Abstract
Background
Hepatic and pancreatic surgery is rarely performed in patients with end-stage renal disease (ESRD). The present authors used a national clinical database to characterize outcomes and perioperative risk in ESRD patients who require hepatic or pancreatic resection.
Methods
The 2005–2011 National Surgical Quality Improvement Program database was queried for all patients undergoing hepatic or pancreatic resection. Patients were classified by the presence or absence of ESRD. The independent effects of ESRD on outcomes were assessed after propensity score adjustment and multivariable logistic regression.
Results
Of the 27 376 patients submitted to hepatic or pancreatic procedures identified in the database, 101 patients were found to have preoperative ESRD. Patients with ESRD experienced perioperative mortality at a rate similar to that in those without ESRD (5.0% versus 2.3%; P = 0.08). After risk adjustment, the presence of ESRD was associated with three-fold higher odds of postoperative sepsis (adjusted odds ratio: 2.98, P = 0.014), but no significant differences in mortality or major complication rates.
Conclusions
Hepatic and pancreatic resections can be performed safely in selected patients with ESRD. These patients may have an increased risk for the development of postoperative sepsis. Further study is needed to characterize modifiable risk factors that impact outcomes in patients with ESRD who require hepatic or pancreatic resection.
Introduction
Hepatic and pancreatic surgery is rarely performed in patients with end-stage renal disease (ESRD), in part as a result of perceptions of prohibitive operative risk. Although the safety of hepatic and pancreatic resections has improved significantly in the modern era, these complex abdominal procedures are still associated with significant perioperative morbidity even in healthy patients. Patients with ESRD are at increased risk for cardiovascular complications and exhibit abnormalities in immunologic function and coagulation, which have significant implications on their ability to tolerate and recover from major abdominal surgery.1,2 These considerations are likely to contribute to reluctance on the part of surgeons to recommend major abdominal surgery to these patients.
However, the ESRD population in the USA continues to grow. There are currently an estimated 500 000 patients with ESRD on dialysis and this number is projected to grow by 7% per year.3 As the life expectancy of patients with ESRD improves, it is likely that more patients will present with hepatic and pancreatic pathologies requiring operative intervention.
In view of the relative paucity of studies examining the outcomes of complex abdominal surgery in ESRD patients, the present authors sought to explore this topic specifically in relation to hepatic and pancreatic resection. The primary goals of this study were to characterize perioperative outcomes of hepatic or pancreatic resection in patients with ESRD and to compare these outcomes with those in patients without ESRD. As patients in this population frequently have numerous other comorbidities, risk adjustment using propensity-based techniques was performed.
Materials and methods
The 2005–2011 American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) Participant User Files (PUFs) were queried for all patients undergoing hepatic or pancreatic resection. Patients were included for analysis using Current Procedural Terminology (CPT) codes for partial hepatectomy (47120, 47122, 47125 and 47130), distal pancreatectomy (48140, 48145, 48146) and pancreaticoduodenectomy (Whipple-type procedure: 48150, 48152–48154). Patients coded as submitted to an emergent operation and patients with American Society of Anesthesiologists (ASA) class 5 status, defined as moribund and not expected to survive without the operation, were excluded from analysis. The main predictor variable of interest was the presence of ESRD, defined as requiring treatment with peritoneal dialysis, haemodialysis, haemofiltration, haemodiafiltration or ultrafiltration within the 2 weeks prior to surgery. The primary outcome measure was rate of major complications, defined a priori as the occurrence of one or more of the available major complications captured by NSQIP, which roughly correspond to complications of Clavien–Dindo Grade III or higher.
Statistical analysis
For continuous and categorical variables, measures of central tendency and proportions were assessed, respectively. Preoperative baseline patient characteristics and specific postoperative complications and endpoints were included in univariate comparisons of patients with and without ESRD, using Pearson’s chi-squared test for categorical variables and Student’s t-test for continuous variables. Because of the significant differences in baseline characteristics between patients with and without ESRD, patients were matched according to their propensity for undergoing an operation in the setting of ESRD, using a 3:1 optimized nearest-neighbour algorithm.4–6 To adjust for residual imbalance following matching, a non-parsimonious multivariable logistic regression model was then created in order to estimate the independent effect of ESRD on the study’s primary and secondary endpoints. As a post hoc sensitivity analysis focusing on patients submitted to the highest-risk procedures, two subsets of the original study population were then examined separately: (i) patients submitted to major hepatectomy (defined as trisegmentectomy or right hepatectomy), and (ii) patients submitted to pancreaticoduodenectomy.
