Percutaneous stone surgery using a tubeless technique with fibrin sealant: report of our first 107 cases

Authors


James O. L'Esperance, Department of Urology, 34800 Bob Wilson Drive, Naval Medical Center San Diego, San Diego, CA 92134, USA. e-mail: james.lesperance@med.navy.mil

Abstract

Study Type – Therapy (case series)

Level of Evidence 4

What's known on the subject? and What does the study add?

Small case series support the safety and efficacy of tubeless PCNL with fibrin sealant. However, there is a paucity of data from larger case series supporting this approach.

To our knowledge, this is among the largest tubeless PCNL series. We found the use of fibrin sealant for tubeless PCNL was associated with excellent stone-free rates (approaching 90%), short hospitalisation, and low complication rates. Tubeless PCNL with nephrostomy tract fibrin sealant appears to be viable option for appropriately select patients.

OBJECTIVE

  • • To report on our first 107 cases of tubeless percutaneous nephrolithotomy (PCNL) using fibrin sealant as a haemostatic agent within the access tract. PCNL is the preferred treatment for patients with large renal stones, and the tubeless technique with the use of fibrin sealant has recently gained popularity.

PATIENTS AND METHODS

  • • We performed a retrospective review of single-access, PCNL cases performed without a nephrostomy tube from January 2002 to July 2008.
  • • Nephrostomy tracts were sealed at the conclusion of each procedure with fibrin-containing haemostatic agents.
  • • We evaluated demographic variables, tracked complications, and compared pre- and postoperative haemoglobin, haematocrit and creatinine levels.
  • • On postoperative day 1 computed tomography was used to determine stone-free rates.
  • • Student's t-test calculations were used to determine statistical significance at P≤ 0.05.

RESULTS

  • • In all, 59 men and 48 women with a mean age of 43 years were included in the analysis of 107 cases. The mean stone size was 2.9 cm2 and the average hospital stay was 1.07 days.
  • • Pre- and postoperative changes in serum haemoglobin and serum creatinine were not statistically different. Postoperative haematocrit declined by a mean of 4.5% (P≤ 0.05), but no patients required a transfusion.
  • • Stone-free rates were 72% overall, and 90% when excluding patients with residual fragments of <4 mm.
  • • Complications included seven asymptomatic subcapsular haematomas, one pseudoaneurysm requiring selective embolization, one urine leak, and five return visits to the emergency room for pain.

CONCLUSIONS

  • • The use of fibrin sealant in this large tubeless PCNL series was associated with favourable stone-free rates, short hospital stays, and low complication rates with no significant bleeding.
  • • Tubeless PCNL with nephrostomy tract fibrin sealant appears to be a viable option for appropriately selected patients, but future randomised trials are warranted.
Abbreviations
PCNL

percutaneous nephrolithotomy

NT

nephrostomy tube

INTRODUCTION

Percutaneous nephrolithotomy (PCNL) was first described in 1955 and then more widely adopted in 1976 for removal of large renal stones [1]. This method of removing renal stones was a dramatic improvement over open surgical procedures and was shown to shorten hospital stay and decrease complications [1]. PCNL has now become the primary method for removal of large upper tract stones. PCNL has traditionally required placement of a nephrostomy tube (NT) at the conclusion of the surgery, which can be removed postoperatively in most cases ≤5 days of the procedure. This technique serves to ensure proper renal drainage, control bleeding, and facilitate a second look in the event of postoperative haemorrhage or residual calculi. Michel et al. [2] published an article in 2007 detailing complication rates in >1000 of their own PCNLs and reviewed previously published results. They reported overall complication rates of 83% with most being clinically insignificant. Major complications requiring intervention included haemorrhage (0.6–1.4%) and sepsis (0.9–4.7%).

In an effort to maintain PCNL effectiveness while decreasing complications and patient discomfort associated with prolonged NT drainage, innovative urologists began to experiment with tubeless PCNL in the late 1990s. Since then, only a few tubeless PCNL series have been reported. Omission of the postoperative NT has several advantages over the traditional approach, including decreased hospital confinement and less postoperative pain resulting in decreased analgesic requirements [1,3–6]. Investigators have since begun to embrace the tubeless PCNL concept for select patients, with or without the use of haemostatic agents [7]. In the present study we report on our first 107 patients undergoing tubeless PCNL augmented with fibrin sealant.

