ORIGINAL CONTRIBUTIONS
Effects of bariatric surgery in different obesity phenotypes: Tehran Obesity Treatment Study (TOTS)
Maryam Barzin1&Shayan Aryannezhad1&Alireza Khalaj2&Maryam Mahdavi1&Majid Valizadeh1&Sahar Ghareh3&
Feridoun Azizi4&Farhad Hosseinpanah1
#Springer Science+Business Media, LLC, part of Springer Nature 2019 Abstract
BackgroundNot all morbid obese patients suffer from metabolic co-morbidities; thus, a sub-group of metabolically healthy morbid obese (MHMO) individuals are identified. However, the role of bariatric surgery is not well understood in this subgroup.
Methods A total of 2244 morbid obese individuals aged 18–65 years undergoing bariatric surgery were selected. Patients were considered MHMO according to the joint interim statement (JIS) definition, as having two or less abnormalities in these five parameters: waist circumference (WC), triglyceride (TG), high density lipoprotein cholesterol (HDL-C), systolic or diastolic blood pressure (SBP or DBP), and fasting plasma glucose (FPG). Otherwise, they were considered metabolically unhealthy morbid obese (MUMO). Follow-up data were collected at 6, 12, and 24 months post-surgery.
Results Prior to surgery, 36.2% of participants were MHMO and had significantly lower BMI, WC, TG, FPG, SBP, and DBP and higher HDL-C compared to MUMO. Both MHMO and MUMO participants showed a significant decrease in BMI, WC, TG, SBP, DBP, and FPG and increase in HDL-C and the percentage of excess weight loss (%EWL). Two-year post-operative changes (from baseline) of BMI, WC, and %EWL were greater in MHMO subjects and changes of TG, HDL-C, DBP, SBP, and FPG were greater in MUMO subjects. Further multivariate regression analysis for delta (Δ) change in these characteristics revealed that only the delta (Δ) changes of WC and %EWL were statistically different between the two phenotypes and were greater in MHMO subjects, 2 years after the surgery (−3.077 cm decrease in WC and + 3.612% higher %EWL compared to MUMO subjects).
ConclusionBariatric surgery is an effective method for reduction of metabolic abnormalities and weight loss in both MUMO and MHMO phenotypes. Benefits of this intervention are comparable between patients with these two obesity phenotypes.
Keywords Bariatric surgery . Morbid obesity . Obesity phenotype
Introduction
Morbid obesity (defined as body mass index [BMI] > 40 or BMI > 35 kg m2in the presence of significant co-morbidities)
has a well-demonstrated association with many adverse health conditions including impaired glucose tolerance and diabetes mellitus, hypertension, dyslipidemia, and metabolic syndrome [1], making a large proportion of individuals with morbid obesity metabolically unhealthy morbid obese (MUMO).
However, a subgroup of metabolically healthy morbid obese (MHMO) individuals are also identified who are protected against these cardiometabolic co-morbidities [2]. Different definitions have led to the description of a “metabolically healthy”state in the MHMO phenotype, which is character- ized by the absence of metabolic abnormalities such as dys- lipidemia, insulin resistance, hypertension, and an unfavor- able inflammatory profile [3]. Nevertheless, long-term prog- nostic value of the MHMO phenotype is questionable, and while this phenotype was originally regarded to be a static condition, it is becoming increasingly evident that the MHMO status has a transitional nature [4,5]. Besides, having a healthy metabolic profile in obese patients cannot
* Farhad Hosseinpanah [email protected]
1 Obesity Research Center, Research Institute for Endocrine Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran
2 Tehran Obesity Treatment Center, Department of Surgery, Faculty of Medicine, Shahed University, Tehran, Iran
3 Mashhad Medical Branch, Faculty of Medicine, Islamic Azad University, Mashhad, Islamic Republic of Iran
4 Endocrine Research Center, Research Institute for Endocrine Sciences, Shahid Beheshti University of Medical Science, Tehran, Iran
https://doi.org/10.1007/s11695-019-04182-y
necessarily prevent many co-morbidities such as chronic back pain, osteoarthritis, asthma, and sleep apnea [6]. Therefore, interventions for weight reduction have potential benefits in the quality of life of both MHMO and MUMO patients.
Comparing different strategies for long-term manage- ment of morbid obesity, bariatric surgery is considered to be more effective than other non-surgical alternatives such as weight loss through lifestyle modification, nu- tritional diet, and pharmacologic interventions [7–9].
Furthermore, use of some dietary interventions in meta- bolically healthy obese subjects for management of obe- sity and its co-morbidities is not supported by the evi- dence [10]. In addition to significant weight loss, bar- iatric surgery has other beneficial effects on improve- ment of many obesity-related co-morbidities as well [11]. However, the beneficial role of bariatric surgery on the MHMO subgroup has remained unclear, and it is not fully understood how this procedure could affect weight loss and cardiometabolic health status of individ- uals with this phenotype. Studies addressing the role of anti-obesity treatments for MHMO phenotype are con- troversial, while some concluded that weight loss strat- egies may not be able to significantly reduce obesity- related risks of MHMO [12, 13], few studies showed that bariatric surgery could improve the metabolic status of this subgroup as well [14, 15].
