Ten-Year Weight Regain after Bariatric Surgery: Prevalence, Predictors, and Metabolic Impact

Abstract

Introduction

Bariatric surgery (BS) is an effective intervention for severe obesity, but long-term weight regain (WR) can occur and may impact comorbidity outcomes.

Methods

In this retrospective study, we analyzed patients who underwent BS with a 10-year follow-up. WR was calculated as the percentage of maximum weight loss (%MWL) that had been regained from the nadir weight. Patients were categorized as “Maintainers” (WR ≤20% of %MWL) or “Regainers” (WR >20%).

Results

Among the 353 included patients, 317 underwent Roux-en-Y gastric bypass (RYGB) and 36 underwent sleeve gastrectomy (SG), 90.4% were female, with a mean age of 42 ± 11 years and a mean body mass index of 44.6 kg/m². Mean WR at 10 years was 28% (±25), higher in SG vs. RYGB (41.37% vs. 26.17%, p = 0.03). Overall, 56.7% were Regainers. Baseline type 2 diabetes mellitus (T2D) was associated with an approximately 40% reduced risk of WR >20% (OR = 0.59; 95% CI: 0.38–0.93; p = 0.023). WR was not significantly associated with the recurrence or remission of T2D, hypertension, or dyslipidemia.

Conclusion

WR is common a decade after BS, particularly following SG. While WR does not appear to significantly impact comorbidity recurrence, its clinical relevance warrants further study. Standardized definitions are urgently needed to guide long-term management.

Introduction

Obesity is one of the most prevalent chronic conditions worldwide, affecting over 1 billion individuals according to the World Health Organization (WHO), with this number expected to continue rising [1]. Bariatric surgery (BS) is considered the most effective long-term treatment for severe obesity, and the number of procedures has increased globally over recent decades [2]. In addition to promoting sustained weight loss, BS has been shown to improve or resolve obesity-related comorbidities, such as type 2 diabetes mellitus (T2D), hypertension (HTN), and dyslipidemia, reduce overall mortality, and lower the risk of future cardiovascular events [3].

However, weight regain (WR) is a well-recognized concern following BS and can occur as early as the first few postoperative years. A considerable proportion of patients are estimated to experience significant WR after reaching their nadir weight, although reported prevalence varies widely depending on the definition used and the duration of follow-up [4–6]. WR has been associated with the recurrence or worsening of comorbidities, including T2D and HTN, as well as with reduced quality of life [7–9]. In some cases, revisional BS may be considered; however, these procedures are associated with higher complication and mortality rates compared to primary interventions [10].

WR is considered a multifactorial phenomenon, influenced by behavioral, hormonal, and anatomical factors [4, 11–14]. Although it has been investigated in several studies with short- and medium-term follow-up, the long-term prevalence of WR, along with associated predictors and impact on obesity-related comorbidities, remains insufficiently characterized [15]. Recognizing the potential of WR to undermine the metabolic benefits of BS, we aimed to assess its long-term prevalence, identify associated predictors, and evaluate its relationship with the recurrence of obesity-related comorbidities 10 years after surgery in a real-world cohort.

Materials and Methods

Study Design and Participants

We conducted a retrospective observational study with a 10-year follow-up, including adult patients with obesity who were followed-up by the Multidisciplinary Group for Surgical Obesity Management at our center and underwent BS. Criteria for BS were body mass index (BMI) ≥40 kg/m² or BMI between 35 and 40 kg/m² with obesity-related comorbidities.

Our BS program, established in 2008, included Roux-en-Y gastric bypass (RYGB), sleeve gastrectomy (SG), and gastric banding surgeries. At the time of the surgeries included in this study, our hospital was the only nationally designated reference center in the country. All RYGB procedures used a standardized technique: a 70-cm biliopancreatic limb, a 120-cm alimentary limb, and an 8-cm gastric pouch calibrated with a 36-F Foucher tube. SG was performed with a 46-F bougie, starting approximately 6 cm from the pylorus. Postoperatively, patients had scheduled visits at 6 months, 1 year, and annually up to 3 years, followed by routine annual follow-up with their general practitioner or hospital-based care if comorbidities. Fasting blood samples were collected before each appointment.

