Computed tomography-based correction of the body mass index improves malnutrition detection in patients with cirrhosis and ascites

Table 1 General characteristics and nutrition-related variables of the study population (n = 100), n (%)/median (Q1, Q3)

Variable	Value
Male	73 (73.0)
Female	27 (27.0)
Age, years	52.00 (44.00, 63.00)
Body weight at admission, kg	70.00 (60.00, 78.00)
Actual BMI, kg/m2	24.22 (21.85, 26.15)
White blood cell count, × 109/L	2.73 (2.04, 3.75)
Lymphocyte count, × 109/L	0.57 (0.38, 0.89)
Prealbumin, mg/L	91.60 (62.92, 113.25)
Albumin, g/L	25.94 (23.59, 30.29)
Total bilirubin, μmol/L	18.30 (11.67, 30.56)
INR	1.23 (1.10, 1.42)
Ascites volume, mL	1513.37 (627.29, 2565.20)
Child-Pugh score	8.00 (6.00, 9.00)
Length of hospital stay, days	9.00 (7.00, 13.00)
Admission complications	78 (78.0)
Nutritional intervention during hospitalization	25 (25.0)
Readmission	69 (69.0)
One-year mortality	12/80 (15.0)
AMC (n = 85), cm	23.29 (22.54, 24.17)
Reduced AMC	8/85 (9.4)

BMI: Body mass index; INR: International normalized ratio; AMC: Arm muscle circumference.

Table 2 Comparison of body mass index, low body mass index, and Global Leadership Initiative on Malnutrition-defined malnutrition among the three methods, n (%)/median (Q1, Q3)

Outcome	Actual body weight method	CT-corrected method	CT-sensitive corrected method
BMI (kg/m2)	24.22 (21.85, 26.15)	23.67 (20.73, 25.48)	23.39 (20.53, 25.48)
Low BMI	6 (6.0)	10 (10.0)	12 (12.0)
GLIM-defined malnutrition	19 (19.0)	22 (22.0)	24 (24.0)

The body mass index was compared using the Wilcoxon signed-rank test, and the detection rates of a low body mass index and Global Leadership Initiative on Malnutrition-defined malnutrition were compared using the McNemar test without continuity correction. CT: Computed tomography; BMI: Body mass index; GLIM: Global Leadership Initiative on Malnutrition.

Table 3 Agreement analysis of Global Leadership Initiative on Malnutrition-defined malnutrition classification among the different methods

Comparison	Paired sample size	Observed agreement (%)	Kappa value
Actual body weight method vs CT-corrected method	100	97.0	0.908
Actual body weight method vs CT-sensitive corrected method	100	95.0	0.852
CT-corrected method vs CT-sensitive corrected method	100	98.0	0.944

Observed agreement refers to the percentage of concordant classifications between the two compared methods. Agreement was evaluated using kappa statistics. CT: Computed tomography.

Table 4 Comparison of the clinical characteristics between patients with and without low-body mass index reclassification (actual body weight method vs computed tomography -sensitive corrected method), n (%)/median (Q1, Q3)

Variable	No reclassification (n = 94)	Reclassification (n = 6)	P value
Age, years	52.00 (44.00, 63.00)	52.50 (39.50, 62.50)	0.722
Sex (male/female)	70/24	3/3	0.339
Length of hospital stay, days	11.00 (7.00, 15.00)	11.50 (6.25, 14.50)	0.716
Actual BMI, kg/m2	24.39 (22.77, 26.36)	19.53 (18.82, 20.68)	0.002
CT-corrected BMI, kg/m2	23.71 (21.62, 25.52)	17.84 (17.55, 18.93)	< 0.001
CT-sensitive corrected BMI, kg/m2	23.67 (21.55, 25.52)	17.84 (17.55, 18.10)	< 0.001
AMC, cm	23.54 (22.56, 24.26)	21.23 (20.92, 21.73)	0.025
Albumin, g/L	33.20 (27.75, 39.10)	33.55 (32.00, 34.80)	0.845
Total bilirubin, μmol/L	17.80 (11.82, 30.40)	19.10 (11.62, 121.52)	0.706
INR	1.21 (1.10, 1.41)	1.46 (1.17, 1.66)	0.242
Ascites volume, mL	1437.49 (612.18, 2327.16)	5252.04 (3918.76, 6262.42)	0.001
Liver function grade (A/B/C)	26/52/16	1/3/2	0.621
Severe lower-limb edema	6 (6.4)	3 (50.0)	0.009

