Pyeloplasty for hydronephrosis: Issues of double J stent versus nephrostomy tube as drainage technique.

AIMS: To compare the efficacy, complications, cost analysis and hospital stay between two methods of drainage of the kidney: double J (DJ) stent versus nephrostomy tube following open pyeloplasty for ureteropelvic junction obstruction hydronephrosis. PATIENTS AND METHODS: This was a prospective randomized study of 20 patients in each group over 14 months. Pre and post-operative (3 months) function and drainage were assessed by ethylenedicysteine scan and intravenous urogram.
RESULTS: Both groups showed similar good improvement in function and drainage. Nephrostomy group had significantly longer hospital stay (P < 0.001) but incurred less cost. Complications with nephrostomy included tube breakage (n = 1) and urine leak after tube removal (n = 2). DJ stents were associated with stent migration (n = 4), increased frequency of micturition (n = 9), dysuria (n = 4) and urinary tract infection (n = 1).
CONCLUSION: Both methods of drainage did not interfere with improvement after pyeloplasty. Minor complications were more with DJ stent (P = 0.0003). Although overall cost of treatment was more with stents, they reduced length of hospital stay. Optimal length of stent is essential to reduce complications secondary to migration and bladder irritation.

RCT Entities:   Population Interventions Outcomes

INTRODUCTION

The Anderson-Hynes dismembered pyeloplasty is the most commonly performed procedure for pediatric ureteropelvic junction obstruction (UPJO) since its first description.[1] However, the type of drainage method is still debated and has only occasionally been compared with each other.[234] We conducted this study to compare the efficacy and specific complications of nephrostomy tube versus double J (DJ) stent as a method of urinary diversion in open pyeloplasty for UPJO hydronephrosis. In addition, we analyzed the cost and length of hospital stay.

PATIENTS AND METHODS

This was a prospective study of 40 patients with UPJO from July 2012 to November 2013. Patients underwent open Anderson-Hynes pyeloplasty with either DJ stent or a nephrostomy tube for drainage of urine. Patients were allocated by manual randomization to either group. Approval was obtained from the Institute Ethics Committee before start of the study.

Children aged up to 12 yrs, of either sex, having obstructed drainage pattern on renal dynamic scan (Technetium-99m-L, L-ethylenedicysteine (EC Scan)) and intravenous urogram (IVU) with unilateral or bilateral obstruction were included. Patients who had previously undergone percutaneous nephrostomy (PCN) or DJ stenting were excluded. Patients were admitted a day prior to surgery and informed consent taken including discussion about both methods of diversion. Rough estimate of the required length of DJ stent was determined by measuring the distance from the tip of the 11th rib to ipsilateral pubic tubercle in centimeters.

Modified Anderson-Hynes pyeloplasty was performed through an anterolateral extra peritoneal flank incision. The length of the retained ureter was measured using the 5F (or larger) feeding tube till it hit the ureterovesical junction (UVJ) and DJ stent inserted. Its position was confirmed by the flow of urine through the stent holes. In the nephrostomy group, a Malecot tube (size 8-12 F) was inserted through the lower pole of the kidney before closure of pelvis. No perinephric drain was placed in any patient.

Postoperative antibiotics and analgesia were provided as required. A bedside abdominal radiograph was performed on the evening of the surgery in the DJ stent group to confirm its position. The patients were discharged on the 2nd or 3rd post operative day (POD) depending on adequacy of oral intake. Urine specimen voided per urethra was taken for cultures on 7th POD on outpatient basis. The stent was removed 2 weeks later under short general anesthesia as a day care procedure.

In patients with a nephrostomy diversion, the tube was elevated slowly from the 2nd-3rd POD in the absence of hematuria. Urine specimen for culture was sent on 7th POD from nephrostomy tube as well as per urethra. The nephrostogram was usually done on the 7th POD in the Radiology department under fluoroscopy. If there was free drainage of contrast, the tube was clamped. If there was slow drainage, nephrostogram was repeated again after a week. The patients were discharged home the next day after the tube removal. In case of persistent leak of urine, a trial of catheterization was first tried, failing which a DJ stent was inserted.

