Correlation between serum phosphorus level and muscle mass in patients with type 2 diabetes mellitus Authors: Xiangyu Gao, Meijian Wang, Wenjie Ma, Rui Wang, Wenchao Hu Published: 2026 ABSTRACT This study aimed to evaluate the correlation between serum phosphorus level and low muscle mass in patients with type 2 diabetes mellitus (T2DM). In this cross-sectional study, 1078 T2DM inpatients were recruited. Low muscle mass was defined as a skeletal muscle index (SMI) < 7.0 kg/m2 for men and < 5.4 kg/m2 for women, which measured by dual-energy X-ray absorptiometry. Serum phosphorus (P) was measured using the phosphomolybdate method. Clinical information and other biochemical characteristics were measured and recorded. The prevalence of low muscle mass was 20.2% in men and 13.6% in women. A higher serum phosphorus level was associated with a significantly lower risk of low muscle mass in both genders after multivariable adjustment. A nonlinear, threshold-dependent positive correlation was observed between serum phosphorus and SMI, with significant associations below cohort-specific thresholds of 0.98 mmol/L in men and 1.04 mmol/L in women. Higher serum phosphorus level is associated with a lower prevalence of low muscle mass (i.e., positively correlated with SMI) in T2DM patients. And this exploratory finding requires validation in longitudinal studies.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1038/s41598-026-56772-3. Introduction Sarcopenia is a clinical syndrome characterized by the progressive loss of skeletal muscle mass and function which often be related to aging1. However, recent researches have revealed a rising prevalence of sarcopenia among younger individuals, especially those with obesity, highlighting the need for greater public awareness of its detrimental effects. Sarcopenia is linked to an increased risk of fractures, malnutrition, and movement disabilities, additionally even reported as a vital chronic complication of diabetes mellitus, with which it associated with increased mortality rates and poor long-term prognosis2,3. According to a British meta-analysis, the incidence of sarcopenia was significantly higher in patients diagnosed with type 2 diabetes mellitus (T2DM) (28.4%) than that in the non-diabetic individuals (18.7%)4. Given the adverse consequences of sarcopenia, early identification and prevention are important, especially in patients with diabetes. According to Asian Working Group for Sarcopenia (AWGS), the proposed diagnostic thresholds for assessing low muscle mass include sex-specific cutoffs of 7.0 kg/m2 (male) and 5.4 kg/m2 (female) when measured through dual-energy X-ray absorptiometry (DXA), while bioelectrical impedance analysis yields comparable values of 7.0 kg/m2 for males and 5.7 kg/m2 for females. Complementary functional criteria consist of grip strength thresholds below 26 kg for men and 18 kg for women, along with reduced mobility indicators defined by walking speeds slower than 0.8 m/s5. Currently, DXA are most commonly used clinically to measure muscle mass and indirectly predict sarcopenia. Moreover, screening for novel serum biomarkers represents a promising approach to predict muscle mass and achieve prevention of sarcopenia. It is important to note that, due to the limitations of our dataset, which assessed only skeletal muscle mass index (SMI), the present study focuses specifically on low muscle mass as the outcome of interest. Phosphorus (P) is a critical macromineral that plays vital roles in various biological functions in human beings. Notably, P plays a vital role in skeletal mineralization and energy metabolism, with its deficiency leading to a significant disruption of the musculoskeletal system6. Phosphorus-containing metabolites are essential components of cellular structures and functions, serving as critical elements in the formation of cell membranes and nucleic acid molecules7. Besides, phosphorus-containing metabolites also play a vital role in metabolic processes and energy metabolism, particularly through the phosphorylation of intermediate metabolites and the storage of energy released in high-energy phosphate bonds, such as those in ATP or phosphocreatine, during oxidative phosphorylation8,9. In skeletal muscles, furthermore, P is essential for the provision of high-energy phosphates required for the contractile activity and structural integrity of the muscular membrane and intracellular organelles, including the sarcoplasmic and mitochondrial reticulum10. These roles suggest that the level of P in the serum may correlate to its level in the skeletal muscle, thereby serving a vital biomarker for predicting and evaluating sarcopenia. Previous studies examining the relationship between serum phosphorus level and muscle-related outcomes have reported conflicting results. For instance, a study in hemodialysis patients found lower serum phosphorus levels in those with low muscle mass 11, whereas research in elderly patients with T2DM reported that higher serum phosphorus was associated with weaker grip strength 12. Furthermore, a systematic review concluded that evidence for an association between phosphorus and sarcopenia in older adults was insufficient due to a limited number of studies 13. Therefore, the relationship between phosphorus and muscle mass specifically in T2DM patients remains underexplored, and the available findings are largely inconsistent. Thus this study was performed to evaluate the correlation of serum P level and low muscle mass in patients with T2DM, in order to explore the potential targets for the prediction and prevention of sarcopenia. Methods Study population In total, 1078 consecutive inpatients (including 579 males and 499 females) from the Department of Endocrinology of Qilu hospital (Qingdao), Shandong University were recruited from September 2017 to September 2019 in this cross-sectional study. All participants were clinically diagnosed with T2DM based on the criteria by the American Diabetic Association, with a fasting plasma glucose (FPG) ≥ 7.0 