butylscopolammonium bromide does not provide additional analgesia when combined with morphine and...

ORIGINAL RESEARCH

Butylscopolammonium bromide does notprovide additional analgesia when combinedwith morphine and ketorolac for acuterenal colicemm_1502 144..150

Sung Wook Song,1 Kyuseok Kim,1 Joong Eui Rhee,1 Jin Hee Lee,1 Gil Joon Seo2 and Hyun Mi Park1

1Department of Emergency Medicine, Seoul National University Bundang Hospital, Sungnam-si, and2Department of Emergency Medicine, Seoul National University Hospital, Seoul, Korea

Abstract

Objective: To evaluate the effect of adding butylscopolammonium bromide (BB) to morphine andketorolac in the treatment of acute renal colic in the ED.

Methods: A prospective, double-blind, randomized controlled trial of i.v. triple therapy (morphine,ketorolac and BB) versus double therapy (morphine and ketorolac) in adult ED patientswith a clinical diagnosis of acute renal colic and a pain rating greater than five on a 10 cmvisual analogue scale (VAS). VAS was recorded at time 0, 20 and 40 min. Patients receivedrescue morphine at 20 or 40 min according to the protocol if needed. We compared painreduction and the need for rescue analgesia at 4 min between two groups.

Results: Eighty-nine patients were randomized over a 13 month period. A total of 46 (51.7%)patients received BB in addition to morphine and ketorolac. The mean difference in changein pain score in the triple therapy group and double therapy group was 7.1 cm (95% CI6.4–7.8) and 5.9 cm (95% CI 5.1–6.7), respectively (P = 0.024). Rescue morphine wasrequired by 7/46 (15.2% [95% CI 4.4–20.6]) patients in the triple therapy group and 14/43(32.6% [95% CI 18.0–47.1]) in the double therapy group (OR 0.37 [95% CI 0.133–1.038]).

Conclusions: Although the addition of BB to morphine and ketorolac appeared to show a statisticallysignificant reduction in pain compared with morphine and ketorolac alone, a reduction of1.2 cm on VAS is unlikely to be clinically significant.

Key words: analgesic, butylscopolammonium bromide, randomized controlled trial, renal colic.

Introduction

Renal colic is one of the most common urological prob-lems in the ED with an incidence estimated as high as12% during the average lifetime.1 The pain of renal colic

is related to acute ureteral obstruction, ureteric smoothmuscle spasm, oedema and inflammation at the level ofthe calculus.2,3 The first priority of treatment in the EDis rapid and effective pain control. However, the idealanalgesic regimen for acute renal colic remains a

Correspondence: Professor Kyuseok Kim, Department of Emergency Medicine, Seoul National University Bundang Hospital, 300Gumi-dong, Bundang-gu, Sungnam-si, Gyeonggi-do 463-707, Korea. Email: [email protected]

Sung Wook Song, MD, Clinical Fellow; Kyuseok Kim, MD, Assistant Professor; Joong Eui Rhee, MD, Associate Professor; Jin Hee Lee, MD,Assistant Professor; Gil Joon Suh, MD, Professor; Hyun Mi Park, Registered Nurse, Clinical Nurse Consultant.

doi: 10.1111/j.1742-6723.2011.01502.xEmergency Medicine Australasia (2012) 24, 144–150

© 2011 The AuthorsEMA © 2011 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine

difficult question in the ED. Intravenous non-steroidalanti-inflammatory drugs (NSAIDs) and opioids havebeen commonly used in acute renal colic as the main-stay treatment.4 The use of anticholinergic agents,such as butylscopolammonium bromide (BB), has beendescribed to decrease ureteral spasm by controllingsmooth muscle spasm.5–8 The concept of balanced anal-gesia, which means a combination of different regimensto achieve a synergistic effect, was suggested in earlystudies.9 A recent randomized controlled study showedthat a combination of i.v. morphine and ketorolacachieved superior pain reduction than either drug alone,and was associated with decreased rescue analgesicdemands.10 However, there has been little evidence inprevious studies to support the addition of drugs withanticholinergic effect when used in combination withNSAIDs and opioids for the treatment of acute renalcolic in the ED.11–16

Our aim was to test our null hypothesis that theaddition of BB to therapy with NSAIDs and opioidswould not reduce the pain of acute renal colic more thanthe dual therapy alone. Our secondary goals were tocompare the rate of change in analgesic response, thedemand for rescue analgesia, and the incidence and typeof adverse effects in each treatment group.

