https://orcid.org/0000-0002-5123-3176
Abstract
Until recently, weight loss in older obese people was feared because of ensuing muscle loss and frailty. Facing overall increasing longevity, high rates of obesity in older individuals (age ≥ 65 years) and a growing recognition of the health and functional cost of the number of obesity years, abetted by evidence that intentional weight loss in older obese people is safe, this approach is gradually, but not unanimously, being replaced by more active principles. Lifestyle interventions that include reduced but sufficient energy intake, age-adequate protein and micronutrient intake, coupled with aerobic and resistance exercise tailored to personal limitations, can induce weight loss with improvement in frailty indices. Sustained weight loss at this age can prevent or ameliorate diabetes. More active steps are controversial. The use of weight loss medications, particularly glucagon-like peptide-1 analogs (liraglutide as the first example), provides an additional treatment tier. Its safety and cardiovascular health benefits have been convincingly shown in older obese patients with type 2 diabetes mellitus. In our opinion, this option should not be denied to obese individuals with prediabetes or other obesity-related comorbidities based on age. Finally, many reports now provide evidence that bariatric surgery can be safely performed in older people as the last treatment tier. Risk-benefit issues should be considered with extreme care and disclosed to candidates. The selection process requires good presurgical functional status, individualized consideration of the sequels of obesity, and reliance on centers that are highly experienced in the surgical procedure as well as short-term and long-term subsequent comprehensive care and support.
With the aging of the population and growing rate of obesity, the prevalence of obesity in older individuals is steadily on the rise (1). However, it is still mostly viewed as a risk factor for future cardiometabolic disease, not a disease on its own, and is usually studied in younger age groups. Definitions of older age vary across studies (2) and organizations (3) and range from 60 years upward. Most commonly the term “elderly” refers to individuals aged 65 years or older. This review addresses daily clinical challenges in older obese patients.
What Is the Definition of Obesity in the Older Population?
Obesity in the older population is usually defined, as in the general population, by a body mass index (BMI) greater than or equal to 30 because of the wide acceptance, simplicity, and relative reproducibility of these measures and despite multiple limitations (1, 4-6). BMI rises with age because of fat mass expansion and decline in spinal height (eg, vertebral compression causing loss of height). BMI-defined obesity misses individuals with “normal weight obesity,” mostly abdominal obesity, with diminished lean weight, who, at the older age, are particularly prone to frailty and disability (7).
Although waist circumference (WC) better correlates with comorbidities, its definition and actual measurement is more variable, and the normal range varies with ethnicity. Fatness (fat mass fraction of total weight) increases with age and muscle/bone mass declines (8-10), reflecting, in part, lower basal metabolic rate, physical activity, testosterone (T) and growth hormone production as well as the responsiveness to thyroid hormone and leptin (9, 11-13). Assessment of body fat requires equipment such as dual-energy x-ray absorptiometry (DEXA) or bioimpedance-based devices. Further, normal ranges for fat mass fraction vary with sex, age, and ethnicity and may overlap (14), so its application is complex.
The American Association of Clinical Endocrinologists and American College of Endocrinology suggested a new term and focus, “adiposity-based chronic disease” (ABCD). This diverts clinicians’ attention from fat excess per se to its health sequelae, mediated by the mass, distribution, and dysfunction of adipose tissue in the context of its comorbid diseases (15). Sarcopenic obesity comprises an important phenotype in older adults (16, 17). It is defined by the combination of obesity (according to BMI, WC, or % fat) with impaired muscle strength (< 27 kg and < 16 kg using a hand grip test in men and women, respectively), lower muscle mass (< 5.67 kg/m2 and < 7.23 kg/m2 of appendicular mass index using DEXA for women and men, respectively), and/or impaired physical function (10). Sarcopenia was recently officially coded in the International Statistical Classification of Diseases and Related Health Problems 10th Revision (18). Compared to obesity alone, obesity with sarcopenia was associated with higher odds for metabolic syndrome (19), decreased survival in patients with several cancer types (20), and osteopenia/osteoporosis (21).
Obesity is linked to a host of complications and comorbidities that include metabolic syndrome (43% of people aged ≥ 60 years) (22), type 2 diabetes mellitus (T2DM), dyslipidemia, heart failure, atherosclerotic cardiovascular disease, atrial fibrillation, stroke, cognitive decline, many types of cancer, nonalcoholic fatty liver disease, arthritis, thromboembolic events, pulmonary abnormalities, sleep apnea, urinary incontinence, decreased quality of life, frailty, impaired motility, and disability (9).
Despite the strong link between obesity and metabolic diseases, the “obesity paradox” suggests that obesity may confer some protective effects such as lowering the risk of mortality in older individuals, especially in heart failure or chronic obstructive pulmonary disease patients or in residents of nursing homes (12). The dissociation between obesity and mortality may be limited to grade I obesity (BMI 30.0-34.9) (23). Low body weight in older people per se is often linked to frailty, undernutrition, or hidden life-shortening disease, the roles of which in the obesity paradox is hard to discern despite attempted adjustments.
