Outcome of Transurethral Resection of Bladder Tumor: Does Antiplatelet Therapy Really Matter? Analysis of a Retrospective Series

Abstract

Introduction:

Transurethral resection of bladder tumor (TURBT) is considered to be at a moderate or high risk of bleeding during surgical procedure. The number of patients on antiplatelet (AP) drugs has been increasing; we wanted to assess their impact on the outcome of patients undergoing scheduled TURBT.

Materials and Methods:

A retrospective assessment of noninferiority of 450 consecutive procedures performed between April 2013 and June 2015 was conducted. Patients were divided in two groups: naive or AP drug users. The primary endpoint was the average length of stay (ALOS). Noninferiority was set at 1 day. A subgroup analysis comparing the acetylsalicylic acid (ASA) group and clopidogrel group to the naive group was performed. Multivariate analysis was performed to find the determinants of a longer ALOS. Chi-square or Fisher tests were used to analyze categorical variables, and Student's or Mann–Whitney tests were used to analyze quantitative variables.

Results:

We included 325 patients who underwent TURBT: 117 received AP drugs (ASA, 85; clopidogrel, 32) and 208 were naive to AP drugs (of whom 117 were consecutively analyzed). The ALOSs were 2.5 days (naive group) and 2.9 days (AP group). The subgroup analysis showed ALOSs of 2.6 days (ASA group) and 3.7 days (clopidogrel group). Clopidogrel therapy (odds ratio = 4.1 [1.7–9.6]) and the duration and depth of resection emerged as determinants of a longer ALOS in multivariate analysis. Perioperative management of AP therapies was achieved according to recommended practices.

Conclusions:

The ALOS of patients receiving AP drugs was not clinically different from naive patients. This result was identical for patients receiving ASA. However, clopidogrel increased the length of stay, making us question its use in perioperative management.

Introduction

More than 50,000 transurethral resections of bladder tumor (TURBTs) are performed each year in France.¹ The number of patients treated with antiplatelet (AP) drugs has been increasing because of the aging population and new indications of primary or secondary prevention of thromboembolic risk.² For example, every year, more than 1 million percutaneous coronary interventions (PCIs) are performed in Europe and the United States,^3,4 with 85% of patients⁵ receiving implants of a bare or covered stent and having a long-term indication for AP.

Moreover, the incidence peaks of urologic cancer, especially bladder cancer, and PCI are similar.⁶ The combination of these elements brings up the question of the antagonistic relationship between the thrombotic risk associated with discontinuation of the treatment and the hemorrhagic risk during an invasive procedure with patients receiving AP drugs.²

TURBT is considered as a surgical procedure associated with a moderate^6,7 or even a high² bleeding risk. Moreover, poorly controlled hemostasis and blood loss are associated with increased morbidity,⁷ a longer length of stay (LoS),⁸ and costs.⁶

Few studies have evaluated the impact of AP drugs in TURBT with a focus on the evaluation of bleeding. We aimed to study the overall impact of AP drugs in the management of patients undergoing TURBT using a concrete and practical determinant: the LoS.

Materials and Methods

This was a noninferiority retrospective study of all TURBTs performed by all surgeons from a tertiary hospital from April 2013 to June 2015. Exclusion criteria were patients hospitalized in the emergency unit, those receiving anticoagulant therapy, and those who underwent another surgical procedure, except internal urethrotomy for an intraoperative indication or repermeabilization of one or two ureters associated with the implementation of a Double-J stent. Our study was approved by our institutional review board.

The primary endpoint was the LoS, defined by the number of days spent in the unit postoperatively (i.e., the day of operation was considered as day 0). We set a clinically significant difference in the LoS as >1 day. Patients were initially divided into two groups: naive or AP drug users, since AP drugs were used during patients' home treatment.

From the average LoS of the first 50 patients, we calculated that we needed at least 160 patients (80 per group) to demonstrate noninferiority (<1 day) of the AP group compared to the naive group (with an α risk of 2.5% and a power study of 80%). A subgroup analysis was performed to compare the naive group with the acetylsalicylic acid (ASA) group and clopidogrel group.