Patients were followed up in the OPD a week after the discharge. Uroprophylaxis was stopped 1 week after nephrostomy or stent removal. After 3 months of removal of nephrostomy/DJ stent, EC scan and IVU were repeated. The statistical analysis was carried out using Statistical Package for Social Sciences (SPSS Inc., Chicago, IL, version 15.0 for Windows).

RESULTS

The patients’ age ranged from 2 months to 12 years with a mean of 2.7 years±3.47 in DJ stent group and 3.76 years±3.21 SD in the nephrostomy group. Both the groups had similar presentation and normal routine hematological and biochemical investigations. Pre-operative urine cultures were positive in 7 patients (Escherichia coli 6, Klebsiella 1). The renal pelvis anteroposterior diameter on USG was 1.0-5.5 cm (mean 2.820 ± 1.261 cm) and 1.1-6.2 cm (mean 2.495 ± 1.115 cm) in DJ stent and nephrostomy group, respectively. UPJO was present on the left side in 24 (9 vs. 15), right 13 (10 vs. 3) and bilateral in 3 (1 vs. 2) patients in DJ stent and nephrostomy group, respectively. There was no statistically significant difference between the two groups.

The hospital stay in the DJ stent group was 03–25 days (mean: 5.15 ± 4.749) compared to 10–16 days (mean: 11.95 ± 1.395) in the nephrostomy group which was statistically significant (P value = 0.001). However, the latter spent Rs. 1310 less on an average [Table 1]. All patients had improved drainage on EC scan and IVU as well as improved function on IVU.

Table 1

Average expenditure incurred by patient (all admissions)

Complications of DJ stent [Figure 1] and nephrostomy are shown in Table 2. There was spontaneous extrusion of entire DJ stent per urethra at the end of the surgery in one. One patient in the DJ stent group required readmission a week after discharge for ipsilateral flank swelling due to perinephric collection and proximal stent migration. Pigtail catheter insertion successfully resolved the issue. The flower of the Malecot catheter broke during removal on one side following bilateral pyeloplasty [Figure 2]. This infant had initially presented in emergency with septic shock secondary to pyonephrosis. Open pyelotomy was required for removal.

Figure 1

Plain radiograph of abdomen showing bilateral distal migration of stents

Table 2

Comparison of complications: DJ stent versus nephrostomy tube

Figure 2

Plain radiograph of abdomen showing retained Malecot flower (left kidney)

DISCUSSION

The Anderson-Hynes dismembered pyeloplasty, the most commonly used type of repair for UPJO hydronephrosis was first described as a stentless procedure.[1] Over the years, drainage techniques were added as perianastomotic leakage of urine and infection were thought to be the cause of stenosis or stricture formation requiring re-operation. Drainage tubes may be external, e.g. nephrostomy tube; completely internal, e.g. DJ stent; or partly external and partly internal, e.g. Salle stent.[456]

Problems are associated with all types of drainage procedures. This can occur during insertion, especially in small children.[47] We were unable to negotiate the UVJ while inserting the DJ stent in a 2-month-old patient for whom a nephrostomy tube was inserted instead. In another 1-year-old, the small intrarenal pelvis could not accommodate the flower of the nephrostomy tube. A DJ stent was inserted instead. McMullin et al. advised retrograde cystoscopic insertion of DJ stent in neonates due to the narrow ureteric size.[3] A gentle traction of the ureter from above to straighten the UVJ has been found to be helpful.[7] We would advocate nephrostomy drainage or not to use any form of drainage in such cases to avoid injury to the UVJ in small infants.

