Abstract
Objective
The clinical impact of underweight on outcomes in patients undergoing surgical aortic valve replacement (SAVR) remains unclear. This study aimed to investigate the prognostic significance of underweight status in patients undergoing SAVR and to identify key determinants of adverse outcomes within this population.
Methods
805 patients underwent SAVR between January 2016 and December 2021 at eight Japanese centers. After excluding 31 patients with missing laboratory data, a total of 774 patients were included. Patients were categorized into underweight group (Body mass index [BMI] < 18.5 kg/m², 79 patients), normal-weight group (BMI 18.5–25 kg/m², 464 patients), and overweight group (BMI ≥25 kg/m², 231 patients). The primary endpoint was all-cause mortality.
Results
The median age was 76 years (interquartile range [IQR]: 71–79), and the median BMI was 23.3 kg/m² (IQR: 20.9–25.8). Underweight patients had the highest 5-year mortality, followed by normal-weight and overweight patients (23.5%, 11.9%, and 7.4%; p = 0.001). In multivariable analysis, BMI was an independent predictor of mortality (1.0 kg/m² increase, adjusted hazard ratio: 0.92; 95% confidence interval: 0.86 to 0.99; p = 0.020). Among underweight patients, serum albumin was independently associated with all-cause mortality (p = 0.002). Underweight patients were stratified by median albumin into low (< 4.0 g/dL, n = 43) and normal (≥ 4.0 g/dL, n = 36) groups, and 5-year mortality was higher in the low albumin group (34.1% vs. 12.6%, p = 0.015).
Conclusions
Low BMI was associated with higher mortality after SAVR, and serum albumin provided additional risk stratification within underweight group.
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1007/s11748-026-02299-3.
Keywords: Body mass index, Albumin, Aortic stenosis, Surgical aortic valve replacement
Introduction
Underweight (body mass index [BMI] < 18.5 kg/m²) is less common in Western countries, and its prognostic impact after surgical aortic valve replacement (SAVR) has been insufficiently studied. In addition, because the patient profile of underweight individuals may differ in Asia, the clinical significance of underweight in Asian SAVR patients may not be directly inferred from Western data. In patients with severe aortic stenosis (AS), accurate preoperative risk assessment is essential for selecting the most appropriate treatment strategy, [1, 2] as SAVR and transcatheter aortic valve replacement (TAVR) are currently established as the two main therapeutic options. The Society of Thoracic Surgeons Predicted Risk of Mortality (STS-PROM) score is widely applied for operative risk assessment [3, 4]. However, because the STS-PROM model was developed with a valid BMI input range of 18–50 kg/m², BMI values < 18 kg/m² are truncated to 18 kg/m². Consequently, patients with BMI values below this threshold cannot be accurately stratified, and their true operative risk remains uncertain. Among underweight patients, some are frail or malnourished, whereas others are constitutionally small but otherwise healthy, leading to heterogeneous clinical outcomes [5]. Accordingly, the present study aimed to examine the prognostic impact of underweight in patients undergoing SAVR and to identify clinical determinants of outcome within the underweight population.
Materials and methods
Study population
The ACTIVIST registry is a multicenter, retrospective registry that enrolled patients with severe AS who underwent either primary isolated SAVR or TAVR between February 2016 and December 2021 across eight centers in Japan. Patient characteristics, clinical risk factors and comorbidities, echocardiographic data, procedural details, and follow-up outcomes were retrospectively collected in a centralized database. Echocardiographic data were collected before surgery and at discharge. The study protocol was approved by the Institutional Review Board or ethics committee of each participating center. Severe AS was defined as an aortic valve area < 1.0 cm², an indexed aortic valve area < 0.6 cm²/m², a mean transvalvular pressure gradient ≥ 40 mmHg, or a peak aortic jet velocity ≥ 4.0 m/s. The registry initially included 2,811 patients. For the present study, 805 patients who underwent isolated SAVR were selected from this cohort. The study was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the Clinical Research Ethics Committee of Osaka University Hospital (Approval No.: 20222 [T2]; Approval Date: December 15, 2021).
Body mass index and albumin classification
Patients were classified according to the World Health Organization BMI criteria: underweight (< 18.5 kg/m²), normal weight (18.5–25 kg/m²), and overweight/obese (≥ 25 kg/m²). Patients with missing laboratory data (n = 31) were excluded from the analysis. Consequently, a total of 774 patients were included in the final analysis. 79 patients were categorized as “underweight,” 464 patients as “normal weight,” and 231 patients as “overweight.” (Fig. 1).
Fig. 1.