TURBTs were performed using monopolar resectors (Ch26; Karl Storz, Tuttlingen, Germany or Olympus, Tokyo, Japan). Parameters collected were demographic data, medical background, LoS, perioperative management with AP drugs, characteristics of the resected tumor, type of anesthesia, blood test results, and postoperative outcome.

Chi-square or Fisher tests were used to analyze categorical variables, and Student's or Mann–Whitney tests were used to analyze quantitative variables. Data are expressed as a mean (standard deviation) or percentage (%). Differences with a p-value <0.05 were considered statistically significant. Logistic multivariate analysis was used to test the association between LoS and the explanatory variables. The explanatory variables with a p-value <0.2 were selected in univariate analysis. The LoS was considered short if it was ≤2 days, whereas it was considered long if it was >2 days. Statistical analysis was performed using SPSS Statistics software, version 22 (IBM Corp., Armonk, NY).

Results

We identified 450 TURBTs performed between April 2013 and June 2015. One hundred twenty-five procedures were excluded according to preestablished criteria. Of the 325 remaining procedures, 208 were performed in naive patients, and 117 were performed in patients receiving AP drugs (ASA, 85; clopidogrel, 32 [including 8 with dual therapy]). We analyzed all 117 patients receiving AP drugs and 117 consecutive patients naive to AP drugs. Characteristics of the patients in each group are presented in Table 1.

Table 1.

Characteristics of Patients in the Naive and Antiplatelet Groups

	Naive group (n = 117)	AP group (n = 117)
Age (years)	66 ± 13	74 ± 9^*
Male/female sex (%)	79/21	85/15
ASA score	2	3
BMI (kg/m²)	28.6	28.4
Medical history (%)
High blood pressure	40	63^*
Ischemic heart disease	1	37^*
Atrial fibrillation	4	11^*
IDDM	0	4^*
NIDDM	16	38^*
COPD	9	17
Tobacco use	26	16
Dyslipidemia	28	54^*
Stroke	3	16^*
Anesthesia (%)
General anesthesia	73	57^*
Spinal anesthesia	27	43^*
Tumor location/localization (%)
Trigona/neck of the urinary bladder	52	58
Left side	30	26
Right side	38	28
Posterior wall	44	39
Dome	20	17
Anterior wall	16	14
Multifocal	59	58
Average duration of resection (minutes)	28 ± 13	27 ± 13
Pathologic findings (%)
pTa	41	50
CIS (alone or associated)	4	13^*
pT1	12	19
pT2	13	9
Other	33	16^*
Muscle described on pathologic findings (%)	62	71
Weight of resected tumor (g)	7.3 ± 11	4.6 ± 5

The values are expressed as a mean (standard deviation) or percentage.

p < 0.05.

AP = antiplatelet; ASA = American Society of Anesthesiologists; BMI = body mass index; IDDM = insulin-dependent diabetes mellitus; NIDDM = non-insulin-dependent diabetes mellitus; COPD = chronic obstructive pulmonary disease; CIS = carcinoma in situ.

The average LoSs were 2.5 ± 1.2 days in the naive group and 2.9 ± 1.8 days in the AP group (p = 0.024). In the subgroup analysis, the average LoS was not statistically different between the two groups (ASA group: 2.6 ± 1.3 days; naive group: 2.5 ± 1.2 days; p = 0.445). However, patients receiving clopidogrel had an average LoS of 3.7 ± 2.5 days (p = 0.008). These differences are illustrated in Figure 1. The demographic characteristics of patients in both subgroups are shown in Table 2.

FIG. 1.

Average length of stay (days) according to the drug therapies. Comparisons are made with the naive/control group.

Table 2.

