|Year : 2019 | Volume
| Issue : 4 | Page : 216-222
Predictors of mortality in outborns with neonatal sepsis: A prospective observational study
Rajkumar Motiram Meshram, Vishal S Gajimwar, Swapnil D Bhongade
Department of Paediatrics, Government Medical College, Nagpur, Maharashtra, India
|Date of Web Publication||4-Oct-2019|
Dr. Swapnil D Bhongade
Department of Paediatrics, Government Medical College, Nagpur, Maharashtra
Source of Support: None, Conflict of Interest: None
Background: Neonatal sepsis-related mortalities are the outcome of a complex interaction of maternal–foetal colonisation, transplacental immunity and physical and cellular defence mechanisms of neonates. Objective: The objective of this study was to evaluate the risk factors of mortality in outborn neonatal sepsis. Materials and Methods: A 1-year prospective observational study was undertaken at a tertiary care centre. All referred neonates with maternal and neonatal risk factors of sepsis were enrolled. Blood culture, sepsis screen and other relevant investigations were performed. Results: The mortality rate of neonatal sepsis among outborns was 38.24%. The common presentations of these neonates were respiratory distress, lethargy and hypothermia. On univariate analysis, significant risk factors for mortality included male sex (P = 0.05), weight on admission <1500 g (P < 0.001), hypothermia (P = 0.003), respiratory distress (P = 0.04), cyanosis (P = 0.001), convulsions (P = 0.02), prolonged capillary refill time (P < 0.001), thrombocytopenia (P < 0.001), abnormal radiological finding (P = 0.01), cerebrospinal fluid cellularity (P = 0.002) and positive C-reactive protein (P < 0.001). Maternal factors such as hypertension in pregnancy (P = 0.001) and antepartum haemorrhage (P = 0.03) were associated with statistically significant mortality. Gestational age (odds ratio [OR]: 0.49, confidence interval [CI]: 0.26–0.90, P = 0.02), weight on admission (OR: 1.57, CI: 1.08–2.27, P = 0.01), age at admission (OR: 0.89, CI: 0.78–0.99, P = 0.04), distance travelled with neonate (OR: 1.01, CI: 1.00–1.01, P = 0.003), duration of hospital stay (OR: 0.69, CI: 0.63–0.74, P < 0.001), hypothermia (OR: 1.87, CI: 1.01–3.42, P = 0.04), convulsion (OR: 2.88, CI: 1.33–6.20, P = 0.007), cyanosis (OR: 2.39, CI: 1.07–5.35, P = 0.03) and prolonged capillary refill time (OR: 3.34, CI: 1.78–6.24, P < 0.001) were the independent predictors of mortality in neonatal sepsis. Conclusion: Gestational age; birth weight; long distance travelled with neonate and presentation with hypothermia, cyanosis, convulsions and prolonged capillary refill time were the independent risk factors for mortality in neonatal sepsis among outborns.
Keywords: Mortality predictors, neonatal sepsis, outborn neonate
|How to cite this article:|
Meshram RM, Gajimwar VS, Bhongade SD. Predictors of mortality in outborns with neonatal sepsis: A prospective observational study. Niger Postgrad Med J 2019;26:216-22
|How to cite this URL:|
Meshram RM, Gajimwar VS, Bhongade SD. Predictors of mortality in outborns with neonatal sepsis: A prospective observational study. Niger Postgrad Med J [serial online] 2019 [cited 2019 Dec 13];26:216-22. Available from: http://www.npmj.org/text.asp?2019/26/4/216/268601
| Introduction|| |
Neonatal period is the most vulnerable time for child survival. Globally, approximately 7000 newborns die every day. In 2016, around 2.6 million deaths occurred during neonatal period. India contributes 24% of the global newborn deaths and has a neonatal mortality rate of 25.4/1000 live births with interstate and rural–urban variations., Infections (36%), prematurity (28%) and birth asphyxia (23%) are the major causes of neonatal deaths in developing countries, whereas prematurity and malformations are mainly responsible for neonatal mortality in developed countries.
