Elective Surgery in Adult Patients with Excess Weight: Can Preoperative Dietary Interventions Improve Surgical Outcomes? A Systematic Review

This systematic review summarises the literature regarding the impact of preoperative dietary interventions on non-bariatric surgery outcomes for patients with excess weight/obesity, a known risk factor for poor surgical outcomes. Four electronic databases were searched for non-bariatric surgery studies that evaluated the surgical outcomes of a preoperative diet that focused on weight/fat loss or improvement of liver steatosis. Meta-analysis was unfeasible due to the extreme heterogeneity of variables. Fourteen studies, including five randomised controlled trials, were selected. Laparoscopic cholecystectomy, hernia repair, and liver resection were most studied. Diet-induced weight loss ranged from 1.4 kg to 25 kg. Preoperative very low calorie diet (≤800 kcal) or low calorie diet (≤900 kcal) for one to three weeks resulted in: reduction in blood loss for two liver resection and one gastrectomy study (−27 to −411 mL, p < 0.05), and for laparoscopic cholecystectomy, reduction of six minutes in operating time (p < 0.05) and reduced difficulty of aspects of procedure (p < 0.05). There was no difference in length of stay (n = 7 studies). Preoperative ≤ 900 kcal diets for one to three weeks could improve surgical outcomes for laparoscopic cholecystectomy, liver resection, and gastrectomy. Multiple randomised controlled trials with common surgical outcomes are required to establish impact on other surgeries.


Introduction
Obesity has tripled worldwide since 1975, with 31% of adults in Australia classified as obese [1]. The increase in obesity and the associated higher volume of visceral fat and liver steatosis is paralleled by the increased number of patients who require elective surgery for conditions that these individuals are at higher risk for, such as gallstones, liver cancer, and colorectal cancer [2,3].
It is widely accepted that the presence of obesity and excess visceral fat is a risk factor for complications arising from a variety of elective surgeries [4][5][6]. Cardiac and pulmonary events, excess blood loss, surgical site infection, and lengthier stays in hospital occur more frequently for surgical patients with obesity when compared to those in the healthy weight range [5,7,8]. Excess visceral fat and a high liver volume due to liver steatosis are known to complicate the technical aspects of surgery within the abdominal cavity [9].

Materials and Methods
This systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [20] and was registered on PROSPERO (the International Prospective Register of Systematic Reviews) with registration number CRD42021235483.

Search Strategy
A systematic search was performed using electronic databases MEDLINE, Embase, CINAHL, and Cochrane Central Register of Controlled Trials using the search terms defined in Table 1. Clinicaltrials.gov was searched for unpublished trials using basic search terms: "obesity", "surgery", and "diet". Reference lists from eligible articles and known grey literature were manually searched for relevant articles. The literature search was performed in February 2021. Table 1. Participants, Interventions, Comparisons, Outcomes, and Study Design (PICOS) criteria.

Participants
Adults with obesity or excess weight/fat who are undergoing elective, non-bariatric surgery obes * or overweight or "body size" or "visceral fat" or "central obesity" or "high body mass index" or adipos * or fat

Interventions
Dietary intervention(s) with/without exercise that were aimed at weight/fat loss prior to surgery diet or "weight reduc *" or "weight loss" or "preoperative care" or "diet therapy" or "healthy lifestyle" Comparisons --

Outcomes
Surgical outcome(s) of interest "postoperative complication *" or "surgery outcome *" or "intraoperative complication *" or "length of operation" or "length of surgery" or "length of stay" or "operat * time" or "intraoperative time" or "surgical procedures, operative" or ischemia or anesthesia or necrosis Study design All study types considered -

Inclusion and Exclusion Criteria
Articles included in the systematic literature review met all of the following criteria: (i) interventional studies with or without a comparator group; (ii) adults (≥18 years) with overweight/obesity or excess fat (such as centrally distributed fat or fatty liver); (iii) examined any type of elective surgery except bariatric surgery; (iv) evaluated the effect of a preoperative dietary intervention, for the purposes of weight loss, visceral fat reduction, and/or favourable liver changes; and (v) reported on relevant outcomes: weight change, liver changes, intra-or postoperative outcomes or complications, or subjective measures such as surgical difficulty. There were no restrictions on date of publication.
Articles that met any of the following criteria were excluded: (i) not written in English; (ii) animal studies; (iii) used pharmacological or exercise interventions without diet change; (iv) it was unclear if participants had lost weight unintentionally; (v) preoperative weight loss intervention was not described, or it was unclear whether it was dietary in nature; (vi) did not report weight loss or surgical outcomes for participants with overweight/obesity separately; or (vii) outcomes reported were not of relevance to the reviewers (e.g., changes in co-morbidities or laboratory values).
Meta-analysis was planned if an adequate number of studies measured and reported on similar surgical outcomes.