Factors That Contribute to the Pathogenesis of Obesity in the Older Population: Practical Aspects
Partly modifiable risk factors for obesity in older age include reduction in energy expenditure due to diminishing physical activity (24) and decline in resting metabolic rate, mostly secondary to declining muscle mass. Age-related muscle loss is accelerated by inflammation, chronic diseases, impaired mobility and nutrition, and declining anabolic hormones (eg, T, growth hormone) (25, 26). The decline in resting metabolic rate is estimated at 2% to 3% per year from age 20 (27).
Although monogenic obesity is expressed in early life, family history of early onset obesity in an older individual can provide clues to genetic background (28), for example, pro-opiomelanocortin and melanocortin receptor mutations. Full exome (next-generation sequencing) testing or obesity gene panels could provide access to developing gene-specific/related treatments (eg, melanocortin receptor mutations) (29). Integration of complex genetic information related to polygenic obesity (eg, contribution by genes such as FTO) (30, 31) into the overall care plan in older obese individuals, however, remains a future challenge.
Lastly, obesity is clearly associated with a lower socioeconomic status (32) and financial hardship (33). Nevertheless, socioeconomic-related factors do not necessarily predict successful weight loss in older patients (34).
Should Some Form of Biological Age Determination Replace Chronological Age in the Decision Tree?
Obesity has been traditionally viewed as an age accelerator (35-43). There is an ongoing effort to construct “objective” biological age scales based on widely accepted parameters (eg, biochemical profile, smoking, hypertension, obesity): metabolic age (35), vascular age (36, 37), cognitive age (38), and epigenetic age (39). Overall, obesity pushes most of these age classifiers upward (35, 39-41), thus biologically accelerating the actual chronological age. A recently developed formula for biological age assessment was applied to clinical and biochemical Third National Health and Nutrition Examination Survey (NHANES III) data and confirmed the layperson belief that “60 is the new 50” (44). Between 1988 and 2010, the mean biological age became lower for older adults, despite the concomitant rise in the prevalence of obesity. Obesity (41), BMI (39, 45, 46), lipid levels, inflammation (47), and nutrition (45) were shown to affect epigenetic age. It is therefore presently uncertain whether we can we push apparent age limits (eg, 70-75 years for bariatric surgery) upward based on individualized favorable assessment of an obese patient’s biological age.
Is Weight Loss per se at This Age Segment Still Safe and Desirable?
For intentional weight loss in older obese patients, it is helpful to conduct a comprehensive nutritional evaluation, covering sufficient intake of energy, protein, and micronutrients; measure body composition to exclude low muscle mass, determine fat mass, and measure fat distribution; and assess muscle function/strength to detect sarcopenia/sarcopenic obesity.
Some observational (but not interventional) studies in older adults have raised concern suggesting weight loss was associated with lower survival (48). We conducted a systematic search in the PubMed search engine using the Medical Subject Headings function in the 10 years preceding October 2020. This focused on individuals younger than 60 years in randomized control trials and meta-analyses of clinical trials targeting the effect of intentional weight loss with a follow-up of at least 3 months and is summarized as follows (Supplementary Tables 1-3) (49).
Weight Loss in Obese Older Adults and its Effect on Cardiovascular Disease, Diabetes, and Cardiometabolic Outcomes
Based on the 155 articles that met the clinical research search criteria for metabolic outcomes and the 47 for functional outcomes (see Supplementary Table 1) (49), it appears that older patients can benefit from intentional weight reduction achieved by diet alone, diet plus exercise or currently US Food and Drug Administration–approved drug therapy. Nevertheless, these studies were generally not designed to evaluate the effects of weight loss on mortality or cardiovascular disease (CVD), excess CVD, or diabetes incidence. Reported benefits of weight loss in older adults (age > 60 years) included obesity-associated metabolic and cardiovascular (CV) factors: Blood lipids, glucose, loss of visceral fat (50-56), blood pressure, cardiac structure, or atrial fibrillation (52-54, 56-58) were all improved by the various weight loss programs, even at older age. Long-term evidence for weight loss efficacy is exemplified by the Look AHEAD trial of older individuals with diabetes. Weight loss both at 1 and 4 years was associated with improved control of hypertension, T2DM, and dyslipidemia (59, 60). Interestingly, after 4 years, older adults (aged 65-76 years) experienced greater weight loss and greater reduction in glycated hemoglobin A1c and increment in high-density lipoprotein cholesterol than younger adults (aged 45-64 years) (55). CVD incidence was lower in patients allocated to liraglutide (61), renal function was preserved in high-risk individuals with diabetes and albuminuria (54), and the loss of muscle mass was compensated by clear evidence for functional improvement (62-64). However, weight loss in older individuals can be linked to decline in lean mass, bone mineral density (65, 66), and a slight increase in risk of fractures (57).
Weight Loss in Obese Older Adults and its Effect on Physical Function (Frailty, Sarcopenia, and Mobility) and on Quality-of-Life Outcomes
Overall, the systematic search presented in Supplementary Table 3 (49) supports the efficacy of weight loss in improving physical function as assessed by subjective and objective results in several mobility tests (62-64, 67-69). This is in spite of the reductions in lean mass that accompanies fat loss. This is a critical point, because muscle strength and function appear to better predict morbidity and mortality than muscle mass (10, 16, 17). Across different studies, weight loss was also associated with improvements in quality of life and physical function (67, 70, 71).
