Introduction

The horseshoe kidney is seen in 1 of every 400 births and is the most common congenital renal fusion anomaly. The incidence is twice as high in males as it is in females (1). In this condition, stone formation is seen in about 21–60% of patients. Renal stones are the most common reason that patients require kidney surgery (2). Abdominal and flank pain (83%) are the most frequently reported symptoms in those who suffer from kidney stones (3). Percutaneous nephrolithotomy (PNL) is the most widely used method in horseshoe kidney patients who have failed other therapy and is the treatment of choice for stones larger than 20 mm (3,4,5). The success and complication rates of PNL in horeshoe kidney patients are within an acceptable range of those for PNL undertaken in kidneys with a normal anatomy (1, 3, 6, 7). In this study, we share our first experience with PNL in horseshoe kidneys.

Materials and Methods

Seven patients with horseshoe kidney with calculi were treated with PNL. Their presenting symptoms were hematuria along with recurrent urinary tract infections (14%), atypical abdominal pain (28%), and loin pain (58%). All patients were evaluated with a detailed medical history and physical examination, complete blood count, serum biochemistry, urinalysis, urine culture, and coagulation tests. Each patient underwent a comprehensive preoperative radiologic assessment, including plain abdominal radiography, urinary tract ultrasonography, and non-contrast computed tomography (CT). The stone sizes (mm2) were calculated with dimensions obtained by geometric projections from millimetric graph paper placed directly on the urinary system radiography of the stones.

Surgical Technique

An open-ended 5-French (5-Fr) ureteral catheter (Marflowtm, Marflow AG, Switzerland) was placed in the ureter and was guided by cystoscopy with the patient in the lithotomy position under general anesthesia. After placement, the patient was moved to a prone position. The anatomy of the pelvicalyceal system was visualized with opaque material that was instilled using a uretheral catheter under C-arm fluoroscopy. A 19.5-gauge percutaneous needle (Percutaneous Access Needle, Boston Scientific Corporation, MA, USA) was introduced into the renal collecting system. Fluoroscopy was used to place a guidewire (Zebratm Nitinol Guidewire, Boston Scientific Corporation, MA, USA) in the collecting system for access. The tract was dilated up to 30-Fr with semi-rigid amplatz dilators (Boston Scientific Microvasive Amplatz TractmasterTM, Boston Scientific Corporation, MA, USA), and an Amplatz sheath was inserted into the collecting system. Stone disintegration was performed using a pneumatic lithotripter (CalculithTM Lithotripter, PCK, Turkey) through a 26-Fr rigid nephroscope (Karl StorzTM Endoscopy-America Inc.). If necessary, additional input was made, and the process was terminated by inserting a nephrostomy tube (open-ended hole 21-Fr soft drain) into the renal collecting system. Complications were graded according to the classification system used by Clavien (8). Postoperative residual stone fragments were evaluated at the end of the first month with non-contrast spiral abdominal computed tomography (CT) in asymptomatic patients.

Results

The average age of patients was 46 years (range: 32-59 years), and the female-to-male ratio was 1: 6. The stones were located on the left side in five patients and on the right side in two patients. Stones were localized simply in the pelvis in four patients, in the renal pelvis and upper calyx (partial-staghorn) in one patient, and in multiple locations (in the pelvis and lower calyx) in two patients. None of the patients had a history of previous ipsilateral open stone renal surgery or prior unsuccessful extracorporeal shock wave lithotripsy (ESWL) sessions. Only one patient had a history of ipsilateral PNL six years previously. The mean operative time was 131 (range: 70-215) minutes including the preparatory phase (assuming the lithotomy position, intubation, cystoscopy-guided ureteral catheter placement, transfer to the prone position, sterile painting, and surgical equipment preparation after draping the patient). The average stone surface area was 1234 (range: 250-2460) mm2. Patients’ peroperative and postoperative findings along with the clinical and demographic measures are listed in Table 1. Blood transfusion (Clavien grade 2) was required in 1 (14%) patient (with partial-staghorn calculi (2460 mm2) and a dual access catheter) postoperatively. Bleeding during the post-operative period was controlled by clamping both nephrostomy tubes. Prolonged urinary leakage from the nephrostomy tract (>48 hours) and severe colic pain (Clavien grade 3) was observed in one patient (14%). The ureteral stone was extracted with a semirigid ureterorenoscopic intervention, and double J ureteral stent were placed. After the patient remained stone-free and the prolonged urinary leakage had been treated, the double J ureteral stent was removed three weeks later. None of the patients developed thoracic complications, neighboring organ injuries, or postoperative fever.