Model diagnostics and balance were assessed; no major model assumptions were violated. In a manner consistent with previous studies using ACS-NSQIP data, laboratory values were categorized as normal or abnormal using standard cut-offs, and missing laboratory data were assigned a third categorical indicator variable.7–9 Missing data for other variables were handled with case-wise deletion, given the substantial completeness of NSQIP data. A P-value of ≤0.05 was used to indicate statistical significance for all comparisons and analyses of outcomes. Statistical analyses were performed using R Version 3.0.1 (R Foundation for Statistical Computing, Vienna, Austria).
Results
A total of 27 376 patients were identified in the ACS-NSQIP database as meeting the study inclusion criteria. Of these, 10 175 (37.2%) underwent partial hepatectomy, 5608 (20.5%) underwent distal pancreatectomy and 11 593 (42.3%) underwent pancreatoduodenectomy. A total of 101 patients were identified as having preoperative ESRD (Fig. 1). Baseline characteristics in all patients stratified by ESRD status are shown in Table 1. Compared with the non-ESRD controls, patients with ESRD were more often male (60.4% versus 48.4%; P = 0.021), had higher ASA scores (P < 0.001), more often met criteria for systemic inflammatory response syndrome preoperatively (5.9% versus 1.5%; P < 0.001), more often had non-independent functional status (13.9% versus 2.0%; P < 0.001), and had lower median haematocrit (35.5% versus 38.8%; P < 0.001). After propensity matching, groups were highly similar without significant differences (Table 2).
Figure 1.
Study design showing patient selection. NSQIP, National Surgical Quality Improvement Program; PUFs, Participant User Files; HPB, hepatopancreatobiliary; ESRD, end-stage renal disease
Table 1.
Baseline characteristics of patients submitted to major hepatopancreatobiliary procedures, stratified by end-stage renal disease (ESRD) status
| Variable | All patients (n = 27 376) | Patients without ESRD (n = 27 275) | Patients with ESRD (n = 101) | P-value |
|---|---|---|---|---|
| Age, years, median (range) | 62 (53–71) | 62 (53–71) | 60 (53–68) | 0.192 |
| Female sex, n (%) | 14 091 (51.5%) | 14 051 (51.6%) | 40 (39.6%) | 0.021 |
| ASA classification | <0.001 | |||
| 1 (no disturbance) | 377 (1.4%) | 377 (1.4%) | 0 | |
| 2 (mild disturbance) | 8616 (31.5%) | 8615 (31.6%) | 1 (1.0%) | |
| 3 (severe disturbance) | 17 137 (62.6%) | 17 078 (62.6%) | 59 (58.4%) | |
| 4 (life-threatening) | 1246 (4.6%) | 1205 (4.4%) | 41 (40.6%) | |
| 5 (moribund) | 0 | 0 | 0 | |
| Procedure type, n (%) | 0.487 | |||
| Distal pancreatectomy | 5608 (20.5%) | 5583 (20.5%) | 25 (24.8%) | |
| Liver resection | 10 175 (37.2%) | 10 137 (37.2%) | 38 (37.6%) | |
| Whipple procedure | 11 593 (42.3%) | 11 555 (42.4%) | 38 (37.6%) | |
