Parenchyma-sparing hepatectomy (PSH) versus non-PSH for bilobar liver metastases of colorectal cancer

Introduction

Within three years after the diagnosis of colorectal cancer (CRC), 29% of patients will have developed liver metastases [1]. If these can be resected, 5-year survival of up to 60% is achievable [2-4]. Curative liver resection is feasible when the metastases can be radically resected whilst sufficient liver volume can be spared [5]. The number of patients eligible for resection has expanded over time, through advances in surgical procedures and the development of multimodal treatment strategies, in which preoperative chemotherapy is used to decrease tumor burden, and portal vein embolization (PVE) is applied to increase the liver volume spared after resection [6,7].

This has also increased the possibilities for resection in the case of bilobar liver metastases. However, bilobar disease inevitably implies balancing between resecting enough liver tissue to achieve radicality and sparing enough parenchyma to prevent postoperative liver failure. When treating bilobar metastases, resection in practice often implies at least a hemihepatectomy, sacrificing one lobe entirely and sparing very little liver tissue. Previous research has shown that parenchyma-sparing surgery for CRC liver metastases can be feasible and does not compromise the outcome when compared to more extensive surgery; indeed, it may even have better results in terms of complications and overall survival [8-10]. However, in these studies most patients had only a few metastases and often merely unilobar spread. Another study, including only patients with at least four metastases and thus often with bilobar spread, found no certain difference in survival based on the extent of resection [11]. The strategy of parenchyma-sparing hepatectomy (PSH) is presumably most relevant in patients with bilobar disease, where it can be hard to preserve a sufficient amount of liver parenchyma. Yet, studies show that advances are being made in the application of parenchyma-sparing liver surgery, even in patients with technically challenging tumor locations, in which intraoperative ultrasound guidance plays an important role [12,13]. Few studies have shown comparable oncological results when applying PSH compared to more extensive surgery (non-PSH) in patients with bilobar disease; whilst PSH might improve perioperative morbidity and mortality [14,15], a true consensus is lacking.

The aim of this retrospective study was to compare the perioperative course and overall survival after liver resection for bilobar CRC metastases in patients who underwent PSH, i.e., segmentectomies and minimal resections in both liver lobes, preserving liver parenchyma, and those who were treated with non-PSH, i.e., an extended hemihepatectomy or hemihepatectomy plus any resection needed on the other lobe.

Patients and methods

Results

Patients

A total of 119 patients with liver resection for bilobar CRC metastases were included. Of these, 59 underwent PSH and 60 patients underwent non-PSH. Both groups had a comparable distribution over time for operation, with 2012 and 2011 as the median years of operation in the PSH group and the non-SPH group, respectively. Patients undergoing PSH had a higher median age (P<0.01), but the two groups were comparable regarding sex, ASA score and staging of their CRC (Table 1). Although the median size of the largest metastasis was comparable between groups, the median number of metastases was significantly higher in the group with non-PSH and this group also included patients who underwent preoperative PVE, whereas the group with PSH did not. Preoperative chemotherapy administration was similar in both groups, and most often entailed oxaliplatin-based treatment, followed by irinotecan-based treatment.

Table 1 Baseline characteristics in patients undergoing PSH or non-PSH for bilobar CRC metastases

Perioperative outcome

The patients treated with PSH had statistically significantly less median intraoperative bleeding, a shorter operation time and a shorter postoperative stay in hospital (Table 2). Although the division among the different complication grades differed slightly between groups, with more patients having grade 0 or I after PSH and grade II being more common after non-PSH, the occurrence of severe complications was comparable. Radical resection was achieved in similar numbers in both groups.

Table 2 Perioperative outcome in patients undergoing PSH or non-PSH for bilobar CRC metastases

Within the entire study population there were no perioperative deaths, yielding a total 30-day and 90-day mortality rate of 0%.

Long-term outcome

The median follow-up time was 35 months among survivors. The median overall unadjusted survival, calculated by the Kaplan-Meier method, was 39 months after PSH and 51 months after non-PSH (P=0.139). The Kaplan-Meier curve for overall survival in the two groups is presented in Fig. 1, which also shows the unadjusted overall survival using Cox regression (P=0.141).

