Impact of a dietitian-led very low calorie diet clinic on perioperative risk for patients with obesity awaiting elective, non-bariatric surgery: A retrospective cohort study

Background: Despite a lack of evidence that intentional weight loss reduces the risk of postoperative complications, adults with obesity are commonly asked to lose weight before elective surgery. We hypothesized that patients undertaking dietitian-led preoperative, very low calorie diet treatment could reduce perioperative surgery risks, as per validated risk scoring systems. The purpose of this study was to measure the impact of a dietitian-led preoperative very low calorie diet clinic on the American Society of Anesthesiologists physical status scores and National Surgical Quality Improvement Program Surgical Risk Calculator scores for patients with obesity awaiting non-bariatric elective surgery. Methods: This retrospective cohort study included patients referred to the preoperative dietitian-led very low calorie diet clinic before elective surgical procedures over a 2-year-9-month period. The dietitian prescribed individualized, very low calorie diet-based treatment. Primary outcomes were changes in the American Society of Anesthesiologists and Surgical Risk Calculator scores from pretreatment until surgery. Results: A total of 141 eligible participants (48 ± 13.4 years, 76% women, body mass index 41.7 ± 6.3 kg/ m 2 ) demonstrated clinically signi ﬁ cant weight loss (mean 7.1 ± 6.1kg, 5.2% body weight, P < .001). Median treatment duration was 13 weeks (interquartile range 6.2 e 19.2 weeks). Five participants (3.5%) avoided surgery due to weight loss-related improvements in their condition. American Society of Anesthesiologists scores improved for 16% (n ¼ 22/141) of participants. Overall, the median surgical risk calculator estimated risk of ‘ serious ’ and ‘ any ’ postoperative complication reduced from 4.8% to 3.9% ( P < .001) and 6% to 5.1% ( P < .001), respectively. Reduction in all Surgical Risk Calculator scores occurred, including surgical site infection, re-admission, and cardiac events ( P < .05). Conclusion: The dietitian-led preoperative, very low calorie diet clinic improved American Society of Anesthesiologists and Surgical Risk Calculator scores for non-bariatric elective surgery patients with obesity. Randomized controlled trials comparing this approach with a control group are warranted.


Introduction
Prevalence of obesity has grown to epidemic proportions throughout the Western world, with worldwide obesity rates tripling since 1975. 1 With this rise in obesity, there is an increasing incidence of obesity-related conditions that require elective surgery, such as reflux, colorectal cancer, oesophageal cancer, and gallbladder disease. 2,36][7] Furthermore, centrally distributed fat and enlarged fatty liver further elevate the risks of procedures within the abdominal cavity. 8,9The higher rate of complications in adults with obesity also increases the cost and resources required for their surgical care. 10he recent focus on the 'prehabilitation' of elective surgical patients aims to improve surgical outcomes by enhancing general health and well-being before major surgery by optimizing modifiable risk factors, 11 such as smoking, alcohol intake, respiratory fitness, body weight, blood glucose control, and physical fitness. 12xcept for smoking, these factors can be clinically improved with 3% to 5% weight loss, 13,14 making weight loss an integral part of prehabilitation for patients with obesity.
The limited literature available shows that short-term very low calorie diets (VLCDs) ( 800 kcal/d) of 1 to 3 weeks can shorten operating time and reduce blood loss in surgeries proximal to the liver (gastrectomy, liver resection, and laparoscopic cholecystectomy). 15The impact on other surgery types is largely unknown.A dietitian-led preoperative model of care using a VLCD-based intervention (the 'VLCD Clinic') 16 has previously been reported as achieving clinically significant (!5%) weight loss, while surgeons reported subjective perioperative risk reduction with improved technical ease and more efficient procedures.
Given the current lack of evidence from well-designed studies and the expense of randomized controlled trials (RCTs), preliminary pre-post trials are needed to inform the design of future RCTs on appropriate tools, estimated risk, and effect sizes.The current study revisited the VLCD Clinic to determine its impact on perioperative risk, measured using the following 2 validated tools: the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) Surgical Risk Calculator (SRC) 17 and the American Society of Anesthesiologists (ASA) Physical Status Classification. 18In addition, weight change and relevant surgical outcome data were collected to estimate effect size to inform adequately powered sample sizes in future RCTs.

