Omission of Hemostatic Agents During Robotic Partial Nephrectomy Does Not Increase Postoperative Bleeding Risk

Abstract

Introduction:

The use of hemostatic agents (HA) during robotic partial nephrectomy (RPN) is largely empiric. We sought to assess the impact of HA on postoperative bleeding after RPN in a contemporary cohort.

Patients and Methods:

Using our institutional RPN database, we identified consecutive patients treated between 2010 and 2015. HA were routinely placed in the nephrectomy bed at the time of renorrhaphy until 2014 when their use was phased out to reduce cost. We compared postoperative bleeding outcomes (blood transfusion and hemoglobin decline) between patients who did and did not receive HA, after excluding patients with preoperative anemia (hemoglobin <11 g/dL) or high estimated blood loss (≥175 mL). The total inflation-adjusted costs (for 2015) of HA were calculated.

Results:

Of 544 cases, HA were used in 240 (44.1%). The mean number of agents per case was 1.4 ± 0.73, including 77 (14.2%), 52 (9.6%), and 39 (7.2%) cases in which cellulose, fibrin, or gelatin-based agents were used alone, respectively, and 72 (13.2%) cases in which multiple agents were used. The mean cost of HA per case was $488 ± 421. Nearly 90% of cases were performed by surgeons who were beyond their learning curves. Overall, 13 (2.4%) patients were transfused, and the median hemoglobin decline was 2.2 g/dL (IQR, 1.4–3.0 g/dL). On univariate analysis, HA use and type of HA were not significantly associated with blood transfusion (p = 0.20 and p = 0.29, respectively), but were associated with hemoglobin decline (p = 0.01 and p = 0.02, respectively). After adjusting for covariates, HA use was no longer significantly associated with postoperative hemoglobin decline.

Conclusions:

In nonanemic patients with minimal intraoperative bleeding, HA use does not alter postoperative bleeding outcomes after RPN, suggesting that their routine use in this setting merits reconsideration. Further research is needed to determine if HA may be useful in certain high-risk situations.

Introduction

Postoperative hemorrhage is one of the most common complications after partial nephrectomy (PN), occurring in ∼5% of cases.^1,2 The primary means of hemostasis during PN is a multilayer sutured closure of the parenchymal defect.³ Traditionally, topical hemostatic agents (HA) were considered an adjunct to, rather than substitute for, good surgical technique.³ HA have been used empirically in renal surgery for more than 35 years, but routine use took hold in the era of minimally invasive surgery.^4,5 Because of the technical difficulty of laparoscopic suturing and the need for rapid hemostasis under warm ischemia, HA were used to supplement or even supplant ligation sutures during laparoscopic PN.^6

–11 Emerging data suggest that HA use may not modify bleeding risk after modern-day PN.^12

–15 Furthermore, robotic PN (RPN), now the most common minimally invasive approach, is associated with improved hemostasis, likely attributable to the improved ease, speed, and precision of intracorporeal suturing offered by the robotic platform.^13,16,17 HA are still widely used during PN despite absence of clear evidence.^12,14,15,18 We sought to examine the impact of HA on postoperative bleeding after RPN in a contemporary cohort.

Patients and Methods

Data source

Data were abstracted from our prospectively maintained, institutional review board-approved RPN database, which includes demographic, clinical, laboratory, pathologic, and outcome data on all excisional RPN cases performed at our institution by five robotic surgeons.