| Preoperative sepsis, n (%) | <0.001 | |||
| None | 26 600 (97.9%) | 26 509 (97.2%) | 91 (90.1%) | |
| SIRS | 425 (1.6%) | 419 (1.5%) | 6 (5.9%) | |
| Sepsis | 121 (0.4%) | 119 (0.4%) | 2 (2.0%) | |
| Septic shock, | 12 (0.0004%) | 10 (0.0004%) | 2 (2.0%) | |
| Dyspnoea, n (%) | 2349 (8.6%) | 2332 (8.5%) | 17 (16.8%) | 0.005 |
| Non-independent functional status, n (%) | 563 (2.1%) | 549 (2.0%) | 14 (13.9%) | <0.001 |
| DNR status | 46 (0.2%) | 45 (0.2%) | 1 (1.0%) | 0.157 |
| Tobacco use, n (%) | 5354 (19.6%) | 5333 (19.6%) | 21 (20.8%) | 0.851 |
| Alcohol use, n (%) | 686 (2.9%) | 686 (2.9%) | 0 | 0.187 |
| Diabetes, n (%) | 5479 (20.0%) | 5431 (19.9%) | 48 (47.5%) | <0.001 |
| COPD, n (%) | 1075 (3.9%) | 1069 (3.9%) | 6 (5.9%) | 0.431 |
| CAD, n (%) | 2221 (9.3%) | 2204 (9.2%) | 17 (19.1%) | 0.003 |
| Bleeding disorder, n (%) | 806 (2.9%) | 800 (2.9%) | 6 (5.9%) | 0.136 |
| Recent steroid use, n (%) | 540 (2.0%) | 535 (2.0%) | 5 (5.0%) | 0.050 |
| Disseminated malignancy, n (%) | 4494 (16.4%) | 4481 (16.4%) | 13 (12.9%) | 0.407 |
| Recent weight loss, n (%) | 3138 (11.5%) | 3124 (11.5%) | 14 (13.9%) | 0.547 |
| Recent chemotherapy (30 days), n (%) | 1104 (4.6%) | 1104 (4.6%) | 0 | 0.036 |
| Contaminated/dirty case, n (%) | 2325 (8.5%) | 2319 (8.5%) | 6 (5.9%) | 0.457 |
| Bilirubin > 4 mg/dl, n (%) | 1671 (6.9%) | 1665 (6.9%) | 6 (6.5%) | 0.999 |
| Haematocrit, %, median (range) | 38.8 (35.4–41.7) | 38.8 (35.4–41.7) | 35.5 (31.1–39.3) | <0.001 |
| Albumin, g/dl, median (range) | 3.9 (3.5–4.3) | 3.9 (3.5–4.3) | 3.4 (2.7–3.9) | <0.001 |
| WBC count, × 103/μl, median (range) | 6.8 (5.5–8.4) | 6.8 (5.5–8.4) | 7.4 (5.6–9.1) | 0.038 |
| Platelets, × 103/μl, median (range) | 241 (192–302) | 241 (192–302) | 213 (173–264) | 0.086 |
ASA, American Society of Anesthesiologists; SIRS, systemic inflammatory response syndrome; DNR, do not resuscitate; COPD, chronic obstructive pulmonary disease; CAD, coronary artery disease; WBC, white blood cell.
Table 2.
Baseline characteristics of patients submitted to major hepatopancreatobiliary procedures, stratified by end-stage renal disease (ESRD) status following 3:1 propensity matching
| Variable | Patients without ESRD (n = 303) | Patients with ESRD (n = 101) | P-value |
|---|---|---|---|
| Age, years, median (range) | 60 (50.5–71) | 60 (53–68) | 0.790 |
| Female sex, n (%) | 124 (40.9%) | 40 (39.6%) | 0.907 |
| ASA classification, n (%) | 0.428 | ||
| 1 (no disturbance) | 0 | 0 | |
| 2 (mild disturbance) | 11 (3.6%) | 1 (1.0%) | |
| 3 (severe disturbance) | 164 (54.1%) | 59 (58.4%) | |
| 4 (life-threatening) | 128 (42.2%) | 41 (40.6%) | |
| 5 (moribund) | 0 | 0 | |
| Procedure type, n (%) | 0.322 | ||
| Distal pancreatectomy | 64 (21.1%) | 25 (24.8%) | |
| Liver resection | 99 (32.7%) | 38 (37.6%) | |
