- Email: [email protected]
The Impact of Pelvic Venous Pressure on Blood Loss during Open Radical Cystectomy and Urinary Diversion: Results of a Secondary Analysis of a Randomized Clinical Trial
Trauma/Reconstruction/Diversion
The Impact of Pelvic Venous Pressure on Blood Loss during Open Radical Cystectomy and Urinary Diversion: Results of a Secondary Analysis of a Randomized Clinical Trial Patrick Y. Wuethrich,* Fiona C. Burkhard, George N. Thalmann, Frank Stueber and Urs E. Studer From the Departments of Urology (FCB, GNT, UES), and Anesthesiology and Pain Medicine (PYW, FS), University Hospital Bern, Bern, Switzerland
Abbreviations and Acronyms CVP ¼ central venous pressure MAP ¼ mean arterial pressure PLND ¼ pelvic lymph node dissection PVP ¼ pelvic venous pressure RC ¼ radical cystectomy UD ¼ urinary diversion Accepted for publication December 31, 2014. Study received Kantonale Ethikkommission Bern approval. Supported by the Else Kr€oner Fresenius Foundation (2010_A80), Bad Homburg, Germany, and institutional research funds of Department of Anesthesiology and Pain Medicine, and Department of Urology, University Hospital Bern, Bern, Switzerland. * Correspondence: Department of Anaesthesiology and Pain Medicine, University Hospital Bern, CH-3010 Bern, Switzerland (telephone: þ41316322725; FAX: þ416320554; e-mail: [email protected]).
Purpose: Blood loss and blood substitution are associated with higher morbidity after major abdominal surgery. During major liver resection low local venous pressure decreases blood loss. Ambiguity persists concerning the impact of local venous pressure on blood loss during open radical cystectomy. We determined the association between intraoperative blood loss and pelvic venous pressure as well as factors affecting pelvic venous pressure. Materials and Methods: In this single center, double-blind, randomized trial pelvic venous pressure was measured in 82 patients in a norepinephrine-low volume group and in 81 controls with liberal hydration. As secondary analysis patients from each arm were stratified into subgroups with pelvic venous pressure less than 5 mm Hg, or 5 or greater as measured after cystectomy, which is the optimal cutoff for identifying patients with relevant blood loss according to the Youden index. Results: Median blood loss was 800 ml (range 300 to 1,600) in 55 of 163 patients (34%) with pelvic venous pressure less than 5 mm Hg and 1,200 ml (range 400 to 3,000) in 108 of 163 (66%) with pelvic venous pressure 5 mm Hg or greater (p <0.0001). Pelvic venous pressure less than 5 mm Hg was measured in 42 of 82 patients (51%) in the norepinephrine-low volume group and in 13 of 81 controls (16%) (p <0.0001). Pelvic venous pressure decreased significantly after removing abdominal packing and abdominal lifting in each group at all time points, that is at the beginning and end of pelvic lymph node dissection, and the end of cystectomy (p <0.0001). No correlation was detected between pelvic venous pressure and central venous pressure. Conclusions: Blood loss was significantly decreased in patients with low pelvic venous pressure. Factors affecting pelvic venous pressure were fluid management and abdominal packing. Key Words: urinary bladder; cystectomy; venous pressure; blood loss, surgical; norepinephrine
DURING open RC combined with PLND and UD bleeding from the local pelvic venous plexus and capillaries surrounding the bladder and prostate may be substantial despite meticulous surgery. Strategies to
146
j
www.jurology.com
decrease blood loss must be developed to reduce the incidence of blood transfusion and improve the postoperative outcome. In the frame of a prospective, randomized trial we recently found that
0022-5347/15/1941-0146/0 THE JOURNAL OF UROLOGY® © 2015 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC.
http://dx.doi.org/10.1016/j.juro.2014.12.094 Vol. 194, 146-152, July 2015 Printed in U.S.A.
PELVIC VENOUS PRESSURE AND BLOOD LOSS DURING CYSTECTOMY AND URINARY DIVERSION
preemptive norepinephrine administered intraoperatively combined with restrictive deferred hydration significantly decreased postoperative complications, blood loss and the need for blood transfusion.1,2 It is not known whether this was due to decreased local venous pressure or to other factors such as vasoconstriction or a minimal change in the concentration of coagulation factors. Local venous pressure in the surgical field may influence blood loss. Low (less than 5 mm Hg) CVP is considered a simple, potent method to minimize intraoperative blood loss during liver resection surgery.3e5 Therefore, we investigated whether blood loss also correlates with PVP (a surrogate for local venous pressure around the bladder and prostate) during open RC combined with PLND and UD. We applied 2 intraoperative fluid management strategies, including 1) preemptive norepinephrine infusion combined with restrictive deferred hydration and 2) more liberal hydration without norepinephrine. We also analyzed the impact of abdominal packing on PVP.
METHODS Trial Design and Participants This randomized, double-blind, single center study was approved by the local ethics committee (Kantonale Ethikkommission Bern) and registered at Clinicaltrials. gov (NCT01276665). Patients with an ASAÒ (American Society of AnesthesiologistsÒ) score of II to III were included in analysis. Exclusion criteria were coagulopathy, significant hepatic and renal dysfunction defined as K/DOQI (Kidney Disease Outcomes Quality Initiative) stage 3 or greater, congestive heart failure and a contraindication to thoracic epidural analgesia. All patients provided written informed consent. A total of 167 consecutive patients were scheduled for PLND, open RC and UD between November 2009 and September 2012. Patients were prospectively randomized into 2 groups, including 1) preemptive administration of norepinephrine combined with a restrictive deferred crystalloid fluid regimen (norepinephrine-low volume group) and 2) more liberal intravenous hydration (controls). In this analysis we evaluated only patients with available measurements of intraoperative PVP (fig. 1).
Initial Randomization and Blinding The protocol and design of the original study were described previously.1 Briefly, patients were randomly allocated to the norepinephrine-low volume group or to the control group by computer generated permuted block randomization with 1:1 allocation. Surgeons, patients and data assessors were blinded to the assigned fluid regimen. Crystalloid bags and perfusion pumps were hidden behind a sterile curtain during surgery. Assessors of postoperative data were blinded to patient anesthesiology patient data.
