Advertisement

Case-controlled study comparing peri-operative and cancer-related outcomes after major hepatectomy and parenchymal sparing hepatectomy for metastatic colorectal cancer

Open ArchivePublished:May 09, 2017DOI:https://doi.org/10.1016/j.hpb.2017.04.007

      Abstract

      Introduction

      Liver resection is potentially curative in selected patients with colorectal liver metastases (CLM). There has been a trend towards parenchyma sparing hepatectomy (PSH) rather than major hepatectomy (MH) due to lower perioperative morbidity. Although data from retrospective series suggest that long-term survival after PSM are similar to MH, these reports may be subject to selection bias. The aim of this study was to compare outcomes of PSH and MH in a case-controlled study.

      Patients and methods

      917 consecutive patients who underwent liver resection for CLM during 2000–2010 were identified from a prospective database. 238 patients who underwent PSH were case-matched with 238 patients who had MH, for age, gender, tumour number, maximum tumour diameter, primary Dukes' stage, synchronicity and chemotherapy status using a propensity scoring system. Peri-operative outcomes, recurrence and long-term survival were compared.

      Results

      Fewer PSH patients received peri-operative blood transfusions (p < 0.0001). MH patients had greater incidence of complications (p = 0.04), grade III/IV complications (p = 0.01) and 90-day mortality (p = 0.03). Hospital stay was greater in the MH group (p = 0.04). There was no difference in overall/disease-free survival.

      Conclusion

      Patients with resectable CLM should be offered PSH if technically feasible. PSH is safer than MH without compromising long-term survival.

      Introduction

      Colorectal cancer is the second commonest cause of cancer-related death in the western world, and more than half of patients will develop metastatic disease, frequently limited to the liver.
      • Karanjia N.D.
      • Lordan J.T.
      • Fawcett W.J.
      • et al.
      Survival and recurrence after neo-adjuvant chemotherapy and liver resection for colorectal metastases – a ten year study.
      • Simmonds P.C.
      • Primrose J.N.
      • Colquitt J.L.
      • et al.
      Surgical resection of hepatic metastases from colorectal cancer: a systematic review of the published studies.
      Without treatment, the prognosis of patients with colorectal cancer liver metastases (CLM) is dismal. For selected patients with CLM, long-term survival and even cure has become feasible due to advances in liver surgical techniques and availability of effective chemotherapeutic agents.
      • Karanjia N.D.
      • Lordan J.T.
      • Fawcett W.J.
      • et al.
      Survival and recurrence after neo-adjuvant chemotherapy and liver resection for colorectal metastases – a ten year study.
      • Simmonds P.C.
      • Primrose J.N.
      • Colquitt J.L.
      • et al.
      Surgical resection of hepatic metastases from colorectal cancer: a systematic review of the published studies.
      • Nordlinger B.
      • Sorbye H.
      • Glimelius B.
      • et al.
      Perioperative chemotherapy with FOLFOX4 and surgery versus surgery alone for resectable liver metastases from colorectal cancer (EORTC Intergroup trial 40983): a randomised controlled trial.
      In recent years, there has been a trend in favour of parenchymal-sparing hepatectomy (PSH) over major hepatectomy (MH) for patients with resectable liver-only disease.
      • Ferrero A.
      • Vigan L.
      • Lo Tesoriere R.
      • et al.
      Bisegmentectomies as alternative to right hepatectomy in the treatment of colorectal liver metastases.
      • Sui C.J.
      • Cao L.
      • Li B.
      • et al.
      Anatomical versus non-anatomical resection for colorectal liver metastases: a meta-analysis.
      Early reports indicated that PSH was associated with higher positive margin rates and worse long-term survival compared to MH.
      • DeMatteo R.P.
      • Palese C.
      • Jarnagin W.R.
      • et al.
      Anatomic segmental hepatic resection is superior to wedge resection as an oncologic operation for colorectal liver metastases.
      • Kokudo N.
      • Tada K.
      • Seki M.
      • et al.
      Anatomical major resection versus nonanatomical limited resection for liver metastases from colorectal carcinoma.
      • Zorzi D.
      • Mullen J.T.
      • Abdalla E.K.
      • et al.
      Comparison between hepatic wedge resection and anatomic resection for colorectal liver metastases.
      These differences were not observed in several recent series,
      • Karanjia N.D.
      • Lordan J.T.
      • Fawcett W.J.
      • et al.
      Survival and recurrence after neo-adjuvant chemotherapy and liver resection for colorectal metastases – a ten year study.
      • Lordan J.T.
      • Karanjia N.D.
      • Quiney N.
      • et al.
      A 10 year study of outcome following hepatic resection for colorectal liver metastases - the effect of evaluation in a multi-disciplinary team setting.
      including a meta-analysis of 1662 patients.
      • Sui C.J.
      • Cao L.
      • Li B.
      • et al.
      Anatomical versus non-anatomical resection for colorectal liver metastases: a meta-analysis.
      Due to the retrospective, uncontrolled nature of these studies, it is feasible that any differences in oncological results between PSH and MH may have been concealed by selection bias.
      • Mise Y.
      • Aloia T.A.
      • Brudvik K.W.
      • et al.
      Parenchymal-sparing hepatectomy in colorectal liver metastasis improves salvageability and survival.
      Risk factors for disease recurrence after resection of CLM, such as tumour size and number, use of perioperative chemotherapy, perioperative blood transfusion and postoperative complications,
      • Karanjia N.D.
      • Lordan J.T.
      • Fawcett W.J.
      • et al.
      Survival and recurrence after neo-adjuvant chemotherapy and liver resection for colorectal metastases – a ten year study.
      • Simmonds P.C.
      • Primrose J.N.
      • Colquitt J.L.
      • et al.
      Surgical resection of hepatic metastases from colorectal cancer: a systematic review of the published studies.
      • Sui C.J.
      • Cao L.
      • Li B.
      • et al.
      Anatomical versus non-anatomical resection for colorectal liver metastases: a meta-analysis.
      • Weber S.M.
      • Jarnagin W.R.
      • DeMatteo R.P.
      • et al.
      Survival after resection of multiple hepatic colorectal metastases.
      must be taken into account when evaluating the relative merits of PSH and MH. Our aim was to perform a case-controlled analysis of the outcomes of patients undergoing parenchymal-sparing or major hepatectomy for colorectal liver metastases in a single high volume institution.

