J Thromb Thrombolysis DOI 10.1007/s11239-014-1124-z
Thrombophilic factor analysis in cirrhotic patients with portal vein thrombosis Bernd Saugel • Marcel Lee • Stephanie Feichtinger Alexander Hapfelmeier • Roland M. Schmid • Jens T. Siveke
•
Ó Springer Science+Business Media New York 2014
Abstract Liver cirrhosis, myeloproliferative disorders (MPDs) and prothrombotic mutations are aetiologic factors for portal vein thrombosis (PVT). The role and frequency of thrombophilic genetic risk factors in cirrhotic patients is not well established. In this case–control study, we investigated the frequency of Janus kinase 2 (JAK2) (JAK2 V617F), Factor V Leiden (FVL G1691A), and Prothrombin (G20210A) mutations in cirrhotic patients with PVT (LCi?/PVT? group, n = 21) in comparison with two control collectives (cirrhotic patients without PVT, LCi?/ PVT- group, n = 43; PVT patients without liver cirrhosis, LCi-/PVT? group, n = 29). In the LCi?/PVT? group, JAK2 V617F was present in 2/21 patients (10 %; p = 0.104 compared to LCi?/PVT-; p = 0.092 compared
to LCi-/PVT?), whereas 0/43 LCi?/PVT- patients (0 %; p \ 0.001 compared to LCi-/PVT?) and 9/29 LCi-/ PVT? patients (31 %) harboured this mutation. The FVL G1691A mutation was identified in 1/21 patients (5 %) in the LCi?/PVT? group, in 5/43 patients (12 %) in the LCi?/PVT- group, and in 2/29 patients (7 %) in the LCi-/PVT? group. The Prothrombin G20210A mutation was present in 0/21 LCi?/PVT? patients (0 %), in 1/43 LCi?/PVT- patients (2 %), and in 4/29 patients (14 %) in the LCi-/PVT? group. This study provides evidence that a relevant proportion of cirrhotic patients with PVT harbours a JAK2 V617F mutation. Keywords Factor V Leiden mutation JAK2 V617F mutation Liver cirrhosis Prothrombin mutation Portal vein thrombosis
Bernd Saugel and Marcel Lee have contributed equally to the study.
Electronic supplementary material The online version of this article (doi:10.1007/s11239-014-1124-z) contains supplementary material, which is available to authorized users. B. Saugel M. Lee S. Feichtinger R. M. Schmid J. T. Siveke II. Medizinische Klinik und Poliklinik, Klinikum rechts der Isar der Technischen Universita¨t Mu¨nchen, Ismaninger Strasse 22, 81675 Munich, Germany Present Address: B. Saugel (&) Department of Anesthesiology, Center of Anesthesiology and Intensive Care Medicine, University Medical Center HamburgEppendorf, Martinistrasse 52, 20246 Hamburg, Germany e-mail: [email protected] A. Hapfelmeier Institut fu¨r Medizinische Statistik und Epidemiologie, Klinikum rechts der Isar der Technischen Universita¨t Mu¨nchen, Ismaninger Strasse 22, 81675 Munich, Germany
Introduction Portal vein thrombosis (PVT) is a common complication in patients with liver cirrhosis [1–3]. Liver cirrhosis can be found in 28 % of patients with PVT and is therefore a major risk factor for development of PVT [3]. Besides other ‘‘local factors’’ (including intraabdominal infection or cancer of abdominal organs), various other factors associated with a higher risk of developing PVT are known [2]. Several prothrombotic factors including protein C and protein S deficiency, antithrombin deficiency, factor V Leiden (FVL) G1691A mutation, Prothrombin G20210A mutation, and myeloproliferative disorders (MPDs) can promote development of PVT [1, 2]. In non-cirrhotic patients, MPDs can be found in about 30–40 % of patients with PVT [2]. In 2005, a somatic point mutation in the Janus kinase 2 (JAK2) gene (gain of
123
B. Saugel et al.
function mutation JAK2 V617F) leading to MPDs was identified [4–6]. Since both liver cirrhosis and MPDs are major aetiologic factors for PVT, the frequency of the JAK2 mutation in patients with liver cirrhosis with or without PVT might be of particular interest. Therefore, the primary aim of our study was to investigate the frequency of the JAK2 V617F mutation in patients with liver cirrhosis and PVT in comparison with (1) patients with liver cirrhosis without PVT and (2) patients with PVT without liver cirrhosis. In addition to JAK2 V617F, we evaluated the frequency of the FVL G1691A and Prothrombin G20210A mutation. We hypothesised that in a clinically relevant proportion of cirrhotic PVT patients the JAK2 V617F mutation might be detectable.
