With the substantial increase in the number of obese Americans over the past several decades, the use of bariatric surgery has increased as well. According to a statistical report from the Agency for Healthcare Research and Quality (AHRQ), the annual number of bariatric surgeries in the United States increased from 13,386 to 121,055 between 1998 and 2004, a change of more than 800% [23]. According to the American Society for Metabolic and Bariatric Surgery (ASMBS), the number of bariatric surgeries performed in the United States increased from 158,000 in 2011 to 256,000 in 2019, an increase of 62%, with the biggest jump occurring between 2017 and 2018 [24].

Weight-loss surgeries most commonly used in the United States are the laparoscopic sleeve gastrectomy (LSG), or "sleeve," and Roux-en-Y gastric bypass (RYGB), surpassing the historically popular laparoscopic adjustable gastric band (LAGB), or "band." RYGB is a mixed restrictive/malabsorptive procedure, while LSG and LAGB are purely restrictive. Based on data from the University HealthSystem Consortium Clinical Database, gastric bypass made up 66% of bariatric surgeries performed at academic medical centers in 2007, while LAGB accounted for 23% [30]. By 2016, LSG had become the leading procedure performed, accounting for 58.1% of bariatric surgeries, compared with 17.8% in 2011 [24]. In 2019, RYGB comprised 17.8% and LAGB made up only 0.9% [24]. Certain other surgeries, previously common, have fallen out of favor due to high complication rates. They are described briefly in this course because patients who had these surgeries will still be seen in primary and specialty care. Many publications regarding bariatric surgery incorporate multiple procedures or variations on RYGB; as much as possible, the original terminology will be used when discussing each study.

Weight-loss surgeries most commonly used in the United States are the laparoscopic sleeve gastrectomy (LSG), or "sleeve," and Roux-en-Y gastric bypass (RYGB), surpassing the historically popular laparoscopic adjustable gastric band (LAGB), or "band." RYGB is a mixed restrictive/malabsorptive procedure, while LSG and LAGB are purely restrictive. Based on data from the University HealthSystem Consortium Clinical Database, gastric bypass made up 66% of bariatric surgeries performed at academic medical centers in 2007, while LAGB accounted for 23% [30]. By 2016, LSG had become the leading procedure performed, accounting for 58.1% of bariatric surgeries, compared with 17.8% in 2011 [24]. In 2019, RYGB comprised 17.8% and LAGB made up only 0.9% [24]. Certain other surgeries, previously common, have fallen out of favor due to high complication rates. They are described briefly in this course because patients who had these surgeries will still be seen in primary and specialty care. Many publications regarding bariatric surgery incorporate multiple procedures or variations on RYGB; as much as possible, the original terminology will be used when discussing each study.

In 2019, the American Association of Clinical Endocrinologists (AACE), the Obesity Society (TOS), and the ASMBS released updated guidelines for the perioperative care of the bariatric surgery patient that increased the number of total recommendations from 74 to 85 [11,33,145].Current selection criteria include BMI >40 if no comorbidities are present, >35 if there is one or more severe obesity-associated comorbidity, or greater than 30 with diabetes or metabolic syndrome [11,145]. Guidelines indicate that the BMI criterion for bariatric procedures should be adjusted for ethnicity (e.g., among Asian patients, 18.5–22.9 is normal range, 23–24.9 is overweight, and ≥25 is obesity) [11]. In addition, patients must have tried and failed non-surgical weight reduction and be willing and able to adhere to postoperative care. Any patient with current alcohol or drug abuse, psychiatric illness that is uncontrolled, or underlying disorder causing the obesity should undergo a formal mental health evaluation. Finally, patients must understand the risks, benefits, alternatives, necessary lifestyle changes, and expected outcomes. Since publication of these criteria, sufficient data were presented to the U.S. Food and Drug Administration (FDA) that led to the approval of more relaxed criteria (i.e., BMI less than 35 with an obesity-related comorbidity [mild obesity]) for LAGB. Researchers have demonstrated comparable safety and efficacy of LAGB between mildly obese and more severely obese patients [11; 34; 145].

The increase in the number of LSG procedures reflects several advantages over other types of bariatric surgery. Initial excess weight loss is slightly less than RYGB (60% to 80% RYGB vs. >50% LSG); however, long-term weight maintenance rates are comparable at approximately 50% [166]. LSG requires no foreign objects in the body, as LAGB does, and does not re-route the food stream, as in RYGB. LSG hospitalization stays are shorter than other procedures, averaging two days [166].

