european urology 51 (2007) 397–402
available at www.sciencedirect.com journal homepage: www.europeanurology.com
Bladder Cancer
Perioperative Complications of Radical Cystectomy in a Contemporary Series Vladimir Novotny a,*, Oliver W. Hakenberg a, Diana Wiessner a, Ulrike Heberling a, Rainer J. Litz b, Sven Oehlschlaeger a, Manfred P. Wirth a a b
Department of Urology, Technical University, Dresden, Germany Department of Anesthesiology, Technical University, Dresden, Germany
Article info
Abstract
Article history: Accepted June 8, 2006 Published online ahead of print on June 27, 2006
Objectives: Radical cystectomy is the preferred standard treatment for patients with muscle-invasive bladder cancer. With improvements in intra- and perioperative care lower complication rates have been reported. We retrospectively evaluated our series of patients who underwent radical cystectomy for advanced bladder cancer for perioperative complications as well as operative time, postoperative hospital stay and transfusion rates. Patients and methods: Between April 1993 and August 2005, 516 radical cystectomies were performed for muscle infiltrating transitional cell carcinoma and other types of neoplastic diseases of the bladder at our institution. The average age was 66.3 yr (31–89). Results: The perioperative mortality rate was 0.8%. A total of 141 patients (27.3%) developed at least one perioperative complication. The most frequent medical complications were subileus in 20 (3.9%) patients, deep venous thrombosis in 24 (4.7%), and enterocolitis in 10 (1.9%). Surgical complications included pelvic lymphoceles in 42 (8.1%) patients, wound dehiscence in 46 (8.9%), pelvic hematoma in 4 (0.8%), peritonitis in 4 (0.8%) and small bowel obstruction in 4 (0.8%). The total early reoperation rate was 6.2%. Operative time, postoperative hospital stay and average number of blood units transfused decreased over the period 1993–2005. Conclusions: Radical cystectomy today is a procedure with an acceptable rate of perioperative morbidity and mortality. Improvements in surgical technique and anaesthesia as well as increased quality of perioperative care in recent years have resulted in reduced morbidity and shorter hospital stay.
Keywords: Bladder cancer Hospital stay Morbidity Mortality Operative time Perioperative care Perioperative complications Radical cystectomy Transitional cell carcinoma Urinary bladder
# 2006 European Association of Urology. Published by Elsevier B.V. All rights reserved. * Corresponding author. Department of Urology, Technical University, Fetscherstrasse 74, 01309 Dresden, Germany. Tel. +49 351 4584158; Fax: +49 351 4584333. E-mail address:
[email protected] (V. Novotny).
0302-2838/$ – see back matter # 2006 European Association of Urology. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.eururo.2006.06.014
398
1.
european urology 51 (2007) 397–402
Introduction
Bladder cancer is the second most common urologic malignancy, with transitional cell carcinoma making up nearly 90% of all primary bladder tumors [1]. The general increase in life span is associated with an increase in the incidence of bladder cancer, which has significantly increased over the last 20 yr [2]. Although the majority of patients present with superficial bladder cancer, 20–40% either present with or develop invasive disease [1]. Radical cystectomy with pelvic lymph node dissection is the gold standard treatment for muscle-invasive bladder carcinoma and remains the most effective method for local control [3]. Radical cystectomy is a major procedure with the potential for serious complications, most of which develop in the early postoperative period. However, improvements in surgical technique, anesthesia, and peri- and postoperative management have reduced the complication and mortality rates previously associated with this operation. In this study we retrospectively evaluated the perioperative morbidity and mortality in patients who underwent radical cystectomy and urinary diversion for bladder cancer in our department over a period of 12 yr. Operative time, length of postoperative hospital stay, and transfusion rates were analysed as well. 2.
Patients and methods
We reviewed the records of all 516 patients with advanced bladder cancer who underwent radical cystectomy and urinary diversion between April 1993 and August 2005 in our department. Patients who underwent cystectomy with pelvic exenteration for advanced bowel or gynaecologic malignancies are not included in this series. Of these 516 patients, 413 (80.0%) were male and 103 (20.0%) female. The mean patient age was 66.3 yr (median: 67 yr; range: 31–89); 64.5% of patients had a preoperative ASA score of 2, 32.3% of 3, and only 2.5% had a preoperative ASA score of 1 and 0.6% of 4, respectively. Patients who received neoadjuvant chemotherapy totaled 4.3%, whereas 0.8% had undergone neoadjuvant radiotherapy. Preoperative workup routinely included chest x-ray, renal ultrasound, excretory urography, routine blood chemistry, computerized tomography of the abdomen and pelvis, and a bone scan in selected cases. All patients received prophylactic low-molecular-weight heparin subcutaneously, initiated 12 hours before surgery and maintained until discharge. For 2 d before surgery, bowel preparation by oral electrolyte solution and intravenous antibiotics were carried out routinely in all patients. A nasogastric tube was placed intraoperatively and removed postoperatively or no later than 24 hours after surgery in almost all cases. All patients were monitored intra- and perioperatively by a central venous line and an arterial line.
