|
||||||||||||||||||||||||||||||||||||||||||||||
Aetiological Agents in Neonatal Nosocomial Sepsis and their Sensitivity Pattern from a Tertiary Care Hospital, Odisha, India: A Cross-sectional Study |
||||||||||||||||||||||||||||||||||||||||||||||
RK Shwetabh, Manas Ranjan Upadhyay, Rajlaxmi Upadhyay 1. Senior Resident, Department of Paediatrics, SCB Medical College and Hospital, Cuttack, Odisha, India. 2. Associate Professor, Department of Paediatrics, F M Medical College, Balasore, Odisha,India. 3. Associate Professor, Department of Pharmacology, Shri Jagannath Medical College, Puri, Odisha, India. |
||||||||||||||||||||||||||||||||||||||||||||||
Correspondence Address : RK Shwetabh, Flat No 105, Siba Plaza, Gajpati Nagar, Bhubaneswar, Odisha, India. E-mail: rk.shwetabh@gmail.com |
||||||||||||||||||||||||||||||||||||||||||||||
ABSTRACT | ||||||||||||||||||||||||||||||||||||||||||||||
: Introduction: Nosocomial infections are a major problem for hospitalised neonates due to increase in morbidity, mortality, duration of hospitalisation and costs of treatment. The magnitude of this problem varies from place to place and is unique to each place as per the organisms and their resistance pattern is concerned. There is need to develop local level surveillance data on incidence of nosocomial sepsis, causative organisms, their sensitivity pattern and periodically review antibiotic policy based on this information. Aim: To determine the aetiological agents of nosocomial sepsis and their antibiotic sensitivity and resistance pattern. Materials and Methods: This was a cross-sectional study in which the cases of nosocomial infections in neonates of > 35 weeks were studied from November 2018-October 2019 at Sriram Chandra Bhanja Medical College and Hospital (S.C.B.M.C.H) and Sardar Vallabh Bhai Patel Postgraduate Institute of Paediatrics (S.V.P.P.G.I.P) based on clinical findings, sepsis screen and blood culture. Blood culture is considered as gold standard for diagnosis of sepsis. Blood sample (1 mL) was collected under strict asepsis in Becton Dickinson and Company (BACTEC) and was sent for performing blood culture. The data was processed and arranged into distribution tables and cross tables using Statistical Package for the Social Sciences (SPSS) version 21.0. Results: Out of total 100 suspected cases of nosocomial sepsis, blood culture was positive in 46 (46%) of cases. Candida spp. was the most common obtained organism 9 (19.5%) followed by Klebsiella pneumoniae 8 (17.3%) and Staphylococcus aureus 8 (17.3%) each respectively. There was increased incidence of bacterial resistance to commonly used antibiotics and combinations of it, like ampicillin, gentamicin, cefotaxime, amikacin and piperacillin+ tazobactam among these bacterial isolates. Among the possible new combinations deduced from the observation, the combination of vancomycin+amikacin had sensitivity of 67.6% and can be considered as initial antibiotic combination of choice while combination with colistin and tigecycline should be reserved only for culture proven resistant cases or babies who continue to be deteriorating and critically sick while on previous combination. Conclusion: Candida spp. is the leading cause of nosocomial sepsis. Among bacterial organisms, Klebsiella pneumoniae and Staphylococcus aureus are most common. Most isolates are resistant to traditional antibiotics, hence new combination like vancomycin+amikacin are more appropriate empiric choice in present context and combination with colistin and tigecycline are reserved only for culture proven resistant sepsis. | ||||||||||||||||||||||||||||||||||||||||||||||
Keywords : Antibiotic combination, Blood culture, Nosocomial sepsis | ||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||
DOI and Others :
DOI: 10.7860/IJNMR/2022/55647.2351
Date of Submission: Feb 13, 2022 Date of Peer Review: Mar 22, 2022 Date of Acceptance: May 30, 2022 Date of Publishing: Sep 30, 2022 AUTHOR DECLARATION: • Financial or Other Competing Interests: None • Was Ethics Committee Approval obtained for this study? Yes • Was informed consent obtained from the subjects involved in the study? Yes • For any images presented appropriate consent has been obtained from the subjects. No PLAGIARISM CHECKING METH |
||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||