Methods

We conducted a prospective, randomized, controlled,double-blind clinical trial of i.v. triple combination (mor-phine, ketorolac and BB) versus double combination(morphine and ketorolac). This study was performed inthe adult ED of a tertiary-care urban academic hospitalin Bundang, Korea, with 67 000 annual visits from 1November 2007 to 30 December 2008. Study enrolmentwas carried out between 09.00 and 17.00 hours on week-days during the study period. The institutional reviewboards of the hospital’s ethics committee approved thestudy.

Our inclusion criteria were patients with a clinicalpresentation of ‘typical renal colic’ rather than ‘con-firmed urinary stone by CT scan’. We chose to use thisclinical inclusion criterion to reflect the real clinical situ-ation faced by emergency physicians. In our worksetting, not all patients with clinically diagnosed renalcolic undergo CT or X-ray imaging acutely. Therefore,we did not want to subject patients to ionizing radiationperhaps unnecessarily. Patients presenting to the EDwith flank pain were screened for inclusion by the triagenurse. The triage nurse identified patients who were at

least 18 years of age and whose flank pain was consistentwith an abrupt onset of severe paroxysmal unilaterallocation. If the patient met these inclusion criteria, thetriage nurse notified the attending physician and clinicalresearch nurse of potentially eligible patients. Theattending physician then confirmed the triage nurse’ssuspicion of renal colic and assessed the patient forexclusion criteria. Patients were excluded from the studyif they met any of the following criteria: (i) patient painrating less than five on a 10 cm visual analogue scale(VAS); (ii) confirmed or suspected pregnancy; (iii) breast-feeding; (iv) contraindication to NSAIDs, opioids or BBs;(v) history of peptic ulcer or renal disease; (vi) use ofanalgesics within 6 h of presentation; (vii) current use ofanticoagulants; (viii) history of bleeding tendency; (ix)suspicious surgical condition; (x) hemodynamic instabil-ity, defined as pulse >110/min and systolic blood pres-sure <100 mmHg; or (xi) previous participation in thestudy. Consent was obtained from potentially eligiblepatients by the clinical research nurse of this study.

Patients who agreed to participate were then dividedinto two treatment groups by the triage nurse using arandomization table.

Standing orders were used to manage the patient’spain after enrolment, allowing ED nursing staff to carefor the patient without specific physician’s orders. Allpatients received standard treatment defined as 1 L ofnormal saline hydration at 240 mL per hour, 30 mgketorolac i.v. and 5 mg morphine i.v. over 5 min at timezero. In addition to the standard treatment, the BBgroup also received 20 mg of BB i.v., whereas theplacebo group was given 50 mL of normal saline solu-tion, also at time zero. BB was diluted with 50 mL ofnormal saline by the treating nurse so the study drug(BB) appeared identical to the placebo. All patients, theattending physician and the clinical research nurse wereblinded to the patient’s treatment arm. The triage nurseand treating nurse were not.

The blinded clinical research nurse prospectively col-lected the following information from patients andrecorded on a preformatted data collection form: loca-tion of pain, pain score through VAS, adverse effectsand need for rescue morphine. Other data were retro-spectively collected from medical records by a clinicalresearch nurse: age, sex, ED arrival time, initial drugadministration time, time of ED discharge, self-reportedmedical history, vital signs, presence of costovertebralangle tenderness, results of blood urea nitrogen, creati-nine, results of laboratory urine test, radiologist’s CTscan reports and radiologist’s reports on i.v. pyelogra-phy after ED discharge.

Butylscopolammonium bromide in renal colic

145© 2011 The AuthorsEMA © 2011 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine

Pain score was measured on a 10 cm VAS at time 0,20 and 40 min after receiving the study drugs.

The rescue analgesia (3 mg morphine i.v.) was admin-istered if the patients felt a persistent pain more thanfive on the VAS or requested analgesia at 20 or 40 minbecause of persistent severe pain.

Adverse effects, including nausea, vomiting, analtered level of consciousness (GCS < 15), respiratorydepression (respiratory rate < 12/min), dizziness andany other unusual events, were observed until thepatient was discharged.

The primary outcome was the pain reductionchanges between the two groups at 40 min. Secondaryoutcomes were the rate of change in analgesic response,the need for rescue morphine and the occurrence ofadverse effects.

