Faculty Bibliography

BACKGROUND:The laparoscopic sleeve gastrectomy (LSG) has become one of the most popular surgical weight loss options. Since its inception as a procedure intended to promote durable weight loss, the association between LSG and gastroesophageal reflux disease (GERD) has been a point of debate. First and foremost, it is known that GERD occurs more frequently in the obese population. With the sleeve gastrectomy growing to be the predominant primary bariatric operation in the United States, it is imperative that we understand the impact of LSG on GERD. OBJECTIVE:To examine the effects of LSG on GERD symptoms. METHODS:One hundred and ninety-one bariatric surgery candidates completed a Gastroesophageal Reflux Disease-Health-Related Quality of Life (GERD-HRQL) questionnaire before and after undergoing elective LSG (mean follow-up time of 20.4 ± 2.7 months). Values were stratified by the presence or absence of preoperative GERD, GERD medications, age, gender, crural repair, patient satisfaction with present condition, and percent total weight loss (%TWL). RESULTS:respectively. Within the overall cohort, there was no significant change in GERD symptoms from before to after surgery (mean GERD-HRQL scores were 6.1 before and after surgery, p = 0.981). However, in a subgroup analysis, patients without GERD preoperatively demonstrated a worsening in mean GERD-HRQL scores after surgery (from 2.4 to 4.5, p = 0.0020). The percentage of change in the usage of medications to treat GERD was not statistically significant (from 37 to 32%, p = 0.233). The percent of patients satisfied with their condition postoperatively was significantly increased in those with preoperative GERD, older age, crural repair intraoperatively, and in those with the highest %TWL. CONCLUSION/CONCLUSIONS:These results suggest that while overall LSG does not significantly affect GERD symptoms, patients without GERD preoperatively may be at risk for developing new or worsening GERD symptoms after surgery. It is important to remark that this is a review of the patient's clinical symptoms of GERD, not related to any endoscopic, pathological, or manometry studies. Such studies are necessary to fully establish the effect of LSG on esophageal health.

Background/UNASSIGNED:. Bariatric surgery reduces progression of CKD to ESKD, but the risk of perioperative complications remains a concern. Methods/UNASSIGNED:The 24-month data spanning 2017-2018 were obtained from the Metabolic and Bariatric Surgery Quality Improvement Program (MBSAQIP) database and analyzed. Surgical complications were assessed on the basis of the length of hospital stay, mortality, reoperation, readmission, surgical site infection (SSI), and worsening of kidney function during the first 30 days after surgery. Results/UNASSIGNED:<0.001) times more likely to die compared with those with normal kidney function. However, absolute mortality rates remained relatively low at 0.53% in those with stage 3b CKD. Furthermore, absolute mortality rates were <0.5% in those with stages 4 and 5 CKD, and these advanced CKD stages were not independently associated with an increased risk of early postoperative mortality. Conclusions/UNASSIGNED:Increased severity of kidney disease was associated with increased complications after bariatric surgery. However, even for the population with advanced CKD, the absolute rates of postoperative complications were low. The mounting evidence for bariatric surgery as a renoprotective intervention in people with and without established kidney disease suggests that bariatric surgery should be considered a safe and effective option for patients with CKD.

Background/Purpose: Symptomatic knee osteoarthritis (SKOA) patients with obesity who undergo bariatric surgery experience knee pain relief, though the reduced mechanical load explains only part of the improvement. A reduction in inflammatory biomarkers from adipose tissue may also impact pain. We previously identified an IL1RN haplotype (TTG; rs419598, rs315952, and rs9005) that associates with OA severity and inflammatory markers. We aimed to determine whether TTG distinguishes patients who lose more weight and have more significant decreases in inflammation with greater knee OA pain relief.
Method(s): From 2013-2019 we enrolled patients >=30 years old with BMI >=30 kg/m 2 and painful knee OA who planned surgical (sleeve gastrectomy, gastric bypass, or laparoscopic band) or medical weight loss (MWL) at Bellevue Hospital or NYU Langone Health. Patients with lupus, rheumatoid arthritis, or psoriatic arthritis were excluded. Weight-bearing knee x-rays assessed OA severity to confirm a Kellgren-Lawrence grade of at least 1 (scale 0-4). Participants completed the Knee Injury and Osteoarthritis Outcomes (KOOS) questionnaire and provided blood at baseline and 1, 3, 6, and 12 months. Patients were genotyped to determine whether they carried 1 or 2 copies of the TTG haplotype (TTG-1/2) or none (TTG-0). Sleeve was the most common weight loss intervention, therefore our analysis is focused on this surgical subset to minimize variable effects on weight and biomarkers.