Therefore, due to the lack of established knowledge regard- ing the effect of bariatric surgery on different obesity pheno- types, we aimed to investigate and compare weight loss and metabolic outcomes of bariatric surgery in MUMO and MHMO individuals.
Material and methods
Study design and participantsThis study was conducted within the framework of the Tehran Obesity Treatment Study (TOTS), an ongoing prospective study in which individuals with morbid obe- sity undergo bariatric surgery in an obesity treatment center. Participants of the study were 18–65 years old, with obesity class II (35 ≤ BMI < 40 kg/m2) who had comorbidities or did not achieve durable weight loss with nonsurgical methods or class III (BMI ≥ 40 kg/
m2). More data regarding the study protocol is available elsewhere [16]. Of the 2620 morbid obese patients who underwent bariatric surgery, 376 individuals were ex- cluded (under 18 years [n = 8], had missing baseline data [n = 123] and did not have any further referral to our center [n = 245]). The remaining 2244 individuals who had one or more post-operation follow-up data (ei- ther at 6, 12, or 24 months after the surgery) made up
the study population. These participants were divided into two groups; 1431 were metabolically unhealthy morbid obese (MUMO) and 813 were metabolically healthy morbid obese (MHMO) at baseline. Follow up rate for 6, 12, and 24 months post-operation were 86.3%, 85.6%, and 65.3%, respectively, with no signif- icant differences between MUMO and MHMO pheno- types. They all underwent bariatric surgery (either s l e e v e g a s t r e c t o m y o r g a s t r i c b y p a s s ) f r o m March 2013 to March 2018 and were assessed for an- thropometrics and metabolic outcomes after the surgery.
Measurements
Trained investigators collected data according to study proto- col. All patients underwent routine medical history and phys- ical examinations (including systolic and diastolic blood pres- sure [BP]) for observation of obesity-related conditions as well as determining the general health status of the individual.
Baseline data including demographics (sex, age) and anthro- pometric indices (weight, height, waist circumference [WC], and body mass index [BMI]) were measured according to WHO guidelines [17].
Blood samples were collected before surgery and 6, 12, and 24 months post-surgery. After 12–14 h of overnight fasting, blood samples were taken and assessed for general blood chemistry (lipid profile including serum triglyceride [TG], to- tal cholesterol and high density lipoprotein cholesterol [HDL- C], and fasting plasma glucose [FPG]). For evaluating weight loss outcomes, the percentage of excess weight loss was cal- culated as %EWL = [(Initial weight) – (Post operation weight)] / [(Initial weight) −(Ideal weight)], in which ideal weight is defined by the weight corresponding to a BMI of 25 kg/m2.
Surgery procedures
All procedures were completed by a single surgical team, under general anesthesia. Patients underwent either laparo- scopic sleeve gastrectomy (SG) or gastric bypass (GB). GB was performed by either the Roux-en-Y (RYGB) method or mini-gastric bypass (MGB): the technical details of these pro- cedures have been published elsewhere [16]. Briefly, SG was performed over a 36-F bougie and reinforced with running sutures. RYGB was performed with construction of a vertical pouch of stomach and anastomosis to an antecolic 150-cm roux limb of jejunum and a side-to-side jejunojejunostomy with a 50-cm biliopancreatic limb. For MGB, a long gastric tube was created using Endo GIA stapler (Endo GIA Auto suture, Covidien, Mansfield, MA, USA) from the incisura angularis to the angle of His over a 36-F bougie. An antecolic loop gastrojejunostomy was performed 160 cm distal to the ligament of Trietz with an Endo GIA stapler and reinforced
with continuous sutures. Of the total 2244 patients, 1451 (64.7%) underwent SG and 793 (35.3%) subjects underwent GB (651 MGB and 142 RYGB).
Definitions
Metabolically healthy status in morbid obese patients was defined based on the joint interim statement (JIS) definition [18]; MHMO phenotype was considered as having two or fewer of the following five parameters:
1. WC≥91 cm in females and≥89 cm in males—based on national cut-offs [19].
2. TG≥150 mg/dl or using lipid lowering medication.
3. HDL-C < 40 in men and < 50 in women or using lipid lowering medication.
4. Systolic blood pressure (SBP)≥135 mmHg and/or dia- stolic blood pressure (DBP)≥85 mmHg or taking anti- hypertensive medication.
5. FPG≥100 mg/dl or anti-diabetic medication use.
Education level was assessed using a questionnaire and categorized into two groups; either completing 12 years of education or less (high school diploma or lower) or complet- ing more than 12 years of education (higher than high school diploma).
Smokers were defined as adults who have smoked 100 cigarettes in their lifetime and currently smoke cigarettes ev- ery day (daily) or some days (nondaily).
Statistical Analysis
Variables with normal distribution have been illustrated as mean ±SD, and skewed variables as geometric mean, CI.