| Whipple procedure | 140 (46.2%) | 38 (37.6%) | |
| Preoperative sepsis, n (%) | 0.114 | ||
| None | 284 (93.7%) | 91 (90.1%) | |
| SIRS | 14 (4.6%) | 6 (5.9%) | |
| Sepsis | 5 (1.7%) | 2 (2.0%) | |
| Septic shock | 0 | 2 (2.0%) | |
| Dyspnoea, n (%) | 40 (13.2%) | 17 (16.8%) | 0.458 |
| Non-independent functional status, n (%) | 37 (12.2%) | 14 (13.9%) | 0.795 |
| DNR status, n (%) | 7 (3.7%) | 1 (1.1%) | 0.443 |
| Tobacco use, n (%) | 55 (18.2%) | 21 (20.8%) | 0.659 |
| Alcohol use, n (%) | 1 (0.5%) | 0 | 0.999 |
| Diabetes, n (%) | 139 (45.9%) | 48 (47.5%) | 0.863 |
| COPD, n (%) | 17 (5.6%) | 6 (5.9%) | 0.999 |
| CAD, n (%) | 26 (15.5%) | 17 (19.1%) | 0.572 |
| Bleeding disorder, n (%) | 24 (7.9%) | 6 (5.9%) | 0.661 |
| Recent steroid use, n (%) | 12 (4.0%) | 5 (5.0%) | 0.775 |
| Disseminated malignancy, n (%) | 44 (14.5%) | 13 (12.9%) | 0.804 |
| Recent weight loss, n (%) | 37 (12.2%) | 14 (13.9%) | 0.795 |
| Recent chemotherapy (30 days), n (%) | 18 (10.7%) | 0 | <0.001 |
| Contaminated/dirty case, n (%) | 24 (7.9%) | 6 (5.9%) | 0.661 |
| Bilirubin >4 mg/dl, n (%) | 33 (11.9%) | 6 (6.5%) | 0.197 |
| Haematocrit, %, median (range) | 35.4 (31.9–39.8) | 35.5 (31.1–39.3) | 0.300 |
| Albumin, g/dl, median (range) | 3.5 (2.9–4.0) | 3.4 (2.7–3.9) | 0.184 |
| WBC count, × 103/μl, median (range) | 7 (5.4–8.868) | 7.4 (5.6–9.1) | 0.568 |
| Platelets, × 103/μl, median (range) | 230.5 (173.2–292.0) | 213.0 (173.0–264.0) | 0.285 |
ASA, American Society of Anesthesiologists; SIRS, systemic inflammatory response syndrome; DNR, do not resuscitate; COPD, chronic obstructive pulmonary disease; CAD, coronary artery disease; WBC, white blood cell.
In an unadjusted comparison referring to the study’s primary outcomes, patients with ESRD did not show increased perioperative mortality (5.0% versus 2.3%; P = 0.08), but did have a significantly higher rate of major complications (45.5% versus 28.5%; P < 0.001) (Table 3). Similarly, patients with ESRD had a higher rate of overall complications (48.5% versus 35.9%; P = 0.011), a longer median length of stay (10 days versus 7 days; P < 0.001), and a higher rate of postoperative sepsis (19.8% versus 8.4%; P < 0.001). Following propensity matching and multivariable logistic regression, ESRD was not found to be associated with increased odds of major complications [adjusted odds ratio (AOR) 0.98, P = 0.95]. Patients with ESRD, however, had nearly three-fold higher odds of sepsis compared with their non-ESRD counterparts (AOR 2.98, P = 0.014). There were no statistically significant differences in other secondary outcomes, including mortality (AOR 0.97, P = 0.96), reoperation (AOR 1.17, P = 0.75), respiratory complications (AOR 1.77, P = 0.14) and surgical site infection (AOR 0.94, P = 0.89) (Table 4).
Table 3.