      Patients and methods

      This was a case controlled comparison analysis of prospectively collected data over an eleven year period (January 2000 to December 2010). All consecutive patients who underwent liver resection for CLM during the study period were identified (n = 917, Table 1). Data were anonymous according to the ethical standards of the Medical Research Council Good Clinical Practice in Clinical Trials.
      Table 1All available cases prior to matching
      Major hepatectomy (n = 634)Parenchymal sparing hepatectomy (n = 283)
      Gender ratio M:F1.7:11.8:1
      Mean age/years (range, SD)67 (21–86, 2.02)67 (31–87, 1.90)
      Chemotherapy
       Yes (%)401 (63.3)170 (60.1)
       No (%)233 (36.8)113 (39.9)
      Number of CLM/n (%)
       Single metastasis382 (60.3)190 (67.1)
       2 metastases124 (19.6)57 (20.1)
       3 metastases78 (12.3)16 (5.7)
       4 metastases34 (5.4)12 (4.2)
       >4 metastases16 (2.5)8 (2.8)
      Mean maximum tumour diameter/mm (range, SD)4.7 (0.1–20, 3.46)3.2 (0.5–14, 1.90)
      Primary tumour Dukes' Stage (%)
       A14 (2.2)13 (4.6)
       B147 (23.2)65 (23.0)
       C398 (62.8)193 (68.2)
       D2 (0.3)1 (0.4)
      Synchronous metastases/n (%)41 (6.5)18 (6.4)
      SD = Standard deviation.
      Patients were grouped according to type of liver resection; those who had major hepatectomies (MH, n = 634) and those who had parenchymal sparing hepatectomy (PSH, n = 283). At the time of surgery, patients were selected for MH or PSH according to the surgeon's preference. MH were defined as liver resections removing ≥3 segments, and PSH had fewer than 3 segments removed. Patients receiving PSH were 1:1 matched to MH on a case by case basis using a propensity scoring system,
      • Rees M.
      • Tekkis P.
      • Welsh F.
      • et al.
      Evaluation of long-term survival after hepatic resection for metastatic colorectal cancer.
      which was guided by a statistician (JH).
      Patients were included if they potentially could have undergone either MH or PSH on review of the preoperative imaging. Exclusion criteria included patients with small isolated peripheral lesions, left lateral segmentectomies, portal vein embolisations, and patients who underwent radiofrequency ablation. Furthermore, patients with large lesions that could not have undergone PSH safely were also excluded. Therefore, 238 PSH patients were included in the study.
      The factors used in the matching were age, gender, tumour number, maximum tumour diameter, primary Dukes' stage, cancer involved resection margins, synchronous metastases and chemotherapy status (238 patients in each group). Age matches were within ±2 years, whist the dichotomous variables (e.g. gender) were matched exactly. Where a PSH patient could potentially be paired with multiple MH patients, the match whose date of surgery was closest chronologically was used.
      The pre-operative imaging of patients who underwent MH were reviewed and those who were potentially eligible for parenchymal-sparing resection were included in the matching process. Standardised differences were calculated for each matched variable, to determine the quality of matching.
      • Aurtin P.
      Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples.
      A standardised difference of <0.1 was deemed to be indicative of a closely matched variable.
      • Aurtin P.
      Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples.
      Parenchymal sparing resections were undertaken in a non-anatomical fashion, with the objective of achieving negative margins. Nine of the PSH patients had a laparoscopic resection.
      Data included patients' demographics, peri-operative blood transfusions, peri-operative complications, 30-day, 90-day and hospital mortality, lengths of critical care (intensive care or high dependency) and hospital stays, resection margin status, hepatic insufficiency (defined by the International Study Group of Liver Surgery),
      • Fukushima K.
      • Fukumoto T.
      • Kuramitsu K.
      • et al.
      Assessment of ISGLS definition of posthepatectomy liver failure and its effect on outcome in patients with hepatocellular carcinoma.
      neo-adjuvant chemotherapy, disease free survival (DFS) and overall survival (OS). Cancer involved resection margins (R1) were defined as <1 mm. Peri-operative complications were graded according to the Clavien classification.
      • Dindo A.
      • Demartines N.
      • P-A C.
      Classification of surgical complications.
      Post-operative mortality was defined as death in hospital or within 90 days of surgery. DFS and OS were defined from the date of index liver resection to the date of first recurrence, death or latest follow up appointment. Patients were followed up six monthly for the first 3 years, and once a year thereafter. Patients underwent redo liver resections if the disease was technically resectable and the patient was deemed fit and willing for further surgery.
      Statistical analyses were performed using SPSS (version 21). Continuous data was reported as means and SD with p values from Paired T tests, and categorical data reported as percentages and p values from McNemar or Fishers exact tests. Survival curves were constructed using the Kaplan–Meier technique with log-rank tests used to compare between groups. Recurrence and death were considered time-to-event end points in the Kaplan–Meier analysis. Odds ratios (OR) and 95% confidence intervals (CI) were estimated and p < 0.05 was considered to be statistically significant throughout.
      All clinically relevant variables were included in a multivariable cox regression model, alongside the type of surgery, in order to account for potentially confounding factors for the entire cohort prior to matching. All statistical analyses were guided by our specialist statistician (Mr James Hodson).