Flowchart depicting the selection of the study population
Endpoints
The primary study endpoint was all-cause death at five years. The secondary endpoints included cardiovascular death and non-cardiovascular death at five years. Both cardiovascular and non-cardiovascular deaths were defined according to the Valve Academic Research Consortium 3 criteria [6].
Statistical analysis
Continuous variables are presented as a median (interquartile range [IQR]), and categorical variables are expressed as counts and percentages. Between-group differences were assessed using one-way analysis of variance or the Kruskal–Wallis test for continuous variables, and the χ2 test or Fisher’s exact test for categorical variables, as appropriate. The cumulative incidence of each endpoint was analyzed using the Kaplan–Meier model, and the differences among the groups were assessed with the log-rank test. Predictors of the primary endpoint were evaluated by Cox proportional hazard regression models. Variables with a p-value < 0.05 in the univariable analysis and those clinically relevant based on previous studies were included in the multivariable analysis. A restricted cubic spline analysis was performed to explore the potential non-linear association between BMI and all-cause mortality in the entire cohort, using the median BMI as the reference. In addition, the same approach was used to examine the association between serum albumin level and all-cause mortality among underweight patients, with 4.0 g/dL as the reference. All data were analyzed using the Statistical Analysis Systems software JMP 18.0 (SAS Institute Inc., Cary, NC, USA) and R software version 4.3.0 (R Foundation for Statistical Computing, Vienna, Austria).
Results
The median age was 76 years (IQR: 71–79), and 52.8% of the patients were female. The median BMI was 23.3 kg/m² (IQR: 20.9–25.8). Patient characteristics are summarized in Table 1. Among the three BMI groups, underweight patients were older and had higher STS-PROM scores, and the prevalence of hypertension, diabetes mellitus, and dyslipidemia was lower in the underweight group. Postoperative echocardiographic findings are shown in Table 2. Among the three BMI groups, there were no significant differences in the incidence of prosthesis–patient mismatch or paravalvular leak.
Table 1.
Baseline patients’ characteristics
| Underweight | Normal weight | Overweight | p value | |
|---|---|---|---|---|
| N = 79 | N = 464 | N = 231 | ||
| Clinical data | ||||
| Age, year | 77 [71–82] | 76.00 [71–80] | 74 [69–78] | 0.002 |
| Female | 49 (62.0) | 228 (49.1) | 132 (57.1) | 0.031 |
| Body surface area, m2 | 1.38 [1.29–1.49] | 1.53 [1.42–1.65] | 1.67 [1.56–1.83] | < 0.001 |
| Body mass index, kg/m2 | 17.0 [15.7–18.0] | 22.4 [20.9–23.7] | 27.7 [26.2–29.7] | < 0.001 |
| STS score, % | 3.36 [1.92–5.08] | 2.95 [1.83–4.82] | 2.12 [1.23–3.22] | < 0.001 |
| NYHA class Ⅲ or Ⅳ | 17 (21.5) | 92 (19.8) | 36 (15.6) | 0.32 |
| Hypertension | 50 (63.3) | 321 (69.2) | 188 (81.4) | < 0.001 |
| Diabetes | 14 (17.7) | 125 (26.9) | 84 (36.4) | 0.003 |
| Hyperlipidemia | 31 (39.2) | 243 (52.4) | 147 (63.6) | < 0.001 |
| Stroke | 4 (5.1) | 44 (9.5) | 16 (6.9) | 0.28 |
| Smoking | 20 (25.3) | 99 (21.3) | 52 (22.5) | 0.72 |
| Coronary artery disease | 4 (5.1) | 58 (12.5) | 30 (13.0) | 0.14 |
| Atrial fibrillation | 10 (12.7) | 79 (17.0) | 22 (9.5) | 0.026 |
| Pacemaker implantation | 3 (3.8) | 16 (3.4) | 2 (0.9) | 0.13 |
| Chronic lung disease ≥moderate | 5 (6.3) | 21 (4.6) | 5 (2.2) | 0.18 |
| Chronic kidney disease | 43 (54.4) | 269 (58.0) | 120 (51.9) | 0.31 |
| eGFR, mL/min/1.73 m2 | 58.4 [40.50–73.4] | 56.4 [41.6–71.4] | 59.3 [49.1–70.1] | 0.12 |