147
Management Standard monitoring included continuous electrocardiogram, heart rate, core temperature, peripheral capillary oxygen saturation, invasive MAP and CVP. Before anesthesia was induced an epidural catheter was placed at the T9-T10 level and activated with bupivacaine 0.25% at a rate of 8 ml per hour until the end of PLND. Anesthesia was induced with propofol, fentanyl and rocuronium, and maintained with isoflurane. Ventilation with an inspired oxygen fraction of 60% was mechanically controlled to maintain normocapnia. Normothermia was maintained by an air warming system and a fluid warmer. In the norepinephrine-low volume group a preemptive norepinephrine infusion was started at 2 mg/kg per hour after anesthesia induction. Balanced crystalloid RingerfundinÒ solution was administered at 1 ml/kg per hour until the bladder was removed, followed by infusion of 3 ml/kg per hour of crystalloid solution until the end of surgery while maintaining norepinephrine infusion. In case of hypotension (ie MAP less than 60 mm Hg) norepinephrine infusion was titrated accordingly after an initial 10 mg bolus. In controls a 6 ml/kg bolus of crystalloid solution was administered during anesthesia induction, followed by balanced crystalloid solution at 6 ml/kg per hour intraoperatively. In this group hypotension episodes were treated with boluses of 250 ml crystalloid solution. In all patients blood units were only transfused when hemoglobin was less than 8.0 gm/dl.6 Patients were in the 30-degree head down position during the entire operative time. The surgical technique was standardized.7,8 At least 1 of 3 senior urologists (FCB, GNT and UES) was present during surgery.
Data Collection and Outcome Measures Blood loss was assessed by anesthesiologists based on the amount of blood in the suction device and the blood volume absorbed by surgical gauze, which was weighed. Blood loss was assessed separately during the PLND, open RC and UD periods.2 The number of patients who received blood unit transfusions during surgery was documented. During surgery the heart rate, invasive MAP and CVP were continuously documented and mean values were calculated. PVP was assessed 4 times invasively with a 20 gauge needle placed in the external iliac vein adjacent to the iliac bifurcation. The pressure transducers were calibrated and zeroed separately and leveled to the external iliac vein. PVP measurements were recorded after preparation of the iliac bifurcation (ie the beginning of PLND), at the end of PLND, after bladder removal (ie the end of cystectomy) and finally before abdominal wall closure (ie the end of UD). Except before abdominal wall closure PVP was measured twice at these defined time points. The first measurement was made with the gauze in the abdominal cavity to remove the bowel from the surgical field (ie abdominal packing). The second measurement was made after removing the gauze with the abdominal wall lifted to analyze the impact of gauze/packing on venous return. Mean PVP was also calculated. Patients were analyzed in subgroups according to whether PVP at the end of the cystectomy part was
148
PELVIC VENOUS PRESSURE AND BLOOD LOSS DURING CYSTECTOMY AND URINARY DIVERSION
Figure 1. Modified CONSORT flowchart
associated with clinically relevant blood loss (ie the need for blood unit transfusion).
Statistical Analysis Data were analyzed using nonparametric statistical models. Data are shown as the median and range or the number and percent. Categorical data were compared with the chi-square test. RR and the 95% CI were also calculated as appropriate. Continuous data were compared using the 2-tailed Mann-Whitney U and Kruskal-Wallis H tests followed by the Bonferroni correction for multiple comparisons. In each group the within patient PVP values with and without gauze were analyzed by the Wilcoxon signed rank test. The Spearman correlation test was applied to test the correlation of blood loss with MAP, CVP and PVP at the different time points. Test results were considered significant at p <0.05. ROC curves were constructed to identify the optimal cutoff values for the association of blood loss with the need for blood unit transfusion, defined as clinically relevant blood loss, and then between mean PVP and clinically relevant blood loss. The optimal cutoff was defined as the
value associated with the highest sum of sensitivity and specificity (the Youden index). Multiple logistic regression analyses using forward selection were applied to identify independent risk factors for relevant blood loss and are reported as the adjusted OR with the 95% CI. Factors were included at p <0.10. The confounders considered were mean MAP, CVP and PVP. No interaction terms were included due to sample size. Model fit and predictive power were assessed by the Hosmer-Lemeshow goodness of fit test and the ROC-AUC.
RESULTS PVP was assessed in 82 patients in the norepinephrine-low volume group and in 81 controls (fig. 1). Baseline data were equally distributed between the subgroups in the groups (supplementary table, http://jurology.com/). The mean PVP ROC-AUC was 0.753 (95% CI 0.659e0.847) according to blood loss requiring blood unit transfusion. The optimal cutoff of mean PVP
PELVIC VENOUS PRESSURE AND BLOOD LOSS DURING CYSTECTOMY AND URINARY DIVERSION
was 5.1 mm Hg, which had 84% sensitivity and 64% specificity to discriminate patients with relevant blood loss. On multiple logistic regression analysis PVP (less than 5 vs 5 mm Hg or greater) was the only independent predictor of clinically relevant blood loss (OR 0.373, 95% CI 0.165e0.842, p ¼ 0.018). MAP (OR 1.054, 95% CI 0.984e1.130, p ¼ 0.130) and CVP (OR 1.094, 95% CI 0.974e1.230, p ¼ 0.115) did not significantly correlate. The model accorded well with the Hosmer-Lemeshow test (p ¼ 0.801) and the ROC-AUC was 0.701. Independent of the groups median blood loss was 800 ml (range 300 to 1,600) in 55 of 163 patients (34%) with PVP less than 5 mm Hg and 1,200 ml (range 400 to 2,800) in 108 of 163 (66%) with PVP 5 mm Hg or greater (p <0.0001). Independent of the groups PVP significantly correlated with blood loss at the end of PLND (r2 ¼ 0.234, 95% CI 0.096e0.397, p ¼ 0.004). At the end of the cystectomy part CVP and PVP significantly correlated with blood loss during that period (r2 ¼ 0.241, 95% CI 0.085e0.384, p ¼ 0.006 and r2 ¼ 0.344, 95% CI 0.155e0.469, p <0.0001, respectively). At the end of the UD period PVP significantly correlated with blood loss during that time (r2 ¼ 0.190, 95% CI 0.022e0.371, p ¼ 0.022, fig. 2). CVP and PVP did not significantly correlate except for a weak correlation at the end of cystectomy (r2 ¼ 0.134, 95% CI 0.019e0.259, p ¼ 0.025). In the norepinephrine-low volume group PVP significantly correlated with blood loss at the end of the cystectomy part and at the end of the UD part (r2 ¼ 0.244, 95% CI 0.021e0.461, p ¼ 0.032 and r2 ¼ 0.269, 95% CI 0.024e0.495, p ¼ 0.019, respectively). In controls no significant correlation was found between PVP and blood loss during the different periods. PVP was significantly lower in the norepinephrine-low volume group during all periods assessed compared to controls (table 1). PVP was less than 5 mm Hg at the end of the cystectomy part in 42 of 82 patients (51%) in the norepinephrine-low volume group and in 13 of 81 controls (16%) (RR 0.58, 95% CI 0.46e0.74, p <0.0001, table 2). PVP less than 5 mm Hg at the end of the cystectomy part was associated with significantly less intraoperative blood loss in the 2 groups. In the norepinephrinelow volume group median total blood loss in the subgroup with PVP less than 5 mm Hg was 800 ml (range 300 to 1,200) compared to 900 ml (range 500 to 1,800) in the subgroup with PVP 5 mm Hg or greater (p ¼ 0.004, table 2). In controls the subgroup with PVP less than 5 mm Hg had a median blood loss of 1,000 ml (range 600 to 1,600) compared to 1,300 ml (range 400 to 3,000) in the subgroup with PVP 5 mm Hg or greater (p ¼ 0.006, table 2 and fig. 3).
