A recent study was conducted to evaluate intraobserver and interobserver agreement when using a 10-point scale of abnormal findings designed to standardize reporting of abdominal radiographs in neonates or infants with suspected necrotizing enterocolitis (NEC).

The study, published in AJR, demonstrates that substantial inter- and intraobserver agreement can be found when radiologists use a 10-point scale to report abnormal findings in children with suspected NEC.

The Study
All patients who underwent anteroposterior and cross-table lateral abdominal radiography for suspected necrotizing enterocolitis over a 1-year period were identified.

Inclusion criteria consisted of age <=2 months at the time of study, frontal and cross-table lateral views obtained, and clinically suspected NEC.

Subjects were selected as sample cases to exemplify the separate classifications of findings as determined on a 10-point scale (the Duke Abdominal Assessment Scale).

The scale was designed so that increasing numbers reflected increasing certainty that a patient had NEC and increasing concern regarding the severity of the patient’s disease, with 0 indicating a normal exam and 10 indicating pneumoperitoneum.

Eighty-eight cases from 49 patients were included in the study for review.

Methodology
The radiographs were reviewed as well as the most recent prior abdominal radiograph by 4 pediatric radiologists who were recruited as study participants. Each participant recorded a single score using the 10-point scale.

Each participant reviewed the cases in the same sequence. The studies were interpreted twice by all participants at least 4 weeks apart to evaluate for intraobserver agreement. Data analysis was performed by a statistician.

Results
The patient mean age in the total study population was 24.9 days. Overall, 47.3% were girls and 52.7% were boys. The 4-weighted kappa values for intraobserver agreement ranged from 0.635 to 0.946 corresponding to substantial intraobserver agreement.

The 24-weighted kappa values for interobserver agreement ranged from 0.574 to 0.898, also corresponding to substantial interobserver agreement. Reader agreement was greatest for a score of 9 and 10, indicating portal venous gas and pneumoperitoneum, respectively.

Agreement was poorest for scores of 4 and 6, indicating separation or focal thickening of bowel loops and possible pneumatosis with other abnormal findings, respectively.

[text_ad]

Conclusions
The investigation served to initiate validation of the Duke Abdominal Assessment Scale by demonstrating consistent substantial intraobserver and interobserver agreement when evaluating radiographs for suspected necrotizing enterocolitis.

Reviewer’s Comments
Although the authors allude to this also, this reviewer feels that this article lacks the correlation with the scoring system to clinical outcomes or change in management that may result depending on the scoring assigned.

They refer to the greatest inter- and intraobserver agreement with portal venous gas and pneumoperitoneum, both signs that often indicate impending surgical treatment.

However, this reviewer wonders how the scale can be helpful when the scoring falls somewhere between 2 and 6.

Thankfully for us, the authors recognize that this is only early data, and follow-up studies will be conducted

Author: Basil Hubbi, MD
Reference:
Coursey CA, Hollingsworth CL, et al. Radiologists’ Agreement When Using a 10-Point Scale to Report Abdominal Radiographic Findings of Necrotizing Enterocolitis in Neonates and Infants. AJR; 2008; 191 (July): 190-197

[text_ad]