| Albumin, g/dL | 3.90 [3.60–4.20] | 4.00 [3.70–4.20] | 4.00 [3.80–4.30] | 0.001 |
| Ankle Brachial Index | ||||
| Right | 1.09 [1.03–1.16] | 1.10 [1.02–1.16] | 1.09 [1.04–1.16] | 0.85 |
| Left | 1.09 [1.04–1.15] | 1.08 [1.01–1.15] | 1.09 [1.04–1.16] | 0.31 |
| Echocardiographic data | ||||
| LVEF, % | 66 [56–74] | 67 [59–73] | 68 [61–72] | 0.66 |
| AVA, cm2 | 0.64 [0.50–0.77] | 0.72 [0.59–0.85] | 0.79 [0.65–0.89] | < 0.001 |
| AVAi, cm2/m2 | 0.46 [0.37–0.57] | 0.47 [0.38–0.55] | 0.46 [0.40–0.53] | 0.77 |
| MPG, mmHg | 52 [41–63] | 49 [40–61] | 48 [41–57] | 0.55 |
| Peak velocity, m/s | 4.68 [4.22–5.20] | 4.60 [4.20–5.10] | 4.55 [4.20–4.96] | 0.27 |
| TRPG, mmHg | 26 [19–34] | 25 [21–30] | 23 [19–29] | 0.10 |
| LVM, g | 187 [151–228] | 194 [159–236] | 208 [174–249] | < 0.001 |
| LVMi, g/m2 | 128.5 [107.6-158.9] | 126.1 [103.5-151.9] | 124.9 [106.3-143.6] | 0.38 |
| E/A | 0.76 [0.59–0.91] | 0.71 [0.58–0.88] | 0.67 [0.55–0.82] | 0.060 |
| E/e’ | 16.3 [12.0-23.3] | 15.0 [11.8–21.0] | 16.2 [12.4–21.0] | 0.69 |
| moderate to severe AR | 20 (25.3) | 66 (14.2) | 32 (13.9) | 0.031 |
| moderate to severe MR | 8 (10.1) | 34 (7.3) | 13 (5.6) | 0.39 |
| moderate to severe TR | 6 (7.6) | 13 (2.8) | 5 (2.2) | 0.048 |
| Bicuspid valve | 18 (22.8) | 111 (23.9) | 50 (21.6) | 0.80 |
| Procedural data | ||||
| Valve size ≤ 21 mm | 54 (68.4) | 279 (60.1) | 126 (54.5) | 0.083 |
| Valve type | ||||
| Bioprosthetic valve | 73 (92.4) | 432 (93.1) | 207 (89.6) | 0.28 |
| Mechanical valve | 6 (7.6) | 32 (6.9) | 24 (10.4) | |
| Clamp time, min. | 92 [70–107] | 96 [79–115] | 97 [76–118] | 0.076 |
| Pump time, min. | 128 [103–157] | 135 [113–164] | 139 [112–168] | 0.083 |
Values are n (%), or median [interquartile range]. Bold indicates p < 0.05.
AR, aortic regurgitation; AVA, aortic valve area; AVAi, aortic valve area indexed to body surface area; eGFR, estimated glomerular filtration rate; LVEF, left ventricular ejection fraction; LVM, left ventricular mass; LVMi, left ventricular mass indexed to body surface area; MPG, mean pressure gradient; MR, mitral regurgitation; NYHA, New York Heart Association; STS, Society of Thoracic Surgeons; TR, tricuspid regurgitation; TRPG, tricuspid regurgitation pressure gradient.
Table 2.
Postoperative echocardiographic data and 30-day clinical outcomes
| Underweight | Normal weight | Overweight | p value | |
|---|---|---|---|---|
| N = 79 | N = 464 | N = 231 | ||
| Echocardiographic data | ||||
| LVEF, % | 63 [54–69] | 63 [56–68] | 62 [58–68] | 0.94 |
| MPG, mmHg | 12 [9–15] | 12 [9–15] | 12 [9–15] | 0.44 |
| EOA, cm2 | 1.39 [1.21–1.58] | 1.47 [1.25–1.74] | 1.55 [1.30–1.84] | 0.004 |
| EOAi, cm2/m2 | 0.99 [0.85–1.16] | 0.96 [0.81–1.14] | 0.92 [0.77–1.07] | 0.031 |
| PPM | 16 (25.0) | 119 (31.6) | 68 (37.4) | 0.15 |
| Moderate PPM | 11 (17.2) | 108 (28.6) | 58 (31.9) | |
| Severe PPM | 5 (7.8) | 11 (2.9) | 10 (5.5) | |
| moderate to severe PVL | 1 (1.3) | 1 (0.2) | 0 (0.0) | 0.28 |
| moderate to severe MR | 0 (0.0) | 13 (2.8) | 1 (0.4) | 0.035 |
| moderate to severe TR | 3 (3.9) | 11 (2.4) | 1 (0.4) | 0.12 |
| 30-day clinical outcomes | ||||
| All-cause death | 1 (1.3) | 3 (0.7) | 0 (0.0) | 0.33 |
| Cardiovascular death | 0 (0.0) | 1 (0.2) | 0 (0.0) | 0.72 |
| Stroke | 0 (0.0) | 8 (1.7) | 4 (1.7) | 0.50 |
| Permanent pacemaker implantation | 1 (1.3) | 9 (1.9) | 6 (2.6) | 0.74 |
| Heart failure rehospitalization | 0 (0.0) | 2 (0.4) | 1 (0.4) | 0.84 |
| Reintervention | 0 (0.0) | 0 (0.0) | 0 (0.0) | N/A |
Values are n (%), or median [interquartile range]. Bold indicates p < 0.05
