Esophageal variceal bleeding is one of the most dreaded complications of cirrhosis because of its high mortality. Each episode of bleeding has 30–50% mortality risk 1.
Besides its role as a neurotransmitter in the central nervous system, serotonin appears to be a central physiologic mediator of many gastrointestinal functions and a mediator of the brain–gut connection 2. Serotonin is one of the most abundant molecules in the gastrointestinal tract, and it plays a crucial role in the regulation of several physiological functions, such as motility, secretion, and visceral sensitivity. Besides this well-documented physiological role, increasing evidence supports the concept that 5-HT is directly involved in pathological mechanisms, as well as in the modulation of immune/inflammatory responses within the gut 3. Recently, Profirovic et al.4 suggested a prominent role of the 5-hydroxytryptamine type 4 receptor (5-HT(4)R) in promoting angiogenesis and identified 5-HT(4)R as a potential therapeutic target for modulating angiogenesis under pathological conditions.
After application of serotonin inhibitors, portal pressure decreased in patients with liver cirrhosis, confirming the importance of serotonin in the pathogenesis of portal hypertension 5,6.
Recently it was reported that plasma serotonin levels are significantly higher in patients with cirrhosis than in controls and represent the degree of liver insufficiency. In addition, platelet poor plasma (PPP) serotonin estimation is a better marker for liver insufficiency compared with platelet serotonin content 6. Moreover, increased levels of free plasma serotonin were hypothized to contribute to variceal development 7.
The purpose of this work was to determine plasma free-serotonin concentrations in patients with esophageal and fundal varices in a trial to assess its diagnostic significance as a noninvasive marker for the presence of gastroesophageal varices and its value in the assessment of disease progression.
Participants and methods
The present study was conducted in the tropical medicine department, Ain Shams University Hospital, from June 2012 to May 2013 on 60 adult patients with liver cirrhosis and 30 age-matched and sex-matched apparently healthy participants serving as a control group, all of whom willingly participated in the study, and informed consent was obtained before enrollment in the study.
The patients included in the study were diagnosed with stigmata of chronic liver disease based on clinical, laboratory, and radiological data presented with upper gastrointestinal symptoms (e.g. nausea, vomiting, hematemesis, melena, etc.). After upper gastrointestinal endoscopy, patients were divided into two groups. Group I included 30 patients with liver cirrhosis complicated with either esophageal varices or fundal varices (mean age of 52.10±6.1 years). There were 27 male and three female patients. This group was further subdivided on the basis of the degree of oesophageal varices (OV) into grade I, II (n=7), grade III (n=7), grade IV (n=8), and OV with fundal extension (n=8). Group II included 30 patients with liver cirrhosis who did not have varices (mean age of 51.2±4.7 years). There were 28 male and two female patients. Patients’ groups were compared with a control group (group III), which included 30 healthy participants without any clinical or laboratory evidence of liver disease (mean age of 50.3±7.0 years, 28 male and two female participants). Participants diagnosed with schizophrenia, hypertension, Huntington’s disease, Duchenne’s muscular dystrophy, carcinoid syndrome, portal vein or splenic vein thrombosis, and hepatocellular carcinoma were excluded from our study.
All patients were subjected to the following:
- Full history taking and thorough clinical examination.
- Laboratory tests: liver function tests including serum liver enzymes (AST, ALT and ALP), total and direct bilirubin, albumin, and international normalized ratio (INR); complete blood count (CBC), erythrocyte sedimentation rate (ESR), and serum α-fetoprotein; kidney function tests including serum blood urea nitrogen, creatinine, sodium (Na), and potassium (K); and determination of plasma free serotonin in PPP samples by enzyme-linked immunosorbent assay (ELISA) technique.
- Abdominal ultrasound to assess liver echogenicity, presence of focal lesion, ascites, portal vein diameter, and spleen size.