A total of 460 patients underwent SG or RYGB at our center between January 2010 and May 2013 and had complete baseline data. Of these, 31 patients were excluded for having undergone revisional procedures, such as conversion of SG to RYGB, conversion of SG to duodenal switch, or conversion of RYGB to distal gastric bypass with shortening of the common channel. The remaining 429 patients underwent primary BS, either laparoscopic SG or RYGB, with no cases of gastric banding included.

At the end of the study period, 76 patients were lost to follow-up, resulting in a final study population of 353 patients. The 10-year overall mortality rate was 1.1%.

All participants were assessed preoperatively, and clinical and anthropometric data were collected at baseline and at 1, 2, 3, 4, and 10 years after surgery to evaluate weight evolution and the long-term metabolic effects of BS. Only patients with complete and documented baseline data and available 10-year follow-up were included.

Clinical outcome data were obtained from electronic medical records maintained by our hospital or from documentation provided by patients’ primary care physicians, with whom our center maintains structured communication channels to retrieve follow-up clinical data when necessary. Variables extracted at each time point included body weight, BMI, diabetes status, blood pressure, lipid profile, and use of antidiabetic, antihypertensive, and lipid-lowering medications.

All procedures involving human participants were conducted in accordance with the ethical standards of the Institutional and/or National Research Ethics Committee and with the principles outlined in the Declaration of Helsinki (1964) and its later amendments.

Definition of WR

WR was defined as the percentage of the maximum weight loss (%MWL) that had been regained at the 10-year follow-up. It was calculated using the formula: 100 × (weight at 10 years − nadir weight)/(preoperative weight − nadir weight) [9]. This approach quantifies the proportion of the initial weight loss (from the preoperative period to the lowest postoperative weight or nadir) that was subsequently regained. Based on this metric, patients were classified into two categories: “Maintainers” defined as those who regained ≤20% of the weight lost and “Regainers” defined as those who regained >20% [9]. %MWL was calculated as the highest relative weight loss from baseline observed at follow-up years. As a sensitivity analysis, we also explored an alternative threshold for WR (≥10% of %MWL), to evaluate whether baseline predictors remained consistent across different definitions.

Definition of Baseline Comorbidities

T2D was defined by fasting plasma glucose ≥126 mg/dL, glycated hemoglobin ≥6.5%, 2-h plasma glucose after a 75-g oral glucose tolerance test ≥200 mg/dL, or the use of antihyperglycemic medications [16]. HTN was defined as systolic blood pressure ≥140 mm Hg, diastolic BP ≥90 mm Hg, or the use of antihypertensive medications [17].

Dyslipidemia was defined by the use of lipid-lowering agents, serum low-density lipoprotein cholesterol (LDL-C) ≥160 mg/dL, serum high-density lipoprotein cholesterol (HDL-C) <40 mg/dL, or triglycerides (TG) ≥200 mg/dL [18, 19]. For subgroup analysis, dyslipidemia was further categorized into: isolated LDL-C abnormality (LDL-C ≥160 mg/dL with normal HDL and TG levels), isolated HDL-C abnormality (HDL <40 mg/dL with normal LDL and TG), and isolated hypertriglyceridemia (TG ≥200 mg/dL with normal LDL and HDL levels) [19] – to explore their distinct metabolic profiles in relation to WR. T2D remission was defined by glycated hemoglobin <5.7% or fasting plasma glucose <100 mg/dL in the absence of diabetes pharmacologic therapy [20].

The presence of T2D, HTN, and dyslipidemia prior to surgery was retrieved and confirmed through review of patients’ electronic medical records. Patients were either previously diagnosed or newly diagnosed during preoperative screening based on internationally accepted criteria [21]. At 10-year follow-up postoperatively, comorbidity status was assessed using the American Society of Metabolic and Bariatric Surgery (ASMBS) guidelines [22].