LVEF, left ventricular ejection fraction; MPG, mean pressure gradient; N/A, not applicable; EOA, effective orifice area; EOAi, effective orifice area indexed to body surface area; PPM, prosthesis–patient mismatch; PVL, paravalvular leak; MR, mitral regurgitation; TR, tricuspid regurgitation
Relation between body mass index and clinical outcomes
Restricted cubic spline analysis showed that the hazard ratio for all-cause mortality increased as BMI decreased below the median BMI of 23.3 kg/m² (Fig. 2). In multivariable Cox proportional hazards analyses, BMI as a continuous variable was significantly associated with all-cause mortality (adjusted hazard ratio [HR]: 0.92; 95% confidence interval [CI]: 0.86 to 0.99; p = 0.020). When BMI was categorized into three groups, underweight was associated with higher all-cause mortality compared with normal weight (adjusted HR: 2.42; 95% CI: 1.25 to 4.69; p = 0.016), whereas overweight was not significantly associated with mortality (adjusted HR: 0.91; 95% CI: 0.47 to 1.73; p = 0.77) (Table 3).
Fig. 2.
Restricted cubic spline analysis illustrating the continuous association between body mass index and all-cause mortality. Hazard ratios (solid line) and 95% confidence intervals (shaded area) are shown. The reference body mass index was set at the median value of 23.3 kg/m²
Table 3.
Association between body mass index and all-cause mortality in multivariable Cox proportional hazards analysis
| Univariable | Multivariable (Model 1) | Multivariable (Model 2) | ||||
|---|---|---|---|---|---|---|
| HR (95% CI) | p value | Adjusted HR (95% CI) | p value | Adjusted HR (95% CI) | p value | |
| BMI, per 1.0 kg/m2 | 0.88 (0.83–0.94) | < 0.001 | 0.92 (0.86–0.99) | 0.020 | ||
| BMI category | ||||||
| Normal weight | 1.00 (Reference) | 1.00 (Reference) | ||||
| Underweight | 2.11 (1.16–3.85) | 0.023 | 2.42 (1.25–4.69) | 0.016 | ||
| Overweight | 0.56 (0.30–1.03) | 0.052 | 0.91 (0.47–1.73) | 0.77 | ||
Multivariable analyses were adjusted for age, sex, Society of Thoracic Surgeons score, coronary artery disease, chronic kidney disease, moderate to severe tricuspid regurgitation, and albumin. Model 1; BMI is included in the model as a continuous variable. Model 2; BMI is included in the model as a categorical variable.
Bold indicates p < 0.05. BMI, body mass index; CI, confidence interval; HR, hazard ratio.
Underweight and clinical outcomes
Thirty-day clinical outcomes according to BMI category are summarized in Table 2. The 5-year incidence of all-cause mortality differed significantly across BMI categories, with the highest incidence observed in underweight patients, followed by normal-weight and overweight patients (23.5%, 11.9%, and 7.4%, respectively; log-rank p = 0.001; Fig. 3A). A similar pattern was observed for cardiovascular death, which was highest in underweight patients (8.7%, 4.9%, and 1.5%, respectively; p = 0.036; Fig. 3B), and for non-cardiovascular death (16.1%, 7.3%, and 6.0%, respectively; p = 0.022; Fig. 3C). In a comparison between underweight and normal-weight patients, all-cause mortality (p = 0.013) and non-cardiovascular death (p = 0.030) were significantly higher in underweight patients, whereas cardiovascular death was not significantly different (p = 0.20).
Fig. 3.
Kaplan–Meier survival curves comparing normal-weight and underweight patients. (A) All-cause mortality, (B) non-cardiovascular mortality, and (C) cardiovascular mortality
The distribution of causes of non-cardiovascular death is summarized in Supplemental Figure E1. Although the distribution of causes did not differ significantly between groups (p = 0.19), infection-related death was frequent among underweight patients (87.5%).