Overall, then, intentional weight loss in older adults with obesity is not only feasible but improves cardiometabolic, functional, and quality-of-life parameters. Furthermore, adherence to lifestyle interventions may actually be better at older ages (72). Evidence from the Weight Loss Maintenance trial suggested that older adults (aged > 60 years) had less weight regain, greater sustained weight loss, and a higher rate of clinically significant weight loss after 3 years compared to younger adults (73).
What Are the Treatment Goals of Obesity in Older Obese Individuals and Which Treatment Modalities Are Backed by Evidence?
There is a common belief that since no correlation exists between BMI and all-cause mortality beyond age 75 (13), obesity treatment should focus on younger elderly individuals. Weight loss might be still desirable, however, if it helps to preserve activity level, self-sufficiency (tying your own shoelaces), and self-esteem and lessen deterioration of existing chronic diseases and their complications.
The main strategies shown to improve comorbidities and CV risk factors in older obese patients are exercise and weight loss of at least 5% from baseline. The combination of weight loss and physical activity outperforms physical activity or diet alone. Caloric restriction in older people requires careful tailoring with attention to nutrients supply. To achieve a weight loss of 0.5 to 1 kg per week or 8% to 10% over 6 months, energy intake should be reduced by 500 to 1000 kcal/d (9, 74).
In 2016, obesity clinical practice guidelines issued by the American Association of Clinical Endocrinologists and the American College of Clinical Endocrinologists (75) clearly supported weight loss for older obese individuals at risk for complications such as diabetes and emphasized the role of lifestyle modification and the need to avoid sarcopenia. The guidelines were somewhat more qualified with respect to the use of weight loss drugs and bariatric surgery. A post hoc analysis of the Look AHEAD trial, which assessed the effects of a comprehensive intensive lifestyle intervention, suggested an association between the magnitude of weight loss and the incidence of CVD in people with T2DM up to age 76 (76).
Diet Therapy for Older Adults—Which Diet and How?
To avoid sarcopenia and bone loss, simple caloric restriction should not be prescribed for older patients (77, 78). Still, mild to moderate weight loss achieved by proper measures can improve or at least maintain physical function. For example, a weight loss of 3 to 4 kg over 1 to 3 years was shown to enhance physical function (79). In a 1-year diet and exercise intervention (mean age = 70 years), Villareal and colleagues reported a loss of 9% of body weight with improved measures of frailty, physical performance, and function (64). This study established that only the combination of exercise with diet achieved weight loss, relative preservation of lean body mass, and hip bone mineral density with improvement in frailty score. A 12-week intervention trial in obese older adults with a mean age of 65.5 years showed that combined exercise and caloric restriction resulted in preserved muscle mass and improvement in insulin resistance and cardiometabolic risk factors (80). In a longer trial in overweight and obese older adults (mean age = 67 years), combining physical activity (150 minutes/week) with diet-induced weight loss (10%) was superior to the physical activity alone or the diet alone for muscle composition and function (52). Improvements in diastolic blood pressure, glucose, and high-density lipoprotein cholesterol were associated with loss of fat mass whereas improvement in triglycerides, insulin, and the homeostatic model assessment of insulin resistance were linked to changes in fat mass and lean mass (52, 81).
In an attempt to counteract age-associated muscle loss due to alterations in metabolic regulation, immune status, and hormonal fluctuations, the Society for Sarcopenia, Cachexia and Wasting and the PROT-AGE Study Group proposed dietary supplementation with protein or amino acids, the presumed building blocks of muscle (82) in older people are as follows: a) 1.0 to 1.6 g of protein/kg/day spread equally throughout the day; b) a leucine-enriched balanced amino acid supplement, especially in older adults who exercise; and c) vitamin D should be supplemented if blood levels are below 100 nmol/L. A minimum of 60 to 90 minutes per week of resistance and aerobic activity was recommended to slow muscle loss.
During catabolic scenarios, essential amino acids can offset significant anabolic resistance in older adults and protect skeletal muscle (83-87). Leucine activates mammalian target of rapamycin, and hence, a leucin-rich diet may maximize the anabolic response and lessen the overall dietary protein requirement (88). This leaves room for sufficient carbohydrate and fat intake and still achieves the weight loss needed (5%-10%) for metabolic improvement and functional independence (89). There is a dose-dependent relationship between amino acid intake and muscle protein synthesis (86, 87, 90-93). Yet, despite increased protein intake (> 1.0 g/kg/day), 20% of the achieved weight loss in older obese adults represents a reduction in lean body mass (94). Another potential anabolic supplement in older adults with sarcopenic obesity is β-hydroxy-β-methylbutyrate (HMB), a metabolite of leucine produced in skeletal muscle; a recommended dose of HMB is 3 g/day (95). While a recent meta-analysis of 3 randomized controlled trials (n = 203) concluded that HMB supplementation improved (+1.2%) or preserved muscle mass and improved function frail/sarcopenic individuals aged 60 years and older (96), explicit recommendation to use HMB awaits further studies (95).