      Results

      During the study period, the number of liver resections performed for CLM in our unit increased progressively from year 1 to year 11 (Fig. 1). The proportion of patients who underwent PSH each year also increased. The decision to undergo MH or PSH was decided preoperatively at a specialist hepatobiliary multidisciplinary meeting, guided by tumour characteristics and the predicted size of the future liver remnant. In individual cases, the procedure may have been modified by the operating surgeon depending on intra-operative findings.
      Figure 1
      Figure 1All consecutive liver resections performed between Jan 2000 and Dec 2010 (n = 917), demonstrating proportion of parenchymal sparing hepatectomy and major hepatectomy each year; Kaplan–Meier curve demonstrating disease free survival of case-matched patients following MH and PSH for CLM (Log rank test p = 0.62); Kaplan–Meier curve demonstrating overall survival of case-matched patients following MH and PSH for CLM (Log rank test p = 0.56)
      There was no difference in patients who received neo-adjuvant chemotherapy between the groups (8.4% v 7.1% respectively, p = 0.25, Table 3). In the PSH group, the type of neo-adjuvant chemotherapy given included oxaliplatin and 5-flurourocil (5-FU) (n = 16) and 5-FU alone (n = 1). Of these patients, 5 received greater than 4 cycles of neo-adjuvant chemotherapy. In the MH group, the type of neo-adjuvant chemotherapy given included oxaliplatin and 5-FU (n = 17), 5-FU alone (n = 2) and Capecitabine/Irinotecan (n = 1). Of these patients, 3 received greater than 4 cycles of neo-adjuvant chemotherapy.
      Table 2Standardised differences of matched variables between MH and PSH groups
      Major hepatectomy (n = 238)Parenchymal sparing hepatectomy (n = 238)Standardised difference
      • Aurtin P.
      Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples.
      Gender ratio M:F1.2:11.3:10.01
      Mean age/years (range, SD)64.8 (24–86, 9.68)65.7 (31–87, 9.99)0.09
      Chemotherapy
       Yes (%)135 (56.7)139 (58.4)<0.01
       No (%)103 (43.3)99 (41.6)
      Number of CLM/n (%)
       Single metastasis161 (67.7)153 (64.3)<0.01
       2 metastases44 (18.5)51 (21.4)<0.01
       3 metastases14 (5.9)15 (6.3)0.01
       4 metastases9 (3.8)12 (5.0)0.01
       >4 metastases9 (3.8)7 (2.9)0.02
      Mean maximum tumour diameter/mm (range, SD)3.2 (0.4–20, 1.8)3.1 (0.5–14, 1.8)0.06
      Primary tumour Dukes' Stage (%)
       A6 (2.5)9 (3.8)0.06
       B62 (26.1)49 (20.6)0.1
       C145 (60.9)132 (55.5)0.09
       D01 (0.4)<0.01
      Synchronous metastases/n (%)14 (5.9)15 (6.3)0.02
      SD = Standard deviation.
      Standardised difference of <0.1 suggests a closely matched variable.
      • Weber S.M.
      • Jarnagin W.R.
      • DeMatteo R.P.
      • et al.
      Survival after resection of multiple hepatic colorectal metastases.
      Table 3Peri-operative outcomes of case-matched patients who underwent major hepatectomy and parenchymal sparing hepatectomy during Jan. 2000 and Dec. 2010
      Major hepatectomy (n = 238)Parenchymal sparing hepatectomy (n = 238)p value
      Incidence of peri-operative blood transfusions (%)66 (27.7)24 (10.1)<0.0001
      Neo-adjuvant chemotherapy (%)20 (8.4)17 (7.1)0.25
      Mean days in ITU/Days (range, SD)1.49 (0–75, 6.25)0.68 (0–18, 2.73)0.02
      Median hospital stay/days (range, SD)7.0 (4–91, 7.27)6.0 (3–68, 5.73)0.4
      Complications (%)88 (37.0)69 (29.0)0.04
      Complications
      • Dindo A.
      • Demartines N.
      • P-A C.
      Classification of surgical complications.
      Grade III & IV (%)22 (9.2)9 (3.8)0.01
      Grade I & II (%)66 (27.7)60 (25.2)
      Peri-operative mortality/n (%)9 (3.8)2 (0.8)0.03
      Hepatic insufficiency/n (%)13 (5.5)0<0.0001
      Involved resection margins/R1/2 (%)18 (7.6)27 (11.3)0.14
      SD = Standard deviation.
      Continuous data reported as means and SD with p values from Paired T tests, and categorical data reported as percentages and p values from Fishers exact test.