mmol/L, a 2-h postprandial plasma glucose ≥ 11.1 mmol/L, or both. Patients were excluded from the study if they were pregnant, undergoing anti-tumor therapy, characterized by severe hepatic and renal failures, experiencing severe bone disease, suffering from acute infectious diseases, and experiencing thyroid and parathyroid dysfunctions. Detailed information on the specific types and dosages of anti-hyperglycemic agents was not systematically collected in this study. Of note, the study period (September 2017 to September 2019) predates the widespread use of newer agents (e.g., GLP-1 receptor agonists and SGLT2 inhibitors) in the local region, which may partially reduce the potential confounding effect of these medications. This study was approved by the ethics committees of Qilu Hospital (Qingdao) of Shandong University, additionally, written informed consent were obtained from all participants. Low muscle mass definition Skeletal muscle index (SMI) was measured using the dual-energy X-ray absorptiometry (DXA, Hologic Discovery A, Waltham, MA, USA). Subsequently, SMI was calculated as appendicular skeletal muscle mass in kilograms divided by the square of the body height in meters (kg/m2). Low muscle mass was defined as SMI < 7.0 kg/m2 in males and SMI < 5.4 kg/m2 in females, based on the consensus by the Asian Working Group for Sarcopenia5. Basic characteristics and biochemical measurements Information on basic epidemiological characteristics, including age, sex, height, weight, blood pressures, and the duration of diabetes were collected and recorded. Body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters (kg/m2), with patients having BMI ≥ 28 was regarded as obesity. Additionally, DXA, previously mentioned, was used to assist in categorizing patients as either ANDROID or GYNOID, which are different types of obesity. Fasting blood samples were collected from all participant and various biochemical measurements were quantified, including serum albumin to globulin ratio (A/G), FPG, total cholesterol (TC), high-density lipoprotein cholesterol (HDL), low-density lipoprotein cholesterol (LDL), triglyceride (TG), serum calcium (Ca), and P. Specifically, serum phosphorus concentration was measured using the phosphomolybdate ultraviolet method on an automated biochemical analyzer. The assay has a detection range of 0.10 – 6.00 mmol/L and a sensitivity of 0.01 mmol/L. Statistical analysis Quantitative data following a normal distribution were presented as either mean ± standard deviation (SD) or median (25th percentile, 75th percentile), while categorical data following a normal distribution were summarized using proportions or percentages of patients. Student’s t-test was used for parametric data, whereas the Mann–Whitney U-test or chi-square test was used for non-parametric data to compare the differences in variables between normal muscle mass and less muscle mass groups. The relationships between the two groups were assessed using the Spearman correlation analysis. Moreover, binary logistic regression analysis was used to determine the predictors for low muscle mass after adjusting for potential confounding variables. Besides, serum phosphorus concentration were categorized into sample-specific quartiles (Q1–Q4) based on the distribution within the study cohort to explore potential non-linear associations. Additionally, the independent association between the serum P level and SMI was assessed using linear regression, with a smooth curve fitting used to explore this association. A multivariate piecewise linear regression was further used to examine the threshold correlation of the serum P level and SMI based on to the smooth curve fit. It is important to note that the inflection points identified by this method represent exploratory, data-driven estimates specific to our study cohort. They should not be considered universal biological thresholds without external validation in independent populations. All P values < 0.05 were considered to be statistically significant. To control for multiple comparisons, we applied the Benjamini–Hochberg False Discovery Rate (FDR) correction (with a nominal FDR of q < 0.05) to all 130 p-values from the primary and exploratory analyses. Results were considered statistically significant if the FDR-adjusted q-value remained below 0.05 (Supplementary Table S1). All analyses were performed using SPSS Version 26.0 (IBM Corp., Armonk, NY, USA). Results The differences of characteristics between patients with low and normal muscle mass A total of 1078 T2DM patients, including 579 males and 499 females were recruited into the study, as indicated in Table 1. Among these participants, 117 males (20.2%) and 68 females (13.6%) had low muscle mass. Moreover, among the male subjects, there tended to be higher age, duration of diabetes, HDL as well as lower A/G ratio, ANDROID, SMI, prevalence of obesity, height, weight, BMI, DBP, TG and serum phosphorus in low muscle mass group compared to normal muscle mass group. While in female subjects, age was significantly higher, while A/G ratio, ANDROID, SMI, prevalence of obesity, height, weight, BMI, DBP, TG, and serum P were significantly lower in the low muscle mass group compared to the normal muscle mass group.Table 1The clinical characteristics differences between T2DM patients with and without low muscle mass.CharacteristicsMale (n = 579)Female (n = 499)Normal muscle mass (n = 462)Low muscle mass (n = 117)P valueNormal muscle mass (n = 431)Low muscle mass (n = 68)P valueAge (years)54.68 ± 11.9460.86 ± 13.47 < 0.00161.11 ± 10.7865.85 ± 11.440.001A/G ratio1.29 ± 0.171.15 ± 0.20 < 0.0011.10 ± 0.151.04 ± 0.180.003ANDROID34.60 ± 6.1831.93 ± 8.230.00140.52 ± 6.1737.51 ± 7.720.003GYNOID26.95 ± 4.5527.63 ± 5.190.19737.05 ± 4.9536.17 ± 5.250.176SMI (kg/m2)7.96(7.54,8.51)6.51(6.27,6.80) < 0.0016.35(5.90,6.84)5.18(4.95,5.34) < 0.001Obesity (n, %)367(79.4%)82(70.1%)0.03319(74%)39(57.4%)0.005Height (cm)173.26 ± 5.83171.04 ± 8.740.001160.67 ± 5.12158.99 ± 5.430.013Weight (Kg)82.95 ± 14.2468.90 ± 