149
Figure 2. Fit lines with 95% CI show PVP by blood loss during PLND (A), cystectomy (B) and UD (C ) (r2 ¼ 0.234, 0.344 and 0.190, respectively). Sphere size indicates number of patients.
150
PELVIC VENOUS PRESSURE AND BLOOD LOSS DURING CYSTECTOMY AND URINARY DIVERSION
Table 1. PVP by randomized norepinephrine-low volume and control groups Median mm Hg PVP (range) Overall
Norepinephrine
No. pts 183 82 PLND: Start 7 (0e16) 5 (0e13)* Start þ abdominal 2 (0e12) wall lifting End 6 (0e15) 4 (0e12)* End þ abdominal 2 (e1e4) wall lifting RC: End 6 (e1e14) 4 (e1e12)* Abdominal wall 1 (2e6) lifting After UD 5 (e1e13) 2 (e1e8)
Control
13 patients (15%) had PVP less than 5 mm Hg vs 24 of 68 (35%) with PVP 5 mm Hg or greater (p ¼ 0.206).
p Value e
81 8 (0e16)* 3 (0e12)
<0.0001 0.007
8 (0e15)* 3 (e2e9)
<0.0001 <0.0001
7 (0e14)* <0.0001 3 (0e8) < 0.0001 6 (0e13)
<0.0001
* Compared to abdominal wall lifting Wilcoxon signed rank test with Bonferroni adjustment p <0.0001
PVP decreased significantly after removing the gauze from the abdominal cavity and lifting the abdominal wall in each group at any time point measured (p <0.0001, table 1). There was a significant difference in the rate of blood unit transfusion between patients with PVP less than 5 mm Hg and patients with PVP 5 mm Hg or greater (5 of 55 or 9% vs 28 of 108 or 26%, RR 0.72, 95% CI 0.10 to 0.79, p ¼ 0.013). However, we found no difference between PVP subgroups in either randomized group. In the norepinephrine-low volume group 3 of 42 patients (7%) had PVP less than 5 mm Hg vs 4 of 40 (10%) with PVP 5 mm Hg or greater (p ¼ 0.709). In the control group 2 of
DISCUSSION Total intraoperative blood loss was significantly less in patients with low PVP (less than 5 mm Hg) compared to that in patients with PVP 5 mm Hg or greater. Significantly more patients with preemptive infusion of norepinephrine combined with restrictive deferred hydration had PVP less than 5 mm Hg. Even in this group increased PVP was associated with increased blood loss. This suggests that fluid management can influence PVP and consequently impact blood loss. Blood loss during open RC with UD impacts the postoperative outcome and increases postoperative morbidity.9e12 Blood unit transfusion after excessive blood loss carries the risk of transfusion associated lung injury, postoperative transfusion associated fluid overload and an increased wound infection rate, and it was postulated to promote cancer recurrence.13e15 Consequently intraoperative strategies to decrease blood loss may have vital consequences. Little is known about the impact of venous pressure at the surgical site on blood loss. Substantial literature has already demonstrated that a significant reduction in blood loss could be achieved by maintaining CVP at lower than 5 mm Hg in patients undergoing liver dissection, thus, making it easier
Table 2. Blood loss, hemodynamic data and intraoperative blood unit transfusion rate in 82 patients in norepinephrine-low volume group and 81 controls by PVP subgroup Median Norepinephrine mm Hg PVP (range) Less Than 5 No. pts Blood loss (ml): Per min PLND RC UD Heart rate (bpm): PLND RC UD MAP (mm Hg): PLND RC UD CVP (mm Hg): Baseline PLND RC UD Crystalloid (ml) *p †p ‡p §p
42 800 (300e1,200)*,† 1.8 (0.7e2.9)*,† 100 (0e400)* 480 (200e850)*,† 100 (0e410)
Median Control mm Hg PVP (range)
5 or Greater
Less Than 5
5 or Greater
p Value
40 900 (500e1,800)‡ 2.0 (1.2e3.5)‡ 145 (0e300)‡ 550 (300e1,200)‡ 200 (0e500)
13 1,000 (600e1,600) 2.7 (1.5e3.7) 150 (50e300) 650 (500e1,200) 150 (50e500)
68 1,300 (400e2,800) 3.2 (2.9e3.5) 200 (50e2,000) 900 (300e2,300) 200 (0e700)
<0.0001 0.005 <0.0001 <0.0001 0.049
66 70 74
(50e94) (47e102) (50e107)
66 67 77
(45e92) (43e100) (41e100)
60 65 75
(52e85) (44e78) (55e90)
65 63 70
(45e90) (40e90) (40e98)
0.167 0.123 0.210
70 70 72
(60e85) (60e85) (62e94)
70 71 72
(60e93) (60e93) (60e100)
70 73 70
(60e84) (60e81) (60e85)
68 70 70
(60e90) (60e100) (60e92)
0.577 0.438 0.684
12 (7e20) 12 (7e19) 11 (4e23) 9 (3e22) 1,500 (700e2,500)*,†
<0.0001 vs control PVP 5 mm Hg or greater. <0.05 vs control PVP less than 5 mm Hg. <0.0001 vs control PVP 5 mm Hg or greater. <0.0001 vs control PVP less than 5 mm Hg.
