The surgical domain can be seen as more complex and high risk in its delivery of care than other non-interventional specialities. It is therefore not surprising that in the majority of studies of adverse events in healthcare, at least 50% occurred within the surgical domain and the majority of these in the operating theatre. Furthermore, at least half of these adverse events were also deemed preventable. Just as the multiple studies in the developed world have similar figures for adverse events in hospitalised patients across all specialities, there appears to be a similar rate of harm in surgery. A review of 14 studies, incorporating more than 16000 surgical patients, quoted an adverse event occurring in 14.4% of surgical patients. This was not simply minor harm; a full 3.6% of these adverse events were fatal, 10% severe and 34% moderately harmful. Gawande, a surgeon from Boston, made one of the first attempts to clarify the source of these adverse events.
This paper pioneered the concept that the majority of these adverse events were not due to lack of technical expertise or surgical skill on the part of the surgeon, finding instead that ‘systems factors’ were the main contributing factor in 86% of adverse events. The most common system factors quoted were related to the people involved and how they were functioning in their environment. Communication breakdown was a factor in 43% of incidents, individual cognitive factors (such as decision-making) were cited in 86%, with excessive workload, fatigue and the design or ergonomics of the environment also contributing.
These findings were confirmed in the systematic review of surgical adverse events, where it was found that errors in what were described as ‘nonoperative management’ were implicated in 8.32% of the study population versus only 2.5% contributed to by technical surgical error. In accordance with other high-risk industries, such as commercial aviation, the majority of these adverse events are therefore not caused by failures of technical skill on the part of the individual surgeon, but rather lie within the wider healthcare team, environment and system. Lapses and errors in communication, teamworking, leadership, situational awareness or decision-making all feature highly in post-hoc analysis of surgical adverse events. This knowledge of error causation has been prominent and acknowledged in most other high-risk industries for many years, but it is only recently that healthcare has appreciated this.
Benign liver tumours are common and are frequently found coincidentally. Most benign liver lesions are asymptomatic, although larger lesions can cause non-specific complaints such as vague abdominal pain. Although rare, some of the benign lesions, e.g. large hepatic adenomas, can cause complications such as rupture or bleeding. Asymptomatic lesions are often managed conservatively by observation. Surgical resection can be performed for symptomatic lesions or when there is a risk of malignant transformation. The type of resection is variable, from small, simple, peripheral resections or enucleations, to large resections or even liver transplantation for severe polycystic liver disease.
Hepatocellular adenomas (HCA) are rare benign hepatic neoplasms in otherwise normal livers with a prevalence of around 0.04% on abdominal imaging. HCAs are predominantly found in women of child-bearing age (2nd to 4th decade) with a history of oral contraceptive use; they occur less frequently in men. The association between oral contraceptive usage and HCA is strong and the risk for a HCA increases if an oral contraceptive with high hormonal potency is used, and if it is used for over 20 months. Long-term users of oral contraceptives have an estimated annual incidence of HCA of 3–4 per 100000. More recently, an increase in incidence in men has been reported, probably related to the increase in obesity, which is reported as another risk factor for developing HCA. In addition, anabolic steroid usage by body builders and metabolic disorders such as diabetes mellitus or glycogen storage disease type I are associated with HCAs. HCAs in men are generally smaller but have a higher risk of developing into a malignancy. In the majority of patients, only one HCA is found, but in a minority of patients more than 10 lesions have been described (also referred to as liver adenomatosis).
Small HCAs are often asymptomatic and found on abdominal imaging being undertaken for other purposes, during abdominal surgery or at autopsy. Some patients present with abdominal discomfort, fullness or (right upper quadrant) pain due to an abdominal mass. It is not uncommon that the initial symptoms of a HCA are acute onset of abdominal pain and hypovolaemic shock due to intraperitoneal rupture. In a series of patients who underwent resection, bleeding was reported in up to 25%. The risk of rupture is related to the size of the adenoma. Exophytic lesions (protruding from the liver) have a higher chance of bleeding compared to intrahepatic or subcapsular lesions (67% vs 11% and 19%, respectively, P<0.001). Lesions in segments II and III are also at higher risk of bleeding compared to lesions in the right liver (35% vs 19%, P = 0.049).
There is no guideline for the treatment of HCAs, although there are general agreements. In men, all lesions should be considered for surgical resection independent of size, given the high risk of malignant transformation, while taking into account comorbidity and location of the lesion. Resection should also be considered in patients with HCAs due to a metabolic disorder. In women, lesions <5 cm can be observed with sequential imaging after cessation of oral contraceptive treatment. In larger tumours, treatment strategies vary. Some clinicians have proposed non-surgical management if hormone therapy is stopped and patients are followed up with serial radiological examinations. The time period of waiting is still under debate, however recent studies indicate that a waiting period of longer than 6 months could be justified.
More recently, the subtypes of the Bordeaux classification of HCA have been studied related to their risk of complications. Some groups report that percutaneous core needle biopsy is of limited value because the therapeutic strategy is based primarily on patient sex and tumour size. Others report a different therapeutic approach based on subtype. Thomeer et al. concluded that there was no evidence to support the use of subtype classification in the stratification and management of individual patients related to risk of bleeding. Size still remains the most important feature to predict those at risk of bleeding during follow-up. However, malignant transformation does seem to be related to differences in subtypes. β-catenin-mutated HCAs trigger a potent mitogenic signalling pathway that is prominent in HCC. Cases of inflammatory HCAs can also show activation of the β-catenin pathway with a risk of developing malignancy. Therefore, β-catenin-mutated and inflammatory HCAs are prone to malignant degeneration, and particularly if >5cm. In these circumstances, invasive treatment should be considered.