Comprehensive standardized report of complications of retropubic and laparoscopic radical prostatectomy.

Abstract

The lack of standardized reporting of the complications of radical prostatectomy in the literature has made it difficult to compare incidences across institutions and across different surgical approaches.
To define comprehensively the incidence, severity, and timing of onset of medical and surgical complications of open retropubic prostatectomy (RP) and laparoscopic radical prostatectomy (LP) using a standardized reporting methodology to facilitate comparison.
Between January 1999 and June 2007, 4592 consecutive patients underwent RP or LP without prior radiation or hormonal therapy. Median follow-up was 36.9 mo (interquartile range: 20.3-60.6).
Open or laparoscopic radical prostatectomy.
All medical and surgical complications of radical prostatectomy were captured and graded according to the modified Clavien classification and classified by timing of onset.
There were 612 medical complications in 467 patients (10.2%) and 1426 surgical complications in 925 patients (20.1%). The overall incidences of early minor and major medical and surgical complications for RP were 8.5% and 1.5% for medical and 11.4% and 4.9% for surgical complications, respectively. The overall incidences of early minor and major medical and surgical complications for LP were 14.2% and 2.3% for medical and 23.1% and 6.6% for surgical complications, respectively. On multivariate analysis, LP approach was associated with a higher incidence of any grade medical and surgical complications but a lower incidence of major surgical complications than RP. Six hundred fifty-two men (14.2%) visited the emergency department, and 240 men (5.2%) required readmission. The main limitation is the retrospective nature.
With standardized reporting, the incidence of some complications is higher than recognized in the literature. Although most complications are minor in severity, medical and surgical complications are observed in approximately 10% and 20% of patients, respectively. Accurate reporting of complications through a standardized methodology is essential for counseling patients regarding risk of complications, for identifying modifiable risk factors, and for facilitating comparison across institutions and approaches.

Full text

Page 1

Platinum Priority – Prostate Cancer
Editorial by Markus Graefen on pp. 387–389 of this issue
Comprehensive Standardized Report of Complications of
Retropubic and Laparoscopic Radical Prostatectomy
Farhang Rabbani*, Luis Herran Yunis, Rodrigo Pinochet, Lucas Nogueira, Kinjal C. Vora,
James A. Eastham, Bertrand Guillonneau, Vincent Laudone, Peter T. Scardino, Karim Touijer
Urology Service, Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY, USA
EUROPEAN UROL OG Y 57 (2010) 371–386
available at www.sciencedirect.com
journal homepage: www.europeanurology.com
Article info
Article history:
Accepted November 17, 2009
Published online ahead of
print on November 25, 2009
Keywords:
Complications