Stratification of underweight patients by serum albumin level
Among underweight patients, the median preoperative serum albumin level was 4.0 g/dL. Restricted cubic spline analysis suggested a continuous inverse association between serum albumin and the hazard of all-cause mortality (Fig. 4). In Cox proportional hazards analyses, serum albumin was significantly associated with all-cause mortality in the univariable model (HR: 0.16; 95% CI: 0.04 to 0.60; p = 0.006). Serum albumin remained independently associated with all-cause mortality in both the minimally adjusted model (Model 1; adjusted HR: 0.13; 95% CI: 0.03 to 0.51; p = 0.005) and the more fully adjusted model (Model 2; adjusted HR: 0.15; 95% CI: 0.03 to 0.69; p = 0.014) (Table 4). Underweight patients were additionally stratified by the median albumin level into lower albumin (< 4.0 g/dL, n = 43) and normal albumin (≥ 4.0 g/dL, n = 36) subgroups. The low albumin subgroup had a higher 5-year all-cause mortality than the normal albumin subgroup (34.1% vs. 12.6%, log-rank p = 0.015). No significant difference in 5-year all-cause mortality was observed between underweight patients with normal albumin and normal-weight patients (12.6% vs. 11.9%, log-rank p = 0.76) (Fig. 5).
Fig. 4.
Restricted cubic spline analysis illustrating the continuous association between serum albumin level and all-cause mortality in underweight patients. Hazard ratios (solid line) and 95% confidence intervals (shaded area) are shown. The reference serum albumin level was set at 4.0 g/dL
Table 4.
Preoperative serum albumin and all-cause mortality in underweight patients: multivariable Cox analysis
| Univariable | Multivariable (Model 1) | Multivariable (Model 2) | ||||
|---|---|---|---|---|---|---|
| HR (95% CI) | p value | Adjusted HR (95% CI) | p value | Adjusted HR (95% CI) | p value | |
| Albumin, per 1.0 g/dL | 0.16 (0.04–0.60) | 0.006 | 0.13 (0.03–0.51) | 0.005 | 0.15 (0.03–0.69) | 0.014 |
| Age, per 1 year | 1.02 (0.95–1.11) | 0.63 | 1.03 (0.95–1.13) | 0.45 | 1.05 (0.96–1.17) | 0.27 |
| Female | 1.17 (0.39–3.49) | 0.78 | 1.64 (0.53–5.10) | 0.39 | 1.60 (0.51-5.00) | 0.42 |
| BMI, per 1.0 kg/m2 | 0.77 (0.56–1.11) | 0.16 | 0.66 (0.43–1.02) | 0.060 | ||
| Chronic kidney disease | 7.47 (1.66–33.57) | 0.009 | 5.84 (1.20-28.32) | 0.029 | ||
| LVEF, per 1% | 0.99 (0.96–1.03) | 0.50 | 1.02 (0.98–1.06) | 0.44 | ||
Model 1 was adjusted for age and sex. Model 2 was adjusted for age, sex, BMI, chronic kidney disease, and LVEF.
Bold indicates p < 0.05. BMI, body mass index; CI, confidence interval; HR, hazard ratio; LVEF, left ventricular ejection fraction.
Fig. 5.
Kaplan–Meier curves for all-cause mortality comparing normal-weight patients, underweight patients with low serum albumin (< 4.0 g/dL), and underweight patients with normal serum albumin (≥ 4.0 g/dL)
Discussion
The main findings of this study are summarized as follows: (1) Underweight patients had highest all-cause mortality: (2) Lower BMI was identified as an independent predictor of all-cause mortality: (3) Among underweight subgroup, lower serum albumin was independently associated with a higher risk of all-cause mortality.
BMI is widely recognized as a fundamental indicator of nutritional assessment for evaluating malnutrition risk, and underweight is considered a form of malnutrition, with a BMI < 18.5 kg/m² commonly used as the threshold [7–9]. Large epidemiological studies, including the Japanese pooled cohort and the contemporary U.S. National Health Interview Survey, have both demonstrated a J-shaped association between BMI and all-cause mortality, with increased risk not only with higher BMI but within the underweight range [10, 11] In patients with chronic heart failure, underweight or low BMI was independently associated with a higher risk of all-cause mortality, as reported in the CHARM program analysis [12]. Several studies suggested that underweight patients experience higher postoperative mortality after cardiac surgery [13–19]. However, these studies were primarily conducted in Europe and North America, where the proportion of underweight patients was extremely low, approximately 1% for those with a BMI < 18.5 kg/m², [17,18] and 1.6–5.6% for those with a BMI < 20 kg/m² [13, 14, 16, 19]. In contrast, in our study, 10.2% of patients were classified as underweight (BMI < 18.5 kg/m²), which is roughly tenfold higher than the rate reported in Western populations. Accordingly, the clinical significance of underweight may also differ from that in Western populations.