      • Weber S.M.
      • Jarnagin W.R.
      • DeMatteo R.P.
      • et al.
      Survival after resection of multiple hepatic colorectal metastases.
      Standardised differences of matched variables were reported in Table 2. There was no significant difference in matched variables except Dukes' stage B between patients who underwent PSH compared to MH. The median time difference between dates of surgery for the matched cases was 17 months (range, 7–22 months).
      MH patients were more likely to receive a perioperative blood transfusion than PSH patients (66 (27.7%) v 24 (10.1%), p < 0.0001). Median length of intensive care stay and hospital stay were significantly higher in MH patients (1.49 vs. 0.68 days; p = 0.02 and 7 vs 6 days p = 0.04 respectively, Table 3).
      There was a significantly greater incidence of peri-operative complications (37.0% vs. 29.0%; p = 0.04), Clavien grade III/IV complications (9.2% vs. 3.8%; p = 0.01) and 90-day mortality (3.8% vs. 0.8%; p = 0.03) in the MH group (Table 3).
      • Dindo A.
      • Demartines N.
      • P-A C.
      Classification of surgical complications.
      Hepatic insufficiency was the only liver related complication which had a significantly different incidence between the two groups (MH 13 (5.5%), PSH 0, p < 0.0001, Table 3).
      A greater number of patients who had PSH underwent repeat hepatectomy due to disease recurrence (redo liver resections), but this was not statistically significant (MH 23 (9.7%), PSH 35 (14.7%), p = 0.09).
      The median long-term follow up for the cohort was 3 years (range, 0.01–12 years). The 1, 3 and 5 year OS in the MH group were 82.3%, 55.6% and 34.9% respectively (Fig. 1). The 1, 3 and 5 year OS in the PSH group were 92.3%, 54.0% and 35.5% respectively (p = 0.56).
      The 1, 3 and 5 year DFS in the MH group were 67.9%, 39.2% and 28.5% respectively (Fig. 1). The 1, 3 and 5 year DFS in the PSH group were 72.2%, 34.8% and 24.8% respectively (p = 0.62).
      The pattern of liver only recurrence, systemic recurrence, or both liver and systemic recurrence were recorded in Table 4. There was a significantly higher incidence of both liver and systemic recurrence in the PSH group (10.5% vs. 5.0%; p = 0.03). However, there was no difference in liver only recurrence or systemic only recurrence. Furthermore, there was no significant difference in recurrence at the resection margin.
      Table 4Pattern of recurrence of disease following major hepatectomy and parenchymal sparing hepatectomy for CLM
      Major hepatectomy (n = 238)Parenchymal sparing hepatectomy (n = 238)p Value
      Total recurrence/n (%)161 (67.7)167 (70.2)0.51
      Liver recurrence/n (%)125 (52.5)112 (47.1)0.25
      Liver recurrence at resection margin/n (%)5 (2.1)7 (2.9)0.42
      Systemic recurrence/n (%)25 (10.5)30 (12.6)0.46
      Liver and systemic recurrence/n (%)12 (5.0)25 (10.5)0.03
      Categorical data reported as percentages and p values from Fishers exact test.
      • Weber S.M.
      • Jarnagin W.R.
      • DeMatteo R.P.
      • et al.
      Survival after resection of multiple hepatic colorectal metastases.
      The multivariable cox regression model of the entire cohort prior to matching (Table 5), found that the main predictors of mortality were involved resection margins (p < 0.001) and increasing numbers of tumours (p = 0.003). In addition to these, increasing age at surgery (p = 0.047) and the use of peri-operative FFP (p = 0.04) were also associated with higher mortality rates. After accounting for all of the factors in the model, the difference between the types of surgery was non-significant (p = 0.343), with a hazard ratio of 0.89 (95% CI = 0.70–1.13) for PSH, relative to MH.
      Table 5Univariable and multivariable analyses of patient survival of the entire cohort before matching
      FactorUnivariableMultivariable
      Hazard ratio

