Abstract
Background Intrahepatic pedicle ligation (IPL) is an alternative to extrahepatic portal dissection (EPD). Although IPL has been well described, concern has arisen over a possible association with increased complication rates.
Methods Patients who underwent hemi-hepatectomy during January 1995 to December 2010 were reviewed and the inflow control technique (IPL versus EPD) documented. Patient, tumour, treatment and outcome variables were compared.
Results A total of 798 patients underwent hemi-hepatectomy, 568 (71.2%) of the right and 230 (28.8%) of the left liver. In univariate analysis, factors associated with the choice of IPL included surgeon, right hepatectomy, preoperative portal vein embolization, diagnosis of colorectal cancer liver metastasis, and smaller tumour size (P < 0.011). In multivariate analysis, right hepatectomy [versus left: hazard ratio (HR) 3.878, 95% confidence interval (CI) 1.15–13.14; P = 0.029] and smaller tumour size (median of 4.5 cm versus 5.5 cm: HR 0.72, 95% CI 0.59–0.88; P = 0.002) were associated with IPL. Pringle manoeuvre time was longer in IPL procedures (40 min versus 29 min; P < 0.001). Complication rates (49.8% in IPL versus 48.4% in EPD; P = 0.706) were similar in both groups, as was the severity of complications; 17.6% of EPD and 22.3% of IPL patients experienced complications of grade ≥3 (P = 0.225).
Conclusions Patients with small tumours undergoing right hepatectomy were more likely to undergo IPL. In selected patients, IPL was not associated with an increased complication rate and thus it should be considered a safe approach.
Introduction
During hemi-hepatectomy, vascular inflow control is a critical step for which several different techniques have been described.1 In the 1950s, the technique of extrahepatic portal dissection (EPD), prior to parenchymal division, became common practice.2 This consists of the dissection and ligation of the ipsilateral hepatic artery and portal vein within the hilus of the liver. This extrahepatic technique is still commonly employed today, with division of the inflow being performed with stapling devices or suture ligation. Control of the bile duct can be performed either in the hilus at the time of vascular ligation, or within the liver during transection of the hepatic parenchyma.3,4 Some authors have noted, however, that this extrahepatic method of inflow control is time-consuming and may result in inadvertent injury to aberrant or contralateral vascular or biliary structures.5,6
The technique of intrahepatic vascular inflow control was first reported by Couinaud,1 and Launois and Jamieson.7,8 It is based on the fact that the structures of the portal triad invaginate Glisson's capsule and enter the liver parenchyma together as a pedicle. Thus, within the liver all three structures of the porta are contained within a well-formed sheath (pedicle), which can be isolated and divided en masse. Herein, this technique is referred to as intrahepatic pedicle ligation (IPL). Outside the liver, no well-characterized sheath exists and therefore the structures are isolated and divided individually. The technique of IPL can be performed using either of two different approaches.7,8 In the posterior approach, hepatotomies are created anterior and posterior to the hilus on the side of resection. An instrument or finger is then used to isolate the pedicle. In the anterior approach, the hepatic parenchyma is initially divided from anterior to posterior until the pedicle is identified. The pedicle is then isolated within the liver. Although the technique has been well described, concern has been raised over the possibility of increased complication rates with IPL, particularly of bile leaks.9
The aims of this study were to analyse factors associated with the selection of either inflow control technique and to evaluate the differences in postoperative outcomes between patients undergoing IPL and EPD, respectively, in a high-volume tertiary centre.
Materials and methods
A retrospective search of the prospectively maintained hepatopancreatobiliary database at Memorial Sloan–Kettering Cancer Center (MSKCC) was carried out to identify patients who underwent hemi-hepatectomy between January 1995 and December 2010. Permission to study these patients was granted by the MSKCC Institutional Review Board according to the institution's policy for protected health information. Patients who were treated for bile duct cancer or bile duct stricture, and those who underwent a bile duct resection for any reason were excluded. Patients who had undergone a formal trisegmentectomy were also excluded. This search strategy identified 848 patients who had undergone a hemi-hepatectomy. A review of operative notes was performed to determine the technique used for vascular inflow ligation; in 826 of the 848 patients (97.4%) the specific technique used for inflow control could be determined. Of these 826 patients, 28 were excluded because a mixed approach (IPL combined with EPD) to inflow control had been taken. This left 798 patients in the study population. To assess changes over time, the cohort was split according to date of operation into two groups defined by 8-year time periods (1995–2002 and 2003–2010).
