Nonalcoholic Fatty Liver Disease May Cause Thrombocytopenia

Nonalcoholic fatty liver disease (NAFLD) is the most common cause of chronic liver injury worldwide [1]. It covers a wide spectrum of hepatic disorders ranging from simple steatosis and steatohepatitis (steatosis with inflammation) to liver cirrhosis. The molecular and cellular mechanisms underlying hepatic injury in NAFLD are not clear. There is evidence to suggest that multiple mechanisms including insulin resistance, oxidative stress, inflammation and genetic factors interact to initiate the development of NAFLD [1]. Thrombocytopenia has been described in some cases of NAFLD [2,3,4], but its prevalence and salient features have not been fully described. Currently, the diagnosis of NAFLD requires liver biopsy, an invasive procedure not free from potential complications; however, several noninvasive diagnostic strategies have been proposed as potential diagnostic alternatives, each with different sensitivities and accuracies [5,6,7]. Among the noninvasive alternatives to liver biopsy, several studies have demonstrated the predictive value and a better benefit-to-risk ratio than biopsy of combinations of simple serum biochemical markers [5].In an effort to define some features of the thrombocytopenia that occurs in NAFLD, we prospectively studied 2 groups of individuals: patients with NAFLD defined by noninvasive methods and patients with thrombocytopenia. Individuals who were consecutively referred to our clinic after March 2008 due to thrombocytopenia (platelet count <100 × 109/l) were prospectively studied. Patients with immune thrombocytopenia [7], overt viral hepatitis or liver cirrhosis were excluded. The study was approved by the Ethics Committee of the Clinica Ruiz and informed consent was obtained from the patients included in both groups. Alpha 2 macroglobulin, haptoglobin, apolipoprotein A, bilirubin, gamma-glutamyl transpeptidase, alanine aminotransferase, aspartate aminotransferase, glucose, cholesterol and triglycerides were measured in all patients. These biochemical markers were then analyzed. FibroTest was used for the quantitative assessment of fibrosis, SteatoTest for the quantitative assessment of steatosis, ActiTest for the quantitative assessment of necroinflammatory activity in chronic viral hepatitis and NashTest for the categorical diagnosis of nonalcoholic steatohepatitis [5]. Patients with a score of >50% in either the SteatoTest or the NashTest coupled with a score of <50% in the FibroTest were defined as having NAFLD. Within the group of 93 consecutive patients studied due to their platelet count <100 × 109/l, 5 were found to fulfill the criteria for NAFLD. The median platelet count in these individuals was 72 × 109/l, (range 48-99), whereas median platelet volume was normal at 9 fl (range 8.8-11.6). All patients had an increased body mass index, in the range of 28-38 (median 36). At the physical examination, all patients had moderate hepatomegaly but none had splenomegaly. There was no evidence of mucocutaneous bleeding in any of the patients. They all denied abuse of alcohol or other substances and the female patient was not pregnant. No patients were taking any drugs or over-the-counter remedies known to cause thrombocytopenia. The manual and citrated platelet counts confirmed the automated counts in all individuals. No platelet clumping or red cell abnormalities were present. There was no anemia, deficiency of vitamin B12 or folic acid, disseminated intravascular coagulation or abnormal kidney function in the analyzed cohort. All patients had normal or negative values for antinuclear antibody titers, hepatitis serologies, human immunodeficiency virus testing, antiphospholipid antibodies and direct Coombs testing. Doppler ultrasonic exploration of the liver revealed changes consonant with fat infiltration. Two patients had abnormally high levels of transferases and 3 had increased gamma-glutamyl transpeptidase levels. Two patients displayed both a decreased level of haptoglobin and an increased level of bilirubin, a finding consonant with a mild degree of hemolysis, with the hemoglobin levels being normal in both these individuals (table 1). Thrombopoietin (TPO) levels were not measured. All 5 patients with both NAFLD and thrombocytopenia were followed for periods of 1 up to 27 months (median 9). Without any specific treatment, the platelet count remained relatively stable and no patient had a count of <40 × 109/l platelets during this period. Figure 1 depicts the platelet count in these 5 patients.NAFLD encompasses a wide spectrum of clinical conditions, actually representing an emerging disease of great clinical interest: the condition is the hepatic expression of metabolic syndrome, the prevalence of which is growing at an alarming rate worldwide. The molecular and cellular mechanisms underlying hepatic injury in NAFLD are not clear: there is evidence to suggest that multiple mechanisms including insulin resistance, oxidative stress, inflammation and genetic factors interact to initiate the development of NAFLD [8]. On the other hand, the presence of peripheral blood pancytopenia is a clearly identified complication of chronic liver disease. The mechanism involved has not been completely elucidated, but hypersplenism, a deficiency of erythropoietin and TPO, bone marrow hypofunction and other factors have been mentioned [9,10]. The most common cytopenia in this context is thrombocytopenia. Although it represents a marked feature of the end stage of liver disease, its prevalence, salient features and significance in NAFLD is less clear. As for the prevalence of thrombocytopenia in NAFLD, very little information exists. In a group of 12 consecutive patients with NAFLD, we found <100 × 109/l platelets in 5 [10]. The cause of thrombocytopenia in this setting is largely unknown. Due to a better understanding of the physiology of hematopoiesis, new insights into the mechanisms of hematologic alterations in liver disease have been elucidated. TPO is crucial for the regulation of megakaryocyte development and platelet production and is synthesized predominantly by the liver, its production being dependent on the functional liver cell mass. Accordingly, thrombocytopenia associated with liver disease may be attributable, in part, to inadequate TPO production in the failing liver [11]. However, on a practical level, there are many contradictory reports showing normal, increased or decreased serum TPO levels in various liver conditions including, but not limited to, liver cirrhosis, acute liver failure and infection with hepatitis B and C [3,4]. It is therefore possible that inadequate TPO production may be related to the presence of thrombocytopenia in NAFLD. In this group of NAFLD patients, we found evidence of a mild ongoing hemolytic condition which could stem from hypersplenism despite the fact that the spleen was not enlarged in any one individual; it is thus possible that hypersplenism could also contribute to the thrombocytopenia of NAFLD. There is very little information about thrombocytopenia in NAFLD. In a retrospective analysis of a series of patients with metabolic syndrome, Dasanu et al. [3 ]documented the presence of persistent moderate thrombocytopenia in patients with NAFLD. The salient features of our findings in this area were: (1) thrombocytopenia presented in less than half the patients with NAFLD, all of whom were overweight, (2) it was mild, usually >40 × 109/l and was not associated with mucocutaneous bleeding and (3) it did not require treatment. A major drawback of our study is that it relied on the definition of NAFLD by means of noninvasive diagnostic strategies [5] whereas ideally, the diagnosis of NAFLD requires a liver biopsy. We believe, however, that this approach is more adapted to real life and we would recommend against having a liver biopsy done as part of the evaluation of patients with thrombocytopenia. Additional studies are needed to further define the features and mechanisms of thrombocytopenia of NAFLD. NAFLD should be considered as a cause of mild thrombocytopenia.