      (95% CI)
      p-ValueHazard ratio

      (95% CI)
      p-Value
      Surgery type (PSH)0.86 (0.96–1.08)0.1920.89 (0.70–1.13)0.343
      Age0.0850.047
       <60
       60–701.12 (0.88–1.42)0.3651.10 (0.86–1.41)0.431
       >701.31 (1.03–1.67)0.0281.36 (1.06–1.76)0.017
      Gender (Female)0.90 (0.74–1.10)0.2890.92 (0.75–1.14)0.452
      Year0.5740.624
       2000–2002
       2003–20051.14 (0.89–1.46)0.3071.19 (0.91–1.55)0.199
       2006–20081.20 (0.91–1.57)0.1911.14 (0.86–1.52)0.356
       2009–20111.24 (0.78–1.97)0.3741.19 (0.74–1.92)0.479
      Periop blood transfusion1.10 (0.89–1.37)0.3750.94 (0.72–1.22)0.651
      Periop FFP1.46 (1.05–2.04)0.0241.56 (1.02–2.38)0.040
      Periop platelets1.25 (0.52–3.02)0.6210.70 (0.27–1.85)0.476
      Days on ITU0.0190.312
       0
       1–51.11 (0.78–1.59)0.5620.98 (0.67–1.43)0.898
       >51.88 (1.21–2.93)0.0051.47 (0.88–2.43)0.140
      Involved resection margins1.78 (1.34–2.31)<0.0011.61 (1.23–2.10)<0.001
      Max tumour size0.0920.078
       <5
       5–101.06 (0.85–1.31)0.6321.04 (0.83–1.30)0.726
       >101.62 (1.05–2.50)0.0291.67 (1.07–2.62)0.024
      Tumour number (>3)1.61 (1.20–2.16)0.0011.61 (1.18–2.19)0.003
      Any complications1.38 (1.12–1.70)0.0031.22 (0.97–1.54)0.096
      Results from Cox regression models, with death as the outcome.
      Bold p-values are significant at p < 0.05.
      PSH = parenchymal sparing hepatectomy.

