Cardiovascular benefits of bariatric surgery in morbidly obese patients

Authors

  • V. G. Athyros,

    1. Second Propedeutic Department of Internal Medicine, Aristotelian University, Hippocration Hospital
    Search for more papers by this author
  • K. Tziomalos,

    1. First Propedeutic Department of Internal Medicine, Aristotelian University, AHEPA Hospital, Thessaloniki, Greece
    Search for more papers by this author
  • A. Karagiannis,

    1. Second Propedeutic Department of Internal Medicine, Aristotelian University, Hippocration Hospital
    Search for more papers by this author
  • D. P. Mikhailidis

    Corresponding author
    1. Department of Clinical Biochemistry (Vascular Disease Prevention Clinic), Royal Free campus, University College London Medical School, University College London (UCL), London, UK
      DP Mikhailidis, Department of Clinical Biochemistry (Vascular Disease Prevention Clinic), Royal Free Hospital, University College London Medical School, Pond Street, London NW3 2QG, UK. E-mail: mikhailidis@aol.com
    Search for more papers by this author

DP Mikhailidis, Department of Clinical Biochemistry (Vascular Disease Prevention Clinic), Royal Free Hospital, University College London Medical School, Pond Street, London NW3 2QG, UK. E-mail: mikhailidis@aol.com

Summary

Morbid obesity is associated with increased morbidity and represents a major healthcare problem with increasing incidence worldwide. Bariatric surgery is considered an effective option for the management of morbid obesity. We searched MEDLINE, Current Contents and the Cochrane Library for papers published on bariatric surgery in English from 1 January 1990 to 20 July 2010. We also manually checked the references of retrieved articles for any pertinent material. Bariatric surgery results in resolution of major comorbidities including type 2 diabetes mellitus, hypertension, dyslipidemia, metabolic syndrome, non-alcoholic fatty liver disease, nephropathy, left ventricular hypertrophy and obstructive sleep apnea in the majority of morbidly obese patients. Through these effects and possibly other independent mechanisms bariatric surgery appears to reduce cardiovascular morbidity and mortality. Laparoscopic Roux-en-Y gastric bypass (LRYGB) appears to be more effective than laparoscopic adjustable gastric banding (LAGB) in terms of weight loss and resolution of comorbidities. Operation-associated mortality rates after bariatric surgery are low and LAGB is safer than LRYGB. In morbidly obese patients bariatric surgery is safe and appears to reduce cardiovascular morbidity and mortality.

Introduction

The prevalence of obesity in the USA has increased by 33% during the last decade (1). Obesity is a leading cause of preventable death, related to more than 400 000 deaths per year (1,2). About 5% of the US population are considered to be morbidly obese (body mass index [BMI] > 40 or BMI > 35 kg m−2 in the presence of significant comorbidities) (3). Therefore, morbid obesity is a major health problem in developed societies. Morbid obesity is closely related to type 2 diabetes mellitus (T2DM), dyslipidemia, hypertension, obstructive sleep apnea (OSA), cardiovascular disease (CVD) and several forms of cancer (4–6). Morbid obesity is responsible for >2.5 million deaths per year worldwide (6) and it has been estimated that a 25-year-old morbidly obese man will lose 12 years of his life because of this condition (7). Morbid obesity is largely refractory to diet and drug therapy but generally responds well to bariatric surgery (3,4,7–11). Laparoscopic adjustable gastric banding (LAGB) and laparoscopic Roux-en-Y gastric bypass (LRYGB) are the two more commonly performed bariatric procedures (9,11).

We summarize the effects of bariatric surgery on comorbidities and CVD mortality as well as the mechanisms underlying these effects (Table 1). The safety and efficacy of LAGB and LRYGB are also compared.

