This was a hospital based cross-sectional study carried among children who presented with chronic suppurative otitis media. The study subjects were children ≤60 months of age.

This study was conducted in Outpatient Department, Nutrition ward and Clinical microbiology Laboratory at St. Mary’s Hospital Lacor Gulu, Northern Uganda. The hospital is located approximately 6 Km along Gulu Nimule Road. Gulu City is located approximately 335 Km north of Kampala City, Uganda. Gulu City is centrally located and the biggest City in Northern Uganda. The coordinates of Lacor Hospital are 02˚46'03''N 32˚15'11''E (Latitude: 2.767500; Longitude: 32.253056). St Mary’s Hospital Lacor serves a large population of Northern Uganda and South Sudan. It therefore receives a large number of patients and it is the sentinel center for WHO in the region.

Children above 60 months of age presenting with CSOM and those who received prior antibiotics for the previous two weeks, or were on antibiotic chemotherapy during the visit were excluded from the study.

Children who were 60 months of age and below, and presented with CSOM, with no history of antibiotic therapy for the last two weeks, and/or not on treatment during the visit to the hospital were recruited in the study. The study participants who satisfied the inclusion criteria were recruited after a written informed consent was obtained from their parents/guardians.

Non-randomized purposive sampling was used to recruit study participants. We collected samples from 108 children who were confirmed to be having CSOM. We relied on only those who were sent to us in the laboratory. The total number screened in the clinic is unknown to us.

Ear pus discharge was taken from external auditory canal using sterile cotton swab after cleaning the outer ear canal using cotton wool moistened with sterile physiological saline or 70% ethanol. Two swabs were taken. The first swab was used for direct Gram staining for presumptive diagnosis, and the second swab was cultured immediately on blood agar (Mast diagnostic UK), Chocolate agar (heated blood agar) and MacConkey agar (Mast diagnostic UK). The inoculated plates were incubated at 37˚C for 24 hours. Chocolate plates were incubated at 37˚C in 5% CO₂ atmosphere for a period of 48 hours. After incubation, colonial/phenotypic characteristics such as morphology, shape of colony, swarming of colony, sugar fermentation, hemolysis of Red Blood Cells, pigment production on solid culture medium, were used for preliminary identification of bacteria. After colonial identifications, Gram stain was done to differentiate Gram positive from Gram negative bacteria. Biochemical identification test such as catalase test was used to differentiate Staphylococcus spp. from Streptococcus spp. Coagulase test was done to differentiate S. aureus from other species of Staphylococcus, Oxidase test was used to differentiate P. aeruginosa from other Gram-negative organisms. Other tests such as citrate utilization test, phenylalanine deaminase test, triple iron sugars (TSI), lysine iron agar, motility test were used to identify Gram-negative rods/bacilli.

Antimicrobial susceptibility test was performed using Kirby-Bauer disc diffusion method according to NCCLS. The following antibiotics were tested. Methicillin (10 µg), Ampicillin (10 µg), Gentamicin (100 µg), Erythromycin (5 µg), Ciprofloxacin (1 µg), Cotrimoxazole (25 µg), Chloramphenicol (10 µg) and tetracycline (10 µg). Reference strains, E. coli (ATCC 25922) for Gram negative oxidase negative organisms, P. aeruginosa (ATCC 27853) for Gram negative oxidase positive organisms and S. aureus (ATCC 25923) for Gram positive cocci were used as quality control during susceptibility testing. Each batch of culture media used in the isolation of microorganism and antibiotics sensitivity testing was first incubated overnight and those plates found with growth/contamination were discarded. All materials used were checked for dates of manufacture and expiry.

The data was entered in Microsoft excel, cleaned and analyzed using STATA version 18.0. Quantitative variables were summarized as either mean with standard deviation or median with extreme values. Chi-square and Pearson correlation tests were used to test for the relationship between variables. A p < 0.005 was taken as statistically significant, and the results were presented as figures and tables.

A total of 108 study participants who presented with CSOM were enrolled. The age ranged from 5 - 60 months, with mean age of 29.9 months, and SD of ±17.1. In all those age groups, 57.41% were male (62/108) and 42.59% were females (46/108) with a ratio of ~2:1. One hundred and seven (99.1%) of study participants visited OPD, while only 1 (0.9%) participant visited nutrition ward. All the study participants, 108 (100.0%) who presented with CSOM were ear swabbed ( Table 1 ).

OPD: Out Patient Department, IPD: Inpatient Department.

The disease was more prevalent in children aged between 13 - 24 months compared to other age groups ( Table 1 ).