Study population

We identified 974 consecutive patients treated with RPN between 2010 and 2015. For all cases, the contemporary RPN technique was used, as described previously.¹⁹ After hilar clamping (n = 493) or no clamp (n = 51), the mass was excised sharply with a small rim of normal parenchyma. The parenchymal defect was sutured closed in two layers. If used, HA were applied to the renorrhaphy site after the capsular closure. Bolsters were not used. Before 2014, HA use and the type of HA used were at the discretion of the surgeon. In 2014, HA use for RPN was phased out and, by 2015, was eliminated entirely for cost reduction. Since our objective was to investigate the effect of HA on postoperative bleeding outcomes, patients with preoperative hemoglobin <11 g/dL, estimated blood loss (EBL) ≥175 mL (approximate actual blood loss ≥250 mL),²⁰ or intraoperative blood transfusion were excluded to avoid confounding from chronic anemia or intraoperative bleeding (n = 430). Antiplatelet medications other than nonsteroidal anti-inflammatory agents and full-strength or therapeutic anticoagulants were not used in the perioperative period.

Study variables

Information on HA use was obtained from hospital billing records. For patients who were charged for a HA, intraoperative use was confirmed by review of the operative notes. HA used at our institution during the study period belonged to one of three classes, namely oxidized cellulose (Surgicel^®), fibrin sealants (Tisseel^®, Evicel^®, and Vitagel™), or gelatin-based sealants (FloSeal^®, Surgiflo™, Gelfoam^®, and Surgifoam™). While each class of HA has its own unique composition and mechanism of action, as previously described, they all support hemostasis by either supplementing or entirely supplanting the internal clotting cascade.²¹ To assess the impact of use vs. nonuse of HA and the impact of HA type, we defined HA using two variables. The first variable categorized use dichotomously as either use of one or more HA from any category or use of no HA. The second variable categorized hemostatic use by the type of HA used, namely (1) no HA, (2) oxidized cellulose, (3) fibrin sealant, (4) gelatin-based sealant, or (5) multiple types of HA.

Demographic variables included patient sex and race (white vs. nonwhite). Clinical variables included age (in years), nonage-adjusted Charlson comorbidity index, body mass index (in kilograms per meter squared), perioperative use of antiplatelet medications or anticoagulants for deep venous thrombosis prophylaxis (yes or no), and preoperative hemoglobin (in grams per deciliter). Disease variables, based on preoperative imaging, included clinical tumor size (in centimeters) and R.E.N.A.L. nephrometry score. Operative variables included the intraoperative intravenous fluid (IVF) volume (in liters), the EBL (in milliliters), and surgeon experience. Surgeon experience was defined based on the surgeon's case volume (≤30 cases vs. >30 cases) at the time of the surgery based on evidence that the initial 30 cases represent the learning curve for RPN.²² The total inflation-adjusted cost of HA was calculated for each case from hospital billing records, using inflation data from the U.S. Department of Labor's Consumer Price Index.

Study outcomes

The primary outcome was postoperative blood transfusion. The secondary outcome was postoperative hemoglobin decline (in g/dL). Hemoglobin decline was calculated as the difference between the preoperative hemoglobin, obtained within one month of surgery, and the postoperative hemoglobin, obtained before transfusion (for those who received a blood transfusion) or at the time of the first follow-up visit. This time point was chosen to minimize the impact of hemodilution from IVFs on hemoglobin concentration.

Statistical analyses

The association between HA use and the outcomes was assessed relative to other bleeding risk factors (covariates). The chi-squared and Fisher exact tests were used for categorical variables, and the Mann–Whitney U and Kruskal–Wallis tests were used for continuous variables. On univariate analysis, EBL was the only factor associated with postoperative transfusion, obviating the need for multivariate analysis of this outcome. Multiple linear regression was used to assess the association between HA use and hemoglobin decline after adjusting for other significant covariates. The two variables describing HA use were analyzed separately to avoid potential spurious results because of collinearity. Statistical tests were performed using SAS^® University Edition (SAS Institute Inc., Cary, NC). p-Values <0.05 were considered statistically significant.

Results

Of 544 cases, HA were used in 240 (44.1%). The mean number of agents used per case was 1.4 ± 0.73, including 77 (14.2%), 52 (9.6%), and 39 (7.2%) cases in which cellulose, fibrin, or gelatin-based agents were used alone, respectively, and 72 (13.2%) cases in which multiple agents were used. The mean cost of HA per case was $488 ± 421.

