Changing practice to stented percutaneous nephrolithotomy (PCNL): a 2 year experience

Results Co-morbidities, anatomical abnormalities, stone burden, operative time, number of punctures, access and infection rates were comparable. The stented technique was employed in complex stones (staghorn n = 13, multiple stones n = 11) and in patients with anatomical abnormalities ( n = 9). Overall stone free rate was 92.8% and 82% for the stented and the conventional groups, respectively. No haemorrhage or ureteric obstruction was reported in the stented group. All stent removals were uneventful. Mean LOS and analgesic requirement of diclofenac were markedly less in stented group. The number of stented procedures tripled from year 1 ( n = 14) to year 2 ( n = 42). Mean LOS dropped steadily from 68.6 h in the first 6 months ( n = 5) to 47.4 h in the last 6 months ( n = 22) with increased experience. Twenty-four patients (42%) were fit for discharge within 35 h. Conclusion In our experience, the stented technique was well tolerated and has reduced the length of hospital stay with no evidence of loss of efficacy or increased morbidity even in complex stones or patients with anatomical abnormalities. Keywords PCNL Ureteric stent Tubeless Nephrostomy Introduction Since the introduction of percutaneous nephrolithotomy in 1976 by Fernstrom and Johannson [1] the procedure has been continuously evolving to be less invasive. Stented PCNL is an increasingly practised technique which involves the placement of a ureteric double-J (JJ) stent antegradely at the end of the procedure with no nephrostomy. Advocates of the stented technique, also known as tubeless PCNL, have concluded that hospitalization time, convalescence and analgesic requirements are significantly lower compared to the conventional technique (nephrostomy post-procedure) without any compromise to patient outcome or safety [2–7] . In our unit, the use of JJ stents without nephrostomy to ensure renal drainage following PCNL has been gradually introduced since October 2005. We present our results from one of the biggest UK series of stented PCNL to assess the impact of this technique on the safety, efficacy, length of stay and post-operative comfort. This study also examines the factors used by the surgeon in selecting a stent over a conventional nephrostomy. Patients and methods A retrospective review of case notes and computer records was performed on all consecutive patients who underwent stented PCNL with a subcostal approach at Sunderland Royal Hospital from October 2005 to October 2007. The use of a stent rather than a nephrostomy was decided at the time of surgery based on the following parameters: straightforward puncture with no complication at time of puncture, minimal renal bleeding and no intra-operative damage to the collecting system. Apparent complete stone clearance was also required for a JJ stent to be placed otherwise a nephrostomy tube was left to allow a second PCNL a week later. The need for multiple tracts was not a contraindication to placing a JJ stent. All patients had routine pre-operative work-up including urine for microbiological analysis and pre-operative imaging in the form of IVU or CT urography with 3D reconstruction. All cases were performed under general anaesthesia and received antibiotics upon induction. A cystoscopy and retrograde pyelogram was performed with placement of a retrograde catheter in the collecting system of the affected side and secured alongside a 16-Fr Foley catheter. The patient was turned into the prone position on a Wilson frame to optimise access. The collecting system was delineated via the retrograde catheter with contrast medium (Urografin 150) containing methylene blue. The collecting system was punctured, followed by the placement of two parallel guide wires down the ureter into the bladder. Further calyceal access routes were established as required with guidewire placement to achieve optimum stone clearance. The primary access tract was dilated with a balloon dilatation system (Nephromax or Maxforce) with placement of an access sheath. Stone extraction was undertaken using stone graspers or basket with secondary tract dilatation or stone fragmentation using ultrasound lithotripter or lithoclast as required. A 6-French ureteric JJ stent was placed antegradely at the end of the procedure. Then the access sheath and guidewires were removed with no nephrostomy. No adjuncts to achieve tract haemostasis were used. The skin was closed with an absorbable suture. The urethral catheter was removed on day 1 postoperative if the patient was mobilizing and apyrexial. The JJ stent was removed 1–4 weeks later using flexible cystoscopy under local anaesthesia. All patients were given a date for stent removal before discharge. Patient demographic data and peri-operative details were collected including co-morbidities, urinary tract anomalies and stone characteristics. Patients were assessed in terms of complications, the length of hospital stay