- To assess the use of local haemostatic agents (HAs) in a prospective multicentre large series of partial nephrectomies (PNs).
Control of blood loss during urological surgery is paramount to the success of patient recovery. Haemostatic agents (HAs) and tissue sealants are used routinely to prevent excess blood loss and in reconstruction during surgical repair.
Over the past decade, an increasing body of data on the indications of nephron-sparing surgery (NSS) for RCC has developed. Today, NSS is an established approach for patients with localised RCC when preservation of renal function is desired.
Bleeding after partial nephrectomy (PN) can be immediate or delayed and may have severe consequences. It is estimated that, including immediate and delayed bleeding, its incidence is 4–6% after laparoscopic PN (LPN) and ≈1.6% after open PN (OPN) [1, 2].
To maximise haemostasis after the excision of the tumour and to reduce the rate of postoperative haemorrhagic events, a wide variety of HAs have been used. Some of the available products include thrombin sealant, fibrin glue, oxidised methylcellulose, and gelatine matrix. Each of these agents differs in mechanism of action, cost, and application.
Haemostasis is a rather complex process. The injury of a blood vessel triggers the following sequence of events: (i) vessel constriction to reduce blood flow; (ii) adherence of circulating platelets to the vessel wall at the site of the trauma; and (iii) platelet activation and aggregation, coupled with an intricate series of enzymatic reactions involving coagulation proteins (≈30) that produce fibrin to form a stable haemostatic plug. The aim of all HAs is to act by imitating, promoting, or bypassing specific steps of the coagulation cascade.
The purpose of the present study was to assess the use of local HAs in a prospective multi-institutional large series of PNs.
Study design: Prospective National Observational Registry on the Practices of Haemostasis in Partial Nephrectomy (NEPHRON). The aim of this registry was to evaluate and assess the use and results of local HAs in PN. The Institute's ethics committee approved this study.
Study sites: the study was conducted in 54 French urological centres from 1 June to 31 December 2010.
Subjects: 570 consecutive patients undergoing a PN were enrolled in this study in a prospective manner.
Data collection: the data was collected prospectively via an electronic case-report form: five different sheets were included for preoperative, perioperative, postoperative and follow-up data, respectively.
Primary objective: the main objective of the study was to describe the process of haemostasis during PN in France and to assess the results at 1 month after PN. The variables used for this purpose were:
Quantitative variables were expressed as the mean (sd) with the range. The distribution of quantitative variables was analysed using the Shapiro–Wilk test. Qualitative variables were presented by frequency and percentile.
Comparison tests were done using an anova for quantitative variables normally distributed or using the non-parametric Kruskal-Wallis test, otherwise the chi-squared test or Fisher's exact test for categorical variables.
A P < 0.05 was considered to indicate statistical significance.
The mean (sd, range) age of the patients was 60.0 (13, 21–90) years. The male/female ratio was of 1:6. The mean (sd) body mass index (BMI) was 26.4 (4.7) kg/m2. In all, 85.6% of patients were asymptomatic at diagnosis (incidental discovery) and 16% of patients were receiving antiplatelet or anticoagulant treatment before PN. Other demographic characteristics are given in Table 1.
|Mean (sd, range) age, years||60.0 (13, 21–90)|
|Sex, n (%)|
|Mean (sd, range):|
|Height, cm||169.5 (8.7, 136–195)|
|Weight, kg||76.0 (15.4, 40–130)|
|BMI, kg/m2||26.4 (4.7, 14.5–44.5)|
|Von Hippel Lindau syndrome||12 (2.1)|
|Birt-Hogg Dube syndrome||1 (0.18)|
|Tuberous sclerosis||2 (0.36)|
|Renal insufficiency (MDRD <60 mL/min/1.73m2)||53 (9.3)|
|Obesity (BMI >30 kg/m2)||129 (22.7)|
|Antiplatelet or anticoagulant treatment||91 (16)|
|Mean (sd, range):|
|Haemoglobin, g/dL||13.82 (1.48, 7.7–19)|
|Creatinine, μmol/L||88.5 (34.8, 37–434)|
The median (range) number of patients included per centre was 9 (1–52) and the mean (sd) tumour size was 3.69 (1.9) cm. In 537 patients (94.4%) the tumour was solitary. The indication for PN was imperative in 132 patients (23.2%). The tumour characteristics are given in Table 2.