13 (6e21) 13 (5e21) 12 (2e23) 11 (4e19) 1,800 (800e4,000)‡,§
14 (5e20) 15 (10e20) 16 (10e20) 15 (9e20) 3,800 (3,000e6,100)
13 (5e21) 15 (5e23) 15 (2e25) 14 (3e29) 4,450 (2,800e6,200)
0.132 <0.0001 <0.0001 <0.0001 <0.0001
PELVIC VENOUS PRESSURE AND BLOOD LOSS DURING CYSTECTOMY AND URINARY DIVERSION
Figure 3. Total blood loss in norepinephrine-low volume (A) and control (B) groups by PVP.
to control vascular injury.3,4 In this study we found that preemptive norepinephrine use with restrictive deferred fluid management led to a significant reduction in PVP and blood loss compared to a more liberal fluid regimen. The question remains of whether the decreased blood loss was attributable to reduced intravenous fluid administration and to vasoconstriction induced by norepinephrine, resulting in lower PVP. The correlation between blood loss and PVP independent of the amount of fluid administered emphasizes that PVP significantly impacts blood loss. Even in the control group, in which patients received more generous fluid administration, PVP less than 5 mm Hg at the end of cystectomy was associated with a significant decrease in blood loss. An additional important observation was that lifting the abdominal wall significantly decreased
151
PVP. This suggests that avoiding abdominal packing, for example using a table fixed, self-retaining retractor to lift the abdominal wall, may help reduce intraoperative PVP. The potential hemodynamic impact of an abdominal wall-lifting retractor compared to conventional abdominal packing in patients undergoing open abdominal surgery is less well studied than in laparoscopic surgery. In laparoscopic surgery an abdominal wall lifting device had a positive effect on blood loss and CVP in animal models and in patients undergoing cholecystectomy.16,17 The abdominal wall lifting device resulted in an increased cardiac index compared to carbon dioxide insufflation without affecting systemic vascular resistance in healthy patients undergoing laparoscopic cholecystectomy and was associated with fewer side effects, such as postoperative nausea and vomiting, and right shoulder pain. For these reasons this approach was recommended in patients at high risk.17e20 Another important observation is the lack of correlation between PVP and CVP except for a weak correlation at the end of cystectomy. Thus, standard CVP assessment may not be a reliable surrogate for PVP, which is a factor that must be considered. In some of our patients negative PVPs were measured, which could potentially indicate a risk of air emboli. However, the 30-degree Trendelenburg position and pelvic packing with wet gauze after bladder removal should limit the risk of air emboli. In addition, no decrease was detected in end tidal carbon dioxide, making clinically relevant air emboli unlikely. The main limitation of this study is the lack of a specific power analysis concerning the reduction in blood loss because this study was a secondary analysis. In addition, PVP was only measured at certain time points and not continuously. However, continuous monitoring of PVP would imply catheterization of the veins, resulting in a higher risk of thrombosis.
CONCLUSIONS Our results suggest that low PVP during PLND, RC and UD significantly decreases intraoperative blood loss. Techniques to decrease PVP, such as a norepinephrine-low volume regimen and avoidance of vena caval compression (eg by intra-abdominal packing), should be established for open RC and UD.
REFERENCES 1. Wuethrich PY, Burkhard FC, Thalmann GN et al: Restrictive deferred hydration combined with preemptive norepinephrine infusion during
radical cystectomy reduces postoperative complications and hospitalization time: a randomized clinical trial. Anesthesiology 2014; 120: 365.
2. Wuethrich PY, Studer UE, Thalmann GN et al: Intraoperative continuous norepinephrine infusion combined with restrictive deferred hydration
152
PELVIC VENOUS PRESSURE AND BLOOD LOSS DURING CYSTECTOMY AND URINARY DIVERSION
significantly reduces the need for blood transfusion in patients undergoing open radical cystectomy: results of a prospective randomised trial. Eur Urol 2014; 66: 352.
9. Linder BJ, Frank I, Cheville JC et al: The impact of perioperative blood transfusion on cancer recurrence and survival following radical cystectomy. Eur Urol 2013; 63: 839.
3. Wang WD, Liang LJ, Huang XQ et al: Low central venous pressure reduces blood loss in hepatectomy. World J Gastroenterol 2006; 12: 935.
10. Chang SS, Smith JA Jr, Wells N et al: Estimated blood loss and transfusion requirements of radical cystectomy. J Urol 2001; 166: 2151.
4. Jones RM, Moulton CE and Hardy KJ: Central venous pressure and its effect on blood loss during liver resection. Br J Surg 1998; 85: 1058.
11. Chang SS, Baumgartner RG, Wells N et al: Causes of increased hospital stay after radical cystectomy in a clinical pathway setting. J Urol 2002; 167: 208.
5. Smyrniotis V, Kostopanagiotou G, Theodoraki K et al: The role of central venous pressure and type of vascular control in blood loss during major liver resections. Am J Surg 2004; 187: 398.
12. Chang SS, Cookson MS, Baumgartner RG et al: Analysis of early complications after radical cystectomy: results of a collaborative care pathway. J Urol 2002; 167: 2012.
6. Klein HG, Spahn DR and Carson JL: Red blood cell transfusion in clinical practice. Lancet 2007; 370: 415.
13. Hendrickson JE and Hillyer CD: Noninfectious serious hazards of transfusion. Anesth Analg 2009; 108: 759.
7. Burkhard FC, Kessler TM, Mills R et al: Continent urinary diversion. Crit Rev Oncol Hematol 2006; 57: 255.
14. Amato A and Pescatori M: Perioperative blood transfusions for the recurrence of colorectal cancer. Cochrane Database Syst Rev 2006; CD005033.
8. Perimenis P and Studer UE: Orthotopic continent urinary diversion an ileal low pressure neobladder with an afferent tubular segment: how I do it. Eur J Surg Oncol 2004; 30: 454.
15. Glance LG, Dick AW, Mukamel DB et al: Association between intraoperative blood transfusion and mortality and morbidity in patients
undergoing noncardiac surgery. Anesthesiology 2011; 114: 283. 16. Lindgren L, Koivusalo AM and Kellokumpu I: Conventional pneumoperitoneum compared with abdominal wall lift for laparoscopic cholecystectomy. Br J Anaesth 1995; 75: 567. 17. Andersson L, Lindberg G, Bringman S et al: Pneumoperitoneum versus abdominal wall lift: effects on central haemodynamics and intrathoracic pressure during laparoscopic cholecystectomy. Acta Anaesthesiol Scand 2003; 47: 838. 18. Galizia G, Prizio G, Lieto E et al: Hemodynamic and pulmonary changes during open, carbon dioxide pneumoperitoneum and abdominal walllifting cholecystectomy. A prospective, randomized study. Surg Endosc 2001; 15: 477. 19. Rademaker BM, Meyer DW, Bannenberg JJ et al: Laparoscopy without pneumoperitoneum. Effects of abdominal wall retraction versus carbon dioxide insufflation on hemodynamics and gas exchange in pigs. Surg Endosc 1995; 9: 797. 20. Rademaker BM, Bannenberg JJ, Kalkman CJ et al: Effects of pneumoperitoneum with helium on hemodynamics and oxygen transport: a comparison with carbon dioxide. J Laparoendosc Surg 1995; 5: 15.
