Abstract
Background
The growth hormone-insulin-like growth factor (GH-IGF-1) axis and its impairment with sarcopenia, frailty, bone health, complications, and prognosis are not well characterized in cirrhosis.
Methods
We investigated the adult decompensated cirrhosis out-patients at a tertiary care institute between 2021 and 2023 for serum GH and IGF-1 levels, and associated them with sarcopenia (CT-SMI in cm2/m2), liver frailty index (LFI), osteodystrophy (DEXA), clinical decompensations (overall, ascites, encephalopathy, infection, and bleed), and survival up to 180 days.
Results
One-hundred-seventy-two patients, 95% males, aged 46.5 years (median). logIGF-1 levels were negatively associated with sarcopenia, osteodystrophy, LFI, CTP, and MELD-Na score (P < 0.05 each). Patients with low IGF-1 levels had a higher incidence of complications (overall, ascites and encephalopathy) than those with intermediate, and high IGF-1 levels (P < 0.05 each). Both logIGF-1 (AUC: 0.686) and MELD (AUC: 0.690) could predict 180-day mortality (P < 0.05, each). Adding logIGF-1 with MELDNa further improved discriminative accuracy of MELDNa (AUC: 0.729) P < 0.001. The increase in IGF-1 on follow-up was associated with better survival and fewer complications.
Conclusion
Reduced IGF-1 levels reflect sarcopenia, frailty, and osteodystrophy in cirrhosis. Low IGF-1 are associated with severity, development of decompensations, and mortality.
Keywords: decompensated cirrhosis, sarcopenia, frailty, osteodystrophy, insulin-like growth factor 1
Cirrhosis, an advanced stage of chronic liver injuries, progresses from an asymptomatic compensated stage to decompensated at a rate of 5%–7% each year.1 The decompensated stage is characterized by the appearance of complications related to the presence of portal hypertension or synthetic dysfunction leading to increased mortality.1 Complications such as malnutrition, sarcopenia, frailty, and impaired bone health are common and contribute to poorer quality of life and mortality in cirrhosis.2, 3, 4 Cross talks between systemic inflammation, altered metabolism, hormonal dysregulation, and muscle wasting underpin the pathophysiology of frailty, however, a detailed mechanobiology remains understudied.3,4
Sarcopenia is characterized by muscle loss and imbalanced processes involving muscle formation and breakdown. Further, it is regulated by two opposing factors, myostatin, and insulin-like growth factor 1 (IGF-1).3 Growth hormone (GH) resistance with low IGF-1 and IGFBP-3 is common in cirrhosis and the severity of GH resistance reflects injury to the liver.5 Nutritional supplements given to improve sarcopenia are often ineffective due to anabolic and GH resistance, which is mediated by impaired IGF-1 signaling and increased expressions of myostatin.6,7
Reductions in IGF-1 and IGFBP-3 are related to the intensity of hepatic dysfunction and associated with degree and severity of cirrhosis.8 Several pre-clinical studies in carbon tetrachloride (CCL4) induced cirrhosis have demonstrated an improvement of liver functions, architecture, enterocyte cytoskeleton, and osteopenia with the administration of IGF-1.9 IGF-1 levels have recently been linked with decompensation and poor prognosis in patients with cirrhosis.10,11 However, no study has focussed on exploring the association of GH-IGF-1 on complications and the development of new-onset complications. We hypothesize that the GH-IGF-1 axis is compromised in patients with cirrhosis and their levels can predict the development of complications and mortality in decompensated cirrhosis. Furthermore, we propose that improvement in IGF-1 levels is associated with enhanced survival outcomes.
Methods
Design
It was a single-center, prospective, cohort study performed at a large tertiary care institute between August 2021 to October 2023. We conformed to the institute's ethics guidelines and Good Clinical Practices. Approval was obtained from the Institute Ethics Committee NK/7952/PHD/471 and we adhered to the STROBE guidelines for reporting the study.
Patients
Patients between the age of 18–80 years with a confirmed diagnosis of decompensated liver cirrhosis presenting to the outpatient clinics of the Department of Hepatology were included after obtaining informed consent. Cirrhosis was diagnosed on clinical, biochemical, radiological, and histological findings.12 Decompensation was defined by the presence or history of ascites, variceal bleeding, or HE. We excluded patients with hepatocellular or any other malignancy, severe cardiac dysfunction (NYHA class III/IV), active infection, ongoing alcohol use within the past three months, human immunodeficiency virus seropositivity, and those with pregnancy or lactation.
Variables
We noted the demographics, clinical details such as symptoms, present or past decompensations, vitals, and laboratory parameters, viz. blood counts, liver functions, renal functions, coagulation parameters, and radiological findings at enrolment. Fasting plasma samples were collected at baseline in all patients and tested within 2 h for GH and IGF-1 levels through standardized chemiluminescence immunoassay (CLIA) using elecsys kits on cobas IGF-1 analyzer following manufacturer's instructions. Another sample at follow-up at 90-day was collected to assess the dynamic change in GH/IGF-1 levels. The degree of ascites was defined according to the International Club of ascites (ICA) guidelines.13 Hepatic encephalopathy (HE) was assessed by the West-Haven grading system.14 Infections, gastrointestinal (GI) bleed, acute kidney injury (AKI), and complications were defined and managed according to the EASL guidelines European Association for the study of the liver (EASL).15 Patients underwent assessment of sarcopenia, frailty, and bone health within a week of sample collection. Methods for the calculation of body mass index (BMI), sarcopenia, liver frailty, and bone health are detailed in supplementary material. We employed an international cut-off (SMI <50 cm2/m2 for males; <39 cm2/m2 for females) for the diagnosis of sarcopenia.16 Frailty was assessed using cut-offs: cut off for LFI: <3.2 robust, 3.2–4.5 pre-frail, >4.5 frail.4 Osteopenia and osteoporosis were defined by T scores as per WHO criteria (Osteopenia: −1 to −2.5; Osteoporosis: ≤−2.5)17 with osteodystrophy combining both.
