Abstract
Introduction
Enhanced Recovery After Surgery protocols have been implemented effectively after liver resection and provide benefits in terms of general morbidity rates. In order to optimise peri-operative care protocols and minimise morbidity, further investigation is required to identify factors associated with poor outcome after liver resection.
Methods
A retrospective analysis of patients undergoing liver resection and enhanced recovery care between January 2006 and September 2012 was conducted. Data were collected on patient outcome and demographics, operative and pathological details. Univariate and multivariate analyses were performed to determine independent predictors of adverse outcome.
Results
603 patients underwent liver resection during the study period. Morbidity and mortality rates were 34.3% and 1.5% respectively. The only predictor of major morbidity was extended resection (OR 4.079; 95% CI 2.177–7.642).
Conclusions
Extended resection is associated with major morbidity. When determining optimum peri-operative care, ERAS protocols must incorporate care components that can mitigate against morbidity associated with extended resection.
Introduction
Liver resection offers definitive surgery for a number of malignant and benign conditions. Traditionally liver resection has been associated with high post-operative mortality and morbidity rates.1 With centralisation of services to high volume centres, mortality rate has steadily declined to an accepted rate of less than five per cent.2 Morbidity rates, however, remain high at up to 45%.3 Not only is peri-operative morbidity delaying discharge, causing patient suffering and increased risk of mortality, but it is also associated with decreased overall long term survival following surgery for malignant disease.4
Therefore the minimisation of morbidity is fundamental to improving outcomes. Enhanced Recovery After Surgery (ERAS) programmes have been utilised extensively in colorectal surgery and have been shown to not only reduce hospital admission time but can reduce morbidity rates and is now established as standard of care.5 In liver resectional surgery there has been increasing interest in ERAS protocols.6 Not only is ERAS after liver surgery deemed safe and feasible but two recent RCTs7, 8 have shown reduced morbidity rates after liver resections.
Predictors of morbidity have been assessed before,3, 9 however, a recent assessment of predictors of outcome after resection of all tumour types, in a general population undergoing enhanced recovery multi-modal peri-operative care is lacking.
The investigating unit is experienced at providing enhanced recovery care after liver surgery.10, 11, 12 In order to formulate successful post-operative ERAS protocols and continue to effectively reduce surgical complications it is critical to evaluate the factors associated with post-operative morbidity and determine areas of care that can be optimised after liver resection. The aim of this study was to quantify outcomes and assess the predictors of morbidity after liver surgery when enhanced recovery care principles have been applied.
Methods
Approval from the NHS Lothian review board was obtained prior to commencing the data collection. All patients who underwent a liver resection between January 2006 and September 2012 within the Royal Infirmary of Edinburgh, UK, were identified from the prospectively collated Lothian Surgical Audit database. The Caldicott Guardian approval was obtained and these principles of data management were adhered to.
Peri-operative protocol
All patients undergoing liver resection underwent review in a multi-disciplinary team meeting where their radiological investigation was assessed by the surgical and radiological team in an attempt to ascertain resectability. This decision was made in concert with the oncology team and a decision made regarding pre-operative chemotherapy, further staging and suitability for surgery if appropriate.
Peri-operative care at the Royal Infirmary of Edinburgh, is based on a protocol first described by van Dam et al. (2008)11 (Table 1) and has been utilised in subsequent clinical trials.10, 12
Table 1.
Enhanced recovery protocol
| Time point | Recovery elements |
|---|---|
| Day before surgery | |
| Normal oral nutrition until midnight No preanesthetic medication |
|
| DOS | |
| Short-acting i.v. anaesthetic agent No nasogastric drainage; if used, remove immediately after surgery Warm i.v. fluids, and upper and lower body air-warming device Prophylactic antibiotics Avoidance of excessive i.v. fluids No routine drainage of the peritoneal cavity Epidural analgesia Restart oral intake of water/nutrition ad libitum |
|
| POD 1 | |
| Arterial and central lines out Patient mobilizes around bed Discontinuation of intravenous fluids if haemodynamically stable and drinks more than 1 L of fluid Normal diet Continue epidural 1000 mg paracetamol every 6 h |
|
| POD 2 | |
| Continue mobilization Patient to mobilize 1000 mg paracetamol every 6 h Urinary catheter out Normal diet Oral analgesia Transfer to general ward |
|
| POD 3 | |
| Epidural out Continue mobilization Normal diet Check discharge criteria |
|
| POD 4 | |
| Check discharge criteria | |
The patients were routinely reviewed in the out-patient clinic approximately 4 weeks post discharge and discussed in the MDT meeting where pathological analysis of the specimen was reported and follow up and adjuvant chemotherapy was decided upon.