11.44 < 0.00170.51 ± 12.4156.38 ± 7.65 < 0.001BMI (kg/m2)27.57 ± 4.0923.69 ± 4.86 < 0.00127.29 ± 4.5022.30 ± 2.86 < 0.001SBP (mmHg)140.40 ± 19.80137.68 ± 21.460.192143.38 ± 20.64142.44 ± 22.130.731DBP (mmHg)82.49 ± 12.9376.20 ± 12.30 < 0.00176.53 ± 12.4474.18 ± 10.960.142Duration (years)7 (4, 10)8 (6, 11)0.0088 (5, 10)8 (5, 10)0.858FPG (mmol/L)8.15 ± 2.777.96 ± 3.200.5647.72 ± 2.897.19 ± 2.740.154TC (mmol/L)4.54 ± 1.154.48 ± 1.090.6064.70 ± 1.354.60 ± 1.170.583HDL (mmol/L)1.14 ± 0.271.24 ± 0.350.0041.29 ± 0.371.35 ± 0.350.192LDL (mmol/L)3.0 ± 0.902.97 ± 0.950.8193.0 ± 0.932.90 ± 1.00.408TG (mmol/L)1.67(1.07,2.52)1.10(0.85,1.77) < 0.0011.47(1.04,2.15)1.32(0.89,1.67)0.004Ca (mmol/L)2.30 ± 0.122.28 ± 0.130.1472.30 ± 0.112.30 ± 0.130.958P (mmol/L)1.21 ± 0.181.14 ± 0.18 < 0.0011.26 ± 0.171.20 ± 0.210.017A/G: Serum albumin to globulin ratio, SMI: skeletal muscle index, BMI: body mass index, SBP: systolic blood pressure, DBP: diastolic blood pressure, FPG: fasting plasma glucose, TC: total cholesterol, HDL: high-density lipoprotein cholesterol, LDL: low-density lipoprotein cholesterol, TG: triglyceride, Ca: serum calcium and P: serum phosphorus. Table 1The clinical characteristics differences between T2DM patients with and without low muscle mass.CharacteristicsMale (n = 579)Female (n = 499)Normal muscle mass (n = 462)Low muscle mass (n = 117)P valueNormal muscle mass (n = 431)Low muscle mass (n = 68)P valueAge (years)54.68 ± 11.9460.86 ± 13.47 < 0.00161.11 ± 10.7865.85 ± 11.440.001A/G ratio1.29 ± 0.171.15 ± 0.20 < 0.0011.10 ± 0.151.04 ± 0.180.003ANDROID34.60 ± 6.1831.93 ± 8.230.00140.52 ± 6.1737.51 ± 7.720.003GYNOID26.95 ± 4.5527.63 ± 5.190.19737.05 ± 4.9536.17 ± 5.250.176SMI (kg/m2)7.96(7.54,8.51)6.51(6.27,6.80) < 0.0016.35(5.90,6.84)5.18(4.95,5.34) < 0.001Obesity (n, %)367(79.4%)82(70.1%)0.03319(74%)39(57.4%)0.005Height (cm)173.26 ± 5.83171.04 ± 8.740.001160.67 ± 5.12158.99 ± 5.430.013Weight (Kg)82.95 ± 14.2468.90 ± 11.44 < 0.00170.51 ± 12.4156.38 ± 7.65 < 0.001BMI (kg/m2)27.57 ± 4.0923.69 ± 4.86 < 0.00127.29 ± 4.5022.30 ± 2.86 < 0.001SBP (mmHg)140.40 ± 19.80137.68 ± 21.460.192143.38 ± 20.64142.44 ± 22.130.731DBP (mmHg)82.49 ± 12.9376.20 ± 12.30 < 0.00176.53 ± 12.4474.18 ± 10.960.142Duration (years)7 (4, 10)8 (6, 11)0.0088 (5, 10)8 (5, 10)0.858FPG (mmol/L)8.15 ± 2.777.96 ± 3.200.5647.72 ± 2.897.19 ± 2.740.154TC (mmol/L)4.54 ± 1.154.48 ± 1.090.6064.70 ± 1.354.60 ± 1.170.583HDL (mmol/L)1.14 ± 0.271.24 ± 0.350.0041.29 ± 0.371.35 ± 0.350.192LDL (mmol/L)3.0 ± 0.902.97 ± 0.950.8193.0 ± 0.932.90 ± 1.00.408TG (mmol/L)1.67(1.07,2.52)1.10(0.85,1.77) < 0.0011.47(1.04,2.15)1.32(0.89,1.67)0.004Ca (mmol/L)2.30 ± 0.122.28 ± 0.130.1472.30 ± 0.112.30 ± 0.130.958P (mmol/L)1.21 ± 0.181.14 ± 0.18 < 0.0011.26 ± 0.171.20 ± 0.210.017A/G: Serum albumin to globulin ratio, SMI: skeletal muscle index, BMI: body mass index, SBP: systolic blood pressure, DBP: diastolic blood pressure, FPG: fasting plasma glucose, TC: total cholesterol, HDL: high-density lipoprotein cholesterol, LDL: low-density lipoprotein cholesterol, TG: triglyceride, Ca: serum calcium and P: serum phosphorus. The clinical characteristics differences between T2DM patients with and without low muscle mass. A/G: Serum albumin to globulin ratio, SMI: skeletal muscle index, BMI: body mass index, SBP: systolic blood pressure, DBP: diastolic blood pressure, FPG: fasting plasma glucose, TC: total cholesterol, HDL: high-density lipoprotein cholesterol, LDL: low-density lipoprotein cholesterol, TG: triglyceride, Ca: serum calcium and P: serum phosphorus. The correlation between low muscle mass and the clinical parameters Spearman correlation analysis was applied to investigate the whole clinical parameters that might be related to low muscle mass. As indicated in Table 2, age, duration of diabetes, and HDL were positively correlated with low muscle mass (r > 0, P < 0.05), while A/G ratio, ANDROID, SMI, prevalence of obesity, height, weight, BMI, DBP, TG and serum phosphorus level were negatively correlated with low muscle mass (r < 0, P < 0.05) in male patients. Among the female subjects, there were positive correlation between age and low muscle mass (r > 0, P < 0.05). While A/G ratio, ANDROID, SMI, prevalence of obesity, height, weight, BMI, and TG were negatively correlated with low muscle mass (r < 0, P < 0.05). Notably, it should be noted that the serum phosphorus showed a weak, non-significant negative correlation with low muscle mass in female subjects (r = − 0.074, p = 0.099), suggesting minimal clinical relevance. We identified SMI as a quantitative indicator of low muscle mass, subsequently, we conducted linear regression analysis to further explore the correlation between SMI and clinical characteristics. As Table 3 shown, SMI was correlated with age, A/G ratio, ANDROID, prevalence of obesity, height, weight, BMI, DBP, duration of diabetes, HDL, TG, serum level of Ca and P (P < 0.05) in male subjects. In female counterparts, SMI was correlated with age, A/G ratio, ANDROID, GYNOID, prevalence of obesity, height, weight, BMI, DBP, HDL, and serum phosphorus level (P < 0.05). Among them, we are most concerned about the consistent positive linear correlation between SMI and phosphorus, with (β = 1.248, 95% CI 0.568–1.929) in male patients and (β = 0.862, 95% CI 0.374–1.349) in female patients, respectively. Furthermore, both male and female patients were categorized into four groups based on the quartiles of the serum P levels: Q1 (< 1.06 mmol/L), Q2 (1.06–1.19 mmol/L), Q3 (1.19–1.33 mmol/L), and Q4 (≥ 1.33 mmol/L) in males and Q1 (< 1.13 mmol/L), Q2 (1.13–1.26 mmol/L), Q3 (1.26–1.36 mmol/L), and Q4 (≥ 1.36 mmol/L) in females. Additionally, the results revealed that elevated P levels reduced the risk of low muscle mass in males (P for trend = 0.001) and in females (P for trend = 0.020).Table 2Correlation of low muscle mass with clinical parameters by Spearman analysis.VariablesMaleFemalerP valuerP valueAge (years)0.199 < 0.0010.1400.002A/G ratio− 0.275 < 0.001− 0.1080.016ANDROID− 0.1440.001− 0.1250.005GYNOID0.0500.234− 