Table 1.  Beneficial cardiovascular effects of bariatric surgery
ComorbidityEffect of bariatric surgery
HypertensionLowering of systolic and diastolic blood pressure
Resolution of hypertension
Type 2 diabetes mellitusReduction in blood glucose and HbA1c levels
Reduction in insulin resistance
Prevention of progression of impaired glucose tolerance to type 2 diabetes mellitus
Resolution of type 2 diabetes mellitus
Reduction in mortality because of type 2 diabetes mellitus
DyslipidemiaLowering of serum low density cholesterol and triglyceride levels
Increase of serum high density lipoprotein cholesterol levels
Resolution of dyslipidemia
HyperuricemiaResolution of hyperuricemia
Metabolic syndromeResolution of metabolic syndrome
Non-alcoholic fatty liver diseaseImprovement in liver steatosis, inflammation and fibrosis
Resolution of non-alcoholic fatty liver disease
Chronic kidney diseaseDecrease in albuminuria and glomerular hyperfiltration
Left ventricular hypertrophyReduction in left ventricular mass index
Obstructive sleep apneaResolution of obstructive sleep apnea
Coronary heart diseaseReduction in mortality because of coronary heart disease

Search method

For this review we searched MEDLINE, Current Contents and the Cochrane Library for papers published on bariatric surgery in English from 1 January 1990 to 20 July 2010, using the following keywords: ‘bariatric surgery’, ‘morbid obesity’, ‘cardiovascular disease’, ‘morbidity’, ‘mortality’, ‘cardiovascular risk’, ‘metabolic effects’, ‘diabetes’, ‘hypertension’, ‘dyslipidemia’, ‘chronic kidney disease’, ‘metabolic syndrome’, ‘obstructive sleep apnea’ and ‘left ventricular hypertrophy’. We also manually checked the references of retrieved articles for any pertinent material.

Safety and efficacy of bariatric surgery

A recent study reported the outcomes of 14 716 patients undergoing gastric bypass surgery in Pennsylvania hospitals between 1999 and 2003 (3). Bariatric surgery had substantial beneficial effects on obesity-related comorbidities. In addition, the inhospital mortality was 0.37% and the 30-day mortality was 1.15%. Controlling for other factors, patients treated by high-volume (procedures/year) surgeons and respective hospitals had lower 30-day mortality rates than low- and medium-volume surgeons and hospitals. Length of hospitalization was also shorter at high-volume hospitals. These results suggest that mortality is decreased when bariatric surgery is performed by surgeons at hospitals that perform more than 100 procedures per year (3).

Regarding long-term mortality after bariatric surgery, a Swedish study retrospectively analysed 12 379 patients (9614 women) with mean age 39.5 ± 10.4 years who underwent 14 768 weight-lowering procedures (12). During a mean follow-up of 10.9 ± 6.3 years, 751 (6.1%) patients died and the cumulative 30-day, 90-day and 1-year mortality rates were 0.2, 0.3 and 0.5%, respectively. Early cumulative mortality was higher in men and in patients older than 50 years of age. Long-term mortality was also higher in men (90 vs. 50/10 000 person years when excluding early deaths; mortality risk ratio [RR] 1.8, 95% confidence interval [CI] 1.5–2.1). There was no difference in early mortality rates when primary procedures were compared with re-operations. Myocardial infarction (MI) and malignancies were the most common causes of late death. These data suggest that anti-obesity surgery can be performed safely in unselected populations of morbidly obese patients with low rates of early and long-term mortality (12).

Similar early mortality rates were observed in a meta-analysis of 136 studies including 16 944 patients (mean BMI 47 kg m−2) treated with bariatric surgery (11). Overall, patients lost 61.2% of pre-operative excess weight (47.5 and 61.6% when AGB and RYGB were performed, respectively). This contributed to the resolution of T2DM, dyslipidemia, hypertension and OSA in the majority of patients. Resolution rates of TD2M and dyslipidemia where higher when RYGB was performed compared with AGB. The 30-day mortality rates ranged 0.1–1.1% depending on the procedure (11).