Upper gastrointestinal tract endoscopy (for group I and II): an upper gastrointestinal tract endoscopy was performed with an endovideo system (Olympus GIF-Q 165, Olympus company, USA). To measure the esophageal varices size, Paquet’s classification was used: in I degree, fewer snake-like mucosal protrusions were seen; in II degree, varices were predominating up to a half of the esophageal lumen radius; in III degree, varices were in contact at some points; and in IV degree, heralds of the imminent rupture (cherry-red spots) were seen. Endoscopic examination showed portal hypertensive gastropathy (snake-skin mucosa) and varices of the gastric fundus 8.
A volume of 10 ml of venous blood was collected under complete aseptic conditions from each participant. The collected blood was divided among an EDTA tube for CBC, a citrated tube for ESR, a citrated tube for coagulation profile and INR estimation, a citrated plastic tube for free plasma serotonin assay, and a plain test tube for serum separation. After clotting, samples were centrifuged at 1000g for 15 min and sera were separated for liver and kidney function tests. Free plasma serotonin samples were centrifuged at room temperature for 10 min at 200g to obtain platelet rich plasma. A volume of 1 ml platelet rich plasma was then transferred to another tube and centrifuged at 1000g for 10 min to obtain PPP and then stored at −80°C until assay. Hemolyzed samples and highlyictric (bilirubin>5 mg/dl) were discarded, and repeated freezing and thawing was avoided.
Liver function tests were conducted: serum liver enzymes (AST, ALT, and ALP), total and direct bilirubin, and albumin were measured on the Synchron CX-9 autoanalyzer (Beckman Instruments Inc.; Scientific Instruments Division, Fullerton, California, USA), and INR was measured using Stago coagulation-analyzers (Diagnostica Stago Inc., USA). CBC was carried out on the Coulter LH 750 hematology analyzer (Beckman Coulter International SA, Nyon, Switzerland). ESR was determined using the Westergren method. Serum α-fetoprotein was determined using the Immulite 2000 XPi immunoassay system (Siemens AG, Erlangen, Germany). It is a solid-phase, two-site sequential chemiluminescent immunometric assay. The solid phase was a polystyrene bead enclosed within the test unit, coated with monoclonal antibody specific for AFP. Kidney function tests were conducted: serum blood urea nitrogen, creatinine, sodium (Na), and potassium (K) were measured on the Synchron CX-9 autoanalyzer (Beckman Instruments Inc.; Scientific Instruments Division). Plasma free serotonin in PPP samples was determined: assays were carried out with the sandwich ELISA technique using reagents provided by DRG Serotonin ELISA kit (DRG International Inc., USA). The assay procedure follows the basic principle of competitive ELISA in which there is competition between biotinylated and nonbiotinylated antigens for a fixed number of antibody binding sites. The amount of biotinylated antigen bound to the antibody is inversely proportional to the analyte concentration of the sample. When the system is in equilibrium, the free biotinylated antigen is removed by a washing step, and the antibody-bound biotinylated antigen is determined using antibiotin alkaline phosphatase as a marker and p-nitrophenyl phosphate as a substrate. The color development is discontinued, and the intensity of the color is measured. Quantification of unknowns is achieved using a standard curve that was constructed by plotting the mean absorbance for each standard on the y-axis (linear) against the concentration on the x-axis (logarithmic). The best-fit curve was drawn through the points on the graph. Because of sample preparation (dilution and acylation), the values have to be multiplied by the factor (×23.5) to obtain the concentration in ng/ml.
IBM SPSS statistics (V. 19.0; IBM Corp., USA) was used for data analysis. Parametric data were expressed as mean and SD, and nonparametric data were expressed as median and interquartile range; categorized data were expressed as both number and percentage. Comparative statistical analysis was performed using the Wilcoxon rank sum and Kruskall–Wallis test in case of nonparametric data and using analysis of variance test in case of parametric data. The χ2-test was used for comparison between independent groups as regards the categorized data. Correlation analysis was performed using Spearman’s rank correlation (rs). P values less than 0.05 were considered significant, whereas values less than 0.01 were considered highly significant. The diagnostic validity test (it includes diagnostic sensitivity, diagnostic specificity, positive predictive value, and negative predictive value) was carried out. The receiver operating characteristic (ROC) was used to obtain the most sensitive and specific cutoff.