Comparing patients who received HA to those who did not receive HA, there were no statistically significant differences between the groups in terms of age, sex, race, Charlson comorbidity index, body mass index, preoperative hemoglobin, clinical tumor size, or R.E.N.A.L. score (Table 1). HA were less likely to be used in patients receiving perioperative antiplatelet therapy or deep vein thrombosis chemoprophylaxis (36.7% vs. 46.7%, p = 0.04) and by experienced surgeons (40.7% vs. 71.7%, p < 0.01); however, these differences reflect changing practice patterns over time. There was no difference in HA use between patients who underwent clamped (44.6%) vs. unclamped (39.2%) RPN (p = 0.46). Mean warm ischemia time did not differ significantly between the HA and non-HA groups, respectively (20.0 minutes vs. 18.6 minutes, p = .18). Mean total OR time was slightly longer for the HA group vs. the non-HA group (3.0 hour vs. 2.8 hour, p = .02).

Table 1.

Characteristics of Patients Who Underwent Robotic Partial Nephrectomy Stratified by Hemostatic Agent Use

	Overall (N = 544)	Hemostatic agent(s) used (n = 240)	Hemostatic agent(s) not used (n = 304)	p
Variables
Age (years)				0.64
Median (IQR)	59 (49–66)	58 (49–66)	59 (49–67)
Mean (SD)	57.6 (12.3)	57.3 (12.0)	57.9 (12.6)
Sex				0.07
Male, n (%)	308 (56.6)	146 (60.8)	162 (53.3)
Female, n (%)	236 (43.4)	94 (39.2)	142 (46.7)
Race				0.46
White, n (%)	460 (84.6)	206 (85.8)	254 (83.6)
Nonwhite, n (%)	84 (15.4)	34 (14.2)	50 (16.4)
Charlson comorbidity index				0.36
Median (IQR)	0 (0–2)	0 (0–2)	0 (0–2)
Mean (SD)	1.0 (1.4)	1.1 (1.5)	0.9 (1.2)
Body mass index (kg/m2)				0.89
Median (IQR)	29.1 (25.4–33.3)	29.2 (25.7–32.9)	29.0 (25.3–33.5)
Mean (SD)	30.1 (6.5)	30.0 (6.1)	30.1 (6.9)
Preoperative hemoglobin (g/dL)				0.93
Median (IQR)	14.1 (13.1–15.0)	14.1 (13.1–15.1)	14.1 (13.2–15.0)
Mean (SD)	14.0 (1.3)	14.0 (1.3)	14.0 (1.4)
Antiplatelet/anticoagulant				0.04
Yes, n (%)	139 (25.6)	51 (21.3)	88 (28.9)
No, n (%)	405 (74.4)	189 (78.8)	216 (71.1)
Clinical tumor size (cm)				0.70
Median (IQR)	2.9 (2.1–4.0)	2.9 (2.0–4.0)	3.0 (2.2–3.9)
Mean (SD)	3.1 (1.5)	3.1 (1.5)	3.2 (1.5)
R.E.N.A.L. score				0.58
Median (IQR)	7 (6–9)	7 (6–9)	7 (6–9)
Mean (SD)	7.2 (2.0)	7.3 (2.0)	7.2 (1.9)
Surgeon experience				<0.01
≤30 cases, n (%)	484 (89.0)	43 (17.9)	17 (5.6)
>30 cases, n (%)	60 (11.0)	197 (82.1)	287 (94.4)

EBL = estimated blood loss; IVF = intravenous fluid.

Overall, 13 (2.4%) patients were transfused postoperatively. On univariate analysis, neither HA use nor the specific type of HA used was significantly associated with blood transfusion (p = 0.20 and p = 0.29, respectively; Table 2). The only significant predictor of postoperative transfusion was EBL, which was slightly higher among those who were transfused (114 ± 84.6 mL vs. 90.4 ± 39.7 mL, p = 0.04). Other risk factors for transfusion, including preoperative hemoglobin (p = 0.20), antiplatelet or anticoagulant use (p = 0.14), tumor size (p = 0.36), R.E.N.A.L. score (p = 0.86), and surgeon experience (p = 0.66), were not associated with a higher risk of transfusion.