Study design
This was a single-center retrospective observational cohort study of non-bariatric elective surgery patients treated within an established dietitian-led preoperative 'VLCD Clinic,' conducted within an outpatient setting at a large public hospital in southeast Queensland, Australia, which facilitates a wide range of elective surgical procedures.The research was approved by the Metro South Health Human Research Ethics Committee (HREC/2022/QMS/ 83772) and reported according to Strengthening The Reporting of Observational Studies in Epidemiology guidelines. 19

Eligibility criteria
Eligibility criteria included adults with obesity (body mass index [BMI] !30 kg/m 2 ) referred by their surgeon with planned elective surgery at the facility and patients having attended at least 1 dietitian appointment, agreed to commence VLCD Clinic treatment and discharged between November 2019 and July 2022.All consultant (lead) surgeons at the research facility who performed elective surgeries (on all adult patients, not just high-risk) were aware and had the ability to refer patients to the clinic.Patients with a BMI of 28 to 30 kg/m 2 were eligible if <4 weeks of VLCD was clinically indicated (eg, to reduce fatty liver volume and/or abdominal adiposity).Patients were ineligible for the VLCD Clinic if they had contraindications to VLCD (malnutrition, metastatic cancer, heart/liver/kidney failure, overt psychosis, pregnancy, or breastfeeding) or only minor procedures were planned (eg, colonoscopy).At an initial appointment, the dietitian assessed patients' nutritional status using the Subjective Global Assessment. 20Those diagnosed or at risk of malnutrition were excluded.Patients who no longer required surgery between their first and second appointments were discharged and excluded from the study.

VLCD Clinic model of care
This preoperative VLCD Clinic model of care was led by a dietitian to facilitate a lower-risk surgical procedure by providing safe, rapid weight loss and/or liver volume reduction.The clinic aimed to achieve time-based targets (eg, 2 weeks of VLCD) or weight loss targets.Time-based targets were given when the surgery was more urgent (limiting preoperative time) or surgeons desired a reduction in liver volume only.Weight loss targets were provided by the surgeon (limited to 10% body weight), based on known benefits to established comorbidities 21 and being a realistic target to achieve within 12 weeks. 22For those who needed to achieve a weight loss target, the model of care aimed for surgery to occur within 6 to 8 weeks if the target weight loss was achieved.For participants who did not achieve their weight loss target and were unlikely to do so, the surgeon clinically assessed whether they proceeded to surgery.

Dietary intervention
Within the VLCD Clinic, the dietitian prescribed a diet that consisted of meal replacement products and additional protein-rich food(s) to meet individualized protein requirements (0.8e1 g/kg of adjusted ideal body weight per day).Dependent on age and ability to adhere to dietary restrictions, participants were placed on phase 1, phase 2, or phase 3 diets previously described. 16In summary, phase 1 consisted of 3 VLCD products providing ~800 kcal (~3,400 kJ) and could be prescribed up to 920 kcal (3,900 kJ) with additional food to ensure individual protein requirements were met.Phase 2 was 2 VLCD products, aiming for ~900 kcal (~3,800 kJ) with food.Phase 3 was 1 VLCD product, aiming for 1,200 kcal (5,300 kJ) with food.The following commercially available VLCD meal replacement products were recommended: Optifast (Nestle Health, Frankfurt am Main, Germany) or Optislim (OptiPharm, Mordialloc, VIC, Australia).Adjustments were made to promote dietary adherence or transition to other dietary phases as required.Participants attended fortnightly dietitian appointments for evidence-based nutrition care 23 and monitoring of weight, dietary intake, and side effects, including pathology and communication with other health professionals if required; encouraging structured physical activity where appropriate; and providing education on progression to general healthy eating after treatment.Participants selfreported adherence to the prescribed diet at each appointment to inform the nutrition care plan.

