In this study, out off 122 paticipants, 56.6% were male and 43.4% female. The male were predominant which agrees with previous reports [25]. The blood culture positivity rate was 59.0%, this was a high blood culture-positivity rate as comparable to other findings [26]. The high prevalence could have been due to the fact that the study site (KILEMBE MINES HOSPITAL ) was the only general hospital around Kasese town, most frequently receiving neonates with complications as well as complicated pregnancies than surrounding low level health facilities.
56.6% of the participants presented with early onset septicaemia and 43.4% with late onset septicaemia which agrees with the high prevalence reported by Islam [27] and [28]. However, a study conducted at Mbarara regional referral hospital [29] indicated EOS of 24% (19/80 neonates) and LOS of 21.3 (7/80 neonates) with blood culture positivity of 32.5% (26/80 neonates). In our study, the positivity rates amongst neonates that presented with EOS and LOS were 41(56.9%) and 31(43.1%) respectively, this could have been due to infections asecending from the perineum of the mother or due to poor infection control during the delivery process. This was higher in male (55%) than female (44.4%) as also reported in other studies [30].
Of the 11 etiological agents identified, GBS ie Streptococcus agalactiae (21%) was the most common amongst the neonates followed by S. aureus 19%, Klebsiella pneumoniae (14%), (Escherichia coli (11%), Acinetobacter spp (11%), Enterobacter aerogenes (10%), Enterobacter aerogenes (7%), Citrobacter freundii (1%), Viridans streptococci (1%), Proteus mirabilis (1%) and Enterococcus.spp.(1%). This was contrary to a study by Maimoona [31] who reported most common pathogens as Klebsiella pneumoniae (35%), followed by Staphylococcus aureus (24.1%). The difference could be due to difference in health care systems, population studied, diagnosis criteria and the case definition between the study sites [13].
Gram-negative and gram-positive septicaemia was encountered in 56.9%(41) and 43.1%(31) of the culture positive cases in this study respectively, which was comparable to a study conducted by Gupta [32] and other studies where gram-negative and gram-positive organisms were responsible for 59% and 41% of the septicaemia cases, respectively as observed by Mugalu [4]. Gram negative organisms 41 (56.9%) were most implicated with neonatal septicaemia. This was also reported in the previous study [26]. However, this was contrary to a study conducted at Mulago hospital which indicated that gram positive organisms were predominant (69.2%) [4].
Gram negative agents most responsible for neonatal septicaemia were Klebsiella pneumoniae 10(24.4%), Escherichia coli 8(19.5%) as reported in other findings [33], Acinetobacter spp 8(19.5%), Enterobacter aerogenes 7(17.1%), Pseudomonas aeruginosa 6(14.6), Citrobacter freundii 1(2.4%) and Proteus mirabilis 1(2.4%). Klebsiella pneumoniae was the predominant isolate (24.4%) among the gram-negative pathogens which correlates with other findings [34]. However, this was contrary to a study which reported Acinetobacterspp (9.5%) as the most predominant gram negative organism followed by Klebsiella pneumoniae (7.7%) [2]. The difference could have been due to changes in causative agents of neonatal septicaemia over time and may vary from place to place [35].
This study r that out of the 31(43.1%) gram-positive organisms identified, majority of these were Streptococcus agalactiae 15(48.4%) as also reported by Nuorti [36] as the leading cause of invasive bacterial infections in newborn babies followed by Staphylococcus aureus 14 (45.2%), Enterococcus.spp. 1(3.2%) and Viridans streptococci1(3.2%).
From the analysis of drug susceptibility profiles according to the WHO recommended first and second-line antibiotics, our study showed that among gram-negative isolates, majority of the isolates (92.9%) were resistant to ampicillin, Cefoxitin (76.7%), Cotrimoxazole (70.6%), Ceftriaxone (64.3%), Netilmicin (59.5%), Gentamicin (58.9%), Amikacin(53.3), Amoxyl/clavulinic acid (46.9%), Cefotaxime (45.5%) and Linezolid (33.3%). The least resistance was observed to imipenem (25.5%) as seen in other studies [37]. Among gram-positive isolates, high resistance was observed to ampicillin (100%) similarly to a study by Mustafa [38], Gentamicin (80.9%), Ceftriaxone (72.3%), Cotrimoxazole (69.1%), Amoxyl/clavulinic acid (50.9%), Cefoxitin (41.9%), Amikacin (25.4%). There was no resistance of Streptococcus agalactiae to Vancomycin as also reported by other studies [39]. Overall, the least resistance was to Netilmicin (8.3%) followed by Linezolid (18.6%) and Cefotaxime (22.4%). Of the aminoglycosides used, amikacin (46.5%), exhibited a verge sensitivity over netilmicin (36.4%) and gentamicin (41.0%) against gram negative organisms as observed in other studies [25].
