The rapid diagnosis of intraamniotic infection with nanopore sequencing.

Abstract

Background: Intraamniotic infection (defined as intraamniotic inflammation with microorganisms) is an important cause of the preterm labor syndrome. Methods for the detection of microorganisms in amniotic fluid are culture and/or polymerase chain reaction. However, both methods take time, and results are rarely available for clinical decision-making. Nanopore sequencing technology offers real-time, long-read sequencing that can produce rapid results.

Objectives: To determine 1) the diagnostic performance of the 16S rDNA nanopore sequencing method for the identification of microorganisms in patients with intraamniotic inflammation; and 2) the relationship between microbial burden and the intensity of the amniotic fluid inflammatory response.

Study design: We performed a prospective cohort study that included singleton pregnancies presenting with symptoms of preterm labor with intact membranes or of preterm prelabor rupture of the membranes. Amniotic fluid samples were obtained for the evaluation of bacteria in the amniotic cavity using cultivation and polymerase chain reaction-based 16S Sanger sequencing methods. Participants were classified into 4 groups according to the results of an amniotic fluid culture, 16S Sanger sequencing, and an amniotic fluid interleukin-6 concentration: 1) no intraamniotic infection and intraamniotic inflammation (interleukin-6 <2.6 ng/mL and no microorganisms in the amniotic cavity determined by culture or 16S Sanger sequencing); 2) microbial invasion of the amniotic cavity without intraamniotic inflammation, defined by the presence of bacteria detected by culture or 16S Sanger sequencing; 3) sterile intraamniotic inflammation (interleukin-6 ≥2.6 ng/mL without microbial invasion of the amniotic cavity); and 4) intraamniotic infection (interkeukin-6 ≥2.6 ng/mL with microbial invasion of the amniotic cavity). Patients who underwent a midtrimester amniocentesis, had no intraamniotic infection or intraamniotic inflammation, and delivered at term represented the control group. 16S rDNA nanopore sequencing was performed and the diagnostic indices for the identification of intraamniotic infection were determined. Bioinformatic analysis was carried out to identify microorganisms, and a read count of at least 100 or a read count exceeding that of the background species from the control group, along with a relative abundance of no less than 1%, was used.

Results: 1) 16S nanopore sequencing had a sensitivity of 88.9% (8/9), specificity of 95.4% (41/43), positive predictive value of 80.0% (8/10), negative predictive value of 97.6% (41/42), positive likelihood ratio of 19.1 (95% CI, 4.8-75.4), negative likelihood ratio of 0.1 (95% CI, 0.02-0.7), and an accuracy of 94.2% (49/52) for the identification of intraamniotic infection [prevalence, 17% (9/52)]; 2) the microbial load determined by 16S nanopore sequencing had a strong positive correlation with the intensity of an intraamniotic inflammatory response (amniotic fluid interleukin-6 concentration; Spearman's correlation 0.9; p = 0.002); and 3) a subgroup of patients with intraamniotic inflammation did not have bacteria determined by culture, Sanger sequencing, or nanopore 16S, thus confirming the existence of sterile intraamniotic inflammation.

Conclusion: 16S nanopore sequencing has high diagnostic indices, predictive values, likelihood ratios, and accuracy in the diagnosis of intraamniotic infection.