Sample size was calculated for our primary outcome.A minimum of 42 patients per group (total of 84) wererequired to detect a 1.8 cm difference in VAS scorebetween groups with a power of 90% at the two-sided5% significance level. This difference was based on anassumption that the addition of BB to morphine andketorolac should reduce the pain as much as the addi-tion of ketorolac to morphine in a previous study. Thestandard deviation used for this calculation was 2.5.10 Afollow-up loss of 10% was also accounted for in the finalrequired number of patients to randomize. However, oursample size did not provide enough statistical power foranalysing our secondary outcomes.

SPSS version 13.0 (SPSS, Chicago, IL, USA) was usedfor data analysis. Continuous variables were presentedas mean � SD and were compared using t-tests fornormal distribution or Wilcoxon rank sum test for non-parametric distribution. Categorical variables were pre-sented as proportions and were compared using Fisher’sexact tests. All statistical analyses were conductedusing two-tailed tests. Significance levels of P < 0.05(two-tailed) were accepted. Mean differences in painreduction were calculated by using mean differences inVAS pain score between initial and at the end of 40 min,and we analysed the response feature. We also analysedthe rate of change of the analgesic response between thetwo groups. We used as our dependent variable thechange in VAS from baseline for each patient and per-formed linear regression for changes in VAS againstminutes after treatment. We calculated the estimate ofthe regression slope parameter for all patients and com-pared this parameter between the two groups. Use ofrescue morphine to control sustained pain was calcu-lated odds ratios using logistic regression, and presenceof adverse events was considered a binomial variable.

Results

During the 13 months between November 2007 andDecember 2008, 115 adult patients suspected of havingacute renal colic were potentially eligible for inclusion inthe trial. Of these, four patients were excluded followingthe attending physician’s examination as the attendingphysician did not think they were suffering from renalcolic. Twelve patients did not want to participate in thetrial and 10 patients were excluded because their painratings were less than five.

The remaining 89 patients were included in the analy-sis consisting of 46 patients in the BB group and 43patients in the placebo group (Fig. 1). Baseline patientcharacteristics between the BB group and placebogroup are shown in Table 1.

The mean pain scores at time 0, 20 and 40 min areshown in Table 2 and characterized in Figure 2.

The mean difference in pain reduction between theplacebo group and the BB group during 20 min was-0.6 cm (95% CI -1.6–0.5, P = 0.287). The mean VASdifference between the placebo group and the BB groupat the end of the protocol was -1.2 cm (95% CI -2.2 to-0.2, P = 0.024) (Table 3, Fig. 3).

The rates of change of the analgesic responsebetween the two groups are showed in Table 4 andFigure 4.

Rescue morphine to control sustained pain was givento 14/43 (32.6% [95% CI 18.0–47.1]) patients in theplacebo group and 7/46 (15.2% [95% CI 4.4–26.0])patients in the BB group (OR 0.372; 95% CI 0.133–1.038,P = 0.059).

The noted adverse effects are summarized in Table 5.

Discussion

A number of previous studies have demonstrated thatwhen used as monotherapy, BB is an inferior analgesicagent in the setting of renal colic when compared withother agents, such as opioids or NSAIDs.11,13,14,17

Recently, ketorolac and morphine when used in com-bination were reported to be more effective in relievingthe pain of renal colic than when these agents wereused as monotherapy.10 The aim of our study was todetermine whether the addition of BB to this dualtherapy regimen would have an additional effect onpain reduction.

We were able to demonstrate a statistically signifi-cant difference in pain scores at 40 min (1.2 cm), and in

SW Song et al.

146 © 2011 The AuthorsEMA © 2011 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine

the rate of reduction of pain in the triple therapy group,but do not consider these differences to be clinicallysignificant. Before the study we had defined a clinicallysignificant difference in pain scores to be 1.8 cm in

keeping with the ketorolac and morphine study.10 Otherauthors have suggested that a difference of more than2.0 cm on a VAS constitutes a minimum clinically mean-ingful change.18–20

Figure 1. Study design summary. BB, butylscopolammonium bromide; VAS, visual analogue scale.