|Mean (sd, range) tumour size, cm||3.69 (1.9, 0.19–15)|
|Cystic tumour||96 (16.9)|
|Renal sinus||109 (19.2)|
|Upper pole||201 (35.6)|
|Lower pole||196 (34.7)|
The surgical approach was OPN, LPN and robot-assisted PN (RAPN) in 63.3%, 21% and 15.7%, respectively. The mean (sd) tumour size by surgical approach was as follows: 3.9 (2) cm for OPN, 3.14 (1.8) cm for LPN and 3.1 (1.3) cm for RAPN. The rate of conversion from LPN/RAPN to OPN was 4.6% and conversion to radical nephrectomy was of 1.9%. Pedicle clamping was used in 70.6% of patients, parenchyma clamping (mechanical clamping) in 19.6% and no clamping in 9.8%. The mean (sd) ischaemia time was of 15.6 (11.8) min and blood loss was 280.4 (340) mL. Perioperative data are given in Table 3.
|Conversion to open surgery||26 (4.6)|
|Conversion to radical nephrectomy||11 (1.9)|
|Type of PN:|
|Wedge resection||148 (26.2)|
|Mean (sd, range) ischaemia time, min||15.6 (11.8, 0–60)|
|Type of clamping:|
|Collecting system closure||199 (35.2)|
|Collecting system drainage (JJ or ureteric catheter)||66 (11.7)|
|Mean (sd, range) operation time, min||146.6 (55.1, 45–385)|
|Haemostatic incision of parenchyma||54 (9.6)|
|Mean (sd, range) blood loss, mL||280.4 (340, 0–4000)|
|Perioperative transfusion||25 (4.4)|
|Units of PRBC:|
The mean (sd) hospital stay was 8.2 (4.6) days. In all, 13 patients (2.4%) needed a renal selective embolization postoperatively, for 1.6% (nine patients) the embolization was carried out during the hospitalisation, and for 0.6% (four patients) after hospital discharge (Tables 4, 5).
|Mean (sd, range) hospital stay, days||8.2 (4.6, 2–56)|
|Mean (sd, range) haemoglobin, g/dL||11.98 (1.72, 4–17.5)|
|Ultrasound and CT||1 (0.2)|
|No studies||297 (57.1)|
|Hospital readmission||23 (4.4)|
|Mean (sd, range):|
|Haemoglobin, g/dL||12.79 (1.37, 8.8–18.1)|
|Creatinine, μmol/L||98.7 (71.3, 41–885)|
A preoperative biopsy was taken in 11.5% of patients; it was diagnostic in 92.3%. Frozen-section analysis was done in 11.9%. The positive margin rate on definitive pathology was of 7.8%. The mean (sd, range) security margin on definitive pathology was 2.9 (3.0, 0–40) mm. The pathological characteristics are given in Table 5. There was a hospital readmission rate of 4.4% and the overall transfusion rate was 12.1% (Table 6).
|Yes (diagnostic)||60 (10.6)|
|Yes (not diagnostic)||5 (0.9)|
|Benign tumour||3 (5.0)|
|Frozen-section analysis||67 (11.9)|
|Benign tumour||92 (16.6)|
|Tumour necrosis||73 (13.2)|
|Positive margin||43 (7.8)|
|Mean (sd, range) security margin, mm||2.9 (3.0, 0–40)|
A HA was primarily used in 71.4% of patients, with a statistically significant difference among surgical approaches: 67.6% for OPN, 75.6% for LPN and 80.9% for RAPN (P = 0.024). The preferentially used HAs were: Floseal® (49.1%), Tachosil® (32.9%), Surgicel® (12.9%), Tisseel® (3.1%), Quixil® (1.5%), and Bioglue® (0.5%). In 91.8% of cases, a single dose of the HA was sufficient for achieving haemostasis. The timing of the HA application varied according to urologist practices, in 47.3% before unclamping, in 23.1% after unclamping, and in 23.4% before closing the parenchyma. The HA was used either alone (13.9%) or in association with sutures (80.3%). One or more additional haemostatic actions were needed in 12.3% of the cases. It included: new HA application in 67.6% of cases (Surgicel 43.4%, Floseal 30.4%, Tachosil 21.7%), coagulation or sutures (22.1%), iterative renal vessels clamping (7.4%) and total nephrectomy (2.9%).