Neonatal sepsis is a clinical syndrome characterised by systemic signs of circulatory compromise caused due to invasion of the bloodstream by bacteria in the 1st month of life, and conventionally, it is classified as early-onset sepsis (EOS) and late-onset sepsis (LOS) with 72 h of life as a common demarcation. Neonatal sepsis is a great masquerader that poses diagnostic challenge due to non-specific clinical manifestations and variable laboratory parameters. In addition, many other conditions can mimic sepsis, which leads to both over- and underdiagnosis and treatment and each has its own hazards. Many screening tests lack the capacity to detect specific pathogens and are unavailable at many centres in developing countries. Positive blood culture is the gold standard for the diagnosis of neonatal sepsis, but it is positive in 50%–80% at best; however, negative blood culture does not rule out the disease.,
Risk factors of neonatal sepsis include interaction of maternal–foetal colonisation, transplacental immunity and physical and cellular defence mechanisms of neonate. Most of the previous studies on neonatal sepsis have been conducted on hospital-born and neonatal intensive care-treated neonates as against the outborn neonates who have been previously admitted at a different health facility or might have been delivered at home and sometimes older at admission and susceptible to community-acquired infections. Data on such neonatal sepsis and those treated in suboptimal care at general paediatric ward are scanty.
With this background, this study was planned to evaluate the predictors of mortality in outborns with neonatal sepsis.
| Materials and Methods|| |
This prospective observational study was undertaken at one of the largest tertiary care teaching government referral hospitals of Central India over a period of 1 year. Outborn neonates are usually treated in a separate neonatal cabinet in general paediatric ward with facilities such as central oxygen pipes, phototherapy units, warmers and bubble continuous positive airway pressure machines. As per the hospital policy, neonates were treated with injectable ampicillin and gentamicin from same source before culture report is available. Specific antibiotics were subsequently commenced according to culture report sensitivity pattern.
The study was approved by the local institutional ethical committee (EC/Pharmac/GMC/NGP/960, dated 10/01/2017). We recruited all outborn referred neonates, with one or more clinical features of sepsis [Appendix 1], admitted through either outpatient or emergency department after informed valid consent from parents. Data were collected following admission, from either the mother or caregiver using a pro forma specifically designed for the study. Those who left the hospital against medical advice or not willing to participate in the study were excluded. The data extracted included maternal diseases, obstetric complications, mode of delivery and place of delivery. Socioeconomic status of the parents was classified on the basis of the Modified Kuppuswamy scale. Neonatal data, including gestational age, gender, age at admission, weight on admission, diagnosis at admission, duration of hospital stay and Apgar scores, were noted from the available referral documents.
All neonates underwent the following investigations.
With all aseptic precautions, 1-ml sample of blood was collected in a blood culture bottle containing 5–10 ml of culture media before starting antibiotic administration. All blood cultures were performed on blood agar and MacConkey's agar, and they were observed for 7 days before reported as negative.
As per the hospital protocol, sepsis screen included complete blood counts, absolute neutrophil count and C-reactive protein (CRP). Anaemia was defined as haemoglobin (Hb) level <10 g/dl, while leucopenia was defined as total blood count <5000 mm 3 and thrombocytopenia was defined as platelet count <100,000/mm 3. CRP was considered positive when it was elevated to above 3 mg/dl. Cerebrospinal fluid (CSF) examination, urine culture, renal function test, chest X-ray, abdominal X-ray, abdominal ultrasound, neurosonogram and computed tomography of brain were performed in indicated neonates.
Clinical sepsis was defined as neonates having symptoms and/or signs of sepsis [Appendix 1]. Probable sepsis was defined as clinical sepsis with a positive sepsis screen, and confirmed sepsis was defined as growth of causative organism in blood culture. EOS was defined as the clinical manifestation of sepsis appearing within 72 h of birth, whereas LOS was defined as clinical manifestation of sepsis after 72 h of birth.
All neonates were observed for clinical events and managed according to our standard protocol, and followed up to discharge or death.