The technique used for vascular inflow control was categorized as either IPL or EPD. Patients operated on with EPD control underwent EPD and ligation of the ipsilateral arterial and portal venous branches within the hilus of the liver, with division of the bile duct either in the hilus or within the liver parenchyma. In general, the latter strategy was the preferred approach. Ligation of the vessels was performed either with suture ligation or using a vascular stapler. The technique for IPL has been previously described in detail.7,8,10 The intrahepatic isolation of the portal pedicle was achieved through either the anterior or posterior approach as described herein. Pedicle transection was performed with a vascular stapler or by clamping and suturing, or with a combination of both.
Patient- and tumour-related variables were analysed to identify selection factors associated with the use of IPL and EPD, respectively. Patient-related factors analysed included age, gender, coexistent comorbidities (presence of any cardiac disease, diabetes or chronic obstructive pulmonary disease, and each individually), and the pathologic status of the hepatic parenchyma. Tumour-related factors included the location, number, size and preoperative diagnosis of tumours. Treatment-related factors included the side of resection, preoperative portal vein embolization (PVE), and whether additional hepatic or extrahepatic procedures were performed. Outcome variables included estimated blood loss (EBL), Pringle manoeuvre time, operative time, margin status, complications, grade of complications, length of hospital stay (LoS), intensive care unit (ICU) admission, hospital readmission, blood product transfusion rate, and the rate of fluid and/or bile collections requiring drainage. The margin was defined as negative (R0) when tumour was histologically found to be ≥1 mm from the cut surface. The presence of tumour cells within 1 mm of the cut surface was considered to indicate a positive margin.
Postoperative complications were graded on a scale of 1–5 according to a previously published and validated grading system as follows: grade 0, no complication; grade 1, complications requiring the administration of oral antibiotics, bowel rest, basic monitoring or supportive care; grade 2, complications requiring the administration of i.v. antibiotics, total parenteral nutrition, transfusions, arrhythmia treated with i.v. medication, chest tube insertion; grade 3, complications requiring interventional radiology drainage, operative drainage, ICU admission, intubation, pacemaker placement, bronchoscopy; grade 4, complications resulting in chronic disability, organ resection, enteral diversion, and grade 5, complications resulting in death.11
Univariate associations between selected factors and inflow control approach were assessed using chi-square tests for categorical covariates and Wilcoxon rank-sum tests for continuous covariates, except for preoperative laboratory data, which were compared using t-tests. The Cochrane–Armitage trend test was used to compare the number of units of blood transfused between EPD and IPL patients using the following categories: 0 units; 1–3 units; 4–6 units, and >6 units. In addition to the comparison of complication rates between the two groups (chi-squared test), differences in the severity of complications were compared (on an ordinal scale with six levels: no complications, grades 1–5) using the Cochrane–Armitage trend test.
Factors that were significant on univariate analysis were put into a multivariate logistic regression model with a random surgeon effect, and backward elimination was used to obtain the final model. This was carried out to avoid the bias of single-surgeon personal preferences in the multivariate model. For the multivariate model, largest tumour size was log-transformed. All statistical tests were two-sided and P-values of < 0.05 were considered to indicate statistical significance. Statistical analyses were performed in sas Version 9.2 (SAS Institute, Inc., Cary, NC, USA) and R Version 2.11.1 (R Foundation for Statistical Computing, Vienna, Austria). The percentage of surgeries performed over time using each litigation technique (EPD with intrahepatic control of the bile duct, EPD with extrahepatic control of the bile duct, anterior IPL, posterior IPL) was graphed using Excel 2007.
Results
Factors associated with inflow control approach
Table 1 details the factors associated with the type of inflow control. In univariate analysis, the use of IPL was significantly associated with right hepatectomy, individual surgeon, a preoperative diagnosis of metastatic colorectal cancer, PVE and small tumour size. In multivariate analysis, the only two factors independently associated with the use of IPL were right hepatectomy [hazard ratio (HR) 3.88, 95% confidence interval (CI) 1.15–13.14; P = 0.029] and smaller tumour size (5.5 cm in EPD versus 4.5 cm in IPL) (for a 1-cm decrease in tumour size: HR 1.39, 95% CI 1.14–1.69; P = 0.002).
Table 1.