Circadian rhythm considerations, time-restricted feeding, and intermittent fasting for older adults: current state and future directions
The circadian rhythm is often disrupted with aging (97) and sleep disruption linked to poor health outcomes. Additionally, circadian clocks within key metabolic operators such as the liver and pancreas regulate the metabolic rhythms of glucose and lipid metabolism. Circadian misalignments have been linked to metabolic disorders such as obesity and T2DM (98). Late evening or night eating dissociates metabolic tissues’ rhythm from the central circadian rhythm. Time-restricted feeding, in which food is consumed during several hours of the day and usually disallowed in the evening and night, may counteract these trends and promote better synchronization between meals and the systemic circadian rhythm (99). Shifting the timing and nutritional content of main meals (morning > evening) comprises a potential strategy in the management of obesity (99) and T2DM (100). This rapidly unfolding and promising approach requires refinement and testing of feasibility vis-à-vis current human behavioral patterns and still lacks significant clinical evidence in obese older adults.
Recent evidence, mostly from animal research, suggests that in addition to time-restricted feeding, intermittent fasting may favorably affect multiple indices of health, intervene in disease progression, and lead to deceleration of aging and increased longevity (101, 102). Intermittent fasting induced improved glucose regulation and resilience to stress through improved capacity for DNA repair, mitochondrial biogenesis, autophagy, and downregulation of inflammation. These effects are apparently mediated through nutritional sensors and cues that are not necessarily attributed to weight loss. Prolonged fasting per se, for example, can reactivate impaired metabolic switching of fuel utilization from carbohydrates to fat (> 12 hours) (101), but data in older obese adults are scarce (103, 104). A recent meta-analysis of several studies with overweight or obese young older adults (upper age range ~ 70 years) found no differences in loss of lean mass due to intermittent fasting compared to continuous energy restriction (104). Finally, a very low-calorie diet coupled with mixed aerobic/resistance training resulted in an impressive 11% weight reduction with some functional improvement within 3 months in older individuals whose ages ranged from 65 to 85 years (105). The long-term impact of this protocol awaits assessment.
Exercise as a Treatment Tool for Weight Management in Older Adults—What Kind and How Much?
The American College of Sports Medicine and the American Health Association recommend that older adults engage in physical activity that includes aerobic, muscle-strengthening, flexibility, and balance exercise (106). They recommended 150 min/week of moderate-intensity exercise, performed “as conditions allow.” Caloric expenditure through exercise is limited in the unfit older population: just 4 kcal/min at 50% of peak oxygen uptake. To translate this into significant caloric deficit, for example, for 3500 kcal, 130 minutes of exercise per day will be needed (78). However, most studies of older individuals included only 30 to 90 minutes of moderate aerobic and/or resistance exercise training 3 to 5 days/week. Hence, although exercise per se showed beneficial effects on physical function, it induced no significant weight loss by itself (78).
Pharmacologic Approach to Obesity in Older Individuals—to Treat or not to Treat?
There are insufficient data to assess whether geriatric patients respond differently to weight loss medications compared with other patients. The treatment of older patients mandates special precautions, particularly in individuals who are already on multiple medications, which may increase the risk of drug interactions.
Of the 4 weight loss medications that are currently US Food and Drug Administration approved for the long-term (≥ 1 year) treatment of obesity (BMI ≥ 30, or BMI ≥ 27 in the presence of T2DM, hypertension, or dyslipidemia), orlistat (Xenical, Roche) (107), phentermine-topiramate (Qsymia, Currax pharmaceutical) (108), liraglutide (SAXENDA, Novonordisc) (109), and bupropion-naltrexone (Contrave, Takeda), the latter was not in tested in trials with older adults.
Xenical is a nonsystemic gastric and pancreatic lipase inhibitor that inhibits the absorption of dietary fat by approximately one-third and can reduce daily energy intake by 100 to 300 kcal. A long-term analysis (2-year randomized study) of an older subpopulation in a primary care setting showed the same effectiveness of orlistat in adults older or younger than 65 years (107), amounting to an approximately 10% weight reduction after 1 year (107, 110). Gastrointestinal side effects, mostly steatorrhea, were also not different among older and younger patients (107, 110) and occurred mostly on high dietary fat intake. This is particularly unacceptable in older individuals with background fecal incontinence. Additionally, orlistat may hinder the absorption of fat-soluble vitamins, especially vitamin D, which may require vitamin D monitoring or supplementation with fat-soluble formulas (2 hours before orlistat).
Phentermine-topiramate (Qsymia) combines phentermine, a centrally acting sympathomimetic appetite suppressant, with topiramate, an antiepileptic agent with incompletely understood anorexigenic properties, but there is insufficient evidence to support its use in older obese patients (111, 112).
The GLP-1 analogs liraglutide and semaglutide can induce weight loss in older individuals with diabetes up to age 80 (61, 113). Despite the impressive weight loss effect of semaglutide, it has not yet been approved for the treatment of obesity. Liraglutide (1.8 mg/day) outperformed other antiobesity drugs in that it reduced CV events and CV-related mortality. In a post hoc analysis of patients older than 75 years, there was a significant positive effect of liraglutide (compared to placebo) in the time to first occurrence of death from CV causes, nonfatal myocardial infarction, or nonfatal stroke. A beneficial effect was also seen in the expanded composite outcome and all-cause mortality (hazard ratio, 0.73; P = .013 vs placebo) (109, 114, 115). In the Scale trials (56 weeks, double-blinded trial), during which liraglutide was given at a dose of 3 mg to nondiabetic obese patients for weight management, approximately 7% of the liraglutide-treated patients were 65 years and older, and 0.5% were 75 years and older (116). There was no requirement for dose adjustment by age. Several reports showed no difference in efficacy or safety between older vs younger patients (117-121).