Definition of Outcomes
Patients were monitored for 180 days. The primary endpoint was the impact of GH-IGF-1 levels on mortality, defined as the occurrence of death or transplant during the follow-up period. Patients lost to follow-up were censored at the last date known alive. Also, the cumulative incidence of the development of complications was monitored. Complications were defined according to consensus definitions-new onset or worsening or development of refractory ascites, acute episode of HE grades 2 or higher, portal hypertension-related bleeding, proven infections, AKI stage 1B or greater or HRS-AKI as detailed in Supplementary Table 1.18
Statistical Methods
Categorical variables were represented as numbers (percentages). The numerical data were described as mean with standard deviation or median with interquartile range depending on the distribution of data. We used the Shapiro–Wilk test to ascertain the distribution of numerical data. To analyze the normally distributed and not-normally distributed numerical variables between two groups Student's t-test and Mann–Whitney U test were used. Categorical variables were compared using Chi-Square or Fisher's exact test. To assess the dynamic changes between variables, %delta (Δ) change was calculated [%Δ change = (follow-up day variable minus the baseline variable)/(baseline variable) × 100]. Kaplan–Meier method was used for survival analysis, and cumulative incidence of complications, and the log Rank test was used for inter-group comparisons. The predictors of mortality were analyzed through multivariable logistic and cox-regression analysis. A P-value of <0.05 was considered significant, and the analysis was carried out using SPSS v.22.0 (SPSS Inc., Chicago, IL).
Results
A total of 172 patients with decompensated cirrhosis 95% males with a median age of 46.5 years, decompensation of 0.5 years, and alcohol (77.3%) as etiology were included. Ascites and HE were present in 95.9% and 7% of patients at enrolment. One patient was listed but none underwent the transplantation. Median serum GH and IGF-1 levels were 3.6 ng/ml (2.02–6.45) and 39.3 ng/ml (28.7–50). Patients were further stratified by IGF-1 (ng/ml) quartiles in low (Q1: <28.7), intermediate (Q2–Q3: 28.7–50), and high (Q4: >50), as per literature.19 Table 1 and Supplementary Table 2 detail the clinical characteristics of patients stratified as per their IGF-1 levels.
Table 1.
Baseline Clinical and Demographical Details of Patients With Decompensated Liver Cirrhosis Stratified According-IGF-1 Levels.
| Parameters | Total cohort (n = 172) | Low-IGF-1 (n = 43) | Intermediate IGF-1 (n = 88) | High IGF-1 (n = 41) | P value |
|---|---|---|---|---|---|
| Age-years | 46.5 (40.0–55.0) | 47.0 (40.0–57.5) | 46.0 (39.8–55.2) | 46.0 (42.0–53.0) | 0.643 |
| Gender-male | 163.0 (94.8) | 40.0 (93.0) | 83.0 (94.3) | 40.0 (97.6) | 0.623 |
| Etiology | 0.033 | ||||
|
133.0 (77.3) | 30.0 (69.8) | 64.0 (72.7) | 39.0 (95.1) | |
|
8.0 (4.7) | 2.0 (4.7) | 6.0 (6.8) | 0 (0.0) | |
|
6.0 (3.5) | 0 (0.0) | 5.0 (5.7) | 1.0 (2.4) | |
|
14.0 (8.1) | 7.0 (16.3) | 6.0 (6.8) | 1.0 (2.4) | |
|
11.0 (6.4) | 4.0 (9.3) | 7.0 (8.0) | 0 (0.0) | |
| Present decompensations | |||||
| Present ascites | 165.0 (95.9) | 42.0 (97.7) | 84.0 (95.5) | 39.0 (95.1) | 0.797 |
| Present ascites grade | 0.114 | ||||
|
55.0 (32) | 8.0 (18.6) | 30.0 (34.1) | 17.0 (41.5) | |
|
77.0 (44.8) | 20.0 (46.5) | 39.0 (44.3) | 18.0 (43.9) | |
|
33.0 (19.2) | 14.0 (32.6) | 15.0 (17.0) | 4.0 (9.8) | |
| Hepatic encephalopathy | 12.0 (7.0) | 2.0 (4.7) | 10.0 (11.4) | 0 (0.0) | 0.049 |
| HE grades | 0.040 | ||||
|
160.0 (93) | 41.0 (95.3) | 78.0 (88.6) | 41.0 (100.0) | |
|
12.0 (7) | 0 (0.0) | 8.0 (9.1) | 0 (0.0) | |
|
4.0 (2.3) | 2.0 (4.7) | 2.0 (2.3) | 0 (0.0) | |
| Clinical investigations | |||||
| Haemoglobin-g/dL | 10.1 (9.0–11.8) | 9.9 (8.8–11.4) | 9.9 (8.5–11.6) | 10.7 (9.4–12.2) | 0.189 |
| TLC-per mm3 | 5.8 (4.8–7.9) | 6.0 (4.7–7.5) | 6.2 (5.0–8.5) | 5.5 (4.2–7.2) | 0.078 |
| NLR | 2.3 (1.7–3.7) | 2.3 (1.8–4.1) | 2.5 (1.6–3.6) | 2.1 (1.5–3.3) | 0.500 |
| Platelet-×103/uL | 102.5 (70.8–144.0) | 104.0 (77.5–141.5) | 105.0 (69.8–140.8) | 93.0 (60.0–148.0) | 0.621 |
| Na-mmol/L | 136.0 (134.0–139.0) | 136.0 (133.0–140.0) | 136.0 (134.0–139.0) | 137.0 (135.0–139.0) | 0.455 |
| Bilirubin-mg/dL | 2.5 (1.6–4.0) | 2.7 (1.7–4.5) | 2.8 (1.6–5.0) | 2.3 (1.6–3.0) | 0.240 |
| Bilirubin conjugated-mg/dL | 1.4 (0.9–2.5) | 1.8 (0.9–2.7) | 1.4 (0.9–2.8) | 1.2 (0.8–1.6) | 0.114 |