Data collection
Demographic details, namely age, gender and comorbidities were collected from the patient case files. Presence of co-morbidity was determined when at least one co-morbidity was described in the pre-operative clinic assessment.
Pre-operative oncological data were obtained from the transcription of the Multi-Disciplinary Team (MDT) meeting prior to resection. From this report primary resectability, portal vein emobilisation (PVE) and neoadjuvant chemotherapy were obtained. Confirmation of PVE was obtained from CT report. Neoadjuvant chemotherapy was also confirmed from the MDT report and deemed to be positive if chemotherapy was commenced prior to resection. Pre-operative blood tests were obtained from the laboratory investigations contained within the electronic patient records.
The extent of the procedure and use of in-flow occlusion were determined with major resection being defined as resection of three or more segments. Extended resection was defined as resection of five or more segments as per the Brisbane criteria.13
Intra-operative and post-operative blood transfusion information was obtained from the Blood Transfusion Service database of prospectively collected data. Day of operation from the operation note and admission dates were compared with the dates of transfusion of RCC and reported as receiving transfusion accordingly.
Admission data were gained from the patient case records and clinical course, complications and index length of stay were documented. Post-operative complications were also gathered from the patent records, namely the discharge letter from the discharging surgeon. The nature of the complication was recorded as per the operating surgeons' discharge documentation. Morbidity severity was subsequently categorised into major morbidity (Clavien Dindo grade three or above) and minor morbidity (Clavien Dindo grade below three).
Histopathological data were gained from the original pathology report. Underlying tumour pathology, size and number were recorded. Abnormalities in the underlying liver parenchyma was also confirmed from the pathology report as was resection margin with a R1 resection being confirmed if the tumour edge was within 1 mm of the resection margin.
Peri-operative mortality was confirmed if patients were reported to have died in hospital or thirty days after operation. Thirty day mortality was confirmed by data from the Information and Statistics Department (Scotland).
Statistical analysis
Continuous data were expressed as median and interquartile range (IQR) and compared using the Mann Whitney U test when not normally distributed. If normal distribution was observed a mean and standard deviation were presented and unpaired t-tests performed to determine differences in means. Categorical data were assessed with Chi squared test, Chi squared test for trend or Fisher's exact test where appropriate. Multivariable logistic regression analysis was performed to assess for independent predictors of morbidity. Factors were entered into the multivariable analysis if they were clinically relevant or achieved significance at the 5% level. P values were two-tailed and considered statistically significant if less than 0.05.
Results
During the study period 603 patients underwent liver resection preoperative data are shown in Table 2. Operative data are shown in Table 3. Pathological outcome data are shown in supplementary table 1.
Table 2.
Pre-operative patient characteristics
| All (n = 603) | Cholangiocarcinoma (n = 50) | HCC (n = 75) | CLM (n = 381) | Other malignant (n = 34) | Benign (n = 63) | |
|---|---|---|---|---|---|---|
| Age (years) | 60.5 ± 13.4 | 60.1 ± 11.8 | 63.0 ± 15.0 | 62.4 ± 11.8 | 66.0 ± 12.0 | 48.8 ± 15.6 |
| Female | 253 (42.0) | 21 (42.0) | 20 (26.7) | 150 (39.4) | 19 (55.9) | 43 (68.3) |
| Comorbidities | 252 (41.8) | 18 (36.0) | 64 (85.3) | 139 (36.5) | 12 (35.3) | 19 (30.2) |
| Neoadjuvant chemotherapy | 168 (27.9) | 0 (0) | 0 (0) | 158 (41.5) | 9 (26.5) | 1 (1.6) |
| PVE | 41 (6.8) | 6 (12.0) | 4 (5.3) | 28 (7.3) | 2 (5.9) | 1 (1.6) |
Data presented as mean ± sd or n(%).