The post hoc analysis of patients from the SCALE Obesity and Prediabetes and SCALE Diabetes trials showed a similar mean and rate of 5% and 10% weight loss in patients younger and older than 65 years, with no differences in secondary outcomes including WC, blood pressure, heart rate, lipid and glycemic control parameters, and quality-of-life questionnaire between the 2 age groups (121). However, older participants had a higher rate of more serious gastrointestinal adverse events in both trials.
Perna et al (119) examined the effects of liraglutide on body composition in overweight and obese older patients with T2DM who were treated with 3 mg liraglutide for 3 months. DEXA showed reduced fat mass and android fat together with the preservation of muscular properties. In individuals who had suffered crush nerve injury, GLP-1 agonist treatment reduced functional disability, electrophysiological dysfunction, and atrophy of the tibialis anterior muscle (122). This supports a potential unique role for GLP-1 analogs in protecting skeletal muscle.
A growing body of evidence suggests a link between obesity and cognitive decline operating through vascular dementia, Alzheimer disease, and Parkinson disease (123-126). In addition to the more widely recognized central activation of GLP-1 receptors to induce satiety (127), emerging evidence suggests GLP-1 receptors also contribute to the control of synaptic plasticity and signaling pathways associated with learning and memory (128-130). GLP-1 agonists prevented tau hyperphosphorylation associated with aging in diabetic rats (131) and improved neuropathological features and cognitive functions in models of Alzheimer disease (132-134). Collectively, the good safety and efficacy of GLP-1 agonists, their ability to preserve skeletal muscle, and putative neuroprotective effects make them rather attractive, but further human testing is needed.
Drug-Drug Interactions Related to Obesity Pharmacotherapies
There is an inherent higher risk for drug-drug interactions in older individuals (135). The use of weight loss drugs must be considered in the context of the older patients’ list of medications. Two key drug-specific examples are the following: a) the potential deleterious effects of orlistat on the absorption of other drugs or supplements, particularly fat-soluble compounds such as vitamin D or other common medications such as thyroxine or anticoagulants; and b) phentermine-topiramate must be administered with much care and attention in older obese patients treated for hypertension and/or heart disease (reviewed in [136]).
Surgical Management of Obesity at Older Age
Although bariatric (metabolic) surgery is considered the most effective means of combating morbid obesity, it may elicit considerable disruption in everyday life for older individuals. Patient selection should consider preoperative functional independence and adaptability to changes in eating and bowel habits as well as the impact of even the temporary perioperative immobilization on overall health (137). Presently, no consensus exists regarding the safety, age cutoff, and preferred bariatric procedure for older adults. Published reports seem to reflect better results with bariatric surgery in older patients, with growing confidence in surgically treating individuals at continuously higher ages. However, attention to local expertise in this age segment is critical.
Older adults seem to have higher rates of surgical complications and longer hospital stay, with a statistically nonsignificant risk of postsurgery death (138, 139) in patients aged 70 years and older (138, 139). In a review of a US national database of nearly 1500 patients aged 70 years and older between 2005 and 2016, the overall rate of morbidity and mortality as well as rates of acute renal failure and myocardial infarction in the patients older than 70 years undergoing bariatric surgery was increased relative those younger than 70 years. The rates of adverse events were increased in age 70+ patients who underwent Roux-en-Y gastric bypass (RYGB) but not sleeve gastrectomy (SG), suggesting that SG may be the preferred procedure for older patients. Impaired preoperative functional status was linked to increased rates of morbidity and mortality (140). Some reports indicate that older patients lose less weight and benefit less from bariatric surgery than younger patients, possibly because impaired metabolic capacity and a greater presence of sarcopenia (141, 142). On the other hand, with the prevailing transition to laparoscopic procedures, better patient selection and preoperative optimization have contributed to decreased perioperative mortality and complications postsurgery also for the older patient (143). When traditional SG or RYGB surgery was performed in patients older than 65 years, the overall complication rate was low and SG had a favorable safety profile compared with RYGB, but the rate for RYGB was not significantly different between the older and younger groups (144). The International Sleeve Gastrectomy Expert Panel Consensus suggested laparoscopic SG as a safer procedure in the older population, but some studies showed that laparoscopic RYGB was equally safe compared to laparoscopic SG and resulted in better weight loss at 1 year (145-149). Concordantly, in a systematic review of the outcomes of the Ontario Bariatric Surgery Registry of more than 3000 patients, outcomes and complication rates of bariatric surgery in patients older than 60 years are comparable to those in a younger population, independent of the type of bariatric procedure performed (143). Another systematic review of 256 publications compared patients younger and older than 60 years and concluded that patients should not be denied bariatric surgery because of age alone (150).
A number of studies specifically evaluated obese patients older than 70 years, and some even older than 75 years, and also reported no significant difference in postoperative complications (137, 147, 148, 151-155). A significant limitation of most of these reports is the small number of patients. After RGYB and SG, weight loss was somewhat lower in the older than 70 years group compared with the younger group, reminiscent of previous studies showing lower weight loss in older adults (156-159).
