Acute Chorioamnionitis Does Not Predict Neonatal Sepsis

Diana Mahoney

November 06, 2012

BOSTON — Despite a strong correlation between histopathologic chorioamnionitis (HC) and neonatal sepsis, the finding is not an independent marker of the infectious complication and should not be used on its own to diagnose early-onset sepsis in full-term infants, a new study has shown.

HC, defined by the presence of acute histologic changes in the amniotic membrane and chorion of the placenta, is one of the most important findings in placental evaluation and is known to have an association with neonatal sepsis, but the specificity of HC for the prediction of clinical sepsis in full-term infants has not been well documented in the literature. Laleh Hakima, DO, from Winthrop University Hospital in Mineola, New York, explained this here at the American Society for Clinical Pathology (ASCP) 2012 Annual Meeting.

The diagnostic utility of HC is controversial, given the dearth of studies evaluating specific histopathologic features of inflammation both from the maternal side (HC) and the fetal response (funisitis or chorionic plate vasculitis [CPV]), she noted.

To investigate the predictive power of HC, funisitis, and CPV in diagnosing early-onset sepsis in full-term infants, Dr. Hakima and colleagues reviewed the surgical pathology records of symptomatic full-term infants admitted to the hospital's neonatal intensive care unit in 2008 and 2009.

They used the Redline classification system to stage the histologic findings, and they obtained clinical data, including laboratory markers for sepsis (defined as a C-reactive protein [CPR] level of 10 or higher and an immature to total neutrophil [I/T] ratio of 0.20 or higher), duration of antibiotic treatment, and blood culture results. Fisher's exact, Chi-square, and Cochran–Armitage trend tests were used to evaluate the association between clinical sepsis and HC, funisitis, and CPV, and regression analysis models assessed the predictive ability of HC in the diagnosis of sepsis.

Of the 388 patients for whom complete data were available, 100 (26%) were clinically septic and 288 (74%) were not, Dr. Hakima reported. Among the septic patients, 78.5% had elevated CRP levels and 36.6% had elevated I/T ratios; among the nonseptic patients, 19.1% had elevated CRP levels and 9.5% had elevated I/T ratios.

Regarding placental findings, "histopathologic chorioamnionitis was present in 47 of the 100 septic infants [47%], compared with 71 of the nonseptic infants [25%], but increasing stages of [HC] did not differentiate between septic and nonseptic patients with statistical significance," she noted. Furthermore, CPV associated with HC was present in more septic than nonseptic infants (24.0% vs 8.7%), as was funisitis (68.0% vs 52.0%). However, increasing stages of funisitis "did not differentiate between septic and nonseptic infants," she explained.

Because there were fewer than 4 cases of stage 3 HC, "the numbers were insufficient for statistical analysis," Dr. Hakima told Medscape Medical News.

In multiple regression analysis of the predictive ability of HC, CRP, and the I/T ratio to diagnosis sepsis, "the area under the curve for CRP and the I/T ratio together increased, relative to the area under the curve for CRP alone, but the ROC curve for CRP, the I/T ratio, and HC overlapped that of CRP and the I/T ratio only, and the area under the curve did not improve the predictive ability of CRP and the I/T ratio to diagnose sepsis," Dr. Hakima reported.

Although the findings confirm a strong correlation between HC and clinical sepsis, and demonstrated that, among patients with HC, septic patients are more likely to have CPV, "the presence of HC alone does not improve the predictive ability of CRP or the I/T ratio to diagnose sepsis," Dr. Hakima stated. "For this reason, [HC] should not be used as an independent marker to diagnose sepsis [and] avoid unnecessary antibiotic treatment and associated complications in full-term infants."

The results are consistent with those of a previous examination of factors associated with HC in a population of low-risk women with term deliveries (PLoS One. 2012;7[3]:e31819). In a secondary analysis among 195 low-risk women with term pregnancies enrolled in that randomized trial, which included histologic and microbiologic evaluation of placentas, the investigators concluded that HC at term is most often a noninfectious inflammatory process. The first author of that study was Drucilla Jane Roberts, MD, from the Massachusetts General Hospital in Boston.

"Our ability to predict fetal infection is low, and we have always thought those with [HC] and fetal inflammation are the ones to worry about," Dr. Roberts, who was not involved in the current study, told Medscape Medical News. These findings "emphasize the need to use Redline staging and grading so that fetal inflammation is always diagnosed. They also point to the need for additional studies examining the most appropriate diagnostic criteria for and clinical responses to acute chorioamnionitis in term pregnancies," she added.

The availability of an accurate method for diagnosing infection during labor could eliminate unnecessary antibiotic treatment for mothers and their infants, she said.

Dr. Hakima and Dr. Roberts have disclosed no relevant financial relationships.

American Society for Clinical Pathology (ASCP) 2012 Annual Meeting: Poster 18. Presented November 3, 2012.