WCC, white cell count; PVE, portal vein embolisation; ALT, alanine transaminase.
Table 3.
Operative characteristics
| Total (n = 603) | Cholangiocarcinoma (n = 50) | HCC (n = 75) | CLM (n = 381) | Other malignant (n = 34) | Benign (n = 63) | |
|---|---|---|---|---|---|---|
| Procedure | ||||||
| Major resection | 336 (55.7) | 46 (92.0) | 25 (33.3) | 209 (54.9) | 17 (50.0) | 39 (61.9) |
| Minor resection | 267 (44.3) | 4 (8.0) | 50 (66.7) | 172 (45.1) | 17 (50.0) | 24 (38.1) |
| Pringle | 157 (26.0) | 5 (10.0) | 20 (26.7) | 110 (28.9) | 8 (23.5) | 14 (22.2) |
| Blood transfusion | ||||||
| Intra-operative | 88 (14.6) | 9 (18.0) | 15 (20.0) | 49 (12.9) | 6 (17.6) | 9 (14.3) |
| No. of units | 2 (2–4) | 2 (2–4) | 2 (1–4) | 2 (2–5) | 2 (1–10) | 2 (2–5) |
| Rest of admission | 74 (12.3) | 10 (20.0) | 12 (16.0) | 37 (9.7) | 4 (11.8) | 11 (17.5) |
| No. of units | 1 (1–2) | 1 (1–2) | 2 (1–3) | 1 (1–2) | 1 (1–2) | 1 (1–2) |
| Redo procedure | 41 (6.8) | 2 (4.0) | 3 (4.0) | 34 (8.9) | 1 (2.9) | 1 (1.6) |
| Laparoscopic resection | 33 (5.5) | 0 (0) | 12 (16.0) | 17 (4.5) | 1 (2.9) | 3 (4.8) |
Data are presented as n(%) or median (IQR).
Post-operative outcome data are shown in Table 4. Supplementary Tables 2 and 3 show individual complication type. Nine (1.5%) patients died in hospital over the review period. All nine patients died in the intensive care unit having suffered a major post-operative complication. Six of these patients died of multi-organ failure secondary to hepatic failure, one died after complications following significant post-operative bleeding and two died following prolonged respiratory failure.
Table 4.
Post-operative outcomes
| Total (n = 603) | Cholangiocarcinoma (n = 50) | HCC (n = 75) | CLM (n = 381) | Other malignant (n = 34) | Benign (n = 63) | |
|---|---|---|---|---|---|---|
| Overall Complications | 207 (34.3) | 27 (54.0) | 28 (37.3) | 124 (32.5) | 12 (35.3) | 16 (25.4) |
| Major complication (CD 3–5) | 72 (11.9) | 13 (26.0) | 9 (12.0) | 42 (11.0) | 1 (2.9) | 7 (11.1) |
| Minor complication (CD 1–2) | 135 (22.4) | 14 (28.0) | 19 (25.3) | 82 (21.5) | 11 (32.4) | 9 (14.3) |
| Bile leak | 30 (5.0) | 3 (6.0) | 1 (1.3) | 19 (5.0) | 2 (5.9) | 5 (7.9) |
| Liver failure | 15 (2.5) | 5 (10.0) | 2 (2.7) | 6 (1.6) | 1 (2.9) | 1 (1.6) |
| Bleeding | 11 (1.8) | 0 (0) | 3 (4.0) | 8 (2.1) | 0 (0) | 0 (0) |
| IAA | 31 (5.1) | 7 (14.0) | 3 (4.0) | 16 (4.2) | 3 (8.8) | 2 (3.2) |
| Length of stay (days) | 7 (5–10) | 12 (7–21) | 7 (5–9) | 7 (5–10) | 7 (6–9) | 7 (5–9) |
| Readmission | 45 (7.5) | 8 (16.0) | 2 (2.7) | 26 (6.8) | 4 (11.8) | 5 (7.9) |
| 30 day mortality | 9 (1.5) | 3 (6.0) | 2 (2.7) | 4 (1.0) | 0 (0) | 0 (0) |
Data are presented as n (%) or median (IQR). Overall complications are inclusive of nine patients who suffered mortality.