Another key point resides in the duration of the follow up. Fortunately, several long-term studies have been conducted with older obese patients who underwent bariatric surgery. One report with a 3-year follow-up that included 451 older obese adults (mean age = 67.92 years, mean BMI = 40.32; 76% of patients had comorbidities) found that only approximately 9% of the patients had perioperative complications. About 1% needed reoperations, approximately 3% patients with operative complications were treated conservatively, 2% were readmitted, and another 3% had exacerbation of chronic lung disease. One year after surgery, the average remission of diseases was approximately 35%, improvement was 50%, and no changes in comorbidities were seen in approximately 16%. All patients lost more than 25% of their excess body weight after the first postoperative year, but older patients lost less weight: Patients aged 70 to 74 years lost about 40%, whereas patients older than 80 years lost about 20% of excess body weight. Importantly, no deaths were reported (160). Another study, with a median follow-up of more than 4 years, showed that bariatric surgery was safe in older patients with effective long-term control of obesity, diabetes, and with improved overall survival (161). A retrospective study of 46 patients aged 60 years and older (mean age = 64 years; mean BMI 49.6) who underwent conventional gastric bypass technique (laparotomy) noted more minor and major complications in older patients (> 65 years), including 2 cases of death in the older but not in the younger group (161).
In summary, age per se is not necessarily a contraindication for bariatric surgery. It can improve the quality of life, but we must recognize that the complication rate may be higher and the benefits less than expected for the young obese. While evidence to that effect is accumulating, much still remains unknown. In the absence of clear recommendations by professional societies, individualized decisions are unavoidable and patients’ awareness of current knowledge gaps and participation in the decision process are mandatory.
Should Common Endocrinopathies Be Treated in Older Adults?
A recent meta-analysis of 68 studies of 19 996 patients with obesity found a considerable prevalence of endocrine disorders in these patients (162): overt and subclinical hypothyroidism: approximately 14% each; hypercortisolism: approximately 1%, and low total or free T: approximately 43% and approximately 33%, respectively. These already considerable rates are known to rise with age, certainly as regards diagnoses based on serum thyrotropin (TSH) (163) and male T levels (164).
Mild Increase in Thyrotropin
As a rule, the active pursuit of normalization of mildly elevated TSH levels in patients age 75 years or older is not recommended (165). Several studies have suggested that in nonhypothyroid patients, obesity, particularly morbid obesity, per se is linked to higher TSH levels than seen in normal-weight individuals (166, 167). In patients already treated for hypothyroidism, overtreatment as reflected by low TSH appears to decrease with aging, so that more levothyroxine users 65 years and older may have normal TSH than those younger than 65 years. Obese individuals who receive levothyroxine show higher degrees of adiposity (168). A meta-analysis of 24 studies showed that bariatric surgery is followed by a significant decrease in TSH, free 3,5,3′-triiodothyronine, and 3,5,3′-triiodothyronine levels, with no effect on free thyroxine or 3,3,5′-triiodothyronine (169). Finally, there is no clear evidence that treating mild “subclinical hypothyroidism” in older obese individuals achieves any clear health benefit. However, there is room for individualization of decisions. The issue of treating “fit” vs “frail” older patients, yet another aspect of functional “biological” age, may be a factor because frail individuals are more susceptible to side effects (170, 171) and frailty is quite common in older obese patients (172). In a prospective follow-up of subclinical hypothyroidism in older patients, there was no association with increased risk of dementia (173).
Hypogonadism
There is a complex interaction between obesity, age, and gonadal status in men. Serum T declines with age, particularly after age 40 years (174-177). Independently, serum T is lower in obese men (176, 178, 179), often in association with increased circulating estradiol (derived mostly from fat cell aromatase). This inhibits a secondary rise in luteinizing hormone secretion in response to the low serum T (180, 181). Sperm count and quality are also lower in obese men (182, 183). Aromatase inhibitors can partially restore T and increase sperm count in obese men (180, 184). Serum T rises in obese men undergoing bariatric surgery or losing weight through diet, lifestyle modification, and even with GLP-1 analog (liraglutide) treatment (185-188). Weight loss, diet, and physical activity each plays a separate role in this setting (189). Around 10% of weight loss is required to achieve a significant increase in circulating T in obese men, which can improve clinical features related to androgen deficiency (190). An increase in T was observed in obese men placed on a Mediterranean diet with high protein content (191, 192). T is even lower in obese men with metabolic syndrome and/or T2DM (193, 194). These associations are bidirectional: T-replacement therapy can improve glycemic indices, decrease visceral fat, and induce fat loss in patients with and without T2DM (195). In a rather extreme example, 260 obese men with metabolic syndrome and low T received T treatment over 5 years, resulting in substantial weight loss along with significant metabolic improvement (196). No adverse CV events were addressed in this report. Finally, in a study of prediabetic men with a mean age at the onset of 59 years and a follow-up of 6.6 years, T significantly prevented T2DM (197).
In contrast to serum T per se, sex hormone–binding globulin (SHBG) increases with age, particularly after age 50 years, thus leaving a gradually smaller fraction of the total circulating T in a biologically available form (198). This relationship is further complicated because not only is SHBG lower in obese men and declines with weight gain (199), but low SHBG is an independent marker of the subsequent development of T2DM in obesity (200), whereas high SHBG predicts CV events in men older than 65 years (201).