0.0630.163SMI (kg/m2)− 0.695 < 0.001− 0.594 < 0.001Obesity (%)− 0.0900.030− 0.1270.005Height (cm)− 0.1190.004− 0.1230.006Weight (Kg)− 0.478 < 0.001− 0.436 < 0.001BMI (kg/m2)− 0.474 < 0.001− 0.429 < 0.001SBP (mmHg)− 0.0510.217− 0.0230.607DBP (mmHg)− 0.199 < 0.001− 0.0590.188Duration (years)0.1110.0080.0080.858FPG (mmol/L)− 0.0470.255− 0.0680.130TC (mmol/L)− 0.0080.844− 0.0180.696HDL (mmol/L)0.1200.0040.0750.095LDL (mmol/L)− 0.0140.728− 0.0370.406TG (mmol/L)− 0.220 < 0.001− 0.1270.004Ca (mmol/L)− 0.0530.201− 0.0170.707P (mmol/L)− 0.1310.002− 0.0740.099A/G: Serum albumin to globulin ratio, SMI: skeletal muscle index, BMI: body mass index, SBP: systolic blood pressure, DBP: diastolic blood pressure, FPG: fasting plasma glucose, TC: total cholesterol, HDL: high-density lipoprotein cholesterol, LDL: low-density lipoprotein cholesterol, TG: triglyceride, Ca: serum calcium and P: serum phosphorus.Table 3The correlation between SMI and clinical characteristics by Linear regression analysis.MaleFemaleβ (95% CI)P valueβ (95% CI)P valueAge (years)− 0.028 (− 0.038, − 0.018) < 0.001− 0.021 (− 0.028, − 0.013) < 0.001A/G ratio1.583 (0.916, 2.250) < 0.0011.004 (0.449, 1.559) < 0.001ANDROID0.044 (0.026, 0.062) < 0.0010.033 (0.020, 0.046) < 0.001GYNOID0.025 (− 0.002, 0.051)0.0680.019 (0.002, 0.036)0.027Obesity (n, %)0.365 (0.067, 0.662)0.0160.342 (0.155, 0.530) < 0.001Height (cm)0.035 (0.016, 0.053) < 0.0010.019 (0.003, 0.036)0.021Weight (Kg)0.050 (0.043, 0.057) < 0.0010.047 (0.042, 0.052) < 0.001BMI (kg/m2)0.151 (0.126, 0.176) < 0.0010.131 (0.117, 0.146) < 0.001SBP (mmHg)0.005 (− 0.001, 0.011)0.1360.003 (− 0.001, 0.007)0.170DBP (mmHg)0.020 (0.011, 0.030) < 0.0010.013 (0.006, 0.020) < 0.001Duration (years)− 0.020 (− 0.039, − 0.001)0.036− 0.003 (− 0.017, 0.011)0.714FPG (mmol/L)0.017 (− 0.027, 0.060)0.452− 0.005 (− 0.035, 0.025)0.726TC (mmol/L)0.031 (− 0.079, 0.141)0.585− 0.009 (− 0.073, 0.056)0.793HDL (mmol/L)− 0.576 (− 1.006, − 0.145)0.009− 0.270 (− 0.504, − 0.035)0.024LDL (mmol/L)0.040 (− 0.098, 0.177)0.5710.003 (− 0.088, 0.094)0.943TG (mmol/L)0.086 (0.026, 0.145)0.0050.055 (− 0.009, 0.118)0.092Ca (mmol/L)1.216 (0.167, 2.265)0.0230.175 (− 0.558, 0.908)0.693P (mmol/L)1.248 (0.568, 1.929) < 0.0010.862 (0.374, 1.349)0.001P for trend0.0010.020A/G: Serum albumin to globulin ratio, SMI: skeletal muscle index, BMI: body mass index, SBP: systolic blood pressure, DBP: diastolic blood pressure, FPG: fasting plasma glucose, TC: total cholesterol, HDL: high-density lipoprotein cholesterol, LDL: low-density lipoprotein cholesterol, TG: triglyceride, Ca: serum calcium and P: serum phosphorus. Table 2Correlation of low muscle mass with clinical parameters by Spearman analysis.VariablesMaleFemalerP valuerP valueAge (years)0.199 < 0.0010.1400.002A/G ratio− 0.275 < 0.001− 0.1080.016ANDROID− 0.1440.001− 0.1250.005GYNOID0.0500.234− 0.0630.163SMI (kg/m2)− 0.695 < 0.001− 0.594 < 0.001Obesity (%)− 0.0900.030− 0.1270.005Height (cm)− 0.1190.004− 0.1230.006Weight (Kg)− 0.478 < 0.001− 0.436 < 0.001BMI (kg/m2)− 0.474 < 0.001− 0.429 < 0.001SBP (mmHg)− 0.0510.217− 0.0230.607DBP (mmHg)− 0.199 < 0.001− 0.0590.188Duration (years)0.1110.0080.0080.858FPG (mmol/L)− 0.0470.255− 0.0680.130TC (mmol/L)− 0.0080.844− 0.0180.696HDL (mmol/L)0.1200.0040.0750.095LDL (mmol/L)− 0.0140.728− 0.0370.406TG (mmol/L)− 0.220 < 0.001− 0.1270.004Ca (mmol/L)− 0.0530.201− 0.0170.707P (mmol/L)− 0.1310.002− 0.0740.099A/G: Serum albumin to globulin ratio, SMI: skeletal muscle index, BMI: body mass index, SBP: systolic blood pressure, DBP: diastolic blood pressure, FPG: fasting plasma glucose, TC: total cholesterol, HDL: high-density lipoprotein cholesterol, LDL: low-density lipoprotein cholesterol, TG: triglyceride, Ca: serum calcium and P: serum phosphorus. Correlation of low muscle mass with clinical parameters by Spearman analysis. A/G: Serum albumin to globulin ratio, SMI: skeletal muscle index, BMI: body mass index, SBP: systolic blood pressure, DBP: diastolic blood pressure, FPG: fasting plasma glucose, TC: total cholesterol, HDL: high-density lipoprotein cholesterol, LDL: low-density lipoprotein cholesterol, TG: triglyceride, Ca: serum calcium and P: serum phosphorus. Table 3The correlation between SMI and clinical characteristics by Linear regression analysis.MaleFemaleβ (95% CI)P valueβ (95% CI)P valueAge (years)− 0.028 (− 0.038, − 0.018) < 0.001− 0.021 (− 0.028, − 0.013) < 0.001A/G ratio1.583 (0.916, 2.250) < 0.0011.004 (0.449, 1.559) < 0.001ANDROID0.044 (0.026, 0.062) < 0.0010.033 (0.020, 0.046) < 0.001GYNOID0.025 (− 0.002, 0.051)0.0680.019 (0.002, 0.036)0.027Obesity (n, %)0.365 (0.067, 0.662)0.0160.342 (0.155, 0.530) < 0.001Height (cm)0.035 (0.016, 0.053) < 0.0010.019 (0.003, 0.036)0.021Weight (Kg)0.050 (0.043, 0.057) < 0.0010.047 (0.042, 0.052) < 0.001BMI (kg/m2)0.151 (0.126, 0.176) < 0.0010.131 (0.117, 0.146) < 0.001SBP (mmHg)0.005 (− 0.001, 0.011)0.1360.003 (− 0.001, 0.007)0.170DBP (mmHg)0.020 (0.011, 0.030) < 0.0010.013 (0.006, 0.020) < 0.001Duration (years)− 0.020 (− 0.039, − 0.001)0.036− 0.003 (− 0.017, 0.011)0.714FPG (mmol/L)0.017 (− 0.027, 0.060)0.452− 0.005 (− 0.035, 0.025)0.726TC (mmol/L)0.031 (− 0.079, 0.141)0.585− 0.009 (− 0.073, 0.056)0.793HDL (mmol/L)− 0.576 (− 1.006, − 0.145)0.009− 0.270 (− 0.504, − 0.035)0.024LDL (mmol/L)0.040 (− 0.098, 0.177)0.5710.003 (− 0.088, 0.094)0.943TG (mmol/L)0.086 (0.026, 0.145)0.0050.055 (− 0.009, 0.118)0.092Ca (mmol/L)1.216 (0.167, 2.265)0.0230.175 (− 0.558, 0.908)0.693P (mmol/L)1.248 (0.568, 1.929) < 0.0010.862 (0.374, 1.349)0.001P for trend0.0010.020A/G: Serum albumin to globulin ratio, SMI: skeletal muscle index, BMI: body mass index, SBP: systolic blood pressure, DBP: diastolic blood pressure, FPG: fasting plasma glucose, TC: total cholesterol, HDL: high-density lipoprotein cholesterol, LDL: low-density lipoprotein cholesterol, TG: triglyceride, Ca: serum calcium and P: serum phosphorus. The correlation between SMI and clinical characteristics by Linear regression analysis. A/G: Serum albumin to globulin ratio, SMI: skeletal muscle index, BMI: body mass index, SBP: systolic blood