Unadjusted outcomes in patients submitted to major hepatopancreatobiliary procedures stratified by end-stage renal disease (ESRD) status
| Variable | All patients (n = 27 376) | Patients without ESRD (n = 27 275) | Patients with ESRD (n = 101) | P-value |
|---|---|---|---|---|
| Mortality within 30 days, n (%) | 621 (2.3%) | 616 (2.3%) | 5 (5.0%) | 0.080 |
| Overall complication rate, n (%) | 9840 (35.9%) | 9791 (35.9%) | 49 (48.5%) | 0.011 |
| Major complication rate, n (%) | 7816 (28.6%) | 7770 (28.5%) | 46 (45.5%) | <0.001 |
| Early return to operating room, n (%) | 1446 (5.3%) | 1436 (5.3%) | 10 (9.9%) | 0.063 |
| Length of stay, days, median (range) | 7 (6–12) | 7 (6–12) | 10 (6–20.5) | <0.001 |
| Operative time, min, median (range) | 277 (191–378) | 277 (191–378) | 269 (190–355) | 0.426 |
| Superficial surgical site infection, n (%) | 1906 (7.0%) | 1901 (7.0%) | 5 (5.0%) | 0.557 |
| Deep surgical site infection, n (%) | 435 (1.6%) | 433 (1.6%) | 2 (2.0%) | 0.676 |
| Organ space surgical site infection, n (%) | 2390 (8.7%) | 2379 (8.7%) | 11 (10.9%) | 0.552 |
| Wound dehiscence, n (%) | 333 (1.2%) | 332 (1.2%) | 1 (1.0%) | 0.999 |
| Sepsis, n (%) | 2298 (8.4%) | 2278 (8.4%) | 20 (19.8%) | <0.001 |
| Septic shock, n (%) | 898 (3.3%) | 896 (3.3%) | 2 (2.0%) | 0.776 |
| Pneumonia, n (%) | 1156 (4.2%) | 1152 (4.2%) | 4 (4.0%) | 0.999 |
| Re-intubation, n (%) | 1092 (4.0%) | 1081 (4.0%) | 11 (10.9%) | 0.001 |
| Prolonged (> 48 h) vent dependence, n (%) | 1195 (4.4%) | 1174 (4.3%) | 21 (20.8%) | <0.001 |
| Pulmonary embolism, n (%) | 317 (1.2%) | 317 (1.2%) | 0 | 0.635 |
| Acute kidney injury, n (%) | 237 (0.9%) | 237 (0.9%) | 0 | 0.999 |
| Urinary tract infection, n (%) | 1223 (4.5%) | 1219 (4.5%) | 4 (4.0%) | 0.999 |
| Stroke, n (%) | 90 (0.3%) | 89 (0.3%) | 1 (1.0%) | 0.283 |
| Coma, n (%) | 44 (0.2%) | 41 (0.2%) | 3 (3.0%) | 0.001 |
| Cardiac arrest, n (%) | 281 (1.0%) | 277 (1.0%) | 4 (4.0%) | 0.020 |
| Myocardial infarction, n (%) | 152 (0.6%) | 150 (0.5%) | 2 (2.0%) | 0.109 |
| Postoperative bleeding, n (%) | 3028 (11.1%) | 3008 (11.0%) | 20 (19.8%) | 0.008 |
| Deep venous thrombosis, n (%) | 577 (2.1%) | 573 (2.1%) | 4 (4.0%) | 0.165 |
Table 4.
Adjusted odds of outcomes in patients submitted to major hepatopancreatobiliary procedures stratified by end-stage renal disease (ESRD) after propensity matching and multivariable logistic regression
| Complication | Odds ratio (95% CI) | P-value |
|---|---|---|
| Mortality within 30 days | 0.97 (0.26–3.6) | 0.959 |
| Early return to the OR | 1.17 (0.43–3.19) | 0.753 |
| Major complication | 0.98 (0.53–1.81) | 0.949 |
| Any complication | 0.97 (0.52–1.79) | 0.914 |
| Respiratory complicationa | 1.77 (0.83–3.78) | 0.142 |
| Surgical site infection | 0.94 (0.43–2.09) | 0.887 |
| Sepsis or septic shock | 2.98 (1.25–7.14) | 0.014 |
Respiratory complications defined as one or more of the following postoperative complications: pneumonia; re-intubation, and prolonged (>48 h) ventilation.
95% CI, 95% confidence interval; OR, operating room.
In a sensitivity analysis of the subgroup of patients undergoing pancreaticoduodenectomy (Fig. 2), ESRD was found to impose two-fold higher odds for the occurrence of a major complication (AOR 2.35, P = 0.025), as well as for a respiratory complication (AOR 2.78, P = 0.011) and sepsis (AOR 2.55, P = 0.018). In the subgroup of patients submitted to right hepatectomy or trisegmentectomy, ESRD imposed no significantly increased odds for any primary or secondary endpoints after adjustment (Fig. 3).
Figure 2.
Adjusted odds for the occurrence of complications in the subgroup of patients with end-stage renal disease submitted to pancreaticoduodenectomy (Whipple technique). OR, operating room
Figure 3.
