results with a recently reported series of laparoscopic and robotic RRP.

Transcription

1 . JOURNAL COMPILATION 2008 BJU INTERNATIONAL Urological Oncology CLAVIEN CLASSIFICATION OF COMPLICATIONS OF OPEN RADICAL PROSTATECTOMY CONSTANTINIDES et al. BJUI BJU INTERNATIONAL Short- and long-term complications of open radical prostatectomy according to the Clavien classification system Constantinos A. Constantinides, Stavros I. Tyritzis, Andreas Skolarikos*, Evangelos Liatsikos, Anastasios Zervas and Charalambos Deliveliotis* 1st and *2nd University Departments of Urology, Athens University Medical School, Laiko and Sismanoglio Hospital, Athens, and Department of Urology, Patras University Medical School, Patras, Greece Accepted for publication 18 June 2008 Study Type Harm (case series) Level of Evidence 4 OBJECTIVE To assess the use of the Clavien classification system in documenting the complications related to open retropubic radical prostatectomy (RRP). PATIENTS AND METHODS The medical records of 995 patients, who had open RRP during a period of 7 years, were reviewed retrospectively. Short- and long-term complications were classified according to the recently revised Clavien classification system. We also compared the results with a recently reported series of laparoscopic and robotic RRP. RESULTS The overall complication rate was 26.9%; Grade I, Id, II, IIIa, IIIb and V complications were recorded in 3.4%, 3.9%, 12.8%, 2.6%, 3.8% and 0.3% of cases, respectively. Rectal injuries (10) and postoperative wound infections (24) were included in the Grade I category. Anastomotic leakage was recorded in 39 patients and rated as Grade Id. Grade II included cases of deep vein thrombosis (11), urinary tract infections (42), lymphorrhoeas (22) and haemorrhage requiring transfusion (53). Anastomotic strictures (26) and incisional hernias (38) were included in Grade IIIa and IIIb, respectively. Pulmonary embolism was fatal for three patients (0.3%) of Grade IV and V. CONCLUSIONS To avoid incoherence in reporting morbidity data, a reproducible and practical classification system is necessary. The Clavien system could provide, after refinement and validation, a common language among urologists. KEYWORDS radical retropubic prostatectomy, prostate cancer, Clavien classification system, complications INTRODUCTION Despite the development of laparoscopic approaches to retropubic radical prostatectomy (RRP) since 1992, when it was first described by Schuessler et al. [1], traditional RRP still remains the reference standard for managing localized prostate cancer [2]. The comparison between the types of this procedure is a source of significant debate among authors, with often conflicting or comparable results. However, the comparison is biased by several variables, e.g. surgeon experience, accumulation of cases, etc. This bias can be overcome by using classification systems in reporting the complications. The classification system of Clavien et al. [3] offers a standardized way of reporting negative outcomes; this system was introduced in 1992 and was initially used for complications associated with cholecystectomy. In the present study of RRPs we used the recently revised Clavien system, as published by Dindo et al. in 2004 [4], with the goal of suggesting a common method for an objective quality assessment in urological surgery worldwide. PATIENTS AND METHODS The medical records of 1161 patients, who had a RRP in three different academic units over a period of 7 years (January 2000 to April 2007), were retrospectively reviewed, with the intention to similarly document immediate, short- and long-term medical and surgical complications. Patients with inadequate data or lost to follow-up were excluded. The total number of functionally evaluable patients was 995; the mean (SD, range) follow-up was 36.8 (16.7, 6 75) months. Patient data were recorded by a senior resident of each department, using a customized patient flow-chart. All laboratory data were recorded in a computerized hospital data system, used in our department by Strict definitions of complications were used for the recording of data in the departments participating in the study. Bleeding and blood transfusion were determined when haemoglobin levels were 9 g/dl. Lymphorrhoea was specified as >100 ml of lymph drainage daily. Anastomotic strictures were established after cysto-urethroscopic evaluation 6 months after RRP. Deep venous thrombosis (DVT) was recorded if diagnosed 336 JOURNAL COMPILATION 2008 BJU INTERNATIONAL 103, doi: /j x x