This study included three age-matched and sex-matched groups. As per Child’s classification, there were seven (23.3%) patients of class A, 13 (43.3%) of class B, and 10 (33.4%) of class C in group I, versus 17 (56.7%), nine (30%), and 13.3%, respectively, in group II. Upper gastrointestinal tract endoscopic findings are shown in Table 1. Estimation of plasma free serotonin showed a high statistically significant difference between groups (P<0.01), with the values being higher in group I (77.79±23.67 ng/ml) compared with group II (47.75 ±30.71 ng/ml, P<0.01) and the control group (13.94±6.67 ng/ml, P<0.01) (Table 2). Figure 1 shows a highly significant stepwise progressive increase in the marker level recorded through grades of OV 2–5 (stage 2, 20.7±7.7 ng/ml; stage 3, 44.3±8.9 ng/ml, P<0.01; stage 4, 100.1±12.4 ng/ml, P<0.01; stage 5, 114.6±14.7 ng/ml, P<0.05). Moreover, a highly significant increase was observed in plasma free serotonin levels in case of gastric varices in comparison with its level in OV alone (P<0.01). Correlation analysis was performed between plasma free serotonin and endoscopic findings in a group of chronic liver disease (CLD) patients with OV, revealing highly significant positive correlation (rs=0.860, P<0.01). Conversely, no statistically significant correlation was found between plasma free serotonin levels neither with demographic data such as age, sex, and smoking nor with laboratory investigations in studied groups (P>0.05 in all). Diagnostic performance of plasma free serotonin description of group I and II versus control was carried out using ROC curve. A serotonin cutoff value of 24 ng/ml was chosen with 100% sensitivity, 96.7% specificity, 98.3 to discriminate all CLD patients from healthy controls (area under the curve=0.981) (Fig. 2a). In addition, for discrimination of esophageal varices patients, ROC curve was constructed; at a cutoff value of 58.0 ng/ml, group I could be differentiated from group II with 80% sensitivity, 86.7% specificity, 88.23% positive predictive, 83.33% negative predictive value, and total efficacy of 85.7% (area under the curve=0.855) (Fig. 2b).
Portal hypertension is the main complication of cirrhosis and is defined as a hepatic venous pressure gradient of more than 5 mmHg. The development of gastroesophageal varices and variceal hemorrhage is the most direct consequence of portal hypertension 9. Moreover, in the study by Vorobioff et al. 9, it was confirmed that the application of ketanserin and ritanserin (serotonergic receptor inhibitors) caused the lowering of portal hypertension in patients with liver cirrhosis 10.
Different factors may affect the concentration of circulating serotonin in liver cirrhosis, such as impaired serotonin metabolism due to higher activity of the monoamino oxidases, impaired metabolism of tryptophan as serotonin precursor, platelet sequestration in the spleen and/or platelet activation 11.
In our study, free or unconjugated serotonin levels were investigated. The levels of free serotonin were higher in patients with liver cirrhosis compared with the healthy control group, similar to the results of Rudić et al. 7. In the study by Beaudry et al.12, the whole-blood serotonin levels were significantly lower in 30 patients with cirrhosis compared with the age-matched controls, and no correlation was found between these levels and the severity of cirrhosis. However, in the same study the unconjugated plasma serotonin levels, an indication of the active form of serotonin, were significantly higher in patients with cirrhosis than in the controls 12. Recently, it has been reported that disturbances in serotonin and melatonin homeostasis observed in patients with liver cirrhosis may be associated with advanced encephalopathy 13.
In a study by Ćulafić et al. (2007), plasma free serotonin levels were significantly higher in Child-Pugh grade A/B than in grade C patients, and this matched our results 6.
Moreover, there was no significant correlation between serotonin concentration in plasma and the platelet count, which is one of the main store of serotonin; This was in agreement with the result of Rudić et al. 7.
Rudić et al. 7 reported no significant difference between the level of serotonin and grades of esophageal varices; our results showed significant difference. The study included only 33 patients with liver cirrhosis and 24 healthy participants who served as the control group; esophageal varices were not detected in five (15.1%) patients and only four patients (12.1%) were of grade IV. However, in our study the mean plasma free serotonin level was higher in patients with esophageal varices compared with patients without varices 7.
