Abstract
Introduction:
Several clinical series of retrograde nephrostomy for percutaneous nephrolithotomy (PCNL) have been published over the past 30 years demonstrating good outcomes and safety. We previously reported our adaptation of the Lawson technique, wherein we deploy the puncture wire through a flexible ureteroscope. We herein aim to clarify the performance characteristics of this nephrostomy creation technique.
Materials and Methods:
Institutional Review Board approval and informed consent were obtained. A ureteroscopy-assisted retrograde nephrostomy (UARN) procedure was performed as described previously. Data were collected prospectively. Multiple patient and operative factors were evaluated for association with UARN success and nephrostomy creation time: body mass index (BMI), skin-to-stone distance, Guy's score, Clinical Research of the Endourological Society nephrolithometric score, hydronephrosis, stone burden, location of nephrostomy, exit from a stone-bearing calix, and use of holmium laser to access calix.
Results:
Nephrostomy was successful in 49/52 UARN procedures (94%). Only single access was placed: upper—18, mid—27, and lower—7. Median BMI was 29 kg/m2 and median time for nephrostomy creation was 39 minutes. Fluoroscopy time for the entire PCNL including nephrostomy creation was 84 and 16 seconds for case numbers 1–25 and 26–52, respectively. By stepwise linear regression, variables correlating with nephrostomy creation time were BMI (r 2=0.219), stone burden (r 2=0.094), use of holmium laser to access calix (r 2=0.104), and total r 2 linear=0.416.
Conclusions:
UARN is an intuitive safe procedure that offers dramatic reductions in fluoroscopy times. UARN is best suited to cases requiring only one nephrostomy tract. Upper pole access is commonly performed with a subcostal technique to navigate the puncture wire below the ribs. Increasing BMI best predicts longer nephrostomy creation times; procedure failure was associated with BMI exceeding 40 kg/m2. UARN is a robust technique for nephrostomy creation in appropriately selected patients.
Introduction
S
We aim to clarify which patient and stone-related factors are associated with success and ease of the ureteroscopy-assisted retrograde nephrostomy (UARN) technique.
Materials and Methods
Institutional Review Board (IRB) approval and informed consent were obtained. The UARN procedure was performed as described previously. Data were collected prospectively. Factors evaluated in relation to procedure success and nephrostomy creation time were body mass index (BMI), skin-to-stone distance (SSD), 9 Guy's stone score, 10 Clinical Research of the Endourological Society (CROES) nephrolithometric score, 11 degree of hydronephrosis, 12 stone burden, 13 location of nephrostomy, exit from a stone-bearing calix, and use of holmium laser to access calix.
Surgeries were performed by a single surgeon and urology residents participated in various portions of the procedures. The primary surgeon was experienced with ureteroscopy, but had no subspecialty training in PCNL. All patients scheduled for PCNL were evaluated for UARN.
This procedure has been previously reported. 8 Eligibility was determined by careful review of a preoperative computerized tomography (CT) scan. We projected a straight line drawn from the renal pelvis through the infundibulum and out of the selected calix. Although a range of puncture angles could be created by deflecting the ureteroscope tip in the calix, we usually maintained a fairly neutral position for the scope tip. Thus, the puncture wire trajectory was primarily set by the line created by the endoscope passing from the renal pelvis through the infundibulum. To evaluate the safety of a potential puncture tract, we visually estimated a probability cone projecting ∼30 degrees in all directions out from the scope tip out of the calix and through the flank. If any adjacent organs fell within this cone, the tract was rejected and an alternate was selected. We separately examined the region surrounding each of the upper, mid, and lower pole for presence of nearby organs.
Our practice had been to place a ureteric stent several days before PCNL to dilate the ureter. This facilitates safe placement of a ureteral access sheath to optimize fluid exchange during the UARN procedure. Preplacement of a stent and use of a ureteral access sheath are not essential to the UARN technique. We currently only preplace a stent if clinically indicated and no longer routinely use a ureteral access sheath during nephrostomy creation.
Patients were placed under general anesthetic and positioned in the Bart's modified supine position. 8,14 After placement of a ureteral safety wire and access sheath, a retrograde pyelogram was performed and the ureteroscope was navigated into the preselected zone of the kidney. For obstructing stones, normal saline pressure irrigation at 150 cm H2O often created space between the stone and urothelium for advancement of our ureteroscope. We applied holmium laser to the stone edge to permit advancement of the endoscope between an impacted stone and adjacent urothelium.
Once in our chosen posterior calix, we repeated a retrograde pyelogram through the endoscope to confirm the position. The deflection of the endoscope tip was then fine-tuned using anterior–posterior plane fluoroscopy to create a trajectory aiming 10 to 30 degrees below horizontal. Lower pole tracts often required directing the endoscope tip through the upper fornix of the lower pole calix to obtain a lateral trajectory (Fig. 1).
