Carteri RB, Marroni CA, Ferreira LF, Pinto LP, Czermainski J, Tovo CV, Fernandes SA. Do Child–Turcotte–Pugh and nutritional assessments predict survival in cirrhosis: A longitudinal study. World J Hepatol 2025; 17(1): 99183 [DOI: 10.4254/wjh.v17.i1.99183]
Corresponding Author of This Article
Sabrina A Fernandes, PhD, Postdoc, Professor, Researcher, Postgraduate in Hepatology, Universidade Federal de Ciências da Saúde de Porto Alegre, Sarmento Leite 245, Porto Alegre 90050-170, Brazil. sabrinaafernandes@gmail.com
Research Domain of This Article
Gastroenterology & Hepatology
Article-Type of This Article
Retrospective Cohort Study
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Randhall B Carteri, Department of Nutrition, Centro Universitário CESUCA, Cachoeirinha 94935-630, Brazil
Randhall B Carteri, Claudio A Marroni, Luis F Ferreira, Letícia P Pinto, Juliana Czermainski, Cristiane V Tovo, Sabrina A Fernandes, Postgraduate in Hepatology, Universidade Federal de Ciências da Saúde de Porto Alegre, Porto Alegre 90050-170, Brazil
Luis F Ferreira, School of Electronics, Electrical Engineering and Computer Science, Queen’s University Belfast, Belfast BT9 5BN, Northern Ireland, United Kingdom
Author contributions: Carteri RB contributed to the methodological development and material support, collected data, interpreted results, conducted data analysis, wrote and revised the manuscript; Tovo CV and Marroni CA contributed to the conception and critical review of the manuscript; Pinto LP collected data and interpreted the results; Czermainski J contributed to the methodological development and material support; Ferreira LF conducted data analysis and interpreted the results; Fernandes SA designed the research project, collaborated in writing, and critically reviewed the manuscript; All authors have read and approved the final manuscript.
Institutional review board statement: This study was conducted in accordance with the Helsinki Declaration and was approved by the ethics and research committee of the Federal University of Health Sciences of Porto Alegre, under the number: 5203619. All participants signed the Informed Consent Form (TCLE) in advance.
Informed consent statement: All study participants or their legal guardian provided informed written consent about personal and medical data collection prior to study enrolment.
Conflict-of-interest statement: The authors declare that they have no conflict of interest.
Data sharing statement: No additional data is available for sharing.
STROBE statement: The authors have read the STROBE Statement—a checklist of items, and the manuscript was prepared and revised according to the STROBE Statement-a checklist of items.
Open-Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/
Corresponding author: Sabrina A Fernandes, PhD, Postdoc, Professor, Researcher, Postgraduate in Hepatology, Universidade Federal de Ciências da Saúde de Porto Alegre, Sarmento Leite 245, Porto Alegre 90050-170, Brazil. sabrinaafernandes@gmail.com
Received: July 16, 2024 Revised: October 23, 2024 Accepted: December 6, 2024 Published online: January 27, 2025 Processing time: 174 Days and 8.9 Hours
Abstract
BACKGROUND
Cirrhotic patients face heightened energy demands, leading to rapid glycogen depletion, protein degradation, oxidative stress, and inflammation, which drive disease progression and complications. These disruptions cause cellular damage and parenchymal changes, resulting in vascular alterations, portal hypertension, and liver dysfunction, significantly affecting patient prognosis.
AIM
To analyze the association between Child–Turcotte–Pugh (CTP) scores and different nutritional indicators with survival in a 15-year follow-up cohort.
METHODS
This was a retrospective cohort study with 129 cirrhotic patients of both sexes aged > 18 years. Diagnosis of cirrhosis was made by liver biopsy. The first year of data collection was 2007, and data regarding outcomes were collected in 2023. Data were gathered from medical records, and grouped by different methods, including CTP, handgrip strength, and triceps skinfold cutoffs. The prognostic values for mortality were assessed using Kaplan–Meier curves and multivariate binary logistic regression models.
RESULTS
The coefficient for CTP was the only statistically significant variable (Wald = 5.193, P = 0.023). This suggests that with a negative change in CTP classification score, the odds of survival decrease 52.6%. The other evaluated variables did not significantly predict survival outcomes in the model. Kaplan–Meier survival curves also indicated that CTP classification was the only significant predictor.
CONCLUSION
Although different classifications showed specific differences in stratification, only CTP showed significant predictive potential. CTP score remains a simple and effective predictive tool for cirrhotic patients even after longer follow-up.
Core Tip: Cirrhosis involves destroying hepatic cells, leading to metabolic changes that disrupt the body’s homeostasis. These alterations adversely affect the patient’s clinical condition and prognosis. Identifying a parameter that can predict significant events is crucial for a more precise approach, reducing mortality and enhancing the quality of life. This study reinforces the predictive value of Child–Turcotte–Pugh for predicting the clinical status of those with chronic liver disease.
Citation: Carteri RB, Marroni CA, Ferreira LF, Pinto LP, Czermainski J, Tovo CV, Fernandes SA. Do Child–Turcotte–Pugh and nutritional assessments predict survival in cirrhosis: A longitudinal study. World J Hepatol 2025; 17(1): 99183
In 2019, cirrhosis caused 2.4% of global deaths, primarily due to hepatitis C, followed by alcohol-related liver disease. However, with better viral hepatitis management and rising obesity and alcohol consumption, cirrhosis cases linked to metabolic-associated steatotic liver disease and alcohol are increasing, while mortality from other causes has declined. Deaths from advanced chronic liver disease are expected to more than double between 2016 and 2030[1,2].