2 CLAVIEN CLASSIFICATION OF COMPLICATIONS OF OPEN RADICAL PROSTATECTOMY Grade I II III IIIa IIIb IV IVa IVb V d Definition Any deviation from the normal course after surgery, with no need for pharmacological treatment or surgical, endoscopic and radiological interventions. Allowed therapeutic regimens are: drugs as anti-emetics, antipyretics, analgesics, diuretics, electrolytes, and physiotherapy. This grade also includes wound infections opened at the bedside. Requiring pharmacological treatment with drugs other than such allowed for grade I. Blood transfusions and total parenteral nutrition are also included. Requiring surgical, endoscopic or radiological intervention Intervention not under general anaesthesia Intervention under general anaesthesia Life-threatening complication (including CNS complications*) requiring IC/ICU management Single-organ dysfunction (including dialysis) Multi-organ dysfunction Death of the patient If the patient has a complication at the time of discharge, suffix d (for disability) is added to the respective grade of complication, indicating the need for a follow-up to fully evaluate the complication TABLE 1 The Clavien classification system revised by Dindo et al. [4] *Brain haemorrhage, ischaemic stroke, subarrachnoidal bleeding, but excluding transient ischaemic attacks. IC, intermediate care; ICU, intensive care unit. TABLE 2 The baseline characteristics of the 995 patients Characteristic Mean (SD, range) or n Age, years 63.2 (50 74) PSA level, ng/ml 5.73 (1.24, ) Stage pt2a 283 pt2b 312 pt3a 212 pt3b 83 pt3c 90 pt4 15 Gleason score by Doppler ultrasonography. All UTIs were characterized as mild infections, diagnosed by a positive urine culture. An obturator-sampling lymphadenectomy was performed in 680 patients (68.3%), who met the following criteria: PSA level 8 ng/ml, Gleason score >7, or intraoperative findings showing disease extent. Complications concerning a lymphadenectomy were adjusted to the number of patients undergoing this operation. Complications were rated as immediate when occurring during RRP, short-term when within the first 30 days after RRP and long-term when beyond that time. All complications were classified according to the Clavien classification system revised by Dindo et al. [4]. The original classification system consisted of four severity grades. The recently revised system emphasizes on the risk and invasiveness of the therapy used to treat a complication and comprises mainly four important modifications: (i) life-threatening complications were differentiated from complications treated in the ward; (ii) CNS complications were included in the same category (Grade IV); (iii) the length of hospital stay is no longer considered in the ranking; and (iv) complications that can potentially lead to long-lasting disability are highlighted by a suffix d (for disability ). Consequently the new classification system comprises five severity grades (Table 1). In the earlier patients (until 2001) the surgical technique comprised a vesico-urethral anastomosis (in 323) with no attempt to preserve the neurovascular bundles, while in later patients (838) we used a urethrourethral anastomosis with a unilateral or bilateral neurovascular bundle-sparing technique, depending on the disease extent. Continence, erectile dysfunction rates and oncological outcomes were not subject of this report. RESULTS Table 2 presents the baseline characteristics of the patients included in the study. According to the revised Clavien classification, there were 268 complications in 995 patients, with an overall complication rate of 26.9%. The incidence of complications during RRP was 1%, with 10 rectal injuries, that were identified and managed immediately with a two-layer suture (mucosa and muscle). The early complication rate was 21.4%, and three patients (0.3%) died during the first week after RRP, from pulmonary embolism. In all, 39 patients (3.9%) had a persistent anastomotic leakage, which was treated by prolonged bladder drainage (>14 days); 