Antibiotic administration reduced intra-amniotic inflammation 7 days after preterm premature rupture of the membranes with intra-amniotic infection

Abstract Objective Intra-amniotic infections increase the risk of preterm delivery and short- and long-term fetal morbidity; however, no consensus exists on the choice of antimicrobial agents as treatment for these infections. We aimed to examine the efficacy of intravenous administration of sulbactam/ampicillin (SBT/ABPC) and azithromycin (AZM) for intra-amniotic infection in patients with preterm premature rupture of membranes (PPROM). Methods This study followed a single-centered retrospective cohort design. We compared changes in interleukin 6 (IL-6) levels and the load of Ureaplasma species DNA in the amniotic fluid between singleton pregnancy patients with intra-amniotic infection (Group A) and without either intra-amniotic inflammation (IAI) or microbial invasion of the amniotic cavity (MIAC) (Group B) who developed PPROM between week 22, day 0 and week 33, day 6 of gestation and maintained pregnancy for ≥7 d after diagnosis (August 2014 to April 2020). Patients in Group A were treated with SBT/ABPC and AZM, whereas those in Group B were treated with ABPC and AZM or clarithromycin. Results Thirty-one patients with IAI and 48 patients without either IAI or MIAC at diagnosis of PPROM underwent pregnancy/delivery management at our hospital. Following the study population selection, we evaluated six patients in Group A and 13 patients in Group B. Amniotic fluid IL-6 concentrations at the initial amniocentesis were high, ranging from 11.7 ng/mL to 139.2 ng/mL, indicating a state of severe IAI in all six patients in Group A. In five of the six patients in Group A, the amniotic fluid cultures during the first amniocentesis included Ureaplasma species only. In both groups, the amniotic fluid IL-6 concentration at the follow-up amniocentesis was lower than that at the initial amniocentesis (Group A: follow-up median 3.06 ng/mL [quartiles, 1.75–6.74], initial median 30.53 ng/mL [quartiles, 15.60–67.07], p＝.03; Group B: follow-up median 0.40 ng/mL [quartiles, 0.18–0.69], initial median 0.96 ng/mL [quartiles, 0.65–1.42], p＝.005); Group A showed a greater decrease than Group B (p < .001). No difference was found between the microbial loads of Ureaplasma species DNA in the initial and follow-up amniocentesis (p = .13). Conclusions In patients with PPROM and intra-amniotic infection, IL-6 levels in the amniotic fluid decreased significantly from before antimicrobial administration to day 7. This decrease is thought to be mainly due to the effects of intravenous AZM. The efficacy of AZM in patients with PPROM needs to be further confirmed via randomized controlled studies in the future.


Introduction
Preterm premature rupture of membranes (PPROM), which refers to premature rupture of membranes occurring before week 37 of gestation, is said to be associated with approximately 1/3 of preterm births [1] and is considered the most common single factor causing preterm birth [2].
The current standard of management for PPROM patients at <34 weeks of gestation is generally elective treatment and expectant management, including the administration of antibiotics and adrenocortical steroids.Antibiotics are reported to prolong the duration of pregnancy, reduce maternal-infant infections, and lower fetal morbidity due to prematurity [10][11][12][13].However, only a few studies have demonstrated the efficacy of antimicrobial therapy in intra-amniotic infections.As an antibiotic treatment approach, concurrent administration of ampicillin (ABPC) and erythromycin (EM) is generally performed.ABPC is ineffective against Ureaplasma species (spp.), which are most commonly detected in intra-amniotic infection, and EM is likely ineffective [14][15][16].Azithromycin (AZM) has shown efficacy in animal experiments [17][18][19] as well as favorable transferability to the human placenta and myometrium [20,21], and treatment cases in humans have been reported [22,23].However, to our knowledge, there have been no studies directly evaluating the efficacy of AZM against intra-amniotic infection caused by Ureaplasma spp.Intravenous administration of clarithromycin (CLR), another macrolide antibiotic, to patients with intra-amniotic infection has been reported to reduce the load of Ureaplasma spp.DNA, eliminate bacteria, and attenuate or normalize IAI [9, 24,25].In addition, among antibiotics administered at the diagnosis of PPROM, the use of sulbactam/ampicillin (SBT/ABPC) for bacteria other than Ureaplasma spp. was reported to result in a longer delivery interval than ABPC [26], and the usefulness of antibiotics with broader antibacterial spectrum has been suggested [27,28].
The present study aimed to elucidate whether intravenous administration of SBT/ABPC and AZM reduces interleukin 6 (IL-6) and the load of Ureaplasma spp.DNA in the amniotic fluid of PPROM patients with intra-amniotic infection.

