The popularity of bariatric surgery has grown in recent years because of exponential increases in the rate of obesity in the United States and the documented safety of bariatric procedures. A body mass index (BMI) between 25 and 29.9 kg/m2 is defined as overweight. Obesity is defined as a BMI ≥ 30 kg/m2. Severe obesity can be defined as either (a) a BMI ≥ 40 kg/m2 or (b) a BMI ≥ 35 kg/m2 with at least one comorbid condition (National Institutes of Health [NIH], 1998). The Centers for Disease Control and Prevention (CDC, 2010) estimate that more than one third of older adults are obese. This figure has increased over the last 15 years especially in adults between 65 and 74 years old (CDC, 2012). Estimates from the Federal Interagency Forum on Aging-Related Statistics (2012) suggest that 44% of individuals between the ages of 65 and 74 and 29% of individuals over 75 are obese. Lang, Llewellyn, Alexander, and Melzer (2008) and Newman (2009) suggest that older adults with excess body weight may have difficulty losing weight. Providers may recommend bariatric surgery for obese older adults when diet and exercise do not yield sufficient weight loss. A recent study found a 60% increase in the number of older adults undergoing bariatric surgery from 2005 to 2009 nationwide (Dorman et al., 2011). However, individual studies examining outcomes of bariatric surgery in older adults tend to have small sample sizes, therefore minimizing the generalizability of their findings. Thus, the purpose of this review is to establish an understanding of the safety and efficacy of bariatric surgery in older adults. This is important information for nurse practitioners (NPs) and other providers caring for these patients.
Background and significance
Many predisposing factors contribute to obesity including hormone-mediated metabolic rate, genetic predisposition, socioeconomic conditions that limit food choice to calorie dense foods, and cultural beliefs of acceptable body habitus (CDC, 2010). Excess body weight increases the risk for type 2 diabetes mellitus (DM), cardiovascular disease, premature death, certain types of cancer, osteoarthritis, and obstructive sleep apnea (Guh et al., 2009; U.S. Department of Health and Human Services, 2001; Villareal, Apovian, Kushner, & Klein, 2005). Osteoarthritis is also common in overweight individuals because of the increased pressure on weight-bearing joints (Newman, 2009).
Obesity has a significant financial burden on the individual and society through direct and indirect medical costs. Indeed, it is estimated that in 2008 obesity cost the United States a staggering $147 billion, which is approximately 21% of the nation's healthcare spending (Cawley & Meyerhoefer, 2012). Direct medical costs included diagnosis and treatment of comorbid conditions and surgeries related to obesity. Indirect costs were associated with work absenteeism and restricted activities (CDC, 2010).
In 2006, the Centers for Medicare & Medicaid Services (CMS, 2009) announced that Medicare beneficiaries who have a BMI ≥ 35 kg/m2 with at least one comorbid condition or those with a BMI ≥ 40 kg/m2 meet the criteria for bariatric surgery coverage. This national coverage determination (NCD) stated that covered procedures include Roux-en-Y gastric bypass (RYGB), laparoscopic adjustable gastric banding (LAGB), and biliopancreatic diversion with duodenal switch (BPD/DS). Sleeve gastrectomy (SG) was not included as a covered procedure. At the time reimbursement was limited to bariatric surgery Centers of Excellence (COE) as designated by either the American College of Surgeons (ACS) or the American Society for Metabolic and Bariatric Surgery (ASMBS; CMS, 2009). COE were defined as institutions that performed a high volume of bariatric surgeries per year (>125 cases) and who abided to strict regulations to maintain this status. However, in 2013 CMS issued an updated decision that eliminated the COE requirement for reimbursement and indicates that local Medicare Administrative Contractors may approve procedures not covered under CMS’ decision (CMS, 2013). The updated NCD has sparked controversy regarding the safety of patients undergoing bariatric surgery at non-COE as these institutions may see an increase in Medicare patients, many of whom are over 65 years old (Dimick, Nicholas, Ryan, Thumma, & Birkmeyer, 2014).