Morbidity of Transurethral Resection of Bladder Tumor Among the Naive, American Society of Anesthesiologists, and Clopidogrel Groups

	Naive group (n = 117)	ASA group (n = 85)	Clopidogrel group (n = 32)
Duration of bladder catheter placement (days)	2.2 ± 0.9	2.2 ± 0.9	2.7 ± 1.3^*
Failure of bladder catheter removal (%)	2	5	22^*
Blood transfusion (%)	1	1	9^*
Hemostatic operations (%)	0	1	13^*
Hb level preoperatively (g/dL)	14.2 ± 1.9	13.6 ± 2.1	13.0 ± 1.6^*
Hb level on day 1 (g/dL)	13.4 ± 1.8	13.2 ± 1.8	12.1 ± 1.7^*
Decrease in the Hb level on day 1 (g/dL)	−0.8 ± 0.8	−0.4 ± 1.1	−0.9 ± 1.1
Rehospitalization in the month following operation (%)	6	5	13
Clavien grade (%) II	9	6	25^*
IIIb	0	2	13^*
CNVE (%)	0.9	0	3.1

Values are expressed as a mean (standard deviation) or percentage.

p < 0.05.

ASA = acetylsalicylic acid; Hb = hemoglobin; CNVE = cardiovascular and neurovascular event

The perioperative management with AP drugs was performed according to the anesthesiologist's and urologist's choice as shown in Table 3.

Table 3.

Preoperative management of Antiplatelet

	Withheld (%)	Withdraw (%)	Relayed by ASA 75 mg (%)
ASA 75 mg (n = 64)	81	19	—
ASA >75 mg (n = 21)	10	19	71
Clopidogrel (n = 32)	0	16	84^a

Of which eight dual AP therapies.

ASA = acetylsalicylic acid.

On average, patients receiving dual therapy (clopidogrel with aspirin) did not have a longer LoS than patients receiving just clopidogrel (3.1 ± 1.9 vs 3.8 ± 2.5 days; p = 0.467).

Sixty-nine of 85 patients in the ASA group were operated on without stopping the 75 mg of ASA therapy, with an average LoS of 2.6 ± 1.4 compared to 2.5 ± 1.2 days in the naive group (p = 0.445).

Clopidogrel was started again 5–7 days after the procedure.

Two patients experienced cardiovascular or neurovascular events in the month following operation, one in the naive group and one in the AP group (clopidogrel). The patient in the naive group was found to have a low-symptom pulmonary embolism during computed tomography staging of his bladder tumor (pT2). The patient in the AP group who presented with an ischemic stroke 23 days postoperatively was treated with thrombolysis.

Logistic regression analysis of 231 procedures was performed. The adjustment variable was a long LoS (>2 days) vs a short LoS (≤2 days). Significant variables in univariate analysis were as follows: the therapeutic status (i.e., naive, ASA, or clopidogrel group), American Society of Anesthesiologists score (reflecting comorbidities), operative time, and muscle presence on histological analysis (comparable to resection depth). In multivariate analysis, resection in those receiving clopidogrel (odds ratio [OR] = 4.1 [1.7–9.6], p = 0.02), a longer operative time (OR = 2 [1.1–3.7], p = 0.02), and depth of resection (muscle described) (OR = 1.9 [1–3.8], p = 0.04) emerged as determinants of a LoS >2 days.

Discussion

This study showed that the average LoS after TURBT in patients receiving AP drugs was different in terms of statistical significance compared to naive patients (2.5 vs 2.9 days). However, this difference was not pertinent in terms of clinical relevance. Analysis of the subgroups showed that taking ASA did not lead to a clinical or statistical difference in the LoS, whereas taking clopidogrel was associated with longer hospitalization.

The absence of a clinical difference found between naive patients and patients taking AP drugs in this study is consistent with the finding in Wenders and colleagues's study.⁹ This is also the conclusion of Picozzi and colleagues¹⁰ who did not find an increased risk of bleeding despite ASA therapy. The link between bleeding and the average LoS seems legitimate because the hemorrhagic problem is directly related to the duration of bladder catheterization and therefore directly affects the LoS. This criterion also enables physicians to pragmatically understand the clinical impact of AP drugs on the patient and to indirectly consider associated comorbidities. In this study, it was logical that the AP group had more cardiovascular comorbidities, and it is interesting that they did not have a clinically meaningful impact on the LoS. This finding is consistent with the 2014 European Society of Cardiology/European Society of Anaesthesiology classification of TURBT as a low cardiovascular risk operation.¹¹