A retrospective cohort US study assessed the long-term mortality (from 1984 to 2002) of 9949 patients who underwent gastric bypass surgery and 9628 severely obese persons who did not (13). From these subjects, 7925 surgical patients and 7925 severely obese control subjects were matched for age, sex and BMI. During a mean follow-up of 7.1 years, adjusted long-term mortality from any cause in the surgery group decreased by 40% compared with the control group (38 vs. 57 deaths/10 000 person-years, P < 0.001). Cause-specific mortality from coronary heart disease (CHD), T2DM and cancer was also lower in the surgery group by 56% (2.6 vs. 5.9/10 000 person-years, P = 0.006), 92% (0.4 vs. 3.4/10 000 person-years, P = 0.005), and 60% (5.5 vs. 13.3/10 000 person-years, P < 0.001), respectively. Interestingly, mortality from cancer was reduced more in men than in women (by 85 vs. 55%, respectively) (13). On the other hand, rates of death not caused by disease (including suicide, accidents not related to drugs, poisonings of undetermined intent and other deaths) were 58% higher in the surgery group (11 vs. 6/10 000 person-years, P = 0.04) (13). This increase appeared to be limited to female patients (13). Mood disorders, which are frequently present in morbidly obese patients and might not resolve after bariatric surgery, might partly explain this increase of 5/10 000 person-years (13). It is also possible that the major change in body image after bariatric surgery and its multiple implications for the social life of patients might aggravate pre-existing mood disorders or induce new-onset psychiatric disease in some patients. On the other hand, suicide rates per se were low and did not differ significantly between the surgery and control group; therefore, another possibility is that the increase in death might represent a chance finding.

The prospective Swedish Obese Subjects (SOS) study compared weight loss and long-term all cause mortality in 2010 morbidly obese subjects who underwent bariatric surgery (surgery group; 376 underwent nonadjustable or adjustable banding, 1369 underwent vertical banded gastroplasty and 265 underwent gastric bypass) and 2037 subjects of similar BMI who received conventional treatment (control group) (14). During an average follow-up of 10.9 years, 129 deaths occurred in the control group and 101 deaths in the surgery group. The unadjusted hazard ratio (HR) in the surgery group was 0.76 (P = 0.04) compared with the control group and the HR adjusted for sex, age and CVD risk factors was 0.71 (P = 0.01). The most common causes of death were MI (13 vs. 25 deaths) and cancer (29 vs. 47 deaths) (14). The reduction in CVD mortality was primarily explained by a reduction in patients with T2DM whereas the mortality did not differ between the control and surgery group in subjects with normal glucose tolerance (14). On the other hand, and in contrast with the findings of the gastric bypass US study mentioned above (13), the reduction in cancer in the SOS trial was significant in women but not in men (15).

A study from Canada included a treatment cohort of 1035 patients who underwent bariatric surgery in a single center between 1986 and 2002 (81.3% underwent RYGB and 18.7% underwent vertical banded gastroplasty) and a control group (n = 5746) of age- and gender-matched severely obese patients who had not undergone weight-reduction surgery (16). Subjects with medical conditions other than morbid obesity were excluded. Patients were followed for a maximum of 5 years. Bariatric surgery resulted in a 67% reduction in excess weight (P < 0.001) and also reduced significantly the risk for cardiovascular and neoplastic diseases. All cause mortality rates were 0.68 and 6.2% in the bariatric surgery cohort and in controls, respectively (RR 0.11, 95% CI 0.04–0.27) (16). Excess weight loss was significantly higher in patients who underwent RYGB but it was not reported whether morbidity and mortality rates differed according to the type of bariatric surgery (16).

Therefore, bariatric surgery in morbid obesity results in substantial and persistent weight loss, improvement or resolution of comorbidities and decreased all cause mortality compared with conventional (non-surgical) treatment.

Comparative safety and efficacy of laparoscopic adjustable gastric banding and laparoscopic Roux-en-Y gastric bypass

The LRYGB induces greater weight loss than LAGB and in a shorter period of time (9,11). Different effects on gut hormones of LRYGB and LAGB might contribute to the greater weight reduction after LRYGB (17). The peptide YY area under the curve (AUC) at week 52 was greater in patients who underwent LRYGB than in those who underwent LAGB (P < 0.01). Glucagon-like peptide-1 levels 30 min postprandially were threefold greater in the LRYGB group (P < 0.001) whereas ghrelin AUC increased after LAGB at week 52 (P < 0.05) but tended to fall after LRYGB. Glucose, insulin and leptin levels as well as insulin resistance assessed with the homeostasis model of assessment (HOMA-IR) decreased in both groups but were lower at week 52 in the LRYGB group (17).

In a single-centre study in 838 consecutive patients undergoing bariatric surgery (765 LRYGB and 73 LAGB), the reduction in total cholesterol (TC) levels at 12 months post-operatively was greater in the LRYGB group (9).