Categorical variables were illustrated as frequency (percent- ages). To assess the significance of differences in baseline characteristics for categorical data, the t test or Mann- Whitney test was used. To assess the significance of differ- ences in the baseline characteristics for continuous data, the Pearson chi-square test was used. Longitudinal differences in BMI, WC, TG, HDL-C, SBP, DBP, FPG, and %EWL throughout the follow-up period were analyzed using the gen- eralized estimated equation (GEE) method. Trends were ana- lyzed using GEE with autoregressive working correlation structures through linear model with identity link function.P values were calculated for trends along the follow up (Ptrend), and interaction tests between the subgroups (P interaction).
Association of change (Δ) in BMI, WC, TG, HDL-C, SBP, DBP, FPG, and %EWL between preoperative and 12 or 24 months postoperative values was performed using linear re- gression model, adjusted for the covariates including sex, type of surgery, and baseline values for each dependent variable.
All analyses were performed using IBM SPSS for Windows,
version 20 (SPSS, Chicago, IL, USA).P value < 0.05 was considered as statistically significant.
Results
Of the study participants at baseline, 813 (36.2%) were iden- tified as MHMO and 1431 (63.8%) as MUMO phenotype.
Subjects were mostly female in both groups, but MHMO pa- tients were younger (34.7 ± 10.5 vs. 41.1 ± 11.1 years).
MHMO patients had a significantly lower BMI, WC, serum TG level, FPG, SBP, and DBP and higher serum HDL-C compared to MUMO patients. Smoking status did not differ between the two groups. MHMO patients had higher educa- tional level than the MUMO. Regarding parameters of healthy metabolic status based on JIS definition, elevated WC was the most common abnormality in our participants, and almost all the individuals in both groups had elevated WC (data not shown). The second most prevalent metabolic abnormality was decreased serum HDL-C level in both groups (28.7% in MHMO vs. 78.4% in MUMO patients). The least prevalent metabolic abnormality in MHMO patients was elevated TG (12.6%) and in MUMO patients, it was elevated BP (57.5%).
Study participants underwent either SG or GB, and the pro- portion of surgery methods used did not differ between the two phenotypes. More details of baseline characteristics of the study population are available in Table1.
Weight loss and metabolic outcomes of surgery at 6, 12, and 24 months post-surgery are available in Table2and Fig.
1. Both MHMO and MUMO patients showed a significant decrease in BMI and WC (Ptrend< .001) which was greater in the MHMO phenotype (Pbetween-groups< .001, Fig.1a, b).
Changes in %EWL were significant for both phenotypes throughout the follow-up (Ptrend< .001, Fig.1h). Two years after the operation the %EWL was 74.3 ± 21.7 in MUMO and 80.2 ± 21.0 in MHMO (Pbetween-groups= 0.006,Pinteraction= 0.006.
Regarding post-operative changes of metabolic parameters, both phenotypes showed a decrease in TG, SBP, DBP, and FPG (Ptrend< .001, Fig.1c, e, f, g) and increase in HDL-C (Ptrend< .001 for MUMO andPtrend= 0.03 for MHMO, Fig.
1d).Pvalues were significant for all these parameters between groups. The absolute changes from baseline of TG, HDL-C, SBP, DBP, and FPG were greater in MUMO subjects. The prevalence of patients with elevated TG, low HDL-C, elevat- ed BP, and elevated FPG showed a decreasing trend through- out the 2-year follow-up in patients with both obesity pheno- types. Compared to MHMO, MUMO patients experienced a greater decrease in the prevalence of elevated TG, low HDL- C, and elevated FPG, whereas MHMO patients experienced a greater decrease in the prevalence of elevated BP compared to MHMO patients.
Between group differences of TG, HDL-C, SBP, and FPG were more prominent in the first phase of follow-up compared to the rest of the follow-ups (significant Pvalues of time interaction).
Table3shows multivariate regression analysis for delta (Δ) change in different characteristics of the study population at 12 and 24 months post-surgery. Standardized beta and P values ofΔare shown for obesity phenotype (MHMO) as an
independent variable and are adjusted for age, sex, and base- line values of each variable and type of bariatric surgery per- formed. At 12 months post-operation, compared to MUMO patients, MHMO patients experienced a greater delta (Δ) change in BMI, WC, TG, and %EWL (Pvalue < 00.5). At 24 months post-operation, however, only the delta (Δ) chang- es in the WC and %EWL were statistically different between the two phenotypes.