Statistical Analysis

Categorical variables are presented as absolute frequencies and percentages. Continuous variables are expressed as mean and standard deviation (SD) or as median and interquartile range (IQR) when appropriated.

Comparisons of continuous variables between two groups were performed using the Student’s t test or one-way ANOVA when more than two groups were involved. Categorical variables were compared using the chi-squared test or Fisher’s exact test, as appropriate.

Logistic regression analysis was used to assess associations between the studied variables and the occurrence of WR >20%. A two-sided p value <0.05 was considered statistically significant.

Results

Baseline Population Characteristics

The baseline characteristics of the study population are summarized in Table 1. Among the 353 patients analyzed, 317 (89.8%) underwent RYGB and 36 (10.2%) underwent SG. The cohort was predominantly female (90.4%), with a mean age of 42 ± 11 years and a mean BMI of 44.6 kg/m². At the time of surgery, approximately one-third of patients had T2D (31.4%), 45.2% had dyslipidemia, and 62.2% had HTN. The mean follow-up duration was 9.9 ± 0.79 years.

Table 1.

Baseline clinical characteristics of study population (n = 353)

Females	319 (90.4)
Age, years	42±11
Type of surgery
SG	36 (10.2)
RYGB	317 (89.8)
Weight, kg	116±16
BMI, kg/m2	44.6±5.3
Education
<4 years	73 (24.3)
5–12 years	189 (62.8)
>12 years	39 (13.0)
Waist circumference, cm	123±12
Hip circumference, cm	133±11
T2D	111 (31.4)
Dyslipidemia	150 (45.2)
HTN	204 (62.2)
Laboratorial tests
HbA1c	5.9±1.05
LDL, mg/dL	129±33
HDL, mg/dL	50±11
Triglycerides, mg/dL	135±62
Medication
Antidepressants	100 (28.7)
Benzodiazepines	62 (67.8)

SG, sleeve gastrectomy; RYGB, Roux-en-Y gastric bypass; BMI, body mass index; T2D, type 2 diabetes mellitus; HTN, hypertension; HbA1c, hemoglobin A1c; LDL, low-density lipoprotein; HDL, high-density lipoprotein.

Values are presented as medians ± SD or numbers (%).

Online supplementary Table 1 (for all online suppl. material, see https://doi.org/10.1159/000549498) compares baseline characteristics of 353 patients who completed follow-up vs. 76 patients lost to follow-up. Patients lost to follow-up more often underwent SG, were more frequently male, and had higher baseline body weight, despite a similar BMI, compared to included patients.

Comparison of Baseline Characteristics by WR Classification

Table 2 compares baseline characteristics between patients categorized as Maintainers and Regainers. Patients in the Regainers group presented a nonsignificant higher baseline weight (117 vs. 114 kg, p = 0.076) and BMI (45 vs. 44 kg/m², p = 0.067) compared to Maintainers. However, the Regainers group had significantly lower prevalence of T2D (26.5% vs. 37.9%, p < 0.05), lower baseline HbA1c levels (5.78% vs. 6.10%, p = 0.01), and less frequent use of metformin (19.6% vs. 32.0%, p < 0.01).

Table 2.

Comparison between WR groups at baseline

	Maintainers WR ≤20% MWL, n = 153	Regainers WR >20% MWL, n = 200	p value
Type of surgery
SG	12 (7.8)	24 (12.0)	0.200
RYGB	141 (92.2)	176 (88.0)	​
Age, years	43±11	41±11	0.180
Sex	​	​	0.530
Female	140 (91.5)	179 (89.5)	​
Male	13 (8.5)	21 (10.5)	​
Education	​	​	0.420
<4 years	36 (27.1)	37 (22.0)	​
5–12 years	78 (58.6)	111 (66.1)	​
>12 years	19 (14.3)	20 (11.9)	​
Initial weight, kg	114±16	117±16	0.076
BMI, kg/m2	44±5.1	45±5.4	0.067
Waist circumference, cm	122±11	123±12	0.410
Hip circumference, cm	133±10	134±12	0.600
Comorbidities
T2D	58 (37.9)	53 (26.5)	0.022
HTN	96 (65.8)	108 (59.3)	0.230
Dyslipidemia	72 (50.0)	78 (41.5)	0.120
Laboratorial tests
HbA1c	6.1±1.22	5.78±0.89	0.010
Glucose, mg/dL	104±38	96±26	0.034
LDL, mg/dL	131±35	127±31	0.210
HDL, mg/dL	51±11	49±11	0.350
Triglycerides, mg/dL	138±58	138±64	0.790
Medication
Metformin	49 (32.0)	39 (19.6)	0.008
Antidepressants	42 (29.1)	56 (28.3)	0.860
Benzodiazepines	23 (15.1)	39 (19.7)	0.280

