Does the presence of significant risk factors affect perioperative outcomes after robot-assisted radical cystectomy?

Authors


Khurshid A. Guru, Director, Robotic Surgery, Department of Urologic Oncology, Roswell Park Cancer Institute, Elm & Carlton Streets, Buffalo, NY 14263, USA.
e-mail: khurshid.guru@roswellpark.org

Abstract

OBJECTIVE

To evaluate the effect of preoperative risk factors on perioperative outcomes up to 3 months after robot-assisted radical cystectomy (RARC), as RC continues to be associated with a high rate of morbidity and mortality.

PATIENTS AND METHODS

From 2005 to 2007, 66 consecutive patients had RARC at Roswell Park Cancer Institute. Patient demographics, preoperative risk factors and complications up to 3 months after RARC were reviewed from a prospective quality-assurance database. Patients were stratified into high- and low risk groups based on age, previous abdominal surgery, chronic obstructive pulmonary disease (COPD), body mass index (BMI), Revised Cardiac Risk Index (RCRI) and American Society of Anesthesiologists (ASA) score.

RESULTS

Age, previous abdominal surgery, COPD, BMI, RCRI score and ASA score did not significantly influence complications during or up to 3 months following RARC (P > 0.05). Advanced age was associated with a higher RCRI score (P = 0.014) and an increased likelihood of admission to the Intensive Care Unit (P = 0.007). A higher ASA score was associated with an increased overall hospital stay (P = 0.039). Previous abdominal surgery was associated with more frequent unscheduled postoperative clinic visits (P = 0.014). Operative duration did not significantly influence complication rates (P > 0.05). Fifteen of 62 patients (24%) had a major complication, while 15 (24%) had minor complications within 3 months of surgery. The reoperation rate was 11% and the overall mortality rate was 1.6%.

CONCLUSIONS

RARC appears to be well tolerated, independent of comorbid risk factors such as age, BMI, RCRI and ASA score.

Abbreviations
(RA)RC

(robot-assisted) radical cystectomy

RCRI

Revised Cardiac Risk Index

ASA

American Society of Anesthesiologists

EBL

estimated blood loss

ICU

Intensive Care Unit

PAS

previous abdominal surgery

COPD

chronic obstructive pulmonary disease

BMI

body mass index.

INTRODUCTION

Radical cystectomy (RC) is associated with a perioperative mortality of 2–4% and a 30% risk of developing at least one complication afterward [1,2]. This is largely because patients with bladder carcinoma are typically elderly with significant comorbidities.

Using receiver operating characteristic curve analysis, Lee et al.[3] found the Revised Cardiac Risk Index (RCRI) to be a better prognostic aid in predicting cardiac complications after RC than previous cardiac risk indices and the American Society of Anesthesiologists (ASA) score. While previous studies examined the association between the ASA score and complications after RC, we used the RCRI score to specifically examine the postoperative cardiac risk.

PATIENTS AND METHODS

Using a prospective quality-assurance database, we reviewed 66 consecutive patients who had robot-assisted RC (RARC) at our institution from October 2005 to December 2007. Four patients whose procedures were aborted due to locally advanced inoperable disease were excluded from the study.

Patients were stratified into high- and low risk groups based on the RCRI score. For the purposes of this study, patients were defined as low risk if they had a RCRI score of <2 and high risk if ≥2 (Table 1). The effects of age (<65 vs ≥65 years), ASA score (2 vs 3), body mass index (BMI; ≤28 vs >28 kg/m2) and operative duration (<360 vs ≥360 min) on total complications were also assessed. Complications (major and minor) after RARC were reported for up to 3 months afterward. Univariate analysis was used to determine which preoperative risk factors correlated with the complication rates after RARC. The complications were graded for severity using the Dindo-Clavien system (Table 2) [5]. RARC was performed using the previously reported technique [6]. Estimated blood loss (EBL) was determined by the anaesthesiologist for each patient.

Table 1.  The RCRI
RCRI total score% Risk of major cardiac complication
  1. Each risk factor is assigned 1 point: High-risk surgery, ischaemic heart disease, congestive heart failure, cerebrovascular disease, insulin therapy for diabetes, preoperative serum creatinine level >2.0 g/dL.

00.4
10.9
27
>211
Table 2.  Complication grades based on the Dindo-Clavien Classification
GradesOrgan SystemDescriptionN complications
IOtherWound infection treated by opening of the wound at the bedside3
IICardiacAtrial fibrillation1
Myocardial infarction1
RespiratoryPulmonary embolism2
GastrointestinalClostridium difficile colitis2
Ileus2
RenalUTI3
OtherSepsis4
Major depression requiring hospitalization2
IIIaRenalUreteric stricture1
Acute renal failure3
OtherLabial abscess1
IIIbGastrointestinalBowel obstruction4
Appendicitis1
RenalUrinary fistula1
 OtherWound dehiscence2
Return to operating room7
IVa NoneNone
V Death1

RESULTS

The mean (range) age of the patients was 67 (36–90) years. Advanced age was associated with both a higher RCRI score and an increased likelihood of Intensive Care Unit (ICU) admission. Accordingly, patients with an RCRI score of 1 had a mean age of 65 years, whereas those with an RCRI score of 2 or 3 had a mean age of 74 years (P = 0.014). Of patients aged <65 years, 44% were admitted to the ICU, vs 63% of patients aged >65 years (P = 0.007). Notaby, 55% of all patients ultimately required an ICU admission, of whom only three stayed for >2 days (Table 3).