Majority of the organisms isolated were Gram-negative, 78 (72.2%) and only 30 (27.8%) were Gram-positive, implying that there was a significant association between organisms isolated and their gram reaction (p < 0.001) ( Table 1 ). Eleven species of bacteria and one unidentified coliform were isolated from 108 children who presented with CSOM ( Table 1 , Figure 1 ). The most prevalent bacteria isolated were P. mirabillis (42.6%, n = 46) followed by S. aureus (19.4%, n = 21), Acinetobacter spp. (9.3%, n = 10), K. pneumoniae (7.4%, n = 8), P. aeruginosa (5.6%, n = 6), E. coli (4.6%, n = 5), Enterococcus faecalis (2.8%, n = 3), Enterobacter spp. (2.8%, n = 3), Haemophilus influenzae (1.9%, n = 2), P. vulgaris (0.9%, n = 1), S. pneumoniae (0.9%, n = 1) and unidentified coliform (0.9%, n = 1).

Methicillin was set for S. aureus only and it showed 100% (21/21) average susceptibility. There was a low level of average bacterial resistance to Optochin, 1.9% (2/108), Ciprofloxacin, 8.3% (9/108), Gentamicin, 12% (13/108) and Erythromycin, 18.5% (20/108). However, very high level of resistance was observed in Tetracycilin, 77.8% (84/108), Chloramphenicol, 73.1% (79/108), Cotrimoxazole, 64.5% (69/108) and Ampicillin, 64.5% (69/108) respectively ( Table 2 , Table 3 ).

A detailed analysis of the susceptibility profiles of Bacteria to antibiotics showed that, all the isolates of Proteus mirabilis were susceptible to ciprofloxacin (100%, n = 46), (8.7%, n = 4) of the isolates were resistant to Gentamicin and only (2.2%, n = 1) isolate was resistant to Erythromycin. However, a very high level of resistance, was observed in Chloramphenicol (82.6%, n = 38), Ampicillin (73.3%, n = 33) and Cotrimoxazole (68.9%, n = 31). Interestingly all the isolates tested with Tetracycline showed resistance (91.3%, n = 42). All the isolates of Staphylococcus aureus were susceptible to Methicillin and Ciprofloxacin (100%, n = 21), while high level resistance was noted with Tetracycline (57.1%, n = 12) and Ampicillin (47.6%, n = 10). We also noted moderate resistance to Cotrimoxazole (42.9%, n = 9), Chloramphenicol (33.3%, n = 7) and Erythromycin (42.9%, n = 9). However, only (4.8%, n = 1) bacterium resisted Gentamicin. Acinetobacter spp. Showed high level of resistance to Cotrimoxazole (90%, n = 9), Chloramphenicol and Tetracycline (80%, n = 8), respectively. Moderate resistance was seen in Ampicillin (60%, n = 6), while (40%, n = 4) of the microorganisms were resistant to Ciprofloxacin and Erythromycin respectively, and only (20%, n = 2) were resistant to Gentamicin.

In Klebsiella pneumoniae, high level of resistance was noted in Chloramphenicol and Ampicillin (87.5%, n = 7), Tetracycline and Cotrimoxazole (75.0%, n = 6), while we noted low level of resistance (25.0%, n = 2) to Ciprofloxacin and finally, all the organisms did not resist Gentamicin. There was very low level of resistance of Klebsiella pneumoniae to Ciprofloxacin (16.7%, n = 1) and (33.3%, n = 2) for Gentamicin. One hundred percent of E. coli were resistant to Tetracycline, Ampicillin and Chloramphenicol, while only one isolate resisted Ciprofloxacin and all the isolates were sensitive to Gentamicin.

Other bacteria with low prevalence also had varying susceptibility patterns. Enterococcus faecalis showed 100% (n = 3) susceptibility to Ciprofloxacin and Ampicillin respectively, However, it showed 100% (n = 3) resistance to Chloramphenicol and Co-trimoxazole. Hemophilus influenza was 100% (n = 2) susceptible to Ciprofloxacin and Gentamicin respectively, but showed 100% (n = 2) resistance to the rest of the antibiotics tested. Enterobacter species showed 100% (n = 3) susceptibility to Ciprofloxacin, while one isolate was resistant to Gentamicin. However, isolates were 100% (n = 3) resistance to the rest of the antibiotics tested. It was interesting to find that Proteus vulgaris was 100% (n = 1) resistant to Tetracycline, Chloramphenicol and Cotrimoxazole and 100% susceptible to the rest of the antibiotics tested. Streptococcus pneumoniae was susceptible to Chloramphenicol, Ampicillin, Tetracycline and Erythromycin (100%, n = 1). An identified coliform was susceptible to a majority of the antibiotics tested ( Table 3 ).