Factors associated with the choice of approach to inflow control
| EPD | IPL | Population | P-value | |
|---|---|---|---|---|
| (n = 318) | (n = 480) | (n = 798) | ||
| Side, n (%) | <0.001 | |||
| Right | 173 (30.5%) | 395 (69.5%) | 568 (100%) | |
| Left | 145 (63.0%) | 85 (37.0%) | 230 (100%) | |
| Surgeon, n (%) | <0.001 | |||
| A | 76 (23.9%) | 38 (7.9%) | 114 (14.3%) | |
| B | 44 (13.8%) | 246 (51.3%) | 290 (36.3%) | |
| C | 87 (27.4%) | 91 (19.0%) | 178 (22.3%) | |
| D | 44 (13.8%) | 58 (12.1%) | 102 (12.8%) | |
| E | 45 (14.2%) | 5 (1.0%) | 50 (6.3%) | |
| Others | 22 (6.9%) | 42 (8.8%) | 64 (8.0%) | |
| Preoperative PVE, n (%) | 15 (4.7%) | 69 (14.4%) | 84 (10.5%) | <0.001 |
| Repeat surgery, n (%) | 22 (6.9%) | 42 (8.8%) | 64 (8.0%) | 0.351 |
| Other liver procedure, n (%) | 103 (32.4%) | 152 (31.7%) | 255 (32.0%) | 0.830 |
| Other procedure, n (%) | 161 (50.6%) | 223 (46.5%) | 384 (48.1%) | 0.248 |
| Variant anatomy, n (%) | 64 (20.1%) | 84 (17.5%) | 148 (18.5%) | 0.350 |
| Pump placement, n (%) | 50 (15.7%) | 84 (17.5%) | 134 (16.8%) | 0.511 |
| Diagnosis, n (%) | 0.011 | |||
| Colorectal metastasis | 183 (57.5%) | 338 (70.4%) | 521 (65.3%) | |
| Other metastasis | 41 (12.9%) | 50 (10.4%) | 91 (11.4%) | |
| Hepatocellular carcinoma | 46 (14.5%) | 47 (9.8%) | 93 (11.7%) | |
| Cholangiocarcinoma/gallbladder cancer | 21 (6.6%) | 18 (3.8%) | 39 (4.9%) | |
| Haemangioma | 5 (1.6%) | 8 (1.7%) | 13 (1.6%) | |
| Benign liver lesions | 11 (3.5%) | 7 (1.5%) | 18 (2.3%) | |
| Other | 11 (3.5%) | 12 (2.5%) | 23 (2.9%) | |
| Number of nodules, n (%) | 0.004 | |||
| <3 | 237 (74.5%) | 311 (64.8%) | 548 (68.7%) | |
| ≥3 | 80 (25.2%) | 167 (34.8%) | 247 (31.0%) | |
| Size of largest lesion, cm, median (range) | 5.5 (0.0–34.5) | 4.5 (0.0−32.0) | 5.5 (0.0–34.5) | <0.001 |
| Pathologic liver parenchyma, n (%) | 180 (56.6%) | 249 (51.9%) | 429 (53.8%) | 0.190 |
| Inflammationa, n (%) | 71 (22.3%) | 97 (20.2%) | 168 (21.1%) | |
| Steatosisa, n (%) | 93 (29.2%) | 129 (26.9%) | 222 (27.8%) | |
| Fibrosisa, n (%) | 26 (8.2%) | 35 (7.3%) | 61 (7.6%) | |
| Cirrhosisa, n (%) | 11 (3.5%) | 8 (1.7%) | 19 (2.4%) | |
| Cholestasisa, n (%) | 7 (2.2%) | 14 (2.9%) | 21 (2.6%) |
Rates presented for selected types of pathologic liver parenchyma. P-values not provided.
EPD, extrahepatic portal dissection; IPL, intrahepatic pedicle ligation; PVE, portal vein embolization.
Perioperative outcome
Demographics, and intraoperative and postoperative results in the 798 patients who underwent EPD or IPL are detailed in Table 2. Of the intraoperative outcomes, the only factor that significantly differed between the two groups was the median Pringle manoeuvre time (40 min in IPL versus 29 min in EPD; P < 0.001). The use of IPL was significantly associated with a higher rate of transfusion (P = 0.002) although EBL (P = 0.684) and red blood cell transfusions (P = 0.189) were similar between the groups. This result probably reflects a significantly higher rate of fresh frozen plasma transfusions associated with IPL (39.2% in IPL versus 25.8% in EPD; P < 0.001). The rate of postoperative complications was similar between the IPL and EPD groups (P = 0.706). The overall 90-day mortality rate was 2.3% and did not vary significantly between the groups.