Types of bariatric surgery
Most guidelines highlight the importance of medical and lifestyle interventions prior to bariatric surgery (ACS, 2014; ASMBS, 2013). Bariatric surgery is considered after interventions such as diet, exercise, and medication management do not produce adequate weight loss. Several procedures, which vary in invasiveness and recovery time, have been developed since the emergence of bariatric surgery in the 1950s. However, ASMBS (2013) stresses the importance of interdisciplinary teams that in addition to surgeons and licensed independent providers provide care to bariatric patients, specially trained registered nurses, dietitians, mental health professionals, and physical therapists.
Common procedures include RYGB, LAGB, SG, and BPD/DS (ASMBS, 2014). During RYGB a pouch is created in the upper portion of the stomach. A small outlet is attached to the pouch, which connects to the small intestine (ASMBS, 2014). Weight loss from RYGB occurs as a result of food being rerouted, thus altering the mechanism of digestion and absorption. This affects insulin-regulating hormones in the stomach and intestines (ASMBS, 2014). LAGB is the least invasive and is thus considered the preferred procedure among older adults. The band is placed around the superior portion of the stomach restricting food intake. A band may be adjusted by accessing a subcutaneous port. Saline can be injected or removed from the port (ASMBS, 2014). SG is an irreversible procedure that divides the stomach vertically and reduces the organ to a quarter of its original size. The stomach retains its function, and digestion remains unaltered because the pyloric valve at the distal end is intact (ASMBS, 2014). Lastly, BPD/DS mimics a malabsorptive state. A portion of the stomach is removed in order to limit food capacity. A gastric pouch is created and allows ingestion of larger meals than RYGB (ASMBS, 2014).
The purpose of this review is to determine if bariatric surgery can be safely performed on adults over the age of 60, and if the results of such surgeries are significant enough for NPs and other healthcare providers to recommend them in this population.
The search strategy was to find the highest level of evidence related to bariatric surgery outcomes in adults over the age of 60. A Health Sciences Librarian was consulted to determine appropriate search strategies. The search and analysis of studies was performed in accordance with the PRISMA 2009 checklist and Whittemore and Knafl's (2005) method of integrative review. Databases searched included PubMed, the Cochrane Database of Systematic Reviews, the Cumulative Index of Nursing & Health Allied Literature (CINAHL), EMBASE, PsycInfo, and the Joanna Briggs Institute. Dissertation Abstracts was also searched to identify unpublished studies. Thirteen studies were identified using this method and an additional two studies were obtained through ancestry search of reference lists of articles retrieved. Key words used in the search included a combination of bariatric surgery, weight loss surgery, older adults, and elderly.
All English language studies published between 2002 and 2013 that compared outcomes of bariatric in adults over the age of 60 to those between the ages of 18 and 60 were included in this review. Studies were excluded if they only had samples composed of adults over 60; had any subjects under the age of 18; defined older adult as starting at any age other than 60. Authors identified a total of 15 studies that met the inclusion criteria. Figure 1 provides a summary of the search. Table S1 provides a detailed summary of the search results.
This review includes 15 studies with a total of 6588 adults over the age of 60. Weight loss, improvement of comorbidities, mortality, surgical complications, and length of stay were the main outcomes evaluated in the 15 studies included in this review. Table 1 details study characteristics, while Table S2 presents a summary of relevant findings.
