A single dose of a non-steroidal anti-inflammatory drug (NSAID) prevents severe pain after ureteric stent removal: a prospective, randomised, double-blind, placebo-controlled trial

Authors


Correspondence: Michael J. Conlin, Oregon Health & Science University, Division of Urology and Renal Transplantation, CH 10U, 3303 Southwest Bond Avenue, Portland, OR 97239, USA.

e-mail: conlinm@ohsu.edu

Abstract

What's known on the subject? and What does the study add?

  • The problem of severe pain after ureteric stent removal, its incidence and prevention, has not been previously described.
  • The present paper is the first to determine the incidence of severe pain after ureteric stent removal. It also shows that a single dose of NSAID before ureteric stent removal can prevent this pain in a significant number of patients.

Objectives

  • To determine the incidence of severe pain after ureteric stent removal.
  • To evaluate the efficacy of a single dose of a non-steroidal anti-inflammatory drug (NSAID) in preventing this complication.

Patients and Methods

  • A prospective, randomised, double-blind, placebo-controlled trial was performed at our institution.
  • Adults with an indwelling ureteric stent after ureteroscopy were randomised to receive either a single dose of placebo or an NSAID (rofecoxib 50 mg) before ureteric stent removal.
  • Pain was measured using a visual analogue scale (VAS) just before and 24 h after stent removal
  • Pain medication use after ureteric stent removal was measured using morphine equivalents.

Results

  • In all, 22 patients were enrolled and randomised into the study before ending the study after interim analysis showed significant decrease in pain level in the NSAID group.
  • The most common indication for ureteroscopy was urolithiasis (14 patients).
  • The proportion of patients with severe pain (VAS score of ≥7) during the 24 h after ureteric stent removal was six of 11 (55%) in the placebo group and it was zero of 10 in the NSAID group (P < 0.01).
  • There were no complications related to the use of rofecoxib.

Conclusions

  • We found a 55% incidence of severe pain after ureteric stent removal.
  • A single dose of a NSAID before stent removal prevents severe pain after ureteric stent removal.
Abbreviations
COX

cyclooxygenase

VAS

visual analogue scale

Introduction

Indwelling ureteric stents are commonly used for many urological procedures. They are a primary cause of morbidity with >80% of patients experiencing stent-related pain that affects their daily activities [1]. We have observed that ureteric stent removal can be followed by transient, severe, renal colic-like pain. This may require additional interventions, e.g. re-admission, i.v. narcotics, and ureteric stent replacement. The incidence and cause of severe pain after ureteric stent removal are unknown. Recent studies have reported methods for improving the discomfort from indwelling stents, but there are no published reports focusing on the prevention and treatment of severe pain after ureteric stent removal [2-5]. NSAIDs are a common and effective renal colic treatment. Cyclooxygenase (COX) inhibition by NSAIDs can decrease renal colic by reducing ureteric contractility and inflammation [6]. The purposes of this double-blind, randomised, placebo-controlled trial was to determine the incidence of severe pain after ureteric stent removal and establish the efficacy of a one-time dose of a NSAID (rofecoxib) in preventing this severe pain.

Patients and Methods

This double-blind, randomised, prospective, placebo-controlled trial was approved by our Institutional Review Board. Patients were eligible if they were aged ≥18 years and had an indwelling ureteric stent after unilateral ureteroscopy. Patients were ineligible if they had advanced renal disease, hepatic impairment, history of gastrointestinal ulcers, coronary artery disease, asthma, mental impairment, an allergy to NSAIDs, were pregnant, or were participating in another clinical trial. Our research pharmacy service provided identical appearing study medication (rofecoxib 50 mg) or matching placebo tablets. Randomisation was performed in a 1:1 ratio. Using a random number generator, tablets were randomly assigned to numbered envelopes, which were sealed and given to the study investigators. Investigators and patients were ‘blinded’ to the randomisation scheme and medication identity until study termination. When a study patient arrived in the outpatient clinic for ureteric stent removal, the next numbered envelope was opened and the tablet given to the patient in the presence of an investigator no less than 15 min before stent removal. The patient filled out a pain questionnaire (with the assistance of a study nurse), which included a pain visual analogue scale (VAS), a record of pain medications taken and potential medication adverse effects. Cystoscopy and ureteric stent removal was then performed in the usual fashion by a single urologist (M.J.C.), with the use of urethral lidocaine jelly but no other additional pain medications or anaesthetics. We contacted the patient the following day, 24 h later, and using the same questionnaire, recorded the maximum pain level in the 24 h after ureteric stent removal, and rescue pain medication use, complications, and adverse effects. All narcotic use was converted into milligrams (mg) of i.v. morphine to facilitate comparison between groups. We reviewed the records of patients who were admitted to the hospital and recorded additional medications and interventions.