The data were entered into Microsoft excel sheet, and analysis was done by using STATA version 14, developed by StataCorp 4905 Lakeway Drive, College Station, Texas 77845-4512 (USA). Continuous variables were presented as mean ± standard deviation. Categorical variables were expressed in frequency and percentages. Continuous variables were compared between survival and nonsurvival by performing independent t-test for normalised data and Mann–Whitney test for nonnormalised data. Categorical variables were compared by Chi-square test. For small numbers, Fisher's exact test was used wherever applicable. Multiple logistic regressions were performed to identify significant risk factors of mortality. Adjusted odds ratio and 95% confidence interval were calculated. P < 0.05 was considered statistically significant.
| Results|| |
A total of 1038 referred neonates were admitted during the study period, among whom 455 (43.83%) were diagnosed as clinical sepsis. The incidence of EOS was 44.40%, whereas that of LOS was 55.60%. The culture-positive sepsis was 27 (5.93%). The male-to-female ratio was 1.6:1, and 61% of the neonates belonged to parents of lower socioeconomic class from rural area. Three hundred and thirty-four (73.41%) neonates were term, 118 (25.93%) were preterm and 3 (0.66%) were post-term. Most of the neonates (96.7%) were delivered by vaginal route at primary, secondary or tertiary healthcare levels. The mean age at admission of non-survivors was 6.60 ± 7.55 days and that of surviving neonates was 8.62 ± 8.37 days, and the difference was statistically significant (P = 0.001). Similarly, the average duration of hospital stay of non-survivors was 4.68 ± 3.86 days compared to 9.67 ± 4.50 days in surviving neonates (P< 0.0001). The average distance travelled with non-surviving neonates was 97.28 ± 82.67 km longer than the surviving neonates (77.26 ± 73.57 km), and this difference was statistically significant (P = 0.007). Three hundred and thirty-six (73.85%) mothers were suffering from anaemia, followed by 60 (13.19%) with hypertension during pregnancy, while pre-eclampsia and eclampsia were the most common obstetric complications during labour [Table 1].
|Table 1: Sociodemographic, maternal and neonatal profile of outborns with sepsis|
Click here to view
Clinical and laboratory characteristics
Respiratory distress (429 [94.29%]), lethargy (378 [83.08%]) and hypothermia (294 [64.62%]) were the common clinical presentations. In thirty neonates who had suspected meningitis, CSF examination was done; abnormal cellularity was found in 21 neonates. On radiological examination, pneumonia was detected in seven neonates on chest radiograph; bronchogenic cyst, rectovaginal fistula and intestinal obstruction were found in one neonate each on ultrasound examination. Ventricular septal defect was detected in one newborn on two-dimensional echocardiography. Other important laboratory features that were recorded include neutropenia, thrombocytopenia, anaemia (Hb < 10 g %) and positive CRP among 178 (39.12%), 185 (40.66%), 154 (33.85%) and 136 (29.89%) neonates, respectively. The mean blood sugar was 97.77 ± 23.08 mg% among survivors and 102.26 ± 29.61 mg% among non-survivors, and the mean serum calcium in survival neonates was 9.11 ± 0.46 mg/dl and 9.09 ± 0.43 mg/dl in non-survival neonates, and this difference was statistically not significant [Table 2]. Microorganisms were isolated in 27 neonates (12 in EOS and 15 in LOS). Amongst the microorganisms, Escherichia More Details coli was the most common in 18 cases (66.67%); Pseudomonas in 6 cases (22.23%) and Klebsiella, Streptococcus pneumoniae and Acinetobacter were isolated in each case [Figure 1].
Risk factors of mortality
Out of 455 neonates with clinical sepsis, 174 died, giving a mortality rate of 38.24%. Mortality rate in EOS (94/202) was 46.53%, whereas that in LOS (80/253) was 31.62%. On univariate analysis, male sex (P = 0.05); preterm/weight <1500 g on admission (P< 0.001) and admission with clinical features such as hypothermia (P = 0.003), respiratory distress (P = 0.04), apnoea (P< 0.001), cyanosis (P = 0.001), prolonged capillary refill time (P< 0.001) and convulsion (P = 0.002) were the significant risk factors for mortality [Table 2]. Maternal factors such as systemic hypertension (P = 0.001), preeclampsia/eclampsia (P = 0.002) and antepartum haemorrhage (P = 0.03) contributed statistically significantly to higher mortality on univariate analysis but not on multiregression model. Other factors such as socioeconomic status of parents, residence, mode/place of delivery, maternal anaemia, pre-gestational/gestational diabetes and antepartum fever were not significantly associated with mortality. Neonates with thrombocytopenia (P = 0.0001), abnormal radiological finding (P = 0.01), cellularity in CSF (P = 0.002) and positive CRP (P ≤ 0.001) had statistically significantly higher mortality on univariate analysis. Long distance travelled with neonates, age at admission and duration of hospital stay were the independent risk factors of mortality in outborn neonatal sepsis [Table 3].