Floseal was used more frequently in RAPN (68.1%) and LPN (56.7%) than in OPN (39.2%) (P < 0.001; Table 7). There was no difference between the surgical approaches for blood loss (P = 0.951), transfusion rate (P = 0.729) and units of packed red blood cells (PRBC) used (P = 0.656).
|Early unclamping (before parenchyma closure)||174 (31)|
|No use of HAs, reasons:||161 (28.6)|
|Surgical haemostasis sufficient||141 (87.5)|
|Due to cost||30 (18.6)|
|No evidence of benefit||25 (15.5)|
|Other reasons||8 (4.9)|
|First-line adjunct HAs used:||402 (71.4)|
|Single dose used to achieve haemostasis||360 (91.8)|
|Additional use of second-line HAs:||46 (8.2)|
|Timing of HA application:|
|Before unclamping||190 (47.3)|
|After unclamping||94 (23.1)|
|Before closing the parenchyma||131 (23.4)|
|HA used alone||56 (13.9)|
|Ratio of HAs used according to surgical approach, n/N (%):|
|Nephrectomy due to bleeding||2 (0.35)|
|Perioperative transfusion||25 (4.4)|
|Mean (sd, range) blood loss, mL||280.4 (340, 0–4000)|
When comparing patients who received a HA with those who did not receive a HA, we found that there was no statistical difference between the groups for tumour size (3.726 [1.91] vs 3.647 [1.90], P = 0.542), collecting system drainage (P = 0.538), hospital stay (P = 0.508), operation time (P = 0.169), blood loss (312.5 [371.9] vs 267 , P = 0.387) and transfusion rate (P = 0.713) (Table 8).
|Variable||Group with HAs (402 patients)||Group without HAs (160 patients)||P|
|Tumour size, cm||3.726 (1.913)||3.647 (1.898)||0.542a|
|Operation time, min||145.2 (64.8)||147.4 (50.9)||0.169a|
|Blood loss, mL||312.5 (371.9)||267.0 (326.2)||0.387a|
|Hospital stay, days||7.6 (2.8)||8.4 (5.2)||0.508a|
|Collection system drainage||21 (13.0%)||45 (11.2%)||0.538b|
|Transfusion rate||8 (5.0%)||17 (4.3%)||0.713b|
When comparing patients who received Floseal as the HA with those who received other HAs, we found that there was no statistically significant differences between the groups for tumour size (P = 0.502), collecting system drainage (P = 0.821), operation time (P = 0.189) and transfusion rate (P = 0.558). While there were statistically significant differences in hospital stay (P = 0.005) and blood loss (P = 0.027) in favour of the Floseal group, this difference is clinically irrelevant (Table 9).
|Variable||Group with Floseal (194 patients)||Group with other HAs (208 patients)||P|
|Tumour size, cm||3.698 (1.925)||3.599 (1.876)||0.502a|
|Hospital stay, days||8.0 (5.3)||8.7 (5.1)||0.005a|
|Operation time, min||151.3 (52.5)||143.9 (49.3)||0.189a|
|Blood loss, mL||257.6 (390.6)||275.6 (254.0)||0.027a|
|Collection system drainage||21 (10.8)||24 (11.5)||0.821b|
|Transfusion rate||7 (3.6)||10 (4.8)||0.558b|
Technical modifications and refinements as well as a variety of haemostatic tools have been added to the armoury of urological surgeons. In particular, HAs and sealants, with >100 years of history, play an important role.