WR, weight regain; MWL, maximum weight loss; SG, sleeve gastrectomy; RYGB, Roux-en-Y gastric bypass; BMI, body mass index; T2D, type 2 diabetes mellitus; HTN, hypertension; HbA1c, hemoglobin A1c; LDL, low-density lipoprotein; HDL, high-density lipoprotein. Bold values highlight results that reached statistical significance, defined as p < 0.05.

Values are presented as medians ± SD or numbers (%) unless otherwise stated.

Longitudinal Weight Trajectory over 10 Years According to WR Classification

At the 10-year follow-up, the mean WR was 28% (±25), with 56.7% of patients experiencing WR greater than 20%. Figure 1 illustrates the evolution of weight over the 10-year postoperative period, stratified by WR classification. Both groups experienced substantial weight loss within the first 2 years, reaching nadir weights of 77 kg in the Maintainers group and 78 kg in the Regainers group. The Maintainers group showed weight stabilization between years 4 and 10 (from 77 kg to 75 kg), while the Regainers group experienced a progressive weight increase during the same period (from 83 kg to 93 kg).

Fig. 1.

Weight by year of follow-up over a 10-year period, stratified by WR category (Maintainers WR ≤20% MWL and Regainers: WR >20% of MWL). NS, nonsignificant; ** p <0.01; ***p < 0.001. MWL, maximum weight loss.

Longitudinal Weight Trajectory over 10 Years and WR According to Type of Surgery

Figure 2a illustrates weight evolution over the 10-year postoperative period, stratified by the type of surgery. From the first postoperative year onward, weight was significantly lower in the RYGB group compared to the SG group. Aligned Rank Transform ANOVA revealed significant effects of time and surgery type on weight (p < 0.05), but with no statistically significant interaction, indicating that the pattern of weight change over time was similar across surgical types (online suppl. Table 2).

Fig. 2.

a Weight trajectories over a 10-year follow-up in patients undergoing SG versus RYGB. b WR at 10-year follow-up in patients undergoing SG versus RYGB. NS, nonsignificant; *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 2b shows the distribution of WR at 10 years, according to surgical technique. Patients who underwent SG had significantly higher WR than those who underwent RYGB (41.37% vs. 26.17%, p = 0.03).

WR According to T2D Status at Baseline

Patients with T2D at baseline experienced less initial weight loss in the early postoperative period (80 ± 13 kg at year 2 vs. 76 ± 13 kg for non-T2D patients, p < 0.05), but no significant difference in absolute body weight was observed between groups at year 10 (85 ± 16 kg vs. 85 ± 17 kg, p = 0.87) (Fig. 3a). Mean WR was significantly lower in T2D patients compared to those without T2D (23.02% vs. 29.88%; p = 0.01) (Fig. 3b).

Fig. 3.

a Evolution of mean body weight from surgery to 10-year follow-up, stratified by the presence of T2D. b WR at 10-year follow-up in patients with T2D at baseline versus those without. NS, nonsignificant; *p < 0.05.

Predictors of WR

Univariate analysis (Table 3) showed no significant association between WR classification and baseline weight, BMI, sex, age, psychotropic medication use, or the presence of HTN or dyslipidemia. In contrast, T2D at baseline emerged as a significant protective factor, associated with a 40% reduced likelihood of being classified as a Regainer (OR = 0.59; 95% CI: 0.38–0.93; p = 0.023). This association remained statistically significant after adjusting for age and %MWL (OR = 0.59; p = 0.037) (Table 4), with greater weight loss also independently associated with lower odds of WR (OR = 0.05; p = 0.037).