Table 1. Study characteristics
|Sleeve gastrectomy (SG)|
|Leivonen et al. (2011)||Prospective||Finland||N = 12 (21.8%)||88%|
|Laparoscopic adjustable gastric banding (LAGB)|
|Alhamdani et al. (2012)||Retrospective||United Kingdom||N = 39 (6.7%)||80%|
|Busetto et al. (2008)||Prospective||Italy||N = 216 (4.1%)||88%|
|Taylor and Layani (2006)||Prospective||Australia||N = 40 (4.4%)||73%|
|Dorman et al. (2011)||Retrospective||United States||N = 1994 (4.1%)||95%|
|Flum et al. (2005)||Retrospective||United States||N = 1517 (9.4%)||95%|
|Livingston and Langert (2006)||Retrospective||United States||N = 196 (0.7%)||90%|
|Sugerman et al. (2004)||Prospective||United States||N = 80 (2.7%)||85%|
|Varela et al. (2006)||Retrospective||United States||N = 1339 (2.7%)||88%|
|Laparoscopic Roux-en-Y gastric bypass (LRYGB) or open RYGB|
|Dunkle-Blatter et al. (2007)||Retrospective||United States||N = 61 (5.7%)||83%|
|Hallowell et al. (2007)||Retrospective||United States||N = 46 (4.9%)||85%|
|Sosa et al. (2004)||Prospective||United States||N = 23 (4.1%)||78%|
|St. Peter et al. (2005)||Prospective||United States||N = 20 (15.3%)||68%|
|Tiwari et al. (2011)||Retrospective||United States||N = 905 (2.4%)||90%|
|Willkomm et al. (2010)||Prospective||United States||N = 100 (6.8%)||90%|
Once the final sample of studies was selected for this review each author utilized the Crowe Critical Appraisal Tool (CCAT) to appraise each study. The CCAT has been demonstrated to be a reliable tool for assessing the quality of research papers (Crowe, Sheppard, & Campbell, 2011). The CCAT appraisal score can be reported as a raw score with a maximum score of 40 or a percentage. Scores on the CCAT were compared between authors to determine agreement of study quality. Two of three authors needed to agree on the quality rating of each study prior to finalizing a score. Any scores that differed by more than 10% were reevaluated by all authors for consistency. Table 1 includes the CCAT percentage score for each study.
The CCAT scores for the 15 studies in this review ranged from 68% to 95%. This indicates a generally moderate to high level of quality throughout. In general high scoring studies displayed strong study designs, standardized data collection methods, addressed study limitations as well as implications for future research, and presented detailed data on patient demographics. Strengths and limitations identified are detailed in the discussion section.
It is important to consider the demographics characteristics of older adults undergoing bariatric surgery in order to determine which individuals may benefit most from these procedures. The average BMI of participants in these 15 studies ranged from 42.2 to 50.4 kg/m2. The mean age ranged from 61.6 to 68. The oldest participant was 83 years old in an Italian study (Busetto et al., 2008). Nine (69.2%) studies reported higher rates of comorbidities and disease severity among older adults at baseline prior to surgical intervention (Busetto et al., 2008; Dorman et al., 2011; Dunkle-Blatter et al., 2007; Hallowell et al., 2007; Leivonen, Juuti, Jaser, & Mustonen, 2011; Sugerman et al., 2004; St. Peter, Craft, Tiede, & Swain, 2005; Varela, Wilson, & Nguyen, 2006; Willkomm, Fisher, Barnes, Kennedy, & Kuhn, 2010). Participants were disproportionately female. The majority of studies included information on gender with the exception three (Sosa, Pombo, Pallavicini, & Ruiz-Rodriguez, 2004; St. Peter et al., 2005; Willkomm et al., 2010). The range of female participants in the remaining 12 studies was from 67.2% to 87% suggesting that female patients represent a larger percentage of older adults undergoing bariatric surgery. Only four (30.8%) of the studies in this sample report on race/ethnicity, which indicates that data on race/ethnicity may not be routinely collected by bariatric surgery programs (Dorman et al., 2011; Sugerman et al., 2004; Tiwari et al., 2011; Varela et al., 2006). In addition, the majority of participants in these four studies identified as white and comprised as much as 73.1%–85% of their samples. This suggests that disparities in access to bariatric surgery may exist among older adults of racial/ethnic minority groups.
