Royal College of Surgeons in Ireland Col

1
Royal College of Surgeons in IrelandColáiste
Ríoga na Máinleá in Éirinn
A Systematic Review and Meta-Analysis of the
Alvarado Score in Predicting Acute Appendicitis
Robert Ohle, Fran OReilly, Kirsty OBrien, Tom
Fahey and Borislav D Dimitrov
2
Overview

• Introduction
• Methods
• Results
• Discrimination analysis
• Calibration analysis
• Discussion
• Summary and implications

3
Introduction
4
Background

• Acute appendicitis is the most common cause of an
  acute abdomen requiring surgery - lifetime risk
  7
• Difficult diagnosis - symptoms of appendicitis
  overlap with a number of other conditions making
  diagnosis a challenge
• Imaging can be useful, but expensive and not
  always available

5
Clinical Prediction Rules

• Clinical tool used to stratify patients according
  to their probability of having a target disorder
  or outcome of interest.
• CPRs quantify the contribution of
• History
• Examination
• Diagnostic tests
• Outcome of interest can be in terms of diagnosis,
  prognosis, referral or treatment

6
Alvarado score
Characteristic Points
Migration of pain 1
Anorexia 1
Nausea 1
Tenderness in right lower quadrant 2
Rebound pain 1
Elevated temperature 1
Leucocytosis 2
Shift of white blood cell count to the left 1
Total 10
1-4
5-6
7-10
Discharge
Observation
Surgery
7
Methods
8
Inclusion/exclusion criteria

• Population
• Emergency department/surgical ward
• Signs and symptoms suggestive of appendicitis
• Study type
• prospective and retrospective studies
• No Restrictions imposed for
• age, sex or size of study population
• language or publication status

9
Outcome measures and quality assessment

• Outcome measures and reference standards
• Histological examination
• Follow up
• No repeat admissions
• Quality assessment QUADAS

10
Statistical methods

• Discrimination analysis (Diagnostic test
  accuracy)
• Admission, 1-4 compared to 5-10
• Surgery, 1-6 compared to 7-10
• Sensitivity, specificity /- likelihood ratios
• STATA 9.1 metandi
• Calibration analysis
• The derived (index) rule was used as a predictive
  model and applied to all validation studies
• Pooled results are presented as risk ratios (RRs)
  with 95 confidence interval
• Subgroup analysis on men, woman and children

11
Results
12
Quality of included studies

• Quality of studies was good for most items
• However, many studies did not actively follow up
  patients who had a low score and were discharged

13
Discrimination analysis
    No. of studies(n) Sensitivity (95 CI) Specificity (95 CI) LR (95 CI) -LR (95 CI)
Admission Men 5 0.96 (0.88-0.99) 0.34 (0.24-0.47) 1.46 (1.24-1.74) 0.11 (0.04-0.35)
Admission Women 5 0.99 (0.92-0.99) 0.35 (0.14-0.64) 1.52 (1.01-2.28) 0.35 (0.008-0.16)
Admission Children 6 0.94 (0.70-0.99) 0.57 (0.31-0.79) 2.15 (1.24-3.75) 0.11 (0.03-0.50)
Surgery Men 6 0.88 (0.75-0.95) 0.57 (0.40-0.73) 2.09 (1.47-2.96) 0.20 (0.1-0.4)
Surgery Women 7 0.86 (0.78-0.92) 0.73 (0.56-0.84) 3.19 (1.92-5.30) 0.19 (0.11-0.32)
Surgery Children 6 0.89 (0.72-0.96) 0.74 (0.40-0.92) 3.37 (1.19-9.51) 0.15 (0.06-0.40)
14
Calibration
15
Low Risk (1-4)
16
Intermediate Risk (5-6)
17
High Risk (7-10)
18
Discussion
19
Clinical Pathway
Suspected Appendicitis
Alvarado Score
1-4
5-6
7-10
Surgical consult
CT
Discharge
20
Summary and implications

• It is a sensitive criteria for admission
• As the sole criteria for surgery, could lead to a
  high number of negative appendectomies
• Accurately predicts appendicitis in men of all
  risk strata
• Consistently over predicts in women of all risk
  strata
• Produces inconsistent results in children
• Potential clinical impact especially in low
  resource settings
• Users should be aware of its limitations in
  certain patient groups

21
Acknowledgements

• Robert Ohle and Fran OReilly
• The HRB Centre for Primary Care Research
• The Health Research Board of Ireland for funding