IAA, intra-abdominal abscess; CD, Clavien Dindo.
Patients having resection for cholangiocarcinoma had the highest major morbidity rates (n=13, 26%; p=0.01) as defined as complications relating to a Clavien Dindo classification of three and above (Table 4).
After these data were gathered, univariate analyses were performed to identify significant factors associated with major morbidity (Table 5).
Table 5.
Univariate and multivariate analysis of factors associated with morbidity (CD grade III–V)
| Total (n = 603) | CD III–V (n = 72) | Univariate analysis p value | OR | 95% CI | Multivariate analysis p value |
|---|---|---|---|---|---|
| Age – 60.5 ± 13.4 | 62.6 ± 14.0 | 0.152a | |||
| Female (n = 253) | 26 (10.3) | 0.284b | |||
| Re-do (n = 41) | 5 (12.2) | 1.000c | |||
| Primarily resectable (n = 512) | 61 (11.9) | 0.962b | |||
| Neoadjuvant chemo (n = 168) | 20 (11.9) | 0.987b | |||
| EHD (n = 61) | 6 (9.8) | 0.593b | |||
| Co-morbidities (n = 252) | 30 (11.9) | 0.982b | |||
| Major resection (n = 336) | 51 (15.2) | 0.006b | 1.122 | 0.597–2.109 | 0.720 |
| Extended resection (n = 96) | 30 (31.2) | <0.0001b | 4.079 | 2.177–7.642 | 0.000 |
| Steatosis (n = 252) | 24 (9.5) | 0.121b | |||
| Steatohepatitis (n = 43) | 5 (11.6) | 1.000c | |||
| Sinusoidal dilatation (n = 67) | 4 (6.0) | 0.159c | |||
| Cirrhosis (n = 34) | 4 (11.8) | 1.000c | |||
| Blood transfusion (n = 88) | 10 (11.4) | 0.856b | |||
| Albumin ≤30 g/L (n = 31) | 8 (25.8) | 0.014b | 2.420 | 0.989–5.922 | 0.053 |
| Platelets ≤ 150 × 109/L (n = 51) | 8 (15.7) | 0.388b | |||
| Bilirubin ≥20 μmol/L (n = 63) | 15 (23.8) | 0.002b | 1.850 | 0.931–3.676 | 0.079 |
Data are presented as n (%) or mean ± sd.
EHD, extra hepatic disease.
Unpaired t-test.
χ2 test.
Fisher's exact test.
Extended resection was the only independent predictor of major adverse outcome. A further analysis of complication types following extended resection was performed and significantly higher levels of liver failure, intra-abdominal abscess and mortality were observed in patients who underwent an extended resection (Table 6).
Table 6.
Adverse outcomes associated with extended resection
| Extended resection (n = 96) | Major, minor resection (n = 507) | p Value | |
|---|---|---|---|
| Liver failure | 6 (6.2) | 9 (1.8) | 0.010a |
| Bile leak | 13 (13.5) | 17 (3.4) | <0.001a |
| Bleeding | 4 (4.2) | 7 (1.4) | 0.081b |
| Intra-abdominal abscess | 10 (10.4) | 21 (4.1) | 0.010a |
| Mortality | 6 (6.2) | 3 (0.6) | 0.001b |
Data are presented as n (%).
χ2 test.
Fisher's exact test.