From 1999 onwards, patients received thoracic epidural application of local anesthetics and opioids intraoperatively with general anaesthesia and postoperatively for pain management. Before then, postoperative analgesia had been effected by patient-controlled intravenous administration of opioids. All patients were routinely admitted to the department’s urologic intensive care unit for 2–5 d with routine cardiovascular monitoring and intensive nursing care. Radical cystectomy included the bladder, distal ureters, seminal vesicles, and prostate in men, whereas urethrectomy was not routinely performed in males. In women the uterus, both ovaries, the anterior vaginal wall, and the urethra were routinely removed with the bladder, except in cases of neobladder diversion. Pelvic lymphadenectomy was performed in all cases with the upper limit of lymphadenectomy extending to the common iliac vessels up to the aortic bifurcation and including clearance of the internal iliac vessels. Ileal anastomoses were done side-by-side by stapler technique and, since 2002, end-to-end by a single seromuscular running suture with the use of double-armed Maxon 3.0. This change in technique was due to costs. For urinary diversion, ileal neobladders were constructed in 190 (36.8%) patients, ileal conduits in 299 (57.9%) patients, and other types of urinary diversion (colon conduits, pouches, ureterocutaneostomies) in 27 (5.2%) patients. All ureterointestinal anastomoses were stented intraoperatively regardless of the form of urinary diversion for 12 d. In ileal neobladders, urethral catheters were removed 3 wk postoperatively after a contrast cystogram had demonstrated no anastomotic leakage. Perioperative complications were defined as any adverse event within 30 d of surgery. Equally, perioperative mortality was defined as death from any cause within 30 d of surgery. For statistical analysis, means were calculated and compared by standard t test.
3.
Results
Tumor stages after transurethral resection and prior to cystectomy were pT1 G3 in 13.4%, pT2 in 70.7%, pT3 in 2.0%, pT4 in 3.7%, and/or pTis in 15.2%. Pathologic stages after cystectomy were pT0 in 18.3%, pTa in 1.3%, pT1 in 10.6%, pT2 in 21.9%, pT3 in 26.4%, pT4 in 12.8%, and/or pTis in 13.4%. Lymph node metastases were found in 22.7% of all patients (pN1 in 8.5%, pN2 in 13.6%, pN3 in 0.6%). Five hundred two patients had transitional cell carcinoma, 6 adenocarcinoma, 4 squamous cell carcinoma, and 1 bladder melanoma; three patients had other types of rare bladder neoplasms. There were four perioperative deaths (0.8%). One 68-year-old man and one 69-year-old woman with severe systemic diseases (diabetes, coronary heart disease, hypertension, renal insufficiency) died of heart failure intraoperatively and on the fourth postoperative day, respectively. One 69-year-old
399
european urology 51 (2007) 397–402
Table 1 – Incidence of perioperative complications in all patients with cystectomy Perioperative complications Medical Deep venous thrombosis Pulmonary embolism Septicemia Subileus (paralytic) Acute respiratory distress Myocardial infarction Enterocolitis Mortality Surgical Small-bowel obstruction Pelvic lymphocele No intervention Percutaneous drainage Peritonitis Wound dehiscence Secondary healing With revision Pelvic hematoma Without revision With revision No. of patients with complications
No. of complications (n (%)) 24 9 7 20 8 7 10 4
(4.7) (1.7) (1.4) (3.9) (1.6) (1.4) (1.9) (0.8)
4 (0.8) 28 (5.4) 14 (2.7) 4 (0.8) 20 (3.9) 26 (5.0) 2 (0.4) 2 (0.4) 141 (27.3)
Perioperative complications in all patients (n = 516). Complications are divided into medical complications with conservative treatment and surgical complications, some of which required surgical revision. The complications given occurred in 141 of 516 patients.
