Association of Bariatric Surgery With Risk of Fracture in Patients With Severe Obesity

Key Points Question Does severe obesity confer a risk of fracture that could be modified with bariatric surgery? Findings In this cohort study of 49 113 bariatric surgery–eligible patients covered by Medicare, patients who were eligible for bariatric surgery but did not undergo surgery had an equal risk of fracture compared with patients undergoing Roux-en-Y gastric bypass but had a greater risk of all site-specific fractures and fractures overall compared with those who underwent sleeve gastrectomy. Among 32 742 patients undergoing bariatric surgery, patients who underwent sleeve gastrectomy were found to be at a significantly reduced risk of developing a humeral fracture or fracture in general compared with those undergoing Roux-en-Y gastric bypass. Meaning Bariatric surgery may be associated with lower odds of fracture in patients with severe obesity, and sleeve gastrectomy might be the best option for bariatric surgery in patients in which fractures could be a concern.


Introduction
Bariatric surgery has increasingly become common as obesity has become a widespread concern in much of the high-income world. [1][2][3][4][5] These interventions have been shown to be associated with lasting and substantial weight loss, correction and protection from obesity-related conditions, and substantial benefits in quality of life and longevity. 2,3,6-11 Among obesity-related conditions, bariatric surgery has been demonstrated to reduce the burden of metabolic and cardiovascular diseases, migraines, and obesity-related risk of some cancers. 8,[12][13][14][15][16][17][18] There is a large body of literature reporting an observational association between higher body mass index (BMI, calculated as weight in kilograms divided by height in meters squared) and higher bone mineral density (BMD) that has implied that high BMI has protective effects on the skeleton and has led to the inference that loss of excessive body weight may result in decreases in BMD. 11,[19][20][21][22] To this end, bariatric surgery might therefore result in a decreased BMD and serve as a contributor to potentially higher rates and risks of fracture.
Furthermore, surgical weight loss approaches that alter the fundamental patterns of alimentary absorption, like Roux-en-Y gastric bypass (RYGB), may serve to hasten this risk and have been associated with the development of metabolic bone disease, resulting in higher bone turnover and long-term declines, disruptions, and deterioration in bone density and bone microarchitecture. [22][23][24][25][26][27][28][29][30] However, the notion that obesity protects from fracture has been challenged. 31 Recent studies have reported heightened rates of fracture among those with greater levels of central obesity and in postmenopausal women with obesity. 32,33 Those studies challenge the value of BMD as a surrogate measure to assess bone fragility in individuals with obesity and furthermore raise questions on the general notion that bariatric surgery is detrimental to the skeleton because it lowers BMD while pointing out the need of long-term studies to assess actual bone fragility after metabolic and weight management surgery. Fracture risk after bariatric surgery has been scarcely studied, and data are contradictory, reporting either no increased risk or limited increased risk. Those inconsistencies are likely due to the wide difference in matching parameters for controls, limited follow-up after surgery, a limited number of participants, and differences in surgical procedures. [23][24][25][26][27]34 Given the complex relationship between body composition, bone density, and bone fragility as well as the correlative nature of the studies that have established the prevailing notion that higher BMIs may be protective against osteopenia, osteoporosis, and, therefore, fracture, here we explored the absolute risk of fracture in patients with severe obesity who did not undergo bariatric surgery, those who underwent surgical interventions with both restrictive and malabsorptive features (RYGB), and those who underwent surgical interventions with less malabsorptive features (sleeve gastrectomy [SG]). or SG were plotted within 3 years following surgery. Log-rank testing was performed to compare fracture risk between patients undergoing RYGB and SG. Significance levels were adjusted using Bonferroni corrections, and significance was set at a 2-tailed P value less than .008. Data were analyzed using R statistical software version 3.42 (The R Foundation).  experienced significantly lower rates of fracture overall at 3 years following surgery (P < .001) (    were no significant differences in the risk associated with developing fractures of the radius or ulna, pelvis, hip, or vertebrae between patients who underwent RYGB and those who underwent SG (Figure 2). Log-rank P = .70 Log-rank P = .50

Discussion
In this matched cohort analysis of 49 113 bariatric surgery-eligible patients, including 16 371 who did not undergo bariatric surgery, 16 371 who underwent RYGB, and 16 371 who underwent SG, it was found that patients who were eligible for bariatric surgery but did not undergo surgery had a similar odds of fracture as patients undergoing RYGB at 3 years following surgery. Furthermore, patients who underwent SG had significantly lower odds of all site-specific fractures and fractures overall compared with bariatric surgery-eligible patients. Among patients undergoing bariatric surgery, patients who underwent SG were found to be at a significantly reduced risk of developing a humeral fracture or fracture in general.
The present study provides valuable clinical information to the field of bariatrics by providing, for the first time to our knowledge, specific analysis of the risk of fracture among patients who undergo bariatric surgery compared with patients who are eligible for bariatric surgery but do not.
Furthermore, this study establishes a timeline for this deferential risk in patients who undergo RYGB compared with patients who undergo SG. In addition, this study demonstrates challenging information to the long-supported idea in medicine that patients with higher BMI (and so patients who are obese) experience a protective benefit against osteoporosis and fractures by illustrating a potential protection against fracture in patients who undergo bariatric surgery. Furthermore, SG was found to be more protective against fracture compared with RYGB.

