Routine Day-Case Laparoscopic Pyeloplasty: A Paradigm Shift?

Introduction

S ince first reported by Schuessler and associates ¹ in 1993, laparoscopic pyeloplasty (LP) has become the treatment of choice for patients with pelviureteral junction obstruction (PUJO) in many centers. In comparison with open surgery, laparoscopic procedures have the advantage of less postoperative pain, perioperative morbidity, and scar formation, shorter duration of hospitalization, and earlier return to routine activities. ² As expertise in laparoscopic technique has improved, many laparoscopic procedures are increasingly performed as day-case surgery (DS). We have developed laparoscopic DS at our center for a number of procedures, including LP, which we first performed in March 2006; indeed, DS is now routine for managing PUJO by LP. The aim of this study is to report our initial experience with DS LP and to assess its feasibility and safety.

Patients and Methods

In this retrospective observational study, we evaluated consecutive patients who were planned for DS LP between March 2006 and January 2010 in a single urologic center. The selection criteria for DS were similar to those needed for other DS procedures. Initially, young, fit, and well-motivated patients were offered the option of DS. All of them had to meet the social criteria for DS: After the procedure, there should be a responsible adult/carer/parent able and willing to care for the patient for at least 24 hours postsurgery, they must have access to a private telephone, escorts must be available to drive or accompany them home in a taxi, the journey home should take no longer than 1 hour. When the feasibility and safety was becoming apparent, social criteria as well as medical criteria were used: Acceptable comorbidities and any coexisting medical condition had to be stable and optimally treated.

The diagnosis of PUJO was confirmed by diuretic renography (furosemide technetium–99m mercaptoacetyltriglycine renal scan) in all cases. We collected data regarding demographic information, American Society of Anesthesiologists (ASA) status, medical comorbidities, preoperative and postoperative symptoms, admission as well as discharge time and date, operating room time, renographic data on diuretic drainage characteristics, and split renal function before and after surgery. The success rate of DS and reasons for unplanned overnight admission and readmission were collected and evaluated.

Every patient who presented with PUJO was generally considered for and informed about the option of DS. All patients planned as DS followed a standard pathway of care. Invitations for admission were sent with instructions to fast from 12 midnight for solids and 6:30 am for fluids. A confirmation phone call was made the day before the procedure. Patients were routinely admitted at approximately 7:30 am to a dedicated DS ward. All patients received 1600 mg of slow-release ibuprofen by mouth an hour before surgery. The procedure was performed during the morning session, usually as the first case, with the patient returning to the ward at about 1:30 pm. If considered necessary, retrograde ureteropyelography was performed before the procedure to confirm the diagnosis.

The LP was performed using a standard transperitoneal approach with four ports on the left side and five ports on the right side. The PUJ was dissected and dismembered, with transposition of the crossing lower vessel when necessary. A Double-J ureteral stent was used in every case. A tube drain was inserted in the first few patients and was later restricted to only those patients for whom the surgeon thought drainage was needed. Urethral catheterization was performed at the beginning of the procedure, and the catheter was removed before recovery from anesthesia. After the initial 10 cases, the urethral catheter was not inserted at all. Intraoperative opioids were avoided, and patients received 1 g intravenous paracetamol toward the end of the procedure. Multimodal antiemesis was provided with dexamethasone and odansetron.

Levobupivacaine 0.5% was instilled around the port sites before trocar placement and at the conclusion of the procedure to minimize postoperative pain.

Postoperatively, patients were monitored in the recovery ward and then in the DS ward. Hourly parameters of pulse rate, blood pressure, temperature, conscious level, pain, drain and urine output were recorded. Oral fluids and diet were introduced postoperatively as tolerated, and the tube drain, when inserted, was removed about 3 hours after surgery if the drainage was less then 30 mL. All patients were reviewed by a senior staff member, between 6 and 8 pm, before being discharged.