Study Selection
Title and abstract screening for eligible articles was completed independently by the lead author/reviewer (S.G.). Two reviewers (S.G. and R.L.) independently completed full-text screening of all articles against the inclusion and exclusion criteria. Results were compared, and discrepancies between reviewers were discussed with the wider research team to reach a final decision. For studies found via Clinicaltrials.gov search, the author was contacted to request results, if available.

Data Extraction
Data were extracted independently by S.G. from each included article. Authors of articles were contacted for relevant missing outcome data. Data extracted included characteristics such as author, year of publication, country, setting, and population group studied. Descriptions of dietary intervention and dietitian involvement and measures of adherence and attendance to treatment were also examined. For studies that did not specifically mention presence of a "dietitian", "dietician", or "nutritionist" or similar within the intervention, it was assumed there was no dietitian involved. Weight and body measurements or composition change were extracted if reported. Surgical outcomes of interest were extracted where reported and included: mortality, wound complications, and subjective outcomes from the perspective of participants or surgeon(s) such as surgical difficulty. For those studies that had a comparator group, outcomes reported by at least four studies were extracted into a table for ease of comparison and collation; this included: intraoperative blood loss, operating time, infection rate, and length of stay. Data units were standardised for comparison between studies, i.e., length of stay (days), operating time (minutes), infection rates (% incidence), where appropriate.

Quality Assessment
Quality assessment was competed independently by S.G. and R.L. using the "Quality Criteria Checklist: Primary Research" tool created by the American Dietetic Association's Evidence Analysis Library [21]. Each article was given a quality rating of either "Positive", indicating issues of bias have clearly been addressed, "Negative", indicating issues of bias have not been adequately addressed, or "Neutral", indicating the study had some bias concerns but was neither exceptionally strong nor weak. Results were compared, and the research team consulted on any disagreements to achieve consensus on the final quality rating.

Data Analysis
Meta-analysis was deemed unfeasible due to extreme heterogeneity of intervention type and duration, surgery type, limited number of articles with a comparator group examining comparable outcomes, and missing data despite attempts to contact authors. In its place, where able, a critical synthesis was performed to depict an overview of the size and effects seen for outcomes measured in multiple studies. To create forest plots, a validated calculation by Hozo et al. (2005) was performed to convert values to mean and standard deviation [22] for studies that reported only median and range. Review Manager (RevMan) Version 5.4, The Cochrane Collaboration, London, United Kingdom, 2020, was used to analyse data using pooled mean differences and the inverse-variance method [23]. Figure 1 outlines the detailed steps of the article selection process as per PRISMA guidelines [20]. From 4597 articles retrieved, 184 duplicates were removed using EndNote X8 (Clarivate Analytics, Philadelphia, PA, USA) and the remaining abstracts screened, with 53 full-text articles assessed for eligibility. A total of 14 studies met inclusion criteria and were included in our systematic review: four randomised controlled trials (RCTs) [24][25][26][27], one pilot RCT [28], five single-cohort [29][30][31][32][33], three retrospective cohort studies [34][35][36], and one retrospective case series [37].  Table 2 shows the characteristics of the included studies. Publication dates ranged over 24 years, with the oldest study published in 1997 [33] and the most recent in 2021 [29], with eight of the 14 studies published in the last five years [24][25][26][28][29][30]36,37].
The intervention locations were mostly in hospital-based or large medical centre outpatient settings, with most settings described as "hospitals" [26,30,33,36] and some described as surgical specialty facilities [31,37], while some appeared to be university settings [27].
The 14 included studies had a total of 1023 enrolled participants (intervention: n = 726, comparator: n = 297). Five studies reported dropout rates, which were generally low, the highest being n = 4/50 (10%) [28]. The number of participants enrolled in each study ranged from 25 to 188. Participants' ages ranged from 39 to 71 years of age. Most studies had majority female participants, except for two studies (7% [36] and 45% female [24]).
Eleven out of the 14 studies set inclusion criteria according to BMI cut-offs [25][26][27][28][29][31][32][33]37]. One of these used a combination of BMI and waist circumference [36]. The lowest BMI cut-off was ≥ 25 kg/m 2 and the highest ≥ 40 kg/m 2 [24,36]. The remaining studies had eligibility criteria based on the need for hepatobiliary-related surgery with the aim of liver shrinkage or reduction in liver steatosis [33][34][35]. Baseline mean BMI across the studies ranged from 26 kg/m 2 [36] to 49 kg/m 2 [31]. The lowest individual BMI was 22.5 kg/m 2 for a participant who met inclusion criteria with waist circumference ≥85 cm undergoing gastrectomy for gastric cancer and the highest was 85 kg/m 2 , for a complex incisional hernia repair.
General surgery was the most frequently represented surgery type, with nine studies examining a specific general surgery procedure (hernia repair n = 3, liver resection n = 3, laparoscopic cholecystectomy n = 2, gastrectomy n = 1). Orthopaedic surgery was limited to knee or hip arthroplasty, examined in two studies [27,37]. The remaining three studies each examined a range of elective procedures. One examined both hernia repairs and laparoscopic cholecystectomies [28], while the other two studies [29,33] each examined a wide range of procedures, including gynaecological, general, orthopaedic, and neurological, but had no control groups. The numbers of participants for these surgeries ranged from six participants in colorectal surgery to 54 participants in gynaecological surgery.