Result(s): We enrolled 113 patients (95 F, 18 M) with painful knee OA prior to their weight loss intervention. The mean age, BMI, and KOOS pain at baseline were 50.3 +/- 12.0 years, 44.8 +/- 8.9 kg/m 2, and 48.4 +/- 18.2 (0-100, with 100 = no pain). Of 113 patients, 48 underwent sleeve, 20 bypass, 9 laparoscopic banding, 12 did not have the surgery, and 24 pursued medical weight loss. The 77 who completed surgery had a mean % excess weight loss (%EWL) of 51.7 after 6 months, with significant decreases in hsCRP (4.4 mg/L) and leptin (32.8 ng/dL), and mean KOOS pain improvement of 22.4 (MCID= 16.7). The corresponding changes for patients who tried various MWL regimens were modest at best. We obtained the IL1RN haplotype for 45 of the 48 sleeve patients, and found 34 (70.8%) carried the TTG-1 or TTG 2 haplotype while 11 were TTG-0 (with similar baseline age, BMI, and KOOS for the two groups). At each follow-up time point through 6 months (Figure 1), TTG-1/2 patients had more difficulty losing weight than the TTG-0 group (p< 0.005 by ANOVA), with corresponding smaller reductions in hs CRP (p=0.36) and leptin (p=0.006). TTG-1/2 carriers also reported less KOOS pain relief relative to the TTG-0 group (p=0.021), markedly at 1 and 3 months with some improvement later (Table 1). All of these findings held true when only plotting data only from the 23 patients (18 TTG-1/2, 5 TTG-0) who completed each of the followup visits.
Conclusion(s): SKOA patients with obesity achieve marked excess weight loss, reductions in inflammatory mediators, and knee pain relief with bariatric surgery. The subset of patients with the TTG-0 IL1RN haplotype demonstrated more significant and/or rapid improvement in each of these outcomes, suggesting a potential predictor of which OA patients will have a more successful response to bariatric surgery

Background/Purpose: High body mass index (BMI, kg/m2) is a modifiable risk factor that has been associated with the development and progression of osteoarthritis (OA) and knee pain. While total knee arthroplasty (TKA) is the gold standard for the treatment of end stage OA, morbidly obese patients (BMI>=40kg/m2) are often required to lose weight prior to TKA due to increased surgical risk and a higher rate of complications. While conservative weight-loss often fails to help these patients, bariatric surgery can be an alternative option. Here we present interim data from the trial entitled "Surgical Weight-loss to Improve Functional Status Trajectories following arthroplasty for painful knee osteoarthritis". This current multi-center, prospective study compares pain and functional outcomes in patients receiving bariatric surgery prior to TKA versus obese patients who go straight to TKA.
Method(s): Patients with BMI >=40 kg/m2 and painful knee osteoarthritis who met the indications for TKA were recruited at four hospital centers. Patients with a BMI >35 kg/m2 were also recruited if they had a qualifying comorbid condition including obstructive sleep apnea, diabetes, hypertension or hyperlipidemia. Patients were assigned to either the bariatric (BAR) or TKA arm based on surgical choice (goal n=150 for each arm), with all bariatric patients having anatomy-altering sleeve gastrectomy or gastric bypass. At baseline and several time points after surgery (Figure 1), we documented height, weight, the Knee injury and Osteoarthritis Outcome Score (KOOS), visual analog pain (VAS) scales, and the Western Ontario and McMaster Universities Arthritis Index (WOMAC), and had patients perform functional assessments (Timed-Up and Go, 30-second Chair Stand and 40-meter fast paced walk test). We targeted minimum detectable change (MDC) in outcomes for the VAS for knee pain (33% reduction), Timed Up and Go (decrease by 2 seconds), 30-second Chair Stand (increase by 2 reps), 40-meter fast paced walk (increase by 0.16 m/s), WOMAC score (16% reduction), and the KOOS pain score (10-point improvement). Using a logistic regression to adjust for age and baseline BMI, we compared the percentage of patients in the two arms who achieved an MDC for the various outcomes.