Materials and methods
This was a single-center retrospective cohort study.All procedures followed were in accordance with the ethical standards of the Institutional Review Board of the NHO Saga Hospital (approval number: R3-17) and with the Declaration of Helsinki and its later amendments.
The eligibility criteria for subjects were as follows: 1. Singleton pregnancy patients who developed PPROM between week 22, day 0, and week 33, day 6 of gestation (from August 2014 to April 2020).
2. Patients who underwent pregnancy/delivery management, including antibiotic administration, at our hospital and maintained pregnancy for �7 d after being diagnosed with PPROM.
The exclusion criteria were as follows: patients with an unclear time of membrane rupture, patients who underwent amniocentesis after amnioinfusion, patients who underwent amniocentesis more than 24 h after being diagnosed with PPROM, patients with unclear results of amniotic fluid culture, patients diagnosed with IAI alone or MIAC alone because of the lack of constant antimicrobial agents and the limited number of cases, patients who delivered <7 d after being diagnosed with PPROM, patients who were not administered previously arranged antibiotics, patients who did not undergo follow-up amniocentesis for any reason, and patients without either IAI or MIAC who had the secondary intra-amniotic infection <7 d after PPROM.
The diagnosis of PPROM was confirmed by a history of watery vaginal discharge and the following: amniotic fluid leakage from the cervical os, vaginal retention of amniotic fluid, and a positive insulin-like growth factor binding protein-1 test after speculum examination.
Indications for amniocentesis included a diagnosis of PPROM and follow-up (7 ± 2 d after PPROM) as a part of the clinical management of singleton pregnancies with PPROM.Amniocentesis was additionally performed in patients suspected of exacerbation of intra-amniotic inflammation or infection.
Amniocentesis was performed after explaining the necessity of the test as well as the complications to the patients and obtaining written informed consent.For the amniotic fluid test, 5 ml of amniotic fluid was collected transabdominal with a 25 G percutaneous transhepatic cholangiography needle under transabdominal ultrasound guidance.
For bacterial screening, aerobic and anaerobic cultures and the Urea Arginine LYO2 Kit (Urea-Arginine LYO2 134, Bio Merieux, France) were used to culture Mycoplasma hominis and Ureaplasmas spp.following Gram staining.For patients who tested positive for Ureaplasma spp.and Mycoplasma hominis, the stored samples were transferred to the Department of Developmental Medicine of the Research Institute at Osaka Women's and Children's Hospital (Osaka, Japan), and genetic tests were performed to identify Ureaplasma urealyticum, Ureaplasma parvum, and Mycoplasma hominis [29].
The load of Ureaplasma spp.DNA (copies/1 ml amniotic fluid) was determined using the following method: DNA was extracted from amniotic fluid samples using an automated DNA extraction system (Maxwell RSC V R instrument, Promega, Japan) with the Maxwell RSCV R Blood DNA kit (Promega).The DNA copy number of Ureaplasma spp. was measured using quantitative realtime polymerase chain reaction (qPCR) in the QuantStudio 5 Real-Time PCR System (Thermo Fisher Scientific, Waltham, MA, USA) according to a previous report [30].Briefly, the following primers and probes were used: 16S-Urea F1 (AGGCATGCGTCTAGGGTAGGA), 16S-Urea R1 (ACGTTCTCGTAGGGATACCTTGTTA), and 16S-Urea-FAM-MGB probe (FAM-CGGTGACTGGAGTTAA-MGB).The amplification reaction mixtures comprised 5 mL of TaqMan TM Fast Advanced Master Mix (Thermo Fisher Scientific), primers (0.9 mM each), probe (Thermo Fisher Scientific, 0.25 mM), and 1 mL of sample DNA and were adjusted to 10 mL with nuclease-free water.The qPCR protocol was 95 � C for 20 s, followed by 50 cycles at 95 � C for 1 s and at 58 � C for 20 s.The standard Ureaplasma 16S rRNA DNA (10-1,000,000 copies) was simultaneously amplified.
Regarding the selection of antibiotics, patients with PPROM were intravenously administered 2 g of ABPC four times a day for two days, followed by oral administration of 1000 mg of amoxicillin (AMPC) for five days þ 400 mg of clarithromycin (CLR) for seven days.Patients with membrane rupture at <27 weeks of gestation were intravenously administered 500 mg of azithromycin (AZM) once daily for three days instead of CLR, regardless of IAI.When IAI or positive Gram staining was detected, patients were intravenously administered 1.5 g of SBT/ABPC four times a day for three days þ 500 mg of AZM once daily for three days, and antibiotics were changed based on the result of amniotic fluid culture (Figure 1).
Maternal steroid administration was performed at the time of diagnosis of PPROM.In principle, tocolytics were not used.Indications for delivery were based on the onset of labor pains, non-reassuring fetal status, exacerbation of intra-amniotic infection, clinical chorioamnionitis, and judgment of the attending physician.
A correlation between chemiluminescent enzyme immunoassay (CLEIA) and electrochemiluminescent immunoassay (ECLIA) for IL-6 concentration was high (r ¼ 1.00), and IAI was defined as an amniotic fluid IL-6 concentration �3.0 ng/mL measured using ECLIA [31], and MIAC was defined as patients positive for aerobic, anaerobic, Mycoplasma hominis, and Ureaplasmas spp.cultures.Intra-amniotic infection was defined as patients positive for both IAI and MIAC.
The primary study endpoints were the change and rate of change in amniotic fluid IL-6 levels between the initial and follow-up amniocentesis.Amniotic fluid IL-6 concentrations at the initial and follow-up amniocentesis in patients with intra-amniotic infection and without MIAC or IAI were analyzed using the Wilcoxon signed-rank sum test.Furthermore, changes in amniotic fluid IL-6 concentration between the two groups were compared by the aligned rank transform for nonparametric factorial analysis of variance (ANOVA).
The secondary study endpoints were the comparison of the changes in the load of Ureaplasma spp.DNA in the amniotic fluid of patients with an intraamniotic infection between the initial and follow-up amniocentesis.Changes in the load of Ureaplasma spp.DNA in the amniotic fluid were compared using the Mann-Whitney U test.
For the comparison of clinical background, pregnancy, and fetal outcomes between the two groups, the Mann-Whitney U test was performed for continuous variables, and Fisher's probability test and v 2 test were performed for nominal variables.p < .05 was considered statistically significant.
EZR software (Saitama Medical Center, Jichi Medical University, Saitama, Japan), a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria), was used for statistical processing [32].