Because RYGB alters the configuration of the digestive tract, it changes the body's response to certain foods. A "dumping syndrome" may occur, particularly with the ingestion of foods with high sugar content. Within a short time after eating, patients with dumping syndrome experience lightheadedness, palpitations, flushing, and diarrhea. Dumping syndrome occurs in 70% or more of gastric bypass patients initially [33,145]. In some, it resolves over time, but others have ongoing intolerance to certain foods. Some experts and patients feel that dumping syndrome is actually an advantage, because it discourages consumption of high-calorie, low-nutrient foods [166]. Reversal of RYGB has been proven as a safe and effective way to treat dumping syndrome [46,167].

In 2007, Buchwald and colleagues conducted a meta-analysis of mortality data using studies published between 1990 and 2006 [67]. Based on a total of 361 studies including 478 treatment arms and 85,048 patients, they found an overall mortality rate of 0.28% within the first 30 days and 0.35% between 31 days and two years. For gastric bypass, 30-day mortality was 0.44% for open procedures and 0.16% for laparoscopic procedures. Mortality from 31 days to two years was 0.69% and 0.09%, respectively. For gastric banding, open procedures had a short-term mortality rate of 0.18%, while the short-term mortality for laparoscopic procedures was 0.06%. The longer-term mortality rates were statistically 0.00% for both groups. For the most part, this analysis found that mortality trended downward with more recent studies, and smaller studies had higher mortality rates than larger ones. Mortality was highest in observational studies (0.7%) compared with other study designs (0.07% to 0.30%) [67]. In addition, mortality among patients who have undergone LSG have shown to be similar to those of the more well-studied procedures, with a mortality rate ranging from 0% to 1.2% depending on study type [178]. Nguyen and colleagues conducted an audit of bariatric surgery cases at 29 institutions participating in the University HealthSystem Consortium Bariatric Surgery Benchmarking Project [30]. For each institution, 40 consecutive cases were examined; a total of 1,144 cases met inclusion criteria, which was age older than 17 years and younger than 65 years, BMI of 35–70, and no previous bariatric surgery. Procedures were primarily gastric bypass (91.7%), with smaller numbers of gastroplasty or gastric banding (8.2%) and BPD (0.1%). For gastric bypass, with about three-fourths of the procedures done laparoscopically, 30-day mortality was 0.4%. Restrictive procedures had a 30-day mortality of 0%, with 92% of procedures done laparoscopically. Data support the low incidence of severe adverse events and mortality.

Other studies have shown that increased physician experience and higher case volumes are associated with lower mortality. For example, lower mortality rates have been reported at hospitals doing more than 100 bariatric surgeries annually compared with hospitals with lower numbers. Length of stay, morbidity, and costs were also lower at the high-volume institutions [68]. Concerns regarding the safety and uneven quality of bariatric surgeries performed across hospitals prompted the American College of Surgeons (ACS) and the ASMBS to implement an accreditation program for bariatric surgery centers—the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program [69]. The general guidelines to receive accreditation vary between programs but typically include a minimum volume of procedures, availability of resources for morbidly obese patients, and submission of outcomes data to a central registry [69,70]. The ASMBS and ACS also partnered with the Society for American Gastrointestinal and Endoscopic Surgeons to establish credentialing guidelines for bariatric surgeons to ensure that surgeons maintain a certain skill level and are prepared for potential complications during bariatric surgery [71].

Some adverse effects are not technically surgical complications, but occur as a result of rapid weight loss. Cholelithiasis is a common result of rapid weight loss and is frequently seen in bariatric surgery patients [60]. Estimates of symptomatic cholelithiasis after RYGB, for example, range from 3% to 28% in various studies [74].

The current recommendation from the American College of Obstetricians and Gynecologists (ACOG) is that women should delay pregnancy for 12 to 24 months after bariatric surgery to ensure that gestation does not occur during the rapid weight-loss phase [78,79]. However, the opportune timing of pregnancy after surgery is unknown. The ACOG also strongly recommends preconception assessment and counseling and education regarding possible complications. Prior to attempting pregnancy, obese patients should be encouraged to undertake a weight-reduction program that includes diet, exercise, and behavior modification. Evaluation for nutritional deficiencies and the need for vitamin supplementation are also recommended [80].

Weight loss after bariatric surgery is usually most rapid in the first year. It may be fastest in the first few months, when caloric restriction is greatest. Weight loss is expected to slow at about six to nine months, and maximal total loss generally occurs at around 12 to 18 months [33].