In all patients, the ureteral catheter was withdrawn immediately after surgery and is taken urethral foley catheter on the first postoperative day. The mean amount of time that nephrostomy tubes remained in place was 3.14 (range: 2-5) days. The mean hospitalization time was 4.14 days (range: 2-8). All patients who were asymptomatic postoperatively were evaluated with noncontrast CT during the first month after surgery, but one symptomatic patient was assessed on postoperative day 4. Any stones smaller than 4 mm were considered clinically insignificant residual stones upon noncontrast CT.

Our success rate was 5/7 (71%) after one session of PNL. These five patients exhibited no clinically significant or insignificant residual stones upon abdominal CT. Residual stones were detected in only two patients. One of these patients became completely stone-free when treated with ureteroscopy (URS). Thus, our success rate increased to 6/7 (86%). The remaining patient who had a residual kidney stone 12 mm in diameter received SWL. The patient also not have success with SWL was periodically controlled. The patient had no complaints or adverse clinical or laboratory findings at the one-year follow-up visit.

Discussion

The treatment methods for calculi in horseshoe kidneys ranges from minimally invasive surgery to open surgery. Although PNL is the most commonly used technique, SWL, URS, and laparoscopy are other minimally invasive treatment methods employed for calculi in horsehoe kidneys (5). Since 1980, PNL has become the standard treatment for kidney stones after other treatment methods, such as SWL, have failed, particularly for large (>2 cm) and complex stones in horseshoe kidney patients (3,4,5). The stone-free rate (SFR) of SWL for the treatment of small stones (<2 cm) in the horseshoe kidney as reported in the literature ranges from 50-79% (9,10). Yohannes et al. found that SWL produced the best results in the treatment of small renal calculi (<2 cm) in the horseshoe kidney, patients who have failed ESWL therapy and is the treatment of choice for stones larger than 20 mm (2).

The residual stone rate is higher in ESWL and ureterorenoscopy than it is for PNL (2). The ESWL procedure is frequently unsuccessful for large stones in the horseshoe kidney. Furthermore, there is a need for more shock waves due to the abnormal renal anatomy and location in these patients. Therefore, PNL has become the ideal treatment technique to produce successful results in PNL (4). Flexible ureterorenoscopy in patients with horseshoe kidney is a good alternative treatment method when problems arise with ESWL or PNL (11). In several recent studies with a small number of patients, those treated with RIRS (Retrograde Intrarenal Surgery) stones in the horseshoe kidney have reported success rates ranging from 85-90% in selected cases (single lithiasis or lithiasis in only two locations and of a size <30 mm), and RIRS has been proposed as an alternative to PNL (12,13).

Complication rates were reported in the literature to range from 0-16.7% in PNL in patients with horseshoe kidney (5,14,15). These patients did not specifically carry an increased risk of bleeding during PNL because of the anteriomedial localization of their vascular structures and the kidney orientation to the posterior of the calyx (5). In the literature, blood transfusions were necessary for 0–13.3% of the patients (5,7). In the present study, only one patient (14%) required a blood transfusion (Clavien grade 2). In our study, prolonged urinary leakage from the nephrostomy tract (Clavien grade 3a) was observed in only one patient, and both complications 2/7 (28%) were associated with a relatively small number of patients.

In patients with normal renal anatomy, upper pole percutaneous access is frequently gained above the 12th rib, which may cause intrathoracic complications. Upper pole percutaneous access is often essential in horseshoe kidney patients and is relatively safe due to the inferior displacement of the kidneys away from the pleura. Raj et al. reported that in patients with a horseshoe kidney, the pneumothorax rate when establishing upper pole access was 6% (1). Supracostal percutaneous access to a horseshoe kidney is less frequenly applied compared to a normal kidney. In our study, supracostal access occurred in 28% of participants. In the literature, this rate varied from 22-42% (3,5,16).