|Table 3: Risk factors for mortality in outborn neonates with sepsis (multivariate analysis)|
Click here to view
| Discussion|| |
Despite the improving maternal/neonatal health services and intensive care management, neonatal sepsis is responsible for one-quarter of the global neonatal deaths. Most of the previous studies on neonatal sepsis were mostly on inborn neonates; however, in our country, majority of childbirth are occurring at home by trained or untrained birth attendants and paramedical persons providing perinatal services at primary and secondary healthcare level. Data on such outborn neonatal sepsis, bacteriological profile and risk factors of mortality are scanty. Hence, this study was planned to evaluate the predictors of mortality in outborn neonates with sepsis in limited resource setting.
In the present study, the incidence of clinical neonatal sepsis was 43.83%. Higher rate of sepsis might be due to poor antenatal care, illiteracy, poverty and poor hygienic level at delivery by paramedical persons at primary/secondary healthcare facilities. Although the reported incidence of sepsis in inborn neonates was around 20%–25%,, similar high incidence of 39% and 49.3% was reported by Jajoo et al. and Iyer et al. from North and South India, respectively.,
Neonatal mortality rate is a reliable yardstick for evaluating the progress of perinatal services. Our hospital is the largest referral centre in Central India and as per the policy of the hospital, outborn neonates were not admitted to the special care neonatal unit, but they were treated in special neonatal cabinet in general paediatrics ward. The mortality in the present study was 38.24%. Similar to our study, high mortality (72%) was reported by Bhutta and Yusuf  and Jumah and Hassan  (44.2%), whereas lower mortality rates were reported by Bharad et al. (11.7%) and Sharma et al. (16%). Most of the studies were performed on outborn neonates, and they were treated in neonatal intensive care unit. These differences in mortality may be due to socioeconomic, geographical and racial factors. The mortality in EOS was 46.53%, and it was higher than in LOS (31.62%); this is similar to findings in other studies. Few studies reported higher mortality in LOS.
In rural India, male gender is more acceptable and with increased biological vulnerability of male neonates, we observed a male preponderance as well as significantly higher mortality, similar to the finding reported by Ogunlesi and Ogunfowora, but Trotman et al. revealed that female gender had poor outcome in neonatal sepsis. In the present study, neonates with admission weight <1500 g were associated with higher mortality; this may be due to deficiencies in humoral and cellular immunity as well as high tendencies for prolonged hospitalisation, which increases the risk of nosocomial infection. Various authors had reported preterm neonates and low-birth weight babies to have increased mortality in neonatal sepsis., Maternal factors such as premature rupture of membranes, intrapartum fever, meconium-stained amniotic fluid, foul-smelling amniotic fluid and instrumental delivery are associated with an increased chance of neonatal sepsis-related mortality, but we could not correlate a significant risk factor for mortality in our study. Hypertensive diseases during pregnancy and antepartum haemorrhage were significantly associated with mortality, owing to the high incidence of prematurity and low-birth weight births associated with these conditions.
Similar to the report of various authors, clinical features such as hypothermia, respiratory distress, cyanosis, convulsion, apnoea and prolonged capillary refill time had a significant association with mortality in neonatal sepsis, and these may be related to the occurrence of cardiovascular collapse and metabolic derangement.,,,,, Pneumonia and meningitis may be complicated with metabolic derangement and cardiovascular collapse requiring ventilatory support. Such critically sick neonates may require bedside radiological examination and ventilatory support. Unfortunately, such facilities are not available in limited resource setting, resulting in high mortality. Similar to Jajoo et al., we found high mortality in neonates with pneumonia and meningitis.