Low T or free T have been reported as predictors of CVD in some (202-205) but not other (206, 207) studies in the general older population. There is also evidence for an association between T and the rate of CV events and mortality in older obese men. In the course of 10 years, in patients with a mean age of 77 years, men with higher T, SHBG, and physical activity had the lowest BMI and WC. Higher T was associated with the lowest risk of incident CVD events, irrespective of physical activity level (208). In Chinese men with T2DM, up to age 90 years, baseline low T was associated with an increased propensity to nonprostate cancer and decline in renal function during follow-up (209). The role of lower biological age and better general health, lifestyle, and body composition in these associations remains to be elucidated.
Considerable disagreement remains as to the potential role and therapeutic effects of T-replacement therapy in aging men with low T. The Endocrine Society’s position paper offers a conservative approach that is based on strict criteria of classical clinical and hormonal hypogonadism (210). The European Menopause and Andropause Society’s position statement offers a different approach, not only by recapitulating the concept of late-onset hypogonadism, but by suggesting a laxer definition of low T (eg, beginning with < 15 nmol/L for decreased libido vs 6.9 nmol by the Endocrine Society) and addressing aging-related potential benefits from T treatment that are not related to “textbook hypogonadism,” including favorable effects on obesity, metabolic syndrome, T2DM, and osteoporosis (211). Despite evidence that low T is associated with increased CV morbidity, T replacement in older men remains controversial. One randomized clinical trial in frail older men showed excess in CV events in men receiving T treatment (212). Several meta-analyses yielded more reassuring outcomes (213-215), but in one, T treatment in patients older than 65 was associated with a 2.9-fold increase in major CV events (216). Hence, the safety concern of chronic T-replacement therapy in older obese men is presently unresolved and remains subject to individualized decisions, with careful monitoring of on-treatment serum T and hematocrit. Finally, T-replacement therapy is contraindicated in many men with prostate cancer and should be offered only to select hypogonadal patients who have a history of definitively treated prostate cancer, as concluded by a recent meta-analysis (217).
Complicated Obesity in an Older, Morbidly Obese Man
A 78-year-old man was admitted to the hospital because of congestive heart failure in late December 2017, and again in early February 2018. At that time his BMI was 42.1. Prior to admission, he was treated with oral furosemide 120 mg daily, intravenous (IV) furosemide, 240 mg once to twice a week, and metolazone 2.5 mg daily. For several decades he was burdened with class III obesity, T2DM, and resistant hypertension. These had been complicated in recent years by advancing renal failure, cardiac cirrhosis, ischemic heart disease requiring stenting, bilateral significant carotid stenosis (90% on the left, 70% on the right), paroxysmal atrial fibrillation, sleep apnea, pulmonary hypertension, and class III diastolic dysfunction. During the 2 preceding years he moved around within his kibbutz yard (eg, home to work, a distance of 300 meters) using an electric utility vehicle. Prior to admission, there was a gradual worsening of heart failure and hypertension, which prompted referral to a heart failure treatment center, where the use of diuretic agents was gradually intensified. Expectedly, serum creatinine gradually rose to 2.6 to 3 mg%. He also had a worsening of obesity-hypoventilation syndrome, with daytime hypopnea and sleepiness (218).
He was admitted with breathlessness, grade 4, bilateral, lower-leg edema, and a weight of 108 kg. At that time, he was unable get up or walk more than 2 meters. Having received intensive daily IV diuresis, his weight dropped to 96 kg. On discharge, he was instructed to adhere to a daily dose of furosemide 240 mg, which was further increased by a weekly furosemide “booster” dose of 240 mg. Within 2 more weeks his weight dropped to 94 kg, but serum creatinine increased to 4.0 mg%.
First postdischarge follow-up: clinical dilemmas:
1) Is obesity still a factor that requires treatment?
Although many of the clinical features and multiorgan damage appeared irreversible, it was felt that functional alleviation might be attainable with weight loss to reduce sleep apnea (219) and pulmonary hypertension (220). Carrying less weight might also improve activities of daily living (getting up, walking, dressing, tying shoelaces without/with less exertion) (221); and, if associated with physical activity, quality of life (222). Intensive diuresis was very effective, but clearly linked to rapid deterioration in renal function, then with an estimated glomerular filtration rate of 15.5 cc/min.
2) Should caloric restriction be applied, at this age, in the presence of cirrhosis (compensated), renal failure, diastolic heart failure, daytime hypoventilation, and T2DM requiring bedtime insulin treatment?
The nutritional requirements for weight loss in older patients include higher protein intake to avoid sarcopenia, protein restriction to slow down the progression of renal failure, sufficient caloric intake to support the extra effort needed for ventilation, but caloric restriction, particularly moderation of carbohydrate intake, to improve glycemic control and achieve some weight loss. Salt restriction and adjusting potassium intake were also necessary. While some of these apparently contrasting needs could possibly be reconciled, the patient reported that he already strictly followed the advice of previous dieticians, apparently to no avail. A body composition study using DEXA showed very good muscle mass, likely related to hard physical labor for several decades in the patient’s earlier life. Still, muscle mass and body fluids are measured as a single (fat-free) compartment by this technique, so muscle mass could be overestimated in the presence of fluid retention. Bariatric surgery was not considered because of his overall condition.
3) Deciphering the major immediate health threats in a very high-risk 78-year-old patient with the listed comorbidities.
