For a kidney transplant candidate, a living-donor kidney transplant is the best option: it offers the longest survival to the recipient. Currently, more than 27 000 living-donor kidney transplants are done each year in the world. More than 25% of all living kidney donors are now considered obese at the time of donation compared with fewer than 8% in the 1970s, and the proportion of obese recipients is doubling every 15 years as a result of the global epidemic of obesity. Obesity exposes both the donor and recipient to multiple complications: first, of an immediate surgical nature in the intra- and postoperative periods, then of a medical nature in the short and long term. Although surgical complications are well known and better mastered thanks to great technical progress, long-term medical complications can be severe and compromise patient survival and that of the transplanted kidney. For the recipient, the benefits of kidney transplant compared with end-stage renal disease are significant despite all the risks and complications caused by obesity. For the donor, however, no risk of severe complications is acceptable, and the main objective of pretransplant explorations is to ensure that the donation of a kidney occurs in complete safety for the donor. In the presence of obesity, if the donor is highly motivated and barring other exclusion criteria, it is mandatory to clearly inform the donor of all potential obesity-related risks, to strongly encourage him/her to lose weight before the donation, and to maintain this weight afterward.
Key words : Complications, Living donor, Obesity, Surgery
By 2013, there were 2 billion individuals worldwide who were overweight or obese, and 62% of the world’s obese population resided in developing countries.1 Obesity affects all countries, whether developed or developing countries, both sexes, and different age groups.2,3 It becomes a real public health problem because of its burdensome complications, including cardiovascular and metabolic consequences, and the high cost generated by its management.4 The global spread of obesity means that kidney donors and transplant candidates will potentially be obese and raises major medical and surgical difficulties. Indeed, over half of all recipients are either overweight (body mass index [BMI] > 25) or obese (BMI > 30), and obesity affects more than 20% of patients in both chronic hemodialysis and peritoneal dialysis.5-7 Even though some studies have shown better survival among obese patients having had kidney transplants (KT) than among obese patients who stayed on dialysis, access to KT remains difficult.8-10 The long-term prognosis of KT is also compromised because of comorbidities associated with obesity, which will occur in both donor and recipient. All these constraints may lead either to a definitive exclusion of the potential living kidney donor (LKD), or the KT candidate, or the performance of a KT with high risk of peri- and postoperative complications. Obesity is one of the main factors of medical exclusion of LKDs, and its prevalence varies from 14.3% to 24%.11-13 The aim of this literature review was to determine the risks and consequences of obesity in both the LKDs and the kidney transplant recipients (KTRs) and discuss the recommendations of different international societies about obesity in KT.
Definition and Evaluation of Obesity
Obesity is most commonly assessed by BMI, a simple index that is commonly used to classify overweight and obesity in adults.
The BMI is defined as a person’s weight in kilograms divided by the square of the person’s height in meters. Overweight is defined as a BMI between 25 and 29.9, and obesity is defined as a BMI ≥ 30. Obesity is usually classified into 3 stages of increasing severity: class I obesity or moderate obesity (BMI of 30-34.9), class II obesity or severe obesity (BMI of 35-39.9), and class III obesity or morbid obesity (BMI of ≥ 40).14,15 The main limitation of BMI is that it is a poor estimate of fat mass distribution and cannot differentiate body fat from lean mass or central from peripheral fat and therefore cannot predict the cardiovascular risk strongly associated with so-called central abdominal obesity. It is should be noted that BMI, originally derived as a measure predictive of population level mortality risk, is now widely used as a surrogate of obesity assessment even though it is far from an optimal and ideal marker of obesity measurement.16 Measurements of central adiposity, such as increased waist circumference (WC), predict cardiometabolic risk, which cannot be directly determined by elevated BMI, and so central adiposity assessment contributes to accurately assessing cardiovascular risk when BMI is relatively low. Waist circumference measurement may be unnecessary in individuals with high BMI (≥ 35) because almost all of these obese individuals already have a high risk of cardiovascular disease and mortality. Thus, in patients with a BMI ranging from 27 to 34.9, the association of BMI and WC allows a better appreciation of the cardiovascular risk, diabetes risk, and overall risk to individuals.17 Otherwise, WC varies considerably by ethnicity, so no universal classification can be applied. Waist circumference of > 102 cm for men and > 88 cm for women, in American and European populations, is considered elevated and indicative of increased cardiovascular and metabolic risk.14 Table 1 reports the main tools used to evaluate obesity in adult populations.
