Tenderness and Risk for Intra-abdominal Injury in Children with - - PowerPoint PPT Presentation

Importance of Costal Margin Tenderness and Risk for Intra-abdominal Injury in Children with Blunt Abdominal Trauma

Pediatric Trauma Society Scottsdale, AZ | November 6-7, 2015 Katherine T. Flynn-O’Brien, MD, MPH Nathan Kupperman, MD, MPH James F. Holmes, MD, MPH

Disclosures

• I have no disclosures.

Background

• Risk of radiation exposure from CT imaging
• Suspected intra-abdominal injury (IAI)
• History & physical exam (H&P) findings assist

physician’s decision

• PECARN prediction rule
• 7 H&P findings to identify children with low risk

for IAI

• Costal margin tenderness (CMT) was not

identified as an independent predictor

Background

IAI

Abdominal wall trauma GCS <14 Abdominal tenderness Costal margin tenderness Thoracic wall trauma c/o Abdominal pain Absent or decreased BS Vomiting

Objective

1)

Determine the risk of IAI in children with:

• Isolated costal margin tenderness (CMT)
• Non-isolated CMT (+other PECARN)

2)

Quantify potentially avoidable CT scans in children with CMT and low risk for IAI

Methods

• Planned secondary analysis
• Large prospective multicenter cohort study
• May 2007-Jan 2010
• 12,044 children <18 years
• Blunt torso trauma/”non-trivial” abd trauma
• 761 children with IAI; 203 IAI with acute

intervention

• Limited to GCS 14-15

Methods

• Descriptive statistics
• Comparison of binomial proportions with

exact confidence intervals (CI)

• 2-sided p-value 0.05 or 1-sided p-value 0.025
• Relative risk
• Risk difference

12,044 patients enrolled 11,045 patients with GCS 14-15

999 excluded for GCS: 995 GCS 3-13 4 missing data for GCS 107 missing data for CMT

9,174 patients with data for costal margin tenderness and all PECARN predictor variables

1,764 excluded due to missing data for PECARN predictors:

• Abd. trauma/seat belt sign (n = 334)
• Abd. tenderness (n = 558)

Thoracic wall trauma (n = 66) C/o abd. pain (n = 1077) Absent/dec. breath sounds ( n = 195) Vomiting (n = 188)

9,174 patients with data for costal margin tenderness and PECARN predictor variables 177 isolated CMT 1,090 CMT and 1+ PECARN variables 7,907 without CMT 1,267 CMT

9,174 patients with data for costal margin tenderness and PECARN predictor variables 177 isolated CMT 1,090 CMT and 1+ PECARN variables 7,907 without CMT 1,267 CMT 3,804 without CMT or PECARN variables 4,103 without CMT; but with 1+ PECARN var.

9,174 patients with data for costal margin tenderness and PECARN predictor variables 177 isolated CMT 1,090 CMT and 1+ PECARN variables 7,907 without CMT 1,267 CMT 3,804 without CMT or PECARN variables 4,103 without CMT; but with 1+ PECARN var.

Results

Isolated CMT Non-isolated CMT

Demographics Age, years 14 (10, 16) 13 (10, 16) Male 112 (63.3) 712 (65.3) Mechanism of Injury MVC 52 (29.4) 263 (24.1) Ped v. MVC 14 (7.9) 117 (10.9) ATV/scooter 5 (2.8) 55 (5.1) Bicycle 6 (3.4) 86 (7.9) Fall from elev 18 (10.2) 119 (10.9) Fall from stairs 7 (4.0) 10 (1.8) Abd struck 23 (13.0) 206 (18.9)

Results

Isolated CMT Non-isolated CMT

Other PECARN variables Abdominal trauma n/a 272 (25.0)

• Abd. tenderness

n/a 837 (76.8) Thoracic wall trauma n/a 445 (40.8) C/o abdominal pain n/a 839 (77.0) Absent/dec. BS n/a 56 (5.1) Vomiting n.a 119 (10.9)