INTRODUCTION | ||||||||||||||||||||||||||||||||||||||||||||||
Neonatal sepsis represents a significant cause of neonatal mortality and long-term morbidity. There are an estimated 1.3 -3.9 million annual neonatal sepsis cases and 4,00,000 to 7,00,000 annual deaths worldwide due to sepsis (1). Severe neonatal infections (including sepsis, meningitis and pneumonia) represent a significant cause of neonatal mortality (24%) and cause short-term and long-term complications, such as preterm birth and neonatal encephalopathy. An estimated 84% of neonatal deaths due to infections could be prevented through measures such as early diagnosis and timely appropriate clinical management. Among hospital born infants, nosocomial infections account for an estimated 4%-56% of all deaths in the neonatal period, depending on the study and geographical area (1). Neonatal sepsis is a clinical syndrome characterised by signs and symptoms of infection with or without accompanying bacteremia in the first month of life. It encompasses various systemic infections of newborn such as septicemia, meningitis, pneumonia, arthritis, osteomyelitis and urinary tract infections. Sepsis in neonates has been broadly classified in two groups-Early onset sepsis (onset of sepsis within 72 hours of birth) and late onset sepsis (onset of sepsis after 72 hours of birth). Late onset sepsis has been again divided into two groups-community acquired and nosocomial sepsis (2). Nosocomial sepsis is defined as any infection causing illness that was not present, or in its incubation period during the time of admission, and which occurs 48 hours after admission to hospital. The risk of nosocomial infection in neonates is the direct consequence of the severity of illness, prematurity, congenital defects, systemic diseases, level of invasive monitoring, indiscriminate use of antibiotics, lapses in sterilisation and disinfection techniques and the nature of diagnostic procedures. The profile of these infections keeps on changing and varies from place to place (3). Blood culture is considered as gold standard for diagnosis of sepsis. The antibiotic susceptibility pattern obtained from blood culture can guide us about appropriate antibiotics which are to be administered. New techniques like BACTEC (Becton Dickinson and Company and BACT/ALERT (bioMerieux) blood culture systems can detect bacterial growth even within 12-24 hours (2). In India very few studies like Kamath S et al., Pathak S et al., etc. have been done previously on nosocomial sepsis but no study have been done about appropriate empirical combination of antibiotics for nosocomial sepsis (3),(4). Hence the present study was done to determine the organisms causing nosocomial sepsis, their antibiotic sensitivity pattern and the best combination of antibiotics for empiric therapy in a tertiary care hospital. | ||||||||||||||||||||||||||||||||||||||||||||||
Material and Methods | ||||||||||||||||||||||||||||||||||||||||||||||
It was a cross-sectional study done in (Neonatal Intensive Care Unit (NICU) and its step down at Sriram Chandra Bhanja Medical College and Hospital and Sardar Vallabh bhai Patel Postgraduate Institute of Paediatrics (S.V.P.P.G.I.P), Cuttack from November 2018 to October 2019 after approval from Ethical Committee (Ethics committee Reg No-ECR/84/Inst/OR/2013/RR-20: application no. 22, dated 7/2/2020). Inclusion criteria: A total of 100 neonates of ?35 weeks who developed new infections while staying for at least 48 hours in hospital or those neonates whose signs or symptoms persisted or worsened even after receiving at least 72 hours of antibiotics were included in the study. Exclusion criteria: Neonates having TORCH infection (Toxoplasomosis, Others like syphilis and hepatitis B, Rubella,Cytomegalovirus, Herpes simplex) or congenital anomaly were excluded from the study. Sample size calculation: Sample size was calculated by the formula: Sample size (n) = 4pq/d2, Where, p=Prevalence; q=100-prevelance; d=0.005 (precision level) Taking into account the study done by Hoseini M et al., on nosocomial sepsis, prevalence rate of nosocomial sepsis was 6.7% (5). Hence, the calculation: n = 4pq/d2 Here, p=0.066 Hence, q=1-0.066=0.934; d=0.05; n = 4*0.066*0.934/0.05*0.05 =98.6 So, minimum 98 samples were required. The present study comprised of 100 participants. Informed consent was taken from