Cirrhotic patients experience increased energy demand, leading to rapid muscle and glycogen depletion, heightened protein degradation, oxidative stress, and inflammation; all of which contribute to disease progression and its complications. This disruption in body homeostasis causes a parenchymal rearrangement, direct related to cellular damage, vascular alterations, portal hypertension, and hepatocellular dysfunction, significantly affecting patient prognosis[3,4].
Monitoring chronic liver disease progression is challenging due to its invasiveness, cost, and data inconsistencies among professionals and healthcare services. This underscores the need for new tools, such as a temporal prognostic index based on the natural history of the disease. Developing reliable predictive tools for chronic diseases like cirrhosis remains difficult. The Child–Turcotte–Pugh (CTP) score, a traditional prognostic tool, is debated for its limited variables and several limitations, including empirically selected components, arbitrary cutoff values, equal weighting of variables, subjective definitions, and exclusion of key prognostic factors such as renal function and markers of portal hypertension[5].
Cirrhotic patients exhibit significant bioenergetic and macronutrient metabolic changes, leading to protein depletion and fluid retention, even in early disease stages (CTP score A)[6]. Disease decompensation is often associated with clinical signs such as ascites, muscle loss, edema, and hepatic encephalopathy, complicating the nutritional diagnosis. These limitations highlight the need for improved prognostic tools that consider the multifaceted nature of chronic liver disease[2,7].
From 2008 to 2014, hospitalization costs for cirrhotic patients increased by 30.2%, with a 36% rise in hospitalizations of compensated cirrhotic patients and a 24% increase in decompensated patients. In 2014, decompensated cirrhosis accounted for 58.6% of total cirrhosis hospitalizations, driven by costlier procedures and clinical complications, which rose by 15%–152%[8]. Thus, cost-effective methods for predicting outcomes are essential. Given the role of malnutrition in chronic liver diseases, nutritional indicators such as body mass index (BMI), triceps skinfold (TSF), and arm circumference (AC) could serve as valuable prognostic tools[9,10].
Therefore, we conducted a cohort study of cirrhotic patients with different etiologies to capture longitudinal trajectories over 15 years. We analyzed the association between CTP scores and different nutritional indicators with survival. Additionally, we classified the patients in accordance with different indicators, and associations with the main complications of cirrhosis, such as ascites, encephalopathy, spontaneous bacterial peritonitis, and bleeding esophageal varices.
MATERIALS AND METHODS
Study population
This was a retrospective cohort study of patients of both genders, aged > 18 years, with cirrhosis, who were treated at the gastroenterology clinic of a hospital complex in Porto Alegre, Brazil. The diagnosis of cirrhosis was confirmed through liver biopsy histology, imaging studies, and biological markers. Exclusion criteria included patients with intestinal malabsorption, AIDS, chronic kidney failure, those receiving enteral nutrition, upper limb neuromuscular disorders, chronic pancreatitis, chronic diarrhea, and mental and/or cognitive impairments. In the survival analysis, patients who underwent transplantation were included up until the time of their transplant. This study adhered to the Helsinki Declaration and received approval from the ethics and research committee of the Federal University of Health Sciences of Porto Alegre, Brazil, under the approval number: 5203619. All participants provided written informed consent prior to participation.
Data collection
Data collection in the first year occurred in January 2007, and data regarding the outcomes were gathered in January 2023, during ambulatory evaluations. Demographic data such as age and sex were collected. The clinical data on patient outcomes included the number of hospitalizations, ascites, paracentesis, encephalopathy, blood pressure, and hepatocellular carcinoma (HCC). The staging of cirrhosis was assessed through the CTP and model for end-stage liver disease (MELD) scores. The laboratory tests considered in this study were those conducted within 3 mo before or after the evaluation, including albumin, aspartate aminotransferase, alanine aminotransferase, g-glutamyl transferase, alkaline phosphatase, bilirubin levels (data not shown), and the international normalized ratio (INR) used to calculate MELD. The length of hospitalization was determined by calculating the period between the admission and discharge dates. Body mass and height were measured using a scale and stadiometer (Filizola, Brazil) and BMI was calculated to classify the athletes according to the criteria of the World Health Organization[11,12]. Anthropometric measurements included the TSF measured using a Cescorf skinfold caliper, and AC measured with a graduated, nonretracting, flexible measuring tape using standard protocols[11,12]. Mid-arm muscle circumference (MAC) was calculated from AC and TSF using the formula: MAC = AC - (3.1415 × TSF), and the cutoff points were determined as proposed by Saueressig et al[13]. The grouping of TSF was as follows: TSF 1 = patients in the 5th and 10th percentiles; TSF 2 = patients in the 25th, 50th, and 75th percentiles; and TSF 3 = patients in the 90th and 95th percentiles[14]. An adjustable-handled mechanical handgrip dynamometer (Baseline Smedley Spring, New York, USA) was used to measure handgrip strength (HS). The highest value recorded across the three attempts was taken as the final result, and the measurement procedures were carried following the recommended by the American Association of Hand Therapists[15]. The grouping of HS was done in below (HS 1) or above (HS 2) the cutoff points, as proposed by Budziareck et al[16], by excluding patients below 5th and above 95th percentile (outliers), and using one standard deviation below the mean as low cutoff (HS 1), and one standard deviation above the mean as the high cutoff (HS 2).