Table 2.

Univariate Analysis of Factors Associated with Postoperative Transfusion After Robotic Partial Nephrectomy

	Transfused (n = 13)	Not transfused (n = 531)	p
Variables
Hemostatic agent			0.20
Yes, n (%)	8 (61.5)	232 (43.7)
No, n (%)	5 (38.5)	299 (56.3)
Hemostatic agent type			0.29
None, n (%)	5 (38.5)	299 (56.3)
Cellulose, n (%)	3 (23.1)	74 (13.9)
Fibrin, n (%)	0	52 (9.8)
Gelatin, n (%)	2 (15.4)	37 (7.0)
Multiple, n (%)	3 (23.1)	69 (13.0)
Age (years)			0.19
Median (IQR)	59 (55–71)	58 (49–66)
Mean (SD)	62.3 (10.1)	57.5 (12.3)
Sex			0.72
Male, n (%)	8 (61.5)	300 (56.5)
Female, n (%)	5 (38.5)	231 (43.5)
Race			0.44
White, n (%)	10 (76.9)	450 (84.7)
Nonwhite, n (%)	3 (23.1)	81 (15.3)
Charlson comorbidity index			0.09
Median (IQR)	1 (0–3)	0 (0–2)
Mean (SD)	1.8 (2.0)	1.0 (1.4)
Body mass index (kg/m2)			0.84
Median (IQR)	29.4 (26.2–31.5)	29.1 (25.4–33.3)
Mean (SD)	29.8 (5.1)	30.1 (6.5)
Preoperative hemoglobin (g/dL)			0.20
Median (IQR)	13.7 (12.2–14.3)	14.1 (13.1–15.0)
Mean (SD)	13.6 (1.7)	14.0 (1.3)
Antiplatelet/anticoagulant			0.14
Yes, n (%)	1 (7.7)	138 (26.0)
No, n (%)	12 (92.3)	393 (74.0)
Clinical tumor size (cm)			0.36
Median (IQR)	2.6 (2.0–3.1)	2.9 (2.1–4.0)
Mean (SD)	2.8 (1.3)	3.1 (1.5)
R.E.N.A.L. score			0.86
Median (IQR)	7 (5–9)	7 (6–9)
Mean (SD)	7.1 (2.2)	7.2 (2.0)
Intraoperative intravenous fluid volume (L)			0.16
Median (IQR)	4.0 (2.8–4.1)	3.4 (2.8–3.9)
Mean (SD)	3.7 (1.0)	3.3 (0.8)
Estimated blood loss volume (mL)			0.04
Median (IQR)	100 (100–150)	100 (50–100)
Mean (SD)	114 (38.6)	90.4 (39.7)
Surgeon experience			0.66
≤30 cases, n (%)	2 (15.4)	58 (10.9)
>30 cases, n (%)	11 (84.6)	473 (89.1)

The median postoperative hemoglobin decline was 2.2 (IQR, 1.4–3.0) g/dL, and the median time to the postoperative hemoglobin was 4 (IQR, 2–8) days. HA use and the type of HA used were significantly associated with hemoglobin decline (p = 0.01 and 0.02, respectively; Table 3). Other factors significantly associated with hemoglobin decline were intraoperative IVF volume (p < 0.01) and EBL (p < 0.01), while surgeon experience approached significance (p = 0.05). After adjusting for these covariates in a multivariate model, neither HA use nor the specific type of HA used was significantly associated with hemoglobin decline (Table 4). EBL and intraoperative IVF volume were the only independent predictors of significant hemoglobin decline. For every milliliter increase in EBL, a 0.0034–0.0035 g/dL decline in postoperative hemoglobin was predicted, or in other words, for every unit (300 mL) of EBL, a 1.0–1.1 mg/dL decline in postoperative hemoglobin was predicted (p = 0.02 and p = 0.03). For every liter increase in intraoperative fluid volume received, a 0.16–0.17 g/dL decline in postoperative hemoglobin was predicted (p = 0.02).