Furthermore, the correlation between plasma serotonin concentration and fundal varices was highly significant in the study by Rudić et al. 7, similar to that found in our study.
In the study by Beaudry et al. 11, the whole-blood serotonin levels were significantly lower in 30 patients with cirrhosis than in the age-matched controls, and no correlation was found between these levels and the severity of cirrhosis. However, in the same study the unconjugated plasma serotonin levels, an indication of the active form of serotonin, were significantly higher in patients with cirrhosis than in the controls 12.
The sensitivity and specificity of plasma free serotonin has been shown to vary with the different cutoff values used. According to our results, at a cutoff of 24 ng/ml the sensitivity was 100% and the specificity was 96.7%, indicating cirrhosis, and at a cutoff of 58 ng/ml the sensitivity was 80% and the specificity was 86.7%, indicating the presence of varices.
In conclusion, plasma free serotonin levels could possibly be used as a noninvasive predictive method for the presence or absence of gastroesophageal varices.
Further studies are recommended to investigate the clinical value for use of serotonin receptor blockers in these patients with CLD and varices.
Conflicts of interest
There are no conflicts of interest.
1. Sharma P, Rakela J. Management of pre-liver transplantation patient-part 2. Liver Transpl 2005; 11:249–260
2. Lesurtel M, Soll C, Graf R, Clavien PA. Role of serotonin in the hepato-gastroIntestinal tract: an old molecule for new perspectives. Cell Mol Life Sci 2008; 65:940–952
3. Cirillo C, VandenBerghe P, Tack J. Role of serotonin in gastrointestinal physiology and pathology. Minerva Endocrinol 2011; 36:311–324
4. Profirovic J, Strekalova E, Urao N, Krbanjevic A, Andreeva AV, Varadarajan S, et al.. A novel regulator of angiogenesis in endothelial cells: 5-hydroxytriptamine 4 receptor. Angiogenesis 2013; 16:15–28
5. Li T, Weng SG, Leng XS, Peng JR, Wei YH, Mou DC, et al.. Effects of 5-hydroxytamine and its antagonists on hepatic stellate cells. Hepatobiliary Pancreat Dis Int 2006; 5:96–100
6. Culafic DM, Mirkovic DS, Vukcevic MD, Rudic JS. Plasma and platelet serotonin levels in patients with liver cirrhosis
. World J Gastroenterol 2007; 13:5750–5753
7. Rudić JS, Culafić DM, Mirković DS, Ješić RS, Krstić MN. Role of serotonin in development of esophageal and gastric fundal varices
. World J Gastroenterol 2010; 16:6135–6138
8. Oberhammer E, Paquet KJ, Diselmaier W. Endoscopic finding and portal hypertension. Therapiweoche 1978; 28:7178–7187. Adopted from Rudić et al
9. Bari K, Garcia-Tsao G. Treatment of portal hypertension. World J Gastroenterol 2012; 18:1166–1175
10. Vorobioff J, Garcia-Tsao G, Groszmann R, Aceves G, Picabea E, Villavicencio R, et al.. Long-term hemodynamic effects of ketanserin, a 5-hydroxytryptamine blocker, in portal hypertensive patients. Hepatology 1989; 9:88–91
11. Laffi G, Marra F, Gresele P, Romagnoli P, Palermo A, Bartolini O, et al.. Evidence for a storage pool defect in platelets from cirrhotic patients with defective aggregation. Gastroenterology 1992; 103:641–646
12. Beaudry P, Hadengue A, Callebert J, Gaudin C, Soliman H, Moreau R. Blood and plasma 5-hydroxytryptamine levels in patients with cirrhosis. Hepatology 1994; 20:800–803
13. Chojnacki C, Walecka-Kapica E, Klupińska G, Wachowska-Kelly P, Żylińska K, Winczyk K, et al.. Serotonin and melatonin secretion and metabolism in patients with liver cirrhosis
. Pol Arch Med Wewn 2012; 122:392–397