Prostatectomy
Laparoscopy
Prostatic neoplasms
Risk factors
Comorbidity
Please visit
www.eu-acme.org/
europeanurology to read and
answer questions on-line.
The EU-ACME credits will
then be attributed
automatically.
Abstract
Background: The lack of standardized reporting of the complications of radical prosta-
tectomy in the literature has made it difficult to compare incidences across institutions
and across different surgical approaches.
Objective: To define comprehensively the incidence, severity, and timing of onset of
medical and surgical complications of open retropubic prostatectomy (RP) and laparo-
scopic radical prostatectomy (LP) using a standardized reporting methodology to
facilitate comparison.
Design, setting, and participants: Between January 1999 and June 2007, 4592 consecu-
tive patients underwent RP or LP without prior radiation or hormonal therapy. Median
follow-up was 36.9 mo (interquartile range: 20.3–60.6).
Intervention: Open or laparoscopic radical prostatectomy.
Measurements: All medical and surgical complications of radical prostatectomy were
captured and graded according to the modified Clavien classification and classified by
timing of onset.
Results and limitations: There were 612 medical complications in 467 patients (10.2%)
and 1426 surgical complications in 925 patients (20.1%). The overall incidences of early
minor and major medical and surgical complications for RP were 8.5% and 1.5% for
medical and 11.4% and 4.9% for surgical complications, respectively. The overall inci-
dences of early minor and major medical and surgical complications for LP were 14.2%
and 2.3% for medical and 23.1% and 6.6% for surgical complications, respectively. On
multivariate analysis, LP approach was associated with a higher incidence of any grade
medical and surgical complications but a lower incidence of major surgical complica-
tionsthanRP. Six hundredfifty-twomen (14.2%)visited the emergencydepartment,and
240 men (5.2%) required readmission. The main limitation is the retrospective nature.
Conclusions: With standardized reporting, the incidence of some complications is
higher than recognized in the literature. Although most complications are minor in
severity,medicaland surgicalcomplicationsareobserved inapproximately10% and20%
of patients, respectively. Accurate reporting of complications through a standardized
methodology is essential for counseling patients regarding risk of complications, for
identifying modifiable risk factors, and for facilitating comparison across institutions
and approaches.
# 2009 European Association of Urology. Published by Elsevier B.V. All rights reserved.
* Corresponding author. Urology Service, Department of Surgery, Memorial Sloan-Kettering Cancer
Center, 1275 York Avenue, New York, NY 10065, USA. Tel. +1 646 422 4385; Fax: +1 212 988 0760.
E-mail address: rabbanif@mskcc.org (F. Rabbani).
0302-2838/$ – see back matter # 2009 European Association of Urology. Published by Elsevier B.V. All rights reserved.doi:10.1016/j.eururo.2009.11.034