Table 3.

Univariate Analysis of Factors Associated with Hemoglobin Decline After Robotic Partial Nephrectomy

Hemoglobin Decline (g/dL)
	Median (IQR)	Mean (SD)	p
Variables
Hemostatic agent			0.01
Yes	2.4 (1.6–3.2)	2.5 (1.3)
No	2.2 (1.4–2.8)	2.2 (1.2)
Hemostatic agent type			0.02
None	2.2 (1.4–2.8)	2.2 (1.2)
Cellulose	2.1 (1.4–3.0)	2.3 (1.4)
Fibrin	2.2 (1.5–3.1)	2.4 (1.2)
Gelatin	2.6 (2.0–3.2)	2.6 (1.2)
Multiple	2.4 (1.8–3.5)	2.7 (1.5)
Sex			0.56
Male	2.2 (1.4–2.9)	2.3 (1.3)
Female	2.2 (1.5–3.0)	2.3 (1.2)
Race			0.88
White	2.3 (1.5–3.0)	2.3 (1.3)
Nonwhite	2.2 (1.3–3.0)	2.3 (1.3)
Chemoprophylaxis			0.33
Yes	2.3 (1.5–3.1)	2.4 (1.2)
No	2.2 (1.4–2.9)	2.3 (1.3)
Surgeon experience			0.05
≤30 cases	2.4 (1.7–3.3)	2.7 (1.5)
>30 cases	2.2 (1.4–2.9)	2.2 (1.2)
	Correlation coefficient	95% CI	p
Age (years)	0.0606	−0.0237–0.1438	0.16
Charlson comorbidity index	0.0242	−0.0600–0.1081	0.57
Body mass index (kg/m2)	−0.0239	−0.1078–0.0604	0.58
Clinical tumor size (cm)	0.0715	−0.0127–0.1546	0.10
R.E.N.A.L. score	0.0734	−0.0121–0.1576	0.09
Intraoperative intravenous fluid volume (L)	0.1558	0.0722–0.2369	<0.01
Estimated blood loss volume (mL)	0.1480	0.0428–0.2087	<0.01

Table 4.

Multiple Linear Regression Analysis of Factors Associated with Hemoglobin Decline After Robotic Partial Nephrectomy

	Correlation coefficient	95% CI	p
Variables
A. Use of any hemostatic agent
Hemostatic agent used (vs. no agent)	0.1560	−0.0741–0.3862	0.18
Surgeon experience >30 cases (vs. ≤30 cases)	−0.3106	−0.6709–0.0497	0.09
Clinical tumor size (cm)	−0.0043	−0.0862–0.0776	0.92
R.E.N.A.L. score	0.0135	−0.0481–0.0750	0.67
Intraoperative intravenous fluid volume (L)	0.1674	0.0320–0.3028	0.02
Estimated blood loss volume (mL)	0.0035	0.0005–0.0064	0.02
B. Use of specific type of hemostatic agent
Hemostatic agent type (vs. no agent)
Cellulose	0.0158	−0.3171–0.3487	0.93
Fibrin	0.0840	−0.3038–04717	0.67
Gelatin	0.2494	0.1935–0.6923	0.27
Multiple	0.3163	−0.0260–0.6587	0.07
Surgeon experience >30 cases (vs. ≤30 cases)	−0.3008	−0.6639–0.0623	0.10
Clinical tumor size (cm)	−0.0014	−0.0835–0.0807	0.97
R.E.N.A.L. score	0.0125	−0.0493–0.0743	0.69
Intraoperative intravenous fluid volume (L)	0.1605	0.0246–0.2964	0.02
Estimated blood loss volume (mL)	0.0034	0.0004–0.0063	0.03