| AST-U/L | 54.0 (43.8–78.8) | 55.0 (44.6–79.0) | 57.8 (43.5–85.4) | 52.5 (44.3–67.1) | 0.675 |
| ALT-U/L | 33.2 (24.4–46.9) | 32.1 (24.0–42.9) | 34.1 (24.9–47.3) | 34.0 (24.0–44.6) | 0.666 |
| ALP-U/L | 145.0 (106.0–193.0) | 139.5 (105.2–226.8) | 145.0 (102.5–180.0) | 145.0 (116.0–188.0) | 0.74 |
| Protein-mg/dL | 7.4 (7.0–7.8) | 7.2 (6.8–7.7) | 7.4 (6.9–7.8) | 7.7 (7.3–8.0) | 0.024 |
| Albumin-mg/dL | 3.3 (2.9–3.7) | 3.1 (2.8–3.5) | 3.3 (2.9–3.7) | 3.6 (3.2–3.9) | 0.024 |
| Creatinine-mg/dL | 0.8 (0.7–1.0) | 0.9 (0.7–1.1) | 0.8 (0.7–1.0) | 0.8 (0.6–0.9) | 0.088 |
| PT-sec | 17.9 (15.4–20.8) | 18.4 (15.8–20.2) | 18.0 (16.0–21.3) | 15.6 (14.3–20.5) | 0.104 |
| INR | 1.5 (1.3–1.8) | 1.6 (1.3–1.7) | 1.5 (1.3–1.8) | 1.4 (1.2–1.7) | 0.128 |
| Vitamin D-ng/mL | 20.4 (10.9–29.5) | 20.0 (9.0–26.9) | 21.1 (11.4–33.4) | 19.5 (11.2–25.5) | 0.431 |
| GH-ng/mL | 3.9 (2.0–6.4) | 3.8 (1.9–5.7) | 3.8 (2.3–6.5) | 4.3 (2.0–7.3) | 0.796 |
| IGF-1-ng/mL | 39.3 (28.7–50) | 22 (17.6–25.9) | 39.6 (35–42.5) | 71.7 (54.8–71.7) | <0.001 |
| Disease severity scores | |||||
| CTP score | 8.0 (7.0–9.2) | 9.0 (7.5–10.0) | 8.0 (7.0–9.2) | 8.0 (7.0–9.0) | 0.037 |
| CTP class | 0.283 | ||||
|
22.0 (12.8) | 5.0 (11.6) | 8.0 (9.1) | 9.0 (22.0) | |
|
107.0 (62.2) | 25.0 (58.1) | 58.0 (65.9) | 24.0 (58.5) | |
|
43.0 (25.0) | 13.0 (30.2) | 22.0 (25.0) | 8.0 (19.5) | |
| MELD score | 15.2 (11.8–18.7) | 16.1 (12.7–20.1) | 15.4 (12.2–18.9) | 13.1 (10.0–16.7) | 0.052 |
| MELD-Na score | 16.3 (12.8–21.6) | 17.4 (13.6–23.0) | 16.1 (13.2–22.4) | 14.7 (11.2–17.7) | 0.029 |
| Anthropometry | |||||
| BMI-kg/cm2 | 22.3 (20.3–25.9) | 24.5 (21.3–27.2) | 21.9 (20.2–24.1) | 21.8 (20.2–25.9) | 0.082 |
| HGS-kg/F | 8.9 (6.2–11.9) | 9.4 (7.1–11.7) | 8.5 (5.6–12.7) | 8.6 (6.3–11.8) | 0.825 |
| MAMC-cm | 24.2 (21.2–26.4) | 23.7 (21.1–26.7) | 23.7 (20.8–26.0) | 24.8 (21.5–27.2) | 0.258 |
| Sarcopenia | |||||
| SMI-cm2/m2 | 48.2 (42.6–54.0) | 48.7 (42.4–52.2) | 47.2 (42.2–54.1) | 49.1 (44.6–55.0) | 0.581 |
| Sarcopenia | 0.233 | ||||
|
51.0 (40.5) | 11.0 (34.4) | 23.0 (37.1) | 17.0 (53.1) | |
|
75.0 (59.5) | 21.0 (65.6) | 39.0 (62.9) | 15.0 (46.9) | |
| Frailty | |||||
| Liver frailty index | 4.2 (3.9–4.5) | 4.3 (4.1–4.6) | 4.2 (4.0–4.6) | 4.1 (3.8–4.2) | 0.009 |
| Frailty class | 0.014 | ||||
|
2.0 (1.3) | 0 (0.0) | 2.0 (2.6) | 0 (0.0) | |
|
113.0 (74.3) | 26.0 (70.3) | 51.0 (66.2) | 36.0 (94.7) | |
|
37.0 (24.3) | 11.0 (29.7) | 24.0 (31.2) | 2.0 (5.3) | |
| Bone health | |||||
| Bone health | 0.085 | ||||
|
76.0 (58) | 16.0 (48.5) | 38.0 (54.3) | 22.0 (78.6) | |
|
46.0 (35.1) | 15.0 (45.5) | 25.0 (35.7) | 6.0 (21.4) | |
|
9.0 (6.9) | 2.0 (6.1) | 7.0 (10.0) | 0 (0.0) | |
| Osteodystrophy | 55.0 (42.0) | 17.0 (51.5) | 32.0 (45.7) | 6.0 (21.4) | 0.039 |
Data are represented as mean ± SD or median (IQR) or n (%) as appropriate.
ACLF: Acute on chronic liver failure, ALP: Alkaline phosphatase, ALT: Alanine aminotransferase, AST: Aspartate aminotransferase, BMI: Body mass index, CTP: Child-Turcotte-Pugh score, GH: Growth hormone, GI: Gastrointestinal bleed, HE: Hepatic encephalopathy, HGS: Handgrip strength, HR: Heart rate, IGF-1: Insulin-like growth factor-1, INR: International normalized ratio, MAMC: Midarm muscle circumference, MAP: mean arterial pressure, MASLD: Metabolic dysfunction-associated steatotic liver disease, MELD: Model for end-stage liver disease, MetALD: Metabolic alcohol-associated liver disease, Na: Sodium, NLR: Neutrophil/lymphocyte ratio, SMI: Skeletal muscle index, sPO2−: Saturation of peripheral oxygen, TLC: Total leucocyte count.
Sarcopenia and GH-IGF-1 Levels
Seventy-five patients (59.5%) satisfied the criteria for sarcopenia. Median SMI was 47.7 cm2/m2 (42.6–54.1). Patients with sarcopenia had statistically insignificant GH levels compared to patients without sarcopenia. However, those with sarcopenia had significantly lower levels of IGF-1 than those without sarcopenia (Supplementary Table 3). On linear regression, we observed a significantly positive association between logIGF-1 and SMI (β = 9.67, P = 0.019) (Figure 1A). The association remained significant after adjusting for age, gender, BMI, and disease severity-MELDNa (β = 10.07, P = 0.003) (Supplementary Table 4). On logistic regression, logIGF-1 levels predicted sarcopenia (OR: 0.094, 0.013–0.676; P = 0.019) suggesting a 91% reduced likelihood and odds of sarcopenia with each unit increase in logIGF-1 levels. This association was significant after adjusting for age, gender, BMI, and MELDNa (aOR: 0.018, 0.001–0.344, P = 0.008) (Supplementary Table 5).