Remarkably, only eight (53.3%) studies reported on weight loss of subjects who underwent bariatric surgery (Alhamdani et al., 2012; Busetto et al., 2008; Dunkle-Blatter et al., 2007; Leivonen et al., 2011; Sosa et al., 2004; St. Peter et al., 2005; Sugerman et al., 2004; Taylor & Layani, 2006). Different measures were used to quantify the weight loss experienced by participants, including percent of excess weight loss (%EWL), change in BMI, and weight loss in kilograms. %EWL and BMI were both evaluated in three studies (Leivonen et al., 2011; Sosa et al., 2004; Sugerman et al., 2004). The remaining five studies evaluated either %EWL or BMI.
Although weight loss was generally lower in older adults across all studies, no statistically significant differences in %EWL or BMI were found in three studies (Alhamdani et al., 2012; Dunkle-Blatter et al., 2007; Leivonen et al., 2011). Only Busetto et al. (2008) found statistically significant differences in weight loss with younger patients experiencing greater weight loss. Both St. Peter et al. (2005) and Sugerman et al. (2004) reported mixed results. St. Peter (2005) found no significant difference in weight loss between age groups, but significant differences were noted when comparing BMI levels. Sugerman et al. (2004) reported that although %EWL was significantly lower in older adults at 1 year following surgery, this difference disappeared at 5 years follow-up. Sosa et al. (2004) and Taylor and Layani (2006) did not provide the statistical significance of their findings. Despite these differences, researchers agreed that any weight loss sustained by older subjects improved their overall health, whether or not weight loss was equal to that of younger participants.
Improvement of comorbidities
The potential for bariatric surgery to improve or resolve comorbidities in obese adults is often touted as one of its advantages. In this sample, seven (46.7%) studies evaluated the impact of bariatric surgery on comorbid conditions of which six studies evaluated the impact of bariatric surgery on DM. The rate of improvement of DM ranged from 61% to 100% (Busetto et al., 2008; Dunkle-Blatter et al., 2007; Leivonen et al., 2011; Sosa et al., 2004; Sugerman et al., 2004; Taylor & Layani, 2006.) Hypertension (HTN) also improved after older adults underwent bariatric surgery. Improvements in HTN were reported in six studies in this review (Busetto et al., 2008; Dunkle-Blatter et al., 2007; Leivonen et al., 2011; Sosa et al., 2004; Sugerman et al., 2004; Taylor & Layani, 2006). HTN rates improved between 50% and 91%. Only Busetto et al. (2008) found a significantly greater improvement in HTN among older adults compared to younger patients. Dyslipidemia and hyperlipidemia (HLD) were evaluated in four studies (Busetto et al., 2008; Leivonen et al., 2011; Sosa et al., 2004; Taylor & Layani, 2006). The rate of improvement of dyslipidemia or HLD was lower than that of DM and HTN at 38%–60%. Busetto et al. (2008) found no significant decreases in dyslipidemia in older adults after bariatric surgery.
Moreover, the results are contradictory when comparing improvement of comorbidities between older adults and adults under 60. Leivonen et al. (2011) found no difference in improvement of DM, HTN, or HLD between the two age groups. On the contrary, Sugerman et al. (2004) found a statistically significant greater improvement of both DM and HTN among younger patients. Although St. Peter et al. (2005) indicate that older adults had a greater reduction in medication use compared to younger patients, this was not statistically significant.
Similar to the studies evaluating weight loss, different measures were used to quantify improvement of comorbidities. Most studies utilized a combination of laboratory values to assess improvement of comorbidities (Busetto et al., 2008; Dunkle-Blatter et al., 2007; Leivonent et al., 2011; Sugerman et al., 2004; St. Peter et al., 2005). However, Taylor and Layani (2006) relied on decreased medications use to evaluate improvement of comorbid conditions while Sosa et al. (2004) did not disclose the method used to measure this outcome.