Obesity and Chronic Kidney Disease
Data published in recent decades suggest that obesity is independently associated with a higher risk of developing chronic kidney disease (CKD), even end-stage renal disease (ESRD).18 The renal effects of obesity in experimental animals and humans include both structural and functional adaptations, such as increased glomerular filtration rate (GFR), increased renal blood flow, and renal hypertrophy.19 Physiological studies indicate that obese patients have elevations of both renal plasma flow and GFR that exceed those of controls by 31% and 51%, respectively, leading to an increased filtration fraction.19 Potential mechanisms for kidney injury in people with obesity may be through hormone activation or the development of other comorbid conditions such as diabetes and hypertension, which may affect kidney function. Hormonal mechanisms include insulin resistance, activation of the renin angiotensin-aldosterone system, low adiponectin levels, high leptin levels, increased synthesis of proinflammatory cytokines, endothelin-1 and insulin-like growth factor-1 and factor-2 and oxidative stress.20,21 Glomerular hyperfiltration seems to be one of the major pathogenic causes for the development of obesity-related glomerulopathy, including focal segmental glomerulosclerosis (FSGS), glomerulomegaly, mesangial hyperplasia, and/or minimal foot process fusion.22,23 In addition, obese individuals have more FSGS and develop obesity-related glomerulopathy whatever the severity of obesity.23 The association of proteinuria with massive obesity was first reported in 1974.24 Obesity-related FSGS is characterized by a lower incidence of nephrotic syndrome compared with idiopathic FSGS, which may relate to differences in the severity of podocyte injury, in the severity and selectivity of proteinuria, and in the ability of the tubules to reabsorb and catabolize the filtered protein.25 Obesity-related FSGS has been described in patients with severe obesity and may be reversible with weight loss.23
Obesity is also a risk factor for kidney insufficiency after unilateral nephrectomy. At 10 years postnephrectomy, 60% of patients whose BMI was > 30 at the time of nephrectomy developed proteinuria (> 3 g/day) and 30% developed kidney insufficiency (creatinine clearance < 70 mL/min).26 These data suggest that nephrectomy in obese patients increases the risk of developing proteinuria and/or kidney insufficiency on the remaining kidney, the seat of FSGS. The risk of ESRD in patients with obesity increases as BMI increases, with an adjusted relative risk for ESRD at 3.6 in those with moderate obesity, 6.1 in those with severe obesity, and 7.1 in those with morbid obesity.18 Stengel and colleagues found that the relative risk of CKD was greater in morbidly obese people than in people of normal weight.27 In addition, comorbidities associated with obesity, such as diabetes and hypertension, can lead to specific kidney damage, accelerate the progression of CKD, and contribute to significantly increase the risk of impaired GFR.
Living Kidney Donors and Obesity
The first successful KT from a living donor was 60 years ago between identical twins. Since then, great technical and therapeutic advances have been made that have allowed KT surgery to be performed at low risk. However, emergence of obesity-related comorbidities that cause CKD, such as diabetes and hypertension, have raised concerns about the long-term prognosis of LKDs.
For a KT candidate, a living-donor transplant is the best option as it offers the longest survival to the recipient, compared with that of using a deceased-donor transplant. More than 27 000 living-donor KTs are now done each year in the world.28 Currently, more than 25% of all LKDs are considered obese at the time of donation compared with fewer than 8% in the 1970s.29 The mean BMI of LKDs in the United States has increased over time, from 24.3 in the 1970s to 27.3 in the 2000s.29 To ensure total safety and to minimize short-and long-term risks to the donor, transplant centers adhere completely to donor exclusion criteria, elaborated by international societies of transplantation. The evaluation of the body weight of the potential donor is one of the major clinical criteria to be considered during the preliminary assessment. This makes it possible to classify the potential donor, in the absence of the other exclusion criteria (hematuria, proteinuria, diabetes, kidney stones, decreased GFR, etc): a risk-free donor has a BMI of less than 25, ie, in the absence of overweight condition and obesity; a low-risk potential donor has a BMI of between 25.1 and 29.9, ie, in the presence of overweight condition, especially when the BMI is greater than 27; a moderate-risk donor has a BMI of between 30 and 31.9; a high-risk donor has a BMI of between 32 and 35; and very high risk is defined by BMI above 35. Body mass index ≥ 30 has long been a relative contraindication to kidney donation, as a BMI at this level is associated with perioperative complications and short- and long-term medical complications.