The Impact of Pelvic Venous Pressure on Blood Loss during Open Radical Cystectomy and Urinary Diversion: Results of a Secondary Analysis of a Randomized Clinical Trial Patrick Y. Wuethrich,* Fiona C. Burkhard, George N. Thalmann, Frank Stueber and Urs E. Studer From the Departments of Urology (FCB, GNT, UES), and Anesthesiology and Pain Medicine (PYW, FS), University Hospital Bern, Bern, Switzerland
Abbreviations and Acronyms CVP ¼ central venous pressure MAP ¼ mean arterial pressure PLND ¼ pelvic lymph node dissection PVP ¼ pelvic venous pressure RC ¼ radical cystectomy UD ¼ urinary diversion Accepted for publication December 31, 2014. Study received Kantonale Ethikkommission Bern approval. Supported by the Else Kr€oner Fresenius Foundation (2010_A80), Bad Homburg, Germany, and institutional research funds of Department of Anesthesiology and Pain Medicine, and Department of Urology, University Hospital Bern, Bern, Switzerland. * Correspondence: Department of Anaesthesiology and Pain Medicine, University Hospital Bern, CH-3010 Bern, Switzerland (telephone: þ41316322725; FAX: þ416320554; e-mail: [email protected]).
Purpose: Blood loss and blood substitution are associated with higher morbidity after major abdominal surgery. During major liver resection low local venous pressure decreases blood loss. Ambiguity persists concerning the impact of local venous pressure on blood loss during open radical cystectomy. We determined the association between intraoperative blood loss and pelvic venous pressure as well as factors affecting pelvic venous pressure. Materials and Methods: In this single center, double-blind, randomized trial pelvic venous pressure was measured in 82 patients in a norepinephrine-low volume group and in 81 controls with liberal hydration. As secondary analysis patients from each arm were stratified into subgroups with pelvic venous pressure less than 5 mm Hg, or 5 or greater as measured after cystectomy, which is the optimal cutoff for identifying patients with relevant blood loss according to the Youden index. Results: Median blood loss was 800 ml (range 300 to 1,600) in 55 of 163 patients (34%) with pelvic venous pressure less than 5 mm Hg and 1,200 ml (range 400 to 3,000) in 108 of 163 (66%) with pelvic venous pressure 5 mm Hg or greater (p <0.0001). Pelvic venous pressure less than 5 mm Hg was measured in 42 of 82 patients (51%) in the norepinephrine-low volume group and in 13 of 81 controls (16%) (p <0.0001). Pelvic venous pressure decreased significantly after removing abdominal packing and abdominal lifting in each group at all time points, that is at the beginning and end of pelvic lymph node dissection, and the end of cystectomy (p <0.0001). No correlation was detected between pelvic venous pressure and central venous pressure. Conclusions: Blood loss was significantly decreased in patients with low pelvic venous pressure. Factors affecting pelvic venous pressure were fluid management and abdominal packing. Key Words: urinary bladder; cystectomy; venous pressure; blood loss, surgical; norepinephrine
DURING open RC combined with PLND and UD bleeding from the local pelvic venous plexus and capillaries surrounding the bladder and prostate may be substantial despite meticulous surgery. Strategies to
146
j
www.jurology.com
decrease blood loss must be developed to reduce the incidence of blood transfusion and improve the postoperative outcome. In the frame of a prospective, randomized trial we recently found that
0022-5347/15/1941-0146/0 THE JOURNAL OF UROLOGY® © 2015 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC.
http://dx.doi.org/10.1016/j.juro.2014.12.094 Vol. 194, 146-152, July 2015 Printed in U.S.A.
PELVIC VENOUS PRESSURE AND BLOOD LOSS DURING CYSTECTOMY AND URINARY DIVERSION
preemptive norepinephrine administered intraoperatively combined with restrictive deferred hydration significantly decreased postoperative complications, blood loss and the need for blood transfusion.1,2 It is not known whether this was due to decreased local venous pressure or to other factors such as vasoconstriction or a minimal change in the concentration of coagulation factors. Local venous pressure in the surgical field may influence blood loss. Low (less than 5 mm Hg) CVP is considered a simple, potent method to minimize intraoperative blood loss during liver resection surgery.3e5 Therefore, we investigated whether blood loss also correlates with PVP (a surrogate for local venous pressure around the bladder and prostate) during open RC combined with PLND and UD. We applied 2 intraoperative fluid management strategies, including 1) preemptive norepinephrine infusion combined with restrictive deferred hydration and 2) more liberal hydration without norepinephrine. We also analyzed the impact of abdominal packing on PVP.
METHODS Trial Design and Participants This randomized, double-blind, single center study was approved by the local ethics committee (Kantonale Ethikkommission Bern) and registered at Clinicaltrials. gov (NCT01276665). Patients with an ASAÒ (American Society of AnesthesiologistsÒ) score of II to III were included in analysis. Exclusion criteria were coagulopathy, significant hepatic and renal dysfunction defined as K/DOQI (Kidney Disease Outcomes Quality Initiative) stage 3 or greater, congestive heart failure and a contraindication to thoracic epidural analgesia. All patients provided written informed consent. A total of 167 consecutive patients were scheduled for PLND, open RC and UD between November 2009 and September 2012. Patients were prospectively randomized into 2 groups, including 1) preemptive administration of norepinephrine combined with a restrictive deferred crystalloid fluid regimen (norepinephrine-low volume group) and 2) more liberal intravenous hydration (controls). In this analysis we evaluated only patients with available measurements of intraoperative PVP (fig. 1).
Initial Randomization and Blinding The protocol and design of the original study were described previously.1 Briefly, patients were randomly allocated to the norepinephrine-low volume group or to the control group by computer generated permuted block randomization with 1:1 allocation. Surgeons, patients and data assessors were blinded to the assigned fluid regimen. Crystalloid bags and perfusion pumps were hidden behind a sterile curtain during surgery. Assessors of postoperative data were blinded to patient anesthesiology patient data.