Table 3.

Predictors of WR 10 years after BS

​	OR	p value
Sex (female)	1.26 (0.61–2.61)	0.528
Age (years)	0.99 (0.97–1.01)	0.173
BMI preop (kg/m2)	1.04 (1.00–1.08)	0.423
MWL (%)	0.10 (0.01–1.48)	0.094
T2D	0.59 (0.38–0.93)	0.023
Dyslipidemia	0.71 (0.46–1.10)	0.123
HTN	0.76 (0.48–1.19)	0.234
Type of surgery	0.62 (0.30–1.29)	0.204
HbA1c preop	0.74 (0.59 to 0.94)	0.011

WR, weight regain; BS, bariatric surgery; MWL, maximum weight loss; BMI, body mass index; preop, preoperative; T2D, type 2 diabetes mellitus; HTN, hypertension; HbA1c, hemoglobin A1c. Bold values highlight results that reached statistical significance, defined as p < 0.05.

Table 4.

Logistic regression for association of T2D with WR >20% adjusted for age and MWL

​	OR	p value
T2D	0.62 (0.38–1.01)	0.056
MWL	0.05 (0.00–0.84)	0.037
Age (years)	0.99 (0.97–1.01)	0.415

T2D, type 2 diabetes mellitus; WR, weight regain; MWL, maximum weight loss. The bold value highlights a result that reached statistical significance, defined as p < 0.05.

Univariate logistic regression using a 10% cutoff for WR (online suppl. Table 3) showed that baseline T2D tended to be protective, although the association did not reach statistical significance (OR = 0.67; p = 0.114).

Association between WR and Comorbidities

Among the 111 patients with baseline T2D, 93 (83.8%) achieved remission at some point during follow-up, though only 55 (49.5%) remained in remission at year 10. Relapse occurred in 40 patients (43.0% of those who had experienced remission). No significant association was found between WR and T2D remission (mean WR: 22.08% in non-remitters vs. 23.97% in remitters, Fig. 4a). Similarly, WR did not differ significantly between patients who relapsed and those who maintained remission at 10 years (23.39% vs. 23.97%, Fig. 4b).

Fig. 4.

a WR at 10-year follow-up in patients with and without T2D remission. b WR at 10-year follow-up in patients with persistent remission of T2D versus those who experienced a relapse after initial remission.

Among the 204 patients diagnosed with HTN at baseline, 54 (26.5%) were in remission at year 10, while 39 (19.1%) experienced a relapse after an initial remission. WR did not differ significantly between those in remission and those with persistent HTN (mean WR 24.29% vs. 25.91%, respectively, p = 0.68) (Fig. 5a) or between patients with stable remission and those who relapsed (mean WR 24.29% vs. 31.64%, p = 0.15) (Fig. 5b).

Fig. 5.

a WR at 10-year follow-up in patients with and without remission of HTN. b WR at 10-year follow-up in patients with persistent remission of HTN versus those who experienced a relapse after initial remission.

For dyslipidemia, 62 of 150 patients (41.3%) were in remission at year 10 and 38 (25.3%) experienced relapse following initial improvement. No significant differences in WR were observed between patients with persistent remission and those with ongoing dyslipidemia (mean WR 23.71% vs. 26.10%, p = 0.51) (Fig. 6a) or between those who maintained remission and those who relapsed (mean WR 23.71% vs. 28.59%, p = 0.34) (Fig. 6b). For isolated LDL-C, 27 of 29 patients (93.1%) were in remission at year 10 and 2 (6.9%) experienced a relapse. For isolated HDL-C, 83 of 106 patients (78.3%) were in remission at year 10 and 16 (15.1%) experienced a relapse. No remission or relapse events were observed in the isolated hypertriglyceridemia subgroup. No significant differences in mean WR were observed in any of these lipid subgroups.