The principal concern for the obese LKDs is the possibility that donation may have an adverse effect on long-term kidney function from conditions such as hypertension, diabetes, and metabolic syndrome, which may lead to CKD and ESRD. In addition, the presence of obesity in kidney donors is associated in some studies with an increase in perioperative complications, although these are mostly relatively minor in nature and have significantly diminished because of the improvement and better control of new surgical techniques such as laparoscopy and robotic surgery. Friedman and colleagues30 studied surgical complications in 6331 kidney donors, with data collected retrospectively in the United States between 1999 and 2005. Obese patients accounted for 2%, and obesity was identified in multivariate analysis as one of the strongest predictors of higher perioperative complication rates.30 The same observation was also found in the series reported by Patel and colleagues,31 who studied surgical complications in 3074 kidney donors with data collected in the United States between 2004 and 2005. Obese patients accounted for 2.43%, and obesity was identified as one of the strongest predictors of higher perioperative complication rates.31 Serrano and associates studied surgical complications in 3752 kidney donors, with data collected retrospectively in the United States between 1975 and 2014. Obese patients accounted for 17.5% of all patients, and obesity was identified in multivariate analysis as one of the strongest predictors of higher perioperative complication rates.32 In that study, mean estimated GFR at donation was 115 mL/min/1.73 m2 in obese donors versus 97 mL/min/1.73 m2 (P < .001) in nonobese donors. Obese donors were more likely to develop diabetes (P < .001), hypertension (P < .01), and impaired glucose fasting (P < .001) after donation. Furthermore, these health problems were diagnosed earlier after donation.
The GFR, generally high at the moment of kidney donation, is deceptively reassuring because it points to a glomerular hypertrophy related to obesity and not to good physiological kidney function. In a single-center retrospective study, Heimbach and colleagues studied perioperative complications and long-term complications in 553 patients who underwent laparoscopic surgery for LKDs; 114 had moderate obesity and 58 had severe obesity.33 In that study, those patients with severe obesity (≥ 35, n = 58) had longer operative times (mean increase of 19 min), more minor perioperative complications (mostly wound complications), but the same low rate of major surgical complications (conversion to open nephrectomy or reoperation) and a similar length of stay as the 2 other groups with low BMI donors (< 25 and 25-29.9). Concerning the evolution at 1 year after intervention, prenephrectomy blood pressures and fasting plasma glucose levels were higher and rose incrementally with BMI (P < .05), whereas serum creatinine levels did not differ between groups.
Most of these large studies were published before 2010 and concerned the results of LKD nephrectomies performed between 1999 and 2005. The current intraoperative and early postoperative results are probably better as a result of great advances in surgical techniques. Currently, at most KT centers laparoscopic kidney removal is performed, possibly even with robotic assistance, and classic open surgery is increasingly abandoned, except at centers where limits on human and technical resources do not allow for options to exclude of open surgery.34,35
If the surgical and postoperative risk seems minimal, then what about the long-term kidney risk, which is a real concern for the transplant nephrologist and a source of major concern for the donor? In the series from Serrano and colleagues, the authors noted that estimated GFR had dropped significantly to 62 mL/min/1.73 m2 in obese LKDs versus 67 mL/min/1.73 m2 in nonobese LKDs (P = .005), 20 to 30 years after donation, and the rate of ESRD was not statistically different between obese LKDs and nonobese LKDs in this cohort.32 In the series from Locke and colleagues,36 the authors compared the incidence of ESRD in obese LKDs and nonobese LKDs and found that the cumulative incidence of ESRD per 10 000 living donors was 3-fold greater among LKDs who were obese at the time of donation compared with their nonobese LKDs counterparts. Furthermore, at 20 years postdonation, obese LKDs had a cumulative incidence of ESRD of 93.9 per 10 000 compared with 39.7 per 10 000 among their nonobese LKD counterparts.36 In that study, the BMI at time of donation was 32.7 in the obese group (n = 20 588) versus 24.8 in the nonobese group (n = 58 004). After controlling for multiple risk factors, the only potentially modifiable factor that remained independently associated with increased risk for development of ESRD postdonation was obesity. In addition, compared with nonobese LKDs, obese LKDs had a 1.9-fold increased risk of ESRD postdonation (adjusted hazard ratio, 1.86; 95% confidence interval [95% CI], 1.05-3.30; P = .04), and for each 1 unit increase in predonation BMI above 27, there was an associated 7% increased risk of ESRD postdonation. Note that, in this cohort of LKDs, all donors were free of diabetes and hypertension at the time of donation. These studies revealed that even overweight donors with BMI > 27 are at risk of developing ESRD as well as obese LKDs with BMI > 30. This is a major element to integrate into the selection for kidney donation, and the donor with a BMI between 27 and 30 should no longer be considered as a risk-free donor.