The LAGB is associated with lower perioperative risks than LRYGB (9,11). However, a recent study suggested an under-reporting of late mortality and re-operations after LAGB (18).

In an analysis of 14 comparative studies (1 randomized), excess weight loss at 1 year was greater when LRYGB was performed (median difference vs. LAGB 26%, range 19–34%, P < 0.001) (19). Resolution of comorbidities was also more frequent after LRYGB. In the highest-quality study, excess body-weight loss was 76% with LRYGB vs. 48% with LAGB, and T2DM resolved in 78% vs. 50% of patients, respectively. Both operating room time and length of hospitalization were shorter in patients undergoing LAGB. Operative mortality was <0.5% in both procedures. Perioperative complications were more common with LRYGB (9 vs. 5%) but long-term re-operation rates were lower after LRYGB (16 vs. 24%). Patient satisfaction rates favored LRYGB (P = 0.006) (19). The authors suggested that LRYGB should remain the primary bariatric procedure for morbid obesity in the USA (19). Other comparative studies reached similar conclusions (20).

Resolution of comorbidities

In an analysis of 5502 morbidly obese patients who underwent bariatric surgery (gastric bypass in 85.7% of the cases) and were followed-up for up to 3 years, significant decreases (P < 0.05) were observed compared with the presurgery period in total CVD (44 vs. 14%), T2DM (20 vs. 8%) (21). The frequency of medication use for CVD also decreased significantly (21). In the SOS trial, patients (mean age 48 years; mean BMI 41) were followed-up for at least 2 years (4047 subjects) or 10 years (1703 subjects) (22,23). The follow-up rate for laboratory examinations was 87% at 2 years and 74% at 10 years. After 2 years, weight had increased by 0.1% in the control group and decreased by 23% in the surgery group (P < 0.001). After 10 years, the weight had increased by 1.6% in the control group and decreased by 16% in the surgery group (P < 0.001). The 2- and 10-year rates of resolution of T2DM, high triglycerides (TGs), low levels of high-density lipoprotein cholesterol (HDL-C), hypertension and hyperuricemia were higher in the surgery group, whereas resolution of hypercholesterolemia did not differ between the groups (22). As mentioned above, most patients in the SOS trial were operated with a gastric banding. Among the 641 patients of the surgery group who were followed-up for 10 years, 156 had undergone nonadjustable or adjustable banding, 451 had undergone vertical banded gastroplasty and 34 had undergone gastric bypass (22). Reduction in weight, waist, diastolic blood pressure (BP), insulin, TG and TC levels was greater in the gastric bypass subgroup than in the banding subgroup (22). However, the small number of patients in the gastric bypass subgroup did not allow comparisons of resolution rates of comorbidities in patients undergoing different types of bariatric surgery (22) (Table 1).

a. Type 2 diabetes mellitus

In patients with morbid obesity and impaired glucose tolerance, most studies report a 99–100% prevention of progression to T2DM with bariatric surgery (24). In subjects with overt T2DM prior surgery, resolution of the disease is reported in 64–93% of cases, occurs as early as the third post-operative month and is sustained long term (11,24–27). Several studies showed that surgical intervention in patients with recently diagnosed T2DM have higher rates of resolution of T2DM than patients with longer duration of the disease (25). Having either T2DM for more than 10 years or more advanced forms of the disease (according to the type of anti-diabetic treatment) is associated with less weight loss after bariatric surgery, smaller improvements in glycemic control and less frequent resolution of T2DM. However, a substantial reduction in oral medications or insulin doses has been observed after bariatric surgery (25). Tight glycemic control reduces the risk for micro- and possibly macro-vascular complications of T2DM (28), especially when combined with BP and lipid control (29,30).

The LRYGB prevents diabetes in 99–100% of patients with impaired glucose tolerance and leads to clinical resolution of 80–90% of newly diagnosed cases of T2DM. Gastric banding procedures are associated with lower rates (50–60%) of clinical remission of T2DM (9,31).