24 patients (2.4%) sustained a wound infection managed successfully with antibiotic treatment and bed-side intervention. DVT was diagnosed in 11 patients (1.1%), and UTI documented and managed conservatively in 42 (4.2%). A blood transfusion was necessary in 53 patients (5.3%), due to intraoperative bleeding and decreased haemoglobin levels, whereas four had persistent asymptomatic lymphorrhoea, following an ilio-obturator lymphadenectomy, which resolved uneventfully with continued suction drainage for a mean 7 days; no open revision was needed in any patient. The late complications were an anastomotic stricture (2.6% incidence) and incisional hernia in 38 of 995 patients (3.8%). Anastomotic strictures were treated by endoscopic bladder neck incision. A stratification of the negative postoperative events by the Clavien system is listed in Table 3. DISCUSSION The evaluation of surgical techniques through their complication rates, using standardized, valid and reliable methods, could be the cornerstone for the development and improvement of these techniques, and for JOURNAL COMPILATION 2008 BJU INTERNATIONAL 337

3 CONSTANTINIDES ET AL. establishing benchmarks for training and experience. However, it is unavoidable that there are major differences in reporting the perioperative negative outcomes, as there is no consensus. According to Touijer and Guillonneau [5], complication rates for RRP and its laparoscopic surrogate (LRP) vary significantly, at %. Also, Artibani et al. [6] concluded that LRP does not provide significant advantages in terms of perioperative morbidity over the traditional retropubic approach, whereas Remzi et al. [7] reported the opposite. Both studies compared the morbidity of RRP and LRP. These are good examples of the incoherence of morbidity data worldwide, and this fact hampers the accurate monitoring and measurement of the surgical adverse effects. In 1997, Dillioglugil et al. [8] introduced a severity score which was assigned to each adverse event as follows: (i) causes symptoms but requires no active treatment and leaves no sequelae; (ii) requires noninvasive treatment or causes morbidity; (iii) requires prolonged or secondary hospitalization, invasive treatment or surgical repair of an intraoperative complication; and (iv) requires intensive treatment or major surgical intervention for a severe, life-threatening or fatal event. The lack of subcategorization could result in an approximation for reporting negative events, as for example, a death would be classified as severity score 4, the same score as with a successful re-operation for bleeding. TABLE 3 Surgical complications stratified by the Clavien classification system Clavien grade/complication n (%) Management During surgery I Rectal injury 10 (1) Two-layer suture Short-term I Wound infection 24 (2.4) Antibiotics; bed-side management Id, Anastomotic leakage 39 (3.9) Prolonged bladder drainage II DVT 11 (1.1) Conservative UTI 42 (4.2) Antibiotics Lymphorrhoea 22 (3.2) Suction drainage Bleeding 53 (5.3) Blood transfusion V Death 3 (0.3) Long-term IIIa Anastomotic stricture 26 (2.6) Endoscopic bladder neck incision IIIb Incisional hernia 38 (3.8) Surgical repair TABLE 4 Comparison of published complications after open RP, LRP and robotic RP LRP Robotic RP RRP Grade [15] [9] [10] [11] [12] [13] [14] Present study I Id II IId 0.3 IIa IIb 5.67 III IIIa IIIb IVa IVb V One of the classification systems that tries to standardize and compare complications using a reliable and valid method is that reported by Clavien et al. [3] in The system was initially intended for low-morbidity procedures. Since then, it has been modified to overcome several disadvantages. The most recent update by Dindo et al. in 2004 [4] focused on the risk and invasiveness of the method used to treat negative sequelae. Moreover, it was applied in a large series of patients (>6000), resulting in the validation of its reproducibility and credibility. The Clavien classification system was used previously by several authors to present their surgical negative events of LRP) and robotic LRP. Guilloneau et al. [9] reported complication rates for Grades I, IIa, IIb and III of 1.95%, 10.52%, 5.67% and 0.17%, respectively; Conzalgo et al. reported 8.1%, 1.2% 3.7% and 0.8% for Grades