Table 1 Baseline characteristics of metabolically healthy morbid obese (MHMO) and metabolical- ly un-healthy morbid obese (MUMO) patients prior to bariat- ric surgery
Variable MUMO (N= 1431 ) MHMO (N= 813 ) Pvalue
Sex,n(%)
Male 272 (19.0) 161 (19.8) 0.646
Female 1159 (81.0) 652 (80.2)
Age (years) 41.1±11.1 34.7±10.5 < .001
Age groups,n(%)
< 40 658 (46.0) 573 (70.5) < .001
≥40 773 (54.0) 240 (29.5)
BMI (kg/m2) 45.4±6.1 44.4±5.6 < .001
BMI group,n(%)
≥35, < 40 221 (15.5) 166 (17.7) < .001
≥40, < 45 527 (36.8) 350 (43.1)
≥45, < 50 403 (28.2) 203 (25.0)
≥50 280 (19.6) 116 (14.3)
WC (cm) 124.5 ±13.4 120.8 ±14.2 < .001
TG(mg/dl), median IQ 25-75 167.0 (163.3-170.7) 107.7 (105.2-110.3) < .001
Elevated TGa,n(%) 1061 (64.5) 102 (12.6) < .001
HDL-C (mg/dl) 45.0±10.9 52.8±11.0 < .001
Decreased HDLb,n(%) 1117 (78.4) 232 (28.7) < .001
SBP (mm Hg) 126.7 ±14.2 119.0±9.1 < .001
DBP (mm Hg) 80.7 ±8.4 76.7±7.2 < .001
Elevated BPc,n(%) 820 (57.5) 107 (13.2) < .001
FPG (mg/dl) 119.5±43.5 93.7±18.6 < .001
Elevated FPGd,n(%) 980 (68.5) 108 (13.3) < .001
Smoking,n(%) 0.065
Smoker 148 (10.9) 104 (13.6)
Non-smoker 1214 (89.1) 663 (86.4)
Education,n(%) 0.005
High school diploma or lower 755 (54.4) 375 (48.1)
Higher than high school diploma 634 (45.6) 404 (51.9) Surgery type,n(%)
SG 914 (63.9) 537 (66.0) 0.198
GB 516 (36.1) 277 (34.0)
RYGB 84 (16.3) 58 (20.9)
MGB 432 (83.7) 219 (79.1)
Data are expressed as total numbers, mean ± SD, median IQ 25-75 or percentages
BMIbody mass index,WCwaist circumference,TGtriglyceride,HDL-Chigh-density lipoproteins-cholesterol, BPblood pressure,FPGfasting plasma glucose,SGsleeve gastrectomy,GBgastric bypass,RYGBRoux-en-Y gastric bypass,MGBmini-gastric bypass
aTG≥150 mg/dl or lipid lowering medication use
bHDL-C < 40 in men and < 50 in woman or lipid lowering medication use
cSystolic BP≥135 mmHg and/or diastolic BP≥85 mmHg or anti-hypertensive medication use
dFPG≥100 mg/dl or anti-diabetic medication use
Table2Differentcharacteristicsofmetabolicallyhealthymorbidobese(MHMO)andmetabolicallyun-healthymorbidobese(MUMO)patientsbeforeandafterbariatricsurgery VariablePhenotypePre-operationPost-operationChangefrombaselinePtrendPbetween- groupsPinteraction 6months12months24months BMI(kg/m2 )MUMO45.4±6.133.5±4.830.7±4.830.8±5.2−15.0±5.1<.001<.0010.131 MHMO44.4±5.532.4±4.729.3±4.429.5±4.9−15.5±4.9<.001 WC(cm)MUMO124.6±13.3101.9±11.496.5±11.896.9±11.9−28.0±12.5<.001<.0010.133 MHMO120.2±13.997.5±11.791.7±11.291.1±11.4−30.5±12.8<.001 TGlevel(mg/dl)MUMO166.9(163.3–170.7)117.3(114.8–119.9)97.4(94.7–100.3)91.2(85.8–97.0)−68.5±67.5<.001<.001<0.001 MHMO107.7(105.1–110.3)89.5(87.2–91.8)76.8(74.3–79.4)70.6(65.4–76.3)−45.3±54.7<.001 HighTG,n(%)MUMO923(75.4)281(25.8)111(16.6)28(15.7)−79.2%a <.001<0.0010.001 MHMO86(12.7)28(4.8)7(1.9)4(4.4)−65.3%a <.001 HDL-Clevel(mg/dl)MUMO45.2±10.945.9±12.550.3±11.452.3±12.36.7±13.1<.001<.001<0.001 MHMO52.7±11.249.4±11.554.3±13.657.3±12.91.7±13.6.030 LowHDL-C,n(%)MUMO969(79.0)690(63.7)302(45.4)71(41.0)−48.1%a <.001<0.001<0.001 MHMO203(29.9)283(49.0)121(33.4)19(21.6)−27.8%a .338 SBP(mmHg)MUMO126.9±14.4116.0±11.2115.3±11.4115.9±12.9−10.6±16.8<.001<.001<0.001 MHMO118.9±9.1113.7±10.2112.8±10.8112.9±9.0−4.9±14.9<.001 DBP(mmHg)MUMO80.7±8.672.1±9.571.6±9.473.1±9.8−6.6±12.6<.001<.0010.194 MHMO76.6±7.070.2±9.069.3±9.470.1±8.9−4.5±11.6<.001 HighBP,n(%)MUMO727(58.4)244(23.2)125(16.3)58(20.8)−64.4%a <.001<.0010.198 MHMO95(13.2)32(5.5)26(6.0)1(0.7)−94.7%a <.001 FPG(mg/dl)MUMO120.4±43.893.6±20.289.5±15.291.8±18.4−29.4±38.4<.001<.001<0.001 MHMO93.6±17.686.1±12.585.1±7.886.1±12.4−9.3±21.9<.001 HighFPG,n(%)MUMO879(69.3)268(23.8)112(16.0)44(21.8)−68.5%a <.001<.001<.001 MHMO88(12.9)31(5.2)9(2.5)5(5.3)−58.9%a <.001 %EWLMUMO–61.5±16.074.9±19.174.3±21.713.3±15.1b <.0010.0060.006 MHMO–65.1±16.781.2±19.380.2±21.017.2±14.7b <.001 Dataareexpressedasmean±SDorgeometricmean,CI BMIbodymassindex,WCwaistcircumference,TGtriglyceride,HDL-Chigh-densitylipoproteins-cholesterol,SBPsystolicbloodpressure,DBPdiastolicbloodpressure,FPGfastingplasmaglucose, %EWLthepercentageofexcessweightloss a Changefrombaselineofprevalenceiscalculatedas[(24monthspost-operationvalue−pre-operationvalue)/pre-operationvalue]×100% b Sincetherewasnobaselinevaluefor%EWL,changefrom6monthswascalculated
