useful in this context (Table 2).
TPN was mainly used for penetrating disease in the study of Bellolio et al. [34], and Yao et al. [38] and Jacobson [35] treated patients with obstructing disease to observe improved immunity and clinical remission. Concerning formulas and timing, data was heterogeneous. Hence, no solid conclusions can be drawn. As a consequence, guidelines on parenteral nutrition [64] should be used for guidance. Schwartz [41] emphasized the need for larger prospective trials to strengthen the evidence. With this respect and due to lacking data, parenteral nutrition should be reserved for patients who are unable to cover their energetic needs by enteral nutrition.
Further considerations regarding routes of administration and associated potential complications have been published before [63].

Particularities in Perioperative Nutrition for Crohn's Disease
Despite the particularities of Crohn's disease and potential clinical discrepancies with the general surgical population including disease flares at time of surgery, exhaustive immunomodulating and medical treatment, and unfavorable baseline conditions, guidelines on enteral and parenteral nutrition including screening modalities, nutritional support strategies, and nutritional follow-up can be extrapolated to Crohn's patients. However, severe malnutrition in high-risk patients or inability to cover energy needs in patients with obstructing or fistulizing disease might impede conventional nutritional support (including oral nutritional supplements and immunonutrition) [4]. In these circumstances, specific nutritional support strategies including EEN or TPN have to be discussed. The following algorithm gives an overview on treatment suggestions considering available guidelines and the evidence of this systematic review (Figure 2).
Several limitations of the present study need to be mentioned. Due to heterogeneity of data and modest study quality of original studies regarding nutritional treatment strategies, solid conclusions cannot be drawn, and further high-quality evidence will be needed. The suggested treatment algorithm ( Figure 2) should thus rather help in decision-making than provide formal recommendations.

Conclusions
Perioperative nutrition in Crohn's patients awaiting surgery should be considered as a mandatory adjunct within preoperative optimization strategies. Guidelines including enteral nutrition and perioperative care for general surgery do also apply to Crohn's patients. Encouraging data for exclusive enteral or total parenteral nutrition, especially regarding induction of surgerypreventing disease remission and decreased recurrence, call for further high-quality studies.

Conclusions
Perioperative nutrition in Crohn's patients awaiting surgery should be considered as a mandatory adjunct within preoperative optimization strategies. Guidelines including enteral nutrition and perioperative care for general surgery do also apply to Crohn's patients. Encouraging data for exclusive enteral or total parenteral nutrition, especially regarding induction of surgery-preventing disease remission and decreased recurrence, call for further high-quality studies.