DJ stents can cause mechanical irritation of the bladder trigone. McMullin noticed urinary urgency in 11.1% patients.[3] Braga et al. noted bladder spasm symptoms in 2.9% patients requiring early stent removal in some because of severity of symptoms.[4] In a study by Ozdemir et al., all 71 patients had intermittent mild hematuria and mild pelvic discomfort until removal of stent.[8] In our study, 85% patients in the DJ stent group had similar complications compared to 25% in the nephrostomy group (P = 0.0003). Increased urine frequency with associated extra length of stent in the bladder was noted in nine patients (45%), normalizing in all after stent removal. Incontinence can be associated with urgency, but can also occur with stent migration into the urethra beyond the external sphincter.[9]

DJ stents have been known to migrate in 2.5% to 16.6% cases.[3410] Stents with full coils are less prone to migrate than those with a J-shape which can occur due to inadequate length. We encountered proximal migration in one and distal migration in three patients. Braga et al. noted urinoma formation in 1.2% in the DJ stent group and 0.4% with pyeloureteral stents.[4] In our study, urinoma formation occurred in one patient due to proximal migration of stent requiring pigtail insertion and prolonged hospital admission. On the other hand, in the nephrostomy group, the only serious complication was tube breakage during removal in one patient. This may have been due to manufacturing defect or pre-existing inflammation causing adhesion of Malecot catheter flanges.

Choosing a correct stent length can therefore avoid several post-operative problems. Mathematic formulae have been proposed.[1112] Pre-operatively, we measure the distance in cm from the tip of the 11th rib to the ipsilateral pubic tubercle for an approximate length of stent. Intra-operatively, the ureter length was measured by passing a 5F feeding tube distally. We have to keep in mind that the stents (in addition to the labeled length) also have 5 cm length tails on each side as a coil. To avoid excess length of stent in the bladder, we now propose the formula:

Length of DJ stent (cm) = length of retained ureter (cm) – 2.

In children with DJ stent, mild to moderate flank pain could be due to reflux which in turn can lead to pyelonephritis.[13] Encrustation or infection of stent can lead to suprapubic pain.[14] We noted persistent fever with positive urine culture in one patient (5%) in the DJ stent group. Bacterial colonization of urine may occur after long-term presence of DJ stent.[15] In one series, readmission due to pyelonephritis was seen in 2.1% patients in the DJ stent group compared to 0.4% in the pyeloureteral stent group.[4]

All patients showed improvement in function and drainage on IVU in our study. This is in contrast to another study where there was better functional improvement in the DJ stent compared to the nephrostomy group.[2] We feel that functional improvement is more dependent on the operating technique and the pre-operative status of the kidney rather than the type of stent used. Although not true in our series, redo pyeloplasties have been performed by others equally with both types of drainage.[34]

In our study, the mean hospital stay was significantly lesser in the DJ stent compared to the nephrostomy group as seen in other studies as well.[23] Patient attendants were not keen on discharge with an external tube in situ due to cost of travel and the worry of dislodgement. Occasionally, the nephrostogram could not be done on the requested day, thereby increasing hospital stay. In two separate studies, patients with nephrostomy tube and a Salle stent could be discharged early as these tubes were removed later on in the outpatients leading to further reduction of cost.[45] Overall, mean cost of procedure in our study was higher in the DJ stent group due to its higher cost per se and removal under anesthesia.

CONCLUSION

Both DJ stent and nephrostomy tube were equally effective for drainage and did not interfere with improvement after pyeloplasty. Although nephrostomy group had statistically significant longer hospital stay, it was more cost effective. Patients preferred to have the tube removed before discharge. The post-operative nephrostogram confirmed the success of the pyeloplasty and allayed patient fears. However, it can be associated with peritubal leak, blockage, urine leak after removal for a few days and rarely breakage during removal. The DJ stent group on the other hand had a high incidence of minor complications which increased with inappropriate length of stent. They required second anesthesia for stent removal and overall spent more. Our current practice now is to use DJ stent using the formula suggested reserving nephrostomy to patients where the UVJ cannot be negotiated or when appropriate stent length is not available. Optimal length of stent is essential to reduce complications secondary to migration and bladder irritation.