Discussion
This study reports morbidity rates of 34.3%. Morbidity after liver surgery has been reported between 22 and 45%.2, 3, 9, 14 Morbidity after liver resection following ERAS protocols has been reported between 11% and 46%6 with overall morbidity being reduced following resection managed with enhanced recovery principles compared to traditional practice. Surgical specific morbidity, however, has not been seen to be affected by ERAS care.6 The findings from the current study corroborate these findings with extended resection being identified as an independent predictor of major morbidity. Significantly higher rates of liver impairment after extended resection compared to major and minor hepatic resections were observed. Despite the use of advanced techniques to prevent liver failure (PVE, tumour volume reduction and two stage procedures), and decreased levels of hepatic insufficiency compared with previous studies, this complication remains an on going concern for major resections. The present study also observed significantly higher rates of bile leak (13.5% vs 3.4%) and abscess formation (10.4% vs 4.1%) after extended resections compared to non-extended resections. Complex resections and post-operative morbidity must however be considered in the context of the survival benefit following major resection for, in many cases, colorectal liver metastases. Increasingly favourable long-term outcomes are being achieved following complex, extensive resection for advanced disease, justifying an aggressive surgical approach.15 Current techniques to reduce post-operative bleeding and bile leak are not incorporated in the investigating unit's peri-operative care pathway and have not routinely been utilised in other reported ERAS protocols.6 CVP minimisation to reduce intra-operative blood loss has been included in only two liver ERAS trials7, 11 despite the well reported adverse impact of blood loss to short term outcomes.3 The current study has identified that despite optimising peri-operative care, extended resection still results in significant morbidity. It is therefore suggested that elements often employed to reduce perioperative bleeding or bile leak could be more formally standardised into operative practice and incorporated into ERAS pathways. Furthermore when addressing non-surgical complications and practice, extended resections provide further challenges. Simple analgesic regimen formulation is challenging and acetaminophen metabolism is currently an unknown quantity after major resection. Moreover, nutrition post-operatively is an important consideration after extended liver resection due to potential nutritional insufficiency related to the increased surgical insult and increased risk of sepsis,16 Further work is required to further investigate the effects of small liver remnant volume on outcome after liver resection and how ERAS protocols need to accommodate this unique aspect of liver surgery that is not translatable from other areas of abdominal surgery.
These results should be taken in context of the limitations of this study. The completeness of the patients' records introduces reporter bias in the analysis. If the complication was not reported then it was not included and there is a possibility that some complications were missed, most likely minor complications and so the outcomes could be biased towards complications of a higher Clavien Dindo classification. A similar picture is also true for reporting of co-morbidities. If no mention of co-morbidities was made then the assumption was made that no significant co-morbidities were present. Intra-operative data were also not universally available. In particular, operation time and estimated blood loss. These data are therefore not included. This was dealt with by looking at those patients who had received a blood transfusion, however the findings are likely to be less discriminatory.
Footnotes
Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.hpb.2015.10.011.
Funding sources
None.
Conflicts of interest
None declared.
Appendix A. Supplementary data
The following is the supplementary data related to this article:
References
- 1.Virani S., Michaelson J.S., Hutter M.M., Lancaster R.T., Warshaw A.L., Henderson W.G. Morbidity and mortality after liver resection: results of the patient safety in surgery study. J Am Coll Surg. 2007;204:1284–1292. doi: 10.1016/j.jamcollsurg.2007.02.067. [DOI] [PubMed] [Google Scholar]