The placebo and study medication groups were compared using t-test for continuous data, Fisher's exact test for binomial data, and Mann–Whitney (rank-sum) test for VAS data. Results were considered significant at P < 0.05. Sample size determination was performed based on our clinical observations of severe pain after ureteric stent removal, because there are no published reports of this incidence. Assuming a 0.5 proportion of these patients experience severe pain, 28 patients would be needed in each arm to detect a 75% reduction in this proportion (two-sided test, α 0.05, β 0.80). We conducted an interim analysis (as part of our patient safety plan) after 10 patients in each group had competed the follow-up. Because of the marked decrease in severe pain in the NSAID group, the study was terminated.

The study medication was provided by Merck (Whitehouse Station, NJ, USA). No additional funding or assistance of any kind was provided by the manufacturer.

Results

In all, 40 patients were assessed for eligibility, and 22 qualifying patients were enrolled into the study and randomised, before the study termination (Fig. 1). Reasons for exclusion can be divided into two categories: contraindications to rofecoxib use [asthma (three), liver or renal impairment (three), pregnant (one), NSAID allergy (one), aged < 18 years (one)] and potential confounding factors [chronic pain (three), possible infection (one), renal transplant (one), i.v. drug abuse (one), current use of COX-2 inhibitor (one), non-english speaking (one), and mental health issues (one)]. In all, 21 (95%) had adequate follow-up data (one patient from the NSAID group could not be contacted within a reasonable time for the final pain questionnaire). Baseline characteristics were not different between the groups (Table 1). The most common indication for ureteroscopy was urolithiasis (14 patients), followed by PUJ obstruction (four) and haematuria (three) and were similar between the groups. The 6 F x 24 cm was the most common ureteric stent size. Stents were in place a mean (range) of 15 (4–34) days in the NSAID group and a mean (range) of 18.6 (5–41) days in the placebo group, which was not significantly different (P = 0.57). The mean (median) pain score before stent removal 2.36 (2) in the placebo group, and 1.65 (1) in the NSAID group (P = 0.31). The mean (median) score of maximum pain after stent removal was 2.7 (3) and 5.5 (7) for the NSAID and placebo groups respectively (P = 0.13; Fig. 2). The proportion of patients with severe pain (VAS score of ≥7) within the 24 h after ureteric stent removal was zero of 10 in the single-dose NSAID group and six of 11 (55%) in the placebo group (P < 0.01). All but one patient with severe pain had initial stent placement after successful ureteroscopic treatment for urolithiasis. The remaining patient had stent placement after diagnostic ureteroscopy. The mean amount of pain medication required during the 24 h after stent removal was 1.67 mg in the NSAID group compared with 4.77 mg in the placebo group (P = 0.19). However, those patients with severe pain (VAS score of >7) in the placebo group used an average 8.19 mg (P = 0.03). Two patients in the placebo group required evaluation and treatment in an emergency department for continued pain after stent removal, one required admission. No patients in the study NSAID group required additional interventions or visits. There were no study medication-related complications.

Figure 1.

Study Flow Diagram.

Figure 2.

Change in VAS score by group.

Table 1. Baseline demographics.
CharacteristicPlacebo groupNSAID groupP
N1110 
Mean age, years45.737.40.20
Gender, n/N (%) men5/11 (45)6/10 (60)0.51
Indication for ureteroscopy, n:   
Stone86 
PUJ obstruction22 
Other12 
Side, n/N (%) right8/11 (72)6/10 (60)0.54
Mean duration, days18.6150.57
Median VAS score210.31