Patients were given verbal and written information regarding the procedure, including important signs and symptoms that might be of concern: Increased pain, nausea, ineffectiveness of analgesia, increase in temperature and pulse, dizziness, lowering of blood pressure, and increasing abdominal distension. They were provided with direct access to the urology ward and to the surgical admissions unit, with a telephone number for advice, and both wards were informed about the patient. Unless contraindicated, the patients were prescribed a 5-day course of slow release ibuprofen and co-codamol to be taken regularly. The district nurse visited the patient on the evening of surgery, on day 1 and 2 postoperatively, and thereafter at the nurse's discretion to monitor temperature, pulse, blood pressure, and wounds. The ureteral stent was removed approximately 4 weeks after surgery, and diuretic renography was obtained between 3 and 6 months postoperatively to document upper tract drainage.

Results

Thirty-two (63%) procedures were planned as DS of the 50 LP performed during the study period. Figure 1 illustrates whether consecutive patients were selected for DS or inpatient.

OPEN IN VIEWER

FIG. 1.

Allocation and outcome of patients selected for day-case or inpatient surgery. DS=day-case surgeries.

Of the DS LP group, there were 20 (62.5%) females and 12 (37.5%) males with a median age of 37 years (range 11–69 y). The PUJO was on the left side in 19 (59.3%) patients and on the right side in 13 (40.6%) patients. The majority of patients were classified according to their physical status as ASA 1 (59.37%), ASA 2 (37.5%), and only one patient as ASA 3. Six patients had a history of diabetes and either cardiac or respiratory disease and eight had high blood pressure; all medical conditions were assessed and stabilized before surgery. Surgical time was available from the notes in only 19 patients, mainly from the beginning of the series, and it varied from 90 to 210 minutes (mean 148.9 min, standard deviation 34.70).

Estimated blood loss was minimal (<100 mL) in 30 (93.75%) patients. Of the two patients with more than 100 mL blood loss, one had had a previous percutaneous nephrolithotomy (PCNL) 9 weeks before LP, and subsequent scarring led to a long procedure (180 min) with an estimated blood loss of 150 mL. The other patient classified as ASA 2, with an estimated blood loss of 150 mL, had a tube drain inserted during the procedure that was removed at 6 pm, before successful discharge. She presented to the surgical assessment unit (SAU) with pain 2 days later and was discharged immediately (see below) (Table 1).

Seven (21.87%) patients failed to be discharged on the same day. The reasons for failed discharge were comorbidities (ASA 3) for one patient, difficult procedure for three patients, uncontrolled pain for two, and social reasons for one. The complexity of the procedure was increased by a previous PCNL on the same side 9 weeks earlier in one patient, by the concomitance of cholecystectomy with LP in the second one, and in the third one, by pyelonephritis in a horseshoe kidney that necessitated a nephrostomy to be inserted 2 months before the procedure and kept in place until surgery. Those three patients were admitted overnight, because it was considered safer and not for any other reason. The two patients who failed discharge because of pain control were young (18 and 20 years old, respectively), ASA 1. One of the seven patients was medically fit for discharge, but she was admitted overnight because her domestic circumstances were not suitable for her to go home. Five of those seven patients were discharged on the first postoperative day and two of them on the second day.

Two (6.25%) patients were readmitted after surgery (Fig. 1). One of the patients was a 69-year-old lady who came to the SAU 2 days after the LP complaining of pain and nausea. Her blood tests were normal, and she was immediately discharged with symptomatic treatment. The other patient was a 33-year-old male with anxiety-depression who was admitted 19 days after LP with pain. Ultrasonography showed no collection, and he was discharged 3 days later.

Symptomatically, all but four patients were improved after the procedure; these four patients had much improved pain. There was no correlation between the procedure's functional results and either readmission or failed discharge. Renography results documenting preoperative and postoperative split renal function and isotope half-life (T_1/2) were available for all but six patients; one patient did not attend the follow-up renography, and three are yet to have their follow-up scans. Two further patients have been followed up in other hospitals. Of those 26 cases with complete data, 19 (73.07%) patients had improved or stable split renal function, with a mean improvement from 40.96% to 43.61%. Twenty-five (96.15%) patients had improved drainage and only one, a 64-year-old woman who underwent concomitant DS LP and cholecystectomy, showed an obstructed renogram pattern after surgery.

