Predictors of Risk and Success of Obesity Surgery

Abstract

<b><i>Background:</i></b> Obesity surgery has proven successful for weight loss and the resolution of comorbidities. There is, however, little evidence on its success and the risk of complications when considering age of onset of obesity (AOO), years of obesity (YOO), preoperative body mass index (BMI), Edmonton obesity staging system (EOSS) score, and age as possible predictors of weight loss, the resolution of comorbidities, and the risk of complications. <b><i>Methods:</i></b> Patients who underwent Roux-en-Y gastric bypass (RYGB) and laparoscopic sleeve gastrectomy (LSG) from a prospective database were analyzed. Multiple regression analyses were used to predict comorbidities and their resolution, percentage excess weight loss (%EWL) and total weight loss (%TWL) 12 months after surgery, and the risk of complications using the predictors AOO, YOO, age, EOSS, and BMI. <b><i>Results:</i></b> 180 patients aged 46.8 ± 11.1 years with a preoperative BMI 49.5 ± 7.5 were included. The number of preoperative comorbidities was higher with older age (β = 0.054; <i>p</i> = 0.023) and a greater BMI (β = 0.040; <i>p</i> = 0.036) but was not related to AOO and YOO. Patients with AOO as a child or adolescent were more likely to have an EOSS score of ≥2. Greater preoperative BMI was negatively associated with %EWL (β = –1.236; <i>p</i> < 0.001) and older age was negatively associated with %TWL (β = –0.344; <i>p</i> = 0.020). Postoperative complications were positively associated with EOSS score (odds ratio [OR] 1.147; <i>p</i> = 0.042) and BMI (OR 1.010; <i>p</i> = 0.020), but not with age. AOO and YOO were not related to postoperative outcome. <b><i>Conclusion:</i></b> Greater BMI was associated with a lower %EWL and age was associated with a low %TWL. YOO and AOO did not influence outcome. Age, BMI, and EOSS score were the most important predictors for risk and success after obesity surgery. Surgery should be performed early enough for optimal outcomes.

Usefulness of the Edmonton Obesity Staging System for stratifying the presence and severity of weight‐related health problems in clinical and community settings: A rapid review of observational studies

“…All of the five studies in clinical populations following bariatric surgery consistently showed that the risk of postoperative complication outcomes, 30‐day mortality and years of life lost increased with EOSS, especially for EOSS 3–4 17–19,30,31 . One of these studies reported an increased odds ratio with the EOSS for hospital length of stay (OR 1.175, p = 0.02), 31 whereas another reported no such association with hospital length of stay, readmission rate and minor complications 30 .…”

“…These included four cohort studies, 21,22,28,29 seven case series 17–20,30–32 and nine cross‐sectional studies 33–41 . Five studies were from the United States, 21,22,29,35,41 four from Canada, 19,20,28,36 four from Germany, 17,18,31,32 two from Italy 39,40 and one each from Australia, 33 Spain, 34 Poland, 30 South Korea 37 and Brazil 38 . Of these, 12 studies were in clinical populations with sample sizes between 100 and 400 in six studies, 19,28,31,32,34,41 500 and 1000 in three studies 17,38,39 and greater than 1000 in two studies 18,20 .…”

“…Five studies were from the United States, 21,22,29,35,41 four from Canada, 19,20,28,36 four from Germany, 17,18,31,32 two from Italy 39,40 and one each from Australia, 33 Spain, 34 Poland, 30 South Korea 37 and Brazil 38 . Of these, 12 studies were in clinical populations with sample sizes between 100 and 400 in six studies, 19,28,31,32,34,41 500 and 1000 in three studies 17,38,39 and greater than 1000 in two studies 18,20 . The other eight were in community populations with sample sizes between 100 and 1000 in three studies, 35,36,40 8000 and 10 000 in two studies 21,33 and 17 000 and 30 000 in three studies 22,29,37 .…”

“…Despite this inconsistency, the proportions of clinical populations within EOSS 0 (1%–9%), 1 (12%–39%), 2 (31%–79%), 3 (4%–27%) and 4 (6%) categories and community populations within EOSS 0 (2%–5%), 1 (0%–42%), 2 (43%–87%), 3 (3%–16%) and 4 (0%–7%) categories were similarly distributed. Outcomes in studies of clinical populations were mostly postoperative complications following bariatric surgery, 17–19,30,31 treatment outcomes including postoperative changes in nonalcoholic fatty liver disease (NAFLD), 19,20,30–32 as well as associations with cesarean rates, 28 disability, 39 cardiovascular disease risk markers, 41 age/sex, BMI 38,39 and lifestyle risk factors 34 . Outcomes in studies of community populations were mortality, 21,22,29 lifestyle risk factors, 35 quality of life, skeletal strength and fitness in young participants, 36 and associations with obesity, 35,37 polypharmacy and health services use, 33 as well as work ability 40 …”

“…We present the results of our quality assessment of each study in the supporting information (Table S3). Cohort studies had eight, 28 nine 21,29 and 10 22 out of 11 quality items; case series studies had six, 30–32 eight, 17 nine, 18,20 and 10 19 out of 10 quality items; and cross‐sectional studies had four, 41 five, 34,35,38,39 six 36,37 and seven 33,40 out of eight quality items.…”