Page 2

1.Introduction
With the introduction of laparoscopic radical prostatecto-
my (LP) and robot-assisted LP (RARP), there have been a
number of publications comparing these more recent
techniques to open radical retropubic prostatectomy (RP)
[1–6]. Differences in patterns of practice make comparison
of complications across different approaches difficult and
even more challenging across different institutions. To
facilitate comparison of complication rates, consistency and
clarity in reporting is necessary, as some authors have
advocated [7].
The more widespread use of grading schemes [8] in
reporting complications has facilitated standardization to
some degree. Martin et al [7] have identified 10 critical
elements of accurate and comprehensive reports of
surgical complications (Appendix A). Although capture of
all complications may be impossible in a retrospective
study due to poor documentation, even in prospective
studies [1], some complication categories may not be
reported. A review of the studies evaluating complication
rates in the urologic oncology literature reported that only
2% met 9 or 10 of Martin et al’s criteria and 21% met 7 or 8
criteria. The most commonly underreported criteria were
complication definitions, complication severity/grade,
outpatient data, comorbidities, and duration of the
reporting period [9].
We sought to apply the standards set forth by Martin
et al [7] in performing a comprehensive report on the
complications of radical prostatectomy at our institution in
an effort to encourage the use of standardized reporting in
the literature and, thus, to facilitate comparison across
institutions and surgical approaches and to aid in patient
counseling.
2.Patients and methods
Between January 1999 and June 2007, 4592 consecutive patients
underwent radical prostatectomy (RP or LP) for adenocarcinoma of the
prostate at our institution. Patients undergoing LP received low-
molecular-weight heparin starting before surgery and continued daily
until discharge; patients undergoing RP did not routinely receive low-
molecular-weight heparin but had sequential compression boots.
Table 1 – Preoperative, intraoperative, and postoperative parameters stratified by prostatectomy approach
ParameterAll patients
(n = 4592) n (%)
RP (n = 3458)
n (%)
LPy(n = 1134)
n (%)
p value#
Preoperative clinical parameters
Age
Median (IQR)
Ethnicity
White
African American
Other/unknown
BMI*
Median (IQR)
Clinical stage
T1
T2
T3
Tx
Gleason score
?6
7
?8
Preoperative PSA, ng/mlz
Median (IQR)
Neoadjuvant hormonal therapy
ASA score§
1
2
3
4
Modified Charlson comorbidity score**
0
1
2
3
4
?5
Hypertension
Hypercholesterolemia/hypertriglyceridemia
Diabetes
Coronary artery disease
Prior MI
Comorbid disease
Cardiac
59.5 (54.7–64.2)59.4 (54.5–64.1)59.7 (55.4–64.7) 0.26
4067 (88.6)
317 (6.9)
208 (4.5)
3081 (89.1)
234 (6.8)
143 (4.1)
986 (86.9)
83 (7.3)
65 (5.7)
0.060
27.6 (25.4–30.3) 27.7 (25.4–30.3)27.6 (25.5–30.2) 0.73
2864 (62.4)
1571 (34.2)
150 (3.3)
7 (0.2)
2056 (59.5)
1275 (36.9)
121 (3.5)
6 (0.2)
808 (71.3)
296 (26.1)
29 (2.6)
1 (0.09)
<0.001
2679 (58.3)
1544 (33.6)
369 (8.0)
2029 (58.7)
1126 (32.6)
303 (8.8)
650 (57.3)
418 (36.9)
66 (5.8)
0.001
5.5 (4.1–7.8)
224 (4.9)
5.6 (4.2–8.0)
190 (5.5)
5.3 (3.8–7.3)
34 (3.0)
0.023
0.001
501 (10.9)
3506 (76.4)
580 (12.6)
1 (0.02)
424 (12.3)
2601 (75.2)
428 (12.4)
1 (0.03)
77 (6.8)
905 (79.8)
152 (13.4)
0 (0)
<0.001
403 (8.8)
1673 (36.4)
1789 (39.0)
563 (12.3)
125 (2.7)
36 (0.8)
1714 (37.3)
1623 (35.3)
311 (6.8)
328 (7.1)
98 (2.1)
317 (9.2)
1263 (36.5)
1362 (39.4)
401 (11.6)
86 (2.5)
26 (0.8)
1255 (36.3)
1125 (32.5)
225 (6.5)
220 (6.4
65 (1.9)
86 (7.6)
410 (36.2)
427 (37.7)
162 (14.3)
39 (3.4)
10 (0.9)
459 (40.5)
498 (43.9)
86 (7.6)
108 (9.5)
33 (2.9)
0.083
0.011
<0.001
0.21
<0.001
0.037
355 (7.7)238 (6.9) 117 (10.3)
<0.001
E U RO P E A N U R O L O GY 57 ( 2 0 1 0 ) 3 7 1 – 3 8 6
372