In a large meta-analysis (32) that included 621 studies in 135 246 patients, resolution rates of T2DM (i.e. resolution of the clinical and laboratory manifestations of T2DM) were reported in 3188 patients. In addition, 19 studies in 11 175 patients reported both weight loss and T2DM resolution separately for the 4070 diabetic patients who were included in these 19 studies. Mean age of the diabetic patients was 40 years, BMI was 48 kg m−2 and 80% were female. Mean weight loss was 38 kg or 56% excess body-weight loss. Overall, 78% of diabetic patients showed complete resolution of T2DM whereas improvement or resolution was observed in 87% of patients. Resolution of T2DM was more frequent in patients undergoing gastric bypass than in those undergoing gastric banding. Fasting glucose, HbA1c and insulin levels declined significantly post-operatively. Weight and T2DM parameters showed little differences within the first 2 post-operative years and afterwards. Therefore, the clinical and laboratory manifestations of T2DM resolve or improve in the great majority of patients after bariatric surgery (32).

b. Hypertension

Hypertension is one of the commonest comorbidities associated with obesity (33). In a recent study, systolic BP significantly decreased after LRYGB from 140 ± 17 mmHg preoperatively to 120 ± 18 mmHg at 12 months (P < 0.01). The mean diastolic BP also significantly decreased from 80 ± 11 mmHg preoperatively to 71 ± 8 mmHg at 12 months (P < 0.01). At 12 months of follow-up, 44 (46%) patients achieved complete resolution of hypertension while 18 (19%) patients showed an improvement in BP. Patients with complete resolution had a shorter duration of hypertension compared with patients without resolution (53 vs. 95 months, respectively, P = 0.01) (33). Overall, weight loss after LRYGB substantially improves and/or resolves hypertension in a substantial number of patients (37–53%) or reduces the need for anti-hypertensive agents (18–36%) (33–36). LRYGB effectively reduces comorbidities including hypertension even in elderly patients (35). Although the percentage of excess body-weight loss is less in patients over 60 years of age and both T2DM and hypertension are more common in this age group, there was a trend towards greater improvement of these comorbidities after LRYGB compared with younger patients (35). On the other hand, in the SOS trial, where only 13.2% of the patients underwent gastric bypass, a rebound in hypertension was observed after 6–8 years of follow-up (22).

It should be mentioned that patients who develop vitamin D deficiency (a common nutritional deficiency after LRYGB) (36) have significantly lower rates of resolution of hypertension compared with those with adequate vitamin D levels (42% vs. 61%; P = 0.008) (37). In addition, vitamin D supplementation (50 000 IU weekly) corrects vitamin D depletion in most patients and frequently results in resolution of hypertension (75% vs. 32%; P = 0.029) (38). Recent data suggest that vitamin D might also reduce CVD risk through BP-independent mechanisms (39).

c. Dyslipidemia

Up to 50% of morbidly obese patients who undergo bariatric surgery have dyslipidemia (40). In turn, dyslipidemia plays a role in the excess morbidity and mortality of the morbidly obese (41). Several randomized trials have shown that lipid-lowering treatment with statins reduces CVD morbidity and mortality (42). Weight loss after gastric banding has also been shown to improve the lipid profile (41). In a study in patients undergoing LRYGB, a 66% excess body-weight loss at 12 months post-operatively was associated with a 31% decrease in low-density lipoprotein cholesterol levels, a 39% increase in HDL-C levels and a 63% decrease in TGs (41). Also, within 1 year, 82% of patients who were on lipid-lowering medications preoperatively did not require such medications (41). A cohort study that included 6235 patients who underwent bariatric surgery (mean age 44 years, 82% women, 34% with T2DM, 80% underwent gastric bypass) yielded similar results (43). Medication use declined significantly at 3 months post-operatively. At 12 months after surgery, medication use for T2DM, hypertension and dyslipidemia declined by 76, 51 and 59%, respectively (43).

