A 7-year-old presents with a couple days of bellyache and nausea. His abdomen is mildly tender without focality, and the treating EP is not impressed. “This kid,” his intuition says, “is probably just FOS.” Can that gut instinct be trusted?
Albert Einstein wrote, “The intuitive mind is a sacred gift and the rational mind is a faithful servant.” Here, the rational mind is victorious, and the treating physician sets aside his intuition for what seems to be the safer option, getting an ultrasound.
The psychologist Kurt Koffka wrote, “It has been said: The whole is more than the sum of its parts. It is more correct to say that the whole is something else than the sum of its parts, because summing is a meaningless procedure, whereas the whole-part relationship is meaningful.” An abdominal ultrasound is quite literally the sum of its pixelated parts, but its interpretation is a whole different sum of pixelated parts and echo amplitudes. In this case, the interpretation returned with the wholly unsatisfying conclusion: “The appendix is not clearly visualized. Equivocal for acute appendicitis.”
Quoting Einstein again, “We have created a society that honors the servant and has forgotten the gift.” The doctor is chagrined—he is a servant to medicolegal considerations, and now has a tricky conversation ahead about management. “If only,” he thinks, “there was some evidence to back up my confidence in my gestalt.”
The Evidence
Clinical gestalt is the 10,000 hours or so of pattern recognition. In the ED, it starts with the eyeball test and the dichotomous sick or not sick, and progresses toward a more nuanced, top-down approach of evaluating the history and physical. Unfortunately, many clinicians believe in their gestalt but not all truly trust it, and there is a paucity of evidence-based literature to give medicolegal support. Prediction rules and algorithms are often not compared with gestalt, perhaps because such rules are devised by wholly rational minds. (Ann Emerg Med. 2017;70 [3]:338.) But one study looked at the diagnostic performance of clinical gestalt in the ED for acute appendicitis in children. (Acad Emerg Med. 2020. https://bit.ly/2Hdsy8r.)
This analysis examined 3426 patients ages 5-20 with right-sided or diffuse abdominal pain for five days or less to determine the accuracy of EP gestalt for acute appendicitis using data from a prospective study of the pediatric Appendicitis Risk Calculator. Treating physicians entered their predicted likelihood of appendicitis on a 1-100 scale prior to ordering ED imaging using a computerized research tool in the EMR. The study collected subsequent ED imaging counts and rates of acute appendicitis, negative appendectomy, and missed appendicitis within one week of the index visit.
Overall, 334 youngsters (9.8%) were diagnosed with acute appendicitis by operative and pathology reports. Physician gestalt had very good accuracy overall (C-statistic, 0.83), but overestimated incidence in the intermediate and higher-likelihood categories (e.g., in the 11-49% gestalt range). The take-home here is we are usually right when we think appendicitis is highly unlikely. Physicians five or more years out of medical school performed better than recent graduates (C-statistic, 0.84 v. 0.74), but physician and hospital characteristics were otherwise not associated with gestalt accuracy.
Discordance with Imaging
Physician gestalt performed particularly well in the very low-risk strata—one percent to 10 percent estimated risk of appendicitis—with an observed appendicitis rate of only one percent, but clinicians still obtained imaging (US and CT) in 23 percent of very low-risk patients. This finding is consistent with other studies suggesting that doctors do not always behave consistently with their gestalt, perhaps because of concerns about adverse consequences or an error of diagnostic omission. (Am J Emerg Med. 2005;23[6]:782.)
These results are bolstered by a similar C-statistic (0.84) for clinician gestalt for pediatric appendicitis in four Australian hospitals, two of them pediatric. (Emerg Med Australas. 2019;31[4]:612.) Here, however, the discordance with imaging was not as profound because CT is rarely used for pediatric abdominal pain evaluation in Australasia.
A large multicenter study suggested that none of us is sufficiently accurate to bypass our need for decision support in ruling in or out acute coronary syndrome. We tend to overcall it when it's not there and undercall it when it is. (Acad Emerg Med. 2020;27[1]:80; https://bit.ly/35Zyexs.) Gestalt, however, fared better in pulmonary embolism when used to estimate pretest probability of patients with possible PE (Chest. 2005;127[5]:1627) and when identifying ED patients with PE who might be safely managed without hospitalization. (Thromb Res. 2018;167:37.)
The legendary Dr. Benjamin Spock once advised us, “Trust yourself. You know more than you think you do.” This may not be true for all clinicians (be cautious if you are a second-year resident) or in all clinical situations (e.g., chest pain with concern for ACS), but it does seem designed for experienced clinicians who have a low index of suspicion for appendicitis in a 7-year-old.
Dr. Vinsonis an emergency physician at Kaiser Permanente Sacramento Medical Center, a chair of the KP CREST (Clinical Research on Emergency Services and Treatment) Network, and an adjunct investigator at the Kaiser Permanente Division of Research (https://www.kpcrest.net/). Dr. Ballardis an emergency physician at San Rafael Kaiser, a past chair of the KP CREST Network, and the medical director for Marin County Emergency Medical Services. Read their past articles athttp://bit.ly/EMN-MedClear.