27 29 31 33 35 37 39 41 43 45 47
0 6 12 18 24
BMI(kg/m2) )naeMlanigraMdetamitsE(
Follow up (Months)
MUMO MHMO MUMO Ptrend= <0.001 MHMO Ptrend= <0.001 Pbetween-groups= <0.001 Pinteracon= 0.131
85 90 95 100 105 110 115 120 125 130
0 6 12 18 24
WC(cm) (Estimated Marginal Mean)
Follow up (Months)
MUMO MHMO MUMO Ptrend= <0.001 MHMO Ptrend= <0.001 Pbetween-groups= <0.001 Pinteracon= 0.133
60 80 100 120 140 160 180 200
0 6 12 18 24
TGlevel(mg/dl) )naeMlanigraMdetamitsE(
Follow up (Months)
MUMO MHMO MUMO Ptrend= <0.001 MHMO Ptrend= <0.001 Pbetween-groups= <0.001 Pinteracon= <0.001
0 10 20 30 40 50 60 70 80
0 6 12 18 24
HDL-C level (mg/dl) (Estimated Marginal Mean)
Follow up (Months)
MUMO MHMO MUMO Ptrend= <0.001 MHMO Ptrend= 0.030 Pbetween-groups= <0.001 Pinteracon= <0.001
a b
c d
110 112 114 116 118 120 122 124 126 128
0 6 12 18 24
SBP (mm Hg) )naeM lanigraM detamitsE(
Follow up (Months)
MUMO MHMO MUMO Ptrend= <0.001 MHMO Ptrend= <0.001 Pbetween-groups= <0.001 Pinteracon= <0.001
50 55 60 65 70 75 80 85 90
0 6 12 18 24
DBP (mm Hg) (Estimated Marginal Mean)
Follow up (Months)
MUMO MHMO MUMO Ptrend= <0.001 MHMO Ptrend= <0.001 Pbetween-groups= <0.001 Pinteracon= 0.194
0 20 40 60 80 100 120 140
0 6 12 18 24
)ld/gm( GPF )naeM lanigraM detamitsE(
Follow up (Months)
MUMO MHMO MUMO Ptrend= <0.001 MHMO Ptrend<0.001 Pbetween-groups= <0.001 Pinteracon= <0.001
0 10 20 30 40 50 60 70 80 90
0 6 12 18 24
% EWL (Estimated Marginal Mean)
Follow up (Months)
MUMO MHMO MUMO Ptrend= <0.001 MHMO Ptrend= <0.001 Pbetween-groups= 0.006 Pinteracon= 0.006
e
f
g
h
Fig. 1 Trend of different characteristics of MHMO and MUMO phenotype in study population during 24 months of follow-up.aBody Mass Index (BMI).bWaist circumference (WC).cTriglyceride (TG).d
High-density lipoproteins-cholesterol (HDL-C).eSystolic blood pressure (SBP).fDiastolic blood pressure (DBP).gFasting plasma glucose (FPG).
hPercentage of excess weight loss (%EWL)
Discussion
In this study, we addressed the benefits of bariatric surgery in different obesity phenotypes. Both MUMO and MHMO pheno- types showed improvements regarding weight loss and metabol- ic parameters, two years after the surgery. In the fully adjusted regression model, 24 months post-operation, changes in all weight loss and metabolic parameters were similar in patients with both obesity phenotypes, except for changes in WC and
%EWL which were greater in the MHMO. However, differences in these two parameters were not clinically important. Hence, we can conclude that outcomes of bariatric surgery are comparable for patients having either of these two phenotypes.
Due to the lack of an accepted definition for healthy metabolic status among individuals with obesity, its prevalence varies in different population settings, ranging from 6 to 75% [20].
Previous studies in the Middle East reported the prevalence of metabolically healthy status from 13 to 37% in obese individuals [21]. In our study, however, we reported the prevalence of met- abolically healthy individuals among morbid obese patients.
Using JIS definition, MHMO prevalence was found to be 36.2% which was slightly high compared to prior findings.
There are some explanations for this finding; for instance, differ- ences in the study settings (our patients were bariatric surgery candidates), gender and age of study participants, and the diver- sity in definition of metabolic health status.
Since MHMO patients are thought to have a benign health status, the role of interventions for management of obesity and its co-morbidities among them has always been questionable.
