Introduction
            With the rapid increase in US obesity rates, growing numbers of patients are presenting post-massive weight loss (MWL).  We sought to characterize demographic and outcome differences between patients who underwent gastric bypass surgery (GBP) and those who lost weight exclusively through diet and exercise (DE).
Methods
            Over six years, 449 patients representing 511 post-MWL body contouring cases were entered into a prospective registry.  Entry into the registry was based on body contouring surgery after a loss of >50 pounds. 
            DE patients were matched in a 1:n fashion to 191 GBP patients based on procedures performed (brachioplasty, breast surgery, thigh lift, lower body lift, upper body lift, cosmetic facial surgery and/or panniculectomy).  Fixed-effects and conditional logistic regression were used to test for differences between groups.  1:1 matching was then performed by nearest neighbor matching to the most similar GBP patient based on gender, age and BMI.  The t-test was used to compare matched patients.
Results
Twenty-nine patients representing 34 cases (6.7%) lost weight exclusively through DE.  Overall, DE patients were not significantly different from GBP patients in age, gender, maximum (pre-MWL) BMI or BMI at time of body contouring.  DE patients had significantly higher pre-operative hematocrit (39.6 versus 37.6; p = 0.02) and hemoglobin (14.4 versus 12.8; p = 0.05).  DE patients tended to have multiple procedures performed in the same case more often (74% versus 61%) and had higher absolute complication rates (50% versus 41%; p=0.3).  Infection rates were the only individual complication to reach statistical significance (18% versus 7%; p = 0.03).  When matched based on procedures performed, DE patients had a higher but not significant complication rate (OR 1.4; p = 0.38).
One-to-one matching resulted in 34 procedure-matched pairs that were not significantly different by age, gender, current BMI, maximum BMI or intraoperative time.  DE patients still had more complications (50% versus 38%; p = 0.3), with higher rates of wound dehiscence and infection (18% versus 9% and 27% versus 12% respectively; p = 0.3 and 0.09), a similar rate of hematomas (3%) and a lower rate of seromas (21% versus 29%). 
Pair-matched DE patients had a higher albumin in the 20 patients with preoperative measurement (3.6 versus 4.0; p = 0.06).  Twenty-two patients received a preoperative consultation from a nutritionist regarding protein intake; 7 of 9 DE patients (78%) exceeded the median daily protein intake (60g) while only 6 of 13 GBP patients did (46%; p = 0.15).  DE patients also had a lower rate of self-reported history of anemia (20% versus 8%).
Discussion
Post-MWL reconstruction is one of the fastest growing areas in plastic surgery.  Postbariatric status, which is well correlated with both caloric and nutritional deficits, has motivated much of the emphasis on monitoring and improving nutrition before surgery in this patient population.
As might be expected, metabolic assessments and protein intake appeared better in the cohort of patients who lost weight through diet and exercise rather than through bypass surgery.  However, this was not borne out in surgical outcomes with non-bypass patients experiencing more complications even after controlling for procedures performed and demographic factors.  Definitive conclusions are difficult due to the relatively small proportion of patients who undergo massive weight loss without surgical intervention, but we found no evidence to suggest that postbariatric status adds additional risk beyond that of massive weight loss itself.  A comprehensive assessment of nutritional status is an important part of patient selection in this population.