Table 2.
Perioperative factors associated with inflow control technique
| EPD | IPL | Population | P-value | |
|---|---|---|---|---|
| (n = 318) | (n = 480) | (n = 798) | ||
| Age, years, median (range) | 61 (0–84) | 61 (1–89) | 61 (0–89) | 0.660 |
| Gender, ratio, female : male | 0.93 | 0.86 | 0.89 | 0.606 |
| Comorbidities, n (%) | 84 (26.4%) | 114 (23.8%) | 198 (24.8%) | 0.353 |
| Cardiac, n (%) | 43 (13.5%) | 62 (12.9%) | 105 (13.2%) | 0.804 |
| Diabetes mellitus, n (%) | 39 (12.3%) | 53 (11.0%) | 92 (11.5%) | 0.598 |
| COPD, n (%) | 17 (5.3%) | 25 (5.2%) | 42 (5.3%) | 0.932 |
| Preoperative laboratory data | ||||
| Total bilirubin, mg/dl, mean ± SD | 0.70 ± 1.15 | 0.65 ± 0.42 | 0.67 ± 0.80 | 0.724 |
| Albumin, g/dl, mean ± SD | 4.17 ± 0.44 | 4.15 ± 0.44 | 4.15 ± 0.44 | 0.298 |
| Creatinine, mg/dl, mean ± SD | 1.00 ± 0.29 | 1.03 ± 0.33 | 1.02 ± 0.31 | 0.423 |
| Pringle time, min, median (range) | 29 (0–77) | 40 (0–103) | 35 (0–103) | <0.001 |
| Operating time, min, median (range) | 245 (110–587) | 248 (70–967) | 247 (70–967) | 0.795 |
| EBL, ml, median (range) | 500 (50–5000) | 500 (50–5600) | 500 (50–5600) | 0.684 |
| Margin, n (%) | 24 (7.5%) | 42 (8.8%) | 66 (8.3%) | 0.563 |
| Transfusion, n (%) | 127 (39.9%) | 244 (50.8%) | 371 (46.5%) | 0.002 |
| Red blood cell transfusion, n (%) | 89 (28.0%) | 162 (33.8%) | 251 (31.5%) | 0.189a |
| LoS, days, median (range) | 8 (4–51) | 8 (4–69) | 8 (0–69) | 0.642 |
| Complications, n (%) | 154 (48.4%) | 239 (49.8%) | 393 (49.2%) | 0.706 |
| 0.225b | ||||
| Grade 1 | 49 (15.4%) | 67 (14.0%) | 116 (14.5%) | |
| Grade 2 | 49 (15.4%) | 65 (13.5%) | 114 (14.3%) | |
| Grade 3 | 50 (15.7%) | 90 (18.8%) | 140 (17.5%) | |
| Grade 4 | 1 (0.3%) | 4 (0.8%) | 5 (0.6%) | |
| Grade 5 (90-day mortality) | 5 (1.6%) | 13 (2.7%) | 18 (2.3%) | 0.423 |
| Interventional radiology drainage, n (%) | 38 (12.0%) | 66 (13.8%) | 104 (13.0%) | 0.460 |
| Postoperative acute biliary complications | 16 (5.0%) | 27 (5.6%) | 43 (5.4%) | 0.716 |
P-value for trend.
EPD, extrahepatic portal dissection; IPL, intrahepatic pedicle ligation; COPD, chronic obstructive pulmonary disease; SD, standard deviation; EBL, estimated blood loss; LoS, length of hospital stay.
Comparison of the approach to the bile duct during EPD and to the pedicle during IPL
A further analysis of the different techniques utilized to control the bile duct during EPD (intra- or extrahepatic) and the different approaches to the pedicle (posterior or anterior) during IPL is shown in Table 3. Side of resection was significantly associated with the choice of approach in both groups. Pringle manoeuvre time was longer in operations utilizing IPL compared with those using EPD (see above); within IPL operations, it was significantly longer in the posterior approach compared with the anterior. Moreover, the IPL anterior approach was significantly associated with a higher EBL compared with the posterior approach. There was no difference in operating time or complications according to approach within either inflow technique group.