Figure 1 Unadjusted overall survival curves showing overall survival in patients undergoing parenchyma-sparing hepatectomy (PSH) or non-PSH for bilobar colorectal cancer metastases, as demonstrated by the Kaplan-Meier method (A) and unadjusted Cox regression (B)

Adjusted overall survival was calculated considering age, sex, the patient variables with significant differences or tendencies to differ between groups (ASA score, synchronous versus metachronous metastases, presence of lung metastases, number of liver metastases, size of the largest metastasis and preoperative treatment with PVE), as well as perioperative parameters including the occurrence of severe complications and radicality of resection. Using Cox proportional hazards regression analysis, no statistically significant differences in survival were found when comparing patients who underwent PSH with those who underwent non-PSH (Fig. 2). Hazard ratios for extensive versus limited resection are displayed in Table 3.

Figure 2 Adjusted overall survival curves showing overall survival in patients undergoing parenchyma-sparing hepatectomy (PSH) or non-PSH for bilobar colorectal cancer metastases, calculated through adjusted Cox regression

Table 3 Cox multivariable regression analysis for overall survival

Discussion

The present study is one of few reports focusing on the influence of the extent of liver surgery on perioperative outcome and long-term survival, specifically in patients with bilobar liver metastases. An important argument against performing PSH when dealing with bilobar CRC liver metastases has been the suggestion that this might have a negative impact on radicality. Radicality, i.e., a tumor-free resection margin, is a prognostic factor for long-term survival [18,20], and some have found a higher rate of positive margins when comparing wedge resections to larger anatomical resections [21]. In the current study, however, no difference in radicality was found between groups. This finding is supported by the study of Memeo et al, who also compared patients with bilobar disease undergoing PSH to those undergoing non-PSH and found similar rates of radical resection [15].

The present study demonstrated differences in perioperative course when comparing both types of surgery. When the resected liver volume was limited, this was associated with a smaller amount of perioperative bleeding, a shorter operation time and shorter hospital stay. Also, the division amongst the different grades of complications was different. However, when the number of severe complications was considered, the groups were comparable. A favorable perioperative course after PSH was also reported by Memeo et al, showing fewer complications, although hospital stay was comparable in their study [15]. In the study by Gold et al [14], groups were separated according to different time periods and this showed that less extensive resections were performed during later periods, and were correlated with a decrease in blood loss, hospital stay and 90-day mortality over time. They also showed that extensive resections were related to a higher morbidity.

Besides the possibly favorable perioperative course, another reason to opt for PSH is the possibility of future resections in case the patient suffers from hepatic recurrence. If achievable in these cases, reresection has been shown to be beneficial [22,23], but is more likely to be ruled out in patients with a previous non-PSH and thus less remaining liver tissue.

This study showed comparable overall survival in patients treated with PSH and those who underwent non-PSH. However, one difficulty in the current study, which was a retrospective study with all its limitations, was the difference between the two study groups regarding certain aspects that may have affected outcomes. First of all, the median age was higher in the group with PSH, reported to negatively influence survival [2]. Also, this group included 10 patients who had lung metastases at the time of liver resection, whilst the group with non-PSH only contained two patients with lung metastases. Since the overall 3-year survival for patients with lung metastases from CRC is reported to be only in the range of 11-14% [24], the presence of lung metastases would thus be considered as a negative prognostic factor. Additionally, the median number of liver metastases differed between the two groups. Several studies found a higher number of liver metastases to be related to worse survival [25-27], whereas others have not judged this to be a negative predictive factor [28]. A higher number of liver metastases might also have influenced the surgeon to consider non-PSH in the first place, and the possible prognostic value of this parameter should be taken into account. To adjust for the abovementioned factors and also taking other marginally differing variables into account, an adjusted survival analysis was performed with Cox regression analysis and this still showed no statistically significant difference in survival between the two groups. This finding is in accordance with previous studies involving patients with bilobar liver metastases, also showing comparable overall survival in patients with extensive and with limited surgery [14,15]. Therefore, the presumption that PSH might negatively influence long-term outcome is contradicted.

In conclusion, when comparing the perioperative outcome and long-term survival in CRC patients who underwent PSH or non-PSH for bilobar liver metastases, our study could not demonstrate any disadvantage when using PSH. Instead, possible advantages compared to non-PSH—e.g., less intraoperative bleeding, shorter operation time and shorter hospital stay—were noted. Furthermore, minimizing the liver resection did not appear to influence radicality. Although extensive resections often are unavoidable because of tumor location and technical aspects, our findings imply that the application of PSH is safe when feasible, thereby sparing liver parenchyma and facilitating possible future liver resections for local recurrences.