Figure 1.
Scatter diagram showing an association between IGF-1 levels with skeletal muscle index (A), liver frailty index (B), CTP-Class (C), MELD-Na (D), and MELD 3.0 (E). logIGF-1 levels were positively associated with skeletal muscle index and negatively associated with frailty and disease severity scores.
Frailty and GH-IGF-1 Levels
The median frailty index was 4.18 (3.93–4.49), reflecting pre-frailty in 75.7% and frailty in 24.3% of patients. No significant differences were identified between GH and the frailty status (Supplementary Table 6). Median IGF-1 levels (ng/ml) [36.5 (26.1–40)] were numerically lower among frail patients than pre-frail [40.1 (29.4–54.1)] and robust patients [36.5 (26.1–40)], (P = 0.059). Linear regression revealed a negative association of logIGF-1 with frailty (β = −0.42, P = 0.042) (Figure 1B). However, the association lost its significance after adjusting for age, gender, BMI, and MELDNa score (Supplementary Table 4).
Bone Health and GH-IGF-1 Levels
The mean T-score of the cohort was −0.78 ± 1.07. 42% of patients had osteodystrophy [osteoporosis: 35.1; osteopenia: 6.9%]. Patients with osteodystrophy had similar GH levels compared to patients with normal bone health (Supplementary Table 7). However, the median IGF-1 levels were lower among patients with osteodystrophy than those without. On logistic regression, logIGF-1 levels were negatively associated with osteodystrophy (OR: 0.101, 0.014–0.724; P = 0.022) indicating a 90% reduced likelihood of osteodystrophy with each unit increase in logIGF-1 levels. This association remained significant after adjusting for age, gender, BMI, and MELDNa score (OR: 0.068, 0.007–0.653, P = 0.020) (Supplementary Table 5).
Severity Scores and GH-IGF-1 Levels
GH levels did not show any significant association with the disease severity parameters. However, linear regression revealed, that logIGF-1 was negatively associated with CTP score (β = −1.48, P = 0.034) (Figure 1C). Further, logIGF-1 was negatively associated with MELDNa (β = −4.55, P = 0.048), MELD3.0 (β = −4.82, P = 0.029) (Figure 1D–E).
Complications of Cirrhosis and GH-IGF-1 Levels
Complications in the Past
GH and IGF-1 levels did not show any significant association with any complications in the past (Table 1).
Complications at Enrolment
No significant association was observed between GH levels and complications at enrolment. IGF-1 levels were significantly reduced in patients progressing from grade I to III ascites [I: 41.7 (35.4–53.6); II: 39.6 (27.5–48.8); III: 29.4 (24.4–41.7) ng/ml, P = 0.048]. Patients with low and intermediate IGF-1 levels had a higher prevalence of HE (grade I–II) compared to high IGF-1 levels [4.7 vs 11.4 vs 0, P = 0.049] (Table 1).
Complications during Follow-Up
During the 90-day follow up 18 patients expired. 26.7% of patients developed at least one of the complications (Supplementary Table 8). No association was observed between GH and complications. However, the probability of overall complications at 90-day was significantly high among patients with low-IGF-1 compared to intermediate IGF-1 and high IGF-1, specifically for the Incidence of new-onset or worsening or development of refractory ascites and HE (Figure 2A–C). The baseline IGF-1 levels (ng/ml) were lowest among patients developing grade 3 ascites [27.4 (23.6–37.2)] than grade 1–2 [40 (34–50)] or no ascites [45.4 (35.1–58.9)]. Likewise, baseline IGF-1 levels were lower in patients developing no control of ascites than those with partial and complete control at 90-days [25.8 (21–28.6) vs 40 (29.6–50) vs 45.4 (34.5–58.8) ng/ml; P = 0.003]. We noticed a higher incidence of infections in patients with low and intermediate IGF-1 compared to high IGF-1 levels. No association was found between IGF-1 and incidence of other decompensations at 90-days.
Figure 2.
Cumulative incidences plot for development of complications during 90-day and 180-day follow-up stratified by IGF-1 levels. Development of overall complications during 90-day (A), during 180-day (D), development of new-onset ascites, worsening of ascites or development of refractory ascites during 90-day (B), and at 180-day (E), development of hepatic encephalopathy during 90-day (C), and 180-day (F). Patients with low IGF-1 levels had a higher incidence of complications (overall, ascites and encephalopathy) than those with intermediate, and high IGF-1 levels (P < 0.05 each) except for overall complications and HE at 90-day follow-up (which was significant among low and high IGF-1 levels only).
During the 180-day follow up 30 patients expired, among 142 surviving patients 133 patients completed the follow-up. 37.8% of patients developed at least one of the complications (Supplementary Table 8). GH levels were not associated with complications. The probability of complications was maximum among patients with low-IGF-1, and intermediate IGF-1 than high IGF-1. Specifically, these complications were ascites and HE (Figure 2D–F). Baseline IGF-1 levels for patients with no ascites control were numerically lower compared to patients with partial and complete control [26.4 (23.7–31.4) vs 40 (34.4–50) vs 44.1 (35.4–55.5) ng/ml; P = 0.066]. We observed a higher probability of infections in patients with low (P = 0.027) and intermediate IGF-1 (P = 0.014) compared to high IGF-1 levels (14% vs 19.3% vs 4.9%). No association was found between IGF-1 and other decompensations at 180-days.
Prognosis
The 90-day and 180-day survival of the cohort was 89.5% and 82.6%. We observed statistically insignificant lower GH levels in the non-survivors at 180-days [3.92 (2.04–7.07) vs 3.16 (1.84–4.33); P = 0.088]. However, 90-day and 180-day survival was significantly lower in patients with low IGF-1 levels (90-day: 76.7%, 180-day: 69.8%) compared to patients with intermediate IGF-1 levels (90-day: 93.2%, 180-day: 85.2%) and high IGF-1 levels (90-day: 95.1%, 180-day: 90.2%), [P = 0.004 and P = 0.010 for 90-day and 180-day] (Figure 3A and B). After adjusting for age, gender, and MELD-Na, logIGF-1 independently predicted 90-day (aHR: 0.06, P = 0.017) and 180-day mortality (aHR: 0.10, P = 0.022) (Table 2), suggesting a 90% reduced hazard of mortality with each unit increase in logIGF-1 levels at 180-days.