Table 3.
Comparison of approaches to the bile duct (BD) during extrahepatic portal dissection (EPD) and to the pedicle during intrahepatic pedicle ligation (IPL)
| Extrahepatic BD extra | Extrahepatic BD intra | EPD | Pedicle anterior | Pedicle posterior | IPL | |
|---|---|---|---|---|---|---|
| (n = 64) | (n = 254) | P-value | (n = 198) | (n = 282) | P-value | |
| Side, n (%) | 0.013 | 0.028 | ||||
| Right | 26 (40.6%) | 147 (57.9%) | 172 (86.9%) | 223 (79.1%) | ||
| Left | 38 (59.4%) | 107 (42.1%) | 26 (13.1%) | 59 (20.9%) | ||
| Pringle manoeuvre time, min median (range) | 25 (0–74) | 30 (0–77) | 0.485 | 35 (0–83) | 40 (0–103) | <0.001 |
| Operating time, min, median (range) | 242 (125–577) | 246 (110–587) | 0.713 | 246 (70–967) | 250 (106–680) | 0.385 |
| EBL, ml, median (range) | 500 (100–4000) | 500 (50−5000) | 0.711 | 600 (50–5600) | 500 (100−5000) | 0.024 |
| Complications, n (%) | 34 (53.1%) | 120 (47.2%) | 0.400 | 102 (51.5%) | 137 (48.6%) | 0.527 |
| Largest lesion size, cm, median (range) | 6.2 (1–23) | 5.5 (0–34.5) | 0.280 | 5.0 (0−20) | 4.2 (0−20) | 0.226 |
Extrahepatic BD extra, extrahepatic dissection with extrahepatic control of the bile duct; Extrahepatic BD intra, extrahepatic dissection with intrahepatic control of the bile duct; EBL, estimated blood loss.
Right hepatectomy
A subanalysis of the 568 patients who underwent right hepatectomy was conducted. Extrahepatic portal dissection and IPL were used in 30.5% (n = 173) and 69.5% (n = 395) of the operations, respectively. In univariate analysis, individual surgeon (P < 0.001), a diagnosis of metastatic colorectal cancer (P = 0.002), preoperative PVE (16.2% in the IPL group versus 6.4% in the EPD group; P < 0.001), and smaller tumour size (4.1 cm in the IPL group versus 6.0 cm in the EPD group; P < 0.001) were associated with IPL. The IPL technique was also associated with a longer Pringle manoeuvre time (median: 40 min versus 30 min; P < 0.001) and a shorter operative time (median: 250 min versus 267 min; P = 0.020). There was no difference in EBL, margin positivity or complication rates between the two techniques.
Left hepatectomy
In the 230 patients who underwent left hepatectomy, IPL and EPD were used in 37.0% (n = 85) and 63.0% (n = 145) of patients, respectively. In univariate analysis, individual surgeon (P < 0.001) and longer Pringle manoeuvre time (median: 42 min in IPL versus 25 min in EPD; P < 0.001) were the only factors associated with IPL. There was no difference in operating room time, EBL, margin positivity or complication rates between the two techniques in this subset. Of note, unlike the other surgical subset, tumour size in this group was not associated with the technique used for inflow control.
Changes over time
The evolution over time of patient characteristics, techniques and operative outcomes during the 15-year study period is depicted in Table 4. Factors significantly associated with the more recent period were an increase in left hepatectomies (P = 0.002), a decrease in EBL (P < 0.001), a decrease in blood transfusion requirement (P = 0.009), and a shorter LoS (P = 0.017). A higher rate of complications was seen (P = 0.014) in the later period. More complex surgical procedures were also associated with the second timeframe: more patients underwent repeated liver surgery (P < 0.001) or simultaneous liver resections (P < 0.001). Furthermore, in the later period, a significant change was seen in surgical technique (P < 0.001), mostly accounted for by an increase in the use of the anterior approach in IPL (Fig. 1).
Table 4.