147
Management Standard monitoring included continuous electrocardiogram, heart rate, core temperature, peripheral capillary oxygen saturation, invasive MAP and CVP. Before anesthesia was induced an epidural catheter was placed at the T9-T10 level and activated with bupivacaine 0.25% at a rate of 8 ml per hour until the end of PLND. Anesthesia was induced with propofol, fentanyl and rocuronium, and maintained with isoflurane. Ventilation with an inspired oxygen fraction of 60% was mechanically controlled to maintain normocapnia. Normothermia was maintained by an air warming system and a fluid warmer. In the norepinephrine-low volume group a preemptive norepinephrine infusion was started at 2 mg/kg per hour after anesthesia induction. Balanced crystalloid RingerfundinÒ solution was administered at 1 ml/kg per hour until the bladder was removed, followed by infusion of 3 ml/kg per hour of crystalloid solution until the end of surgery while maintaining norepinephrine infusion. In case of hypotension (ie MAP less than 60 mm Hg) norepinephrine infusion was titrated accordingly after an initial 10 mg bolus. In controls a 6 ml/kg bolus of crystalloid solution was administered during anesthesia induction, followed by balanced crystalloid solution at 6 ml/kg per hour intraoperatively. In this group hypotension episodes were treated with boluses of 250 ml crystalloid solution. In all patients blood units were only transfused when hemoglobin was less than 8.0 gm/dl.6 Patients were in the 30-degree head down position during the entire operative time. The surgical technique was standardized.7,8 At least 1 of 3 senior urologists (FCB, GNT and UES) was present during surgery.
Data Collection and Outcome Measures Blood loss was assessed by anesthesiologists based on the amount of blood in the suction device and the blood volume absorbed by surgical gauze, which was weighed. Blood loss was assessed separately during the PLND, open RC and UD periods.2 The number of patients who received blood unit transfusions during surgery was documented. During surgery the heart rate, invasive MAP and CVP were continuously documented and mean values were calculated. PVP was assessed 4 times invasively with a 20 gauge needle placed in the external iliac vein adjacent to the iliac bifurcation. The pressure transducers were calibrated and zeroed separately and leveled to the external iliac vein. PVP measurements were recorded after preparation of the iliac bifurcation (ie the beginning of PLND), at the end of PLND, after bladder removal (ie the end of cystectomy) and finally before abdominal wall closure (ie the end of UD). Except before abdominal wall closure PVP was measured twice at these defined time points. The first measurement was made with the gauze in the abdominal cavity to remove the bowel from the surgical field (ie abdominal packing). The second measurement was made after removing the gauze with the abdominal wall lifted to analyze the impact of gauze/packing on venous return. Mean PVP was also calculated. Patients were analyzed in subgroups according to whether PVP at the end of the cystectomy part was
148
PELVIC VENOUS PRESSURE AND BLOOD LOSS DURING CYSTECTOMY AND URINARY DIVERSION
Figure 1. Modified CONSORT flowchart
associated with clinically relevant blood loss (ie the need for blood unit transfusion).
Statistical Analysis Data were analyzed using nonparametric statistical models. Data are shown as the median and range or the number and percent. Categorical data were compared with the chi-square test. RR and the 95% CI were also calculated as appropriate. Continuous data were compared using the 2-tailed Mann-Whitney U and Kruskal-Wallis H tests followed by the Bonferroni correction for multiple comparisons. In each group the within patient PVP values with and without gauze were analyzed by the Wilcoxon signed rank test. The Spearman correlation test was applied to test the correlation of blood loss with MAP, CVP and PVP at the different time points. Test results were considered significant at p <0.05. ROC curves were constructed to identify the optimal cutoff values for the association of blood loss with the need for blood unit transfusion, defined as clinically relevant blood loss, and then between mean PVP and clinically relevant blood loss. The optimal cutoff was defined as the
value associated with the highest sum of sensitivity and specificity (the Youden index). Multiple logistic regression analyses using forward selection were applied to identify independent risk factors for relevant blood loss and are reported as the adjusted OR with the 95% CI. Factors were included at p <0.10. The confounders considered were mean MAP, CVP and PVP. No interaction terms were included due to sample size. Model fit and predictive power were assessed by the Hosmer-Lemeshow goodness of fit test and the ROC-AUC.
RESULTS PVP was assessed in 82 patients in the norepinephrine-low volume group and in 81 controls (fig. 1). Baseline data were equally distributed between the subgroups in the groups (supplementary table, http://jurology.com/). The mean PVP ROC-AUC was 0.753 (95% CI 0.659e0.847) according to blood loss requiring blood unit transfusion. The optimal cutoff of mean PVP
PELVIC VENOUS PRESSURE AND BLOOD LOSS DURING CYSTECTOMY AND URINARY DIVERSION
was 5.1 mm Hg, which had 84% sensitivity and 64% specificity to discriminate patients with relevant blood loss. On multiple logistic regression analysis PVP (less than 5 vs 5 mm Hg or greater) was the only independent predictor of clinically relevant blood loss (OR 0.373, 95% CI 0.165e0.842, p ¼ 0.018). MAP (OR 1.054, 95% CI 0.984e1.130, p ¼ 0.130) and CVP (OR 1.094, 95% CI 0.974e1.230, p ¼ 0.115) did not significantly correlate. The model accorded well with the Hosmer-Lemeshow test (p ¼ 0.801) and the ROC-AUC was 0.701. Independent of the groups median blood loss was 800 ml (range 300 to 1,600) in 55 of 163 patients (34%) with PVP less than 5 mm Hg and 1,200 ml (range 400 to 2,800) in 108 of 163 (66%) with PVP 5 mm Hg or greater (p <0.0001). Independent of the groups PVP significantly correlated with blood loss at the end of PLND (r2 ¼ 0.234, 95% CI 0.096e0.397, p ¼ 0.004). At the end of the cystectomy part CVP and PVP significantly correlated with blood loss during that period (r2 ¼ 0.241, 95% CI 0.085e0.384, p ¼ 0.006 and r2 ¼ 0.344, 95% CI 0.155e0.469, p <0.0001, respectively). At the end of the UD period PVP significantly correlated with blood loss during that time (r2 ¼ 0.190, 95% CI 0.022e0.371, p ¼ 0.022, fig. 2). CVP and PVP did not significantly correlate except for a weak correlation at the end of cystectomy (r2 ¼ 0.134, 95% CI 0.019e0.259, p ¼ 0.025). In the norepinephrine-low volume group PVP significantly correlated with blood loss at the end of the cystectomy part and at the end of the UD part (r2 ¼ 0.244, 95% CI 0.021e0.461, p ¼ 0.032 and r2 ¼ 0.269, 95% CI 0.024e0.495, p ¼ 0.019, respectively). In controls no significant correlation was found between PVP and blood loss during the different periods. PVP was significantly lower in the norepinephrine-low volume group during all periods assessed compared to controls (table 1). PVP was less than 5 mm Hg at the end of the cystectomy part in 42 of 82 patients (51%) in the norepinephrine-low volume group and in 13 of 81 controls (16%) (RR 0.58, 95% CI 0.46e0.74, p <0.0001, table 2). PVP less than 5 mm Hg at the end of the cystectomy part was associated with significantly less intraoperative blood loss in the 2 groups. In the norepinephrinelow volume group median total blood loss in the subgroup with PVP less than 5 mm Hg was 800 ml (range 300 to 1,200) compared to 900 ml (range 500 to 1,800) in the subgroup with PVP 5 mm Hg or greater (p ¼ 0.004, table 2). In controls the subgroup with PVP less than 5 mm Hg had a median blood loss of 1,000 ml (range 600 to 1,600) compared to 1,300 ml (range 400 to 3,000) in the subgroup with PVP 5 mm Hg or greater (p ¼ 0.006, table 2 and fig. 3).