Table 3.  Perioperative outcomes with stratification according to RCRI score
Mean (sd)* or n (%) variableRangeRCRI <2RCRI ≥2P
  • *

    Continuous variables,

  • †Categorical variables, as cumulative frequency (relative frequency).

Operative duration, min178–827364 (124)348 (131)0.593
EBL, mL100–3000507 (349)857 (811)0.079
Transfusion, units0–116 (13)4 (28)0.237
Total complications 21 (45)6 (40)1
Hospital stay, days4–379 (8) 11 (8)0.093
Postop ICU admission, days0–1024 (51)10 (67)0.377
ICU stay, days    
 ≥2 45 (96)14 (93)0.571
 3–5 1 (2)0 
 >5 1 (2)1 (7) 
Early complications 18 (38)5 (33)1
Reoperation 4 (9)3 (20)0. 345
Unscheduled visits 14 (30)2 (13)0.314
Deaths 01 (7)0.242

Age, previous abdominal surgery (PAS), chronic obstructive pulmonary disease (COPD), BMI, ASA score and RCRI score did not significantly influence complications during or after RARC for up to 3 months after surgery (Table 4). A history of PAS was only associated with more frequent unscheduled clinic visits after RARC. Among patients with PAS, 34% reported at least one unscheduled clinic visit, vs only 12% of patients with no such history. A higher ASA score was only associated with a longer hospital stay; patients with an ASA score of 2 had a mean stay of 8 days, whereas those with an ASA score of 3 had a mean stay of 12 days (P = 0.039); the overall mean hospital stay was 9 days.

Table 4.  Univariate analysis of perioperative risk factors and associated complication rates
Variable (range) (n)% totalOdds ratio (95% CI)P
Complications  1
Age, years (36–90)   
 ≤65 (27)37.01.005 (0.355–2.840) 
 >65 (35)37.1 1
ASA (2–3)   
 2 (44)36.41.114 (0.360–3.445) 
 3 (18)38.9 1
RCRI (1–3)   
 1 (47)38.30.806 (0.237–2.740) 
 2, 3 (15)33.3  
PAS (38)44.72.429 (0.790–7.470)0.177
COPD (4)250.546 (0.053–5.575)1
BMI (17–45)   
 ≤28 (31)43.80.551 (0.193–1.571)0.302
 >28 (31)30.0  
Operative duration, min (178–827)   
 <360 (30)32.31.517 (0.538–4.279)0.600
 ≥360 (32)41.9  

The RCRI score was associated with a trend towards increased intraoperative blood loss, although this was not statistically significant (P = 0.079). Patients with an RCRI score of 1 had a mean EBL of 507 mL, vs 857 mL in those with an RCRI score of 2 or 3 (P = 0.079; Table 3). There were no significant differences in outcomes or complication rates when patients were stratified according to RCRI score. Four patients had both major and minor complications, and only six had more than one complication. Among the major complications were four small bowel obstructions, four with sepsis, one myocardial infarction, two pulmonary embolism/deep vein thrombosis, seven returns to the operating room, two with wound dehiscence and one death. The minor complications are listed in Table 2. Only three complications were grade I while 34 were grades II and III, according to the Dindo-Clavien classification (Table 2). There were no intraoperative complications, and operative duration did not significantly influence complication rates (P = 0.600). The reoperation rate was 11%. Among patients requiring reoperation, two had small bowel resections, two needed closure of wound dehiscence, two had haemorrhage after surgery and one had a urinary fistula repaired. Overall, there was just one death (1.7%), in a 90-year-old patient (RCRI score 2) who required reoperation after developing a small bowel obstruction; he developed septic shock and died.

DISCUSSION

Patients presenting for RC are often of advanced age and have significant comorbid conditions. It is critically important to identify those risk factors which place patients at an increased risk of postoperative complications, and where possible, favourably affect those risks before surgery. However, this is often not possible, and it is therefore necessary to stratify a patient’s risk before deciding to proceed with surgery.

McLaughlin et al.[7] found various preoperative risk factors and intraoperative measures that affected complications up to a month after urological surgery. Although not specific for RC, their results showed the importance of focusing on risk factors such as diabetes, congestive heart failure and ischaemic heart disease in determining outcomes after surgery. Assessing specifically the complications after RC, various series evaluated the effects of age, ASA score, operative duration, EBL and transfusion requirement [1,8,9].

Current reports lack consensus on the effect of age and comorbidity on complications after RC. Hollenbeck et al.[1] reported that patients were 1.3 times more likely to develop a complication after RC for every 10-year increase in age. Fairey et al.[10] retrospectively reviewed clinical outcomes in 314 patients who had RC and found that increasing age was not associated with adverse clinical outcomes during a 3-month period. Nonetheless, increased comorbidities were associated with an increased risk of 90-day mortality and of early complications after RC, either major or minor.