in hours and analgesic requirement post-operatively. Stone free rates were recorded. The scoring system of the Society of Cardiovascular and Interventional Radiology was used to grade the complications into minor or major [8] . Minor complications were those requiring observation alone with no active treatment, e.g. self-limiting haematuria and pyrexia within 24 h post-procedure with negative culture, while major complications included patients requiring specific therapy or increased level of care (e.g. infection, ureteric obstruction, bleeding requiring transfusion or embolisation). These parameters were compared to another cohort of 50 patients who underwent PCNL using the conventional nephrostomy tube at the end of the procedure. Results The stented technique (post-procedure JJ stent only) was used in 56 cases performed by 3 surgeons over a 2-year period. The next 50 standard nephrostomy patients were also examined to provide a comparison cohort. Comparative data on the two cohorts is shown in Table 1 . The mean age for patients in the stented group was 54.7 (range 15–80) versus 51.2 (range 20–81) in the nephrostomy group. The gender distribution was predominantly female in the stented cohort with 71% women and 29% men versus 46% women and 54% men in the nephrostomy cohort. The number of stented procedures tripled from year 1 ( n = 14) to year 2 ( n = 42). The number of patients with ASA grades 1 and 2 was 47 and there were 9 patients with ASA grade ≥3 in the stented cohort versus 10 patients in the nephrostomy cohort. Anatomical abnormality was not a contraindication to placing a stent. The abnormalities seen were horseshoe kidney, calyceal diverticulum, bifid pelvis, duplex, solitary kidney and ileal conduit. See Table 2 . The mean operating time was 145.5 min (range 86–253) and 133 min (range 60–240) for the stented and the conventional groups, respectively. A single puncture was required in most cases to access the stone, with 48 single punctures in the stented group and 45 in the conventional group. Eight patients in the stented group required ≥2 punctures to achieve satisfactory stone clearance compared to 5 patients in the conventional group. The maximum was 3 punctures in 2 patients in each group. The lower calyx was the most common location of entry followed by the middle and upper calyces with 59% ( n = 33), 26.7% ( n = 15) and 14% ( n = 8) in the stented group and 82% ( n = 41), 4% ( n = 2) and 14% ( n = 7) in the nephrostomy group, respectively. There were 32 patients with single stones with mean size of 14.7 mm (range 7–25) in the JJ group versus 19 patients in the nephrostomy group with mean size of 13.5 mm (range 6–20). The stented technique was also used in a variety of complex stones when prolonged manipulation within the collecting system was required to achieve stone clearance. There were 11 patients with multiple stones (maximum = 3 stones per unit) with overall mean size of 11.7 mm (range 5–20) in the stented group versus 17 patients (maximum = 4 stones per unit) with mean size of 11.5 mm (range 4–25) in the nephrostomy group. Thirteen patients had staghorn stones with mean size of 33 mm (range 25–42 mm) in the JJ group versus 14 patients in the nephrostomy group with a mean size of 35 mm (range 30–40). Ninety-one percent of patients ( n = 51) were stone free following the primary procedure confirmed on screening and on post-operative plain KUB X-ray at follow up in the JJ cohort versus 72% in the nephrostomy cohort. The overall stone free rates for single, multiple and staghorn stones were 100%, 82% and 85% in the stented group versus 100%, 60% and 50% in the conventional group, respectively. Only one patient required Shock-Wave Lithotripsy (SWL) for a 9 mm residual fragment in the upper pole detected on follow up plain KUB X-ray to achieve complete stone clearance in the JJ group. In the nephrostomy group 10% ( n = 5) required a second procedure to achieve complete stone clearance. Two patients had repeat PCNL, one patient had combined ureteroscopy and repeat PCNL, one patient required ureteroscopic stone extraction and only one patient required ESWL for a 7 mm stone. Second procedure stone free rate rose to 92.8% ( n = 52) in the stented cohort and 82% ( n = 41) in the nephrostomy cohort. In the stented group, 7.2% ( n = 4) in comparison to 18% ( n = 9) in the conventional group had small residual stone burden (largest was 9 mm) which required follow up and no further intervention. The biochemical stone analysis for the stented cohort showed that 60% were calcium oxalate, 12.5% uric acid, 12.5% triple phosphate, 10% calcium phosphate and 1.7% cystine stones. In comparison the nephrostomy cohort showed that 66% were calcium oxalate, 6% calcium phosphate, 6% uric acid, 20% triple phosphate and 2% cystine stones. There was no haemorrhage requiring transfusion and no pneumothorax or other visceral injuries were reported in either group. Only one patient (1.7%) had post-operative urinary tract infection in