Methods Setting, study design, and patients This case–control study was conducted in a German university hospital between December 2009 and August 2011. Cirrhotic patients with PVT but without known MPD (study group, LCi?/ PVT? group) were assessed for the presence of the JAK2 V617F, FVL G1691A, and Prothrombin G20210A mutation. Data from the study group were compared with two control collectives (cirrhotic patients without PVT, control group 1, LCi?/PVT- group; patients with PVT without liver cirrhosis, control group 2, LCi-/PVT? group). Patients with hepatocellular carcinoma were excluded before statistical data analysis. Patients with diagnosed PVT with or without liver cirrhosis previously treated in our department were identified using the electronic database of the hospital. Patients were then contacted and enrolled in the study. In addition, patients with PVT with or without liver cirrhosis and cirrhotic patients without PVT were prospectively identified and enrolled in the study during the study period. The study was approved by the institutional review board and written informed consent was obtained from all patients. Clinical and laboratory data were recorded after analysis of the patients’ medical records. Due to missing laboratory values, the Child-Pugh score and the model for end-stage liver disease (MELD) score were only calculable in 29/64 and 56/64 cirrhotic patients, respectively. For missing laboratory values please see Table S1 (Online Resource 1). In order to avoid selection bias, the study groups were compared with regard to patients’ characteristics including demographic and clinical variables. Detection of JAK2 V617F, Prothrombin G20210A and FVL G1691A mutations DNA collected from blood samples were processed by standard procedures. Detection of JAK2 V617F,
123
Prothrombin G20210A, and FVL G1691A mutation was performed using melting curve analysis with the Roche LightCycler 480 Real-Time PCR System (Roche Diagnostics, Mannheim, Germany) in 96 well plates. Primers and Fluorescence Resonance Energy Transfer (FRET) hybridization probes have been published before [7, 8] and are shown in Table S2 (Online Resource 2). The cycling conditions and settings for the melting curve analysis are specified in Table S3 (Online Resource 3). DNA extracted from HEL cells served as a positive control for the JAK2 V617F mutation. To assess the allelic burden of the JAK2 V617F mutation in carriers, melting curves with dilution series using DNA from HEL cells were obtained and compared to the patient’s melting curve. Minimum allelic burden reliably detectable was 5 %. For FVL G1691A and Prothrombin G20210A mutation, the control template from the Factor V Leiden Kit and Factor II (Prothrombin) G20210A Kit (both Roche Diagnostics) served as positive control, respectively. Representative results of the melting curve analyses are shown in Figure S1 (Online Resource 4). Statistical analysis The distribution of data is presented by measures of location and variability. For quantitative measures depending on the type of distribution these are given by mean ± standard deviation (normally distributed data) or median and percentile ranges (5–95 % percentile) (not normally distributed data). Likewise, corresponding group comparisons were performed by Student’s t test or by the nonparametric Wilcoxon rank-sum test. The distribution of qualitative data is displayed by contingency tables containing absolute and relative frequencies. According to the expected cell counts a comparison of relative frequencies was performed by Pearson’s Chi square test or Fisher’s exact test. Statistical tests were conducted two-sided. A pvalue \ 0.05 was considered statistically significant. Statistical analyses were performed by using IBM SPSS Statistics 19 (SPSS inc., Chicago, IL, USA) and the statistical software package R (The R Foundation for Statistical Computing, Vienna, Austria).
