Major hepatic resection: from volumetry to liver scintigraphy

The minimum future remnant liver (FRL) is required to anticipate postoperative risk, to assess the need for portal vein embolization (PVE), and to modulate the immediate postoperative course of extended liver resection, such as in the technique of Associating Liver Partition and Portal Vein Ligation for Staged Hepatectomy (ALPPS). There is little consensus on the required residual volume. Classical cut-offs vary from 20 to 30% of total liver volume (TLV), to standardized TLV¹ or has been determined at 0.5% according to the ratio to body weight.² There is a tendency to resect increasingly compromised livers, the post-operative function of which and the capacity to regenerate can differ for similar remnant liver mass following previous intensive chemotherapy or associated with obesity. This has led to a reappraisal of volumetric cut-offs to 37.7%³ and to the incorporation of pathological parameters into formulas such as that developed by the Leeds group. These take into account steatosis in calculating a so-called functional remnant volume.⁴

Data show that volumetric evaluation must be combined by functional assessment since the main limitation of using volumetry, as the sole standard for estimating the risk of postoperative liver failure (LF), is that the volume may not reflect the functional mass of the liver after major hepatectomy. However, the most appropriate test or score for estimating liver function and the perioperative risks remains to be validated. Routine blood tests provide indirect or limited information regarding liver function by quantifying substances or markers synthetized by the liver such as albumin or coagulation factors but these reflect only one specific function of hepatocytes and the liver. These tests often record normal levels in patients with normal or injured but non-cirrhotic livers. Other quantitative liver function tests characterize the elimination process of substances cleared by the liver, assessing the global liver function. Clearance of indocyanine green (ICG) has been validated in cirrhotic patients but is useless in patients with colorectal cancer metastases. Metabolism of C-methacetin by the liver analyzed by CO₂ expiration (calculated for total liver, LiMax value), can extrapolate the estimated remnant liver function when expressed as a ratio of computed tomography (CT) volume, so-called residual LiMax.⁵ However, such markers reflect the whole liver function, not the directly-measured function of the FRL itself. These methods suppose a homogeneous distribution of function inside the liver volume but this may not be the case due to vascular or biliary involvement or compression related to the tumour. Furthermore, distribution of fibrosis or chemo-induced injuries is heterogeneous, so volumetry may compare a part of the liver with differing functional values. This is especially true after portal vein embolization (PVE) where the partition of the function between the embolized liver and non-embolized liver is unknown.

The best tool for functional evaluation of the future remnant itself would appear to be hepatobiliary scintigraphy (HBS). Isotopic imaging techniques have now been developed to evaluate liver function from a regional perspective. (99m)Tc-mebrofenin HBS has shown the highest uptake performance and can be associated with static or dynamic single-photon emission CT to assess segmental liver function. HBS can measure functional liver volume inside the FRL, using the outline extraction method calculating the uptake capacity per minute in the region of interest of the future remnant. In a previous paper, the group from Amsterdam showed a threshold of uptake of 2.5%/min/m² below which the risk of death by liver failure was estimated to be increased twenty fold.⁶ Moreover, a report analyzing 24 patients undergoing PVE, determined that HBS uptake could reach 2.69%/min/m² even if volume did not increase beyond 25%, and thus permitting liver resection.⁷ Cieslak et al. have also reported last month the value of this technique in assessing overall liver function that was shown to decline with age although even in a controlled non-operative setting the obtained values were variable.⁸

In this issue, the same group (Cieslak et al.⁹) report the cumulative Amsterdam center's experience of (99m)Tc-mebrofenin HBS and the impact of its implementation in the preoperative management of major liver resection. In 163 patients, HBS appeared superior in the identification of patients at high risk of LF compared to the volumetric methods, resulting in lower rates of LF and LF-related death compared to an historical cohort. Undeniably, HBS may bypass or complement the inadequacies of volumetric evaluation in the preoperative assessment of liver resection. However, there may be potential bias due to the difference use of historical controls and as a result of improvements in the supportive treatment of patients with postoperative LF. This recent study requires validation in a prospective intention-to-treat cohort. Moreover, a precise cut-off such as the change in HBS before and after PVE may be more relevant in appreciating actual liver function, as with results in ALPPS patients¹⁰ in which the change of HBS may assist in postponing the second stage of resection.

Finally, there are limitations of the role of volumetric or functional evaluation of the FLR in assessing the morbidity and mortality after liver resection. Preoperative score or cut-off anticipates the risk of liver failure but do not take into account intraoperative events or immediate postoperative technical complications. The question remains whether preoperative assessment of the function alone can anticipate the behaviour of the liver after resection in the light of such perioperative events. For two similar patients with identical functional FRL, other factors such as massive blood loss, severe ischemia-reperfusion injury or a postoperative event may be more relevant. The multifactorial and complex nature of post resection LF presumably explains the relative relevance of purely biological definitions of LF that only reflect its consequences.¹¹ Given the multifactorial origin of LF, some have proposed composite scores associating biochemical and clinical data, and even including postoperative complications¹² to better determine the immediate postoperative course following liver resection.

The article of Cieslak et al.⁹ seems to show that HBS can be detected more accurately than the volumetric methods alone in those patients at risk of liver failure before extended hepatectomy. HBS may circumvent the need for volumetric cut-offs and may become a mandatory tool in the liver surgeon's armamentarium.