The risk of retrorenal colon is higher in these patients due to the more anterior localization of the horseshoe kidney. Some authors have stated that CT is the recommended imaging modality before PNL because it can aid in planning the proper access route to avoid complications in horseshoe kidney patients (16,17). If available, three-dimensional abdominal CT can provide even more information (17). All patients in our study underwent non-contrast abdominal CT both preoperatively and during the first month after surgery. We did not observe retrorenal colon in any of our patients, but because the posterolateral column is more pronounced in patients with horseshoe kidney, we established percutaneous access from a truly medial position as a precaution. In addition, the pulmonary injury rates in the intercostal percutaneous area are lower in horseshoe kidney patients due to the more caudal placement of their organs compared to a normal kidney (1,16). In our study, none of the patients developed thoracic complications.

Gross et al. established percutaneous access under direct visualization with laparoscopic guidance and reported that this method was effective for avoiding injury of the vessels, bowel, and surrounding tissue (18). The upper pole access tract was recommended, as it allows the physician to reach all calices, the pelvis, and the proximal ureter. Our study created access in the lower calyx, but access can also be established via the upper and middle calyx, which can be used to access the lower calyx. In patients with a difficult-to-reach lower calyx, a sheath was advanced toward the kidney, and the stones were removed using light to moderate manipulation of the nephroscope. Therefore, stones unreachable using a standard nephroscope may require a longer nephroscope (sheath length: 23 cm) or a flexible nephroscope to ensure complete access to the entire length. In our study, two patients who were obese (BMI: 33.6 and 35.7) had a stone in the renal pelvis that could not be reached because the sheath length was too short. In these patients, the sheath was pushed about 2 cm from the skin toward the inside of the intracorporeal and the stones were successfully removed.
Urinary drainage in horseshoe kidney patients is problematic due to the anatomical structure of the lower pole of the kidney. Thus, spontaneous excretion of stones is less likely than it is in normal kidneys (11,19). In a study by Demirkesen et al., 68 renal units in 52 patients who had congenital upper urinary tract abnormalities were treated with ESWL (20). In 25 renal units with clinically insignificant residual fragments after ESWL, recurrence, regrowth, and retreatment rates in the abnormal renal unit were 50%, 37%, and 34%, respectively.

The term “clinically insignificant residual fragments” is controversial because these fragments have the potential to cause ureteral obstruction and are important risk factors for stone growth and recurrence (3,21). In the congenitally anomalous kidney, an outcome with “clinically insignificant fragments” should not be considered a success. Thus, complete removal of the stones from the kidney is generally required to call the procedure a success (3). In addition, stone analysis is recommended to investigate the etiology and to identify the best medical treatment available for inpatients with horseshoe kidney as well as to prevent stone recurrence (22).

In the literature, ultimately stone-free rates following PNL in the horseshoe kidney were reported to be within the range of %72-91% (5,6,14). In our present series, the final stone-free rate was 6/7 (86%). The success rate of our present series was comparable to those of previous studies. Tepeler et al. investigated the success and complication rates of PNL in horseshoe kidney patients and reported the incidence to be 91% and 17%, respectively. Complications are less affected by abnormal anatomy and demographic and operative parameters, while stone complexity and multiplicity have been reported to play an important role in achieving stone-free status (5). Ozden et al. conducted a comprehensive study and concluded that despite the presence of anatomic abnormalities, PNL is a safe and effective treatment modality for calculi both in horseshoe and normal kidneys with similar success and complication rates (3). The main limitations of our study were its retrospective design and small sample size. However, our findings for PNL in horseshoe kidneys confirmed the results of previous studies.

Conclusions

PNL is difficult to carry out in the renal unit in patients with horseshoe kidneys. The success and complication rates of PNL are similiar both in patients with horseshoe kidneys and normal kidneys. Therefore, consistent with the literatüre, PNL is a method that can be safely applied, and its success has been demonstrated in our study in a limited number of patients.

Conflict of interest

There are no conflicts of interest.