Puncture of the upper fornix of lower pole calix to achieve lateral wire tract (flank elevation rotates kidney; imaged is a posterior calix).
While the primary surgeon held the ureteroscope in this precise position, the assistant introduced the 3F puncture wire/sheath ensemble (Ref. 087000; COOK® Medical, Bloomington, IN) through the ureteroscopes' working channel until the wire and its sheath emerged from the endoscope. The assistant opened the pin vise lock and advanced the puncture wire under intermittent spot fluoroscopy through the kidney and flank. The C-arm was moved laterally to follow the wire to the skin. The wire tip usually tented the flank skin (Fig. 2); the skin was incised and the wire was drawn 35 cm out of the flank. Any kinking of the wire tip was removed sharply with heavy scissors or wire cutters.
Puncture wire tenting flank skin. Incision is made to deliver wire.
Only punctures that emerged behind the posterior axillary line qualified for dilation (Fig. 2). We then checked the puncture site in relation to the tip of the 11th or 12th rib and reviewed the CT for corresponding anatomic relationships between the tip of this rib and nearby organs to confirm tract safety. If satisfied, we advanced a 30-cm, 5F transitional dilating coaxial catheter (Ref. 7950; Vascular Solutions, Minneapolis, MN) over the puncture wire at the flank until the tip intubated the ureteral access sheath. After removing the puncture wire and inner catheter, a standard 0.035′′ working wire was advanced through the outer exchange catheter (Fig. 3). 8 The PCNL working wire was thus through and through, from flank to urethra. The ureteral access sheath was then reloaded over our safety wire. Nephrostomy balloon and sheath positioning was performed under direct ureteroscopic vision to reduce fluoroscopy requirements. We did not reposition the patient for tract dilation/PCNL.
View in ureteral access sheath of outer coaxial, ready for antegrade passage of 0.035′′ working wire.
In the event of an unsatisfactory wire angle on fluoroscopy, we removed the puncture wire/sheath ensemble from the ureteroscope, repositioned the wire tip just inside the sheath, relocked the pin vise lock, and then reintroduced the puncture wire ensemble into the ureteroscopes' working channel.
During upper pole access, we often pursued a subcostal puncture to avoid hitting the ribs. Anesthesiology assisted by inspiring the lung just before wire advancement to inferiorly displace the kidney in relation to the ribs.
Stone burden was estimated by multiplying the stones largest dimension on axial CT by the length 90 degree orthogonal to this. Where two stones were present, the two areas were added together. 13 SSD was measured as previously described. 9
Data were analyzed using IBN SPSS 19.0 for Windows (IBM SPSS, Armonk, NY).
Results
Fifty-two patients were evaluated; all were qualified for and underwent our UARN procedure 8 (Table 1).
BMI=body mass index.
Perioperative results are listed in Table 2, and complications are listed in Table 3. The one complication possibly attributable to UARN is a proximal right ureteric stricture ultimately requiring robotic ureteroureterostomy. This may have been due to the movement of the ureteric access sheath during puncture wire deployment, although this is not clear.
CROES=Clinical Research of the Endourological Society; PCNL=percutaneous nephrolithotomy.
DVT=deep vein thrombosis.
The patient who required transfusion began bleeding after we switched from a rigid nephroscope to a flexible nephroscope and torqued this scope to reach a difficult anterior calix. This also inadvertently opened the collecting system. We do not believe this bleed was due to our nephrostomy tract creation. By puncturing through papillae under vision during nephrostomy creation, we avoid injury to the interlobar arteries, which encircle the infundibulae. By dilating the nephrostomy balloon and advancing the nephrostomy sheath under vision, we avoid injury to the narrow infundibulae and the renal pelvis. We believe that these steps together contributed to minimal postoperative hematuria and supported our short length of stay.
In our failed cases (3/52, 6%), mean BMI was 44 kg/m2 (standard deviation [SD] 6.6) and SSD was 15.5 cm (SD 0.22). These values are higher than the BMI and SSD in successful cases—28.6 kg/m2 (SD 6.9) and 10.8 cm (SD 2.45); p<0.001 and p=0.002, respectively. We have a long nephrostomy sheath and nephroscope available in case of long tracts in obese patients. Where the nephrostomy sheath ends at the skin, we find that a Kelly clamp placed on the edge of the sheath provides good control in case the sheath migrates a few millimeters inside the skin. In the first of our failed cases, we had previously established retrograde access using 6 seconds of fluoro time in 45 minutes; the second, failed tract was to help access a staghorn stone. We were able to clear the stone using the first tract with flexible nephroscopy and a subsequent ureteroscopy. The other two failed cases were managed effectively with staged ureteroscopy.