The AHRQ evidence report related that, in published bariatric surgery case series, diabetes improved or resolved in 69% to 100% of cases [23]. In the meta-analysis by Buchwald and colleagues, among studies that reported resolution of diabetes, 76.8% of patients had complete resolution [7]. In studies that also reported improvement, 86.0% had either resolution or improvement. A 2007 review found that diabetes resolved in more than 75% of bariatric surgery patients [9].

In 2008, researchers published data from a randomized controlled trial comparing lifestyle change, including the option of medication to treat obesity, to LAGB in patients with type 2 diabetes [101]. Out of 60 patients enrolled, 55 were followed to two years. Starting BMI was between 30 and 40, and diabetes diagnosis was recent, having been made within the past two years. The surgery group lost 62.5% of excess body weight, and 73% experienced remission of type 2 diabetes. In the non-surgical group, excess weight loss was 4.3% and diabetes remission was 13%. Remission of diabetes correlated with weight loss and also with lower hemoglobin A_1c (HbA_1c) levels at baseline. A study published in 2012 showed remission of type 2 diabetes in 62% of RYGB patients at a six-year follow-up [180].

While malabsorptive procedures may have additional mechanisms of action against diabetes, similar changes in gut hormones do not occur with purely restrictive procedures [111]. The reduction in diabetes associated with LAGB appears to be due to weight loss alone.

Studies with this length of follow-up are uncommon, but some additional evidence is available on blood pressure several years after surgery. White and colleagues used data from a single surgeon's gastric bypass cases (variations on RYGB), collected over 14 years, to examine outcomes including the resolution of hypertension [112]. With a median follow-up of just over four years, 62% of previously hypertensive patients had normal blood pressure and 25% showed improvement.

Changes in lipids are also widely seen in follow-up studies of bariatric surgery patients, although long-term data are somewhat mixed [10]. In the SOS cohort, rates of recovery from hypertriglyceridemia and from low high-density lipoprotein (HDL) were better in surgery patients than in the control group at both 2 and 10 years of follow-up. Recovery from hypercholesterolemia, on the other hand, was not statistically different in surgical patients compared with controls at either time point. The incidence of hypercholesterolemia was similar as well.

As discussed, in the short term there is a small but definite mortality risk associated with bariatric surgery. However, long-term mortality data suggest that, compared with obese controls, patients who choose surgery experience a reduced risk of premature death. Data from the SOS study show that, with an average of 10.9 years of follow-up, there were 129 deaths in the control group and 101 deaths in the surgery group. The unadjusted overall hazard ratio was 0.76 in the surgery group [129]. A review of data from the SOS study at 20 years follow-up found a long-term reduction in overall mortality as well as decreased incidences of diabetes, myocardial infarction, stroke, and cancer with bariatric surgery compared with usual care [84].

Similar benefit was noted in a retrospective cohort study that compared mortality among 9,949 gastric bypass patients and 9,628 severely obese controls [130]. Matching for age, sex, and BMI was achieved in 7,925 of each group, and the mean follow-up was 7.1 years. Adjusted long-term mortality from any cause decreased by 40% in the surgery group compared with the controls, with 37.6 and 57.1 deaths, respectively, per 10,000 person-years. Cause-specific mortality in the surgery group decreased by 56% for coronary artery disease, by 92% for diabetes, and by 60% for cancer. Rates of death not caused by disease, such as accidents and suicide, were 58% higher in the surgery group than in the control group.

Symptomatic thiamine deficiency after bariatric surgery is not usual, but cases of Wernicke-Korsakoff syndrome, a degenerative brain disorder, after both malabsorptive and restrictive procedures have appeared in the literature [135]. Patients who have unresolved nausea and vomiting may be particularly at risk. Case reports of beriberi have also been published, and the ASMBS guideline notes that occurrence may, in fact, not be rare [137,138]. Beriberi can cause irreversible neuromuscular disorders as well as defects in memory. Preoperative deficiency of thiamine has been estimated at <1% to 49% depending on time frame and type of weight-loss surgery [135].