Intervention and Comparator Characteristics
Across the 14 studies, preoperative dietary interventions ranged greatly in duration, method of delivery, and type, but VLCD or a variation of VLCD was most used. See Table 3 for details of interventions. For the 11 studies that reported kcal (kilocalorie) prescription, diets ranged from 458 kcal to 1500 kcal. When kcal of the diets were compared against the accepted definition of a VLCD [38] (≤800 kcal per day), six studies utilised a VLCD [24,26,28,30,31,33], while others used "VLCD-based" (800-1200 kcal) [29] or LCD (900-1190 kcal) [27,34,35]. One study reported using a VLCD, but when examined, only one VLCD meal replacement per day was prescribed, with the remaining two meals of the participants' choice [36], which does not meet VLCD criteria. Two studies used what appeared to be general healthy eating advice [25,37], and one used a "weight loss program" which was based on a 1500 kcal/day, "low carbohydrate" diet [32]. Surgeries in proximity to the liver utilised VLCD or LCDs exclusively [24,26,30,34,36].
Studies reported wide variations in intervention durations, ranging from 1 week [24,34] to 26 weeks [23], with an outlier of 68 weeks [31]. The median duration was 9 weeks when including the outlier. The duration of the dietary intervention differed between participants in five studies due to the need to lose a certain amount of weight (7% [25] or 10% [35] body weight) or reach a certain BMI target prior to surgery (BMI < 40 kg/m 2 [31,37] or <28 kg/m 2 [32]). One of these studies aimed to achieve a BMI of <40 kg/m 2 before complex incisional hernia repair [31], requiring lengthy dieting time frames (68 weeks on foodbased VLCD), likely due to the high baseline BMI of participants (49 ± 10 kg/m 2 , range 36-85 kg/m 2 ). An additional two studies provided individualised weight-loss targets, and thus duration was based on planned operation, weight-loss progress, and judgement of surgeon [29,33]. All five studies that examined surgery types in proximity to the liver used either VLCD or LCD interventions of less than three weeks [24,26,30,34,36], with the exception of one study, which aimed at weight loss prior to liver donation [35] and used VLCD for a median duration of 7.3 weeks. The two remaining studies had set durations of eight weeks [28] or 12 weeks [27] prior to general surgical procedures and hip/knee replacements, respectively.
Eight studies had a dietitian involved in their interventions: the interventions were led solely by a dietitian (five studies) [24,28,29,36,37]; in the other studies, the dietitian provided dietary advice once at the beginning of the intervention [35], on demand via telephone and a diet sheet [26], and within a wider multidisciplinary team [25]. In three studies, it was unclear who provided dietary instruction in a once-off appointment using verbal advice alone [27,32] or a "diet sheet" [30]. The remaining three studies had instructions delivered solely by their surgeon [34], a "medical weight-loss specialist" [31], or a "therapist" [33]. Interventions of longer than three weeks appeared to incorporate fortnightly or monthly contact with a dietitian or the multidisciplinary team. Dietary protein prescriptions were based on individuals' body weights in three studies [28,29,31], with other prescriptions ranging from 33 g [34] to 90 g per day [35], and six studies not reporting the amount of protein prescribed [25,26,30,32,36,37].      3 shakes, small amount of low-CHO vegetables, 2 L water/lowenergy fluids. "Re-feeding" period commenced 1 month prior to surgeryreduced to 1 shake per day for 1 week, then to normal food. Mean 14 weeks (range 7-24), guided by initial weight, planned operation, and weight-loss progress.
"Therapist" provided care, tailored appointments, including "nutritional education and behaviour therapy to create new eating habits and to prevent weight gain".