These findings are in contrast with those of the SOS study (22,23). The SOS reported no differences in TC levels between control patients and patients who underwent gastric surgery for morbid obesity after 2- and 10-years of follow-up (22). Within the surgical group, patients who had gastric bypass showed greater reductions in TC levels compared with those who underwent gastric banding or vertical banded gastroplasty (22). There was a 13% reduction in TC levels observed at 10 years post-operatively in the patients who underwent gastric bypass (22). This difference in TC fall might be because of the greater weight loss in the LRYGB group (73%) compared with the banding group (28%) and vertical-banded-gastroplasty group (35%). In the entire study population, a 16% decrease in TGs and a 24% increase in HDL-C levels were observed. Therefore, some CVD risk benefit might also be expected from these changes in TG and HDL-C levels (44–47).

d. Metabolic syndrome

Metabolic syndrome (MetS) is an important public health problem given its increasing prevalence and its association with the development of T2DM and CVD (48,49). In a small study (n = 46), 61% met the definition of the MetS before gastric bypass (50). After 2 years, resolution of hypertension, dysglycemia, dyslipidemia and MetS were observed in 85, 94, 96 and 84% of patients (50). A larger study in 827 patients who underwent LRYGB yielded similar results (51).

In a comparative study, 180 morbidly obese patients underwent LRYGB and 157 patients followed a weight-reduction programme but did not undergo surgery (52). After bariatric surgery, the number of patients with MetS decreased by 58% (from 87 to 29%). In the non-surgical group, MetS prevalence decreased by 10% (from 85 to 75%) (52). The relative risk reduction for MetS was 0.59 with bariatric surgery (95% CI 0.48–0.67, P < 0.001). The number needed to treat with surgery to resolve one case of MetS was 2.1 (52). These results did not change when patients with T2DM or CVD were excluded from the analysis or when diagnostic criteria for MetS other than BMI were applied (52). The resolution of MetS after bariatric surgery appears to depend mostly on the extent of excess weight loss (52,53). Inflammatory markers such as high sensitivity C-reactive protein (hsCRP) and white blood count (WBC) might represent useful markers of CVD risk in patients with MetS (54–57). Both baseline WBC and hsCRP are elevated in morbidly obese patients (56,58,59). The weight reduction 1 year after bariatric surgery was associated with a marked reduction in hsCRP and WBC levels (58). The hsCRP and WBC levels were within the normal range in 94% and 98% of patients after surgery, respectively (58). Although patients who underwent LRYGB lost more weight than patients who underwent LAGB, resolution rates of elevated CRP and WBC 1 year after surgery were similar in the two types of surgery (58). Anti-obesity agents (orlistat, rimonabant and sibutramine) also reduce inflammatory markers in morbidly obese patients; however, safety considerations lead to the withdrawal from the markets of the latter two agents (60,61).

e. Non-alcoholic fatty liver disease

Non-alcoholic fatty liver disease (NAFLD) is present in >30% of morbidly obese patients (62). The NAFLD appears to be an independent CVD risk factor and also increases liver disease-related morbidity and mortality (63). In a study with a 28-year follow-up, patients with NAFLD had 69% higher all cause mortality rates than the general population and this excess risk was mainly because of CVD and to a lesser degree to liver diseases (63). Patients with nonalcoholic steatohepatitis (NASH) (i.e. a more advanced form of NAFLD) had an 86% increase in all cause mortality (63). Lipid-lowering agents appear to be useful in the treatment of NAFLD (64,65) whereas in morbidly obese patients bariatric surgery is another option (62). In a study in 70 patients who underwent laparoscopic bariatric surgery, mainly LRYGB, a 59% mean loss of excess body weight was observed 15 months post-operatively (62). In addition, there was a reduction in the prevalence of MetS from 70 to 14% (P < 0.001) and a marked improvement in liver steatosis (from 88 to 8%), inflammation (from 23 to 2%) and fibrosis (from 31 to 13%; P < 0.001 for all comparisons vs. baseline). Inflammation and fibrosis resolved in 37% and 20% of patients, respectively, and the grade and stage of liver disease improved by 82 and 39%, respectively (P < 0.001 for both comparisons vs. baseline) (62). In another study in 35 patients who had evidence of NASH in liver biopsy at the time of gastric bypass, significant improvements in steatosis, lobular inflammation, portal and lobular fibrosis were observed post-operatively (66). The NASH resolved in 89% of patients (66). Other studies that evaluated the effects of RYGB on NASH showed an improvement and complete remission of necroinflammation in approximately 90 and 70% of patients within 18 months or more (67–69). Thus, surgical weight loss, mainly using RYGB, significantly improves liver histology in morbidly obese patients at all stages of NAFLD, even the more advanced. Whether these beneficial changes will reduce CVD or liver disease-related events is currently unknown.