Figure 3.
A–B). Kaplan Meier curves showing cumulative survival rates during 90-day and 180-day follow-up stratified by IGF-1 levels. 90-day and 180-day survival was significantly lower in patients with low IGF-1 compared to intermediate and high IGF-1 levels. C–D). The time-dependent area under the receiving operating characteristics curves (AUROC) shows sensitivity and specificity for mortality prediction at 90-day and 180-day follow-up. Both logIGF-1 and MELD could predict 90 and 180-day mortality (P < 0.05, each). Adding logIGF-1 with MELD could further improve the discriminative accuracy of MELD for mortality at 90-day and 180-day follow-up (P < 0.001).
Table 2.
Model 1: Univariable and Multivariable Analysis for Prediction of 90-Day Mortality.
| Variable | Univariable HR (95% CI) | P value | Multivariable HR (95% CI) | P value |
|---|---|---|---|---|
| Gender | 1.08 (0.14–8.15) | 0.937 | 1.58 (0.20–12.69) | 0.668 |
| Age | 1.11 (1.03–1.19) | 0.005 | 1.11 (1.03–1.20) | 0.010 |
| MELDNa score | 0.04 (0.01–0.35) | 0.003 | 0.06 (0.01–0.59) | 0.017 |
| logIGF-1 | 1.01 (0.97–1.06) | 0.588 | 1.03 (0.97–1.08) | 0.313 |
| Model 2: Univariableand multivariableanalysis for prediction of 180-day mortality | ||||
| Variable | Univariable HR (95% CI) | Pvalue | Multivariable HR (95% CI) | Pvalue |
| Gender | 1.04 (0.25–4.36) | 0.959 | 0.87 (0.17–4.35) | 0.864 |
| Age | 0.99 (0.95–1.03) | 0.598 | 1.00 (0.95–1.04) | 0.847 |
| MELDNa score | 1.14 (1.07–1.22) | <0.001 | 1.13 (1.06–1.21) | 0.001 |
| logIGF-1 | 0.09 (0.02–0.51) | 0.006 | 0.10 (0.01–0.72) | 0.022 |
CI: Confidence interval, HR: Hazard ratio, IGF-1: Insulin-like growth factor-1, MELD: Model for end-stage liver disease, Na: Sodium.
Both logIGF-1 (AUC: 0.726, P = 0.001) and MELDNa (AUC: 0.689, P = 0.002) could predict 90-day mortality. Similarly, logIGF-1 (AUC: 0.690, P = 0.001) and MELDNa (AUC: 0.696, P < 0.001) predicted 180-day mortality. Adding logIGF-1 with MELDNa score further improved the discriminative ability of MELDNa for 90-day (AUC: 0.761, P < 0.001) and 180-day mortality (AUC: 0.729, P < 0.001) (Figure 3C–D).
Dynamic Change in GH-IGF1 Levels with Mortality and Complications at 180-days
Percentage-delta changes in GH and IGF-1 levels observed at 90-day follow-up (Supplementary Table 9) showed a significant increase of IGF-1 among survivors than non-survivors (+45% vs −32%, P = 0.030), while no difference was observed in GH levels (Figure 4A). %Δ-IGF-1 levels could predict 180-day mortality (AUC: 0.791) with the best cut-off at −45%. Adding %Δ-IGF-1 and %Δ-MELDNa enhanced the %Δ-MELDNa's mortality prediction ability by 20% (Figure 4B). Δ-IGF-1 was higher among patients free from complications than those with complications (+75% vs +12%, P < 0.001) (Figure 4C). Patients free from new-onset, worsening, or refractory ascites showed an increase in Δ-IGF-1 levels (+48.5% vs −8.4%, P = 0.030) (Figure 4D). Likewise, patients free from infections had improvement in Δ-IGF-1 levels (+55.7% vs +2.8%, P = 0.002) (Figure 4E). No change in IGF-1 was noted among those developing HE and GI bleed (Supplementary Figure 1).
Figure 4.
A). Bar graphs showing %Δ change in IGF-1 levels among survivors and non-survivors at 180-day follow-up. Significant difference was observed in the %Δ change between survivors and non-survivors. B). The time-dependent area under the receiving operating characteristics curves (AUROC) shows sensitivity and specificity for mortality prediction at 180-day follow-up. %Δ change in IGF-1 combined with the %Δ change in MELDNA improved the discriminative ability of MELDNa for mortality prediction from AUC 0.611 to 0.813. C–E). Bar graphs showing %Δ change in IGF-1 levels at 90-day in patients who developed complications at 180-day follow-up.
Discussion
Our study reveals a novel association between IGF-1 levels and complications and new-onset complications in cirrhosis. We demonstrated a negative relation between IGF-1 levels with sarcopenia, frailty, osteodystrophy, and disease severity. Development of complications and mortality during follow-up was higher among patients with low IGF-1, particularly ascites and HE. This was further supplemented by better prognostic ability of MELDNa combined with IGF-1 and dynamically increased IGF-1 levels among survivors. These findings highlight IGF-1's multifaceted role in cirrhosis progression and its potential as a prognostic and therapeutic marker for preventing complications and mortality in cirrhosis.
Sarcopenia, a multifactorial pathologic process of muscle loss in cirrhosis, involves hyperammonemia, increased autophagy, insulin resistance, hyperinflammation, anabolic resistance, and lower branched-chain amino acids.3 Cirrhosis is characterised by low IGF-1 levels that further relates to elevated myostatin,7 sarcopenia, hyperammonemia, and impaired protein synthesis.20 IGF-1 prevents sarcopenia through activating the mTOR pathway and satellite cell recruitment. IGF-1/Akt axis also reduces muscle protein breakdown. It inhibits muscle atrophy-inducing cytokine and myostatin signaling via inhibition of the NF-κΒ and Smad pathways.21 However, cirrhosis is characterised by low IGF-1 levels that further relates to elevated myostatin,7 sarcopenia, hyperammonemia, and impaired protein synthesis.20 Not surprisingly, a negative association between IGF-1 levels and sarcopenia in our study aligns with a previous study from Japan.19 Interestingly, these findings paves way for exogenous GH therapy to restore IGF-1 levels and improve sarcopenia in cirrhosis.22
Frailty is an important complication negatively associated with outcomes in cirrhosis. We found a negative association between IGF-1 and frailty, however, it lost significance after adjusting with BMI and disease severity. Likewise, another study by Ha et al. did not find an association between IGF-1 and frailty.23 However, studies in older adults link low IGF-1 levels with frailty.24 We believe frailty in cirrhosis is nuanced and apart from endocrine perturbations, neuromuscular, immune dysregulation and gut dysbiosis contribute to frailty.25 Future research is needed to explore the intricate interplay between these factors for better underpinning the pathophysiology of frailty in cirrhosis.