Statistical analysis
Continuous variables are expressed as mean ± SD, while categorical variables are reported as both absolute and relative frequencies (n and %). Group comparisons were conducted using one-way analysis of variance with Tukey’s post hoc analysis, or Student’s t test for independent samples, depending on the context. The relationship between independent variables and survival likelihood was also explored with logistic regression analysis, and odds ratio (OR) % was calculated as (OR – 1) × 100%. To evaluate the time-to-event outcomes, we used Kaplan–Meier survival analysis, and patients who did not experience the event by the end of the study period were censored at the time of their last follow-up. Survival curves were generated for different variables and the log-rank test was used to compare the survival distributions between groups. All tests were made using statistical product and service solutions software, version 25.0 (IBM Corp., Armonk, NY, USA).
RESULTS
Most patients died during the 15-year follow-up (Table 1). Even with bad outcomes, with most patients being overweight, and a third of the sample having systemic arterial hypertension and/or diabetes mellitus, < 10% of the patients developed HCC.
Table 1 Sample characterization (n = 129), mean ± SD/ n (%).
Descriptive variables
Gender
Male
73 (56.59)
Female
56 (43.41)
Death
Yes
74 (57.36)
No
55 (42.64)
Transplanted
No
111 (86.05)
Yes
18 (13.95)
Cirrhosis etiology
Alcohol
31 (24.03)
Autoimmune
5 (3.88)
Cryptogenic NASH
5 (3.88)
Virus C + alcohol
14 (10.85)
Virus B + C + alcohol
12 (9.3)
Virus B + alcohol
1 (0.78)
Virus C
1 (0.78)
Virus B
56 (43.41)
Other
4 (3.1)
Child–Turcotte–Pugh
1
85 (65.89)
2
33 (25.58)
3
11 (8.53)
Continuous variables
Age (yr)
56.32 ± 11.252
Survival time (mo)
110.901 ± 73.139
Survival time (yr)
9.115 ± 6.011
Height (cm)
164.438 ± 8.979
First assessment weight (kg)
74.467 ± 14.692
Body mass index
27.528 ± 4.85
Arm circumference (cm)
30.538 ± 4.237
Muscle arm circumference (cm/mm)
24.791 ± 3.221
Bioimpedance analysis-phase angle (°)
6.62 ± 2.953
MELD
10.987 ± 4.699
In the TSF classification, it was possible to identify some significant differences, such as gender, where the vast majority of men were in the TSF 3 classification, while the majority of women were in TSF 2. In BMI, the vast majority of overweight patients were TSF 3, while the majority of underweight were TSF 1.
The continuous data are presented in Tables 2 and 3, following the same categorizations in TSF and HS. Body weight, BMI, AC, and MAC were significantly higher in TSF 3, while bioimpedance (BIA) resistance and urea levels were higher in TSF 2. In the categorization by HS, all significant data verified higher values in HS 1, namely, height, weight, AC, MAC and sodium levels. When grouping by HS, it was possible to verify only two significant differences: in MAC, where patients in HS 1 presented > 80% below the cutoff point, while only two-thirds of HS 2 showed the same behavior, and, in HCC, nine cases were identified in HS 1 (one-seventh of the sample), and only two in HS 2 (a little more than 2%).
Table 2 Sample characterization categorized by Triceps skinfold and Child–Turcotte–Pugh score (n = 129), mean ± SD/n (%).
Descriptive variables
Categorized by TSF
P value
Categorized by Child–Turcotte–Pugh
P value
1
2
3
1
2
3
24 (18.6)
53 (41.09)
52 (40.31)
85 (65.89)
33 (25.58)
11 (8.53)
Gender
0.005
0.404
Male
9 (37.5)
26 (49.06)
38 (73.08)
45 (52.94)
20 (60.61)
8 (72.73)
Female
15 (62.5)
27 (50.94)
14 (26.92)
40 (47.06)
13 (39.39)
3 (27.27)
Death
0.817
0.013
Yes
15 (62.5)
29 (54.72)
30 (57.69)
41 (48.24)
25 (75.76)
8 (72.73)
No
9 (37.5)
24 (45.28)
22 (42.31)
44 (51.76)
8 (24.24)
3 (27.27)
Transplanted