At this phase, rapid and permanent (acute on chronic) renal deterioration leading to reaccumulation of fluids removed by intensive diuresis was perceived as the most immediate threat. Was there, possibly, an alternative way?
4) Fasting-induced diuresis and intermittent fasting
The phenomenon of fasting-induced diuresis has been recognized for several decades, but remains incompletely understood (223, 224). In brief, fasting is followed by several days of increased natriuresis, which can be attenuated/reversed by carbohydrate administration. While this has been studied mostly in the context of obesity, in which the drastic caloric reduction is often followed by initial rapid weight loss related to initial salt and water excretion, fasting diuresis was successfully applied more than 60 years ago in nonobese patients with congestive heart failure. Bloom and Mitchell noted that 5 to 7 days of complete fasting was associated with diuresis, natriuresis, and weight loss and appeared an effective mode of treatment of the failing heart in “fully digitalized,” otherwise treatment-refractory, individuals (225, 226). One of Bloom’s originally described fasting patients with heart failure lost 7.3 kg in 5 days, and the average daily weight loss of his 10 original patients was 0.85 kg to 1.5 kg per day, way above what might have been expected by the negative caloric balance on fat and lean tissue loss. To our knowledge, there are no data on the effects of intermittent fasting in severely complicated obesity, which is presently under study at our center. Although multiple benefits have been ascribed to intermittent fasting or a fasting-mimicking diet, including weight loss in obesity, increased longevity, autophagy, and mitochondrial biogenesis (101), most were based on experimental data. A controlled trial of intermittent complete fasting or a fasting-mimicking diet in obesity did not show a weight loss advantage over the intermittent protocols (227, 228). However, this was not assessed in the context of heart/renal failure, where fasting diuresis might be critical to reduce the need of pharmacological diuresis.
Our patient was then placed on an intermittent fasting protocol, in which complete caloric fasting with free access to water and calorie-free fluids is exercised for 36 hours twice a week. Fasting began after an 8 pm regular dinner (eg, Saturday night), skipping all meals over the next day, and resuming regular food consumption with an 8 am breakfast the following day (eg, Monday morning). This protocol was repeated twice a week, every week, with adjustment of drugs and glucose-lowering treatment, including insulin, during the fasting days. Serum electrolytes, magnesium, and renal function were monitored. Within several weeks the furosemide dose could be lowered from 240 mg to 80 mg daily, the weekly IV treatment with furosemide was discontinued, as was the use of metolazone. Peripheral edema nearly completely resolved. His serum creatinine varied between 1.6 and 2.1 mg%, and the lowest recorded serum creatinine in recent weeks was 1.6 mg%, equivalent to levels seen nearly 4 years earlier. The patient lost an additional 8 kg within the next 6 months, returned to part-time work in his kibbutz, resumed walking, and indeed undertook a daily 3-km walk, under supervision, with several 5- to 10-minute breaks. He refused to discontinue the intermittent fasting protocol and did well for 18 months. Within this time, he uneventfully underwent a partial colectomy for an 8-cm sessile villous adenoma. Several months later he began to gradually reaccumulate fluids. He died suddenly 2 years after the initial admission, approaching age 80.
5) Take-home message
Late complications of obesity are often underappreciated as such, and attention is focused on the most immediate threat, such as heart failure or renal failure requiring dialysis. Even facing what appeared to be a dead end of irreversible multiple severe cardiac, renal, pulmonary, hepatic, and vascular complications of obesity, weight management in an apparently terminally ill morbidly obese diabetic individual, here in an unorthodox manner, was valuable in temporarily (1.5 years) allowing resumption of active life. In this case, intermittent caloric restriction (fasting) was applied first, with improvement in fluid overload. Some physical activity, despite severe limitations, could be added later.
Conclusion
Until recent years the concept of intentional weight loss in obese patients aged 65 years or older faced significant resistance, mainly because of concerns focused on muscle loss. Sufficient evidence has now accumulated to justify a more active approach, as summarized in the algorithm in Fig. 1. A combination of caloric restriction with an emphasis on adequate supply of energy, protein, and micronutrients along with programmed physical activity tailored to personal limitations comprises the first strategy to be considered. There is good evidence that drug therapy, particularly liraglutide, can be a legitimate second-tier strategy with concrete health benefits. Lastly, bariatric surgery is now less feared and can be safely applied in select patients treated at truly experienced centers.
A proposed algorithm for the treatment of obesity in older individuals.
Abbreviations
- BMI
body mass index
- CV
cardiovascular
- CVD
cardiovascular disease
- DEXA
dual-energy x-ray absorptiometry
- GLP-1
glucagon-like peptide-1
- HMB
β-hydroxy-β-methylbutyrate
- IV
intravenous
- RYGB
Roux-en-Y gastric bypass
- SG
sleeve gastrectomy
- T
testosterone
- SHBG
sex hormone–binding globulin
- T2DM
type 2 diabetes mellitus
- TSH
thyrotropin
- WC
waist circumference.
Acknowledgments
In memory of Amichai Resnik, a pioneer in the Negev desert, whose brave and unwinnable battle with the devastating late complications of obesity inspired this article.
Author Contributions: All authors wrote and edited the manuscript and provided final approval of the version to be published. All people designated as authors qualify for authorship, and all those who qualify for authorship are listed.
Additional Information
Disclosures: The authors have nothing to disclose.
Data Availability
Not applicable.