Discussion

Postoperative hemorrhage is one of the most common, if not most feared, complications after PN.^1,2 Consequently, HA have been used routinely during PN to minimize the risk of delayed postoperative bleeding.^5,18 The evidence behind this practice is weak, derived mostly from small retrospective, poorly designed, and/or outdated studies.^{6,7,9,10,12,14,15,18,23} Because of the improved speed and precision of robotic suturing over earlier laparoscopic techniques, we hypothesized that HA may not offer an additional benefit over standard hemostatic suturing when performed robotically.

In a large single-center study of contemporary RPN cases, HA use was not associated with postoperative transfusion or hemoglobin decline. We investigated several HA types, namely gelatin-based sealants (e.g., FloSeal), oxidized cellulose (e.g., Surgicel), fibrin sealants (e.g., Tisseel), and combinations of these types, which represent the most commonly used HA during PN.⁵ Neither HA use nor HA type was significantly associated with postoperative bleeding.

We chose postoperative transfusion as our primary endpoint given its clinical relevance and accounted for the impact of preoperative anemia and intraoperative blood loss (greater than one unit) on this endpoint by excluding patients with these characteristics from our study cohort. Given the low incidence of transfusion in modern times, we selected postoperative hemoglobin decline as a secondary endpoint to further understand the effect of HA on postoperative bleeding. To minimize the impact of hemodilution on this endpoint, we used hemoglobin data from median postoperative day four, by which time most patients have begun to mobilize third-spaced fluids, and adjusted for intraoperative IVF volume in our analysis.

To address the potential bias introduced by our exclusion criteria, we performed a subanalysis of the overall cohort without excluding for preoperative anemia or intraoperative bleeding (data not shown). On multivariate analysis, preoperative anemia and CCI were the only significant predictors of postoperative transfusion, and HA use predicted increased, not decreased, postoperative hemoglobin decline even after adjusting for preoperative anemia and EBL. These results confirm the substantial confounding effect of preoperative anemia and intraoperative bleeding on postoperative bleeding outcomes, supporting our decision to exclude based on these factors.

Only two prospective studies have investigated the efficacy of HA for PN. In the only randomized controlled trial, Siemer and coworkers found a shorter intraoperative time to hemostasis (primary endpoint) for TachoSil^® vs. standard renorrhaphy, but no difference in transfusion rates or hemoglobin concentration of drainage fluid (secondary endpoints) in patients undergoing open PN.²³ In a prospective observational study, Lang and colleagues found no difference in EBL or intraoperative transfusion rates associated with HA use; however, the robotic approach only accounted for 16% of cases.¹⁸ Neither study assessed postoperative hemostasis, focusing instead on intraoperative bleeding parameters (i.e., bleeding time, EBL, and intraoperative transfusion), which may be influenced more so by surgeon expertise rather than HA efficacy. These studies contained little, if any, data on the robotic approach, rendering them less relevant to the present discussion.

Laparoscopic series account for the majority of studies, particularly those reporting positive results for HA.^6

–11 The greatest limitation of these studies, aside from their observational designs, is that they do not reflect the improved quality of robotic renorrhaphy techniques.²⁴ Since a multilayer sutured renorrhaphy is still the primary means of hemostasis during PN, these studies may not apply to contemporary practice, in which RPN is the dominant minimally invasive approach.^13,23

