Relationship among peripheral leukocyte counts, etiologic agents and clinical manifestations in acute otitis media

Abstract

Objectives: To analyze the peripheral leukocyte counts of children with acute otitis media (AOM) in relation to etiology, age, clinical symptoms and signs, prior antibiotic treatment, previous AOM history and pathogen eradication during antibiotic therapy.

Patients and methods: Leukocyte counts were determined at diagnosis and on days 4 to 6 of therapy in patients age 3 to 36 months with AOM enrolled in double tympanocentesis studies. Clinical status was determined by a clinical score evaluating severity of fever, irritability and tympanic membrane redness and bulging.

Results: Of 771 enrolled patients, culture-positive middle ear fluid was reported in 590 (77%): 294 (50%) Haemophilus influenzae; 150 (25%) Streptococcus pneumoniae; 127 (21%) H. influenzae and S. pneumoniae together; 9 (2%) Moraxella catarrhalis; and 10 (2%) others. Mean leukocyte count +/- sd in patients with AOM caused by S. pneumoniae (15.7 +/- 6.7 cells x 10/mm) was significantly higher than those of patients with AOM caused by H. influenzae (13.7 +/- 5.8 cells x 10/mm) and patients with culture-negative AOM (13.8 +/- 5.6 cells x 10/mm), P < 0.01 for each comparison. Mean absolute neutrophil count (ANC) +/- sd was higher in patients with AOM caused by S. pneumoniae (8.6 +/- 5.1 cells x 10/mm) than in patients with AOM caused by H. influenzae (6.5 +/- 4.0 cells x 10/mm) or culture-negative patients (6.5 +/- 4.4 cells x 10/mm), P < 0.05 for each comparison. When multivariant regression analysis was used, S. pneumoniae etiology and fever (body temperature > or = 38 degrees Celsius) were independent factors, each significantly associated with leukocyte and ANCs. A significant correlation was found between leukocyte and ANCs and higher clinical scores in patients with pneumococcal AOM (P = 0.01, r = 0.21 and P < 0.01, r = 0.27, respectively). The mean leukocyte count on Days 4 to 6 was lower than on Day 1 (11.5 +/- 4.1 compared with 14.2 +/- 6.0 cells x 10/mm, P < 0.01). When paired (Day 1 and Days 4 to 6) examinations were analyzed, the leukocyte counts were lower on Days 4 to 6 irrespective of bacterial eradication or persistence.

Conclusions: Significantly higher leukocyte counts and ANCs were found in pneumococcal AOM than in AOM caused by H. influenzae or in culture-negative AOM. A significant decrease in leukocyte counts was found during antibiotic therapy for AOM, regardless of etiology and bacteriologic outcome. Isolation of S. pneumoniae and fever were each significantly associated with increased peripheral leukocyte and ANCs.