pressure, DBP: diastolic blood pressure, FPG: fasting plasma glucose, TC: total cholesterol, HDL: high-density lipoprotein cholesterol, LDL: low-density lipoprotein cholesterol, TG: triglyceride, Ca: serum calcium and P: serum phosphorus. The association between the serum P level and low muscle mass The logistic regression analysis further confirmed the association between the serum P level and low muscle mass, as shown in Table 4 and Fig. 1. Notably, the serum phosphorus was associated with a decreased risk of low muscle mass in both male (OR = 0.117, 95% CI 0.036–0.382, P < 0.001) and female (OR = 0.166, 95% CI 0.037–0.735, P = 0.018) T2DM participants with no adjustment (model 1). After adjusting for age, serum phosphorus was found to be negatively associated with low muscle mass in males (OR = 0.237, 95% CI 0.069–0.819, P = 0.023) and females (OR = 0.170, 95% CI 0.036–0.794, P = 0.024) (model 2). And after adjusting for basic clinical parameters that lacked laboratory data such as age, SBP, DBP and duration of diabetes, phosphorus was found to be correlated with a reduced risk of low muscle mass in males (OR = 0.214, 95% CI 0.060–0.755, P = 0.017) and females (OR = 0.172, 95% CI 0.037–0.800, P = 0.025) (model 3). Furthermore, serum phosphorus was still inversely correlated with low muscle mass after adjusting for age, A/G ratio, SBP, DBP, duration of diabetes, Ca, FPG, TC, TG, HDL, and LDL both in male patients (OR = 0.187, 95% CI 0.045–0.785, P = 0.022) and female patients (OR = 0.161, 95% CI 0.032–0.816, P = 0.027) (model 4). After FDR correction, all associations that were nominally significant (p < 0.05) remained significant at q < 0.05, indicating a robust association between serum phosphorus and low muscle mass across the tested models. Table 4The association between serum phosphorus and low muscle mass.MaleFemaleOR (95%CI)P valueOR (95%CI)P valueModel 10.117(0.036–0.382) < 0.001 0.166(0.037–0.735) 0.018 Model 20.237(0.069–0.819) 0.023 0.170(0.036–0.794) 0.024 Model 30.214(0.060–0.755) 0.017 0.172(0.037–0.800) 0.025 Model 40.187(0.045–0.785) 0.022 0.161(0.032–0.816) 0.027 Model 1: not adjusted.Model 2: adjusted for age.Model 3: adjusted for age, SBP, DBP, duration of diabetes.Model 4: adjusted for age, A/G ratio, SBP, DBP, duration of diabetes, Ca, FPG, TC, TG, HDL, LDL.A/G: Serum albumin to globulin ratio, SBP: systolic blood pressure, DBP:diastolic blood pressure, FPG: fasting plasma glucose, TC: total cholesterol, HDL: high-density lipoprotein cholesterol, LDL: low-density lipoprotein cholesterol, TG: triglyceride, Ca: serum calcium and P: serum phosphorus. Table 4The association between serum phosphorus and low muscle mass.MaleFemaleOR (95%CI)P valueOR (95%CI)P valueModel 10.117(0.036–0.382) < 0.001 0.166(0.037–0.735) 0.018 Model 20.237(0.069–0.819) 0.023 0.170(0.036–0.794) 0.024 Model 30.214(0.060–0.755) 0.017 0.172(0.037–0.800) 0.025 Model 40.187(0.045–0.785) 0.022 0.161(0.032–0.816) 0.027 Model 1: not adjusted.Model 2: adjusted for age.Model 3: adjusted for age, SBP, DBP, duration of diabetes.Model 4: adjusted for age, A/G ratio, SBP, DBP, duration of diabetes, Ca, FPG, TC, TG, HDL, LDL.A/G: Serum albumin to globulin ratio, SBP: systolic blood pressure, DBP:diastolic blood pressure, FPG: fasting plasma glucose, TC: total cholesterol, HDL: high-density lipoprotein cholesterol, LDL: low-density lipoprotein cholesterol, TG: triglyceride, Ca: serum calcium and P: serum phosphorus. The association between serum phosphorus and low muscle mass. Model 1: not adjusted. Model 2: adjusted for age. Model 3: adjusted for age, SBP, DBP, duration of diabetes. Model 4: adjusted for age, A/G ratio, SBP, DBP, duration of diabetes, Ca, FPG, TC, TG, HDL, LDL. A/G: Serum albumin to globulin ratio, SBP: systolic blood pressure, DBP: diastolic blood pressure, FPG: fasting plasma glucose, TC: total cholesterol, HDL: high-density lipoprotein cholesterol, LDL: low-density lipoprotein cholesterol, TG: triglyceride, Ca: serum calcium and P: serum phosphorus. Fig. 1The correlation between clinical characteristics and low muscle mass in male (a) and female (b) T2DM patients. Abbreviations: A/G: Serum albumin to globulin ratio, SBP: systolic blood pressure, DBP: diastolic blood pressure, FPG: fasting plasma glucose, TC: total cholesterol, HDL: high-density lipoprotein cholesterol, LDL: low-density lipoprotein cholesterol, TG: triglyceride, P: serum phosphorus. The correlation between clinical characteristics and low muscle mass in male (a) and female (b) T2DM patients. Abbreviations: A/G: Serum albumin to globulin ratio, SBP: systolic blood pressure, DBP: diastolic blood pressure, FPG: fasting plasma glucose, TC: total cholesterol, HDL: high-density lipoprotein cholesterol, LDL: low-density lipoprotein cholesterol, TG: triglyceride, P: serum phosphorus. The association between serum phosphorus and SMI Participants were categorized into four groups based on the quartiles of the serum P level as previously stated and displayed in Table 5. We found that SMI were significantly increased with ascending serum P level, suggesting a positive correlation between phosphorus and SMI in both genders. Furthermore, a potential nonlinear association between P and SMI was indentified in both male and female subjects, which shown in Fig. 2. To further analyze this relationship, a two-piecewise linear regression model was applied (Fig. 3). Moreover, our model estimated an inflection point at approximately 0.98 mmol/L in males. When the level of phosphorus was ≤ 0.98, it was significantly positive correlated with SMI in male patients (β = 3.299, 95% CI 0.635–5.962, P = 0.015). However, when P was > 0.98, there was no significant correlation between P and SMI (P > 0.05), as shown in Table 6. Similarly, segmented regression suggested a potential threshold at 1.04 mmol/L in females within this cohort. A positive correlation between P and SMI was observed when P was ≤ 1.04 (β = 3.217, 95% CI 1.105–5.329, P = 0.003). Conversely, when P was > 1.04, there was no statistical significance, as