Among our successful cases (49/52, 94%), BMI (p=0.002) and SSD (p=0.009) were both significantly correlated with nephrostomy creation time, however, BMI had a slightly stronger correlation. Stepwise linear regression was performed to determine the correlation of measured variables with nephrostomy creation time (42/49 cases). BMI correlated best (r 2=0.219) followed by stone burden 13 (r 2=0.094) and use of holmium laser to access calix (r 2=0.104), for a total r 2 linear=0.416 (Fig. 4). Notably, none of the CROES score, Guy's score, nephrostomy exit site, or hydronephrosis were correlated with nephrostomy creation time.
Stepwise regression analysis of BMI, stone burden, and use of laser to access calix in relation to nephrostomy creation time. BMI=body mass index.
Exiting from a stone-bearing calix (30/52 cases, 58%) required holmium laser lithotripsy, only one-third of the time (10/30, 33%). In only 2/22 cases where we did not exit from a stone-bearing calix, laser was required (9%), which approached significance—33% versus 9%, p=0.051. Nephrostomy creation time was longer when laser was required (63 minutes, n=12) than when laser was not used (38 minutes, n=40), p<0.01.
Intraoperative ultrasound was employed to confirm a safe tract in two patients (2/52, 4%) for right-sided supra-11th rib accesses.
Discussion
We have found UARN to be an intuitive approach to nephrostomy creation. Our nephrostomy creation fluoroscopy times are the lowest we have seen in the literature, even among series employing reduced radiation protocols for PCNL or even ureteroscopy. 15 –20 This has greatest significance for children undergoing PCNL and for surgical teams performing many PCNLs. Our low fluoroscopy times are understood by considering (1) we select our calix under direct vision; (2) advance of our puncture wire from the calix to the flank requires only minimal fluoroscopy; (3) the 5F coaxial exchange catheter has a smooth tapered tip and penetrates fascia and renal capsule easily; (4) positioning and inflation of the nephrostomy balloon are often performed under ureteroscopic vision; (5) we prestent most patients easing the placement of our ureteric access sheath (typically requires 5 seconds of fluoroscopy); (6) pulse settings for C-arm reduce our fluoroscopy times by an estimated 50%; and (7) avoidance of continuous fluoroscopy.
Although we did not specifically evaluate the learning curve in this article, we estimate it to be under 10 cases based on our experience training residents and fellows. This will be the subject of further study.
Notably, we found that UARN becomes more difficult as the BMI and SSD increase. We attribute this to the mechanical properties of the puncture wire; the puncture wire is flexible enough that it does not significantly reduce deflection of our scope tip. In thin patients (e.g., BMI<30 kg/m2), the puncture wire maintained a straight course through the subcutaneous tissues and emerged at the skin without difficulty. In patients with BMI of ∼35 to 40 kg/m2, the wire sometimes deflected caudad as it passes through the subcutaneous fat. We believe this occurs because of a longer exposed wire length, and subcutaneous adipose tissue is less supportive of the puncture wire. We aim for a horizontal tract to prevent this and any caudal wire deflection was usually mild and insignificant. In approximately three cases where BMI exceeded 35 kg/m2, the caudad deflection was more pronounced. Although this is not dangerous, as the caudal deflection occurs in the subcutaneous space, we were able to correct these as follows: the wire was delivered through the skin and held with a clamp. The wire's subcutaneous path was palpated in a cephalad direction—at the uppermost point where the wire was still palpable, a counterincision was made and the wire was delivered at this site. Any kinked segment of wire was removed with heavy scissors before loading the exchange catheter. In morbidly obese patients (BMI>40 kg/m2), the puncture wire did not have enough pushability to travel the distance to the skin. We no longer perform UARN in patients with BMI above 40 kg/m2.
To avoid kinking the puncture wire, the wire is advanced by the assistant in no more than 1.5 cm increments. The coaxial exchange catheter must also be loaded gently over the puncture wire at the flank to prevent kinking.
The outer diameter of the sheath protecting the puncture wire is 3F. This reduces irrigation flow through the working channel; however, because we did not introduce the device into our ureteroscope until it was in final position for puncture, we did not notice any reduction in visualization.
UARN does require advanced flexible ureteroscopy skills. Challenges include navigating the ureteroscope into a specific calix, occasionally past a large stone, while maintaining good positional orientation in the kidney. We consider our access creation time a worthwhile investment toward simplifying the second phase of our PCNL: UARN provides precise surgeon selected access with through-and-through wire control, both of which enable the subsequent PCNL. 21 Our surgical assistant required only minimal skills to advance the puncture wire under the direction of the surgeon.