Result(s): To date, 25 BAR and 28 TKA patients have completed their follow-up visits through at least 6 months. Although there was a similar sex distribution, the bariatric group was younger (52 vs 60 years old, p=0.0023) with a higher baseline BMI (47.0 vs 41.6 p=0.0006). Most bariatric patients achieved comparable improvement to the TKA cohort with regards to the benchmarks of the 30-second Chair Stand (TKA 54% vs BAR 33%, p=0.156), KOOS pain score (TKA 91% vs BAR 67%, p=0.130), the Visual Analog Pain Scale (TKA 50% vs BAR 39%, p=0.466), Timed Up and Go test (TKA 43% vs BAR 22%, p=0.141) and the 40-meter fast paced walk (TKA 61% vs BAR 35%, p=0.073). The TKA cohort had a greater percent with a MDC for the WOMAC (TKA 88% vs BAR 54%, p=0.030).
Conclusion(s): In morbidly obese patients who are eligible for TKA, bariatric surgery may result in modest improvements in knee outcomes and may eventually delay the need for a TKA

In their relative infancy, intragastric balloons (IGB) offer a solution to patients who do not qualify for bariatric surgery due to their body mass index (BMI) or those reticent about major surgery with a promise of 10-15% of total body weight loss (TBWL%). Given a short implantation period of 6 months, weight regain following IGB removal has been commonly noted. This caveat prompted analysis of the addition of weight loss medications for improved efficacy in achieving sustained weight loss results post-IGB removal. In a single-center retrospective analysis from 2015 to 2018, 18 patients (mean age 39.72, 5 males, 13 females) with a saline-filled single intragastric balloon were evaluated for 12 months following IGB insertion. Exactly half of the patients (n=9) were on weight loss medications before, during, or after placement of IGB (IGB-M cohort) to compare to patients with IGBs only (IGB-O cohort). All patients were >18 years old, non-pregnant, and with no previous bariatric surgeries. Data was collected at 0, 3, 6, and 12 month intervals. Mean weight at baseline was 198.33lbs and 223lbs (p=0.814) and mean BMI was 32.79 kg/m2 versus 35.5 kg/m2 (p=0.546), in the IGB-O cohort versus the IGB-M cohort, respectively. At six months, the TBWL% in the IGB-O cohort versus the IGB-M cohort was 12.7% versus 13.1%, while mean BMI was 28.42 versus 31.62 (p=0.645), respectively. Attrition rate was 11.1% by 6 months and 72.2% by 12 months for both cohorts combined. At 12 months, TBWL% in the IGB-O cohort versus IGB-M cohort was 2.8% and 10.7%, while mean BMI was 33.77 and 29.17 (p=0.4), respectively. The most common class of medications utilized were glucagon-like peptide-1 (GLP-1) agonists (37.5%). Phentermine was the single most commonly prescribed medication (25%). The mean number of medications needed for a patient was 1.8. The TBWL% at six months demonstrated a slightly greater 0.4 TBWL% in the IGB-M cohort. This meant weight loss achieved with a balloon or weight loss pharmacotherapy was essentially equivalent in our study at the time of IGB removal at 6 months. However, a stark variance is noted at the 12 month mark in the IGB-O cohort with patients either partially regaining their previously lost weight or losing marginally further with a mean 2.8 TBWL% post IGB removal. Comparatively, the IGB-M cohort patients continued their weight loss or maintained their initial weight loss with a 10.7% TBWL creating a 7.9 TBWL% difference at the 12 month follow up between the two arms. Administration of medications might increase follow-up post IGB removal and decrease attrition rates. Results illustrate a two-pronged approach of combining weight loss medications with IGBs would culminate in a more clinically significant TBWL% with long term sustainability post IGB removal. Larger multi-center studies are recommended in order to achieve significant conclusions. [Formula presented] [Formula presented]
Copyright

Purpose: High body mass index (BMI, kg/m²) is a modifiable risk factor that has been associated with the development and progression of osteoarthritis (OA) and knee pain. The gold standard for the treatment of end stage knee OA remains total knee arthroplasty (TKA), though patients with morbid obesity (BMI>=40 kg/m²) are increasingly required to lose weight prior to TKA due to increased surgical risk and a higher rate of complications. Conservative weight-loss approaches often fail to help patients lose the recommended weight prior to TKA, potentially making them ineligible for surgery. Recent studies have shown that patients who undergo bariatric surgery experience short-term improvements in knee functionality and decreased pain, either obviating the need for arthroplasty - or at least improving the success of subsequent total knee replacements. Here we present interim data from the trial entitled "Surgical Weight-loss to Improve Functional Status Trajectories following arthroplasty for painful knee osteoarthritis". This current multi-center, prospective study compares pain and functional outcomes in patients receiving bariatric surgery prior to TKA versus obese patients who go straight to TKA.