149
Figure 2. Fit lines with 95% CI show PVP by blood loss during PLND (A), cystectomy (B) and UD (C ) (r2 ¼ 0.234, 0.344 and 0.190, respectively). Sphere size indicates number of patients.
150
PELVIC VENOUS PRESSURE AND BLOOD LOSS DURING CYSTECTOMY AND URINARY DIVERSION
Table 1. PVP by randomized norepinephrine-low volume and control groups Median mm Hg PVP (range) Overall
Norepinephrine
No. pts 183 82 PLND: Start 7 (0e16) 5 (0e13)* Start þ abdominal 2 (0e12) wall lifting End 6 (0e15) 4 (0e12)* End þ abdominal 2 (e1e4) wall lifting RC: End 6 (e1e14) 4 (e1e12)* Abdominal wall 1 (2e6) lifting After UD 5 (e1e13) 2 (e1e8)
Control
13 patients (15%) had PVP less than 5 mm Hg vs 24 of 68 (35%) with PVP 5 mm Hg or greater (p ¼ 0.206).
p Value e
81 8 (0e16)* 3 (0e12)
<0.0001 0.007
8 (0e15)* 3 (e2e9)
<0.0001 <0.0001
7 (0e14)* <0.0001 3 (0e8) < 0.0001 6 (0e13)
<0.0001
* Compared to abdominal wall lifting Wilcoxon signed rank test with Bonferroni adjustment p <0.0001
PVP decreased significantly after removing the gauze from the abdominal cavity and lifting the abdominal wall in each group at any time point measured (p <0.0001, table 1). There was a significant difference in the rate of blood unit transfusion between patients with PVP less than 5 mm Hg and patients with PVP 5 mm Hg or greater (5 of 55 or 9% vs 28 of 108 or 26%, RR 0.72, 95% CI 0.10 to 0.79, p ¼ 0.013). However, we found no difference between PVP subgroups in either randomized group. In the norepinephrine-low volume group 3 of 42 patients (7%) had PVP less than 5 mm Hg vs 4 of 40 (10%) with PVP 5 mm Hg or greater (p ¼ 0.709). In the control group 2 of
DISCUSSION Total intraoperative blood loss was significantly less in patients with low PVP (less than 5 mm Hg) compared to that in patients with PVP 5 mm Hg or greater. Significantly more patients with preemptive infusion of norepinephrine combined with restrictive deferred hydration had PVP less than 5 mm Hg. Even in this group increased PVP was associated with increased blood loss. This suggests that fluid management can influence PVP and consequently impact blood loss. Blood loss during open RC with UD impacts the postoperative outcome and increases postoperative morbidity.9e12 Blood unit transfusion after excessive blood loss carries the risk of transfusion associated lung injury, postoperative transfusion associated fluid overload and an increased wound infection rate, and it was postulated to promote cancer recurrence.13e15 Consequently intraoperative strategies to decrease blood loss may have vital consequences. Little is known about the impact of venous pressure at the surgical site on blood loss. Substantial literature has already demonstrated that a significant reduction in blood loss could be achieved by maintaining CVP at lower than 5 mm Hg in patients undergoing liver dissection, thus, making it easier
Table 2. Blood loss, hemodynamic data and intraoperative blood unit transfusion rate in 82 patients in norepinephrine-low volume group and 81 controls by PVP subgroup Median Norepinephrine mm Hg PVP (range) Less Than 5 No. pts Blood loss (ml): Per min PLND RC UD Heart rate (bpm): PLND RC UD MAP (mm Hg): PLND RC UD CVP (mm Hg): Baseline PLND RC UD Crystalloid (ml) *p †p ‡p §p
42 800 (300e1,200)*,† 1.8 (0.7e2.9)*,† 100 (0e400)* 480 (200e850)*,† 100 (0e410)
Median Control mm Hg PVP (range)
5 or Greater
Less Than 5
5 or Greater
p Value
40 900 (500e1,800)‡ 2.0 (1.2e3.5)‡ 145 (0e300)‡ 550 (300e1,200)‡ 200 (0e500)
13 1,000 (600e1,600) 2.7 (1.5e3.7) 150 (50e300) 650 (500e1,200) 150 (50e500)
68 1,300 (400e2,800) 3.2 (2.9e3.5) 200 (50e2,000) 900 (300e2,300) 200 (0e700)
<0.0001 0.005 <0.0001 <0.0001 0.049
66 70 74
(50e94) (47e102) (50e107)
66 67 77
(45e92) (43e100) (41e100)
60 65 75
(52e85) (44e78) (55e90)
65 63 70
(45e90) (40e90) (40e98)
0.167 0.123 0.210
70 70 72
(60e85) (60e85) (62e94)
70 71 72
(60e93) (60e93) (60e100)
70 73 70
(60e84) (60e81) (60e85)
68 70 70
(60e90) (60e100) (60e92)
0.577 0.438 0.684
12 (7e20) 12 (7e19) 11 (4e23) 9 (3e22) 1,500 (700e2,500)*,†
<0.0001 vs control PVP 5 mm Hg or greater. <0.05 vs control PVP less than 5 mm Hg. <0.0001 vs control PVP 5 mm Hg or greater. <0.0001 vs control PVP less than 5 mm Hg.