Only three observational studies have examined the efficacy of HA during RPN. In a negative study of 114 patients who underwent RPN, Cohen and associates found no difference in postoperative transfusion rates in patients who received Evicel vs. controls; however, the study was underpowered with only one patient in the control group necessitating transfusion.¹² In a propensity-matched study, Antonelli and coworkers investigated the efficacy of FloSeal vs. TachoSil vs. no HA in terms of transfusion rates and postoperative day 3 hemoglobin change, finding no benefit from HA.¹⁵ Although the study was well designed, it suffered from several limitations, including small sample sizes (only 66 patients per group), inability to distinguish between the laparoscopic and robotic approaches, and inclusion of a large percentage of simple enucleations in the study cohort. In a large multicenter study of 515 robotic partial nephrectomies between 2009 and 2015, Peyronnet and colleagues found no difference in hemorrhagic complications, defined as the occurrence of pseudoaneurysm, arteriovenous fistula, or hematoma necessitating transfusion, associated with the use of various HA.¹⁴ Two out of three components of their primary endpoint (i.e., pseudoaneurysm and arteriovenous fistula formation) are influenced heavily by surgical technique and tumor location, factors for which they did not adjust, making them poor measures of HA efficacy.

In our study, the average cost of HA was nearly $500 per case. Given that HA are used routinely in about 75% of cases, there is significant room for national cost savings by eliminating unnecessary HA use.^5,18 That being said, there may be certain situations, that is, inexperienced surgeons, complex cases, or patients at high risk for bleeding, in which HA use may be useful, but this remains to be determined.

The main limitation of our study is its retrospective nonrandomized design. The decision to use a HA was not randomized and thus susceptible to bias. Initially, HA use was considered standard practice, but as the study period progressed, use was phased out and ultimately eliminated to reduce cost. Consequently, opportunistic HA use or changing surgeon experience over time may have introduced bias into our study. To minimize the impact of these potential biases, we adjusted for antiplatelet or anticoagulant use, surgeon experience, tumor size, tumor complexity, EBL, and other risk factors for postoperative bleeding, in our multivariate analysis. Furthermore, all cases were performed using the contemporary RPN technique, eliminating technique as a source of bias.¹⁹ Second, our study evaluated multiple types of HA, each with their own mechanisms of action. This heterogeneity is both a strength and weakness of our study, increasing its clinical relevance, but also potentially obscuring the efficacy of individual HA. To address this heterogeneity, we analyzed the efficacy of HA both collectively and individually by type. Finally, although our study represents the largest of its kind to date, the relatively small sample sizes, particularly for the different HA groups, and low number of transfusion events may have limited our ability to detect a significant difference. To account for the low number of transfusion events, we assessed the impact of HA use on postoperative hemoglobin decline, one of the principal drivers of transfusion, and obtained similarly negative results.

Conclusions

In nonanemic patients with minimal intraoperative bleeding (preoperative hemoglobin ≥11 g/dL, EBL <175 mL), HA use does not alter postoperative bleeding risk after RPN. While the selective use of HA in certain situations may be indicated, the routine empiric use of HA for RPN merits reconsideration.

Footnotes

Author Disclosure Statement

We deny any financial or material research support for this work. J.K. is a consultant for Endocare, Inc. We deny any other potential direct or indirect conflicts of interest.

Abbreviations Used

References

Joudi

, Allareddy

, Kane

, et al. Analysis of complications following partial and total nephrectomy for renal cancer in a population based sample. J Urol, 2007; 177:1709.

Spana

, Haber

, Dulabon

, et al. Complications after robotic partial nephrectomy at centers of excellence: Multi-institutional analysis of 450 cases. J Urol, 2011; 186:417.

Campbell

, Novick

. Surgical technique and morbidity of elective partial nephrectomy. Semin Urol Oncol, 1995; 13:281.

Urlesberger

, Rauchenwald

, Henning

. Fibrin adhesives in surgery of the renal parenchyma. Eur Urol, 1979; 5:260.

Breda

, Stepanian

, Lam

, et al. Use of haemostatic agents and glues during laparoscopic partial nephrectomy: A multi-institutional survey from the United States and Europe of 1347 cases. Eur Urol, 2007; 52:798.

Richter

, Schnorr

, Deger

, et al. Improvement of hemostasis in open and laparoscopically performed partial nephrectomy using a gelatin matrix-thrombin tissue sealant (FloSeal). Urology, 2003; 61:73.