Another quality criterion of this study was that there was no apparent difference in the surgical practices with respect to literature, with a similar mean time of resection,¹⁰ equivalent duration of bladder catheterization placement,⁹ and no evidence of increased bleeding.¹⁰ Carmignani and coworkers¹² hypothesized that bladder tumors may become symptomatic, through hematuria, earlier and more intensely in patients receiving AP drugs. In the AP group of our study, the mean weight of the resected tumor (4.6 vs 7.3 g) and T2 rate (9% vs 13%) was lower than those in the naive group, which suggests that our management occurred earlier, although these differences were not statistically significant.

Stopping ASA results in a dramatic rebound in the pro-inflammatory and pro-thrombotic states.⁶ When prescribed for secondary prevention, withdrawal is associated with an excess risk of acute coronary syndrome and death (OR = 3.14), and the excess risk increases (OR = 89.9) when withdrawal is achieved in patients with a coronary stent.¹³ Therefore, the guidelines^2,7 recommend the continuation of low-dose ASA in patients for secondary prevention and relay (5 days before operations) in patients taking clopidogrel. Our practices were in accordance with these recommendations.

Results of subgroup analysis in this study are reinsuring for patients in the ASA group; however, it was not for those receiving clopidogrel for whom LoS was significantly increased (clinically and statistically). This increased LoS was also found by Carmignani and colleagues¹² who reported an average LoS of 4 days, which is similar to that in our study (3.7 days). This raises the question of relevance in the 5-day relay period of clopidogrel, and it could be explained by its pharmacokinetic and pharmacodynamic properties. Indeed, clopidogrel is a thienopyridine (P2Y12 receptor inhibitor), which has a variability of responses related to the variability of its hepatic metabolic conversion but also with factors such as body weight or presence of diabetes. A near-normal platelet function is found in most patients 4–7 days after treatment discontinuation, but some patients need a longer delay.¹⁴ This variability can explain the heterogeneity within the population receiving clopidogrel in our study, with a standard deviation in the LoS of 2.5 days, which is much higher than that in naive patients (1.2 days) and those taking ASA (1.3 days).

Analysis of the postoperative morbidity suggested that the increase in LoS in the clopidogrel group was mainly due to its hemorrhagic component. Indeed, the duration of bladder catheterization placement was longer in relation to prolonged hematuria, and the rate of urinary retention after catheter removal was higher (mainly blood clots) and close to the rate of 25% found by Carmignani and associates.¹² In addition, there was a more frequent need for transfusion or reintervention for hematuria. In a previous study, Virseda-Rodríguez and colleagues¹⁵ did not find more patients receiving clopidogrel among those who needed a transfusion, but their analysis was performed in a small subgroup (15 patients). Carmignani and colleagues,¹² in a series of 12 patients taking dual therapies, including 8 without withdrawal, did not report major complications but a high rate of acute bladder retention consecutive to hematuria.

Inclusion objectives were largely achieved. The ASA group fulfilled the effective requirements previously calculated to be statistically interpretable (n > 80), thereby providing a reliable representation of the situation. If the clopidogrel group appeared to be underpowered in relation to the initial hypothesis, the discovery of such a statistical difference for small numbers was not an argument for an absence of power. These results should still be confirmed in a prospective analysis for a stronger validation; a clopidogrel group should be included in a prospective analysis to comparing different relay management.

The present study has some limitations. LoS and taking AP drugs or a naive status could have been collected for every patient. However, complications such as rehospitalization and cardiovascular or neurovascular events have been only identified in patients receiving support at our hospital, but this bias equally affected naive patients such as those taking AP drugs. Finally, the criteria of LoS were composite by nature and considered many different parameters. Therefore, the criteria did not enable us to perform a detailed analysis that would be much more relevant in daily clinical practice.