PATIENTS AND METHODS

Under an Institutional Review Board-approved protocol using our kidney stone database, we analysed 107 tubeless PCNL procedures performed with the use of a fibrin sealant at our institution between January 2002 and July 2008. During this period, 129 PCNL procedures were performed. All cases were performed by our group of staff surgeons, as well as urology resident physicians. Although, no specific inclusion or exclusion criteria were used to select patients for a tubeless approach, those with significant intraoperative haemorrhage, collecting system injury and those with more than one access tract were not selected for this approach. Although stone composition was not a criterion for inclusion or exclusion in our series per se, complex branched stones, e.g. infection stones made of struvite, often required multiple access tracts and therefore were not candidates for tubeless PCNL.

Percutaneous access was achieved by the urologist in 20 cases (19%) and by an interventional radiologist in 87 cases (81%). In the operating room after Foley catheter placement, balloon dilatation was performed over a super-stiff working guidewire. A 30 F access sheath was then advanced to maintain nephrostomy tract access throughout the procedure. Stone fragmentation was achieved with a pneumatic/ultrasonic combination lithotrite, ultrasonic lithrotrite alone, or ultrasonic lithotrite combined with holmium laser. An intraoperative nephrostogram was taken to ensure stone clearance was achieved and that there was not significant collecting system injury.

Antegrade JJ ureteric stent placement was performed with confirmation of proper location by fluoroscopy and direct nephroscopic visualisation. Then using nephroscopic visualisation, the percutaneous access sheath was withdrawn to the junction of the collecting system with the parenchymal junction. With a 17.8-cm, 18 G needle, 2–5 mL of fibrin sealant (Tisseel®, Baxter International, Deerfield, IL or EvicelTM, Ethicon, Inc., Somerville, NJ, USA) was instilled into the renal cortical defect as the access sheath was withdrawn ≈2 cm. The cost of the Evicel fibrin sealant was $501.19/unit and the cost of the Tisseel fibrin sealant was $58.19/unit. Only one unit was typically required per procedure. An obturator was advanced under fluoroscopic guidance down the sheath to level of the fibrin sealant plug to maintain its position while the sheath and guidewire were removed. The skin was closed using absorbable sutures.

Postoperative three-phase haematuria protocol CT was obtained in all patients on the morning of postoperative day 1, to evaluate for residual stone fragments, significant renal haematoma, and urinary extravasation. The Foley catheter was generally removed on postoperative day 1 before discharge. Stents were removed via office cystoscopy 5–7 days postoperatively, provided patients were stone free.

We recorded demographic and stone parameters into the database, including age, gender, stone size, and location. We also made a subjective assessment of bandage soilage on postoperative day 1. We evaluated pre- and postoperative creatinine, haemoglobin and haematocrit to calculate significant changes in these values. In-hospital complications as well as emergency department visits and re-admissions were reviewed and recorded in all patients. Stone size was reported using stone area and calculated from CT by multiplying the length x width in centimetres at the level of the greatest length of the stone. Surgical success was defined by the absence of residual fragments of ≥4 mm on the postoperative day 1 CT scan. This CT was used to calculate stone-free rates. The mean and median calculations were performed on continuous variables. Student's t-test calculations were used to analyse data and statistical significance for pre- and postoperative laboratory values was determined by P≤ 0.05.

RESULTS

Demographic variables and perioperative data from our 107 patients are shown in Table 1. The mean (range) patient age was 43 (20–77) years and mean stone size was 2.9 (0.45–12.0) cm2. Pre- and postoperative PCNL creatinine and haemoglobin were not statistically different, with a mean change in creatinine of 0.04 mg/dL, and a mean change in haemoglobin of 1.4 g/dL. The pre- to postoperative change in haematocrit was statistically significant with a mean declinep of 4.5%, P≤ 0.05. However, this change had little clinical impact, as all patients remained haemodynamically stable and there were no postoperative blood transfusions. Assessment of the 5 × 5-cm coverlet dressing showed minimal or no fluid staining on the central portion of the dressing in all patients, suggesting negligible access tract drainage.

Table 1. Demographics and perioperative data of patients that underwent tubeless PCNL
VariableValue
Number of patients107
Mean (range) age, years43 (20–77)
Gender, male/female, n59/48
Mean (range) preoperative: 
 Serum creatinine, mg/dL0.92 (0.4–1.8)
 Haemoglobin, g/dL14.1 (10.0–17.8)
 Haematocrit %41.4 (30.3–53)
Mean (range) stone area, cm22.9 (0.45–12.0)
Stone location, n: 
 Renal pelvis64
 Upper pole2
 Middle pole8
 Lower pole46
Mean (range) postoperative: 
 Serum creatinine, mg/dL0.96 (0.3–2.3)
 Haemoglobin, g/dL12.7 (8.1–16.4)
 Hematocrit %36.9 (23.7–49.3)
Creatinine change, mg/dL0.04
Haemoglobin change, g/dL1.4
Haematocrit change %4.5
Hospital stay, days1.07
Patients with hospital stay >24 h, n (%)8 (7.3)

Of the 107 patients, 72% were stone free on postoperative day 1 as assessed by three-phase CT. When using our criteria of procedural success, defined as target stone clearance and absence of residual fragments of ≥4 mm, our stone-free rate increased to 90%.