Data collection and outcomes
Potential participants were identified from an established referral database and screened for eligibility.All data, including demographic data, were collected retrospectively from electronic medical records.Change (pre/post) data were measured from the start of VLCD Clinic treatment (pre) and at the time of surgery, or, if surgery did not occur, at discharge from VLCD Clinic (post).Primary outcomes were changes in ASA and SRC scores.
The ASA score is a subjective assessment of a patient's overall health, based on the following 5 classes (IeV)dI: patient is a healthy fit patient, II: patient has mild systemic disease, III: patient has severe systemic disease that is not incapacitating, IV: patient has incapacitating disease that is a constant threat to life, and V: a moribund patient not expected to survive 24 hours with or without surgery. 18The scoring system is accompanied by a guide that gives examples to guide decision-making, for example, poorly versus well-controlled comorbidities (Supplementary Figure S1).American Society of Anesthesiologists scores have been closely related to the rate of postoperative complications, with higher scores having a higher chance of complications. 24As ASA scores are not highly sensitive to changes in modifiable risk factors and do not account for specific procedural risks, pre/post-SRC scores were also measured.The SRC is a risk prediction suite for 1,557 procedures. 17t uses 21 variables, including medical history, demographic data, ASA score, BMI, comorbidities, use of related medications, and procedures planned, to create scores.Surgical Risk Calculator scores are expressed as '% risk' of 30-day adverse postoperative events.Two of the SRC% risk scores relate to 'serious' complications and 'any' complications.These are useful to compare across multiple surgery types, as other complications, such as urinary tract infections, may be more influenced by the type of procedure.The SRC also provides an estimated length of hospital stay.
Pretreatment ASA scores were collected from the nearest documented in-person anesthetist's assessment before the patient started VLCD treatment.Post-treatment ASA scores were collected from an anesthetist's assessment, which generally occurred 1 to 2 weeks before the planned surgery.The principal investigator (PI) checked all pre-and post-ASA scores in the medical chart against the ASA scoring criteria, 18 using the participant's medical data at that time point.If the PI had a conflicting score from that documented, when compared against examples and definitions provided within the system, or if no ASA score was documented, the specialist anesthetist was asked to provide the score using both medical data from the electronic medical record and their own clinical judgment.The anesthetist was not blinded to the participants' VLCD Clinic intervention, as they had access to the electronic medical record that showed dietetic intervention notes.
Pre-and post-treatment NSQIP SRC scores were calculated via input of medical chart data collected at pre-and post-time points into the web-based SRC. 17 Procedure codes were provided by a specialist anesthetist when the PI was unable to determine.Pre and post-scores were both based on the procedure planned at baseline, not the procedure that actually occurred (if it differed) for consistency across calculations.The 'surgeon adjustment of risks' function was set at '1 -no adjustment necessary' for all calculations to minimize inaccuracy resulting from adjustments based on retrospective data.Scores, expressed as '% risk' for all 11 specific complications, were collected, as well as overall scores for 'serious' and 'any' complication and predicted length of stay.
Weight data recorded in participants' medical charts up to 12 months post-surgery were collected.Thirty-day postoperative complications were collected from medical charts using simplified methods published by NSQIP SRC. 17 Standard procedures in place at the research site meant that some study participants were randomly selected by trained NSQIP reviewers who documented any complications in the 30-day postoperative period, and these documented data were used in those cases.If the PI was unable to determine whether a complication occurred, the NSQIP reviewer provided judgment.Length of stay was calculated as days crossing midnight in the hospital.Unplanned laparoscopic to open conversion was collected for planned laparoscopic procedures.Operating time was defined as the earlier time of either the surgical preparation of the participant or the commencement of the skin preparation until the time all instrument and sponge counts were verified, all dressings and drains were secured, and the surgical team had completed all procedure-related activities.Total operating theater time was calculated as the time the participant entered and left the operating theater.Anesthetic induction time was the time from preparation of anesthetic induction to the start of operating time.Estimated blood loss and the presence of fatty or enlarged liver were collected from documentation by the surgeon.
Secondary outcomes included weight and BMI change, clinic outcomes including 'failure to attend' rate, surgical outcomes including 30-day postoperative complications, length of stay, and estimated blood loss (only collected for gynecology procedures due to absent blood loss data from other surgical types).

Statistical analysis
Data were entered into Excel, version 2212 (Microsoft Corp, Redmond, WA) and presented as frequency (number,%), mean ± SD, or median with IQR.Pre/post outcomes were analyzed using SPSS Version 27.0 (IBM SPSS, Inc, Armonk, NY).Data were assessed for normality using ShapiroeWilk and tested with non-parametric (Wilcoxon rank) and parametric tests (paired t test, 2-tailed) as appropriate, apart from ASA change, which was determined using a 1-sided t test.

Results
Over the study period, 23 consultant (lead) surgeons from all 7 eligible (major) surgical disciplines at the research site referred patients to the VLCD Clinic.From 209 potentially eligible patients, 68 were excluded, resulting in n ¼ 141 participants (Figure 1).

