Title

Will the peripheral leukocyte count enable this adult emergency department patient with possible appendicitis to either be safely discharged home with follow-up or taken confidently to surgery, as judged by comparison of WBC values in such patients with pathology reports or prolonged follow-up without return of symptoms or need for surgery?

Reviewed by

Brian K Snyder, MD, and Stephen R Hayden, MD

Date

September 1, 1998

Expiration date

September 1, 2000

Clinical bottom line

If the threshold for probable surgery is taken as a likelihood of appendicitis of 80% or greater, and the threshold for discharge from the ED as below 10%, the total WBC count could contribute significantly to clinical decisionmaking for patients with WBC counts greater than 19.0 or less than 7.0 ×10³/μL. This would constitute about 18% of a population comprised of patients with possible, but not clinically unequivocal, acute appendicitis.

Search strategy

MEDLINE 1966–1998 via Melvyl: using MeSH term “appendicitis” AND Subheading “diagnosis,” AND “explode: leukocyte count,” limited to human studies.

Citation

Dueholm S, Bagi P, Bud M: Laboratory aid in the diagnosis of acute appendicitis: A blinded prospective trial concerning diagnostic value of leukocyte count, neutrophil differential count, and C-reactive protein. Dis Col Rectum 1989;32:855-859.

Summary of study

Population 204 adult patients, ages 15–45 years, 35% male presenting with signs/symptoms suggestive of acute appendicitis. One hundred patients went to laparotomy, with 59 having appendicitis for prevalence of the disease in the population of 29%. Thirty-three women with clinically identifiable gynecologic disease excluded. Median duration of symptoms 24 hours.

Intervention WBC count by automated counter together with other cell counts and C-reactive protein.

Outcomes Presence or absence of appendicitis judged by pathology reports or observation for return of symptoms or need for surgery during hospital stay.

Dueholm S, Bagi P, Bud M: Laboratory aid in the diagnosis of acute appendicitis: A blinded prospectivetrial concerning diagnostic value of leukocyte count, neutrophil differential count, and C-reactive protein. Dis Col Rectum 1989;32:855-859.

Guide

Comments

Are the results valid?

Was there an independent “blind” comparison with a reference standard?

Yes: The reference standard was histologic evaluation of the appendix or in-hospital observation until symptoms resolved or another disease process was identified. The result of the laboratory CBC was unlikely to have influenced the pathologist’s report.

Did the patient sample include an appropriate spectrum of patients to whom the test will be applied in clinical practice?

Yes: All patients between 15 and 45 years of age admitted with right lower quadrant abdominal pain, including all patients who were operated on and those who were observed for suspected appendicitis were included.

Did the result of the test being evaluated influence the decision to perform the reference standard?

No: The WBC count was not available to the surgeons during the entire hospital stay.

Were the methods of the test described in sufficient detail to permit replication?

The WBC counts were determined with a Technicon Hemalog cell counter.

Overall validity rating (weak–fair–good–excellent)=Good

What are the results?

What were the likelihood ratios? (calculated from data in Dueholm et al³)

LR for WBC 4–7^*=.10 (0–.39)^† LR for WBC 7–9=.52 (0–1.57) LR for WBC 9–11=.29 (0–.62) LR for WBC 11–13=2.8 (1.2–4.4) LR for WBC 13–15=1.7 (0–3.6) LR for WBC 15–17=2.8 (0–6.0) LR for WBC 17–19=3.5 (0–10) LR for WBC 19–22=∞ (NA)

Will the results improve my patient care?

Will the reproducibility of the test result and its interpretation be satisfactory in my setting?

Yes: Automated WBC count is reproducible across most hospital laboratory settings.

Are the results applicable to my patients?

Yes: The study population observed by Dueholm was very similar in important respects to the average, otherwise healthy, adult patient evaluated in an ED setting for possible appendicitis.

Will the results change my management?

Probably not. The LRs for WBC counts other than extreme values, using the interval cutoff method, are not in the range that fundamentally alters clinical assessment in an intermediate risk population. Fewer than one 1 of 5 patients in the study by Dueholm et al had extreme WBC count values.

Will patients be better off as a result of the test?

Probably not, because the LRs are so poor. Cost, risk, and patient discomfort are minimal.

^*Unit of measure=×10³/μL. ^†95% CI indicated in parentheses.