Recent studies show a high prevalence of hepatic osteodystrophy in cirrhosis 66–95%.2 In the present study, osteodystrophy was present in 42% of patients. We found significantly lower IGF-1 levels, potentially contributing to osteodystrophy. Existing studies also report increased bone turnover with low IGF-1 levels.26 Reduced IGF-1 levels are shown to cause osteoblast dysfunction and subsequent osteodystrophy in cirrhosis.2 Future research should delve deeper in the complexities of bone turnover and therapeutically explore GH and IGF-1 for managing bone health in cirrhosis.
In alignment with prior investigations, our study revealed a proportional decrease in IGF-1 levels with increasing disease severity.19 Additionally, IGF-1 exhibited a significant negative association with prognostic scores. Previous studies also correlated MELD and CTP scores, suggesting IGF-1 as an index of cirrhosis severity and marker of liver dysfunction.8,27,28
We found a significant association between low IGF-1 levels and ascites, consistent with a prior study.28 Another study reported that individuals with HE at admission demonstrated lower IGF-I levels.8 Further, our findings indicate that patients with low IGF-1 were more prone to develop future complications, particularly ascites and HE. In compensated cirrhosis, IGF-1 levels have recently been linked with decompensation.10 Such observations underscore the prognostic relevance of IGF-1 in predicting the occurrence of specific complications in cirrhosis. Hence, IGF-1 levels can be used as a prognostic marker to stratify the patients at higher risk for developing complications of cirrhosis. This will help to prioritize monitoring and interventions for those at greatest risk.
Our findings suggest that adding IGF-1 levels to the existing MELDNa score can enhance its predictive accuracy for outcomes in cirrhosis, improving risk stratification and aiding in making more informed clinical decisions regarding patient management and treatment planning. Assy et al. also highlighted IGF-I levels <10 nmol/L with lower one-year survival in cirrhosis.29 Moreover, the dynamic improvement in IGF-1 levels among patients surviving at follow-up, and the development of lesser complications highlights a strong potential for therapeutic manipulation of IGF-1 levels in cirrhosis. Emerging literature also supports safety and improvement in survival with GH and colony-stimulating factors. However, the GH dose (1 U/day) utilized in the previous study was suboptimal and was not titrated with IGF-1 levels.30 Another study utilizing GH showed its safety and improvement in malnutrition among patients with decompensated cirrhosis.22 Based on the results, we believe GH administration with IGF-1 optimization can improve sarcopenia, bone health, complications, and outcomes in cirrhosis.
The liver's ability to produce IGF-1 is impaired due to significant liver damage and loss of functioning liver units among patients with decompensated cirrhosis, As a result, circulating levels of IGF-1 can decrease.9 Previous studies in cirrhosis patients have demonstrated severe GH resistance, characterized by low circulating levels of IGF-1 and IGFBP-3. The severity of GH resistance parallels with severity of liver disease.5,8,31 Growth hormone exerts its effects both directly and indirectly. Direct effects result from growth hormone binding to its receptors on target cells. GH stimulates the lipolysis in adipose tissue.32 GH contributes to the liver regeneration through the Foxm1b pathway and increases the levels of cyclin proteins (Cdc25A, Cdc25B, and cyclin B1) and reduces (cyclin-dependent kinase p27).33 It also, contributes to liver regeneration by activating hepatocyte nuclear factor and hepatocyte growth factor, through Ras-Raf-MEK-ERK and PI3K-PKB-mTOR pathways.34 However, a majority of effects of growth hormone are through its downstream mediator, IGF-I acting on its target cells.32 Together GH and IGF-1 has anabolic and growth-promoting effects.32 IGF-1 also, plays a crucial role in liver physiology. Preclinical studies have demonstrated the anti-fibrotic and anti-inflammatory action of IGF-1 in the liver. In rats with liver cirrhosis, IGF-1 improves liver function, reduces oxidative liver damage, and decreases collagen levels in the liver.35 IGF-I enhances intestinal barrier function and reduces portal pressure, endotoxemia, and bacterial translocation in cirrhotic rats.36 IGF-1 also limits fibrosis by directly inactivating HSCs in a p53-dependent manner.37 IGF-I replacement for short-duration increased albumin levels and improved energy metabolism in liver cirrhosis patients.38 These basic and clinical research findings indicate that IGF-1 is closely involved in hepatic inflammation and fibrosis regulation; hence, advanced disease stage and impaired liver functional reserve reduce IGF-1 levels, which may in turn exacerbate the disease conditions and cause poor prognosis. However, still establishing a causal relationship between low IGF-1 levels and poor outcomes in cirrhosis is indeed complex. As per Bradford Hill criteria,39 (detailed in the Supplementary data), we propose that low IGF-1 may be a causal factor. We demonstrated strong and consistent associations between low IGF-1 levels and increased mortality, sarcopenia, and osteodystrophy. Furthermore, the temporality, biological gradient, plausibility, and coherence of our findings reinforce the potential causative role of IGF-1 in these adverse outcomes. Future experimental studies are needed to confirm this causal relationship.
Strengths of our study include a comprehensive exploration of the multifaceted relationship between the GH-IGF-1 axis and cirrhosis; its complications, disease severity, and outcomes. It provides a longitudinal perspective on the association between IGF-1 and the development of complications and prognosis. Limitations include a single center design and, a lack of measurements of mediators such as insulin-like factor binding protein and acid-labile unit. Further studies are needed to elucidate the therapeutic effects of GH and IGF-1 administration on pathological axes with response-guided titration of GH in patients with decompensated cirrhosis.
In conclusion, IGF-1 levels reflect sarcopenia and osteodystrophy and could serve as liquid muscle and bone health biomarkers in cirrhosis. Incremental improvement in the predictive accuracy of MELDNa with the addition of IGF-1 suggests it to be a companion biomarker for enhanced prognostication in cirrhosis. IGF-1 levels predicted complications and progression with dynamic increase suggesting survival in cirrhosis. Thus, there is a strong potential for modulating the GH-IGF-1 axis to improve outcomes in patients with cirrhosis.