Fig. 6.

a WR at 10-year follow-up in patients with and without remission of dyslipidemia. b WR at 10-year follow-up in patients with persistent remission of dyslipidemia versus those who experienced a relapse after initial remission.

Association between WR and Metabolic Parameters

Table 5 summarizes the association between WR classification and metabolic parameters. No significant differences were observed between groups in HbA1c, blood pressure, HDL-C, or triglyceride levels. However, LDL-C was slightly but significantly higher in the Regainers group compared to the Maintainers group (101 vs. 95 mg/dL, p < 0.05).

Table 5.

Association between weight and metabolic parameters

​	Maintainers WR ≤20% MWL		Regainers WR >20% MWL		p value (Maintainers vs. Regainers)
	year 0	year 10	year 0	year 10
HbA1c	6.09±1.22	5.78±0.76	5,78±0.89	5.80±0.62	0.911
Systolic BP, mm Hg	132±18	126±16	132±17	129±17	0.180
Diastolic BP, mm Hg	83±12	77±10	82±11	79±11	0.282
HDL, mg/dL	51±11	63±18	49±11	61±17	0.240
LDL, mg/dL	131±35	95±26	127±31	101±27	0.040
Triglycerides, mg/dL	138±58	86±40	136±64	93±34	0.089

WR, weight regain; MWL, maximum weight loss; HbA1c, hemoglobin A1c; BP, blood pressure; LDL, low-density lipoprotein; HDL, high-density lipoprotein.

Bold values highlight results that reached statistical significance, defined as p < 0.05.

Discussion

BS remains the most effective long-term intervention for severe obesity, yielding substantial and sustained weight loss as well as significant metabolic improvements [4, 23, 24]. However, WR remains a persistent concern that may undermine these benefits [25, 26].

A major challenge in studying WR lies in the lack of a universally accepted definition. Definitions vary – ranging from absolute weight regained to percentages of MWL or nadir weight regained – with thresholds typically between 10% and 25% or more [4, 25]. This heterogeneity hampers cross-study comparisons and the identification of consistent predictors and clinical outcomes. In our study, WR was defined as >20% of %MWL, a threshold that has been shown to correlate well with adverse clinical outcomes, including the recurrence of comorbidities and reduced quality of life [9]. Nevertheless, this definition, like others, has inherent limitations. What qualifies as “significant” WR remains debatable: is it based on metabolic relapse, BMI thresholds, or quality of life metrics? These uncertainties highlight the need for a standardized, multidimensional, and clinically grounded definition [4]. To address these issues, we conducted an exploratory sensitive analysis with a 10% cutoff, to assess whether baseline predictors remained consistent across different definitions.

Reported WR prevalence in the literature varies widely – from 3% to 76% – depending on surgical technique, WR metric, and follow-up length [27–30]. Most studies focus on short- or mid-term outcomes, and long-term data (>7–10 years) are scarce. Lauti et al. [9] and King et al. [30] emphasize this gap, calling for longer-term studies with standardized WR metrics.

In our 10-year follow-up study, 56.7% of the cohort experienced WR >20%, with a mean WR of 28%. Patients who underwent SG reached a higher nadir weight and had significantly more WR – averaging 41% of MWL – compared to 26% in the RYGB group. These findings align with those of Cooper et al. [31] who reported a mean WR of 23.4% of MWL at 7 years post-RYGB, with 37% of patients regaining more than 25% of their total weight lost. Similarly, Monaco-Ferreira et al. [32] found that 41% of RYGB patients met the criteria for significant WR (≥15% MWL) at 10 years post-surgery. For SG, the data are even more heterogenous. A systematic review reported WR rates as high as 75% by year 6, depending on the definition used [30]. Although the number of SG in our cohort was small, limiting definitive conclusions, our findings suggest that SG may be associated with both lower initial weight loss and greater long-term WR compared to RYGB. This greater WR may be partly explained by the smaller initial weight loss observed, as well as by anatomical factors such as gastric tube dilation and hormonal adaptations, including higher pre-prandial ghrelin and lower post-prandial GLP-1 levels [10, 29, 30]. This trend underscores the importance of careful surgical decision-making.