Vitamin D is absorbed in the ileum and jejunum, suggesting that deficiency of this nutrient would not be severe following RYGB. However, studies of vitamin D deficiency before and after bariatric surgery suggest that suboptimal levels of vitamin D are quite common preoperatively, making supplementation an issue. In a 2007 series of 95 patients, 54% were vitamin D deficient (<50 nmol/L) and another 34% had suboptimal levels (50–79 nmol/L) [139]. In another study, 80% of preoperative patients had 25-OH vitamin D (the storage form of the vitamin) levels less than 32 ng/mL [140]. The ASMBS estimates that as many as 90% of obese patients may have low levels of vitamin D preoperatively [135]. Although supplementation has been shown to increase levels following surgery, a pilot study involving 45 post-RYGB patients suggests that, for many patients, current levels of supplementation may not be high enough to normalize levels [141]. A 2017 study indicated that 96.2% of pre-procedure patients were deficient in vitamin D and, after four years, 86% still had a deficiency [64]. In addition, the ASMBS estimates that up to 100% of post-bariatric surgery patients have a vitamin D deficiency [135].

Calcium is primarily absorbed in the duodenum and proximal jejunum. Low calcium intake and low levels of vitamin D can both contribute to deficiency in whole-body calcium, leading to increased bone resorption and potentially osteoporosis [135]. One study, a prospective design with one year of follow-up in a small group of patients, found a strong association between declining bone mineral density at the hip and degree of weight loss after RYGB [142]. Intake of both calcium and vitamin D increased after surgery, but most patients continued to have levels of vitamin D less than 30 ng/mL. Deficits in calcium and vitamin D, with associated increases in bone resorption, may also occur after LAGB [135]. The long-term significance of bone density changes is unknown, however. For calcium supplementation, calcium citrate, which does not require high acidity for absorption, may be a better choice than calcium carbonate, particularly in RYGB and LSG patients and others with reduced gastric acid.

After RYGB, changes in the physical structure of the GI tract can influence the absorption of certain medications. Extended-release formulations that are designed to remain in the intestine for long periods may not be absorbed as well or according to the expected time course [134]. Immediate-release formulations are generally recommended in these patients; however, healthcare providers are not always aware of recommended vitamin regimens, dosages, and appropriate formulations. A retrospective study conducted from 2006 through 2007 in patients with a history of bariatric surgery examined vitamin/nutrient supplements and medication dosage formulations given upon admission. Daily multivitamin, calcium, iron, vitamin B12, and folic acid supplementation were evaluated. Of 133 patient admissions, 88% had a history of a malabsorptive procedure. Approximately 33% of patients were given a multivitamin; 5.1% were given supplemental vitamin B12; 7.7% received supplemental calcium; 11.1% received additional folic acid; and 12% received iron. Inappropriate formulations (e.g., non-immediate-release, enteric-coated) were ordered in 61.5% of patients. Fifty percent of patients were discharged with inappropriate formulations [146].

Although medication absorption in bariatric surgery patients is not well studied, the reduction in acid due to structural changes in the stomach may alter absorption of medications that require an acidic environment. More pharmacokinetic clinical studies are needed to address the specific effects of various bariatric procedures on drug absorption [147].

However, some patients will be able to lose weight and keep it off through increased physical activity and healthier eating. The U.S. Preventive Services Task Force (USPSTF) recommends intensive counseling and behavioral interventions to promote sustained weight loss for obese adults [149]. Counseling may address diet, exercise, or both, and behavioral interventions may be aimed at skill development, motivation, and support. The USPSTF concluded that these strategies could result in a typical weight loss of about 3–5 kg (6.6–11 lbs), lasting one year or more. Some patients will struggle to lose this amount of weight, while others will be able to lose substantially more.

Some patients will benefit from pharmacotherapy to aid in weight loss. Approval criteria for anti-obesity drugs include a 5% or more mean placebo-subtracted weight loss after one year of therapy or a minimum of 35% of participants achieving more than 5% weight loss. The European Medicines Agency guideline requirements are similar. Both agencies also call for evidence of improvements in metabolic comorbidities [152]. At present, five weight-loss drugs are FDA-approved for long-term use: orlistat, phentermine/topiramate, bupropion/naltrexone, semaglutide, and liraglutide [152,153,154]. Weight loss achieved through the use of medication tends to be modest, and weight is often regained when the drugs are stopped [155,156,157].

Orlistat inhibits nutrient absorption. Orlistat has been shown to increase weight loss and improve cardiovascular risk factors. Primary side effects are gastrointestinal discomfort and a decrease in absorption of fat-soluble vitamins [152]. Independent reports of liver injuries (including six cases of liver failure between 1999 and 2008) prompted the FDA to approve a label revision for orlistat that includes a warning of possible severe liver injury [152]. However, the risk of severe liver injury is low, and this risk should be weighed against potential benefits [153]. Orlistat is indicated for the treatment of obesity in conjunction with a reduced-calorie diet [152].