• 
• 

Three Part Question

In [children with acute abdominal pain] does [analgesia before surgical consultation] affect [surgical diagnostic accuracy]?

Clinical Scenario

A 9 year old boy presents with severe right iliac fossa pain. You contact the surgical team who are currently in theatre and will not be able to attend for at least twenty minutes. You wonder if administering morphine to the boy will hinder or delay diagnosis.

Search Strategy

Dialog Datastar interface in Medline and Embase
MEDLINE (1950 - date):(abdominal ADJ pain OR acute ADJ abdomen) AND (analges$ OR pain ADJ relief) AND diagnosis AND LG=EN AND HUMAN=YES AND (CHILD# OR ADOLESCENT.DE. OR INFANT#)

EMBASE (1974 - date):(abdominal ADJ pain OR acute ADJ abdomen) AND (analges$ OR pain ADJ relief) AND diagnosis AND LG=EN AND HUMAN=YES AND CHILD=YES

Search Outcome

MEDLINE yielded 56 papers and EMBASE yielded 100 papers. BestBETs yielded 1 BET, but although the clinical scenario involved the assessment of a child, all of the evidence related to studies performed in adults.

After removal of duplicates, 134 abstracts were scanned. Four papers were found to be relevant to the three-part question

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Kim et al. 2002 USA	60 children aged 5-18 years with abdominal pain for <5 days scoring 5 or more on a visual analog scale. 0.1mg/kg morphine IV vs same volume 0.9% NaCl IV. Patients examined before and after administration of study drug by paediatric emergency physicians and surgeons	Randomised, double-blind, placebo-controlled trial (level 2b)	Pain score	Median difference in reduction of pain score between two groups of points (P=0.002)	Small sample size. Post-hoc power calculation performed but not fully reported.
			Change in mean number of areas of tenderness to palpation before and after study drug	Pediatric emergency physicians: Morphine; 0.9 (95% CI 0.1 to 1.8). Placebo; 0.1 (95% CI -0.6 to 0.7) Surgeons: Morphine; 0.1 (95% CI -0.6 to 0.7). Placebo; 0.3 (95% CI -0.1 to 0.6)
			Change in mean number of areas of tenderness to percussion before and after study drug	Pediatric emergency physicians: Morphine; 1.0 (95% CI 0.1 to 1.9 ). Placebo; 0.0 (95% CI -0.3 to 0.4) Surgeons: Morphine; 0.2 (95% CI -0.1 to 0.6). Placebo; -0.2 (95% CI -0.7 to 0.4)
			Difference in diagnostic accuracy (true surgical causes and true non-surgical causes as a proportion of all results) between morphine and placebo groups	Pediatric emergency physicians before study group: 1.8% (95% CI 0.1 to 2.0). After study drug: 5.4% (95% CI -0.1 to 0.3) Surgeons before study drug: 11.6% (95% CI 0.1 to 2.0). After study drug: 11.8% (95% CI 0.1 to 2.0).
Kokki et al. 2005 Finland	63 children aged 4-15 years with abdominal pain scoring 5 or more out of 10 on a visual analog scale. 0.1 mg/kg buccal oxycodone vs same volume buccal 0.9% NaCI. Patients examined before and after administration of study drug 0.1mg/kg buccal oxycodone vs same volume buccal 0.9% NaCl. Patients examined before and after administration of study drug.	Randomised, double-blind placebo-controlled Trial (level 2b)	Mean summed pain intensity difference over 7 observations at half-hourly intervals. Diagnostic accuracy (true surgical causes and true non-surgical causes as a proportion of all results)	Oxycodone group: 22 (SD 18). Placebo group: 9 (SD 12). Mean difference: 13 (95% CI 2 to 24; p=0.004). Oxycodone group: Before analgesia 72%. After analgesia 88% (p = 0.12). Placebo group: Before analgesia 84%. After analgesia 84% (p value not reported)	Small number of patients. Powered to detect significant change in pain scores but not diagnostic accuracy.
Bailey et al, 2007, Canada	90 children aged 8-18 years with presumptive appendicitis and pain scoring >5 out of 10 on a verbal numeric scale 0.1 mg/kg iv morphine vs similar looking placebo. Patients examined before and after administration of study drug.	Randomized, double-blind placebo controlled trial (level 2b)	Difference between time of arrival in ED and time of surgical decision for disposition of patient	Morphine group: 269 min. Placebo group: 307 min. Mean difference:-34 min (95%CI - 105 to 40 min)	Sample size required to detect significant reduction in pain (estimated as -20mm was calculated at 152). Sample size required to detect difference of 1 hr in time between arrival and surgical disposition was calculated at 184. Study terminated early due to slow enrolment and interim analysis showing no difference.
			Decrease in pain intensity on 100 mm VAS after study drug administration	Morphine group: 24 (SD 23 ) mm. Placebo group: 20 (SD 18 mm Mean difference: 4 mm (95% CI -5 to 12 mm).
Green et al. 2005 Canada	108 children aged 5-16 years with abdominal pain of <48 hours duration of possible surgical origin. 0.05mg/kg morphine IV vs same volume 0.9% NaCl IV. Patients examined before and after administration of study drug by paediatric emergency physician and afterwards only by paediatric surgeon.	Double-blind, randomised, placebo-controlled trial (level 2b)	Pain score using colour analogue score	Mean pain score reduction with morphine 2.2 vs 1.2 with placebo (P=0.015)	No power calculation.
			Physician confidence in diagnosis (0-100%) before and after study drug	Morphine group: 68.9% (before) vs 69.5% (after) (effect size: 1.2%; 95% CI -2.9% to 5.3%). Placebo group: 65.5% (before) vs 70.9% (after) (effect size; 5.3% 95% CI 2.7% to 7.9%)
			Appendicitis at laparotomy in those children undergoing surgical intervention	Morphine group:24/25. Placebo group: 22/24 (p=0.25)
			Surgeon confidence in diagnosis (0-100%) after study drug	Morphine group: 73.8%. Placebo group: 73.6% (effect size; 0.01%; 95% CI -0.39% to 0.40%).