Fortnightly.
Listed for those who experienced adverse outcomes/problems with adherence. n = 1 depression, n = 3 excess alcohol intake. n = 6 "suspected" intake of normal food. n = 2 ceased VLCD within "first weeks", reason NR. Only six studies measured adherence to the diet, using a variety of measures from urinary ketone levels [28] to verbal reports from the participant to the surgeon on the day of surgery [34]. Three studies measured adherence via written food records [24,26,27] (ranging from three days to two weeks). Two studies utilised VLCD (≤800 kcal) and found a mean intake of 805 kcal [24] and 947 kcal [26]. The other study utilised the reduction of usual intake by 500 kcal for the intervention group (reduction of 543 kcal achieved) but had similar kcal intake between groups (1290 kcal in comparator versus 1190 kcal in intervention). Three studies included prescribed structured exercise as part of the intervention [25,27,35], and, in one study, structured activity was encouraged throughout [29]. Table 4 outlines the body composition and weight change resulting from the interventions, reported in all except two studies that chose to focus on liver-related changes such as hepatocyte glycogen/steatosis [24] and blood loss [34]. Mean weight loss across the 11 studies that reported mean or median weight change was 8.6 kg (range of 1.4 kg [30] to 24 kg [31]), with a mean of 1.7 kg loss per week over an average of 6.5 weeks for VLCDs (when the outlier of 68 weeks was removed) and 0.4 kg per week for other diets. One study reported BMI loss alone, which was 4.4 kg/m 2 [35]. Only two studies measured lean body mass change [27,28], for a general surgery cohort and joint replacements, with both finding no statistically significant change from baseline.      [39]; β Hepatocyte glycogen = storage of carbohydrate, lost with ketosis. One gram of glycogen binds to four grams of water resulting in significant loss of liver volume when glycogen stores reduced. Table 5 shows the most reported surgical outcomes with a comparator group to allow for analysis. Operating time and blood loss were the only objective surgical outcomes that were found to be impacted significantly. Operating time was reduced in one study by six minutes (p < 0.01) for laparoscopic cholecystectomy participants given a VLCD for two weeks preoperatively [26]. This study was the only study out of six that was adequately powered to detect this difference as a primary outcome. Operating time was the only outcome with sufficient data across multiple studies to enable collation into a forest plot (see Supplementary Figure S1 for forest plot). Three of four studies showed reduced blood loss: two liver resection studies that used VLCD or LCD (900 kcal) for two weeks [34] and one gastrectomy study that used a VLCD for three weeks [36]. Blood-loss reduction ranged from −27 mL for gastrectomy [36] to −411 mL for liver resection [24] (p < 0.05). Infection rate was measured in four studies, but only two had sufficient data for statistical comparison between groups, and no statistically significant difference in infection rates was found [28,34]. Two other studies reported having no infection occurrences in either group [25,27]. Length of hospital stay was reported in seven studies [24][25][26][27][28]34,35], and none showed statistical significance between groups.