Method(s): We recruited patients with BMI >=40 kg/m² and painful knee OA who met the indications for TKA from the bariatrics, orthopedics and rheumatology clinics at the four hospital centers. Patients with a BMI >35 kg/m² were also recruited if they had a qualifying comorbid condition including obstructive sleep apnea, diabetes, hypertension or hyperlipidemia. Patients were assigned to the bariatric (BAR) or TKA (TKA) arms based on surgical choice (goal n=150 for each arm), with all bariatric patients having anatomy-altering sleeve gastrectomy or gastric bypass, and not laparoscopic gastric banding. At baseline and several time points after surgery (see figure), we documented height, weight, the Knee injury and Osteoarthritis Outcome Score (KOOS), visual analog pain (VAS) scales, and the Western Ontario and McMaster Universities Arthritis Index (WOMAC), and had patients perform functional assessments (Timed-Up and Go, 30-second Chair Stand and 40-meter fast paced walk test). We targeted minimum detectable change (MDC) in outcomes for the VAS for knee pain (33% reduction), Timed Up and Go (decrease by 2 seconds), 30-second Chair Stand (increase by 2 reps), 40-meter fast paced walk (increase by 0.16 m/s), WOMAC score (16% reduction), and the KOOS pain score (10-point improvement). Using a logistic regression to adjust for age and baseline BMI, we compared the percentage of patients in the two arms who achieve an MDC for the various outcomes.
Result(s): For the current analysis, 20 bariatric surgery patients and 22 TKA patients completed their 6-month follow-up visits. Each subcohort had a similar sex distribution (15% vs 14%, p=0.999) but the bariatric group was younger (53 vs 61 years old, p=0.0045) and had a higher baseline BMI (47.1 vs 40.9, p=0.0009). Many bariatric patients achieved significant improvement in knee pain and function as evidenced by the MDC in the various tests, though the percent with MDC consistently tended to be higher in the TKA cohort. These MDC benchmarks included the 30-second Chair Stand (TKA 60% vs BAR 29%, p=0.085), KOOS pain score (TKA 89% vs BAR 65%, p=0.130), the Visual Analog Pain Scale (TKA 50% vs BAR 34%, p=0.390), Timed Up and Go (TKA 45% vs BAR 21%, p=0.205), the 40-meter fast paced walk (TKA 61% vs BAR 42%, p=0.319), and the WOMAC (TKA 81% vs BAR 59%, p=0.225). Many of the bariatric patients have had subsequent arthroplasty, but not enough time has passed to report levels of improvement.
Conclusion(s): In morbidly obese patients who are eligible for TKA, bariatric surgery may result in modest improvements in knee pain and function when compared to TKA at 6 months post operatively. In addition to potentially altering the need for and timing of TKA, surgical weight loss may result in improved TKA outcomes. This study was supported by the Investigator Initiated Study Program of Ethicon-Endo Surgery, Inc. (Grant Number: 14-621). [Formula presented]
Copyright

Background/Purpose : Weight loss in obese patients can reduce knee osteoarthritis (OA) pain, even when physical therapy and intra-articular injections have failed. The impacts of either non-surgical or surgical weight loss on knee OA pain have been reported separately, but few studies have assessed them conjointly. While the decrease in mechanical load helps, the contribution of metabolic changes is less clear. We aimed to compare biomarker changes with weight loss as predictors of knee pain improvement, and consider a threshold of total weight loss necessary for these changes. Methods : Patients from the NYU Langone Weight Management program were screened for knee pain prior to bariatric surgery or the start of a medical weight loss (MWL) regimen. We excluded patients with autoimmune disease, recent malignancy, recent intra-articular knee injections, and lack of OA by Kellgren-Lawrence (KL) x-ray grading. The BMI, Knee Injury and Osteoarthritis Outcome Score (KOOS) for pain, and blood samples were obtained at baseline and 1, 3, 6 and 12 months for evaluation of pain and biomarker levels. Results : Of 140 patients screened, 81 were eligible and enrolled (82.7% female; BMI 45.2+/-9.6 kg/m2, 31-74; age 52+/-12 years, 30-80). A total of 49 