A Prospective study was conducted at the neonatal intensive care unit, Department of Paediatrics and Department of Microbiology, Rama Medical College, Kanpur, India, after due permission of the ethical committee of our institute. In this analysis, we analysed the data of culture and sensitivity pattern of cases of neonatal sepsis from 1st April 2011 to 31st March 2013. Neonates with clinical features of sepsis were included in the study with age 0-28 days. Details of obstetric history, maternal risk factors, and physical examination were recorded meticulously. All the cases of suspected sepsis were screened by using C-reactive protein, TLC, ANC, I/T ratio, micro ESR and blood culture. The CRP was done after 24 hours of life in case of intramural babies with risk factors (with or without symptoms) and also for extramural babies. Blood culture was done by standard microbiological techniques (BACTEC Method) in all the cases. CSF analysis as well as culture was done only in suspected cases of meningitis and the late onset sepsis. Urine examination and culture were performed only for selected cases.

Neonates with congenital anomalies, acute bilirubin encephalopathy, grade III perinatal asphyxia, neonates on antibiotics or those whose mothers have received antibiotics before delivery, were excluded from the present study. Empirical antibiotics were started after taking blood for culture and sensitivity and then changed accordingly.

A total of 364 neonates were included in the present study, 226 (62.08%) were male and 138 (37.92%) females (M: F ratio 1.63:1). Amongst them 205 (56.32%) were aged < 72 hours (early onset) and 159 (43.68%) were aged >72 hours (late onset).

All the cases were screened for sepsis; 254 (69.78%) were positive for sepsis but only 137 (37.63%) yielded positive cultures.

Among the culture positive neonates, 48 were delivered at hospital and the rest elsewhere. Culture positivity rate was high among preterm babies (47.04%) as compared to term babies (35.84%) [Table/Fig-1].

Out of these 137 bacterial isolates, 76 (55.48%) were gram negative organisms and the rest were gram positive bacteriae, mostly comparising of Staphylococcus aureus. This is followed by Klebsiella pneumoniae 37 (27.01%) & Escherichia coli 27 (19.70%). All other pathogen were responsible for less than 16% of cases [Table/Fig-2].

All isolates showed low sensitivity to ampicillin, ciprofloxacin and gentamicin, good sensitivity to cefotaxime, and maximum sensitivity to amikacin, cefepime, meropenem and vancomycin. Staphylococcus aureus was absolutely resistant to penicillin but showed 100% sensitivity to vancomycin. A good sensitivity to cefotaxime (76.47%) and amikacin (82.35%) was seen but very low sensitivity to other commonly used antibiotics [Table/Fig-3].

The klebsiella isolates were 100% sensitive to meropenem, while the cefepime as well as amikacin were also quite effective. E. coli displayed 100% sensitivity to not only meropenem but also to cefepime & amikacin. Pseudomonas aeruginosa isolated from cases were also 100% sensitive to Amikacin, Cefepime and Meropenem. The sensitivity pattern of acinetobacter and enterobacter was similar with a few minor differences. All isolates of enterobacter were sensitive to amikacin as against 66.67% of acinetobacter. Acinetobacter, however, displayed 66.67% sensitivity to gentamicin but all the isolates of enterobacter were resistant.

The streptococcal species were sensitive to most penicillins, cephalosporins, ciprofloxacin and erythromycin with a comparatively high degree of resistance to aminoglycosides (50% to 100%). The four cases of Coagulase Negative Staphylococcus (CoNS) were resistant to all antibiotics except vancomycin.

None of the gram negative organisms were resistant to meropenem and negligible resistance was seen with cefepime and amikacin. All the gram positive organisms were sensitive to vancomycin.

Blood culture has remained the gold standard for the confirmation of sepsis [7]. In our study culture positivity rate was 37.63% while that in Shah AJ et al., [8] (2012) study was 31.75%, Shaw CK et al., [9] (2007) study was 54.64%, Bhattacharjee et al., [10] study was 32%. In advanced centres, blood culture is positive in 80% of genuine sepsis [11]. Thus culture positivity rate is highly variable from place to place.

The rate of admission of early and late onset sepsis as well as the prevalence of organisms and their sensitivity patterns were much similar [Table/Fig-4]. This may be due to the fact that not only the vertical transmission but also the horizontal spread of infection may play a part in the early onset of sepsis in hospitalized neonates [12,13]. A male predominance was found in our study which is found in almost all the studies of sepsis in newborn [8,9].

The most common organism identified in our study was staphylococcus aureus (37.22%) [n=51] which is very similar to recent study of Shaw CK et al., [9]. The other gram positive organisms to be isolated were streptococcal species and coagulase negative staphylococcus. Most of the studies have found a preponderance of gram negative organisms like klebsiella, pseudomonas, and enterobacter species [14–17]. However, staphylococcus was the commonest gram positive organism to be isolated in most of the studies [14,18]. In western countries, Group B streptococcus is mainly responsible for neonatal sepsis but this is not observed in this part of the world [19]. CoNS are usually associated with indwelling catheters or central lines. In our study, CoNS was isolated only in late onset sepsis group. All of them were sensitive to vancomycin only. Similar findings were reported by a study in Nepal [9].

Klebsiella pneumoniae was the most common gram negative organism (27.01%) and the second most frequent after staphylococcus aureus in the study. This finding is not in accordance with NNPD 2002 – 03 data, where the most common organisms causing neonatal sepsis was klebsiella pneumoniae followed by staphylococcus aureus and pseudomonas [3]. In this study, culture positivity rate was found to be high among preterm neonates as compared to term (43.48% Vs.35.66%), suggesting prematurity to be a risk factor associated with neonatal sepsis. This finding was similar to the study by Monjur F et al., [20].

In our study, all the isolates were resistant to penicillin. Ampicillin, gentamicin & ciprofloxacin had lowest sensitivity to all bacterial isolates. Highest sensitivity was recorded with meropenem and vancomycin followed by amikacin and cefepime. Vancomycin and meropenem showed sensitivity of 100%. As far as cephalosporins are concerned, moderate sensitivity was observed for third generation cephalosporins i.e., cefotaxime while higher sensitivity was documented for fourth generation cephalosporins i.e. cefepime. Low sensitivity of commonly used antibiotics and fair sensitivity to amikacin was also observed by other authors [16,21,22] Tallur et al., [22] concur with us that most isolates were resistant to ampicillin, gentamicin and cotrimoxazole.

Almost all the isolates in our study were sensitive to either cefotaxime or amikacin and hence a co-prescription of these two antibiotics appear prudent as the initial choice while awaiting for the blood culture reports. This combination has given us the best results in our neonatal intensive care unit.

Neonatal sepsis is a leading cause of neonatal admissions, morbidity and mortality in developing countries. Bacterial spectrum for sepsis could be different in different regions. Sensitivity pattern also differs accordingly. The antibiotic susceptibility pattern in our study suggested that, initial empirical choice of cefotaxime in combination with amikacin was the most appropriate as maximum isolates were sensitive to either cefotaxime or amikacin. A low susceptibility to commonly used antibiotics like ampicillin and gentamicin is a cause for concern. The knowledge of prevailing strains and the antibiotic sensitivity patterns in the region is mandatory for each center due to temporal changes in the causative organisms and their antibiotic susceptibility. Periodic evaluations not only reveals the recent trend of increasing resistance to commonly used antibiotics but also helps in implementation of a rational empirical therapy.