Results Patients and patients’ characteristics During the study period 144 patients were confirmed to be eligible for the study. Thirty-two patients refused to give informed consent (6 patients in LCi?/PVT? group, 24 patients in LCi?/PVT- group, 2 patients in LCi-/PVT? group). Nineteen patients were excluded from the statistical
Thrombophilic factor analysis Fig. 1 Patients enrolled in the study group and the control groups. LCi Liver cirrhosis, PVT portal vein thrombosis
All patients n=93
Patients with PVT n=50
Patients with LCi and PVT (LCi+/PVT+ group) n=21 STUDY GROUP
Patients with PVT without LCi (LCi-/PVT+ group) n=29 CONTROL GROUP 2
Patients with LCi without PVT (LCi+/PVT- group) n=43 CONTROL GROUP 1
analysis because of the presence of hepatocellular carcinoma (7 patients in LCi?/PVT? group, 12 patients in LCi?/PVT- group). Finally, 93 patients were analysed in the present study (Fig. 1, Table 1).
groups are shown in Fig. 2 and Table 2. In one patient from the LCi-/PVT? group the FVL G1691A and the JAK2 V617F mutation were simultaneously present.
Portal vein thrombosis
Discussion
In 50 patients (LCi?/PVT? group and LCi-/PVT? group) PVT was present (Fig. 1, Table 1). Diagnosis of PVT was primarily based on ultrasonography in 38 patients (76 %), on computed tomography in 11 patients (22 %), and on magnetic resonance tomography in 1 patient (2 %).
We investigated the frequency of the JAK2 V617F, FVL G1691A, and Prothrombin G20210A mutation in cirrhotic patients with PVT compared with cirrhotic patients without PVT and with patients with PVT in the absence of liver cirrhosis. In previous studies evaluating the prevalence of the JAK2 mutation in patients with splanchnic vein thrombosis, the presence of liver cirrhosis was a predefined exclusion criterion [9–14]. In contrast, our study was deliberately aimed to investigate the frequency of these mutations in patients with PVT and liver cirrhosis for two main reasons. First, cirrhotic patients with these mutations may have a particular high risk of PVT development. As cirrhotic patients may undergo liver transplantation, minimising the risk of PVT and subsequent development of portal vein transformation is pivotal as such patients are not eligible for surgery. Second, MPDs caused by JAK2 mutations may be missed in cirrhotic patients because of the multifactorial causes for blood cell suppression in this disease. Thus, MPD may not easily be recognised by standard blood cell count [1, 15–17]. In our study, about 30 % of patients with PVT in the absence of cirrhosis of the liver, harboured a JAK2 V617F mutation. This finding is in accordance with previous data showing that JAK2 mutation-positive MPDs can be found frequently in patients with non-cirrhotic PVT [2]. Whereas in none of the 43 cirrhotic patients without PVT from this study the JAK2 V617F mutation was identified, a JAK2 V617F mutation could be detected in 2/21 patients (10 %) with PVT and liver cirrhosis. Both
Liver cirrhosis Data regarding aetiology and severity of liver cirrhosis for patients in the study group and control group 1 are presented in Table 1. JAK2 V617F mutation Among all patients evaluated in this study, in 11 patients the JAK2 V617F mutation was identified. In the LCi?/PVT? group, the JAK2 V617F mutation was present in 2/21 patients (10 %) (Fig. 2, Table 2). Whereas no patient in the LCi?/PVT- group showed the JAK2 V617F mutation (0/ 43, 0 %), the mutation was observed in 9/29 patients (31 %) in the LCi-/PVT? group. The allelic burden of the JAK2 V617F mutation in carriers ranged from 5 to 50 % (Table 3). The blood cell counts of the 11 patients with the JAK2 V617F mutation are presented in Table 3. FVL G1691A mutation and Prothrombin G20210A mutation The frequency of the FVL G1691A mutation and the Prothrombin G20210A mutation in the different study
123
B. Saugel et al. Table 1 Patients’ characteristics Characteristics of patients in the study group (SG), control group 1 (CG1), and control group 2 (CG2) LCi?/ PVT?(SG)
LCi?/ PVT-(CG1)
LCi-/ PVT?(CG2)
p-value of group comparisons SG vs. CG1
Number of patients, n
SG vs. CG2
CG1 vs. CG2
21
43
29
–
Age, years
55 ± 9
58 ± 10
55 ± 15
0.296
0.959
0.357
Sex, female
6 (29 %)
17 (40 %)
11 (38 %)
0.391
0.490
0.891 –
Demographic data
Clinical data Portal vein thrombosis, n