Results

• 1267 with CMT
• 187 patients with IAI
• 36 patients with IAI requiring acute

intervention

Results: CMT +/- PECARN vars

• 1267 with CMT
• 187 patients with IAI
• 36 patients with IAI requiring acute

intervention

177 isolated CMT 1,090 CMT and 1+ PECARN variables

0%

17.2% (15-19.5%) IAI 3.3% (2.3-4.5%) IAI

requiring acute intervention

Results: Any IAI (n = 507)

No./No. % (95% CI) Relative Risk (95% CI) Isolated CMT 0/177 0.0 (0.0-2.2)

• CMT, 1+

PECARN 187/1090 17.2 (15.0-19.5) 16.3 (11.7-22.8) No CMT, 1+ PECARN 280/4103 6.8 (6.1-7.6) 6.5 (4.7-9.0) No CMT, no PECARN 40/3804 1.1 (0.8-1.4) Ref

Results: IAI acute interv. (n = 113)

No./No. % (95% CI) Relative Risk (95% CI)

Isolated CMT 0/177 0.0 (0.0-2.2)

• CMT, 1+

PECARN 36/1090 3.3 (2.3-4.5) 20.9 (8.8-49.6) No CMT, 1+ PECARN 71/4103 1.7 (1.4-2.2) 11.0 (4.8-25.3) No CMT, no PECARN 6/3804 0.2 (0.0-0.3) Ref

Results: Risk difference

Exposed Unexposed Cases a b Non-cases c d

RD = Risk exposed – Risk unexposed

CMT + PECARN var PECARN var only IAI a b No IAI c d

RD = Risk CMT/PECARN – Risk PECARN variable only

Results: Risk difference

Riske Risku RD (95 CI%) p-value

• Abd. trauma

8.3% 3.0% 5% (10,21) 0.297

• Abd. tenderness

7.9% 5.1% 3% (-6,12) 0.487

Thoracic wall tr.

3.4% 1.4% 2% (-1,5) 0.063

C/o abd. pain

5.3% 2.5% 3% (-5,10) 0.336

Absent/dec BS

0% 3.4%

• 3% (-10,3)

0.673

Vomiting

0% 1.8%

• 2% (-3,-.2)

0.700

Attributable risk of IAI due to CMT = low

Results: CT scan use

• 3,947 scans of the abdomen & pelvis
• 40.7% of the population
• Isolated CMT
• 36/177 (20.3; 14.6-27.0)

Conclusions

• Isolated CMT is associated with very low

risk of IAI

• Attributable risk due to CMT for IAI above

and beyond other PECARN risk factors is likely minimal

• In the setting of isolated CMT, abdominal

CT scan for evaluation of IAI can likely be avoided

Limitations

• Missing data
• Small n, limited power
• Secondary analysis, increase type I error
• Generalizability

Thank you

Questions? flynnobr@uw.edu

Results

N Any IAI N = 507 No./No. (%; 95% CI) IAI req. acute intervention N = 113 No./No. (%; 95% CI)

Isolated CMT 177 0/177 (0.0; 0.0-2.2) 0/177 (0.0; 0.0-2.2) CMT and only 1 PECARN var.

• Abd. trauma

12 1/12 (8.3; 0.2-38.5) 1/12 (8.3; 0.2-38.5)

• Abd. tenderness

38 3/38 (7.9; 1.7-21.4) 0/38 (0.0; 0.0-9.3) Thoracic wall tr. 119 4/119 (3.4; 0.9-8.4) 1/119 (0.8; 0.0-4.6) C/o abd. pain 38 2/38 (5.3; 0.6-17.7) 0/38 (0.0; 0.0-9.3) Absent/dec BS 5 0/5 (0.0; 0.0-52.2) 0/5 (0.0; 0.0-52.2) Vomiting 8 0/8 (0.0; 0.0-36.9) 0/8 (0.0; 0.0-36.9)