Page 3

Table 1 (Continued)
ParameterAll patients
(n = 4592) n (%)
RP (n = 3458)
n (%)
LPy(n = 1134)
n (%)
p value#
Vascular
Valvular heart disease
Dysrhythmia
Endocrine
Neurologic
Procoagulable disorder***
Coagulopathy
Immunologic
Upper GI
Lower GI
Pulmonary
Renal
Hepatic
Musculoskeletal
398 (8.7)
20 (0.4)
133 (2.9)
400 (8.7)
19 (0.4)
55 (1.2)
6 (0.1)
26 (0.6)
357 (7.8)
83 (1.8)
339 (7.4)
21 (0.5)
78 (1.7)
200 (4.4)
270 (7.8)
16 (0.5)
106 (3.1)
294 (8.5)
14 (0.4)
45 (1.3)
5 (0.1)
17 (0.5)
259 (7.5)
53 (1.5)
251 (7.3)
11 (0.3)
61 (1.8)
157 (4.5)
128 (11.3)
4 (0.4)
27 (2.4)
106 (9.3)
5 (0.4)
10 (0.9)
1 (0.09)
9 (0.8)
98 (8.6)
30 (2.6)
88 (7.8)
10 (0.9)
17 (1.5)
43 (3.8)
<0.001
0.80
0.23
0.38
0.80
0.26
1.0
0.24
0.21
0.015
0.58
0.015
0.55
0.28
Pathologic parameters
Pathologic stage
pT0
pT2
pT3
pT4
Nodal status
N0
N1–N2
Nx
Seminal vesicle invasion
Positive surgical margins
Pathologic Gleason score
2–6
7
8–10
Not graded due to pT0
Not graded due to treatment effect^
Specimen weight, g****
Median (IQR)
35 (0.8)
3186 (69.4)
1272 (27.7)
99 (2.2)
28 (0.8)
2385 (69.0)
961 (27.8)
84 (2.4)
7 (0.6)
801 (70.6)
311 (27.4)
15 (1.3)
0.13
4060 (88.4)
218 (4.7)
314 (6.8)
302 (6.6)
634 (13.8)
3227 (93.3)
171 (4.9)
60 (1.7)
260 (7.5)
506 (14.6)
833 (73.5)
47 (4.1)
254 (22.4)
42 (3.7)
128 (11.3)
<0.001
<0.001
0.005
1681 (36.6)
2489 (54.2)
324 (7.1)
35 (0.8)
63 (1.4)
1338 (38.7)
1773 (51.3)
261 (7.5)
28 (0.8)
58 (1.7)
343 (30.2)
716 (63.1)
63 (5.6)
7 (0.6)
5 (0.4)
<0.001
50 (41–63) 50 (41–63)50 (41.5–63) 0.77
Intraoperative and postoperative parameters
Bilateral PLND performed
Node count
Median (IQR)
Operative time, min
Median (IQR)
Estimated blood loss, ml
Median (IQR)
Blood transfusion
Mean units of blood transfused, no.
Allogenic blood transfusion
Mean units of allogenic blood transfused, no.
Mean drop in serum HGB from preoperative
to discharge, g/dl
Length of stay, d
Median (IQR)
Routine postoperative cystography performed
Drain fluid sent for creatinine
Catheter duration, d
Median (IQR)
Prolonged pelvic drainage, >7 d
Follow-up, mo
Median (IQR)
4278 (93.2) 3398 (98.3) 880 (77.6)
<0.001
10 (7–15) 10 (6–15)12 (8–16)
<0.001
213 (176–255)202 (170–240)241 (210–285)
<0.001
900 (500–1500)
1945 (42.4)
0.90
581 (12.7)
0.31
3.9
1100 (800–1600)
1900 (55)
1.2
538 (15.6)
0.37
4.2
250 (200–400)
45 (4.0)
0.12
43 (3.8)
0.12
2.9
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
3 (2–3)
187 (4.1%)
775 (16.9%)
3 (3–4)
61 (1.8)
472 (13.6)
2 (1–2)
126 (11.1)
303 (26.7)
<0.001
<0.001
<0.001
11 (9–15)
114 (2.5)
13 (10–15)
67 (1.9)
7 (7–10)
47 (4.1)
<0.001
<0.001
36.9 (20.3–60.6)45.5 (24.3–67.7)24.1 (13.7–35.9)
<0.001
ASA = American Society of Anesthesiologists; BMI = body mass index; HGB = hemoglobin; IQR = interquartile range; LP = laparoscopic prostatectomy;
MI = myocardial infarction; PLND = pelvic lymph node dissection; PSA = prostate-specific antigen; RP = open radical retropubic prostatectomy.
yIncludes 97 robotic-assisted LPs.
* BMI missing in 284 patients.
zPSA missing in one patient.
§ASA score missing in four patients.
** Missing in three patients.
*** Includes history of deep venous thrombosis, pulmonary embolus.
**** Specimen weight missing in 225 patients.
^Not graded due to prior hormonal therapy precluding assignment of Gleason score.
#p values for categorical variables are for differences in distributions between RP and LP approaches.
EUROPEAN UR OLOGY 57 (2010) 371– 386
373