Today, NSS is an established approach for patients with localised RCC when preservation of renal function is desired. PN is a technically demanding surgery. Significant technical limitations remain for the control of bleeding and closure of the collecting system. An attractive approach on the horizon of local haemostatic and wound control is the use of local tissue sealants. In a multi-institutional LPN study, >80% of urologists used sealants as an adjunct . A recent survey on the current practice for urologist performing LPN, confirmed that HAs were used in 77.4% of cases and were mainly represented by gelatine-matrix thrombin (Floseal), fibrin gel (Tisseel), BSA (Bioglue), oxidised regenerated cellulose (Surgicel) or combinations of these. The overall postoperative bleeding requiring transfusion and urine leakage rates were 2.7% and 1.9%, respectively [4-7]. Triaca et al.  were the first to describe the use of thrombin sealant with no renal artery occlusion in OPN and concluded that PN with no temporary arterial occlusion is technically feasible and safe. While Hidas et al.  compared changes in renal function after NSS using tissue adhesive only (24 patients) vs NSS using standard suturing technique (32 patients), as measured by technetium-99m dimercaptosuccinic acid (99mTc-DMSA) uptake by the kidney. In the tissue-sealant group after NSS there was a mean individual renal functional loss of 11% vs 20% in the suture group (P = 0.02).
While no prospective randomised study in humans has established the role of tissue sealants in NSS, it is widely used in this indication. Knowing that an observational study is the first step before more analytical studies, the aim of NEPHRON was to evaluate and assess the use and results of local HAs used in PN in a large multi-institutional cohort.
The results of the registry showed that a HA was used as an adjunct for PN in 71.4% of patients. Floseal was the preferentially used HA. Among surgical approaches, a HA was used more frequently in RAPN/LPN than OPN, while the tumour sizes were smaller in the RAPN/LPN groups than in the OPN group, this may be due to the fact that in OPN, the control of bleeding is easier for the surgeon. Floseal was used more frequently in RAPN/LPN than OPN, this may be explained by the fact that other synthetic HAs (Tachosil) are difficult to introduce by trocar and that once applied, the surgeon has to hold the sealed parts in the desired position for at least 3–5 min to ensure that the setting glue adheres firmly to the surrounding tissue. This manoeuver may be difficult in LPN/RAPN.
In the present study, HAs were used in association with sutures in 80.3% of cases. Although HAs have been applied with good results, other haemostatic aids, e.g. bipolar coagulation, ultracision, clip suture or sutures, are often used in combination, making an evaluation of the relative merit of HAs difficult and confusing.
When comparing patients who received HAs with those who did not receive a HA, we found no statistical difference in the main variables assessed. Knowing that, the present study was not designed to compare groups and that the practice may have been influenced by knowledge of the patients' treatment status, these results must be taken with caution.
Pharmaco-economic studies are required to better define the place of HAs in improving the safety and simplicity of surgical procedures and decreasing costs.
There are numerous biological adhesives and HAs available to the urological surgeon. As newer biological HAs and sealants come onto the market, the urological surgeon must be critical of the agents' efficacy, safety, biodegradability and cost.
There are some limitations of the present study. As the data collection dates back to 2010 and the techniques of PN have been rapidly evolving, the results of the present study may not precisely reflect the current use of HAs. Also, in the present study, there was no anatomical classification or nephrometry scoring used. Finally, we have to note that 83% of the PNs were done in public hospitals (university hospital, general hospital) and 17% in a private hospital.
In conclusion, HAs are widely used by French urologists during PN. HAs are used significantly more during LPN and Floseal is the more commonly used agent. Progress is needed for standardising HA application, especially regarding timing of application. For the time being, the role of the HA in NSS still needs to be critically evaluated.
The work was funded by Baxter in association with the French Association of Urology. The funding sources had a role in the design and conduct of the study; collection of the data, and no role in the analysis, interpretation of the data; preparation, review, or approval of the manuscript. The authors' work was independent of the funder.
American Society of Anesthesiologists)
body mass index)
Eastern Cooperative Oncology Group)
Prospective National Observational Registry on the Practices of Haemostasis in Partial Nephrectomy)
(laparoscopic) (open) (robot-assisted) partial nephrectomy)
packed red blood cells