patients had surgery (10 bypass, 30 sleeve, 9 LapBand) and 24 medical weight loss. 33 patients completed visits up to 6 months (2 bypass, 18 sleeve, 6 LapBand, 7 MWL). By 1 month, the surgical patients had lost much more total weight than the MWL group (9.8% vs 4.1 %, p=0.001), and realized marked pain relief (p< 0.001). By 6 months both groups had continued to lose weight, proportionately greater for surgical patients with further pain improvement. (Figure 1) Leptin levels dropped at 1 and then 6 months with both methods of weight loss. The pro-inflammatory protein IL-1Ra decreased significantly by 6 months in the bariatric patients, but increased with the medical regimen across both time points. Soluble vascular adhesion protein 1 (sVAP-1), another pro-inflammatory protein that facilitates leukocyte infiltration, decreased at both the 1 and 6 month intervals -but much more in MWL than in surgical patients. Consistent with the literature, the anti-inflammatory soluble receptor for advanced glycation endproducts (sRAGE) mirrored KOOS pain improvement only in surgical patients and stabilized after 1 month, but did not change in the MWL group. (Table 1) In a subgroup analysis, the 14 surgical patients who lost at least 10% of Figure 1. Surgical and medical outcomes for % total weight loss (TWL), knee pain and biomarkers Table 1. total weight by 1 month had significantly less pain at 6 months than the 12 who did not meet the threshold (DELTAKOOS 47.5 vs 29.9) but the biomarker levels were similar. (Figure 2) Conclusion : Surgical and medically supervised weight loss both lead to significant decreases in adiposity, but only those having bariatric surgery realize significant pain relief. The anatomical changes of surgical (vs. medical) weight loss result in different metabolic cascades given divergent biomarker trends. Bariatric patients who lose more than 10 percent of total body weight within the first month are more likely to have better pain relief by 6 months, but the biomarker changes reflect anatomic intervention -and are not dependent on the degree of surgical weight loss

INTRODUCTION: Bariatric surgery has become a common therapeutic approach to obesity. However, bariatric procedures may affect bowel habits due to changes in dietary intake as well as altered anatomy. To date, few studies have evaluated the impact of bariatric surgery on post-operative constipation. The aim of this study is to determine if patients experience a greater rate of constipation after bariatric surgery compared to non-bariatric controls.
METHOD(S): A retrospective chart review at New York Langone Hospital (NYU) was performed on 160 bariatric surgery patients who had surgery in the year 2012 and 160 control patients with BMI < 30 kg/m2 seen in primary care in 2012. Reports of constipation were recorded up until November 2018. Exclusion criteria included those with inflammatory bowel disease or hereditary colorectal cancer syndromes prior to age 50. The primary outcome was diagnosis of post-operative constipation. Secondary outcomes included rates of constipation according to surgical procedure. Presence of constipation was recorded if listed on the problem list or if medications for constipation were prescribed. Logistic regression and chi-squared testing was used to assess differences in groups.
RESULT(S): Table 1 shows patient characteristics of the study population. The average age of bariatric surgery patients was 64.1 years compared to 69.8 in the control group. Overall, 20% of bariatric patients were diagnosed with constipation compared to 15% of controls (P = 0.239). Constipation rates post-bariatric surgery were 17.9% in lap band, 20.6% in sleeve gastrectomy, and 7.1% in gastric bypass patients (P = 0.256). A logistic regression controlling for age, sex, and Charlson co-morbidity index was performed between controls and post-surgery subjects. This revealed no significant difference in rates of constipation between the two groups (OR 1.158, 95% CI 0.790 - 1.696 P-value = 0.45). There was, however, a significant difference in constipation rates between the bariatric group pre-surgery (13.8%) and post-surgery (17.5%) (P = < 0.001) (Table 2).