Evolution over time and population of the study
| 1995–2002 | 2003–2010 | P-value | |
|---|---|---|---|
| (n = 389) | (n = 409) | ||
| Age, years, median (range) | 62 (0–86) | 59 (5−89) | 0.368 |
| Gender, ratio, female : male | 1.00 | 0.77 | 0.061 |
| Comorbidities, n (%) | 90 (23.1%) | 108 (26.4%) | 0.285 |
| Side, right/left, n (%) | 297/92 (76.3/23.7%) | 271/138 (66.3/33.7%) | 0.002 |
| Pringle time, min, median (range) | 32 (0–103) | 35 (0–80) | 0.620 |
| Operating time, min, median (range) | 247 (110–680) | 247 (70–967) | 0.647 |
| EBL, ml, median (range) | 600 (50–5000) | 575 (50–5600) | <0.001 |
| Margin, n (%) | 25 (6.4%) | 41 (10.0%) | 0.065 |
| Blood transfusion, n (%) | 141 (36.2%) | 113 (27.6%) | 0.009a |
| LoS, days, median (range) | 8 (4–51) | 8 (4–69) | 0.017 |
| Complications, n (%) | 180 (46.3%) | 213 (52.1%) | 0.014 |
| Grade 1 | 58 (14.9%) | 58 (14.2%) | 0.061a |
| Grade 2 | 52 (13.4%) | 62 (15.2%) | |
| Grade 3 | 60 (15.4%) | 80 (19.6%) | |
| Grade 4 | 1 (0.3%) | 0 | |
| Grade 5 | 9 (2.3%) | 9 (2.2%) | |
| Interventional radiology drainage, n (%) | 44 (11.3%) | 60 (14.7%) | 0.159 |
| Postoperative acute biliary complications, n (%) | 21 (5.4%) | 22 (5.4%) | 0.990 |
| Preoperative PVE, n (%) | 10 (2.6%) | 74 (18.1%) | <0.001 |
| Repeat surgery, n (%) | 13 (3.3%) | 51 (12.5%) | <0.001 |
| Other liver procedure, n (%) | 102 (26.2%) | 153 (37.4%) | <0.001 |
| Ligation technique, n (%) | <0.001 | ||
| Extrahepatic BD extra | 41 (10.5%) | 23 (5.6%) | |
| Extrahepatic BD intra | 123 (31.6%) | 131 (32%) | |
| Pedicle anterior | 56 (14.4%) | 142 (34.7%) | |
| Pedicle posterior | 169 (43.4%) | 113 (27.6%) |
P-value for trend.
EBL, estimated blood loss; LoS, length of hospital stay; PVE, portal vein embolization; Extrahepatic BD extra, extrahepatic dissection with extrahepatic control of the bile duct; Extrahepatic BD intra, extrahepatic dissection with intrahepatic control of the bile duct.
Figure 1.
Changes in choice of surgical technique over time (percentages of resections performed using each technique, respectively, per year). Extra BD extra, extrahepatic dissection with extrahepatic control of the bile duct; extra BD intra, extrahepatic dissection with intrahepatic control of the bile duct
Discussion
Over the past two decades, there have been significant reductions in morbidity and mortality following hepatic resection, with major centres now reporting mortality rates of 3–5%.12,13,14 Although the reasons for these improvements in outcome are multifactorial, advances in both the understanding of hepatic anatomy and operative technique are thought to have played pivotal roles. One technical aspect of hepatic resection that has undergone significant evolution is vascular inflow control. This evolution has been secondary to advances in technique that are direct results of the improved understanding of hepatic anatomy.
The concept of vascular inflow control with digital compression was initially reported by J. Hogarth Pringle in 1908.15 In the 1950s, the use of extrahepatic arterial and portal venous ligation prior to parenchymal transection became common practice.2 Control of the bile duct using this technique can be performed either outside the liver or within the liver parenchyma. The technique of IPL is an alternative to EPD that was first reported in the 1980s.16 Since this description, many surgeons have felt that the en masse transection of the pedicle for inflow control would decrease the time required for dissection and decrease the risk for injury to contralateral biliary or vascular structures,17 although an increase in bile leaks during left hepatectomy using IPL has been reported.9
The first large series of patients to undergo IPL during hemi-hepatectomy was reported by Launois and Jamieson in 1992.8 In this series, 70 patients underwent hepatic resection with vascular inflow control achieved using IPL. The authors concluded that IPL was a safe technique that resulted in decreased operative blood loss.8 The first direct comparison between EPD and IPL was reported from Japan by Nakai et al. in 1999.9 In this study of 90 patients, the only difference shown between the two techniques was a higher rate of postoperative bile leak in the IPL group (23% versus 7%; P = 0.03), with most of the leaks occurring in patients who had undergone left hepatectomy.9 A subsequent prospective randomized trial did not show any differences in safety or effectiveness between the two techniques. Specifically, there was no difference in postoperative morbidity or bile leakage. The trial was, however, constrained by small numbers of patients and limited statistical power.18
The present study of 798 patients undergoing right or left hepatectomy was conducted to evaluate potential differences in postoperative outcomes between patients undergoing IPL and EPD, respectively, in a high-volume tertiary centre, to verify the safety of IPL and to identify factors associated with the selection of technique at one institution.