Credit authorship contribution statement
Parminder Kaur: Methodology, Software, Analysis, Investigation, analysis, Writing - Original draft, Project administration, Visualisation. Nipun Verma: Conceptualisation, Software, Analysis, Resources, Writing and Editing, Supervision, Project administration, Funding acquisition. Aishani Wadhawan: Methodology. Pratibha Garg: Methodology, Samonee Ralmilay: Methodology, Naveen Kalra: Supervision, Writing - Review and Editing. Abhiman Baloji: Methodology. Pinaki Dutta: Supervision, Writing - Review and Editing, Gaurav Sharma: Supervision, Writing-Review and Editing, Sahaj Rathi: Writing - Review and Editing, Arka De: Writing - Review and Editing, Madhumita Premkumar: Writing - Review and Editing. Sunil Taneja: Writing - Review and Editing. Ajay Duseja: Writing - Review and Editing, Virendra Singh: Supervision, Writing - Review and Editing.
Declaration of Competing Interest
None.
Acknowledgments
None.
Data availability statement
Data can be available by requesting the corresponding author.
Informed consent
Informed consent was taken from all participants.
Financial support and sponsorship
This study was partly supported by the Indian Council of Medical Research (ICMR) (NCD/Adhoc/123/2022-23).
Footnotes
The abstract of this article has been submitted and accepted for poster presentation at the EASL Congress 2024 (European Association for the Study of Liver Disease).
Supplementary data to this article can be found online at https://doi.org/10.1016/j.jceh.2024.102402.
Contributor Information
Nipun Verma, Email: nipun29j@gmail.com.
Virendra Singh, Email: virendrasingh100@hotmail.com.
Appendix A. Supplementary data
The following is the Supplementary data to this article:
References
- 1.D'Amico G., Garcia-Tsao G., Pagliaro L. Natural history and prognostic indicators of survival in cirrhosis: a systematic review of 118 studies. J Hepatol. 2006;44:217–231. doi: 10.1016/j.jhep.2005.10.013. [DOI] [PubMed] [Google Scholar]
- 2.Jeong H.M., Kim D.J. Bone diseases in patients with chronic liver disease. Int J Mol Sci. 2019;20 doi: 10.3390/ijms20174270. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Dasarathy S. Consilience in sarcopenia of cirrhosis. J Cachexia Sarcopenia Muscle. 2012;3:225–237. doi: 10.1007/s13539-012-0069-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Tandon P., Montano-Loza A.J., Lai J.C., Dasarathy S., Merli M. Sarcopenia and frailty in decompensated cirrhosis. J Hepatol. 2021;75 Suppl 1(suppl 1):S147–S162. doi: 10.1016/j.jhep.2021.01.025. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Donaghy A., Ross R., Gimson A., Hughes S.C., Holly J., Williams R. Growth hormone, insulinlike growth factor-1, and insulinlike growth factor binding proteins 1 and 3 in chronic liver disease. Hepatology. 1995;21:680–688. [PubMed] [Google Scholar]
- 6.Liu W., Thomas S.G., Asa S.L., Gonzalez-Cadavid N., Bhasin S., Ezzat S. Myostatin is a skeletal muscle target of growth hormone anabolic action. J Clin Endocrinol Metab. 2003;88:5490–5496. doi: 10.1210/jc.2003-030497. [DOI] [PubMed] [Google Scholar]
- 7.García P.S., Cabbabe A., Kambadur R., Nicholas G., Csete M. Elevated myostatin levels in patients with liver disease: a potential contributor to skeletal muscle wasting. Anesth Analg. 2010;111:707–709. doi: 10.1213/ANE.0b013e3181eac1c9. [DOI] [PubMed] [Google Scholar]
- 8.Ronsoni M.F., Lazzarotto C., Fayad L., et al. IGF-I and IGFBP-3 serum levels in patients hospitalized for complications of liver cirrhosis. Ann Hepatol. 2013;12:456–463. [PubMed] [Google Scholar]
- 9.de la Garza R.G., Morales-Garza L.A., Martin-Estal I., Castilla-Cortazar I. Insulin-like growth factor-1 deficiency and cirrhosis establishment. J Clin Med Res. 2017;9:233–247. doi: 10.14740/jocmr2761w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Saeki C., Kanai T., Ueda K., et al. Insulin-like growth factor 1 predicts decompensation and long-term prognosis in patients with compensated cirrhosis. Front Med. 2023;10 doi: 10.3389/fmed.2023.1233928. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Yao Y., Yang D., Huang Y., Dong M. Predictive value of insulin-like growth factor 1-Child-Turcotte-Pugh score for mortality in patients with decompensated cirrhosis. Clin Chim Acta. 2020;505:141–147. doi: 10.1016/j.cca.2020.02.031. [DOI] [PubMed] [Google Scholar]
- 12.Procopet B., Berzigotti A. Diagnosis of cirrhosis and portal hypertension: imaging, non-invasive markers of fibrosis and liver biopsy. Gastroenterology Report. 2017;5:79–89. doi: 10.1093/gastro/gox012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Arroyo V., Ginès P., Gerbes A.L., et al. Definition and diagnostic criteria of refractory ascites and hepatorenal syndrome in cirrhosis. International Ascites Club. Hepatology. 1996;23:164–176. doi: 10.1002/hep.510230122. [DOI] [PubMed] [Google Scholar]
- 14.Ferenci P., Lockwood A., Mullen K., Tarter R., Weissenborn K., Blei A.T. Hepatic encephalopathy--definition, nomenclature, diagnosis, and quantification: final report of the working party at the 11th World Congresses of Gastroenterology, Vienna, 1998. Hepatology. 2002;35:716–721. doi: 10.1053/jhep.2002.31250. [DOI] [PubMed] [Google Scholar]
- 15.EASL Clinical Practice Guidelines on nutrition in chronic liver disease. J Hepatol. 2019;70:172–193. doi: 10.1016/j.jhep.2018.06.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Carey E.J., Lai J.C., Wang C.W., et al. A multicenter study to define sarcopenia in patients with end-stage liver disease. Liver Transpl. 2017;23:625–633. doi: 10.1002/lt.24750. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Sözen T., Özışık L., Başaran N. An overview and management of osteoporosis. Eur J Rheumatol. 2017;4:46–56. doi: 10.5152/eurjrheum.2016.048. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Solà E., Pose E., Campion D., et al. Endpoints and design of clinical trials in patients with decompensated cirrhosis: position paper of the LiverHope Consortium. J Hepatol. 2021;74:200–219. doi: 10.1016/j.jhep.2020.08.009. [DOI] [PubMed] [Google Scholar]