We observed that time had a significant effect on WR across both surgical groups, reinforcing the notion that WR is a chronic and progressive phenomenon. Notably, although weight trajectories for Maintainers and Regainers were similar during the first 3 years postoperatively, they began to diverge from the fourth year onward. In our hospital, routine follow-up is typically maintained until the fourth postoperative year, after which patients are discharged to primary care. This pattern suggests that the period starting after the third year may represent a critical window for identifying early signs of WR. As such, enhanced surveillance and earlier intervention during this phase may help prevent more substantial long-term WR. These findings highlight the importance of long-term follow-up extending beyond the commonly reported 5-year timeframe, particularly to detect and address emerging trends in WR.

Despite extensive research, consistent predictors of WR remain elusive, likely reflecting the multifactorial nature of the phenomenon [33–35]. In our cohort, traditionally implicated factors, such as age, baseline BMI, HTN, and HDL-C [36–38], did not emerge as significant predictors. This contrasts with previous findings and may be due to differences in population characteristics, sample size, or surgical distribution. Notably, our patients had a lower baseline BMI and fewer individuals with BMI >50 kg/m² than in earlier reports [36, 37], which may partially explain these discrepancies.

A notable and unexpected finding was that patients with baseline T2D experienced a more stable long-term weight trajectory. Although their initial weight loss was less pronounced – consistent with prior studies [39–41] – they regained significantly less weight over time. This association was significant after adjustment for age and %MWL (p = 0.037). In the sensitivity analysis using a 10% cutoff, a similar but nonsignificant protective trend was observed (p = 0.114). These findings challenge traditional assumptions that T2D predicts poorer surgical outcomes and suggest that despite more modest early weight loss, patients with T2D may be more likely to sustain weight over time. This could reflect several factors, including more structured clinical follow-up, greater adherence to lifestyle interventions, and potentially, pharmacological support. Indeed, medications commonly prescribed for T2D, such as glucagon-like peptide-1 receptor agonists (GLP-1 RAs) and sodium-glucose co-transporter-2 (SGLT2) inhibitors, are known to facilitate weight loss and prevent WR [42]. Although we were unable to systematically capture medication use in our cohort, the increasing adoption of these agents in post-surgical care may partially contribute to the observed differences in weight trajectories. Future studies should prioritize the collection of detailed pharmacological data to clarify these relationships.

In terms of glycemic outcomes, 83.8% of patients with T2D at baseline achieved remission at some point during follow-up, and 47.7% remained in remission at 10 years, similar to the 50% 12-year post-RYGB remission rate reported by Adams et al. [43]. Notably, WR did not significantly differ between those who relapsed and those who maintained remission, suggesting that glycemic durability may be mediated by mechanisms beyond weight trajectory. These may include improved insulin sensitivity, beta-cell preservation, and early metabolic adaptation, supporting the idea of a “metabolic memory effect.”

For HTN, although the limited sample size restricted the strength of our conclusions, we observed no significant association between WR and HTN status at 10 years, despite a trend toward higher WR among those with relapse. This contrasts with previous studies reporting WR as a significant predictor of HTN relapse as early as 3 years postoperatively [42]. Conversely, HTN remission and relapse have been reported to occur at similar rates following SG and RYGB, despite the well-established differences in weight loss outcomes between these procedures [25, 44]. These findings, consistent with our data, suggest that HTN remission may be influenced by weight-independent mechanisms, including reduced systemic inflammation, or neurohormonal regulation [44, 45].

A similar trend was observed in patients with dyslipidemia. Those who experienced relapse had slightly higher WR compared to those with sustained remission, though again, this difference did not reach statistical significance. Similarly, no significant differences in WR were observed across isolated LDL-C and HDL-C categories. Patients with WR >20% also had marginally higher LDL-C levels. These observations suggest a potential link between WR and lipid control, but the relationship appears to be complex and likely mediated by additional factors such as dietary habits and genetic predisposition [25, 43].