Adjusted odds for the occurrence of complications in the subgroup of patients with end-stage renal disease submitted to major hepatectomy (right hepatectomy or trisegmentectomy). OR, operating room
Discussion
Very few patients with ESRD undergo complex abdominal surgery, particularly hepatic or pancreatic resection. The reasons for this are likely to be multifactorial, but a major factor is the common perception that procedures of this magnitude carry a prohibitively high risk in ESRD patients.
A few recent studies have examined the effects of ESRD in a wide spectrum of elective general and vascular surgical procedures using large national clinical databases.3,10,11 These studies have demonstrated significantly increased risks for perioperative mortality and morbidity in the ESRD population. Gajdos et al. included hepatic and pancreatic procedures in an analysis among several other general surgery operations, but did not characterize specific outcomes in this population.3
Several factors may contribute to the increased operative risk observed in this population. Patients with ESRD frequently have other significant comorbidities, including cardiovascular disease and diabetes, which make them susceptible to cardiovascular complications such as myocardial infarction and stroke in the perioperative period. Moreover, these patients commonly demonstrate coagulation abnormalities related to uraemic platelet dysfunction and are thus susceptible to bleeding complications.2,12 In addition, patients with ESRD have impaired immune function, characterized by the induction of a proinflammatory state, diminished number and function of lymphoid cells, and phagocyte dysfunction.1,13,14 This has significant implications for the development of surgical site infection and sepsis. Finally, ESRD patients are also at increased risk for pulmonary complications, possibly related to hypoxaemia and hypocapnia induced by heart disease, contributing to atelectasis and pneumonia.3 Beyond these significant concerns, there are additional considerations in complex hepatic or pancreatic resection. These procedures are characterized by large fluid shifts both intraoperatively and in the early postoperative period. Patients with ESRD are less able to accommodate such large fluid shifts, which can potentially lead to complications related to fluid overload.
In light of these numerous risk factors, the primary focus of the present study was to characterize perioperative outcomes in patients with ESRD submitted to complex hepatopancreaticobiliary procedures. Most of the previous literature on this topic consists of relatively small single-institution series, which have generally demonstrated the feasibility of performing safe hepatic or pancreatic resection in highly selected patients with renal dysfunction and ESRD.15–25 More recently, Squires and colleagues examined the effects of preoperative renal insufficiency on outcomes in a large series of patients undergoing pancreatic resection.26 In this study, renal insufficiency was not associated with increased postoperative mortality, but was an independent predictor of postoperative complications and respiratory failure by multivariate analysis.26
The findings of the present study support the suggestion that hepatic and pancreatic resections are not prohibitively high risk in selected patients with ESRD, but are frequently associated with a more difficult postoperative course. Similar rates of perioperative mortality were observed in ESRD patients and non-ESRD patients, but the former group showed a significantly increased complication rate. After propensity matching to isolate the independent effect of ESRD, followed by multivariate analysis, the present analysis failed to find significantly increased odds of mortality, major complications or reoperation in ESRD patients. Of note, the analysis did find that ESRD patients had a three-fold increased risk for the development of sepsis, which is consistent with findings highlighted by Gajdos et al.3 and Drolet et al.,10 and perhaps reflects the abnormalities of immune function in this population.
In a specific examination of the subset of patients with ESRD submitted to pancreaticoduodenectomy, significantly increased odds for the occurrence of major complications, respiratory complications and sepsis in the postoperative period emerged. This may reflect the reduced capacity of ESRD patients to tolerate these lengthy, complex operations with their associated physiologic stress and significant potential for the occurrence of septic complications in the postoperative period. Interestingly, no such trend emerged among patients submitted to major liver resection and an adjusted analysis of this group showed no increased risk for any of the study parameters. These findings should be interpreted with caution because the numbers of patients in each of these subsets are limited, but they do give some indication that patients with ESRD who undergo pancreaticoduodenectomy are likely to experience a more difficult postoperative course with greater potential for septic complications.
There are some important limitations of this study to consider. Firstly, it is a retrospective study and is thus subject to the inherent limitations of this type of analysis. Secondly, the patients in this study with ESRD who underwent hepatic or pancreatic resection represent a highly selected group of patients. However, it was not possible to ascertain the specific selection criteria employed by surgical teams in their decisions to proceed with surgery. It is possible that this population of ESRD patients was healthier than patients in previous studies as a result of the careful preoperative evaluation performed prior to hepatic or pancreatic resection. Thirdly, there are limitations in the data available for analysis in the NSQIP database. Specifically, it was not possible to ascertain the duration of renal failure (acute versus chronic) and the mode of dialysis (peritoneal versus haemodialysis) in patients with ESRD.