shown in Table 7. Table 5The comparison of SMI by quartiles of serum phosphorus level.MaleFemaleGroupsNSMI (kg/m2) H P NSMI (kg/m2) H P Q11327.49 (6.77–8.11)19.05 < 0.001 1156.06 (5.61–6.63)9.76 0.021 Q21547.78 (7.09–8.47)1276.06 (5.64–6.67)Q31457.75 (7.17–8.21)1266.35 (5.73–6.79)Q41487.89 (7.30–8.64)1316.37 (5.82–6.89)SMI: skeletal muscle index.Serum phosphorus level in males was classified as Q1 (< 1.06 mmol/L), Q2 (1.06–1.19 mmol/L), Q3 (1.19–1.33 mmol/L), and Q4 (≥ 1.33 mmol/L).Serum phosphorus level in females was classified as Q1 (< 1.13 mmol/L), Q2 (1.13–1.26 mmol/L), Q3 (1.26–1.36 mmol/L), and Q4 (≥ 1.36 mmol/L). Table 5The comparison of SMI by quartiles of serum phosphorus level.MaleFemaleGroupsNSMI (kg/m2) H P NSMI (kg/m2) H P Q11327.49 (6.77–8.11)19.05 < 0.001 1156.06 (5.61–6.63)9.76 0.021 Q21547.78 (7.09–8.47)1276.06 (5.64–6.67)Q31457.75 (7.17–8.21)1266.35 (5.73–6.79)Q41487.89 (7.30–8.64)1316.37 (5.82–6.89)SMI: skeletal muscle index.Serum phosphorus level in males was classified as Q1 (< 1.06 mmol/L), Q2 (1.06–1.19 mmol/L), Q3 (1.19–1.33 mmol/L), and Q4 (≥ 1.33 mmol/L).Serum phosphorus level in females was classified as Q1 (< 1.13 mmol/L), Q2 (1.13–1.26 mmol/L), Q3 (1.26–1.36 mmol/L), and Q4 (≥ 1.36 mmol/L). The comparison of SMI by quartiles of serum phosphorus level. SMI: skeletal muscle index. Serum phosphorus level in males was classified as Q1 (< 1.06 mmol/L), Q2 (1.06–1.19 mmol/L), Q3 (1.19–1.33 mmol/L), and Q4 (≥ 1.33 mmol/L). Serum phosphorus level in females was classified as Q1 (< 1.13 mmol/L), Q2 (1.13–1.26 mmol/L), Q3 (1.26–1.36 mmol/L), and Q4 (≥ 1.36 mmol/L). Fig. 2The relationship between serum phosphorus and SMI in male and female T2DM patients. Abbreviations: SMI: skeletal muscle index, P: serum phosphorus. The relationship between serum phosphorus and SMI in male and female T2DM patients. Abbreviations: SMI: skeletal muscle index, P: serum phosphorus. Fig. 3The relationship between P and SMI by smooth curve fitting in male (a) and female (b) T2DM patients. Threshold values (P ≤ 0.98 mmol/L in males: P ≤ 1.04 in females). Abbreviations: SMI: skeletal muscle index, P: serum phosphorus. The relationship between P and SMI by smooth curve fitting in male (a) and female (b) T2DM patients. Threshold values (P ≤ 0.98 mmol/L in males: P ≤ 1.04 in females). Abbreviations: SMI: skeletal muscle index, P: serum phosphorus. Table 6The correlation between P and SMI in male subjects.Modelsβ (95% CI)P value Model I One line slope0.427 (− 0.072, 0.926)0.094 Model II Turning point0.98 ≤ 0.98 slope 13.299 (0.635, 5.962) 0.015 > 0.98 slope 20.143 (− 0.418, 0.704)0.618P: serum phosphorus, SMI: skeletal muscle index. Table 6The correlation between P and SMI in male subjects.Modelsβ (95% CI)P value Model I One line slope0.427 (− 0.072, 0.926)0.094 Model II Turning point0.98 ≤ 0.98 slope 13.299 (0.635, 5.962) 0.015 > 0.98 slope 20.143 (− 0.418, 0.704)0.618P: serum phosphorus, SMI: skeletal muscle index. The correlation between P and SMI in male subjects. P: serum phosphorus, SMI: skeletal muscle index. Table 7The correlation between P and SMI in female subjects.Modelsβ (95% CI)P value Model I One line slope0.862 (0.375, 1.348) < 0.001 Model II Turning point1.04 ≤ 1.04 slope 13.217 (1.105, 5.329) 0.003 > 1.04 slope 20.524 (− 0.043, 1.091)0.071P: serum phosphorus, SMI: skeletal muscle index. Table 7The correlation between P and SMI in female subjects.Modelsβ (95% CI)P value Model I One line slope0.862 (0.375, 1.348) < 0.001 Model II Turning point1.04 ≤ 1.04 slope 13.217 (1.105, 5.329) 0.003 > 1.04 slope 20.524 (− 0.043, 1.091)0.071P: serum phosphorus, SMI: skeletal muscle index. The correlation between P and SMI in female subjects. P: serum phosphorus, SMI: skeletal muscle index. Discussion In recent years, sarcopenia, which often be characterized by loss of skeletal muscle mass, reduced muscle strength and/or impaired physical performance, has been increasingly recognized as one of vital chronic complications of diabetes mellitus. Low muscle mass is closely associated with poor prognosis of diabetes. Consequently, an increasing number of studies have focused on exploring the epidemiology and the underlying mechanisms of low skeletal muscle mass in diabetic patients. Therefore, the early detection and diagnosis of low muscle mass is particularly important to facilitate timely follow-up intervention, which helps in mitigating the deterioration of syndrome. In our previous researches, we have found that ALT/AST was inversely correlated with muscle mass in patients with T2DM14. Besides, creatine kinase (CK) is negatively correlated with low muscle mass among patients with T2DM15. TyG index, a novel marker of insulin resistance, was inversely correlated with the presence of sarcopenia in T2DM patients16. Moreover, higher TG/HDL-C ratio was also demonstrated to be correlated with skeletal muscle mass in T2DM patients17. However, there remains a significant gap in literature given that only a limited number of studies have reported the relationship between serum electrolyte level and muscle mass. Our present study indicated that higher serum P level was correlated with a reduced risk of low muscle mass in patients with T2DM, that is, phosphorus was positively associated with SMI. The adjusted odds ratio (OR) for serum phosphorus in males (OR = 0.187, 95% CI 0.045–0.785, P = 0.022) indicates an 81.3% reduction in low muscle mass risk per unit increase in serum phosphorus. And there may be an 83.9% reduction in low muscle mass risk per unit increase in serum phosphorus among female patients (OR = 0.161, 95% CI 0.032–0.816, P = 0.027). Besides, the β-coefficient for P in T2DM patients reflects a clinically meaningful increase in P with higher SMI, with β = 1.248 in males (P < 0.001) and β = 0.862 in females (P = 0.001), respectively. Similarly, another study based on hemodialysis patients reported that serum P level was lower in the low muscle mass group compared to the normal muscle mass group (P = 0.015), this observation is consistent with our findings to some