The safety of retrograde nephrostomy has been confirmed in over 400 recently published cases of the traditional Lawson retrograde nephrostomy, without injury to adjacent organs. 2,7 Our series adds to these data. This favorable safety profile is understood in light of the posterior rotation of the kidneys in the paravertebral gutters; a line drawn from the renal pelvis into a posterior calix consistently projects a nephrostomy path behind the colon and spleen. This is true even for many anterior calyces when punctured through the posterior fornix; nonetheless, we always preferred puncturing through a posterior calix. Our ureteral stricture may have been due to the open end of the access sheath contacting the ureteral mucosa during puncture wire advancement, however this is not certain. We feel the benefits of using a ureteral access sheath outweigh the risks, especially when the ureter has been pre-dilated with a stent to reduce ureteral trauma associated with sheath insertion. Benefits offered by the ureteral access sheath include optimized visualization from greater fluid exchange, delivery of the nephrostomy wire inserted through the flank out of the urethra (through the access sheath), and easy re-introduction of the flexible ureteroscope for direct-vision nephrostomy balloon position, inflation, and nephrostomy sheath introduction. As such, we still insert a ureteral stent 5–7 days prior to the PCNL for safer and easier access sheath insertion, and are extremely careful not to apply any pressure to the access sheath during our PCNL procedure.
Our stone-free rates are slightly lower than predicted by the CROES nomogram. 11 We believe this reflects (1) our heavy stone burden 11,22 –25 ; (2) use of single access 26,27 ; and (3) our deferral of any secondary procedures to outpatient ureteroscopy. 28 –30 This secondary procedure could just as well have been Extracorporeal Shock Wave Lithotripsy in most cases. Had we performed a second look procedure during our first procedure hospitalization, we would have approached 90% stone-free status at discharge from hospital. Thus, we do not consider our stone-free rates reflective of our nephrostomy technique as much as our case strategy, especially in that we succeeded in obtaining access through the upper, interpolar, or lower pole as desired. In the future, we will consider the use of multiple accesses for complex staghorn stones.
We performed flexible ureteroscopy during PCNL to retrieve fragments in only two cases—this requires further study to assess its impact on stone-free rates. We found that the irrigation through the flexible ureteroscope channel with a basket in place was not sufficient for good visualization. This problem was solved by irrigating through the nephroscope at the flank while performing flexible ureteroscopy.
When gaining access through an upper pole calix, the puncture wire often contacts a rib and does not advance. When this occurs, the wire is drawn back, reset, and then readvanced during inspiration to help deliver the wire subcostally. 7,28 –30 This contact of a rib during upper pole punctures accounts for a nonsignificant trend toward more wire advances during upper pole nephrostomy creation than for lower pole nephrostomy (Table 2). Of our 18 upper pole accesses, 11 (61%) were subcostal. 7,31 –33 Lower pole tract creation may benefit from slight cephalad deflection of the scope tip in the upper fornix to create a more horizontal tract (Fig. 1).
We did not perform multiple accesses due to the time required to create our tracts. Instead, we used a flexible nephroscope to access stones located at an angle from our nephrostomy sheath. Patients with full staghorn stones underwent more than one PCNL and may have been better served with multiple accesses during a single procedure. In two cases, we performed flexible ureteroscopy from below to remove fragments not accessible through our tract.
The true working channel lengths of most flexible ureteroscopes are ∼85 cm. The protective sheath covering the Lawson puncture wire is 85 cm long; this needs to emerge from the endoscope channel to safely advance the puncture wire without damaging the scope. For the Storz Flex X2, we used a blue ACMI seal (ACMI CS-B612) to cap the working channel port. The Olympus URF Type P5 has an external laser guide that creates a working channel length in excess of 85 cm. After positioning this scope into the calix, we removed the external laser guide and advanced the puncture wire through the working channel without any attachments on the port. We recommend preoperative bench testing of each flexible ureteroscope with the puncture wire and nipple attachments.
Although we have not performed a financial analysis of this technique, it does add the cost and maintenance of a flexible ureteroscope and adds to the PCNL operative time. This must be weighed against the common alternate scenario of interventional radiology creating the nephrostomy tract and the costs incurred therefrom.
Conclusions
UARN is an intuitive safe procedure that offers dramatic reductions in fluoroscopy times. UARN is best suited to cases requiring only one nephrostomy tract. Upper pole access is commonly performed with a subcostal technique to navigate the puncture wire below the ribs. Increasing BMI predicts longer nephrostomy creation times; procedure failure was associated with BMI exceeding 40 kg/m2. UARN is a robust technique for nephrostomy creation in appropriately selected patients.
Footnotes
Acknowledgment
Wayne State University IRB, HIC numbers: 020111M1F.
Disclosure Statement
No competing financial interests exist.