In the current study, it was seen that the mortality rate was higher when neonates had thrombocytopenia, positive CRP, CSF cellularity and abnormal radiological findings; similar findings were reported by Ahmed et al. and Jumah and Hassan, but unlike their reports, we could not demonstrate a significant correlation with anaemia, neutropenia, serum calcium and blood sugar with increased mortality. Although the isolation of bacteria is the gold standard in the diagnosis of neonatal sepsis, in a large number of cases, the blood culture remained sterile despite the presence of clinical and laboratory signs, which may be attributed to small inoculums or prior antibiotic exposure. The culture-positive rate (5.93%) in our study was very low. This might be due to the fact that most of the neonates had received antibiotics before referral at primary or secondary healthcare level. Gram-negative sepsis remains an important cause of neonatal sepsis in developing countries. Similar to Bharad et al., E. coli was the common isolate in our study followed by Pseudomonas and Klebsiella, but most of the authors had reported Klebsiella and Pseudomonas as the predominant isolates.,,
On logistic regression analysis, the predictive factors for mortality were gestational age; birth weight; age at admission; distance travelled by neonate and factors such as hypothermia, convulsion, cyanosis and prolonged capillary refill time.
| Conclusion|| |
Mortality rate in neonatal sepsis among outborns was 38.24% in the present study. Prematurity; weight <1500 g on admission; long distance travelled by neonate and hypothermia, cyanosis, convulsion and prolonged capillary refill time on admission were the independent risk factors for mortality in outborn neonatal sepsis.
- Maternal health services should be strengthened by obstetricians at primary and secondary healthcare levels for effective antenatal care and training of birth attendants in safe vaginal delivery
- Provision of safe and timely transport of neonates in ambulance with skilled workforce
- And facilities such as warmer, oxygen and portable ventilation should be made
- Governments should establish neonatal intensive care facilities for such critically sick neonates at all tertiary care centres on an urgent basis.
We are sincerely thankful to all parents who participated in the study.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| Appendix|| |
Appendix 1: One or more of the following symptoms/signs were considered as clinical sepsis:
(a) Hypothermia or fever; (b) lethargy, poor cry, refusal to suck; (c) poor perfusion and prolonged capillary refill time; (d) hypotonia and absent neonatal reflex; (e) brady/tachycardia; (f) respiratory distress, apnoea and gasping respiration; (g) hypo/hyperglycaemia; (h) metabolic acidosis; (i) history of previous admission/invasive procedure/mechanical ventilation/administration of parenteral fluids.
Specific features related to various systems:
Central nervous system: Bulging anterior fontanelle, vacant stare, high–pitched cry, excess irritability, stupor/come, seizures, neck retraction. Presence of these features should raise a clinical suspicion of meningitis.
Cardiac: Hypotension, poor perfusion and shock.
Gastrointestinal: Feed intolerance, vomiting, diarrhoea, abdominal distension, paralytic ileus and necrotising enterocolitis.
Hepatic: Hepatomegaly and direct hyperbilirubinaemia.
Renal: Acute renal failure.
Haematological: Bleeding, petechiae and purpura.
Skin changes: Multiple pustules, abscess, sclerema, mottling, umbilical redness and discharge.
With one or more of the following risk factor: (1) Febrile illness in the mother with evidence of bacterial infection within 2 weeks prior to delivery. (2) Foul-smelling and/or meconium-stained liquor. (3) Rupture of membrane >24 h. (4) Single unclean or >3 sterile vaginal examination (s) during labour. (5) Prolonged labour (sum of the 1st stage and 2nd stage of labour ≥24 h). (6) History suggestive of poor hygiene, poor cord care, bottle feeding and prelactal feeds. (7) Spontaneous prematurity (8) Perinatal asphyxia (Apgar score <4 at 1 min).
| References|| |
United Nations Inter-Agency Group for Child Mortality – UNICEF Data; 2018. Available from: https://data.unicef.org
. [Last accessed on 2019 Mar 27].
Sankar MJ, Neogi SB, Sharma J, Chauhan M, Srivastava R, Prabhakar PK, et al.
State of newborn health in India. J Perinatol 2016;36:S3-S8.
Black RE, Cousens S, Johnson HL, Lawn JE, Rudan I, Bassani DG, et al.
