Sigmoid Diverticulitis and POCUS

 

POCUS procedure and definitions

A sonographic evaluation was performed using commercially available portable scanners with a low-frequency convex probe (2.5-6 Mhz) and a high-frequency linear probe (up to 13 or 15 Mhz). Panoramic views were performed with the convex probe searching for free fluid, collections, etc. and the linear probe, with better resolution, ...

Diverticulitis was diagnosed on POCUS based on any one of the following criteria: bowel wall thickness ≥ 4mm, pain on graded compression, pericolic fat changes (increased echogenicity), and diverticula; while abscess, fistula, the presence of free fluid and the absence of peristalsis defined complicated diverticulitis . 

A modified Hinchey classification based on POCUS findings was used:

 0 - mild diverticulitis, 

Ia - localized pericolic inflammation/phlegmon, 

Ib - localized pericolic abscess,

II - pelvic or distant abdominal abscess, 

III - purulent peritonitis, and

IV - fecal peritonitis. The primary diagnosis was categorized as uncomplicated diverticulitis, complicated diverticulitis, or other diagnoses.

Accuracy of POCUS versus CT

Using CT as the reference standard, there were 36 true positive findings, four true negatives, and five false positives where CT detected a different pathology. There were no false-negative diagnoses. POCUS sensitivity was 100% (95% CI, 90.2-100%) with 44.4% specificity (95% CI, 13.7-78.8%). POCUS had an overall accuracy of 88.8% (95% CI, 75.95-96.2%). The positive predictive value (PPV) was 87.8% and the negative predictive value (NPV) was 100%. Cohen’s kappa coefficient was calculated as 0.56, which confirms a moderate agreement between radiology and POCUS regarding the overall diagnosis.

False positives and discordant cases

The five false positive cases where POCUS had diagnosed diverticulitis, were diagnosed on CT as colitis in 3, terminal ileitis in one, and one case of diverticulosis with pancreatic cancer. For Hinchey staging, POCUS and CT agreed in 27 of 36 (75%) instances. There were nine discordant cases (25%) where POCUS underestimated the stage in five patients and overestimated it in four. 

Interval between POCUS and CT report

Between POCUS and the formal radiology CT report, the mean duration (in hours) was calculated as 9.14 hours (median 4.5 hours) with a range of up to 43.5 hours.