The art and science of diagnosing acute appendicitis.

In this month's article by Adams et al, the early diagnosis of acute appendicitis comes full circle. Appendectomy for acute appendicitis remains one of the most common intra-abdominal surgeries performed in the United States. For Americans, the lifetime risk of developing appendicitis is 8.6% for males and 6.7% for females, and the lifetime risk of undergoing an appendectomy (all causes) is 12% for males and 23.1% for females. (1)

Traditional surgical teaching stated that it was acceptable to have a 20% negative appendectomy rate--it was expected that 20% of all patients undergoing appendectomy for acute appendicitis would have a normal appendix. This was tolerated and even expected due to the ramifications of missed appendicitis and allowing the appendix to rupture. Such ramifications include protracted hospital course, wound complications, intra-abdominal sepsis and abscesses, and even death.

However, with growing technology and medical advances, this rate of negative appendectomy is no longer the norm. Negative appendectomy rates have been reported to be less than 5% in some series, with perforated appendicitis rates reduced to less than 10%. (2) While diagnostic laparoscopy or even an open appendectomy for a normal appendix seems relatively benign, the financial costs, risk of hernia, risk of adhesions and bowel obstruction, risk of general anesthesia, and risk of unnecessary surgery accumulate. (3)

Some recent algorithms suggest that near-routine CT scans on all patients with suspected appendicitis can achieve a sensitivity, specificity, and accuracy of up to 98%. (4) Patients in these types of studies have typically already been sub-selected as likely having appendicitis. Injudicious applications of these types of studies have resulted in widespread use of CT scanning for all patients with abdominal pain and a departure from fundamental medical skills of a thorough history and physical examination.

The pain associated with appendicitis is classically periumbilical and diffuse that eventually migrates to the right lower quadrant (RLQ) over the following hours (sensitivity 81%, specificity 53%). (5) This pain is often not severe, especially in the early stages of appendicitis. While RLQ pain is one of the most sensitive signs of appendicitis, atypical pain or minimal pain will occasionally be the initial presentation.

Patients with acute appendicitis also have associated gastrointestinal symptoms that follow the onset of pain (sensitivity 100%, specificity 64%). Gastrointestinal symptoms before the onset of pain suggest a different etiology such as gastroenteritis. These symptoms include nausea (sensitivity 58%, specificity 36%), vomiting (sensitivity 51%, specificity 45%), and anorexia (sensitivity 68%, specificity 36%). Often, young men will continue to remain hungry even during the peaks of pain. (5)

Evidence of fever or elevated white blood cell count or a left shift is commonly associated with appendicitis (sensitivity 67%, specificity 79%). (5) The typical fever or white blood cell count with acute appendicitis is not severe; extremely high fever or high WBC suggests rupture or a different etiology.

On physical examination, the RLQ pain should be approached gently so as not to exacerbate the patient's pain. Conversely, patients often will not present with severe pain, rebound or guarding, especially in the early hours of acute appendicitis. In examining the abdomen, one should begin by observing the patient's routine movements, taking note of respiratory effort, distension, scars, hernias, vasculature, and color. Palpation should be initiated gently and as far as possible from the source of the pain. Initial palpation should be light touch followed by a deeper examination if the patient tolerates the softer initial exploration. Rebound should be initially checked by gentle percussion, not pushing into the abdomen and suddenly stopping. The patient should be asked to verbally express the location and level of pain during the examination.

Other signs suggestive of acute appendicitis include Rovsing sign (RLQ pain with left lower quadrant palpation), obturator sign (pain with internal rotation of the right thigh), psoas sign (pain with extension of the right thigh), heel tap or jumping on one leg, and right-sided pain with rectal examination.

Excluding other obvious pathology (especially in female patients where gynecologic history, beta human chorionic gonadotropin, and pelvic examination need to be clearly delineated), the combination of RLQ pain, gastrointestinal symptoms following the onset of pain, and fever or elevated WBC or left shift is all that is required to operate on a patient.

This article by Adams et al reminds us that radiologic imaging is only an adjunct to clinical diagnosis and the foundation begins with a thorough history and physical. Seeing and observing the patient, listening to the patient, and carefully examining the patient still plays the dominant role of being a physician. All patients with acute abdominal pain should be seen by a surgeon who then may or may not recommend a CT scan.

References

1. Addiss DG, Shaffer N, Fowler BS, et al. The epidemiology of appendicitis and appendectomy in the United States. Am J Epidemiol 1990;132:910-925.

2. Jones K, Pena AA, Dunn EL, et al. Are negative appendectomies still acceptable. Am J Surg 2004; 188:748-754.

3. Flum DR. Koepsell T. The clinical and economic correlates of misdiagnosed appendicitis: nationwide analysis. Arch Surg 2002;137:799-804.

4. Rao PM, Rhea JT, Novelline RA, et al. Effect of computed tomography on of the appendix on treatment of patients and use of hospital resources. N Engl J Med 1998;338:141-146.

5. Wagner JM, McKinney P, Carpenter JL. Does this patient have appendicitis? JAMA 1996;276:1589-1594.

Mike K. Liang, MD

From the Department of Surgery, New York University School of Medicine, Bellevue Hospital, New York, NY.

Reprint requests to Mike K. Liang, MD, Department of Surgery, New York University School of Medicine, Bellevue Hospital, 445 East 68th Street 9M, New York City, NY 10021. Email: [e-mail]

Accepted August 23, 2005.