Results
A flow chart of the study population selection is shown in Figure 2. Thirty-one patients with intraamniotic infection and 48 patients without either IAI or MIAC at PPROM diagnosis underwent pregnancy/ delivery management at our hospital.Twenty-four patients with intra-amniotic infection and 26 without either IAI or MIAC delivered <7 d after PPROM; following the study population selection, we evaluated six patients with intra-amniotic infection (Group A) and 13 patients without either IAI or MIAC (Group B).Overall, 4 of 13 patients in Group B experienced PPROM at <27 weeks of gestation.
All bacterial species identified in amniotic fluid cultures during the initial amniocentesis are shown in Supplemental Table 1.Mycoplasma hominis or Ureaplasma spp.were found in 19 of 31 patients with intra-amniotic infection (61%).Additionally, 6 of 16 patients with intra-amniotic infection caused solely by Ureaplasma spp.delivered >7 d after PPROM, whereas only 1 of 12 patients with intraamniotic infection caused by aerobic and anaerobic bacteria delivered >7 d after PPROM.
Patient background, pregnancy outcomes, and perinatal data are shown in Table 1.Amniotic fluid IL-6 concentrations at the initial amniocentesis were high, ranging from 11.7 ng/mL to 139.2 ng/mL, indicating a state of severe IAI in all six patients in Group A [31,33].We note that there was no statistically significant difference in neonatal outcomes between the two groups.
Background, pregnancy outcomes, and perinatal data for patients who delivered <7 d after PPROM are shown in Table 2. Amniotic fluid IL-6 concentrations at the initial amniocentesis were high (median: 42.53 ng/mL [quartiles, 14.70-83.87] in patients with intra-amniotic Patients with intra-amniotic infection had a lower gestational age at birth and a higher prevalence of maternal inflammatory response, fetal inflammatory response, and fetal inflammatory response syndrome than those without either IAI or MIAC. In both groups, the amniotic fluid IL-6 concentration at the follow-up amniocentesis was lower than that at the initial amniocentesis (Group A: follow-up median 3.06 ng/mL [quartiles, 1.75-6.74],initial median 30.53 ng/mL [quartiles, 15.60-67.07],p＝.03; Group B: follow-up median 0.40 ng/mL [quartiles, 0.18-0.69],initial median 0.96 ng/mL [quartiles, 0.65-1.42],p＝.005) (Figure 3).
Changes in amniotic fluid IL-6 concentration between the two groups were compared by the aligned rank transform for nonparametric factorial ANOVA, and a significantly greater decrease was found in patients in Group A than in Group B (p < .001).
Figure 4 shows the changes in the load of Ureaplasma spp.DNA in the amniotic fluid of four patients in Group A between the initial and follow-up amniocentesis.No difference was found in the concentrations of Ureaplasma spp.DNA in the amniotic fluid from initial and follow-up amniocenteses, possibly because of the extremely small sample size (n ¼ 4).
Table 3 shows the details of the clinical course of the six patients in Group A. The amniotic fluid cultures during the initial amniocentesis identified Ureaplasma spp. in five patients.Four of five patients showed negative amniotic fluid bacterial cultures during follow-up with antibiotic treatment; however, two of these patients showed positive amniotic fluid cultures during the subsequent amniocentesis.Two patients showed persistent positive amniotic fluid cultures with Ureaplasma spp. at 14 and 49 d.