13 (6e21) 13 (5e21) 12 (2e23) 11 (4e19) 1,800 (800e4,000)‡,§
14 (5e20) 15 (10e20) 16 (10e20) 15 (9e20) 3,800 (3,000e6,100)
13 (5e21) 15 (5e23) 15 (2e25) 14 (3e29) 4,450 (2,800e6,200)
0.132 <0.0001 <0.0001 <0.0001 <0.0001
PELVIC VENOUS PRESSURE AND BLOOD LOSS DURING CYSTECTOMY AND URINARY DIVERSION
Figure 3. Total blood loss in norepinephrine-low volume (A) and control (B) groups by PVP.
to control vascular injury.3,4 In this study we found that preemptive norepinephrine use with restrictive deferred fluid management led to a significant reduction in PVP and blood loss compared to a more liberal fluid regimen. The question remains of whether the decreased blood loss was attributable to reduced intravenous fluid administration and to vasoconstriction induced by norepinephrine, resulting in lower PVP. The correlation between blood loss and PVP independent of the amount of fluid administered emphasizes that PVP significantly impacts blood loss. Even in the control group, in which patients received more generous fluid administration, PVP less than 5 mm Hg at the end of cystectomy was associated with a significant decrease in blood loss. An additional important observation was that lifting the abdominal wall significantly decreased
151
PVP. This suggests that avoiding abdominal packing, for example using a table fixed, self-retaining retractor to lift the abdominal wall, may help reduce intraoperative PVP. The potential hemodynamic impact of an abdominal wall-lifting retractor compared to conventional abdominal packing in patients undergoing open abdominal surgery is less well studied than in laparoscopic surgery. In laparoscopic surgery an abdominal wall lifting device had a positive effect on blood loss and CVP in animal models and in patients undergoing cholecystectomy.16,17 The abdominal wall lifting device resulted in an increased cardiac index compared to carbon dioxide insufflation without affecting systemic vascular resistance in healthy patients undergoing laparoscopic cholecystectomy and was associated with fewer side effects, such as postoperative nausea and vomiting, and right shoulder pain. For these reasons this approach was recommended in patients at high risk.17e20 Another important observation is the lack of correlation between PVP and CVP except for a weak correlation at the end of cystectomy. Thus, standard CVP assessment may not be a reliable surrogate for PVP, which is a factor that must be considered. In some of our patients negative PVPs were measured, which could potentially indicate a risk of air emboli. However, the 30-degree Trendelenburg position and pelvic packing with wet gauze after bladder removal should limit the risk of air emboli. In addition, no decrease was detected in end tidal carbon dioxide, making clinically relevant air emboli unlikely. The main limitation of this study is the lack of a specific power analysis concerning the reduction in blood loss because this study was a secondary analysis. In addition, PVP was only measured at certain time points and not continuously. However, continuous monitoring of PVP would imply catheterization of the veins, resulting in a higher risk of thrombosis.
CONCLUSIONS Our results suggest that low PVP during PLND, RC and UD significantly decreases intraoperative blood loss. Techniques to decrease PVP, such as a norepinephrine-low volume regimen and avoidance of vena caval compression (eg by intra-abdominal packing), should be established for open RC and UD.
REFERENCES 1. Wuethrich PY, Burkhard FC, Thalmann GN et al: Restrictive deferred hydration combined with preemptive norepinephrine infusion during
radical cystectomy reduces postoperative complications and hospitalization time: a randomized clinical trial. Anesthesiology 2014; 120: 365.
2. Wuethrich PY, Studer UE, Thalmann GN et al: Intraoperative continuous norepinephrine infusion combined with restrictive deferred hydration
152
PELVIC VENOUS PRESSURE AND BLOOD LOSS DURING CYSTECTOMY AND URINARY DIVERSION
significantly reduces the need for blood transfusion in patients undergoing open radical cystectomy: results of a prospective randomised trial. Eur Urol 2014; 66: 352.
9. Linder BJ, Frank I, Cheville JC et al: The impact of perioperative blood transfusion on cancer recurrence and survival following radical cystectomy. Eur Urol 2013; 63: 839.
3. Wang WD, Liang LJ, Huang XQ et al: Low central venous pressure reduces blood loss in hepatectomy. World J Gastroenterol 2006; 12: 935.
10. Chang SS, Smith JA Jr, Wells N et al: Estimated blood loss and transfusion requirements of radical cystectomy. J Urol 2001; 166: 2151.
4. Jones RM, Moulton CE and Hardy KJ: Central venous pressure and its effect on blood loss during liver resection. Br J Surg 1998; 85: 1058.
11. Chang SS, Baumgartner RG, Wells N et al: Causes of increased hospital stay after radical cystectomy in a clinical pathway setting. J Urol 2002; 167: 208.
5. Smyrniotis V, Kostopanagiotou G, Theodoraki K et al: The role of central venous pressure and type of vascular control in blood loss during major liver resections. Am J Surg 2004; 187: 398.
12. Chang SS, Cookson MS, Baumgartner RG et al: Analysis of early complications after radical cystectomy: results of a collaborative care pathway. J Urol 2002; 167: 2012.
6. Klein HG, Spahn DR and Carson JL: Red blood cell transfusion in clinical practice. Lancet 2007; 370: 415.
13. Hendrickson JE and Hillyer CD: Noninfectious serious hazards of transfusion. Anesth Analg 2009; 108: 759.
7. Burkhard FC, Kessler TM, Mills R et al: Continent urinary diversion. Crit Rev Oncol Hematol 2006; 57: 255.
14. Amato A and Pescatori M: Perioperative blood transfusions for the recurrence of colorectal cancer. Cochrane Database Syst Rev 2006; CD005033.
8. Perimenis P and Studer UE: Orthotopic continent urinary diversion an ileal low pressure neobladder with an afferent tubular segment: how I do it. Eur J Surg Oncol 2004; 30: 454.
15. Glance LG, Dick AW, Mukamel DB et al: Association between intraoperative blood transfusion and mortality and morbidity in patients
undergoing noncardiac surgery. Anesthesiology 2011; 114: 283. 16. Lindgren L, Koivusalo AM and Kellokumpu I: Conventional pneumoperitoneum compared with abdominal wall lift for laparoscopic cholecystectomy. Br J Anaesth 1995; 75: 567. 17. Andersson L, Lindberg G, Bringman S et al: Pneumoperitoneum versus abdominal wall lift: effects on central haemodynamics and intrathoracic pressure during laparoscopic cholecystectomy. Acta Anaesthesiol Scand 2003; 47: 838. 18. Galizia G, Prizio G, Lieto E et al: Hemodynamic and pulmonary changes during open, carbon dioxide pneumoperitoneum and abdominal walllifting cholecystectomy. A prospective, randomized study. Surg Endosc 2001; 15: 477. 19. Rademaker BM, Meyer DW, Bannenberg JJ et al: Laparoscopy without pneumoperitoneum. Effects of abdominal wall retraction versus carbon dioxide insufflation on hemodynamics and gas exchange in pigs. Surg Endosc 1995; 9: 797. 20. Rademaker BM, Bannenberg JJ, Kalkman CJ et al: Effects of pneumoperitoneum with helium on hemodynamics and oxygen transport: a comparison with carbon dioxide. J Laparoendosc Surg 1995; 5: 15.