0.683
0.025
No
22 (91.67)
45 (84.91)
44 (84.62)
78 (91.76)
24 (72.73)
9 (81.82)
Yes
2 (8.33)
8 (15.09)
8 (15.38)
7 (8.24)
9 (27.27)
2 (18.18)
Cirrhosis etiology
0.981
0.178
Alcohol
4 (16.67)
13 (24.53)
14 (26.92)
25 (29.41)
4 (12.12)
2 (18.18)
Autoimmune
0 (0)
3 (5.66)
2 (3.85)
5 (5.88)
0 (0)
0 (0)
Cryptogenic-NASH
2 (8.33)
2 (3.77)
1 (1.92)
3 (3.53)
2 (6.06)
0 (0)
Virus C + alcohol
4 (16.67)
4 (7.55)
6 (11.54)
9 (10.59)
5 (15.15)
0 (0)
Virus B + C + alcohol
3 (12.5)
6 (11.32)
3 (5.77)
5 (5.88)
4 (12.12)
3 (27.27)
Virus B + alcohol
1 (4.17)
0 (0)
0 (0)
0 (0)
0 (0)
1 (9.09)
Virus C
0 (0)
0 (0)
1 (1.92)
0 (0)
1 (3.03)
0 (0)
Virus B
10 (41.67)
23 (43.4)
23 (44.23)
36 (42.35)
15 (45.45)
5 (45.45)
Other
0 (0)
2 (3.77)
2 (3.85)
2 (2.35)
2 (6.06)
0 (0)
Ascites
0.519
0.001
Yes
12 (50)
34 (64.15)
28 (53.85)
40 (47.06)
25 (75.76)
9 (81.82)
No
12 (50)
19 (35.85)
23 (44.23)
45 (52.94)
7 (21.21)
2 (18.18)
Paracentesis
0.980
0.311
Yes
8 (33.33)
17 (32.08)
14 (26.92)
20 (23.53)
13 (39.39)
6 (54.55)
No
16 (66.67)
35 (66.04)
36 (69.23)
62 (72.94)
20 (60.61)
5 (45.45)
Esophageal varices
0.679
0.599
Yes
19 (79.17)
41 (77.36)
37 (71.15)
62 (72.94)
27 (81.82)
8 (72.73)
No
5 (20.83)
12 (22.64)
15 (28.85)
23 (27.06)
6 (18.18)
3 (27.27)
Varices bleeding
0.276
0.457
Yes
4 (16.67)
7 (13.21)
3 (5.77)
10 (11.76)
3 (9.09)
1 (9.09)
No
19 (79.17)
46 (86.79)
48 (92.31)
75 (88.24)
28 (84.85)
10 (90.91)
Encephalopathy
0.476
0.001
Yes
10 (41.67)
20 (37.74)
15 (28.85)
22 (25.88)
14 (42.42)
9 (81.82)
No
14 (58.33)
33 (62.26)
37 (71.15)
63 (74.12)
19 (57.58)
2 (18.18)
Encephalopathy level
0.299
0.001
0
14 (58.33)
33 (62.26)
37 (71.15)
63 (74.12)
19 (57.58)
2 (18.18)
1
4 (16.67)
6 (11.32)
8 (15.38)
11 (12.94)
4 (12.12)
3 (27.27)
2
2 (8.33)
11 (20.75)
5 (9.62)
6 (7.06)
8 (24.24)
4 (36.36)
3
4 (16.67)
3 (5.66)
1 (1.92)
4 (4.71)
2 (6.06)
2 (18.18)
4
0 (0)
0 (0)
1 (1.92)
1 (1.18)
0 (0)
0 (0)
Developed HCC
0.079
0.576
Yes
21 (87.5)
52 (98.11)
45 (86.54)
77 (90.59)
30 (90.91)
11 (100)
No
3 (12.5)
1 (1.89)
7 (13.46)
8 (9.41)
3 (9.09)
0 (0)
Continuous variables
TSF
P value
Child-Turcotte-Pugh
P value
1
2
3
1
2
3
24 (18.6)
53 (41.09)
52 (40.31)
85 (65.89)
33 (25.58)
11 (8.53)
Age (yr)
57.671 ± 8.837
56.156 ± 12.206
55.865 ± 11.382
0.804
55.942 ± 11.825
58.424 ± 10.417
52.937 ± 8.385
0.328
Survival time (mo)
98.715 ± 78.11
116.068 ± 74.444
111.258 ± 70.185
0.631
128.237 ± 68.005
81.32 ± 70.682
65.676 ± 77.103
0.001
Survival time (yr)
8.114 ± 6.42
9.54 ± 6.119
9.144 ± 5.769
0.631
10.54 ± 5.589
6.684 ± 5.809
5.398 ± 6.337
0.001
Height (cm)
161.333 ± 8.731
163.83 ± 8.733
166.49 ± 9
0.053
164.082 ± 8.597
164.955 ± 9.518
165.636 ± 10.847
0.806
First assessment weight (kg)
74.3 ± 10.434
66.517 ± 10.935
82.648 ± 15.389
0.000
72.445 ± 12.54
78.991 ± 18.539
76.527 ± 15.337
0.083
Body mass index
28.574 ± 4.052
24.816 ± 3.76
29.811 ± 4.872
0.000
26.868 ± 4.17
29.131 ± 6.005
27.821 ± 5.228
0.073
Arm circumference (cm)
31.222 ± 3.001
27.674 ± 3.968
33 ± 3.269
0.000
30.556 ± 4.122
30.778 ± 4.3
28.3 ± 5.078
0.262
Hospital admissions (n)
1.542 ± 2.167
1.811 ± 2.426
1.942 ± 2.653
0.807
1.576 ± 2.291
2.394 ± 3.02
1.909 ± 1.64
0.269
Length of stay (d)
9.94 ± 11.803
5.985 ± 8.494
5.751 ± 7.901
0.137
6.588 ± 9.23
6.692 ± 8.481
6.727 ± 10.031
0.998
MAC (cm/mm)
25.032 ± 2.515
23.975 ± 3.596
25.511 ± 2.957
0.045
24.779 ± 3.319
24.853 ± 2.903
24.696 ± 3.64
0.989
BIA-phase angle (°)
6.82 ± 3.912
6.348 ± 2.777
6.805 ± 2.645
0.686
6.903 ± 3.249
6.275 ± 2.42
5.474 ± 1.323
0.238
MELD
12.444 ± 5.217
10.704 ± 5.125
10.62 ± 3.899
0.350
8.647 ± 3.513
13.683 ± 3.028
17.208 ± 5.302
0.000
Table 3 Sample characterization categorized by subjective global assessment, muscle-arm circumference and handgrip strength (n = 129), mean ± SD/n (%).