It is important to emphasize that, while the recipient gains several benefits from receiving a KT, the donor must incur no risk and even less that of ESRD. It is important to note that, particularly with overweight and obese potential donors, predonation counseling should aim to clearly and explicitly explain all potential risks of surgical and medical complications throughout the perioperative, short-term, and long-term evolution.
The decision to accept the obese LKD must be carefully considered and account for all clinical and laboratory criteria, the real risk to the donor, and the real benefit to the recipient. If the decision is taken to accept the obese LKD, the following step should include discussions of lifestyle modifications and weight loss. There are 2 options: losing weight in an intensive weight loss program, with or without pharmacological treatment, or bariatric surgery, which remains indicated for BMI > 40, or BMI > 35 if medical measures fail. Nguyen and colleagues37 retrospectively studied a cohort of 675 LKDs, from data collected between 2011 and 2017. In this group, 175 were obese and 22 had benefited from bariatric surgery before donation.37 The mean BMI was 46.2 versus 30.6, respectively, before and after bariatric surgery, and the authors noted that bariatric surgery allowed the 22 morbidly obese LKDs who would have otherwise been discouraged from kidney donation to become candidates without complicating their subsequent laparoscopic living-donor nephrectomy or affecting their early postoperative outcomes.
An important meta-analysis, comprising 29 studies, determined a quality of life in all LKDs without differentiating the obese from the nonobese.38 In 17 of these studies, many donor scores were similar or better than for the general population. Additionally, quality of life scores were no different or better for donors who had a laparoscopic nephrectomy compared with those who underwent an open procedure. In a recent study, Klop and colleagues studied a multivariate analysis of health-related quality of life in 501 donors after kidney donation.39 In this study, the authors found an inverse proportional relationship between BMI and quality of life, and even when adjusting for baseline, this effect was still significant, suggesting a diminished quality of life for donors who had an increase of BMI of 5 at baseline. However, no significant relationship was observed between BMI and the mental component score. An important question is whether lifestyle changes can decrease the long-term postdonation risk of diabetes and hypertension, especially among obese LKDs. This is especially important in the younger obese donors, whose long-term sequelae from donation are contingent on future lifestyle choices (smoking, diet, weight management, pregnancy, routine medical care), and there is no certainty about this subject.Kidney Transplant Recipients and Obesity.