parents. Detailed history was taken, vital signs were monitored and detailed examination was performed initially and repeated every day. Sepsis screen and blood culture were done before starting or changing antibiotics. Blood Culture Blood sample was collected under strict asepsis. After wearing sterile gloves a patch of skin of approximately 5 cm in diameter was disinfected, over the proposed venipuncture site with 70% isopropyl alcohol, followed by povidone iodine and again followed by alcohol. Povidone iodine was applied in concentric circles moving outward from the centre. The skin was allowed to dry for at least one minute before sample collection. Blood sample (1 mL) of blood was collected in BACTEC and sent for culture (2). Antibiotic combination Antibiotic combination was taken keeping in view covering both gram positive and gram negative organisms (6) (Table/Fig 1). The interpretation was done in following way. Hence, if an organism was sensitive to at least any one of the antibiotic the combination was taken to be sensitive, which was logical also as combination was given to the patient as well. On the other hand if the organism was resistant to both the antibiotics than the combination was taken to be resistant. Statistical Analysis The data were processed and arranged into distribution tables and cross tables using SPSS version 21.0. | ||||||||||||||||||||||||||||||||||||||||||||||
Results | ||||||||||||||||||||||||||||||||||||||||||||||
Total of 100 cases were enrolled, out of which 67 (67%) were males. Positive blood culture was found in 46% cases. Among 100 cases taken in the present study 32 (32%) cases were of gestational age 35-37 weeks, 67 (67%) were between 37-42 week and 1 (1%) case was post-term (>42 weeks) (Table/Fig 2). Most common sign/symptom for changing/adding antibiotic was poor feeding/lethargy 43 (43%) followed by respiratory distress 25 (25%) (Table/Fig 3). Out of 46 isolates 9 (19.5%) were fungal (Candida spp.) and 37 (80.5%) were bacterial, out of which 20 (54.1%) were gram positive and 17 (45.9%) were gram negative. Out of total 37 bacterial isolates, Staphylococcus aureus 8 (21.6%), Klebsiella pneumoniae 8 (21.6%), Acinetobacter baumannii 7(18.9%) followed by Staphylococcus haemolyticus with 6 (16.2%) were important bacterial isolates. Klebsiella pneumoniae was the most common gram negative isolate followed by Acinetobacter baumannii while Staphylococcus aureus and Staphylococcal haemolyticus lead in gram positive isolates (Table/Fig 4). Candida species obtained in the present study were 100% sensitive to voriconazole, flucytosine and amphotericin B. The sensitivity with capsofungin and micafungin was 88.8% while sensitivity to fluconazole was 66.7% (Table/Fig 5). Antibiotic sensitivity among gram positive organisms was tigecycline (100%) followed by vancomycin (89.5%), cotrimoxazole (87.5%), colistin and amikacin (80%) each, linezolid (77.7%), teicoplanin (65%), ciprofloxacin (30.7%) and for gram negative organisms was colistin (87.5%) followed by tigecycline (75%), amikacin (35.2%), gentamicin (18.75%) and ciprofloxacin (13.3%) (Table/Fig 6). Overall vancomycin (89.5%) had maximum sensitivity, followed by colistin (85.7%), tigecycline (84%), linezolid (77.7%) and teicoplanin (65%) (Table/Fig 6). Overall, the combination of teicoplanin+colistin, tigecycline+vancomycin, colistin+vancomycin, tigecycline+colistin and colistin+linezolid had sensitivity of 82.8%, 85.7%,88.8%, 90% and 93.75% respectively. Even when four most important bacterial isolates are taken (Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii and Staphylococcus haemolyticus), which constitute 78.3%(29/37) of all bacterial isolates then also the sensitivity to teicoplanin+ colistin, tigecycline+ vancomycin, colistin+ vancomycin, tigecycline+colistin and colistin+linezolid were at 79.3%, 81.5%, 89.2%, 86.9% and 92% respectively, which is comparable to the overall sensitivity to all bacterial isolates taken together (Table/Fig 7). | ||||||||||||||||||||||||||||||||||||||||||||||
Discussion | ||||||||||||||||||||||||||||||||||||||||||||||