Page 4

Data were collected from a prospective prostatectomy database
together with a prospective institutional morbidity database as well
as retrospectively from all postoperative inpatient and outpatient
billing records, inpatient and outpatient medical records including
physician notes, operative notes, discharge summaries, nursing notes,
and correspondence with local physicians outside our institution.
Institutional review board approval was obtained for performing the
study.
All complications other than erectile dysfunction and incontinence,
both medical and surgical, occurring at any time after surgery were
captured including the inpatient stay as well as in the outpatient setting.
Theywereclassifiedasearly(onset:?30d),intermediate(onset:31–90d),
or late (onset: >90 d) complications, depending on the date of onset.
For late complications, those deemed to be related or possibly related to
the prostatectomy were captured, regardless of how long after prostatec-
tomy the onset occurred. All complications were recorded with a grade
(I, II, IIIa, IIIb, IVa, IVb, or V) assigned according to the modified Clavien
classification [8] and were grouped into subcategories of medical and
surgical complications.
Data collected and included in the multivariate analysis as
covariates included prostatectomy approach, whether pelvic lymph
node dissection (PLND) was performed, age, ethnicity, body mass
index
prostate-specific antigen (PSA), prior treatment (including androgen-
deprivation therapy [ADT]), specific comorbidities as well as American
Society of Anesthesiologists (ASA) physical status classification score [10]
and modified Charlson score [11], pathologic stage and nodal status,
surgical margin status, specimen weight, operative time, and estimated
blood loss (EBL). Additional data collected included preoperative and
postoperative serum hemoglobin levels, transfusion data, intensive care
unit(ICU)admissions,reoperations,lengthofstay(LOS),emergencyroom
(ER) visits or readmissions either at our institution or elsewhere, vital
status, and date of death.
Appendix B clarifies the definition of some of the complications. The
primary indication for allogenic blood transfusion was symptomatic
anemia.
(BMI),clinical stage,biopsy Gleasonscore, preoperative
2.1.Statistical analysis
The Pearson x2test, the median test, and the Kruskal-Wallis test were
used to evaluate the distributions of categorical and continuous
variables, as appropriate [12]. Correlation was tested with the Spearman
correlation coefficient. Multivariate Cox proportional hazards analysis
was used with forward stepwise variable selection to assess significant
Table 2 – Incidence and main indications for reoperation, emergency department visit, and readmission
Quality-of-care measure
n (%)Main indications (n)
Reoperation during initial admission37 (0.8):
RP 27 (0.8)
LP 10 (0.9), p = 0.74
Postoperative bleeding (24)
Cystoscopic fulguration of bleeder (5)
Retained drain (4)
Sigmoid perforation (1)
Suspected small bowel obstruction (1)
Ureteral reimplant (1)
Bilateral percutaneous nephrostomy placement (1)
Incisional hernia/dehiscence (21)
Bowel ischemia/injury (6)
Anastomotic disruption (5)
Retained drain (3)
Bladder calculi (3)
Ureteral injury (2)
Suture granuloma (2)
Bladder clot evacuation (1)
Replacement Foley catheter (1)
Postoperative bleeding (1)
Abscess drainage (1)
Aortic valve replacement (1)
Urinary retention/bladder neck contracture/decreased urine output (115)
Lower extremity swelling/deep venous thrombosis (101)
Drain/catheter complaints (89)
Fever (74)
Lymphocele (52)
Wound complaints (48)
Abdominal/flank pain (45)
Urinary tract infection (36)
Shortness of breath/pulmonary embolism (23)
GI complaints (23)
Penile/scrotal swelling (19)
Lymphocele (57)
Fever (46)
Urinary tract infection (32)
Urinary retention (17)
Deep venous thrombosis (17)
Pulmonary embolism (14)
Abscess (15)
Urinoma/urine leak (11)
Abdominal pain (10)
Wound infection (7)
Reoperation after discharge
(other than incision of
BNC or internal urethrotomy)
44 (1.0):
RP 21 (0.6)
LP 23 (2.0), p < 0.001
ER visit 652 (14.2):
RP 454 (13.1)
LP 198 (17.5), p < 0.001
Readmission240 (5.2):
RP 154 (4.5)
LP 86 (7.6), p < 0.001
BNC = bladder neck contracture; ER = emergency room; GI = gastrointestinal; LP = laparoscopic prostatectomy; RP = open radical retropubic prostatectomy.
E U RO P E A N U R O L O GY 57 ( 2 0 1 0 ) 3 7 1 – 3 8 6
374