      Discussion

      To the authors' knowledge, there have been no previous randomised or case-matched analyses regarding the peri-operative and long term outcomes of parenchymal sparing versus major hepatectomy for CLM. Furthermore, this was the largest comparative analysis in the literature to date.
      Peri-operative blood transfusions and morbidity, especially complications requiring intervention, have been shown to independently adversely affect long-term survival in patients following liver resection for CLM.
      • Karanjia N.D.
      • Lordan J.T.
      • Fawcett W.J.
      • et al.
      Survival and recurrence after neo-adjuvant chemotherapy and liver resection for colorectal metastases – a ten year study.
      • Simmonds P.C.
      • Primrose J.N.
      • Colquitt J.L.
      • et al.
      Surgical resection of hepatic metastases from colorectal cancer: a systematic review of the published studies.
      • Sui C.J.
      • Cao L.
      • Li B.
      • et al.
      Anatomical versus non-anatomical resection for colorectal liver metastases: a meta-analysis.
      • Fong Y.
      • Fortner J.
      • RL S.
      Clinical scores for predicting recurrence after hepatic resection for metastatic colorectal cancer; analysis of 1001 consecutive cases.
      • Muratore A.
      • Ribero D.
      • Zimmitti G.
      • et al.
      Resection margin and recurrence-free survival after liver resection of colorectal metastases.
      • Wei A.C.
      • Greig P.D.
      • Grant D.
      • et al.
      Survival after hepatic resection for colorectal metastases: a 10-year experience.
      The significantly fewer complications and blood transfusions in patients who underwent PSH may have reflected technical difficulties in the MH group.
      • Sui C.J.
      • Cao L.
      • Li B.
      • et al.
      Anatomical versus non-anatomical resection for colorectal liver metastases: a meta-analysis.
      This may also explain the significantly lower 90-day mortality seen in the PSH group.
      The incidence of PSH increased throughout the study period. It is possible, therefore, that the improved peri-operative outcome in this group may have reflected the overall increase in the experience of the liver unit over time. Yet, the patients who underwent MH, who were matched, were selected throughout the same study period, which would suggest that experience in the unit would benefit these patients to a similar degree. However, the patients were not necessarily matched according to each year, which may have introduced bias into the study.
      Studies have shown that cancer involved resection margins independently predicts poor outcome following liver resection for CLM.
      • Muratore A.
      • Ribero D.
      • Zimmitti G.
      • et al.
      Resection margin and recurrence-free survival after liver resection of colorectal metastases.
      • Lordan J.T.
      • N.D K.
      ‘Close Shave’ in liver resection for colorectal liver metastases.
      While early studies recommended MH,
      • DeMatteo R.P.
      • Palese C.
      • Jarnagin W.R.
      • et al.
      Anatomic segmental hepatic resection is superior to wedge resection as an oncologic operation for colorectal liver metastases.
      pre- and peri-operative imaging have advanced over the years to allow more accurate parenchymal transection.
      • Lordan J.T.
      • Stenson K.M.
      • ND K.
      The value of intraoperative ultrasound and preoperative imaging, individually and in combination, in liver resection for metastatic colorectal cancer.
      This may explain why there was no difference in involved resection margins between the groups in this study, and suggests that there may be no oncological difference between MH and PSH. However, it was not possible to determine if the recurrence within the liver was a new liver metastasis, or a recurrence at the resection margin due to cancer involved margins.
      The groups were matched on a case-by-case basis in an attempt to achieve highly accurate matching, using a propensity score.
      • Rees M.
      • Tekkis P.
      • Welsh F.
      • et al.
      Evaluation of long-term survival after hepatic resection for metastatic colorectal cancer.
      However, data were not available to study certain previously described risk factors, including ASA, primary tumour lymph node status, underlying liver parenchyma histopathological status and CEA levels.
      • Rees M.
      • Tekkis P.
      • Welsh F.
      • et al.
      Evaluation of long-term survival after hepatic resection for metastatic colorectal cancer.
      • Fong Y.
      • Fortner J.
      • RL S.
      Clinical scores for predicting recurrence after hepatic resection for metastatic colorectal cancer; analysis of 1001 consecutive cases.
      Further potential bias may has been introduced due to the less than perfect matching according to age and Dukes' stage C, despite matching all the variables as closely as possible between groups. In the assessment of all proposed clinical risk scores the Fong score was shown to be particularly unhelpful in stratifying DFS and DSS amongst patients undergoing CRLM resection with an actual follow up of 10 years minimum.
