DISCUSSION
In this study, we evaluated a large sample of subjects submitted to real-time US-guided PLB in an outpatient setting over 5 years. Viral hepatitis was the main indication for biopsy, similar to previous reports 12,14,15. As the prevalence of HBV chronic infection in our country is low 16, HCV-related biopsies were more frequent in our sample. Some studies carried out in other geographic areas enrolled more commonly HBV patients15,17. Likewise, post-transplant follow-up was the most common reason for PLB at transplant centers 18. In the pediatric population, the major reported indications were elevated liver enzymes, AIH, and cholestatic diseases19.
In our study, the reason for PLB has undergone a transformation over the years. It was possible to notice a remarkable drop in the amount of HCV-related procedures and a progressive increase in MAFLD-related biopsies. As we have mentioned, a reduction in the need for liver biopsy due to HCV was expected, but other liver diseases still require histological evaluation to help on the management and follow-up1 – notably, AIH, small-duct primary sclerosing cholangitis, anti-mitochondrial antibody–negative primary biliary cholangitis, drug or herb-induced liver injuries, some HBV patients, systemic diseases with hepatic involvement, and MAFLD, which is being intensely studied. As promising therapies are awaited, it is feasible to predict that histological staging will help physicians in choosing the best approach to MAFLD patients.
As supposed, pain was the most frequent consequence of PLB1,10,19. A study involving 54 patients described an 84% incidence of pain immediately following the PLB and 39% of the patients still complained of pain after 24 hours 14. This symptom was associated with anxiety pre-procedure and the female gender 14. We found a 53.6% incidence of pain and a significant association with the female gender but we have not studied patients’ complaints prior to the biopsy. Although pain is generally mild and has a good response to analgesics, local anesthetics do not seem to prevent it, instead of the use of midazolam or nitrous oxide14. The real-time image-guided setting may reduce the incidence of pain, as it provides anatomical ascertain, reducing trauma to the subcutaneous tissue and hepatic capsule4,17,18.
According to the literature, the patient’s position after the PLB does not affect the rate and severity of complications 20, so our patients decided a comfortable position to rest. At our center, the protocol is to maintain observation for up to 4 to 6 hours after PLB, but the most appropriate period has not been well defined in the literature and it can last from 2 to 8 hours4,9,21-23.
In our study, the incidence of low blood pressure following the PLB was low (1.5%), similar to previous reports 7,20. It may occur due to bleeding, vasovagal reflex associated with pain, or even as a side effect of sedative drugs, such as midazolam4,7,8. We did not find an association between low blood pressure and bleeding or pain, possibly due to the low frequency of such complications in our sample.
The rate of bleeding was also low in our study, similar to the literature 3,4,24. It has already been described that patients older than 50 years are more likely to progress with bleeding25,26, but no statistical association between age and hemorrhage was found in our study. Parente et al. 8reported a higher rate of bleeding after liver biopsies (2.3%); however, the studied population had neoplasia, which is more susceptible to hemorrhage. A higher risk of bleeding is also expected in patients with platelets count < 50,000-70,000/mm³ and/or INR > 1.3-1.5 3,10,25-27. Most of our patients had normal INR and platelets count and there was no bleeding in the subjects with tests out of these limits, so we did not find any association between coagulation tests and bleeding.
The need for hospital admission in our sample (1.9%) was similar to previous reports, ranging from 1 to 3% 3,4,18. In addition, there were no more serious complications, such as perforation of the gallbladder, colon, kidneys, hemobilia, abscess, and intraperitoneal bleeding. Death following a PLB is extremely rare (up to 0.1%) 1,18 and it did not occur in our study. However, it is essential to notice that, in general, serious complications and death are more frequent in patients with severe comorbidities, suspected neoplasia, and cirrhosis, and such individuals usually undergo in-hospital biopsies at our center.
The real-time US-guided setting seems to reduce the risk of PLB-related serious complications 1,3,25,26. This is not entirely proven, since some studies have found different results24,28-30. Other factors must be reminded in order to improve the safety of PLB, such as the operator experience and a careful selection of individuals to undergo it in an outpatient setting31. Although radiologists most commonly perform US-guided liver biopsies 5, gastroenterologists and hepatologists can also safely execute it 6. Our results showed a low incidence of serious AEs, even in a scenario of senior residents in training, supervised by a hepatologist.
This study has some limitations: 1) the single-center design may be associated with bias on patients selection, 2) the retrospective design and non-uniform descriptions in the medical records may have changed the rate of AEs, 3) the data collected exclusively from the medical records may have missed outcomes of patients eventually admitted to other hospitals, and 4) a non-standardized dose of midazolam for sedation may have affected the incidence of pain.