Table 1. Comparison of baseline characteristics of the two study groups

Placebo (n = 43) BB (n = 46) P-values

Age (years), mean (�SD) 41.9 (�9.6) 38.8 (�9.8) 0.132†Male, n (%) 34 (79.1) 38 (82.6) 0.789‡Weight (kg), mean (�SD) 71.1 (�10.1) 70.6 (�10.6) 0.818†Ureteral stone history, n (%) 14 (32.6) 15 (32.6) 1.000‡Initial vital sign

SBP (mmHg), mean (�SD) 142.5 (�22.4) 144.5 (�19.6) 0.667†DBP(mmHg), mean (�SD) 78.5 (�13.7) 79.8 (�18.7) 0.717†HR (/min), mean (�SD) 80.0 (�2.5) 76.5 (�11.3) 0.161†RR (/min), mean (�SD) 19.6 (�1.5) 20.1 (�1.5) 0.176†

Presence of CVAT, n (%) 41 (95.4) 42 (91.3) 0.678‡Microscopic haematuria, n (%) 43 (100) 44 (95.7) 0.167‡BUN (mg/dL), mean (�SD) 14.6 (�3.7) 14.9 (�3.6) 0.681†Creatinine (mg/dL), mean (�SD) 1.1 (�0.2) 1.1 (�0.2) 0.477†Initial VAS, mean (�SD) 8.4 (�1.4) 8.5 (�1.4) 0.823†CT scan in the ED, n (%) 8 (18.6) 10 (21.7) 0.459‡Positive for stone, n (%) 8 (18.6) 9 (19.6) 0.562‡IVP after ED discharge, n (%) 28 (65.1) 37 (80.4) 0.134‡Positive for stone, n (%) 14 (32.6) 22 (47.8) 0.166‡

†t-tests. ‡Fisher’s exact tests. BB, butylscopolammonium bromide; BUN, blood urea nitrogen; CVAT, costovertebral angle tenderness;DBP, diastolic blood pressure; HR, heart rate; IVP, intravenous pyelography; RR, respiratory rate; SBP, systolic blood pressure; VAS, visualanalogue scale.



Although we demonstrated that twice as manypatients in the placebo group required rescue morphinewhen compared with the BB group (32.6% vs 15.2%, OR0.372 [95% CI 0.133–1.038]), this was not statisticallysignificant. However, our sample size calculations were

based on the VAS outcome rather than need for rescueanalgesia. A larger study would be required to deter-mine if there is a clinically significant difference in thisoutcome.

Only 2/42 (4.7% [95% CI -1.9–11.2]) patients in theplacebo group and 3/46 (6.5% [95% CI -0.8–13.9]) in theBB group suffered adverse effects. All patients weredischarged without significant adverse events or theneed for intensive medical care.

Other studies examining the effects of anticholinergicagents used in combination with opioids or NSAIDshave shown conflicting results. Stankov et al.13 reported

Table 2. Mean pain scores between the two study groups

Placebo (n = 43) BB (n = 46) P-value

0 min VAS (�SD) 8.4 (�1.4) 8.4 (�1.4) 0.82320 min VAS (�SD) 3.1 (�2.4) 2.6 (�2.4) 0.34340 min VAS (�SD) 2.5 (�2.6) 1.3 (�1.9) 0.023

BB, butylscopolammonium bromide; VAS, visual analoguescale.

Figure 2. Pain scores for butylscopolammonium bromide (BB)and placebo groups at 0, 20 and 40 min.

Table 3. A response feature analysis between the two studygroups

Minutes after treatment

20 40

Placebo (n = 43)Mean change from baseline VAS 5.2 5.9Standard error 0.362 0.39995% confidential interval 4.5 to 6.0 5.1 to 6.7

BB (n = 46)Mean change from baseline VAS 5.8 7.1Standard error 0.372 0.33795% confidential interval 5.0 to 6.5 6.4 to 7.8

Mean differencePlacebo – BB VAS -0.6 -1.295% confidential interval -1.6 to 0.5 -2.2 to -0.2P-values 0.287 0.024

BB, butylscopolammonium bromide; VAS, visual analoguescale.

Figure 3. Mean visual analogue scale change plots for eachgroup. ( ) Butylscopolammonium bromide group (n = 46), ( )placebo group (n = 43).

Table 4. Comparison of regression slope in visual analoguescale change against time

Placebo(n = 43)

BB(n = 46)

P-values

Estimates of regression slopeMean 0.147 0.177 0.02425th–75th percentile 0.115–0.200 0.138–0.23395% confidentialinterval

0.127–0.168 0.160–0.194

BB, butylscopolammonium bromide.

