Rick, Anne-Marie
(2021)
Assessment of Neonatal Risk Stratification Methods for the Detection of Early Onset Neonatal Sepsis.
Doctoral Dissertation, University of Pittsburgh.
(Unpublished)
This is the latest version of this item.
Abstract
Early onset sepsis (EOS) occurs infrequently in newborns but can result in life-long deficits or even death. There is tremendous uncertainty about how to best identify infected infants. We aimed to validate the new obstetric diagnoses for intraamniotic infection, collectively known as Triple I, for their ability to identify EOS among infants >35 weeks gestational age and compare it to other approaches. We first determined that the obstetric diagnosis, suspected intraamniotic infection, modestly improves identification of infants with EOS compared to clinical chorioamnionitis with a numerically higher sensitivity and significantly higher area under the receiver operating curve (AUC). This solidifies use of this diagnosis in obstetric and pediatric practice over previous criteria. However, its test characteristics were suboptimal with a sensitivity of only 53% (95%CI: 40-66) and an AUC of 0.752 (95%CI: 0.682-0.821). Next, we combined diagnosis of suspected intraamniotic infection with the infant’s clinical appearance after birth and assessed test characteristics of this categorical approach for EOS and compared it to the multivariate EOS risk calculator, an alternative, evidence-based approach to EOS screening. We identified that the categorical approach had sensitivity of 90% (95%CI: 79-96%) and AUC of 0.875 (0.825-0.924). While this approach identified EOS better than the calculator, the calculator maintained higher specificity. We then evaluated if placenta data can enhance specificity of the categorical approach. Among infants >35 weeks gestational age exposed to suspected intraamniotic infection in utero, we identified that combining absence of umbilical cord inflammation and placenta culture growth could successfully rule-out 90% of non-infected but exposed infants. However, the maximum benefit of incorporating placenta data occurs if it is obtained shortly after delivery, a practice that is not commonly done. In conclusion, we successfully validated that a categorical approach combining diagnosis of maternal suspected intraamniotic infection and infant clinical appearance will identify the majority of EOS cases. However, it lacks specificity. While this can be improved using placenta histopathology and culture, it would require significant practice change. As institutions re-consider their approach to EOS screening given recent guideline changes, it is necessary to evaluate the strengths and limitations of each approach.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
26 May 2021 |
| Date Type: |
Publication |
| Defense Date: |
14 April 2021 |
| Approval Date: |
26 May 2021 |
| Submission Date: |
29 April 2021 |
| Access Restriction: |
2 year -- Restrict access to University of Pittsburgh for a period of 2 years. |
| Number of Pages: |
91 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
School of Medicine > Clinical and Translational Science |
| Degree: |
PhD - Doctor of Philosophy |
| Thesis Type: |
Doctoral Dissertation |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
early onset sepsis; chorioamnionitis; infants |
| Date Deposited: |
26 May 2021 04:26 |
| Last Modified: |
26 May 2023 05:15 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/41098 |
Available Versions of this Item
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[feed]](/images/feed-icon-32x32.png){kind=icon}