      • Roberts K.J.
      • White A.
      • Cockbain A.
      • et al.
      Performance of prognostic scores in predicting long-term outcome following resection of colorectal liver metastases.
      Thus comparing patients by clinical risk score is probably of limited benefit.
      The literature has reported evidence that neo-adjuvant chemotherapy followed by liver resection may confer significant long term benefit for patients with CLM.
      • Karanjia N.D.
      • Lordan J.T.
      • Fawcett W.J.
      • et al.
      Survival and recurrence after neo-adjuvant chemotherapy and liver resection for colorectal metastases – a ten year study.
      • Nordlinger B.
      • Sorbye H.
      • Glimelius B.
      • et al.
      Perioperative chemotherapy with FOLFOX4 and surgery versus surgery alone for resectable liver metastases from colorectal cancer (EORTC Intergroup trial 40983): a randomised controlled trial.
      • Karanjia N.D.
      • Lordan J.T.
      • Quiney N.
      • et al.
      A comparison of right and extended right hepatectomy with all other hepatic resections for colorectal liver metastases: a ten-year study.
      However, the hepato-toxic effects of chemotherapeutic drugs have been shown to be cumulative.
      • Vauthey J.N.
      • Pawlik T.M.
      • Ribero D.
      • et al.
      Chemotherapy regimen predicts steatohepatitis and an increase in 90-day mortality after surgery for hepatic colorectal metastases.
      As hepatopancreatobiliary (HPB) units increasingly treat patients with neo-adjuvant chemotherapy, the reported associated increased peri-operative morbidity and 90-day mortality,
      • Vauthey J.N.
      • Pawlik T.M.
      • Ribero D.
      • et al.
      Chemotherapy regimen predicts steatohepatitis and an increase in 90-day mortality after surgery for hepatic colorectal metastases.
      and in particular hepatic insufficiency,
      • Fisher S.B.
      • Kneuertz P.J.
      • RM D
      A comparison of right posterior sectorectomy with formal right hepatectomy: a dual-institution study.
      may put further pressure on HPB surgeons to perform PSH. The data from this study suggest that while this may be appropriate from a peri-operative perspective, HPB surgeons and patients must be aware of the potentially increased risk of liver recurrence following PSH compared with MH.
      Although these data suggest that PSH is safer during the peri-operative period than MH, it is important to note that this study was conducted over a 10 year period whereby there was a shift in tendency towards performing PSH. As a result of the non-contemporaneous data, bias may has been introduced. This may account for the greater requirement of peri-operative transfusion in the MH group from the earlier period of the study.
      Furthermore, as liver surgery becomes safer in the peri-operative period, redo liver resections are becoming more feasible. Indeed, this study did not show a difference in patients undergoing redo resections in both groups. Redo liver resections for CLM have been shown to be safe and to confer long term benefit.
      • de Jong M.C.
      • Mayo S.C.
      • Pulitano C.
      • et al.
      Repeat curative intent liver surgery is safe and effective for recurrent colorectal liver metastasis: results from an international multi-institutional analysis.
      Also, the similar long term survival rates seen in the current study may have reflected the advances in adjuvant chemotherapy, in particular oxaliplatin, irinotecan and monoclonal antibody based regimens.
      • Giacchetti S.
      • Perpoint B.
      • R Z
      Phase III multicentre randomised trial of oxaliplatin added tp chronomodulated fluorouracil-leucovorin as first-line treatment of metastatic colorectal cancer.
      • Giantonio B.J.
      • Levy D.E.
      • O'dwyer P.J.
      • et al.
      A phase II study of high-dose bevacizumab in combination with irinotecan, 5-fluorouracil, leucovorin, as initial therapy for advanced colorectal cancer: results from the Eastern Cooperative Oncology Group study E2200.
      • Cunningham D.
      • Humblet Y.
      • Siena S.
      • et al.
      Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer.
      • Goldberg R.M.
      • Sargent D.J.
      • RF M
      A randomised controlled trial of fluorouracil plus leucovorin, irinotecan, oxaliplatin combinations in patients with previously untreated metastatic colorectal cancer.
      For PSH, it would be advisable to have a low threshold for consideration of redo surgery for liver recurrence. As such, with a more aggressive policy to perform redo liver resections, further studies may show a significantly greater long term OS in patients who undergo PSH versus MH for CLM.
      In conclusion, for patients with CLM, oncologic clearance can be achieved by parenchymal-sparing/minor hepatectomy in selected patients with colorectal liver metastases with significantly lower peri-operative morbidity and 90-day mortality compared to major hepatectomy. We believe parenchymal-sparing/minor resection should be considered if technically feasible, even for patients with multifocal disease, regardless of treatment with or without chemotherapy.

