الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 208 |  |  |
| | | from | 208 | | |
Abstract
Sepsis is still one of the most important causes of mortality and morbidity in neonates. Diagnosis of sepsis is not an easy issue. The early signs of sepsis could be inconclusive and differ according to the gestational age of the newborns. The diagnosis of sepsis depending on the blood culture usually delayed. Recent studies focused on early diagnosis of sepsis by detection of early sepsis biomarkers that could be detected at time of initial assessment. But studies that focus on detection of early response to given antibiotics and outcome of sepsis are few. Studies that focus on the role of lactate in prediction of outcome in neonatal sepsis are even fewer. As many studies relied on lactate in prediction of tissue recovery after hypoxia or trauma.
The aim of this study is to evaluate the ability of serial neonatal sepsis and hence the early prediction of the effectiveness of the management used with these patients.
So our study was conducted over 70 newborns with suspected or proven sepsis admitted to NICU of Menoufia University hospital according to hematological and clinical sepsis score. Patients are classified into 2 groups according to the outcome; group 1(37 survivors) and group 2(33 non survivors).
Each neonate was subjected to:
1) History taking: which included prenatal, natal and postnatal events
2) Thorough clinical examination: that included general and systematic examination.
3) Laboratory investigations: that included (CBC, blood gases, blood glucose, liver and kidney functions, serum albumin and blood culture). Specific investigations including (CRP, initial lactate level and lactate clearance).
Initial serum lactate levels showed a highly significant difference (p value<0.001) between the two groups as higher levels (48.4mg/dl±16.8SD) were noticed in group 2 (the non survivors) compared to group 1(28.5mg/dl±7.4SD).
There was a highly significant difference between the two groups as regards to the delayed lactate serum levels which were significantly higher in the non survivors (46.3mg/dl±13.1SD) when compared to the survivors (17.7mg/dl±6.03SD) with p value≤0.001.
In our study 24 hours lactate clearance % showed highly significant difference (p value˂0.001) between the two groups. The non survivors showed lower levels of lactate clearance in 24 hours (8.1%±11.2SD) when compared the survivors (52.7%±98.1SD).
Initial CRP levels withdrawn at clinical diagnosis or suspicion of sepsis showed a significant difference (p value= 0.02) between the two groups as higher values observed in the non survivors (42.1mg/dl±27.1SD) than in the survivor group (29.1mg/dl±16.7).
There was no significant difference between the initial CRP level and the delayed CRP level obtained after starting treatment. But there was a highly significant difference (p value<0.001) between the initial (37.9mg/dl±16.1SD) and delayed (31.2mg/dl±17.4SD) serum lactate levels being lower after 24 hours of starting treatment.
On studying the relation between the CRP levels, lactate levels and lactate clearance % with the type of bacteria isolated we found that there was no significant difference between the Gram positive and Gram negative bacteria as regards to the level of initial CRP level and delayed CRP levels. But there was a significant difference between the Gram positive and Gram negative bacterial blood cultures as regards to the lactate clearance % as lower levels of lactate clearance were noticed in patients with Gram negative blood cultures (10.8%±12.3SD) than in Gram positive ones (17.15%±15.9SD) with p value=0.03.
As regards to the initial lactate levels there was a significant difference (p value= 0.003) between the Gram positive and Gram negative bacteria as higher levels of initial lactate was noticed with Gram negative bacteria (56.6mg/dl±18.9SD) than in Gram positive ones (35.1mg/dl±14.5SD). Also a significant difference between the 2 groups was noticed in the delayed lactate serum levels as higher levels were noticed in Gram negative infection (46.1mg/dl±9.07SD) than in Gram positive ones (32.2mg/dl±14.8SD).
There was a significant negative correlation between the lactate clearance % and I/T ratio and SGOT serum level. While there was as a significant negative correlation between the lactate clearance % and serum Na+ level. Other laboratory parameters showed no significant correlation with lactate clearance %.
Correlation between initial lactate levels and other parameters showed a significant negative correlation between the initial lactate levels and birth weight of the newborns, gestational age, serum hemoglobin level, platelets level and serum albumin level. But there was a significant positive correlation between the initial serum lactate levels and IT ratio, HSS score and clinical sepsis score.
Correlation between delayed lactate levels and other parameters shows that there was a significant negative correlation between delayed lactate serum levels and birth weight, gestational age, mean blood pressure, platelets count, serum albumin level and prothrombin time. But there was a significant positive correlation between the delayed lactate serum levels and initial CRP, IM ratio, IT ratio, HSS score and clinical sepsis score.
ROC curve analysis of the CRP level as prognostic factor for neonatal sepsis mortality shows area under the curve (AUC) =0.63 with a significant cut-off point ≥18mg/dl that was significant to predict mortality with a sensitivity of 84% and a weak specificity of 40% that might be due to non-specific rise of CRP levels with various conditions. Positive predictive value (PPV) of CRP value was 58% and negative predictive value (NPV) was 71%.
ROC curve analysis of initial serum lactate level shows area under the curve of 0.89 and a cut-off point of 33.3 mg/dl which was highly significant in prediction of mortality. The sensitivity of this point was 90.9% (high) and the specificity was moderate (76%). And positive predictive value was 99% and negative predictive value of 80%.
ROC curve analysis of serum 24hours lactate clearance showed an area under the curve of 0.88. The cut-off point of lactate clearance was 24.5% which was highly significant (p value < 0.001) in prognosis of mortality. The lactate clearance at this point was 93.9% in prediction of mortality but had a moderate specificity of 76% as lactate levels can be affected by many factors including hypoxia and also half life of serum lactate is not quite known. The ability of lactate clearance to detect positive cases of mortality was good (PPV =79.3%). The ability to detect negative cases was more potent (NPV=90.9%).
Roc curve analysis of delayed serum lactate determining its specificity and sensitivity in prediction of mortality showed a cutoff point ≥26.35mg/dl. This showed a sensitivity of 100% and a specificity of 83.8% with a negative predictive value of 98.2% and positive predictive value of 84.3%.
When combing the markers of our study we found that when combing the initial CRP and the initial serum lactate together there was 93.9% sensitivity and 64.9% specificity with negative predictor value of 13.3% and positive predictor value of 56.3%.