The proportion of obese recipients is doubling every 15 years as a result of the global epidemic of obesity, exposing recipients to early postoperative, short-term, and long-term complications that may affect global survival and/or graft survival.5,40 Concerning mortality, it has recently been reported that KT confers survival benefit in obese recipients that is similar to those with a normal BMI when compared with maintenance dialysis in the first year after transplant.9
Hill and colleagues conducted an important meta-analysis about outcomes after KT in a context of recipient obesity.41 In this meta-analysis, which included 17 studies, recipient obesity was not associated with an increased risk of death compared with recipients with a normal BMI, and the authors concluded that potential recipients should, therefore, not be excluded from KT solely based on obesity. However, obese KTRs had a marginally greater risk of death-censored graft loss and were more likely to develop delayed graft function. Obese individuals undergoing abdominal surgery for KT also have more frequent anesthetic perioperative complications and longer hospital stays, compared with the nonobese population.42,43 This may have an impact upon early graft outcome measures, such as the incidence of delayed graft function, and could result in premature graft loss.44,45
Schwarznau and colleagues retrospectively studied 81 LKRs, performed between 2000 and 2004, in which 30.8% of all recipients had BMI > 25, and the prevalence of graft survival was lower in the group with BMI > 25 (76%) than in the group with BMI < 25 (94.8%) (P < .001).46 In multivariate analysis of this study, the BMI was an independent risk factor for graft loss within the first year. After analysis of body weight of the donor and the recipient, the authors found that the quotient of the squared value of recipient body weight divided by the value of donor body weight was also an independent risk factor for graft loss. Ditonno and colleagues conducted a retrospective study that included 563 KTRs (collected between 2000 and 2008) and noted that delayed graft function was observed in 29.4%, 26.4%, 30%, 50%, and 40%, respectively, in groups A (BMI < 18.5), B (BMI 18.6-24.9), C (BMI 25-29.9), D (BMI 30-34.9), and E (BMI ≥ 35), without a statistically significant difference.47 In this study, obese recipients (groups D and E) showed significantly higher mean serum creatinine values and worse renal function at 6 months, 1 year, and 3 years (P < .001), but no statistically significant differences were observed with regard to patient or graft survivals at 6, 12, 36, or 60 months. Patients with a BMI > 30 were at increased risk of acute rejection compared with nonobese recipients (P = .01) and were more susceptible to cardiovascular and metabolic complications (P = .01). Also, subjects in group E showed a higher incidence of postsurgical complications (P = .002).
Weissenbacher and colleagues conducted a retrospective study that included 1113 KTRs, collected from 2000 to 2009, and noted in multivariate logistic regression analysis that recipient BMI was independently associated with delayed graft function (odds ratio, 1.09; 95% CI, 1.04-1.13; P < .0001).48 These last 2 studies were conducted in a context of deceased donors, in which several factors such as donor age, immunological compatibility, cold ischemia time, associated comorbidities, etc. intervene and explain a part of these results. Concerning transplanted kidneys, damage may be caused by specific pathophysiological mechanisms similar to those which occur in the native kidneys of obese patients (mentioned above).
As for peri- and postoperative complications, Greze and colleagues studied a retrospective cohort of 245 KTRs performed from 2008 to 2014 in France and found no statistically significant difference concerning delay in graft function recovery, duration of postgraft hospitalization, rejection, bacterial infection, cardiovascular complications, postgraft diabetes, vascular complications of the graft, and urinary tract complications.49 In addition, there is also some evidence that obesity may alter the metabolism and bioavailability of immunosuppressive medications with a potential error in the estimation of the dosage of immunosuppressive drugs, exposing the kidney allograft to underexposure or overexposure, thus increasing the risk of many and serious complications. One of the most probable mechanisms is the changes of expression levels of drug-metabolizing enzymes and drug transporters. One possibility is that inadequate dosing adjustments are being made to maintain immunosuppressive medications according to patient weight. In fact, if one aims to maintain the usual recommended doses, then the daily doses may well turn out to be too high in obesity classes I and II and too low in obesity class III.
Some studies have shown that the risk of acute rejection has increased in KTRs with high BMI, but the mechanisms underlying this relationship are unclear.50,51 The risk of chronic immunological injury is also increased in this context.52 Rodrigo and colleagues studied the link between high tacrolimus doses and recipient BMI in data of 85 KTRs collected from 1994 to 2003 and noted, after logistic regression, that age (relative risk, 1.05; 95% CI, 1.01-1.08; P = .02) and BMI (relative risk, 1.18; 95% CI, 1.01-1.37; P = .03) remained significant risk factors for developing high initial blood levels of tacrolimus.53 In this study, the mean BMI was 27± 4 versus 23 ± 3 and the mean dose of tacrolimus was 13.7 ± 3.3 mg/day versus 11.5 ± 3.4 mg/day, respectively, in the group of KTRs with tacrolimus concentration > 15 ng/ml and the group of KTRs with tacrolimus concentration < 15 ng/ml. Flabouris and colleagues analyzed the link between BMI, weight-adjusted immunosuppression, and the risk of acute rejection and infection after KT in their cohort of 7919 KTRs.54 In this study, 3.06% of patients were underweight and 21.1% were obese, of whom 81.8% had class I obesity (BMI 30-34.9), 15.5% had class II obesity (BMI 43.5-39.9), and 2.69% had class III obesity (BMI ≥ 40). The authors found that overweight and obese patients were at increased risk of acute rejection and, compared with those with a BMI of 25, patients with a BMI of 35 were at a higher risk of developing acute rejection. Obesity exposes the patient to high doses of immunosuppressive drugs, particularly in the case of the anticalcineurins. Dosages are based on body weight, and any overdose incurs the risk of an acute and/or chronic nephrotoxicity that may cause or contribute to graft loss. Despite all of these risks, and whatever the level of risk for the future KTRs, it remains acceptable in relation to the risk of ESRD, especially in young patients with a long life expectancy. Table 2 reports the most important risks of complications occurring in obese donors related to nephrectomy and in KTRs related to obesity.