Knowledge about comparing the effects of various weight loss strategies among different obesity phenotypes is scarce. Non- surgical methods (such as life style modification and restricted energy diet) are believed to cause similar weight loss in both
healthy and unhealthy obese patients; however, metabolic status and co-morbidity improvements are only detectable in individ- uals with metabolically unhealthy obesity [22,23].
While it is well known that bariatric surgery is one of the best interventions for treatment of morbid obesity and its co-morbid- ities, there is a knowledge gap about its role on the MHMO subgroup; only a few studies with limited samples sizes and length of follow-up have compared the effects of bariatric surgery on MUMO and MHMO patients [14,15,24,25]. These studies differ in study settings and sample size, post-surgical assess- ments, type of bariatric surgery performed, and definition of MHMO phenotype; therefore, their results could vary with each other and with our findings as well. In 2014, Goday et al.
assessed the improvements of cardiovascular risk factors in MHMO patients who underwent bariatric surgery [24]. In a pro- spective cohort study, bariatric surgery (RYGB or SG) was per- formed on 42 MHMO and 180 MUMO patients. One year after surgery, both phenotypes showed significant improvements in all risk factors. However, they concluded that metabolic changes associated with surgery were less prominent in MHMO than in MUMO subjects, particularly in reduction of glucose, HbA1c, HOMA-IR (homeostasis model assessment for insulin-resis- tance), and triglycerides and increase of HDL-C. In another study conducted in France, 21 MHMO and 81 MHMO patients underwent RYGB and completed the 2-year follow-up [15].
They observed a higher rate of excess BMI loss in the MHMO group after surgery. While all metabolic parameters were signif- icantly improved after 2 years from surgery in the MUMO group, the MHMO group only benefited from changes in HOMA-IR, C-reactive protein (CRP), and TG, whereas FPG and HDL-C remained unchanged. Two more studies have compared the ef- fects of bariatric surgery on insulin-sensitive and insulin-resistant morbidly obese patients, showing that while all morbid obese Table 3 Multivariate linear regression analysisafor delta (Δ) change in different characteristics of study population at 12 and 24 months post-operation
Dependent Independent 12 months post-operation 24 months post-operation
Standardized beta B(SE) Pvalue Standardized beta B(SE) Pvalue
ΔBMI MHMO −0.039 −0.353(0.174) 0.043 −0.029 −0.309(0.334) 0.356
ΔWC MHMO −0.064 −1.572 (0.581) 0.007 −0.117 −3.077 (1.069) 0.004
ΔTG MHMO −0.050 −6.832(2.539) 0.007 −0.041 −5.492 (5.149) 0.287
ΔHDL-C MHMO 0.054 1.387(0.770) 0.072 0.052 1.466 (1.768) 0.408
ΔSBP MHMO 0.002 0.053(0.693) 0.939 −0.015 −0.526 (1.271) 0.679
ΔDBP MHMO −0.034 −0.842(0.593) 0.156 −0.054 −1.404 (1.036) 0.176
ΔFPG MHMO −0.001 −0.088(0.831) 0.915 −0.005 −0.390 (1.928) 0.840
Δ% EWL MHMO 0.124 2.412 (0.562) < 0.001 0.114 3.612 (1.408) 0.011 SEstandard error,BMIbody mass index,WCwaist circumference,TGtriglyceride,HDL-Chigh-density lipoproteins-cholesterol,SBPsystolic Blood pressure,DBPdiastolic blood pressure,FPGfasting plasma glucose,%EWLthe percentage of excess weight loss.
aAdjusted for age, sex, surgery group, and baseline values for each dependent variable
patients gain similar weight loss post-surgery, changes in meta- bolic abnormalities are greater in insulin-resistant subjects [14, 25]. The key finding of these studies [14,15,24,25] was that while both obesity phenotypes benefit from bariatric surgery in terms of weight loss outcomes, in metabolic parameters, MHMO patients gain either equal or less favorable results compared to MUMO patients, which is in contrast to our findings.
Compared to those mentioned studies, our study benefits from a larger sample size and reasonable length of post-surgical fol- low-up. Additionally, we adjusted our results for baseline value of each parameter and other confounders such as age, sex, and methods of bariatric surgery. Another distinction of our study is the different criteria used for defining MHMO, race of our par- ticipants, and gender ratio. In our study, all the participants gained noticeable benefits from bariatric surgery. In both MUMO and MHMO groups, improvements in all the metabolic and weight loss parameters were seen post-operatively (BMI loss, decrease in WC, TG, FPG, SBP, and DBP and also increase in HDL-C and %EWL, with a significantPvalue of trend throughout the follow-up). Our primary GEE analysis (without any adjustments for co-founders) revealed that the absolute changes in WC, BMI, and %EWL were greater in MHMO patients and in TG, HDL-C, DBP, BPP, and FPG were greater in MUMO patients. Regarding changes in the prevalence of metabolic abnormalities, patients with MUMO phenotype experienced a greater decrease in the prevalence of high TG, low HDL-C, and high FPG, and patients with MHMO phenotype experienced a greater decrease in the prevalence of high BP.