To our knowledge, this is one of the few studies to evaluate the prevalence, predictors, and impact of WR on obesity-related comorbidities at 10 years post-BS, in a real-world cohort with a substantial sample size. Our findings confirm that WR is common over the long term but also show that many patients retain comorbidity remission. These results reinforce the idea that WR is just one of several interacting elements shaping long-term outcomes after BS and that metabolic benefits may be sustained through weight-independent mechanisms, such as improved insulin sensitivity or metabolic memory. In line with this, Sjöström [46] demonstrated lasting reductions in cardiovascular events and mortality despite partial WR. These findings advocate for a broader definition of success after BS – one that goes beyond weight change and includes metabolic, functional, and quality-of-life outcomes.

Limitations

This study has several limitations. First, its retrospective design and reliance on clinical records may introduce selection bias and limit the completeness of data, particularly regarding pharmacological therapy and behavioral factors. In addition, our analysis focused primarily on clinical and biochemical parameters, without including anatomical (e.g., pouch dilation, sleeve enlargement), behavioral (e.g., dietary patterns, snacking, physical activity), or psychological (e.g., emotional or loss-of-control eating) determinants, which are known to substantially influence weight trajectories after surgery. Future research should aim to integrate these multidimensional aspects to provide a more comprehensive understanding of WR mechanisms. Another limitation is that loss to follow-up was associated with specific baseline characteristics, namely, male sex and undergoing SG rather than RYGB, which may introduce selection bias and could have led to underestimation of long-term WR. The number of SG cases was small, which limited the power of comparative analyses. Although the 10-year follow-up is a key strength, some loss to follow-up, especially in comorbidity tracking, may have influenced subgroup analyses. The long interval between years 4 and 10, during which no data were collected, may limit the interpretation of both weight and metabolic trajectories. Moreover, patients who underwent revisional surgery were excluded from this study, due to incomplete baseline data and procedural heterogeneity, precluding the identification of potential early regainers, an important subgroup that may follow distinct clinical trajectories and carry predictors of surgical failure. Finally, the generalizability of the results may be limited due to the single-center nature of the study and the predominantly female population.

Conclusion

This study confirms the high prevalence of WR 10 years after BS. While WR is frequently viewed as a key determinant of long-term outcomes, our findings suggest that weight alone may not fully explain the trajectory of comorbidity resolution or relapse.

There is an urgent need for standardized and clinically meaningful definitions of WR to improve research comparability and guide clinical practice. Incorporating body composition data, distinguishing fat from lean mass, may provide more precise insights into the health implications of postoperative weight changes. Further research should explore the role of adjunctive pharmacotherapies, particularly GLP-1 receptor agonists and SGLT2 inhibitors, in sustaining weight loss and metabolic benefits. Lastly, evaluating the relationship between WR and broader outcomes, such as quality of life, morbidity, and mortality, will be essential to define a clinically meaningful WR and to optimize long-term care strategies for bariatric patients.

Statement of Ethics

All procedures performed were in accordance with the ethical standards of the Institutional and/or National Research Committee and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. This study protocol was reviewed and approved by the Ethic Commission of Centro Hospitalar Universitário de São João, Approval No. 165/22. Written informed consent from participants was not required in accordance with the local/national guidelines.

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Funding Sources

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author Contributions

S.R. and P.F. are equally contributing first authors. P.F., S.R., A.L., I.M., J.M., J.G., and H.U.F. designed the study and collected clinical data. B.L. did the statistical analysis. S.R. and P.F. wrote the main manuscript text. T.M., M.B.-C., M.M.S., V.G., J.P., A.V., D.F.S., P.S.F., E.L.C., and J.Q. provided clinical feedback in interpreting the results and contributed critically to subsequent revisions. E.L.C. and J.Q. are equally contributing senior authors. All authors approved the final version of the manuscript.

Funding Statement

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data Availability Statements

All data generated or analyzed during this study are included in this article and its supplementary material files. Further inquiries can be directed to the corresponding author.