Conclusions
Hepatic and pancreatic resections can be safely performed in selected patients with ESRD and these patients should not be categorically excluded from consideration for these complex procedures. It is important to clearly outline and describe the increase in complications these patients might expect when counselling patients and family in the preoperative setting. Further research is necessary to investigate which patients with ESRD best tolerate hepatic or pancreatic resection, and to identify risk factors that can be modified in order to optimize outcomes.
Conflicts of interest
None declared.
References
- Betjes MG. Immune cell dysfunction and inflammation in end-stage renal disease. Nat Rev Nephrol. 2013;9:255–265. doi: 10.1038/nrneph.2013.44. [DOI] [PubMed] [Google Scholar]
- Jalal DI, Chonchol M, Targher G. Disorders of haemostasis associated with chronic kidney disease. Semin Thromb Hemost. 2010;36:34–40. doi: 10.1055/s-0030-1248722. [DOI] [PubMed] [Google Scholar]
- Gajdos C, Hawn MT, Kile D, Robinson TN, Henderson WG. Risk of major nonemergent inpatient general surgical procedures in patients on long-term dialysis. JAMA Surg. 2013;148:137–143. doi: 10.1001/2013.jamasurg.347. [DOI] [PubMed] [Google Scholar]
- Arbogast P, Seeger JD DEcIDE Methods Center Summary Variable Working Group. 2012. Summary variables in observational research: propensity scores and disease risk scores. Effective Health Care Program Research Report No 33. ( ) Available at http://effectivehealthcare.ahrq.gov/reports/final.cfm (last accessed September 2013)
- Brookhart MA, Schneeweiss S, Rothman KJ, Glynn RJ, Avorn J, Sturmer T. Variable selection for propensity score models. Am J Epidemiol. 2006;163:1149–1156. doi: 10.1093/aje/kwj149. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pearl J. Causal inference in statistics: an overview. Stat Surv. 2009;3:96–146. [Google Scholar]
- Khuri SF, Daley J, Henderson W, Hur K, Gibbs JO, Barbour G, et al. Risk adjustment of the postoperative mortality rate for the comparative assessment of the quality of surgical care: results of the National Veterans Affairs Surgical Risk Study. J Am Coll Surg. 1997;185:315–327. [PubMed] [Google Scholar]
- Cohen ME, Dimick JB, Bilimoria KY, Ko CY, Richards K, Hall BL. Risk adjustment in the American College of Surgeons National Surgical Quality Improvement Program: a comparison of logistic versus hierarchical modelling. J Am Coll Surg. 2009;209:687–693. doi: 10.1016/j.jamcollsurg.2009.08.020. [DOI] [PubMed] [Google Scholar]
- Dimick JB, Osborne NH, Hall BL, Ko CY, Birkmeyer JD. Risk adjustment for comparing hospital quality with surgery: how many variables are needed? J Am Coll Surg. 2010;210:503–508. doi: 10.1016/j.jamcollsurg.2010.01.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Drolet S, Maclean AR, Myers RP, Shaheen AA, Dixon E, Donald Buie W. Morbidity and mortality following colorectal surgery in patients with end-stage renal failure: a population-based study. Dis Colon Rectum. 2010;53:1508–1516. doi: 10.1007/DCR.0b013e3181e8fc8e. [DOI] [PubMed] [Google Scholar]
- Gajdos C, Hawn MT, Kile D, Henderson WG, Robinson T, McCarter M, et al. The risk of major elective vascular surgical procedures in patients with end-stage renal disease. Ann Surg. 2013;257:766–773. doi: 10.1097/SLA.0b013e3182686b87. [DOI] [PubMed] [Google Scholar]
- Lutz J, Menke J, Sollinger D, Schinzel H, Thurmel K. Haemostasis in chronic kidney disease. Nephrol Dial Transplant. 2014;29:29–40. doi: 10.1093/ndt/gft209. [DOI] [PubMed] [Google Scholar]
