As per World Health Organization (WHO) sepsis/infection is one of the most common causes of neonatal mortality and morbidity [1]. It has been estimated that 7.6million children younger than five years of age died in 2010; of these deaths, 64% were attributed to infectious causes, and neonates contributed to a significant proportion (40.3%)[2].

Owing to the immaturity of the immune system, newborn infants are highly susceptible to systemic infection [3-6]. Studies have demonstrated a significant deficit across both innate and adaptive immunity. Neonatal adaptive immune function is hampered by deficiencies in T-cell function and B-cell function (weak immunoglobulin production) and by underdeveloped secondary lymphoid tissues [3, 6]. The innate immune system of these neonates is compromised by deficits in barrier integrity; circulating complement components; expression of antimicrobial proteins and peptides; quantitative and qualitative impairments in neutrophil, monocyte, macrophage, and dendritic cell functions; and decreased response to most Toll-like receptor agonists [5, 6].

The mortality rate can reach 60% in very low birth weight infants (VLBWI, birth weight < 1500 g) [7]. Early diagnosis, timely administration of appropriate antibiotics, and proper supportive therapy are crucial to improve survival and reduce long-term sequelae [8, 9]. Unfortunately, neonatal sepsis can progress rapidly to septic shock, occurring in 1.3% of neonates hospitalized in a neonatal intensive care unit (NICU), with an overall mortality of 40%, reaching 71% in neonates weighing less than 1000 g at the onset of sepsis [10].

Case reports published in the medical literature in the 1970s [11, 12]. reporting the effective use of exchange transfusion (ET) in severe neonatal infection with sclerema prompted some authors to use this procedure as rescue therapy in neonates with severe sepsis in subsequent years [13–15]. The rationale for the use of ET using fresh, whole, adult blood is to remove bacteria, bacterial toxins, and circulating pro-inflammatory cytokines; to improve perfusion and tissue oxygenation; to correct the plasma coagulation system; and to enhance immunological defence mechanisms (increase in circulating levels of C3, immunoglobulins, improvement in the opsonic activity against the pathogen, enhancement of neutrophil function) [16-18].

4. All neonates having severe sepsis with RDS
5. All neonates having severe sepsis with MAS
6. All neonates having weight <1000 gms

Major variable: Weight, Gestational age, Severe sepsis

Outcome variable: Mortality rate, Organ dysfunction, Duration of hospital stay, duration of antibiotic therapy, Adverse events during exchange transfusion

Data entry and analysis: Data entry was done in Excel and analysis was done using SPSS 20.0 software. Wherever, possible percentage, Chi-square test and logistic regression were applied.

Neonatal sepsis defined as any sign and symptoms of bacteremia with any two or more of the following septic screening (WBC count <5000/mm³, ANC < 1800/mm³, Immature to total neutrophil ratio >0.2, micro ESR >15 mm in 1^st hour, CRP > 10 mg/L. Signs and symptoms of bacteremia include hypothermia/hyperthermia, lethargy, poor cry, refusal to suck, poor perfusion prolonged capillary refill time, hypotonia, absent neonatal reflexes, bradycardia/tachycardia, respiratory distress, apnea, gasping respiration, hypoglycemia/hyperglycemia, metabolic acidosis.

Severe neonatal sepsis defined as neonatal sepsis with sclerema. Sclerema defined as diffuse hardening of subcutaneous tissue with minimal inflammation.

Those neonates that satisfied the inclusion criteria with severe sepsis enrolled and after obtaining written and informed consent from their parents/legal guardian, all basic demographic and clinical details recorded in a pre-designed proforma. Proforma contained information of neonates, gender, age, weight, DOB, DOA, GA, maturity, risk factors, vitals before and after exchange transfusion, investigation reports. Each neonate followed until the outcome (discharge from NICU or death).

In our study around half of the study subjects were 23 (54.76%) were preterm and the rest were term neonates. In the study two-third of 28 (66.67%) were outborn and one third were inborn admissions who underwent exchange transfusion.

Table 3 shows maternal risk factor in study subjects, 17 (40.48%) had PROM, 5 (11.90%) had > 3 vaginal exam and 3 (7.14%) each had maternal fever, foul smelling discharge and 2 (4.76%) each had uterine tenderness, hyperleukocytosis.

Table-4: Treatment outcome of study subjects

Table 4 shows the treatment outcome of stud subjects, 24 (57.14%) have died and 18(42.86%) were discharged after treatment.