Descriptive variables
SGA
P value
MAC
P value
Handgrip strength
P value
1
2
1
2
1
2
124 (96.12)
5 (3.88)
93 (72.09)
36 (27.91)
53 (41.09)
76 (58.91)
Gender
0.699
0.599
0.016
Male
70 (56.45)
3 (60)
45 (48.39)
28 (77.78)
34 (64.15)
39 (51.32)
Female
54 (43.55)
2 (40)
48 (51.61)
8 (22.22)
19 (35.85)
37 (48.68)
Death
0.850
0.640
0.204
Yes
72 (58.06)
2 (40)
54 (58.06)
20 (55.56)
32 (60.38)
42 (55.26)
No
52 (41.94)
3 (60)
39 (41.94)
16 (44.44)
21 (39.62)
34 (44.74)
Transplanted
0.029
0.000
0.642
No
106 (85.48)
5 (100)
82 (88.17)
29 (80.56)
46 (86.79)
65 (85.53)
Yes
18 (14.52)
0 (0)
11 (11.83)
7 (19.44)
7 (13.21)
11 (14.47)
Cirrhosis etiology
0.002
0.539
0.346
Alcohol
30 (24.19)
1 (20)
21 (22.58)
10 (27.78)
12 (22.64)
19 (25)
Autoimmune
4 (3.23)
1 (20)
5 (5.38)
0 (0)
2 (3.77)
3 (3.95)
Cryptogenic-NASH
4 (3.23)
1 (20)
5 (5.38)
0 (0)
5 (9.43)
0 (0)
Virus C + alcohol
12 (9.68)
2 (40)
11 (11.83)
3 (8.33)
5 (9.43)
9 (11.84)
Virus B + C + alcohol
12 (9.68)
0 (0)
7 (7.53)
5 (13.89)
6 (11.32)
6 (7.89)
Virus B + alcohol
1 (0.81)
0 (0)
0 (0)
1 (2.78)
0 (0)
1 (1.32)
Virus C
1 (0.81)
0 (0)
1 (1.08)
0 (0)
1 (1.89)
0 (0)
Virus B
56 (45.16)
0 (0)
41 (44.09)
15 (41.67)
22 (41.51)
34 (44.74)
Other
4 (3.23)
0 (0)
2 (2.15)
2 (5.56)
0 (0)
4 (5.26)
Ascites
0.009
0.000
0.295
Yes
73 (58.87)
1 (20)
51 (54.84)
23 (63.89)
33 (62.26)
41 (53.95)
No
50 (40.32)
4 (80)
41 (44.09)
13 (36.11)
19 (35.85)
35 (46.05)
Paracentesis
0.000
0.789
0.453
Yes
39 (31.45)
0 (0)
27 (29.03)
12 (33.33)
18 (33.96)
21 (27.63)
No
82 (66.13)
5 (100)
64 (68.82)
23 (63.89)
34 (64.15)
53 (69.74)
Esophageal varices
0.260
0.004
0.006
Yes
94 (75.81)
3 (60)
70 (75.27)
27 (75)
43 (81.13)
54 (71.05)
No
30 (24.19)
2 (40)
23 (24.73)
9 (25)
10 (18.87)
22 (28.95)
Varices bleeding
0.067
0.095
0.214
Yes
14 (11.29)
0 (0)
10 (10.75)
4 (11.11)
4 (7.55)
10 (13.16)
No
108 (87.1)
5 (100)
81 (87.1)
32 (88.89)
48 (90.57)
65 (85.53)
Encephalopathy
0.000
0.007
0.215
Yes
45 (36.29)
0 (0)
33 (35.48)
12 (33.33)
17 (32.08)
28 (36.84)
No
79 (63.71)
5 (100)
60 (64.52)
24 (66.67)
36 (67.92)
48 (63.16)
Encephalopathy level
0.001
0.256
0.555
0
79 (63.71)
5 (100)
60 (64.52)
24 (66.67)
36 (67.92)
48 (63.16)
1
18 (14.52)
0 (0)
13 (13.98)
5 (13.89)
6 (11.32)
12 (15.79)
2
18 (14.52)
0 (0)
13 (13.98)
5 (13.89)
8 (15.09)
10 (13.16)
3
8 (6.45)
0 (0)
6 (6.45)
2 (5.56)
3 (5.66)
5 (6.58)
4
1 (0.81)
0 (0)
1 (1.08)
0 (0)
0 (0)
1 (1.32)
Developed HCC
0.105
0.126
0
Yes
114 (91.94)
4 (80)
83 (89.25)
35 (97.22)
44 (83.02)
74 (97.37)
No
10 (8.06)
1 (20)
10 (10.75)
1 (2.78)
9 (16.98)
2 (2.63)
Continuous variables
SGA
P value
MAC
P value
Handgrip strength
P value
1
2
1
2
1
2
124 (96.12)
5 (3.88)
93 (72.09)
36 (27.91)
53 (41.09)
76 (58.91)
Age (yr)
56.924 ± 10.633
41.355 ± 16.836
0.262
56.966 ± 11.719
54.651 ± 9.903
0.360
55.802 ± 11.576
56.789 ± 11.119
0.735
Survival time (mo)
109.674 ± 73.255
141.327 ± 70.428
0.105
108.091 ± 74.661
118.159 ± 69.541
0.139
104.957 ± 72.813
116.397 ± 73.1
0.945
Survival time (yr)
9.014 ± 6.021
11.616 ± 5.789
0.105
8.884 ± 6.136
9.712 ± 5.716
0.139
8.627 ± 5.985
9.567 ± 6.008
0.945
Height (cm)
164.444 ± 9.069
164.3 ± 7.12
0.200
164.102 ± 9.222
165.306 ± 8.38
0.419
166.84 ± 8.149
162.947 ± 9.121
0.242
First assessment weight (kg)
74.963 ± 14.722
62.18 ± 6.82
0.106
73.43 ± 14.914
77.147 ± 13.947
0.950
78.138 ± 14.441
71.853 ± 14.499
0.412
Body mass index
27.706 ± 4.836
23.114 ± 2.845
0.250
27.212 ± 4.794
28.347 ± 4.965
0.823
27.921 ± 3.898
27.161 ± 5.399
0.076
Arm circumference (cm)
30.579 ± 4.286
26.8 ± 2.28
0.116
30.155 ± 4.439
30.966 ± 3.887
0.447
32.095 ± 3.427
29.123 ± 4.472
0.060
Hospital admissions
1.887 ± 2.483
0.027
1.634 ± 2.254
2.278 ± 2.914
0.054
1.792 ± 2.634
1.813 ± 2.364
0.351
Length of stay (d)
6.894 ± 9.123
0.019
6.843 ± 9.741
6.068 ± 7.011
0.411
6.396 ± 8.498
6.758 ± 9.515
0.925
MAC (cm/mm)
24.895 ± 3.232
22.202 ± 1.43
0.085
24.527 ± 3.377
25.472 ± 2.702
0.139
25.92 ± 3.176
24.011 ± 3.049
0.590
BIA-phase angle (°)
6.625 ± 3.007
6.49 ± 1.023
0.373
6.612 ± 2.802
6.641 ± 3.355
0.745
6.839 ± 2.923
6.471 ± 3.004
0.818
MELD
11.096 ± 4.607
8.41 ± 6.854
0.370
10.872 ± 4.954
11.294 ± 4.009
0.431
10.378 ± 4.136
11.353 ± 5.069
0.206