Discussion
In patients with PPROM with intra-amniotic infection who received SBT/ABPC þ AZM and could maintain their pregnancy for >7 d, the decrease in IL-6 levels in the amniotic fluid from before antimicrobial administration to day 7 was more pronounced than that in those without either IAI or MIAC.
A previous study evaluated the effect of intravenous administration of CLR, a macrolide antimicrobial agent similar to AZM, on intra-amniotic infection, assessed by amniotic fluid culture and polymerase chain reaction [6].It was reported that 40 of 270 patients who had developed PPROM between week 24, day 0, and week 33, day 6 of gestation successfully   underwent follow-up amniocentesis �7 days after the initial Furthermore, CLR was administered intravenously to seven patients with intra-amniotic infection and seven patients with aseptic IAI, and patients without either MIAC or IAI (n ¼ 26) were treated with benzylpenicillin or clindamycin.This resulted in extensively lower amniotic fluid IL-6 concentrations at the follow-up amniocentesis than at the initial amniocentesis in all cases.Scholars found Ureaplasma spp. in six of seven patients with intraamniotic infection, and the load of Ureaplasma spp.DNA in amniotic fluid was markedly lower at the follow-up amniocentesis than at the initial amniocentesis, suggesting that intravenous administration of CLR to patients with intra-amniotic infection/inflammation can alleviate inflammation and eradicate bacteria.AZM, which is a macrolide antibiotic semi-synthesized from erythromycin-like CLR, has excellent tissue transferability and a long half-life in the blood.Ureaplasma spp.Has been reported to be highly susceptible to AZM, as with CLR.Regarding the therapeutic effect of AZM on intra-amniotic infection caused by Ureaplasma spp., there have been useful reports of the effect of AZM administration during surgery (mainly intravenous administration) on preventing postoperative infections in patients with emergency Cesarean section (with premature rupture of membranes or after the onset of labor pains), who are at high risk of intra-amniotic infection [34][35][36][37].A multicenter joint study in the United States conducted a randomized controlled trial (RCT) enrolling approximately 2000 patients undergoing emergency Cesarean section and administered conventional antibiotics alone or conventional antibiotics with intravenous administration of AZM and reported a low postoperative infection risk of 0.51 in the AZM administration group [38].
We noted that AZM has favorable tissue transferability even in the human placenta, and its concentration is maintained for a long period of time.The AZM concentration in the placenta and myometrium 24-72 h after its administration was shown to be approximately 20 times higher than the maternal serum concentration [20].The effectiveness of AZM treatment for intra-amniotic infection caused by Ureaplasma spp. is likely due to its favorable transferability to the placenta and long-term maintenance of high drug concentrations.Regarding intra-amniotic infection caused mainly by aerobic and anaerobic bacteria, pregnancy could be maintained for �7 d in only one of 12 patients (Lactobacillus acidophilus) in the present study, and in one of 17 patients (Anaerococcus tetradius) in the study by Kacerovsky et al. [6], suggesting that of pregnancy with SBT/ ABPC treatment may be difficult.
In this and previous studies, patients without either MIAC or IAI exhibited considerably lower amniotic fluid IL-6 concentrations at the follow-up amniocentesis than at the initial amniocentesis.Although the effect of maternal steroid administration could not be ruled out, a decrease in IL-6 concentration was observed seven days after steroid administration; this was found even in patients without steroid administration, suggesting that the observed decrease was due to the anti-inflammatory actions of AZM and CAM themselves [6, 24,25].
In the present study, a considerably greater decrease in amniotic fluid IL-6 concentration was observed in patients with intra-amniotic infection than in patients without either MIAC or IAI, suggesting that inflammation was controlled by the bactericidal effect of AZM.
Strengths of this study include the following: primarily, this is the first study to demonstrate the efficacy of AZM and SBT/ABPC administration in PPROM patients with intra-amniotic infection, using serial evaluation of amniotic fluid IL-6 concentration and a load of Ureaplasma spp.DNA.Second, we compared the change in IL-6 concentration between the initial and follow-up amniocentesis and between patients with intra-amniotic infection and without either MIAC or IAI.
The limitations of this study should be considered while interpreting the findings of the current research.First, we could not evaluate the effect of antibiotics in patients with a PPROM-to-delivery interval of fewer than 7 days.It could be that antimicrobial therapy was less effective in the 24/31 patients with intraamniotic infection who delivered in <7 d.However, even in the 48 patients without either IAI or MIAC, 26 delivered in <7 d, suggesting that the 24 patients with intra-amniotic infection may have delivered early due to causes other than intra-amniotic infection.Therefore, this needs to be investigated in future research.Second, the load of Ureaplasma spp.DNA in the amniotic fluid could be measured in a few patients with intra-amniotic infection, resulting in no statistically significant difference.Third, we did not use molecular methods to detect bacteria and fungi.
In conclusion, IL-6 levels in the amniotic fluid decreased significantly from before antimicrobial administration to day 7 in patients with PPROM and intra-amniotic infection.This decrease is thought to be mainly due to the effects of intravenous AZM.The  efficacy of AZM in patients with PPROM needs to be confirmed in the future via RCTs.