Bak

, Singh

, Shekarriz

. Use of gelatin matrix thrombin tissue sealant as an effective hemostatic agent during laparoscopic partial nephrectomy. J Urol, 2004; 171:780.

Pruthi

, Chun

, Richman

. The use of a fibrin tissue sealant during laparoscopic partial nephrectomy. BJU Int, 2004; 93:813.

Gill

, Ramani

, Spaliviero

, et al. Improved hemostasis during laparoscopic partial nephrectomy using gelatin matrix thrombin sealant. Urology, 2005; 65:463.

10.

Finley

, Lee

, Eichel

, et al. Fibrin glue-oxidized cellulose sandwich for laparoscopic wedge resection of small renal lesions. J Urol, 2005; 173:1477.

11.

Triaca

, Zagha

, Libertino

. Does thrombin sealant allow nephron-sparing surgery with no renal artery occlusion? A description of technique and initial results. BJU Int, 2005; 95:1273.

12.

Cohen

, Jayram

, Mullins

, et al. Do fibrin sealants impact negative outcomes after robot-assisted partial nephrectomy?. J Endourol, 2013; 27:1236.

13.

Ghani

, Sukumar

, Sammon

, et al. Practice patterns and outcomes of open and minimally invasive partial nephrectomy since the introduction of robotic partial nephrectomy: Results from the nationwide inpatient sample. J Urol, 2014; 191:907.

14.

Peyronnet

, Oger

, Khene

, et al. The use of hemostatic agents does not prevent hemorrhagic complications of robotic partial nephrectomy. World J Urol, 2015; 33:1815.

15.

Antonelli

, Minervini

, Mari

, et al. TriMatch comparison of the efficacy of FloSeal versus TachoSil versus no hemostatic agents for partial nephrectomy: Results from a large multicenter dataset. Int J Urol, 2015; 22:47.

16.

Zargar

, Allaf

, Bhayani

, et al. Trifecta and optimal perioperative outcomes of robotic and laparoscopic partial nephrectomy in surgical treatment of small renal masses: A multi-institutional study. BJU Int, 2015; 116:407.

17.

Faria

, Caputo

, Wood

, et al. Robotic partial nephrectomy shortens warm ischemia time, reducing suturing time kinetics even for an experienced laparoscopic surgeon: A comparative analysis. World J Urol, 2014; 32:265.

18.

Lang

, Mouracade

, Gimel

, et al: National prospective study on the use of local haemostatic agents during partial nephrectomy. BJU Int, 2014; 113:E56.

19.

Kaouk

, Khalifeh

, Hillyer

, et al. Robot-assisted laparoscopic partial nephrectomy: Step-by-step contemporary technique and surgical outcomes at a single high-volume institution. Eur Urol, 2012; 62:553.

20.

McCullough

, Roth

, Ginsberg

, et al. Estimated blood loss underestimates calculated blood loss during radical retropubic prostatectomy. Urol Int, 2004; 72:13.

21.

Dalpiaz

, Neururer

, Bartsch

, et al. Haemostatic sealants in nephron-sparing surgery: What surgeons need to know. BJU Int, 2008; 102:1502.

22.

Mottrie

, De Naeyer

, Schatteman

, et al. Impact of the learning curve on perioperative outcomes in patients who underwent robotic partial nephrectomy for parenchymal renal tumours. Eur Urol, 2010; 58:127.

23.

Siemer

, Lahme

, Altziebler

, et al. Efficacy and safety of TachoSil as haemostatic treatment versus standard suturing in kidney tumour resection: A randomised prospective study. Eur Urol, 2007; 52:1156.

24.

Aboumarzouk

, Stein

, Eyraud

, et al. Robotic versus laparoscopic partial nephrectomy: A systematic review and meta-analysis. Eur Urol, 2012; 62:1023.