- 19.Saeki C., Kanai T., Nakano M., et al. Low serum branched-chain amino acid and insulin-like growth factor-1 levels are associated with sarcopenia and slow gait speed in patients with liver cirrhosis. J Clin Med. 2020;9 doi: 10.3390/jcm9103239. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Nishikawa H., Enomoto H., Ishii A., et al. Elevated serum myostatin level is associated with worse survival in patients with liver cirrhosis. J Cachexia Sarcopenia Muscle. 2017;8:915–925. doi: 10.1002/jcsm.12212. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Yoshida T., Delafontaine P. Mechanisms of IGF-1-mediated regulation of skeletal muscle hypertrophy and atrophy. Cells. 2020;9 doi: 10.3390/cells9091970. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Kumari S., De A., Kalra N., Singh V. Growth hormone therapy in decompensated cirrhosis: an open-label, randomized control trial. Off J Am Coll Gastroenterol | ACG. 2024;119:116–126. doi: 10.14309/ajg.0000000000002300. [DOI] [PubMed] [Google Scholar]
- 23.Ha N.B., Seetharaman S., Kent D.S., et al. Serum and plasma protein biomarkers associated with frailty in patients with cirrhosis. Liver Transpl. 2023;29:1089–1099. doi: 10.1097/LVT.0000000000000128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Doi T., Makizako H., Tsutsumimoto K., et al. Association between insulin-like growth factor-1 and frailty among older adults. J Nutr Health Aging. 2018;22:68–72. doi: 10.1007/s12603-017-0916-1. [DOI] [PubMed] [Google Scholar]
- 25.Laube R., Wang H., Park L., et al. Frailty in advanced liver disease. Liver Int. 2018;38:2117–2128. doi: 10.1111/liv.13917. [DOI] [PubMed] [Google Scholar]
- 26.Mitchell R., McDermid J., Ma M.M., Chik C.L. MELD score, insulin-like growth factor 1 and cytokines on bone density in end-stage liver disease. World J Hepatol. 2011;3:157–163. doi: 10.4254/wjh.v3.i6.157. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Khoshnood A., Nasiri Toosi M., Faravash M.J., et al. A survey of correlation between insulin-like growth factor-I (igf-I) levels and severity of liver cirrhosis. Hepat Mon. 2013;13 doi: 10.5812/hepatmon.6181. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.El-Toukhy N., Khidr M.A.E.-H., Youssef M. Association between insulin-like growth factor-1 and severity of liver disease in patients with liver cirrhosis. Afro-Egyptian Journal of Infectious and Endemic Diseases. 2021;11:208–215. [Google Scholar]
- 29.Assy N., Hochberg Z., Enat R., Baruch Y. Prognostic value of generation of growth hormone-stimulated insulin-like growth factor-I (IGF-I) and its binding protein-3 in patients with compensated and decompensated liver cirrhosis. Dig Dis Sci. 1998;43:1317–1321. doi: 10.1023/a:1018828412631. [DOI] [PubMed] [Google Scholar]
- 30.Verma N., Kaur A., Sharma R., et al. Outcomes after multiple courses of granulocyte colony-stimulating factor and growth hormone in decompensated cirrhosis: a randomized trial. Hepatology. 2018;68:1559–1573. doi: 10.1002/hep.29763. [DOI] [PubMed] [Google Scholar]
- 31.Kratzsch J., Blum W.F., Schenker E., Keller E. Regulation of growth hormone (GH), insulin-like growth factor (IGF)I, IGF binding proteins -1, -2, -3 and GH binding protein during progression of liver cirrhosis. Exp Clin Endocrinol Diabetes. 1995;103:285–291. doi: 10.1055/s-0029-1211366. [DOI] [PubMed] [Google Scholar]
- 32.Olarescu N.C., Gunawardane K., Hansen T.K., Møller N., Jørgensen J.O.L. 2015. Normal physiology of growth hormone in adults. [Google Scholar]
- 33.Krupczak-Hollis K., Wang X., Dennewitz M.B., Costa R.H. Growth hormone stimulates proliferation of old-aged regenerating liver through forkhead box m1b. Hepatology. 2003;38:1552–1562. doi: 10.1016/j.hep.2003.08.052. [DOI] [PubMed] [Google Scholar]
- 34.Abu Rmilah A.A., Zhou W., Nyberg S.L. Hormonal contribution to liver regeneration. Mayo Clin Proc Innov Qual Outcomes. 2020;4:315–338. doi: 10.1016/j.mayocpiqo.2020.02.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Castilla-Cortazar I., Garcia M., Muguerza B., et al. Hepatoprotective effects of insulin-like growth factor I in rats with carbon tetrachloride-induced cirrhosis. Gastroenterology. 1997;113:1682–1691. doi: 10.1053/gast.1997.v113.pm9352873. [DOI] [PubMed] [Google Scholar]
- 36.Lorenzo-Zúñiga V., Rodríguez-Ortigosa C.M., Bartolí R., et al. Insulin-like growth factor I improves intestinal barrier function in cirrhotic rats. Gut. 2006;55:1306–1312. doi: 10.1136/gut.2005.079988. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Nishizawa H., Iguchi G., Fukuoka H., et al. IGF-I induces senescence of hepatic stellate cells and limits fibrosis in a p53-dependent manner. Sci Rep. 2016;6 doi: 10.1038/srep34605. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Conchillo M., de Knegt R.J., Payeras M., et al. Insulin-like growth factor I (IGF-I) replacement therapy increases albumin concentration in liver cirrhosis: results of a pilot randomized controlled clinical trial. J Hepatol. 2005;43:630–636. doi: 10.1016/j.jhep.2005.03.025. [DOI] [PubMed] [Google Scholar]
- 39.Fedak K.M., Bernal A., Capshaw Z.A., Gross S. Applying the Bradford Hill criteria in the 21st century: how data integration has changed causal inference in molecular epidemiology. Emerg Themes Epidemiol. 2015;12:14. doi: 10.1186/s12982-015-0037-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data can be available by requesting the corresponding author.