Fracture Risk in Bariatric Surgery and Obesity
Increasing BMI and obesity have been long associated with higher BMD and lower incidences of fracture. 7,22,28,36-43 Very recently, the Look AHEAD trial, 37-39 a multicenter randomized clinical trial that was designed to determine whether intentional weight loss reduces cardiovascular morbidity and mortality in overweight individuals with T2D, suggested that even relatively small percentages of weight loss (6% to 9%) are associated with significant reductions in BMD and with increased risk of hip, pelvis, and upper arm fracture. Given that most studies that report total body weight loss following RYGB and SG have reported a mean percentage of total weight loss of more than 25% and 18%, respectively, at 5 years, the notion that this kind of major weight loss itself results in an increase in the fracture risk profile of patients would reasonably follow. 7,17,[44][45][46][47][48] Emerging data have started to challenge the notion that obesity is protective against all fractures and has supported that patterns of fat deposition (body fat sites and ratios of visceral fat to subcutaneous fat) and body compositions (muscle mass) may have a greater influence on BMD and fracture risk profiles than BMI alone. [49][50][51][52][53][54][55][56][57] The present study assessed the risk of fractures of the humerus, radius or ulna, pelvis, hip, and vertebrae in patients undergoing bariatric surgery and those who were eligible to undergo bariatric surgery but did not and found that obesity conferred a significantly greater risk of all fracture types and fractures overall compared with patients undergoing SG and similar risks for all fracture types and fractures overall compared with patients undergoing RYGB, providing evidence for a potential protective effect of weight loss against the risk of fractures. In the case of RYGB, a malabsorptive mechanism and complex bone metabolism changes may serve to further complicate this relationship. This does not mean that BMD does not change in these patients; in fact, several studies have described long-term changes in BMD following weight loss surgery as measured by BMD scans and markers of bone turnover, but this further underscores a more multifaceted and multifactorial risk profile for fractures in patients with severe obesity. [58][59][60][61][62][63][64][65][66][67][68][69][70]

Risk of Fracture in RYGB vs SG
Surgical weight loss approaches that alter the fundamental patterns of alimentary absorption, like RYGB, have been shown to hasten this risk and have even been associated with the development of metabolic bone disease, resulting in higher bone turnover and long-term declines, disruptions, and deterioration in bone density and bone microarchitecture. However, it is important to note that many of the clinical studies that have studied postoperative fracture risk and BMD changes following RYGB  [74][75][76][77] Unfortunately, there is a paucity of evidence or information regarding both the independent risks of fracture imparted by SG or comparing bone loss with fracture risk in patients undergoing RYGB, with the very few studies available either providing conflicting information with regards to BMD loss or being underpowered. 72,[78][79][80] These limited clinical data, when interpreted in the context of the limited animal studies comparing bone loss seen following RYGB vs SG, seem to show that the rate of bone loss following SG is less than that observed in RYGB. 81 The present study, to our knowledge, demonstrates for the first time a comparative increase in fracture risk, albeit demonstrated to be lower than their bariatric equivalents, in patients undergoing RYGB vs SG.

JAMA Network Open | Surgery
Specifically, this increased risk was found to be significant in fractures overall and humerus fractures but was not significant when comparing the risk of fractures of the radius or ulna, pelvis, hip, or vertebrae. To this end, it appears that the risk of fracture in patients following SG is less than that observed with RYGB, although additional investigation is required to better elucidate this risk and, further, the expressed relationship between BMD loss and fracture following RYGB and SG.

Limitations
Administrative data allow access to more medical visits nationwide and longitudinal tracking of these patients through distinct identifiers based on a standardized coding system; however, important limitations in the use of these data must be considered. First, administrative data are intended for financial and administrative use rather than research purposes and therefore may vary in detail and accuracy. Second, administrative data also do not provide qualifiable details on the severity of disease states or patient-reported outcome scores or allow for standardization of treatment protocols or surgeon technique or expertise, which may mask certain confounding factors. Third, specifically for the purposes of this article, administrative data limit the assessment of the specific weight loss a patient may experience as a result of bariatric surgery and so we are unable to directly assess any potential associations between absolute weight loss and fracture risk.

Conclusions
The generally accepted notion that obesity is protective when considering the risks of fracture may not be as straightforward as previously thought. The relationship between BMI, body composition, and bone density may play an important role when evaluating the risk of fracture in patients with obesity. Severe obesity status alone might be associated with an increased risk of fracture, and there is a role for weight loss surgery in augmenting this risk. Specifically, SG might be the best option for weight loss in patients in whom fractures could be a concern, as RYGB may be associated with an increased fracture risk compared with SG. Additional studies are needed to not only further characterize the risk profile of obesity on rates of fracture but also to access fracture risk and benefits of different surgical weight loss options.

JAMA Network Open | Surgery
Association of Bariatric Surgery With Risk of Fracture in Patients With Severe Obesity