the stented group versus 2 patients (4%) in the nephrostomy group. Only one patient required early removal of the stent on the second post-operative day due to severe stent related symptoms associated with antegrade migration. None of the patients required early re-admission for stent removal. All patients had the stent removed uneventfully as planned. No ureteric obstruction was reported in our series even after removal of JJ stent. In the conventional group, 8 patients developed significant pain on clamping the nephrostomy and underwent a nephrostogram to exclude ureteric obstruction. Four patients were found to have UPJ obstruction secondary to a clot or oedema, 3 of them underwent insertion of JJ stent antegradely and in the fourth patient the nephrostomy was removed a week later. Two other patients had upper ureteric residual fragments. One required emergency antegrade ureteric stenting and further ureteroscopy and stone extraction. The other had a second look PCNL via the same access a week later. One patient was admitted 2 weeks after conventional PCNL with a partially obstructed infected system secondary to a missed ureteric fragment and required emergency retrograde insertion of JJ stent. Four weeks later, she underwent uretroscopic stone extraction and repeat PCNL for a 20 mm residual fragment. No ureteric obstruction was reported in the stented group even after removal of JJ stent. Overall complications can be seen in Table 3 . The mean requirements of morphine, diclofenac and tramadol in the first 48 h post-operatively were 12.65, 25 and 11.8 mg for the stented cohort versus 12.67, 83.7 and 53.5 mg for the conventional group, respectively. The overall mean length of hospital stay (LOS) was 57.8 h (range 27–149 h) in the JJ cohort versus 124.9 h (range 31–298 h) in the nephrostomy cohort. Table 4 shows the significant reduction in LOS in the stented group regardless of the stone burden. This could be noted especially in the single stone patients where complete clearance was achieved in both groups however the mean LOS in the stented group was 58 h compared to 120 h in the conventional group. The mean LOS in the first year was 68.5 h (2.8 days) ( n = 14) compared to 46.7 h (1.9 days) for the last 14 patients in our series of stented patients. The projection in Fig. 1 shows that the mean LOS dropped steadily from 68.6 h in the first 6 months ( n = 5) to 47.4 h in the last 6 months ( n = 22) with increased experience. Twenty-four patients (42%) had a mean LOS of 31 h (1.25 days, range 27–35 h). Discussion Previous studies have concluded that stented PCNL (also known as tubeless PCNL) is associated with lower morbidity and quicker recovery time than conventional PCNL. We have confirmed this finding as demonstrated by decreased post-operative requirements of diclofenac and tramadol between our two cohorts. A recent prospective randomised comparative study between stented and conventional PCNL by Madhu et al., with 101 patients in each arm, demonstrated statistically significant advantages of stented PCNL in terms of postoperative comfort, complications and hospital stay. The inclusion criteria for their study were single tract procedure with complete clearance and minimal bleeding at completion. The mean LOS in this study was 21.8 h (0.9 day) in the stented group compared to 54.2 h (2.25 days) in the conventional group [9] . Our cohort was not randomised and we have also included patients with multiple punctures, complex stones and abnormal anatomy. There are several reports on totally tubeless PCNL (no nephrostomy, no ureteric stent) as an option in selected patients [10,11] . A recent case series of 100 patients who underwent totally tubeless PCNL had a mean LOS of 2.9 days [12] which is longer than in our stented cohort (see Table 5 ). Over the course of our study period there was an increase in the use of JJ stents over nephrostomy placements, 36% (14/40 cases) in the first year to 64% (42/66 cases) of all PCNLs in the second year. The rise in the number of stented procedures correlated with reduced overall mean LOS (stented + conventional) in Year 2 (72 h) in comparison to Year 1 (116.8 h). A satisfactory learning curve is achieved after 50 cases of stented PCNLs [13] . In our series, 56 stented cases were performed. The trend in Fig. 1 suggests continuing decreasing LOS with increased experience. More recently studies have suggested that stented PCNL can be safely and effectively employed based on intra-operative factors without pre-operative contraindications [7] . Authors only advocate the use of a nephrostomy tube at the end of PCNL if a second look via the same entry site is required for high residual stone burden, if significant intra-operative bleeding occurred from the access tract with failure to control with Amplatz Sheath, or if more than three punctures were required to access the collecting system [6] . A study by Zaman et al., with 170 cases of stented PCNL, reported that five patients developed ureteric