- 2.Cescon M., Vetrone G., Grazi G.L., Ramacciato G., Ercolani G., Ravaioli M. Trends in perioperative outcome after hepatic resection: analysis of 1500 consecutive unselected cases over 20 years. Ann Surg. 2009;249:995–1002. doi: 10.1097/SLA.0b013e3181a63c74. [DOI] [PubMed] [Google Scholar]
- 3.Jarnagin W.R., Gonen M., Fong Y., DeMatteo R.P., Ben-Porat L., Little S. Improvement in perioperative outcome after hepatic resection: analysis of 1,803 consecutive cases over the past decade. Ann Surg. 2002;236:397–406. doi: 10.1097/01.SLA.0000029003.66466.B3. discussion -7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Mavros M.N., de Jong M., Dogeas E., Hyder O., Pawlik T.M. Impact of complications on long-term survival after resection of colorectal liver metastases. Br J Surg. 2013;100:711–718. doi: 10.1002/bjs.9060. [DOI] [PubMed] [Google Scholar]
- 5.Kehlet H., Wilmore D.W. Evidence-based surgical care and the evolution of fast-track surgery. Ann Surg. 2008;248:189–198. doi: 10.1097/SLA.0b013e31817f2c1a. [DOI] [PubMed] [Google Scholar]
- 6.Hughes M.J., McNally S., Wigmore S.J. Enhanced recovery following liver surgery: a systematic review and meta-analysis. HPB: Off J Int Hepato Pancreato Biliary Assoc. 2014;16:699–706. doi: 10.1111/hpb.12245. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Jones C., Kelliher L., Dickinson M., Riga A., Worthington T., Scott M.J. Randomized clinical trial on enhanced recovery versus standard care following open liver resection. Br J Surg. 2013;100:1015–1024. doi: 10.1002/bjs.9165. [DOI] [PubMed] [Google Scholar]
- 8.Ni C.Y., Yang Y., Chang Y.Q., Cai H., Xu B., Yang F. Fast-track surgery improves postoperative recovery in patients undergoing partial hepatectomy for primary liver cancer: a prospective randomized controlled trial. Eur J Surg Oncol: J Eur Soc Surg Oncol Br Assoc Surg Oncol. 2013;39:542–547. doi: 10.1016/j.ejso.2013.03.013. [DOI] [PubMed] [Google Scholar]
- 9.Poon R.T., Fan S.T., Lo C.M., Liu C.L., Lam C.M., Yuen W.K. Improving perioperative outcome expands the role of hepatectomy in management of benign and malignant hepatobiliary diseases: analysis of 1222 consecutive patients from a prospective database. Ann Surg. 2004;240:698–708. doi: 10.1097/01.sla.0000141195.66155.0c. discussion -10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Revie E.J., McKeown D.W., Wilson J.A., Garden O.J., Wigmore S.J. Randomized clinical trial of local infiltration plus patient-controlled opiate analgesia vs. epidural analgesia following liver resection surgery. HPB: Off J Int Hepato Pancreato Biliary Assoc. 2012;14:611–618. doi: 10.1111/j.1477-2574.2012.00490.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.van Dam R.M., Hendry P.O., Coolsen M.M., Bemelmans M.H., Lassen K., Revhaug A. Initial experience with a multimodal enhanced recovery programme in patients undergoing liver resection. Br J Surg. 2008;95:969–975. doi: 10.1002/bjs.6227. [DOI] [PubMed] [Google Scholar]
- 12.Hendry P.O., van Dam R.M., Bukkems S.F., McKeown D.W., Parks R.W., Preston T. Randomized clinical trial of laxatives and oral nutritional supplements within an enhanced recovery after surgery protocol following liver resection. Br J Surg. 2010;97:1198–1206. doi: 10.1002/bjs.7120. [DOI] [PubMed] [Google Scholar]
- 13.Strasberg S.M. Nomenclature of hepatic anatomy and resections: a review of the Brisbane 2000 system. J Hepato-biliary-pancreatic Surg. 2005;12:351–355. doi: 10.1007/s00534-005-0999-7. [DOI] [PubMed] [Google Scholar]
- 14.Zimmitti G., Roses R.E., Andreou A., Shindoh J., Curley S.A., Aloia T.A. Greater complexity of liver surgery is not associated with an increased incidence of liver-related complications except for bile leak: an experience with 2,628 consecutive resections. J Gastrointest Surg: Off J Soc Surg Alimentary Tract. 2013;17:57–64. doi: 10.1007/s11605-012-2000-9. discussion p.-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Lam V.W., Laurence J.M., Johnston E., Hollands M.J., Pleass H.C., Richardson A.J. A systematic review of two-stage hepatectomy in patients with initially unresectable colorectal liver metastases. HPB: Off J Int Hepato Pancreato Biliary Assoc. 2013;15:483–491. doi: 10.1111/j.1477-2574.2012.00607.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Sandstrom R., Drott C., Hyltander A., Arfvidsson B., Schersten T., Wickstrom I. The effect of postoperative intravenous feeding (TPN) on outcome following major surgery evaluated in a randomized study. Ann Surg. 1993;217:185–195. doi: 10.1097/00000658-199302000-00013. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.