A comparative study of current reports in minimally invasive surgery provides an insight into various factors responsible for outcomes. Zhao et al.[11] reported no increase in complications in patients aged >75 years undergoing laparoscopic urological procedures, other than an increased hospital stay. Matin et al.[12] reported similar findings in patients aged >65 years undergoing laparoscopic renal and adrenal surgery. While in the present series we did not find advanced age to be associated with an increased hospital stay, we did find an increased likelihood of an ICU admission. ICU admission was initially part of routine postoperative care. Subsequently, the decision was based on the severity of comorbidities, EBL, length of procedure and overall intraoperative course. ICU admission was probably confounded by the increased risk of comorbidity with advanced age. In the present series there was also a positive association between age and the RCRI score.

The ASA score has been shown to be a significant predictor of morbidity in surgical patients [13]. Hollenbeck et al.[1] showed that patients with a higher ASA score were 1.3 times more likely to develop a complication than those with a low ASA score. However, the ASA score itself is an imperfect indicator of physical status, as it neglects the effect of surgery on patient outcomes, and is inconsistently interpreted by clinicians [14,15]. Nonetheless, its predictive utility is well established [13,16]. Interestingly, only the length of hospital stay was affected by a higher ASA score in the present series. Similar to the findings of Matin et al.[12], the ASA score had no effect on the complication rate in the present patients. Also patients undergoing RC were 1.2 times more likely to develop a complication for every extra hour of operative time [1]. The present series had no difference in complications based on total operative duration (range 178–827 min).

EBL and transfusion requirements have often been cited as two of the most significant predictive variables for postoperative complications [1]. Patients with a transfusion during or after surgery were 1.4 times more likely to develop a complication than those who were not transfused [1]. A minimally invasive surgical approach might decrease complications after surgery by significantly reducing intraoperative blood loss, and thus the need for transfusions. In the present series the mean EBL was 591 mL with an overall transfusion rate of 16%, which can be favourably compared to 38% in a recent open series of 553 patients [4].

In the present series, PAS, COPD and BMI had no significant associations with complications. Several series have reported on the effects of an elevated BMI on open RC. With a high BMI, Lee et al.[17] reported increased blood loss, prolonged surgery and increased complication rates. Similarly, Chang et al.[18] reported that BMI was associated with increased blood loss, but ultimately found no association between BMI and complication rates or length of hospital stay. In the present series, the lack of influence of obesity on the complication rate is encouraging. A recent study by Butt et al.[19] found no differences in operative times, EBL and complication rates across various ranges of BMI in patients who had RARC.

The benefits of minimally invasive surgery, including decreased EBL, decreased postoperative pain and opiate use, earlier mobility and return of bowel function, might also lead to better outcomes in RARC [20,21]. An 11% rate of repeat surgery for inpatient complications in open-assisted (laparoscopic cystectomy and open diversion), vs 29% in purely laparoscopic RC, was reported by Haber et al.[22]. Only seven of the present patients needed reoperation.

Minimizing complications after surgery results in a significant decrease in length of stay and total hospital charges. Konety et al.[23] concluded that the occurrence of each complication appeared to increase total charges by 35%, and there was a significant increase of 216% in patients with three complications. In their series, patients with persistent postoperative fistula had the highest increase in total charges, while patients with a respiratory complication and a postoperative infection had the highest risk of death. None of the present patients had a respiratory complication, while only one had a urinary fistula that needed reoperation.

Lowrance et al.[4] reviewed 553 consecutive patients after open RC and reported rates of 38% and 7.6% for minor and major complications up to a month after RC. In the present series, the major complication rate was 24%, with an equal minor complication rate, based on a similar strict analysis, with a prolonged follow-up of up to 3 months after surgery. Colombo et al.[24], during their analysis of all laparoscopic urological surgery, found completely laparoscopic RC to be an independent risk factor for postoperative complications. They found that the development to open-assisted laparoscopic RC effectively decreased the complication rates. The present series advocates a similar approach using RARC and extended lymph node dissection, while using an open incision for urinary diversion. Our major complication rate is higher than those published previously and reflects our early experience with robot-assisted surgery for bladder cancer. Several early open series have also had longer operative times, increased blood loss, prolonged hospital stay and higher rate of complications, which over time have significantly reduced and lead to an optimal open RC standard [1,2,4,8].

The present study has several limitations. All data were derived from one institution and with relatively few patients (compared with major high-volume centres), reflecting our early experience, thus limiting the power of our study. RARC is still in its infancy and collaboration to obtain combined multicentre data from various academic institutions is being pursued under the auspices of the International Robotic Cystectomy Consortium [25].

In conclusion, RARC appears to be well tolerated, with complication rates independent of comorbid risk factors such as age, BMI, RCRI and ASA score.

CONFLICT OF INTEREST

Khurshid A. Guru is a paid speaker for Intuitive Surgical Inc.

Ancillary