Estimating the mortality rate of patients over 60 undergoing bariatric surgery is important when considering whether these procedures should be recommended in this population. As such, mortality was evaluated in 13 (86.7%) studies in this sample. The findings on mortality are contradictory. There were no reported deaths in either age group in five studies (Alhamdani et al., 2012; Hallowell et al., 2007; Sugerman et al., 2004; Taylor & Layani, 2006; Willkomm et al., 2010). Three studies found higher, but not statistically significant mortality rates, among older adults (Busetto et al., 2008; Dorman et al., 2011; Dunkle-Blatter et al., 2007). On the other hand, five studies reported higher mortality rates among older adults at statistically significant rates (Flum et al., 2005; Livingston & Langert, 2006; Sosa et al., 2004; Tiwari et al., 2011; Varela et al., 2006). However, with a lower than expected mortality, Varela et al. (2006) assert that bariatric surgery in older adults is as safe as other gastrointestinal surgeries.
Furthermore, surgical complications were evaluated in 13 (86.7%) studies. The most common surgical complications reported included bleeding, hematoma, wound infections, pneumonia, pulmonary embolism, gastrointestinal complications, vitamin deficiencies, cardiac arrhythmias, chest pain, urinary retention, and hospital readmissions. Approximately 46.1% of the studies that evaluated surgical complications found no difference in complications by age group (Busetto et al., 2008; Dunkle-Blatter et al., 2007; Hallowell et al., 2007; Sosa et al., 2004; St. Peter et al., 2005; Taylor & Layani, 2006; Willkomm et al., 2010). Of the studies that evaluated surgical complications, both Alhamdani et al. (2012) and Dorman et al. (2011) reported lower rates among older adults. Dorman et al. (2011) found the lowest complication rates among individuals greater than 70 years old. Only four studies reported greater complications among older adults. Leivonen et al. (2011) reported significantly greater early complications in older adults but found that late complications were more common in younger patients. Livingston and Langert (2006), Tiwari et al. (2011), and Varela et al. (2006) found greater complication rates among older adults. Readmission rates were evaluated by Tiwari et al. (2011), Varela et al. (2006), and Willkomm et al. (2010) with only Varela et al. (2006) reporting significantly higher readmission rates among older adults.
Length of stay
Some studies reported postoperative length of stay following bariatric surgery. Of the studies included in this sample, nine (60%) evaluated length of stay of older adults compared to younger patients. The results of this analysis are contradictory with five studies reporting statistically significant higher lengths of stay among older adults (Dorman et al., 2011; Leivonen et al., 2011; Tiwari et al., 2011; Varela et al., 2006; Willkomm et al., 2010) and four studies reporting no difference (Alhamdani et al., 2012; Hallowell et al., 2007; Sosa et al., 2004; St. Peter et al., 2005).
Although operation time was another outcome discussed in the literature, only four (26.7%) studies evaluated this outcome. In general the operative time for older adults was similar to that of younger patients (Alhamdani et al., 2012; Leivonen et al., 2011; Willkomm et al., 2010). Only Hallowell et al. (2007) reported a difference with older adults experiencing an average operative time of 17 min less than younger patients (Hallowell et al., 2007).
The studies in this review suggest that bariatric surgery can be safely performed in older adults. Only three (20%) studies advise against bariatric surgery in older adults (Flum et al., 2005; Livingston & Langert, 2006; Tiwari et al., 2011). It is important to note that the data presented by Flum et al. (2005) and Livingston and Langert (2006) were collected prior to the rise in popularity of laparoscopic procedures, particularly LAGB as well as the development of reimbursement codes for these procedures. Tiwari et al. (2011) focused exclusively on laparoscopic RYGB. However, most evidence suggests that LAGB is the preferred procedure for older adults (Alhamdani et al., 2012; Busetto et al., 2008; Dorman et al., 2011).
Most studies provided detailed information about surgical technique, preoperative and postoperative management of bariatric surgery patients (Alhamdani et al., 2012; Busetto et al., 2008; Dunkle-Blatter et al., 2007; Hallowell et al., 2007; Leivonen et al. 2011; Sosa et al., 2004; Sugerman et al., 2004; Taylor & Layani, 2006; Willkomm et al., 2010). This standardized management of bariatric surgery patients is perhaps a response to the changes that occurred following CMS’ 2006 NCD and the development of COE. For the most part studies also reported data on mean BMI, gender, and age of participants.