- Kurts C, Panzer U, Anders HJ, Rees AJ. The immune system and kidney disease: basic concepts and clinical implications. Nat Rev Immunol. 2013;13:738–753. doi: 10.1038/nri3523. [DOI] [PubMed] [Google Scholar]
- Vaziri ND, Pahl MV, Crum A, Norris K. Effect of uraemia on structure and function of immune system. J Ren Nutr. 2012;22:149–156. doi: 10.1053/j.jrn.2011.10.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Nishizaki T, Kanematsu T, Matsumata T, Yasunaga C, Takenaka K, Onoyama K, et al. Maintenance haemodialysis poses few limitations for liver surgery. Int Surg. 1989;74:101–103. [PubMed] [Google Scholar]
- Yeh CN, Lee WC, Chen MF. Hepatic resection for hepatocellular carcinoma in end-stage renal disease patients: two decades of experience at Chang Gung Memorial Hospital. World J Gastroenterol. 2005;11:2067–2071. doi: 10.3748/wjg.v11.i14.20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cheng SB, Wu CC, Shu KH, Ho WL, Chen JT, Yeh DC, et al. Liver resection for hepatocellular carcinoma in patients with end-stage renal failure. J Surg Oncol. 2001;78:241–246. doi: 10.1002/jso.1160. ; discussion 246–247. [DOI] [PubMed] [Google Scholar]
- Cheng SB, Yeh DC, Shu KH, Wu CC, Wen MC, Liu TJ, et al. Liver resection for hepatocellular carcinoma in patients who have undergone prior renal transplantation. J Surg Oncol. 2006;93:273–278. doi: 10.1002/jso.20465. [DOI] [PubMed] [Google Scholar]
- Kaibori M, Matsui Y, Kwon AH, Tokoro T, Kamiyama Y. Prognosis of hepatocellular carcinoma after hepatectomy in patients with renal dysfunction. World J Surg. 2005;29:375–381. doi: 10.1007/s00268-004-7515-0. [DOI] [PubMed] [Google Scholar]
- Lui SL, Yip PS, Lam MF, Lo WK. Feasibility of reinstitution of CAPD after partial hepatectomy in patients with malignant hepatic tumours. Perit Dial Int. 2003;23:504–506. [PubMed] [Google Scholar]
- Orii T, Takayama T, Haga I, Fukumori T, Amada N. Efficacy of a liver resection for hepatocellular carcinoma in patients with chronic renal failure. Surg Today. 2008;38:329–334. doi: 10.1007/s00595-007-3634-1. [DOI] [PubMed] [Google Scholar]
- Toshima T, Shirabe K, Yoshiya S, Muto J, Ikegami T, Yoshizumi T, et al. Outcome of hepatectomy for hepatocellular carcinoma in patients with renal dysfunction. HPB. 2012;14:317–324. doi: 10.1111/j.1477-2574.2012.00452.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Uchida H, Shibata K, Kai S, Iwaki K, Ohta M, Kitano S. Pylorus-preserving pancreaticoduodenectomy in patients undergoing chronic haemodialysis. Surg Today. 2008;38:1152–1154. doi: 10.1007/s00595-008-3746-2. [DOI] [PubMed] [Google Scholar]
- Yamagata M, Kanematsu T, Matsumata T, Nishizaki T, Utsunomiya T, Sugimachi K, et al. Possibility of hepatic resection in patients on maintenance haemodialysis. Hepatogastroenterology. 1993;40:249–252. [PubMed] [Google Scholar]
- Yasuda K, Tahara K, Kume M, Tsutsui S, Higashi H, Kitano S. Risk factors for morbidity and mortality following abdominal surgery in patients on maintenance haemodialysis. Hepatogastroenterology. 2007;54:2282–2284. [PubMed] [Google Scholar]
- Squires MH, 3rd, Mehta VV, Fisher SB, Lad NL, Kooby DA, Sarmiento JM, et al. Effect of preoperative renal insufficiency on postoperative outcomes after pancreatic resection: a single institution experience of 1061 consecutive patients. J Am Coll Surg. 2014;218:92–101. doi: 10.1016/j.jamcollsurg.2013.09.012. [DOI] [PMC free article] [PubMed] [Google Scholar]