Table 4 presents the logistic regression coefficients, SE, Wald statistics, P values, and OR with 95% confidence interval (CI) for each predictor variable. The coefficient for CTP was -07.46, which was the only statistically significant (Wald = 5.193, P = 0.023). This suggests that with a negative change in the CTP classification score, the odds of survival decrease by 52.6%. The other evaluated variables did not significantly predict survival outcomes in the model. The survival experience of patients was analyzed using Kaplan–Meier survival curves (Figure 1). The CTP classification was the only significant predictor. The median survival time for group 1 was 129.99 ± 7.49 (95%CI: 115.295–144.676), compared to 82.178 ± 12.34 (95%CI: 57.991–106.365) for group 2 and 76.224 ± 26.738 (95%CI: 23.819–128.63) for group 3, respectively. The log-rank test was used to compare survival distributions between the groups, showing significant differences (χ² = 12.45, P = 0.002).
This study assessed 129 cirrhotic patients to identify potential tools for predicting outcomes over 15 years. Although different classifications showed specific differences in patients’ stratification, only CTP showed significant predictive potential in the present study.
The typical progression of cirrhosis in patients shows an average survival time of around 10 years. However, depending on individual clinical conditions, some patients may experience specific complications of cirrhosis that adversely affect morbidity and mortality[12-14]. Early diagnosis or treatment of these complications can extend survival, enhance quality of life, and lower the mortality rate for those awaiting liver transplantation.
The demographic characteristics of the population in this study are consistent with those observed in Belarmino’s research, which involved 134 cirrhotic patients with an average age of 54.3 years (SD 10.10 years), predominantly male[17]. Similar demographics were reported in Ruiz-Margáin et al’s cohort[18] which included 136 cirrhotic patients, again mostly male, with an average age of 54.5 years (SD 10.0 years). However, it is important to note that the population in our study presented with compensated cirrhosis. In contrast, Belarmino et al[17], reported that only 18% of their patients were classified as CTP A, 55% as B, and 27% as C, with an average MELD score of 14.11 (SD 4.95). Ruiz-Margáin et al’s study featured a homogeneous sample according to the CTP classification, with 34.1% in class B, 33.3% in class C, and an average MELD score of 14 (SD 6)[18].
Malnutrition in cirrhosis occurs in 5%–92% of patients and needs more attention by healthcare professionals, since it is associated with increased mortality risk[19]. Nutritional assessment in cirrhosis is challenging due to the absence of validated screening tools and the impact of liver dysfunction on body composition measurements[20]. Estimating the prognosis of cirrhosis patients is complex due to variations in disease etiologies, presence of portal hypertension, liver function, and disease reversibility. Conventional prognostic scoring systems, like the CTP score and MELD are commonly used; although revised models seeking to enhance prognostic accuracy often omit nutritional status evaluations despite malnutrition being a prevalent complication affecting cirrhosis patient survival. However, the clinical relevance of these cutoffs in relation to standard malnutrition definitions in cirrhosis remains unclear due to the lack of a gold standard method. BIA has demonstrated a significant correlation with CTP scores, with a phase angle of 5.44° suggested as a new parameter for nutritional classification[12,21]. However, the effectiveness of BIA in assessing malnutrition in cirrhosis has been questioned[22]. While some studies found no association between malnutrition and disease severity using Child–Pugh scores, others reported significant correlations between HS, MAC, and disease severity[22]. Importantly, malnutrition and hypermetabolism have been identified as potentially treatable presurgical factors that can adversely affect survival of these patients[23,24].