Among patients undergoing hemi-hepatectomy with IPL, the overall complication rate was 49.8%. Sixty-six (13.8%) patients developed a postoperative fluid collection that required drainage and 27 (5.7%) patients developed a postoperative biloma requiring drainage. These results were similar to those in patients undergoing EPD and are comparable with complication rates reported in the literature.12,19,20,21 Operative outcome variables, such as occurrence of major complications, operative time and EBL, were also similar between the groups. Patients who underwent IPL, on average, required longer times for the Pringle manoeuvre compared with patients undergoing EPD (40 min versus 29 min; P < 0.001). These results are not surprising because, in both the anterior and posterior approaches in IPL, the Pringle manoeuvre may be used during the isolation and division of the pedicle.
In this study, patients who underwent EPD had tumours that were significantly larger than those in patients who underwent IPL (4.5 cm in the IPL group versus 5.5 cm in the EPD group; P < 0.001). Tumour size was independently associated with the choice of technique. This may be related to tumour proximity to the porta hepatis, in which the IPL approach may have had a high risk for a positive margin. The other factor independently associated with the choice of inflow control technique was the side of the liver undergoing resection. Patients undergoing right hepatectomy were more likely to undergo IPL than those undergoing left hepatectomy (69.5% of patients undergoing right hepatectomy and 37.0% of those undergoing left hepatectomy underwent IPL). This may be explained by the long extrahepatic course of the left portal structures and the relative ease of extrahepatic dissection. In addition, as Nakai et al. noted,9 drainage of the right posterior sectoral ducts into the main left hepatic ductal system may occur and this aberrant anatomy is at risk for injury during IPL.17 Results of the subanalyses performed in the right and left hepatectomy subgroups did not significantly differ from those for the whole cohort, although a few differences emerged. The impact of tumour size in the choice of approach was also significant in the analysis of the subgroup of patients who underwent a right hepatectomy. Tumour size, however, was not significantly associated with the approach in the left hepatectomy subgroup. Of note, among the operative factors, in the right hepatectomy subgroup, operative time was significantly shorter during IPL than during EPD.
Over time, there has been a significant change in the degree of complexity associated with surgical procedures. More patients now undergo repeat liver surgery or other liver resections or ablation at the time of hemi-hepatectomy. This may explain the higher rate of complications seen in the more recent period of the study. Surgical technique also changed, mainly in terms of the increased use of IPL with an anterior approach. This may reflect an increase in the number of patients who have had prior liver surgery, in whom resulting adhesions in the hepatic hilus make individual vessel dissection more difficult.
In summary, patients were more likely to undergo IPL during a right hepatectomy for smaller tumours, and IPL was not associated with an increased rate of postoperative complications. In the setting of smaller, peripherally located tumours, IPL is a safe and effective method of gaining vascular inflow control.
Conflicts of interest
None declared.