      Conflicts of interest

      None to declare.

      References

        • Karanjia N.D.
        • Lordan J.T.
        • Fawcett W.J.
        • et al.
        Survival and recurrence after neo-adjuvant chemotherapy and liver resection for colorectal metastases – a ten year study.
        Eur J Surg Oncol. 2009; 35: 838-843
        • Simmonds P.C.
        • Primrose J.N.
        • Colquitt J.L.
        • et al.
        Surgical resection of hepatic metastases from colorectal cancer: a systematic review of the published studies.
        Br J Cancer. 2006; 94: 982-999
        • Nordlinger B.
        • Sorbye H.
        • Glimelius B.
        • et al.
        Perioperative chemotherapy with FOLFOX4 and surgery versus surgery alone for resectable liver metastases from colorectal cancer (EORTC Intergroup trial 40983): a randomised controlled trial.
        Lancet. 2008; 371: 1007-1016
        • Ferrero A.
        • Vigan L.
        • Lo Tesoriere R.
        • et al.
        Bisegmentectomies as alternative to right hepatectomy in the treatment of colorectal liver metastases.
        Hepatogastroenterology. 2009; 56: 1429-1435
        • Sui C.J.
        • Cao L.
        • Li B.
        • et al.
        Anatomical versus non-anatomical resection for colorectal liver metastases: a meta-analysis.
        Int J Colorectal Dis. 2012; 27: 939-946
        • DeMatteo R.P.
        • Palese C.
        • Jarnagin W.R.
        • et al.
        Anatomic segmental hepatic resection is superior to wedge resection as an oncologic operation for colorectal liver metastases.
        J Gastrointest Surg. 2000; 4: 178-184
        • Kokudo N.
        • Tada K.
        • Seki M.
        • et al.
        Anatomical major resection versus nonanatomical limited resection for liver metastases from colorectal carcinoma.
        Am J Surg. 2001; 181: 153-159
        • Zorzi D.
        • Mullen J.T.
        • Abdalla E.K.
        • et al.
        Comparison between hepatic wedge resection and anatomic resection for colorectal liver metastases.
        J Gastrointest Surg. 2006; 10: 86-94
        • Lordan J.T.
        • Karanjia N.D.
        • Quiney N.
        • et al.
        A 10 year study of outcome following hepatic resection for colorectal liver metastases - the effect of evaluation in a multi-disciplinary team setting.
        Eur J Surg Oncol. 2009; 35: 302-306
        • Mise Y.
        • Aloia T.A.
        • Brudvik K.W.
        • et al.
        Parenchymal-sparing hepatectomy in colorectal liver metastasis improves salvageability and survival.
        Ann Surg. 2016; 263: 146-152
        • Weber S.M.
        • Jarnagin W.R.
        • DeMatteo R.P.
        • et al.
        Survival after resection of multiple hepatic colorectal metastases.
        Ann Surg Oncol. 2000; 7: 643-650
      1. MRC Good clinical Practice in clinical Trials. 1998
        • Rees M.
        • Tekkis P.
        • Welsh F.
        • et al.
        Evaluation of long-term survival after hepatic resection for metastatic colorectal cancer.
        Ann Surg. 2008; 247: 125-135
        • Aurtin P.
        Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples.
        Stat Med. 2009; 28: 3083-3107
        • Fukushima K.
        • Fukumoto T.
        • Kuramitsu K.
        • et al.
        Assessment of ISGLS definition of posthepatectomy liver failure and its effect on outcome in patients with hepatocellular carcinoma.
        J Gastrointest Surg. 2014; : 18
        • Dindo A.
        • Demartines N.
        • P-A C.
        Classification of surgical complications.
        Ann Surg Oncol. 2004; 240: 205-213
        • Fong Y.
        • Fortner J.
        • RL S.
        Clinical scores for predicting recurrence after hepatic resection for metastatic colorectal cancer; analysis of 1001 consecutive cases.
        Ann Surg Oncol. 1999; 230: 309-318
        • Muratore A.
        • Ribero D.
        • Zimmitti G.
        • et al.
        Resection margin and recurrence-free survival after liver resection of colorectal metastases.
        Ann Surg Oncol. 2010; 17: 1324-1329
        • Wei A.C.
        • Greig P.D.
        • Grant D.
        • et al.
        Survival after hepatic resection for colorectal metastases: a 10-year experience.
        Ann Surg Oncol. 2006; 13: 668-676
        • Lordan J.T.
        • N.D K.
        ‘Close Shave’ in liver resection for colorectal liver metastases.
        Eur J Surg Oncol. 2010; 36: 47-51
        • Lordan J.T.
        • Stenson K.M.
        • ND K.
        The value of intraoperative ultrasound and preoperative imaging, individually and in combination, in liver resection for metastatic colorectal cancer.
        Ann R Coll Surg Engl. 2011; 93: 246-249
        • Roberts K.J.
        • White A.
        • Cockbain A.
        • et al.
        Performance of prognostic scores in predicting long-term outcome following resection of colorectal liver metastases.
        BJS. 2014; 101: 856-866
        • Karanjia N.D.
        • Lordan J.T.
        • Quiney N.
        • et al.
        A comparison of right and extended right hepatectomy with all other hepatic resections for colorectal liver metastases: a ten-year study.
        Eur J Surg Oncol. 2009; 35: 65-70
        • Vauthey J.N.
        • Pawlik T.M.
        • Ribero D.
        • et al.
        Chemotherapy regimen predicts steatohepatitis and an increase in 90-day mortality after surgery for hepatic colorectal metastases.
        J Clin Oncol. 2006; 24: 2065-2072
        • Fisher S.B.
        • Kneuertz P.J.
        • RM D
        A comparison of right posterior sectorectomy with formal right hepatectomy: a dual-institution study.
        HPB. 2013; 15: 753-762
        • de Jong M.C.
        • Mayo S.C.
        • Pulitano C.
        • et al.
        Repeat curative intent liver surgery is safe and effective for recurrent colorectal liver metastasis: results from an international multi-institutional analysis.
        J Gastrointest Surg. 2009; 13: 2141-2151
        • Giacchetti S.
        • Perpoint B.
        • R Z
        Phase III multicentre randomised trial of oxaliplatin added tp chronomodulated fluorouracil-leucovorin as first-line treatment of metastatic colorectal cancer.
        J Clin Oncol. 2000; 18: 136-147
        • Giantonio B.J.
        • Levy D.E.
        • O'dwyer P.J.
        • et al.
        A phase II study of high-dose bevacizumab in combination with irinotecan, 5-fluorouracil, leucovorin, as initial therapy for advanced colorectal cancer: results from the Eastern Cooperative Oncology Group study E2200.
        Ann Oncol. 2006; 17 ([Clinical Trial, Phase II. Journal Article. Multicenter Study. Research Support, N.I.H., Extramural]): 1399-1403
        • Cunningham D.
        • Humblet Y.
        • Siena S.
        • et al.
        Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer.
        N Engl J Med. 2004; 35: 337-345
        • Goldberg R.M.
        • Sargent D.J.
        • RF M
        A randomised controlled trial of fluorouracil plus leucovorin, irinotecan, oxaliplatin combinations in patients with previously untreated metastatic colorectal cancer.
        J Clin Oncol. 2004; 22: 23-30