PubMed1987
A study was conducted to examine the usefulness of blunt probing for evaluation of penetrating abdominal trauma. Ten anesthetized adult beagles weighing 8 to 17 kg were sacrificed by ventricular fibrillation following a standardized resuscitation protocol. Immediately thereafter, the abdomen was shaved and ten stab wounds were placed in the abdomen of each dog with #11 blade. A separate individual probed the wounds sequentially with a cotton swab and predicted on the basis of this examination whether the wound penetrated the parietal peritoneum (deep) or not (superficial). The abdomen then was opened and reflected, by a midline incision, and peritoneal penetration was verified visually by two observers. Statistical analysis of data was performed using the z test for binomial proportions. Fifty-nine of the 100 wounds were found to penetrate the parietal peritoneum. Examiners predicted 35 correctly, but called 24 of 59 superficial. Thirty-eight of 41 superficial wounds were identified correctly; three were incorrectly called deep. Sensitivity was .59 (z = 1.432, P greater than .5); specificity was .93 (z = 5.466, P less than .001). Blunt probing may be a useful technique if peritoneal penetration is determined, but it is inaccurate in predicting peritoneal nonpenetration. If violation of the peritoneum is not determined with this technique, other methods must be used.