In patients undergoing TAVR, the clinical impact of underweight has been examined in several studies. The proportion of underweight patients (BMI < 18.5 kg/m²) was reported to be 1.7–2.6% in Western populations, [20–22] whereas it was markedly higher, at 14.3–15.1% in Japanese cohorts [1,23, 24]. All of these studies consistently demonstrated that underweight status was associated with increased mortality, [1, 6, 20–23] and several further suggested that an excess of non-cardiovascular deaths may have contributed to this increased risk [23, 24]. Similarly, in the present study, underweight was associated with approximately threefold higher mortality after SAVR compared with normal-weight patients. Moreover, this excess mortality appeared to result primarily from a higher incidence of non-cardiovascular death. In particular, infection-related death was more frequent among underweight patients, suggesting that impaired nutritional status and frailty may have contributed.
However, some underweight Japanese patients are not frail or malnourished but are constitutionally small and healthy, and such patients may not be at high risk for SAVR. In a further analysis among underweight patients, this study showed that serum albumin level effectively stratified their prognosis. Serum Albumin is the main blood protein that maintains fluid balance, transports substances, and protects against inflammation and clotting, and low levels are often caused by poor nutrition or chronic inflammation [25]. Previous studies have highlighted the prognostic importance of nutritional status in patients undergoing cardiovascular surgery. Lee et al. reported that the prognostic nutritional index, which incorporates serum albumin, was useful for predicting early clinical outcomes after adult cardiac surgery [26]. In addition, Hayashi et al. showed that a low prognostic nutritional index was associated with postoperative complications and worse survival in patients undergoing cardiovascular surgery, including an Asian population similar to that in the present study [27]. Although these studies evaluated the prognostic nutritional index rather than the isolated effect of serum albumin, they support the clinical importance of preoperative nutritional assessment in cardiovascular surgery. Our findings suggest that serum albumin alone may also be useful for risk stratification in underweight Japanese patients undergoing SAVR. Although a serum albumin level of 3.5 g/dL is commonly used as the cutoff for hypoalbuminemia, higher thresholds have also been applied [25, 28]. In the present study, we used the median serum albumin level of 4.0 g/dL as the cutoff to stratify underweight patients. Even among underweight patients, those with preserved serum albumin levels (≥ 4.0 g/dL) exhibited favorable survival outcomes. These findings indicate that a subset of underweight Japanese patients experience favorable outcomes after SAVR. Incorporating serum albumin into preoperative risk assessment could help identify underweight patients most likely to benefit from SAVR. In contrast, underweight patients with low serum albumin showed significantly higher mortality, indicating a particularly high-risk subgroup. In such patients, less invasive approaches, including TAVR and minimally invasive surgical AVR, may be considered depending on overall risk profile, anatomy, and the treating center’s experience [12]. Overall, our findings suggest that serum albumin measurement provides a simple and effective tool for risk stratification in underweight patients undergoing SAVR. Therefore, in underweight patients with severe AS, comprehensive risk evaluation incorporating serum albumin should be considered when determining the optimal treatment strategy.
Study limitation
This study has several limitations. First, it used a retrospective design and is therefore subject to selection bias. Second, the study population consisted only of Japanese patients with relatively small body size, and potential racial or ethnic differences were not assessed. Therefore, the generalizability of these findings to other populations with larger average body sizes may be limited. Third, data regarding body-weight changes within the six months prior to surgery were unavailable. Although weight loss represents an important criterion for nutritional assessment, it is often based on patient self-report, making it difficult to obtain objective documentation. Fourth, BMI reflects overall body size but does not capture body composition, such as muscle mass. Although serum albumin was a useful prognostic marker in our analyses, it can be influenced by fluid balance and acute inflammatory states, including infection. Therefore, BMI and serum albumin should be interpreted cautiously as indicators of nutritional status or frailty. Fifth, the number of events in the underweight subgroup was limited, and therefore the multivariable analyses may still have been affected by overfitting despite careful restriction of the number of covariates.
Conclusion
Underweight was associated with higher mortality after SAVR. Serum albumin provided additional risk stratification within the underweight group, with preserved albumin associated with more favorable outcomes and low albumin identifying a particularly high-risk subgroup.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
The authors would like to thank Kunitaka Kumagai, MD, Shohei Yamada, MD, Yusuke Misumi, MD, PhD, Kizuku Yamashita, MD, PhD, Arudo Hiraoka, MD, PhD, Daisuke Yoshioka, MD, PhD, Kazuo Shimamura, MD, PhD, Haruhiko Kondoh, MD, PhD, and Hidenori Yoshitaka, MD, PhD, for their valuable contributions to data collection, and critical discussions related to this study.