21
0
29
–
–
Liver cirrhosis, n
21
43
0
–
–
Alcoholic liver cirrhosis, n (%)a
8 (38 %)
36 (84 %)
–
Thrombophilic factor analysis in cirrhotic patients with portal vein thrombosis Bernd Saugel • Marcel Lee • Stephanie Feichtinger Alexander Hapfelmeier • Roland M. Schmid • Jens T. Siveke
•
Ó Springer Science+Business Media New York 2014
Abstract Liver cirrhosis, myeloproliferative disorders (MPDs) and prothrombotic mutations are aetiologic factors for portal vein thrombosis (PVT). The role and frequency of thrombophilic genetic risk factors in cirrhotic patients is not well established. In this case–control study, we investigated the frequency of Janus kinase 2 (JAK2) (JAK2 V617F), Factor V Leiden (FVL G1691A), and Prothrombin (G20210A) mutations in cirrhotic patients with PVT (LCi?/PVT? group, n = 21) in comparison with two control collectives (cirrhotic patients without PVT, LCi?/ PVT- group, n = 43; PVT patients without liver cirrhosis, LCi-/PVT? group, n = 29). In the LCi?/PVT? group, JAK2 V617F was present in 2/21 patients (10 %; p = 0.104 compared to LCi?/PVT-; p = 0.092 compared
to LCi-/PVT?), whereas 0/43 LCi?/PVT- patients (0 %; p \ 0.001 compared to LCi-/PVT?) and 9/29 LCi-/ PVT? patients (31 %) harboured this mutation. The FVL G1691A mutation was identified in 1/21 patients (5 %) in the LCi?/PVT? group, in 5/43 patients (12 %) in the LCi?/PVT- group, and in 2/29 patients (7 %) in the LCi-/PVT? group. The Prothrombin G20210A mutation was present in 0/21 LCi?/PVT? patients (0 %), in 1/43 LCi?/PVT- patients (2 %), and in 4/29 patients (14 %) in the LCi-/PVT? group. This study provides evidence that a relevant proportion of cirrhotic patients with PVT harbours a JAK2 V617F mutation. Keywords Factor V Leiden mutation JAK2 V617F mutation Liver cirrhosis Prothrombin mutation Portal vein thrombosis
Bernd Saugel and Marcel Lee have contributed equally to the study.
Electronic supplementary material The online version of this article (doi:10.1007/s11239-014-1124-z) contains supplementary material, which is available to authorized users. B. Saugel M. Lee S. Feichtinger R. M. Schmid J. T. Siveke II. Medizinische Klinik und Poliklinik, Klinikum rechts der Isar der Technischen Universita¨t Mu¨nchen, Ismaninger Strasse 22, 81675 Munich, Germany Present Address: B. Saugel (&) Department of Anesthesiology, Center of Anesthesiology and Intensive Care Medicine, University Medical Center HamburgEppendorf, Martinistrasse 52, 20246 Hamburg, Germany e-mail: [email protected] A. Hapfelmeier Institut fu¨r Medizinische Statistik und Epidemiologie, Klinikum rechts der Isar der Technischen Universita¨t Mu¨nchen, Ismaninger Strasse 22, 81675 Munich, Germany
Introduction Portal vein thrombosis (PVT) is a common complication in patients with liver cirrhosis [1–3]. Liver cirrhosis can be found in 28 % of patients with PVT and is therefore a major risk factor for development of PVT [3]. Besides other ‘‘local factors’’ (including intraabdominal infection or cancer of abdominal organs), various other factors associated with a higher risk of developing PVT are known [2]. Several prothrombotic factors including protein C and protein S deficiency, antithrombin deficiency, factor V Leiden (FVL) G1691A mutation, Prothrombin G20210A mutation, and myeloproliferative disorders (MPDs) can promote development of PVT [1, 2]. In non-cirrhotic patients, MPDs can be found in about 30–40 % of patients with PVT [2]. In 2005, a somatic point mutation in the Janus kinase 2 (JAK2) gene (gain of
123
B. Saugel et al.
function mutation JAK2 V617F) leading to MPDs was identified [4–6]. Since both liver cirrhosis and MPDs are major aetiologic factors for PVT, the frequency of the JAK2 mutation in patients with liver cirrhosis with or without PVT might be of particular interest. Therefore, the primary aim of our study was to investigate the frequency of the JAK2 V617F mutation in patients with liver cirrhosis and PVT in comparison with (1) patients with liver cirrhosis without PVT and (2) patients with PVT without liver cirrhosis. In addition to JAK2 V617F, we evaluated the frequency of the FVL G1691A and Prothrombin G20210A mutation. We hypothesised that in a clinically relevant proportion of cirrhotic PVT patients the JAK2 V617F mutation might be detectable.