SW Song et al.


greater pain relief and need for rescue analgesiawhen using a combination of tramadol and butylscopo-lamine compared with either agent as monotherapy.Holdgate,21 however, reported that the addition of BB tomorphine and i.v. fluid, with or without indomethacin,did not reduce the opioid requirements to achieveadequate analgesia or the need for rescue analgesia.Similarly, a third study16 failed to demonstrate any addi-tional benefit of sublingual hyoscyamine over ketorolacalone in terms of pain scores and need for rescue anal-gesia. Our study is unique in that it measured reduc-tions in VAS pain scores, the rate of reduction in VASand the need for rescue analgesia. It also includedpatients that were diagnosed clinically with renal colic,

rather than patients with radiologically proven uretericcalculi.

There are several limitations to our study. Patientswere only recruited between 09.00 and 17.00 hours onweekdays because of the availability of the clinicalresearch nurse. The study was only conducted in the EDof one tertiary-care urban hospital. The results might notbe generalized to other institutions. CT scan was notperformed on all patients at the time of ED presentation,or used as part of the inclusion criteria. We chose to usea presentation of ‘typical renal colic’ as the inclusioncriteria rather than confirmed ureteric stone. This assess-ment was undertaken by both the triage nurse and theattending physician before enrolment in the study. Theattending physician was able to exclude patients withsurgical conditions that often mimic renal colic (e.g.abdominal aortic aneurysm, biliary colic, appendicitis).All patients therefore had a typical renal colic presenta-tion, and 87/89 (97.8% [96% CI 94.6–100]) had haema-turia on urinalysis. The two patients without haematuriahad CT-proven ureteric calculi. We believe our approachto inclusion criteria reflects the real world assessmentand management of renal colic. The presumptive diag-nosis is made on clinical grounds (classic symptoms,physical examination and haematuria) with treatmentinstituted as a result of this assessment. Analgesia isusually not withheld pending a confirmatory CT scan.

Imaging studies were not carried out on all patients.This is the usual practice in our institution where we

Figure 4. Comparison of the regression slope for changes in visual analogue scale (VAS) against minutes after treatment between thetwo study groups.

Table 5. Adverse effects between the two groups

Placebo(n = 43)

BB(n = 46)

Total(n = 89)

Nausea, n (%) 1 (2.3) 0 1 (1.1)Vomiting, n (%) 1 (2.3) 0 1 (1.1)Altered level of

consciousness, n (%)0 0 0

Respiratory depression,n (%)

0 0 0

Dizziness, n (%) 1 (2.3) 2 (4.3) 3 (3.4)Others, n (%) 0 1 (2.2)† 1 (1.1)

†One patient was sleepy, but there was not any significantmental alteration. BB, butylscopolammonium bromide.



have concerns about exposing patients to ionizing radia-tion. However, of the 18 patients who had a CT scanperformed in the ED, 17 were positive for a calculus(94.4% [95% CI 82.7–100]). A further 57 patients under-went i.v. pyelography following their discharge from theED, and 33 of these were positive for calculi (57.9% [95%CI 44.7–71.1]). It is well known that patients who sufferan acute attack of renal colic might subsequently passtheir stone while waiting for a follow-up imaging study,and therefore, delayed imaging might be negative. Four-teen patients did not undergo any imaging procedure.

Conclusions

In this study, the addition of BB to a regimen of ketorolacplus morphine did not appear to reduce the pain of renalcolic sufficiently for it to be considered a clinically impor-tant addition to therapy. A larger study is required todetermine whether the addition of BB to ketorolac andmorphine would reduce the need for rescue analgesia.

Author contributions

KK, JER and GJS conceived the study. JHL and HMPsupervised the data collection. KK, SWS, HMP reviewedthe collected data. KK and SWS analysed, interpreteddata and performed statistical analysis. SWS draftedthe manuscript, and all authors contributed substan-tially to its revision. KK takes responsibility for thepaper as a whole.

Competing interests

None declared.

Accepted 19 September 2011

References

1. Manthey DE, Teichman J. Nephrolithiasis. Emerg. Med. Clin.North Am. 2001; 19: 633–54, viii.

2. Segura JW, Preminger GM, Assimos DG et al. Ureteral StonesClinical Guidelines Panel summary report on the management ofureteral calculi. The American Urological Association. J. Urol.1997; 158: 1915–21.