Page 5

Table 3 – Association of preoperative, pathologic, and intraoperative parameters with medical and surgical complications
ParameterMedical complications Surgical complications
Present n (%) Absent n (%)
p value*Present n (%)Absent n (%)
p value*
Age
Median (IQR) 60.3 (56.0–65.1)59.4 (54.6–64.1) 0.01160.1 (55.7–64.9)59.3 (54.6–64.1)0.008
Ethnicity
White
African American
Other/unknown
402 (9.9)
42 (13.2)
23 (11.1)
3665 (90.1)
275 (86.8)
185 (88.9)
0.15 803 (19.7)
80 (25.2)
42 (20.2)
3264 (80.3)
237 (74.8)
166 (79.8)
0.063
BMI*
Median (IQR)27.9 (25.6–30.6) 27.6 (25.3–30.2)0.30 28.1 (25.5–30.9)27.5 (25.3–30.1) 0.003
Clinical stage
T1
T2
T3
Tx
312 (10.9)
133 (8.5)
21 (14)
1 (14.3)
2552 (89.1)
1438 (91.5)
129 (86)
6 (85.7)
0.027 585 (20.4)
303 (19.3)
36 (24)
1 (14.3)
2279 (79.6)
1268 (80.7)
114 (86)
6 (85.7)
0.50
Biopsy Gleason score
?6
7
?8
Preoperative PSA, ng/mlz
Median (IQR)
247 (9.2)
173 (11.2)
47 (12.7)
2432 (80.8)
1371 (88.8)
322 (87.3)
0.028523 (19.5)
319 (20.7)
83 (22.5)
2156 (80.5)
1225 (79.3)
286 (87.5)
0.34
5.7 (4.2–8.4)5.5 (4.1–7.7) 0.235.7 (4.2–8.1) 5.4 (4.1–7.7)0.022
Neoadjuvant hormonal therapy
Yes
No
20 (8.9)
447 (10.2)
204 (90.1)
3921 (89.8)
0.53 45 (20.1)
880 (20.1)
179 (79.9)
3488 (79.9)
0.98
ASA score
1
2
3–4
50 (10.0)
344 (9.8)
73 (12.6)
451 (90.0)
3162 (90.2)
508 (87.4)
0.13 83 (16.6)
698 (19.9)
144 (24.8)
418 (83.4)
2808 (80.1)
437 (75.2)
0.003
Modified Charlson score
0
1
2
3
4
?5
Hypertension
Yes
No
33 (8.2)
144 (8.6)
194 (10.8)
71 (12.6)
19 (15.2)
6 (16.7)
370 (91.8)
1529 (91.4)
1595 (89.2)
492 (87.4)
106 (84.8)
30 (83.3)
0.02565 (16.1)
299 (17.9)
379 (21.2)
142 (25.2)
28 (22.4)
11 (30.6)
338 (83.9)
1374 (82.1)
1410 (78.8)
421 (74.8)
97 (77.6)
25 (69.4)
<0.001
200 (11.7)
267 (9.3)
1514 (88.3)
2611 (90.7)
0.010 378 (22.1)
547 (19.0)
1336 (77.9)
2331 (81.0)
0.013
Hypercholesterolemia/Hypertriglyceridemia
Yes
No
164 (10.1)
303 (10.2)
1459 (89.9)
2666 (89.8)
0.91 313 (19.3)
612 (20.6)
1310 (80.7)
2357 (79.4)
0.28
Diabetes
Yes
No
45 (14.5)
422 (9.9)
266 (85.5)
3859 (90.1)
0.009 78 (25.1)
847 (19.8)
233 (74.9)
3434 (80.2)
0.025
Coronary artery disease
Yes
No
33 (10.1)
434 (10.2)
295 (89.9)
3830 (89.8)
0.9568 (20.7)
857 (20.1)
260 (79.3)
3407 (89.9)
0.78
Prior MI
Yes
No
11 (11.2)
456 (10.1)
87 (88.8)
4038 (89.9)
0.73 16 (16.3)
909 (20.2)
82 (83.7)
3585 (79.8)
0.34
Valvular heart disease
Yes
No
6 (30.0)
461 (10.1)
14 (70.0)
4111 (89.9)
0.003
0.006
9 (45.0)
916 (20.0)
11 (55.0)
3656 (80.0)
0.005
Dysrhythmia
Yes
No
23 (17.3)
444 (10.0)
110 (82.7)
4015 (90.0)
0.015
0.48
30 (22.6)
895 (20.1)
103 (77.4)
3564 (79.9)
0.48
Procoagulable disorder
Yes
No
11 (20.0)
456 (10.1)
44 (80.0)
4081 (89.9)
<0.001
0.16
19 (34.5)
906 (20.0)
36 (65.5)
3631 (80.0)
0.007
EUROPEAN UR OLOGY 57 (2010) 371– 386
375