Impact on Surgical Outcomes
Subjective outcomes were collected in five studies. Four studies surveyed the surgeon regarding the difficulty of the procedure [24,26,29,30]. Due to a lack of validated tools for this purpose, different scales and surveys were used in each study, making it difficult to compare results. Regardless, two of these studies surveyed surgeons performing a laparoscopic cholecystectomy [26,30] who were blinded to groups. The surgeons were asked to evaluate the dissection or visualisation of Calot's triangle, a critical point in the procedure that can be made technically difficult due to enlarged steatotic left hepatic lobe, excess intra-abdominal fat, and a more friable, fatty liver which is prone to bleeding [40]. In one study, surgeons found visualisation was easier for the intervention group, and in the other study, surgeons found dissection easier in the intervention group (p < 0.05 for both studies). One study found a statistically significant increase in health-related quality of life post-intervention from baseline for the intervention group [28].
Eighteen other surgical outcomes were reported on, including mortality, deep vein thrombosis, bile leak, conversion to open procedure, abscess, wound dehiscence, haematoma, and hernia recurrence (for hernia repair), but were infrequently reported across studies. Most had no comparator group, and when able to compare between groups, showed no differences.  Figure 2 shows quality assessment outcomes for the included studies against quality criteria. Three studies met all 10 criteria for low-level risk of bias [24,26,28], and six studies received an overall "positive" quality rating. Eight studies received an overall "neutral" quality rating, mostly due to differences between or lack of a comparator group and poor or unclear reporting on whether measurement of outcomes was blinded. Nine studies had at least one criterion considered indicative of a high risk of bias [25,27,[30][31][32][33][34][35][36].