Moreover, our further linear regression analysis for delta (Δ) change allowed us to assess the impact of bariatric surgery on the outcomes of different obesity phenotypes, adjusted for differences in baseline values of each parameter, methods of bariatric surgery, age, and gender of participants. We found that 2 years after bariatric surgery, changes in BMI, TG, HDL- C, SBP, DBP, and FPG were independent of obesity pheno- type. For WC and %EWL, MHMO phenotype caused a great- er delta (Δ) change from baseline. However, it is worth men- tioning that while the WC decrease (−3.077 cm) and higher
%EWL (+ 3.612%) detected in MHMO patients are statisti- cally significant, we believe the abovementioned values have no clinical importance. Besides, the greater %EWL observed in MHMO might be due to the higher baseline BMI of MUMO patients; therefore, they need a larger weight loss to accomplish the same %EWL. In contrast to the common belief that severe obese patients who suffer from obesity-related co- morbidities are the best candidates for bariatric surgery [26], we found that weight loss and metabolic outcomes of MHMO subjects undergoing bariatric surgery are similar to MUMO subjects and even better at some aspects (WC reduction and higher %EWL). In our regression analysis, before age adjust- ment, we observed a remarkably greater benefit of surgery in MHMO patients compared to MUMO (in five parameters:
BMI, WC, TG, DBP, and %EWL. Data not shown).
However, this finding disappeared after age adjustment, sug- gesting that lower pre-operative age might be associated with achieving more benefits from bariatric surgery, since MHMO patients were significantly younger than MUMO patients.
We are aware that our study has some limitations. We did not assess any inflammatory serum markers (such as CRP) and insulin resistance (by HOMA-IR) in our participants, while we know that a more favorable inflammatory status and insulin sensitivity are positively associated with metabolic health in individuals with obesity, and they are also used as criteria for defining the MHMO phenotype [3,27]. Another limitation of our study was using different methods of bariatric surgery (SG or GB) with possibly different outcomes on par- ticipants. However, we overcame this pitfall by making sure that the number of patients in both phenotype groups which underwent each bariatric surgery method at baseline was sta- tistically the same, and we also adjusted our results for this confounder by linear regression analysis. Finally, our follow- up rate of 65.3% at the 24th month was not favorable.
Despite these limitations, our study has some noteworthy strengths too. To the best of our knowledge, this study was the first prospective cohort with a relatively large sample size to assess the outcomes of bariatric surgery on different obesity phenotypes. Our follow-up period was reasonably long com- pared to similar studies. We also used nationally based cut- offs for definition of excess WC. Another advantage of our study was the use of GEE analysis for statistical modeling, which is an advanced method for handling missing data. We also benefited from regression analysis to assess the role of obesity phenotype on 1- and 2-year delta (Δ) change of each metabolic and weight loss variable adjusted for confounders (differences in baseline values of each variable between phe- notypes, bariatric surgery method, age and gender).
In conclusion, bariatric surgery is an effective method for reduction of metabolic abnormalities and weight loss in both MUMO and MHMO obesity phenotypes. Although individ- uals with MHMO phenotype are considered healthy in regard of many metabolic disorders, the benefits they gain from this intervention is considerable and comparable to individuals with MUMO phenotype. Considering similar responses of healthy and un-healthy morbid obese patients to bariatric sur- gery, we believe that awareness of healthy or unhealthy status should not play an important role while we are making deci- sion for performing bariatric surgery. Taking into account the importance of post-surgical patient care and assessment in bariatric surgery outcomes, we recommend future observa- tional studies with designs to fully match the two obesity phenotypes in terms of medical care, frequency of checkup visits, dietary intake, physical activity, etc. Moreover, in order to assess the impact of bariatric surgery on various outcomes of MUMO and MHMO patients, randomized control trials (RCTs) comparing different bariatric surgery methods, strati- fied by obesity phenotype status, are needed.
AcknowledgmentsThe authors would like to thank the hospital staff, study assistants, and coordinators that took part in this research. The authors would also like to thank Dr. Forough Ghanbari for critically editing English grammar and syntax of the manuscript.
Funding information This work was funded by the Research Institute for Endocrine Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran.
Compliance with Ethical Standards
Conflict of Interest The authors declare that they have no conflict of interest.
Statement of Informed Consent Informed consent was obtained from all individual participants included in the study.
Human Rights/Ethical Approval This study has been approved by the Human Research Review Committee of the Endocrine Research Center, Shahid Beheshti University of Medical Sciences, No. 2ECRIES 93/03/
13.
References
1. Van Gaal LF, Mertens IL, De Block CE. Mechanisms linking obe- sity with cardiovascular disease. Nature. 2006;444(7121):875–80.
2. Karelis AD. Metabolically healthy but obese individuals. Lancet (London, England). 2008;372(9646):1281–3.
3. Phillips CM. Metabolically healthy obesity: definitions, determi- nants and clinical implications. Reviews in endocrine & metabolic disorders. 2013;14(3):219–27.
4. Jung CH, Lee WJ, Song KH. Metabolically healthy obesity: a friend or foe? The Korean journal of internal medicine.
2017;32(4):611–21.
5. Phillips CM. Metabolically healthy obesity across the life course:
epidemiology, determinants, and implications. Annals of the New York Academy of Sciences. 2017;1391(1):85–100.