Global, regional, and national causes of child mortality in 2008: A systematic analysis. Lancet 2010;375:1969-87.
Edmond K, Zaidi A. New approaches to preventing, diagnosing, and treating neonatal sepsis. PLoS Med 2010;7:e1000213.
Sepsis in Newborn: Agarwal R, Deorari A, Paul V, Sankar MJ, Sachdeva A. AIIMS Protocol in Neonatology. 2nd
edition, Noble Vision, New Delhi (India); 2019. P. P. 305-15.
Upadhaya A, Aggarwal R, Kapil A, Sinch S, Paul VK, Deorari AK. Profile of neonatal sepsis in a tertiary care neonatal unit from India: a retrospective study. J Neonatol India 2006;20:50-7.
Datta S, Oberoi JK, Chgh TD. Laboratory diagnosis of neonatal sepsis. J Neonatol India 2006;20:16-23.
Gerdes JS. Diagnosis and management of bacterial infections in the neonate. Pediatr Clin North Am 2004;51:939-59, viii-ix.
Bairwa M, Rajput M, Sachdeva S. Modified Kuppuswamy's socioeconomic scale: Social researcher should include updated income criteria, 2012. Indian J Community Med 2013;38:185-6.
] [Full text]
Verma P, Sadawarte K. Neonatal septicemia: Its etiological agents and clinical associates. Indian J Child Health 2015;2:113-7.
Kumar R, Kumari A, Kumari A, Verma N. Evaluation of perinatal factors in neonatal sepsis at tertiary centre. Int J Reprod Contracept Obstet Gynecol 2017;6:4981-5.
Jajoo M, Kapoor K, Garg LK, Manchanda V, Mittal SK. To study the incidence and risk factors of early onset sepsis in an out born neonatal intensive care unit of India. J Clin Neonatol 2015;4:91-5. [Full text]
Iyer CR, Naveen G, Suma HR, Kumarguru BN, Sweta K, Janakiraman. Clinical profile and outcome of neonates with suspected sepsis form a rural medical college hospital of South India. Int J Contemp Pediatr 2018;5:55-60.
Bhutta ZA, Yusuf K. Neonatal sepsis in Karachi: Factors determining outcome and mortality. J Trop Pediatr 1997;43:65-70.
Jumah DS, Hassan MK. Predictors of mortality outcome in neonatal sepsis. MJBU 2007;25:11-8.
Bharad RV, Singh CS, Singh LR. Risk factors and immediate outcome of early onset neonatal sepsis. J Med Sci Res 2017;5:21050-6.
Sharma R, Soni TN, Rathore R, Rajput JS. A prospective study of risk factors of mortality in neonatal sepsis patients. Int J Med Res 2018;3:72-4.
Stoll BJ, Hansen N, Fanaroff AA, Wright LL, Carlo WA, Ehrenkranz RA, et al.
Late-onset sepsis in very low birth weight neonates: The experience of the NICHD neonatal research network. Pediatrics 2002;110:285-91.
Ogunlesi TA, Ogunfowora OB. Predictors of mortality in neonatal septicemia in an underresourced setting. J Natl Med Assoc 2010;102:915-21.
Trotman H, Bell Y, Thame M, Nicholson AM, Barton M. Predictors of poor outcome in neonates with bacterial sepsis admitted to the university hospital of the West Indies. West Indian Med J 2006;55:80-4.
Liang LD, Kotadia N, English L, Kissoon N, Ansermino JM, Kabakyenga J, et al.
Predictors of mortality in neonates and infants hospitalized with sepsis or serious infections in developing countries: A systematic review. Front Pediatr 2018;6:277.
Kardana IM. Incidence and factors associated with mortality of neonatal sepsis. Paediatr Indones 2011;51:144-8.
Klinger G, Levy I, Sirota L, Boyko V, Lerner-Geva L, Reichman B. Outcome of early-onset sepsis in a national cohort of very low birth weight infants. Pediatrics 2010;125:e736-40.
Ahmed A, Ahmad I, Ahmad S, Iqbal J, Din WN, Bashir Y, et al
. Hematological and biochemical predictors of mortality in neonatal sepsis. Natl J Med Res 2015;5:169-73.
[Table 1], [Table 2], [Table 3]