woman with coronary heart disease and chronic obstructive lung disease died of myocardial infarction on postoperative day 17, and one 83-year-old man died of septicemia on postoperative day 4. A total of 141 (27.3%) patients including the four perioperative deaths developed at least one perioperative complication (Table 1). One hundred four (20.2%) patients developed only one complication each, while 37 (7.8%) patients developed two or more perioperative complications. We divided complications into medical and surgical complications. Paralytic subileus that was successfully managed conservatively (and therefore defined as a medical complication) developed in 20 (3.9%) patients, deep venous thrombosis in 24 (4.7%) patients, and enterocolitis in 10 (1.9%) patients. These were the most frequent medical complica-
tions. Surgical complications included the formation of pelvic lymphoceles in 42 (8.1%) patients, wound dehiscence in 46 (8.9%), pelvic hematoma in 4 (0.8%), peritonitis in 4 (0.8%) and small bowel obstruction in 4 (0.8%). Revision laparotomy was required for small-bowel obstruction in four (0.8%) patients and for pelvic hematoma in two (0.4%). Twenty-six (5.0%) patients underwent surgical revision for wound dehiscence. The total reoperation rate was therefore 6.2%. Mean operative time was 6.1 1.3 h (median: 6) and depended on the type of diversion (Table 2). On average, cystectomies with neobladder diversions lasted longer than those with ileal conduit diversion (6.5 vs. 5.8 h, p < 0.001). Operative time, length of postoperative hospital stay, and average number of blood units given perioperatively were evaluated according to the date of surgery (Table 3). The mean length of postoperative hospital stay was 21.2 6.8 d (median: 19) and decreased significantly over the period analyzed. The average postoperative hospital stays were 23.8 7.4 d (median: 22) in patients who underwent surgery before April 1999, 22.5 6.4 d (median: 21) between May 1999 and July 2002, and 17.4 4.7 d (median: 15) between August 2002 and August 2005 ( p < 0.001). As expected, the postoperative hospital stay was significantly longer in patients with complications (24.9 8.8 d [median: 23], p < 0.001) compared with patients without complications who had a mean postoperative hospital stay of 19.8 5.7 d (median: 18). Complications thus caused a mean prolongation of the average postoperative hospital treatment of 5.1 d. The mean postoperative hospital stay in patients with septicemia was 30.1 18 d, in those with enterocolitis 33.1 7.5 d, in those with wound dehiscence 26.2 8.6 d, and in those with pelvic hematoma 31.3 10.3 d. The postoperative hospital stay was longer in patients with ileal neobladders than in patients with ileal conduits (24.3 vs 19.0 d, p < 0.001; Table 2). The average intraoperative blood loss was 1208 ml. A total of 425 (82.4%) patients received intra- and/or postoperative blood transfusions, with 329 (63.8%) receiving transfusions intraoperatively (on average:
Table 2 – Operative time and postoperative hospital stay in neobladder urinary diversion versus ileal conduits
Operative time (h) Postoperative hospital stay (d)
Ileal neobladder (n = 190)
Ileal conduit (n = 299)
Significance
6.5 1.1 (median: 6.5) 24.3 6.6 (median: 25)
5.8 1.3 (median: 5.7) 19.0 5.3 (median: 17)
p < 0.001 p < 0.001
Surgical time and postoperative length of hospital stay in this series in the two frequently performed types of urinary diversion. Data are means SD (median).
400
european urology 51 (2007) 397–402
Table 3 – Operative time, length of postoperative hospital stay, and average number of blood units transfused in relation to date of surgery Date of surgery
Operative time (hours)
Postoperative hospital stay (days)
Blood units per patient
Group A April 1993–April 1999 (n = 172)
6.6 1.2 (median: 6.7)
23.8 7.4 (median: 22)
2.7 2.4 (median: 2)
Group B May 1999–July 2002 (n = 172)
5.8 1.3 (median: 5.8)
22.5 6.4 (median: 21)
2.3 1.8 (median: 2)
Group C August 2002–August 2005 (n = 172)
5.7 1.1 (median: 5.6)
17.4 4.7 (median: 15)
1.4 1.4 (median: 0)
Surgical time, blood requirements, and postoperative length of hospital stay over the time period analyzed. The total group of 516 patients was divided into three consecutive groups of equal size (n = 172 each) corresponding to the periods of April 1993–April 1999, May 1999–July 2002, and August 2002–August 2005. For surgical time, the differences between groups A and B as well as between groups A and C were significant ( p < 0.001 for each); for blood requirements and the length of postoperative hospital stay, the differences between groups A and C and between groups B and C were significant ( p < 0.001 for each).
2.6 2.4 units) and 305 (59.1%) postoperatively (on average: 1.6 1.4 units). Thus, an overall mean of 2.1 1.8 units of blood was given per patient (range: 0–24; median: 2). Female patients needed more transfusions than males (2.8 vs 2.0, p < 0.001). The need for perioperative blood transfusions decreased over the time period analysed. On average, up to April 1999, 2.7 units of blood per patient were transfused, 2.3 units per patient between May 1999 and July 2002, and 1.4 units per patient since August 2002. This decrease was significant ( p < 0.001; Table 3).
4.
Discussion
Radical cystectomy is the treatment of choice for patients with invasive bladder cancer. Although the surgical routine with this procedure has improved and even less invasive laparoscopic techniques can now be applied to radical cystectomy [4], it remains an operative procedure with significant morbidity and potentially life-threatening complications. Although the morbidity of radical cystectomy is clearly lower than in previous decades, probably because of more sophisticated postoperative care as well as improved anaesthesiologic and surgical techniques, the rate remains higher than 30% in the early postoperative period [1,5]. In our series, we observed a 30-day mortality rate of 0.8%. The reported perioperative mortality rate in cystectomy patients ranges between 0.3% and 4.5% [1,5–8]. Ghoneim et al. [3] reported a perioperative mortality of 4% in 1026 cases, and in another large series, Stein et al. [1] observed 3% perioperative deaths in 1054 patients. In an update of the latter series published by Quek et al. [9], a 30-day mortality of 2% in 1359 patients was reported. Chang et al. [5] reported the lowest mortality with only one death in 304 patients (0.3%). Perioperative mortality is higher
in older patients. Stroumbakis et al. [7] reported a mortality of 4.5% in patients aged >80 yr, and Figueroa et al. [6] reported mortality of 2.8% in 404 patients >70 yr. In the same series, the mortality was 2% in patients