International Recommendations for Kidney Donation
Almost all of the literature suggests that transplant programs generally exclude as donors individuals with a BMI > 35 and 10% of programs exclude donors with a BMI > 30.55 Most guidelines set the value of BMI 35 as the threshold for absolute contraindication to kidney donation. However, potential donors with morbid obesity, in the absence of other donation exclusion criteria and if the donors are highly motivated, may fully adhere to weight loss programs or benefit from bariatric surgery to achieve a BMI < 35.
The threshold of BMI 35 is not a definitive absolute contraindication but should be considered as a temporary contraindication. Kidney Disease Improving Global Outcomes, in their recommendation published in 2017, advocate that the decision to approve donor candidates with obesity and BMI > 30 should be individualized based on demographic and health profiles in relation to the transplant program’s acceptable risk threshold, but without specifying the upper threshold beyond which the donation is no longer authorized.56 European Renal Best Practice, published in 2015, state that a BMI of 35 is a contraindication to living kidney donation without giving more details.57 The recommendations of the British Society, published in 2018, state that there is no formal upper limit of BMI for donation, but increased risk with a BMI of 30 to 35 and limited safety data when it is greater than 35.58 The International Forum on the Care of the Live Kidney Donor, including over 100 experts and leaders in transplantation representing more than 40 countries from around the world, which gathered in Amsterdam in 2004, stated that patients with a BMI > 35 should be discouraged from donating, especially when other comorbid conditions are present, and that obese donors should be encouraged to lose weight prior to kidney donation and should be advised not to donate if they have other associated comorbid conditions.59 Australian and New Zealand guidelines advise measurement of WC within the assessment of overweight and obese donor candidates and state that obesity (BMI > 30) should be considered a relative contraindication to donation.60 The Canadian Council for Donation and Transplantation states that there is debate regarding the eligibility of those with BMI > 35 and that little is known about either the long-term risks to such donors or the long-term outcomes of kidneys from such donors.61 In addition, all of these guidelines recommended the following: obese patients should be informed of both acute and long-term risks, especially when other comorbid conditions are present; healthy lifestyle education should be available to all living donors; and these donors should be advised to pursue weight loss before donation and to maintain a healthy body weight after donation. Figure 1 illustrates a global approach to assessment of the potential kidney donor who is overweight or obese.
Obesity is a common, and potentially modifiable, condition in the KT population. It cannot be considered at the outset as a definitive and absolute contraindication to KT. Rather, it should be used to encourage the patient who is motivated and without other associated exclusion criteria, to lose weight, aiming to bring the BMI down to less than 30 if initial BMI is < 35 or to < 32 if initial BMI is > 35. The data on the effects of recipient and living donor BMI on graft survival, long-term complications such as metabolic or cardiovascular disorders, and on survival are controversial and require more rigorous and precise research to study the different aspects of this subject.
DOI : 10.6002/ect.2020.0074
From the 1Nephrology and Kidney Transplantation Unit, University Hospital
Mohammed VI, Oujda, University Mohammed First, Oujda, Morocco; and the
2Laboratory of Epidemiology, Clinical Research and Public Health, Medical
School, University Mohammed First, Oujda, Morocco
Acknowledgements: The author has not received any funding or grants in support of the presented research or for the preparation of this work and has no potential declarations of interest.
Corresponding author: Yassamine Bentata, Avenue Hassan II, rue Kadissia, numéro 12, Oujda, Morocco
Table 1. The 3 Main Tools to Estimate Obesity in Adult Populations
Table 2. The Most Important Risks of Complications Occurring in Obese Donors Related to Nephrectomy and in Recipients Related to Obesity
Figure 1. A Global Approach to Risk Assessment of the Potential Kidney Donor According to Body Mass Index