The management of perioperative AP is an important and daily challenge faced by both surgical and anesthesia teams because of the aging population for whom the life prescription of AP drugs for primary or secondary prevention is common. Our study was reassuring in the clinical impact of AP drugs taken during TURBT, which is especially true for low-dose ASA. However, the question arose regarding the duration of clopidogrel withdrawal or relay, which could be extended to 7 days before operation, to expect a postoperative decrease in hemorrhagic morbidity.

Conclusions

Even if there is a statistically significant difference in the LoS among patients taking AP compared to naive patients undergoing TURBT, it seems that this difference is not clinically relevant. More precisely, ASA does not increase the LoS or morbidity of TURBT, and this trend was the same with continued therapy at a dosage of 75 mg preoperatively. Conversely, clopidogrel increases the LoS and postoperative complications, making us question its use in the perioperative management of these patients. These results must be confirmed with a prospective study.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

Abbreviations Used

References

ATIH. Agence technique de l'information sur l'hospitalisation. Available at: http://atih.sante.fr/ (Accessed September 2, 2015 ).

Haute Autorité de santé. Recommandations_antiagregant_plaquettaire.pdf. Available at: http://has-sante.fr/portail/upload/docs/application/pdf/2012-10/recommandations_antiagregant_plaquettaire.pdf (Accessed: September 2, 2015).

Roger

, Go

, Lloyd-Jones

, et al. Heart disease and stroke statistics—2011 update: A report from the American Heart Association. Circulation, 2011; 123:e18–e209.

Moschovitis

, Cook

, Meier

. Percutaneous coronary interventions in Europe in 2006. EuroIntervention, 2010; 6:189–194.

Savonitto

, Caracciolo

, Cattaneo

, et al. Management of patients with recently implanted coronary stents on dual antiplatelet therapy who need to undergo major surgery. J Thromb Haemost, 2011; 9:2133–2142.

Gupta

, Streiff

, Resar

, et al. Coronary stent management in elective genitourinary surgery. BJU Int, 2012; 110:480–484.

Urofrance. Article. Available at: http://urofrance.org/nc/science-et-recherche/base-bibliographique/article/html/chirurgie-ambulatoire-en-urologie-texte-des-recommandations-afu.html (Accessed September 2, 2015 ).

Dunne

, Malone

, Tracy

, et al. Perioperative anemia: An independent risk factor for infection, mortality, and resource utilization in surgery. J Surg Res, 2002; 102:237–244.

Wenders

, Wenzel

, Nitzke

, et al. Perioperative platelet inhibition in transurethral interventions: TURP/TURB. Int Braz J Urol, 2012; 38:606–610.

10.

Picozzi

, Marenghi

, Ricci

, et al. Risks and complications of transurethral resection of bladder tumor among patients taking antiplatelet agents for cardiovascular disease. Surg Endosc, 2014; 28:116–121.

11.

Kristensen

, Knuuti

, Saraste

, et al. 2014 ESC/ESA Guidelines on non-cardiac surgery: Cardiovascular assessment and management: The Joint Task Force on non-cardiac surgery: Cardiovascular assessment and management of the European Society of Cardiology (ESC) and the European Society of Anaesthesiology (ESA). Eur J Anaesthesiol, 2014; 31:517–573.

12.

Carmignani

, Picozzi

, Stubinski

, et al. Endoscopic resection of bladder cancer in patients receiving double platelet antiaggregant therapy. Surg Endosc, 2011; 25:2281–2287.

13.

Biondi-Zoccai

GGL

, Lotrionte

, Agostoni

, et al. A systematic review and meta-analysis on the hazards of discontinuing or not adhering to aspirin among 50,279 patients at risk for coronary artery disease. Eur Heart J, 2006; 27:2667–2674.

14.

Fourcade

, Geeraerts

, Minville

, et al. Traité d'anesthésie et de réanimation, 4th edition. Paris, France: Médecine Sciences Publications, 2014.

15.

Virseda-Rodríguez

, Padilla-Fernández

, López-Parra

, et al. Influence of antiplatelet-anticoagulant drugs on the need of blood components transfusion after vesical transurethral resection. Arch Ital Urol Androl, 2015; 87:136–140.