Discussion
This is the first review to summarise the literature on preoperative dietary weightloss interventions for adult non-bariatric elective surgery patients and their impact on surgical outcomes. Prescription of a preoperative VLCD or low calorie diet (≤900 kcal) that utilised meal replacements for one to three weeks appeared to reduce operating time for laparoscopic cholecystectomy and reduce blood loss for liver resection and gastrectomy. This was supported by the subjective reduced difficulty of dissection and critical view of Calot's triangle for laparoscopic cholecystectomy. Length of stay did not appear to be affected. These results should be interpreted with caution as no surgery types had multiple RCTs to confirm results. No conclusions could be made regarding the other surgery types examined due to heterogeneity of interventions, surgery types, surgical endpoints, and lack of comparator groups throughout the remaining studies. Unfortunately, there were no studies that specifically examined gynaecological, colorectal, endocrine, urological, or cardiothoracic surgery, which represents a substantial gap in the literature.
It has been established that a 14-day VLCD prior to bariatric surgery can reduce the risk of 30-day postoperative complications [14,41]. The present review highlighted that the same intervention could potentially be utilised prior to elective laparoscopic cholecystectomies, liver resections, and gastrectomy to reduce visceral fat and liver steatosis, leading to reduced blood loss and operating time. Whether it reduces 30-day postoperative complications in the same way as bariatric surgery is still unclear and requires further research. Measuring patients' visceral fat or liver steatosis is not currently part of standard care for patients awaiting these types of surgery but has the potential to better predict postoperative complications than BMI [42] and may be worth considering to facilitate direction of additional preoperative resources to those patients who would most benefit.
There have been two systematic reviews published in this area, and both examined bariatric and non-bariatric surgery studies together [15,16]. Due to lack of reporting on common surgical endpoints, they were both unable to draw conclusions on benefits to surgical outcomes, and one review attempted meta-analysis [16], but only three non-bariatric studies were included. In this respect, the present review sheds new light to demonstrate potential risk reduction for three surgical procedures (liver resection, laparoscopic cholecystectomy, and gastrectomy) but also highlights the major gaps in the literature for other surgery types. Given the increasing pressure placed on hospital systems and surgeons to optimise their surgical patients with excess weight due to the rising prevalence of obesity, this research should be prioritised.
Bariatric surgery patients can achieve 2.8 kg to 14.8 kg weight loss with preoperative VLCDs [43]. The present review reflects the same degree of weight loss is possible for nonbariatric surgery patients, with VLCD or VLCD-based intervention achieving 1.4 kg (7-day duration) to 24 kg loss (17-month duration) and at the most rapid rate (1.4 kg/week). The results of the present review parallel conclusions made in numerous studies on VLCD that weight loss appears to correlate with the duration of dietary intervention rather than degree of caloric restriction [15]. The suitability of time-frames and the efficacy of a preoperative weight-loss intervention is important to consider, as delaying elective surgery has been shown to result in deteriorations in health, worsened quality of life, increased disability, and decreased work capability [44,45]. Further, delaying surgeries such as hernia repairs or laparoscopic cholecystectomy can result in unplanned emergency surgery [46,47], which has an inherently higher risk of complications than elective surgery [48]. Due to these risks, it may be prudent to consider a VLCD as the most appropriate dietary choice for adults awaiting surgery for conditions that may worsen over time due to the rapidity of weight loss it provides when compared with other diets.
Few studies in the current review measured common surgical endpoints such as infection or respiratory events, despite the known increased risk of these events for patients with obesity. When measured, very low incidences of these outcomes were found across studies but could not be linked to the preoperative intervention due to underpowered sample sizes and lack of comparative groups. The most commonly reported endpoint in studies with a comparator group was length of stay, with all seven studies showing no significant difference. Despite this, length of stay should continue to be measured in future studies due to the profound implications it has for healthcare facilities and the small sample sizes and number of studies in this review.
Increased BMI is associated with increased costs of surgical care [49,50]. The two outcomes that were found to be impacted by preoperative dietary interventions (blood loss and operating time) have great cost-saving potential. Using operating time as an example, standard operating theatre costs in Australian public hospital facilities, not including equipment costs, are approximately AUD $7 per minute [51]. Therefore, even a seemingly insignificant reduction in operating time per laparoscopic cholecystectomy procedure could have major benefits, with 49,874 laparoscopic cholecystectomy procedures performed in Australia annually [52]. Furthermore, other hospital-acquired postoperative complications can incur punitive financial penalties at some Australian healthcare facilities [53]. A preoperative intervention that potentially reduces operating time and postoperative complications for adults with excess weight could be considered an investment in surgical care, but studies are required to show definitive economic benefit.
There is a clear association between depletion of lean body mass and poor surgical outcomes [54][55][56], including delayed wound healing, infectious complications, intensive care unit admissions, and longer hospital stays [57]. Despite this, there was a distinct lack of reporting on change to lean body mass in the included studies. Although the two studies that measured this reported no excess loss of lean body mass, this result cannot be assumed across all population groups, such as those with malignancy or awaiting organ transplants who may have underlying dysfunctional metabolic processes. Loss of lean body mass can be mitigated by providing adequate dietary protein [18], but interestingly, only three studies individualised their protein prescription based on body weight, and one study provided only 33 g of protein per day, which would be inadequate to meet protein requirements for any adult with obesity when based on minimum 0.8 g/kg. This highlights the need to ensure adequate protein provision in preoperative weight-loss interventions as a matter of urgency, especially considering the increasing pressure placed on patients to lose weight preoperatively [12,13]. Loss of lean body mass could potentially be avoided by utilising qualified dietitians for dietary prescription and monitoring, but due to the lack of studies in this area that involve dietitians and measure lean body mass, this is still unknown and requires exploration.
Study design was a major limitation of the included studies. For example, the lack of comparative data was a source of bias in single-cohort studies [29][30][31][32][33] and retrospective studies [28] and contributed to a lower level of evidence from this review. The nonrandomised comparative studies had potential sources of major bias: methods often poorly described with opportunistic recruitment, with historical or unmatched comparator groups. The high number of studies without a comparator group (six studies) and the differing surgery types and length of intervention duration reduced the capacity of the authors to draw conclusions on the impact of their interventions or to facilitate meta-analysis. For future studies, it would seem pertinent to establish a list of relevant surgical endpoints for inclusion to facilitate meta-analysis and definitive conclusions. The American College of Surgeons National Surgical Quality Improvement Program (NSQIP) [58] has developed a list of postoperative complications that are collected and analysed at participating hospitals, which may provide a scaffold for future studies to collect common outcome data.

Conclusions
Preoperative VLCD and LCD interventions provided for one to three weeks to patients with overweight, obesity, higher amounts of visceral fat, or liver steatosis have potential to reduce operating time and blood loss during non-bariatric elective surgeries within the abdominal cavity, such as laparoscopic cholecystectomy, liver resection, and gastrectomy. Although the benefits were likely in studies that used dietary interventions longer than three weeks in other surgery types, they were less clear due to the lack of comparator groups and common surgical endpoints.
To determine whether preoperative dietary interventions for weight loss improve surgical outcomes, larger well-designed randomised controlled trials need to examine other surgery types and common surgical endpoints. Furthermore, safety of dietary interventions through preservation of lean body mass, particularly for patients with malignancy or underlying metabolic processes, should be confirmed to ensure there is no added risk to patients due to restrictive dieting in the preoperative period.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/nu13113775/s1, Figure S1. Forest plot of operating time outcome for studies that conducted preoperative dietary intervention and compared time between groups. Funding: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.