6. Guh DP, Zhang W, Bansback N, et al. The incidence of co- morbidities related to obesity and overweight: a systematic review and meta-analysis. BMC public health. 2009;9:88.
7. Mingrone G, Panunzi S, De Gaetano A, et al. Bariatric surgery versus conventional medical therapy for type 2 diabetes. The New England journal of medicine. 2012;366(17):1577–85.
8. McTigue KM, Harris R, Hemphill B, et al. Screening and interven- tions for obesity in adults: summary of the evidence for the U.S.
Preventive Services Task Force. Annals of internal medicine.
2003;139(11):933–49.
9. Picot J, Jones J, Colquitt JL, et al. The clinical effectiveness and cost- effectiveness of bariatric (weight loss) surgery for obesity: a system- atic review and economic evaluation. Health technology assessment (Winchester, England). 2009;13(41):1–190. 215-357, iii-iv 10. Stelmach-Mardas M, Walkowiak J. Dietary interventions and
changes in cardio-metabolic parameters in metabolically healthy obese subjects: a systematic review with meta-analysis. Nutrients.
2016;8:8.
11. Athyros VG, Tziomalos K, Karagiannis A, et al. Cardiovascular benefits of bariatric surgery in morbidly obese patients. Obesity reviews : an official journal of the International Association for the Study of Obesity. 2011;12(7):515–24.
12. Karelis AD, Messier V, Brochu M, Rabasa-Lhoret R. Metabolically healthy but obese women: effect of an energy-restricted diet.
Diabetologia. 2008;51(9):1752-4.
13. Kantartzis K, Machann J, Schick F, et al. Effects of a lifestyle intervention in metabolically benign and malign obesity.
Diabetologia. 2011;54(4):864–8.
14. Sesti G, Folli F, Perego L, et al. Effects of weight loss in metabol- ically healthy obese subjects after laparoscopic adjustable gastric banding and hypocaloric diet. PloS one. 2011;6(3):e17737.
15. Pelascini E, Disse E, Pasquer A, et al. Should we wait for metabolic complications before operating on obese patients? Gastric bypass outcomes in metabolically healthy obese individuals. Surgery for obesity and related diseases : official journal of the American Society for Bariatric Surgery. 2016;12(1):49–56.
16. Barzin M. Hosseinpanah F. Bariatric surgery for morbid obesity:
Tehran Obesity Treatment Study (TOTS) rationale and study de- sign. 2016;5(1):e8.
17. Gösta S. Physical status: the use and interpretation of anthropome- try. WHO Technical Report Series. Acta Paediatrica. 1997;86(3):
280.
18. Alberti KG, Eckel RH, Grundy SM, et al. Harmonizing the meta- bolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention;
National Heart, Lung, and Blood Institute; American Heart A s s o c i a t i o n ; Wo r l d H e a r t F e d e r a t i o n ; I n t e r n a t i o n a l Atherosclerosis Society; and International Association for the Study of Obesity. Circulation. 2009;120(16):1640–5.
19. Delavari A, Forouzanfar MH, Alikhani S, et al. First nationwide study of the prevalence of the metabolic syndrome and optimal cutoff points of waist circumference in the Middle East: the national survey of risk factors for noncommunicable diseases of Iran.
Diabetes care. 2009;32(6):1092–7.
20. Rey-Lopez JP, de Rezende LF, Pastor-Valero M, et al. The preva- lence of metabolically healthy obesity: a systematic review and critical evaluation of the definitions used. Obesity reviews : an official journal of the International Association for the Study of Obesity. 2014;15(10):781–90.
21. Lin H, Zhang L, Zheng R, et al. The prevalence, metabolic risk and effects of lifestyle intervention for metabolically healthy obesity: a systematic review and meta-analysis: a PRISMA-compliant article.
Medicine. 2017;96(47):e8838.
22. Karelis AD, Messier V, Brochu M, et al. Metabolically healthy but obese women: effect of an energy-restricted diet. Diabetologia.
2008;51(9):1752–4.
23. Arsenault BJ, Cote M, Cartier A, et al. Effect of exercise training on cardiometabolic risk markers among sedentary, but metabolically healthy overweight or obese post-menopausal women with elevated blood pressure. Atherosclerosis. 2009;207(2):530–3.
24. Goday A, Benaiges D, Parri A, et al. Can bariatric surgery improve cardiovascular risk factors in the metabolically healthy but morbid- ly obese patient? Surgery for obesity and related diseases : official journal of the American Society for Bariatric Surgery. 2014;10(5):
871–6.
25. Jimenez A, Perea V, Corcelles R, et al. Metabolic effects of bariatric surgery in insulin-sensitive morbidly obese subjects. Obesity sur- gery. 2013;23(4):494–500.
26. Gastrointestinal surgery for severe obesity. Consensus statement National Institutes of Health Consensus Development Conference. 1991;9(1):1-20.
27. Phillips CM, Perry IJ. Does inflammation determine metabolic health status in obese and nonobese adults? The Journal of clinical endocrinology and metabolism. 2013;98(10):E1610–9.
Publisher’s NoteSpringer Nature remains neutral with regard to jurisdic- tional claims in published maps and institutional affiliations.