II, IIIa, IIIb and IVa complications, respectively [10]. Stolzenburg et al. [11] applied the Clavien system in 900 patients who had an endoscopic extraperitoneal RP, reporting rates of 4.6%, 2% 6.2% and 0.3% for Grade I, II, III and IVa. The group from the Vattikuti Urology Institute using robotic LRP also used the Clavien classification system. Kaul et al. [12], in 154 patients, reported rates of 2.5%, 5% 0.6% and 2% for Grades Id, I, IIb and IIIa. Bhandari et al. [13], in a study with 300 patients, recorded 3.7% and 2% of Grade I and II complications, respectively. Similar results stratified by the Clavien system were also reported by Atug et al. [14] and Permpongkosol et al. [15]; the summarized comparison of these results is shown in Table 4. As RRP remains the reference standard for managing localized prostate cancer, we decided to use the same classification system in the present series of patients. To our knowledge, this the first study to use the Clavien system to stratify the recorded morbidity of RRP. Our results are comparable with those reported for LRP, but there are two distinct limitations, which also often bias other studies. First, the present study was retrospective, incorporating different techniques of the same operation. As noted, the later patients had a urethro-urethral anastomosis, which reduced the leakage of the anastomosis and dramatically improved the continence rates. Second, the patients were not operated by the same surgeon. This latter fact introduces a bias in every study, because of the major influence of the surgeon s experience, as this is reflected by the volume of operations, in the outcome [16]. Indeed, surgeon experience is a predictor of the severity of complications as suggested by Hu et al. [17]. 338 JOURNAL COMPILATION 2008 BJU INTERNATIONAL

4 CLAVIEN CLASSIFICATION OF COMPLICATIONS OF OPEN RADICAL PROSTATECTOMY The Clavien classification systems has two shortcomings; it cannot evaluate the longterm aspects of the patients quality of life, e.g. continence, potency and disease recurrence (as presented by positive surgical margins and subsequent biochemical recurrence), which might be more important. The other shortcoming is that the system does not include the comorbidity of the patient, which is a stronger predictor than age of almost all categories of early complications after RP, as suggested by Alibhai et al. [18] in their study of patients. The same assumption was recently reported by the CaPSURE study [19]. Other possible predictors of the severity of complications after surgery are prostate size and the body mass index of the patient [20], and the interval between the prostate biopsy and the RP [21]. Another variable that could affect the long-term complications and especially those associated with potency and continence, is the additional therapy during the follow-up of the patient, e.g. radiation or hormonal therapy [22]. Finally, the Clavien classification system commits the investigator to report insignificant events after surgery, e.g. fever or an episode of vomiting. These events do not represent a clear deviation from the natural course. The system should focus on the major events, those that test the surgical technique and ability, and whose presentation will reflect on the patient s quality of life after surgery. A future modification of the Clavien system should include these issues and correct these possible drawbacks. A recent effort was made by Kocak et al. [23] to modify the Clavien system for use in livedonor nephrectomy, a procedure with several singularities. Unfortunately it is not so practical to create several modifications of the same system to fully cover the spectrum of urological surgery. Our objective was to create a platform for discussion and stimulate further analysis of this system, and not to impose one. Up- or down-rating of the complications and final improvement of the Clavien classification system should be decided after peer review. However, the need remains for a widely accepted and reliable algorithm that will allow a more subjective criticism. In conclusion, incoherence of the reported complications and several other variables of surgery has always been an obstacle to an objective evaluation. A possible starting effort has been made