3.5 Differences in cytopenia by cirrhosis etiology
The rate of leukopenia was notably higher and the rate of erythropenia was lower in HBV-related cirrhotic patients than alcohol-related cirrhotic patients (leukopenia: 18.33% vs 7.26%, P = 0.034; erythropenia: 48.14% vs 64.18%, P = 0.018). No significant difference was observed in the rate of thrombocytopenia between the two groups (Table 5). After adjusting for confounders, the risk of erythropenia (OR = 0.13, 95% CI: 0.02–0.85, P = 0.033) was higher in alcohol-related cirrhotic patients, while the risk of leukopenia (OR = 13.39, 95% CI: 1.80–99.76, P = 0.011) was higher in HBV-related cirrhotic patients (Table 6).
4 DISCUSSION
In the present study, HBV-related cirrhotic patients had higher adjusted risks of HCC and hypersplenism than alcohol-related cirrhotic patients. In contrast, the adjusted risks of HE and ACLF were notably greater in alcohol-related cirrhotic patients. The differences in complications between cirrhotic patients with different etiologies indicate that cirrhosis is not a single disease.
A retrospective cohort study indicated that the rate of HCC in HBV-related cirrhotic patients was higher than that in alcohol-related cirrhotic patients (32.6% vs 6.0%).18 Consistently, we observed that the rate of HCC was notably higher in HBV-related cirrhotic patients than alcohol-related cirrhotic patients (39.55% vs 1.45%). Another study revealed that viral hepatitis promoted the development of HCC more than alcoholic hepatitis, which meant that patients with HCC due to viral hepatitis had worse outcomes.19 Therefore, rigorous HCC surveillance is needed in HBV-related cirrhotic patients.
We also found that HBV-related cirrhotic patients had a higher rate of hypersplenism than alcohol-related cirrhotic patients (45.84% vs 28.99%). As one of the most common cirrhotic complications, hypersplenism often causes rapid and premature destruction of blood cells, especially platelets and leukocytes, which can lead to infection and bleeding.20 A study reported that leukopenia in cirrhotic patients led to a much higher risk of infection and poor prognosis.21 We discovered that the rate of leukopenia was substantially higher in HBV-related cirrhotic patients than alcohol-related cirrhotic patients (18.33% vs 7.26%). Thus, more attention should be paid to preventing infection among patients with HBV-related cirrhosis. In contrast, erythropenia was more common in alcohol-related cirrhotic patients (64.18% vs 48.14%). This may be because alcohol suppresses erythropoiesis in the bone.22,23
We found that the rate of HE was significantly higher in alcohol-related cirrhotic patients than HBV-related cirrhotic patients (15.94% vs 4.49%). Similarly, a retrospective cohort study of 598 cirrhotic patients by Vaz et al24 reported that alcohol-related cirrhotic patients had a higher rate of HE than hepatitis C virus-related cirrhotic patients (11.0% vs 5.0%). This may be because chronic alcohol abuse may cause more severe neocortical injury and cognition impairment than chronic hepatitis virus infection.25 Furthermore, the severity of HE has been shown to be associated with the serum ammonia level 26and we found that alcohol-related cirrhotic patients with HE had a higher serum ammonia level than HBV-related cirrhotic patients with HE (134.00 ± 62.99 vs 82.16 ± 17.82 μmol/L), indicating more severe HE in alcohol-related cirrhotic patients. Therefore, early detection and treatment of HE in alcohol-related cirrhotic patients should be considered.
Additionally, we found that the rate of ACLF was higher in alcohol-related cirrhotic patients than HBV-related cirrhotic patients (7.25% vs 2.28%). Our result was similar to that of research by Axley et al 27, which demonstrated that the rate of ACLF was clearly higher in alcohol-related cirrhotic patients than hepatitis virus-related cirrhotic patients (7.2% vs 4.1%). Moreover, another study reported that alcohol-related cirrhotic patients exhibited more severe ACLF than patients with viral liver disease.28Thus, it is necessary to be aware of the high risk of ACLF when treating patients with alcohol-related cirrhosis.
There were, admittedly, several limitations in our study, particularly the small sample size. Additionally, we only included hospitalized patients (who may have had greater disease severity), which may have caused selection bias. Furthermore, several potential confounders, such as diet (especially a high-protein diet), that may influence the occurrence of HE in cirrhotic patients, were not taken into consideration. Future studies with large sample sizes are warranted to confirm our findings.
5 CONCLUSION
HBV-related cirrhotic patients had increased risks of HCC and hypersplenism, whereas alcohol-related cirrhotic patients more readily developed HE and ACLF. Therefore, there should be a greater focus on regular screening for HCC and hypersplenism in HBV-related cirrhotic patients and for HE and ACLF in alcohol-related cirrhotic patients.