Table 2 details complications identified in our cohort. Complications occurred in 13% of our cohort, with postoperative pain being the most common. Major complications included one pseudoaneurysm and five return visits to the emergency department ≤1 week of surgery for pain. The patient with the pseudoaneurysm presented 13-days postoperatively with complaints of haematuria. Selective renal angiography showed the pseudoaneurysm, which was embolized without further incident. Minor complications included seven asymptomatic subcapsular haematomas noted incidentally on CT on postoperative day 1, and one urine leak as shown on the contrast phase of the CT, which was resolved by stent re-positioning.

Table 2. Surgical outcomes and complications from 107 tubeless PCNLs
Variablen/N (%)
Stone-free rate: 
 Immediate stone-free status77/107 (72)
 Immediate residual fragment <4 mm96/107 (90)
Minor complications: 
 Subcapsular haematoma7/107 (6.5)
 Urine leak1/107 (0.9)
Major complications: 
 Blood transfusion0
 Pseudoaneurysm1/107 (0.9)
 Pain (return visit to ER for management)5/107 (4.6)
Total no. complications14/107 (13)

DISCUSSION

This retrospective study of 107 tubeless PCNLs for large renal stones represents the second largest published tubeless PCNL experience. We selected patients for a tubeless approach based on a single percutaneous access, absence of significant renal haemorrhage or perforation, and adequate clearance of the target stone. Using these criteria, nearly 83% (107/129) of patients at our institution underwent a tubeless procedure. With a mean stone size of 2.9 cm2, we achieved an absolute stone free rate of 72%, and 90% were stone free when excluding fragments of <4 mm. These results are consistent with published data using rigorous criteria and based on immediate postoperative CT (Table 3) [4,5,8,9]. Patients were discharged on postoperative day 1 in most cases. We identified no significant changes in postoperative haemoglobin or creatinine using this technique, but did find a significant decrease in mean haematocrit of 4.5% that was not clinically meaningful. Complications occurred in 13% of the present cohort, with postoperative pain being the most common complaint. There were no transfusions required, but one patient returned with a pseudoaneurysm requiring angioembolization. Minor complications included one urine leak and seven subcapsular haematomas. Tubeless PCNL in the present series was associated with favourable stone-free rates, a short hospital stay, and low complication rates with no significant bleeding.

Table 3. Complications and stone-free rates in tubeless PCNL as reported in other series
ReferenceYearNumber of patientsFibrin sealantComplications, %Stone-free rate, %Stone-free rate measured by:
  1. KUB, plain abdominal radiograph of the kidneys, ureters and bladder.

Present series2011107Yes1390CT
Sofer et al. [9]2010200No1491CT
Shah et al. [8]200617Yes1891X-ray
Limb and Bellman [4]200286No793KUB or CT
Portis et al. [7]200137No595Not specified

Since the first percutaneous renal surgery performed for stone removal 30 years ago, the technique has become safer and more effective for patients with renal stones. PCNL is currently the primary method for removal of large upper tract stones [1]. While PCNL was traditionally performed with postoperative NT placement, use of this measure has not proven necessary in many patients undergoing PCNL. Portis et al. [7] concluded that the need for NT placement can be assessed accurately intraoperatively when those with significant haemorrhage or residual stone fragmentation requiring a ‘second look’ are encountered.

Over the last decade several case reports and small series have been published on the tubeless PCNL technique. Table 3 identifies the largest tubeless PCNL series and compares perioperative variables. The present study compares favourably to those punlished with an excellent stone-free rate and a low transfusion and complication rate. The tubeless approach also appears to limit pain and leads to a shorter hospital stay, a finding that we and others have identified. Sofer et al. [9] reported on 200 (71%) tubeless PCNLs from their series of 281 cases. With a mean (range) stone burden of 924 (400–3150) mm2, they had a mean (range) operative duration of 99 (45–210) min, a complication rate of 14%, and an immediate stone-free rate of 91%. The transfusion rate was 1% and the median (range) hospital stay was 1 (1–15) days.