Comment(s)

Classic teaching in general surgery has suggested that administration of analgesia in children with acute abdominal pain should be deferred until after a definitive surgical treatment plan has been formulated. Theoretically, analgesia may mask pain and lessen the examination findings that would normally suggest a surgical cause for abdominal pain. All but one of the studies found that opioid analgesia was effective at reducing pain scores in children with acute abdominal pain. Bailey et al state that morphine was not more effective than placebo in diminshing pain. This study suffers from being significantly underpowered regarding this outcome, but this does not fully explain the result, which appears to be due to a high placebo response compared to the other studies rather than a lack of response to morphine. The reasons for such a high response are likely to be complex and beyond the scope of this commentary. The studies identified all report that administration of analgesia to children with acute abdominal pain did not significantly interfere with diagnosis. Diagnostic accuracy was defined in two studies as true surgical and true non-surgical diagnoses as a proportion of all results. One study detected no difference (Kokki) whilst Kim et al noted a difference when children were examined by one subgroup of doctors, although the confidence intervals are borderline, and the authors' other measure of diagnostic accuracy (reduction in mean number of areas of abdominal tenderness) was unaffected by the administration of analgesia. One study (Green et al) used the doctors estimation of confidence in diagnosis as their measure of diagnostic accuracy. The remaining study Bailey et al used the time between arrival in the ED and the surgical decision. Other proxy measures of diagnostic accuracy, such as differences in time to operating theatre and perforation rates, where recorded, are also reported as being unaffected. These results are in keeping with what is known in adult patients; a recent Cochrane review of this topic concluded that the use of opioid analgesics in patients with acute abdominal pain does not delay treatment decisions. None of the studies identified included children less than 5 years old. In infants and preschool children, acute abdomen is uncommon, examination findings may be non-specific and as a result diagnosis may be difficult. Generalising findings from older children to this age group may therefore be detrimental. The clinical assessors were reported as blinded to whether the children had received analgesia or placebo. However, as the same assessor was responsible for examining the child before and after administration of the study drug, no mechanism existed to prevent bias introduced by the assessor remembering the previous clinical findings. Only in the study by Green et al were the children examined after administration of medication by another assessor (a paediatric surgeon) who was naive to the initial examination findings and whose confidence in diagnosis was the same for both the analgesia and the placebo groups. The studies all suffer from being underpowered to detect true differences in diagnostic ability. Post-hoc power calculations performed on the papers by Green et al and Kokki et al indicate that in order to attain a power of 80% over 1000 patients would need to be recruited into each arm of a trial. This would be a significant undertaking and subjecting several thousand children in pain to placebo analgesia to identify a difference in diagnostic accuracy that may have little clinical impact is ethically suspect. Certainly, none of the studies above identified major morbidity or mortality as a result of early treatment with analgesia. Since such a trial is therefore very unlikely to be performed, what remains is to make an informed decision based on the current evidence, and with the patient’s interests foremost. Taking these into account, children with acute abdominal pain should be treated promptly and adequately with analgesia unless future studies suggest evidence of harm.

Clinical Bottom Line

Surgical diagnostic accuracy is not affected by pre-assessment analgesia. (Grade B) Early analgesia in children with suspected surgical abdominal pain is effective; administration should not be withheld pending surgical consultation. (Grade B)

References

1. Kim MK, Strait RT, Sato TT, Hennes HM A Randomised Clinical Trial of Analgesia in Children with Acute Abdominal Pain Acad Emerg Med 2002;9:281-287
2. Kokki H, Lintula H, Vanamo K, Heiskanen M, Eskelinen M Oxycodone vs Placebo in Children With Undifferentiated Abdominal Pain Arch Pediatr Adolesc Med 2005;159:320-325.
3. Bailey B, Bergeron S, Gravel J, et al. Efficacy and impact of intravenous morphine before surgical consultation in children with right lower quadrant pain suggestive of appendicitis: a randomised controlled trial. Ann Emerg Med 2007;50:371-8.
4. Green R, Bulloch B, Kabani A, Hancock BJ, Tenenbein M Early Analgesia for Children With Acute Abdominal Pain Pediatrics 2005;116:978-983.