For instance, HS is a simple and inexpensive measure of muscle strength that has been shown to be a strong predictor of mortality in cirrhotic patients. HS is independently associated with increased mortality, even after adjusting for other prognostic factors such as age, sex, CTP score, and MELD score[25], and studies have shown that HS consistently identifies the highest prevalence of malnutrition in cirrhotic patients, ranging from 58.8% to 88.8%[22].
One particular study focused on assessing malnutrition in nonhospitalized cirrhotic patients using MAC and HS[26]. Among 352 cirrhotic patients, 56% were malnourished according to the subjective global assessment (SGA). Malnutrition prevalence was 27% based on MAC and 42% based on HS. The HS strength had the highest diagnostic accuracy for malnutrition with an area under the curve of 0.82 (95%CI: 0.78–0.86, P = 0.001), compared to MAC 0.60 (95%CI: 0.55–0.64, P = 0.001). Thus, the HS strength showed a sensitivity of 67%, specificity of 95%, and diagnostic accuracy of 87%, making it a superior predictor of malnutrition in cirrhotic patients compared to MAC. Similarly, a study by Alvares-da-Silva and Reverbel da Silveira[27] with 50 cirrhotic patients with major complications (where 88% of patients were CTP A and 12% were CTP B) showed that malnutrition prevalence in cirrhosis was 28% by SGA, 18.7% by prognostic nutritional index, and 63% by HS. The HS was the only method predicting poorer clinical outcomes, with 65.5% of malnourished patients developing major complications compared to 11.8% of well-nourished patients. No significant differences in transplantation or death rates were observed.
While these studies suggest that HS is a valuable prognostic predictor for cirrhotic patients, in our 15-year follow-up, only CTP score showed predictive potential for survival. The CTP classification was historically used for liver transplant allocation but had three major limitations: (1) It involved subjective grading of ascites and encephalopathy; (2) it did not consider renal function; and (3) it provided only 10 discrete scores[28]. This last limitation was particularly problematic as it failed to adequately differentiate patients based on disease severity, making wait time a significant factor in prioritization. For example, a patient with an INR of 6 and bilirubin of 14 could receive the same score as a patient with an INR of 2.3 and bilirubin of 4.0[28]. To address these issues, the MELD score was developed, offering a broader range of continuous variable values. Initially, the MELD score included bilirubin level, creatinine level, INR, and the cause of liver disease[28]. It has since evolved to exclude the cause of liver disease and incorporate serum sodium levels and whether the patient has renal impairment. A recent study aimed to validate and compare the performance of several prognostic scoring systems for predicting in-hospital outcomes in 330 cirrhotic patients with acute variceal bleeding where only the clinical Rockall score and CTP score showed clinically acceptable discriminative ability for predicting in-hospital rebleeding, while CTP and other scores were effective for predicting in-hospital mortality[29].
In contrast, Chawla et al[30] compared the effectiveness of MELD, CTP, and the creatinine-modified CTP score in predicting mortality among 102 Indian patients with cirrhosis, showing that all scores had excellent diagnostic accuracy for predicting mortality (c-statistics > 0.85). However, MELD was more accurate than CTP for predicting short-term and intermediate-term mortality in cirrhosis patients in this particular study. Importantly, a systematic review and meta-analysis including 119 studies comprising 269 comparisons: 44 favored MELD, 16 favored CTP, while others did not report statistical significance. Also, 42 papers were included in the meta-analysis, showing discrepant results. In acute-on-chronic liver failure patients, CTP had higher sensitivity but lower specificity than MELD. In critical patients, MELD had a smaller negative likelihood ratio and higher sensitivity than Child–Pugh score. For surgical patients, CTP had higher specificity than MELD. The authors concluded that while CTP and MELD scores generally offer similar prognostic values, but their efficacy varies in specific conditions. A study by Boursier et al[31] evaluating the accuracy of the CTP score, MELD, and the MELD-serum sodium (Na) score (which combines MELD and Na) in predicting 6-month mortality in 308 cirrhotic patients showed no significant differences in predicting survival during the 6-month follow-up. However, significant differences were observed comparing CTP score with MELD-Na in the decompensated cirrhotic patient’s subgroup analysis. The authors concluded that, while the CTP score remains a simple and effective predictive tool for cirrhotic patients, MELD and MELD-Na were more suitable for those with decompensated cirrhosis.
Although the superiority of the MELD score over the CTP score for predicting outcomes in chronic liver disease patients is still debated, the present study in an open-world setting with a 15-year follow-up showed that CTP was the best predictive tool, reinforcing its importance in the assessment of cirrhotic patients. Notably, studies show that MAC, HS, and SGA may predict severity and short-term survival in cirrhotic patients, while Child–Pugh score and MELD score are robust predictors of mortality in end-stage liver disease patients[13,32,33]. Child–Pugh score has a higher sensitivity and lower specificity than MELD score in assessing the prognosis of liver cirrhosis patients, and the MELD score can present accuracy decrease for long-term predictions which can explain the results of the present study[34,35].