Table-5: Association b/w maturity and treatment outcome of study subjects

Table 5 shows the association b/w treatment outcome and maturity of study subjects, out of 23 preterm neonates 15 (65.2%) died and 8 (34.8%) were discharged after treatment and out of 19 term neonates 9 (47.4%) died and 10 (52.6%) were discharged after treatment.

Table-6: Association b/w duration of hospital stay and treatment outcome of study subjects.

This is suggestive that early preterm premature rupture is one of the major risk factors for sepsis in children. The onset of sepsis and sclerema in study subjects: In the present study majority 30 (71.43%) had EOS and 12 (28.57%) had LOS. All the study subjects who had exchange blood transfusion developed sclerema. Pugni, L et al. (2016) in their study reported that newborns who underwent exchange transfusion 54% had early onset of sepsis and 46% had late onset of sepsis [21].

Duration of mean hospital stay days is 11.73 ±7.48 days. Those who died had a mean duration of hospital stay was 6.5 days and those who survived had 18.7 days. Similarly the median duration of days in hospital was 11 days. In our study out of 42 study subjects 24 (57.14%) have died and 8 (42.86%) were discharged after exchange transfusion treatment. Of those who died 15 (62.5%) were preterm and of those discharged 10 (55.6%) were term neonates (p=0.349). Outborn neonates more died as compare to inborn though this was also not significant (p=0.133). Neonates with exchange transfusion who had 2-5 days of antibiotic had more death 13/24 (54.16%) as compared to 6-8 days 8/24 (33.33%) and it was statistically significant (p=0.000).

Neonates who stayed for a fewer number of days in hospitals had more dying as compared to those who stayed for longer duration and this was also statistically significant (p=0.000). Morality in 2-5 days and 5-10 days of hospital duration was 100% and in more than 10 days it was 16.67%. Vain NE et al (1980) study the role of exchange transfusion in the treatment of severe septicemia. Ten critically ill newborn infants presenting with septicemia were treated with exchange transfusions. Seven of the ten (70%) infants showed immediate improvement and ultimately survived [14]. Bossi et al. (1981) study the role of exchange transfusion for severe neonatal septicemia. They treated 35 neonates with severe sepsis, 22 of them treated with ET, and 13 with standard therapy alone.

The survival rate was similar between the two groups (ET, 54.5% vs. ScT, 53.8%; p= No Significant). [22] Gross S.J. et al (1982) reported from a controlled trial, that the mortality rate was the same among those who received ET and those who did not receive it [23]. Dalvi R et al (1991) study the exchange transfusions in neonatal sepsis. The mean time for recovery following ET was 19.6 +/- 12.4 h (range: 1-48 h).

It may reduce mortality and is a safe procedure in severely sick and septic neonates [26].

In our study out of 36 culture-positive samples 15 (41.7%) were discharged after treatment and 21 (58.3%) have died. Of 36 culture positives maximum of 10 were positive for Klebsiella, 8 for E.coli, 6 each for S. aureus and P. aeruginosa, 4 were positive for Acinobactor and 2 were positive for Enterococcus. Pugni, L et al. (2016) reported that in the exchange transfusion group of neonates out of 26 culture-positive samples,5 (10.6%) were Gram-positive organisms and 21 (44.4%) were Gram-negative organisms. In Gram-positive organisms maximum were S. epidermidis and in Gram-negative organisms maximum were P. aeruginosa, E. coli and K. pneumonia [21].

In our study maximum 32 (76.19%) had cardiovascular dysfunction, 28 (66.67%) had respiratory system dysfunction and 11 (26.19%) had renal system dysfunction. In a similar study Aradhya AS et al (2015) in Chandigarh reported that 88% of cases had cardiovascular dysfunction, 56% had Hematological dysfunction and 34% had renal dysfunction [20]. Similarly Pugni, L et al. (2016) reported in their study that in (96%) cases the respiratory system has some dysfunction, in (90%) cases had CVS dysfunction and in (48%) cases had renal system dysfunction [21].

The study showed that in study subjects who underwent exchange transfusion 57.14% died and 42.86% were discharged after exchange transfusion treatment. Of those who died around two-third were preterm also, outborn neonates more died as compare to inborn though this was also not significant. Before exchange transfusion all the newborns had sclerema and that improved in 55% cases after exchange transfusion.

In the study 76.19% had cardiovascular dysfunction, 66.67% had respiratory system dysfunction and 26.19% had renal system dysfunction. On vitals except for blood pressure (p=0.000) none of the vitals i.e. temperature, PR, Spo2, Random blood sugar had a significant mean difference in their mean value before and after exchange transfusion. Neonates who stayed for a fewer number of days in hospitals had more dying as compared to those who stayed for longer duration and this was also statistically significant.