Methods Setting, study design, and patients This case–control study was conducted in a German university hospital between December 2009 and August 2011. Cirrhotic patients with PVT but without known MPD (study group, LCi?/ PVT? group) were assessed for the presence of the JAK2 V617F, FVL G1691A, and Prothrombin G20210A mutation. Data from the study group were compared with two control collectives (cirrhotic patients without PVT, control group 1, LCi?/PVT- group; patients with PVT without liver cirrhosis, control group 2, LCi-/PVT? group). Patients with hepatocellular carcinoma were excluded before statistical data analysis. Patients with diagnosed PVT with or without liver cirrhosis previously treated in our department were identified using the electronic database of the hospital. Patients were then contacted and enrolled in the study. In addition, patients with PVT with or without liver cirrhosis and cirrhotic patients without PVT were prospectively identified and enrolled in the study during the study period. The study was approved by the institutional review board and written informed consent was obtained from all patients. Clinical and laboratory data were recorded after analysis of the patients’ medical records. Due to missing laboratory values, the Child-Pugh score and the model for end-stage liver disease (MELD) score were only calculable in 29/64 and 56/64 cirrhotic patients, respectively. For missing laboratory values please see Table S1 (Online Resource 1). In order to avoid selection bias, the study groups were compared with regard to patients’ characteristics including demographic and clinical variables. Detection of JAK2 V617F, Prothrombin G20210A and FVL G1691A mutations DNA collected from blood samples were processed by standard procedures. Detection of JAK2 V617F,
123
Prothrombin G20210A, and FVL G1691A mutation was performed using melting curve analysis with the Roche LightCycler 480 Real-Time PCR System (Roche Diagnostics, Mannheim, Germany) in 96 well plates. Primers and Fluorescence Resonance Energy Transfer (FRET) hybridization probes have been published before [7, 8] and are shown in Table S2 (Online Resource 2). The cycling conditions and settings for the melting curve analysis are specified in Table S3 (Online Resource 3). DNA extracted from HEL cells served as a positive control for the JAK2 V617F mutation. To assess the allelic burden of the JAK2 V617F mutation in carriers, melting curves with dilution series using DNA from HEL cells were obtained and compared to the patient’s melting curve. Minimum allelic burden reliably detectable was 5 %. For FVL G1691A and Prothrombin G20210A mutation, the control template from the Factor V Leiden Kit and Factor II (Prothrombin) G20210A Kit (both Roche Diagnostics) served as positive control, respectively. Representative results of the melting curve analyses are shown in Figure S1 (Online Resource 4). Statistical analysis The distribution of data is presented by measures of location and variability. For quantitative measures depending on the type of distribution these are given by mean ± standard deviation (normally distributed data) or median and percentile ranges (5–95 % percentile) (not normally distributed data). Likewise, corresponding group comparisons were performed by Student’s t test or by the nonparametric Wilcoxon rank-sum test. The distribution of qualitative data is displayed by contingency tables containing absolute and relative frequencies. According to the expected cell counts a comparison of relative frequencies was performed by Pearson’s Chi square test or Fisher’s exact test. Statistical tests were conducted two-sided. A pvalue \ 0.05 was considered statistically significant. Statistical analyses were performed by using IBM SPSS Statistics 19 (SPSS inc., Chicago, IL, USA) and the statistical software package R (The R Foundation for Statistical Computing, Vienna, Austria).