References
- Couinaud CM. A simplified method for controlled left hepatectomy. Surgery. 1985;97:358–361. [PubMed] [Google Scholar]
- Foster JH. History of liver surgery. Arch Surg. 1991;126:381–387. doi: 10.1001/archsurg.1991.01410270131020. [DOI] [PubMed] [Google Scholar]
- Fong Y, Blumgart LH. Useful stapling techniques in liver surgery. J Am Coll Surg. 1997;185:93–100. doi: 10.1016/s1072-7515(01)00889-4. [DOI] [PubMed] [Google Scholar]
- McEntee GP, Nagorney DM. Use of vascular staplers in major hepatic resections. Br J Surg. 1991;78:40–41. doi: 10.1002/bjs.1800780114. [DOI] [PubMed] [Google Scholar]
- DeMatteo RP, Fong Y, Jarnagin WR, Blumgart LH. Recent advances in hepatic resection. Semin Surg Oncol. 2000;19:200–207. doi: 10.1002/1098-2388(200009)19:2<200::aid-ssu11>3.0.co;2-m. [DOI] [PubMed] [Google Scholar]
- Cresswell AB, Welsh FK, John TG, Rees M. Evaluation of intrahepatic, extra-Glissonian stapling of the right porta hepatis vs. classical extrahepatic dissection during right hepatectomy. HPB. 2009;11:493–498. doi: 10.1111/j.1477-2574.2009.00083.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Launois B, Jamieson GG. The posterior intrahepatic approach for hepatectomy or removal of segments of the liver. Surg Gynecol Obstet. 1992;174:155–158. [PubMed] [Google Scholar]
- Launois B, Jamieson GG. The importance of Glisson's capsule and its sheaths in the intrahepatic approach to resection of the liver. Surg Gynecol Obstet. 1992;174:7–10. [PubMed] [Google Scholar]
- Nakai T, Koh K, Funai S, Kawabe T, Okuno K, Yasutomi M. Comparison of controlled and Glisson's pedicle transections of hepatic hilum occlusion for hepatic resection. J Am Coll Surg. 1999;189:300–304. doi: 10.1016/s1072-7515(99)00127-1. [DOI] [PubMed] [Google Scholar]
- Blumgart LH. Liver resection for benign disease and for liver and biliary tumours. In: Blumgart LH, Fong Y, editors. Surgery of the Liver and Biliary Tract. London: W. B. Saunders Co; 2000. pp. 1639–1713. [Google Scholar]
- Martin RC, Brennan MF, Jaques DP. Quality of complication reporting in the surgical literature. Ann Surg. 2002;235:803–813. doi: 10.1097/00000658-200206000-00007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Jarnagin WR, Gönen M, Fong Y, DeMatteo RP, Ben-Porat L, Little S, et al. Improvement in perioperative outcome after hepatic resection: analysis of 1803 consecutive cases over the past decade. Ann Surg. 2002;236:397–406. doi: 10.1097/01.SLA.0000029003.66466.B3. discussion 406–407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Belghiti J, Hiramatsu K, Benoist S, Massault P, Sauvanet A, Farges O. Seven hundred forty-seven hepatectomies in the 1990s: an update to evaluate the actual risk of liver resection. J Am Coll Surg. 2000;191:38–46. doi: 10.1016/s1072-7515(00)00261-1. [DOI] [PubMed] [Google Scholar]
- Gayowski TJ, Iwatsuki S, Madariaga JR, Selby R, Todo S, Irish W, et al. Experience in hepatic resection for metastatic colorectal cancer: analysis of clinical and pathologic risk factors. Surgery. 1994;116:703–710. discussion 710–711. [PMC free article] [PubMed] [Google Scholar]
- Pringle JH. Notes on the arrest of hepatic haemorrhage due to trauma. Ann Surg. 48:541–549. doi: 10.1097/00000658-190810000-00005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Couinaud CM. Anatomical and Technical Study. Paris: C. Couinaud, University of California; 1981. Controlled hepatectomies and exposure of the intrahepatic bile ducts; pp. 9–27. [Google Scholar]
- Takasaki K. Glissonian pedicle transection method for hepatic resection: a new concept of liver segmentation. J Hepatobiliary Pancreat Surg. 1998;5:286–291. doi: 10.1007/s005340050047. [DOI] [PubMed] [Google Scholar]
- Figueras J, Lopez-Ben S, Lladó L, Rafecas A, Torras J, Ramos E, et al. Hilar dissection versus the ‘Glissonian’ approach and stapling of the pedicle for major hepatectomies: a prospective, randomized trial. Ann Surg. 2003;238:111–119. doi: 10.1097/01.SLA.0000074981.02000.69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Andersson R, Saarela A, Tranberg KG, Bengmark S. Intra-abdominal abscess formation after major liver resection. Acta Chir Scand. 1990;156:707–710. [PubMed] [Google Scholar]
- Fong Y, Brennan MF, Brown K, Heffernan N, Blumgart LH. Drainage is unnecessary after elective liver resection. Am J Surg. 1996;171:158–162. doi: 10.1016/s0002-9610(99)80092-0. [DOI] [PubMed] [Google Scholar]
- Agrawal S, Belghiti J. Oncologic resection for malignant tumours of the liver. Ann Surg. 2011;253:656–665. doi: 10.1097/SLA.0b013e3181fc08ca. [DOI] [PubMed] [Google Scholar]