extent11. However, one Japanese research reported that serum P concentration in weakened grip strength group was significantly higher than that of patients without weakened grip strength (3.8 vs. 3.3 mg/dL, P < 0.01), demonstrated that serum P concentration was negatively correlated with grip strength. Additionally, this negative association was also confirmed after adjusting for the confounding factors12, which was contrary to our conclusion. Several factors may explain this apparent differences. Firstly, the measurement of muscle mass were fundamentally different. We assessed muscle mass (a structural quantity via DXA), whereas Tanaka’s team assessed grip strength (a functional performance). These two aspects of sarcopenia are only modestly correlated, and their relationship with phosphorus may be distinct. Besides, the study populations were diverse. Our cohort consisted of inpatients, who may have a more advanced metabolic state, while Tanaka et al. studied community-dwelling elderly. Differences in diabetes duration, complication burden, and nutritional status could alter the balance between beneficial and detrimental effects of serum phosphorus. More importantly, our study extended the analysis by sex, we performed stratified analyses and found the protective association against low muscle mass to be statistically significant in both genders after full adjustment. Tanaka’s analysis did not examine sex-specific associations, reporting only the combined effect for their cohort. This underscores the value of our sex-stratified approach in revealing that the phosphorus-muscle relationship may be relevant across both genders in the T2DM population, at least in terms of muscle mass preservation. In our sex-stratified Spearman correlation analysis, serum P showed a significant negative correlation with low muscle mass in males (r = − 0.131, p = 0.002). A similar negative correlation was also observed in females, although it did not reach statistical significance (r = − 0.074, p = 0.099). This consistent direction of association across sexes in our study suggests a potential biological link between serum phosphorus and muscle mass, warranting further investigation in larger female cohorts. Furthermore, an intriguing finding emerged from our sex-stratified, segmented regression analyses, which revealed a non-linear relationship between serum phosphorus and SMI in both sexes, characterized by a threshold effect. In both genders, serum phosphorus showed a strong, positive association with SMI below a specific concentration (turning point: 0.98 mmol/L in males and 1.04 mmol/L in females). However, above these thresholds, the association attenuated substantially and lost statistical significance (P = 0.618 in males and P = 0.071 in females). This shared pattern suggests a potential physiological plateau or saturation effect, where the positive influence of circulating phosphorus on muscle mass may diminish beyond an optimal range. And the inflection points identified in our analysis offer a quantitative hypothesis for a potential saturable or threshold effect in the phosphorus-muscle mass relationship. However, these values are inherently cohort-specific, derived from the distribution and characteristics of our T2DM patient sample. Factors such as age range, diabetes duration, renal function (even within our exclusion criteria), dietary habits, and the specific assay used for serum phosphorus measurement could all influence the precise numerical estimate. Therefore, these thresholds must be considered exploratory and require rigorous external validation in prospective, diverse, and independent cohorts before any clinical or physiological implications can be drawn. Future research should aim to replicate these findings and determine whether a consistent threshold exists across populations, or if it varies by specific patient subgroups. The difference in the threshold values between sexes (1.04 mmol/L in females vs. 0.98 mmol/L in males) invites the mechanistic speculation. We hypothesize that sex-specific factors may modulate this “optimal range” could be explained by several factors, including hormonal regulation and body composition differences. Hormonal Regulation: Estrogen is known to influence renal phosphate reabsorption and bone metabolism. Its presence in premenopausal females may alter phosphorus homeostasis, potentially allowing a higher serum threshold before the association with muscle plateaus, compared to males. Body Composition Differences: Males typically have greater absolute muscle mass, which may be more sensitive to phosphorus availability at lower serum levels. Conversely, the higher average adiposity and different fat distribution in females might introduce greater metabolic complexity or measurement bias, potentially contributing to the weaker (borderline) correlation observed above the threshold. It is worth noting that the borderline p-value (0.071) for females above the threshold should not be dismissed as “no association”, but rather indicates a markedly weakened trend within the context of this non-linear model. This parallels the non-significant finding in males above their threshold, reinforcing the concept of a ceiling effect common to both genders, but with the ceiling level itself possibly being sex-modulated. These observations shift the research question from a simple linear correlation to exploring why the beneficial association plateaus, and why the plateau point may differ by sex. These discrepancies underscore the potential influence of sex-related physiological mechanisms on phosphorus metabolism and muscle homeostasis, which warrant further investigation. Future mechanistic studies are needed to explore these hypotheses. Furthermore, findings from a systematic review of Netherlands – specifically in studies exploring the potential role of minerals in the prevention and treatment of sarcopenia in the elderly–failed to establish a relationship between Phosphorus with muscle-reducing outcomes, given that analysis incorporated