Abbreviations
- BMI
Body mass index
- SAVR
Surgical aortic valve replacement
- AS
Aortic stenosis
- TAVR
Transcatheter aortic valve replacement
- STS-PROM
The Society of Thoracic Surgeons Predicted Risk of Mortality
- IQR
Interquartile range
Funding
Open Access funding provided by The University of Osaka. None.
Declarations
Conflict of interest
TO and KH have received institutional research grants and support from Edwards Lifesciences. KM has received institutional research grants and support from Medtronic and Edwards Lifesciences. KM and KS serve as proctors for the SAPIEN and Evolut valve series.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Maeda K, Kumamaru H, Kohsaka S, Shimamura K, Mizote I, Yamashita K, et al. A risk model for 1-year mortality after transcatheter aortic valve replacement from the J-TVT registry. JACC Asia. 2022;2:635–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Szeto WY, Svensson LG, Rajeswaran J, Ehrlinger J, Suri RM, Smith CR, et al. Appropriate patient selection or health care rationing? Lessons from surgical aortic valve replacement in the placement of aortic transcatheter valves I trial. J Thorac Cardiovasc Surg. 2015;150:557–68. [DOI] [PubMed] [Google Scholar]
- 3.Shahian DM, Jacobs JP, Badhwar V, Kurlansky PA, Furnary AP, Cleveland JC, et al. The society of thoracic surgeons 2018 adult cardiac surgery risk models: part 1-background, design considerations, and model development. Ann Thorac Surg. 2018;105:1411–8. [DOI] [PubMed] [Google Scholar]
- 4.O’Brien SM, Feng L, He X, Xian Y, Jacobs JP, Badhwar V, et al. The society of thoracic surgeons 2018 adult cardiac surgery risk models: part 2-statistical methods and results. Ann Thorac Surg. 2018;105:1419–28. [DOI] [PubMed] [Google Scholar]
- 5.Singh R, Nyamasege CK, Hawks SR, Wagatsuma Y. Health-related lifestyles in relation to body mass index among young and middle-aged women in Japan. Womens Health Rep (New Rochelle). 2022;3:887–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.VARC-3 WRITING COMMITTEE:, Généreux P, Piazza N, Alu MC, Nazif T, Hahn RT, Pibarot P, et al. Valve academic research consortium 3: updated endpoint definitions for aortic valve clinical research. J Am Coll Cardiol. 2021;77:2717–46. [DOI] [PubMed] [Google Scholar]
- 7.Cederholm T, Jensen GL, Correia MITD, Gonzalez MC, Fukushima R, Higashiguchi T, et al. GLIM criteria for the diagnosis of malnutrition - A consensus report from the global clinical nutrition community. Clin Nutr. 2019;38:1–9. [DOI] [PubMed] [Google Scholar]
- 8.Kappetein AP, Head SJ, Généreux P, Piazza N, van Mieghem NM, Blackstone EH, et al. Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve academic research consortiuasm-2 consensus document (VARC-2). Eur J Cardiothorac Surg. 2012;42:S45–60. [DOI] [PubMed] [Google Scholar]
- 9.Dent E, Wright ORL, Woo J, Hoogendijk EO. Malnutrition in older adults. Lancet. 2023;401:951–66. [DOI] [PubMed] [Google Scholar]
- 10.Sasazuki S, Inoue M, Tsuji I, Sugawara Y, Tamakoshi A, Matsuo K, et al. Body mass index and mortality from all causes and major causes in Japanese: results of a pooled analysis of 7 large-scale cohort studies. J Epidemiol. 2011;21:417–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Visaria A, Setoguchi S. Body mass index and all-cause mortality in a 21st century U.S. population: a national health interview survey analysis. PLoS ONE. 2023;18:e0287218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kenchaiah S, Pocock SJ, Wang D, Finn PV, Zornoff LA, Skali H, et al. Body mass index and prognosis in patients with chronic heart failure: insights from the Candesartan in Heart failure: assessment of reduction in mortality and morbidity (CHARM) program. Circulation. 2007;116:627–36. [DOI] [PubMed] [Google Scholar]