3. Engineer RS, Peacock WF. Chapter 96. Urologic stone disease. In:Tintinalli JE, Stapczynski JS, Ma OJ, Cline DM, eds. Tintinalli’sEmergency Medicine: A Comprehensive Study Guide, 6th edn.New York: Mcgraw-Hill, 2004; 620–5.

4. Holdgate A, Pollock T. Nonsteroidal anti-inflammatory drugs(NSAIDs) versus opioids for acute renal colic. Cochrane DatabaseSyst. Rev. 2005; (2): CD004137.

5. Menon M, Resnick MI. Urinary lithiasis: etiology, diagnosis, andmedical management. In: Walsh PC, Retik AB, Vaughan ED Jr,Wein AJ, eds. Campbell’s Urology. 9th ed. Philadelphia, Pa: WBSaunders Co; 2002:3229–305.

6. Maclean A, Dunn R. Renal/urology. In: Dunn R, Beach W, eds.The Emergency Medicine Manual. Adelaide: Venom Publishing,2000; 437–39.

7. Brenner PC. Urological trauma and emergencies. In: Fulde GWO,ed. Emergency Medicine: The Principles of Practice. Sydney:Churchill Livingstone, 2004; 206–14.

8. Davenport K, Timoney AG, Keeley FX. Conventional and alter-native methods for providing analgesia in renal colic. BJU Int.2005; 95: 297–300.

9. Cordell WH, Wright SW, Wolfson AB et al. Comparison of intra-venous ketorolac, meperidine, and both (balanced analgesia) forrenal colic. Ann. Emerg. Med. 1996; 28: 151–8.

10. Safdar B, Degutis LC, Landry K, Vedere SR, Moscovitz C,D’Onofrio G. Intravenous morphine plus ketorolac is superior toeither drug alone for treatment of acute renal colic. Ann. Emerg.Med. 2006; 48: 173–81, 181.e1.

11. Jones JB, Dula DJ. The efficacy of sublingual hyoscyaminesulfate and intravenous ketorolac tromethamine in the relief ofureteral colic. Am. J. Emerg. Med. 1998; 16: 557–9.

12. Giamberardino MA, Valente R, De Bigontina P, Iezzi S, VecchietL. Effects of spasmolytic and/or non-steroidal antiinflammatorydrugs on muscle hyperalgesia of ureteral origin in rats. Eur. J.Pharmacol. 1995; 278: 97–101.

13. Stankov G, Schmieder G, Zerle G, Schinzel S, Brune K.Double-blind study with dipyrone versus tramadol and butylsco-polamine in acute renal colic pain. World J. Urol. 1994; 12: 155–61.

14. Benyajati C. Comparative study of Baralgan and hyoscine-N-methyl bromide in the treatment of intestinal and renal colicypain. J. Med. Assoc. Thai. 1986; 69: 569–73.

15. Ross JA, Edmond P, Griffiths JM. The action of drugs on theintact human ureter. Br. J. Urol. 1967; 39: 26–30.

16. Jones JB, Giles BK, Brizendine EJ, Cordell WH. Sublingual hyos-cyamine sulfate in combination with ketorolac tromethamine forureteral colic: a randomized, double-blind, controlled trial. Ann.Emerg. Med. 2001; 37: 141–6.

17. al-Waili NS, Saloom KY. Intravenous tenoxicam to treat acuterenal colic: comparison with buscopan compositum. J Pak MedAssoc. 1998; 48: 370–2.

18. Todd KH. Clinical versus statistical significance in the assess-ment of pain relief. Ann Emerg Med. 1996; 27: 439–41.

19. Gallagher EJ, Liebman M, Bijur PE. Prospective validation ofclinically important changes in pain severity measured on avisual analog scale. Ann Emerg Med. 2001; 38: 633–8.

20. Kelly AM. Setting the benchmark for research in the manage-ment of acute pain in emergency departments. Emerg Med(Fremantle). 2001; 13: 57–60.

21. Holdgate A, Oh CM. Is there a role for antimuscarinics in renalcolic? A randomized controlled trial. J Urol. 2005; 174: 572–5;discussion 75.

SW Song et al.


butylscopolammonium bromide does not provide additional analgesia when combined with morphine and...

Documents