Baseline characteristics
Most (76%) participants were female sex (Table I); 60% had a BMI !40 kg/m 2 , and 37% had hypertension, with 92% of these using anti-hypertensive medications at baseline.Participants with type 2 diabetes (16%) were mostly managed with oral hypoglycemic agents (83%), and 3 were using insulin (10%).Of the 27 participants diagnosed with obstructive sleep apnea, 48% were not using continuous positive airway pressure therapy despite it being prescribed.Gynecological surgery (38%, n ¼ 54), specifically hysterectomy (72%, n ¼ 39), was the most common procedure.Five (3.5%, n ¼ 5/141) procedures were for the primary treatment of cancer (colorectal tumor resections).See Supplementary Table S1 for a full breakdown of surgical procedures.Table I also outlines VLCD Clinic intervention data.

ASA and SRC outcomes
American Society of Anesthesiologists and SRC outcomes are shown in Table II.Overall, 16% (n ¼ 22/141) of the cohort improved their ASA scores from baseline.All of these participants moved 1 ASA leveldmostly from ASA III to ASA II, with 2 improving from II to I.However, for 84% (n ¼ 118/141), their ASA score remained unchanged.One participant (1%) increased from ASA II to III: their weight did not reduce during VLCD treatment but increased in the 12-month wait for surgery.Compared to the baseline, overall (improvement) in mean ASA scores for the entire cohort was statistically significant (one-sided t test, pre 2.68 ± 0.5 to post 2.53 ± 0.56 (change À0.15, P < .001).
Overall, 36% (n ¼ 52/141) of the cohort improved their % risk of complication in 1 or more of the 13 categories of SRC complications, with most of these (32%, n ¼ 45/141) demonstrating improved scores in both 'any' or 'serious' complication categories as well as in 1 or more other categories.When separated, all complication risk categories had a statistically significant improvement at posttreatment compared to baseline.Seven (5%) participants had an increase in 1 or more categories due to cessation of some medications because of weight loss, which may have a protective effect against some risks (eg, oral hypoglycemic agents and cardiac event risk).One participant had overall worsened SRC scoresdthe same participant whose ASA changed from II to III.

Surgical outcomes
Overall, 104 participants underwent elective surgery (Table III).Four participants had emergency (non-elective) surgery, and their data were not included in the analysis of surgical outcomes due to elevated risk profiles.Thirteen participants (9%) did not achieve sufficient weight loss to proceed to surgery.Five participants (3.5%) requested that their surgery be canceled due to weight loss-related improvements in their condition.These participants included 2 hiatus hernia for repair (reflux symptoms improved), 2 knee osteoarthritis for knee replacement (knee pain reduced due to reduced load on a joint), and 1 menorrhagia for laparoscopic hysterectomy (symptoms reduced due to improved hormonal balance from significant weight loss).Other reasons for not receiving surgery included the following: the resolution of a condition unrelated to weight loss (n ¼ 5), lost to follow-up/moved away (n ¼ 5), becoming unfit for the procedure due to worsening health conditions (n ¼ 3), and surgery delayed to the outside of study period (n ¼ 2).Twenty participants (14%) experienced pandemic-related surgery delays due to surgery cancellations and/or staff and participant illness.Table III outlines intraoperative and postoperative outcomes for 104 participants who underwent elective procedures.Some participants had their procedures completed at a private hospital due to coronavirus disease 2019 disruptions; therefore, postoperative outcome data were unable to be collected.One laparoscopic cholecystectomy was converted to a subtotal cholecystectomy due to difficulties relating to body habitus, gallbladder inflammation, and aberrant anatomy.One hiatus hernia repair patient was an extreme outlier for operating time, with a duration of 706 minutes (11.7  hours).This participant's surgery was unexpectedly complex, resulting in conversion to an open procedure and an unplanned intensive care unit admission.One laparoscopic bilateral salpingectomy was abandoned intraoperatively due to technical difficulty and body habitus.This participant's weight had increased over the course of VLCD Clinic and then from discharge to surgery.The median surgical procedure time for all (n ¼ 98) surgeries able to be collected was 159 minutes (IQR 109e206 minutes), with subset analysis per surgery type shown in Table III.Overall, the incidence of 1 or more postoperative complications was 13.9% (n ¼ 14/101).The most common postoperative complication was readmission to hospital within 30 days (7.9%), followed by surgical site infection (6.9%) and urinary tract infection (3%).No participants had cardiac complications, renal failure, discharge to a nursing home, or required a rehabilitation facility.The mortality rate was 0%.