This study exhibits several strengths and weaknesses worth noting. On the positive side, the study’s 15-year follow-up period allows for a comprehensive longitudinal analysis, providing valuable insights into the long-term outcomes of patients. Additionally, the sample size is robust, enhancing the statistical power and reliability of the findings. The inclusion of a variety of etiologies adds to the generalizability of the results, making them applicable to a broader patient population. Furthermore, the utilization of real-world data ensures that the findings are relevant and reflective of actual clinical scenarios. However, there are notable weaknesses as well. One significant challenge is the difficulty in obtaining complete and consistent data from all patients over such an extended period, which can lead to potential biases and missing data issues. This limitation may affect the accuracy and completeness of the study’s conclusions. Also, the present study excluded patients with comorbid conditions such as intestinal malabsorption and chronic kidney failure, which can limit the generalizability of findings, since such comorbidities significantly impact nutritional status and outcomes, which highlights the need for inclusive and comprehensive assessment strategies. Despite these challenges, the study’s extensive follow-up, substantial sample size, diverse etiologies, and real-world data contribute significantly to its value in advancing our understanding of the disease.
CONCLUSION
This study assessed 129 cirrhotic patients to identify the best method for comparisons of patients’ characteristics as well as potential tools for predicting outcomes over a 15-year period. Although different classifications showed specific differences in patients’ stratification, only CTP showed significant predictive potential in the present study. Therefore, CTP score remains a simple and effective predictive tool for cirrhotic patients even in a longer follow-up period.
Footnotes
Provenance and peer review: Invited article; Externally peer reviewed.
Peer-review model: Single blind
Specialty type: Gastroenterology and hepatology
Country of origin: Brazil
Peer-review report’s classification
Scientific Quality: Grade C
Novelty: Grade B
Creativity or Innovation: Grade B
Scientific Significance: Grade C
P-Reviewer: Kimta N S-Editor: Fan M L-Editor: Kerr C P-Editor: Zhao YQ
Estes C, Anstee QM, Arias-Loste MT, Bantel H, Bellentani S, Caballeria J, Colombo M, Craxi A, Crespo J, Day CP, Eguchi Y, Geier A, Kondili LA, Kroy DC, Lazarus JV, Loomba R, Manns MP, Marchesini G, Nakajima A, Negro F, Petta S, Ratziu V, Romero-Gomez M, Sanyal A, Schattenberg JM, Tacke F, Tanaka J, Trautwein C, Wei L, Zeuzem S, Razavi H. Modeling NAFLD disease burden in China, France, Germany, Italy, Japan, Spain, United Kingdom, and United States for the period 2016-2030.J Hepatol. 2018;69:896-904.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 776][Cited by in F6Publishing: 1150][Article Influence: 164.3][Reference Citation Analysis (0)]
Marfell-Jones M, Olds T, Stewart A, Carter JEL.
International Standards for Anthropometric Assessment. Potchefstroom: North-West University, 2006. ISBN: 0-620-36207-3.
[PubMed] [DOI][Cited in This Article: ]
Ruiz-Margáin A, Xie JJ, Román-Calleja BM, Pauly M, White MG, Chapa-Ibargüengoitia M, Campos-Murguía A, González-Regueiro JA, Macias-Rodríguez RU, Duarte-Rojo A. Phase Angle From Bioelectrical Impedance for the Assessment of Sarcopenia in Cirrhosis With or Without Ascites.Clin Gastroenterol Hepatol. 2021;19:1941-1949.e2.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 20][Cited by in F6Publishing: 32][Article Influence: 8.0][Reference Citation Analysis (0)]
Pinto LP, Marroni CA, Czermainski J, Dahlem MLF, Carteri RB, Fernandes SA. Role of the phase angle in the prognosis of the cirrhotic patient: 15 years of follow-up.World J Methodol. 2023;13:238-247.
[PubMed] [DOI][Cited in This Article: ][Reference Citation Analysis (2)]
Gangadhar S, Ramanjanaiah LKV, Kamuni A, Morupoju G, Chopade B, Angadi S, Giri S, Nanjegowda SK, Bhrugumalla S. Assessment of nutritional status in cirrhotics using standard nutrition assessment tools and their correlation with severity of the disease.J Clin Exp Hepatol. 2022;12:S39.
[PubMed] [DOI][Cited in This Article: ]
Alvares-da-Silva MR, Reverbel da Silveira T. Comparison between handgrip strength, subjective global assessment, and prognostic nutritional index in assessing malnutrition and predicting clinical outcome in cirrhotic outpatients.Nutrition. 2005;21:113-117.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 299][Cited by in F6Publishing: 303][Article Influence: 15.2][Reference Citation Analysis (0)]
Tsoris A, Marlar CA.
Use Of The Child Pugh Score In Liver Disease. 2023 Mar 13. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2024 Jan-.
[PubMed] [DOI][Cited in This Article: ]
Salerno F, Merli M, Cazzaniga M, Valeriano V, Rossi P, Lovaria A, Meregaglia D, Nicolini A, Lubatti L, Riggio O. MELD score is better than Child-Pugh score in predicting 3-month survival of patients undergoing transjugular intrahepatic portosystemic shunt.J Hepatol. 2002;36:494-500.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 202][Cited by in F6Publishing: 206][Article Influence: 9.0][Reference Citation Analysis (0)]