insufficient number of studies, resulting in a low quality of evidence13. The results of the current limited studies on low muscle mass remain inconsistent, underscoring the need for further randomized controlled trials to determine the influence of serum P on muscle mass in patients with T2DM. As mentioned above, P plays a vital role in skeletal mineralization and energy metabolism, and P-containing metabolites are key elements in the formation of cell membranes and nucleic acid molecules, serving as essential components of cellular structures and functions7. P-containing metabolites also participates in oxidative phosphorylation through the phosphorylation of intermediate metabolites and the storage of energy released in high-energy phosphate bonds, such as those in ATP or phosphocreatine8,9. Besides, P is essential for the provision of high-energy phosphates required for the contractile activity and structural integrity of the muscular membrane and intracellular organelles in skeletal muscles10. Thus P is essential for skeletal muscle bioenergetics and structural integrity, while its role in T2DM-specific metabolic derangements need further exploration. Insulin resistance, a typical feature of T2DM, may impair phosphorus-mediated pathways critical for muscle maintenance. In addition to the association between phosphorus and muscle mass, our study also showed that patients with T2DM in the low muscle mass group, both males and females, exhibited a higher age, as well as lower A/G ratio, ANDROID, prevalence of obesity, height, weight, BMI, and TG compared to the normal muscle mass group. It is well-established that muscle mass decreases with advancing age, a phenomenon that is supported by numerous studies previously mentioned. Additionally, obesity is recognized as a typical risk factor for sarcopenia. A multicenter study in Japan conducted among community-dwelling older adults showed that obesity and hypertension are independent predictors of sarcopenia among this cohort18. Additionally, previous studies have demonstrated that the metabolic risk factors, including BMI, are significantly related to loss of skeletal muscle or occurrence of sarcopenia19. Contrastingly, findings from a cross-sectional study which assessed older people residing in 11 long-term nursing homes in Australia indicated that low BMI is a predictive risk factor for sarcopenia20. A four-year follow-up longitudinal study in China showed that a higher BMI serves a protective factor against muscle mass loss, but increases the risk factor of reduced gait speed21. As for patients with T2DM, several studies have indicated that BMI is positively correlated with muscle mass, in other words, the prevalence of sarcopenia significantly increases as the level of BMI decreases in patients with T2DM22–25. Consistently, this study demonstrates that T2DM patients with low muscle mass exhibit lower BMI compared to those with normal muscle mass. These findings may initially appear contradictory, as obesity is typically linked to adverse metabolic outcomes. However, our results align with emerging evidence on sarcopenic obesity, where higher BMI may mask underlying muscle loss. Furthermore, BMI inherently reflects total body mass, which includes both fat and muscle components. Thus, while BMI and SMI are positively correlated, the preservation of muscle mass in T2DM patients with higher BMI may mitigate sarcopenia risk, even in the presence of excess adiposity. While obesity is associated with an elevated risk of diabetic complications, it is also inversely correlated with the risk of developing sarcopenia in people with T2DM. This study provides exploratory evidence of an inverse correlation between serum phosphates and low muscle mass in patients with T2DM. However, this study has several limitations inherent to its cross-sectional and exploratory design. Firstly, the observational nature precludes causal inference. To address the risk of Type I error from multiple testing, we applied the Benjamini–Hochberg False Discovery Rate (FDR) correction (q < 0.05) to all 130 p-values from our primary and exploratory analyses. All associations that were nominally significant (P < 0.05) remained significant after FDR correction (q < 0.05), confirming the robustness of our findings. Despite this, the exploratory nature of some sex-stratified and segmented regression analyses warrants cautious interpretation. Secondly, data on several key confounders were unavailable, including physical activity levels, detailed dietary intake (e.g., of protein and phosphate), vitamin D status, and specific glucose-lowering medications. Thirdly, the estimated serum phosphorus thresholds, while statistically identified and biologically plausible, are specific to our cohort and need external validation as discussed above. Furthermore, although patients with severe renal failure such as those on dialysis or eGFR < 30 mL/min/1.73m2 were excluded, comprehensive renal function parameters (continuous eGFR, parathyroid hormone, etc.) were not available for all participants, which may affect phosphorus homeostasis. The influence of these unmeasured factors and methodological choices warrants consideration when interpreting the results. Future research must move beyond cross-sectional designs to establish causality. We specifically recommend prospective, longitudinal cohort studies that can determine whether low serum phosphorus levels predict future muscle loss or the incidence of sarcopenia in patients with T2DM. Such studies should include repeated measurements of serum phosphorus levels, muscle mass, and detailed time-varying confounders mentioned above. Additionally, mechanistic studies are warranted to explore the underlying pathways, such as tissue-specific phosphorus dynamics, interactions with insulin signaling, and sex-specific effects in the diabetic state. Conclusion In conclusion, serum phosphorus level is inversely correlated with low muscle mass in patients with T2DM. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1