obstruction and required postoperative emergency nephrostomy insertion [14] . In our case series, no ureteric obstruction was reported even after removal of stent. This was attributed to careful screening of the ureter to ensure the absence of missed ureteric fragments before the JJ stent was deployed. Shah et al., in a randomised prospective study of 46 patients, concluded that stented PCNL can be performed even in patients with solitary kidney, up to 3 renal punctures or supracostal access [6,17] . However in our series we did not use a JJ stent in supracostal punctures preferring a nephrostomy tube to ensure renal drainage did not occur into the thoracic cavity. Bellman et al. also concluded that stented PCNL is effective and safe in patients with moderate stone burden [4] . Malcolm et al., in a case series of 47 stented PCNL for complex renal stone disease, found the technique to be safe and effective [16] . In our case series, stented technique was employed safely in patients with ASA grades 3 and 4, complex stones, anatomical abnormalities, multiple punctures and residual stone burden. Singh and co-workers reported that only 2/30 stented patients had discomfort related to the stent on day 3 postoperative in comparison to 11/30 that had discomfort related to a 22-Fr nephrostomy tube in the conventional group [15] . The stented technique was well tolerated with decreased analgesic requirement post-operatively during the hospital stay. None of the patients required early admission with stent related symptoms. The results of our study should be interpreted within the context of its limitations as it is an observational study. However, it showed that the stented technique was safe without compromising the outcome of PCNL when compared to a similar cohort of nephrostomy treated patients. Although we accept that stenting patients requires a second procedure to remove the stent at a later date, the cost of this was offset by decreased hospital stay and increased patient comfort and convenience. Conclusion In our experience the stented technique was well tolerated and reduced the LOS without compromising patient safety or the outcome of PCNL even in the complex stone scenarios. Therefore, our current practice is to employ the stented technique in all cases unless there is significant bleeding, contrast extravasation from the collecting system or residual stone burden requiring a second look PCNL through the same access. Conflict of interest None declared. References [1] I. Fernstrom B. Johannson Percutaneous pyelolithotomy. A new extraction technique Scand J Urol Nephrol 10 1976 257 259 [2] G.C. Bellman R. Davidoff J. Candella J. Gerspach S. Kurtz L. Stout Tubeless percutaneous renal surgery J Urol 157 1997 1578 1582 [3] K.M. Delnay R.W. Wake Safety and efficacy of tubeless percutaneous nephrolithotomy World J Urol 16 1998 375 377 [4] J. Limb G.C. Bellman Tubeless percutaneous renal surgery: review of first 112 patients Urology 59 2002 527 531 [5] M.R. Desai R.A. Kukreja M.M. Desai S.S. Mhaskar A prospective randomized comparison of type of nephrostomy drainage following percutaneous nephrolithotomy: large bore vs small bore vs tubeless J Urol 172 2004 565 567 [6] H.N. Shah V.B. Kausik S.S. Hedge J.N. Shah M.B. Bansal Tubeless PCNL a prospective feasibility study & review of previous reports BJU Int 96 2005 879 883 [7] M. Sofer Beri A. Friedman G. Aviram N.J. Mabjeesh J. Chen Extending the application of tubeless PCNL J Urol 172 2 2004 565 567 [8] P. Ramchandani J. Cardella C. Grassi A. Roberts D. Sacks M. Schwartzberg C. Lewis Quality improvement guidelines for percutaneous nephrostomy J Vasc Interv Radiol 12 2001 1247 1251 [9] S. Madhu Agrawal Mayank A randomized comparison of tubeless and standard percutaneous nephrolithotomy J Endourol 22 3 2008 1 4 [10] S.M. Aghamir S.R. Hosseini S. Gooran Totally tubeless percutaneous nephrolithotomy J Endourol 18 2004 647 648 [11] V. Gupta Sadasukhi K.K. Sharma R.G. Yadav R. Mathur Tubeless and stentless percutaneous nephrolithotomy BJU Int 95 2005 905 906 [12] T.J. Crook C.R. Lockyer S.R. Keoghane B.H. Wamsley Totally tubeless percutaneous nephrolithotomy J Endourol 22 2008 267 271 [13] Sofer M, Mabjeesh J, Lidawi G, Keren Paz G, Chen J, Kaver I, et al. Defining the learning curve for tubeless PCNL. EAU March 2007 ABST [873]. [14] Z.R. Zaman S.S. Kommu N.A. Watkin C.R. Jones P.J. Boyd Tubeless percutaneous nephrolithotomy in the management of renal calculi BJU Int 97 2006 43 BAUS Abstracts P032 [15] I. Singh A. Singh G. Mittal Tubeless percutaneous nephrolithotomy: is it really less morbid? J Endourol 22 3 2008 427 433 [16] J.B. Malcolm I.H. Derweesh E.K. Brightbill R. Mehrazin C.J. DiBlasio R.W. Wake Tubeless percutaneous nephrolithotomy for complex renal stone disease: single center experience Can J Urol 15 June (3) 2008 4072 4077 [17] H.N. Shah S.S. Hegde J.N. Shah M.B. Bansal Safety and efficacy of supracostal access in tubeless percutaneous nephrolithotomy J Endourol 20 12 2006 1016 1021 Dec