Figure 3 .
Figure 3. Concentrations of interleukin-6 in amniotic fluid from a subset of patients with preterm premature rupture of the membranes with intraamniotic infection (A) and without either microbial invasion of the amniotic cavity or intraamniotic inflammation (B) who underwent a follow-up amniocentesis.Patients who received steroids are indicated by solid lines, whereas those who did not receive steroids are indicated by dotted lines.

Figure 4 .
Figure 4. Load of Ureaplasma spp.DNA in amniotic fluid in patients with preterm premature rupture of the membranes with intra-amniotic infection who underwent a follow-up amniocentesis.

Table 1 .
Demographic and clinical characteristics and perinatal data of the study patients.
Data are expressed as medians [interquartile ranges] and numbers (%) where appropriate.IAI: intra-amniotic inflammation; MIAC: microbial invasion of the amniotic cavity.

Table 2 .
Demographic and clinical characteristics and perinatal data of the patients who < 7 d after PPROM.
Data are expressed as medians [interquartile ranges] and numbers (%) where appropriate.IAI: intra-amniotic inflammation; MIAC: microbial invasion of the amniotic cavity.

Table 3 .
Clinical characteristics and outcomes of six patients with intra-amniotic infection who were treated with SBT/ABPC and AZM and delivered after 7 d or more following PPROM.