Nosocomial sepsis is a serious problem especially in developing countries like India. According to World Health Organization (WHO), the estimated incidence of hospital acquired sepsis in neonates was 112.9 cases per 1000 Intensive Care Unt (ICU) treated neonates (1). Among hospital born infants, hospital acquired infections account for an estimated 4% to 56% of all deaths in the neonatal period, depending on the study and geographical area investigated (1). Neonatal nosocomial sepsis account for 56.6% of all the nosocomial infections among all the age groups (neonates, pediatric and adults) (1). In the present study for nosocomial infections, fungal (Candida spp.) sepsis was present in significant number of cases 9 (19.5%), compared to a lower incidence noted by Hosseini M et al., at 10.1% and Lopez Sastre JB et al., at 12% (5),(7). In the present context fungal sepsis should be an important etiology of nosocomial sepsis in hospital setting like ours. Among the fungal isolates no resistance was found against Amphotericin B but one-third of isolates were resistant to Fluconazole, which is in contrast to the observation made by Caggiano G et al., where sensitivity to both the antifungal stand at 100% (8). Fluconazole was the 1st line antifungal and mostly used in Very Low Birth Weight (VLBW) infants, as a prophylactic and many cases as empiric choice as antifungal, so the present observation emphasises the need to avoid empiric use of antifungals and restrict use of antifungal drugs in culture positive fungal sepsis. Among bacterial isolates (37) Staphylococcus aureus (21.6%), Klebsiella pneumonia (21.6%), Acinetobacter baumannii (18.9%) and Staphylococcus haemolyticus (16.2%) were common.Similar observation was made by Kamath S et al., (3) where Staphylococcus aureus (12.3%) and Klebsiella pneumoniae (16.4%) were commonly isolated (3). In a similar study Pathak S et al., also observed Staphylococcus aureus (26%) and Klebsiella pneumonia (33.3%) followed by Acinetobacter baumannii (13.3%) to be common gram positive and gram negative isolates (4). The sensitivity of Klebsiella pneumoniae to colistin was 85.7% and similar to observation in all type of neonatal sepsis i.e 88.8% and 91.67% by Pokhrel B et al., And Hashmi MA et al., respectively (9),(10). Sensitivity to amikacin, gentamicin, meropenem and piperacillin+tazobactam was just 50%, 37.5%, 16.4% and 14.2% in the present study, suggesting increasing resistance to the commonly used antibiotics. The study done by Saritha Kamath S et al., and Pathak S et al., suggested sensitivity of Klebsiella pneumoniae to amikacin 55.6% and 70% respectively while that to gentamicin was found to be 16.7% and 10% respectively (3),(4) (Table/Fig 8)a. Acinetobacter baumannii was sensitive to tigecycline in 100% cases, followed by Colistin in 85.7%. The study conducted by Pokhrel B et al., and Hashmi MA et al., found the sensitivity of Acinetobacter baumannii to colistin as 80% and 100% respectively (9),(10). Pokhrel B et al., also found the sensitivity to tigecycline as 66.7% while Abdelhamiod SM, found it to be 75% (11). All these difference is due to different levels of use and misuse of these antibiotics. In study done by Kamath S et al., the sensitivity was observed to be 100% with amikacin and 95.5% with cotrimoxazole while that to gentamicin was 22.7% (3). However in the present study the sensitivity to amikacin and cotrimoxazole was just 14.2% each and there was 100%resistance to gentamicin (Table/Fig 8)b. For gram negative organisms sensitivity was maximum with colistin (87.5%) followed by tigecycline (75%) in the present study. Comparing it with other studies, Pokherl B et al., and Abdelhamid SM, found sensitivity of gram negative organism against tigecycline to be 85.7% and 86.7% respectively (9),(11). Pokhrel B et al., also found sensitivity of gram negative organisms against colistin to be 82.4% (9). Amikacin, gentamicin, cefotaxime, meropenem and ciprofloxacin were only 35.2%, 18.75%,18.75%,14.3% and 13.3% sensitive respectively in the present study. Hosseini M et al., concluded from their study that ciprofloxacin (85.7%) and chloramphenicol (84.6%) are more sensitive against gram negative bacteria while sensitivity for gentamicin was just 18.1% (5). Hasmi et al., found sensitivity of gram negative organisms to Colistin and Meropenem to be 96.55% and 28.6% respectively (10). The sensitivity of Staphylococcus aureus to vancomycin is 85.7% in the present study while in other studies done by Kamath S et al., Hosseini M et al., and Pathak S et al., it was 100% each (3),(4),(5) (Table/Fig 