Results Patients and patients’ characteristics During the study period 144 patients were confirmed to be eligible for the study. Thirty-two patients refused to give informed consent (6 patients in LCi?/PVT? group, 24 patients in LCi?/PVT- group, 2 patients in LCi-/PVT? group). Nineteen patients were excluded from the statistical
Thrombophilic factor analysis Fig. 1 Patients enrolled in the study group and the control groups. LCi Liver cirrhosis, PVT portal vein thrombosis
All patients n=93
Patients with PVT n=50
Patients with LCi and PVT (LCi+/PVT+ group) n=21 STUDY GROUP
Patients with PVT without LCi (LCi-/PVT+ group) n=29 CONTROL GROUP 2
Patients with LCi without PVT (LCi+/PVT- group) n=43 CONTROL GROUP 1
analysis because of the presence of hepatocellular carcinoma (7 patients in LCi?/PVT? group, 12 patients in LCi?/PVT- group). Finally, 93 patients were analysed in the present study (Fig. 1, Table 1).
groups are shown in Fig. 2 and Table 2. In one patient from the LCi-/PVT? group the FVL G1691A and the JAK2 V617F mutation were simultaneously present.
Portal vein thrombosis
Discussion
In 50 patients (LCi?/PVT? group and LCi-/PVT? group) PVT was present (Fig. 1, Table 1). Diagnosis of PVT was primarily based on ultrasonography in 38 patients (76 %), on computed tomography in 11 patients (22 %), and on magnetic resonance tomography in 1 patient (2 %).
We investigated the frequency of the JAK2 V617F, FVL G1691A, and Prothrombin G20210A mutation in cirrhotic patients with PVT compared with cirrhotic patients without PVT and with patients with PVT in the absence of liver cirrhosis. In previous studies evaluating the prevalence of the JAK2 mutation in patients with splanchnic vein thrombosis, the presence of liver cirrhosis was a predefined exclusion criterion [9–14]. In contrast, our study was deliberately aimed to investigate the frequency of these mutations in patients with PVT and liver cirrhosis for two main reasons. First, cirrhotic patients with these mutations may have a particular high risk of PVT development. As cirrhotic patients may undergo liver transplantation, minimising the risk of PVT and subsequent development of portal vein transformation is pivotal as such patients are not eligible for surgery. Second, MPDs caused by JAK2 mutations may be missed in cirrhotic patients because of the multifactorial causes for blood cell suppression in this disease. Thus, MPD may not easily be recognised by standard blood cell count [1, 15–17]. In our study, about 30 % of patients with PVT in the absence of cirrhosis of the liver, harboured a JAK2 V617F mutation. This finding is in accordance with previous data showing that JAK2 mutation-positive MPDs can be found frequently in patients with non-cirrhotic PVT [2]. Whereas in none of the 43 cirrhotic patients without PVT from this study the JAK2 V617F mutation was identified, a JAK2 V617F mutation could be detected in 2/21 patients (10 %) with PVT and liver cirrhosis. Both
Liver cirrhosis Data regarding aetiology and severity of liver cirrhosis for patients in the study group and control group 1 are presented in Table 1. JAK2 V617F mutation Among all patients evaluated in this study, in 11 patients the JAK2 V617F mutation was identified. In the LCi?/PVT? group, the JAK2 V617F mutation was present in 2/21 patients (10 %) (Fig. 2, Table 2). Whereas no patient in the LCi?/PVT- group showed the JAK2 V617F mutation (0/ 43, 0 %), the mutation was observed in 9/29 patients (31 %) in the LCi-/PVT? group. The allelic burden of the JAK2 V617F mutation in carriers ranged from 5 to 50 % (Table 3). The blood cell counts of the 11 patients with the JAK2 V617F mutation are presented in Table 3. FVL G1691A mutation and Prothrombin G20210A mutation The frequency of the FVL G1691A mutation and the Prothrombin G20210A mutation in the different study
123
B. Saugel et al. Table 1 Patients’ characteristics Characteristics of patients in the study group (SG), control group 1 (CG1), and control group 2 (CG2) LCi?/ PVT?(SG)
LCi?/ PVT-(CG1)
LCi-/ PVT?(CG2)
p-value of group comparisons SG vs. CG1
Number of patients, n
SG vs. CG2
CG1 vs. CG2
21
43
29
–
Age, years
55 ± 9
58 ± 10
55 ± 15
0.296
0.959
0.357
Sex, female
6 (29 %)
17 (40 %)
11 (38 %)
0.391
0.490
0.891 –
Demographic data
Clinical data Portal vein thrombosis, n
21
0
29
–
–
Liver cirrhosis, n
21
43
0
–
–
Alcoholic liver cirrhosis, n (%)a
8 (38 %)
36 (84 %)
–