- 13.Forgie K, Bozso SJ, Hong Y, Norris CM, Ishaque A, Gill RS, et al. The effects of body mass index on outcomes for patients undergoing surgical aortic valve replacement. BMC Cardiovasc Disord. 2020;20:255. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Voigtländer L, Twerenbold R, Schäfer U, Conradi L, Balaban Ü, Bekeredjian R, et al. Prognostic Impact of Underweight (Body Mass Index < 20 kg/m2) in patients with severe aortic valve stenosis undergoing transcatheter aortic valve implantation or surgical aortic valve replacement (from the German Aortic Valve Registry [GARY]). Am J Cardiol. 2020;129:79–86. [DOI] [PubMed] [Google Scholar]
- 15.Smith RL, Herbert MA, Dewey TM, Brinkman WT, Prince SL, Ryan WH, et al. Does body mass index affect outcomes for aortic valve replacement surgery for aortic stenosis? Ann Thorac Surg. 2012;93:742–6. [DOI] [PubMed] [Google Scholar]
- 16.Johnson AP, Parlow JL, Whitehead M, Xu J, Rohland S, Milne B. Body mass index, outcomes, and mortality following cardiac surgery in Ontario, Canada. J Am Heart Assoc. 2015;4:e002140. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hartrumpf M, Kuehnel RU, Albes JM. The obesity paradox is still there: a risk analysis of over 15000 cardiosurgical patients based on body mass index. Interact Cardiovasc Thorac Surg. 2017;25:18–24. [DOI] [PubMed] [Google Scholar]
- 18.Gao M, Sun J, Young N, Boyd D, Atkins Z, Li Z, et al. Impact of body mass index on outcomes in cardiac surgery. J Cardiothorac Vasc Anesth. 2016;30:1308–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Kang JJ, Bozso SJ, El-Andari R, Alam AS, Boe DE, Hong Y, et al. Relation of body mass index to long-term survival and cardiac remodelling for patients undergoing mitral valve replacement surgery. Nutr Metab Cardiovasc Dis. 2024;34:1371–80. [DOI] [PubMed] [Google Scholar]
- 20.van Nieuwkerk AC, Santos RB, Sartori S, Regueiro A, Tchétché D, Mehran R, et al. Impact of body mass index on outcomes in patients undergoing transfemoral transcatheter aortic valve implantation. JTCVS Open. 2021;6:26–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Sharma A, Lavie CJ, Elmariah S, Borer JS, Sharma SK, Vemulapalli S, et al. Relationship of Body Mass Index With Outcomes After Transcatheter Aortic Valve Replacement: Results From the National Cardiovascular Data-STS/ACC TVT Registry. Mayo Clin Proc. 2020;95:57–68. [DOI] [PubMed] [Google Scholar]
- 22.Madanat L, Jabri A, Hanson ID, Khalili H, Rodés-Cabau J, Pilgrim T, et al. Obesity Paradox in Transcatheter Aortic Valve Replacement. Curr Cardiol Rep. 2024;26:1005–9. [DOI] [PubMed] [Google Scholar]
- 23.Tezuka T, Higuchi R, Hagiya K, Saji M, Takamisawa I, Nanasato M, et al. Midterm Outcomes of Underweight Patients Undergoing Transcatheter Aortic Valve Implantation: Insight From the LAPLACE-TAVR Registry. JACC Asia. 2022;3:78–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Handa K, Maeda K, Pak K, Shimamura K, Inoue K, Yamada S, et al. Severity of thinness amplifies mortality risk in patients with transcatheter aortic valve implantation. Interdiscip Cardiovasc Thorac Surg. 2025;40:120. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Manolis AA, Manolis TA, Melita H, Mikhailidis DP, Manolis AS. Low serum albumin: A neglected predictor in patients with cardiovascular disease. Eur J Intern Med. 2022;102:24–39. [DOI] [PubMed] [Google Scholar]
- 26.Lee SI, Ko KP, Choi CH, Park CH, Park KY, Son KH. Does the prognostic nutritional index have a predictive role in the outcomes of adult cardiac surgery? J Thorac Cardiovasc Surg. 2020;160:145–e1533. [DOI] [PubMed] [Google Scholar]
- 27.Hayashi J, Uchida T, Ri S, Hamasaki A, Kuroda Y, Yamashita A, et al. Clinical significance of the prognostic nutritional index in patients undergoing cardiovascular surgery. Gen Thorac Cardiovasc Surg. 2020;68:774–9. [DOI] [PubMed] [Google Scholar]
- 28.Berbel-Franco D, Lopez-Delgado JC, Putzu A, Esteve F, Torrado H, Farrero E, et al. The influence of postoperative albumin levels on the outcome of cardiac surgery. J Cardiothorac Surg. 2020;15:78. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.