Discussion
This study is the first to examine the impact of intentional preoperative weight loss on the estimated risk of perioperative complications using validated measures (ASA 25 and SRC 17 scores).The dietitian-led preoperative VLCD Clinic model of care had a significant and positive impact on the perioperative risk profile of elective surgery patients with obesity, demonstrating improved ASA scores and improved SRC scores.VLCD Clinic treatment for a median of 13 weeks produced significant weight loss and resulted in the cessation of anti-hypertensive medication and diabetes medication for some participants.Further, 5 participants requested cancellation of their procedure due to weight-loss-related improvements to their condition.Therefore, the VLCD Clinic has been observed to be a valuable prehabilitation method for patients with obesity awaiting nonbariatric elective surgery.
Current prehabilitation literature on elective surgery patients focuses on oncological surgical patients, prioritizes exercise-based interventions, and does not specifically target obesity and its known associated perioperative risk factors. 26,27Yet obesity rates continue to increase, and BMI restrictions are in place to prohibit those at increased risk from some procedures and reduce surgical costs. 28,29Just over one-half of our cohort lost !5% body weight.Importantly, 12% of those who did not manage to lose at least 5% of body weight (considered clinically significant) were still able to lose sufficient weight to achieve a BMI below 40 kg/m 2 .Where  ASA physical classification status scoresdI: completely healthy fit patient, II: mild systemic disease, III: systemic disease that is not incapacitating, IV: incapacitating disease that is a constant threat to life, V: a moribund patient not expected to survive 24 hours with or without surgery.ASA, American Society for Anesthesiologists; LOS, length of Stay; NSQIP, National Surgery Quality Improvement Project; VLCD, very low calorie diet.
guidelines and hospital facilities commonly use BMI 40 kg/m 2 as a cut-off point to prohibit surgery, 29 this result is significant for the planning of future perioperative pathways using VLCD intervention.
Up to 21% of participants with hypertension or diabetes were able to cease medications for these conditions due to improved blood pressure and blood glucose levels from VLCD Clinic treatment, which improved their perioperative risk scores.Clinically significant improvements in comorbidities, such as sleep apnea, diabetes, and hypertension, have also been observed in VLCD studies for bariatric surgery patients and the general population with obesity. 30,31With a lack of studies examining the effect of preoperative weight loss interventions on non-bariatric surgery outcomes, 15 our study provides valuable evidence for the use of VLCD to improve modifiable risk factors in a non-bariatric surgical cohort and the potential for VLCD Clinics to be used in isolation or within multi-modal prehabilitation programs for patients with obesity and associated comorbidities.
American Society of Anesthesiologists scores were reduced for 16% of participants.This result alone is clinically significant.However, SRC was also included to ensure changes in comorbidities and weight were captured.The SRC includes the use of diabetes and anti-hypertensive medication in its calculation, allowing those patients with improved management of their condition but whose ASA may not have changed to be identified as having reduced risk.Using both tools was important to ensure that the results accurately reflected risk.For example, we found that those who did not achieve at least 5% body weight loss still managed to improve ASA and/ or SRC risk scores, either through reduction to a BMI <40 kg/m 2 , comorbidity improvement, or both.Beyond just weight loss, nutritional adequacy was not measured in this study but was potentially enhanced by dietitian-led treatment.The use of a dietitian to prescribe and monitor a preoperative VLCD is vital to ensure no risk is being added through loss of lean body mass via inadequate dietary protein provision.Future trials should utilize dietitian-led VLCDs and measure lean body mass change pre/post to ensure safety.
Although mean risk reduction in SRC scores for 'any complication' in the entire cohort, and up to 4.4% for some participants, is a seemingly small reduction in risk, these improvements could have major benefits given the large number of elective surgical patients in many countries.For example, Australian public hospitals facilitate over 754,600 elective surgical procedures annually, 32 with !31% of these for patients with obesity. 33Patients have shown a high degree of motivation to change behavior in the preoperative period but a comparative lack of confidence, indicating a need for structured support, 11 which was provided by the VLCD Clinic in our study.Additionally, the surgical site infection rate was lower in our colorectal patients (18.8%) compared with 25.3% for BMI !40kg/m 2 elective colorectal surgery patients in England. 34Conversely, for gynecology surgery, surgical site infection rates (8.3%) were higher in our study than in a large multi-site study (5.3% for patients with BMI !30 kg/m 2 ). 35However, our cohort was considered to have significant perioperative risk to necessitate a referral to an intensive preoperative intervention and, therefore, likely had an inherently higher risk of complications at baseline than other studies.The use of control groups in future studies would assist in confirming whether complication rates can be improved through VLCD clinic access compared to other care.
The primary purpose of the VLCD Clinics is to 'prehabilitate' or sufficiently improve patients' risk profiles to allow them to proceed to surgery.Yet 5 (3.5%) participants were able to avoid hernia and arthroplasty surgeries altogether due to objective improvements in health and subjective improvements in symptoms.Literature supports thisdexcessive body weight is significantly associated with the severity of hiatus hernia symptoms due to increased intra-abdominal pressure, 36 and studies have shown that weight loss of >7.5% in adults with osteoarthritis reduces the risk of hip or knee total joint replacement surgery. 37,38Given this potential to significantly reduce surgical costs through these means, examining the cost-effectiveness of the dietitian-led VLCD Clinic should be prioritized.Due to the influence of the coronavirus disease 2019 pandemic over the 2-year 9-month study period, some participants' surgical care was transferred to a private hospital, resulting in less available data on surgical endpoints and thus underpowered analysis of secondary outcomes.The pandemic also influenced treatment timeframes and caused a delay in surgery for 14% of participants, which likely impacted participants' ability to adhere to the intervention.Although the pandemic could not be controlled for, our results still reflect a successful intervention whereby most participants lost weight and progressed to surgery.

