Resection for hepatic hemangiomas

Abstract

A retrospective analysis of outcomes in 34 patients referred to the Academic Medical Center of the University of Amsterdam for management of hepatic hemangiomas appears in this edition of the Journal.1 Fourteen patients underwent resection of hemangiomas because of severe symptoms or diagnostic uncertainty. Only one patient had persisting abdominal pain after resection; however, the follow-up period in this article was relatively short. Liver surgery is becoming increasingly safe as a result of surgical specialization, improvements in the technology used for liver resection, changes in anesthetic practice, and improvements in postoperative care. Consequently, the indications for liver surgery have been extended beyond malignant disease. Increasingly, there are reports in the literature from large institutions describing experience in the management of hepatic hemangiomas.2 The evolution of safe laparoscopic surgery is likely to further this trend. In skilled hands, the argument is no longer about the morbidity and mortality of operative intervention because open and laparoscopic surgical techniques in such units are now of well proven safety, even in the presence of significant parenchymal liver disease. The issue is now appropriate selection of patients for resection of benign liver tumors. It is important to clearly define the indications for surgery for hepatic hemangiomas. In the Amsterdam experience,1 Erdogan et al. state that the indications for resection of hemangiomas remain controversial. It could also be stated that the indications remain limited. For example, it is not entirely clear from their article what is meant by ‘progressive abdominal pain’. Is this continuous abdominal pain, or worsening of abdominal pain, or both? Nor is it clear what is meant by ‘mechanical complaints’. Certainly, the need to exclude other causes of abdominal pain should be stressed, given that the majority of patients who have hepatic hemangiomas will, in fact, have another cause for abdominal pain if pain is present. Erdogan et al. also recommend surgical intervention in patients where there is diagnostic doubt.1 There is no clear statement about what earlier CT scans or other investigations may have found. No mention is made of the role of PET scanning in cases of diagnostic doubt, particularly where an earlier PET has shown glucose avidity of the primary tumor. Clearly an algorithm is required for the investigation of hepatic hemangiomas, especially where risk factors for hepatocellular carcinoma are present, or when there is a past history of gastrointestinal cancer. Furthermore, the importance of reviewing old investigations, particularly previous imaging and operation reports should be emphasized. All these should be seen as routine aspects of the functioning of appropriately specialized hepatobiliary and liver units. Hepatic hemangiomas tend to follow a benign course and some authors recommend a nonoperative approach for the majority of these lesions.3 Many readers of Journal of Gastroenterology and Hepatology will be part of networks of those involved in detection, investigation, and referral of such patients, yet be outside the specialized units which manage them. Clinicians therefore need to be aware of the appropriate algorithms for investigation of liver lesions thought to be hemangiomas and of the factors leading to selection for surgical intervention. They should also be aware of the risks of invasive investigations, especially fine needle aspiration cytology or core biopsy. Finally, they should be aware of the need to refer such patients to hepatobiliary units able to address all the issues encountered in managing these problems. In doing so, clinicians also need to avoid compromising patient outcomes by inappropriate referrals, delays, or invasive investigations. A recent systematic review has given rise to the recommendation for large long-term randomized clinical trials.4 A trial would have to be focused on the issues of relevance, including the various aspects of case selection and the investigation algorithm, a well-defined use of the concept of ‘diagnostic doubt’, and complications of the underlying disease process rather than the complications of surgical therapy and long-term symptom resolution. Unfortunately, largely for logistic reasons, it is unlikely that such trials will be performed. However, some information will accumulate from reported individual unit experiences and comparisons across these. This will improve our understanding of the management of this disease, provided the units involved are appropriately sophisticated and the data collection is of adequate quality.