f. Chronic kidney disease

During the last years, obesity has been associated with increased risk for developing chronic kidney disease (CKD). Experimental and clinical data suggest that obesity may lead to glomerular hyperfiltration (70). Obesity-related dyslipidemia, hyperinsulinemia and increase in leptin and angiotensin levels may contribute to the development of hyperfiltration (71–74). Glomerular hyperfiltration increases the risk of microalbuminuria and proteinuria in obese patients without overt renal disease (12) or may accelerate the progression of pre-existing renal disease (71,75,76). Reducing glomerular hyperfiltration therefore could prevent or delay the incidence of obesity-associated renal damage (70). Weight loss with a hypocaloric diet and concurrent treatment with an angiotensin-converting enzyme inhibitor may reduce obesity-related proteinuria (77). However, gastric bypass produces greater weight loss and appears to be more effective in controlling not only hypertension, T2DM and dyslipidemia but also renal damage (70,78,79). In a study in 61 extremely obese patients who underwent gastric bypass, a significant decrease in albuminuria was observed during the 24th month of follow-up compared with the first year. The percentage of patients with glomerular hyperfiltration and albuminuria were significantly reduced at 24 months compared with baseline (70). Another study evaluated 140 patients submitted to LRYGB (32% men, mean BMI 46). Glomerular filtration rate before surgery was 149 mL min−1 and decreased to 114 mL min−1 (P < 0.0001) post-operatively (78). Similar results were reported in morbidly obese diabetic patients who underwent LRYGB (79). Thus, in morbidly obese patients, the substantial weight loss after bariatric surgery decreases glomerular hyperfiltration and albuminuria, which could slow the progression towards irreversible renal damage (70,78,79).

g. Left ventricular hypertrophy

Left ventricular hypertrophy (LVH) is independently associated with CVD events, particularly in hypertensive patients (80). The LVH is mostly a marker of target-organ damage but may also contribute directly to CVD through changes in cardiac structure (80). Obesity is associated with an increase in left ventricular mass (81). In a small study in 15 morbidly obese hypertensive women who underwent bariatric surgery (10 underwent LRYGB and 5 underwent LAGB), left ventricular mass index (LVMI) continued to decrease linearly during the 24-month follow-up period whereas weight loss, lean and fat mass loss, and resting energy expenditure reduction reached a plateau at 3 months (P < 0.001 for all) (81). A retrospective analysis of 57 morbidly obese patients who underwent gastric bypass showed a substantial decrease in left ventricular mass (LVM) and LVMI after 3.6 years of follow-up compared with matched controls who had not undergone bariatric surgery (82). In the study population as a whole, multivariate analysis showed a positive correlation between the change in body weight and ventricular septum thickness (r = 0.33), posterior wall thickness (r = 0.31) and LVM (r = 0.38; P < 0.05 for all correlations) (82). Another study evaluated 16 women (mean BMI 44 kg m−2; mean age 44 years) scheduled for bariatric surgery who underwent volumetric cardiovascular magnetic resonance imaging before, 3 and 17 months after surgery (83). Weight loss following bariatric surgery (LRYGB in 12 patients and LAGB in 4 patients) was associated with a reduction of left and right ventricular mass whereas left ventricular ejection fraction did not change. The decrease in LVM was linearly related to the reduction in BMI and independent of changes in BP (83). Interestingly, a pilot study showed that bariatric surgery induced beneficial changes in cardiac autonomic control, which might reduce the risk of life-threatening ventricular arrhythmia (84). In this study, gastric bypass and LAGB had similarly beneficial effects on cardiac autonomic control (84).

h. Obstructive sleep apnea

The OSA is an increasingly prevalent disease that appears to increase the risk for CVD (85). The OSA is characterized by successive episodes of obstruction or decrease in respiratory airflow (86). Obesity is a major risk factor for OSA and the prevalence of OSA in morbidly obese patients is approximately 70%. Treatment is based on the use of continuous positive airway pressure and weight loss in obese patients (86). Bariatric surgery has beneficial effects on the severity of OSA with most studies evaluating gastric bypass (87–89). In a meta-analysis of 136 studies in 22 094 morbidly obese patients treated with bariatric surgery, OSA resolution rates were analyzed in 1195 patients; OSA resolved after surgery in 86% of patients and resolved or improved in 95% (11). Among the 56 and 896 patients who underwent gastric bypass and gastric banding, respectively, resolution rates were 80.4 and 95.0%, respectively (11).