Study limitations
Several study limitations were observed.It is likely that patients deemed by their surgeons as 'higher risk' were referred to the VLCD Clinic to undertake an intensive dietary program, and, therefore, results may not be generalizable to the entire population of patients with obesity awaiting elective surgery.However, it may also be said that surgeons are best placed to use clinical judgment and experience to determine the level of risk for referral to the VLCD Clinic or operate on those patients who would not benefit from weight loss intervention.The VLCD Clinic had been established at the research site for several years and was well known amongst consultant surgeons performing elective surgery, with 27 surgeons referring patients over the study period.It is anticipated that most patients eligible per their surgeons' clinical judgment would have been offered a referral to the VLCD Clinic.Using retrospective data within an electronic record to calculate ASA and SRC scores is prone to error due to potential data inaccuracy and availability.To reduce this bias, the data collector had extensive experience in interpreting clinical records at the research site and collected all data with assistance from an experienced Specialist Anesthetist where required.Although both tools are validated across multiple surgery types, 39,40 they are most commonly used on an individual basis to assist surgical decision-making or as a demographic descriptor in surgical studies.Also, the ASA score is used as a variable in the SRC calculation, which may have introduced bias due to the subjective nature of the ASA, causing variable inter-observer reliability. 41To mitigate this risk, one investigator collected and compared all scores against static examples and classifications.Consultancy from a single experienced specialist anesthetist also occurred when required.As a preliminary study in an area with scarce evidence, the research team chose to include all types of elective surgeries.However, the broad spectrum of types of procedures and techniques (both open and laparoscopic) makes study conclusions more difficult to interpret.Future studies should focus on specific surgical procedures to increase the applicability of the results.Regardless, the results of this study are a valuable addition to the literature and can be generalized to elective, non-bariatric surgical procedures.
In conclusion, the present study has demonstrated that a dietitian-led preoperative VLCD Clinic model of care can lead to improvements in perioperative risk scores for patients with obesity preparing for elective general, gynecology, colorectal, and orthopedic surgery.These results provide valuable evidence to support incorporating this intervention into perioperative models of care, which focus on prehabilitation or improving risk profiles to facilitate elective surgery.Future studies should include a control group to evaluate the effect of this intervention on surgical outcomes and costs of surgical care.

Figure 1 .
Figure 1.Flowchart of eligible patients treated in the preoperative dietitian-led very low calorie diet (VLCD) Clinic included in the study.

Table II
ASA physical status classification system and NSQIP risk calculator scores pre/post VLCD Clinic treatment (n ¼ 141)