Continuous variables are presented as the median (Q1, Q3) and were compared using the Wilcoxon rank-sum test. Categorical variables are presented as counts or percentages and were compared using Fisher’s exact test. CT: Computed tomography; BMI: Body mass index; INR: International normalized ratio; AMC: Arm muscle circumference.

Table 5 Comparison of the clinical characteristics between patients with and without Global Leadership Initiative on Malnutrition-defined malnutrition reclassification (actual body weight method vs computed tomography -sensitive corrected method), n (%)/median (Q1, Q3)

Variable	No reclassification (n = 95)	Reclassification (n = 5)	P value
Age, years	52.00 (44.00, 63.00)	47.00 (37.00, 58.00)	0.393
Actual BMI, kg/m2	24.34 (22.56, 26.32)	20.03 (19.03, 20.90)	0.006
CT-corrected BMI, kg/m2	23.70 (21.57, 25.51)	17.74 (17.49, 19.27)	0.001
CT-sensitive corrected BMI, kg/m2	23.65 (21.44, 25.51)	17.74 (17.49, 18.16)	0.001
AMC, cm	23.54 (22.54, 24.23)	21.42 (21.01, 21.82)	0.084
Albumin, g/L	33.20 (27.80, 38.70)	36.50 (32.00, 38.30)	0.840
Total bilirubin, μmol/L	18.10 (11.77, 30.56)	17.00 (13.80, 121.52)	0.929
INR	1.22 (1.10, 1.41)	1.46 (1.17, 1.75)	0.174
Ascites volume, mL	1438.40 (617.12, 2324.45)	6003.70 (4500.38, 6348.66)	0.001
Severe lower-limb edema	6 (6.3)	3 (60.0)	0.005

Continuous variables are presented as the median (Q1, Q3) and were compared using the Wilcoxon rank-sum test. Categorical variables are presented as counts or percentages and were compared using Fisher’s exact test. CT: Computed tomography; BMI: Body mass index; INR: International normalized ratio; AMC: Arm muscle circumference.

Table 6 Comparison of additional clinical and nutrition-related variables according to Global Leadership Initiative on Malnutrition reclassification, n (%)/median (Q1, Q3)

Variable	Non-reclassified patients	Reclassified patients	P value
Length of hospital stay, days	9.00 (6.00-13.00)	19.00 (16.00-19.00)	0.003
White blood cell count, × 109/L	2.68 (2.02-3.66)	3.70 (2.85-3.94)	0.181
Lymphocyte count, × 109/L	0.58 (0.38-0.90)	0.45 (0.42-0.70)	0.548
Prealbumin, mg/L	93.50 (63.70-115.00)	57.00 (53.00-62.10)	0.019
Albumin, g/L	26.39 (24.19-30.47)	20.31 (20.10-21.02)	0.001
Nutritional intervention during hospitalization	24/95 (25.3)	1/5 (20.0)	1.000
Admission complications	73/95 (76.8)	5/5 (100.0)	0.583
Readmission	67/95 (70.5)	2/5 (40.0)	0.171
One-year mortality	11/76 (14.5)	1/4 (25.0)	0.485

Global Leadership Initiative on Malnutrition (GLIM) reclassification was defined as patients who were not classified as having GLIM-defined malnutrition using the actual body mass index but were reclassified as having GLIM-defined malnutrition using the computed tomography-sensitive corrected body mass index. Continuous variables were compared using the Mann-Whitney U test, and categorical variables were compared using Fisher’s exact test. Patients lost to follow-up were excluded from the one-year mortality analysis.