Discussion

Pain caused by indwelling ureteric stents as well as their removal is relatively common and can be quite severe. Although we knew from experience that some patients experience severe pain of a renal colic nature and occasionally admission for pain control after ureteric stent removal, we could not find this described in the literature. How frequently this complication occurred was previously unknown. In the present study, we found the proportion of patients with severe pain (VAS score of ≥7) during the 24 h after ureteric stent removal was six of 11 (55%) in the placebo group. None of the present patients in the NSAID group experienced severe pain, showing that a single pre-stent removal dose of an NSAID can prevent severe pain after stent removal. In the postoperative period, narcotic use and additional visits also decreased in the NSAID arm. The narcotic requirements decreased from 8.19 mg morphine equivalents in those patients with severe pain, to 1.67 mg in the NSAID group. In all, six of 10 patients in the NSAID group required no pain medications after stent removal. Two patients in the placebo arm visited the emergency department after stent removal for pain control. It appears that the NSAID given before stent removal did not just replace some of the pain medication needed by these patients, but prevented the development of the most severe pain.

Direct contact of the stent against the urothelium and detrusor muscle spasm are thought to be mechanisms that can cause pain in patients with indwelling ureteric stents [7]. Mild reflux probably contributes to this discomfort as well [8]. These mechanisms only explain possible causes from an indwelling stent, and are not likely to cause pain after removal. Inflammation and oedema also presumably play a role in causing pain after stent removal, but this alone does not explain the substantial decrease in pain that the present patients experienced with only a single dose of rofecoxib. Oedema and inflammation can take many hours to days to resolve, and the present study population saw an immediate reduction in pain when they were given the study drug before stent removal.

Mechanisms that may help explain these results include reduced ureteric contractility and renal blood flow. Several studies have shown NSAIDs can decrease renal pain by inhibiting renal prostaglandin synthesis, which decreases renal blood flow and consequently lowers the pressure in the renal pelvis and ureter. Both COX-1 and COX-2 receptors have been found in human urothelium and COX-2 inhibitors also have been shown to decrease contractility in unobstructed pig ureters in vivo [9]. Nakada et al. [10] showed that a selective COX-2 inhibitor reduced ureteric contractility as effectively as indomethacin (a nonselective COX inhibitor) in both porcine and human ureteric segments in vitro. Decreased ureteric contractility and spasm may be the chief mechanism in reducing pain in the present patients, but further study is warranted to better characterise these mechanisms.

Novel stent materials, α1-blockers, and adjustment of stent length and size have decreased indwelling stent discomfort, although the oral administration of α-blockers has shown the greatest reduction in stent related morbidity [2]. Beddingfield et al. [4] showed that alfuzosin (α1-blocker) improved kidney/back/loin pain as well as pain with voiding associated with ureteric stents but did not decrease the amount of narcotics used while the stent was in place. Even injection of botulinum toxin into the ureteric orifice has been shown to decrease indwelling stent discomfort [3]. Unfortunately, none of these studies were designed to evaluate pain after stent removal. To our knowledge, this is the first preliminary study to directly evaluate an intervention designed to ameliorate the renal colic-like pain after ureteric stent removal.

At the time of the present study, rofecoxib was the NSAID of choice due to its selective COX-2 inhibition and decreased gastrointestinal adverse effects. After the present study was completed, rofecoxib was removed from the market owing to increased cardiovascular complications [11]. None of the present study participants had any cardiovascular or other adverse events secondary to the single dose rofecoxib administration during the study. However, relatively strict exclusion criteria were used to avoid potential adverse events related to the use of rofecoxib for this previously undescribed indication. At our institution, we now use a single dose of naproxen 220 mg (a nonselective COX inhibitor) as our NSAID of choice. Anecdotally, we have seen similar results using naproxen, although we have not confirmed this with a prospective study. It is unknown whether both COX pathways are responsible for the decreased pain in these patients or if the COX-2 pathway predominates. This should be considered a preliminary study with further investigation needed to determine the effectiveness of currently available non-selective COX inhibitors. A potential limitation of the present study is the small sample size. The study was closed to further recruitment after we found a significant difference in pain at the interim analysis, and was still powered appropriately to show a statistically significant difference in the incidence of severe pain between the groups.

In summary, we found a 55% incidence of severe pain after ureteric stent removal. A single dose of a NSAID before stent removal is a simple and cost-effective method of preventing this complication.

Acknowledgements

The study medication was provided by Merck (Whithouse Station, NJ, USA). No additional funding or assistance of any kind was provided by the manufacturer.

Conflict of Interest

None declared.

Ancillary