Our study revealed that Staphylococcus aureus was more sensitive to netilmicin (100%) contraly to a study by Peterside [40] where ciprofloxacin was 90.9% effective. However, a study conducted by Lamba agrees to our study that gram-positive isolates that include Staphylococcus aureus have good sensitivity to linezolid and vancomycin [41]. Enterococcus.spp.were equally sensitive to amoxycalvulinic acid, amikacin, linezolid, cefoxitin, vancomycin and netilmicin i.e., 100% dispite its resistance especially when the organisms are in large numbers as reported in other studies [42].
Of the gram positive isolates, imipenem was found to be more effective to Enterobacter aerogene, Pseudomonas aeruginosa and Acinetobacter spp. This agrees with other findings [43]. Different studies [25] agree with the findings of our study indicating that imepenem had the overall best sensitivity (74.5%) among gram-negative organisms.
Maternal factors associated neonatal septicaemia found in this study were PROM and UTI during pregnancy. In this study, neonates born to mothers with these factors were more likely to develop septicaemia. This is consistent in earlier studies conducted in different parts of the world.[44]. Mothers with early PROM and prolonged labor had increased chances of microorganisms ascending from the birth canal into the amniotic sac which could cause fetal compromise as well as septicaemia during the neonatal period. This also explains the rationale for giving prophylactic antibiotic therapy to neonates born to mothers with a history of PROM during pregnancy which could increase chances of antimicrobial resistance. [45]. SVD was also associated with neonatal septicaemia, here babies could have been exposed to maternal vaginal and fecal bacteria [46]. which was contrary to other studies that showed cesarean section was more associated with culture-positive cases [47]. This justifies the need for infection control practices and improving mother hygiene as reported by Ahmed [48]. In other studies, mothers who attended ANC late were not associated with neonatal septicaemia [4], this was contrary to our study where mothers who had ANC attendance of more than 4 times were more associated with neonatal septicaemia, though ANC utilization is vital in reducing the risk factors to neonatal septicaemia but that was not the case in our study and this could have been due to over crowding at ANC (that handled both ANC and postnatal services), use of only one toilet for all the out patients ie where mothers could have contracted infections and exposing their babies, using only one weiging scale for all the babies without decontamination or washing hands between babies. Foul Smelling liquor was also associated with neonatal septicaemia as similarly to other studies [49].
Our findings revealed that Apgarscore in the 1st minute and 5th minutes of ≥6 were highly associated with neonatal septicaemia which was not the case for a study conducted by Abdulhakeem [50]. Since majority of the neonates in our study had adaptated well to extra uterine life without much stress experienced during labour, association of Apgarscore ≥6 to septicaemia was likely to be due to the fact that when babies are in good health, a lot of people want to touch or carry the baby little knowing that they are exposing the baby to infections. Also unhygienic practices and not following guidelines by health workers when handling babies could expose neonates to infections [51]. Neonatal birth weight of ≥2500g and resustation of newborn babies were greatly associated with septicaemia. Similar findings were also observed in other previous studies in Ghana [51]. This could have been nosocomial infections or use of non-sterile equipments during resustation.
Mother's occupation (employed) and urban residence also had an influence on neonatal septicaemia. This was contrary to other findings [50] where mother’s occupation status and urban residence were not found to be predictors of septicaemia. This could have been attributed to nosocomial infections, poor infection prevention control measures or congested wards.
Limitations
i. Our study size and study period were not enough to yield statistically precise estimates for most of the less common etiological agents for septicaemia.
ii. We might have underestimated the proportion of neonates with septicaemia because blood culture itself has a poor sensitivity especial with the small volumes of blood that were collected from the neonates [52].