We have previously reported on a retrospective series of PCNL patients with and without the adjunctive use of the NT. No adverse effects related to fibrin sealants were noted in the tubeless group, yet the patients in the tubeless group were discharged from the hospital on average 1.2 days earlier than the group with the NT [3]. The present series of 107 tubeless cases using fibrin sealant suggests no clinically significant changes in serum creatinine, haemoglobin or haematocrit, nor complications related to the use of sealants.

Haemostatic agents and tissue adhesives have been used for many years across all surgical disciplines. Urologists routinely use fibrin sealants and gelatin matrix thrombin for tissue adhesion, haemostasis, and urinary tract sealing. In 1998, Tisseel Fibrin Sealant was approved by the USA Food and Drug Administration for use in the USA. Evans and Morey [10] list the various uses for tissue sealants in urological surgery. When using haemostatic agents in the collecting system, knowledge of their reaction when in contact with urine must be considered. Uribe et al. [11] placed several different haemostatic agents in contact with urine in vitro, including Tisseel. It was seen to form a solid upon first contact with urine that changed to a semisolid gelatinous state over the subsequent 5 days. The significance of this finding in vivo was not concluded at that time of that study; however, concern for potential urinary tract obstruction by the fibrin sealant have been raised if placed directly into the collecting system.

The use of fibrin sealants for tubeless PCNL is well described on both porcine and human renal units. Kim et al. [12] concluded that tissue sealants, including Tisseel amongst others, directly injected into the renal pelvis was associated with obstruction of the collecting system and development of urinoma in the porcine model, which did not resolve over 5 days. Recommendations to avoid obstructive complications include placement of an occlusion balloon into the infundibulum of the access calyx to prevent inadvertent entry of haemostatic agent into the collecting system. This finding has not been reported in human use. Mikhail et al. [13] first reported on the use of fibrin glue sealants in 20 patients undergoing PCNL and concluded that the use of the glue was safe and was associated with decreased hospital stay. Similar to the present findings, no statistical or clinical difference was found in postoperative haematocrit when compared with tubeless cases without the use of fibrin glue. Noller et al. [14] used a fibrin sealant after PCNL in eight patients and found it safe and feasible in sealing the renal parenchymal defect. Nagele et al. [15] have reported that the use of fibrin sealants after tubeless micro-PCNL when compared with NT placement was associated with decreased hospital stay and decreased postoperative pain.

Investigators have begun to challenge the use of haemostatic agents at the conclusion of tubeless PCNL, showing equivalence in postoperative stone-free rates, hospital stays, and estimated blood loss without the expense of the agent. Shah et al. [8] when comparing tubeless PCNL with and without tissue sealants favours the use of tissue sealants for the reduction of postoperative pain.

Limitations of the present study include the retrospective nature of the project, the difficulties in measuring intraoperative blood loss, lack of historical data on the incidence of asymptomatic subcapsular haematomas after PCNL, and no direct comparison group of NT patients or tubeless PCNL without fibrin sealant adjunct. The present stone-free rates for this tubeless PCNL cohort fall in the middle range of other PCNL series. The present results are probably impacted by the selection criteria of patients used for the tubeless approach and the rigorous criteria used to identify residual fragments. All of the present patients underwent postoperative CT on the morning of postoperative day 1. This is a high standard and helped to identify not only residual stones, but also subcapsular haematomas and urine leaks. Stone-free rates in previous series have more typically used plain abdominal radiograph of the kidneys, ureters and bladder or delayed CT evaluations. Finally, although we are unable to assess the specific additional benefit of haemostatic agents for tubeless PCNL in this retrospective study, the present results are similar to others using this technique. Ultimately, we hope to begin a prospective randomised trial to determine the necessity of fibrin products for nephrostomy tract sealing at the time of PCNL.

In conclusion, the present series of 107 tubeless PCNL cases using fibrin glue between 2002 and 2008 represents one of the largest. We found this approach was associated with a short hospital stay of only 1 day in most cases, a finding noted in previous studies. Decreased hospital stay may translate into lower healthcare costs. Tubeless PCNL was associated with few postoperative complications, and minimal changes in haemoglobin, haematocrit, or creatinine. Thus, we have found tubeless PCNL with fibrin sealant to be safe, effective, and associated with few postoperative complications. At our institution it has become the method of choice for treatment of large renal stones in properly selected patients. The retrospective nature of the present series warrants additional prospective and randomised studies to compare tubeless PCNL techniques with traditional PCNL as well as without the use of fibrin sealant.

CONFLICT OF INTEREST

None declared. Disclosure: The views expressed in this article are those of the authors and do not reflect the official policy or position of the Department of the Navy, Department of Defense, or the United States Government.

Ancillary