by a few authors to initiate a standardized comparison of the studies published by using the Clavien classification system. More studies of the application and modification of this system are necessary for its validation. CONFLICT OF INTEREST None declared. REFERENCES 1 Schessler WW, Schulam PG, Clayman RV, Kavoussi LR. Laparoscopic radical prostatectomy: initial short-term experience. Urology 1997; 50: Aus G, Abbou CC, Bolla M et al. EAU guidelines on prostate cancer. Eur Urol 2005; 48: Clavien PA, Sanabria JR, Strasberg SM. Proposed classification of complications of surgery with examples of utility in cholecystectomy. Surgery 1992; 111: Dindo D, Demartines N, Clavien PA. Classification of surgical complications. A new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004; 240: Touijer K, Guillonneau B. Laparoscopic radical prostatectomy. A critical analysis of surgical quality. Eur Urol 2006; 49: Artibani W, Grossob G, Novaraa G et al. Is laparoscopic radical prostatectomy better than traditional retropubic radical prostatectomy? An analysis of perioperative morbidity in two contemporary series in Italy. Eur Urol 2003; 44: Remzi M, Klinger HC, Tinzl MV, Fong YK, Kiss B, Marberger M. Morbidity of laparoscopic extraperitoneal versus transperitoneal radical prostatectomy versus open retropubic radical prostatectomy. Eur Urol 2005; 48: Dillioglugil O, Leibman BD, Leibman NS, Kattan MW, Rosas AL, Scardino PT. Risk factors for complication and morbidity after radical retropubic prostatectomy. J Urol 1997; 157: Guilloneau B, Rozet F, Cathelineau X et al. Perioperative complications of laparoscopic radical prostatectomy: the Montsouris 3-year experience. J Urol 2002; 167: Gonzalgo ML, Pavlovich CP, Trock BJ, Link RE, Sullivan W, Su LM. Classification and trends of perioperative morbidities following laparoscopic radical prostatectomy. J Urol 2005; 174: Stolzenburg JU, Rabenalt R, Do M et al. Categorization of complications of endoscopic extraperitoneal and laparoscopic radical prostatectomy. World J Urol 2006; 24: Kaul S, Savera A, Badani K, Fumo M, Bhandari A, Menon M. Functional outcomes and oncological efficacy of Vattikuti Institute prostatectomy with Veil of Aphrodite nerve-sparing: an analysis of 154 consecutive patients. BJU Int 2005; 97: Bhandari A, McIntire L, Kaul S, Hemal AK, Peabody JO, Menon M. Perioperative complications of robotic radical prostatectomy after the learning curve. J Urol 2005; 174: Atug F, Castle P, Srivastav SK, Burgess SV, Thomas R, Davis R. Prospective evaluation of concomitant lymphadenectomy in robot-assisted radical prostatectomy: preliminary analysis of outcomes. J Endourol 2006; 20: Permpongkosol S, Link RE, Su LM et al. Complications of 2775 urological laparoscopic procedures: J Urol 2007; 177: Denberg TD, Flanigan RC, Kim FJ, Hoffman RM, Steiner JF. Self-reported volume of radical prostatectomies among urologists in the USA. BJU Int 2007; 99: Hu JC, Elkin EP, Pasta DJ et al. Predicting quality of life after radical prostatectomy. Results from CaPSURE. J Urol 2004; 171: Alibhai SM Leach H, Tomlinson M et al. 30-day mortality and major complications after radical prostatectomy: influence of age and comorbidity. J Natl Cancer Inst 2005; 97: Arredondo SA, Elkin EP, Marr PL et al. Impact of comorbidity on health-related quality of life in men undergoing radical prostatectomy: data from CaPSURE. Urology 2006; 67: Singh A, Fagin R, Shah G, Shekarriz B. Impact of prostate size and body mass index on perioperative morbidity after laparoscopic radical prostatectomy. J Urol 2005; 173: JOURNAL COMPILATION 2008 BJU INTERNATIONAL 339

5 CONSTANTINIDES ET AL. 21 Lee DK, Allareddy V, O donnell MA, Williams RD, Konety BR. Does the interval between prostate biopsy and radical prostatectomy affect the immediate postoperative outcome? BJU Int 2006; 97: Simoneau AR. Treatment- and diseaserelated complications of prostate cancer. Rev Urol 2006; 8: S Kocak B, Koffron AJ, Baker TB et al. Proposed classification of complications after live donor nephrectomy. Urology 2006; 67: Correspondence: Stavros I. Tyritzis, 17 Agiou Thoma str., Athens, Greece. statyr@fre .gr Abbreviations: (L)(R)RP, (laparoscopic) (retropubic) radical prostatectomy; DVT, deep venous thrombosis. 340 JOURNAL COMPILATION 2008 BJU INTERNATIONAL