Mechanisms of the beneficial effects of bariatric surgery on cardiovascular disease risk

The major mechanism through which weight loss after bariatric surgery reduces CVD risk is the resolution of the obesity-related comorbidities analyzed above. However, it is also possible that the drastic weight loss might reduce CVD risk through other mechanisms.

A prospective study enrolled 50 consecutive patients with morbid obesity who underwent LRYGB (90). At 24 months, mean BMI decreased from 47 to 29 kg m−2 (P < 0.001), mean carotid intima-media thickness from 0.84 to 0.50 mm (P < 0.001) and mean flow-mediated dilation improved from 6 to 15% (P < 0.05). In addition, hsCRP levels decreased from 1.23 to 0.65 mg L−1 at 6 months (P < 0.001) and to 0.35 mg L−1 at 24 months (P < 0.001). Thus, LRYGB resulted in significant improvements in markers of inflammation, subclinical atherosclerosis and endothelial dysfunction (90). In another study, vertical banded gastroplasty resulted in an increase in serum nitric oxide concentration, which might also exert atheroprotective effects (91). In 36 morbidly obese patients, markers of thrombosis and fibrinolysis including tissue factor, tissue factor pathway inhibitor, plasminogen activator inhibitor-1 and prothrombin fragment 1.2 improved 2 years after bariatric surgery (92). Thrombin generation, a marker of the overall activity of the coagulation cascade, also decreased post-operatively (92). Bariatric surgery was also shown to have beneficial effects on the levels of several adipokines, including leptin, adiponectin, resistin and visfatin, and these changes may contribute to the reduction in CVD risk (57,93). Furthermore, it was reported that the substantial and sustained weight loss after gastric bypass and the resolution of comorbidities improve quality of life and self-reported functional status; this might also beneficially affect CVD risk (94).

Effects of bariatric surgery on estimated cardiovascular disease risk

In 109 consecutive morbidly obese patients who underwent LRYGB, risk of CHD was estimated using the Framingham risk score (FRS) (95). Estimated CHD risk declined from 11 to 5% in men and from 6 to 3% in women (P < 0.0001 for both changes) (95). In another study (96), the reduction in FRS was greater in the gastric bypass group (from 7.0 to 3.5%; P < 0.001) than in control patients who did not undergo surgery (from 7.1 to 6.5%; P = 0.13). The reduction in estimated CVD risk using the Prospective Cardiovascular Munster Heart Study risk engine was also greater in the gastric bypass group (from 4.1 to 2.0%; P < 0.001) than in the control group (from 4.4 to 3.8%; P = 0.08). In another study in 500 patients the effects of gastric bypass operation on FRS and actual CVD risk were compared (97). A 69% excess body-weight loss was observed after 1 year. The average 10-year absolute risk of CVD events decreased from 5.4% at baseline to 2.7% after 1 year (P = 0.001). The risk reduction was similar in diabetic and non-diabetic patients and in both genders. The actual rate of CHD events in the study cohort was 1% (5/500). At a 5-year projection, this rate was considerably lower (P = 0.001) than the predicted rate before gastric bypass (97). Therefore, estimated CVD risk reduction appears to accurately reflect the actual CVD risk reduction and could be used to evaluate the benefits of bariatric surgery.

Conclusions

Bariatric surgery is both safe and beneficial in morbidly obese patients. Perioperative mortality is low (0.35%) and tends to decline further, particularly in experienced centers (98). In addition, bariatric surgery appears to reduce CVD mortality and this is probably mostly because of the resolution of obesity-related comorbidities, such as T2DM, hypertension, dyslipidemia, MetS, CKD, LVH, NAFLD and OSA. Among the different operations, LRYGB appears to be the most beneficial and has an acceptable safety profile.

Conflict of Interest Statement

This review was written independently. The authors did not receive financial or professional help with the preparation of the manuscript. The authors have given talks, attended conferences and participated in advisory boards and trials sponsored by various pharmaceutical companies.

Ancillary