In patients with preserved liver function and limited disease, hepatic resection continues to provide the primary therapeutic modality in the treatment of HCC. Advantages over liver transplantation include the lack of restrictions based on stringent size criteria, the avoidance of prolonged waiting times for organ availability, and the avoidance of chronic immunosuppressive medications with potentially hazardous side-effects that are required to maintain graft survival, which, in the setting of malignancy, are frankly unknown.
The ability to perform a safe and effective resection is highly dependent on two factors. The first is the importance of performing a complete microscopic resection (R0). The second factor is the condition of the underlying liver remnant. While a precancerous field defect is obviously not eliminated, safe resections are dependent on the ability of the underlying future liver remnant (FLR) to undergo adequate regeneration, which is dependent on the two factors described above. Preservation of an adequate FLR remains the major obstacle to the safe conduct of hepatic resection in patients with HCC.31-34 In patients with adequate FLR volumes, the liver is able to regenerate at a remarkable capacity and over a short period of time (days to weeks), during which time a transient decline in hepatic function may be observed. However, in the event that major resections are undertaken, FLR volumes are compromised beyond a threshold at which point regeneration may not be possible. Combined with the presence of underlying liver dysfunction, cirrhosis, or fibrosis, permanent liver dysfunction may result following even minor liver resections. With an insidious decline in hepatic function, fulminant liver failure, and potentially death may ensue.
While arbitrary FLRs of 20% (in healthy liver) and 40% (cirrhotic liver) are often quoted, the ability of liver to regenerate may be further impeded by preoperative, intraoperative, and postoperative complications, such as major hemorrhage during liver resection, which is more common with concomitant hepatic dysfunction secondary to abnormal coagulation parameters. Incidentally, major hemorrhage is an independent predictor for tumor recurrence and death in HCC.35 Serious liver failure as a result of liver resection continues to provide the most significant obstacle to liver resection in HCC, since patients often have significant parenchymal damage (fibrosis, NAFLD, and cirrhosis) at presentation.36 Careful assessment of liver function using objective tests such as the Child–Pugh classification cannot be overemphasized, followed by careful preoperative planning to include liver volumetric studies in cases where FLR volumes may be close to borderline thresholds for regeneration. As a general rule, Child–Pugh classes B and C are excluded from hepatic resection as well as patients with clinically relevant portal hypertension. A platelet count below 100,000 is indicative of significant portal hypertensive disease and is used as an arbitrary threshold. Additional intraoperative tests such as indocyanine green (ICG) clearance and portal pressure measurement (cutoff of 10 mmHg) have been used to help identify patients who are likely to suffer from postoperative hepatic failure, particularly in clinically occult liver cirrhosis.37 Interestingly, a MELD score cutoff of 9 has been validated as predictive of the development of postoperative hepatic failure, scores above which have been shown to result in mortality rates of 13% and morbidity rates of 50% in those patients.38
In attempts to optimize the FLR volume, preoperative portal vein embolization (PVE) was introduced in the 1980s. The portal vein supplying the portion of liver to be resected is embolized, which results in ipsilateral atrophy and compensatory contralateral hypertrophy of the remnant liver lobe over a period of approximately 6 weeks. By increasing the volume of the FLR, the risk of postoperative hepatic failure would be expected to be reduced, and studies have since validated the effectiveness of this strategy in reducing postoperative morbidity following hepatic resection.39 In a study by Palavecino et al,40 the authors showed that while use of preoperative PVE resulted in equivalent survival outcomes to patients who did not undergo the hypertrophy-inducing procedure, there were no deaths recorded in their PVE group, compared with an 18% mortality rate in the group who did not undergo PVE. Furthermore, in recent analyses by Shindoh et al,41 remnant liver growth after PVE was shown to predict posthepatectomy outcomes better than conventional static liver measurements.
Associated liver partition with portal vein ligation for staged hepatectomy (ALPPS) is a novel two-step technique for ensuring adequate liver volume in patients with limited anatomic reserve, instead of PVE.42 A liver splitting procedure is performed during the first stage, along with operative portal vein ligation, which results in a pronounced short-term parenchymal hypertrophy over approximately 1 week, likely due to the transection of bridging veins between liver segments, which does not occur with PVE alone. The utility of ALPPS over conventional PVE has recently been strongly debated.43
Except in very select circumstances, patients who present with extrahepatic disease are excluded from resection, as they would be from liver transplantation. Relative contraindications to resection include tumor involvement within the portal vein or vena cava.44 Preoperative planning should, therefore, include careful analysis of tumor size, disease multifocality, the number of nodules present, the tumor's proximity to major vessels, and the extent of margin required to achieve R0 resection. Three-dimensional volumetric studies using CT or MRI preoperatively are more sophisticated than ever, and aid in determining the extent of liver resection and volume of the FLR (see Chapters 123 and 124).
Reassuringly, while 1-cm margins have traditionally been sought during the resection of HCC, narrow margins have been shown to exhibit similar recurrence rates and offer no additional advantage from wider margins.45,46 Anatomic resections, where possible, are preferred, and confer higher survival than patients who undergo nonanatomic resections, although this should not be performed at the expense of FLR volumes.47 Recurrent resections, if focal, may offer extended disease-free intervals and are warranted if repeat R0 resections can be achieved.
Although with strict adherence to Milan criteria transplantation survival rates have been high, partial hepatectomy rates have not been able to match those of transplantation, despite improvement in the safety of liver resection, particularly at specialized institutions where such operations are repeatedly performed. It is important to note, however, that patients who undergo hepatic resection are more liberally screened than patients who undergo transplantation, and include larger tumors, with multifocal disease, which may often be associated with early vascular invasion. As such, outcomes following transplantation may be superior based on differing risk categories of patients enrolled. In fact, when patients with more limited disease and favorable tumor characteristics who have undergone hepatic resection are analyzed, the results of hepatic resection approach those of liver transplantation.48-51 Furthermore, disease recurrence in resected patients was confined to the liver in the majority of cases, nearly all of whom were within criteria for subsequent salvage liver transplantation following close surveillance in the postoperative period.52 Reports to confirm the safety of salvage liver transplantation exist and confer equivalent short- and long-term outcomes to patients who undergo initial liver transplantation alone.53-55 As such, hepatic resection may then be reasonably viewed as an alternative bridge to transplantation, which is a concept that is currently being evaluated. This would provide a feasible strategy to halt disease progression while awaiting liver transplantation, but additionally provides important tumor histopathologic data from the resected specimen, which may help exclude patients who are unlikely to benefit from organ transplantation.
Of note, options for adjuvant chemotherapy following resection are limited. However, in the case of HCC arising secondary to chronic HBV cirrhosis, antiviral therapy with nucleotide analogues reduces recurrence rates following curative recurrence in patients with high HBV loads.56,57 Furthermore, elevated preoperative hepatitis B surface antigen (HbsAg) levels greater than 1000 IU/mL has been shown to be an independent predictor of recurrence of HCC.58,59 The use of interferon in HCV-related HCC is similarly thought to reduce recurrence following curative resection.60 With the advent of newer therapies targeting HCV cure, recurrence rates may be reduced even further. Given the continued underlying cirrhotic field defect following liver resection, however, close surveillance following curative resection to monitor for new tumors, as well as tumor recurrence, is imperative.