Of the four patients with residual pain after LP, three had improved renographic split function and improved renal drainage; T_1/2 decreased from 30 to 20 minutes for the first patient, from 30 to 9 minutes for the second patient, and from 30 to 8 minutes for the third one. The fourth symptomatic patient was a 38-year-old woman with a small, grossly hydronephrotic right kidney. The split function decreased from 40% to 16% in spite of the improved renal drainage, with the T_1/2 decreasing from 30 to 9 minutes. Arterial embolization was performed to treat ongoing pain but was ineffective, leading to a laparoscopic nephrectomy 14 months after the LP.

Discussion

PUJO has been managed for years with open pyeloplasty with good and durable results. ³ Alternatives, such as endopyelotomy, have been developed, but LP looks to be the natural successor of open pyeloplasty as the gold standard for the management of PUJO. ⁴ LP has been consistently shown to offer the success rate of open pyeloplasty with the advantages of a minimally invasive technique. ^{4 –7}

DS LP appears to be a logical development. Other laparoscopic interventions have made the step from inpatient to DS: These include cholecystectomy, hernia repair, appendectomy, Nissen fundoplication, and adrenalectomy. ^{8 –12} A Cochrane review concluded that DS laparoscopic cholecystectomy seems to be safe and effective. ¹³ DS allows the patient to recuperate at home, reduces hospitalization costs, and releases inpatient beds for emergencies or other procedures. ¹⁴ But before considering health economy advantages, we first need to consider patient safety, which is of paramount importance.

As regards DS LP, this is the second published article that addresses the subject. The first was published by Rubinstein and coworkers ¹⁵ in 2005. They report a series of six patients who successfully underwent dismembered pyeloplasty using the retroperitoneal or transperitoneal (one patient) approach. The authors concluded that in their well-selected patient population, the procedure was feasible and safe.

The senior author performed our first DS LP in March 2006. With increasing experience, the proportion of patients planned as DS increased from 56% in the first half of the present series to 72% in the second half. The fact that patients with more complex conditions have been planned as DS could explain the increased number of patients who failed to be discharged on the same day observed in the last half of the present series (Fig. 1).

Modifications of the standard technique are thought to be important for successful DS LP. Such modifications include avoiding retrograde ureterography and using antegrade rather than retrograde ureteral stent placement, thus avoiding lower urinary tract instrumentation. We believe that these measures, along with avoidance of a postoperative urethral catheter, is important for promoting spontaneous voiding postoperatively. We also avoid a tube drain, which is a considerable source of pain. This lack of postoperative drainage may be criticized, but we do not believe that harm has been caused by this measure. Indeed, this change seems to follow the trend reported by articles that describe the tubeless PCNL and by those that state that Double-J stents are not necessary after every ureteroscopy. ^16,17 Randomized studies are necessary to provide us with evidenced-based data regarding this aspect.

Seven (21.87%) patients (five women and two men) failed to be discharged on the same day, giving a success rate of 78.12%. This is comparable with the results reported for recent series of laparoscopic cholecystectomy, which are between 86% and 95%. Early experience of DS laparoscopic cholecystectomy had a success rate of 56%. ¹⁸ The readmission rate, 6.25%, is similar to the values reported for laparoscopic cholecystectomy, which ranges between 1.5% and 8%. ^{14,19 –21} Both readmitted patients presented with pain, and they actually needed reassurance.

For DS laparoscopic cholecystectomy, there are criteria that can predict an increased risk for failing the DS procedure: Age more than 65 years, ASA>2, and comorbidity. ^14,22 Analyzing our series, we noticed that one of the two readmitted patients was more than 65 years old (in total, there were five patients more than 65 years old), and the only patient classified as ASA 3 from our series failed to be discharged on the same day.

We acknowledge that there are a number of limitations to the present article, caused in part by the study design (retrospective and nonrandomized). Another source of bias is that the surgical technique has evolved during the study period, and the cases are not from a single surgeon. Regarding our functional results, it appears that they are comparable with other studies addressing LP, but it is beyond the scope of this article to comment further.

Conclusions

The DS LP is feasible and safe. To improve the success rate and to decrease the readmission rate, objective preoperative, intraoperative, and discharge criteria should be developed for DS and validated in randomized studies.