Despite these strengths, the 15 studies included in this review had disparate numbers of older adults with a range of 12–1994. Most studies had small sample sizes with older adults representing only between 2.4% and 21.3% of all patients in individual studies. Generalization of the findings of this review to racial/ethnic minority groups is limited. Only four studies disclosed information regarding race/ethnicity. This is an important area of research to explore as Tiwari et al. (2011) found significant disparities in outcomes particularly among blacks. Bariatric surgery programs should be encouraged to maintain robust databases that collect information on age, gender, race/ethnicity, and insurance status. Increased reporting of these factors may lead to an improved understanding of how bariatric surgery outcomes differ among diverse populations of older adults.
Another limitation is that different procedures were evaluated in the literature under the broad term of bariatric surgery. Indeed not all surgeries evaluated in this review have the same postoperative course. For instance, Dorman et al. (2011) report a significantly greater risk of mortality among older adults undergoing RYGB versus LAGB. Similarly Varela et al. (2006) reveal that in their study older adults had a greater rate of complications following RYGB compared to LAGB and gastroplasty. The number of participants lost to follow-up was greater than 30% in three studies and may create bias in the results reported (Alhamdani et al., 2012; Sugerman et al., 2004; Willkomm et al., 2010). It is interesting to note that Sugerman et al. (2004) and Willkomm et al. (2010) revealed higher follow-up rates among older adults. Only one study reported outcomes of SG in older adults compared to younger patients (Leivonen et al., 2011). The findings indicate that, apart from an increased length of stay, outcomes do not differ greatly between the two groups. Bayham, Greenway, and Bellanger (2012) and Schirmer (2011) assert that SG in older adults is associated with increased weight loss compared to LAGB and fewer complications than RYGB. Outcomes of SG among older adults must be explored in the future.
Moreover, another area that needs further study is determining the adequate length of follow-up. Courcoulas et al. (2013) identify a lack of studies evaluating long-term outcomes of bariatric surgery. Most studies of bariatric surgery patients follow patients for less than 2 years. There is a need for studies that evaluate the long-term outcomes of bariatric surgery because most studies follow patients for 2 years or less. Of the 15 studies in this review, only five studies (30%) reported on findings of subjects followed for more than 2 years (Alhamdani et al., 2012; Busetto et al., 2008; Flum et al., 2005; Sugerman et al., 2004; Taylor & Layani, 2006). There are data that suggest that outcomes vary significantly several years after bariatric surgery. Sugerman et al. (2004) found that significant differences in weight loss were observed between the two age groups at 1 year but vanished after 5 years follow-up. This highlights the importance of long-term follow-up.
The evidence presented in these 15 studies suggests that bariatric surgery is safe and effective in adults over 60 years old. NPs and other providers can use these data to help inform their practice and whether they would recommend bariatric surgery to older adult patients. Future studies should place an emphasis on inclusion of diverse subjects, disclosure of subject demographics, long-term follow-up of bariatric surgery patients, and strategies to retain subjects. The use of qualitative studies that address the patient experience would add much needed depth to understanding the experience of older adult bariatric surgery patients as well as factors that contribute to patient adherence and continued follow-up.
A majority of the studies in this review support the use of bariatric surgery among individuals over the age of 60. As with any surgical procedure in older adults, the risks need to be carefully considered. The role of interdisciplinary teams in managing bariatric surgery patients will become more important as the number of older adults undergoing these procedures is expected to increase. This review provides NPs and other providers with a better understanding of the current state of the science regarding bariatric surgery in older adults.
We would like to thank Dr. Leslie-Faith Morritt Taub, PhD, ANP-C, GNP-BC, CDE, CBSM, FAANP, for her editorial assistance and guidance in preparing this manuscript.