CONCLUSION(S): Bariatric surgery patients experience significantly higher rates of constipation after surgery compared to prior to surgery, but similar rates to controls without obesity. Constipation impairs quality of life and is associated with significant health care costs. Further studies investigating the mechanisms underlying this increase in constipation after bariatric surgery and effective measures to treat it are warranted. (Figure Presented)

Introduction: The gastrojejunostomy (GJ) during Roux-en-Y gastric bypass (RYGB) can be performed by stapled or hand-sewn techniques, and is at risk for anastomotic stricture, reported in the literature at rates from 0 to 33%. This study reviews a single center's experience with anastomotic stricture and intervention required. Methods and Procedures: A retrospective chart review was performed of 904 patients who underwent RYGB as primary or revisional surgery at a single institution from October 2000 through September 2017. There were 182 patients excluded for follow up duration of less than 1 year, 5 for an esophagojejunostomy rather than GJ, and 1 for gastroparesis as the surgical indication rather than morbid obesity. This left 716 patients to be included in the study. Demographic and operative data were collected including technique for GJ, postoperative follow up, and complications, with a focus on GJ stricture and subsequent interventions.
Result(s): Gastrojejunostomy (GJ) was performed with a 25 CEEA stapler in 674 (94.1%) patients, with a linear stapler in 25 (3.5%), was hand-sewn in 7 (1.3%), and the technique was unknown in the remaining 8 (1.1%). Roux-en-Y gastric bypass was performed as a primary surgery in 522 (72.9%) patients and as a revisional surgery in 194 (27.1%). Stricture of the GJ was diagnosed in 29 (4.1%) patients. The average time to diagnosis of early strictures occurring prior to 3 months was 40.3 days, and for late strictures was 871 days. By technique, stricture was diagnosed in 26 (3.9%) patients in the 25CEEAgroup, 1 (4%) in the linear stapler group, and 2 (22.2%) in the hand-sewn anastomosis group. In primaryRYGBpatients stricture was diagnosed in 20 (3.8%) patients, and in revisionalRYGB in 9 (4.6%) patients (p = 0.626). Esophagogastroduodenoscopy (EGD) with dilation was performed at least once (1-9 times) in 26 patients, 2 with concomitant stenting, 2 required operative intervention, and 1 patient awaits operative intervention. Both patients who required surgery also had marginal ulcers, and possible gastro-gastric fistula at time of surgery.
Conclusion(s): The results of this study show that the 25 CEEA circular stapler is a reasonable technique for performance of the GJ anastomosis in RYGB, with a stricture rate of 3.9%. There is also a slightly increased stricture rate in revisional surgical patients, though not statistically significant

BACKGROUND:When administered as a continuous infusion, ketamine is known to be a potent analgesic and general anaesthetic. Recent studies suggest that a single low-dose administration of ketamine can provide a long-lasting effect on mood, but its effects when given in the postoperative period have not been studied. OBJECTIVE:We hypothesised that a single low-dose administration of ketamine after bariatric surgery can improve pain and mood scores in the immediate postoperative period. DESIGN/METHODS:We performed a randomised, double-blind, placebo-controlled study to compare a single subanaesthetic dose of ketamine (0.4 mg kg) with a normal saline placebo in the postanaesthesia care unit after laparoscopic gastric bypass and gastrectomy. SETTING/METHODS:Single-centre, tertiary care hospital, October 2014 to January 2018. PATIENTS/METHODS:A total of 100 patients were randomised into the ketamine and saline groups. INTERVENTION/METHODS:Patients in the ketamine group received a single dose of ketamine infusion (0.4 mg kg) in the postanaesthesia care unit. Patients in the placebo groups received 0.9% saline. OUTCOME MEASURES/METHODS:The primary outcome was the visual analogue pain score. A secondary outcome was performance on the short-form McGill's Pain Questionnaire (SF-MPQ). RESULTS:There were no significant differences in visual analogue pain scores between groups (group-by-time interaction P = 0.966; marginal group effect P = 0.137). However, scores on the affective scale of SF-MPQ (secondary outcome) significantly decreased in the ketamine group as early as postoperative day (POD) 2 [mean difference = -2.2 (95% bootstrap CI -2.9 to 1.6), Bonferroni adjusted P < 0.001], compared with placebo group in which the scores decreased only by POD 7. Scores on the total scale of SF-MPQ for the ketamine group were smaller compared with the placebo group (P = 0.034). CONCLUSION/CONCLUSIONS:Although there was no significant difference between ketamine and placebo for the primary outcome measure, patients who received ketamine experienced statistically and clinically significant improvement in their comprehensive evaluation of pain, particularly the affective component of pain, on POD 2. However, future studies are needed to confirm the enduring effects of ketamine on the affective response to postoperative pain. CLINICAL TRIAL REGISTRATION/BACKGROUND:NCT02452060.