8)c.Two-third of staphylococcus aureus were resistant to linezolid while sensitivity to amikacin was 80%.Comparing it to study done by Dalal P et al., and Hashmi MA et al., the sensitivity of staphylococcus aureus to linezolid was 87% and 88.8% respectively (10),(12). Though in the current hospital teicoplanin is not commonly used still we observed high degree of resistance to teicoplanin (62.5% resistance) as compared to Kamath S et al., and Dalal P et al., where they found no resistance to Teicoplanin at all (3),(12) (Table/Fig 8)c. It was observed that the sensitivity of stahylococcus haemolyticus to be 100% against cotrimoxazole, linezolid, tigecycline and vancomycin, while it was 83.3% against teicoplanin. Tessema B et al., in their study of neonatal sepsis observed sensitivity of staphylococcus haemolyticus to be 100% against vancomycin and 93.3% against linezolid and 73.3% against teicoplanin (13) (Table/Fig 8)d. Sensitivity among gram positive organisms was tigecycline (100%), followed by vancomycin (89.5%), cotrimoxazole (87.5%), colistin and amikacin (80%) each, linezolid (77.7%), teicoplanin (65%) but ciprofloxacin and cefotaxime has just 30.7% and 20% sensitivity. Similarly, Hosseini M et al., observed that most sensitive antibiotics against gram positive bacteria was vancomycin (94.1%). Pokhrel B et al., observed sensitivity of vancomycin and linezolid against gram positive bacteria to be 100% each (5),(8). Hasmi M et al., (11) also observed sensitivity of gram positive bacteria against vancomycin and linezolid to be 83.78% and 81.08% respectively (10). Dalal P et al., and Li X et al., found sensitivity to teicoplanin as 100% and 98.9% respectively while Li C et al., also found sensitivity to tigecycline be 100% against gram positive organisms (14). Considering lack of isolation of causative agents in a significant number of patients and the inherent delay in getting the culture report, sick patients were to be started empiric antibiotic. With the present resistant pattern observed in both gram positive and gram negative organisms its necessary to go for more appropriate antibiotic combination to improve survival. For empiric antibiotic combination traditional antibiotic combination are losing their importance with combinations like piperacillin+tazobactam+gentamicin, ampicillin+gentamicin, cefotaxime+gentamicin and cefotaxime+ amikacin having sensitivity of 24.2%, 28.6% ,31.2% and 50% only. The new empiric combination which can be used in desperately sick neonates are teicoplanin+ colistin, tigecycline+vancomycin, tigecycline+colistin, colistin+ vancomycin, and colistin+linezolid having sensitivity of 80.5%, 85.7%, 87.1%, 88.8% and 91.1% respectively. Comparing it to other studies, Al-Mouqdad MM et al., found a combination of amikacin + cloxacillin to be a better empiric combination for late onset sepsis (15). Sivanandan S et al., had suggested a combination of cloxacillin+gentamicin as empiric antibiotic combination for late onset sepsis, but has also recommended a combination of vancomycin+cefotaxime to be given empirically to neonates with cardio-respiratory instability and in the areas where MRSA (Methicillin Resistant Staphylococcus Aureus) is prevalent (6). Even if four most common bacterial isolates are taken (Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii and Staphylococcus haemolyticus), then the sensitivity to teicoplanin+colistin, tigecycline+vancomycin, colistin+vancomycin, tigecycline+colistin and colistin+linezolid is 79.3%, 81.5%, 89.2%, 86.9% and 88.4% respectively, which is comparable to the overall sensitivity. And, finally the combination of vancomycin+amikacin has sensitivity of 67.6% and hence can be considered as 1st line empiric therapy. Resistance is a continuous phenomenon however antibiotic stewardship is also necessary to be taken care of, so it is important to follow a fixed infection preventive strategy and use the appropriate combination of antibiotics with minimum side effects for the treatment of nosocomial sepsis and de-escalate or change the antibiotics after getting culture report. Limitation(s) The present study had the limitation that extremely preterm neonates were not taken into account and long-term follow-up was not done. | ||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||
Original article / research
|